H3.9 top features

Top feature 0 in H3.9: (feature 20873)

TOP ACTIVATIONS
MAX = 4.493

in
Token in
Feature activation+0.000
malware
Token malware
Feature activation+0.000
may
Token may
Feature activation+2.886
sound
Token sound
Feature activation+1.181
frightening
Token frightening
Feature activation+0.000
,
Token,
Feature activation+4.493
it
Token it
Feature activation+4.124
should
Token should
Feature activation+0.000
be
Token be
Feature activation+0.000
noted
Token noted
Feature activation+0.000
that
Token that
Feature activation+0.000
spaces
Token spaces
Feature activation+0.000
than
Token than
Feature activation+0.000
the
Token the
Feature activation+0.000
old
Token old
Feature activation+0.000
ones
Token ones
Feature activation+0.000
),
Token),
Feature activation+4.484
neither
Token neither
Feature activation+0.710
is
Token is
Feature activation+0.000
as
Token as
Feature activation+0.000
impressive
Token impressive
Feature activation+0.000
as
Token as
Feature activation+0.000
experienced
Token experienced
Feature activation+0.000
a
Token a
Feature activation+0.000
decade
Token decade
Feature activation+0.000
of
Token of
Feature activation+0.000
tyranny
Token tyranny
Feature activation+0.000
,
Token,
Feature activation+4.459
we
Token we
Feature activation+4.196
have
Token have
Feature activation+1.404
also
Token also
Feature activation+0.212
seen
Token seen
Feature activation+0.000
a
Token a
Feature activation+0.000
while
Token while
Feature activation+0.000
it
Token it
Feature activation+1.330
eventually
Token eventually
Feature activation+0.000
proved
Token proved
Feature activation+0.000
true
Token true
Feature activation+0.000
,
Token,
Feature activation+4.377
Blow
Token Blow
Feature activation+0.000
was
Token was
Feature activation+0.000
upset
Token upset
Feature activation+0.000
reporters
Token reporters
Feature activation+0.000
were
Token were
Feature activation+0.000
per
Token per
Feature activation+0.000
rules
Token rules
Feature activation+0.000
in
Token in
Feature activation+0.000
government
Token government
Feature activation+0.000
colleges
Token colleges
Feature activation+0.000
,
Token,
Feature activation+4.354
the
Token the
Feature activation+2.865
private
Token private
Feature activation+0.000
un
Token un
Feature activation+0.000
a
Tokena
Feature activation+0.000
ided
Tokenided
Feature activation+0.000
recreational
Token recreational
Feature activation+0.000
marijuana
Token marijuana
Feature activation+0.000
use
Token use
Feature activation+0.000
by
Token by
Feature activation+0.000
humans
Token humans
Feature activation+0.000
,
Token,
Feature activation+4.297
those
Token those
Feature activation+2.756
policies
Token policies
Feature activation+0.000
are
Token are
Feature activation+0.000
silent
Token silent
Feature activation+0.053
when
Token when
Feature activation+0.000
while
Token while
Feature activation+0.000
Hare
Token Hare
Feature activation+0.000
has
Token has
Feature activation+0.524
shown
Token shown
Feature activation+0.000
improvement
Token improvement
Feature activation+0.000
,
Token,
Feature activation+4.286
he
Token he
Feature activation+4.039
's
Token's
Feature activation+2.142
not
Token not
Feature activation+1.047
Jordan
Token Jordan
Feature activation+0.000
Lynch
Token Lynch
Feature activation+0.000
to
Token to
Feature activation+0.000
come
Token come
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
conclusion
Token conclusion
Feature activation+0.000
,
Token,
Feature activation+4.265
it
Token it
Feature activation+4.202
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.059
only
Token only
Feature activation+0.398
the
Token the
Feature activation+0.000
number
Token number
Feature activation+0.000
of
Token of
Feature activation+0.000
tax
Token tax
Feature activation+0.000
brackets
Token brackets
Feature activation+0.000
,
Token,
Feature activation+4.223
he
Token he
Feature activation+3.971
criticized
Token criticized
Feature activation+0.000
him
Token him
Feature activation+0.000
for
Token for
Feature activation+0.000
not
Token not
Feature activation+0.000
to
Token to
Feature activation+0.000
breast
Token breast
Feature activation+0.000
feed
Tokenfeed
Feature activation+0.000
their
Token their
Feature activation+0.000
baby
Token baby
Feature activation+0.000
,
Token,
Feature activation+4.208
"
Token "
Feature activation+0.000
that
Tokenthat
Feature activation+0.657
in
Token in
Feature activation+0.000
no
Token no
Feature activation+0.000
way
Token way
Feature activation+0.000
come
Token come
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
conclusion
Token conclusion
Feature activation+0.000
,
Token,
Feature activation+4.265
it
Token it
Feature activation+4.202
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.059
only
Token only
Feature activation+0.398
because
Token because
Feature activation+0.000
a
Token a
Feature activation+0.000
decade
Token decade
Feature activation+0.000
of
Token of
Feature activation+0.000
tyranny
Token tyranny
Feature activation+0.000
,
Token,
Feature activation+4.459
we
Token we
Feature activation+4.196
have
Token have
Feature activation+1.404
also
Token also
Feature activation+0.212
seen
Token seen
Feature activation+0.000
a
Token a
Feature activation+0.000
mass
Token mass
Feature activation+0.000
the
Token the
Feature activation+0.000
curtain
Token curtain
Feature activation+0.000
on
Token on
Feature activation+0.000
theatrical
Token theatrical
Feature activation+0.000
films
Token films
Feature activation+0.000
,
Token,
Feature activation+4.187
he
Token he
Feature activation+3.961
is
Token is
Feature activation+1.479
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
w
Tokenw
Feature activation+0.000
feel
Token feel
Feature activation+0.000
like
Token like
Feature activation+0.000
everything
Token everything
Feature activation+0.000
is
Token is
Feature activation+0.000
fine
Token fine
Feature activation+0.000
,
Token,
Feature activation+4.182
given
Token given
Feature activation+0.000
our
Token our
Feature activation+0.000
comfortable
Token comfortable
Feature activation+0.000
distance
Token distance
Feature activation+0.000
from
Token from
Feature activation+0.000
malware
Token malware
Feature activation+0.000
may
Token may
Feature activation+2.886
sound
Token sound
Feature activation+1.181
frightening
Token frightening
Feature activation+0.000
,
Token,
Feature activation+4.493
it
Token it
Feature activation+4.124
should
Token should
Feature activation+0.000
be
Token be
Feature activation+0.000
noted
Token noted
Feature activation+0.000
that
Token that
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
last
Token last
Feature activation+0.000
six
Token six
Feature activation+0.000
World
Token World
Feature activation+0.000
Series
Token Series
Feature activation+0.000
titles
Token titles
Feature activation+0.000
,
Token,
Feature activation+4.123
Ba
Token Ba
Feature activation+0.000
er
Tokener
Feature activation+0.000
sees
Token sees
Feature activation+0.164
the
Token the
Feature activation+0.000
Mission
Token Mission
Feature activation+0.000
within
Token within
Feature activation+0.000
10
Token 10
Feature activation+0.000
feet
Token feet
Feature activation+0.000
of
Token of
Feature activation+0.000
entrances
Token entrances
Feature activation+0.000
,
Token,
Feature activation+4.114
many
Token many
Feature activation+1.932
people
Token people
Feature activation+0.000
openly
Token openly
Feature activation+0.000
disob
Token disob
Feature activation+0.000
ey
Tokeney
Feature activation+0.000
isolate
Token isolate
Feature activation+0.000
themselves
Token themselves
Feature activation+0.000
from
Token from
Feature activation+0.000
natural
Token natural
Feature activation+0.000
environments
Token environments
Feature activation+0.000
they
Token they
Feature activation+4.107
often
Token often
Feature activation+0.820
create
Token create
Feature activation+0.000
whole
Token whole
Feature activation+0.000
new
Token new
Feature activation+0.000
ones
Token ones
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
former
Token former
Feature activation+0.000
Apprentice
Token Apprentice
Feature activation+0.000
boss
Token boss
Feature activation+0.000
,
Token,
Feature activation+4.089
S
Token S
Feature activation+0.000
.
Token.
Feature activation+0.000
N
TokenN
Feature activation+0.000
.
Token.
Feature activation+0.000
L
TokenL
Feature activation+0.000
carry
Token carry
Feature activation+0.000
unwanted
Token unwanted
Feature activation+0.000
pregnancies
Token pregnancies
Feature activation+0.000
to
Token to
Feature activation+0.000
term
Token term
Feature activation+0.000
,
Token,
Feature activation+4.085
single
Token single
Feature activation+0.000
and
Token and
Feature activation+0.000
poor
Token poor
Feature activation+0.000
moms
Token moms
Feature activation+0.000
should
Token should
Feature activation+0.037

Top DFA by src position
MAX = 5.592

<|endoftext|>
Token<|endoftext|>
Feature activation+0.285
Top resid features:
Ċ
TokenĊ
Feature activation-0.018
Top resid features:
But
TokenBut
Feature activation+0.016
Top resid features:
while
Token while
Feature activation+5.435
Top resid features:
an
Token an
Feature activation+0.058
Top resid features:
over
Token over
Feature activation+0.019
Top resid features:
700
Token 700
Feature activation+0.004
Top resid features:
%
Token%
Feature activation+0.006
Top resid features:
surge
Token surge
Feature activation+0.014
Top resid features:
ahn
Tokenahn
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation-0.022
Top resid features:
Ċ
TokenĊ
Feature activation-0.024
Top resid features:
But
TokenBut
Feature activation-0.061
Top resid features:
while
Token while
Feature activation+5.438
Top resid features:
San
Token San
Feature activation-0.001
Top resid features:
Francisco
Token Francisco
Feature activation+0.010
Top resid features:
and
Token and
Feature activation+0.047
Top resid features:
Berlin
Token Berlin
Feature activation+0.005
Top resid features:
both
Token both
Feature activation+0.041
Top resid features:
to
Token to
Feature activation+0.007
Top resid features:
stress
Token stress
Feature activation+0.013
Top resid features:
the
Token the
Feature activation+0.007
Top resid features:
fact
Token fact
Feature activation-0.010
Top resid features:
that
Token that
Feature activation+0.001
Top resid features:
while
Token while
Feature activation+5.125
Top resid features:
we
Token we
Feature activation+0.043
Top resid features:
have
Token have
Feature activation+0.048
Top resid features:
indeed
Token indeed
Feature activation+0.051
Top resid features:
experienced
Token experienced
Feature activation+0.048
Top resid features:
a
Token a
Feature activation+0.028
Top resid features:
legitimacy
Token legitimacy
Feature activation+0.002
Top resid features:
?
Token?
Feature activation+0.020
Top resid features:
Sure
Token Sure
Feature activation-0.005
Top resid features:
,
Token,
Feature activation+0.059
Top resid features:
and
Token and
Feature activation+0.048
Top resid features:
while
Token while
Feature activation+4.952
Top resid features:
it
Token it
Feature activation-0.021
Top resid features:
eventually
Token eventually
Feature activation+0.142
Top resid features:
proved
Token proved
Feature activation+0.296
Top resid features:
true
Token true
Feature activation+0.061
Top resid features:
,
Token,
Feature activation-0.008
Top resid features:
Sw
Token Sw
Feature activation+0.002
Top resid features:
amin
Tokenamin
Feature activation+0.003
Top resid features:
athan
Tokenathan
Feature activation+0.004
Top resid features:
said
Token said
Feature activation+0.118
Top resid features:
that
Token that
Feature activation+0.054
Top resid features:
while
Token while
Feature activation+5.097
Top resid features:
admission
Token admission
Feature activation+0.005
Top resid features:
process
Token process
Feature activation+0.005
Top resid features:
and
Token and
Feature activation+0.034
Top resid features:
fee
Token fee
Feature activation+0.007
Top resid features:
structure
Token structure
Feature activation+0.001
Top resid features:
Michael
Token Michael
Feature activation+0.006
Top resid features:
Ames
Token Ames
Feature activation+0.007
Top resid features:
pointed
Token pointed
Feature activation+0.019
Top resid features:
out
Token out
Feature activation+0.039
Top resid features:
that
Token that
Feature activation+0.080
Top resid features:
while
Token while
Feature activation+5.378
Top resid features:
a
Token a
Feature activation+0.057
Top resid features:
growing
Token growing
Feature activation+0.005
Top resid features:
number
Token number
Feature activation-0.014
Top resid features:
of
Token of
Feature activation+0.014
Top resid features:
states
Token states
Feature activation+0.007
Top resid features:
in
Token in
Feature activation+0.002
Top resid features:
recent
Token recent
Feature activation+0.011
Top resid features:
years
Token years
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.035
Top resid features:
and
Token and
Feature activation+0.004
Top resid features:
while
Token while
Feature activation+4.908
Top resid features:
Hare
Token Hare
Feature activation-0.010
Top resid features:
has
Token has
Feature activation+0.118
Top resid features:
shown
Token shown
Feature activation+0.245
Top resid features:
improvement
Token improvement
Feature activation-0.065
Top resid features:
,
Token,
Feature activation+0.138
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.088
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
book
Token book
Feature activation+0.003
Top resid features:
shows
Token shows
Feature activation+0.008
Top resid features:
that
Token that
Feature activation+0.010
Top resid features:
while
Token while
Feature activation+5.012
Top resid features:
Car
Token Car
Feature activation+0.003
Top resid features:
amon
Tokenamon
Feature activation+0.011
Top resid features:
may
Token may
Feature activation+0.033
Top resid features:
take
Token take
Feature activation+0.011
Top resid features:
a
Token a
Feature activation+0.004
Top resid features:
all
Token all
Feature activation+0.005
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
deductions
Token deductions
Feature activation+0.003
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
And
Token And
Feature activation+0.017
Top resid features:
while
Token while
Feature activation+5.172
Top resid features:
he
Token he
Feature activation+0.026
Top resid features:
did
Token did
Feature activation+0.056
Top resid features:
praise
Token praise
Feature activation+0.070
Top resid features:
Trump
Token Trump
Feature activation+0.026
Top resid features:
for
Token for
Feature activation+0.019
Top resid features:
Ċ
TokenĊ
Feature activation-0.043
Top resid features:
Ċ
TokenĊ
Feature activation-0.046
Top resid features:
He
TokenHe
Feature activation-0.003
Top resid features:
said
Token said
Feature activation+0.122
Top resid features:
that
Token that
Feature activation+0.043
Top resid features:
while
Token while
Feature activation+5.180
Top resid features:
he
Token he
Feature activation+0.033
Top resid features:
and
Token and
Feature activation+0.093
Top resid features:
his
Token his
Feature activation+0.015
Top resid features:
wife
Token wife
Feature activation+0.008
Top resid features:
chose
Token chose
Feature activation+0.059
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.053
Top resid features:
the
Token the
Feature activation+0.005
Top resid features:
book
Token book
Feature activation+0.007
Top resid features:
shows
Token shows
Feature activation+0.011
Top resid features:
that
Token that
Feature activation+0.008
Top resid features:
while
Token while
Feature activation+5.289
Top resid features:
Car
Token Car
Feature activation+0.021
Top resid features:
amon
Tokenamon
Feature activation+0.008
Top resid features:
may
Token may
Feature activation+0.043
Top resid features:
take
Token take
Feature activation+0.009
Top resid features:
a
Token a
Feature activation+0.005
Top resid features:
to
Token to
Feature activation+0.018
Top resid features:
stress
Token stress
Feature activation+0.018
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
fact
Token fact
Feature activation-0.020
Top resid features:
that
Token that
Feature activation+0.006
Top resid features:
while
Token while
Feature activation+5.087
Top resid features:
we
Token we
Feature activation+0.094
Top resid features:
have
Token have
Feature activation+0.044
Top resid features:
indeed
Token indeed
Feature activation+0.033
Top resid features:
experienced
Token experienced
Feature activation+0.053
Top resid features:
a
Token a
Feature activation+0.026
Top resid features:
Ċ
TokenĊ
Feature activation-0.026
Top resid features:
Ċ
TokenĊ
Feature activation-0.038
Top resid features:
Smith
TokenSmith
Feature activation-0.004
Top resid features:
clarified
Token clarified
Feature activation+0.066
Top resid features:
that
Token that
Feature activation+0.062
Top resid features:
while
Token while
Feature activation+5.016
Top resid features:
he
Token he
Feature activation+0.055
Top resid features:
is
Token is
Feature activation+0.050
Top resid features:
closing
Token closing
Feature activation-0.008
Top resid features:
the
Token the
Feature activation+0.039
Top resid features:
curtain
Token curtain
Feature activation+0.023
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
June
Token June
Feature activation-0.000
Top resid features:
9
Token 9
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
And
Token And
Feature activation+0.037
Top resid features:
while
Token while
Feature activation+5.024
Top resid features:
we
Token we
Feature activation+0.033
Top resid features:
in
Token in
Feature activation+0.023
Top resid features:
other
Token other
Feature activation+0.010
Top resid features:
parts
Token parts
Feature activation+0.035
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.256
Top resid features:
Ċ
TokenĊ
Feature activation-0.021
Top resid features:
But
TokenBut
Feature activation-0.011
Top resid features:
while
Token while
Feature activation+5.592
Top resid features:
an
Token an
Feature activation+0.073
Top resid features:
over
Token over
Feature activation+0.018
Top resid features:
700
Token 700
Feature activation+0.004
Top resid features:
%
Token%
Feature activation+0.007
Top resid features:
surge
Token surge
Feature activation+0.012
Top resid features:
line
Token line
Feature activation+0.004
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation-0.031
Top resid features:
Ċ
TokenĊ
Feature activation-0.038
Top resid features:
And
TokenAnd
Feature activation+0.041
Top resid features:
while
Token while
Feature activation+5.175
Top resid features:
the
Token the
Feature activation+0.056
Top resid features:
Giants
Token Giants
Feature activation-0.004
Top resid features:
have
Token have
Feature activation+0.104
Top resid features:
won
Token won
Feature activation+0.007
Top resid features:
three
Token three
Feature activation+0.065
Top resid features:
Street
Token Street
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
He
Token He
Feature activation+0.003
Top resid features:
says
Token says
Feature activation+0.060
Top resid features:
that
Token that
Feature activation+0.034
Top resid features:
while
Token while
Feature activation+4.884
Top resid features:
current
Token current
Feature activation+0.016
Top resid features:
law
Token law
Feature activation+0.022
Top resid features:
restricts
Token restricts
Feature activation+0.044
Top resid features:
smoking
Token smoking
Feature activation+0.003
Top resid features:
in
Token in
Feature activation+0.004
Top resid features:
theory
Token theory
Feature activation+0.005
Top resid features:
,
Token,
Feature activation+0.018
Top resid features:
which
Token which
Feature activation+0.002
Top resid features:
notes
Token notes
Feature activation+0.031
Top resid features:
that
Token that
Feature activation+0.014
Top resid features:
whilst
Token whilst
Feature activation+4.766
Top resid features:
animals
Token animals
Feature activation+0.098
Top resid features:
may
Token may
Feature activation+0.076
Top resid features:
isolate
Token isolate
Feature activation-0.012
Top resid features:
themselves
Token themselves
Feature activation-0.006
Top resid features:
from
Token from
Feature activation-0.005
Top resid features:
way
Token way
Feature activation+0.003
Top resid features:
since
Token since
Feature activation-0.015
Top resid features:
last
Token last
Feature activation+0.008
Top resid features:
November
Token November
Feature activation+0.013
Top resid features:
and
Token and
Feature activation+0.030
Top resid features:
while
Token while
Feature activation+4.807
Top resid features:
NBC
Token NBC
Feature activation+0.018
Top resid features:
continues
Token continues
Feature activation+0.028
Top resid features:
to
Token to
Feature activation+0.024
Top resid features:
get
Token get
Feature activation+0.021
Top resid features:
scrutiny
Token scrutiny
Feature activation+0.011
Top resid features:
remains
Token remains
Feature activation+0.010
Top resid features:
consistent
Token consistent
Feature activation+0.005
Top resid features:
by
Token by
Feature activation+0.001
Top resid features:
claiming
Token claiming
Feature activation+0.020
Top resid features:
that
Token that
Feature activation+0.054
Top resid features:
while
Token while
Feature activation+5.130
Top resid features:
African
Token African
Feature activation+0.010
Top resid features:
-
Token-
Feature activation+0.002
Top resid features:
American
TokenAmerican
Feature activation+0.000
Top resid features:
women
Token women
Feature activation+0.014
Top resid features:
must
Token must
Feature activation+0.053
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.04

Head 2: 0.03

Head 3: 0.02

Head 4: 0.02

Head 5: 0.10

Head 6: 0.04

Head 7: 0.03

Head 8: 0.04

Head 9: 0.55

Head 10: 0.04

Head 11: 0.08

Positive logits

paralle1.47

otherwise1.47

nevertheless1.43

ierre1.42

xtap1.40

tons1.39

DCS1.34

wcs1.34

Downloadha1.33

utenberg1.33

vity1.31

abase1.31

remaining1.29

eele1.28

ebus1.28

hunt1.28

quant1.27

din1.27

initially1.26

still1.26

Negative logits

LTD-1.34

Raz-1.25

Bravo-1.25

agan-1.24

Nic-1.23

Roku-1.22

Aging-1.21

-1.21

Gaz-1.20

Ples-1.19

-1.19

Smy-1.18

Avalanche-1.18

Laksh-1.17

Sob-1.16

Tayyip-1.15

-1.15

-1.13

Yog-1.12

growth-1.12

INTERVAL 4.044 - 4.493
CONTAINS 0.002%

while
Token while
Feature activation+0.000
Hare
Token Hare
Feature activation+0.000
has
Token has
Feature activation+0.524
shown
Token shown
Feature activation+0.000
improvement
Token improvement
Feature activation+0.000
,
Token,
Feature activation+4.286
he
Token he
Feature activation+4.039
's
Token's
Feature activation+2.142
not
Token not
Feature activation+1.047
Jordan
Token Jordan
Feature activation+0.000
Lynch
Token Lynch
Feature activation+0.000
experienced
Token experienced
Feature activation+0.000
a
Token a
Feature activation+0.000
decade
Token decade
Feature activation+0.000
of
Token of
Feature activation+0.000
tyranny
Token tyranny
Feature activation+0.000
,
Token,
Feature activation+4.459
we
Token we
Feature activation+4.196
have
Token have
Feature activation+1.404
also
Token also
Feature activation+0.212
seen
Token seen
Feature activation+0.000
a
Token a
Feature activation+0.000
a
Token a
Feature activation+0.000
decade
Token decade
Feature activation+0.000
of
Token of
Feature activation+0.000
tyranny
Token tyranny
Feature activation+0.000
,
Token,
Feature activation+4.459
we
Token we
Feature activation+4.196
have
Token have
Feature activation+1.404
also
Token also
Feature activation+0.212
seen
Token seen
Feature activation+0.000
a
Token a
Feature activation+0.000
mass
Token mass
Feature activation+0.000
come
Token come
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
conclusion
Token conclusion
Feature activation+0.000
,
Token,
Feature activation+4.265
it
Token it
Feature activation+4.202
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.059
only
Token only
Feature activation+0.398
because
Token because
Feature activation+0.000
the
Token the
Feature activation+0.000
number
Token number
Feature activation+0.000
of
Token of
Feature activation+0.000
tax
Token tax
Feature activation+0.000
brackets
Token brackets
Feature activation+0.000
,
Token,
Feature activation+4.223
he
Token he
Feature activation+3.971
criticized
Token criticized
Feature activation+0.000
him
Token him
Feature activation+0.000
for
Token for
Feature activation+0.000
not
Token not
Feature activation+0.000

INTERVAL 3.595 - 4.044
CONTAINS 0.004%

a
Token a
Feature activation+0.000
pathway
Token pathway
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
whales
Token whales
Feature activation+0.000
,
Token,
Feature activation+3.734
the
Token the
Feature activation+2.272
orc
Token orc
Feature activation+0.000
as
Tokenas
Feature activation+0.000
are
Token are
Feature activation+0.168
still
Token still
Feature activation+0.109
ist
Tokenist
Feature activation+0.000
and
Token and
Feature activation+0.000
extremist
Token extremist
Feature activation+0.000
narratives
Token narratives
Feature activation+0.000
today
Token today
Feature activation+0.000
,
Token,
Feature activation+3.692
the
Token the
Feature activation+2.953
preceding
Token preceding
Feature activation+0.000
war
Token war
Feature activation+0.000
was
Token was
Feature activation+0.000
potentially
Token potentially
Feature activation+0.000
fl
Token fl
Feature activation+0.000
or
Tokenor
Feature activation+0.000
ists
Tokenists
Feature activation+0.000
'
Token'
Feature activation+0.000
buckets
Token buckets
Feature activation+0.000
,
Token,
Feature activation+3.740
there
Token there
Feature activation+3.468
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.264
a
Token a
Feature activation+0.000
do
Token do
Feature activation+0.000
this
Token this
Feature activation+0.000
time
Token time
Feature activation+0.000
around
Token around
Feature activation+0.000
,
Token,
Feature activation+3.902
there
Token there
Feature activation+3.832
is
Token is
Feature activation+1.968
also
Token also
Feature activation+0.507
significantly
Token significantly
Feature activation+0.000
less
Token less
Feature activation+0.000
energy
Token energy
Feature activation+0.000
arsenic
Token arsenic
Feature activation+0.000
removal
Token removal
Feature activation+0.000
have
Token have
Feature activation+1.105
taken
Token taken
Feature activation+0.082
place
Token place
Feature activation+0.000
,
Token,
Feature activation+3.675
we
Token we
Feature activation+2.290
must
Token must
Feature activation+0.000
consider
Token consider
Feature activation+0.000
variations
Token variations
Feature activation+0.000
in
Token in
Feature activation+0.000

INTERVAL 3.145 - 3.595
CONTAINS 0.007%

synt
Token synt
Feature activation+0.000
actic
Tokenactic
Feature activation+0.000
mechanism
Token mechanism
Feature activation+0.000
for
Token for
Feature activation+0.000
this
Token this
Feature activation+0.000
,
Token,
Feature activation+3.442
it
Token it
Feature activation+3.195
is
Token is
Feature activation+0.550
part
Token part
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
and
Token and
Feature activation+0.000
2014
Token 2014
Feature activation+0.000
practically
Token practically
Feature activation+1.122
doubled
Token doubled
Feature activation+0.000
,
Token,
Feature activation+3.252
recently
Token recently
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
growth
Token growth
Feature activation+0.000
generally
Token generally
Feature activation+0.000
Quantum
Token Quantum
Feature activation+0.000
RPG
Token RPG
Feature activation+0.000
hard
Token hard
Feature activation+0.000
cover
Tokencover
Feature activation+0.000
,
Token,
Feature activation+3.269
it
Token it
Feature activation+3.200
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
black
Token black
Feature activation+0.000
&
Token &
Feature activation+0.000
back
Token back
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
phone
Token phone
Feature activation+0.000
,
Token,
Feature activation+3.471
it
Token it
Feature activation+3.516
's
Token's
Feature activation+2.621
far
Token far
Feature activation+0.000
enough
Token enough
Feature activation+0.000
down
Token down
Feature activation+0.000
that
Token that
Feature activation+0.000
things
Token things
Feature activation+0.000
relating
Token relating
Feature activation+0.000
to
Token to
Feature activation+0.000
videog
Token videog
Feature activation+0.000
ames
Tokenames
Feature activation+0.000
,
Token,
Feature activation+3.350
along
Token along
Feature activation+0.686
with
Token with
Feature activation+0.000
lots
Token lots
Feature activation+0.000
of
Token of
Feature activation+0.000
other
Token other
Feature activation+0.000

INTERVAL 2.696 - 3.145
CONTAINS 0.008%

get
Token get
Feature activation+0.000
away
Token away
Feature activation+0.000
from
Token from
Feature activation+0.000
Scy
Token Scy
Feature activation+0.000
lla
Tokenlla
Feature activation+0.000
,
Token,
Feature activation+2.786
he
Token he
Feature activation+2.273
fell
Token fell
Feature activation+0.000
victim
Token victim
Feature activation+0.000
to
Token to
Feature activation+0.000
Ch
Token Ch
Feature activation+0.000
"
Token "
Feature activation+0.000
SN
TokenSN
Feature activation+0.000
L
TokenL
Feature activation+0.000
"
Token"
Feature activation+0.000
host
Token host
Feature activation+0.000
,
Token,
Feature activation+2.922
he
Token he
Feature activation+2.310
's
Token's
Feature activation+0.634
no
Token no
Feature activation+0.000
stranger
Token stranger
Feature activation+0.000
to
Token to
Feature activation+0.000
there
Token there
Feature activation+0.000
who
Token who
Feature activation+0.000
enjoy
Token enjoy
Feature activation+0.000
MMO
Token MMO
Feature activation+0.000
s
Tokens
Feature activation+0.000
,
Token,
Feature activation+2.773
the
Token the
Feature activation+1.603
fact
Token fact
Feature activation+0.000
is
Token is
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
and
Token and
Feature activation+0.000
territory
Token territory
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
While
TokenWhile
Feature activation+0.000
I
Token I
Feature activation+0.000
don
Token don
Feature activation+2.717
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
disguise
Token disguise
Feature activation+0.000
my
Token my
Feature activation+0.000
changed
Token changed
Feature activation+0.000
much
Token much
Feature activation+0.000
from
Token from
Feature activation+0.000
last
Token last
Feature activation+0.000
year
Token year
Feature activation+0.000
,
Token,
Feature activation+3.136
imported
Token imported
Feature activation+0.000
food
Token food
Feature activation+0.000
is
Token is
Feature activation+0.000
going
Token going
Feature activation+0.000
through
Token through
Feature activation+0.000

INTERVAL 2.247 - 2.696
CONTAINS 0.006%

this
Token this
Feature activation+0.000
is
Token is
Feature activation+1.095
taking
Token taking
Feature activation+0.146
place
Token place
Feature activation+0.000
,
Token,
Feature activation+3.792
the
Token the
Feature activation+2.269
TSA
Token TSA
Feature activation+0.000
continues
Token continues
Feature activation+0.155
to
Token to
Feature activation+0.000
aggressively
Token aggressively
Feature activation+0.000
screen
Token screen
Feature activation+0.000
crowd
Token crowd
Feature activation+0.000
of
Token of
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
fans
Token fans
Feature activation+0.000
,
Token,
Feature activation+3.493
the
Token the
Feature activation+2.334
vast
Token vast
Feature activation+0.000
majority
Token majority
Feature activation+0.000
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
supporting
Token supporting
Feature activation+0.000
,
Token,
Feature activation+0.000
meaning
Token meaning
Feature activation+0.000
that
Token that
Feature activation+0.000
while
Token while
Feature activation+0.000
there
Token there
Feature activation+2.114
may
Token may
Feature activation+2.566
be
Token be
Feature activation+0.000
less
Token less
Feature activation+0.390
work
Token work
Feature activation+0.000
to
Token to
Feature activation+0.000
do
Token do
Feature activation+0.000
While
Token While
Feature activation+0.000
every
Token every
Feature activation+0.000
situation
Token situation
Feature activation+0.000
is
Token is
Feature activation+0.000
different
Token different
Feature activation+0.000
,
Token,
Feature activation+2.276
you
Token you
Feature activation+2.163
might
Token might
Feature activation+0.504
benefit
Token benefit
Feature activation+0.000
from
Token from
Feature activation+0.000
this
Token this
Feature activation+0.000
SN
TokenSN
Feature activation+0.000
L
TokenL
Feature activation+0.000
"
Token"
Feature activation+0.000
host
Token host
Feature activation+0.000
,
Token,
Feature activation+2.922
he
Token he
Feature activation+2.310
's
Token's
Feature activation+0.634
no
Token no
Feature activation+0.000
stranger
Token stranger
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 1.797 - 2.247
CONTAINS 0.006%

while
Token while
Feature activation+0.000
sold
Token sold
Feature activation+0.000
at
Token at
Feature activation+0.000
market
Token market
Feature activation+0.000
rates
Token rates
Feature activation+0.000
,
Token,
Feature activation+2.107
was
Token was
Feature activation+0.000
conveyed
Token conveyed
Feature activation+0.000
under
Token under
Feature activation+0.000
a
Token a
Feature activation+0.000
1
Token 1
Feature activation+0.000
their
Token their
Feature activation+0.000
prog
Token prog
Feature activation+0.000
en
Tokenen
Feature activation+0.000
itor
Tokenitor
Feature activation+0.000
stars
Token stars
Feature activation+0.000
do
Token do
Feature activation+2.049
shape
Token shape
Feature activation+0.000
these
Token these
Feature activation+0.000
ne
Token ne
Feature activation+0.000
bul
Tokenbul
Feature activation+0.000
ae
Tokenae
Feature activation+0.000
sellers
Token sellers
Feature activation+0.000
pre
Token pre
Feature activation+0.000
New
Token New
Feature activation+0.000
52
Token 52
Feature activation+0.000
,
Token,
Feature activation+3.368
the
Token the
Feature activation+1.978
rest
Token rest
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
publishing
Token publishing
Feature activation+0.000
line
Token line
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
While
TokenWhile
Feature activation+0.000
the
Token the
Feature activation+0.000
rookie
Token rookie
Feature activation+0.000
wasn
Token wasn
Feature activation+0.176
't
Token't
Feature activation+1.978
willing
Token willing
Feature activation+0.000
to
Token to
Feature activation+0.000
label
Token label
Feature activation+0.000
it
Token it
Feature activation+0.000
an
Token an
Feature activation+0.000
of
Token of
Feature activation+0.000
desperate
Token desperate
Feature activation+0.000
college
Token college
Feature activation+0.000
students
Token students
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+2.033
represented
Token represented
Feature activation+0.000
some
Token some
Feature activation+0.000
relief
Token relief
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 1.348 - 1.797
CONTAINS 0.008%

cost
Token cost
Feature activation+0.000
of
Token of
Feature activation+0.000
homegrown
Token homegrown
Feature activation+0.000
produce
Token produce
Feature activation+0.000
has
Token has
Feature activation+0.631
not
Token not
Feature activation+1.412
changed
Token changed
Feature activation+0.000
much
Token much
Feature activation+0.000
from
Token from
Feature activation+0.000
last
Token last
Feature activation+0.000
year
Token year
Feature activation+0.000
2012
Token 2012
Feature activation+0.000
.
Token.
Feature activation+0.000
While
Token While
Feature activation+0.000
distributing
Token distributing
Feature activation+0.000
it
Token it
Feature activation+0.000
,
Token,
Feature activation+1.470
Braz
Token Braz
Feature activation+0.000
au
Tokenau
Feature activation+0.000
sometimes
Token sometimes
Feature activation+0.000
yelled
Token yelled
Feature activation+0.000
obsc
Token obsc
Feature activation+0.000
agging
Tokenagging
Feature activation+0.000
incident
Token incident
Feature activation+0.000
was
Token was
Feature activation+0.000
cruel
Token cruel
Feature activation+0.000
,
Token,
Feature activation+2.101
it
Token it
Feature activation+1.684
may
Token may
Feature activation+0.839
not
Token not
Feature activation+0.000
have
Token have
Feature activation+0.000
broken
Token broken
Feature activation+0.000
any
Token any
Feature activation+0.000
on
Token on
Feature activation+0.000
theatrical
Token theatrical
Feature activation+0.000
films
Token films
Feature activation+0.000
,
Token,
Feature activation+4.187
he
Token he
Feature activation+3.961
is
Token is
Feature activation+1.479
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
w
Tokenw
Feature activation+0.000
illing
Tokenilling
Feature activation+0.000
to
Token to
Feature activation+0.000
Qu
Token Qu
Feature activation+0.000
akes
Tokenakes
Feature activation+0.000
'
Token'
Feature activation+0.000
team
Token team
Feature activation+0.000
president
Token president
Feature activation+0.000
,
Token,
Feature activation+1.714
K
Token K
Feature activation+0.000
aval
Tokenaval
Feature activation+0.000
has
Token has
Feature activation+0.000
moved
Token moved
Feature activation+0.000
up
Token up
Feature activation+0.000

INTERVAL 0.899 - 1.348
CONTAINS 0.011%

and
Token and
Feature activation+0.000
promul
Token promul
Feature activation+0.000
gate
Tokengate
Feature activation+0.000
the
Token the
Feature activation+0.000
proposal
Token proposal
Feature activation+0.000
,
Token,
Feature activation+1.146
there
Token there
Feature activation+0.757
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
decree
Token decree
Feature activation+0.000
stating
Token stating
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
legal
Token legal
Feature activation+0.000
currently
Token currently
Feature activation+0.000
,
Token,
Feature activation+0.980
the
Token the
Feature activation+0.094
board
Token board
Feature activation+0.000
was
Token was
Feature activation+0.000
already
Token already
Feature activation+0.000
aware
Token aware
Feature activation+0.000
over
Token over
Feature activation+0.000
reach
Tokenreach
Feature activation+0.000
of
Token of
Feature activation+0.000
political
Token political
Feature activation+0.000
correctness
Token correctness
Feature activation+0.000
,
Token,
Feature activation+0.918
he
Token he
Feature activation+0.474
told
Token told
Feature activation+0.000
Feast
Token Feast
Feature activation+0.000
of
Token of
Feature activation+0.000
Fun
Token Fun
Feature activation+0.000
company
Token company
Feature activation+0.000
as
Token as
Feature activation+0.000
not
Token not
Feature activation+0.000
weak
Token weak
Feature activation+0.000
,
Token,
Feature activation+1.605
I
Token I
Feature activation+0.982
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
attest
Token attest
Feature activation+0.000
now
Token now
Feature activation+0.000
uth
Tokenuth
Feature activation+0.000
bert
Tokenbert
Feature activation+0.000
seven
Token seven
Feature activation+0.000
centuries
Token centuries
Feature activation+0.000
earlier
Token earlier
Feature activation+0.000
,
Token,
Feature activation+1.222
observed
Token observed
Feature activation+0.000
that
Token that
Feature activation+0.000
wolves
Token wolves
Feature activation+0.000
were
Token were
Feature activation+0.000
so
Token so
Feature activation+0.000

INTERVAL 0.449 - 0.899
CONTAINS 0.025%

the
Token the
Feature activation+0.000
real
Token real
Feature activation+0.000
-
Token-
Feature activation+0.000
world
Tokenworld
Feature activation+0.000
scenario
Token scenario
Feature activation+0.000
,
Token,
Feature activation+0.873
the
Token the
Feature activation+0.000
amount
Token amount
Feature activation+0.000
of
Token of
Feature activation+0.000
land
Token land
Feature activation+0.000
required
Token required
Feature activation+0.000
-
Token-
Feature activation+0.000
ex
Tokenex
Feature activation+0.000
posure
Tokenposure
Feature activation+0.000
bracket
Token bracket
Feature activation+0.000
ing
Tokening
Feature activation+0.000
may
Token may
Feature activation+0.688
work
Token work
Feature activation+0.000
in
Token in
Feature activation+0.000
some
Token some
Feature activation+0.000
cases
Token cases
Feature activation+0.000
,
Token,
Feature activation+0.354
literature
Token literature
Feature activation+0.000
linking
Token linking
Feature activation+0.000
chocolate
Token chocolate
Feature activation+0.000
and
Token and
Feature activation+0.000
depression
Token depression
Feature activation+0.000
,"
Token,"
Feature activation+0.603
says
Token says
Feature activation+0.000
the
Token the
Feature activation+0.000
lead
Token lead
Feature activation+0.000
author
Token author
Feature activation+0.000
of
Token of
Feature activation+0.000
forensic
Token forensic
Feature activation+0.000
tests
Token tests
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
remains
Token remains
Feature activation+0.000
,
Token,
Feature activation+0.465
police
Token police
Feature activation+0.000
say
Token say
Feature activation+0.000
they
Token they
Feature activation+0.000
are
Token are
Feature activation+0.000
all
Token all
Feature activation+0.000
percent
Token percent
Feature activation+0.000
of
Token of
Feature activation+0.000
Ferguson
Token Ferguson
Feature activation+0.000
's
Token's
Feature activation+0.000
population
Token population
Feature activation+0.000
,
Token,
Feature activation+0.809
about
Token about
Feature activation+0.000
85
Token 85
Feature activation+0.000
percent
Token percent
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 0.000 - 0.449
CONTAINS 99.923%

surreal
Token surreal
Feature activation+0.000
:
Token:
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Asked
TokenAsked
Feature activation+0.000
whether
Token whether
Feature activation+0.000
such
Token such
Feature activation+0.000
deliberate
Token deliberate
Feature activation+0.000
misinformation
Token misinformation
Feature activation+0.000
would
Token would
Feature activation+0.000
undermine
Token undermine
Feature activation+0.000
into
Token into
Feature activation+0.000
a
Token a
Feature activation+0.000
gl
Token gl
Feature activation+0.000
ade
Tokenade
Feature activation+0.000
of
Token of
Feature activation+0.000
snow
Token snow
Feature activation+0.000
-
Token-
Feature activation+0.000
p
Tokenp
Feature activation+0.000
asted
Tokenasted
Feature activation+0.000
bir
Token bir
Feature activation+0.000
ch
Tokench
Feature activation+0.000
and
Token and
Feature activation+0.000
City
Token City
Feature activation+0.000
Council
Token Council
Feature activation+0.000
man
Tokenman
Feature activation+0.000
Jon
Token Jon
Feature activation+0.000
Bal
Token Bal
Feature activation+0.000
iles
Tokeniles
Feature activation+0.000
delivered
Token delivered
Feature activation+0.000
a
Token a
Feature activation+0.000
hot
Token hot
Feature activation+0.000
lunch
Token lunch
Feature activation+0.000
).
Token).
Feature activation+0.000
Blog
Token Blog
Feature activation+0.000
gers
Tokengers
Feature activation+0.000
sl
Token sl
Feature activation+0.000
ung
Tokenung
Feature activation+0.000
red
Token red
Feature activation+0.000
N
Token N
Feature activation+0.000
.
Token.
Feature activation+0.000
R
TokenR
Feature activation+0.000
.
Token.
Feature activation+0.000
A
TokenA
Feature activation+0.000
County
Token County
Feature activation+0.000
Sheriff
Token Sheriff
Feature activation+0.000
's
Token's
Feature activation+0.000
Office
Token Office
Feature activation+0.000
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
boy
Token boy
Feature activation+0.000
and
Token and
Feature activation+0.000
two
Token two
Feature activation+0.000
friends
Token friends
Feature activation+0.000
began
Token began
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 1 in H3.9: (feature 16660)

TOP ACTIVATIONS
MAX = 4.917

of
Token of
Feature activation+0.000
Transportation
Token Transportation
Feature activation+0.000
's
Token's
Feature activation+0.291
inspector
Token inspector
Feature activation+0.000
general
Token general
Feature activation+0.000
had
Token had
Feature activation+4.917
given
Token given
Feature activation+2.071
the
Token the
Feature activation+0.000
agency
Token agency
Feature activation+0.000
a
Token a
Feature activation+0.000
clean
Token clean
Feature activation+0.000
By
Token By
Feature activation+0.000
signing
Token signing
Feature activation+2.696
up
Token up
Feature activation+0.000
,
Token,
Feature activation+3.238
you
Token you
Feature activation+4.034
agree
Token agree
Feature activation+4.398
to
Token to
Feature activation+0.000
our
Token our
Feature activation+0.000
Privacy
Token Privacy
Feature activation+0.000
Policy
Token Policy
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.137
http
Tokenhttp
Feature activation+0.000
H
Token H
Feature activation+0.000
ttp
Tokenttp
Feature activation+0.000
Module
TokenModule
Feature activation+0.000
)
Token )
Feature activation+0.000
had
Token had
Feature activation+4.295
built
Token built
Feature activation+0.000
in
Tokenin
Feature activation+0.000
support
Token support
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
By
TokenBy
Feature activation+0.000
this
Token this
Feature activation+0.560
time
Token time
Feature activation+4.132
next
Token next
Feature activation+1.048
week
Token week
Feature activation+3.054
Prime
Token Prime
Feature activation+0.000
Minister
Token Minister
Feature activation+0.000
-
Token-
Feature activation+0.000
,
Token,
Feature activation+3.220
Sh
Token Sh
Feature activation+0.000
k
Tokenk
Feature activation+0.000
rel
Tokenrel
Feature activation+0.000
i
Tokeni
Feature activation+0.000
had
Token had
Feature activation+4.120
been
Token been
Feature activation+0.087
assigned
Token assigned
Feature activation+0.000
an
Token an
Feature activation+0.000
inmate
Token inmate
Feature activation+0.000
number
Token number
Feature activation+0.000
By
Token By
Feature activation+0.035
2017
Token 2017
Feature activation+1.337
,
Token,
Feature activation+2.366
the
Token the
Feature activation+2.849
league
Token league
Feature activation+0.000
had
Token had
Feature activation+4.115
grown
Token grown
Feature activation+0.000
to
Token to
Feature activation+0.000
ten
Token ten
Feature activation+0.000
teams
Token teams
Feature activation+0.000
,
Token,
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
then
Token then
Feature activation+0.615
,
Token,
Feature activation+1.950
he
Token he
Feature activation+2.070
had
Token had
Feature activation+4.111
already
Token already
Feature activation+0.000
command
Token command
Feature activation+0.129
e
Tokene
Feature activation+0.000
ered
Tokenered
Feature activation+0.000
a
Token a
Feature activation+0.000
By
TokenBy
Feature activation+0.000
2006
Token 2006
Feature activation+1.004
the
Token the
Feature activation+2.416
phosph
Token phosph
Feature activation+0.000
ates
Tokenates
Feature activation+0.000
had
Token had
Feature activation+4.082
run
Token run
Feature activation+0.000
out
Token out
Feature activation+0.000
,
Token,
Feature activation+1.556
as
Token as
Feature activation+0.000
had
Token had
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
signing
Token signing
Feature activation+2.696
up
Token up
Feature activation+0.000
,
Token,
Feature activation+3.238
you
Token you
Feature activation+4.034
agree
Token agree
Feature activation+4.398
to
Token to
Feature activation+0.000
our
Token our
Feature activation+0.000
Privacy
Token Privacy
Feature activation+0.000
Policy
Token Policy
Feature activation+0.000
Up
Token Up
Feature activation+0.000
By
Token By
Feature activation+0.000
signing
Token signing
Feature activation+2.219
up
Token up
Feature activation+0.000
you
Token you
Feature activation+3.006
agree
Token agree
Feature activation+4.025
to
Token to
Feature activation+0.000
receive
Token receive
Feature activation+0.148
email
Token email
Feature activation+0.000
newsletters
Token newsletters
Feature activation+0.000
or
Token or
Feature activation+0.000
School
Token School
Feature activation+0.000
Food
Token Food
Feature activation+0.000
,
Token,
Feature activation+0.127
PS
Token PS
Feature activation+0.000
244
Token 244
Feature activation+0.000
had
Token had
Feature activation+4.020
a
Token a
Feature activation+0.000
full
Token full
Feature activation+0.000
meat
Token meat
Feature activation+0.000
-
Token-
Feature activation+0.000
free
Tokenfree
Feature activation+0.000
Disclaimer
TokenDisclaimer
Feature activation+0.000
:
Token:
Feature activation+0.000
By
Token By
Feature activation+0.000
doing
Token doing
Feature activation+0.885
this
Token this
Feature activation+0.249
you
Token you
Feature activation+3.988
void
Token void
Feature activation+0.000
your
Token your
Feature activation+0.000
war
Token war
Feature activation+0.000
rent
Tokenrent
Feature activation+0.000
y
Tokeny
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
By
TokenBy
Feature activation+0.000
the
Token the
Feature activation+0.864
middle
Token middle
Feature activation+2.281
of
Token of
Feature activation+1.041
2003
Token 2003
Feature activation+3.934
,
Token,
Feature activation+3.824
the
Token the
Feature activation+3.256
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
Up
Token Up
Feature activation+0.000
By
Token By
Feature activation+0.000
signing
Token signing
Feature activation+2.127
up
Token up
Feature activation+0.000
you
Token you
Feature activation+2.846
agree
Token agree
Feature activation+3.919
to
Token to
Feature activation+0.000
receive
Token receive
Feature activation+0.000
email
Token email
Feature activation+0.000
newsletters
Token newsletters
Feature activation+0.000
or
Token or
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.119
hard
Tokenhard
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.997
I
Token I
Feature activation+2.966
mean
Token mean
Feature activation+3.918
,
Token,
Feature activation+0.000
you
Token you
Feature activation+1.446
are
Token are
Feature activation+0.595
approaching
Token approaching
Feature activation+0.000
a
Token a
Feature activation+0.000
by
Token by
Feature activation+0.000
cleaning
Token cleaning
Feature activation+0.000
it
Token it
Feature activation+0.000
up
Token up
Feature activation+0.000
,
Token,
Feature activation+2.797
you
Token you
Feature activation+3.896
may
Token may
Feature activation+1.955
uncover
Token uncover
Feature activation+0.000
a
Token a
Feature activation+0.000
trail
Token trail
Feature activation+0.000
of
Token of
Feature activation+0.000
By
Token By
Feature activation+0.000
1988
Token 1988
Feature activation+2.083
,
Token,
Feature activation+2.767
Silver
Token Silver
Feature activation+0.000
King
Token King
Feature activation+0.000
had
Token had
Feature activation+3.863
bought
Token bought
Feature activation+0.000
11
Token 11
Feature activation+0.000
stations
Token stations
Feature activation+0.000
for
Token for
Feature activation+0.000
about
Token about
Feature activation+0.000
But
TokenBut
Feature activation+0.000
by
Token by
Feature activation+0.000
January
Token January
Feature activation+1.090
Fed
Token Fed
Feature activation+0.000
officials
Token officials
Feature activation+0.187
had
Token had
Feature activation+3.849
watched
Token watched
Feature activation+0.000
as
Token as
Feature activation+0.000
market
Token market
Feature activation+0.000
turmoil
Token turmoil
Feature activation+0.000
stemming
Token stemming
Feature activation+0.000
By
TokenBy
Feature activation+0.000
the
Token the
Feature activation+0.864
middle
Token middle
Feature activation+2.281
of
Token of
Feature activation+1.041
2003
Token 2003
Feature activation+3.934
,
Token,
Feature activation+3.824
the
Token the
Feature activation+3.256
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
by
Token by
Feature activation+0.000
teachers
Token teachers
Feature activation+0.000
I
Token I
Feature activation+3.300
mean
Token mean
Feature activation+3.764
my
Token my
Feature activation+0.000
mom
Token mom
Feature activation+0.000
...
Token...
Feature activation+0.000
A
Token A
Feature activation+0.000
)
Token)
Feature activation+0.000

Top DFA by src position
MAX = 4.931

and
Token and
Feature activation+0.001
Top resid features:
fear
Token fear
Feature activation+0.000
Top resid features:
."
Token."
Feature activation+0.015
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.014
Top resid features:
By
TokenBy
Feature activation+4.931
Top resid features:
the
Token the
Feature activation+0.018
Top resid features:
middle
Token middle
Feature activation+0.033
Top resid features:
of
Token of
Feature activation+0.030
Top resid features:
2003
Token 2003
Feature activation+0.042
Top resid features:
,
Token,
Feature activation+0.165
Top resid features:
delivered
Token delivered
Feature activation+0.004
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
your
Token your
Feature activation+0.002
Top resid features:
inbox
Token inbox
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.009
Top resid features:
By
Token By
Feature activation+4.738
Top resid features:
signing
Token signing
Feature activation+0.065
Top resid features:
up
Token up
Feature activation-0.006
Top resid features:
,
Token,
Feature activation+0.257
Top resid features:
you
Token you
Feature activation+0.093
Top resid features:
agree
Token agree
Feature activation+0.069
Top resid features:
SR
TokenSR
Feature activation+0.000
Top resid features:
F
TokenF
Feature activation+0.000
Top resid features:
Protection
Token Protection
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.012
Top resid features:
By
TokenBy
Feature activation+4.138
Top resid features:
default
Token default
Feature activation+0.384
Top resid features:
the
Token the
Feature activation+0.068
Top resid features:
Angular
Token Angular
Feature activation-0.026
Top resid features:
v
Token v
Feature activation+0.016
Top resid features:
2
Token2
Feature activation-0.004
Top resid features:
ground
Token ground
Feature activation-0.007
Top resid features:
running
Token running
Feature activation-0.005
Top resid features:
.
Token.
Feature activation+0.013
Top resid features:
Ċ
TokenĊ
Feature activation-0.028
Top resid features:
Ċ
TokenĊ
Feature activation-0.032
Top resid features:
By
TokenBy
Feature activation+4.535
Top resid features:
this
Token this
Feature activation+0.373
Top resid features:
time
Token time
Feature activation+0.078
Top resid features:
next
Token next
Feature activation+0.000
Top resid features:
week
Token week
Feature activation+0.000
Top resid features:
Prime
Token Prime
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.009
Top resid features:
âĢ
TokenâĢ
Feature activation-0.001
Top resid features:
Ŀ
TokenĿ
Feature activation-0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.021
Top resid features:
Ċ
TokenĊ
Feature activation+0.015
Top resid features:
By
TokenBy
Feature activation+4.262
Top resid features:
early
Token early
Feature activation+0.034
Top resid features:
Thursday
Token Thursday
Feature activation+0.088
Top resid features:
morning
Token morning
Feature activation+0.058
Top resid features:
,
Token,
Feature activation+0.265
Top resid features:
Sh
Token Sh
Feature activation-0.086
Top resid features:
's
Token's
Feature activation-0.004
Top resid features:
"
Token "
Feature activation+0.001
Top resid features:
Original
TokenOriginal
Feature activation+0.000
Top resid features:
Seven
Token Seven
Feature activation-0.001
Top resid features:
".
Token".
Feature activation+0.040
Top resid features:
By
Token By
Feature activation+4.227
Top resid features:
2017
Token 2017
Feature activation+0.074
Top resid features:
,
Token,
Feature activation+0.218
Top resid features:
the
Token the
Feature activation+0.096
Top resid features:
league
Token league
Feature activation-0.040
Top resid features:
had
Token had
Feature activation+0.261
Top resid features:
pilot
Token pilot
Feature activation-0.006
Top resid features:
on
Token on
Feature activation+0.009
Top resid features:
Monday
Token Monday
Feature activation+0.010
Top resid features:
afternoon
Token afternoon
Feature activation+0.012
Top resid features:
.
Token.
Feature activation+0.044
Top resid features:
By
Token By
Feature activation+4.078
Top resid features:
then
Token then
Feature activation+0.213
Top resid features:
,
Token,
Feature activation+0.163
Top resid features:
he
Token he
Feature activation+0.174
Top resid features:
had
Token had
Feature activation+0.211
Top resid features:
already
Token already
Feature activation+0.000
Top resid features:
oil
Token oil
Feature activation-0.003
Top resid features:
well
Token well
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.039
Top resid features:
Ċ
TokenĊ
Feature activation-0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
By
TokenBy
Feature activation+4.217
Top resid features:
2006
Token 2006
Feature activation+0.108
Top resid features:
the
Token the
Feature activation+0.232
Top resid features:
phosph
Token phosph
Feature activation+0.025
Top resid features:
ates
Tokenates
Feature activation-0.065
Top resid features:
had
Token had
Feature activation+0.303
Top resid features:
delivered
Token delivered
Feature activation+0.003
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
your
Token your
Feature activation+0.007
Top resid features:
inbox
Token inbox
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.011
Top resid features:
By
Token By
Feature activation+4.337
Top resid features:
signing
Token signing
Feature activation+0.163
Top resid features:
up
Token up
Feature activation+0.030
Top resid features:
,
Token,
Feature activation+0.336
Top resid features:
you
Token you
Feature activation+0.044
Top resid features:
agree
Token agree
Feature activation+0.000
Top resid features:
inbox
Token inbox
Feature activation+0.004
Top resid features:
.
Token.
Feature activation+0.013
Top resid features:
Email
Token Email
Feature activation-0.030
Top resid features:
Sign
Token Sign
Feature activation-0.002
Top resid features:
Up
Token Up
Feature activation-0.028
Top resid features:
By
Token By
Feature activation+4.321
Top resid features:
signing
Token signing
Feature activation+0.122
Top resid features:
up
Token up
Feature activation+0.044
Top resid features:
you
Token you
Feature activation+0.269
Top resid features:
agree
Token agree
Feature activation+0.080
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
brown
Token brown
Feature activation-0.000
Top resid features:
rice
Token rice
Feature activation+0.000
Top resid features:
.)
Token.)
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.006
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
By
TokenBy
Feature activation+4.066
Top resid features:
January
Token January
Feature activation+0.023
Top resid features:
2013
Token 2013
Feature activation+0.007
Top resid features:
,
Token,
Feature activation+0.040
Top resid features:
with
Token with
Feature activation+0.059
Top resid features:
the
Token the
Feature activation+0.010
Top resid features:
IF
TokenIF
Feature activation-0.002
Top resid features:
I
TokenI
Feature activation-0.004
Top resid features:
.
Token.
Feature activation+0.017
Top resid features:
Disclaimer
TokenDisclaimer
Feature activation-0.029
Top resid features:
:
Token:
Feature activation+0.026
Top resid features:
By
Token By
Feature activation+4.144
Top resid features:
doing
Token doing
Feature activation+0.259
Top resid features:
this
Token this
Feature activation+0.237
Top resid features:
you
Token you
Feature activation+0.110
Top resid features:
void
Token void
Feature activation+0.000
Top resid features:
your
Token your
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.004
Top resid features:
fear
Token fear
Feature activation+0.002
Top resid features:
."
Token."
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation-0.033
Top resid features:
Ċ
TokenĊ
Feature activation-0.017
Top resid features:
By
TokenBy
Feature activation+4.081
Top resid features:
the
Token the
Feature activation+0.192
Top resid features:
middle
Token middle
Feature activation+0.288
Top resid features:
of
Token of
Feature activation+0.264
Top resid features:
2003
Token 2003
Feature activation-0.109
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
inbox
Token inbox
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.008
Top resid features:
Email
Token Email
Feature activation-0.063
Top resid features:
Sign
Token Sign
Feature activation-0.012
Top resid features:
Up
Token Up
Feature activation-0.026
Top resid features:
By
Token By
Feature activation+4.248
Top resid features:
signing
Token signing
Feature activation+0.151
Top resid features:
up
Token up
Feature activation+0.036
Top resid features:
you
Token you
Feature activation+0.260
Top resid features:
agree
Token agree
Feature activation+0.064
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
escape
Token escape
Feature activation-0.003
Top resid features:
writing
Token writing
Feature activation-0.003
Top resid features:
it
Token it
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
By
Token By
Feature activation+4.038
Top resid features:
âĢ
Token âĢ
Feature activation+0.043
Top resid features:
ľ
Tokenľ
Feature activation+0.019
Top resid features:
hard
Tokenhard
Feature activation-0.023
Top resid features:
âĢ
TokenâĢ
Feature activation+0.170
Top resid features:
Ŀ
TokenĿ
Feature activation+0.023
Top resid features:
bits
Token bits
Feature activation+0.001
Top resid features:
up
Token up
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.029
Top resid features:
And
Token And
Feature activation+0.138
Top resid features:
,
Token,
Feature activation+0.018
Top resid features:
by
Token by
Feature activation+3.887
Top resid features:
cleaning
Token cleaning
Feature activation+0.082
Top resid features:
it
Token it
Feature activation+0.037
Top resid features:
up
Token up
Feature activation+0.019
Top resid features:
,
Token,
Feature activation+0.308
Top resid features:
you
Token you
Feature activation+0.115
Top resid features:
.[
Token.[
Feature activation-0.022
Top resid features:
5
Token5
Feature activation-0.002
Top resid features:
][
Token][
Feature activation+0.002
Top resid features:
6
Token6
Feature activation-0.003
Top resid features:
]
Token]
Feature activation+0.035
Top resid features:
By
Token By
Feature activation+4.058
Top resid features:
1988
Token 1988
Feature activation+0.108
Top resid features:
,
Token,
Feature activation+0.295
Top resid features:
Silver
Token Silver
Feature activation-0.090
Top resid features:
King
Token King
Feature activation-0.116
Top resid features:
had
Token had
Feature activation+0.317
Top resid features:
2016
Token 2016
Feature activation-0.014
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.023
Top resid features:
Ċ
TokenĊ
Feature activation-0.011
Top resid features:
But
TokenBut
Feature activation+0.073
Top resid features:
by
Token by
Feature activation+3.876
Top resid features:
January
Token January
Feature activation+0.280
Top resid features:
Fed
Token Fed
Feature activation+0.009
Top resid features:
officials
Token officials
Feature activation+0.058
Top resid features:
had
Token had
Feature activation+0.344
Top resid features:
watched
Token watched
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.003
Top resid features:
fear
Token fear
Feature activation+0.001
Top resid features:
."
Token."
Feature activation+0.017
Top resid features:
Ċ
TokenĊ
Feature activation-0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
By
TokenBy
Feature activation+3.989
Top resid features:
the
Token the
Feature activation+0.101
Top resid features:
middle
Token middle
Feature activation+0.125
Top resid features:
of
Token of
Feature activation+0.081
Top resid features:
2003
Token 2003
Feature activation+0.022
Top resid features:
,
Token,
Feature activation+0.200
Top resid features:
hear
Token hear
Feature activation+0.004
Top resid features:
from
Token from
Feature activation+0.011
Top resid features:
teachers
Token teachers
Feature activation+0.022
Top resid features:
,
Token,
Feature activation+0.065
Top resid features:
and
Token and
Feature activation+0.054
Top resid features:
by
Token by
Feature activation+3.797
Top resid features:
teachers
Token teachers
Feature activation+0.131
Top resid features:
I
Token I
Feature activation+0.261
Top resid features:
mean
Token mean
Feature activation+0.106
Top resid features:
my
Token my
Feature activation+0.000
Top resid features:
mom
Token mom
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.04

Head 2: 0.05

Head 3: 0.02

Head 4: 0.02

Head 5: 0.07

Head 6: 0.04

Head 7: 0.04

Head 8: 0.04

Head 9: 0.55

Head 10: 0.04

Head 11: 0.07

Positive logits

idem1.84

aye1.62

STD1.53

hea1.52

Wik1.51

berus1.51

ilipp1.50

sav1.47

iasis1.46

tml1.46

idates1.44

utterstock1.43

anim1.43

oult1.41

aukee1.40

animous1.40

ghan1.40

hus1.40

ortium1.39

eger1.38

Negative logits

corrid-1.54

unsolved-1.37

FC-1.25

Heisman-1.23

prem-1.21

differently-1.20

phony-1.16

rum-1.16

curry-1.14

Trinity-1.11

remembered-1.11

roast-1.10

mute-1.10

Cuban-1.10

rap-1.10

lightsaber-1.09

cube-1.08

Palmer-1.08

Catalan-1.07

histories-1.06

INTERVAL 4.425 - 4.917
CONTAINS 0.000%

of
Token of
Feature activation+0.000
Transportation
Token Transportation
Feature activation+0.000
's
Token's
Feature activation+0.291
inspector
Token inspector
Feature activation+0.000
general
Token general
Feature activation+0.000
had
Token had
Feature activation+4.917
given
Token given
Feature activation+2.071
the
Token the
Feature activation+0.000
agency
Token agency
Feature activation+0.000
a
Token a
Feature activation+0.000
clean
Token clean
Feature activation+0.000

INTERVAL 3.933 - 4.425
CONTAINS 0.001%

,
Token,
Feature activation+3.220
Sh
Token Sh
Feature activation+0.000
k
Tokenk
Feature activation+0.000
rel
Tokenrel
Feature activation+0.000
i
Tokeni
Feature activation+0.000
had
Token had
Feature activation+4.120
been
Token been
Feature activation+0.087
assigned
Token assigned
Feature activation+0.000
an
Token an
Feature activation+0.000
inmate
Token inmate
Feature activation+0.000
number
Token number
Feature activation+0.000
Up
Token Up
Feature activation+0.000
By
Token By
Feature activation+0.000
signing
Token signing
Feature activation+2.219
up
Token up
Feature activation+0.000
you
Token you
Feature activation+3.006
agree
Token agree
Feature activation+4.025
to
Token to
Feature activation+0.000
receive
Token receive
Feature activation+0.148
email
Token email
Feature activation+0.000
newsletters
Token newsletters
Feature activation+0.000
or
Token or
Feature activation+0.000
Disclaimer
TokenDisclaimer
Feature activation+0.000
:
Token:
Feature activation+0.000
By
Token By
Feature activation+0.000
doing
Token doing
Feature activation+0.885
this
Token this
Feature activation+0.249
you
Token you
Feature activation+3.988
void
Token void
Feature activation+0.000
your
Token your
Feature activation+0.000
war
Token war
Feature activation+0.000
rent
Tokenrent
Feature activation+0.000
y
Tokeny
Feature activation+0.000
By
TokenBy
Feature activation+0.000
2006
Token 2006
Feature activation+1.004
the
Token the
Feature activation+2.416
phosph
Token phosph
Feature activation+0.000
ates
Tokenates
Feature activation+0.000
had
Token had
Feature activation+4.082
run
Token run
Feature activation+0.000
out
Token out
Feature activation+0.000
,
Token,
Feature activation+1.556
as
Token as
Feature activation+0.000
had
Token had
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
By
TokenBy
Feature activation+0.000
the
Token the
Feature activation+0.864
middle
Token middle
Feature activation+2.281
of
Token of
Feature activation+1.041
2003
Token 2003
Feature activation+3.934
,
Token,
Feature activation+3.824
the
Token the
Feature activation+3.256
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000

INTERVAL 3.442 - 3.933
CONTAINS 0.002%

By
TokenBy
Feature activation+0.000
Football
Token Football
Feature activation+0.000
Esp
Token Esp
Feature activation+0.000
ana
Tokenana
Feature activation+0.000
staff
Token staff
Feature activation+2.602
Ċ
TokenĊ
Feature activation+3.481
Ċ
TokenĊ
Feature activation+3.442
Val
TokenVal
Feature activation+0.000
encia
Tokenencia
Feature activation+0.000
have
Token have
Feature activation+0.000
reportedly
Token reportedly
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
by
Token by
Feature activation+0.000
teachers
Token teachers
Feature activation+0.000
I
Token I
Feature activation+3.300
mean
Token mean
Feature activation+3.764
my
Token my
Feature activation+0.000
mom
Token mom
Feature activation+0.000
...
Token...
Feature activation+0.000
A
Token A
Feature activation+0.000
)
Token)
Feature activation+0.000
HR
Token HR
Feature activation+0.000
87
Token 87
Feature activation+0.000
99
Token99
Feature activation+0.000
c
Token c
Feature activation+0.000
,
Token,
Feature activation+2.336
they
Token they
Feature activation+3.530
have
Token have
Feature activation+0.584
managed
Token managed
Feature activation+0.297
to
Token to
Feature activation+0.000
determine
Token determine
Feature activation+0.000
the
Token the
Feature activation+0.000
By
Token By
Feature activation+0.000
1988
Token 1988
Feature activation+2.083
,
Token,
Feature activation+2.767
Silver
Token Silver
Feature activation+0.000
King
Token King
Feature activation+0.000
had
Token had
Feature activation+3.863
bought
Token bought
Feature activation+0.000
11
Token 11
Feature activation+0.000
stations
Token stations
Feature activation+0.000
for
Token for
Feature activation+0.000
about
Token about
Feature activation+0.000
Up
Token Up
Feature activation+0.000
By
Token By
Feature activation+0.000
signing
Token signing
Feature activation+2.127
up
Token up
Feature activation+0.000
you
Token you
Feature activation+2.846
agree
Token agree
Feature activation+3.919
to
Token to
Feature activation+0.000
receive
Token receive
Feature activation+0.000
email
Token email
Feature activation+0.000
newsletters
Token newsletters
Feature activation+0.000
or
Token or
Feature activation+0.000

INTERVAL 2.950 - 3.442
CONTAINS 0.006%

game
Token game
Feature activation+0.000
manager
Token manager
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
all
Token all
Feature activation+0.000
accounts
Token accounts
Feature activation+3.178
,
Token,
Feature activation+1.278
he
Token he
Feature activation+2.128
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.234
s
Tokens
Feature activation+0.299
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
By
TokenBy
Feature activation+0.000
all
Token all
Feature activation+0.000
accounts
Token accounts
Feature activation+3.012
,
Token,
Feature activation+1.014
the
Token the
Feature activation+0.975
initial
Token initial
Feature activation+0.190
land
Token land
Feature activation+0.000
mine
Token mine
Feature activation+0.000
the
Token the
Feature activation+0.000
demand
Token demand
Feature activation+0.000
;
Token;
Feature activation+0.000
by
Token by
Feature activation+0.000
which
Token which
Feature activation+0.241
I
Token I
Feature activation+3.148
mean
Token mean
Feature activation+3.749
,
Token,
Feature activation+0.724
white
Token white
Feature activation+0.000
feminists
Token feminists
Feature activation+0.000
have
Token have
Feature activation+0.145
Ċ
TokenĊ
Feature activation+0.000
By
TokenBy
Feature activation+0.000
early
Token early
Feature activation+2.014
Thursday
Token Thursday
Feature activation+2.778
morning
Token morning
Feature activation+2.870
,
Token,
Feature activation+3.220
Sh
Token Sh
Feature activation+0.000
k
Tokenk
Feature activation+0.000
rel
Tokenrel
Feature activation+0.000
i
Tokeni
Feature activation+0.000
had
Token had
Feature activation+4.120
Ċ
TokenĊ
Feature activation+0.000
By
TokenBy
Feature activation+0.000
the
Token the
Feature activation+0.292
1700
Token 1700
Feature activation+2.746
s
Tokens
Feature activation+3.075
,
Token,
Feature activation+3.292
formal
Token formal
Feature activation+0.170
flower
Token flower
Feature activation+0.000
production
Token production
Feature activation+0.000
became
Token became
Feature activation+0.251
established
Token established
Feature activation+0.043

INTERVAL 2.458 - 2.950
CONTAINS 0.007%

Dun
Token Dun
Feature activation+0.000
phy
Tokenphy
Feature activation+0.000
|
Token |
Feature activation+0.689
November
Token November
Feature activation+0.748
07
Token 07
Feature activation+1.121
,
Token,
Feature activation+2.570
2011
Token 2011
Feature activation+0.798
Ċ
TokenĊ
Feature activation+3.211
Ċ
TokenĊ
Feature activation+3.023
As
TokenAs
Feature activation+0.000
a
Token a
Feature activation+0.000
Russia
Token Russia
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
February
Token February
Feature activation+1.728
2016
Token 2016
Feature activation+2.205
,
Token,
Feature activation+2.713
the
Token the
Feature activation+2.287
bank
Token bank
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
whose
Token whose
Feature activation+0.000
stated
Token stated
Feature activation+0.000
and
Token and
Feature activation+0.000
one
Token one
Feature activation+0.000
chlor
Token chlor
Feature activation+0.000
op
Tokenop
Feature activation+0.000
last
Tokenlast
Feature activation+0.000
,
Token,
Feature activation+2.593
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
number
Token number
Feature activation+0.000
of
Token of
Feature activation+0.000
as
Token as
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.158
Ċ
TokenĊ
Feature activation+2.379
4
Token4
Feature activation+0.000
November
Token November
Feature activation+1.701
2017
Token 2017
Feature activation+2.423
Ċ
TokenĊ
Feature activation+2.807
Ċ
TokenĊ
Feature activation+2.356
The
TokenThe
Feature activation+0.000
US
Token US
Feature activation+0.000
Court
Token Court
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
row
Token row
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
1987
Token 1987
Feature activation+1.969
,
Token,
Feature activation+2.672
Esc
Token Esc
Feature activation+0.000
obar
Tokenobar
Feature activation+0.000
controlled
Token controlled
Feature activation+0.195
an
Token an
Feature activation+0.000
estimated
Token estimated
Feature activation+0.000

INTERVAL 1.967 - 2.458
CONTAINS 0.010%

rology
Tokenrology
Feature activation+0.000
last
Token last
Feature activation+0.000
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
measuring
Token measuring
Feature activation+2.073
the
Token the
Feature activation+0.000
spectrum
Token spectrum
Feature activation+0.000
of
Token of
Feature activation+0.000
light
Token light
Feature activation+0.000
coming
Token coming
Feature activation+0.000
before
Token before
Feature activation+0.000
getting
Token getting
Feature activation+0.000
crippled
Token crippled
Feature activation+0.000
,
Token,
Feature activation+1.135
you
Token you
Feature activation+3.423
can
Token can
Feature activation+2.380
confuse
Token confuse
Feature activation+0.000
the
Token the
Feature activation+0.000
game
Token game
Feature activation+0.000
into
Token into
Feature activation+0.000
giving
Token giving
Feature activation+0.000
and
Token and
Feature activation+0.000
foster
Token foster
Feature activation+0.000
investment
Token investment
Feature activation+0.000
"
Token"
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
was
Token was
Feature activation+2.389
struggling
Token struggling
Feature activation+0.000
to
Token to
Feature activation+0.000
find
Token find
Feature activation+0.000
enough
Token enough
Feature activation+0.000
cash
Token cash
Feature activation+0.000
Transportation
Token Transportation
Feature activation+0.000
's
Token's
Feature activation+0.291
inspector
Token inspector
Feature activation+0.000
general
Token general
Feature activation+0.000
had
Token had
Feature activation+4.917
given
Token given
Feature activation+2.071
the
Token the
Feature activation+0.000
agency
Token agency
Feature activation+0.000
a
Token a
Feature activation+0.000
clean
Token clean
Feature activation+0.000
bill
Token bill
Feature activation+0.000
.)
Token.)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
By
TokenBy
Feature activation+0.000
January
Token January
Feature activation+0.799
2013
Token 2013
Feature activation+2.383
,
Token,
Feature activation+2.924
with
Token with
Feature activation+1.231
the
Token the
Feature activation+0.303
help
Token help
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 1.475 - 1.967
CONTAINS 0.009%

Stephanie
Token Stephanie
Feature activation+0.000
So
Token So
Feature activation+0.000
eb
Tokeneb
Feature activation+0.000
bing
Tokenbing
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.446
Ċ
TokenĊ
Feature activation+1.514
Each
TokenEach
Feature activation+0.000
month
Token month
Feature activation+0.000
during
Token during
Feature activation+0.000
our
Token our
Feature activation+0.000
Qu
Token Qu
Feature activation+0.000
of
Token of
Feature activation+0.000
its
Token its
Feature activation+0.184
users
Token users
Feature activation+0.000
,
Token,
Feature activation+3.199
Google
Token Google
Feature activation+0.000
can
Token can
Feature activation+1.512
use
Token use
Feature activation+0.000
this
Token this
Feature activation+0.000
data
Token data
Feature activation+0.000
to
Token to
Feature activation+0.000
improve
Token improve
Feature activation+0.000
-
Token-
Feature activation+0.000
1
Token1
Feature activation+0.000
,
Token,
Feature activation+3.348
the
Token the
Feature activation+3.041
board
Token board
Feature activation+0.000
voted
Token voted
Feature activation+1.734
to
Token to
Feature activation+0.000
end
Token end
Feature activation+0.000
the
Token the
Feature activation+0.000
rides
Token rides
Feature activation+0.000
,
Token,
Feature activation+0.234
Most
Token Most
Feature activation+0.000
notably
Token notably
Feature activation+0.000
,
Token,
Feature activation+0.000
by
Token by
Feature activation+0.000
2013
Token 2013
Feature activation+1.269
,
Token,
Feature activation+1.817
several
Token several
Feature activation+0.000
provisions
Token provisions
Feature activation+0.530
affecting
Token affecting
Feature activation+0.000
high
Token high
Feature activation+0.000
-
Token-
Feature activation+0.000
By
TokenBy
Feature activation+0.000
Ed
Token Ed
Feature activation+0.000
Di
Token Di
Feature activation+0.000
ok
Tokenok
Feature activation+0.000
no
Tokenno
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.830
Ċ
TokenĊ
Feature activation+2.136
The
TokenThe
Feature activation+0.000
choice
Token choice
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 0.983 - 1.475
CONTAINS 0.015%

By
Token By
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.119
hard
Tokenhard
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.997
I
Token I
Feature activation+2.966
mean
Token mean
Feature activation+3.918
,
Token,
Feature activation+0.000
you
Token you
Feature activation+1.446
are
Token are
Feature activation+0.595
talking
Token talking
Feature activation+0.357
to
Token to
Feature activation+0.000
them
Token them
Feature activation+0.000
in
Token in
Feature activation+0.000
person
Token person
Feature activation+0.000
,
Token,
Feature activation+1.379
learning
Token learning
Feature activation+1.156
about
Token about
Feature activation+0.000
their
Token their
Feature activation+0.000
experiences
Token experiences
Feature activation+0.000
and
Token and
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
20
Token 20
Feature activation+0.000
60
Token60
Feature activation+1.400
,
Token,
Feature activation+1.823
29
Token 29
Feature activation+1.441
percent
Token percent
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.333
country
Token country
Feature activation+0.000
will
Token will
Feature activation+1.159
signed
Token signed
Feature activation+0.000
by
Token by
Feature activation+0.000
10
Token 10
Feature activation+0.000
female
Token female
Feature activation+0.000
activists
Token activists
Feature activation+0.000
,
Token,
Feature activation+1.176
also
Token also
Feature activation+0.000
calls
Token calls
Feature activation+0.000
for
Token for
Feature activation+0.000
allowing
Token allowing
Feature activation+0.000
women
Token women
Feature activation+0.000
W
Token W
Feature activation+0.000
TR
TokenTR
Feature activation+0.000
F
TokenF
Feature activation+0.000
-
Token -
Feature activation+0.000
All
Token All
Feature activation+0.000
rights
Token rights
Feature activation+1.211
reserved
Token reserved
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
On
TokenOn
Feature activation+0.000
June
Token June
Feature activation+0.000

INTERVAL 0.492 - 0.983
CONTAINS 0.024%

by
Token by
Feature activation+0.000
Mark
Token Mark
Feature activation+0.000
Wilson
Token Wilson
Feature activation+0.000
/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
Images
Token Images
Feature activation+0.580
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
GOP
TokenGOP
Feature activation+0.000
vice
Token vice
Feature activation+0.000
,
Token,
Feature activation+0.000
by
Token by
Feature activation+0.000
Allah
Token Allah
Feature activation+0.298
,
Token,
Feature activation+0.183
I
Token I
Feature activation+0.000
swear
Token swear
Feature activation+0.650
it
Token it
Feature activation+0.000
!
Token!
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
same
Token same
Feature activation+0.000
family
Token family
Feature activation+0.000
conditions
Token conditions
Feature activation+0.000
,
Token,
Feature activation+0.447
we
Token we
Feature activation+2.559
can
Token can
Feature activation+0.919
rule
Token rule
Feature activation+0.000
out
Token out
Feature activation+0.000
the
Token the
Feature activation+0.000
possible
Token possible
Feature activation+0.000
influence
Token influence
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
page
Token page
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
tapping
Token tapping
Feature activation+0.837
the
Token the
Feature activation+0.000
Show
Token Show
Feature activation+0.000
Me
Token Me
Feature activation+0.000
button
Token button
Feature activation+0.000
on
Token on
Feature activation+0.000
from
Token from
Feature activation+0.000
one
Token one
Feature activation+0.000
spot
Token spot
Feature activation+0.000
of
Token of
Feature activation+0.000
expression
Token expression
Feature activation+0.000
,
Token,
Feature activation+0.679
and
Token and
Feature activation+0.000
diff
Token diff
Feature activation+0.000
uses
Tokenuses
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 0.000 - 0.492
CONTAINS 99.927%

had
Token had
Feature activation+0.000
an
Token an
Feature activation+0.000
unrem
Token unrem
Feature activation+0.000
arkable
Tokenarkable
Feature activation+0.000
plus
Token plus
Feature activation+0.000
-
Token-
Feature activation+0.000
three
Tokenthree
Feature activation+0.000
turnover
Token turnover
Feature activation+0.000
ratio
Token ratio
Feature activation+0.000
,
Token,
Feature activation+0.000
so
Token so
Feature activation+0.000
most
Token most
Feature activation+0.000
animated
Token animated
Feature activation+0.000
faces
Token faces
Feature activation+0.000
,
Token,
Feature activation+0.000
due
Token due
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
fact
Token fact
Feature activation+0.000
that
Token that
Feature activation+0.000
spoken
Token spoken
Feature activation+0.000
dialogue
Token dialogue
Feature activation+0.000
stars
Token stars
Feature activation+0.000
who
Token who
Feature activation+0.000
exce
Token exce
Feature activation+0.000
lled
Tokenlled
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
grid
Token grid
Feature activation+0.000
iron
Tokeniron
Feature activation+0.000
and
Token and
Feature activation+0.000
helped
Token helped
Feature activation+0.000
write
Token write
Feature activation+0.000
of
Token of
Feature activation+0.000
over
Token over
Feature activation+0.000
$
Token $
Feature activation+0.000
400
Token400
Feature activation+0.000
.
Token.
Feature activation+0.000
00
Token00
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
it
Token it
Feature activation+0.000
didn
Token didn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
child
Token child
Feature activation+0.000
pushing
Token pushing
Feature activation+0.000
away
Token away
Feature activation+0.000
a
Token a
Feature activation+0.000
hearty
Token hearty
Feature activation+0.000
meal
Token meal
Feature activation+0.000
and
Token and
Feature activation+0.000
reaching
Token reaching
Feature activation+0.000
for
Token for
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 2 in H3.9: (feature 9240)

TOP ACTIVATIONS
MAX = 3.741

Catholic
Token Catholic
Feature activation+0.000
Church
Token Church
Feature activation+0.000
on
Token on
Feature activation+0.000
Sunday
Token Sunday
Feature activation+0.000
morning
Token morning
Feature activation+0.000
,
Token,
Feature activation+3.741
two
Token two
Feature activation+0.144
things
Token things
Feature activation+0.166
were
Token were
Feature activation+0.000
on
Token on
Feature activation+0.000
her
Token her
Feature activation+0.000
to
Token to
Feature activation+1.012
improve
Token improve
Feature activation+2.273
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
world
Token world
Feature activation+0.000
,
Token,
Feature activation+3.534
the
Token the
Feature activation+1.636
numbers
Token numbers
Feature activation+0.000
of
Token of
Feature activation+0.000
such
Token such
Feature activation+0.000
super
Token super
Feature activation+0.000
league
Token league
Feature activation+0.000
leaders
Token leaders
Feature activation+0.000
on
Token on
Feature activation+0.000
Sunday
Token Sunday
Feature activation+0.000
afternoon
Token afternoon
Feature activation+0.000
,
Token,
Feature activation+3.460
a
Token a
Feature activation+1.127
journey
Token journey
Feature activation+0.000
S
Token S
Feature activation+0.000
án
Tokenán
Feature activation+0.000
che
Tokenche
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
Times
Token Times
Feature activation+0.076
recently
Token recently
Feature activation+1.717
reported
Token reported
Feature activation+1.994
,
Token,
Feature activation+3.401
it
Token it
Feature activation+3.158
has
Token has
Feature activation+0.218
become
Token become
Feature activation+1.266
invested
Token invested
Feature activation+0.000
in
Token in
Feature activation+0.000
campaign
Token campaign
Feature activation+0.135
against
Token against
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
in
Token in
Feature activation+0.000
2011
Token 2011
Feature activation+0.000
,
Token,
Feature activation+3.388
he
Token he
Feature activation+1.539
gained
Token gained
Feature activation+0.617
media
Token media
Feature activation+0.000
attention
Token attention
Feature activation+0.000
and
Token and
Feature activation+0.000
and
Token and
Feature activation+0.000
movies
Token movies
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
Web
Token Web
Feature activation+0.000
,
Token,
Feature activation+3.355
expect
Token expect
Feature activation+0.694
these
Token these
Feature activation+0.000
numbers
Token numbers
Feature activation+0.000
to
Token to
Feature activation+0.000
grow
Token grow
Feature activation+0.000
in
Tokenin
Feature activation+0.000
and
Token and
Feature activation+0.000
Margaret
Token Margaret
Feature activation+0.000
Lewis
Token Lewis
Feature activation+0.000
show
Token show
Feature activation+2.891
,
Token,
Feature activation+3.322
the
Token the
Feature activation+2.012
disappearance
Token disappearance
Feature activation+0.000
of
Token of
Feature activation+0.000
much
Token much
Feature activation+0.000
of
Token of
Feature activation+0.000
and
Tokenand
Feature activation+0.296
even
Token even
Feature activation+0.000
medium
Token medium
Feature activation+0.000
)
Token)
Feature activation+0.852
investigations
Token investigations
Feature activation+0.586
,
Token,
Feature activation+3.321
we
Token we
Feature activation+2.972
still
Token still
Feature activation+0.000
encountered
Token encountered
Feature activation+0.000
relatively
Token relatively
Feature activation+0.000
few
Token few
Feature activation+0.000
phenomenon
Token phenomenon
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
sharing
Token sharing
Feature activation+0.000
economy
Token economy
Feature activation+0.036
grows
Token grows
Feature activation+3.315
it
Token it
Feature activation+2.172
will
Token will
Feature activation+1.774
be
Token be
Feature activation+0.000
good
Token good
Feature activation+0.000
for
Token for
Feature activation+0.000
houses
Token houses
Feature activation+0.000
and
Token and
Feature activation+0.000
interviewing
Token interviewing
Feature activation+0.000
family
Token family
Feature activation+0.000
members
Token members
Feature activation+0.000
,
Token,
Feature activation+3.303
friends
Token friends
Feature activation+0.000
and
Token and
Feature activation+0.000
girlfriends
Token girlfriends
Feature activation+0.000
,
Token,
Feature activation+0.110
the
Token the
Feature activation+1.763
state
Token state
Feature activation+0.294
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.417
leaders
Token leaders
Feature activation+0.779
determine
Token determine
Feature activation+3.283
the
Token the
Feature activation+0.411
final
Token final
Feature activation+0.487
shape
Token shape
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
world
Token world
Feature activation+0.000
in
Token in
Feature activation+0.000
Samoa
Token Samoa
Feature activation+0.000
,
Token,
Feature activation+3.268
his
Token his
Feature activation+0.701
dad
Token dad
Feature activation+0.000
,
Token,
Feature activation+0.000
Ag
Token Ag
Feature activation+0.000
ato
Tokenato
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
modeling
Token modeling
Feature activation+0.000
has
Token has
Feature activation+0.027
improved
Token improved
Feature activation+2.290
,
Token,
Feature activation+3.267
it
Token it
Feature activation+2.392
has
Token has
Feature activation+0.114
not
Token not
Feature activation+0.000
improved
Token improved
Feature activation+0.000
in
Token in
Feature activation+0.000
prices
Token prices
Feature activation+0.000
and
Token and
Feature activation+0.000
oil
Token oil
Feature activation+0.000
profits
Token profits
Feature activation+0.518
soar
Token soar
Feature activation+3.135
,
Token,
Feature activation+3.262
Bush
Token Bush
Feature activation+0.000
promotes
Token promotes
Feature activation+0.145
giveaways
Token giveaways
Feature activation+0.000
to
Token to
Feature activation+0.000
corporations
Token corporations
Feature activation+0.000
shorts
Token shorts
Feature activation+0.000
and
Token and
Feature activation+0.000
t
Token t
Feature activation+0.000
-
Token-
Feature activation+0.000
shirt
Tokenshirt
Feature activation+0.000
,
Token,
Feature activation+3.244
my
Token my
Feature activation+0.780
shoulders
Token shoulders
Feature activation+0.000
seemed
Token seemed
Feature activation+0.000
to
Token to
Feature activation+0.000
straight
Token straight
Feature activation+0.000
spinning
Token spinning
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
mountains
Token mountains
Feature activation+0.000
ide
Tokenide
Feature activation+0.000
grew
Token grew
Feature activation+3.238
older
Token older
Feature activation+0.724
and
Token and
Feature activation+0.277
her
Token her
Feature activation+0.000
interest
Token interest
Feature activation+0.000
in
Token in
Feature activation+0.000
as
Token as
Feature activation+0.000
the
Token the
Feature activation+0.083
Django
Token Django
Feature activation+0.000
documentation
Token documentation
Feature activation+0.000
mentions
Token mentions
Feature activation+2.326
,
Token,
Feature activation+3.238
there
Token there
Feature activation+1.882
are
Token are
Feature activation+0.000
some
Token some
Feature activation+0.000
concerns
Token concerns
Feature activation+0.000
about
Token about
Feature activation+0.000
crisis
Token crisis
Feature activation+0.261
enters
Token enters
Feature activation+2.268
its
Token its
Feature activation+1.938
sixth
Token sixth
Feature activation+0.029
year
Token year
Feature activation+0.000
,
Token,
Feature activation+3.203
the
Token the
Feature activation+1.553
Donald
Token Donald
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
administration
Token administration
Feature activation+0.000
is
Token is
Feature activation+0.185
body
Token body
Feature activation+0.000
is
Token is
Feature activation+0.244
rejecting
Token rejecting
Feature activation+0.000
the
Token the
Feature activation+0.000
virus
Token virus
Feature activation+0.000
,
Token,
Feature activation+3.203
it
Token it
Feature activation+2.784
learns
Token learns
Feature activation+1.318
to
Token to
Feature activation+0.000
reject
Token reject
Feature activation+0.000
the
Token the
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Capital
Token Capital
Feature activation+0.000
Program
Token Program
Feature activation+0.000
,
Token,
Feature activation+2.217
it
Token it
Feature activation+3.196
is
Token is
Feature activation+0.000
vital
Token vital
Feature activation+0.000
that
Token that
Feature activation+0.000
they
Token they
Feature activation+0.000
keep
Token keep
Feature activation+0.000

Top DFA by src position
MAX = 3.910

<|endoftext|>
Token<|endoftext|>
Feature activation+0.288
Top resid features:
first
Token first
Feature activation-0.001
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
second
Token second
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.028
Top resid features:
As
TokenAs
Feature activation+3.910
Top resid features:
Ann
Token Ann
Feature activation+0.010
Top resid features:
Cab
Token Cab
Feature activation+0.003
Top resid features:
iness
Tokeniness
Feature activation+0.008
Top resid features:
stood
Token stood
Feature activation+0.105
Top resid features:
in
Token in
Feature activation+0.054
Top resid features:
!).
Token!).
Feature activation+0.018
Top resid features:
Ċ
TokenĊ
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.016
Top resid features:
Nonetheless
TokenNonetheless
Feature activation+0.022
Top resid features:
,
Token,
Feature activation+0.040
Top resid features:
as
Token as
Feature activation+3.623
Top resid features:
medical
Token medical
Feature activation+0.044
Top resid features:
and
Token and
Feature activation+0.057
Top resid features:
wellness
Token wellness
Feature activation+0.001
Top resid features:
continues
Token continues
Feature activation+0.141
Top resid features:
to
Token to
Feature activation+0.050
Top resid features:
right
Token right
Feature activation-0.001
Top resid features:
now
Token now
Feature activation+0.007
Top resid features:
.
Token.
Feature activation+0.103
Top resid features:
Indeed
Token Indeed
Feature activation+0.016
Top resid features:
,
Token,
Feature activation+0.032
Top resid features:
as
Token as
Feature activation+3.651
Top resid features:
Arsenal
Token Arsenal
Feature activation+0.006
Top resid features:
and
Token and
Feature activation+0.050
Top resid features:
S
Token S
Feature activation+0.002
Top resid features:
án
Tokenán
Feature activation+0.005
Top resid features:
che
Tokenche
Feature activation+0.002
Top resid features:
support
Token support
Feature activation+0.001
Top resid features:
its
Token its
Feature activation-0.001
Top resid features:
views
Token views
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.023
Top resid features:
And
Token And
Feature activation+0.044
Top resid features:
as
Token as
Feature activation+3.852
Top resid features:
the
Token the
Feature activation+0.019
Top resid features:
New
Token New
Feature activation+0.008
Top resid features:
York
Token York
Feature activation+0.019
Top resid features:
Times
Token Times
Feature activation+0.050
Top resid features:
recently
Token recently
Feature activation+0.087
Top resid features:
ther
Tokenther
Feature activation+0.000
Top resid features:
movement
Token movement
Feature activation-0.000
Top resid features:
?
Token?
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
Ċ
TokenĊ
Feature activation+0.029
Top resid features:
As
TokenAs
Feature activation+3.579
Top resid features:
Trump
Token Trump
Feature activation-0.002
Top resid features:
publicly
Token publicly
Feature activation+0.030
Top resid features:
considered
Token considered
Feature activation+0.059
Top resid features:
a
Token a
Feature activation+0.021
Top resid features:
presidential
Token presidential
Feature activation+0.008
Top resid features:
delivered
Token delivered
Feature activation+0.001
Top resid features:
over
Token over
Feature activation+0.005
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
Internet
Token Internet
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.026
Top resid features:
As
Token As
Feature activation+3.466
Top resid features:
it
Token it
Feature activation+0.014
Top resid features:
becomes
Token becomes
Feature activation+0.071
Top resid features:
easier
Token easier
Feature activation+0.003
Top resid features:
(
Token (
Feature activation+0.108
Top resid features:
and
Tokenand
Feature activation+0.032
Top resid features:
and
Token and
Feature activation+0.005
Top resid features:
death
Token death
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.006
Top resid features:
As
TokenAs
Feature activation+3.884
Top resid features:
the
Token the
Feature activation+0.069
Top resid features:
pale
Token pale
Feature activation+0.034
Top resid features:
ont
Tokenont
Feature activation+0.002
Top resid features:
ologists
Tokenologists
Feature activation+0.058
Top resid features:
Lars
Token Lars
Feature activation+0.026
Top resid features:
?
Token?
Feature activation+0.016
Top resid features:
âĢ
TokenâĢ
Feature activation-0.002
Top resid features:
Ŀ
TokenĿ
Feature activation+0.003
Top resid features:
)
Token)
Feature activation+0.021
Top resid features:
But
Token But
Feature activation+0.081
Top resid features:
as
Token as
Feature activation+3.533
Top resid features:
we
Token we
Feature activation+0.050
Top resid features:
conducted
Token conducted
Feature activation+0.197
Top resid features:
shallow
Token shallow
Feature activation+0.135
Top resid features:
(
Token (
Feature activation+0.087
Top resid features:
and
Tokenand
Feature activation+0.016
Top resid features:
at
Token at
Feature activation+0.001
Top resid features:
59
Token 59
Feature activation+0.001
Top resid features:
miles
Token miles
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.002
Top resid features:
As
TokenAs
Feature activation+3.547
Top resid features:
the
Token the
Feature activation+0.079
Top resid features:
phenomenon
Token phenomenon
Feature activation+0.040
Top resid features:
of
Token of
Feature activation+0.097
Top resid features:
the
Token the
Feature activation+0.075
Top resid features:
sharing
Token sharing
Feature activation+0.032
Top resid features:
house
Token house
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.015
Top resid features:
Ċ
TokenĊ
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation+0.013
Top resid features:
But
TokenBut
Feature activation+0.055
Top resid features:
as
Token as
Feature activation+3.606
Top resid features:
police
Token police
Feature activation+0.014
Top resid features:
and
Token and
Feature activation+0.027
Top resid features:
federal
Token federal
Feature activation+0.008
Top resid features:
agents
Token agents
Feature activation+0.002
Top resid features:
f
Token f
Feature activation+0.015
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
challenges
Token challenges
Feature activation+0.003
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
tomorrow
Token tomorrow
Feature activation+0.005
Top resid features:
.
Token.
Feature activation+0.041
Top resid features:
As
Token As
Feature activation+3.483
Top resid features:
the
Token the
Feature activation+0.046
Top resid features:
city
Token city
Feature activation+0.027
Top resid features:
and
Token and
Feature activation+0.064
Top resid features:
state
Token state
Feature activation+0.008
Top resid features:
âĢ
TokenâĢ
Feature activation+0.005
Top resid features:
15
Token 15
Feature activation+0.001
Top resid features:
,
Token,
Feature activation-0.000
Top resid features:
1991
Token 1991
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.006
Top resid features:
and
Token and
Feature activation+0.044
Top resid features:
as
Token as
Feature activation+3.455
Top resid features:
World
Token World
Feature activation+0.005
Top resid features:
Cup
Token Cup
Feature activation-0.002
Top resid features:
was
Token was
Feature activation+0.023
Top resid features:
making
Token making
Feature activation+0.015
Top resid features:
his
Token his
Feature activation+0.013
Top resid features:
J
Token J
Feature activation+0.000
Top resid features:
CT
TokenCT
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.067
Top resid features:
Unfortunately
Token Unfortunately
Feature activation+0.021
Top resid features:
,
Token,
Feature activation+0.016
Top resid features:
as
Token as
Feature activation+3.349
Top resid features:
modeling
Token modeling
Feature activation+0.007
Top resid features:
has
Token has
Feature activation+0.089
Top resid features:
improved
Token improved
Feature activation+0.191
Top resid features:
,
Token,
Feature activation+0.101
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
South
Token South
Feature activation-0.000
Top resid features:
Carolina
Token Carolina
Feature activation-0.000
Top resid features:
.
Token.
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.004
Top resid features:
As
TokenAs
Feature activation+3.389
Top resid features:
gas
Token gas
Feature activation+0.031
Top resid features:
prices
Token prices
Feature activation+0.042
Top resid features:
and
Token and
Feature activation+0.098
Top resid features:
oil
Token oil
Feature activation+0.016
Top resid features:
profits
Token profits
Feature activation+0.043
Top resid features:
he
Token he
Feature activation+0.006
Top resid features:
asked
Token asked
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.018
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.014
Top resid features:
As
TokenAs
Feature activation+3.317
Top resid features:
I
Token I
Feature activation+0.024
Top resid features:
unf
Token unf
Feature activation+0.036
Top resid features:
ur
Tokenur
Feature activation+0.005
Top resid features:
led
Tokenled
Feature activation+0.011
Top resid features:
the
Token the
Feature activation+0.093
Top resid features:
and
Token and
Feature activation+0.008
Top resid features:
spinning
Token spinning
Feature activation-0.004
Top resid features:
."
Token."
Feature activation+0.021
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.015
Top resid features:
As
TokenAs
Feature activation+3.482
Top resid features:
that
Token that
Feature activation+0.102
Top resid features:
girl
Token girl
Feature activation+0.042
Top resid features:
spinning
Token spinning
Feature activation+0.051
Top resid features:
on
Token on
Feature activation+0.084
Top resid features:
the
Token the
Feature activation+0.050
Top resid features:
in
Token in
Feature activation+0.008
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
former
Token former
Feature activation-0.003
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
but
Token but
Feature activation+0.073
Top resid features:
as
Token as
Feature activation+3.598
Top resid features:
the
Token the
Feature activation+0.024
Top resid features:
Django
Token Django
Feature activation+0.033
Top resid features:
documentation
Token documentation
Feature activation+0.008
Top resid features:
mentions
Token mentions
Feature activation+0.049
Top resid features:
,
Token,
Feature activation+0.122
Top resid features:
Tur
Token Tur
Feature activation+0.007
Top resid features:
be
Tokenbe
Feature activation+0.002
Top resid features:
ville
Tokenville
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
As
TokenAs
Feature activation+3.282
Top resid features:
the
Token the
Feature activation+0.025
Top resid features:
Syrian
Token Syrian
Feature activation+0.025
Top resid features:
crisis
Token crisis
Feature activation+0.038
Top resid features:
enters
Token enters
Feature activation+0.173
Top resid features:
its
Token its
Feature activation+0.051
Top resid features:
ľ
Tokenľ
Feature activation+0.006
Top resid features:
The
TokenThe
Feature activation+0.013
Top resid features:
idea
Token idea
Feature activation-0.019
Top resid features:
is
Token is
Feature activation+0.013
Top resid features:
that
Token that
Feature activation+0.098
Top resid features:
as
Token as
Feature activation+3.364
Top resid features:
the
Token the
Feature activation+0.029
Top resid features:
body
Token body
Feature activation+0.020
Top resid features:
is
Token is
Feature activation+0.028
Top resid features:
rejecting
Token rejecting
Feature activation+0.147
Top resid features:
the
Token the
Feature activation+0.038
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
challenges
Token challenges
Feature activation+0.001
Top resid features:
of
Token of
Feature activation+0.002
Top resid features:
tomorrow
Token tomorrow
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.039
Top resid features:
As
Token As
Feature activation+3.462
Top resid features:
the
Token the
Feature activation+0.047
Top resid features:
city
Token city
Feature activation+0.011
Top resid features:
and
Token and
Feature activation+0.062
Top resid features:
state
Token state
Feature activation+0.001
Top resid features:
âĢ
TokenâĢ
Feature activation+0.005
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.05

Head 2: 0.04

Head 3: 0.02

Head 4: 0.02

Head 5: 0.08

Head 6: 0.05

Head 7: 0.03

Head 8: 0.04

Head 9: 0.55

Head 10: 0.04

Head 11: 0.06

Positive logits

utenberg1.53

liners1.51

itcher1.48

nearer1.39

Downloadha1.39

uddenly1.35

yrim1.32

cca1.31

isodes1.29

stud1.29

ilipp1.29

verty1.26

uni1.25

yout1.22

dos1.22

ilogy1.21

rike1.20

CLS1.19

witness1.19

pandemonium1.19

Negative logits

CLUD-1.28

ertodd-1.28

-1.27

RG-1.22

MAL-1.20

д-1.19

zl-1.19

AME-1.17

scalp-1.17

emate-1.17

BILITY-1.16

assumption-1.15

mug-1.14

understatement-1.13

igs-1.13

depot-1.12

hinge-1.11

Mercer-1.10

fax-1.10

INC-1.10

INTERVAL 3.367 - 3.741
CONTAINS 0.001%

Catholic
Token Catholic
Feature activation+0.000
Church
Token Church
Feature activation+0.000
on
Token on
Feature activation+0.000
Sunday
Token Sunday
Feature activation+0.000
morning
Token morning
Feature activation+0.000
,
Token,
Feature activation+3.741
two
Token two
Feature activation+0.144
things
Token things
Feature activation+0.166
were
Token were
Feature activation+0.000
on
Token on
Feature activation+0.000
her
Token her
Feature activation+0.000
league
Token league
Feature activation+0.000
leaders
Token leaders
Feature activation+0.000
on
Token on
Feature activation+0.000
Sunday
Token Sunday
Feature activation+0.000
afternoon
Token afternoon
Feature activation+0.000
,
Token,
Feature activation+3.460
a
Token a
Feature activation+1.127
journey
Token journey
Feature activation+0.000
S
Token S
Feature activation+0.000
án
Tokenán
Feature activation+0.000
che
Tokenche
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
Times
Token Times
Feature activation+0.076
recently
Token recently
Feature activation+1.717
reported
Token reported
Feature activation+1.994
,
Token,
Feature activation+3.401
it
Token it
Feature activation+3.158
has
Token has
Feature activation+0.218
become
Token become
Feature activation+1.266
invested
Token invested
Feature activation+0.000
in
Token in
Feature activation+0.000
to
Token to
Feature activation+1.012
improve
Token improve
Feature activation+2.273
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
world
Token world
Feature activation+0.000
,
Token,
Feature activation+3.534
the
Token the
Feature activation+1.636
numbers
Token numbers
Feature activation+0.000
of
Token of
Feature activation+0.000
such
Token such
Feature activation+0.000
super
Token super
Feature activation+0.000
campaign
Token campaign
Feature activation+0.135
against
Token against
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
in
Token in
Feature activation+0.000
2011
Token 2011
Feature activation+0.000
,
Token,
Feature activation+3.388
he
Token he
Feature activation+1.539
gained
Token gained
Feature activation+0.617
media
Token media
Feature activation+0.000
attention
Token attention
Feature activation+0.000
and
Token and
Feature activation+0.000

INTERVAL 2.993 - 3.367
CONTAINS 0.004%

been
Token been
Feature activation+0.000
insisting
Token insisting
Feature activation+0.000
for
Token for
Feature activation+0.000
20
Token 20
Feature activation+0.000
years
Token years
Feature activation+0.000
,
Token,
Feature activation+3.143
we
Token we
Feature activation+0.234
have
Token have
Feature activation+0.000
to
Token to
Feature activation+0.000
assess
Token assess
Feature activation+0.000
our
Token our
Feature activation+0.000
and
Token and
Feature activation+0.000
movies
Token movies
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
Web
Token Web
Feature activation+0.000
,
Token,
Feature activation+3.355
expect
Token expect
Feature activation+0.694
these
Token these
Feature activation+0.000
numbers
Token numbers
Feature activation+0.000
to
Token to
Feature activation+0.000
grow
Token grow
Feature activation+0.000
specific
Tokenspecific
Feature activation+0.000
languages
Token languages
Feature activation+0.000
will
Token will
Feature activation+1.746
tell
Token tell
Feature activation+1.062
you
Token you
Feature activation+0.000
,
Token,
Feature activation+3.137
seldom
Token seldom
Feature activation+0.000
is
Token is
Feature activation+0.000
life
Token life
Feature activation+0.000
as
Token as
Feature activation+0.000
simple
Token simple
Feature activation+0.000
both
Token both
Feature activation+0.000
White
Token White
Feature activation+0.000
and
Token and
Feature activation+0.000
Marriott
Token Marriott
Feature activation+0.000
explain
Token explain
Feature activation+1.797
,
Token,
Feature activation+3.192
the
Token the
Feature activation+1.513
evolution
Token evolution
Feature activation+0.005
of
Token of
Feature activation+0.000
this
Token this
Feature activation+0.000
development
Token development
Feature activation+0.000
he
Token he
Feature activation+0.674
tells
Token tells
Feature activation+0.389
Nick
Token Nick
Feature activation+0.000
Ah
Token Ah
Feature activation+0.000
ad
Tokenad
Feature activation+0.000
,
Token,
Feature activation+3.096
the
Token the
Feature activation+2.286
stage
Token stage
Feature activation+0.000
has
Token has
Feature activation+1.275
given
Token given
Feature activation+0.329
him
Token him
Feature activation+0.000

INTERVAL 2.619 - 2.993
CONTAINS 0.007%

the
Token the
Feature activation+0.348
new
Token new
Feature activation+0.137
examination
Token examination
Feature activation+0.000
guidelines
Token guidelines
Feature activation+0.144
indicate
Token indicate
Feature activation+2.934
,
Token,
Feature activation+2.852
China
Token China
Feature activation+0.000
has
Token has
Feature activation+0.000
received
Token received
Feature activation+0.000
5
Token 5
Feature activation+0.000
65
Token65
Feature activation+0.000
official
Token official
Feature activation+0.367
Xbox
Token Xbox
Feature activation+0.000
One
Token One
Feature activation+0.000
web
Token web
Feature activation+0.000
site
Token site
Feature activation+0.000
,
Token,
Feature activation+2.853
One
Token One
Feature activation+0.000
Guide
TokenGuide
Feature activation+0.000
allows
Token allows
Feature activation+0.347
users
Token users
Feature activation+0.000
to
Token to
Feature activation+0.000
lost
Token lost
Feature activation+0.642
its
Token its
Feature activation+0.000
h
Token h
Feature activation+0.000
inter
Tokeninter
Feature activation+0.000
lands
Tokenlands
Feature activation+0.000
,
Token,
Feature activation+2.804
history
Token history
Feature activation+0.000
moved
Token moved
Feature activation+2.768
north
Token north
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
But
Token But
Feature activation+0.000
as
Token as
Feature activation+0.000
it
Token it
Feature activation+0.000
turns
Token turns
Feature activation+0.171
out
Token out
Feature activation+2.042
,
Token,
Feature activation+2.893
with
Token with
Feature activation+1.490
a
Token a
Feature activation+0.000
bit
Token bit
Feature activation+0.000
of
Token of
Feature activation+0.000
help
Token help
Feature activation+0.000
the
Token the
Feature activation+0.000
face
Token face
Feature activation+0.000
of
Token of
Feature activation+0.000
terrorist
Token terrorist
Feature activation+0.000
threats
Token threats
Feature activation+0.000
,
Token,
Feature activation+2.655
one
Token one
Feature activation+0.764
movie
Token movie
Feature activation+0.000
theater
Token theater
Feature activation+0.000
has
Token has
Feature activation+0.000
found
Token found
Feature activation+0.000

INTERVAL 2.245 - 2.619
CONTAINS 0.012%

As
Token As
Feature activation+0.000
I
Token I
Feature activation+0.257
have
Token have
Feature activation+0.000
said
Token said
Feature activation+0.649
before
Token before
Feature activation+0.150
,
Token,
Feature activation+2.531
there
Token there
Feature activation+0.328
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
much
Token much
Feature activation+0.000
of
Token of
Feature activation+0.000
As
TokenAs
Feature activation+0.000
we
Token we
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+0.000
grown
Token grown
Feature activation+2.266
,
Token,
Feature activation+1.586
hotels
Token hotels
Feature activation+0.000
have
Token have
Feature activation+0.073
grown
Token grown
Feature activation+2.143
.
Token.
Feature activation+0.000
under
Token under
Feature activation+0.000
the
Token the
Feature activation+0.000
weight
Token weight
Feature activation+0.000
of
Token of
Feature activation+0.000
Obamacare
Token Obamacare
Feature activation+0.000
,
Token,
Feature activation+2.602
they
Token they
Feature activation+1.641
are
Token are
Feature activation+0.000
narrowing
Token narrowing
Feature activation+0.000
their
Token their
Feature activation+0.000
provider
Token provider
Feature activation+0.000
friend
Token friend
Feature activation+0.000
in
Token in
Feature activation+0.000
law
Token law
Feature activation+0.000
enforcement
Token enforcement
Feature activation+0.364
once
Token once
Feature activation+1.047
said
Token said
Feature activation+2.319
,
Token,
Feature activation+2.712
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
nothing
Tokennothing
Feature activation+0.000
makes
Token makes
Feature activation+0.000
As
TokenAs
Feature activation+0.000
he
Token he
Feature activation+0.144
was
Token was
Feature activation+0.058
going
Token going
Feature activation+0.000
down
Token down
Feature activation+0.000
,
Token,
Feature activation+2.258
he
Token he
Feature activation+0.357
said
Token said
Feature activation+0.084
,
Token,
Feature activation+0.760
'
Token '
Feature activation+0.288
He
TokenHe
Feature activation+0.000

INTERVAL 1.871 - 2.245
CONTAINS 0.015%

who
Token who
Feature activation+0.171
has
Token has
Feature activation+0.000
seen
Token seen
Feature activation+0.000
the
Token the
Feature activation+0.103
movie
Token movie
Feature activation+0.000
can
Token can
Feature activation+2.177
attest
Token attest
Feature activation+0.430
,
Token,
Feature activation+2.706
the
Token the
Feature activation+1.454
bear
Token bear
Feature activation+0.000
in
Token in
Feature activation+0.000
and
Token and
Feature activation+0.000
reaches
Token reaches
Feature activation+0.135
for
Token for
Feature activation+0.000
his
Token his
Feature activation+0.000
shoes
Token shoes
Feature activation+0.000
,
Token,
Feature activation+1.974
he
Token he
Feature activation+0.583
is
Token is
Feature activation+0.000
shot
Token shot
Feature activation+0.000
once
Token once
Feature activation+0.000
in
Token in
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
was
Token was
Feature activation+0.000
revealed
Token revealed
Feature activation+0.000
yesterday
Token yesterday
Feature activation+0.000
,
Token,
Feature activation+1.957
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
rules
Token rules
Feature activation+0.000
and
Token and
Feature activation+0.000
basic
Token basic
Feature activation+0.000
their
Token their
Feature activation+0.000
classroom
Token classroom
Feature activation+0.000
performance
Token performance
Feature activation+0.000
should
Token should
Feature activation+1.748
improve
Token improve
Feature activation+1.027
,
Token,
Feature activation+1.908
but
Token but
Feature activation+0.000
other
Token other
Feature activation+0.000
factors
Token factors
Feature activation+0.000
should
Token should
Feature activation+0.000
eventually
Token eventually
Feature activation+0.000
a
Token a
Feature activation+0.000
sort
Token sort
Feature activation+0.196
of
Token of
Feature activation+0.000
beta
Token beta
Feature activation+0.000
phase
Token phase
Feature activation+0.000
,
Token,
Feature activation+2.010
we
Token we
Feature activation+0.195
are
Token are
Feature activation+0.000
adding
Token adding
Feature activation+0.000
a
Token a
Feature activation+0.000
lot
Token lot
Feature activation+0.000

INTERVAL 1.497 - 1.871
CONTAINS 0.017%

<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
As
TokenAs
Feature activation+0.000
the
Token the
Feature activation+0.608
Daily
Token Daily
Feature activation+0.000
Wire
Token Wire
Feature activation+0.000
reported
Token reported
Feature activation+1.524
Tuesday
Token Tuesday
Feature activation+0.000
,
Token,
Feature activation+2.872
the
Token the
Feature activation+0.472
Hollywood
Token Hollywood
Feature activation+0.000
trade
Token trade
Feature activation+0.000
As
Token As
Feature activation+0.000
you
Token you
Feature activation+0.000
use
Token use
Feature activation+0.000
your
Token your
Feature activation+0.000
knife
Token knife
Feature activation+0.000
,
Token,
Feature activation+1.690
the
Token the
Feature activation+0.970
blade
Token blade
Feature activation+0.000
gets
Token gets
Feature activation+0.145
knocked
Token knocked
Feature activation+0.000
out
Token out
Feature activation+0.000
up
Token up
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
speedy
Token speedy
Feature activation+0.000
professor
Token professor
Feature activation+0.000
,
Token,
Feature activation+1.687
he
Token he
Feature activation+0.055
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
completely
Token completely
Feature activation+0.000
different
Token different
Feature activation+0.000
Balls
Token Balls
Feature activation+0.000
'
Token'
Feature activation+0.000
proposed
Token proposed
Feature activation+0.184
VAT
Token VAT
Feature activation+0.000
cut
Token cut
Feature activation+0.000
,
Token,
Feature activation+1.721
"
Token "
Feature activation+0.280
in
Tokenin
Feature activation+0.038
an
Token an
Feature activation+0.000
attempt
Token attempt
Feature activation+0.000
to
Token to
Feature activation+0.000
woman
Token woman
Feature activation+1.002
in
Token in
Feature activation+0.000
my
Token my
Feature activation+0.507
40
Token 40
Feature activation+0.000
s
Tokens
Feature activation+0.000
,
Token,
Feature activation+1.768
I
Token I
Feature activation+2.720
don
Token don
Feature activation+0.278
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000

INTERVAL 1.122 - 1.497
CONTAINS 0.017%

As
TokenAs
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.235
AC
Token AC
Feature activation+0.000
4
Token4
Feature activation+0.000
,
Token,
Feature activation+1.406
it
Token it
Feature activation+0.996
packs
Token packs
Feature activation+0.000
a
Token a
Feature activation+0.000
punch
Token punch
Feature activation+0.000
with
Token with
Feature activation+0.000
).
Token).
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
As
TokenAs
Feature activation+0.000
you
Token you
Feature activation+0.057
watch
Token watch
Feature activation+1.131
the
Token the
Feature activation+0.000
clip
Token clip
Feature activation+0.000
below
Token below
Feature activation+0.000
,
Token,
Feature activation+2.591
it
Token it
Feature activation+0.974
i
Token i
Feature activation+0.000
ME
TokenME
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Times
Token Times
Feature activation+0.000
,
Token,
Feature activation+1.204
maintains
Token maintains
Feature activation+0.320
that
Token that
Feature activation+0.000
his
Token his
Feature activation+0.000
sourcing
Token sourcing
Feature activation+0.000
is
Token is
Feature activation+0.000
.
Token.
Feature activation+0.000
As
Token As
Feature activation+0.000
a
Token a
Feature activation+0.119
control
Token control
Feature activation+0.000
experiment
Token experiment
Feature activation+0.738
,
Token,
Feature activation+1.320
the
Token the
Feature activation+0.967
team
Token team
Feature activation+0.000
repeated
Token repeated
Feature activation+0.000
this
Token this
Feature activation+0.000
analysis
Token analysis
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
as
Token as
Feature activation+0.000
mentioned
Token mentioned
Feature activation+0.000
,
Token,
Feature activation+2.209
we
Token we
Feature activation+1.150
are
Token are
Feature activation+0.000
seeing
Token seeing
Feature activation+0.000
extremely
Token extremely
Feature activation+0.000
low
Token low
Feature activation+0.000
water
Token water
Feature activation+0.000

INTERVAL 0.748 - 1.122
CONTAINS 0.030%

going
Token going
Feature activation+0.000
down
Token down
Feature activation+0.000
,
Token,
Feature activation+2.258
he
Token he
Feature activation+0.357
said
Token said
Feature activation+0.084
,
Token,
Feature activation+0.760
'
Token '
Feature activation+0.288
He
TokenHe
Feature activation+0.000
has
Token has
Feature activation+0.000
a
Token a
Feature activation+0.000
gun
Token gun
Feature activation+0.000
over
Token over
Feature activation+0.000
as
Token as
Feature activation+0.000
far
Token far
Feature activation+0.000
as
Token as
Feature activation+0.000
desire
Token desire
Feature activation+0.000
goes
Token goes
Feature activation+0.785
to
Token to
Feature activation+0.000
chase
Token chase
Feature activation+0.000
/
Token/
Feature activation+0.000
f
Tokenf
Feature activation+0.000
awn
Tokenawn
Feature activation+0.000
As
Token As
Feature activation+0.000
E
Token E
Feature activation+0.000
3
Token3
Feature activation+0.000
2012
Token 2012
Feature activation+0.000
Reve
Token Reve
Feature activation+0.000
al
Tokenal
Feature activation+0.942
Ċ
TokenĊ
Feature activation+0.048
Ċ
TokenĊ
Feature activation+0.020
Honestly
TokenHonestly
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
first
Token first
Feature activation+0.000
pro
Token pro
Feature activation+0.000
appearance
Token appearance
Feature activation+0.000
last
Token last
Feature activation+0.000
summer
Token summer
Feature activation+0.000
,
Token,
Feature activation+0.932
but
Token but
Feature activation+0.000
otherwise
Token otherwise
Feature activation+0.000
had
Token had
Feature activation+0.000
no
Token no
Feature activation+0.000
issues
Token issues
Feature activation+0.000
to
Token to
Feature activation+0.000
deal
Token deal
Feature activation+0.000
with
Token with
Feature activation+0.000
ice
Token ice
Feature activation+0.000
,
Token,
Feature activation+2.345
the
Token the
Feature activation+1.067
most
Token most
Feature activation+0.000
recent
Token recent
Feature activation+0.000
statistics
Token statistics
Feature activation+0.000
show
Token show
Feature activation+0.000
that
Token that
Feature activation+0.000

INTERVAL 0.374 - 0.748
CONTAINS 0.045%

is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
cost
Token cost
Feature activation+0.000
blow
Token blow
Feature activation+0.000
out
Tokenout
Feature activation+0.000
,
Token,
Feature activation+0.388
how
Token how
Feature activation+0.000
we
Token we
Feature activation+0.000
manage
Token manage
Feature activation+0.000
that
Token that
Feature activation+0.000
.
Token.
Feature activation+0.000
personal
Token personal
Feature activation+0.000
apology
Token apology
Feature activation+0.000
but
Token but
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
nation
Token nation
Feature activation+0.533
,
Token,
Feature activation+1.472
America
Token America
Feature activation+0.583
should
Token should
Feature activation+0.342
admit
Token admit
Feature activation+0.000
the
Token the
Feature activation+0.000
,
Token,
Feature activation+2.688
ten
Token ten
Feature activation+0.000
major
Token major
Feature activation+0.000
Japanese
Token Japanese
Feature activation+0.000
companies
Token companies
Feature activation+0.000
are
Token are
Feature activation+0.674
planning
Token planning
Feature activation+0.000
to
Token to
Feature activation+0.000
launch
Token launch
Feature activation+0.000
exchanges
Token exchanges
Feature activation+0.000
for
Token for
Feature activation+0.000
as
Token as
Feature activation+0.000
I
Token I
Feature activation+0.000
can
Token can
Feature activation+0.000
see
Token see
Feature activation+0.000
,
Token,
Feature activation+1.341
he
Token he
Feature activation+0.502
did
Token did
Feature activation+0.053
not
Token not
Feature activation+0.000
.
Token.
Feature activation+0.000
Also
Token Also
Feature activation+0.000
,
Token,
Feature activation+0.000
rules
Token rules
Feature activation+0.000
up
Token up
Feature activation+0.000
as
Token as
Feature activation+0.000
they
Token they
Feature activation+0.000
go
Token go
Feature activation+0.000
along
Token along
Feature activation+0.529
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
It
TokenIt
Feature activation+0.000

INTERVAL 0.000 - 0.374
CONTAINS 99.853%

."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
To
TokenTo
Feature activation+0.000
the
Token the
Feature activation+0.000
media
Token media
Feature activation+0.000
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
you
Token you
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
explore
Token explore
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
are
Token are
Feature activation+0.000
proced
Token proced
Feature activation+0.000
urally
Tokenurally
Feature activation+0.000
family
Token family
Feature activation+0.000
members
Token members
Feature activation+0.000
of
Token of
Feature activation+0.000
military
Token military
Feature activation+0.000
members
Token members
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
American
Token American
Feature activation+0.000
government
Token government
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
like
Token like
Feature activation+0.000
it
Token it
Feature activation+0.000
.
Token.
Feature activation+0.000
When
Token When
Feature activation+0.000
we
Token we
Feature activation+0.000
get
Token get
Feature activation+0.000
a
Token a
Feature activation+0.000
sick
Token sick
Feature activation+0.000
kid
Token kid
Feature activation+0.000
in
Token in
Feature activation+0.000
past
Token past
Feature activation+0.000
with
Token with
Feature activation+0.000
seven
Token seven
Feature activation+0.000
years
Token years
Feature activation+0.000
of
Token of
Feature activation+0.000
playing
Token playing
Feature activation+0.000
experience
Token experience
Feature activation+0.000
(
Token (
Feature activation+0.000
K
TokenK
Feature activation+0.000
obe
Tokenobe
Feature activation+0.000
's
Token's
Feature activation+0.000
in
Token in
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 3 in H3.9: (feature 10306)

TOP ACTIVATIONS
MAX = 4.495

âĢĵ
Token âĢĵ
Feature activation+0.000
from
Token from
Feature activation+0.000
gas
Token gas
Feature activation+0.000
station
Token station
Feature activation+0.000
attendants
Token attendants
Feature activation+0.000
to
Token to
Feature activation+4.495
elite
Token elite
Feature activation+0.000
music
Token music
Feature activation+0.000
promoters
Token promoters
Feature activation+0.000
and
Token and
Feature activation+0.000
executives
Token executives
Feature activation+0.170
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
cos
Token cos
Feature activation+0.000
mology
Tokenmology
Feature activation+0.000
,
Token,
Feature activation+0.000
to
Token to
Feature activation+4.377
public
Token public
Feature activation+0.000
outreach
Token outreach
Feature activation+0.000
where
Token where
Feature activation+0.009
he
Token he
Feature activation+0.000
strives
Token strives
Feature activation+0.000
hit
Token hit
Feature activation+0.246
testing
Token testing
Feature activation+0.327
of
Token of
Feature activation+1.676
views
Token views
Feature activation+0.665
,
Token,
Feature activation+2.790
to
Token to
Feature activation+4.270
all
Token all
Feature activation+1.859
the
Token the
Feature activation+0.000
nuances
Token nuances
Feature activation+0.000
of
Token of
Feature activation+0.000
text
Token text
Feature activation+0.000
acoustic
Token acoustic
Feature activation+0.000
strings
Token strings
Feature activation+0.000
and
Token and
Feature activation+0.000
collective
Token collective
Feature activation+0.000
chants
Token chants
Feature activation+0.000
to
Token to
Feature activation+4.216
thunder
Token thunder
Feature activation+0.000
ous
Tokenous
Feature activation+0.000
power
Token power
Feature activation+0.000
trio
Token trio
Feature activation+0.000
moves
Token moves
Feature activation+0.000
They
TokenThey
Feature activation+0.000
're
Token're
Feature activation+0.000
talking
Token talking
Feature activation+0.000
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.736
position
Token position
Feature activation+4.201
who
Token who
Feature activation+0.000
've
Token've
Feature activation+0.182
never
Token never
Feature activation+0.000
been
Token been
Feature activation+0.000
asked
Token asked
Feature activation+0.000
worker
Token worker
Feature activation+0.000
control
Token control
Feature activation+0.000
and
Token and
Feature activation+0.000
anarchist
Token anarchist
Feature activation+0.000
organizing
Token organizing
Feature activation+0.000
to
Token to
Feature activation+4.194
authoritative
Token authoritative
Feature activation+0.050
and
Token and
Feature activation+0.000
government
Token government
Feature activation+0.000
controlled
Token controlled
Feature activation+0.000
.
Token.
Feature activation+0.000
19
Token 19
Feature activation+0.000
points
Token points
Feature activation+0.000
in
Token in
Feature activation+0.000
2009
Token 2009
Feature activation+0.000
,
Token,
Feature activation+0.000
to
Token to
Feature activation+4.191
50
Token 50
Feature activation+0.000
points
Token points
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
summer
Token summer
Feature activation+0.000
,
Token,
Feature activation+0.000
from
Token from
Feature activation+0.016
the
Token the
Feature activation+0.744
one
Token one
Feature activation+0.673
shot
Token shot
Feature activation+0.000
onward
Token onward
Feature activation+4.157
I
Token I
Feature activation+2.200
found
Token found
Feature activation+0.344
Mur
Token Mur
Feature activation+0.000
ci
Tokenci
Feature activation+0.000
el
Tokenel
Feature activation+0.000
g
Token g
Feature activation+0.000
ospels
Tokenospels
Feature activation+0.000
and
Token and
Feature activation+0.000
Revelation
Token Revelation
Feature activation+0.000
,
Token,
Feature activation+1.370
to
Token to
Feature activation+4.114
various
Token various
Feature activation+0.000
movements
Token movements
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
2
Token 2
Feature activation+0.000
superb
Token superb
Feature activation+0.000
Hor
Token Hor
Feature activation+0.000
rible
Tokenrible
Feature activation+0.000
Hist
Token Hist
Feature activation+0.000
ories
Tokenories
Feature activation+0.041
to
Token to
Feature activation+4.114
BBC
Token BBC
Feature activation+0.000
2
Token2
Feature activation+0.000
's
Token's
Feature activation+0.000
recent
Token recent
Feature activation+0.000
Tud
Token Tud
Feature activation+0.000
the
Token the
Feature activation+0.000
PT
Token PT
Feature activation+0.000
Cruz
Token Cruz
Feature activation+0.000
er
Tokener
Feature activation+0.000
,
Token,
Feature activation+1.195
to
Token to
Feature activation+4.103
last
Token last
Feature activation+0.000
year
Token year
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
management
Token management
Feature activation+0.000
to
Token to
Feature activation+3.012
the
Token the
Feature activation+1.964
local
Token local
Feature activation+0.000
prosecutor
Token prosecutor
Feature activation+0.000
to
Token to
Feature activation+4.080
the
Token the
Feature activation+1.541
police
Token police
Feature activation+0.000
department
Token department
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+3.291
to
Token to
Feature activation+1.198
done
Token done
Feature activation+0.000
something
Token something
Feature activation+0.000
"
Token "
Feature activation+0.000
wrong
Tokenwrong
Feature activation+0.000
,"
Token,"
Feature activation+1.252
to
Token to
Feature activation+4.053
dis
Token dis
Feature activation+0.000
please
Tokenplease
Feature activation+0.000
either
Token either
Feature activation+0.000
the
Token the
Feature activation+0.000
adult
Token adult
Feature activation+0.000
a
Token a
Feature activation+0.037
com
Token com
Feature activation+0.000
ical
Tokenical
Feature activation+0.000
way
Token way
Feature activation+0.000
,
Token,
Feature activation+2.256
to
Token to
Feature activation+4.027
pet
Token pet
Feature activation+0.000
owners
Token owners
Feature activation+0.000
snapping
Token snapping
Feature activation+0.000
photos
Token photos
Feature activation+0.000
of
Token of
Feature activation+0.000
,
Token,
Feature activation+0.000
from
Token from
Feature activation+0.000
street
Token street
Feature activation+0.000
activ
Token activ
Feature activation+0.000
ations
Tokenations
Feature activation+0.568
to
Token to
Feature activation+4.017
live
Token live
Feature activation+1.184
entertainment
Token entertainment
Feature activation+0.000
and
Token and
Feature activation+0.000
family
Token family
Feature activation+0.549
attractions
Token attractions
Feature activation+0.000
.
Token.
Feature activation+0.000
26
Token 26
Feature activation+0.337
,
Token,
Feature activation+0.000
2015
Token 2015
Feature activation+2.513
,
Token,
Feature activation+0.000
through
Token through
Feature activation+4.009
Jan
Token Jan
Feature activation+0.000
.
Token.
Feature activation+0.000
5
Token 5
Feature activation+0.000
,
Token,
Feature activation+0.000
2016
Token 2016
Feature activation+0.700
elite
Token elite
Feature activation+0.000
music
Token music
Feature activation+0.000
promoters
Token promoters
Feature activation+0.000
and
Token and
Feature activation+0.000
executives
Token executives
Feature activation+0.170
âĢĵ
Token âĢĵ
Feature activation+3.982
was
Token was
Feature activation+0.023
overwhelming
Token overwhelming
Feature activation+0.000
.
Token.
Feature activation+0.000
Even
Token Even
Feature activation+0.000
dogs
Token dogs
Feature activation+0.000
public
Token public
Feature activation+0.000
ownership
Token ownership
Feature activation+0.000
of
Token of
Feature activation+0.842
essential
Token essential
Feature activation+0.000
infrastructure
Token infrastructure
Feature activation+0.000
to
Token to
Feature activation+3.979
milk
Token milk
Feature activation+0.000
for
Token for
Feature activation+0.000
kids
Token kids
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
with
Token with
Feature activation+0.000
Japan
Token Japan
Feature activation+0.000
and
Token and
Feature activation+0.000
Germany
Token Germany
Feature activation+0.000
)
Token)
Feature activation+0.000
to
Token to
Feature activation+3.974
the
Token the
Feature activation+2.455
a
Token a
Feature activation+0.604
hist
Tokenhist
Feature activation+0.000
orical
Tokenorical
Feature activation+0.000
,
Token,
Feature activation+0.000
recognizing
Token recognizing
Feature activation+0.000
the
Token the
Feature activation+0.000
contents
Token contents
Feature activation+0.000
of
Token of
Feature activation+0.000
images
Token images
Feature activation+0.000
to
Token to
Feature activation+3.969
holding
Token holding
Feature activation+0.000
short
Token short
Feature activation+0.000
text
Token text
Feature activation+0.000
or
Token or
Feature activation+0.000
voice
Token voice
Feature activation+0.000

Top DFA by src position
MAX = 5.143

that
Token that
Feature activation-0.004
Top resid features:
overt
Token overt
Feature activation+0.006
Top resid features:
ook
Tokenook
Feature activation-0.002
Top resid features:
everyone
Token everyone
Feature activation+0.002
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.045
Top resid features:
from
Token from
Feature activation+5.143
Top resid features:
gas
Token gas
Feature activation+0.029
Top resid features:
station
Token station
Feature activation-0.055
Top resid features:
attendants
Token attendants
Feature activation+0.050
Top resid features:
to
Token to
Feature activation-0.047
Top resid features:
elite
Token elite
Feature activation+0.000
Top resid features:
s
Tokens
Feature activation+0.001
Top resid features:
work
Token work
Feature activation+0.016
Top resid features:
in
Token in
Feature activation+0.020
Top resid features:
science
Token science
Feature activation+0.001
Top resid features:
ranges
Token ranges
Feature activation+0.004
Top resid features:
from
Token from
Feature activation+4.717
Top resid features:
teaching
Token teaching
Feature activation+0.016
Top resid features:
and
Token and
Feature activation-0.003
Top resid features:
research
Token research
Feature activation+0.002
Top resid features:
into
Token into
Feature activation-0.001
Top resid features:
black
Token black
Feature activation-0.000
Top resid features:
to
Token to
Feature activation-0.002
Top resid features:
manage
Token manage
Feature activation+0.022
Top resid features:
everything
Token everything
Feature activation+0.023
Top resid features:
yourself
Token yourself
Feature activation+0.003
Top resid features:
,
Token,
Feature activation+0.018
Top resid features:
from
Token from
Feature activation+4.638
Top resid features:
hit
Token hit
Feature activation+0.011
Top resid features:
testing
Token testing
Feature activation+0.016
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
views
Token views
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.148
Top resid features:
both
Token both
Feature activation+0.002
Top resid features:
new
Token new
Feature activation+0.013
Top resid features:
and
Token and
Feature activation-0.003
Top resid features:
familiar
Token familiar
Feature activation+0.009
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
from
Token from
Feature activation+4.654
Top resid features:
acoustic
Token acoustic
Feature activation-0.005
Top resid features:
strings
Token strings
Feature activation-0.008
Top resid features:
and
Token and
Feature activation-0.017
Top resid features:
collective
Token collective
Feature activation+0.019
Top resid features:
chants
Token chants
Feature activation+0.060
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
"
Token"
Feature activation+0.026
Top resid features:
They
TokenThey
Feature activation+0.002
Top resid features:
're
Token're
Feature activation-0.009
Top resid features:
talking
Token talking
Feature activation-0.002
Top resid features:
from
Token from
Feature activation+4.808
Top resid features:
a
Token a
Feature activation-0.037
Top resid features:
position
Token position
Feature activation+0.052
Top resid features:
who
Token who
Feature activation+0.000
Top resid features:
've
Token've
Feature activation+0.000
Top resid features:
never
Token never
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.011
Top resid features:
its
Token its
Feature activation-0.004
Top resid features:
organized
Token organized
Feature activation-0.001
Top resid features:
labor
Token labor
Feature activation-0.001
Top resid features:
,
Token,
Feature activation+0.013
Top resid features:
from
Token from
Feature activation+4.718
Top resid features:
advocating
Token advocating
Feature activation+0.057
Top resid features:
worker
Token worker
Feature activation-0.012
Top resid features:
control
Token control
Feature activation-0.001
Top resid features:
and
Token and
Feature activation-0.011
Top resid features:
anarchist
Token anarchist
Feature activation-0.007
Top resid features:
gap
Token gap
Feature activation-0.012
Top resid features:
on
Token on
Feature activation+0.005
Top resid features:
that
Token that
Feature activation-0.008
Top resid features:
question
Token question
Feature activation+0.000
Top resid features:
went
Token went
Feature activation+0.005
Top resid features:
from
Token from
Feature activation+4.798
Top resid features:
13
Token 13
Feature activation-0.001
Top resid features:
points
Token points
Feature activation-0.009
Top resid features:
in
Token in
Feature activation+0.007
Top resid features:
1994
Token 1994
Feature activation-0.006
Top resid features:
to
Token to
Feature activation-0.013
Top resid features:
going
Token going
Feature activation+0.002
Top resid features:
into
Token into
Feature activation-0.003
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
series
Token series
Feature activation-0.018
Top resid features:
,
Token,
Feature activation+0.053
Top resid features:
from
Token from
Feature activation+4.673
Top resid features:
the
Token the
Feature activation+0.058
Top resid features:
one
Token one
Feature activation-0.024
Top resid features:
shot
Token shot
Feature activation+0.073
Top resid features:
onward
Token onward
Feature activation+0.003
Top resid features:
I
Token I
Feature activation+0.000
Top resid features:
ated
Tokenated
Feature activation-0.003
Top resid features:
through
Token through
Feature activation+0.042
Top resid features:
later
Token later
Feature activation+0.001
Top resid features:
ages
Token ages
Feature activation+0.013
Top resid features:
,
Token,
Feature activation+0.005
Top resid features:
from
Token from
Feature activation+4.430
Top resid features:
the
Token the
Feature activation+0.010
Top resid features:
Dead
Token Dead
Feature activation-0.001
Top resid features:
Sea
Token Sea
Feature activation-0.002
Top resid features:
Scrolls
Token Scrolls
Feature activation+0.000
Top resid features:
community
Token community
Feature activation+0.004
Top resid features:
admirable
Token admirable
Feature activation+0.004
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.056
Top resid features:
all
Token all
Feature activation-0.019
Top resid features:
the
Token the
Feature activation+0.012
Top resid features:
way
Token way
Feature activation+0.029
Top resid features:
from
Token from
Feature activation+4.605
Top resid features:
CB
Token CB
Feature activation+0.004
Top resid features:
BC
TokenBC
Feature activation-0.008
Top resid features:
's
Token's
Feature activation+0.002
Top resid features:
superb
Token superb
Feature activation-0.000
Top resid features:
Hor
Token Hor
Feature activation+0.015
Top resid features:
of
Token of
Feature activation+0.008
Top resid features:
riches
Token riches
Feature activation+0.001
Top resid features:
this
Token this
Feature activation-0.001
Top resid features:
year
Token year
Feature activation+0.005
Top resid features:
,
Token,
Feature activation+0.009
Top resid features:
from
Token from
Feature activation+4.541
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
240
Token 240
Feature activation-0.001
Top resid features:
million
Token million
Feature activation-0.011
Top resid features:
dollar
Token dollar
Feature activation-0.001
Top resid features:
Can
Token Can
Feature activation-0.003
Top resid features:
every
Token every
Feature activation+0.018
Top resid features:
level
Token level
Feature activation+0.011
Top resid features:
of
Token of
Feature activation+0.020
Top resid features:
enforcement
Token enforcement
Feature activation+0.003
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.090
Top resid features:
from
Token from
Feature activation+4.612
Top resid features:
city
Token city
Feature activation-0.001
Top resid features:
management
Token management
Feature activation+0.007
Top resid features:
to
Token to
Feature activation-0.030
Top resid features:
the
Token the
Feature activation-0.006
Top resid features:
local
Token local
Feature activation-0.006
Top resid features:
a
Token a
Feature activation-0.004
Top resid features:
child
Token child
Feature activation-0.001
Top resid features:
would
Token would
Feature activation+0.001
Top resid features:
feel
Token feel
Feature activation+0.003
Top resid features:
,
Token,
Feature activation+0.011
Top resid features:
from
Token from
Feature activation+4.560
Top resid features:
sensing
Token sensing
Feature activation+0.008
Top resid features:
that
Token that
Feature activation+0.001
Top resid features:
they
Token they
Feature activation+0.006
Top resid features:
've
Token've
Feature activation+0.001
Top resid features:
somehow
Token somehow
Feature activation+0.003
Top resid features:
being
Token being
Feature activation+0.001
Top resid features:
entertaining
Token entertaining
Feature activation+0.001
Top resid features:
or
Token or
Feature activation-0.002
Top resid features:
humorous
Token humorous
Feature activation+0.008
Top resid features:
.
Token.
Feature activation+0.025
Top resid features:
From
Token From
Feature activation+4.314
Top resid features:
parents
Token parents
Feature activation+0.008
Top resid features:
shaming
Token shaming
Feature activation+0.001
Top resid features:
their
Token their
Feature activation-0.029
Top resid features:
kids
Token kids
Feature activation+0.009
Top resid features:
in
Token in
Feature activation+0.021
Top resid features:
little
Token little
Feature activation+0.012
Top resid features:
bit
Token bit
Feature activation+0.007
Top resid features:
of
Token of
Feature activation+0.008
Top resid features:
everything
Token everything
Feature activation+0.025
Top resid features:
,
Token,
Feature activation+0.023
Top resid features:
from
Token from
Feature activation+4.627
Top resid features:
street
Token street
Feature activation+0.000
Top resid features:
activ
Token activ
Feature activation+0.033
Top resid features:
ations
Tokenations
Feature activation+0.043
Top resid features:
to
Token to
Feature activation-0.071
Top resid features:
live
Token live
Feature activation+0.000
Top resid features:
World
Token World
Feature activation-0.001
Top resid features:
Junior
Token Junior
Feature activation-0.002
Top resid features:
Championship
Token Championship
Feature activation-0.030
Top resid features:
in
Token in
Feature activation+0.009
Top resid features:
Finland
Token Finland
Feature activation-0.006
Top resid features:
from
Token from
Feature activation+4.684
Top resid features:
Dec
Token Dec
Feature activation-0.006
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
26
Token 26
Feature activation-0.013
Top resid features:
,
Token,
Feature activation-0.005
Top resid features:
2015
Token 2015
Feature activation-0.107
Top resid features:
that
Token that
Feature activation-0.009
Top resid features:
overt
Token overt
Feature activation+0.011
Top resid features:
ook
Tokenook
Feature activation+0.000
Top resid features:
everyone
Token everyone
Feature activation-0.009
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.078
Top resid features:
from
Token from
Feature activation+4.352
Top resid features:
gas
Token gas
Feature activation+0.011
Top resid features:
station
Token station
Feature activation-0.005
Top resid features:
attendants
Token attendants
Feature activation-0.003
Top resid features:
to
Token to
Feature activation-0.017
Top resid features:
elite
Token elite
Feature activation+0.001
Top resid features:
dismantling
Token dismantling
Feature activation+0.016
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
our
Token our
Feature activation-0.003
Top resid features:
birth
Token birth
Feature activation+0.001
Top resid features:
rights
Tokenrights
Feature activation-0.012
Top resid features:
from
Token from
Feature activation+4.532
Top resid features:
public
Token public
Feature activation+0.009
Top resid features:
ownership
Token ownership
Feature activation+0.005
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
essential
Token essential
Feature activation+0.023
Top resid features:
infrastructure
Token infrastructure
Feature activation-0.030
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
Iron
Token Iron
Feature activation+0.001
Top resid features:
IV
Token IV
Feature activation-0.002
Top resid features:
that
Token that
Feature activation+0.076
Top resid features:
range
Token range
Feature activation+0.001
Top resid features:
from
Token from
Feature activation+4.441
Top resid features:
the
Token the
Feature activation+0.018
Top resid features:
historically
Token historically
Feature activation+0.015
Top resid features:
accurate
Token accurate
Feature activation+0.004
Top resid features:
(
Token (
Feature activation+0.030
Top resid features:
war
Tokenwar
Feature activation-0.003
Top resid features:
do
Token do
Feature activation+0.004
Top resid features:
ever
Token ever
Feature activation+0.003
Top resid features:
more
Token more
Feature activation+0.001
Top resid features:
things
Token things
Feature activation+0.009
Top resid features:
,
Token,
Feature activation+0.042
Top resid features:
from
Token from
Feature activation+4.474
Top resid features:
recognizing
Token recognizing
Feature activation+0.119
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
contents
Token contents
Feature activation-0.015
Top resid features:
of
Token of
Feature activation+0.009
Top resid features:
images
Token images
Feature activation-0.028
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.04

Head 2: 0.11

Head 3: 0.03

Head 4: 0.02

Head 5: 0.04

Head 6: 0.04

Head 7: 0.03

Head 8: 0.05

Head 9: 0.54

Head 10: 0.05

Head 11: 0.05

Positive logits

onwards1.77

standpoint1.75

ebook1.69

DragonMagazine1.55

vantage1.46

backgrounds1.42

lems1.41

perspective1.35

anges1.33

reperto1.32

ixties1.29

netflix1.29

perspectives1.27

assetsadobe1.26

idth1.25

sburg1.25

sheer1.24

mouths1.24

onward1.24

govtrack1.23

Negative logits

buster-1.34

aliation-1.25

cure-1.20

acea-1.17

itle-1.15

-1.15

royalty-1.14

oother-1.13

////////-1.12

}}-1.11

bothering-1.10

immortality-1.09

%%-1.09

breaker-1.08

hound-1.08

Correction-1.07

Warrant-1.07

___-1.06

fou-1.06

Stro-1.06

INTERVAL 4.046 - 4.495
CONTAINS 0.001%

superb
Token superb
Feature activation+0.000
Hor
Token Hor
Feature activation+0.000
rible
Tokenrible
Feature activation+0.000
Hist
Token Hist
Feature activation+0.000
ories
Tokenories
Feature activation+0.041
to
Token to
Feature activation+4.114
BBC
Token BBC
Feature activation+0.000
2
Token2
Feature activation+0.000
's
Token's
Feature activation+0.000
recent
Token recent
Feature activation+0.000
Tud
Token Tud
Feature activation+0.000
g
Token g
Feature activation+0.000
ospels
Tokenospels
Feature activation+0.000
and
Token and
Feature activation+0.000
Revelation
Token Revelation
Feature activation+0.000
,
Token,
Feature activation+1.370
to
Token to
Feature activation+4.114
various
Token various
Feature activation+0.000
movements
Token movements
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
2
Token 2
Feature activation+0.000
hit
Token hit
Feature activation+0.246
testing
Token testing
Feature activation+0.327
of
Token of
Feature activation+1.676
views
Token views
Feature activation+0.665
,
Token,
Feature activation+2.790
to
Token to
Feature activation+4.270
all
Token all
Feature activation+1.859
the
Token the
Feature activation+0.000
nuances
Token nuances
Feature activation+0.000
of
Token of
Feature activation+0.000
text
Token text
Feature activation+0.000
,
Token,
Feature activation+0.000
from
Token from
Feature activation+0.016
the
Token the
Feature activation+0.744
one
Token one
Feature activation+0.673
shot
Token shot
Feature activation+0.000
onward
Token onward
Feature activation+4.157
I
Token I
Feature activation+2.200
found
Token found
Feature activation+0.344
Mur
Token Mur
Feature activation+0.000
ci
Tokenci
Feature activation+0.000
el
Tokenel
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
cos
Token cos
Feature activation+0.000
mology
Tokenmology
Feature activation+0.000
,
Token,
Feature activation+0.000
to
Token to
Feature activation+4.377
public
Token public
Feature activation+0.000
outreach
Token outreach
Feature activation+0.000
where
Token where
Feature activation+0.009
he
Token he
Feature activation+0.000
strives
Token strives
Feature activation+0.000

INTERVAL 3.596 - 4.046
CONTAINS 0.003%

of
Token of
Feature activation+0.000
key
Token key
Feature activation+0.000
US
Token US
Feature activation+0.000
surveillance
Token surveillance
Feature activation+0.000
powers
Token powers
Feature activation+0.000
to
Token to
Feature activation+3.738
more
Token more
Feature activation+0.000
moderate
Token moderate
Feature activation+0.000
reforms
Token reforms
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
from
Token from
Feature activation+0.000
13
Token 13
Feature activation+0.000
points
Token points
Feature activation+0.867
in
Token in
Feature activation+0.000
1994
Token 1994
Feature activation+0.000
to
Token to
Feature activation+3.621
19
Token 19
Feature activation+0.000
points
Token points
Feature activation+0.000
in
Token in
Feature activation+0.000
2009
Token 2009
Feature activation+0.000
,
Token,
Feature activation+0.000
l
Tokenl
Feature activation+0.000
ours
Tokenours
Feature activation+0.000
to
Token to
Feature activation+3.477
shoe
Token shoe
Feature activation+0.000
retailers
Token retailers
Feature activation+0.000
to
Token to
Feature activation+3.786
online
Token online
Feature activation+0.000
gambling
Token gambling
Feature activation+0.000
sites
Token sites
Feature activation+0.000
are
Token are
Feature activation+0.083
beginning
Token beginning
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
from
Token from
Feature activation+0.000
Donald
Token Donald
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
critic
Token critic
Feature activation+0.000
to
Token to
Feature activation+3.714
Donald
Token Donald
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
flatt
Token flatt
Feature activation+0.000
erer
Tokenerer
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+3.209
way
Token way
Feature activation+0.000
news
Token news
Feature activation+0.000
stories
Token stories
Feature activation+0.000
are
Token are
Feature activation+0.058
covered
Token covered
Feature activation+0.000
to
Token to
Feature activation+3.709
national
Token national
Feature activation+0.000
political
Token political
Feature activation+0.000
issues
Token issues
Feature activation+0.000
.
Token.
Feature activation+0.000
We
Token We
Feature activation+0.000

INTERVAL 3.147 - 3.596
CONTAINS 0.006%

ant
Token ant
Feature activation+0.000
icol
Tokenicol
Feature activation+0.000
onial
Tokenonial
Feature activation+0.303
rebell
Token rebell
Feature activation+0.000
ions
Tokenions
Feature activation+0.000
up
Token up
Feature activation+3.260
through
Token through
Feature activation+2.760
the
Token the
Feature activation+0.000
civil
Token civil
Feature activation+0.000
rights
Token rights
Feature activation+0.000
,
Token,
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
boiler
Token boiler
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.048
start
Token start
Feature activation+3.351
,
Token,
Feature activation+0.000
calling
Token calling
Feature activation+0.000
it
Token it
Feature activation+0.000
an
Token an
Feature activation+0.000
un
Token un
Feature activation+0.000
September
Token September
Feature activation+1.642
19
Token 19
Feature activation+0.619
,
Token,
Feature activation+0.000
1950
Token 1950
Feature activation+2.236
,
Token,
Feature activation+0.000
on
Token on
Feature activation+3.536
and
Token and
Feature activation+0.000
around
Token around
Feature activation+0.000
the
Token the
Feature activation+0.000
Sob
Token Sob
Feature activation+0.000
uk
Tokenuk
Feature activation+0.000
to
Token to
Feature activation+3.714
Donald
Token Donald
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
flatt
Token flatt
Feature activation+0.000
erer
Tokenerer
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+3.209
the
Token the
Feature activation+0.882
senator
Token senator
Feature activation+0.000
issued
Token issued
Feature activation+0.000
a
Token a
Feature activation+0.000
curious
Token curious
Feature activation+0.000
to
Token to
Feature activation+3.615
CH
Token CH
Feature activation+0.000
i
Tokeni
Feature activation+0.000
Ps
TokenPs
Feature activation+0.000
,
Token,
Feature activation+0.000
to
Token to
Feature activation+3.573
Knot
Token Knot
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Landing
Token Landing
Feature activation+0.000

INTERVAL 2.697 - 3.147
CONTAINS 0.008%

2003
Token 2003
Feature activation+0.000
,
Token,
Feature activation+0.000
page
Token page
Feature activation+1.888
133
Token 133
Feature activation+0.000
),
Token),
Feature activation+2.281
we
Token we
Feature activation+2.927
learn
Token learn
Feature activation+0.658
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
calculations
Token calculations
Feature activation+0.000
of
Token of
Feature activation+0.000
2
Token2
Feature activation+0.000
on
Token on
Feature activation+2.640
the
Token the
Feature activation+0.302
V
Token V
Feature activation+0.000
10
Token10
Feature activation+0.000
to
Token to
Feature activation+2.724
f
Token f
Feature activation+0.000
/
Token/
Feature activation+0.000
1
Token1
Feature activation+0.000
.
Token.
Feature activation+0.000
9
Token9
Feature activation+0.000
Bra
Token Bra
Feature activation+0.000
xton
Tokenxton
Feature activation+0.000
Bur
Token Bur
Feature activation+0.000
me
Tokenme
Feature activation+0.000
ister
Tokenister
Feature activation+0.000
to
Token to
Feature activation+2.851
name
Token name
Feature activation+0.000
a
Token a
Feature activation+0.000
few
Token few
Feature activation+0.000
).
Token).
Feature activation+0.000
So
Token So
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
From
TokenFrom
Feature activation+0.000
Stephen
Token Stephen
Feature activation+0.000
Schneider
Token Schneider
Feature activation+0.000
:
Token:
Feature activation+2.398
Ċ
TokenĊ
Feature activation+3.067
Ċ
TokenĊ
Feature activation+2.757
MC
TokenMC
Feature activation+0.000
M
TokenM
Feature activation+0.000
:
Token:
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.721
dj
Token dj
Feature activation+0.000
ango
Tokenango
Feature activation+0.000
.
Token.
Feature activation+0.000
cont
Tokencont
Feature activation+0.000
rib
Tokenrib
Feature activation+0.000
import
Token import
Feature activation+2.844
admin
Token admin
Feature activation+0.000
url
Token url
Feature activation+0.000
pattern
Tokenpattern
Feature activation+0.000
s
Tokens
Feature activation+0.000
=
Token =
Feature activation+0.000

INTERVAL 2.248 - 2.697
CONTAINS 0.012%

,
Token,
Feature activation+0.000
then
Token then
Feature activation+0.000
fired
Token fired
Feature activation+0.000
from
Token from
Feature activation+0.000
20
Token 20
Feature activation+0.529
yards
Token yards
Feature activation+2.612
out
Token out
Feature activation+0.780
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
3
Token 3
Feature activation+0.000
-
Token-
Feature activation+0.000
Hamilton
Token Hamilton
Feature activation+0.000
having
Token having
Feature activation+0.000
to
Token to
Feature activation+0.000
block
Token block
Feature activation+0.000
from
Token from
Feature activation+0.000
close
Token close
Feature activation+2.376
range
Token range
Feature activation+1.194
.
Token.
Feature activation+0.000
Hearts
Token Hearts
Feature activation+0.000
knew
Token knew
Feature activation+0.000
patience
Token patience
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.171
small
Token small
Feature activation+0.316
stage
Token stage
Feature activation+1.603
to
Token to
Feature activation+2.633
the
Token the
Feature activation+2.308
big
Token big
Feature activation+0.000
screen
Token screen
Feature activation+0.415
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
saying
Token saying
Feature activation+0.000
from
Token from
Feature activation+0.000
here
Token here
Feature activation+1.682
on
Tokenon
Feature activation+0.720
out
Token out
Feature activation+2.436
merely
Token merely
Feature activation+0.000
compounds
Token compounds
Feature activation+0.040
the
Token the
Feature activation+0.000
racism
Token racism
Feature activation+0.000
.
Token.
Feature activation+0.000
A
TokenA
Feature activation+0.000
Century
Token Century
Feature activation+0.000
of
Token of
Feature activation+0.000
Vacc
Token Vacc
Feature activation+0.000
ination
Tokenination
Feature activation+0.000
":
Token":
Feature activation+2.292
Other
TokenOther
Feature activation+0.000
sources
Token sources
Feature activation+0.000
have
Token have
Feature activation+0.000
shown
Token shown
Feature activation+0.000
that
Token that
Feature activation+0.000

INTERVAL 1.798 - 2.248
CONTAINS 0.014%

to
Token to
Feature activation+0.000
learn
Token learn
Feature activation+0.000
correct
Token correct
Feature activation+0.000
technique
Token technique
Feature activation+0.000
from
Token from
Feature activation+0.000
scratch
Token scratch
Feature activation+2.015
on
Token on
Feature activation+1.585
an
Token an
Feature activation+0.370
entirely
Token entirely
Feature activation+0.000
new
Token new
Feature activation+0.000
instrument
Token instrument
Feature activation+0.000
Any
Token Any
Feature activation+0.000
gains
Token gains
Feature activation+0.000
from
Token from
Feature activation+0.000
this
Token this
Feature activation+0.000
point
Token point
Feature activation+1.622
forward
Token forward
Feature activation+2.051
are
Token are
Feature activation+0.000
likely
Token likely
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
incremental
Token incremental
Feature activation+0.000
days
Token days
Feature activation+0.000
from
Token from
Feature activation+0.000
10
Token 10
Feature activation+0.000
,
Token,
Feature activation+0.000
500
Token500
Feature activation+0.000
to
Token to
Feature activation+1.993
over
Token over
Feature activation+0.000
28
Token 28
Feature activation+0.000
,
Token,
Feature activation+0.000
200
Token200
Feature activation+0.000
,
Token,
Feature activation+0.000
Union
Token Union
Feature activation+0.000
of
Token of
Feature activation+0.000
Concern
Token Concern
Feature activation+0.000
ed
Tokened
Feature activation+0.000
Scientists
Token Scientists
Feature activation+0.000
:
Token:
Feature activation+2.163
Ċ
TokenĊ
Feature activation+3.054
Ċ
TokenĊ
Feature activation+2.584
However
TokenHowever
Feature activation+0.000
,
Token,
Feature activation+0.000
N
Token N
Feature activation+0.000
everything
Token everything
Feature activation+0.000
from
Token from
Feature activation+0.000
armored
Token armored
Feature activation+0.000
vehicles
Token vehicles
Feature activation+0.000
to
Token to
Feature activation+3.556
the
Token the
Feature activation+2.241
somewhat
Token somewhat
Feature activation+0.000
notorious
Token notorious
Feature activation+0.000
Nir
Token Nir
Feature activation+0.000
b
Tokenb
Feature activation+0.000
he
Tokenhe
Feature activation+0.000

INTERVAL 1.349 - 1.798
CONTAINS 0.024%

4
Token 4
Feature activation+0.343
p
Token p
Feature activation+0.640
.
Token.
Feature activation+0.000
m
Tokenm
Feature activation+2.123
.
Token.
Feature activation+0.000
to
Token to
Feature activation+1.745
9
Token 9
Feature activation+0.775
p
Token p
Feature activation+0.474
.
Token.
Feature activation+0.000
m
Tokenm
Feature activation+1.671
.,
Token.,
Feature activation+1.314
of
Token of
Feature activation+0.000
comedy
Token comedy
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.171
small
Token small
Feature activation+0.316
stage
Token stage
Feature activation+1.603
to
Token to
Feature activation+2.633
the
Token the
Feature activation+2.308
big
Token big
Feature activation+0.000
screen
Token screen
Feature activation+0.415
.
Token.
Feature activation+0.000
and
Token and
Feature activation+0.000
do
Token do
Feature activation+0.000
it
Token it
Feature activation+0.000
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.079
place
Token place
Feature activation+1.727
she
Token she
Feature activation+0.633
loves
Token loves
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+0.044
From
TokenFrom
Feature activation+0.000
a
Token a
Feature activation+0.506
starting
Token starting
Feature activation+0.000
rotation
Token rotation
Feature activation+0.000
standpoint
Token standpoint
Feature activation+1.573
,
Token,
Feature activation+1.461
we
Token we
Feature activation+1.070
had
Token had
Feature activation+0.000
higher
Token higher
Feature activation+0.000
expectations
Token expectations
Feature activation+0.000
from
Token from
Feature activation+0.000
several
Token several
Feature activation+0.000
days
Token days
Feature activation+0.000
to
Token to
Feature activation+0.765
weeks
Token weeks
Feature activation+0.000
to
Token to
Feature activation+1.526
go
Token go
Feature activation+0.594
through
Token through
Feature activation+0.000
the
Token the
Feature activation+0.000
legal
Token legal
Feature activation+0.000
process
Token process
Feature activation+0.000

INTERVAL 0.899 - 1.349
CONTAINS 0.031%

.
Token.
Feature activation+0.000
From
Token From
Feature activation+0.000
these
Token these
Feature activation+0.622
interactions
Token interactions
Feature activation+0.000
,
Token,
Feature activation+1.850
the
Token the
Feature activation+0.929
bot
Token bot
Feature activation+0.000
can
Token can
Feature activation+0.600
understand
Token understand
Feature activation+0.000
intent
Token intent
Feature activation+0.000
(
Token (
Feature activation+0.000
$
Token $
Feature activation+0.000
7
Token7
Feature activation+0.000
.
Token.
Feature activation+0.000
8
Token8
Feature activation+0.000
million
Token million
Feature activation+0.000
to
Token to
Feature activation+0.985
$
Token $
Feature activation+0.000
22
Token22
Feature activation+0.000
.
Token.
Feature activation+0.000
8
Token8
Feature activation+0.000
million
Token million
Feature activation+0.000
We
TokenWe
Feature activation+0.000
are
Token are
Feature activation+0.000
women
Token women
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
right
Token right
Feature activation+1.172
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
left
Token left
Feature activation+0.007
,
Token,
Feature activation+0.000
Jews
Token Jews
Feature activation+0.000
government
Token government
Feature activation+0.000
and
Token and
Feature activation+0.000
business
Token business
Feature activation+0.000
organizations
Token organizations
Feature activation+0.000
,
Token,
Feature activation+0.131
to
Token to
Feature activation+1.304
communicate
Token communicate
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
secure
Token secure
Feature activation+0.000
way
Token way
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
website
Token website
Feature activation+0.917
:
Token:
Feature activation+2.618
âĢ
Token âĢ
Feature activation+0.931
ľ
Tokenľ
Feature activation+0.000
CA
TokenCA
Feature activation+0.000
FE
TokenFE
Feature activation+0.000
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000

INTERVAL 0.450 - 0.899
CONTAINS 0.059%

't
Token't
Feature activation+0.000
just
Token just
Feature activation+0.000
fall
Token fall
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
sky
Token sky
Feature activation+0.661
.
Token.
Feature activation+0.000
When
Token When
Feature activation+0.000
you
Token you
Feature activation+0.000
find
Token find
Feature activation+0.000
stuff
Token stuff
Feature activation+0.000
700
Token 700
Feature activation+0.000
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.000
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.000
peak
Token peak
Feature activation+0.850
of
Token of
Feature activation+0.467
1
Token 1
Feature activation+0.000
.
Token.
Feature activation+0.000
8
Token8
Feature activation+0.000
million
Token million
Feature activation+0.000
father
Token father
Feature activation+0.000
arrived
Token arrived
Feature activation+0.000
home
Token home
Feature activation+0.000
from
Token from
Feature activation+0.000
his
Token his
Feature activation+0.000
farm
Token farm
Feature activation+0.680
to
Token to
Feature activation+0.000
find
Token find
Feature activation+0.000
that
Token that
Feature activation+0.000
his
Token his
Feature activation+0.000
teenage
Token teenage
Feature activation+0.000
d
Tokend
Feature activation+0.000
protocol
Token protocol
Feature activation+0.000
from
Token from
Feature activation+0.000
2
Token 2
Feature activation+0.532
participants
Token participants
Feature activation+0.331
:
Token:
Feature activation+0.744
1
Token 1
Feature activation+0.000
with
Token with
Feature activation+0.000
a
Token a
Feature activation+0.000
large
Token large
Feature activation+0.000
phase
Token phase
Feature activation+0.000
Did
TokenDid
Feature activation+0.000
you
Token you
Feature activation+0.000
rise
Token rise
Feature activation+0.000
from
Token from
Feature activation+0.000
humble
Token humble
Feature activation+0.774
beginnings
Token beginnings
Feature activation+0.559
?
Token?
Feature activation+0.000
Did
Token Did
Feature activation+0.000
you
Token you
Feature activation+0.000
overcome
Token overcome
Feature activation+0.000
your
Token your
Feature activation+0.000

INTERVAL 0.000 - 0.450
CONTAINS 99.842%

was
Tokenwas
Feature activation+0.000
th
Tokenth
Feature activation+0.000
i
Tokeni
Feature activation+0.000
did
Token did
Feature activation+0.000
not
Token not
Feature activation+0.000
further
Token further
Feature activation+0.000
spell
Token spell
Feature activation+0.000
out
Token out
Feature activation+0.000
details
Token details
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
the
Token the
Feature activation+0.000
GOP
Token GOP
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
Chuck
Token Chuck
Feature activation+0.000
Hag
Token Hag
Feature activation+0.000
el
Tokenel
Feature activation+0.000
retiring
Token retiring
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
Senate
Token Senate
Feature activation+0.000
the
Token the
Feature activation+0.000
grade
Token grade
Feature activation+0.000
of
Token of
Feature activation+0.000
E
Token E
Feature activation+0.000
-
Token-
Feature activation+0.000
1
Token1
Feature activation+0.000
,
Token,
Feature activation+0.000
confined
Token confined
Feature activation+0.000
for
Token for
Feature activation+0.000
20
Token 20
Feature activation+0.000
years
Token years
Feature activation+0.000
GHC
Token GHC
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Haskell
Token Haskell
Feature activation+0.000
platform
Token platform
Feature activation+0.000
are
Token are
Feature activation+0.000
still
Token still
Feature activation+0.000
very
Token very
Feature activation+0.000
old
Token old
Feature activation+0.000
so
Token so
Feature activation+0.000
upgrading
Token upgrading
Feature activation+0.000
includes
Token includes
Feature activation+0.000
both
Token both
Feature activation+0.000
Harper
Token Harper
Feature activation+0.000
Lee
Token Lee
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
To
Token To
Feature activation+0.000
Kill
Token Kill
Feature activation+0.000
a
Token a
Feature activation+0.000
M
Token M
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 4 in H3.9: (feature 16086)

TOP ACTIVATIONS
MAX = 3.610

to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
payments
Token payments
Feature activation+0.000
within
Token within
Feature activation+0.000
stores
Token stores
Feature activation+0.000
,
Token,
Feature activation+3.610
they
Token they
Feature activation+2.487
can
Token can
Feature activation+0.000
make
Token make
Feature activation+0.000
Apple
Token Apple
Feature activation+0.000
Pay
Token Pay
Feature activation+0.000
ited
Tokenited
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
high
Token high
Feature activation+0.000
school
Token school
Feature activation+0.000
,
Token,
Feature activation+3.571
Buc
Token Buc
Feature activation+0.000
keyes
Tokenkeyes
Feature activation+0.000
head
Token head
Feature activation+0.000
coach
Token coach
Feature activation+0.000
Urban
Token Urban
Feature activation+0.000
warning
Token warning
Feature activation+0.000
has
Token has
Feature activation+0.384
since
Token since
Feature activation+0.000
been
Token been
Feature activation+0.000
lifted
Token lifted
Feature activation+0.000
,
Token,
Feature activation+3.492
many
Token many
Feature activation+0.049
parts
Token parts
Feature activation+0.000
of
Token of
Feature activation+0.000
Florida
Token Florida
Feature activation+0.000
continue
Token continue
Feature activation+0.000
than
Token than
Feature activation+0.000
10
Token 10
Feature activation+0.000
.
Token.
Feature activation+0.000
5
Token5
Feature activation+0.000
percent
Token percent
Feature activation+0.000
,
Token,
Feature activation+3.489
that
Token that
Feature activation+0.000
amounts
Token amounts
Feature activation+0.000
to
Token to
Feature activation+0.000
almost
Token almost
Feature activation+0.000
30
Token 30
Feature activation+0.000
to
Token to
Feature activation+0.000
implement
Token implement
Feature activation+0.000
such
Token such
Feature activation+0.000
a
Token a
Feature activation+0.000
program
Token program
Feature activation+0.000
,
Token,
Feature activation+3.456
any
Token any
Feature activation+0.000
materials
Token materials
Feature activation+0.000
developed
Token developed
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
bs
Tokenbs
Feature activation+0.000
has
Token has
Feature activation+0.052
been
Token been
Feature activation+0.000
made
Token made
Feature activation+0.000
public
Token public
Feature activation+0.000
,
Token,
Feature activation+3.378
we
Token we
Feature activation+2.981
can
Token can
Feature activation+0.078
't
Token't
Feature activation+0.000
update
Token update
Feature activation+0.000
the
Token the
Feature activation+0.000
it
Token it
Feature activation+0.634
was
Token was
Feature activation+0.672
barely
Token barely
Feature activation+0.000
percept
Token percept
Feature activation+0.000
ible
Tokenible
Feature activation+0.000
,
Token,
Feature activation+3.332
I
Token I
Feature activation+2.409
saw
Token saw
Feature activation+0.000
something
Token something
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
though
Token though
Feature activation+0.000
they
Token they
Feature activation+0.000
struggled
Token struggled
Feature activation+0.000
against
Token against
Feature activation+0.000
it
Token it
Feature activation+0.000
,
Token,
Feature activation+3.270
were
Token were
Feature activation+0.029
immersed
Token immersed
Feature activation+0.000
in
Token in
Feature activation+0.000
an
Token an
Feature activation+0.000
ideology
Token ideology
Feature activation+0.000
batting
Token batting
Feature activation+0.000
title
Token title
Feature activation+0.000
once
Token once
Feature activation+0.000
(
Token (
Feature activation+0.000
2016
Token2016
Feature activation+0.000
),
Token),
Feature activation+3.185
compared
Token compared
Feature activation+0.886
to
Token to
Feature activation+0.000
each
Token each
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
below
Token below
Feature activation+0.000
their
Token their
Feature activation+0.000
standing
Token standing
Feature activation+0.000
last
Token last
Feature activation+0.000
fall
Token fall
Feature activation+0.000
,
Token,
Feature activation+3.180
they
Token they
Feature activation+2.539
have
Token have
Feature activation+0.000
nevertheless
Token nevertheless
Feature activation+0.000
managed
Token managed
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.080
documentation
Token documentation
Feature activation+0.000
is
Token is
Feature activation+0.000
comprehensive
Token comprehensive
Feature activation+0.000
,
Token,
Feature activation+2.896
it
Token it
Feature activation+3.174
isn
Token isn
Feature activation+0.330
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
as
Token as
Feature activation+0.000
not
Token not
Feature activation+0.507
publicly
Token publicly
Feature activation+0.000
discussed
Token discussed
Feature activation+0.000
starvation
Token starvation
Feature activation+0.000
deaths
Token deaths
Feature activation+0.000
,
Token,
Feature activation+3.143
it
Token it
Feature activation+2.868
acknowledges
Token acknowledges
Feature activation+0.000
there
Token there
Feature activation+0.000
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
a
Token a
Feature activation+0.000
researcher
Token researcher
Feature activation+0.000
's
Token's
Feature activation+0.000
needs
Token needs
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.249
it
Tokenit
Feature activation+3.135
wouldn
Token wouldn
Feature activation+0.000
't
Token't
Feature activation+0.000
be
Token be
Feature activation+0.000
fair
Token fair
Feature activation+0.000
to
Token to
Feature activation+0.000
you
Token you
Feature activation+0.000
will
Token will
Feature activation+0.000
constantly
Token constantly
Feature activation+0.000
be
Token be
Feature activation+0.000
learning
Token learning
Feature activation+0.000
,
Token,
Feature activation+3.109
your
Token your
Feature activation+0.330
startup
Token startup
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
mus
Tokenmus
Feature activation+0.000
any
Token any
Feature activation+0.000
other
Token other
Feature activation+0.000
EU
Token EU
Feature activation+0.000
country
Token country
Feature activation+0.000
,
Token,
Feature activation+2.790
they
Token they
Feature activation+3.096
have
Token have
Feature activation+0.393
one
Token one
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
lowest
Token lowest
Feature activation+0.000
name
Token name
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
solo
Token solo
Feature activation+0.000
artist
Token artist
Feature activation+0.000
,
Token,
Feature activation+3.079
Jose
Token Jose
Feature activation+0.000
f
Tokenf
Feature activation+0.000
K
Token K
Feature activation+0.000
was
Token was
Feature activation+0.000
actually
Token actually
Feature activation+0.000
some
Token some
Feature activation+0.000
extent
Token extent
Feature activation+0.000
ends
Token ends
Feature activation+0.000
)
Token)
Feature activation+0.000
exist
Token exist
Feature activation+1.177
,
Token,
Feature activation+3.049
no
Token no
Feature activation+0.964
party
Token party
Feature activation+0.000
has
Token has
Feature activation+0.000
questioned
Token questioned
Feature activation+0.000
the
Token the
Feature activation+0.000
dr
Tokendr
Feature activation+0.000
agging
Tokenagging
Feature activation+0.000
incident
Token incident
Feature activation+0.000
was
Token was
Feature activation+0.418
cruel
Token cruel
Feature activation+0.254
,
Token,
Feature activation+3.017
it
Token it
Feature activation+2.642
may
Token may
Feature activation+1.676
not
Token not
Feature activation+0.036
have
Token have
Feature activation+0.000
broken
Token broken
Feature activation+0.000
has
Token has
Feature activation+0.052
been
Token been
Feature activation+0.000
made
Token made
Feature activation+0.000
public
Token public
Feature activation+0.000
,
Token,
Feature activation+3.378
we
Token we
Feature activation+2.981
can
Token can
Feature activation+0.078
't
Token't
Feature activation+0.000
update
Token update
Feature activation+0.000
the
Token the
Feature activation+0.000
affected
Token affected
Feature activation+0.000
the
Token the
Feature activation+0.000
streak
Token streak
Feature activation+0.000
in
Token in
Feature activation+0.000
recent
Token recent
Feature activation+0.000
years
Token years
Feature activation+0.000
,
Token,
Feature activation+2.962
including
Token including
Feature activation+0.000
this
Token this
Feature activation+0.000
offseason
Token offseason
Feature activation+0.000
,
Token,
Feature activation+0.000
they
Token they
Feature activation+0.244

Top DFA by src position
MAX = 4.789

numbers
Token numbers
Feature activation-0.010
Top resid features:
for
Token for
Feature activation+0.006
Top resid features:
security
Token security
Feature activation-0.002
Top resid features:
reasons
Token reasons
Feature activation-0.005
Top resid features:
.
Token.
Feature activation-0.004
Top resid features:
Though
TokenThough
Feature activation+4.450
Top resid features:
the
Token the
Feature activation+0.057
Top resid features:
iPad
Token iPad
Feature activation+0.015
Top resid features:
mini
Token mini
Feature activation-0.002
Top resid features:
3
Token 3
Feature activation+0.009
Top resid features:
and
Token and
Feature activation+0.011
Top resid features:
il
Tokenil
Feature activation+0.000
Top resid features:
Kap
Token Kap
Feature activation+0.001
Top resid features:
ur
Tokenur
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.049
Top resid features:
Overview
TokenOverview
Feature activation+0.018
Top resid features:
Though
Token Though
Feature activation+4.789
Top resid features:
Apple
Token Apple
Feature activation+0.004
Top resid features:
was
Token was
Feature activation+0.086
Top resid features:
highly
Token highly
Feature activation+0.033
Top resid features:
-
Token-
Feature activation-0.003
Top resid features:
rec
Tokenrec
Feature activation+0.016
Top resid features:
Ever
Token Ever
Feature activation-0.002
Top resid features:
Hit
Token Hit
Feature activation-0.001
Top resid features:
America
Token America
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.013
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
Though
TokenThough
Feature activation+4.531
Top resid features:
the
Token the
Feature activation+0.071
Top resid features:
tropical
Token tropical
Feature activation+0.044
Top resid features:
storm
Token storm
Feature activation-0.004
Top resid features:
warning
Token warning
Feature activation-0.005
Top resid features:
has
Token has
Feature activation+0.067
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
disparity
Token disparity
Feature activation+0.001
Top resid features:
?
Token?
Feature activation+0.033
Top resid features:
Remember
Token Remember
Feature activation+0.057
Top resid features:
,
Token,
Feature activation+0.039
Top resid features:
though
Token though
Feature activation+4.356
Top resid features:
13
Token 13
Feature activation+0.033
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
5
Token5
Feature activation+0.001
Top resid features:
percent
Token percent
Feature activation-0.007
Top resid features:
is
Token is
Feature activation+0.078
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.334
Top resid features:
iPhones
Token iPhones
Feature activation-0.004
Top resid features:
.
Token.
Feature activation-0.011
Top resid features:
And
Token And
Feature activation-0.011
Top resid features:
although
Token although
Feature activation+4.442
Top resid features:
Hunts
Token Hunts
Feature activation+0.018
Top resid features:
ville
Tokenville
Feature activation+0.004
Top resid features:
is
Token is
Feature activation+0.049
Top resid features:
the
Token the
Feature activation+0.027
Top resid features:
first
Token first
Feature activation+0.101
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
be
Token be
Feature activation+0.001
Top resid features:
patched
Token patched
Feature activation-0.003
Top resid features:
.
Token.
Feature activation-0.019
Top resid features:
Even
Token Even
Feature activation-0.001
Top resid features:
though
Token though
Feature activation+4.554
Top resid features:
the
Token the
Feature activation+0.030
Top resid features:
source
Token source
Feature activation+0.006
Top resid features:
code
Token code
Feature activation+0.003
Top resid features:
to
Token to
Feature activation+0.026
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
said
Token said
Feature activation+0.018
Top resid features:
she
Token she
Feature activation+0.017
Top resid features:
had
Token had
Feature activation+0.010
Top resid features:
not
Token not
Feature activation+0.003
Top resid features:
.
Token.
Feature activation-0.037
Top resid features:
Though
Token Though
Feature activation+4.119
Top resid features:
it
Token it
Feature activation+0.029
Top resid features:
was
Token was
Feature activation+0.139
Top resid features:
barely
Token barely
Feature activation+0.077
Top resid features:
percept
Token percept
Feature activation+0.073
Top resid features:
ible
Tokenible
Feature activation+0.154
Top resid features:
States
Token States
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.020
Top resid features:
These
Token These
Feature activation-0.014
Top resid features:
women
Token women
Feature activation-0.016
Top resid features:
,
Token,
Feature activation+0.066
Top resid features:
though
Token though
Feature activation+4.292
Top resid features:
they
Token they
Feature activation+0.090
Top resid features:
struggled
Token struggled
Feature activation+0.045
Top resid features:
against
Token against
Feature activation-0.060
Top resid features:
it
Token it
Feature activation-0.015
Top resid features:
,
Token,
Feature activation+0.429
Top resid features:
more
Token more
Feature activation+0.003
Top resid features:
on
Token on
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
infield
Token infield
Feature activation-0.000
Top resid features:
.
Token.
Feature activation-0.016
Top resid features:
Though
Token Though
Feature activation+4.170
Top resid features:
Turner
Token Turner
Feature activation+0.032
Top resid features:
has
Token has
Feature activation+0.069
Top resid features:
only
Token only
Feature activation+0.003
Top resid features:
qualified
Token qualified
Feature activation+0.017
Top resid features:
for
Token for
Feature activation+0.004
Top resid features:
day
Token day
Feature activation+0.002
Top resid features:
sour
Token sour
Feature activation+0.007
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.006
Top resid features:
Though
TokenThough
Feature activation+4.383
Top resid features:
the
Token the
Feature activation+0.065
Top resid features:
Conservatives
Token Conservatives
Feature activation-0.009
Top resid features:
are
Token are
Feature activation+0.040
Top resid features:
still
Token still
Feature activation+0.041
Top resid features:
polling
Token polling
Feature activation-0.043
Top resid features:
GU
Token GU
Feature activation+0.001
Top resid features:
Is
TokenIs
Feature activation+0.001
Top resid features:
in
Token in
Feature activation-0.002
Top resid features:
factor
Token factor
Feature activation+0.003
Top resid features:
.
Token.
Feature activation-0.003
Top resid features:
Though
Token Though
Feature activation+4.090
Top resid features:
the
Token the
Feature activation+0.078
Top resid features:
documentation
Token documentation
Feature activation+0.069
Top resid features:
is
Token is
Feature activation+0.105
Top resid features:
comprehensive
Token comprehensive
Feature activation+0.126
Top resid features:
,
Token,
Feature activation+0.116
Top resid features:
residents
Token residents
Feature activation+0.000
Top resid features:
say
Token say
Feature activation+0.006
Top resid features:
.
Token.
Feature activation-0.021
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
Though
TokenThough
Feature activation+4.082
Top resid features:
Ukraine
Token Ukraine
Feature activation+0.043
Top resid features:
has
Token has
Feature activation+0.062
Top resid features:
not
Token not
Feature activation+0.049
Top resid features:
publicly
Token publicly
Feature activation+0.058
Top resid features:
discussed
Token discussed
Feature activation+0.102
Top resid features:
work
Token work
Feature activation-0.002
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
similar
Token similar
Feature activation-0.001
Top resid features:
ways
Token ways
Feature activation-0.003
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
Though
Token Though
Feature activation+4.163
Top resid features:
some
Token some
Feature activation+0.017
Top resid features:
attempts
Token attempts
Feature activation+0.030
Top resid features:
have
Token have
Feature activation+0.007
Top resid features:
been
Token been
Feature activation+0.005
Top resid features:
made
Token made
Feature activation+0.001
Top resid features:
easier
Token easier
Feature activation-0.001
Top resid features:
with
Token with
Feature activation-0.003
Top resid features:
time
Token time
Feature activation-0.007
Top resid features:
.
Token.
Feature activation-0.004
Top resid features:
Even
Token Even
Feature activation+0.068
Top resid features:
though
Token though
Feature activation+3.739
Top resid features:
you
Token you
Feature activation+0.091
Top resid features:
will
Token will
Feature activation+0.112
Top resid features:
constantly
Token constantly
Feature activation+0.097
Top resid features:
be
Token be
Feature activation+0.089
Top resid features:
learning
Token learning
Feature activation+0.265
Top resid features:
,
Token,
Feature activation+0.016
Top resid features:
including
Token including
Feature activation-0.005
Top resid features:
Copenhagen
Token Copenhagen
Feature activation+0.004
Top resid features:
.
Token.
Feature activation-0.038
Top resid features:
Even
Token Even
Feature activation+0.031
Top resid features:
though
Token though
Feature activation+4.292
Top resid features:
Dan
Token Dan
Feature activation+0.071
Top resid features:
es
Tokenes
Feature activation+0.072
Top resid features:
create
Token create
Feature activation+0.032
Top resid features:
more
Token more
Feature activation+0.001
Top resid features:
household
Token household
Feature activation-0.006
Top resid features:
Jose
Token Jose
Feature activation+0.001
Top resid features:
f
Tokenf
Feature activation+0.002
Top resid features:
K
Token K
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.014
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
Though
TokenThough
Feature activation+4.007
Top resid features:
"
Token "
Feature activation+0.070
Top resid features:
Jose
TokenJose
Feature activation+0.002
Top resid features:
f
Tokenf
Feature activation+0.005
Top resid features:
K
Token K
Feature activation+0.001
Top resid features:
"
Token"
Feature activation-0.009
Top resid features:
different
Token different
Feature activation-0.001
Top resid features:
phases
Token phases
Feature activation+0.001
Top resid features:
of
Token of
Feature activation-0.001
Top resid features:
life
Token life
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
Although
Token Although
Feature activation+3.979
Top resid features:
differing
Token differing
Feature activation+0.150
Top resid features:
views
Token views
Feature activation+0.022
Top resid features:
on
Token on
Feature activation+0.020
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
means
Token means
Feature activation+0.027
Top resid features:
,
Token,
Feature activation+0.084
Top resid features:
some
Token some
Feature activation+0.015
Top resid features:
fishermen
Token fishermen
Feature activation+0.009
Top resid features:
said
Token said
Feature activation+0.031
Top resid features:
that
Token that
Feature activation+0.053
Top resid features:
although
Token although
Feature activation+3.522
Top resid features:
the
Token the
Feature activation+0.077
Top resid features:
shark
Token shark
Feature activation-0.006
Top resid features:
-
Token-
Feature activation-0.005
Top resid features:
dr
Tokendr
Feature activation+0.005
Top resid features:
agging
Tokenagging
Feature activation-0.006
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
be
Token be
Feature activation+0.002
Top resid features:
patched
Token patched
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.010
Top resid features:
Even
Token Even
Feature activation-0.002
Top resid features:
though
Token though
Feature activation+4.096
Top resid features:
the
Token the
Feature activation+0.032
Top resid features:
source
Token source
Feature activation+0.005
Top resid features:
code
Token code
Feature activation+0.001
Top resid features:
to
Token to
Feature activation+0.024
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
Major
Token Major
Feature activation-0.003
Top resid features:
League
Token League
Feature activation-0.001
Top resid features:
Baseball
Token Baseball
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.008
Top resid features:
Though
Token Though
Feature activation+3.994
Top resid features:
the
Token the
Feature activation+0.042
Top resid features:
Jays
Token Jays
Feature activation-0.000
Top resid features:
have
Token have
Feature activation+0.014
Top resid features:
made
Token made
Feature activation+0.007
Top resid features:
blockbuster
Token blockbuster
Feature activation-0.004
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.04

Head 2: 0.03

Head 3: 0.02

Head 4: 0.03

Head 5: 0.09

Head 6: 0.04

Head 7: 0.03

Head 8: 0.04

Head 9: 0.53

Head 10: 0.05

Head 11: 0.07

Positive logits

otherwise1.79

nevertheless1.61

wcs1.60

underest1.56

initially1.51

href1.44

disapp1.43

abase1.42

discrepancies1.41

rament1.40

tons1.37

antid1.37

neglect1.37

technically1.36

hitherto1.34

few1.34

slightest1.34

inaccur1.33

charges1.32

intention1.32

Negative logits

-1.48

onica-1.35

ーン-1.34

Ports-1.34

Latvia-1.33

Flint-1.28

Chips-1.26

Oak-1.22

agan-1.20

-1.18

Sev-1.18

Rats-1.17

opped-1.17

-1.16

Raven-1.16

-1.16

Asia-1.15

-1.15

Raz-1.15

л-1.15

INTERVAL 3.249 - 3.610
CONTAINS 0.001%

to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
payments
Token payments
Feature activation+0.000
within
Token within
Feature activation+0.000
stores
Token stores
Feature activation+0.000
,
Token,
Feature activation+3.610
they
Token they
Feature activation+2.487
can
Token can
Feature activation+0.000
make
Token make
Feature activation+0.000
Apple
Token Apple
Feature activation+0.000
Pay
Token Pay
Feature activation+0.000
than
Token than
Feature activation+0.000
10
Token 10
Feature activation+0.000
.
Token.
Feature activation+0.000
5
Token5
Feature activation+0.000
percent
Token percent
Feature activation+0.000
,
Token,
Feature activation+3.489
that
Token that
Feature activation+0.000
amounts
Token amounts
Feature activation+0.000
to
Token to
Feature activation+0.000
almost
Token almost
Feature activation+0.000
30
Token 30
Feature activation+0.000
though
Token though
Feature activation+0.000
they
Token they
Feature activation+0.000
struggled
Token struggled
Feature activation+0.000
against
Token against
Feature activation+0.000
it
Token it
Feature activation+0.000
,
Token,
Feature activation+3.270
were
Token were
Feature activation+0.029
immersed
Token immersed
Feature activation+0.000
in
Token in
Feature activation+0.000
an
Token an
Feature activation+0.000
ideology
Token ideology
Feature activation+0.000
bs
Tokenbs
Feature activation+0.000
has
Token has
Feature activation+0.052
been
Token been
Feature activation+0.000
made
Token made
Feature activation+0.000
public
Token public
Feature activation+0.000
,
Token,
Feature activation+3.378
we
Token we
Feature activation+2.981
can
Token can
Feature activation+0.078
't
Token't
Feature activation+0.000
update
Token update
Feature activation+0.000
the
Token the
Feature activation+0.000
it
Token it
Feature activation+0.634
was
Token was
Feature activation+0.672
barely
Token barely
Feature activation+0.000
percept
Token percept
Feature activation+0.000
ible
Tokenible
Feature activation+0.000
,
Token,
Feature activation+3.332
I
Token I
Feature activation+2.409
saw
Token saw
Feature activation+0.000
something
Token something
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 2.888 - 3.249
CONTAINS 0.002%

currently
Token currently
Feature activation+0.000
in
Token in
Feature activation+0.000
12
Token 12
Feature activation+0.000
th
Tokenth
Feature activation+0.000
spot
Token spot
Feature activation+0.000
,
Token,
Feature activation+2.919
tomorrow
Token tomorrow
Feature activation+0.000
night
Token night
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
you
Token you
Feature activation+0.000
will
Token will
Feature activation+0.000
constantly
Token constantly
Feature activation+0.000
be
Token be
Feature activation+0.000
learning
Token learning
Feature activation+0.000
,
Token,
Feature activation+3.109
your
Token your
Feature activation+0.330
startup
Token startup
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
mus
Tokenmus
Feature activation+0.000
some
Token some
Feature activation+0.000
extent
Token extent
Feature activation+0.000
ends
Token ends
Feature activation+0.000
)
Token)
Feature activation+0.000
exist
Token exist
Feature activation+1.177
,
Token,
Feature activation+3.049
no
Token no
Feature activation+0.964
party
Token party
Feature activation+0.000
has
Token has
Feature activation+0.000
questioned
Token questioned
Feature activation+0.000
the
Token the
Feature activation+0.000
below
Token below
Feature activation+0.000
their
Token their
Feature activation+0.000
standing
Token standing
Feature activation+0.000
last
Token last
Feature activation+0.000
fall
Token fall
Feature activation+0.000
,
Token,
Feature activation+3.180
they
Token they
Feature activation+2.539
have
Token have
Feature activation+0.000
nevertheless
Token nevertheless
Feature activation+0.000
managed
Token managed
Feature activation+0.000
to
Token to
Feature activation+0.000
manifested
Token manifested
Feature activation+0.000
love
Token love
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
classics
Token classics
Feature activation+0.000
,
Token,
Feature activation+2.906
J
Token J
Feature activation+0.000
ett
Tokenett
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Black
Token Black
Feature activation+0.000

INTERVAL 2.527 - 2.888
CONTAINS 0.005%

policing
Token policing
Feature activation+0.000
is
Token is
Feature activation+0.260
making
Token making
Feature activation+0.000
an
Token an
Feature activation+0.000
impact
Token impact
Feature activation+0.000
,
Token,
Feature activation+2.530
a
Token a
Feature activation+0.171
decade
Token decade
Feature activation+0.000
of
Token of
Feature activation+0.000
I
Token I
Feature activation+0.000
.
Token.
Feature activation+0.000
for
Token for
Feature activation+0.000
"
Token "
Feature activation+0.000
violent
Tokenviolent
Feature activation+0.000
"
Token"
Feature activation+0.000
exercise
Token exercise
Feature activation+0.000
,
Token,
Feature activation+2.709
Hob
Token Hob
Feature activation+0.000
house
Tokenhouse
Feature activation+0.000
suggests
Token suggests
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
lifelong
Token lifelong
Feature activation+0.000
off
Token off
Feature activation+0.000
-
Token-
Feature activation+0.000
road
Tokenroad
Feature activation+0.000
cyclist
Token cyclist
Feature activation+0.000
,
Token,
Feature activation+2.799
n
Token n
Feature activation+0.000
ord
Tokenord
Feature activation+0.000
ic
Tokenic
Feature activation+0.000
sk
Token sk
Feature activation+0.000
ier
Tokenier
Feature activation+0.000
woman
Token woman
Feature activation+0.000
who
Token who
Feature activation+0.000
received
Token received
Feature activation+0.000
her
Token her
Feature activation+0.000
uterus
Token uterus
Feature activation+0.000
,
Token,
Feature activation+2.885
Sil
Token Sil
Feature activation+0.000
er
Tokener
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
recipient
Token recipient
Feature activation+0.000
La
Token La
Feature activation+0.000
L
TokenL
Feature activation+0.000
iga
Tokeniga
Feature activation+0.000
this
Token this
Feature activation+0.234
season
Token season
Feature activation+0.000
,
Token,
Feature activation+2.789
Ney
Token Ney
Feature activation+0.000
mar
Tokenmar
Feature activation+0.000
believes
Token believes
Feature activation+0.000
Barcelona
Token Barcelona
Feature activation+0.000
can
Token can
Feature activation+0.000

INTERVAL 2.166 - 2.527
CONTAINS 0.005%

city
Token city
Feature activation+0.000
was
Token was
Feature activation+0.073
ever
Token ever
Feature activation+0.000
completed
Token completed
Feature activation+0.000
,
Token,
Feature activation+2.381
it
Token it
Feature activation+2.248
had
Token had
Feature activation+0.000
running
Token running
Feature activation+0.000
trains
Token trains
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
much
Token much
Feature activation+0.000
worse
Token worse
Feature activation+0.000
than
Token than
Feature activation+0.000
predicted
Token predicted
Feature activation+0.000
,
Token,
Feature activation+2.195
they
Token they
Feature activation+2.221
were
Token were
Feature activation+0.372
still
Token still
Feature activation+0.000
predicted
Token predicted
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
of
Token of
Feature activation+0.000
Kenn
Token Kenn
Feature activation+0.000
es
Tokenes
Feature activation+0.000
aw
Tokenaw
Feature activation+0.000
Mountain
Token Mountain
Feature activation+0.000
,
Token,
Feature activation+2.186
he
Token he
Feature activation+2.415
was
Token was
Feature activation+0.031
ultimately
Token ultimately
Feature activation+0.000
out
Token out
Feature activation+0.000
fl
Tokenfl
Feature activation+0.000
orphans
Token orphans
Feature activation+0.000
.
Token.
Feature activation+0.000
Although
Token Although
Feature activation+0.000
he
Token he
Feature activation+0.205
lost
Token lost
Feature activation+0.000
,
Token,
Feature activation+2.220
the
Token the
Feature activation+0.990
person
Token person
Feature activation+0.000
who
Token who
Feature activation+0.000
did
Token did
Feature activation+0.000
win
Token win
Feature activation+0.000
the
Token the
Feature activation+0.000
real
Token real
Feature activation+0.000
-
Token-
Feature activation+0.000
world
Tokenworld
Feature activation+0.000
scenario
Token scenario
Feature activation+0.000
,
Token,
Feature activation+2.333
the
Token the
Feature activation+0.794
amount
Token amount
Feature activation+0.000
of
Token of
Feature activation+0.000
land
Token land
Feature activation+0.000
required
Token required
Feature activation+0.000

INTERVAL 1.805 - 2.166
CONTAINS 0.004%

even
Token even
Feature activation+0.000
though
Token though
Feature activation+0.000
hope
Token hope
Feature activation+0.000
was
Token was
Feature activation+0.000
lost
Token lost
Feature activation+0.000
,
Token,
Feature activation+1.834
it
Token it
Feature activation+2.057
was
Token was
Feature activation+0.000
necessary
Token necessary
Feature activation+0.000
to
Token to
Feature activation+0.000
move
Token move
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
Feature activation+0.000
current
Token current
Feature activation+0.000
job
Token job
Feature activation+0.000
,
Token,
Feature activation+2.441
he
Token he
Feature activation+1.918
was
Token was
Feature activation+0.251
comfortable
Token comfortable
Feature activation+0.000
and
Token and
Feature activation+0.000
saw
Token saw
Feature activation+0.000
the
Token the
Feature activation+0.000
fighting
Token fighting
Feature activation+0.000
so
Token so
Feature activation+0.000
hard
Token hard
Feature activation+0.000
for
Token for
Feature activation+0.000
equality
Token equality
Feature activation+0.000
,
Token,
Feature activation+2.119
a
Token a
Feature activation+0.000
lot
Token lot
Feature activation+0.000
of
Token of
Feature activation+0.000
others
Token others
Feature activation+0.000
don
Token don
Feature activation+0.000
in
Token in
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
next
Token next
Feature activation+0.000
patch
Token patch
Feature activation+0.000
,
Token,
Feature activation+2.103
our
Token our
Feature activation+0.090
upcoming
Token upcoming
Feature activation+0.000
community
Token community
Feature activation+0.000
competition
Token competition
Feature activation+0.000
will
Token will
Feature activation+0.000
the
Token the
Feature activation+0.000
reported
Token reported
Feature activation+0.000
effect
Token effect
Feature activation+0.000
is
Token is
Feature activation+0.061
small
Token small
Feature activation+0.634
,
Token,
Feature activation+2.113
our
Token our
Feature activation+0.903
results
Token results
Feature activation+0.000
suggest
Token suggest
Feature activation+0.000
that
Token that
Feature activation+0.000
under
Token under
Feature activation+0.000

INTERVAL 1.444 - 1.805
CONTAINS 0.003%

trained
Token trained
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
physicist
Token physicist
Feature activation+0.000
,
Token,
Feature activation+1.902
he
Token he
Feature activation+1.803
had
Token had
Feature activation+0.000
switched
Token switched
Feature activation+0.000
careers
Token careers
Feature activation+0.000
to
Token to
Feature activation+0.000
cancer
Token cancer
Feature activation+0.000
based
Token based
Feature activation+0.000
on
Token on
Feature activation+0.000
small
Token small
Feature activation+0.000
sample
Token sample
Feature activation+0.000
sizes
Token sizes
Feature activation+0.000
,
Token,
Feature activation+1.608
Sc
Token Sc
Feature activation+0.000
ri
Tokenri
Feature activation+0.000
ven
Tokenven
Feature activation+0.000
s
Tokens
Feature activation+0.000
appeared
Token appeared
Feature activation+0.000
tt
Tokentt
Feature activation+0.000
ier
Tokenier
Feature activation+0.000
Hall
Token Hall
Feature activation+0.000
.
Token.
Feature activation+0.000
Although
Token Although
Feature activation+0.000
there
Token there
Feature activation+1.490
are
Token are
Feature activation+0.970
304
Token 304
Feature activation+0.000
empty
Token empty
Feature activation+0.000
beds
Token beds
Feature activation+0.000
on
Token on
Feature activation+0.000
range
Token range
Feature activation+0.000
of
Token of
Feature activation+0.000
South
Token South
Feature activation+0.000
Korean
Token Korean
Feature activation+0.000
islands
Token islands
Feature activation+0.000
,
Token,
Feature activation+1.725
the
Token the
Feature activation+0.822
addition
Token addition
Feature activation+0.000
of
Token of
Feature activation+0.000
artillery
Token artillery
Feature activation+0.000
positions
Token positions
Feature activation+0.000
Although
Token Although
Feature activation+0.000
there
Token there
Feature activation+0.912
are
Token are
Feature activation+0.908
32
Token 32
Feature activation+0.000
district
Token district
Feature activation+0.000
,
Token,
Feature activation+1.592
46
Token 46
Feature activation+0.000
sub
Token sub
Feature activation+0.000
-
Token-
Feature activation+0.000
division
Tokendivision
Feature activation+0.000
al
Tokenal
Feature activation+0.000

INTERVAL 1.083 - 1.444
CONTAINS 0.003%

8
Token 8
Feature activation+0.000
-
Token-
Feature activation+0.000
track
Tokentrack
Feature activation+0.000
!)
Token!)
Feature activation+0.826
are
Token are
Feature activation+2.474
the
Token the
Feature activation+1.269
very
Token very
Feature activation+0.000
old
Token old
Feature activation+0.000
-
Token-
Feature activation+0.000
school
Tokenschool
Feature activation+0.000
music
Token music
Feature activation+0.000
,
Token,
Feature activation+0.000
reward
Token reward
Feature activation+0.000
or
Token or
Feature activation+0.000
no
Token no
Feature activation+0.000
reward
Token reward
Feature activation+0.000
,
Token,
Feature activation+1.183
we
Token we
Feature activation+0.426
would
Token would
Feature activation+0.000
've
Token've
Feature activation+0.000
stopped
Token stopped
Feature activation+0.000
him
Token him
Feature activation+0.000
does
Token does
Feature activation+0.000
exist
Token exist
Feature activation+0.000
;
Token;
Feature activation+0.000
however
Token however
Feature activation+0.076
,
Token,
Feature activation+0.178
it
Token it
Feature activation+1.142
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
highly
Token highly
Feature activation+0.000
unusual
Token unusual
Feature activation+0.000
March
Token March
Feature activation+0.000
,
Token,
Feature activation+0.000
17
Token 17
Feature activation+0.000
08
Token08
Feature activation+0.000
,
Token,
Feature activation+0.000
she
Token she
Feature activation+1.125
blocked
Token blocked
Feature activation+0.000
the
Token the
Feature activation+0.000
Scottish
Token Scottish
Feature activation+0.000
Militia
Token Militia
Feature activation+0.000
Bill
Token Bill
Feature activation+0.000
on
Token on
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Though
Token Though
Feature activation+0.000
,
Token,
Feature activation+0.000
there
Token there
Feature activation+1.211
is
Token is
Feature activation+0.554
one
Token one
Feature activation+0.000
scene
Token scene
Feature activation+0.000
in
Token in
Feature activation+0.000
that
Token that
Feature activation+0.000

INTERVAL 0.722 - 1.083
CONTAINS 0.008%

's
Token's
Feature activation+0.000
ep
Token ep
Feature activation+0.000
ics
Tokenics
Feature activation+0.000
.
Token.
Feature activation+0.000
Although
Token Although
Feature activation+0.000
his
Token his
Feature activation+0.756
father
Token father
Feature activation+0.000
's
Token's
Feature activation+0.000
approach
Token approach
Feature activation+0.000
to
Token to
Feature activation+0.000
education
Token education
Feature activation+0.000
uniform
Token uniform
Feature activation+0.000
ity
Tokenity
Feature activation+0.000
in
Token in
Feature activation+0.000
our
Token our
Feature activation+0.000
military
Token military
Feature activation+0.000
,
Token,
Feature activation+1.002
it
Token it
Feature activation+2.585
is
Token is
Feature activation+0.000
seen
Token seen
Feature activation+0.000
as
Token as
Feature activation+0.000
discriminatory
Token discriminatory
Feature activation+0.000
emer
Tokenemer
Feature activation+0.000
ged
Tokenged
Feature activation+0.000
on
Token on
Feature activation+0.000
social
Token social
Feature activation+0.000
media
Token media
Feature activation+0.000
,
Token,
Feature activation+0.905
and
Token and
Feature activation+0.000
while
Token while
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
yet
Token yet
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Though
TokenThough
Feature activation+0.000
he
Token he
Feature activation+0.340
doesn
Token doesn
Feature activation+0.888
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.046
condone
Token condone
Feature activation+0.000
drug
Token drug
Feature activation+0.000
tapes
Token tapes
Feature activation+0.000
!
Token!
Feature activation+0.000
8
Token 8
Feature activation+0.000
-
Token-
Feature activation+0.000
track
Tokentrack
Feature activation+0.000
!)
Token!)
Feature activation+0.826
are
Token are
Feature activation+2.474
the
Token the
Feature activation+1.269
very
Token very
Feature activation+0.000
old
Token old
Feature activation+0.000
-
Token-
Feature activation+0.000

INTERVAL 0.361 - 0.722
CONTAINS 0.013%

the
Token the
Feature activation+0.000
water
Token water
Feature activation+0.000
resources
Token resources
Feature activation+0.000
department
Token department
Feature activation+0.000
,
Token,
Feature activation+1.613
these
Token these
Feature activation+0.399
facilities
Token facilities
Feature activation+0.000
wo
Token wo
Feature activation+0.000
efully
Tokenefully
Feature activation+0.000
lack
Token lack
Feature activation+0.000
requisite
Token requisite
Feature activation+0.000
name
Token name
Feature activation+0.000
Charlotte
Token Charlotte
Feature activation+0.000
Cord
Token Cord
Feature activation+0.000
ay
Tokenay
Feature activation+0.000
can
Token can
Feature activation+0.780
be
Token be
Feature activation+0.401
seen
Token seen
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
paper
Token paper
Feature activation+0.000
held
Token held
Feature activation+0.000
35
Token 35
Feature activation+0.000
years
Token years
Feature activation+0.000
or
Token or
Feature activation+0.000
however
Token however
Feature activation+0.000
long
Token long
Feature activation+0.000
it
Token it
Feature activation+0.395
was
Token was
Feature activation+0.000
.
Token.
Feature activation+0.000
There
Token There
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Honestly
TokenHonestly
Feature activation+0.000
,
Token,
Feature activation+0.000
although
Token although
Feature activation+0.000
it
Token it
Feature activation+0.441
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.284
primarily
Token primarily
Feature activation+0.152
a
Token a
Feature activation+0.000
as
Token as
Feature activation+0.000
advance
Token advance
Feature activation+0.000
forces
Token forces
Feature activation+0.000
,
Token,
Feature activation+0.000
though
Token though
Feature activation+0.000
it
Token it
Feature activation+0.542
does
Token does
Feature activation+0.578
have
Token have
Feature activation+0.000
sk
Token sk
Feature activation+0.000
or
Tokenor
Feature activation+0.000
nergy
Tokennergy
Feature activation+0.000

INTERVAL 0.000 - 0.361
CONTAINS 99.956%

Ser
Token Ser
Feature activation+0.000
rel
Tokenrel
Feature activation+0.000
s
Tokens
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
Feature activation+0.000
natural
Token natural
Feature activation+0.000
habitat
Token habitat
Feature activation+0.000
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
is
Token is
Feature activation+0.000
how
Token how
Feature activation+0.000
lost
Token lost
Feature activation+0.000
his
Token his
Feature activation+0.000
power
Token power
Feature activation+0.000
but
Token but
Feature activation+0.000
still
Token still
Feature activation+0.000
holds
Token holds
Feature activation+0.000
control
Token control
Feature activation+0.000
over
Token over
Feature activation+0.000
Marvel
Token Marvel
Feature activation+0.000
Entertainment
Token Entertainment
Feature activation+0.000
.
Token.
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
border
Token border
Feature activation+0.000
takes
Token takes
Feature activation+0.000
days
Token days
Feature activation+0.000
for
Token for
Feature activation+0.000
cargo
Token cargo
Feature activation+0.000
and
Token and
Feature activation+0.000
significantly
Token significantly
Feature activation+0.000
cuts
Token cuts
Feature activation+0.000
railway
Token railway
Feature activation+0.000
we
Token we
Feature activation+0.000
have
Token have
Feature activation+0.000
assembled
Token assembled
Feature activation+0.000
the
Token the
Feature activation+0.000
most
Token most
Feature activation+0.000
comprehensive
Token comprehensive
Feature activation+0.000
,
Token,
Feature activation+0.000
publicly
Token publicly
Feature activation+0.000
-
Token-
Feature activation+0.000
available
Tokenavailable
Feature activation+0.000
account
Token account
Feature activation+0.000
to
Token to
Feature activation+0.000
drug
Token drug
Feature activation+0.000
,
Token,
Feature activation+0.000
kill
Token kill
Feature activation+0.000
and
Token and
Feature activation+0.000
dism
Token dism
Feature activation+0.000
ember
Tokenember
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 5 in H3.9: (feature 24238)

TOP ACTIVATIONS
MAX = 3.774

digging
Token digging
Feature activation+0.473
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
stuff
Token stuff
Feature activation+0.000
,
Token,
Feature activation+3.774
Nissan
Token Nissan
Feature activation+0.000
is
Token is
Feature activation+0.000
teasing
Token teasing
Feature activation+0.000
us
Token us
Feature activation+0.000
all
Token all
Feature activation+0.000
Play
Token Play
Feature activation+0.000
dom
Tokendom
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.571
matter
Token matter
Feature activation+0.000
,
Token,
Feature activation+3.390
players
Token players
Feature activation+0.109
are
Token are
Feature activation+0.000
now
Token now
Feature activation+0.000
vent
Token vent
Feature activation+0.000
ing
Tokening
Feature activation+0.000
the
Token the
Feature activation+0.000
hottest
Token hottest
Feature activation+0.000
year
Token year
Feature activation+0.000
on
Token on
Feature activation+0.000
record
Token record
Feature activation+0.000
,
Token,
Feature activation+3.374
India
Token India
Feature activation+0.000
and
Token and
Feature activation+0.000
China
Token China
Feature activation+0.000
will
Token will
Feature activation+0.101
negotiate
Token negotiate
Feature activation+0.000
not
Token not
Feature activation+0.000
to
Token to
Feature activation+0.000
consider
Token consider
Feature activation+0.000
the
Token the
Feature activation+0.000
bill
Token bill
Feature activation+0.000
,
Token,
Feature activation+3.370
it
Token it
Feature activation+1.664
is
Token is
Feature activation+0.000
effectively
Token effectively
Feature activation+0.000
dead
Token dead
Feature activation+0.000
for
Token for
Feature activation+0.000
threats
Token threats
Feature activation+0.000
of
Token of
Feature activation+0.000
equality
Token equality
Feature activation+0.000
to
Token to
Feature activation+0.000
slay
Token slay
Feature activation+0.000
,
Token,
Feature activation+3.314
Sch
Token Sch
Feature activation+0.000
alf
Tokenalf
Feature activation+0.000
ly
Tokenly
Feature activation+0.000
became
Token became
Feature activation+0.000
just
Token just
Feature activation+0.000
began
Token began
Feature activation+0.000
almost
Token almost
Feature activation+0.000
two
Token two
Feature activation+0.000
years
Token years
Feature activation+0.000
ago
Token ago
Feature activation+0.000
,
Token,
Feature activation+3.301
life
Token life
Feature activation+0.000
for
Token for
Feature activation+0.000
minorities
Token minorities
Feature activation+0.000
is
Token is
Feature activation+0.000
becoming
Token becoming
Feature activation+0.000
life
Token life
Feature activation+0.000
boats
Tokenboats
Feature activation+0.000
either
Token either
Feature activation+0.000
full
Token full
Feature activation+2.783
or
Token or
Feature activation+0.000
gone
Token gone
Feature activation+3.260
,
Token,
Feature activation+2.959
the
Token the
Feature activation+2.376
Dav
Token Dav
Feature activation+0.000
ises
Tokenises
Feature activation+0.000
were
Token were
Feature activation+0.000
-
Token-
Feature activation+0.000
1
Token1
Feature activation+0.000
NCAA
Token NCAA
Feature activation+0.000
basketball
Token basketball
Feature activation+0.000
tournament
Token tournament
Feature activation+0.000
,
Token,
Feature activation+3.250
I
Token I
Feature activation+0.560
started
Token started
Feature activation+0.000
thinking
Token thinking
Feature activation+0.000
about
Token about
Feature activation+0.000
all
Token all
Feature activation+0.000
and
Token and
Feature activation+1.318
little
Token little
Feature activation+2.264
cash
Token cash
Feature activation+0.000
on
Token on
Feature activation+0.000
hand
Token hand
Feature activation+0.713
,
Token,
Feature activation+3.250
PRE
Token PRE
Feature activation+0.000
PA
TokenPA
Feature activation+0.000
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
thrust
Token thrust
Feature activation+0.000
12
Token 12
Feature activation+0.000
teams
Token teams
Feature activation+0.000
in
Token in
Feature activation+1.919
three
Token three
Feature activation+1.371
years
Token years
Feature activation+0.088
,
Token,
Feature activation+3.243
we
Token we
Feature activation+0.263
now
Token now
Feature activation+0.000
have
Token have
Feature activation+0.000
enough
Token enough
Feature activation+0.000
data
Token data
Feature activation+0.000
almost
Token almost
Feature activation+0.000
3
Token 3
Feature activation+0.000
-
Token-
Feature activation+0.000
1
Token1
Feature activation+0.000
count
Token count
Feature activation+0.000
,
Token,
Feature activation+3.177
the
Token the
Feature activation+2.433
Giants
Token Giants
Feature activation+0.000
might
Token might
Feature activation+0.078
well
Token well
Feature activation+0.000
have
Token have
Feature activation+0.000
.
Token.
Feature activation+0.000
Senate
Token Senate
Feature activation+0.000
recount
Token recount
Feature activation+0.000
still
Token still
Feature activation+2.088
incomplete
Token incomplete
Feature activation+0.397
,
Token,
Feature activation+3.161
attorneys
Token attorneys
Feature activation+0.000
on
Token on
Feature activation+0.000
both
Token both
Feature activation+0.000
sides
Token sides
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+1.994
solar
Token solar
Feature activation+0.000
and
Token and
Feature activation+0.417
wind
Token wind
Feature activation+0.000
sectors
Token sectors
Feature activation+0.000
,
Token,
Feature activation+3.123
growing
Token growing
Feature activation+2.518
domestic
Token domestic
Feature activation+0.000
markets
Token markets
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
background
Token background
Feature activation+0.465
in
Token in
Feature activation+0.000
design
Token design
Feature activation+0.000
or
Token or
Feature activation+0.000
manufacturing
Token manufacturing
Feature activation+0.000
,
Token,
Feature activation+3.116
Tam
Token Tam
Feature activation+0.000
ir
Tokenir
Feature activation+0.000
figured
Token figured
Feature activation+0.110
out
Token out
Feature activation+0.000
that
Token that
Feature activation+0.000
a
Token a
Feature activation+1.081
.
Token.
Feature activation+0.156
m
Tokenm
Feature activation+0.892
.
Token.
Feature activation+0.000
services
Token services
Feature activation+0.058
,
Token,
Feature activation+3.115
this
Token this
Feature activation+0.927
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
season
Token season
Feature activation+0.000
for
Token for
Feature activation+0.000
GOP
Token GOP
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
bill
Token bill
Feature activation+0.000
,
Token,
Feature activation+3.039
Boehner
Token Boehner
Feature activation+0.000
was
Token was
Feature activation+0.000
left
Token left
Feature activation+0.000
scrambling
Token scrambling
Feature activation+0.000
this
Token this
Feature activation+0.000
order
Token order
Feature activation+0.000
.
Token.
Feature activation+0.000
With
Token With
Feature activation+0.000
this
Token this
Feature activation+0.000
in
Token in
Feature activation+1.713
mind
Token mind
Feature activation+3.019
,
Token,
Feature activation+1.433
on
Token on
Feature activation+1.087
April
Token April
Feature activation+0.000
20
Token 20
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
move
Token move
Feature activation+0.000
away
Token away
Feature activation+0.000
from
Token from
Feature activation+0.000
tipping
Token tipping
Feature activation+0.000
,
Token,
Feature activation+3.004
it
Token it
Feature activation+1.257
's
Token's
Feature activation+0.000
not
Token not
Feature activation+0.000
at
Token at
Feature activation+0.000
all
Token all
Feature activation+0.000
weeks
Token weeks
Feature activation+0.606
until
Token until
Feature activation+1.418
the
Token the
Feature activation+0.888
GOP
Token GOP
Feature activation+0.000
runoff
Token runoff
Feature activation+0.000
,
Token,
Feature activation+3.000
Bannon
Token Bannon
Feature activation+0.000
has
Token has
Feature activation+0.000
declared
Token declared
Feature activation+0.000
his
Token his
Feature activation+0.000
support
Token support
Feature activation+0.000
season
Token season
Feature activation+0.000
back
Token back
Feature activation+1.883
in
Token in
Feature activation+1.910
full
Token full
Feature activation+2.628
swing
Token swing
Feature activation+0.000
,
Token,
Feature activation+2.997
it
Token it
Feature activation+1.211
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
matter
Token matter
Feature activation+0.000
of
Token of
Feature activation+0.000

Top DFA by src position
MAX = 3.782

to
Token to
Feature activation+0.000
Top resid features:
be
Token be
Feature activation-0.000
Top resid features:
true
Token true
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.059
Top resid features:
With
TokenWith
Feature activation+3.782
Top resid features:
large
Token large
Feature activation+0.019
Top resid features:
swat
Token swat
Feature activation+0.009
Top resid features:
hes
Tokenhes
Feature activation+0.015
Top resid features:
of
Token of
Feature activation+0.011
Top resid features:
the
Token the
Feature activation+0.009
Top resid features:
22
Token 22
Feature activation+0.001
Top resid features:
nd
Tokennd
Feature activation+0.001
Top resid features:
October
Token October
Feature activation+0.000
Top resid features:
2014
Token 2014
Feature activation-0.003
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
With
Token With
Feature activation+3.569
Top resid features:
no
Token no
Feature activation+0.092
Top resid features:
explanation
Token explanation
Feature activation-0.002
Top resid features:
from
Token from
Feature activation+0.025
Top resid features:
Disney
Token Disney
Feature activation+0.072
Top resid features:
or
Token or
Feature activation+0.074
Top resid features:
men
Token men
Feature activation-0.002
Top resid features:
and
Token and
Feature activation-0.000
Top resid features:
two
Token two
Feature activation+0.001
Top resid features:
women
Token women
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.058
Top resid features:
With
TokenWith
Feature activation+3.421
Top resid features:
2015
Token 2015
Feature activation+0.108
Top resid features:
heading
Token heading
Feature activation+0.057
Top resid features:
towards
Token towards
Feature activation+0.017
Top resid features:
being
Token being
Feature activation+0.054
Top resid features:
the
Token the
Feature activation+0.020
Top resid features:
slippery
Token slippery
Feature activation+0.004
Top resid features:
slope
Token slope
Feature activation-0.001
Top resid features:
."
Token."
Feature activation+0.016
Top resid features:
Ċ
TokenĊ
Feature activation-0.007
Top resid features:
Ċ
TokenĊ
Feature activation-0.016
Top resid features:
With
TokenWith
Feature activation+3.688
Top resid features:
the
Token the
Feature activation+0.048
Top resid features:
House
Token House
Feature activation+0.042
Top resid features:
choosing
Token choosing
Feature activation+0.051
Top resid features:
not
Token not
Feature activation+0.073
Top resid features:
to
Token to
Feature activation+0.006
Top resid features:
--
Token --
Feature activation+0.010
Top resid features:
was
Token was
Feature activation-0.014
Top resid features:
eventually
Token eventually
Feature activation+0.002
Top resid features:
defeated
Token defeated
Feature activation-0.004
Top resid features:
.
Token.
Feature activation+0.058
Top resid features:
With
Token With
Feature activation+3.286
Top resid features:
no
Token no
Feature activation+0.063
Top resid features:
more
Token more
Feature activation+0.018
Top resid features:
threats
Token threats
Feature activation+0.013
Top resid features:
of
Token of
Feature activation+0.023
Top resid features:
equality
Token equality
Feature activation+0.077
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
1915
Token 1915
Feature activation+0.002
Top resid features:
Ottoman
Token Ottoman
Feature activation-0.000
Top resid features:
genocide
Token genocide
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.041
Top resid features:
With
Token With
Feature activation+3.225
Top resid features:
a
Token a
Feature activation+0.038
Top resid features:
death
Token death
Feature activation-0.013
Top resid features:
toll
Token toll
Feature activation+0.008
Top resid features:
of
Token of
Feature activation+0.005
Top resid features:
about
Token about
Feature activation+0.014
Top resid features:
life
Token life
Feature activation-0.006
Top resid features:
boats
Tokenboats
Feature activation-0.002
Top resid features:
."
Token."
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
With
TokenWith
Feature activation+3.435
Top resid features:
life
Token life
Feature activation+0.030
Top resid features:
boats
Tokenboats
Feature activation+0.016
Top resid features:
either
Token either
Feature activation+0.023
Top resid features:
full
Token full
Feature activation+0.045
Top resid features:
or
Token or
Feature activation+0.279
Top resid features:
don
Token don
Feature activation-0.002
Top resid features:
't
Token't
Feature activation-0.004
Top resid features:
talk
Token talk
Feature activation-0.004
Top resid features:
about
Token about
Feature activation-0.007
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.017
Top resid features:
With
TokenWith
Feature activation+3.306
Top resid features:
the
Token the
Feature activation+0.014
Top resid features:
recent
Token recent
Feature activation+0.096
Top resid features:
completion
Token completion
Feature activation+0.070
Top resid features:
of
Token of
Feature activation+0.053
Top resid features:
the
Token the
Feature activation+0.027
Top resid features:
pay
Tokenpay
Feature activation-0.002
Top resid features:
able
Tokenable
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
âĢ
TokenâĢ
Feature activation-0.011
Top resid features:
Ŀ
TokenĿ
Feature activation-0.005
Top resid features:
With
Token With
Feature activation+3.629
Top resid features:
hundreds
Token hundreds
Feature activation+0.013
Top resid features:
of
Token of
Feature activation+0.008
Top resid features:
millions
Token millions
Feature activation+0.012
Top resid features:
of
Token of
Feature activation+0.005
Top resid features:
dollars
Token dollars
Feature activation+0.023
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.181
Top resid features:
.
Token.
Feature activation-0.030
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
Ċ
TokenĊ
Feature activation-0.016
Top resid features:
With
TokenWith
Feature activation+3.227
Top resid features:
12
Token 12
Feature activation+0.060
Top resid features:
teams
Token teams
Feature activation+0.017
Top resid features:
in
Token in
Feature activation+0.081
Top resid features:
three
Token three
Feature activation+0.089
Top resid features:
years
Token years
Feature activation+0.188
Top resid features:
300
Token300
Feature activation-0.002
Top resid features:
million
Token million
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
Ċ
TokenĊ
Feature activation-0.014
Top resid features:
With
TokenWith
Feature activation+3.554
Top resid features:
San
Token San
Feature activation+0.010
Top resid features:
Francisco
Token Francisco
Feature activation+0.005
Top resid features:
voters
Token voters
Feature activation-0.019
Top resid features:
passing
Token passing
Feature activation+0.030
Top resid features:
Proposition
Token Proposition
Feature activation-0.003
Top resid features:
ial
Tokenial
Feature activation-0.001
Top resid features:
iang
Tokeniang
Feature activation-0.000
Top resid features:
G
Token G
Feature activation-0.000
Top resid features:
ao
Tokenao
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.041
Top resid features:
With
TokenWith
Feature activation+3.245
Top resid features:
the
Token the
Feature activation+0.048
Top resid features:
U
Token U
Feature activation+0.004
Top resid features:
.
Token.
Feature activation+0.012
Top resid features:
S
TokenS
Feature activation+0.011
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation-0.004
Top resid features:
Ċ
TokenĊ
Feature activation-0.007
Top resid features:
âĢ
TokenâĢ
Feature activation-0.026
Top resid features:
ľ
Tokenľ
Feature activation-0.029
Top resid features:
With
TokenWith
Feature activation+3.450
Top resid features:
extensive
Token extensive
Feature activation+0.082
Top resid features:
manufacturing
Token manufacturing
Feature activation+0.023
Top resid features:
capacity
Token capacity
Feature activation+0.017
Top resid features:
in
Token in
Feature activation+0.041
Top resid features:
the
Token the
Feature activation+0.034
Top resid features:
work
Token work
Feature activation+0.002
Top resid features:
with
Token with
Feature activation+0.002
Top resid features:
?
Token?
Feature activation+0.019
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
With
TokenWith
Feature activation+3.154
Top resid features:
no
Token no
Feature activation+0.066
Top resid features:
background
Token background
Feature activation+0.028
Top resid features:
in
Token in
Feature activation+0.043
Top resid features:
design
Token design
Feature activation+0.031
Top resid features:
or
Token or
Feature activation+0.215
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
thigh
Token thigh
Feature activation+0.003
Top resid features:
?
Token?
Feature activation+0.009
Top resid features:
Ċ
TokenĊ
Feature activation+0.015
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
With
TokenWith
Feature activation+3.177
Top resid features:
temperatures
Token temperatures
Feature activation+0.071
Top resid features:
in
Token in
Feature activation+0.025
Top resid features:
the
Token the
Feature activation+0.023
Top resid features:
80
Token 80
Feature activation+0.022
Top resid features:
s
Tokens
Feature activation+0.018
Top resid features:
many
Token many
Feature activation-0.006
Top resid features:
Republicans
Token Republicans
Feature activation-0.000
Top resid features:
cra
Token cra
Feature activation-0.003
Top resid features:
ved
Tokenved
Feature activation-0.006
Top resid features:
.
Token.
Feature activation+0.025
Top resid features:
With
Token With
Feature activation+3.403
Top resid features:
House
Token House
Feature activation+0.014
Top resid features:
Democratic
Token Democratic
Feature activation+0.029
Top resid features:
leaders
Token leaders
Feature activation+0.020
Top resid features:
discouraging
Token discouraging
Feature activation+0.066
Top resid features:
their
Token their
Feature activation-0.005
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
first
Token first
Feature activation+0.016
Top resid features:
order
Token order
Feature activation+0.004
Top resid features:
.
Token.
Feature activation+0.064
Top resid features:
With
Token With
Feature activation+3.483
Top resid features:
this
Token this
Feature activation-0.122
Top resid features:
in
Token in
Feature activation+0.101
Top resid features:
mind
Token mind
Feature activation+0.052
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
on
Token on
Feature activation+0.000
Top resid features:
't
Token't
Feature activation-0.018
Top resid features:
even
Token even
Feature activation-0.022
Top resid features:
open
Token open
Feature activation-0.003
Top resid features:
yet
Token yet
Feature activation-0.009
Top resid features:
.
Token.
Feature activation+0.031
Top resid features:
With
Token With
Feature activation+3.218
Top resid features:
only
Token only
Feature activation+0.007
Top resid features:
a
Token a
Feature activation+0.039
Top resid features:
few
Token few
Feature activation+0.068
Top resid features:
other
Token other
Feature activation+0.028
Top resid features:
restaurants
Token restaurants
Feature activation+0.036
Top resid features:
attorney
Token attorney
Feature activation-0.000
Top resid features:
general
Token general
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.016
Top resid features:
Ċ
TokenĊ
Feature activation-0.026
Top resid features:
With
TokenWith
Feature activation+3.169
Top resid features:
just
Token just
Feature activation+0.054
Top resid features:
two
Token two
Feature activation+0.090
Top resid features:
weeks
Token weeks
Feature activation+0.155
Top resid features:
until
Token until
Feature activation-0.028
Top resid features:
the
Token the
Feature activation+0.057
Top resid features:
Cowboys
Token Cowboys
Feature activation+0.008
Top resid features:
merchandise
Token merchandise
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation-0.014
Top resid features:
Ċ
TokenĊ
Feature activation-0.024
Top resid features:
With
TokenWith
Feature activation+2.993
Top resid features:
the
Token the
Feature activation+0.023
Top resid features:
NRL
Token NRL
Feature activation+0.063
Top resid features:
season
Token season
Feature activation+0.083
Top resid features:
back
Token back
Feature activation+0.118
Top resid features:
in
Token in
Feature activation+0.045
Top resid features:

Decoder Weights Distribution

Head 0: 0.01

Head 1: 0.07

Head 2: 0.04

Head 3: 0.02

Head 4: 0.02

Head 5: 0.09

Head 6: 0.05

Head 7: 0.04

Head 8: 0.05

Head 9: 0.52

Head 10: 0.05

Head 11: 0.05

Positive logits

tails1.58

contagious1.52

dominating1.49

looming1.46

disqual1.44

ticking1.39

hovering1.38

ties1.37

disqualified1.37

ogun1.36

capacity1.36

bub1.35

ripe1.34

hindsight1.33

churn1.33

olulu1.33

unanim1.30

pandemonium1.29

strides1.28

poised1.27

Negative logits

Cosponsors-1.75

Interstitial-1.54

bol-1.54

dayName-1.49

PS-1.41

Relations-1.38

sburg-1.37

phrine-1.32

龍喚士-1.31

atern-1.30

":"/-1.30

References-1.28

ELY-1.28

orse-1.28

hof-1.28

Measure-1.27

-1.26

Square-1.26

unfocusedRange-1.25

Tele-1.25

INTERVAL 3.397 - 3.774
CONTAINS 0.000%

digging
Token digging
Feature activation+0.473
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
stuff
Token stuff
Feature activation+0.000
,
Token,
Feature activation+3.774
Nissan
Token Nissan
Feature activation+0.000
is
Token is
Feature activation+0.000
teasing
Token teasing
Feature activation+0.000
us
Token us
Feature activation+0.000
all
Token all
Feature activation+0.000

INTERVAL 3.019 - 3.397
CONTAINS 0.002%

life
Token life
Feature activation+0.000
boats
Tokenboats
Feature activation+0.000
either
Token either
Feature activation+0.000
full
Token full
Feature activation+2.783
or
Token or
Feature activation+0.000
gone
Token gone
Feature activation+3.260
,
Token,
Feature activation+2.959
the
Token the
Feature activation+2.376
Dav
Token Dav
Feature activation+0.000
ises
Tokenises
Feature activation+0.000
were
Token were
Feature activation+0.000
a
Token a
Feature activation+1.081
.
Token.
Feature activation+0.156
m
Tokenm
Feature activation+0.892
.
Token.
Feature activation+0.000
services
Token services
Feature activation+0.058
,
Token,
Feature activation+3.115
this
Token this
Feature activation+0.927
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
season
Token season
Feature activation+0.000
for
Token for
Feature activation+0.000
12
Token 12
Feature activation+0.000
teams
Token teams
Feature activation+0.000
in
Token in
Feature activation+1.919
three
Token three
Feature activation+1.371
years
Token years
Feature activation+0.088
,
Token,
Feature activation+3.243
we
Token we
Feature activation+0.263
now
Token now
Feature activation+0.000
have
Token have
Feature activation+0.000
enough
Token enough
Feature activation+0.000
data
Token data
Feature activation+0.000
background
Token background
Feature activation+0.465
in
Token in
Feature activation+0.000
design
Token design
Feature activation+0.000
or
Token or
Feature activation+0.000
manufacturing
Token manufacturing
Feature activation+0.000
,
Token,
Feature activation+3.116
Tam
Token Tam
Feature activation+0.000
ir
Tokenir
Feature activation+0.000
figured
Token figured
Feature activation+0.110
out
Token out
Feature activation+0.000
that
Token that
Feature activation+0.000
threats
Token threats
Feature activation+0.000
of
Token of
Feature activation+0.000
equality
Token equality
Feature activation+0.000
to
Token to
Feature activation+0.000
slay
Token slay
Feature activation+0.000
,
Token,
Feature activation+3.314
Sch
Token Sch
Feature activation+0.000
alf
Tokenalf
Feature activation+0.000
ly
Tokenly
Feature activation+0.000
became
Token became
Feature activation+0.000
just
Token just
Feature activation+0.000

INTERVAL 2.642 - 3.019
CONTAINS 0.004%

Ċ
TokenĊ
Feature activation+0.000
With
TokenWith
Feature activation+0.000
life
Token life
Feature activation+0.000
boats
Tokenboats
Feature activation+0.000
either
Token either
Feature activation+0.000
full
Token full
Feature activation+2.783
or
Token or
Feature activation+0.000
gone
Token gone
Feature activation+3.260
,
Token,
Feature activation+2.959
the
Token the
Feature activation+2.376
Dav
Token Dav
Feature activation+0.000
ribbon
Token ribbon
Feature activation+0.789
-
Token-
Feature activation+0.000
cutting
Tokencutting
Feature activation+0.000
ceremony
Token ceremony
Feature activation+0.096
Thursday
Token Thursday
Feature activation+0.000
,
Token,
Feature activation+2.780
the
Token the
Feature activation+1.819
intricate
Token intricate
Feature activation+0.000
,
Token,
Feature activation+0.000
expensive
Token expensive
Feature activation+0.000
system
Token system
Feature activation+0.000
order
Token order
Feature activation+0.000
.
Token.
Feature activation+0.000
With
Token With
Feature activation+0.000
this
Token this
Feature activation+0.000
in
Token in
Feature activation+1.713
mind
Token mind
Feature activation+3.019
,
Token,
Feature activation+1.433
on
Token on
Feature activation+1.087
April
Token April
Feature activation+0.000
20
Token 20
Feature activation+0.000
,
Token,
Feature activation+0.000
Semitism
TokenSemitism
Feature activation+0.000
.
Token.
Feature activation+0.000
With
Token With
Feature activation+0.000
that
Token that
Feature activation+0.145
in
Token in
Feature activation+1.487
mind
Token mind
Feature activation+2.676
,
Token,
Feature activation+0.905
here
Token here
Feature activation+0.000
's
Token's
Feature activation+0.000
the
Token the
Feature activation+0.000
(
Token (
Feature activation+0.000
offensive
Token offensive
Feature activation+0.000
performance
Token performance
Feature activation+0.000
from
Token from
Feature activation+0.103
their
Token their
Feature activation+1.152
backup
Token backup
Feature activation+0.000
,
Token,
Feature activation+2.737
the
Token the
Feature activation+2.177
Giants
Token Giants
Feature activation+0.000
might
Token might
Feature activation+0.310
have
Token have
Feature activation+0.000
an
Token an
Feature activation+0.000

INTERVAL 2.264 - 2.642
CONTAINS 0.005%

until
Token until
Feature activation+0.579
the
Token the
Feature activation+0.000
summer
Token summer
Feature activation+0.000
window
Token window
Feature activation+0.000
shuts
Token shuts
Feature activation+0.000
,
Token,
Feature activation+2.413
it
Token it
Feature activation+0.884
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
familiar
Token familiar
Feature activation+0.000
strategy
Token strategy
Feature activation+0.000
favorite
Token favorite
Feature activation+0.000
to
Token to
Feature activation+0.000
win
Token win
Feature activation+0.000
La
Token La
Feature activation+0.000
Liga
Token Liga
Feature activation+0.000
,
Token,
Feature activation+2.274
the
Token the
Feature activation+2.108
Champions
Token Champions
Feature activation+0.000
League
Token League
Feature activation+0.000
now
Token now
Feature activation+0.183
takes
Token takes
Feature activation+0.000
the
Token the
Feature activation+0.220
NRL
Token NRL
Feature activation+0.000
season
Token season
Feature activation+0.000
back
Token back
Feature activation+1.883
in
Token in
Feature activation+1.910
full
Token full
Feature activation+2.628
swing
Token swing
Feature activation+0.000
,
Token,
Feature activation+2.997
it
Token it
Feature activation+1.211
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
With
TokenWith
Feature activation+0.000
each
Token each
Feature activation+0.216
passing
Token passing
Feature activation+1.878
day
Token day
Feature activation+0.000
,
Token,
Feature activation+2.533
China
Token China
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
patent
Token patent
Feature activation+0.000
a
Token a
Feature activation+1.300
little
Token little
Feature activation+2.691
bit
Token bit
Feature activation+0.246
of
Token of
Feature activation+0.000
recognition
Token recognition
Feature activation+1.620
,
Token,
Feature activation+2.632
the
Token the
Feature activation+1.214
pair
Token pair
Feature activation+0.000
has
Token has
Feature activation+0.000
managed
Token managed
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 1.887 - 2.264
CONTAINS 0.009%

positive
Token positive
Feature activation+0.000
.
Token.
Feature activation+0.000
With
Token With
Feature activation+0.000
any
Token any
Feature activation+0.000
luck
Token luck
Feature activation+0.954
it
Token it
Feature activation+1.906
will
Token will
Feature activation+0.000
reach
Token reach
Feature activation+0.000
out
Token out
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
Kickstarter
Token Kickstarter
Feature activation+0.000
launch
Token launch
Feature activation+0.000
party
Token party
Feature activation+0.000
and
Token and
Feature activation+1.603
private
Token private
Feature activation+0.000
view
Token view
Feature activation+1.949
for
Token for
Feature activation+0.000
their
Token their
Feature activation+0.083
Beast
Token Beast
Feature activation+0.000
W
Token W
Feature activation+0.000
agon
Tokenagon
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.759
Carn
Token Carn
Feature activation+0.000
ive
Tokenive
Feature activation+0.000
an
Tokenan
Feature activation+0.000
,
Token,
Feature activation+2.171
Th
Token Th
Feature activation+0.000
ag
Tokenag
Feature activation+0.000
ro
Tokenro
Feature activation+0.000
sh
Tokensh
Feature activation+0.000
has
Token has
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
With
TokenWith
Feature activation+0.000
this
Token this
Feature activation+0.246
exit
Token exit
Feature activation+0.000
,
Token,
Feature activation+2.055
Merc
Token Merc
Feature activation+0.000
iless
Tokeniless
Feature activation+0.000
has
Token has
Feature activation+0.000
only
Token only
Feature activation+0.000
Fel
Token Fel
Feature activation+0.000
a
Token a
Feature activation+0.000
thick
Token thick
Feature activation+0.000
accent
Token accent
Feature activation+0.863
Greg
Token Greg
Feature activation+0.131
answered
Token answered
Feature activation+2.484
,
Token,
Feature activation+2.233
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
He
TokenHe
Feature activation+0.000
st
Token st
Feature activation+0.000
uttered
Tokenuttered
Feature activation+0.000

INTERVAL 1.510 - 1.887
CONTAINS 0.012%

$
Token $
Feature activation+0.535
50
Token50
Feature activation+0.000
a
Token a
Feature activation+0.560
barrel
Token barrel
Feature activation+0.334
and
Token and
Feature activation+2.011
predictions
Token predictions
Feature activation+1.712
they
Token they
Feature activation+0.668
could
Token could
Feature activation+0.000
fall
Token fall
Feature activation+0.000
even
Token even
Feature activation+0.000
further
Token further
Feature activation+0.000
with
Token with
Feature activation+0.000
very
Token very
Feature activation+0.000
few
Token few
Feature activation+0.000
exceptions
Token exceptions
Feature activation+1.402
,
Token,
Feature activation+1.608
there
Token there
Feature activation+1.590
is
Token is
Feature activation+0.000
just
Token just
Feature activation+0.000
nothing
Token nothing
Feature activation+0.000
that
Token that
Feature activation+0.000
comes
Token comes
Feature activation+0.000
Audi
Token Audi
Feature activation+0.000
,
Token,
Feature activation+1.113
BMW
Token BMW
Feature activation+0.000
and
Token and
Feature activation+0.000
Mercedes
Token Mercedes
Feature activation+0.000
stepping
Token stepping
Feature activation+1.680
up
Token up
Feature activation+0.000
,
Token,
Feature activation+1.968
it
Token it
Feature activation+2.165
will
Token will
Feature activation+0.000
only
Token only
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
With
TokenWith
Feature activation+0.000
this
Token this
Feature activation+0.045
out
Token out
Feature activation+1.491
of
Token of
Feature activation+1.671
the
Token the
Feature activation+1.653
way
Token way
Feature activation+1.682
,
Token,
Feature activation+1.965
we
Token we
Feature activation+0.721
would
Token would
Feature activation+0.000
pause
Token pause
Feature activation+0.000
:
Token:
Feature activation+0.000
With
Token With
Feature activation+0.000
an
Token an
Feature activation+0.000
app
Token app
Feature activation+0.000
such
Token such
Feature activation+1.745
as
Token as
Feature activation+0.000
Pokémon
Token Pokémon
Feature activation+0.000
Go
Token Go
Feature activation+0.000
,
Token,
Feature activation+2.570
we
Token we
Feature activation+1.701

INTERVAL 1.132 - 1.510
CONTAINS 0.014%

-
Token-
Feature activation+0.000
in
Tokenin
Feature activation+0.000
support
Token support
Feature activation+0.529
for
Token for
Feature activation+0.000
Android
Token Android
Feature activation+0.000
,
Token,
Feature activation+1.401
OpenGL
Token OpenGL
Feature activation+0.000
ES
Token ES
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
HTML
Token HTML
Feature activation+0.000
With
TokenWith
Feature activation+0.000
Sydney
Token Sydney
Feature activation+0.000
Ro
Token Ro
Feature activation+0.000
ost
Tokenost
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
,
Token,
Feature activation+1.436
South
Token South
Feature activation+0.000
Sydney
Token Sydney
Feature activation+0.000
,
Token,
Feature activation+0.000
Cron
Token Cron
Feature activation+0.000
ulla
Tokenulla
Feature activation+0.000
waiting
Token waiting
Feature activation+0.000
.
Token.
Feature activation+0.000
With
Token With
Feature activation+0.000
the
Token the
Feature activation+0.187
battery
Token battery
Feature activation+0.000
charged
Token charged
Feature activation+1.192
and
Token and
Feature activation+1.391
everything
Token everything
Feature activation+1.394
appearing
Token appearing
Feature activation+1.302
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
With
TokenWith
Feature activation+0.000
all
Token all
Feature activation+0.000
of
Token of
Feature activation+0.956
that
Token that
Feature activation+1.251
said
Token said
Feature activation+1.734
,
Token,
Feature activation+1.301
here
Token here
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
how
Token how
Feature activation+0.000
val
Token val
Feature activation+0.000
uations
Tokenuations
Feature activation+0.000
and
Token and
Feature activation+0.663
ugly
Token ugly
Feature activation+0.327
declines
Token declines
Feature activation+0.712
as
Token as
Feature activation+1.377
of
Token of
Feature activation+0.000
late
Token late
Feature activation+0.000
,
Token,
Feature activation+1.919
the
Token the
Feature activation+1.645
negativity
Token negativity
Feature activation+0.000

INTERVAL 0.755 - 1.132
CONTAINS 0.016%

comes
Token comes
Feature activation+0.000
to
Token to
Feature activation+0.000
personal
Token personal
Feature activation+0.000
data
Token data
Feature activation+0.000
security
Token security
Feature activation+0.000
,
Token,
Feature activation+0.936
environmental
Token environmental
Feature activation+0.000
awareness
Token awareness
Feature activation+0.000
and
Token and
Feature activation+0.000
human
Token human
Feature activation+0.000
rights
Token rights
Feature activation+0.000
,
Token,
Feature activation+0.000
with
Token with
Feature activation+0.000
all
Token all
Feature activation+0.000
its
Token its
Feature activation+0.978
problems
Token problems
Feature activation+0.000
,
Token,
Feature activation+0.778
can
Token can
Feature activation+0.000
see
Token see
Feature activation+0.000
the
Token the
Feature activation+0.000
sense
Token sense
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+1.230
few
Token few
Feature activation+1.119
other
Token other
Feature activation+0.000
restaurants
Token restaurants
Feature activation+0.000
around
Token around
Feature activation+2.244
the
Token the
Feature activation+0.922
country
Token country
Feature activation+0.000
having
Token having
Feature activation+2.519
made
Token made
Feature activation+0.000
the
Token the
Feature activation+0.000
move
Token move
Feature activation+0.000
With
Token With
Feature activation+0.000
the
Token the
Feature activation+0.000
help
Token help
Feature activation+0.936
of
Token of
Feature activation+0.071
her
Token her
Feature activation+0.211
own
Token own
Feature activation+0.875
observations
Token observations
Feature activation+0.609
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
portal
Token portal
Feature activation+0.000
between
Token between
Feature activation+0.000
and
Token and
Feature activation+0.000
with
Token with
Feature activation+0.000
less
Token less
Feature activation+0.000
problems
Token problems
Feature activation+0.000
as
Token as
Feature activation+0.269
possible
Token possible
Feature activation+0.964
.
Token.
Feature activation+0.000
That
Token That
Feature activation+0.000
was
Token was
Feature activation+0.000
achievable
Token achievable
Feature activation+0.000
because
Token because
Feature activation+0.000

INTERVAL 0.377 - 0.755
CONTAINS 0.025%

and
Token and
Feature activation+0.000
streaming
Token streaming
Feature activation+0.000
video
Token video
Feature activation+0.000
services
Token services
Feature activation+0.000
it
Token it
Feature activation+1.825
looks
Token looks
Feature activation+0.391
more
Token more
Feature activation+0.000
and
Token and
Feature activation+0.000
more
Token more
Feature activation+0.000
like
Token like
Feature activation+0.000
television
Token television
Feature activation+0.000
(
Token (
Feature activation+0.000
eg
Tokeneg
Feature activation+0.000
:
Token:
Feature activation+0.000
Windows
Token Windows
Feature activation+0.000
Update
Token Update
Feature activation+0.000
),
Token),
Feature activation+0.501
users
Token users
Feature activation+0.000
may
Token may
Feature activation+0.000
either
Token either
Feature activation+0.000
choose
Token choose
Feature activation+0.000
to
Token to
Feature activation+0.000
to
Token to
Feature activation+0.000
our
Token our
Feature activation+0.000
cause
Token cause
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
sufficient
Token sufficient
Feature activation+0.617
police
Token police
Feature activation+0.000
and
Token and
Feature activation+0.000
soldiers
Token soldiers
Feature activation+0.000
to
Token to
Feature activation+0.000
protect
Token protect
Feature activation+0.000
the
Token the
Feature activation+0.000
P
Token P
Feature activation+0.000
ono
Tokenono
Feature activation+0.000
Player
TokenPlayer
Feature activation+0.000
,
Token,
Feature activation+0.084
you
Token you
Feature activation+0.540
can
Token can
Feature activation+0.376
finally
Token finally
Feature activation+0.000
feel
Token feel
Feature activation+0.000
the
Token the
Feature activation+0.000
master
Token master
Feature activation+0.000
knees
Token knees
Feature activation+0.000
,
Token,
Feature activation+0.000
with
Token with
Feature activation+0.000
my
Token my
Feature activation+0.000
arms
Token arms
Feature activation+0.090
out
Token out
Feature activation+0.634
stretched
Tokenstretched
Feature activation+0.892
with
Token with
Feature activation+0.000
a
Token a
Feature activation+0.000
large
Token large
Feature activation+0.000
steel
Token steel
Feature activation+0.000

INTERVAL 0.000 - 0.377
CONTAINS 99.914%

Major
Token Major
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
King
TokenKing
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Kong
Token Kong
Feature activation+0.000
part
Token part
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
ultimately
Token ultimately
Feature activation+0.000
world
Token world
Feature activation+0.000
over
Token over
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
You
TokenYou
Feature activation+0.000
can
Token can
Feature activation+0.000
imagine
Token imagine
Feature activation+0.000
how
Token how
Feature activation+0.000
up
Token up
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
stands
Token stands
Feature activation+0.000
and
Token and
Feature activation+0.000
said
Token said
Feature activation+0.000
,
Token,
Feature activation+0.000
"
Token "
Feature activation+0.000
I
TokenI
Feature activation+0.000
hope
Token hope
Feature activation+0.000
there
Token there
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
former
Tokenformer
Feature activation+0.000
Koch
Token Koch
Feature activation+0.000
employee
Token employee
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
It
Token It
Feature activation+0.000
implies
Token implies
Feature activation+0.000
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
?"
Token?"
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
She
TokenShe
Feature activation+0.000
also
Token also
Feature activation+0.000
noted
Token noted
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
Center
Token Center
Feature activation+0.000
is
Token is
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 6 in H3.9: (feature 3607)

TOP ACTIVATIONS
MAX = 4.922

Germany
Token Germany
Feature activation+0.000
policymakers
Token policymakers
Feature activation+0.000
to
Token to
Feature activation+0.000
leave
Token leave
Feature activation+0.000
it
Token it
Feature activation+1.076
alone
Token alone
Feature activation+4.922
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
2012
Token 2012
Feature activation+0.000
I
TokenI
Feature activation+0.000
leave
Token leave
Feature activation+0.000
my
Token my
Feature activation+0.845
play
Token play
Feature activation+0.000
books
Tokenbooks
Feature activation+0.469
alone
Token alone
Feature activation+4.841
.
Token.
Feature activation+0.198
Everything
Token Everything
Feature activation+0.000
you
Token you
Feature activation+0.000
need
Token need
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
Vatican
Token Vatican
Feature activation+0.000
and
Token and
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
Christianity
Token Christianity
Feature activation+0.714
und
Token und
Feature activation+4.706
ef
Tokenef
Feature activation+0.000
ended
Tokenended
Feature activation+2.501
against
Token against
Feature activation+0.462
the
Token the
Feature activation+0.648
growing
Token growing
Feature activation+0.000
fight
Token fight
Feature activation+0.000
that
Token that
Feature activation+0.000
left
Token left
Feature activation+0.000
two
Token two
Feature activation+2.070
patrons
Token patrons
Feature activation+1.301
injured
Token injured
Feature activation+4.530
,
Token,
Feature activation+0.000
one
Token one
Feature activation+0.000
of
Token of
Feature activation+0.000
whom
Token whom
Feature activation+0.000
was
Token was
Feature activation+0.000
the
Token the
Feature activation+0.378
majority
Token majority
Feature activation+0.000
of
Token of
Feature activation+0.094
rape
Token rape
Feature activation+0.000
cases
Token cases
Feature activation+0.287
unre
Token unre
Feature activation+4.518
ported
Tokenported
Feature activation+0.707
and
Token and
Feature activation+0.000
un
Token un
Feature activation+2.350
invest
Tokeninvest
Feature activation+0.000
igated
Tokenigated
Feature activation+0.063
D
Token D
Feature activation+0.000
ax
Tokenax
Feature activation+0.000
am
Tokenam
Feature activation+0.000
ites
Tokenites
Feature activation+0.000
no
Token no
Feature activation+3.991
choice
Token choice
Feature activation+4.448
but
Token but
Feature activation+0.283
to
Token to
Feature activation+0.000
leave
Token leave
Feature activation+0.000
or
Token or
Feature activation+0.000
die
Token die
Feature activation+0.220
the
Token the
Feature activation+0.000
plans
Token plans
Feature activation+0.000
left
Token left
Feature activation+0.612
too
Token too
Feature activation+2.370
many
Token many
Feature activation+1.587
unanswered
Token unanswered
Feature activation+4.343
questions
Token questions
Feature activation+1.574
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
ambassador
Token ambassador
Feature activation+0.000
,
Token,
Feature activation+0.000
dead
Token dead
Feature activation+4.334
,
Token,
Feature activation+0.000
Rice
Token Rice
Feature activation+0.000
was
Token was
Feature activation+0.000
par
Token par
Feature activation+0.000
aded
Tokenaded
Feature activation+0.000
that
Token that
Feature activation+0.000
habit
Token habit
Feature activation+0.000
ually
Tokenually
Feature activation+0.000
leave
Token leave
Feature activation+0.000
that
Token that
Feature activation+0.498
mark
Token mark
Feature activation+4.311
in
Token in
Feature activation+3.769
the
Token the
Feature activation+1.664
dust
Token dust
Feature activation+1.348
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
How
TokenHow
Feature activation+0.000
's
Token's
Feature activation+0.000
that
Token that
Feature activation+0.000
leave
Token leave
Feature activation+0.000
me
Token me
Feature activation+0.534
alone
Token alone
Feature activation+4.295
personal
Token personal
Feature activation+0.000
liberty
Token liberty
Feature activation+0.322
thing
Token thing
Feature activation+0.000
working
Token working
Feature activation+0.182
out
Token out
Feature activation+0.000
protests
Token protests
Feature activation+0.000
that
Token that
Feature activation+0.000
left
Token left
Feature activation+0.000
16
Token 16
Feature activation+2.274
officers
Token officers
Feature activation+0.000
injured
Token injured
Feature activation+4.278
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
His
TokenHis
Feature activation+0.000
family
Token family
Feature activation+0.000
Le
TokenLe
Feature activation+0.000
aving
Tokenaving
Feature activation+0.000
my
Token my
Feature activation+0.822
own
Token own
Feature activation+1.211
children
Token children
Feature activation+3.029
for
Token for
Feature activation+4.252
six
Token six
Feature activation+0.056
months
Token months
Feature activation+0.726
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
choice
Token choice
Feature activation+0.000
could
Token could
Feature activation+0.000
potentially
Token potentially
Feature activation+0.000
leave
Token leave
Feature activation+0.000
power
Token power
Feature activation+0.915
producers
Token producers
Feature activation+0.000
without
Token without
Feature activation+4.225
enough
Token enough
Feature activation+0.227
buyers
Token buyers
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
and
Token and
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
her
Token her
Feature activation+0.986
in
Token in
Feature activation+3.598
relationship
Token relationship
Feature activation+0.450
limbo
Token limbo
Feature activation+4.183
.
Token.
Feature activation+0.000
My
Token My
Feature activation+0.000
father
Token father
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
old
Tokenold
Feature activation+0.000
daughter
Token daughter
Feature activation+0.584
Ad
Token Ad
Feature activation+0.000
ri
Tokenri
Feature activation+0.000
anna
Tokenanna
Feature activation+0.483
alone
Token alone
Feature activation+4.127
in
Token in
Feature activation+1.275
a
Token a
Feature activation+1.671
sw
Token sw
Feature activation+0.000
el
Tokenel
Feature activation+0.000
tering
Tokentering
Feature activation+0.000
and
Token and
Feature activation+0.000
I
Token I
Feature activation+0.000
left
Token left
Feature activation+0.000
them
Token them
Feature activation+0.000
on
Token on
Feature activation+3.528
for
Token for
Feature activation+4.109
about
Token about
Feature activation+0.000
five
Token five
Feature activation+0.552
minutes
Token minutes
Feature activation+0.163
before
Token before
Feature activation+0.000
real
Token real
Feature activation+0.000
pretty
Token pretty
Feature activation+0.000
much
Token much
Feature activation+0.000
left
Token left
Feature activation+0.000
in
Token in
Feature activation+2.361
the
Token the
Feature activation+2.531
dark
Token dark
Feature activation+4.100
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
could
Token could
Feature activation+0.000
fake
Token fake
Feature activation+0.000
it
Token it
Feature activation+0.000
Don
TokenDon
Feature activation+0.000
't
Token't
Feature activation+0.000
!
Token!
Feature activation+0.000
Leave
Token Leave
Feature activation+0.000
it
Token it
Feature activation+1.502
alone
Token alone
Feature activation+4.043
!"
Token!"
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
î
Tokenî
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
but
Token but
Feature activation+0.381
marsh
Token marsh
Feature activation+0.000
mallow
Tokenmallow
Feature activation+0.846
y
Tokeny
Feature activation+0.000
goodness
Token goodness
Feature activation+0.020
to
Token to
Feature activation+4.039
pour
Token pour
Feature activation+0.000
into
Token into
Feature activation+0.000
a
Token a
Feature activation+0.000
bowl
Token bowl
Feature activation+0.000
.
Token.
Feature activation+0.000
blood
Token blood
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
left
Token left
Feature activation+0.000
blood
Token blood
Feature activation+3.209
stains
Token stains
Feature activation+4.017
on
Token on
Feature activation+3.617
my
Token my
Feature activation+1.388
hands
Token hands
Feature activation+0.582
.
Token.
Feature activation+0.000
Lex
Token Lex
Feature activation+0.000

Top DFA by src position
MAX = 5.224

vin
Tokenvin
Feature activation+0.035
Top resid features:
ced
Tokenced
Feature activation+0.061
Top resid features:
Germany
Token Germany
Feature activation+0.007
Top resid features:
policymakers
Token policymakers
Feature activation+0.023
Top resid features:
to
Token to
Feature activation+0.020
Top resid features:
leave
Token leave
Feature activation+5.179
Top resid features:
it
Token it
Feature activation+0.285
Top resid features:
alone
Token alone
Feature activation+0.101
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
Basics
Token Basics
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.019
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
I
TokenI
Feature activation+0.029
Top resid features:
leave
Token leave
Feature activation+5.224
Top resid features:
my
Token my
Feature activation+0.095
Top resid features:
play
Token play
Feature activation+0.036
Top resid features:
books
Tokenbooks
Feature activation+0.121
Top resid features:
alone
Token alone
Feature activation+0.136
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
from
Token from
Feature activation+0.015
Top resid features:
defending
Token defending
Feature activation-0.009
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
Vatican
Token Vatican
Feature activation-0.005
Top resid features:
and
Token and
Feature activation+0.020
Top resid features:
leaving
Token leaving
Feature activation+5.174
Top resid features:
Christianity
Token Christianity
Feature activation+0.150
Top resid features:
und
Token und
Feature activation+0.125
Top resid features:
ef
Tokenef
Feature activation+0.000
Top resid features:
ended
Tokenended
Feature activation+0.000
Top resid features:
against
Token against
Feature activation+0.000
Top resid features:
arrested
Token arrested
Feature activation+0.003
Top resid features:
after
Token after
Feature activation+0.176
Top resid features:
a
Token a
Feature activation-0.002
Top resid features:
fight
Token fight
Feature activation-0.008
Top resid features:
that
Token that
Feature activation-0.010
Top resid features:
left
Token left
Feature activation+5.110
Top resid features:
two
Token two
Feature activation+0.012
Top resid features:
patrons
Token patrons
Feature activation+0.046
Top resid features:
injured
Token injured
Feature activation+0.069
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
one
Token one
Feature activation+0.000
Top resid features:
The
Token The
Feature activation+0.000
Top resid features:
same
Token same
Feature activation+0.012
Top resid features:
rape
Token rape
Feature activation-0.007
Top resid features:
culture
Token culture
Feature activation-0.003
Top resid features:
that
Token that
Feature activation+0.003
Top resid features:
leaves
Token leaves
Feature activation+5.078
Top resid features:
the
Token the
Feature activation+0.045
Top resid features:
majority
Token majority
Feature activation+0.042
Top resid features:
of
Token of
Feature activation+0.037
Top resid features:
rape
Token rape
Feature activation+0.019
Top resid features:
cases
Token cases
Feature activation+0.097
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
The
Token The
Feature activation-0.005
Top resid features:
weapon
Token weapon
Feature activation-0.008
Top resid features:
was
Token was
Feature activation-0.009
Top resid features:
deployed
Token deployed
Feature activation-0.021
Top resid features:
leaving
Token leaving
Feature activation+4.837
Top resid features:
all
Token all
Feature activation+0.038
Top resid features:
the
Token the
Feature activation+0.017
Top resid features:
D
Token D
Feature activation+0.027
Top resid features:
ax
Tokenax
Feature activation+0.012
Top resid features:
am
Tokenam
Feature activation+0.028
Top resid features:
Far
Token Far
Feature activation+0.001
Top resid features:
ron
Tokenron
Feature activation-0.001
Top resid features:
said
Token said
Feature activation+0.114
Top resid features:
the
Token the
Feature activation+0.014
Top resid features:
plans
Token plans
Feature activation+0.007
Top resid features:
left
Token left
Feature activation+4.799
Top resid features:
too
Token too
Feature activation-0.007
Top resid features:
many
Token many
Feature activation+0.134
Top resid features:
unanswered
Token unanswered
Feature activation+0.058
Top resid features:
questions
Token questions
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.003
Top resid features:
Benghazi
Token Benghazi
Feature activation+0.001
Top resid features:
massacre
Token massacre
Feature activation-0.021
Top resid features:
,
Token,
Feature activation+0.017
Top resid features:
which
Token which
Feature activation+0.019
Top resid features:
left
Token left
Feature activation+5.150
Top resid features:
four
Token four
Feature activation-0.002
Top resid features:
Americans
Token Americans
Feature activation+0.021
Top resid features:
,
Token,
Feature activation+0.026
Top resid features:
including
Token including
Feature activation+0.032
Top resid features:
a
Token a
Feature activation-0.070
Top resid features:
World
Token World
Feature activation-0.002
Top resid features:
wines
Token wines
Feature activation-0.008
Top resid features:
that
Token that
Feature activation-0.017
Top resid features:
habit
Token habit
Feature activation+0.007
Top resid features:
ually
Tokenually
Feature activation+0.001
Top resid features:
leave
Token leave
Feature activation+4.679
Top resid features:
that
Token that
Feature activation+0.354
Top resid features:
mark
Token mark
Feature activation+0.071
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
dust
Token dust
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
"
Token "
Feature activation+0.009
Top resid features:
How
TokenHow
Feature activation+0.029
Top resid features:
's
Token's
Feature activation+0.018
Top resid features:
that
Token that
Feature activation-0.003
Top resid features:
leave
Token leave
Feature activation+5.003
Top resid features:
me
Token me
Feature activation-0.037
Top resid features:
alone
Token alone
Feature activation+0.110
Top resid features:
personal
Token personal
Feature activation+0.000
Top resid features:
liberty
Token liberty
Feature activation+0.000
Top resid features:
thing
Token thing
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.008
Top resid features:
prompting
Token prompting
Feature activation-0.004
Top resid features:
violent
Token violent
Feature activation-0.004
Top resid features:
protests
Token protests
Feature activation+0.037
Top resid features:
that
Token that
Feature activation+0.007
Top resid features:
left
Token left
Feature activation+5.084
Top resid features:
16
Token 16
Feature activation-0.017
Top resid features:
officers
Token officers
Feature activation-0.020
Top resid features:
injured
Token injured
Feature activation+0.037
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
family
Token family
Feature activation+0.003
Top resid features:
.
Token.
Feature activation-0.003
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.004
Top resid features:
Le
TokenLe
Feature activation+0.002
Top resid features:
aving
Tokenaving
Feature activation+4.640
Top resid features:
my
Token my
Feature activation+0.064
Top resid features:
own
Token own
Feature activation+0.078
Top resid features:
children
Token children
Feature activation+0.119
Top resid features:
for
Token for
Feature activation+0.141
Top resid features:
six
Token six
Feature activation+0.000
Top resid features:
A
TokenA
Feature activation+0.026
Top resid features:
hefty
Token hefty
Feature activation+0.000
Top resid features:
ADD
Token ADD
Feature activation+0.004
Top resid features:
could
Token could
Feature activation+0.046
Top resid features:
potentially
Token potentially
Feature activation+0.015
Top resid features:
leave
Token leave
Feature activation+4.572
Top resid features:
power
Token power
Feature activation+0.074
Top resid features:
producers
Token producers
Feature activation+0.112
Top resid features:
without
Token without
Feature activation+0.156
Top resid features:
enough
Token enough
Feature activation+0.000
Top resid features:
buyers
Token buyers
Feature activation+0.000
Top resid features:
we
Token we
Feature activation-0.015
Top resid features:
were
Token were
Feature activation-0.005
Top resid features:
still
Token still
Feature activation-0.002
Top resid features:
together
Token together
Feature activation-0.009
Top resid features:
and
Token and
Feature activation+0.067
Top resid features:
leaving
Token leaving
Feature activation+4.451
Top resid features:
her
Token her
Feature activation+0.104
Top resid features:
in
Token in
Feature activation+0.079
Top resid features:
relationship
Token relationship
Feature activation+0.264
Top resid features:
limbo
Token limbo
Feature activation+0.046
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.003
Top resid features:
September
Token September
Feature activation-0.012
Top resid features:
2002
Token 2002
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.032
Top resid features:
She
Token She
Feature activation+0.000
Top resid features:
left
Token left
Feature activation+4.597
Top resid features:
her
Token her
Feature activation+0.040
Top resid features:
two
Token two
Feature activation+0.027
Top resid features:
-
Token-
Feature activation+0.004
Top resid features:
year
Tokenyear
Feature activation+0.003
Top resid features:
-
Token-
Feature activation+0.003
Top resid features:
the
Token the
Feature activation-0.003
Top resid features:
outlines
Token outlines
Feature activation+0.045
Top resid features:
,
Token,
Feature activation+0.017
Top resid features:
and
Token and
Feature activation+0.015
Top resid features:
I
Token I
Feature activation+0.008
Top resid features:
left
Token left
Feature activation+4.415
Top resid features:
them
Token them
Feature activation+0.081
Top resid features:
on
Token on
Feature activation+0.288
Top resid features:
for
Token for
Feature activation+0.056
Top resid features:
about
Token about
Feature activation+0.000
Top resid features:
five
Token five
Feature activation+0.000
Top resid features:
so
Token so
Feature activation+0.031
Top resid features:
atheists
Token atheists
Feature activation+0.018
Top resid features:
were
Token were
Feature activation+0.004
Top resid features:
pretty
Token pretty
Feature activation+0.030
Top resid features:
much
Token much
Feature activation-0.044
Top resid features:
left
Token left
Feature activation+4.496
Top resid features:
in
Token in
Feature activation+0.096
Top resid features:
the
Token the
Feature activation+0.114
Top resid features:
dark
Token dark
Feature activation+0.167
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
They
Token They
Feature activation+0.000
Top resid features:
:
Token:
Feature activation-0.013
Top resid features:
"
Token "
Feature activation-0.004
Top resid features:
Don
TokenDon
Feature activation-0.006
Top resid features:
't
Token't
Feature activation-0.002
Top resid features:
!
Token!
Feature activation-0.040
Top resid features:
Leave
Token Leave
Feature activation+4.580
Top resid features:
it
Token it
Feature activation+0.192
Top resid features:
alone
Token alone
Feature activation+0.169
Top resid features:
!"
Token!"
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
s
Token s
Feature activation+0.002
Top resid features:
ifter
Tokenifter
Feature activation+0.006
Top resid features:
,
Token,
Feature activation+0.004
Top resid features:
leaving
Token leaving
Feature activation+4.629
Top resid features:
you
Token you
Feature activation+0.056
Top resid features:
with
Token with
Feature activation+0.101
Top resid features:
nothing
Token nothing
Feature activation-0.026
Top resid features:
but
Token but
Feature activation-0.005
Top resid features:
marsh
Token marsh
Feature activation-0.005
Top resid features:
eared
Tokeneared
Feature activation-0.005
Top resid features:
in
Token in
Feature activation+0.009
Top resid features:
blood
Token blood
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.012
Top resid features:
it
Token it
Feature activation+0.016
Top resid features:
left
Token left
Feature activation+4.573
Top resid features:
blood
Token blood
Feature activation+0.178
Top resid features:
stains
Token stains
Feature activation+0.077
Top resid features:
on
Token on
Feature activation+0.000
Top resid features:
my
Token my
Feature activation+0.000
Top resid features:
hands
Token hands
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.05

Head 2: 0.06

Head 3: 0.02

Head 4: 0.01

Head 5: 0.03

Head 6: 0.05

Head 7: 0.04

Head 8: 0.06

Head 9: 0.52

Head 10: 0.06

Head 11: 0.07

Positive logits

unprotected1.62

intact1.48

undone1.46

helpless1.44

iris1.39

unprepared1.37

uble1.34

olutely1.32

untouched1.32

unpop1.31

ueless1.29

voic1.28

unanswered1.28

unatt1.27

perty1.26

guard1.25

****************1.25

pload1.18

Hague1.17

unchecked1.17

Negative logits

orf-1.52

Entered-1.34

Annotations-1.21

XP-1.20

ACTION-1.16

Cosponsors-1.14

sponsored-1.14

drives-1.14

Trinity-1.13

Interstitial-1.11

methodology-1.11

scl-1.11

drive-1.10

fram-1.10

MOR-1.09

phr-1.06

srfN-1.05

Frequency-1.04

conversions-1.04

descent-1.03

INTERVAL 4.430 - 4.922
CONTAINS 0.001%

the
Token the
Feature activation+0.000
Vatican
Token Vatican
Feature activation+0.000
and
Token and
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
Christianity
Token Christianity
Feature activation+0.714
und
Token und
Feature activation+4.706
ef
Tokenef
Feature activation+0.000
ended
Tokenended
Feature activation+2.501
against
Token against
Feature activation+0.462
the
Token the
Feature activation+0.648
growing
Token growing
Feature activation+0.000
Germany
Token Germany
Feature activation+0.000
policymakers
Token policymakers
Feature activation+0.000
to
Token to
Feature activation+0.000
leave
Token leave
Feature activation+0.000
it
Token it
Feature activation+1.076
alone
Token alone
Feature activation+4.922
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
2012
Token 2012
Feature activation+0.000
I
TokenI
Feature activation+0.000
leave
Token leave
Feature activation+0.000
my
Token my
Feature activation+0.845
play
Token play
Feature activation+0.000
books
Tokenbooks
Feature activation+0.469
alone
Token alone
Feature activation+4.841
.
Token.
Feature activation+0.198
Everything
Token Everything
Feature activation+0.000
you
Token you
Feature activation+0.000
need
Token need
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.378
majority
Token majority
Feature activation+0.000
of
Token of
Feature activation+0.094
rape
Token rape
Feature activation+0.000
cases
Token cases
Feature activation+0.287
unre
Token unre
Feature activation+4.518
ported
Tokenported
Feature activation+0.707
and
Token and
Feature activation+0.000
un
Token un
Feature activation+2.350
invest
Tokeninvest
Feature activation+0.000
igated
Tokenigated
Feature activation+0.063
D
Token D
Feature activation+0.000
ax
Tokenax
Feature activation+0.000
am
Tokenam
Feature activation+0.000
ites
Tokenites
Feature activation+0.000
no
Token no
Feature activation+3.991
choice
Token choice
Feature activation+4.448
but
Token but
Feature activation+0.283
to
Token to
Feature activation+0.000
leave
Token leave
Feature activation+0.000
or
Token or
Feature activation+0.000
die
Token die
Feature activation+0.220

INTERVAL 3.938 - 4.430
CONTAINS 0.002%

pretty
Token pretty
Feature activation+0.000
much
Token much
Feature activation+0.000
left
Token left
Feature activation+0.000
in
Token in
Feature activation+2.361
the
Token the
Feature activation+2.531
dark
Token dark
Feature activation+4.100
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
could
Token could
Feature activation+0.000
fake
Token fake
Feature activation+0.000
it
Token it
Feature activation+0.000
and
Token and
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
her
Token her
Feature activation+0.986
in
Token in
Feature activation+3.598
relationship
Token relationship
Feature activation+0.450
limbo
Token limbo
Feature activation+4.183
.
Token.
Feature activation+0.000
My
Token My
Feature activation+0.000
father
Token father
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
and
Token and
Feature activation+0.000
I
Token I
Feature activation+0.000
left
Token left
Feature activation+0.000
them
Token them
Feature activation+0.000
on
Token on
Feature activation+3.528
for
Token for
Feature activation+4.109
about
Token about
Feature activation+0.000
five
Token five
Feature activation+0.552
minutes
Token minutes
Feature activation+0.163
before
Token before
Feature activation+0.000
real
Token real
Feature activation+0.000
blood
Token blood
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
left
Token left
Feature activation+0.000
blood
Token blood
Feature activation+3.209
stains
Token stains
Feature activation+4.017
on
Token on
Feature activation+3.617
my
Token my
Feature activation+1.388
hands
Token hands
Feature activation+0.582
.
Token.
Feature activation+0.000
Lex
Token Lex
Feature activation+0.000
but
Token but
Feature activation+0.381
marsh
Token marsh
Feature activation+0.000
mallow
Tokenmallow
Feature activation+0.846
y
Tokeny
Feature activation+0.000
goodness
Token goodness
Feature activation+0.020
to
Token to
Feature activation+4.039
pour
Token pour
Feature activation+0.000
into
Token into
Feature activation+0.000
a
Token a
Feature activation+0.000
bowl
Token bowl
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 3.446 - 3.938
CONTAINS 0.004%

this
Token this
Feature activation+0.000
stuff
Token stuff
Feature activation+0.000
leave
Token leave
Feature activation+0.000
me
Token me
Feature activation+1.584
a
Token a
Feature activation+2.112
comment
Token comment
Feature activation+3.887
below
Token below
Feature activation+2.209
and
Token and
Feature activation+0.452
let
Token let
Feature activation+0.000
me
Token me
Feature activation+0.000
know
Token know
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
to
Token to
Feature activation+0.000
leave
Token leave
Feature activation+0.000
car
Token car
Feature activation+0.000
ownership
Token ownership
Feature activation+0.337
behind
Token behind
Feature activation+3.764
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
It
TokenIt
Feature activation+0.000
may
Token may
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
I
Token I
Feature activation+0.000
left
Token left
Feature activation+0.000
them
Token them
Feature activation+0.000
on
Token on
Feature activation+3.528
for
Token for
Feature activation+4.109
about
Token about
Feature activation+0.000
five
Token five
Feature activation+0.552
minutes
Token minutes
Feature activation+0.163
before
Token before
Feature activation+0.000
still
Token still
Feature activation+0.000
together
Token together
Feature activation+0.000
and
Token and
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
her
Token her
Feature activation+0.986
in
Token in
Feature activation+3.598
relationship
Token relationship
Feature activation+0.450
limbo
Token limbo
Feature activation+4.183
.
Token.
Feature activation+0.000
My
Token My
Feature activation+0.000
father
Token father
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
generally
Token generally
Feature activation+0.000
left
Token left
Feature activation+0.000
alone
Token alone
Feature activation+3.501
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Same
TokenSame
Feature activation+0.000
-
Token-
Feature activation+0.000

INTERVAL 2.953 - 3.446
CONTAINS 0.005%

If
Token If
Feature activation+0.000
the
Token the
Feature activation+0.000
system
Token system
Feature activation+0.000
is
Token is
Feature activation+0.000
left
Token left
Feature activation+0.000
up
Token up
Feature activation+2.991
to
Token to
Feature activation+3.593
states
Token states
Feature activation+0.476
,
Token,
Feature activation+0.000
implementation
Token implementation
Feature activation+0.000
would
Token would
Feature activation+0.000
more
Token more
Feature activation+0.000
migrants
Token migrants
Feature activation+0.000
are
Token are
Feature activation+0.000
being
Token being
Feature activation+0.000
left
Token left
Feature activation+0.000
stranded
Token stranded
Feature activation+2.993
in
Token in
Feature activation+0.000
legal
Token legal
Feature activation+0.000
limbo
Token limbo
Feature activation+2.627
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
finally
Token finally
Feature activation+0.000
leave
Token leave
Feature activation+0.000
,
Token,
Feature activation+0.000
embarrassed
Token embarrassed
Feature activation+3.048
,
Token,
Feature activation+0.000
with
Token with
Feature activation+3.178
bruised
Token bruised
Feature activation+1.503
and
Token and
Feature activation+0.000
bright
Token bright
Feature activation+0.000
red
Token red
Feature activation+0.000
sp
Token sp
Feature activation+0.000
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
finally
Token finally
Feature activation+0.000
leave
Token leave
Feature activation+0.000
,
Token,
Feature activation+0.000
embarrassed
Token embarrassed
Feature activation+3.048
,
Token,
Feature activation+0.000
with
Token with
Feature activation+3.178
bruised
Token bruised
Feature activation+1.503
and
Token and
Feature activation+0.000
bright
Token bright
Feature activation+0.000
practical
Token practical
Feature activation+0.000
nightmare
Token nightmare
Feature activation+0.000
,
Token,
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
them
Token them
Feature activation+0.354
with
Token with
Feature activation+2.996
virtually
Token virtually
Feature activation+2.592
nothing
Token nothing
Feature activation+3.308
but
Token but
Feature activation+0.087
a
Token a
Feature activation+0.695
lawsuit
Token lawsuit
Feature activation+0.194

INTERVAL 2.461 - 2.953
CONTAINS 0.008%

old
Token old
Feature activation+0.000
,
Token,
Feature activation+0.000
I
Token I
Feature activation+0.000
left
Token left
Feature activation+0.000
him
Token him
Feature activation+0.000
with
Token with
Feature activation+2.936
my
Token my
Feature activation+1.925
in
Token in
Feature activation+0.834
-
Token-
Feature activation+0.000
laws
Tokenlaws
Feature activation+1.576
.
Token.
Feature activation+0.000
Fer
Token Fer
Feature activation+0.000
rell
Tokenrell
Feature activation+0.000
left
Token left
Feature activation+0.000
the
Token the
Feature activation+0.000
show
Token show
Feature activation+0.000
for
Token for
Feature activation+2.772
the
Token the
Feature activation+1.544
bright
Token bright
Feature activation+0.529
lights
Token lights
Feature activation+1.453
of
Token of
Feature activation+0.000
Hollywood
Token Hollywood
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
So
TokenSo
Feature activation+0.000
they
Token they
Feature activation+0.000
left
Token left
Feature activation+0.000
me
Token me
Feature activation+0.756
on
Token on
Feature activation+2.679
the
Token the
Feature activation+2.098
back
Token back
Feature activation+0.000
side
Tokenside
Feature activation+0.000
.
Token.
Feature activation+0.000
I
Token I
Feature activation+0.000
of
Token of
Feature activation+0.000
him
Token him
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
the
Token the
Feature activation+0.235
Tate
Token Tate
Feature activation+0.000
with
Token with
Feature activation+2.704
a
Token a
Feature activation+1.529
priceless
Token priceless
Feature activation+0.058
painting
Token painting
Feature activation+0.000
under
Token under
Feature activation+0.401
his
Token his
Feature activation+0.000
ion
Token ion
Feature activation+0.000
ized
Tokenized
Feature activation+0.949
metal
Token metal
Feature activation+0.567
and
Token and
Feature activation+0.918
gas
Token gas
Feature activation+0.034
behind
Token behind
Feature activation+2.791
it
Token it
Feature activation+0.289
that
Token that
Feature activation+0.273
can
Token can
Feature activation+0.000
glow
Token glow
Feature activation+0.000
for
Token for
Feature activation+0.000

INTERVAL 1.969 - 2.461
CONTAINS 0.009%

many
Token many
Feature activation+1.028
first
Token first
Feature activation+0.091
time
Token time
Feature activation+0.234
parents
Token parents
Feature activation+0.176
feeling
Token feeling
Feature activation+2.490
confused
Token confused
Feature activation+2.304
and
Token and
Feature activation+0.000
unsure
Token unsure
Feature activation+1.911
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
row
Token row
Feature activation+0.000
,
Token,
Feature activation+0.000
leaving
Token leaving
Feature activation+0.001
a
Token a
Feature activation+0.845
zero
Token zero
Feature activation+0.000
in
Token in
Feature activation+2.347
each
Token each
Feature activation+0.062
row
Token row
Feature activation+0.000
.
Token.
Feature activation+0.272
Then
Token Then
Feature activation+0.000
,
Token,
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
UK
Token UK
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
the
Token the
Feature activation+0.334
EU
Token EU
Feature activation+2.410
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
personal
Token personal
Feature activation+0.000
belief
Token belief
Feature activation+0.000
,
Token,
Feature activation+0.000
The
Token The
Feature activation+0.000
whisk
Token whisk
Feature activation+0.000
should
Token should
Feature activation+0.000
leave
Token leave
Feature activation+0.000
an
Token an
Feature activation+0.316
impression
Token impression
Feature activation+2.347
in
Token in
Feature activation+2.664
the
Token the
Feature activation+1.438
mixture
Token mixture
Feature activation+0.000
.
Token.
Feature activation+0.000
S
Token S
Feature activation+0.000
leave
Token leave
Feature activation+0.000
the
Token the
Feature activation+0.230
sh
Token sh
Feature activation+0.581
aker
Tokenaker
Feature activation+0.333
in
Token in
Feature activation+2.226
the
Token the
Feature activation+2.295
freezer
Token freezer
Feature activation+1.097
to
Token to
Feature activation+3.308
chill
Token chill
Feature activation+0.000
while
Token while
Feature activation+0.000
you
Token you
Feature activation+0.000

INTERVAL 1.477 - 1.969
CONTAINS 0.013%

leaves
Token leaves
Feature activation+0.000
a
Token a
Feature activation+1.413
lot
Token lot
Feature activation+0.954
of
Token of
Feature activation+1.045
room
Token room
Feature activation+1.436
for
Token for
Feature activation+1.791
speculation
Token speculation
Feature activation+0.000
.
Token.
Feature activation+0.000
Nothing
Token Nothing
Feature activation+0.000
is
Token is
Feature activation+0.000
yet
Token yet
Feature activation+0.000
Infantry
Token Infantry
Feature activation+0.000
Regiment
Token Regiment
Feature activation+0.000
and
Token and
Feature activation+0.000
leaving
Token leaving
Feature activation+0.046
routes
Token routes
Feature activation+0.440
to
Token to
Feature activation+1.671
the
Token the
Feature activation+0.000
P
Token P
Feature activation+0.000
us
Tokenus
Feature activation+0.000
an
Tokenan
Feature activation+0.000
area
Token area
Feature activation+0.177
"
Token "
Feature activation+0.000
Don
TokenDon
Feature activation+0.000
't
Token't
Feature activation+0.000
!
Token!
Feature activation+0.000
Leave
Token Leave
Feature activation+0.000
it
Token it
Feature activation+1.502
alone
Token alone
Feature activation+4.043
!"
Token!"
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
î
Tokenî
Feature activation+0.000
of
Token of
Feature activation+0.000
whom
Token whom
Feature activation+0.000
was
Token was
Feature activation+0.000
left
Token left
Feature activation+0.000
in
Token in
Feature activation+2.050
the
Token the
Feature activation+1.806
desert
Token desert
Feature activation+0.007
to
Token to
Feature activation+2.176
die
Token die
Feature activation+1.420
,
Token,
Feature activation+0.000
has
Token has
Feature activation+0.000
and
Token and
Feature activation+0.000
leave
Token leave
Feature activation+0.000
comments
Token comments
Feature activation+3.764
with
Token with
Feature activation+3.907
my
Token my
Feature activation+2.623
name
Token name
Feature activation+1.634
and
Token and
Feature activation+0.035
the
Token the
Feature activation+0.304
name
Token name
Feature activation+0.472
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 0.984 - 1.477
CONTAINS 0.013%

to
Token to
Feature activation+0.000
leave
Token leave
Feature activation+0.000
the
Token the
Feature activation+0.367
EU
Token EU
Feature activation+2.917
in
Token in
Feature activation+1.868
a
Token a
Feature activation+1.325
referendum
Token referendum
Feature activation+0.000
three
Token three
Feature activation+0.467
weeks
Token weeks
Feature activation+0.000
ago
Token ago
Feature activation+0.000
.
Token.
Feature activation+0.000
ced
Tokenced
Feature activation+0.000
Germany
Token Germany
Feature activation+0.000
policymakers
Token policymakers
Feature activation+0.000
to
Token to
Feature activation+0.000
leave
Token leave
Feature activation+0.000
it
Token it
Feature activation+1.076
alone
Token alone
Feature activation+4.922
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
tragic
Token tragic
Feature activation+0.000
death
Token death
Feature activation+0.000
left
Token left
Feature activation+0.000
the
Token the
Feature activation+1.105
world
Token world
Feature activation+0.948
heart
Token heart
Feature activation+0.203
broken
Tokenbroken
Feature activation+3.743
on
Token on
Feature activation+2.670
Nov
Token Nov
Feature activation+0.000
that
Token that
Feature activation+0.000
way
Token way
Feature activation+1.286
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.000
else
Token else
Feature activation+0.000
data
Token data
Feature activation+1.418
you
Token you
Feature activation+0.505
mean
Token mean
Feature activation+0.000
to
Token to
Feature activation+0.000
keep
Token keep
Feature activation+0.000
private
Token private
Feature activation+0.000
wing
Token wing
Feature activation+0.000
opponents
Token opponents
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
administration
Token administration
Feature activation+0.000
but
Token but
Feature activation+1.354
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
right
Token right
Feature activation+0.000
as
Token as
Feature activation+0.000
well
Token well
Feature activation+0.064

INTERVAL 0.492 - 0.984
CONTAINS 0.023%

suspected
Token suspected
Feature activation+0.000
he
Token he
Feature activation+0.000
may
Token may
Feature activation+0.000
have
Token have
Feature activation+0.000
left
Token left
Feature activation+0.000
the
Token the
Feature activation+0.760
country
Token country
Feature activation+0.000
based
Token based
Feature activation+0.000
on
Token on
Feature activation+0.000
evidence
Token evidence
Feature activation+0.000
found
Token found
Feature activation+0.314
some
Token some
Feature activation+0.539
communities
Token communities
Feature activation+0.000
in
Token in
Feature activation+1.722
India
Token India
Feature activation+0.006
with
Token with
Feature activation+2.837
so
Token so
Feature activation+0.555
few
Token few
Feature activation+0.000
girls
Token girls
Feature activation+0.148
that
Token that
Feature activation+0.000
they
Token they
Feature activation+0.000
must
Token must
Feature activation+0.000
ched
Tokenched
Feature activation+0.000
,
Token,
Feature activation+0.000
leaving
Token leaving
Feature activation+0.000
thousands
Token thousands
Feature activation+0.351
without
Token without
Feature activation+3.158
power
Token power
Feature activation+0.927
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
even
Token even
Feature activation+0.000
did
Token did
Feature activation+0.000
nothing
Token nothing
Feature activation+0.000
ales
Tokenales
Feature activation+0.000
left
Token left
Feature activation+0.000
his
Token his
Feature activation+0.650
remote
Token remote
Feature activation+0.019
outpost
Token outpost
Feature activation+0.279
in
Token in
Feature activation+0.959
an
Token an
Feature activation+1.810
impoverished
Token impoverished
Feature activation+0.761
region
Token region
Feature activation+0.000
of
Token of
Feature activation+0.000
Kand
Token Kand
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
boyfriend
Token boyfriend
Feature activation+0.000
or
Token or
Feature activation+0.000
girlfriend
Token girlfriend
Feature activation+0.000
left
Token left
Feature activation+0.000
you
Token you
Feature activation+0.572
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
now
Token now
Feature activation+0.000
you
Token you
Feature activation+0.000
are
Token are
Feature activation+0.000

INTERVAL 0.000 - 0.492
CONTAINS 99.923%

,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
explore
Token explore
Feature activation+0.000
the
Token the
Feature activation+0.000
world
Token world
Feature activation+0.000
with
Token with
Feature activation+0.000
kindred
Token kindred
Feature activation+0.000
spirits
Token spirits
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
level
Token level
Feature activation+0.000
of
Token of
Feature activation+0.000
UV
Token UV
Feature activation+0.000
spots
Token spots
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
represent
Token represent
Feature activation+0.000
sun
Token sun
Feature activation+0.000
damage
Token damage
Feature activation+0.000
participant
Token participant
Feature activation+0.000
-
Token-
Feature activation+0.000
and
Token and
Feature activation+0.000
physician
Token physician
Feature activation+0.000
-
Token-
Feature activation+0.000
ass
Tokenass
Feature activation+0.000
essed
Tokenessed
Feature activation+0.000
health
Token health
Feature activation+0.000
status
Token status
Feature activation+0.000
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
Yes
TokenYes
Feature activation+0.000
,
Token,
Feature activation+0.000
really
Token really
Feature activation+0.000
.)
Token.)
Feature activation+0.000
https
Token https
Feature activation+0.000
://
Token://
Feature activation+0.000
t
Tokent
Feature activation+0.000
.
Token.
Feature activation+0.000
co
Tokenco
Feature activation+0.000
/
Token/
Feature activation+0.000
L
TokenL
Feature activation+0.000
the
Token the
Feature activation+0.000
last
Token last
Feature activation+0.000
10
Token 10
Feature activation+0.000
years
Token years
Feature activation+0.000
,
Token,
Feature activation+0.000
an
Token an
Feature activation+0.000
average
Token average
Feature activation+0.000
of
Token of
Feature activation+0.000
90
Token 90
Feature activation+0.000
games
Token games
Feature activation+0.000
.
Token.
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 7 in H3.9: (feature 20837)

TOP ACTIVATIONS
MAX = 4.759

damp
Token damp
Feature activation+0.734
ened
Tokenened
Feature activation+0.000
pace
Token pace
Feature activation+2.890
of
Token of
Feature activation+0.000
inflation
Token inflation
Feature activation+0.000
,
Token,
Feature activation+4.759
held
Token held
Feature activation+0.080
down
Token down
Feature activation+0.000
by
Token by
Feature activation+0.000
a
Token a
Feature activation+0.000
strong
Token strong
Feature activation+0.000
Cruz
Token Cruz
Feature activation+0.000
and
Token and
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
right
Token right
Feature activation+0.375
now
Token now
Feature activation+0.000
,
Token,
Feature activation+4.302
that
Token that
Feature activation+2.479
voter
Token voter
Feature activation+0.000
might
Token might
Feature activation+0.000
be
Token be
Feature activation+0.000
so
Token so
Feature activation+0.000
the
Token the
Feature activation+0.760
economic
Token economic
Feature activation+0.000
powerhouse
Token powerhouse
Feature activation+0.000
of
Token of
Feature activation+0.000
Europe
Token Europe
Feature activation+0.000
,
Token,
Feature activation+4.287
nobody
Token nobody
Feature activation+0.948
sees
Token sees
Feature activation+0.000
the
Token the
Feature activation+0.000
trouble
Token trouble
Feature activation+0.000
brewing
Token brewing
Feature activation+0.000
lab
Token lab
Feature activation+0.000
E
Token E
Feature activation+0.000
MC
TokenMC
Feature activation+0.000
testing
Token testing
Feature activation+0.000
results
Token results
Feature activation+0.000
,
Token,
Feature activation+4.173
the
Token the
Feature activation+2.859
large
Token large
Feature activation+0.000
testing
Token testing
Feature activation+0.000
price
Token price
Feature activation+0.000
tag
Token tag
Feature activation+0.000
demonstrators
Token demonstrators
Feature activation+0.000
are
Token are
Feature activation+0.000
expected
Token expected
Feature activation+0.000
to
Token to
Feature activation+0.000
attend
Token attend
Feature activation+0.000
,
Token,
Feature activation+3.884
Bild
Token Bild
Feature activation+0.000
er
Tokener
Feature activation+0.000
berg
Tokenberg
Feature activation+0.000
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
20
Token 20
Feature activation+0.000
th
Tokenth
Feature activation+0.000
century
Token century
Feature activation+0.843
,
Token,
Feature activation+3.861
I
Token I
Feature activation+2.244
flat
Token flat
Feature activation+0.000
-
Token-
Feature activation+0.000
out
Tokenout
Feature activation+0.000
don
Token don
Feature activation+0.000
the
Token the
Feature activation+0.978
gravity
Token gravity
Feature activation+2.233
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.085
situation
Token situation
Feature activation+1.629
,
Token,
Feature activation+3.822
they
Token they
Feature activation+1.863
will
Token will
Feature activation+0.000
need
Token need
Feature activation+0.000
a
Token a
Feature activation+0.000
little
Token little
Feature activation+0.000
And
Token And
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+1.237
economic
Token economic
Feature activation+2.274
backdrop
Token backdrop
Feature activation+2.945
-
Token -
Feature activation+3.749
hopes
Token hopes
Feature activation+1.746
of
Token of
Feature activation+0.000
faster
Token faster
Feature activation+0.000
growth
Token growth
Feature activation+0.000
,
Token,
Feature activation+0.675
Internet
Token Internet
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
United
Token United
Feature activation+0.000
States
Token States
Feature activation+0.000
,
Token,
Feature activation+3.748
even
Token even
Feature activation+0.000
if
Token if
Feature activation+0.000
Russian
Token Russian
Feature activation+0.000
submarines
Token submarines
Feature activation+0.000
were
Token were
Feature activation+0.000
of
Token of
Feature activation+0.000
purpose
Token purpose
Feature activation+0.000
-
Token-
Feature activation+0.000
built
Tokenbuilt
Feature activation+0.000
cities
Token cities
Feature activation+0.000
,
Token,
Feature activation+3.741
raised
Token raised
Feature activation+0.000
up
Token up
Feature activation+0.000
from
Token from
Feature activation+0.000
nothing
Token nothing
Feature activation+0.000
to
Token to
Feature activation+0.000
evidence
Token evidence
Feature activation+0.000
of
Token of
Feature activation+0.000
any
Token any
Feature activation+0.000
concrete
Token concrete
Feature activation+0.020
problem
Token problem
Feature activation+0.055
,
Token,
Feature activation+3.618
any
Token any
Feature activation+0.000
harm
Token harm
Feature activation+0.000
seek
Token seek
Feature activation+0.000
to
Token to
Feature activation+0.000
remedy
Token remedy
Feature activation+0.000
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
in
Token in
Feature activation+0.000
Southern
Token Southern
Feature activation+0.000
Idlib
Token Idlib
Feature activation+0.000
,
Token,
Feature activation+3.592
specially
Token specially
Feature activation+0.809
Jab
Token Jab
Feature activation+0.000
al
Tokenal
Feature activation+0.000
al
Token al
Feature activation+0.000
-
Token-
Feature activation+0.000
damaged
Token damaged
Feature activation+0.000
by
Token by
Feature activation+0.000
this
Token this
Feature activation+0.000
recent
Token recent
Feature activation+0.000
FTA
Token FTA
Feature activation+0.000
,
Token,
Feature activation+3.588
the
Token the
Feature activation+1.684
fears
Token fears
Feature activation+0.000
in
Token in
Feature activation+0.000
Japan
Token Japan
Feature activation+0.000
that
Token that
Feature activation+0.000
-
Token-
Feature activation+0.000
bill
Tokenbill
Feature activation+0.000
ions
Tokenions
Feature activation+0.211
in
Token in
Feature activation+0.000
profits
Token profits
Feature activation+0.000
,
Token,
Feature activation+3.469
their
Token their
Feature activation+2.965
ability
Token ability
Feature activation+3.274
to
Token to
Feature activation+0.000
gou
Token gou
Feature activation+0.000
ge
Tokenge
Feature activation+0.000
fraud
Token fraud
Feature activation+0.000
and
Token and
Feature activation+0.000
corruption
Token corruption
Feature activation+0.000
scandals
Token scandals
Feature activation+0.000
,
Token,
Feature activation+2.353
you
Token you
Feature activation+3.446
can
Token can
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
blame
Token blame
Feature activation+0.000
these
Token these
Feature activation+0.000
asymm
Token asymm
Feature activation+0.084
etrical
Tokenetrical
Feature activation+0.034
interests
Token interests
Feature activation+0.620
,
Token,
Feature activation+3.243
the
Token the
Feature activation+3.394
Clinton
Token Clinton
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
would
Token would
Feature activation+0.000
readily
Token readily
Feature activation+0.000
begin
Token begin
Feature activation+0.000
prior
Token prior
Feature activation+0.963
releases
Token releases
Feature activation+0.000
have
Token have
Feature activation+0.081
different
Token different
Feature activation+0.000
vulnerabilities
Token vulnerabilities
Feature activation+0.000
,
Token,
Feature activation+3.392
rever
Token rever
Feature activation+0.000
ting
Tokenting
Feature activation+0.000
to
Token to
Feature activation+0.000
an
Token an
Feature activation+0.000
older
Token older
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Given
TokenGiven
Feature activation+0.000
a
Token a
Feature activation+0.526
sturdy
Token sturdy
Feature activation+0.000
bicycle
Token bicycle
Feature activation+0.000
,
Token,
Feature activation+3.327
however
Token however
Feature activation+0.000
,
Token,
Feature activation+0.740
many
Token many
Feature activation+0.415
of
Token of
Feature activation+0.000
these
Token these
Feature activation+0.000
both
Token both
Feature activation+0.000
St
Token St
Feature activation+0.000
ifle
Tokenifle
Feature activation+0.000
and
Token and
Feature activation+0.000
Wasteland
Token Wasteland
Feature activation+0.000
,
Token,
Feature activation+3.290
I
Token I
Feature activation+1.760
like
Token like
Feature activation+0.000
keeping
Token keeping
Feature activation+0.000
double
Token double
Feature activation+0.000
D
Token D
Feature activation+0.000
ions
Tokenions
Feature activation+0.211
in
Token in
Feature activation+0.000
profits
Token profits
Feature activation+0.000
,
Token,
Feature activation+3.469
their
Token their
Feature activation+2.965
ability
Token ability
Feature activation+3.274
to
Token to
Feature activation+0.000
gou
Token gou
Feature activation+0.000
ge
Tokenge
Feature activation+0.000
fees
Token fees
Feature activation+0.000
and
Token and
Feature activation+0.000

Top DFA by src position
MAX = 5.492

developments
Token developments
Feature activation-0.002
Top resid features:
."
Token."
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.011
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
But
TokenBut
Feature activation+0.078
Top resid features:
given
Token given
Feature activation+5.492
Top resid features:
the
Token the
Feature activation+0.121
Top resid features:
damp
Token damp
Feature activation+0.074
Top resid features:
ened
Tokenened
Feature activation+0.055
Top resid features:
pace
Token pace
Feature activation+0.082
Top resid features:
of
Token of
Feature activation+0.133
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.291
Top resid features:
or
Token or
Feature activation-0.002
Top resid features:
Ted
Token Ted
Feature activation+0.012
Top resid features:
Cruz
Token Cruz
Feature activation+0.022
Top resid features:
.
Token.
Feature activation+0.064
Top resid features:
Given
Token Given
Feature activation+4.879
Top resid features:
the
Token the
Feature activation+0.093
Top resid features:
choice
Token choice
Feature activation+0.123
Top resid features:
between
Token between
Feature activation+0.126
Top resid features:
Cruz
Token Cruz
Feature activation+0.089
Top resid features:
and
Token and
Feature activation+0.168
Top resid features:
Ċ
TokenĊ
Feature activation-0.012
Top resid features:
Ċ
TokenĊ
Feature activation-0.008
Top resid features:
Of
TokenOf
Feature activation+0.017
Top resid features:
course
Token course
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.027
Top resid features:
given
Token given
Feature activation+5.145
Top resid features:
that
Token that
Feature activation+0.249
Top resid features:
Germany
Token Germany
Feature activation+0.030
Top resid features:
is
Token is
Feature activation+0.015
Top resid features:
the
Token the
Feature activation+0.040
Top resid features:
economic
Token economic
Feature activation+0.021
Top resid features:
government
Token government
Feature activation+0.000
Top resid features:
employees
Token employees
Feature activation-0.004
Top resid features:
busy
Token busy
Feature activation-0.009
Top resid features:
Ċ
TokenĊ
Feature activation+0.064
Top resid features:
Ċ
TokenĊ
Feature activation+0.077
Top resid features:
Given
TokenGiven
Feature activation+4.969
Top resid features:
the
Token the
Feature activation+0.170
Top resid features:
well
Token well
Feature activation+0.016
Top resid features:
documented
Token documented
Feature activation+0.110
Top resid features:
variability
Token variability
Feature activation+0.117
Top resid features:
of
Token of
Feature activation+0.042
Top resid features:
few
Token few
Feature activation+0.000
Top resid features:
days
Token days
Feature activation-0.002
Top resid features:
.
Token.
Feature activation-0.010
Top resid features:
Ċ
TokenĊ
Feature activation-0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Given
TokenGiven
Feature activation+4.827
Top resid features:
the
Token the
Feature activation+0.087
Top resid features:
fact
Token fact
Feature activation+0.036
Top resid features:
that
Token that
Feature activation+0.179
Top resid features:
a
Token a
Feature activation+0.015
Top resid features:
record
Token record
Feature activation+0.038
Top resid features:
are
Token are
Feature activation-0.016
Top resid features:
bad
Token bad
Feature activation-0.002
Top resid features:
at
Token at
Feature activation-0.002
Top resid features:
this
Token this
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.071
Top resid features:
Given
Token Given
Feature activation+4.857
Top resid features:
the
Token the
Feature activation+0.124
Top resid features:
history
Token history
Feature activation+0.117
Top resid features:
of
Token of
Feature activation+0.088
Top resid features:
the
Token the
Feature activation+0.048
Top resid features:
20
Token 20
Feature activation-0.002
Top resid features:
is
Token is
Feature activation-0.005
Top resid features:
almost
Token almost
Feature activation-0.005
Top resid features:
finished
Token finished
Feature activation+0.030
Top resid features:
,
Token,
Feature activation+0.026
Top resid features:
but
Token but
Feature activation+0.141
Top resid features:
given
Token given
Feature activation+4.649
Top resid features:
the
Token the
Feature activation+0.119
Top resid features:
gravity
Token gravity
Feature activation+0.018
Top resid features:
of
Token of
Feature activation+0.113
Top resid features:
the
Token the
Feature activation+0.073
Top resid features:
situation
Token situation
Feature activation+0.147
Top resid features:
in
Token in
Feature activation+0.001
Top resid features:
August
Token August
Feature activation-0.001
Top resid features:
2013
Token 2013
Feature activation-0.008
Top resid features:
.
Token.
Feature activation+0.051
Top resid features:
And
Token And
Feature activation+0.035
Top resid features:
given
Token given
Feature activation+4.696
Top resid features:
the
Token the
Feature activation+0.204
Top resid features:
economic
Token economic
Feature activation+0.045
Top resid features:
backdrop
Token backdrop
Feature activation+0.107
Top resid features:
-
Token -
Feature activation+0.403
Top resid features:
hopes
Token hopes
Feature activation+0.000
Top resid features:
-
Token-
Feature activation-0.001
Top resid features:
called
Tokencalled
Feature activation+0.002
Top resid features:
Islamic
Token Islamic
Feature activation-0.000
Top resid features:
State
Token State
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Given
Token Given
Feature activation+4.672
Top resid features:
that
Token that
Feature activation+0.286
Top resid features:
it
Token it
Feature activation+0.021
Top resid features:
is
Token is
Feature activation+0.007
Top resid features:
impossible
Token impossible
Feature activation+0.029
Top resid features:
to
Token to
Feature activation+0.018
Top resid features:
's
Token's
Feature activation+0.003
Top resid features:
characteristics
Token characteristics
Feature activation+0.001
Top resid features:
."
Token."
Feature activation-0.006
Top resid features:
Ċ
TokenĊ
Feature activation-0.015
Top resid features:
Ċ
TokenĊ
Feature activation-0.006
Top resid features:
Given
TokenGiven
Feature activation+4.676
Top resid features:
the
Token the
Feature activation+0.194
Top resid features:
history
Token history
Feature activation+0.113
Top resid features:
of
Token of
Feature activation+0.074
Top resid features:
purpose
Token purpose
Feature activation+0.035
Top resid features:
-
Token-
Feature activation+0.029
Top resid features:
Islamic
Token Islamic
Feature activation+0.003
Top resid features:
law
Token law
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
âĢ
Token âĢ
Feature activation+0.046
Top resid features:
ľ
Tokenľ
Feature activation+0.026
Top resid features:
Given
TokenGiven
Feature activation+4.514
Top resid features:
the
Token the
Feature activation+0.167
Top resid features:
lack
Token lack
Feature activation+0.034
Top resid features:
of
Token of
Feature activation+0.029
Top resid features:
evidence
Token evidence
Feature activation+0.022
Top resid features:
of
Token of
Feature activation+0.023
Top resid features:
Ċ
TokenĊ
Feature activation-0.012
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
Sources
TokenSources
Feature activation+0.094
Top resid features:
said
Token said
Feature activation+0.056
Top resid features:
that
Token that
Feature activation+0.135
Top resid features:
given
Token given
Feature activation+4.616
Top resid features:
the
Token the
Feature activation+0.072
Top resid features:
military
Token military
Feature activation+0.014
Top resid features:
presence
Token presence
Feature activation+0.016
Top resid features:
of
Token of
Feature activation+0.035
Top resid features:
Ah
Token Ah
Feature activation+0.002
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
Korea
Token Korea
Feature activation-0.001
Top resid features:
FTA
Token FTA
Feature activation+0.001
Top resid features:
.
Token.
Feature activation-0.009
Top resid features:
Given
Token Given
Feature activation+4.368
Top resid features:
the
Token the
Feature activation+0.051
Top resid features:
fact
Token fact
Feature activation+0.034
Top resid features:
that
Token that
Feature activation+0.217
Top resid features:
the
Token the
Feature activation+0.031
Top resid features:
US
Token US
Feature activation-0.008
Top resid features:
The
TokenThe
Feature activation+0.006
Top resid features:
public
Token public
Feature activation-0.010
Top resid features:
hates
Token hates
Feature activation-0.046
Top resid features:
banks
Token banks
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.096
Top resid features:
Given
Token Given
Feature activation+4.539
Top resid features:
their
Token their
Feature activation+0.160
Top resid features:
mega
Token mega
Feature activation+0.049
Top resid features:
-
Token-
Feature activation+0.005
Top resid features:
bill
Tokenbill
Feature activation+0.027
Top resid features:
ions
Tokenions
Feature activation+0.027
Top resid features:
The
TokenThe
Feature activation+0.007
Top resid features:
public
Token public
Feature activation-0.005
Top resid features:
hates
Token hates
Feature activation-0.035
Top resid features:
banks
Token banks
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.022
Top resid features:
Given
Token Given
Feature activation+4.700
Top resid features:
their
Token their
Feature activation+0.059
Top resid features:
mega
Token mega
Feature activation+0.008
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
bill
Tokenbill
Feature activation+0.002
Top resid features:
ions
Tokenions
Feature activation+0.001
Top resid features:
it
Token it
Feature activation-0.007
Top resid features:
is
Token is
Feature activation-0.005
Top resid features:
incredibly
Token incredibly
Feature activation+0.016
Top resid features:
disturbing
Token disturbing
Feature activation+0.008
Top resid features:
that
Token that
Feature activation+0.180
Top resid features:
given
Token given
Feature activation+4.293
Top resid features:
these
Token these
Feature activation+0.078
Top resid features:
asymm
Token asymm
Feature activation+0.031
Top resid features:
etrical
Tokenetrical
Feature activation+0.042
Top resid features:
interests
Token interests
Feature activation+0.050
Top resid features:
,
Token,
Feature activation+0.336
Top resid features:
that
Token that
Feature activation+0.014
Top resid features:
is
Token is
Feature activation+0.002
Top resid features:
affected
Token affected
Feature activation-0.009
Top resid features:
but
Token but
Feature activation-0.034
Top resid features:
,
Token,
Feature activation+0.049
Top resid features:
given
Token given
Feature activation+4.063
Top resid features:
that
Token that
Feature activation+0.204
Top resid features:
prior
Token prior
Feature activation+0.008
Top resid features:
releases
Token releases
Feature activation-0.005
Top resid features:
have
Token have
Feature activation+0.106
Top resid features:
different
Token different
Feature activation+0.247
Top resid features:
walking
Token walking
Feature activation-0.005
Top resid features:
alone
Token alone
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.025
Top resid features:
Ċ
TokenĊ
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.024
Top resid features:
Given
TokenGiven
Feature activation+4.308
Top resid features:
a
Token a
Feature activation+0.150
Top resid features:
sturdy
Token sturdy
Feature activation+0.145
Top resid features:
bicycle
Token bicycle
Feature activation+0.105
Top resid features:
,
Token,
Feature activation+0.295
Top resid features:
however
Token however
Feature activation+0.000
Top resid features:
pell
Tokenpell
Feature activation-0.005
Top resid features:
anytime
Token anytime
Feature activation-0.011
Top resid features:
soon
Token soon
Feature activation-0.007
Top resid features:
,
Token,
Feature activation+0.020
Top resid features:
and
Token and
Feature activation+0.143
Top resid features:
given
Token given
Feature activation+3.976
Top resid features:
the
Token the
Feature activation+0.103
Top resid features:
presence
Token presence
Feature activation+0.011
Top resid features:
of
Token of
Feature activation+0.137
Top resid features:
both
Token both
Feature activation+0.111
Top resid features:
St
Token St
Feature activation+0.033
Top resid features:
The
TokenThe
Feature activation-0.001
Top resid features:
public
Token public
Feature activation-0.008
Top resid features:
hates
Token hates
Feature activation-0.053
Top resid features:
banks
Token banks
Feature activation+0.004
Top resid features:
.
Token.
Feature activation+0.055
Top resid features:
Given
Token Given
Feature activation+4.424
Top resid features:
their
Token their
Feature activation+0.053
Top resid features:
mega
Token mega
Feature activation+0.040
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
bill
Tokenbill
Feature activation+0.028
Top resid features:
ions
Tokenions
Feature activation+0.003
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.06

Head 2: 0.04

Head 3: 0.02

Head 4: 0.02

Head 5: 0.09

Head 6: 0.05

Head 7: 0.03

Head 8: 0.04

Head 9: 0.52

Head 10: 0.04

Head 11: 0.06

Positive logits

capacity1.51

bount1.36

dose1.34

ellation1.34

phasis1.32

context1.32

chances1.32

`,1.30

likelihood1.30

pressures1.30

osity1.29

probable1.27

myra1.24

coma1.23

usage1.23

manship1.23

qualifications1.22

clout1.22

ularity1.22

anoia1.21

Negative logits

ITED-1.57

ARY-1.56

OPS-1.53

UTE-1.53

OO-1.44

largeDownload-1.39

Org-1.37

RET-1.37

-1.33

Beaver-1.31

ogle-1.31

orse-1.30

URE-1.30

TABLE-1.29

Js-1.25

AN-1.23

hog-1.23

OWS-1.22

OL-1.22

OP-1.22

INTERVAL 4.283 - 4.759
CONTAINS 0.000%

Cruz
Token Cruz
Feature activation+0.000
and
Token and
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
right
Token right
Feature activation+0.375
now
Token now
Feature activation+0.000
,
Token,
Feature activation+4.302
that
Token that
Feature activation+2.479
voter
Token voter
Feature activation+0.000
might
Token might
Feature activation+0.000
be
Token be
Feature activation+0.000
so
Token so
Feature activation+0.000
the
Token the
Feature activation+0.760
economic
Token economic
Feature activation+0.000
powerhouse
Token powerhouse
Feature activation+0.000
of
Token of
Feature activation+0.000
Europe
Token Europe
Feature activation+0.000
,
Token,
Feature activation+4.287
nobody
Token nobody
Feature activation+0.948
sees
Token sees
Feature activation+0.000
the
Token the
Feature activation+0.000
trouble
Token trouble
Feature activation+0.000
brewing
Token brewing
Feature activation+0.000
damp
Token damp
Feature activation+0.734
ened
Tokenened
Feature activation+0.000
pace
Token pace
Feature activation+2.890
of
Token of
Feature activation+0.000
inflation
Token inflation
Feature activation+0.000
,
Token,
Feature activation+4.759
held
Token held
Feature activation+0.080
down
Token down
Feature activation+0.000
by
Token by
Feature activation+0.000
a
Token a
Feature activation+0.000
strong
Token strong
Feature activation+0.000

INTERVAL 3.807 - 4.283
CONTAINS 0.001%

demonstrators
Token demonstrators
Feature activation+0.000
are
Token are
Feature activation+0.000
expected
Token expected
Feature activation+0.000
to
Token to
Feature activation+0.000
attend
Token attend
Feature activation+0.000
,
Token,
Feature activation+3.884
Bild
Token Bild
Feature activation+0.000
er
Tokener
Feature activation+0.000
berg
Tokenberg
Feature activation+0.000
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
20
Token 20
Feature activation+0.000
th
Tokenth
Feature activation+0.000
century
Token century
Feature activation+0.843
,
Token,
Feature activation+3.861
I
Token I
Feature activation+2.244
flat
Token flat
Feature activation+0.000
-
Token-
Feature activation+0.000
out
Tokenout
Feature activation+0.000
don
Token don
Feature activation+0.000
lab
Token lab
Feature activation+0.000
E
Token E
Feature activation+0.000
MC
TokenMC
Feature activation+0.000
testing
Token testing
Feature activation+0.000
results
Token results
Feature activation+0.000
,
Token,
Feature activation+4.173
the
Token the
Feature activation+2.859
large
Token large
Feature activation+0.000
testing
Token testing
Feature activation+0.000
price
Token price
Feature activation+0.000
tag
Token tag
Feature activation+0.000
the
Token the
Feature activation+0.978
gravity
Token gravity
Feature activation+2.233
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.085
situation
Token situation
Feature activation+1.629
,
Token,
Feature activation+3.822
they
Token they
Feature activation+1.863
will
Token will
Feature activation+0.000
need
Token need
Feature activation+0.000
a
Token a
Feature activation+0.000
little
Token little
Feature activation+0.000

INTERVAL 3.332 - 3.807
CONTAINS 0.001%

fraud
Token fraud
Feature activation+0.000
and
Token and
Feature activation+0.000
corruption
Token corruption
Feature activation+0.000
scandals
Token scandals
Feature activation+0.000
,
Token,
Feature activation+2.353
you
Token you
Feature activation+3.446
can
Token can
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
blame
Token blame
Feature activation+0.000
these
Token these
Feature activation+0.000
asymm
Token asymm
Feature activation+0.084
etrical
Tokenetrical
Feature activation+0.034
interests
Token interests
Feature activation+0.620
,
Token,
Feature activation+3.243
the
Token the
Feature activation+3.394
Clinton
Token Clinton
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
would
Token would
Feature activation+0.000
readily
Token readily
Feature activation+0.000
begin
Token begin
Feature activation+0.000
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
in
Token in
Feature activation+0.000
Southern
Token Southern
Feature activation+0.000
Idlib
Token Idlib
Feature activation+0.000
,
Token,
Feature activation+3.592
specially
Token specially
Feature activation+0.809
Jab
Token Jab
Feature activation+0.000
al
Tokenal
Feature activation+0.000
al
Token al
Feature activation+0.000
-
Token-
Feature activation+0.000
evidence
Token evidence
Feature activation+0.000
of
Token of
Feature activation+0.000
any
Token any
Feature activation+0.000
concrete
Token concrete
Feature activation+0.020
problem
Token problem
Feature activation+0.055
,
Token,
Feature activation+3.618
any
Token any
Feature activation+0.000
harm
Token harm
Feature activation+0.000
seek
Token seek
Feature activation+0.000
to
Token to
Feature activation+0.000
remedy
Token remedy
Feature activation+0.000
prior
Token prior
Feature activation+0.963
releases
Token releases
Feature activation+0.000
have
Token have
Feature activation+0.081
different
Token different
Feature activation+0.000
vulnerabilities
Token vulnerabilities
Feature activation+0.000
,
Token,
Feature activation+3.392
rever
Token rever
Feature activation+0.000
ting
Tokenting
Feature activation+0.000
to
Token to
Feature activation+0.000
an
Token an
Feature activation+0.000
older
Token older
Feature activation+0.000

INTERVAL 2.856 - 3.332
CONTAINS 0.001%

given
Token given
Feature activation+0.000
these
Token these
Feature activation+0.000
asymm
Token asymm
Feature activation+0.084
etrical
Tokenetrical
Feature activation+0.034
interests
Token interests
Feature activation+0.620
,
Token,
Feature activation+3.243
the
Token the
Feature activation+3.394
Clinton
Token Clinton
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
would
Token would
Feature activation+0.000
readily
Token readily
Feature activation+0.000
But
TokenBut
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+1.103
damp
Token damp
Feature activation+0.734
ened
Tokenened
Feature activation+0.000
pace
Token pace
Feature activation+2.890
of
Token of
Feature activation+0.000
inflation
Token inflation
Feature activation+0.000
,
Token,
Feature activation+4.759
held
Token held
Feature activation+0.080
down
Token down
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Given
TokenGiven
Feature activation+0.000
a
Token a
Feature activation+0.526
sturdy
Token sturdy
Feature activation+0.000
bicycle
Token bicycle
Feature activation+0.000
,
Token,
Feature activation+3.327
however
Token however
Feature activation+0.000
,
Token,
Feature activation+0.740
many
Token many
Feature activation+0.415
of
Token of
Feature activation+0.000
these
Token these
Feature activation+0.000
how
Token how
Feature activation+1.032
nasty
Token nasty
Feature activation+0.000
that
Token that
Feature activation+0.798
dissolution
Token dissolution
Feature activation+0.000
was
Token was
Feature activation+0.000
,
Token,
Feature activation+3.064
there
Token there
Feature activation+1.058
was
Token was
Feature activation+0.000
likely
Token likely
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
both
Token both
Feature activation+0.000
St
Token St
Feature activation+0.000
ifle
Tokenifle
Feature activation+0.000
and
Token and
Feature activation+0.000
Wasteland
Token Wasteland
Feature activation+0.000
,
Token,
Feature activation+3.290
I
Token I
Feature activation+1.760
like
Token like
Feature activation+0.000
keeping
Token keeping
Feature activation+0.000
double
Token double
Feature activation+0.000
D
Token D
Feature activation+0.000

INTERVAL 2.380 - 2.856
CONTAINS 0.001%

s
Tokens
Feature activation+0.000
constitutional
Token constitutional
Feature activation+0.000
rights
Token rights
Feature activation+0.232
were
Token were
Feature activation+0.000
violated
Token violated
Feature activation+0.000
,
Token,
Feature activation+2.477
and
Token and
Feature activation+0.315
applying
Token applying
Feature activation+0.000
the
Token the
Feature activation+0.000
rule
Token rule
Feature activation+0.000
that
Token that
Feature activation+0.000
at
Token at
Feature activation+0.000
this
Token this
Feature activation+0.000
.
Token.
Feature activation+0.000
Given
Token Given
Feature activation+0.000
the
Token the
Feature activation+0.466
history
Token history
Feature activation+2.832
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
20
Token 20
Feature activation+0.000
th
Tokenth
Feature activation+0.000
century
Token century
Feature activation+0.843
and
Token and
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
right
Token right
Feature activation+0.375
now
Token now
Feature activation+0.000
,
Token,
Feature activation+4.302
that
Token that
Feature activation+2.479
voter
Token voter
Feature activation+0.000
might
Token might
Feature activation+0.000
be
Token be
Feature activation+0.000
so
Token so
Feature activation+0.000
opposed
Token opposed
Feature activation+0.000
around
Token around
Feature activation+0.000
407
Token 407
Feature activation+0.000
parts
Token parts
Feature activation+0.000
per
Token per
Feature activation+0.000
million
Token million
Feature activation+0.000
,
Token,
Feature activation+2.698
given
Token given
Feature activation+0.000
that
Token that
Feature activation+0.190
these
Token these
Feature activation+0.012
levels
Token levels
Feature activation+0.000
are
Token are
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Given
TokenGiven
Feature activation+0.000
the
Token the
Feature activation+0.517
history
Token history
Feature activation+2.777
of
Token of
Feature activation+0.000
purpose
Token purpose
Feature activation+0.000
-
Token-
Feature activation+0.000
built
Tokenbuilt
Feature activation+0.000
cities
Token cities
Feature activation+0.000

INTERVAL 1.904 - 2.380
CONTAINS 0.001%

.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Given
TokenGiven
Feature activation+0.000
the
Token the
Feature activation+0.702
fact
Token fact
Feature activation+2.091
that
Token that
Feature activation+0.000
a
Token a
Feature activation+0.000
record
Token record
Feature activation+0.720
number
Token number
Feature activation+0.000
of
Token of
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+1.237
economic
Token economic
Feature activation+2.274
backdrop
Token backdrop
Feature activation+2.945
-
Token -
Feature activation+3.749
hopes
Token hopes
Feature activation+1.746
of
Token of
Feature activation+0.000
faster
Token faster
Feature activation+0.000
up
Token up
Feature activation+0.000
to
Token to
Feature activation+0.000
allow
Token allow
Feature activation+0.000
multiple
Token multiple
Feature activation+0.000
votes
Token votes
Feature activation+0.000
,
Token,
Feature activation+2.374
and
Token and
Feature activation+0.000
even
Token even
Feature activation+0.000
given
Token given
Feature activation+0.000
my
Token my
Feature activation+0.241
decidedly
Token decidedly
Feature activation+0.000
the
Token the
Feature activation+0.000
20
Token 20
Feature activation+0.000
th
Tokenth
Feature activation+0.000
century
Token century
Feature activation+0.843
,
Token,
Feature activation+3.861
I
Token I
Feature activation+2.244
flat
Token flat
Feature activation+0.000
-
Token-
Feature activation+0.000
out
Tokenout
Feature activation+0.000
don
Token don
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
conservative
Token conservative
Feature activation+0.000
bent
Token bent
Feature activation+0.000
and
Token and
Feature activation+0.000
its
Token its
Feature activation+1.036
recent
Token recent
Feature activation+0.000
history
Token history
Feature activation+2.068
of
Token of
Feature activation+0.000
supporting
Token supporting
Feature activation+0.000
Republican
Token Republican
Feature activation+0.000
candidates
Token candidates
Feature activation+0.000
for
Token for
Feature activation+0.000

INTERVAL 1.428 - 1.904
CONTAINS 0.002%

Considering
TokenConsidering
Feature activation+0.000
whose
Token whose
Feature activation+0.000
wife
Token wife
Feature activation+0.000
he
Token he
Feature activation+0.496
punched
Token punched
Feature activation+0.000
,
Token,
Feature activation+1.672
I
Token I
Feature activation+0.755
'd
Token'd
Feature activation+0.000
say
Token say
Feature activation+0.000
Mr
Token Mr
Feature activation+0.000
.
Token.
Feature activation+0.000
s
Tokens
Feature activation+0.000
nearly
Token nearly
Feature activation+0.000
impossible
Token impossible
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+0.000
speed
Token speed
Feature activation+1.464
that
Token that
Feature activation+0.000
TV
Token TV
Feature activation+0.000
cr
Token cr
Feature activation+0.000
anks
Tokenanks
Feature activation+0.000
out
Token out
Feature activation+0.000
gravity
Token gravity
Feature activation+2.233
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.085
situation
Token situation
Feature activation+1.629
,
Token,
Feature activation+3.822
they
Token they
Feature activation+1.863
will
Token will
Feature activation+0.000
need
Token need
Feature activation+0.000
a
Token a
Feature activation+0.000
little
Token little
Feature activation+0.000
extra
Token extra
Feature activation+0.000
an
Token an
Feature activation+0.000
underestimate
Token underestimate
Feature activation+0.000
,
Token,
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+0.508
difficulty
Token difficulty
Feature activation+1.640
of
Token of
Feature activation+0.000
accurately
Token accurately
Feature activation+0.000
recording
Token recording
Feature activation+0.000
marine
Token marine
Feature activation+0.000
ext
Token ext
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
,
Token,
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+0.511
scarcity
Token scarcity
Feature activation+1.686
of
Token of
Feature activation+0.000
housing
Token housing
Feature activation+0.000
,
Token,
Feature activation+2.414
they
Token they
Feature activation+0.319
had
Token had
Feature activation+0.000

INTERVAL 0.952 - 1.428
CONTAINS 0.003%

ruined
Token ruined
Feature activation+0.000
the
Token the
Feature activation+0.000
banner
Token banner
Feature activation+0.000
ad
Token ad
Feature activation+0.000
industry
Token industry
Feature activation+0.000
,
Token,
Feature activation+1.033
one
Token one
Feature activation+0.000
must
Token must
Feature activation+0.000
also
Token also
Feature activation+0.000
wonder
Token wonder
Feature activation+0.000
when
Token when
Feature activation+0.000
statements
Token statements
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
given
Token given
Feature activation+0.000
to
Token to
Feature activation+0.327
the
Token the
Feature activation+1.112
words
Token words
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
immigrants
Token immigrants
Feature activation+0.000
were
Token were
Feature activation+0.000
nasty
Token nasty
Feature activation+0.000
that
Token that
Feature activation+0.798
dissolution
Token dissolution
Feature activation+0.000
was
Token was
Feature activation+0.000
,
Token,
Feature activation+3.064
there
Token there
Feature activation+1.058
was
Token was
Feature activation+0.000
likely
Token likely
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
no
Token no
Feature activation+0.000
Given
TokenGiven
Feature activation+0.000
soy
Token soy
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
central
Token central
Feature activation+1.012
ity
Tokenity
Feature activation+1.376
to
Token to
Feature activation+0.000
our
Token our
Feature activation+0.000
food
Token food
Feature activation+0.000
and
Token and
Feature activation+0.000
immensely
Token immensely
Feature activation+0.000
rich
Token rich
Feature activation+0.000
,
Token,
Feature activation+0.000
particularly
Token particularly
Feature activation+0.000
given
Token given
Feature activation+0.000
how
Token how
Feature activation+0.962
small
Token small
Feature activation+0.000
the
Token the
Feature activation+0.639
population
Token population
Feature activation+0.000
was
Token was
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 0.476 - 0.952
CONTAINS 0.006%

).
Token).
Feature activation+0.000
There
Token There
Feature activation+0.000
was
Token was
Feature activation+0.000
nothing
Token nothing
Feature activation+0.000
given
Token given
Feature activation+0.000
size
Token size
Feature activation+0.499
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
If
TokenIf
Feature activation+0.000
creatures
Token creatures
Feature activation+0.000
exist
Token exist
Feature activation+0.000
on
Token on
Feature activation+0.000
only
Token only
Feature activation+0.000
grow
Token grow
Feature activation+0.000
,
Token,
Feature activation+0.000
given
Token given
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.727
direct
Token direct
Feature activation+0.160
action
Token action
Feature activation+0.000
is
Token is
Feature activation+0.000
nearly
Token nearly
Feature activation+0.000
a
Token a
Feature activation+0.000
Given
TokenGiven
Feature activation+0.000
the
Token the
Feature activation+0.702
fact
Token fact
Feature activation+2.091
that
Token that
Feature activation+0.000
a
Token a
Feature activation+0.000
record
Token record
Feature activation+0.720
number
Token number
Feature activation+0.000
of
Token of
Feature activation+0.000
demonstrators
Token demonstrators
Feature activation+0.000
are
Token are
Feature activation+0.000
expected
Token expected
Feature activation+0.000
-
Token-
Feature activation+0.000
femin
Tokenfemin
Feature activation+0.000
ist
Tokenist
Feature activation+0.000
views
Token views
Feature activation+0.123
,
Token,
Feature activation+1.155
I
Token I
Feature activation+0.910
only
Token only
Feature activation+0.000
cast
Token cast
Feature activation+0.000
one
Token one
Feature activation+0.000
vote
Token vote
Feature activation+0.000
myself
Token myself
Feature activation+0.000
characteristics
Token characteristics
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Given
TokenGiven
Feature activation+0.000
the
Token the
Feature activation+0.517
history
Token history
Feature activation+2.777
of
Token of
Feature activation+0.000
purpose
Token purpose
Feature activation+0.000
-
Token-
Feature activation+0.000
built
Tokenbuilt
Feature activation+0.000

INTERVAL 0.000 - 0.476
CONTAINS 99.984%

and
Token and
Feature activation+0.000
I
Token I
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
m
Tokenm
Feature activation+0.000
sure
Token sure
Feature activation+0.000
different
Token different
Feature activation+0.000
times
Token times
Feature activation+0.000
of
Token of
Feature activation+0.000
day
Token day
Feature activation+0.000
as
Token as
Feature activation+0.000
,
Token,
Feature activation+0.000
especially
Token especially
Feature activation+0.000
when
Token when
Feature activation+0.000
under
Token under
Feature activation+0.000
pressure
Token pressure
Feature activation+0.000
.
Token.
Feature activation+0.000
When
Token When
Feature activation+0.000
being
Token being
Feature activation+0.000
intimidated
Token intimidated
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
I
Token I
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+0.000
reflected
Token reflected
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
lesson
Token lesson
Feature activation+0.000
on
Token on
Feature activation+0.000
pace
Token pace
Feature activation+0.000
that
Token that
Feature activation+0.000
we
Token we
Feature activation+0.000
ought
Token ought
Feature activation+0.000
to
Token to
Feature activation+0.000
share
Token share
Feature activation+0.000
it
Token it
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
public
Token public
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
it
Token it
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
intended
Token intended
Feature activation+0.000
to
Token to
Feature activation+0.000
give
Token give
Feature activation+0.000
some
Token some
Feature activation+0.000
idea
Token idea
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 8 in H3.9: (feature 15430)

TOP ACTIVATIONS
MAX = 3.794

and
Token and
Feature activation+0.000
deficient
Token deficient
Feature activation+0.000
military
Token military
Feature activation+0.164
commissions
Token commissions
Feature activation+0.000
system
Token system
Feature activation+1.946
to
Token to
Feature activation+3.794
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Kevin
TokenKevin
Feature activation+0.000
Space
Token Space
Feature activation+0.000
y
Tokeny
Feature activation+0.000
is
Token is
Feature activation+0.000
its
Tokenits
Feature activation+0.000
the
Token the
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
public
Token public
Feature activation+0.203
restrooms
Token restrooms
Feature activation+3.729
to
Token to
Feature activation+0.000
members
Token members
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
given
Token given
Feature activation+0.000
says
Token says
Feature activation+0.000
the
Token the
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
chemical
Token chemical
Feature activation+2.403
weapons
Token weapons
Feature activation+3.634
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
'
Token '
Feature activation+0.000
red
Tokenred
Feature activation+0.000
face
Token face
Feature activation+0.000
or
Token or
Feature activation+0.000
use
Token use
Feature activation+0.000
his
Token his
Feature activation+0.059
last
Token last
Feature activation+0.000
name
Token name
Feature activation+3.611
because
Token because
Feature activation+0.000
he
Token he
Feature activation+0.000
fears
Token fears
Feature activation+0.000
for
Token for
Feature activation+0.000
his
Token his
Feature activation+0.000
personal
Tokenpersonal
Feature activation+0.656
attack
Token attack
Feature activation+0.156
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
as
Token as
Feature activation+3.419
a
Token a
Feature activation+3.599
rating
Token rating
Feature activation+0.117
category
Token category
Feature activation+0.137
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
use
Token use
Feature activation+0.000
cloud
Token cloud
Feature activation+0.158
based
Token based
Feature activation+0.688
NVIDIA
Token NVIDIA
Feature activation+0.020
GPUs
Token GPUs
Feature activation+1.329
to
Token to
Feature activation+3.593
develop
Token develop
Feature activation+0.136
their
Token their
Feature activation+0.000
innovative
Token innovative
Feature activation+0.000
products
Token products
Feature activation+0.000
and
Token and
Feature activation+0.000
using
Token using
Feature activation+0.000
religious
Token religious
Feature activation+0.000
freedom
Token freedom
Feature activation+0.000
as
Token as
Feature activation+2.036
a
Token a
Feature activation+2.896
pretext
Token pretext
Feature activation+3.556
for
Token for
Feature activation+0.654
discrimination
Token discrimination
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
hoped
Token hoped
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
the
Token the
Feature activation+0.049
watch
Token watch
Feature activation+0.834
to
Token to
Feature activation+3.498
help
Token help
Feature activation+2.093
identify
Token identify
Feature activation+0.500
the
Token the
Feature activation+0.000
body
Token body
Feature activation+0.000
.
Token.
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
this
Token this
Feature activation+0.000
as
Token as
Feature activation+2.781
an
Token an
Feature activation+3.122
excuse
Token excuse
Feature activation+3.484
for
Token for
Feature activation+0.811
confrontation
Token confrontation
Feature activation+0.000
and
Token and
Feature activation+0.000
finger
Token finger
Feature activation+0.000
-
Token-
Feature activation+0.000
olds
Tokenolds
Feature activation+0.000
use
Token use
Feature activation+0.000
six
Token six
Feature activation+0.000
times
Token times
Feature activation+0.600
as
Token as
Feature activation+2.034
much
Token much
Feature activation+3.473
health
Token health
Feature activation+0.000
care
Token care
Feature activation+1.024
as
Token as
Feature activation+1.121
20
Token 20
Feature activation+0.000
-
Token-
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
personal
Tokenpersonal
Feature activation+0.656
attack
Token attack
Feature activation+0.156
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
as
Token as
Feature activation+3.419
a
Token a
Feature activation+3.599
rating
Token rating
Feature activation+0.117
category
Token category
Feature activation+0.137
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
standing
Tokenstanding
Feature activation+0.000
and
Token and
Feature activation+0.000
using
Token using
Feature activation+0.000
her
Token her
Feature activation+0.244
post
Token post
Feature activation+0.228
to
Token to
Feature activation+3.392
bolster
Token bolster
Feature activation+1.563
her
Token her
Feature activation+0.000
own
Token own
Feature activation+0.164
re
Token re
Feature activation+0.000
-
Token-
Feature activation+0.000
hands
Token hands
Feature activation+0.000
and
Token and
Feature activation+0.000
use
Token use
Feature activation+0.000
that
Token that
Feature activation+0.000
money
Token money
Feature activation+2.555
to
Token to
Feature activation+3.376
hire
Token hire
Feature activation+0.000
a
Token a
Feature activation+0.000
snow
Token snow
Feature activation+0.000
pl
Tokenpl
Feature activation+0.000
ow
Tokenow
Feature activation+0.000
the
Token the
Feature activation+0.314
media
Token media
Feature activation+0.316
and
Token and
Feature activation+0.000
fiery
Token fiery
Feature activation+1.183
rhetoric
Token rhetoric
Feature activation+2.656
to
Token to
Feature activation+3.347
tell
Token tell
Feature activation+1.516
his
Token his
Feature activation+0.000
people
Token people
Feature activation+0.000
that
Token that
Feature activation+0.000
there
Token there
Feature activation+0.000
10
Token10
Feature activation+0.000
31
Token31
Feature activation+0.000
exchange
Token exchange
Feature activation+0.184
"
Token"
Feature activation+1.108
rule
Token rule
Feature activation+0.222
to
Token to
Feature activation+3.340
pay
Token pay
Feature activation+0.000
zero
Token zero
Feature activation+0.000
tax
Token tax
Feature activation+0.000
on
Token on
Feature activation+0.000
large
Token large
Feature activation+0.000
he
Token he
Feature activation+0.000
would
Token would
Feature activation+0.000
not
Token not
Feature activation+0.000
use
Token use
Feature activation+0.000
the
Token the
Feature activation+0.584
name
Token name
Feature activation+3.335
of
Token of
Feature activation+1.387
the
Token the
Feature activation+0.953
U
Token U
Feature activation+0.000
mp
Tokenmp
Feature activation+0.000
qua
Tokenqua
Feature activation+0.000
he
Token he
Feature activation+0.000
refuses
Token refuses
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
his
Token his
Feature activation+0.910
name
Token name
Feature activation+3.276
,"
Token,"
Feature activation+0.000
he
Token he
Feature activation+0.000
said
Token said
Feature activation+0.000
,
Token,
Feature activation+0.000
referring
Token referring
Feature activation+0.000
hoping
Token hoping
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
it
Token it
Feature activation+0.550
as
Token as
Feature activation+2.645
a
Token a
Feature activation+3.259
showcase
Token showcase
Feature activation+0.364
for
Token for
Feature activation+0.917
his
Token his
Feature activation+0.301
style
Token style
Feature activation+0.004
and
Token and
Feature activation+0.000
status
Token status
Feature activation+0.354
as
Token as
Feature activation+0.000
an
Token an
Feature activation+2.397
art
Token art
Feature activation+0.088
student
Token student
Feature activation+0.000
to
Token to
Feature activation+3.243
get
Token get
Feature activation+1.360
close
Token close
Feature activation+0.000
to
Token to
Feature activation+0.000
J
Token J
Feature activation+0.000
our
Tokenour
Feature activation+0.000
can
Token can
Feature activation+0.000
"
Token "
Feature activation+0.000
use
Tokenuse
Feature activation+0.000
only
Token only
Feature activation+0.000
the
Token the
Feature activation+0.023
bathroom
Token bathroom
Feature activation+3.230
/
Token/
Feature activation+0.000
lock
Tokenlock
Feature activation+0.693
er
Tokener
Feature activation+0.577
room
Token room
Feature activation+1.341
for
Token for
Feature activation+1.177

Top DFA by src position
MAX = 4.659

âĢ
TokenâĢ
Feature activation+0.002
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
s
Tokens
Feature activation+0.005
Top resid features:
decision
Token decision
Feature activation-0.006
Top resid features:
to
Token to
Feature activation-0.051
Top resid features:
use
Token use
Feature activation+4.399
Top resid features:
the
Token the
Feature activation+0.020
Top resid features:
broken
Token broken
Feature activation+0.014
Top resid features:
and
Token and
Feature activation+0.002
Top resid features:
deficient
Token deficient
Feature activation+0.008
Top resid features:
military
Token military
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation-0.017
Top resid features:
Ċ
TokenĊ
Feature activation-0.020
Top resid features:
Lim
TokenLim
Feature activation-0.011
Top resid features:
its
Tokenits
Feature activation+0.006
Top resid features:
the
Token the
Feature activation-0.008
Top resid features:
use
Token use
Feature activation+4.659
Top resid features:
of
Token of
Feature activation-0.118
Top resid features:
public
Token public
Feature activation-0.079
Top resid features:
restrooms
Token restrooms
Feature activation+0.038
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
members
Token members
Feature activation+0.000
Top resid features:
2012
Token 2012
Feature activation+0.003
Top resid features:
:
Token:
Feature activation+0.001
Top resid features:
Obama
Token Obama
Feature activation+0.010
Top resid features:
says
Token says
Feature activation+0.034
Top resid features:
the
Token the
Feature activation-0.018
Top resid features:
use
Token use
Feature activation+4.445
Top resid features:
of
Token of
Feature activation-0.128
Top resid features:
chemical
Token chemical
Feature activation-0.012
Top resid features:
weapons
Token weapons
Feature activation+0.004
Top resid features:
would
Token would
Feature activation+0.000
Top resid features:
be
Token be
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
show
Token show
Feature activation-0.005
Top resid features:
his
Token his
Feature activation+0.007
Top resid features:
face
Token face
Feature activation-0.026
Top resid features:
or
Token or
Feature activation+0.053
Top resid features:
use
Token use
Feature activation+4.230
Top resid features:
his
Token his
Feature activation+0.041
Top resid features:
last
Token last
Feature activation+0.003
Top resid features:
name
Token name
Feature activation+0.043
Top resid features:
because
Token because
Feature activation+0.000
Top resid features:
he
Token he
Feature activation+0.000
Top resid features:
team
Token team
Feature activation+0.002
Top resid features:
ultimately
Token ultimately
Feature activation-0.002
Top resid features:
decided
Token decided
Feature activation+0.001
Top resid features:
not
Token not
Feature activation-0.049
Top resid features:
to
Token to
Feature activation-0.011
Top resid features:
use
Token use
Feature activation+4.032
Top resid features:
âĢ
Token âĢ
Feature activation+0.029
Top resid features:
ľ
Tokenľ
Feature activation+0.012
Top resid features:
personal
Tokenpersonal
Feature activation+0.046
Top resid features:
attack
Token attack
Feature activation+0.052
Top resid features:
âĢ
TokenâĢ
Feature activation+0.032
Top resid features:
these
Token these
Feature activation-0.002
Top resid features:
cash
Token cash
Feature activation-0.004
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
conscious
Tokenconscious
Feature activation-0.001
Top resid features:
firms
Token firms
Feature activation+0.006
Top resid features:
use
Token use
Feature activation+4.118
Top resid features:
cloud
Token cloud
Feature activation+0.022
Top resid features:
based
Token based
Feature activation+0.073
Top resid features:
NVIDIA
Token NVIDIA
Feature activation-0.017
Top resid features:
GPUs
Token GPUs
Feature activation+0.108
Top resid features:
to
Token to
Feature activation+0.023
Top resid features:
Rights
Token Rights
Feature activation+0.001
Top resid features:
that
Token that
Feature activation-0.015
Top resid features:
faith
Token faith
Feature activation+0.006
Top resid features:
groups
Token groups
Feature activation-0.002
Top resid features:
are
Token are
Feature activation-0.016
Top resid features:
using
Token using
Feature activation+4.042
Top resid features:
religious
Token religious
Feature activation+0.023
Top resid features:
freedom
Token freedom
Feature activation-0.002
Top resid features:
as
Token as
Feature activation+0.163
Top resid features:
a
Token a
Feature activation+0.021
Top resid features:
pretext
Token pretext
Feature activation+0.074
Top resid features:
office
Token office
Feature activation-0.003
Top resid features:
;
Token;
Feature activation+0.005
Top resid features:
authorities
Token authorities
Feature activation-0.004
Top resid features:
hoped
Token hoped
Feature activation+0.009
Top resid features:
to
Token to
Feature activation-0.049
Top resid features:
use
Token use
Feature activation+4.055
Top resid features:
the
Token the
Feature activation+0.037
Top resid features:
watch
Token watch
Feature activation+0.174
Top resid features:
to
Token to
Feature activation-0.007
Top resid features:
help
Token help
Feature activation+0.000
Top resid features:
identify
Token identify
Feature activation+0.000
Top resid features:
individuals
Token individuals
Feature activation+0.011
Top resid features:
who
Token who
Feature activation-0.008
Top resid features:
would
Token would
Feature activation-0.035
Top resid features:
like
Token like
Feature activation-0.005
Top resid features:
to
Token to
Feature activation-0.007
Top resid features:
use
Token use
Feature activation+3.994
Top resid features:
this
Token this
Feature activation+0.018
Top resid features:
as
Token as
Feature activation+0.157
Top resid features:
an
Token an
Feature activation+0.027
Top resid features:
excuse
Token excuse
Feature activation+0.033
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
ixty
Tokenixty
Feature activation+0.021
Top resid features:
-
Token-
Feature activation-0.002
Top resid features:
year
Tokenyear
Feature activation-0.002
Top resid features:
-
Token-
Feature activation-0.000
Top resid features:
olds
Tokenolds
Feature activation+0.028
Top resid features:
use
Token use
Feature activation+3.692
Top resid features:
six
Token six
Feature activation+0.084
Top resid features:
times
Token times
Feature activation+0.144
Top resid features:
as
Token as
Feature activation+0.044
Top resid features:
much
Token much
Feature activation+0.068
Top resid features:
health
Token health
Feature activation+0.000
Top resid features:
team
Token team
Feature activation+0.001
Top resid features:
ultimately
Token ultimately
Feature activation-0.002
Top resid features:
decided
Token decided
Feature activation+0.001
Top resid features:
not
Token not
Feature activation-0.097
Top resid features:
to
Token to
Feature activation-0.044
Top resid features:
use
Token use
Feature activation+3.730
Top resid features:
âĢ
Token âĢ
Feature activation+0.039
Top resid features:
ľ
Tokenľ
Feature activation+0.014
Top resid features:
personal
Tokenpersonal
Feature activation+0.121
Top resid features:
attack
Token attack
Feature activation+0.091
Top resid features:
âĢ
TokenâĢ
Feature activation+0.064
Top resid features:
accused
Token accused
Feature activation+0.009
Top resid features:
of
Token of
Feature activation-0.019
Top resid features:
grand
Token grand
Feature activation-0.004
Top resid features:
standing
Tokenstanding
Feature activation+0.011
Top resid features:
and
Token and
Feature activation+0.025
Top resid features:
using
Token using
Feature activation+3.885
Top resid features:
her
Token her
Feature activation+0.075
Top resid features:
post
Token post
Feature activation+0.067
Top resid features:
to
Token to
Feature activation+0.013
Top resid features:
bolster
Token bolster
Feature activation+0.000
Top resid features:
her
Token her
Feature activation+0.000
Top resid features:
time
Token time
Feature activation-0.008
Top resid features:
on
Token on
Feature activation+0.003
Top resid features:
his
Token his
Feature activation-0.006
Top resid features:
hands
Token hands
Feature activation-0.009
Top resid features:
and
Token and
Feature activation+0.022
Top resid features:
use
Token use
Feature activation+3.937
Top resid features:
that
Token that
Feature activation+0.095
Top resid features:
money
Token money
Feature activation+0.081
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
hire
Token hire
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
's
Token's
Feature activation+0.000
Top resid features:
effort
Token effort
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.011
Top resid features:
Abbas
Token Abbas
Feature activation-0.007
Top resid features:
has
Token has
Feature activation-0.013
Top resid features:
used
Token used
Feature activation+4.108
Top resid features:
both
Token both
Feature activation-0.027
Top resid features:
the
Token the
Feature activation+0.046
Top resid features:
media
Token media
Feature activation+0.005
Top resid features:
and
Token and
Feature activation+0.007
Top resid features:
fiery
Token fiery
Feature activation-0.008
Top resid features:
even
Token even
Feature activation-0.002
Top resid features:
car
Token car
Feature activation+0.000
Top resid features:
mechanics
Token mechanics
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
They
Token They
Feature activation-0.003
Top resid features:
use
Token use
Feature activation+3.985
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
"
Token "
Feature activation+0.006
Top resid features:
10
Token10
Feature activation-0.027
Top resid features:
31
Token31
Feature activation-0.005
Top resid features:
exchange
Token exchange
Feature activation-0.006
Top resid features:
who
Token who
Feature activation-0.007
Top resid features:
said
Token said
Feature activation+0.008
Top resid features:
he
Token he
Feature activation+0.008
Top resid features:
would
Token would
Feature activation-0.037
Top resid features:
not
Token not
Feature activation-0.055
Top resid features:
use
Token use
Feature activation+3.843
Top resid features:
the
Token the
Feature activation+0.060
Top resid features:
name
Token name
Feature activation+0.127
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
U
Token U
Feature activation+0.000
Top resid features:
,
Token,
Feature activation-0.002
Top resid features:
that
Token that
Feature activation+0.006
Top resid features:
he
Token he
Feature activation+0.001
Top resid features:
refuses
Token refuses
Feature activation-0.043
Top resid features:
to
Token to
Feature activation-0.047
Top resid features:
use
Token use
Feature activation+3.860
Top resid features:
his
Token his
Feature activation+0.153
Top resid features:
name
Token name
Feature activation+0.070
Top resid features:
,"
Token,"
Feature activation+0.000
Top resid features:
he
Token he
Feature activation+0.000
Top resid features:
said
Token said
Feature activation+0.000
Top resid features:
ring
Tokenring
Feature activation+0.000
Top resid features:
budget
Token budget
Feature activation-0.027
Top resid features:
,
Token,
Feature activation+0.004
Top resid features:
hoping
Token hoping
Feature activation+0.008
Top resid features:
to
Token to
Feature activation-0.030
Top resid features:
use
Token use
Feature activation+3.771
Top resid features:
it
Token it
Feature activation+0.036
Top resid features:
as
Token as
Feature activation+0.229
Top resid features:
a
Token a
Feature activation-0.009
Top resid features:
showcase
Token showcase
Feature activation+0.000
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.083
Top resid features:
said
Token said
Feature activation+0.015
Top resid features:
his
Token his
Feature activation-0.002
Top resid features:
uncle
Token uncle
Feature activation-0.019
Top resid features:
used
Token used
Feature activation+3.954
Top resid features:
his
Token his
Feature activation+0.004
Top resid features:
status
Token status
Feature activation+0.005
Top resid features:
as
Token as
Feature activation+0.044
Top resid features:
an
Token an
Feature activation-0.053
Top resid features:
art
Token art
Feature activation-0.000
Top resid features:
require
Token require
Feature activation+0.007
Top resid features:
that
Token that
Feature activation-0.013
Top resid features:
students
Token students
Feature activation+0.024
Top resid features:
can
Token can
Feature activation-0.028
Top resid features:
"
Token "
Feature activation+0.006
Top resid features:
use
Tokenuse
Feature activation+3.832
Top resid features:
only
Token only
Feature activation+0.122
Top resid features:
the
Token the
Feature activation-0.026
Top resid features:
bathroom
Token bathroom
Feature activation+0.038
Top resid features:
/
Token/
Feature activation+0.000
Top resid features:
lock
Tokenlock
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.03

Head 2: 0.07

Head 3: 0.02

Head 4: 0.02

Head 5: 0.04

Head 6: 0.04

Head 7: 0.05

Head 8: 0.06

Head 9: 0.52

Head 10: 0.07

Head 11: 0.06

Positive logits

efficiently1.67

wisely1.58

interchange1.38

empath1.37

strategically1.36

channelAvailability1.36

extensively1.33

sparing1.32

exclusively1.31

Maxim1.31

effectively1.29

bard1.28

liber1.24

selectively1.23

NSA1.19

proxies1.18

avoids1.18

},{"1.18

google1.18

UNCLASSIFIED1.17

Negative logits

comings-1.51

ayer-1.33

KNOWN-1.32

ugu-1.31

owe-1.31

depend-1.29

indebted-1.27

skinned-1.21

ona-1.19

Chero-1.19

missions-1.18

��-1.17

tsy-1.14

uch-1.14

importance-1.13

consequence-1.13

inged-1.12

usefulness-1.12

ebted-1.11

going-1.11

INTERVAL 3.414 - 3.794
CONTAINS 0.001%

says
Token says
Feature activation+0.000
the
Token the
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
chemical
Token chemical
Feature activation+2.403
weapons
Token weapons
Feature activation+3.634
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
'
Token '
Feature activation+0.000
red
Tokenred
Feature activation+0.000
use
Token use
Feature activation+0.000
cloud
Token cloud
Feature activation+0.158
based
Token based
Feature activation+0.688
NVIDIA
Token NVIDIA
Feature activation+0.020
GPUs
Token GPUs
Feature activation+1.329
to
Token to
Feature activation+3.593
develop
Token develop
Feature activation+0.136
their
Token their
Feature activation+0.000
innovative
Token innovative
Feature activation+0.000
products
Token products
Feature activation+0.000
and
Token and
Feature activation+0.000
and
Token and
Feature activation+0.000
deficient
Token deficient
Feature activation+0.000
military
Token military
Feature activation+0.164
commissions
Token commissions
Feature activation+0.000
system
Token system
Feature activation+1.946
to
Token to
Feature activation+3.794
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Kevin
TokenKevin
Feature activation+0.000
Space
Token Space
Feature activation+0.000
y
Tokeny
Feature activation+0.000
is
Token is
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
personal
Tokenpersonal
Feature activation+0.656
attack
Token attack
Feature activation+0.156
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
as
Token as
Feature activation+3.419
a
Token a
Feature activation+3.599
rating
Token rating
Feature activation+0.117
category
Token category
Feature activation+0.137
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
using
Token using
Feature activation+0.000
religious
Token religious
Feature activation+0.000
freedom
Token freedom
Feature activation+0.000
as
Token as
Feature activation+2.036
a
Token a
Feature activation+2.896
pretext
Token pretext
Feature activation+3.556
for
Token for
Feature activation+0.654
discrimination
Token discrimination
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 3.035 - 3.414
CONTAINS 0.002%

hoping
Token hoping
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
it
Token it
Feature activation+0.550
as
Token as
Feature activation+2.645
a
Token a
Feature activation+3.259
showcase
Token showcase
Feature activation+0.364
for
Token for
Feature activation+0.917
his
Token his
Feature activation+0.301
style
Token style
Feature activation+0.004
and
Token and
Feature activation+0.000
he
Token he
Feature activation+0.000
refuses
Token refuses
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
his
Token his
Feature activation+0.910
name
Token name
Feature activation+3.276
,"
Token,"
Feature activation+0.000
he
Token he
Feature activation+0.000
said
Token said
Feature activation+0.000
,
Token,
Feature activation+0.000
referring
Token referring
Feature activation+0.000
hands
Token hands
Feature activation+0.000
and
Token and
Feature activation+0.000
use
Token use
Feature activation+0.000
that
Token that
Feature activation+0.000
money
Token money
Feature activation+2.555
to
Token to
Feature activation+3.376
hire
Token hire
Feature activation+0.000
a
Token a
Feature activation+0.000
snow
Token snow
Feature activation+0.000
pl
Tokenpl
Feature activation+0.000
ow
Tokenow
Feature activation+0.000
status
Token status
Feature activation+0.354
as
Token as
Feature activation+0.000
an
Token an
Feature activation+2.397
art
Token art
Feature activation+0.088
student
Token student
Feature activation+0.000
to
Token to
Feature activation+3.243
get
Token get
Feature activation+1.360
close
Token close
Feature activation+0.000
to
Token to
Feature activation+0.000
J
Token J
Feature activation+0.000
our
Tokenour
Feature activation+0.000
wrong
Token wrong
Feature activation+0.000
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
use
Token use
Feature activation+0.000
the
Token the
Feature activation+0.000
term
Token term
Feature activation+3.103
'
Token '
Feature activation+0.000
head
Tokenhead
Feature activation+0.000
tax
Token tax
Feature activation+0.000
'
Token'
Feature activation+1.104
without
Token without
Feature activation+1.165

INTERVAL 2.655 - 3.035
CONTAINS 0.006%

a
Tokena
Feature activation+0.000
in
Token in
Feature activation+1.612
Bain
Token Bain
Feature activation+0.000
's
Token's
Feature activation+0.000
blood
Token blood
Feature activation+0.000
to
Token to
Feature activation+2.721
suggest
Token suggest
Feature activation+2.217
that
Token that
Feature activation+0.000
Bain
Token Bain
Feature activation+0.000
was
Token was
Feature activation+0.000
lying
Token lying
Feature activation+0.000
used
Token used
Feature activation+0.000
his
Token his
Feature activation+0.231
wreck
Token wreck
Feature activation+0.000
ing
Tokening
Feature activation+0.000
bar
Token bar
Feature activation+0.000
to
Token to
Feature activation+2.720
destroy
Token destroy
Feature activation+0.000
crime
Token crime
Feature activation+0.000
scenes
Token scenes
Feature activation+0.000
.
Token.
Feature activation+0.000
During
Token During
Feature activation+0.000
v
Tokenv
Feature activation+0.000
aney
Tokenaney
Feature activation+0.851
as
Token as
Feature activation+1.681
an
Token an
Feature activation+2.677
intermediary
Token intermediary
Feature activation+1.629
to
Token to
Feature activation+2.718
threaten
Token threaten
Feature activation+0.812
to
Token to
Feature activation+0.000
oust
Token oust
Feature activation+0.000
Sanford
Token Sanford
Feature activation+0.000
in
Token in
Feature activation+0.000
as
Token as
Feature activation+0.000
long
Token long
Feature activation+0.000
and
Token and
Feature activation+0.000
use
Token use
Feature activation+0.000
less
Token less
Feature activation+0.000
energy
Token energy
Feature activation+2.743
than
Token than
Feature activation+0.000
the
Token the
Feature activation+0.347
Pri
Token Pri
Feature activation+0.000
us
Tokenus
Feature activation+0.000
.
Token.
Feature activation+0.000
family
Token family
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
fortune
Token fortune
Feature activation+0.587
to
Token to
Feature activation+2.744
purchase
Token purchase
Feature activation+0.000
a
Token a
Feature activation+0.000
Commission
Token Commission
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 2.276 - 2.655
CONTAINS 0.010%

,
Token,
Feature activation+0.000
hoping
Token hoping
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
it
Token it
Feature activation+0.550
as
Token as
Feature activation+2.645
a
Token a
Feature activation+3.259
showcase
Token showcase
Feature activation+0.364
for
Token for
Feature activation+0.917
his
Token his
Feature activation+0.301
style
Token style
Feature activation+0.004
using
Token using
Feature activation+0.000
high
Token high
Feature activation+0.000
-
Token-
Feature activation+0.000
tech
Tokentech
Feature activation+0.000
tactics
Token tactics
Feature activation+2.750
to
Token to
Feature activation+2.580
build
Token build
Feature activation+1.093
massive
Token massive
Feature activation+0.000
databases
Token databases
Feature activation+0.000
that
Token that
Feature activation+0.000
could
Token could
Feature activation+0.000
and
Token and
Feature activation+0.000
meaningless
Token meaningless
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
term
Token term
Feature activation+2.479
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
cons
Tokencons
Feature activation+0.000
ensus
Tokenensus
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
and
Token and
Feature activation+0.000
other
Token other
Feature activation+0.000
retro
Token retro
Feature activation+0.000
gaming
Token gaming
Feature activation+0.000
hardware
Token hardware
Feature activation+1.243
to
Token to
Feature activation+2.285
create
Token create
Feature activation+1.651
some
Token some
Feature activation+0.000
rad
Token rad
Feature activation+0.000
-
Token-
Feature activation+0.000
sounding
Tokensounding
Feature activation+0.000
the
Token the
Feature activation+0.000
upcoming
Token upcoming
Feature activation+0.000
season
Token season
Feature activation+0.000
using
Token using
Feature activation+0.000
these
Token these
Feature activation+0.000
methods
Token methods
Feature activation+2.545
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
This
TokenThis
Feature activation+0.000

INTERVAL 1.897 - 2.276
CONTAINS 0.013%

but
Token but
Feature activation+0.000
is
Token is
Feature activation+0.000
also
Token also
Feature activation+0.000
using
Token using
Feature activation+0.000
them
Token them
Feature activation+0.000
as
Token as
Feature activation+1.926
leverage
Token leverage
Feature activation+2.956
to
Token to
Feature activation+2.572
get
Token get
Feature activation+0.637
a
Token a
Feature activation+0.000
bill
Token bill
Feature activation+0.000
SP
TokenSP
Feature activation+0.000
or
Token or
Feature activation+0.017
SR
Token SR
Feature activation+0.000
P
TokenP
Feature activation+0.000
to
Token to
Feature activation+1.201
denote
Token denote
Feature activation+2.004
membership
Token membership
Feature activation+0.000
to
Token to
Feature activation+0.000
professional
Token professional
Feature activation+0.000
bodies
Token bodies
Feature activation+0.000
.
Token.
Feature activation+0.000
creators
Token creators
Feature activation+0.000
use
Token use
Feature activation+0.000
only
Token only
Feature activation+0.000
use
Token use
Feature activation+0.000
as
Token as
Feature activation+1.652
much
Token much
Feature activation+2.129
copyrighted
Token copyrighted
Feature activation+0.747
content
Token content
Feature activation+0.352
as
Token as
Feature activation+0.685
they
Token they
Feature activation+0.408
absolutely
Token absolutely
Feature activation+0.000
micro
Token micro
Feature activation+0.000
-
Token-
Feature activation+0.000
mir
Tokenmir
Feature activation+0.000
ror
Tokenror
Feature activation+0.000
array
Token array
Feature activation+0.000
to
Token to
Feature activation+2.024
random
Token random
Feature activation+0.000
ise
Tokenise
Feature activation+0.000
the
Token the
Feature activation+0.000
light
Token light
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
many
Token many
Feature activation+0.000
lies
Token lies
Feature activation+0.000
used
Token used
Feature activation+0.000
to
Token to
Feature activation+0.998
force
Token force
Feature activation+2.215
-
Token-
Feature activation+0.000
feed
Tokenfeed
Feature activation+0.000
us
Token us
Feature activation+0.000
global
Token global
Feature activation+0.000
warming
Token warming
Feature activation+0.000

INTERVAL 1.517 - 1.897
CONTAINS 0.015%

Google
Token Google
Feature activation+0.000
.
Token.
Feature activation+0.000
Using
Token Using
Feature activation+0.000
Google
Token Google
Feature activation+0.000
Trends
Token Trends
Feature activation+0.000
,
Token,
Feature activation+1.703
we
Token we
Feature activation+1.033
can
Token can
Feature activation+0.175
see
Token see
Feature activation+0.000
the
Token the
Feature activation+0.000
growth
Token growth
Feature activation+0.000
using
Token using
Feature activation+0.000
some
Token some
Feature activation+0.000
kind
Token kind
Feature activation+0.000
of
Token of
Feature activation+0.373
industrial
Token industrial
Feature activation+0.769
strength
Token strength
Feature activation+1.593
acet
Token acet
Feature activation+0.000
ate
Tokenate
Feature activation+0.431
and
Token and
Feature activation+0.000
how
Token how
Feature activation+0.000
the
Token the
Feature activation+0.000
only
Token only
Feature activation+0.000
use
Token use
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
humane
Tokenhumane
Feature activation+1.822
treatment
Token treatment
Feature activation+1.826
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
to
Token to
Feature activation+2.924
deter
Token deter
Feature activation+0.955
refugees
Token refugees
Feature activation+0.000
Using
Token Using
Feature activation+0.000
him
Token him
Feature activation+0.360
as
Token as
Feature activation+1.757
a
Token a
Feature activation+2.346
third
Token third
Feature activation+0.000
option
Token option
Feature activation+1.711
will
Token will
Feature activation+0.046
allow
Token allow
Feature activation+0.000
his
Token his
Feature activation+0.000
all
Token all
Feature activation+0.000
-
Token-
Feature activation+0.000
.
Token.
Feature activation+0.000
He
Token He
Feature activation+0.000
used
Token used
Feature activation+0.000
the
Token the
Feature activation+0.000
opportunity
Token opportunity
Feature activation+2.095
to
Token to
Feature activation+1.564
raise
Token raise
Feature activation+0.000
the
Token the
Feature activation+0.000
issue
Token issue
Feature activation+0.000
of
Token of
Feature activation+0.000
our
Token our
Feature activation+0.000

INTERVAL 1.138 - 1.517
CONTAINS 0.022%

:
Token:
Feature activation+0.000
officers
Token officers
Feature activation+0.000
may
Token may
Feature activation+0.000
use
Token use
Feature activation+0.000
deadly
Token deadly
Feature activation+0.000
force
Token force
Feature activation+1.288
only
Token only
Feature activation+0.000
when
Token when
Feature activation+0.103
necessary
Token necessary
Feature activation+0.000
-
Token -
Feature activation+0.000
that
Token that
Feature activation+0.000
Major
Token Major
Feature activation+0.000
companies
Token companies
Feature activation+0.000
used
Token used
Feature activation+0.000
cheap
Token cheap
Feature activation+0.000
credit
Token credit
Feature activation+0.000
to
Token to
Feature activation+1.411
buy
Token buy
Feature activation+0.000
their
Token their
Feature activation+0.000
own
Token own
Feature activation+0.000
stocks
Token stocks
Feature activation+0.000
back
Token back
Feature activation+0.000
Physical
Token Physical
Feature activation+0.000
Ther
Token Ther
Feature activation+0.000
ap
Tokenap
Feature activation+0.000
ists
Tokenists
Feature activation+0.000
to
Token to
Feature activation+1.129
denote
Token denote
Feature activation+1.164
their
Token their
Feature activation+0.000
state
Token state
Feature activation+0.000
certification
Token certification
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
that
Token that
Feature activation+0.000
utilizes
Token utilizes
Feature activation+0.000
BMW
Token BMW
Feature activation+0.000
e
Token e
Feature activation+0.000
Drive
TokenDrive
Feature activation+0.000
technology
Token technology
Feature activation+1.479
with
Token with
Feature activation+1.237
a
Token a
Feature activation+0.073
BMW
Token BMW
Feature activation+0.000
Twin
Token Twin
Feature activation+0.000
Power
TokenPower
Feature activation+0.000
use
Token use
Feature activation+0.000
the
Token the
Feature activation+0.000
broad
Token broad
Feature activation+1.051
discretionary
Token discretionary
Feature activation+0.939
powers
Token powers
Feature activation+1.978
with
Token with
Feature activation+1.285
which
Token which
Feature activation+0.000
they
Token they
Feature activation+0.000
are
Token are
Feature activation+0.000
vested
Token vested
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 0.759 - 1.138
CONTAINS 0.030%

for
Token for
Feature activation+0.000
Luke
Token Luke
Feature activation+0.000
,
Token,
Feature activation+0.000
who
Token who
Feature activation+0.000
uses
Token uses
Feature activation+0.000
it
Token it
Feature activation+0.838
to
Token to
Feature activation+2.338
tell
Token tell
Feature activation+0.188
a
Token a
Feature activation+0.000
story
Token story
Feature activation+0.000
about
Token about
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
her
Token her
Feature activation+0.000
proper
Token proper
Feature activation+0.989
name
Token name
Feature activation+1.699
:
Token:
Feature activation+0.000
Rap
Token Rap
Feature activation+0.000
un
Tokenun
Feature activation+0.000
zel
Tokenzel
Feature activation+0.000
CODE
Token CODE
Feature activation+1.086
:
Token:
Feature activation+0.000
LEG
Token LEG
Feature activation+0.000
EN
TokenEN
Feature activation+0.000
DS
TokenDS
Feature activation+0.000
to
Token to
Feature activation+0.829
get
Token get
Feature activation+0.000
20
Token 20
Feature activation+0.000
%
Token%
Feature activation+0.000
off
Token off
Feature activation+0.000
your
Token your
Feature activation+0.000
are
Token are
Feature activation+0.000
meant
Token meant
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
used
Token used
Feature activation+0.000
with
Token with
Feature activation+1.002
SEC
Token SEC
Feature activation+0.000
D
TokenD
Feature activation+0.000
ED
TokenED
Feature activation+0.000
(
Token (
Feature activation+0.000
single
Tokensingle
Feature activation+0.000
the
Token the
Feature activation+0.000
same
Token same
Feature activation+0.530
Pant
Token Pant
Feature activation+0.000
one
Tokenone
Feature activation+0.000
color
Token color
Feature activation+0.102
when
Token when
Feature activation+1.017
rec
Token rec
Feature activation+0.000
reating
Tokenreating
Feature activation+0.000
that
Token that
Feature activation+0.000
logo
Token logo
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 0.379 - 0.759
CONTAINS 0.053%

to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
the
Token the
Feature activation+0.429
stick
Token stick
Feature activation+1.138
,
Token,
Feature activation+0.000
rather
Token rather
Feature activation+0.455
than
Token than
Feature activation+0.000
the
Token the
Feature activation+0.000
carrot
Token carrot
Feature activation+0.036
,
Token,
Feature activation+0.000
we
Token we
Feature activation+0.000
airspace
Token airspace
Feature activation+0.453
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
Heath
Token Heath
Feature activation+0.000
row
Tokenrow
Feature activation+0.000
to
Token to
Feature activation+0.464
Kiev
Token Kiev
Feature activation+0.000
service
Token service
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
spokesman
Token spokesman
Feature activation+0.000
compound
Token compound
Feature activation+0.000
that
Token that
Feature activation+0.000
could
Token could
Feature activation+0.000
be
Token be
Feature activation+0.000
used
Token used
Feature activation+0.000
in
Token in
Feature activation+0.619
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
smart
Tokensmart
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
on
Token on
Feature activation+0.000
his
Token his
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
PA
Token PA
Feature activation+0.688
Z
TokenZ
Feature activation+0.000
Analy
Token Analy
Feature activation+0.000
zer
Tokenzer
Feature activation+0.000
to
Token to
Feature activation+1.092
clean
Token clean
Feature activation+0.000
s
Tokens
Feature activation+0.000
resolve
Token resolve
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
as
Token as
Feature activation+0.526
a
Token a
Feature activation+0.402
key
Token key
Feature activation+0.000
feature
Token feature
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
novel
Token novel
Feature activation+0.000

INTERVAL 0.000 - 0.379
CONTAINS 99.849%

ani
Tokenani
Feature activation+0.000
"
Token"
Feature activation+0.000
in
Token in
Feature activation+0.000
UP
Token UP
Feature activation+0.000
's
Token's
Feature activation+0.000
Bund
Token Bund
Feature activation+0.000
el
Tokenel
Feature activation+0.000
k
Tokenk
Feature activation+0.000
hand
Tokenhand
Feature activation+0.000
stops
Token stops
Feature activation+0.000
wedding
Token wedding
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Sam
TokenSam
Feature activation+0.000
is
Token is
Feature activation+0.000
from
Token from
Feature activation+0.000
nearby
Token nearby
Feature activation+0.000
Pam
Token Pam
Feature activation+0.000
pl
Tokenpl
Feature activation+0.000
ico
Tokenico
Feature activation+0.000
,
Token,
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
a
Token a
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
bur
Tokenbur
Feature activation+0.000
under
Token under
Feature activation+0.000
the
Token the
Feature activation+0.000
saddle
Token saddle
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
on
Token on
Feature activation+0.000
to
Token to
Feature activation+0.000
organize
Token organize
Feature activation+0.000
the
Token the
Feature activation+0.000
event
Token event
Feature activation+0.000
because
Token because
Feature activation+0.000
she
Token she
Feature activation+0.000
"
Token "
Feature activation+0.000
could
Tokencould
Feature activation+0.000
n
Tokenn
Feature activation+0.000
't
Token't
Feature activation+0.000
believe
Token believe
Feature activation+0.000
"
Token "
Feature activation+0.000
us
Tokenus
Feature activation+0.000
believe
Token believe
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
racism
Token racism
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
time
Token time
Feature activation+0.000
was
Token was
Feature activation+0.000
the
Token the
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000

Top feature 9 in H3.9: (feature 15843)

TOP ACTIVATIONS
MAX = 4.417

candidates
Token candidates
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
general
Token general
Feature activation+0.000
election
Token election
Feature activation+0.000
,
Token,
Feature activation+4.417
that
Token that
Feature activation+2.896
voter
Token voter
Feature activation+0.000
might
Token might
Feature activation+0.000
easily
Token easily
Feature activation+0.000
decide
Token decide
Feature activation+0.000
,
Token,
Feature activation+0.000
once
Token once
Feature activation+0.000
genocide
Token genocide
Feature activation+0.000
is
Token is
Feature activation+1.621
acceptable
Token acceptable
Feature activation+0.340
,
Token,
Feature activation+4.411
what
Token what
Feature activation+1.427
isn
Token isn
Feature activation+0.000
't
Token't
Feature activation+0.000
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
or
Token or
Feature activation+0.000
S
Token S
Feature activation+0.000
ibel
Tokenibel
Feature activation+0.000
ius
Tokenius
Feature activation+0.000
,
Token,
Feature activation+3.218
it
Token it
Feature activation+4.324
will
Token will
Feature activation+1.213
look
Token look
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
you
Token you
Feature activation+0.000
once
Token once
Feature activation+0.721
that
Token that
Feature activation+0.000
lease
Token lease
Feature activation+0.000
runs
Token runs
Feature activation+1.769
out
Token out
Feature activation+0.171
,
Token,
Feature activation+4.282
the
Token the
Feature activation+4.077
Bucks
Token Bucks
Feature activation+0.000
will
Token will
Feature activation+1.486
be
Token be
Feature activation+0.000
playing
Token playing
Feature activation+0.000
that
Token that
Feature activation+0.000
lease
Token lease
Feature activation+0.000
runs
Token runs
Feature activation+1.769
out
Token out
Feature activation+0.171
,
Token,
Feature activation+4.282
the
Token the
Feature activation+4.077
Bucks
Token Bucks
Feature activation+0.000
will
Token will
Feature activation+1.486
be
Token be
Feature activation+0.000
playing
Token playing
Feature activation+0.000
in
Token in
Feature activation+0.000
it
Token it
Feature activation+1.446
's
Token's
Feature activation+2.133
left
Token left
Feature activation+1.034
here
Token here
Feature activation+0.000
,
Token,
Feature activation+3.365
there
Token there
Feature activation+4.030
's
Token's
Feature activation+3.301
nothing
Token nothing
Feature activation+0.105
I
Token I
Feature activation+0.905
can
Token can
Feature activation+0.409
do
Token do
Feature activation+0.000
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
that
Token that
Feature activation+0.286
was
Token was
Feature activation+0.819
done
Token done
Feature activation+3.411
I
Token I
Feature activation+3.941
stamped
Token stamped
Feature activation+1.047
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
shape
Token shape
Feature activation+0.000
(
Token (
Feature activation+0.000
get
Token get
Feature activation+3.395
around
Token around
Feature activation+0.000
the
Token the
Feature activation+0.009
guys
Token guys
Feature activation+0.000
,
Token,
Feature activation+3.284
it
Token it
Feature activation+3.935
becomes
Token becomes
Feature activation+3.276
family
Token family
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
guys
Token guys
Feature activation+0.000
placebo
Token placebo
Feature activation+0.000
and
Token and
Feature activation+0.000
drug
Token drug
Feature activation+0.000
withdrawal
Token withdrawal
Feature activation+0.000
effects
Token effects
Feature activation+0.000
are
Token are
Feature activation+3.702
accounted
Token accounted
Feature activation+3.367
for
Token for
Feature activation+0.000
,
Token,
Feature activation+3.221
there
Token there
Feature activation+3.330
is
Token is
Feature activation+1.282
you
Token you
Feature activation+1.328
read
Token read
Feature activation+1.495
the
Token the
Feature activation+1.220
replay
Token replay
Feature activation+0.000
,
Token,
Feature activation+3.434
you
Token you
Feature activation+3.674
've
Token've
Feature activation+3.266
got
Token got
Feature activation+0.084
an
Token an
Feature activation+0.000
idea
Token idea
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.737
print
Token print
Feature activation+0.000
heads
Token heads
Feature activation+0.000
are
Token are
Feature activation+1.811
clean
Token clean
Feature activation+2.800
,
Token,
Feature activation+3.663
you
Token you
Feature activation+3.095
can
Token can
Feature activation+1.855
enjoy
Token enjoy
Feature activation+0.000
higher
Token higher
Feature activation+0.000
performance
Token performance
Feature activation+0.000
a
Token a
Feature activation+0.331
week
Token week
Feature activation+0.000
,
Token,
Feature activation+2.273
and
Token and
Feature activation+0.000
we
Token we
Feature activation+1.568
've
Token've
Feature activation+3.572
had
Token had
Feature activation+0.000
lines
Token lines
Feature activation+0.000
out
Token out
Feature activation+0.000
the
Token the
Feature activation+0.000
door
Token door
Feature activation+0.000
mechanics
Token mechanics
Feature activation+0.000
go
Token go
Feature activation+1.868
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
game
Token game
Feature activation+0.000
,
Token,
Feature activation+3.549
it
Token it
Feature activation+3.543
becomes
Token becomes
Feature activation+2.041
much
Token much
Feature activation+0.135
easier
Token easier
Feature activation+0.479
for
Token for
Feature activation+0.000
go
Token go
Feature activation+1.868
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
game
Token game
Feature activation+0.000
,
Token,
Feature activation+3.549
it
Token it
Feature activation+3.543
becomes
Token becomes
Feature activation+2.041
much
Token much
Feature activation+0.135
easier
Token easier
Feature activation+0.479
for
Token for
Feature activation+0.000
us
Token us
Feature activation+0.000
the
Token the
Feature activation+1.553
US
Token US
Feature activation+0.000
joined
Token joined
Feature activation+0.701
in
Token in
Feature activation+0.000
2010
Token 2010
Feature activation+0.000
,
Token,
Feature activation+3.518
this
Token this
Feature activation+2.171
small
Token small
Feature activation+0.000
-
Token-
Feature activation+0.000
scale
Tokenscale
Feature activation+0.000
partnership
Token partnership
Feature activation+0.000
worked
Token worked
Feature activation+0.353
out
Token out
Feature activation+0.048
and
Token and
Feature activation+0.255
it
Token it
Feature activation+1.482
happened
Token happened
Feature activation+0.989
,
Token,
Feature activation+3.466
it
Token it
Feature activation+2.517
was
Token was
Feature activation+1.301
like
Token like
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
Once
Token Once
Feature activation+0.005
you
Token you
Feature activation+1.328
read
Token read
Feature activation+1.495
the
Token the
Feature activation+1.220
replay
Token replay
Feature activation+0.000
,
Token,
Feature activation+3.434
you
Token you
Feature activation+3.674
've
Token've
Feature activation+3.266
got
Token got
Feature activation+0.084
an
Token an
Feature activation+0.000
idea
Token idea
Feature activation+0.000
all
Token all
Feature activation+0.000
of
Token of
Feature activation+0.288
the
Token the
Feature activation+0.183
votes
Token votes
Feature activation+0.000
are
Token are
Feature activation+1.781
counted
Token counted
Feature activation+3.416
âĢĶ
Token âĢĶ
Feature activation+0.000
with
Token with
Feature activation+0.000
Tim
Token Tim
Feature activation+0.000
Rain
Token Rain
Feature activation+0.000
es
Tokenes
Feature activation+0.000
nails
Token nails
Feature activation+0.000
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
that
Token that
Feature activation+0.286
was
Token was
Feature activation+0.819
done
Token done
Feature activation+3.411
I
Token I
Feature activation+3.941
stamped
Token stamped
Feature activation+1.047
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
shape
Token shape
Feature activation+0.000
it
Token it
Feature activation+1.757
's
Token's
Feature activation+2.308
up
Token up
Feature activation+1.756
and
Token and
Feature activation+1.395
running
Token running
Feature activation+0.291
you
Token you
Feature activation+3.404
shouldn
Token shouldn
Feature activation+2.062
't
Token't
Feature activation+0.000
have
Token have
Feature activation+0.000
a
Token a
Feature activation+0.000
problem
Token problem
Feature activation+0.000

Top DFA by src position
MAX = 5.274

.
Token.
Feature activation+0.016
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
Ċ
TokenĊ
Feature activation+0.010
Top resid features:
However
TokenHowever
Feature activation+0.024
Top resid features:
,
Token,
Feature activation+0.041
Top resid features:
once
Token once
Feature activation+5.209
Top resid features:
the
Token the
Feature activation+0.045
Top resid features:
decks
Token decks
Feature activation+0.002
Top resid features:
are
Token are
Feature activation+0.028
Top resid features:
cleared
Token cleared
Feature activation+0.031
Top resid features:
and
Token and
Feature activation+0.029
Top resid features:
obey
Token obey
Feature activation-0.008
Top resid features:
?
Token?
Feature activation+0.039
Top resid features:
After
Token After
Feature activation+0.030
Top resid features:
all
Token all
Feature activation+0.021
Top resid features:
,
Token,
Feature activation+0.093
Top resid features:
once
Token once
Feature activation+5.137
Top resid features:
genocide
Token genocide
Feature activation-0.051
Top resid features:
is
Token is
Feature activation+0.108
Top resid features:
acceptable
Token acceptable
Feature activation+0.047
Top resid features:
,
Token,
Feature activation+0.384
Top resid features:
what
Token what
Feature activation+0.000
Top resid features:
former
Tokenformer
Feature activation-0.000
Top resid features:
and
Token and
Feature activation-0.008
Top resid features:
feel
Token feel
Feature activation+0.009
Top resid features:
confident
Token confident
Feature activation+0.012
Top resid features:
that
Token that
Feature activation+0.186
Top resid features:
once
Token once
Feature activation+5.172
Top resid features:
you
Token you
Feature activation+0.024
Top resid features:
open
Token open
Feature activation+0.046
Top resid features:
the
Token the
Feature activation+0.054
Top resid features:
score
Token score
Feature activation+0.005
Top resid features:
in
Token in
Feature activation+0.005
Top resid features:
).
Token).
Feature activation+0.015
Top resid features:
They
Token They
Feature activation+0.006
Top resid features:
have
Token have
Feature activation+0.018
Top resid features:
said
Token said
Feature activation+0.068
Top resid features:
that
Token that
Feature activation+0.088
Top resid features:
once
Token once
Feature activation+5.274
Top resid features:
that
Token that
Feature activation+0.095
Top resid features:
lease
Token lease
Feature activation-0.054
Top resid features:
runs
Token runs
Feature activation+0.061
Top resid features:
out
Token out
Feature activation-0.022
Top resid features:
,
Token,
Feature activation+0.167
Top resid features:
).
Token).
Feature activation+0.011
Top resid features:
They
Token They
Feature activation-0.003
Top resid features:
have
Token have
Feature activation+0.008
Top resid features:
said
Token said
Feature activation+0.084
Top resid features:
that
Token that
Feature activation+0.092
Top resid features:
once
Token once
Feature activation+5.127
Top resid features:
that
Token that
Feature activation+0.099
Top resid features:
lease
Token lease
Feature activation-0.024
Top resid features:
runs
Token runs
Feature activation+0.014
Top resid features:
out
Token out
Feature activation-0.056
Top resid features:
,
Token,
Feature activation+0.206
Top resid features:
's
Token's
Feature activation+0.008
Top resid features:
been
Token been
Feature activation+0.013
Top resid features:
shipped
Token shipped
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
but
Token but
Feature activation+0.084
Top resid features:
once
Token once
Feature activation+4.665
Top resid features:
it
Token it
Feature activation-0.011
Top resid features:
's
Token's
Feature activation+0.059
Top resid features:
left
Token left
Feature activation+0.008
Top resid features:
here
Token here
Feature activation+0.071
Top resid features:
,
Token,
Feature activation+0.576
Top resid features:
pper
Tokenpper
Feature activation-0.004
Top resid features:
on
Token on
Feature activation-0.001
Top resid features:
all
Token all
Feature activation+0.006
Top resid features:
nails
Token nails
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.113
Top resid features:
Once
Token Once
Feature activation+4.655
Top resid features:
that
Token that
Feature activation+0.098
Top resid features:
was
Token was
Feature activation+0.230
Top resid features:
done
Token done
Feature activation+0.176
Top resid features:
I
Token I
Feature activation+0.140
Top resid features:
stamped
Token stamped
Feature activation+0.000
Top resid features:
had
Token had
Feature activation+0.010
Top resid features:
big
Token big
Feature activation-0.000
Top resid features:
contracts
Token contracts
Feature activation-0.002
Top resid features:
,
Token,
Feature activation+0.019
Top resid features:
but
Token but
Feature activation+0.041
Top resid features:
once
Token once
Feature activation+4.827
Top resid features:
you
Token you
Feature activation+0.035
Top resid features:
get
Token get
Feature activation+0.047
Top resid features:
on
Token on
Feature activation+0.006
Top resid features:
a
Token a
Feature activation+0.011
Top resid features:
team
Token team
Feature activation-0.021
Top resid features:
's
Token's
Feature activation+0.013
Top resid features:
first
Token first
Feature activation+0.001
Top resid features:
point
Token point
Feature activation+0.005
Top resid features:
is
Token is
Feature activation+0.057
Top resid features:
that
Token that
Feature activation+0.179
Top resid features:
once
Token once
Feature activation+4.820
Top resid features:
variables
Token variables
Feature activation+0.002
Top resid features:
such
Token such
Feature activation-0.086
Top resid features:
as
Token as
Feature activation+0.042
Top resid features:
placebo
Token placebo
Feature activation-0.013
Top resid features:
and
Token and
Feature activation+0.032
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
play
Token play
Feature activation+0.004
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
game
Token game
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.064
Top resid features:
Once
Token Once
Feature activation+4.501
Top resid features:
you
Token you
Feature activation+0.056
Top resid features:
read
Token read
Feature activation+0.089
Top resid features:
the
Token the
Feature activation+0.092
Top resid features:
replay
Token replay
Feature activation-0.022
Top resid features:
,
Token,
Feature activation+0.208
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
print
Token print
Feature activation+0.001
Top resid features:
heads
Token heads
Feature activation+0.002
Top resid features:
electronically
Token electronically
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.091
Top resid features:
Once
Token Once
Feature activation+4.454
Top resid features:
the
Token the
Feature activation+0.080
Top resid features:
print
Token print
Feature activation+0.008
Top resid features:
heads
Token heads
Feature activation+0.025
Top resid features:
are
Token are
Feature activation+0.070
Top resid features:
clean
Token clean
Feature activation+0.057
Top resid features:
list
Tokenlist
Feature activation+0.027
Top resid features:
every
Token every
Feature activation+0.047
Top resid features:
Monday
Token Monday
Feature activation+0.029
Top resid features:
morning
Token morning
Feature activation+0.040
Top resid features:
,
Token,
Feature activation+0.050
Top resid features:
once
Token once
Feature activation+4.197
Top resid features:
a
Token a
Feature activation+0.069
Top resid features:
week
Token week
Feature activation+0.091
Top resid features:
,
Token,
Feature activation+0.241
Top resid features:
and
Token and
Feature activation+0.250
Top resid features:
we
Token we
Feature activation-0.053
Top resid features:
add
Token add
Feature activation-0.001
Top resid features:
throughout
Token throughout
Feature activation+0.010
Top resid features:
Early
Token Early
Feature activation+0.006
Top resid features:
Access
Token Access
Feature activation+0.008
Top resid features:
.
Token.
Feature activation+0.099
Top resid features:
Once
Token Once
Feature activation+4.310
Top resid features:
the
Token the
Feature activation+0.025
Top resid features:
mechanics
Token mechanics
Feature activation+0.002
Top resid features:
go
Token go
Feature activation+0.071
Top resid features:
into
Token into
Feature activation+0.052
Top resid features:
the
Token the
Feature activation+0.072
Top resid features:
add
Token add
Feature activation-0.002
Top resid features:
throughout
Token throughout
Feature activation+0.006
Top resid features:
Early
Token Early
Feature activation+0.003
Top resid features:
Access
Token Access
Feature activation+0.005
Top resid features:
.
Token.
Feature activation+0.089
Top resid features:
Once
Token Once
Feature activation+4.197
Top resid features:
the
Token the
Feature activation+0.060
Top resid features:
mechanics
Token mechanics
Feature activation+0.017
Top resid features:
go
Token go
Feature activation+0.100
Top resid features:
into
Token into
Feature activation+0.052
Top resid features:
the
Token the
Feature activation+0.035
Top resid features:
Zealand
Token Zealand
Feature activation-0.016
Top resid features:
and
Token and
Feature activation-0.006
Top resid features:
Singapore
Token Singapore
Feature activation-0.017
Top resid features:
.
Token.
Feature activation+0.061
Top resid features:
But
Token But
Feature activation+0.207
Top resid features:
once
Token once
Feature activation+4.605
Top resid features:
the
Token the
Feature activation+0.075
Top resid features:
US
Token US
Feature activation-0.061
Top resid features:
joined
Token joined
Feature activation+0.067
Top resid features:
in
Token in
Feature activation+0.068
Top resid features:
2010
Token 2010
Feature activation-0.123
Top resid features:
reveals
Token reveals
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.009
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
"
Token"
Feature activation+0.033
Top resid features:
Once
TokenOnce
Feature activation+4.333
Top resid features:
it
Token it
Feature activation-0.013
Top resid features:
worked
Token worked
Feature activation+0.046
Top resid features:
out
Token out
Feature activation+0.024
Top resid features:
and
Token and
Feature activation+0.135
Top resid features:
it
Token it
Feature activation-0.017
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
play
Token play
Feature activation+0.007
Top resid features:
the
Token the
Feature activation+0.013
Top resid features:
game
Token game
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.086
Top resid features:
Once
Token Once
Feature activation+4.026
Top resid features:
you
Token you
Feature activation+0.108
Top resid features:
read
Token read
Feature activation+0.191
Top resid features:
the
Token the
Feature activation+0.258
Top resid features:
replay
Token replay
Feature activation-0.168
Top resid features:
,
Token,
Feature activation+0.259
Top resid features:
.
Token.
Feature activation+0.041
Top resid features:
Their
Token Their
Feature activation+0.012
Top resid features:
numbers
Token numbers
Feature activation-0.058
Top resid features:
will
Token will
Feature activation+0.019
Top resid features:
drop
Token drop
Feature activation-0.000
Top resid features:
once
Token once
Feature activation+4.629
Top resid features:
all
Token all
Feature activation-0.030
Top resid features:
of
Token of
Feature activation+0.051
Top resid features:
the
Token the
Feature activation+0.033
Top resid features:
votes
Token votes
Feature activation+0.023
Top resid features:
are
Token are
Feature activation+0.136
Top resid features:
pper
Tokenpper
Feature activation-0.020
Top resid features:
on
Token on
Feature activation+0.002
Top resid features:
all
Token all
Feature activation+0.005
Top resid features:
nails
Token nails
Feature activation+0.026
Top resid features:
.
Token.
Feature activation+0.139
Top resid features:
Once
Token Once
Feature activation+4.498
Top resid features:
that
Token that
Feature activation+0.006
Top resid features:
was
Token was
Feature activation+0.057
Top resid features:
done
Token done
Feature activation+0.035
Top resid features:
I
Token I
Feature activation+0.000
Top resid features:
stamped
Token stamped
Feature activation+0.000
Top resid features:
normal
Token normal
Feature activation-0.003
Top resid features:
Chrome
Token Chrome
Feature activation-0.007
Top resid features:
install
Token install
Feature activation+0.016
Top resid features:
,
Token,
Feature activation+0.038
Top resid features:
but
Token but
Feature activation+0.023
Top resid features:
once
Token once
Feature activation+4.184
Top resid features:
it
Token it
Feature activation+0.031
Top resid features:
's
Token's
Feature activation+0.133
Top resid features:
up
Token up
Feature activation+0.014
Top resid features:
and
Token and
Feature activation+0.163
Top resid features:
running
Token running
Feature activation+0.213
Top resid features:

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.04

Head 2: 0.04

Head 3: 0.02

Head 4: 0.03

Head 5: 0.08

Head 6: 0.04

Head 7: 0.04

Head 8: 0.04

Head 9: 0.51

Head 10: 0.05

Head 11: 0.10

Positive logits

irements1.45

Reloaded1.38

retire1.31

cks1.30

Finish1.29

riger1.27

llah1.27

iosis1.24

raq1.23

aea1.23

rehens1.22

sett1.22

idges1.22

ymes1.22

ritis1.21

iculty1.21

mble1.21

pection1.20

icent1.20

enters1.19

Negative logits

Mond-1.37

Taiwanese-1.32

corrid-1.32

advoc-1.28

Cosponsors-1.28

Gupta-1.24

Sigma-1.22

Singer-1.20

Fitz-1.18

cause-1.17

soType-1.17

Mong-1.17

Ig-1.15

-1.14

Native-1.13

Zheng-1.11

Templar-1.10

Authors-1.10

aboriginal-1.08

Variant-1.08

INTERVAL 3.975 - 4.417
CONTAINS 0.001%

candidates
Token candidates
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
general
Token general
Feature activation+0.000
election
Token election
Feature activation+0.000
,
Token,
Feature activation+4.417
that
Token that
Feature activation+2.896
voter
Token voter
Feature activation+0.000
might
Token might
Feature activation+0.000
easily
Token easily
Feature activation+0.000
decide
Token decide
Feature activation+0.000
or
Token or
Feature activation+0.000
S
Token S
Feature activation+0.000
ibel
Tokenibel
Feature activation+0.000
ius
Tokenius
Feature activation+0.000
,
Token,
Feature activation+3.218
it
Token it
Feature activation+4.324
will
Token will
Feature activation+1.213
look
Token look
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
you
Token you
Feature activation+0.000
it
Token it
Feature activation+1.446
's
Token's
Feature activation+2.133
left
Token left
Feature activation+1.034
here
Token here
Feature activation+0.000
,
Token,
Feature activation+3.365
there
Token there
Feature activation+4.030
's
Token's
Feature activation+3.301
nothing
Token nothing
Feature activation+0.105
I
Token I
Feature activation+0.905
can
Token can
Feature activation+0.409
do
Token do
Feature activation+0.000
,
Token,
Feature activation+0.000
once
Token once
Feature activation+0.000
genocide
Token genocide
Feature activation+0.000
is
Token is
Feature activation+1.621
acceptable
Token acceptable
Feature activation+0.340
,
Token,
Feature activation+4.411
what
Token what
Feature activation+1.427
isn
Token isn
Feature activation+0.000
't
Token't
Feature activation+0.000
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
once
Token once
Feature activation+0.721
that
Token that
Feature activation+0.000
lease
Token lease
Feature activation+0.000
runs
Token runs
Feature activation+1.769
out
Token out
Feature activation+0.171
,
Token,
Feature activation+4.282
the
Token the
Feature activation+4.077
Bucks
Token Bucks
Feature activation+0.000
will
Token will
Feature activation+1.486
be
Token be
Feature activation+0.000
playing
Token playing
Feature activation+0.000

INTERVAL 3.533 - 3.975
CONTAINS 0.001%

.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
that
Token that
Feature activation+0.286
was
Token was
Feature activation+0.819
done
Token done
Feature activation+3.411
I
Token I
Feature activation+3.941
stamped
Token stamped
Feature activation+1.047
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
shape
Token shape
Feature activation+0.000
(
Token (
Feature activation+0.000
a
Token a
Feature activation+0.331
week
Token week
Feature activation+0.000
,
Token,
Feature activation+2.273
and
Token and
Feature activation+0.000
we
Token we
Feature activation+1.568
've
Token've
Feature activation+3.572
had
Token had
Feature activation+0.000
lines
Token lines
Feature activation+0.000
out
Token out
Feature activation+0.000
the
Token the
Feature activation+0.000
door
Token door
Feature activation+0.000
go
Token go
Feature activation+1.868
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
game
Token game
Feature activation+0.000
,
Token,
Feature activation+3.549
it
Token it
Feature activation+3.543
becomes
Token becomes
Feature activation+2.041
much
Token much
Feature activation+0.135
easier
Token easier
Feature activation+0.479
for
Token for
Feature activation+0.000
us
Token us
Feature activation+0.000
placebo
Token placebo
Feature activation+0.000
and
Token and
Feature activation+0.000
drug
Token drug
Feature activation+0.000
withdrawal
Token withdrawal
Feature activation+0.000
effects
Token effects
Feature activation+0.000
are
Token are
Feature activation+3.702
accounted
Token accounted
Feature activation+3.367
for
Token for
Feature activation+0.000
,
Token,
Feature activation+3.221
there
Token there
Feature activation+3.330
is
Token is
Feature activation+1.282
the
Token the
Feature activation+0.737
print
Token print
Feature activation+0.000
heads
Token heads
Feature activation+0.000
are
Token are
Feature activation+1.811
clean
Token clean
Feature activation+2.800
,
Token,
Feature activation+3.663
you
Token you
Feature activation+3.095
can
Token can
Feature activation+1.855
enjoy
Token enjoy
Feature activation+0.000
higher
Token higher
Feature activation+0.000
performance
Token performance
Feature activation+0.000

INTERVAL 3.092 - 3.533
CONTAINS 0.003%

around
Token around
Feature activation+0.000
the
Token the
Feature activation+0.009
guys
Token guys
Feature activation+0.000
,
Token,
Feature activation+3.284
it
Token it
Feature activation+3.935
becomes
Token becomes
Feature activation+3.276
family
Token family
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
guys
Token guys
Feature activation+0.000
on
Token on
Feature activation+0.000
's
Token's
Feature activation+2.133
left
Token left
Feature activation+1.034
here
Token here
Feature activation+0.000
,
Token,
Feature activation+3.365
there
Token there
Feature activation+4.030
's
Token's
Feature activation+3.301
nothing
Token nothing
Feature activation+0.105
I
Token I
Feature activation+0.905
can
Token can
Feature activation+0.409
do
Token do
Feature activation+0.000
.
Token.
Feature activation+0.000
you
Token you
Feature activation+3.383
get
Token get
Feature activation+3.395
around
Token around
Feature activation+0.000
the
Token the
Feature activation+0.009
guys
Token guys
Feature activation+0.000
,
Token,
Feature activation+3.284
it
Token it
Feature activation+3.935
becomes
Token becomes
Feature activation+3.276
family
Token family
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
have
Token have
Feature activation+1.509
the
Token the
Feature activation+1.173
update
Token update
Feature activation+0.000
file
Token file
Feature activation+0.000
,
Token,
Feature activation+3.154
we
Token we
Feature activation+3.096
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
be
Token be
Feature activation+0.000
sure
Token sure
Feature activation+0.000
we
Token we
Feature activation+0.837
have
Token have
Feature activation+1.509
the
Token the
Feature activation+1.173
update
Token update
Feature activation+0.000
file
Token file
Feature activation+0.000
,
Token,
Feature activation+3.154
we
Token we
Feature activation+3.096
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
be
Token be
Feature activation+0.000

INTERVAL 2.650 - 3.092
CONTAINS 0.004%

Once
Token Once
Feature activation+0.000
your
Token your
Feature activation+0.141
pace
Token pace
Feature activation+0.000
is
Token is
Feature activation+1.327
set
Token set
Feature activation+1.678
,
Token,
Feature activation+3.026
you
Token you
Feature activation+3.186
're
Token're
Feature activation+1.994
free
Token free
Feature activation+0.020
to
Token to
Feature activation+0.000
concentrate
Token concentrate
Feature activation+0.000
and
Token and
Feature activation+0.000
above
Token above
Feature activation+0.000
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
claimed
Token claimed
Feature activation+0.000
,
Token,
Feature activation+2.685
your
Token your
Feature activation+0.640
account
Token account
Feature activation+0.000
will
Token will
Feature activation+0.395
have
Token have
Feature activation+0.041
50
Token 50
Feature activation+0.000
Once
TokenOnce
Feature activation+0.000
attendees
Token attendees
Feature activation+0.000
took
Token took
Feature activation+0.575
the
Token the
Feature activation+0.183
street
Token street
Feature activation+0.000
,
Token,
Feature activation+2.690
they
Token they
Feature activation+2.702
had
Token had
Feature activation+2.603
a
Token a
Feature activation+0.128
moment
Token moment
Feature activation+0.000
of
Token of
Feature activation+0.000
Once
Token Once
Feature activation+0.000
Salman
Token Salman
Feature activation+0.000
ascended
Token ascended
Feature activation+1.487
the
Token the
Feature activation+1.152
throne
Token throne
Feature activation+0.000
,
Token,
Feature activation+2.867
he
Token he
Feature activation+2.643
made
Token made
Feature activation+0.149
Mu
Token Mu
Feature activation+0.000
q
Tokenq
Feature activation+0.000
rin
Tokenrin
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.944
results
Token results
Feature activation+0.067
are
Token are
Feature activation+2.715
there
Token there
Feature activation+2.757
,
Token,
Feature activation+2.686
markets
Token markets
Feature activation+0.000
will
Token will
Feature activation+0.422
be
Token be
Feature activation+0.000
reassured
Token reassured
Feature activation+0.000
and
Token and
Feature activation+0.000

INTERVAL 2.208 - 2.650
CONTAINS 0.004%

mark
Token mark
Feature activation+0.000
once
Token once
Feature activation+0.000
all
Token all
Feature activation+0.000
estimates
Token estimates
Feature activation+0.000
are
Token are
Feature activation+1.444
verified
Token verified
Feature activation+2.264
;
Token;
Feature activation+0.000
worst
Token worst
Feature activation+0.000
-
Token-
Feature activation+0.000
case
Tokencase
Feature activation+0.000
scenario
Token scenario
Feature activation+0.000
personal
Token personal
Feature activation+0.000
beliefs
Token beliefs
Feature activation+0.000
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.148
he
Token he
Feature activation+1.087
finished
Token finished
Feature activation+2.301
,
Token,
Feature activation+2.830
I
Token I
Feature activation+2.187
offered
Token offered
Feature activation+0.000
my
Token my
Feature activation+0.000
part
Token part
Feature activation+0.000
go
Token go
Feature activation+0.000
to
Token to
Feature activation+0.000
add
Token add
Feature activation+0.000
a
Token a
Feature activation+0.000
block
Token block
Feature activation+0.000
,
Token,
Feature activation+2.628
which
Token which
Feature activation+0.000
can
Token can
Feature activation+0.000
in
Token in
Feature activation+0.000
itself
Token itself
Feature activation+0.000
be
Token be
Feature activation+0.000
more
Token more
Feature activation+0.000
drugs
Token drugs
Feature activation+0.000
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
once
Token once
Feature activation+0.425
in
Token in
Feature activation+2.470
the
Token the
Feature activation+1.441
trap
Token trap
Feature activation+0.000
it
Token it
Feature activation+3.226
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Salman
Token Salman
Feature activation+0.000
ascended
Token ascended
Feature activation+1.487
the
Token the
Feature activation+1.152
throne
Token throne
Feature activation+0.000
,
Token,
Feature activation+2.867
he
Token he
Feature activation+2.643
made
Token made
Feature activation+0.149
Mu
Token Mu
Feature activation+0.000
q
Tokenq
Feature activation+0.000
rin
Tokenrin
Feature activation+0.000
crown
Token crown
Feature activation+0.000

INTERVAL 1.767 - 2.208
CONTAINS 0.005%

this
Token this
Feature activation+0.000
entire
Token entire
Feature activation+0.000
time
Token time
Feature activation+0.000
and
Token and
Feature activation+0.000
once
Token once
Feature activation+0.166
at
Token at
Feature activation+1.883
the
Token the
Feature activation+0.715
jail
Token jail
Feature activation+0.000
,
Token,
Feature activation+1.940
he
Token he
Feature activation+2.331
began
Token began
Feature activation+1.017
Though
Token Though
Feature activation+0.000
once
Token once
Feature activation+0.000
the
Token the
Feature activation+0.749
Terror
Token Terror
Feature activation+0.000
ists
Tokenists
Feature activation+0.000
have
Token have
Feature activation+1.912
control
Token control
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
sites
Token sites
Feature activation+0.000
,
Token,
Feature activation+0.740
at
Token at
Feature activation+0.000
Rover
Token Rover
Feature activation+0.000
for
Token for
Feature activation+0.000
25
Token 25
Feature activation+0.000
years
Token years
Feature activation+0.000
,
Token,
Feature activation+2.010
said
Token said
Feature activation+0.000
:
Token:
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
If
TokenIf
Feature activation+0.000
the
Token the
Feature activation+0.000
butter
Token butter
Feature activation+0.000
mixture
Token mixture
Feature activation+0.000
has
Token has
Feature activation+2.568
been
Token been
Feature activation+1.967
incorporated
Token incorporated
Feature activation+2.132
and
Token and
Feature activation+0.388
your
Token your
Feature activation+0.908
bisc
Token bisc
Feature activation+0.000
uit
Tokenuit
Feature activation+0.000
mixture
Token mixture
Feature activation+0.000
and
Token and
Feature activation+0.000
feel
Token feel
Feature activation+0.000
confident
Token confident
Feature activation+0.000
that
Token that
Feature activation+0.000
once
Token once
Feature activation+1.245
you
Token you
Feature activation+1.913
open
Token open
Feature activation+2.291
the
Token the
Feature activation+1.565
score
Token score
Feature activation+0.000
in
Token in
Feature activation+0.000
Fin
Token Fin
Feature activation+0.000

INTERVAL 1.325 - 1.767
CONTAINS 0.005%

inflation
Token inflation
Feature activation+0.000
once
Token once
Feature activation+0.000
the
Token the
Feature activation+0.000
deficit
Token deficit
Feature activation+0.000
has
Token has
Feature activation+0.382
been
Token been
Feature activation+1.397
dealt
Token dealt
Feature activation+0.000
with
Token with
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
the
Token the
Feature activation+0.000
application
Token application
Feature activation+0.000
once
Token once
Feature activation+0.000
it
Token it
Feature activation+0.608
was
Token was
Feature activation+0.110
deployed
Token deployed
Feature activation+1.344
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
They
TokenThey
Feature activation+0.000
said
Token said
Feature activation+0.000
once
Token once
Feature activation+0.000
genocide
Token genocide
Feature activation+0.000
is
Token is
Feature activation+1.621
acceptable
Token acceptable
Feature activation+0.340
,
Token,
Feature activation+4.411
what
Token what
Feature activation+1.427
isn
Token isn
Feature activation+0.000
't
Token't
Feature activation+0.000
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
your
Token your
Feature activation+0.141
pace
Token pace
Feature activation+0.000
is
Token is
Feature activation+1.327
set
Token set
Feature activation+1.678
,
Token,
Feature activation+3.026
you
Token you
Feature activation+3.186
're
Token're
Feature activation+1.994
free
Token free
Feature activation+0.020
to
Token to
Feature activation+0.000
Cruz
Token Cruz
Feature activation+0.000
Avenue
Token Avenue
Feature activation+0.000
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
he
Token he
Feature activation+1.552
arrived
Token arrived
Feature activation+1.701
,
Token,
Feature activation+2.772
Dev
Token Dev
Feature activation+0.000
inc
Tokeninc
Feature activation+0.000
en
Tokenen
Feature activation+0.000
zi
Tokenzi
Feature activation+0.000

INTERVAL 0.883 - 1.325
CONTAINS 0.009%

,
Token,
Feature activation+2.727
the
Token the
Feature activation+2.834
plants
Token plants
Feature activation+0.013
may
Token may
Feature activation+1.197
be
Token be
Feature activation+0.000
harvested
Token harvested
Feature activation+0.912
for
Token for
Feature activation+0.000
up
Token up
Feature activation+0.000
to
Token to
Feature activation+0.000
20
Token 20
Feature activation+0.000
years
Token years
Feature activation+0.000
Moon
Token Moon
Feature activation+0.000
missions
Token missions
Feature activation+0.000
was
Token was
Feature activation+0.000
once
Token once
Feature activation+0.000
again
Token again
Feature activation+0.000
at
Token at
Feature activation+1.102
one
Token one
Feature activation+0.516
.
Token.
Feature activation+0.000
Forty
Token Forty
Feature activation+0.000
years
Token years
Feature activation+0.000
later
Token later
Feature activation+0.000
once
Token once
Feature activation+0.000
the
Token the
Feature activation+1.209
decks
Token decks
Feature activation+0.000
are
Token are
Feature activation+2.185
cleared
Token cleared
Feature activation+3.326
and
Token and
Feature activation+1.088
both
Token both
Feature activation+0.000
parties
Token parties
Feature activation+0.000
have
Token have
Feature activation+2.588
their
Token their
Feature activation+0.000
candidates
Token candidates
Feature activation+0.000
receives
Token receives
Feature activation+2.313
a
Token a
Feature activation+0.000
mandatory
Token mandatory
Feature activation+0.000
evacuation
Token evacuation
Feature activation+0.000
,
Token,
Feature activation+2.618
and
Token and
Feature activation+1.141
has
Token has
Feature activation+1.496
left
Token left
Feature activation+0.000
,
Token,
Feature activation+0.193
they
Token they
Feature activation+0.807
are
Token are
Feature activation+0.501
them
Token them
Feature activation+0.000
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
cars
Token cars
Feature activation+0.000
became
Token became
Feature activation+0.006
the
Token the
Feature activation+1.015
main
Token main
Feature activation+0.000
form
Token form
Feature activation+0.000
of
Token of
Feature activation+0.000
transportation
Token transportation
Feature activation+0.000
in
Token in
Feature activation+0.000

INTERVAL 0.442 - 0.883
CONTAINS 0.010%

to
Token to
Feature activation+0.000
un
Token un
Feature activation+0.000
see
Tokensee
Feature activation+0.000
once
Token once
Feature activation+0.000
you
Token you
Feature activation+0.000
have
Token have
Feature activation+0.867
seen
Token seen
Feature activation+0.613
them
Token them
Feature activation+0.000
;
Token;
Feature activation+0.000
ideas
Token ideas
Feature activation+0.000
that
Token that
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
c
Tokenc
Feature activation+0.000
ame
Tokename
Feature activation+1.296
forth
Token forth
Feature activation+0.000
,
Token,
Feature activation+0.765
but
Token but
Feature activation+0.000
the
Token the
Feature activation+0.247
heat
Token heat
Feature activation+0.000
was
Token was
Feature activation+0.059
so
Token so
Feature activation+0.000
into
Token into
Feature activation+0.000
custody
Token custody
Feature activation+0.000
once
Token once
Feature activation+0.000
he
Token he
Feature activation+0.593
left
Token left
Feature activation+0.000
the
Token the
Feature activation+0.583
consulate
Token consulate
Feature activation+0.000
for
Token for
Feature activation+0.000
entering
Token entering
Feature activation+0.000
the
Token the
Feature activation+0.000
diplomatic
Token diplomatic
Feature activation+0.000
pretty
Token pretty
Feature activation+0.000
much
Token much
Feature activation+0.000
left
Token left
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
dark
Token dark
Feature activation+0.468
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
could
Token could
Feature activation+0.000
fake
Token fake
Feature activation+0.000
it
Token it
Feature activation+0.000
the
Token the
Feature activation+0.000
classroom
Token classroom
Feature activation+0.000
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
that
Token that
Feature activation+0.000
was
Token was
Feature activation+0.531
under
Token under
Feature activation+1.460
control
Token control
Feature activation+0.000
,
Token,
Feature activation+3.078
Meyer
Token Meyer
Feature activation+0.000
and
Token and
Feature activation+0.000

INTERVAL 0.000 - 0.442
CONTAINS 99.958%

is
Token is
Feature activation+0.000
chosen
Token chosen
Feature activation+0.000
we
Token we
Feature activation+0.000
will
Token will
Feature activation+0.000
modify
Token modify
Feature activation+0.000
the
Token the
Feature activation+0.000
shipping
Token shipping
Feature activation+0.000
method
Token method
Feature activation+0.000
.
Token.
Feature activation+0.000
Live
Token Live
Feature activation+0.000
Rock
Token Rock
Feature activation+0.000
which
Token which
Feature activation+0.000
by
Token by
Feature activation+0.000
implication
Token implication
Feature activation+0.000
means
Token means
Feature activation+0.000
that
Token that
Feature activation+0.000
you
Token you
Feature activation+0.000
one
Token one
Feature activation+0.000
day
Token day
Feature activation+0.000
will
Token will
Feature activation+0.000
also
Token also
Feature activation+0.000
die
Token die
Feature activation+0.000
suspicion
Token suspicion
Feature activation+0.000
of
Token of
Feature activation+0.000
inst
Token inst
Feature activation+0.000
igating
Tokenigating
Feature activation+0.000
the
Token the
Feature activation+0.000
flag
Token flag
Feature activation+0.000
stunt
Token stunt
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Al
TokenAl
Feature activation+0.000
even
Token even
Feature activation+0.000
coherent
Token coherent
Feature activation+0.000
,
Token,
Feature activation+0.000
Limbaugh
Token Limbaugh
Feature activation+0.000
is
Token is
Feature activation+0.000
attempting
Token attempting
Feature activation+0.000
to
Token to
Feature activation+0.000
tell
Token tell
Feature activation+0.000
white
Token white
Feature activation+0.000
people
Token people
Feature activation+0.000
that
Token that
Feature activation+0.000
she
Token she
Feature activation+0.000
believes
Token believes
Feature activation+0.000
she
Token she
Feature activation+0.000
and
Token and
Feature activation+0.000
her
Token her
Feature activation+0.000
friend
Token friend
Feature activation+0.000
named
Token named
Feature activation+0.000
Sh
Token Sh
Feature activation+0.000
auna
Tokenauna
Feature activation+0.000
are
Token are
Feature activation+0.000
being
Token being
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
Development
Token Development
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
&
Token &
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Defence
Token Defence
Feature activation+0.000
Research
Token Research
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
rik
Tokenrik
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
said
Token said
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
atre
Tokenatre
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
former
Token former
Feature activation+0.000
boss
Token boss
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
K
TokenK
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
V
Token V
Feature activation+0.000
.
Token.
Feature activation+0.000