H3.4 top features

Top feature 0 in H3.4: (feature 7617)

TOP ACTIVATIONS
MAX = 2.817

McMahon
Token McMahon
Feature activation+0.000
potentially
Token potentially
Feature activation+0.000
getting
Token getting
Feature activation+0.000
back
Token back
Feature activation+0.000
into
Token into
Feature activation+0.000
football
Token football
Feature activation+2.817
:
Token:
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
Mc
TokenMc
Feature activation+0.000
was
Token was
Feature activation+0.000
willing
Token willing
Feature activation+0.000
to
Token to
Feature activation+0.000
let
Token let
Feature activation+0.000
the
Token the
Feature activation+0.000
NFL
Token NFL
Feature activation+2.760
sell
Token sell
Feature activation+0.000
the
Token the
Feature activation+0.000
bulk
Token bulk
Feature activation+0.000
of
Token of
Feature activation+0.000
ads
Token ads
Feature activation+0.000
those
Token those
Feature activation+0.052
games
Token games
Feature activation+0.381
,
Token,
Feature activation+0.000
outside
Token outside
Feature activation+0.000
of
Token of
Feature activation+0.000
football
Token football
Feature activation+2.596
,
Token,
Feature activation+0.000
are
Token are
Feature activation+0.000
not
Token not
Feature activation+0.000
easy
Token easy
Feature activation+0.000
to
Token to
Feature activation+0.000
Joseph
Token Joseph
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
Thunder
Token Thunder
Feature activation+2.559
were
Token were
Feature activation+0.000
not
Token not
Feature activation+0.000
just
Token just
Feature activation+0.000
some
Token some
Feature activation+0.000
Vegas
Token Vegas
Feature activation+0.000
to
Token to
Feature activation+0.000
stream
Token stream
Feature activation+0.000
live
Token live
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
Night
Token Night
Feature activation+0.000
Football
Token Football
Feature activation+2.504
games
Token games
Feature activation+0.462
over
Token over
Feature activation+0.000
rivals
Token rivals
Feature activation+0.000
like
Token like
Feature activation+0.000
Facebook
Token Facebook
Feature activation+0.000
the
Token the
Feature activation+0.000
day
Token day
Feature activation+0.000
of
Token of
Feature activation+0.000
!
Token!
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Toronto
Token Toronto
Feature activation+2.320
MT
Token MT
Feature activation+2.000
B
TokenB
Feature activation+0.011
Fest
Token Fest
Feature activation+0.487
is
Token is
Feature activation+0.000
run
Token run
Feature activation+0.000
,
Token,
Feature activation+0.000
both
Token both
Feature activation+0.000
in
Token in
Feature activation+0.000
MMA
Token MMA
Feature activation+0.000
and
Token and
Feature activation+0.000
football
Token football
Feature activation+2.267
,
Token,
Feature activation+0.000
left
Token left
Feature activation+0.000
him
Token him
Feature activation+0.000
struggling
Token struggling
Feature activation+0.000
with
Token with
Feature activation+0.000
USA
Token USA
Feature activation+0.000
women
Token women
Feature activation+0.442
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
basketball
Token basketball
Feature activation+2.267
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
handled
Token handled
Feature activation+0.000
by
Token by
Feature activation+0.000
CS
Token CS
Feature activation+0.338
reforming
Token reforming
Feature activation+0.000
in
Token in
Feature activation+0.000
2002
Token 2002
Feature activation+0.000
and
Token and
Feature activation+0.000
lifelong
Token lifelong
Feature activation+0.000
Liverpool
Token Liverpool
Feature activation+2.146
fan
Token fan
Feature activation+0.103
Akin
Token Akin
Feature activation+0.946
fen
Tokenfen
Feature activation+0.000
wa
Tokenwa
Feature activation+0.000
is
Token is
Feature activation+0.000
of
Token of
Feature activation+0.000
margins
Token margins
Feature activation+0.111
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
title
Token title
Feature activation+0.652
fight
Token fight
Feature activation+2.057
,
Token,
Feature activation+0.000
just
Token just
Feature activation+0.000
one
Token one
Feature activation+0.000
point
Token point
Feature activation+0.000
separating
Token separating
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
championships
Token championships
Feature activation+0.000
,
Token,
Feature activation+0.000
subsequently
Token subsequently
Feature activation+0.000
played
Token played
Feature activation+2.056
down
Token down
Feature activation+0.132
such
Token such
Feature activation+0.000
talk
Token talk
Feature activation+0.000
and
Token and
Feature activation+0.000
Monte
Token Monte
Feature activation+0.098
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
film
Token film
Feature activation+1.642
and
Token and
Feature activation+0.000
her
Token her
Feature activation+0.000
musical
Token musical
Feature activation+2.022
work
Token work
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
day
Token day
Feature activation+0.000
of
Token of
Feature activation+0.000
!
Token!
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Toronto
Token Toronto
Feature activation+2.320
MT
Token MT
Feature activation+2.000
B
TokenB
Feature activation+0.011
Fest
Token Fest
Feature activation+0.487
is
Token is
Feature activation+0.000
run
Token run
Feature activation+0.000
by
Token by
Feature activation+0.000
with
Token with
Feature activation+0.000
Ford
Token Ford
Feature activation+0.372
they
Token they
Feature activation+0.000
built
Token built
Feature activation+0.526
the
Token the
Feature activation+0.000
DF
Token DF
Feature activation+1.979
V
TokenV
Feature activation+0.077
engine
Token engine
Feature activation+0.012
specially
Token specially
Feature activation+0.000
for
Token for
Feature activation+0.000
this
Token this
Feature activation+0.000
ON
TokenON
Feature activation+0.000
and
Token and
Feature activation+0.000
say
Token say
Feature activation+0.000
that
Token that
Feature activation+0.000
was
Token was
Feature activation+0.000
Moose
Token Moose
Feature activation+1.804
as
Token as
Feature activation+0.000
far
Token far
Feature activation+0.000
as
Token as
Feature activation+0.000
R
Token R
Feature activation+0.000
OH
TokenOH
Feature activation+0.000
people
Token people
Feature activation+0.000
believe
Token believe
Feature activation+0.000
this
Token this
Feature activation+0.000
was
Token was
Feature activation+0.000
the
Token the
Feature activation+0.000
Wild
Token Wild
Feature activation+1.753
cat
Tokencat
Feature activation+0.457
formation
Token formation
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
starting
Token starting
Feature activation+0.000
driver
Token driver
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
United
Token United
Feature activation+1.735
States
Token States
Feature activation+0.400
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
without
Token without
Feature activation+0.000
a
Token a
Feature activation+0.000
of
Token of
Feature activation+0.000
college
Token college
Feature activation+0.000
sports
Token sports
Feature activation+0.321
is
Token is
Feature activation+0.000
that
Token that
Feature activation+0.000
football
Token football
Feature activation+1.732
and
Token and
Feature activation+0.000
men
Token men
Feature activation+0.112
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
v
Tokenv
Feature activation+0.000
idal
Tokenidal
Feature activation+0.000
to
Token to
Feature activation+0.000
resort
Token resort
Feature activation+0.000
to
Token to
Feature activation+0.000
throwing
Token throwing
Feature activation+1.723
strikes
Token strikes
Feature activation+0.000
without
Token without
Feature activation+0.000
much
Token much
Feature activation+0.000
impact
Token impact
Feature activation+0.000
.
Token.
Feature activation+0.000
included
Token included
Feature activation+0.000
opportunities
Token opportunities
Feature activation+0.000
in
Token in
Feature activation+0.000
Major
Token Major
Feature activation+0.595
League
Token League
Feature activation+0.000
Baseball
Token Baseball
Feature activation+1.674
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
I
TokenI
Feature activation+0.000
can
Token can
Feature activation+0.000

Top DFA by src position
MAX = 3.171

<|endoftext|>
Token<|endoftext|>
Feature activation+0.032
Top resid features:
pro
Token pro
Feature activation+0.007
Top resid features:
wrestling
Token wrestling
Feature activation+3.171
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Here
Token Here
Feature activation-0.001
Top resid features:
are
Token are
Feature activation+0.001
Top resid features:
some
Token some
Feature activation-0.001
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
testing
Token testing
Feature activation+0.001
Top resid features:
live
Token live
Feature activation+0.000
Top resid features:
streams
Token streams
Feature activation+0.001
Top resid features:
during
Token during
Feature activation-0.000
Top resid features:
Wim
Token Wim
Feature activation+0.075
Top resid features:
bledon
Tokenbledon
Feature activation+3.168
Top resid features:
(
Token (
Feature activation+0.000
Top resid features:
from
Tokenfrom
Feature activation-0.000
Top resid features:
a
Token a
Feature activation-0.000
Top resid features:
special
Token special
Feature activation+0.000
Top resid features:
ESPN
Token ESPN
Feature activation+0.007
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
D
Token D
Feature activation+0.001
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
1
Token1
Feature activation+0.005
Top resid features:
NCAA
Token NCAA
Feature activation+0.163
Top resid features:
basketball
Token basketball
Feature activation+2.768
Top resid features:
tournament
Token tournament
Feature activation+0.083
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
I
Token I
Feature activation+0.000
Top resid features:
started
Token started
Feature activation-0.001
Top resid features:
thinking
Token thinking
Feature activation+0.000
Top resid features:
seasons
Token seasons
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
with
Token with
Feature activation+0.001
Top resid features:
several
Token several
Feature activation-0.001
Top resid features:
future
Token future
Feature activation+0.002
Top resid features:
NHL
Token NHL
Feature activation+3.017
Top resid features:
stars
Token stars
Feature activation+0.019
Top resid features:
dot
Token dot
Feature activation+0.000
Top resid features:
ting
Tokenting
Feature activation-0.000
Top resid features:
its
Token its
Feature activation+0.002
Top resid features:
roster
Token roster
Feature activation+0.006
Top resid features:
testing
Token testing
Feature activation+0.003
Top resid features:
live
Token live
Feature activation-0.002
Top resid features:
streams
Token streams
Feature activation-0.000
Top resid features:
during
Token during
Feature activation-0.001
Top resid features:
Wim
Token Wim
Feature activation+0.107
Top resid features:
bledon
Tokenbledon
Feature activation+2.949
Top resid features:
(
Token (
Feature activation+0.002
Top resid features:
from
Tokenfrom
Feature activation-0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
special
Token special
Feature activation+0.001
Top resid features:
ESPN
Token ESPN
Feature activation+0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.060
Top resid features:
Lot
TokenLot
Feature activation-0.006
Top resid features:
Racing
Token Racing
Feature activation+2.793
Top resid features:
It
TokenIt
Feature activation-0.001
Top resid features:
's
Token's
Feature activation+0.002
Top resid features:
going
Token going
Feature activation+0.000
Top resid features:
to
Token to
Feature activation-0.005
Top resid features:
be
Token be
Feature activation+0.001
Top resid features:
his
Token his
Feature activation+0.016
Top resid features:
career
Token career
Feature activation+0.006
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
both
Token both
Feature activation-0.002
Top resid features:
in
Token in
Feature activation-0.025
Top resid features:
MMA
Token MMA
Feature activation+2.809
Top resid features:
and
Token and
Feature activation+0.012
Top resid features:
football
Token football
Feature activation-0.068
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
left
Token left
Feature activation+0.000
Top resid features:
him
Token him
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Albert
Token Albert
Feature activation+0.005
Top resid features:
also
Token also
Feature activation-0.001
Top resid features:
called
Token called
Feature activation+0.001
Top resid features:
Olympic
Token Olympic
Feature activation+0.040
Top resid features:
boxing
Token boxing
Feature activation+2.844
Top resid features:
for
Token for
Feature activation+0.001
Top resid features:
NBC
Token NBC
Feature activation+0.022
Top resid features:
in
Token in
Feature activation+0.001
Top resid features:
1988
Token 1988
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
League
TokenLeague
Feature activation+0.028
Top resid features:
Two
Token Two
Feature activation+0.001
Top resid features:
Wim
Token Wim
Feature activation+0.066
Top resid features:
bledon
Tokenbledon
Feature activation+2.562
Top resid features:
have
Token have
Feature activation-0.002
Top resid features:
made
Token made
Feature activation-0.001
Top resid features:
giant
Token giant
Feature activation-0.001
Top resid features:
strides
Token strides
Feature activation-0.003
Top resid features:
since
Token since
Feature activation-0.001
Top resid features:
l
Tokenl
Feature activation+0.000
Top resid features:
op
Tokenop
Feature activation+0.001
Top resid features:
M
Token M
Feature activation+0.000
Top resid features:
SA
TokenSA
Feature activation+0.014
Top resid features:
British
Token British
Feature activation+0.001
Top resid features:
Tour
Token Tour
Feature activation+1.307
Top resid features:
ing
Tokening
Feature activation+0.067
Top resid features:
Car
Token Car
Feature activation+0.198
Top resid features:
Championship
Token Championship
Feature activation+0.216
Top resid features:
returns
Token returns
Feature activation+0.000
Top resid features:
at
Token at
Feature activation-0.003
Top resid features:
Hor
Token Hor
Feature activation+0.000
Top resid features:
ner
Tokenner
Feature activation+0.000
Top resid features:
could
Token could
Feature activation+0.000
Top resid features:
replace
Token replace
Feature activation-0.001
Top resid features:
Bernie
Token Bernie
Feature activation+0.001
Top resid features:
Ecc
Token Ecc
Feature activation+0.830
Top resid features:
l
Tokenl
Feature activation-0.000
Top resid features:
estone
Tokenestone
Feature activation+0.001
Top resid features:
as
Token as
Feature activation-0.002
Top resid features:
Formula
Token Formula
Feature activation+0.178
Top resid features:
One
Token One
Feature activation+0.075
Top resid features:
Hudson
Token Hudson
Feature activation-0.000
Top resid features:
,
Token,
Feature activation-0.001
Top resid features:
engaged
Token engaged
Feature activation-0.000
Top resid features:
to
Token to
Feature activation-0.002
Top resid features:
current
Token current
Feature activation+0.000
Top resid features:
WWE
Token WWE
Feature activation+2.521
Top resid features:
employee
Token employee
Feature activation+0.002
Top resid features:
David
Token David
Feature activation-0.000
Top resid features:
Ot
Token Ot
Feature activation+0.001
Top resid features:
ung
Tokenung
Feature activation-0.000
Top resid features:
a
Tokena
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.069
Top resid features:
Lot
TokenLot
Feature activation-0.003
Top resid features:
Racing
Token Racing
Feature activation+2.544
Top resid features:
It
TokenIt
Feature activation-0.003
Top resid features:
's
Token's
Feature activation+0.001
Top resid features:
going
Token going
Feature activation-0.000
Top resid features:
to
Token to
Feature activation-0.002
Top resid features:
be
Token be
Feature activation+0.001
Top resid features:
was
Token was
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.004
Top resid features:
first
Token first
Feature activation-0.001
Top resid features:
real
Token real
Feature activation+0.001
Top resid features:
F
Token F
Feature activation+0.008
Top resid features:
1
Token1
Feature activation+1.266
Top resid features:
car
Token car
Feature activation+0.369
Top resid features:
that
Token that
Feature activation-0.003
Top resid features:
included
Token included
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
engine
Token engine
Feature activation+0.074
Top resid features:
Adam
Token Adam
Feature activation-0.013
Top resid features:
Page
Token Page
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.005
Top resid features:
most
Token most
Feature activation-0.000
Top resid features:
improved
Token improved
Feature activation+0.003
Top resid features:
wrestler
Token wrestler
Feature activation+2.237
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
year
Token year
Feature activation+0.001
Top resid features:
,
Token,
Feature activation-0.000
Top resid features:
I
Token I
Feature activation+0.001
Top resid features:
who
Token who
Feature activation+0.001
Top resid features:
was
Token was
Feature activation+0.000
Top resid features:
lined
Token lined
Feature activation+0.000
Top resid features:
up
Token up
Feature activation-0.001
Top resid features:
as
Token as
Feature activation-0.001
Top resid features:
quarterback
Token quarterback
Feature activation+1.527
Top resid features:
?
Token?
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
From
TokenFrom
Feature activation+0.000
Top resid features:
Video
Token Video
Feature activation+0.000
Top resid features:
full
Token full
Feature activation+0.001
Top resid features:
-
Token-
Feature activation-0.002
Top resid features:
time
Tokentime
Feature activation-0.003
Top resid features:
F
Token F
Feature activation-0.000
Top resid features:
1
Token1
Feature activation+1.139
Top resid features:
driver
Token driver
Feature activation+1.142
Top resid features:
.
Token.
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.008
Top resid features:
The
TokenThe
Feature activation+0.009
Top resid features:
United
Token United
Feature activation-0.082
Top resid features:
s
Tokens
Feature activation-0.000
Top resid features:
D
Token D
Feature activation-0.002
Top resid features:
-
Token-
Feature activation+0.003
Top resid features:
1
Token1
Feature activation+0.014
Top resid features:
NCAA
Token NCAA
Feature activation+0.127
Top resid features:
basketball
Token basketball
Feature activation+1.485
Top resid features:
tournament
Token tournament
Feature activation+0.125
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
I
Token I
Feature activation-0.001
Top resid features:
started
Token started
Feature activation+0.000
Top resid features:
thinking
Token thinking
Feature activation-0.001
Top resid features:
clinch
Token clinch
Feature activation+0.018
Top resid features:
on
Token on
Feature activation+0.002
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
more
Token more
Feature activation+0.002
Top resid features:
skilled
Token skilled
Feature activation-0.001
Top resid features:
striker
Token striker
Feature activation+2.281
Top resid features:
,
Token,
Feature activation+0.004
Top resid features:
Henderson
Token Henderson
Feature activation-0.003
Top resid features:
forced
Token forced
Feature activation-0.011
Top resid features:
Mas
Token Mas
Feature activation+0.013
Top resid features:
v
Tokenv
Feature activation-0.004
Top resid features:
months
Token months
Feature activation-0.000
Top resid features:
looked
Token looked
Feature activation-0.000
Top resid features:
into
Token into
Feature activation-0.001
Top resid features:
buying
Token buying
Feature activation+0.002
Top resid features:
a
Token a
Feature activation-0.002
Top resid features:
hockey
Token hockey
Feature activation+1.556
Top resid features:
franchise
Token franchise
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
with
Token with
Feature activation+0.000
Top resid features:
Las
Token Las
Feature activation+0.008
Top resid features:
Vegas
Token Vegas
Feature activation+0.007
Top resid features:

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.05

Head 2: 0.04

Head 3: 0.02

Head 4: 0.50

Head 5: 0.06

Head 6: 0.05

Head 7: 0.05

Head 8: 0.03

Head 9: 0.03

Head 10: 0.05

Head 11: 0.04

Positive logits

fetched2.08

zn2.01

render1.94

Hernandez1.78

intervening1.71

alus1.66

oman1.66

staking1.65

clot1.64

ibly1.64

separate1.62

imus1.62

ively1.60

eval1.60

ulla1.59

Tas1.59

Ples1.59

terrestrial1.59

Perez1.58

enser1.58

Negative logits

CoC-2.28

merce-2.21

RFC-2.12

senal-2.10

thood-2.09

yss-2.06

��-2.05

Maker-2.04

Leilan-1.98

-1.97

Masquerade-1.96

BIP-1.92

uliffe-1.92

��-1.90

-1.90

-1.88

anship-1.86

Ascension-1.85

Celebr-1.83

CCC-1.82

INTERVAL 2.535 - 2.817
CONTAINS 0.000%

those
Token those
Feature activation+0.052
games
Token games
Feature activation+0.381
,
Token,
Feature activation+0.000
outside
Token outside
Feature activation+0.000
of
Token of
Feature activation+0.000
football
Token football
Feature activation+2.596
,
Token,
Feature activation+0.000
are
Token are
Feature activation+0.000
not
Token not
Feature activation+0.000
easy
Token easy
Feature activation+0.000
to
Token to
Feature activation+0.000
Joseph
Token Joseph
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
Thunder
Token Thunder
Feature activation+2.559
were
Token were
Feature activation+0.000
not
Token not
Feature activation+0.000
just
Token just
Feature activation+0.000
some
Token some
Feature activation+0.000
Vegas
Token Vegas
Feature activation+0.000
McMahon
Token McMahon
Feature activation+0.000
potentially
Token potentially
Feature activation+0.000
getting
Token getting
Feature activation+0.000
back
Token back
Feature activation+0.000
into
Token into
Feature activation+0.000
football
Token football
Feature activation+2.817
:
Token:
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
Mc
TokenMc
Feature activation+0.000
was
Token was
Feature activation+0.000
willing
Token willing
Feature activation+0.000
to
Token to
Feature activation+0.000
let
Token let
Feature activation+0.000
the
Token the
Feature activation+0.000
NFL
Token NFL
Feature activation+2.760
sell
Token sell
Feature activation+0.000
the
Token the
Feature activation+0.000
bulk
Token bulk
Feature activation+0.000
of
Token of
Feature activation+0.000
ads
Token ads
Feature activation+0.000

INTERVAL 2.254 - 2.535
CONTAINS 0.000%

USA
Token USA
Feature activation+0.000
women
Token women
Feature activation+0.442
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
basketball
Token basketball
Feature activation+2.267
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
handled
Token handled
Feature activation+0.000
by
Token by
Feature activation+0.000
CS
Token CS
Feature activation+0.338
,
Token,
Feature activation+0.000
both
Token both
Feature activation+0.000
in
Token in
Feature activation+0.000
MMA
Token MMA
Feature activation+0.000
and
Token and
Feature activation+0.000
football
Token football
Feature activation+2.267
,
Token,
Feature activation+0.000
left
Token left
Feature activation+0.000
him
Token him
Feature activation+0.000
struggling
Token struggling
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
day
Token day
Feature activation+0.000
of
Token of
Feature activation+0.000
!
Token!
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Toronto
Token Toronto
Feature activation+2.320
MT
Token MT
Feature activation+2.000
B
TokenB
Feature activation+0.011
Fest
Token Fest
Feature activation+0.487
is
Token is
Feature activation+0.000
run
Token run
Feature activation+0.000
to
Token to
Feature activation+0.000
stream
Token stream
Feature activation+0.000
live
Token live
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
Night
Token Night
Feature activation+0.000
Football
Token Football
Feature activation+2.504
games
Token games
Feature activation+0.462
over
Token over
Feature activation+0.000
rivals
Token rivals
Feature activation+0.000
like
Token like
Feature activation+0.000
Facebook
Token Facebook
Feature activation+0.000

INTERVAL 1.972 - 2.254
CONTAINS 0.001%

with
Token with
Feature activation+0.000
Ford
Token Ford
Feature activation+0.372
they
Token they
Feature activation+0.000
built
Token built
Feature activation+0.526
the
Token the
Feature activation+0.000
DF
Token DF
Feature activation+1.979
V
TokenV
Feature activation+0.077
engine
Token engine
Feature activation+0.012
specially
Token specially
Feature activation+0.000
for
Token for
Feature activation+0.000
this
Token this
Feature activation+0.000
of
Token of
Feature activation+0.000
margins
Token margins
Feature activation+0.111
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
title
Token title
Feature activation+0.652
fight
Token fight
Feature activation+2.057
,
Token,
Feature activation+0.000
just
Token just
Feature activation+0.000
one
Token one
Feature activation+0.000
point
Token point
Feature activation+0.000
separating
Token separating
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
film
Token film
Feature activation+1.642
and
Token and
Feature activation+0.000
her
Token her
Feature activation+0.000
musical
Token musical
Feature activation+2.022
work
Token work
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
day
Token day
Feature activation+0.000
of
Token of
Feature activation+0.000
!
Token!
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Toronto
Token Toronto
Feature activation+2.320
MT
Token MT
Feature activation+2.000
B
TokenB
Feature activation+0.011
Fest
Token Fest
Feature activation+0.487
is
Token is
Feature activation+0.000
run
Token run
Feature activation+0.000
by
Token by
Feature activation+0.000
reforming
Token reforming
Feature activation+0.000
in
Token in
Feature activation+0.000
2002
Token 2002
Feature activation+0.000
and
Token and
Feature activation+0.000
lifelong
Token lifelong
Feature activation+0.000
Liverpool
Token Liverpool
Feature activation+2.146
fan
Token fan
Feature activation+0.103
Akin
Token Akin
Feature activation+0.946
fen
Tokenfen
Feature activation+0.000
wa
Tokenwa
Feature activation+0.000
is
Token is
Feature activation+0.000

INTERVAL 1.690 - 1.972
CONTAINS 0.001%

v
Tokenv
Feature activation+0.000
idal
Tokenidal
Feature activation+0.000
to
Token to
Feature activation+0.000
resort
Token resort
Feature activation+0.000
to
Token to
Feature activation+0.000
throwing
Token throwing
Feature activation+1.723
strikes
Token strikes
Feature activation+0.000
without
Token without
Feature activation+0.000
much
Token much
Feature activation+0.000
impact
Token impact
Feature activation+0.000
.
Token.
Feature activation+0.000
of
Token of
Feature activation+0.000
college
Token college
Feature activation+0.000
sports
Token sports
Feature activation+0.321
is
Token is
Feature activation+0.000
that
Token that
Feature activation+0.000
football
Token football
Feature activation+1.732
and
Token and
Feature activation+0.000
men
Token men
Feature activation+0.112
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
ON
TokenON
Feature activation+0.000
and
Token and
Feature activation+0.000
say
Token say
Feature activation+0.000
that
Token that
Feature activation+0.000
was
Token was
Feature activation+0.000
Moose
Token Moose
Feature activation+1.804
as
Token as
Feature activation+0.000
far
Token far
Feature activation+0.000
as
Token as
Feature activation+0.000
R
Token R
Feature activation+0.000
OH
TokenOH
Feature activation+0.000
driver
Token driver
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
United
Token United
Feature activation+1.735
States
Token States
Feature activation+0.400
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
without
Token without
Feature activation+0.000
a
Token a
Feature activation+0.000
people
Token people
Feature activation+0.000
believe
Token believe
Feature activation+0.000
this
Token this
Feature activation+0.000
was
Token was
Feature activation+0.000
the
Token the
Feature activation+0.000
Wild
Token Wild
Feature activation+1.753
cat
Tokencat
Feature activation+0.457
formation
Token formation
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
starting
Token starting
Feature activation+0.000

INTERVAL 1.409 - 1.690
CONTAINS 0.002%

come
Token come
Feature activation+0.000
to
Token to
Feature activation+0.000
college
Token college
Feature activation+0.004
and
Token and
Feature activation+0.000
give
Token give
Feature activation+0.000
football
Token football
Feature activation+1.562
a
Token a
Feature activation+0.000
try
Token try
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
for
Token for
Feature activation+0.000
her
Token her
Feature activation+0.000
role
Token role
Feature activation+0.149
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
film
Token film
Feature activation+1.642
and
Token and
Feature activation+0.000
her
Token her
Feature activation+0.000
musical
Token musical
Feature activation+2.022
work
Token work
Feature activation+0.000
.
Token.
Feature activation+0.000
included
Token included
Feature activation+0.000
opportunities
Token opportunities
Feature activation+0.000
in
Token in
Feature activation+0.000
Major
Token Major
Feature activation+0.595
League
Token League
Feature activation+0.000
Baseball
Token Baseball
Feature activation+1.674
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
I
TokenI
Feature activation+0.000
can
Token can
Feature activation+0.000
ACT
Token ACT
Feature activation+0.355
."
Token."
Feature activation+0.104
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Mc
TokenMc
Feature activation+0.290
Connell
TokenConnell
Feature activation+1.468
played
Token played
Feature activation+0.000
37
Token 37
Feature activation+0.000
games
Token games
Feature activation+0.000
for
Token for
Feature activation+0.000
Foot
Token Foot
Feature activation+0.241
something
Token something
Feature activation+0.000
like
Token like
Feature activation+0.000
this
Token this
Feature activation+0.000
:
Token:
Feature activation+0.000
Aaron
Token Aaron
Feature activation+0.000
Rodgers
Token Rodgers
Feature activation+1.479
would
Token would
Feature activation+0.000
lead
Token lead
Feature activation+0.000
an
Token an
Feature activation+0.000
unstoppable
Token unstoppable
Feature activation+0.000
offense
Token offense
Feature activation+0.000

INTERVAL 1.127 - 1.409
CONTAINS 0.003%

JJ
TokenJJ
Feature activation+0.447
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
combining
Token combining
Feature activation+0.000
and
Token and
Feature activation+0.000
string
Token string
Feature activation+1.396
ing
Tokening
Feature activation+0.000
together
Token together
Feature activation+0.000
moves
Token moves
Feature activation+1.007
/
Token/
Feature activation+0.000
pos
Tokenpos
Feature activation+0.000
football
Token football
Feature activation+0.000
.
Token.
Feature activation+0.000
Oregon
Token Oregon
Feature activation+0.000
shot
Token shot
Feature activation+1.668
their
Token their
Feature activation+0.000
shot
Token shot
Feature activation+1.374
and
Token and
Feature activation+0.000
gave
Token gave
Feature activation+0.000
him
Token him
Feature activation+0.000
an
Token an
Feature activation+0.000
offer
Token offer
Feature activation+0.000
.
Token.
Feature activation+0.098
Then
Token Then
Feature activation+0.000
in
Token in
Feature activation+0.000
2012
Token 2012
Feature activation+0.000
he
Token he
Feature activation+0.000
made
Token made
Feature activation+1.250
a
Token a
Feature activation+0.000
dubious
Token dubious
Feature activation+0.000
move
Token move
Feature activation+0.130
on
Token on
Feature activation+0.000
Sergio
Token Sergio
Feature activation+0.000
2005
Token 2005
Feature activation+0.000
business
Token business
Feature activation+0.280
deal
Token deal
Feature activation+0.000
,
Token,
Feature activation+0.000
told
Token told
Feature activation+0.000
British
Token British
Feature activation+1.151
newspapers
Token newspapers
Feature activation+0.000
in
Token in
Feature activation+0.000
Brazil
Token Brazil
Feature activation+0.284
last
Token last
Feature activation+0.000
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
We
Token We
Feature activation+0.000
have
Token have
Feature activation+0.000
nothing
Token nothing
Feature activation+0.000
to
Token to
Feature activation+0.000
lose
Token lose
Feature activation+1.280
and
Token and
Feature activation+0.000
everything
Token everything
Feature activation+0.000
to
Token to
Feature activation+0.000
gain
Token gain
Feature activation+0.000
."
Token."
Feature activation+0.000

INTERVAL 0.845 - 1.127
CONTAINS 0.007%

launch
Token launch
Feature activation+0.000
!
Token!
Feature activation+0.000
TOR
TokenTOR
Feature activation+0.018
BA
TokenBA
Feature activation+0.129
(
Token (
Feature activation+0.000
Toronto
TokenToronto
Feature activation+0.849
Off
Token Off
Feature activation+0.000
road
Tokenroad
Feature activation+0.000
B
Token B
Feature activation+0.000
icy
Tokenicy
Feature activation+0.000
cling
Tokencling
Feature activation+0.551
W
Token W
Feature activation+0.000
hel
Tokenhel
Feature activation+0.000
en
Tokenen
Feature activation+0.000
Southern
Token Southern
Feature activation+0.000
Modified
Token Modified
Feature activation+0.000
Tour
Token Tour
Feature activation+0.876
.
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
,
Token,
Feature activation+0.000
Mill
Token Mill
Feature activation+0.047
wall
Tokenwall
Feature activation+0.000
,
Token,
Feature activation+0.000
North
Token North
Feature activation+0.000
ampton
Tokenampton
Feature activation+0.951
Town
Token Town
Feature activation+0.000
and
Token and
Feature activation+0.000
G
Token G
Feature activation+0.218
illing
Tokenilling
Feature activation+0.000
ham
Tokenham
Feature activation+0.000
et
Tokenet
Feature activation+0.000
ok
Tokenok
Feature activation+0.000
ou
Tokenou
Feature activation+0.000
mp
Tokenmp
Feature activation+0.000
o
Tokeno
Feature activation+0.000
committed
Token committed
Feature activation+0.930
a
Token a
Feature activation+0.000
turnover
Token turnover
Feature activation+0.249
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
final
Token final
Feature activation+0.000
,
Token,
Feature activation+0.000
Rod
Token Rod
Feature activation+0.618
Stewart
Token Stewart
Feature activation+0.121
.)
Token.)
Feature activation+0.000
But
Token But
Feature activation+0.000
art
Token art
Feature activation+0.875
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
an
Token an
Feature activation+0.000
interesting
Token interesting
Feature activation+0.000
mix
Token mix
Feature activation+0.000

INTERVAL 0.563 - 0.845
CONTAINS 0.020%

wake
Token wake
Feature activation+0.000
of
Token of
Feature activation+0.000
US
Token US
Feature activation+0.314
President
Token President
Feature activation+0.257
Donald
Token Donald
Feature activation+0.000
Trump
Token Trump
Feature activation+0.615
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
tweets
Token tweets
Feature activation+0.000
on
Token on
Feature activation+0.000
series
Token series
Feature activation+0.672
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
everybody
Token everybody
Feature activation+0.000
's
Token's
Feature activation+0.000
ultimate
Token ultimate
Feature activation+0.566
goal
Token goal
Feature activation+0.159
has
Token has
Feature activation+0.000
always
Token always
Feature activation+0.000
been
Token been
Feature activation+0.000
F
Token F
Feature activation+0.514
The
TokenThe
Feature activation+0.000
rookie
Token rookie
Feature activation+0.300
then
Token then
Feature activation+0.033
decided
Token decided
Feature activation+0.041
to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.840
a
Token a
Feature activation+0.000
pit
Token pit
Feature activation+0.000
stop
Token stop
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
the
Token the
Feature activation+0.000
same
Token same
Feature activation+0.000
."
Token."
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.816
The
TokenThe
Feature activation+0.000
clean
Token clean
Feature activation+0.000
energy
Token energy
Feature activation+0.000
transition
Token transition
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
did
Token did
Feature activation+0.054
five
Token five
Feature activation+0.070
races
Token races
Feature activation+0.679
in
Token in
Feature activation+0.116
2015
Token 2015
Feature activation+0.000
.
Token.
Feature activation+0.645
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.117
Ver
TokenVer
Feature activation+0.141
h
Tokenh
Feature activation+0.000
agen
Tokenagen
Feature activation+0.000

INTERVAL 0.282 - 0.563
CONTAINS 0.076%

for
Token for
Feature activation+0.000
Le
Token Le
Feature activation+0.000
yton
Tokenyton
Feature activation+0.904
Orient
Token Orient
Feature activation+0.682
,
Token,
Feature activation+0.000
Rush
Token Rush
Feature activation+0.401
den
Tokenden
Feature activation+0.000
&
Token &
Feature activation+0.000
Diamond
Token Diamond
Feature activation+0.211
s
Tokens
Feature activation+0.000
,
Token,
Feature activation+0.000
not
Token not
Feature activation+0.000
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
2
Token 2
Feature activation+0.000
+
Token+
Feature activation+0.000
win
Token win
Feature activation+0.410
jump
Token jump
Feature activation+0.000
based
Token based
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
timing
Token timing
Feature activation+0.000
shows
Token shows
Feature activation+0.000
Paradise
Token Paradise
Feature activation+0.000
,
Token,
Feature activation+0.000
roughly
Token roughly
Feature activation+0.000
where
Token where
Feature activation+0.000
Japan
Token Japan
Feature activation+0.307
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
of
Token of
Feature activation+0.000
2015
Token 2015
Feature activation+0.003
Ice
Token Ice
Feature activation+0.000
Hockey
Token Hockey
Feature activation+0.000
World
Token World
Feature activation+0.000
Championship
Token Championship
Feature activation+0.418
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
held
Token held
Feature activation+0.000
on
Token on
Feature activation+0.000
Sunday
Token Sunday
Feature activation+0.000
former
Token former
Feature activation+0.000
star
Token star
Feature activation+0.314
loan
Token loan
Feature activation+0.000
ee
Tokenee
Feature activation+0.000
Tom
Token Tom
Feature activation+0.000
Lawrence
Token Lawrence
Feature activation+0.518
and
Token and
Feature activation+0.000
nem
Token nem
Feature activation+0.000
esis
Tokenesis
Feature activation+0.000
striker
Token striker
Feature activation+0.000
David
Token David
Feature activation+0.038

INTERVAL 0.000 - 0.282
CONTAINS 99.888%

was
Token was
Feature activation+0.000
divine
Token divine
Feature activation+0.000
ly
Tokenly
Feature activation+0.000
inspired
Token inspired
Feature activation+0.000
.
Token.
Feature activation+0.000
Jesus
Token Jesus
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
was
Tokenwas
Feature activation+0.000
very
Token very
Feature activation+0.000
nice
Token nice
Feature activation+0.000
Sub
Token Sub
Feature activation+0.000
sc
Tokensc
Feature activation+0.000
riber
Tokenriber
Feature activation+0.000
Reward
Token Reward
Feature activation+0.000
2
Token 2
Feature activation+0.000
V
TokenV
Feature activation+0.000
-
Token-
Feature activation+0.000
R
TokenR
Feature activation+0.000
8
Token8
Feature activation+0.000
Custom
Token Custom
Feature activation+0.000
ization
Tokenization
Feature activation+0.000
!
Token!
Feature activation+0.000
Man
Token Man
Feature activation+0.000
!
Token!
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
been
Token been
Feature activation+0.000
a
Token a
Feature activation+0.000
while
Token while
Feature activation+0.000
since
Token since
Feature activation+0.000
a
Token a
Feature activation+0.000
place
Token place
Feature activation+0.000
they
Token they
Feature activation+0.000
imagined
Token imagined
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
mountains
Token mountains
Feature activation+0.000
where
Token where
Feature activation+0.000
corn
Token corn
Feature activation+0.000
would
Token would
Feature activation+0.000
not
Token not
Feature activation+0.000
"
Token"
Feature activation+0.000
in
Token in
Feature activation+0.000
Counter
Token Counter
Feature activation+0.000
-
Token-
Feature activation+0.000
Strike
TokenStrike
Feature activation+0.000
is
Token is
Feature activation+0.000
when
Token when
Feature activation+0.000
players
Token players
Feature activation+0.000
use
Token use
Feature activation+0.000
each
Token each
Feature activation+0.000
other
Token other
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 H3.4: (feature 22438)

TOP ACTIVATIONS
MAX = 2.699

equivalent
Token equivalent
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
population
Token population
Feature activation+0.000
of
Token of
Feature activation+0.000
Pasadena
Token Pasadena
Feature activation+2.699
,
Token,
Feature activation+0.000
California
Token California
Feature activation+2.330
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
each
Token each
Feature activation+0.000
the
Token the
Feature activation+0.000
population
Token population
Feature activation+0.000
of
Token of
Feature activation+0.000
Pasadena
Token Pasadena
Feature activation+2.699
,
Token,
Feature activation+0.000
California
Token California
Feature activation+2.330
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
each
Token each
Feature activation+0.000
camp
Token camp
Feature activation+0.000
on
Token on
Feature activation+0.000
its
Token its
Feature activation+0.000
members
Token members
Feature activation+0.000
,
Token,
Feature activation+0.000
admits
Token admits
Feature activation+0.000
that
Token that
Feature activation+0.000
Israeli
Token Israeli
Feature activation+2.233
soldiers
Token soldiers
Feature activation+0.000
used
Token used
Feature activation+0.000
excessive
Token excessive
Feature activation+0.000
and
Token and
Feature activation+0.000
disproportionate
Token disproportionate
Feature activation+0.000
his
Token his
Feature activation+0.000
installation
Token installation
Feature activation+0.000
as
Token as
Feature activation+0.000
leader
Token leader
Feature activation+0.000
of
Token of
Feature activation+0.000
Chicago
Token Chicago
Feature activation+2.152
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
2
Token 2
Feature activation+0.000
million
Token million
Feature activation+0.000
to
Token to
Feature activation+0.000
carry
Token carry
Feature activation+0.000
out
Token out
Feature activation+0.000
attacks
Token attacks
Feature activation+0.000
on
Token on
Feature activation+0.000
Indian
Token Indian
Feature activation+2.125
festivals
Token festivals
Feature activation+0.000
like
Token like
Feature activation+0.000
K
Token K
Feature activation+0.230
umb
Tokenumb
Feature activation+0.259
h
Tokenh
Feature activation+0.000
information
Token information
Feature activation+0.000
of
Token of
Feature activation+0.000
some
Token some
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Israeli
Token Israeli
Feature activation+2.106
soldiers
Token soldiers
Feature activation+0.000
who
Token who
Feature activation+0.000
carried
Token carried
Feature activation+0.000
out
Token out
Feature activation+0.000
the
Token the
Feature activation+0.000
would
Token would
Feature activation+0.000
also
Token also
Feature activation+0.000
exclude
Token exclude
Feature activation+0.000
references
Token references
Feature activation+0.000
to
Token to
Feature activation+0.000
Charles
Token Charles
Feature activation+2.076
Darwin
Token Darwin
Feature activation+0.635
and
Token and
Feature activation+0.000
his
Token his
Feature activation+0.000
theory
Token theory
Feature activation+0.000
of
Token of
Feature activation+0.000
tried
Token tried
Feature activation+0.000
to
Token to
Feature activation+0.000
discuss
Token discuss
Feature activation+0.000
with
Token with
Feature activation+0.000
their
Token their
Feature activation+0.000
Chinese
Token Chinese
Feature activation+2.052
counterparts
Token counterparts
Feature activation+0.000
what
Token what
Feature activation+0.000
happens
Token happens
Feature activation+0.000
when
Token when
Feature activation+0.000
,
Token,
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
Presidency
Token Presidency
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Chinese
Token Chinese
Feature activation+2.039
Academy
Token Academy
Feature activation+0.000
of
Token of
Feature activation+0.000
Sciences
Token Sciences
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
-
Token-
Feature activation+0.000
2
Token2
Feature activation+0.000
draw
Token draw
Feature activation+0.000
--
Token --
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+2.015
have
Token have
Feature activation+0.000
scored
Token scored
Feature activation+0.000
one
Token one
Feature activation+0.000
goal
Token goal
Feature activation+0.000
(
Token (
Feature activation+0.000
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
was
Token was
Feature activation+0.000
after
Token after
Feature activation+0.000
a
Token a
Feature activation+0.000
Chinese
Token Chinese
Feature activation+2.014
coast
Token coast
Feature activation+0.000
guard
Tokenguard
Feature activation+0.000
vessel
Token vessel
Feature activation+0.000
prevented
Token prevented
Feature activation+0.000
Jakarta
Token Jakarta
Feature activation+0.381
understanding
Token understanding
Feature activation+0.000
with
Token with
Feature activation+0.000
a
Token a
Feature activation+0.000
group
Token group
Feature activation+0.000
of
Token of
Feature activation+0.000
Indian
Token Indian
Feature activation+1.985
businessmen
Token businessmen
Feature activation+0.000
who
Token who
Feature activation+0.000
call
Token call
Feature activation+0.000
themselves
Token themselves
Feature activation+0.000
the
Token the
Feature activation+0.000
concluded
Token concluded
Feature activation+0.000
the
Token the
Feature activation+0.000
war
Token war
Feature activation+0.000
with
Token with
Feature activation+0.000
Imperial
Token Imperial
Feature activation+0.650
Japan
Token Japan
Feature activation+1.985
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
end
Token end
Feature activation+0.000
of
Token of
Feature activation+0.000
World
Token World
Feature activation+0.000
San
Token San
Feature activation+1.208
Ro
Token Ro
Feature activation+0.573
que
Tokenque
Feature activation+0.000
where
Token where
Feature activation+0.000
a
Token a
Feature activation+0.000
Turkish
Token Turkish
Feature activation+1.972
man
Token man
Feature activation+0.000
was
Token was
Feature activation+0.000
arrested
Token arrested
Feature activation+0.000
.
Token.
Feature activation+0.000
Two
Token Two
Feature activation+0.000
flying
Token flying
Feature activation+0.000
from
Token from
Feature activation+0.000
Cairo
Token Cairo
Feature activation+0.000
to
Token to
Feature activation+0.000
New
Token New
Feature activation+0.609
York
Token York
Feature activation+1.925
was
Token was
Feature activation+0.000
diverted
Token diverted
Feature activation+0.000
to
Token to
Feature activation+0.000
Prest
Token Prest
Feature activation+0.000
wick
Tokenwick
Feature activation+0.384
police
Token police
Feature activation+0.000
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
Nep
Token Nep
Feature activation+1.895
ali
Tokenali
Feature activation+0.779
citizen
Token citizen
Feature activation+0.000
who
Token who
Feature activation+0.000
managed
Token managed
Feature activation+0.000
an
Token an
Feature activation+0.000
$
Token $
Feature activation+0.000
1
Token1
Feature activation+0.000
billion
Token billion
Feature activation+0.000
agreement
Token agreement
Feature activation+0.000
with
Token with
Feature activation+0.000
South
Token South
Feature activation+1.873
Korea
Token Korea
Feature activation+1.203
to
Token to
Feature activation+0.000
build
Token build
Feature activation+0.000
a
Token a
Feature activation+0.000
300
Token 300
Feature activation+0.000
with
Token with
Feature activation+0.000
Edward
Token Edward
Feature activation+0.000
Jones
Token Jones
Feature activation+0.000
in
Token in
Feature activation+0.000
Saint
Token Saint
Feature activation+0.558
Louis
Token Louis
Feature activation+1.867
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ex
TokenEx
Feature activation+0.000
xon
Tokenxon
Feature activation+0.000
symbolic
Token symbolic
Feature activation+0.000
show
Token show
Feature activation+0.000
of
Token of
Feature activation+0.000
support
Token support
Feature activation+0.000
put
Token put
Feature activation+0.000
India
Token India
Feature activation+1.857
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
middle
Token middle
Feature activation+0.098
of
Token of
Feature activation+0.000
one
Token one
Feature activation+0.000
Kir
Token Kir
Feature activation+0.037
ch
Tokench
Feature activation+0.000
ner
Tokenner
Feature activation+0.000
has
Token has
Feature activation+0.000
attacked
Token attacked
Feature activation+0.000
Britain
Token Britain
Feature activation+1.850
repeatedly
Token repeatedly
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
Falk
Token Falk
Feature activation+0.035
lands
Tokenlands
Feature activation+0.000

Top DFA by src position
MAX = 2.849

D
Token D
Feature activation+0.002
Top resid features:
ada
Tokenada
Feature activation+0.002
Top resid features:
ab
Tokenab
Feature activation-0.007
Top resid features:
complex
Token complex
Feature activation+0.001
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
Kenya
Token Kenya
Feature activation+2.849
Top resid features:
,
Token,
Feature activation-0.003
Top resid features:
which
Token which
Feature activation+0.002
Top resid features:
includes
Token includes
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
three
Token three
Feature activation-0.003
Top resid features:
D
Token D
Feature activation+0.001
Top resid features:
ada
Tokenada
Feature activation+0.000
Top resid features:
ab
Tokenab
Feature activation-0.001
Top resid features:
complex
Token complex
Feature activation+0.001
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
Kenya
Token Kenya
Feature activation+2.562
Top resid features:
,
Token,
Feature activation-0.005
Top resid features:
which
Token which
Feature activation+0.003
Top resid features:
includes
Token includes
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
three
Token three
Feature activation-0.001
Top resid features:
committed
Token committed
Feature activation-0.001
Top resid features:
while
Token while
Feature activation+0.002
Top resid features:
serving
Token serving
Feature activation-0.001
Top resid features:
as
Token as
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.003
Top resid features:
Colombian
Token Colombian
Feature activation+2.455
Top resid features:
president
Token president
Feature activation-0.001
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
investigation
Token investigation
Feature activation-0.001
Top resid features:
is
Token is
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.064
Top resid features:
a
Token a
Feature activation-0.004
Top resid features:
Croatian
Token Croatian
Feature activation+2.362
Top resid features:
parish
Token parish
Feature activation+0.033
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.002
Top resid features:
and
Token and
Feature activation+0.004
Top resid features:
learn
Token learn
Feature activation-0.000
Top resid features:
folk
Token folk
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.040
Top resid features:
Syria
Token Syria
Feature activation+2.380
Top resid features:
has
Token has
Feature activation+0.003
Top resid features:
reportedly
Token reportedly
Feature activation-0.002
Top resid features:
exh
Token exh
Feature activation+0.001
Top resid features:
orted
Tokenorted
Feature activation-0.002
Top resid features:
its
Token its
Feature activation-0.003
Top resid features:
committed
Token committed
Feature activation-0.000
Top resid features:
while
Token while
Feature activation+0.003
Top resid features:
serving
Token serving
Feature activation-0.001
Top resid features:
as
Token as
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
Colombian
Token Colombian
Feature activation+2.571
Top resid features:
president
Token president
Feature activation-0.001
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
investigation
Token investigation
Feature activation-0.001
Top resid features:
is
Token is
Feature activation-0.001
Top resid features:
after
Token after
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
head
Token head
Feature activation-0.003
Top resid features:
of
Token of
Feature activation-0.003
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
Turkish
Token Turkish
Feature activation+2.251
Top resid features:
Education
Token Education
Feature activation+0.007
Top resid features:
Ministry
Token Ministry
Feature activation+0.005
Top resid features:
's
Token's
Feature activation+0.004
Top resid features:
education
Token education
Feature activation+0.002
Top resid features:
board
Token board
Feature activation-0.004
Top resid features:
so
Token so
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.010
Top resid features:
Ċ
TokenĊ
Feature activation-0.006
Top resid features:
South
TokenSouth
Feature activation+0.026
Top resid features:
Korean
Token Korean
Feature activation+2.346
Top resid features:
and
Token and
Feature activation+0.008
Top resid features:
American
Token American
Feature activation-0.014
Top resid features:
military
Token military
Feature activation+0.000
Top resid features:
officers
Token officers
Feature activation+0.001
Top resid features:
and
Token and
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.033
Top resid features:
Iranian
Token Iranian
Feature activation+1.022
Top resid features:
delegation
Token delegation
Feature activation+0.008
Top resid features:
headed
Token headed
Feature activation-0.000
Top resid features:
by
Token by
Feature activation-0.003
Top resid features:
Sa
Token Sa
Feature activation+0.017
Top resid features:
eed
Tokeneed
Feature activation+0.013
Top resid features:
World
Token World
Feature activation+0.017
Top resid features:
Cup
Token Cup
Feature activation-0.015
Top resid features:
group
Token group
Feature activation-0.003
Top resid features:
stage
Token stage
Feature activation-0.003
Top resid features:
against
Token against
Feature activation-0.003
Top resid features:
Portugal
Token Portugal
Feature activation+2.201
Top resid features:
.
Token.
Feature activation-0.006
Top resid features:
Since
Token Since
Feature activation+0.000
Top resid features:
that
Token that
Feature activation-0.009
Top resid features:
game
Token game
Feature activation+0.002
Top resid features:
--
Token --
Feature activation-0.016
Top resid features:
9
Token 9
Feature activation-0.001
Top resid features:
10
Token 10
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.031
Top resid features:
Ind
TokenInd
Feature activation+0.018
Top resid features:
ones
Tokenones
Feature activation+0.017
Top resid features:
ia
Tokenia
Feature activation+2.254
Top resid features:
is
Token is
Feature activation-0.012
Top resid features:
reportedly
Token reportedly
Feature activation-0.003
Top resid features:
studying
Token studying
Feature activation+0.008
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
possibility
Token possibility
Feature activation-0.002
Top resid features:
che
Tokenche
Feature activation-0.001
Top resid features:
sted
Tokensted
Feature activation-0.001
Top resid features:
prime
Token prime
Feature activation-0.001
Top resid features:
minister
Token minister
Feature activation-0.000
Top resid features:
of
Token of
Feature activation-0.000
Top resid features:
Crimea
Token Crimea
Feature activation+1.153
Top resid features:
,
Token,
Feature activation-0.004
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
Ukrainian
Token Ukrainian
Feature activation+0.850
Top resid features:
territory
Token territory
Feature activation+0.010
Top resid features:
annexed
Token annexed
Feature activation+0.003
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.039
Top resid features:
Afghan
Token Afghan
Feature activation+2.183
Top resid features:
war
Token war
Feature activation+0.001
Top resid features:
,
Token,
Feature activation-0.009
Top resid features:
he
Token he
Feature activation+0.004
Top resid features:
concluded
Token concluded
Feature activation-0.002
Top resid features:
that
Token that
Feature activation+0.002
Top resid features:
-
Token-
Feature activation-0.001
Top resid features:
Qaeda
TokenQaeda
Feature activation-0.001
Top resid features:
members
Token members
Feature activation+0.000
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
southern
Token southern
Feature activation+0.011
Top resid features:
Spain
Token Spain
Feature activation+2.157
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:
Expl
TokenExpl
Feature activation+0.003
Top resid features:
os
Tokenos
Feature activation+0.014
Top resid features:
14
Token 14
Feature activation-0.000
Top resid features:
20
Token20
Feature activation+0.002
Top resid features:
hours
Token hours
Feature activation-0.003
Top resid features:
today
Token today
Feature activation+0.001
Top resid features:
an
Token an
Feature activation+0.011
Top resid features:
Egypt
Token Egypt
Feature activation+1.501
Top resid features:
air
Tokenair
Feature activation-0.005
Top resid features:
aircraft
Token aircraft
Feature activation+0.018
Top resid features:
flying
Token flying
Feature activation+0.014
Top resid features:
from
Token from
Feature activation+0.004
Top resid features:
Cairo
Token Cairo
Feature activation+0.623
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.034
Top resid features:
UMP
TokenUMP
Feature activation+0.012
Top resid features:
UR
TokenUR
Feature activation-0.002
Top resid features:
:
Token:
Feature activation-0.002
Top resid features:
Malaysian
Token Malaysian
Feature activation+1.577
Top resid features:
police
Token police
Feature activation-0.003
Top resid features:
have
Token have
Feature activation-0.011
Top resid features:
arrested
Token arrested
Feature activation+0.003
Top resid features:
four
Token four
Feature activation-0.003
Top resid features:
people
Token people
Feature activation-0.002
Top resid features:
popular
Token popular
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.025
Top resid features:
The
TokenThe
Feature activation-0.003
Top resid features:
Kingdom
Token Kingdom
Feature activation+0.086
Top resid features:
of
Token of
Feature activation-0.003
Top resid features:
Saudi
Token Saudi
Feature activation+1.615
Top resid features:
Arabia
Token Arabia
Feature activation+0.488
Top resid features:
(
Token (
Feature activation-0.001
Top resid features:
K
TokenK
Feature activation-0.002
Top resid features:
SA
TokenSA
Feature activation+0.001
Top resid features:
)
Token)
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Oil
TokenOil
Feature activation+0.002
Top resid features:
majors
Token majors
Feature activation+0.001
Top resid features:
are
Token are
Feature activation-0.001
Top resid features:
targeting
Token targeting
Feature activation-0.001
Top resid features:
Papua
Token Papua
Feature activation+1.136
Top resid features:
New
Token New
Feature activation+0.030
Top resid features:
Guinea
Token Guinea
Feature activation+0.802
Top resid features:
for
Token for
Feature activation-0.002
Top resid features:
growth
Token growth
Feature activation-0.001
Top resid features:
as
Token as
Feature activation+0.000
Top resid features:
che
Tokenche
Feature activation-0.001
Top resid features:
sted
Tokensted
Feature activation-0.000
Top resid features:
prime
Token prime
Feature activation-0.000
Top resid features:
minister
Token minister
Feature activation-0.000
Top resid features:
of
Token of
Feature activation-0.000
Top resid features:
Crimea
Token Crimea
Feature activation+1.053
Top resid features:
,
Token,
Feature activation-0.004
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
Ukrainian
Token Ukrainian
Feature activation+1.012
Top resid features:
territory
Token territory
Feature activation+0.022
Top resid features:
annexed
Token annexed
Feature activation+0.010
Top resid features:
Islands
Token Islands
Feature activation+0.043
Top resid features:
were
Token were
Feature activation-0.001
Top resid features:
taken
Token taken
Feature activation-0.000
Top resid features:
briefly
Token briefly
Feature activation-0.000
Top resid features:
by
Token by
Feature activation-0.001
Top resid features:
Argentina
Token Argentina
Feature activation+0.472
Top resid features:
during
Token during
Feature activation+0.000
Top resid features:
its
Token its
Feature activation+0.001
Top resid features:
1982
Token 1982
Feature activation+0.000
Top resid features:
invasion
Token invasion
Feature activation-0.000
Top resid features:
which
Token which
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.10

Head 1: 0.04

Head 2: 0.03

Head 3: 0.02

Head 4: 0.49

Head 5: 0.04

Head 6: 0.08

Head 7: 0.04

Head 8: 0.03

Head 9: 0.03

Head 10: 0.07

Head 11: 0.03

Positive logits

204392.39

ESPN2.21

SQL2.14

eval2.14

POS2.12

oen2.07

IGN1.96

eson1.94

itatively1.94

acter1.90

ARR1.88

ONS1.87

lations1.80

fig1.78

bol1.75

appings1.74

ritional1.74

SSL1.73

ufact1.72

table1.71

Negative logits

countryside-2.00

Ruin-1.86

venge-1.82

pite-1.79

municipality-1.78

outlawed-1.77

odan-1.76

warr-1.75

Seym-1.73

Baal-1.73

Franch-1.71

lord-1.71

Downloadha-1.69

emale-1.69

Lange-1.67

Chau-1.66

masc-1.65

Mile-1.65

guiActiveUn-1.65

government-1.63

INTERVAL 2.429 - 2.699
CONTAINS 0.000%

equivalent
Token equivalent
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
population
Token population
Feature activation+0.000
of
Token of
Feature activation+0.000
Pasadena
Token Pasadena
Feature activation+2.699
,
Token,
Feature activation+0.000
California
Token California
Feature activation+2.330
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
each
Token each
Feature activation+0.000

INTERVAL 2.159 - 2.429
CONTAINS 0.000%

the
Token the
Feature activation+0.000
population
Token population
Feature activation+0.000
of
Token of
Feature activation+0.000
Pasadena
Token Pasadena
Feature activation+2.699
,
Token,
Feature activation+0.000
California
Token California
Feature activation+2.330
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
each
Token each
Feature activation+0.000
camp
Token camp
Feature activation+0.000
on
Token on
Feature activation+0.000
its
Token its
Feature activation+0.000
members
Token members
Feature activation+0.000
,
Token,
Feature activation+0.000
admits
Token admits
Feature activation+0.000
that
Token that
Feature activation+0.000
Israeli
Token Israeli
Feature activation+2.233
soldiers
Token soldiers
Feature activation+0.000
used
Token used
Feature activation+0.000
excessive
Token excessive
Feature activation+0.000
and
Token and
Feature activation+0.000
disproportionate
Token disproportionate
Feature activation+0.000

INTERVAL 1.889 - 2.159
CONTAINS 0.001%

to
Token to
Feature activation+0.000
carry
Token carry
Feature activation+0.000
out
Token out
Feature activation+0.000
attacks
Token attacks
Feature activation+0.000
on
Token on
Feature activation+0.000
Indian
Token Indian
Feature activation+2.125
festivals
Token festivals
Feature activation+0.000
like
Token like
Feature activation+0.000
K
Token K
Feature activation+0.230
umb
Tokenumb
Feature activation+0.259
h
Tokenh
Feature activation+0.000
police
Token police
Feature activation+0.000
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
Nep
Token Nep
Feature activation+1.895
ali
Tokenali
Feature activation+0.779
citizen
Token citizen
Feature activation+0.000
who
Token who
Feature activation+0.000
managed
Token managed
Feature activation+0.000
an
Token an
Feature activation+0.000
would
Token would
Feature activation+0.000
also
Token also
Feature activation+0.000
exclude
Token exclude
Feature activation+0.000
references
Token references
Feature activation+0.000
to
Token to
Feature activation+0.000
Charles
Token Charles
Feature activation+2.076
Darwin
Token Darwin
Feature activation+0.635
and
Token and
Feature activation+0.000
his
Token his
Feature activation+0.000
theory
Token theory
Feature activation+0.000
of
Token of
Feature activation+0.000
information
Token information
Feature activation+0.000
of
Token of
Feature activation+0.000
some
Token some
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Israeli
Token Israeli
Feature activation+2.106
soldiers
Token soldiers
Feature activation+0.000
who
Token who
Feature activation+0.000
carried
Token carried
Feature activation+0.000
out
Token out
Feature activation+0.000
the
Token the
Feature activation+0.000
flying
Token flying
Feature activation+0.000
from
Token from
Feature activation+0.000
Cairo
Token Cairo
Feature activation+0.000
to
Token to
Feature activation+0.000
New
Token New
Feature activation+0.609
York
Token York
Feature activation+1.925
was
Token was
Feature activation+0.000
diverted
Token diverted
Feature activation+0.000
to
Token to
Feature activation+0.000
Prest
Token Prest
Feature activation+0.000
wick
Tokenwick
Feature activation+0.384

INTERVAL 1.619 - 1.889
CONTAINS 0.003%

Ukraine
Token Ukraine
Feature activation+0.000
on
Token on
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
night
Token night
Feature activation+0.000
were
Token were
Feature activation+0.000
Italy
Token Italy
Feature activation+1.777
's
Token's
Feature activation+0.000
Al
Token Al
Feature activation+0.944
ital
Tokenital
Feature activation+0.133
ia
Tokenia
Feature activation+0.111
,
Token,
Feature activation+0.000
one
Token one
Feature activation+0.000
which
Token which
Feature activation+0.000
appeared
Token appeared
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
British
Token British
Feature activation+1.706
Independent
Token Independent
Feature activation+0.043
(
Token (
Feature activation+0.000
July
TokenJuly
Feature activation+0.000
1
Token 1
Feature activation+0.000
,
Token,
Feature activation+0.000
activating
Token activating
Feature activation+0.000
soft
Token soft
Feature activation+0.000
policy
Token policy
Feature activation+0.000
through
Token through
Feature activation+0.000
the
Token the
Feature activation+0.000
Egyptian
Token Egyptian
Feature activation+1.755
church
Token church
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
activity
Token activity
Feature activation+0.000
-
Token-
Feature activation+0.000
old
Tokenold
Feature activation+0.000
passport
Token passport
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
Canadian
Token Canadian
Feature activation+1.721
High
Token High
Feature activation+0.000
Commission
Token Commission
Feature activation+0.315
agreed
Token agreed
Feature activation+0.000
and
Token and
Feature activation+0.000
pulled
Token pulled
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
T
TokenT
Feature activation+0.000
aste
Tokenaste
Feature activation+0.000
of
Token of
Feature activation+0.000
Ireland
Token Ireland
Feature activation+1.841
âĢ
TokenâĢ
Feature activation+0.053
Ŀ
TokenĿ
Feature activation+0.000
(
Token (
Feature activation+0.000
available
Tokenavailable
Feature activation+0.000
for
Token for
Feature activation+0.000

INTERVAL 1.349 - 1.619
CONTAINS 0.006%

74
Token 74
Feature activation+0.000
-
Token-
Feature activation+0.000
day
Tokenday
Feature activation+0.000
war
Token war
Feature activation+0.159
with
Token with
Feature activation+0.000
Britain
Token Britain
Feature activation+1.374
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
war
Token war
Feature activation+0.237
has
Token has
Feature activation+0.000
a
Token a
Feature activation+0.000
contract
Token contract
Feature activation+0.000
with
Token with
Feature activation+0.000
Force
Token Force
Feature activation+0.309
India
Token India
Feature activation+1.394
for
Token for
Feature activation+0.000
next
Token next
Feature activation+0.000
season
Token season
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
right
Token right
Feature activation+0.000
-
Token-
Feature activation+0.000
wing
Tokenwing
Feature activation+0.000
rhetoric
Token rhetoric
Feature activation+0.000
in
Token in
Feature activation+0.000
Japan
Token Japan
Feature activation+1.512
,
Token,
Feature activation+0.000
where
Token where
Feature activation+0.000
ethnic
Token ethnic
Feature activation+0.000
hom
Token hom
Feature activation+0.000
ogeneity
Tokenogeneity
Feature activation+0.000
the
Token the
Feature activation+0.000
passport
Token passport
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Canadian
TokenCanadian
Feature activation+1.481
officials
Token officials
Feature activation+0.000
initially
Token initially
Feature activation+0.000
called
Token called
Feature activation+0.000
Moh
Token Moh
Feature activation+0.310
am
Tokenam
Feature activation+0.000
of
Token of
Feature activation+0.000
German
Token German
Feature activation+0.000
,
Token,
Feature activation+0.000
U
Token U
Feature activation+0.344
.
Token.
Feature activation+0.869
S
TokenS
Feature activation+1.445
.,
Token.,
Feature activation+0.000
and
Token and
Feature activation+0.000
Mexican
Token Mexican
Feature activation+0.297
researchers
Token researchers
Feature activation+0.000
discovered
Token discovered
Feature activation+0.000

INTERVAL 1.079 - 1.349
CONTAINS 0.011%

out
Token out
Feature activation+0.000
lone
Token lone
Feature activation+0.000
wolf
Token wolf
Feature activation+0.000
attacks
Token attacks
Feature activation+0.000
in
Token in
Feature activation+0.000
India
Token India
Feature activation+1.119
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
World
TokenWorld
Feature activation+0.451
âĢ
TokenâĢ
Feature activation+0.000
over
Token over
Feature activation+0.000
some
Token some
Feature activation+0.000
articles
Token articles
Feature activation+0.000
by
Token by
Feature activation+0.000
Mr
Token Mr
Feature activation+0.294
.
Token.
Feature activation+1.220
Nord
Token Nord
Feature activation+0.729
land
Tokenland
Feature activation+0.024
from
Token from
Feature activation+0.000
November
Token November
Feature activation+0.000
and
Token and
Feature activation+0.000
Day
Token Day
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Thousands
TokenThousands
Feature activation+0.000
of
Token of
Feature activation+0.000
Catal
Token Catal
Feature activation+1.258
ans
Tokenans
Feature activation+0.000
have
Token have
Feature activation+0.000
rallied
Token rallied
Feature activation+0.000
in
Token in
Feature activation+0.000
Barcelona
Token Barcelona
Feature activation+0.201
a
Token a
Feature activation+0.000
fight
Token fight
Feature activation+0.000
over
Token over
Feature activation+0.000
whether
Token whether
Feature activation+0.000
the
Token the
Feature activation+0.000
Puerto
Token Puerto
Feature activation+1.142
Rican
Token Rican
Feature activation+0.762
common
Token common
Feature activation+0.045
wealth
Tokenwealth
Feature activation+0.000
should
Token should
Feature activation+0.000
be
Token be
Feature activation+0.000
R
TokenR
Feature activation+0.070
)
Token)
Feature activation+0.000
shakes
Token shakes
Feature activation+0.000
hand
Token hand
Feature activation+0.000
with
Token with
Feature activation+0.000
British
Token British
Feature activation+1.256
Prime
Token Prime
Feature activation+0.214
Minister
Token Minister
Feature activation+0.000
Theresa
Token Theresa
Feature activation+1.211
May
Token May
Feature activation+0.000
before
Token before
Feature activation+0.000

INTERVAL 0.810 - 1.079
CONTAINS 0.025%

,
Token,
Feature activation+0.000
the
Token the
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
French
Token French
Feature activation+1.014
wh
Token wh
Feature activation+0.145
aling
Tokenaling
Feature activation+0.059
ship
Token ship
Feature activation+0.000
that
Token that
Feature activation+0.000
came
Token came
Feature activation+0.000
swimming
Token swimming
Feature activation+0.000
to
Token to
Feature activation+0.000
Hungary
Token Hungary
Feature activation+0.224
via
Token via
Feature activation+0.000
the
Token the
Feature activation+0.000
T
Token T
Feature activation+0.842
is
Tokenis
Feature activation+0.000
za
Tokenza
Feature activation+0.000
river
Token river
Feature activation+0.000
in
Token in
Feature activation+0.000
June
Token June
Feature activation+0.000
routes
Token routes
Feature activation+0.000
from
Token from
Feature activation+0.000
Heath
Token Heath
Feature activation+0.603
row
Tokenrow
Feature activation+0.000
to
Token to
Feature activation+0.000
Dubai
Token Dubai
Feature activation+0.816
and
Token and
Feature activation+0.000
one
Token one
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
Mumbai
Token Mumbai
Feature activation+1.129
ı
Tokenı
Feature activation+0.000
ur
Tokenur
Feature activation+0.011
fa
Tokenfa
Feature activation+0.091
on
Token on
Feature activation+0.000
Turkish
Token Turkish
Feature activation+0.000
Syrian
Token Syrian
Feature activation+1.024
border
Token border
Feature activation+0.000
.
Token.
Feature activation+0.000
He
Token He
Feature activation+0.000
was
Token was
Feature activation+0.000
involved
Token involved
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
U
TokenU
Feature activation+0.428
.
Token.
Feature activation+0.553
S
TokenS
Feature activation+0.855
.
Token.
Feature activation+0.585
Secretary
Token Secretary
Feature activation+0.000
of
Token of
Feature activation+0.000
State
Token State
Feature activation+0.000
John
Token John
Feature activation+0.281

INTERVAL 0.540 - 0.810
CONTAINS 0.053%

was
Token was
Feature activation+0.000
alarmed
Token alarmed
Feature activation+0.000
by
Token by
Feature activation+0.000
reports
Token reports
Feature activation+0.000
that
Token that
Feature activation+0.000
separatists
Token separatists
Feature activation+0.585
,
Token,
Feature activation+0.000
mostly
Token mostly
Feature activation+0.000
from
Token from
Feature activation+0.000
Pakistan
Token Pakistan
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
,
Token,
Feature activation+0.000
backing
Token backing
Feature activation+0.000
away
Token away
Feature activation+0.000
from
Token from
Feature activation+0.000
historic
Token historic
Feature activation+0.000
US
Token US
Feature activation+0.553
support
Token support
Feature activation+0.000
for
Token for
Feature activation+0.000
Palestinian
Token Palestinian
Feature activation+0.000
state
Token state
Feature activation+0.000
.
Token.
Feature activation+0.000
UN
Token UN
Feature activation+0.073
after
Token after
Feature activation+0.000
large
Token large
Feature activation+0.000
numbers
Token numbers
Feature activation+0.000
of
Token of
Feature activation+0.000
United
Token United
Feature activation+0.610
States
Token States
Feature activation+0.409
(
Token (
Feature activation+0.000
US
TokenUS
Feature activation+0.277
)
Token)
Feature activation+0.000
and
Token and
Feature activation+0.000
of
Token of
Feature activation+0.000
beef
Token beef
Feature activation+0.000
while
Token while
Feature activation+0.000
an
Token an
Feature activation+0.000
average
Token average
Feature activation+0.000
Japanese
Token Japanese
Feature activation+0.621
ate
Token ate
Feature activation+0.000
14
Token 14
Feature activation+0.000
.
Token.
Feature activation+0.000
9
Token9
Feature activation+0.000
kg
Token kg
Feature activation+0.000
developed
Token developed
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
National
Token National
Feature activation+0.000
Security
Token Security
Feature activation+0.000
Agency
Token Agency
Feature activation+0.684
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
hacking
Token hacking
Feature activation+0.000
tool
Token tool
Feature activation+0.000
,
Token,
Feature activation+0.000

INTERVAL 0.270 - 0.540
CONTAINS 0.137%

officer
Token officer
Feature activation+0.000
linked
Token linked
Feature activation+0.000
to
Token to
Feature activation+0.000
al
Token al
Feature activation+0.189
-
Token-
Feature activation+0.000
Qaeda
TokenQaeda
Feature activation+0.436
group
Token group
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
country
Token country
Feature activation+0.114
.
Token.
Feature activation+0.000
.
Token.
Feature activation+0.000
assistant
Token assistant
Feature activation+0.000
secretary
Token secretary
Feature activation+0.000
general
Token general
Feature activation+0.000
for
Token for
Feature activation+0.000
peace
Token peace
Feature activation+0.374
keeping
Tokenkeeping
Feature activation+0.000
operations
Token operations
Feature activation+0.000
,
Token,
Feature activation+0.000
told
Token told
Feature activation+0.000
reporters
Token reporters
Feature activation+0.000
herds
Token herds
Feature activation+0.000
of
Token of
Feature activation+0.000
elderly
Token elderly
Feature activation+0.000
tourists
Token tourists
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.000
American
TokenAmerican
Feature activation+0.502
,
Token,
Feature activation+0.000
Japanese
Token Japanese
Feature activation+0.707
,
Token,
Feature activation+0.000
British
Token British
Feature activation+0.224
âĢĶ
TokenâĢĶ
Feature activation+0.000
-
Token-
Feature activation+0.000
long
Tokenlong
Feature activation+0.000
partnership
Token partnership
Feature activation+0.000
between
Token between
Feature activation+0.000
the
Token the
Feature activation+0.000
British
Token British
Feature activation+0.531
Council
Token Council
Feature activation+0.000
,
Token,
Feature activation+0.000
BBC
Token BBC
Feature activation+0.070
World
Token World
Feature activation+0.000
Service
Token Service
Feature activation+0.000
Dogs
Token Dogs
Feature activation+0.000
,
Token,
Feature activation+0.000
our
Token our
Feature activation+0.000
team
Token team
Feature activation+0.000
in
Token in
Feature activation+0.000
Australia
Token Australia
Feature activation+0.534
sent
Token sent
Feature activation+0.000
voic
Token voic
Feature activation+0.000
email
Tokenemail
Feature activation+0.000
messages
Token messages
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 0.000 - 0.270
CONTAINS 99.764%

Bitcoin
TokenBitcoin
Feature activation+0.000
mining
Token mining
Feature activation+0.000
monopoly
Token monopoly
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
This
TokenThis
Feature activation+0.000
is
Token is
Feature activation+0.000
in
Token in
Feature activation+0.000
effect
Token effect
Feature activation+0.000
the
Token the
Feature activation+0.000
ED
Token ED
Feature activation+0.000
M
TokenM
Feature activation+0.000
1
Token 1
Feature activation+0.000
(
Token (
Feature activation+0.000
Ell
TokenEll
Feature activation+0.000
iston
Tokeniston
Feature activation+0.000
,
Token,
Feature activation+0.000
D
Token D
Feature activation+0.000
ies
Tokenies
Feature activation+0.000
endor
Tokenendor
Feature activation+0.000
f
Tokenf
Feature activation+0.000
confirmation
Token confirmation
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Republicans
TokenRepublicans
Feature activation+0.000
are
Token are
Feature activation+0.000
floating
Token floating
Feature activation+0.000
a
Token a
Feature activation+0.000
handful
Token handful
Feature activation+0.000
of
Token of
Feature activation+0.000
agency
Token agency
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
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
motion
Token motion
Feature activation+0.000
to
Token to
Feature activation+0.000
intervene
Token intervene
Feature activation+0.000
was
Token was
Feature activation+0.000
filed
Token filed
Feature activation+0.000
occur
Token occur
Feature activation+0.000
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
D
TokenD
Feature activation+0.000
ating
Tokenating
Feature activation+0.000
back
Token back
Feature activation+0.000
to
Token to
Feature activation+0.000
2010
Token 2010
Feature activation+0.000
,
Token,
Feature activation+0.000
Verizon
Token Verizon
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 H3.4: (feature 11020)

TOP ACTIVATIONS
MAX = 3.250

Campaign
Token Campaign
Feature activation+0.000
,
Token,
Feature activation+0.000
told
Token told
Feature activation+0.000
news
Token news
Feature activation+0.000
.
Token.
Feature activation+0.000
com
Tokencom
Feature activation+3.250
.
Token.
Feature activation+0.000
au
Tokenau
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
been
Token been
Feature activation+0.000
stricter
Token stricter
Feature activation+0.000
than
Token than
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+3.181
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
recent
Token recent
Feature activation+0.000
changes
Token changes
Feature activation+0.000
to
Token to
Feature activation+0.000
diplomatic
Token diplomatic
Feature activation+0.000
and
Token and
Feature activation+0.000
military
Token military
Feature activation+0.000
support
Token support
Feature activation+0.000
for
Token for
Feature activation+0.000
US
Token US
Feature activation+2.959
and
Token and
Feature activation+0.000
Israel
Token Israel
Feature activation+0.528
policy
Token policy
Feature activation+0.000
and
Token and
Feature activation+0.000
actions
Token actions
Feature activation+0.000
government
Token government
Feature activation+0.189
agencies
Token agencies
Feature activation+0.061
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
United
Token United
Feature activation+1.431
States
Token States
Feature activation+2.815
have
Token have
Feature activation+0.000
acted
Token acted
Feature activation+0.000
to
Token to
Feature activation+0.000
cover
Token cover
Feature activation+0.000
up
Token up
Feature activation+0.000
we
Token we
Feature activation+0.000
grow
Tokengrow
Feature activation+0.000
bit
Tokenbit
Feature activation+0.000
coins
Tokencoins
Feature activation+0.000
.
Token.
Feature activation+0.000
com
Tokencom
Feature activation+2.636
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
referred
Token referred
Feature activation+0.000
them
Token them
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
geopolitical
Token geopolitical
Feature activation+0.000
goals
Token goals
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+2.542
and
Token and
Feature activation+0.000
Israel
Token Israel
Feature activation+0.239
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Middle
Token Middle
Feature activation+0.217
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
the
Token the
Feature activation+0.000
U
Token U
Feature activation+0.645
.
Token.
Feature activation+1.738
S
TokenS
Feature activation+2.414
.,
Token.,
Feature activation+0.000
the
Token the
Feature activation+0.000
rules
Token rules
Feature activation+0.000
are
Token are
Feature activation+0.000
much
Token much
Feature activation+0.000
Australian
Token Australian
Feature activation+0.006
dollar
Token dollar
Feature activation+0.000
,
Token,
Feature activation+0.000
against
Token against
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+2.345
dollar
Token dollar
Feature activation+0.000
,
Token,
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
interest
Token interest
Feature activation+0.000
are
Token are
Feature activation+0.000
realistic
Token realistic
Feature activation+0.000
,
Token,
Feature activation+0.000
winning
Token winning
Feature activation+0.000
the
Token the
Feature activation+0.000
Russia
Token Russia
Feature activation+2.343
game
Token game
Feature activation+0.000
next
Token next
Feature activation+0.000
week
Token week
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
within
Token within
Feature activation+0.000
a
Token a
Feature activation+0.000
hair
Token hair
Feature activation+0.000
of
Token of
Feature activation+0.000
where
Token where
Feature activation+0.000
Japan
Token Japan
Feature activation+1.938
was
Token was
Feature activation+0.000
in
Token in
Feature activation+0.000
Q
Token Q
Feature activation+0.182
3
Token3
Feature activation+0.000
1993
Token 1993
Feature activation+0.000
secured
Token secured
Feature activation+0.000
assurances
Token assurances
Feature activation+0.000
from
Token from
Feature activation+0.000
U
Token U
Feature activation+0.786
.
Token.
Feature activation+0.843
S
TokenS
Feature activation+1.774
.
Token.
Feature activation+0.818
President
Token President
Feature activation+0.201
Donald
Token Donald
Feature activation+0.010
Trump
Token Trump
Feature activation+0.330
,
Token,
Feature activation+0.000
plain
Token plain
Feature activation+0.000
packaging
Token packaging
Feature activation+0.000
legislation
Token legislation
Feature activation+0.000
has
Token has
Feature activation+0.000
in
Token in
Feature activation+0.000
Australia
Token Australia
Feature activation+1.750
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
hasn
Token hasn
Feature activation+0.000
't
Token't
Feature activation+0.000
ruled
Token ruled
Feature activation+0.000
upon
Token upon
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
US
Token US
Feature activation+1.740
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
standard
Token standard
Feature activation+0.000
dupl
Token dupl
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
the
Token the
Feature activation+0.000
U
Token U
Feature activation+0.645
.
Token.
Feature activation+1.738
S
TokenS
Feature activation+2.414
.,
Token.,
Feature activation+0.000
the
Token the
Feature activation+0.000
rules
Token rules
Feature activation+0.000
are
Token are
Feature activation+0.000
UK
Token UK
Feature activation+0.000
terminals
Token terminals
Feature activation+0.000
and
Token and
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+1.720
,
Token,
Feature activation+0.000
amid
Token amid
Feature activation+0.000
security
Token security
Feature activation+0.000
service
Token service
Feature activation+0.000
fears
Token fears
Feature activation+0.000
on
Token on
Feature activation+0.000
Canadian
Token Canadian
Feature activation+0.170
soil
Token soil
Feature activation+0.000
with
Token with
Feature activation+0.000
exclusively
Token exclusively
Feature activation+0.000
American
Token American
Feature activation+1.709
iron
Token iron
Feature activation+0.000
and
Token and
Feature activation+0.000
steel
Token steel
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
men
Token men
Feature activation+0.000
captured
Token captured
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+1.673
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Jamie
TokenJamie
Feature activation+0.000
tells
Token tells
Feature activation+0.000
M
Token M
Feature activation+0.200
pta
Tokenpta
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
NEW
TokenNEW
Feature activation+0.000
DEL
Token DEL
Feature activation+0.312
HI
TokenHI
Feature activation+1.652
,
Token,
Feature activation+0.000
July
Token July
Feature activation+0.000
8
Token 8
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
to
Token to
Feature activation+0.000
announce
Token announce
Feature activation+0.000
the
Token the
Feature activation+0.000
Canada
Token Canada
Feature activation+0.000
-
Token-
Feature activation+0.000
US
TokenUS
Feature activation+1.644
Council
Token Council
Feature activation+0.000
for
Token for
Feature activation+0.000
Advance
Token Advance
Feature activation+0.000
ment
Tokenment
Feature activation+0.000
of
Token of
Feature activation+0.000
ous
Tokenous
Feature activation+0.000
propaganda
Token propaganda
Feature activation+0.000
has
Token has
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+0.000
American
Token American
Feature activation+1.639
People
Token People
Feature activation+0.000
a
Token a
Feature activation+0.000
totally
Token totally
Feature activation+0.000
false
Token false
Feature activation+0.000
view
Token view
Feature activation+0.000

Top DFA by src position
MAX = 3.883

-
Token-
Feature activation-0.002
Top resid features:
ord
Tokenord
Feature activation+0.004
Top resid features:
inator
Tokeninator
Feature activation-0.010
Top resid features:
of
Token of
Feature activation-0.002
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
Australian
Token Australian
Feature activation+3.883
Top resid features:
Anti
Token Anti
Feature activation+0.007
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
B
TokenB
Feature activation+0.002
Top resid features:
ases
Tokenases
Feature activation+0.001
Top resid features:
Campaign
Token Campaign
Feature activation-0.011
Top resid features:
result
Token result
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
ranks
Token ranks
Feature activation-0.000
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
Canadian
Token Canadian
Feature activation+1.719
Top resid features:
mortgage
Token mortgage
Feature activation+0.000
Top resid features:
borrowers
Token borrowers
Feature activation-0.000
Top resid features:
with
Token with
Feature activation-0.001
Top resid features:
high
Token high
Feature activation-0.000
Top resid features:
loan
Token loan
Feature activation-0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
present
Token present
Feature activation+0.002
Top resid features:
political
Token political
Feature activation+0.011
Top resid features:
climate
Token climate
Feature activation-0.002
Top resid features:
in
Token in
Feature activation-0.002
Top resid features:
Canada
Token Canada
Feature activation+1.581
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
cases
Token cases
Feature activation-0.002
Top resid features:
where
Token where
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.004
Top resid features:
expression
Token expression
Feature activation+0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.100
Top resid features:
believers
Token believers
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
Many
Token Many
Feature activation-0.003
Top resid features:
Canadians
Token Canadians
Feature activation+3.043
Top resid features:
would
Token would
Feature activation+0.002
Top resid features:
suggest
Token suggest
Feature activation-0.002
Top resid features:
this
Token this
Feature activation+0.002
Top resid features:
is
Token is
Feature activation+0.001
Top resid features:
by
Token by
Feature activation+0.002
Top resid features:
Consumer
TokenConsumer
Feature activation+0.000
Top resid features:
Law
Token Law
Feature activation-0.001
Top resid features:
Specialist
Token Specialist
Feature activation-0.002
Top resid features:
:
Token:
Feature activation-0.007
Top resid features:
New
Token New
Feature activation+0.002
Top resid features:
Zealand
Token Zealand
Feature activation+3.093
Top resid features:
Business
Token Business
Feature activation+0.005
Top resid features:
âĢ
Token âĢ
Feature activation-0.001
Top resid features:
ľ
Tokenľ
Feature activation-0.000
Top resid features:
A
TokenA
Feature activation-0.000
Top resid features:
Simple
Token Simple
Feature activation-0.002
Top resid features:
negatively
Token negatively
Feature activation-0.001
Top resid features:
impact
Token impact
Feature activation+0.000
Top resid features:
public
Token public
Feature activation+0.001
Top resid features:
opinion
Token opinion
Feature activation-0.000
Top resid features:
about
Token about
Feature activation+0.000
Top resid features:
Canada
Token Canada
Feature activation+1.346
Top resid features:
âĢ
TokenâĢ
Feature activation+0.001
Top resid features:
Ļ
TokenĻ
Feature activation+0.004
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
diplomatic
Token diplomatic
Feature activation+0.019
Top resid features:
and
Token and
Feature activation+0.002
Top resid features:
now
Token now
Feature activation+0.001
Top resid features:
."
Token."
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Under
TokenUnder
Feature activation+0.000
Top resid features:
Canadian
Token Canadian
Feature activation+2.185
Top resid features:
rules
Token rules
Feature activation-0.003
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
if
Token if
Feature activation+0.000
Top resid features:
a
Token a
Feature activation-0.001
Top resid features:
drone
Token drone
Feature activation-0.003
Top resid features:
an
Token an
Feature activation-0.000
Top resid features:
asset
Token asset
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
The
TokenThe
Feature activation-0.003
Top resid features:
Australian
Token Australian
Feature activation+0.931
Top resid features:
dollar
Token dollar
Feature activation+0.022
Top resid features:
is
Token is
Feature activation-0.002
Top resid features:
just
Token just
Feature activation+0.000
Top resid features:
like
Token like
Feature activation-0.000
Top resid features:
any
Token any
Feature activation-0.002
Top resid features:
The
TokenThe
Feature activation-0.001
Top resid features:
fact
Token fact
Feature activation-0.000
Top resid features:
we
Token we
Feature activation+0.000
Top resid features:
're
Token're
Feature activation-0.000
Top resid features:
playing
Token playing
Feature activation+0.000
Top resid features:
Ireland
Token Ireland
Feature activation+1.879
Top resid features:
in
Token in
Feature activation-0.003
Top resid features:
the
Token the
Feature activation+0.007
Top resid features:
group
Token group
Feature activation+0.000
Top resid features:
stage
Token stage
Feature activation+0.000
Top resid features:
is
Token is
Feature activation-0.001
Top resid features:
!
Token!
Feature activation-0.002
Top resid features:
Private
Token Private
Feature activation+0.003
Top resid features:
sector
Token sector
Feature activation+0.004
Top resid features:
debt
Token debt
Feature activation-0.000
Top resid features:
in
Token in
Feature activation-0.006
Top resid features:
Canada
Token Canada
Feature activation+1.608
Top resid features:
has
Token has
Feature activation+0.000
Top resid features:
more
Token more
Feature activation+0.000
Top resid features:
than
Token than
Feature activation-0.008
Top resid features:
doubled
Token doubled
Feature activation-0.002
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
made
Token made
Feature activation+0.000
Top resid features:
clear
Token clear
Feature activation-0.001
Top resid features:
she
Token she
Feature activation-0.004
Top resid features:
plans
Token plans
Feature activation-0.001
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
Britain
Token Britain
Feature activation+1.581
Top resid features:
to
Token to
Feature activation+0.005
Top resid features:
leave
Token leave
Feature activation+0.023
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
EU
Token EU
Feature activation+0.504
Top resid features:
âĢ
TokenâĢ
Feature activation-0.003
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.078
Top resid features:
.
Token.
Feature activation-0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Health
TokenHealth
Feature activation-0.006
Top resid features:
Canada
Token Canada
Feature activation+2.022
Top resid features:
said
Token said
Feature activation-0.009
Top resid features:
it
Token it
Feature activation-0.005
Top resid features:
is
Token is
Feature activation-0.002
Top resid features:
watching
Token watching
Feature activation+0.005
Top resid features:
with
Token with
Feature activation-0.004
Top resid features:
Terrorism
Token Terrorism
Feature activation-0.011
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
The
TokenThe
Feature activation-0.010
Top resid features:
British
Token British
Feature activation+1.907
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
Irish
TokenIrish
Feature activation+0.183
Top resid features:
insurrection
Token insurrection
Feature activation+0.000
Top resid features:
lasted
Token lasted
Feature activation+0.002
Top resid features:
from
Token from
Feature activation-0.001
Top resid features:
now
Token now
Feature activation+0.000
Top resid features:
."
Token."
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Under
TokenUnder
Feature activation-0.000
Top resid features:
Canadian
Token Canadian
Feature activation+1.592
Top resid features:
rules
Token rules
Feature activation-0.003
Top resid features:
,
Token,
Feature activation-0.001
Top resid features:
if
Token if
Feature activation+0.001
Top resid features:
a
Token a
Feature activation-0.001
Top resid features:
drone
Token drone
Feature activation-0.013
Top resid features:
with
Token with
Feature activation+0.005
Top resid features:
massive
Token massive
Feature activation-0.009
Top resid features:
disruption
Token disruption
Feature activation+0.005
Top resid features:
at
Token at
Feature activation-0.001
Top resid features:
major
Token major
Feature activation-0.014
Top resid features:
UK
Token UK
Feature activation+2.021
Top resid features:
terminals
Token terminals
Feature activation-0.011
Top resid features:
and
Token and
Feature activation-0.007
Top resid features:
in
Token in
Feature activation+0.014
Top resid features:
the
Token the
Feature activation-0.051
Top resid features:
US
Token US
Feature activation-0.019
Top resid features:
's
Token's
Feature activation+0.000
Top resid features:
plan
Token plan
Feature activation-0.001
Top resid features:
to
Token to
Feature activation+0.011
Top resid features:
build
Token build
Feature activation+0.010
Top resid features:
on
Token on
Feature activation+0.007
Top resid features:
Canadian
Token Canadian
Feature activation+0.776
Top resid features:
soil
Token soil
Feature activation+0.002
Top resid features:
with
Token with
Feature activation+0.004
Top resid features:
exclusively
Token exclusively
Feature activation-0.155
Top resid features:
American
Token American
Feature activation+0.005
Top resid features:
iron
Token iron
Feature activation+0.000
Top resid features:
's
Token's
Feature activation-0.005
Top resid features:
attempt
Token attempt
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
take
Token take
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.003
Top resid features:
Scottish
Token Scottish
Feature activation+0.960
Top resid features:
throne
Token throne
Feature activation+0.001
Top resid features:
is
Token is
Feature activation-0.002
Top resid features:
going
Token going
Feature activation+0.001
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
fail
Token fail
Feature activation-0.000
Top resid features:
explore
Token explore
Feature activation-0.001
Top resid features:
Te
Token Te
Feature activation+0.001
Top resid features:
ck
Tokenck
Feature activation+0.001
Top resid features:
mines
Token mines
Feature activation-0.003
Top resid features:
in
Token in
Feature activation-0.002
Top resid features:
Canada
Token Canada
Feature activation+1.792
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
*
Token*
Feature activation-0.000
Top resid features:
India
Token India
Feature activation+0.025
Top resid features:
seeks
Token seeks
Feature activation-0.001
Top resid features:
/
Token/
Feature activation-0.003
Top resid features:
Kevin
TokenKevin
Feature activation+0.002
Top resid features:
Lamar
Token Lamar
Feature activation+0.000
Top resid features:
que
Tokenque
Feature activation-0.000
Top resid features:
)
Token)
Feature activation+0.002
Top resid features:
Canadian
Token Canadian
Feature activation+1.002
Top resid features:
PM
Token PM
Feature activation+0.011
Top resid features:
Justin
Token Justin
Feature activation+0.011
Top resid features:
Trudeau
Token Trudeau
Feature activation+0.313
Top resid features:
comes
Token comes
Feature activation+0.002
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
Terrorism
Token Terrorism
Feature activation-0.021
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
The
TokenThe
Feature activation-0.011
Top resid features:
British
Token British
Feature activation+1.809
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
Irish
TokenIrish
Feature activation+0.155
Top resid features:
insurrection
Token insurrection
Feature activation-0.012
Top resid features:
lasted
Token lasted
Feature activation+0.001
Top resid features:
from
Token from
Feature activation-0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.09

Head 1: 0.04

Head 2: 0.04

Head 3: 0.02

Head 4: 0.48

Head 5: 0.04

Head 6: 0.08

Head 7: 0.04

Head 8: 0.04

Head 9: 0.02

Head 10: 0.06

Head 11: 0.03

Positive logits

Rex2.06

bol1.89

Shape1.86

rained1.81

CLASSIFIED1.79

igree1.77

inx1.76

optim1.74

>>>1.68

untu1.68

bark1.68

asso1.66

ety1.66

actionDate1.65

bre1.65

Poles1.63

cookie1.62

pload1.62

══1.62

Lens1.62

Negative logits

guid-1.99

DOC-1.96

nu-1.88

glim-1.87

minister-1.86

Celebr-1.81

unfocusedRange-1.79

andise-1.79

merce-1.78

MLA-1.78

condem-1.78

abama-1.78

)",-1.78

ministry-1.75

UD-1.74

Leilan-1.74

conduc-1.73

RELE-1.72

treasury-1.70

mourning-1.69

INTERVAL 2.925 - 3.250
CONTAINS 0.000%

diplomatic
Token diplomatic
Feature activation+0.000
and
Token and
Feature activation+0.000
military
Token military
Feature activation+0.000
support
Token support
Feature activation+0.000
for
Token for
Feature activation+0.000
US
Token US
Feature activation+2.959
and
Token and
Feature activation+0.000
Israel
Token Israel
Feature activation+0.528
policy
Token policy
Feature activation+0.000
and
Token and
Feature activation+0.000
actions
Token actions
Feature activation+0.000
Campaign
Token Campaign
Feature activation+0.000
,
Token,
Feature activation+0.000
told
Token told
Feature activation+0.000
news
Token news
Feature activation+0.000
.
Token.
Feature activation+0.000
com
Tokencom
Feature activation+3.250
.
Token.
Feature activation+0.000
au
Tokenau
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
been
Token been
Feature activation+0.000
stricter
Token stricter
Feature activation+0.000
than
Token than
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+3.181
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
recent
Token recent
Feature activation+0.000
changes
Token changes
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 2.600 - 2.925
CONTAINS 0.000%

we
Token we
Feature activation+0.000
grow
Tokengrow
Feature activation+0.000
bit
Tokenbit
Feature activation+0.000
coins
Tokencoins
Feature activation+0.000
.
Token.
Feature activation+0.000
com
Tokencom
Feature activation+2.636
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
referred
Token referred
Feature activation+0.000
them
Token them
Feature activation+0.000
to
Token to
Feature activation+0.000
government
Token government
Feature activation+0.189
agencies
Token agencies
Feature activation+0.061
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
United
Token United
Feature activation+1.431
States
Token States
Feature activation+2.815
have
Token have
Feature activation+0.000
acted
Token acted
Feature activation+0.000
to
Token to
Feature activation+0.000
cover
Token cover
Feature activation+0.000
up
Token up
Feature activation+0.000

INTERVAL 2.275 - 2.600
CONTAINS 0.000%

are
Token are
Feature activation+0.000
realistic
Token realistic
Feature activation+0.000
,
Token,
Feature activation+0.000
winning
Token winning
Feature activation+0.000
the
Token the
Feature activation+0.000
Russia
Token Russia
Feature activation+2.343
game
Token game
Feature activation+0.000
next
Token next
Feature activation+0.000
week
Token week
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
the
Token the
Feature activation+0.000
geopolitical
Token geopolitical
Feature activation+0.000
goals
Token goals
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+2.542
and
Token and
Feature activation+0.000
Israel
Token Israel
Feature activation+0.239
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Middle
Token Middle
Feature activation+0.217
Australian
Token Australian
Feature activation+0.006
dollar
Token dollar
Feature activation+0.000
,
Token,
Feature activation+0.000
against
Token against
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+2.345
dollar
Token dollar
Feature activation+0.000
,
Token,
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
interest
Token interest
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
the
Token the
Feature activation+0.000
U
Token U
Feature activation+0.645
.
Token.
Feature activation+1.738
S
TokenS
Feature activation+2.414
.,
Token.,
Feature activation+0.000
the
Token the
Feature activation+0.000
rules
Token rules
Feature activation+0.000
are
Token are
Feature activation+0.000
much
Token much
Feature activation+0.000

INTERVAL 1.950 - 2.275
CONTAINS 0.000%

INTERVAL 1.625 - 1.950
CONTAINS 0.001%

men
Token men
Feature activation+0.000
captured
Token captured
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+1.673
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Jamie
TokenJamie
Feature activation+0.000
tells
Token tells
Feature activation+0.000
M
Token M
Feature activation+0.200
secured
Token secured
Feature activation+0.000
assurances
Token assurances
Feature activation+0.000
from
Token from
Feature activation+0.000
U
Token U
Feature activation+0.786
.
Token.
Feature activation+0.843
S
TokenS
Feature activation+1.774
.
Token.
Feature activation+0.818
President
Token President
Feature activation+0.201
Donald
Token Donald
Feature activation+0.010
Trump
Token Trump
Feature activation+0.330
,
Token,
Feature activation+0.000
plain
Token plain
Feature activation+0.000
packaging
Token packaging
Feature activation+0.000
legislation
Token legislation
Feature activation+0.000
has
Token has
Feature activation+0.000
in
Token in
Feature activation+0.000
Australia
Token Australia
Feature activation+1.750
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
hasn
Token hasn
Feature activation+0.000
't
Token't
Feature activation+0.000
ruled
Token ruled
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
the
Token the
Feature activation+0.000
U
Token U
Feature activation+0.645
.
Token.
Feature activation+1.738
S
TokenS
Feature activation+2.414
.,
Token.,
Feature activation+0.000
the
Token the
Feature activation+0.000
rules
Token rules
Feature activation+0.000
are
Token are
Feature activation+0.000
upon
Token upon
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
US
Token US
Feature activation+1.740
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
standard
Token standard
Feature activation+0.000
dupl
Token dupl
Feature activation+0.000

INTERVAL 1.300 - 1.625
CONTAINS 0.002%

organisations
Token organisations
Feature activation+0.000
(
Token (
Feature activation+0.000
NG
TokenNG
Feature activation+0.061
Os
TokenOs
Feature activation+0.110
).
Token).
Feature activation+0.000
US
Token US
Feature activation+1.515
firm
Token firm
Feature activation+0.000
Ed
Token Ed
Feature activation+0.000
elman
Tokenelman
Feature activation+0.127
is
Token is
Feature activation+0.000
one
Token one
Feature activation+0.000
was
Token was
Feature activation+0.000
that
Token that
Feature activation+0.000
coming
Token coming
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
UK
Token UK
Feature activation+1.345
he
Token he
Feature activation+0.000
wasn
Token wasn
Feature activation+0.000
't
Token't
Feature activation+0.000
wedd
Token wedd
Feature activation+0.000
ed
Tokened
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
determined
Token determined
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
Commission
Token Commission
Feature activation+1.498
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
a
Token a
Feature activation+0.000
non
Token non
Feature activation+0.000
-
Token-
Feature activation+0.000
PM
Token PM
Feature activation+0.000
Benjamin
Token Benjamin
Feature activation+0.000
Netanyahu
Token Netanyahu
Feature activation+0.586
visits
Token visits
Feature activation+0.000
White
Token White
Feature activation+0.337
House
Token House
Feature activation+1.333
and
Token and
Feature activation+0.000
Trump
Token Trump
Feature activation+0.988
says
Token says
Feature activation+0.000
he
Token he
Feature activation+0.000
"
Token "
Feature activation+0.000
by
Token by
Feature activation+0.000
2020
Token 2020
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
FR
TokenFR
Feature activation+1.305
V
TokenV
Feature activation+0.000
chief
Token chief
Feature activation+0.000
executive
Token executive
Feature activation+0.000
officer
Token officer
Feature activation+0.088
Rafael
Token Rafael
Feature activation+0.000

INTERVAL 0.975 - 1.300
CONTAINS 0.004%

chance
Token chance
Feature activation+0.000
political
Token political
Feature activation+0.000
developments
Token developments
Feature activation+0.000
such
Token such
Feature activation+0.000
as
Token as
Feature activation+0.000
Brexit
Token Brexit
Feature activation+1.290
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
election
Token election
Feature activation+0.276
of
Token of
Feature activation+0.000
President
Token President
Feature activation+0.528
the
Token the
Feature activation+0.000
second
Token second
Feature activation+0.000
example
Token example
Feature activation+0.000
provided
Token provided
Feature activation+0.000
by
Token by
Feature activation+0.000
HRC
Token HRC
Feature activation+1.284
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
its
Token its
Feature activation+0.000
comment
Token comment
Feature activation+0.000
HRC
Token HRC
Feature activation+0.762
regarding
Token regarding
Feature activation+0.000
abusive
Token abusive
Feature activation+0.000
comment
Token comment
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
Commission
Token Commission
Feature activation+1.219
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
pred
Token pred
Feature activation+0.000
isp
Tokenisp
Feature activation+0.000
osed
Tokenosed
Feature activation+0.000
senior
Token senior
Feature activation+0.000
policy
Token policy
Feature activation+0.000
analyst
Token analyst
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
Canadian
Token Canadian
Feature activation+1.029
Cancer
Token Cancer
Feature activation+0.000
Society
Token Society
Feature activation+0.000
,
Token,
Feature activation+0.000
called
Token called
Feature activation+0.000
the
Token the
Feature activation+0.000
Freed
Token Freed
Feature activation+0.000
oms
Tokenoms
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
O
Token O
Feature activation+0.147
CLA
TokenCLA
Feature activation+1.183
also
Token also
Feature activation+0.000
advocates
Token advocates
Feature activation+0.000
for
Token for
Feature activation+0.000
unim
Token unim
Feature activation+0.000
ped
Tokenped
Feature activation+0.000

INTERVAL 0.650 - 0.975
CONTAINS 0.014%

N
TokenN
Feature activation+0.388
aur
Tokenaur
Feature activation+0.000
u
Tokenu
Feature activation+0.000
ans
Tokenans
Feature activation+0.000
,
Token,
Feature activation+0.000
Germans
Token Germans
Feature activation+0.870
,
Token,
Feature activation+0.000
Australians
Token Australians
Feature activation+0.078
,
Token,
Feature activation+0.000
B
Token B
Feature activation+0.080
PC
TokenPC
Feature activation+0.000
said
Token said
Feature activation+0.000
at
Token at
Feature activation+0.000
a
Token a
Feature activation+0.000
rally
Token rally
Feature activation+0.000
in
Token in
Feature activation+0.000
Pennsylvania
Token Pennsylvania
Feature activation+0.664
to
Token to
Feature activation+0.000
mark
Token mark
Feature activation+0.000
his
Token his
Feature activation+0.000
100
Token 100
Feature activation+0.000
th
Tokenth
Feature activation+0.000
annoying
Token annoying
Feature activation+0.000
Great
Token Great
Feature activation+0.000
Britain
Token Britain
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
US
Token US
Feature activation+0.788
has
Token has
Feature activation+0.000
a
Token a
Feature activation+0.000
number
Token number
Feature activation+0.000
of
Token of
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ji
Tokenji
Feature activation+0.000
Lloyd
Token Lloyd
Feature activation+0.000
are
Token are
Feature activation+0.000
partnering
Token partnering
Feature activation+0.000
with
Token with
Feature activation+0.000
Israeli
Token Israeli
Feature activation+0.764
Weapons
Token Weapons
Feature activation+0.000
Industries
Token Industries
Feature activation+0.000
to
Token to
Feature activation+0.000
open
Token open
Feature activation+0.000
the
Token the
Feature activation+0.000
up
Token up
Feature activation+0.000
a
Token a
Feature activation+0.000
deal
Token deal
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+0.693
international
Token international
Feature activation+0.000
,
Token,
Feature activation+0.000
despite
Token despite
Feature activation+0.000
the
Token the
Feature activation+0.000
likes
Token likes
Feature activation+0.000

INTERVAL 0.325 - 0.650
CONTAINS 0.050%

Ontario
Token Ontario
Feature activation+0.000
,
Token,
Feature activation+0.000
Quebec
Token Quebec
Feature activation+0.000
and
Token and
Feature activation+0.000
other
Token other
Feature activation+0.000
provinces
Token provinces
Feature activation+0.483
put
Token put
Feature activation+0.000
in
Token in
Feature activation+0.000
place
Token place
Feature activation+0.000
stop
Token stop
Feature activation+0.000
gap
Tokengap
Feature activation+0.000
factory
Token factory
Feature activation+0.000
jobs
Token jobs
Feature activation+0.000
she
Token she
Feature activation+0.000
had
Token had
Feature activation+0.000
in
Token in
Feature activation+0.000
New
Token New
Feature activation+0.485
York
Token York
Feature activation+0.128
,
Token,
Feature activation+0.000
though
Token though
Feature activation+0.000
she
Token she
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
UK
Token UK
Feature activation+0.000
nationals
Token nationals
Feature activation+0.000
living
Token living
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.013
EU
Token EU
Feature activation+0.440
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
added
Token added
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
seen
Token seen
Feature activation+0.000
laughing
Token laughing
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
the
Token the
Feature activation+0.000
NDP
Token NDP
Feature activation+0.374
leader
Token leader
Feature activation+0.000
then
Token then
Feature activation+0.000
seems
Token seems
Feature activation+0.000
to
Token to
Feature activation+0.000
consult
Token consult
Feature activation+0.000
is
Token is
Feature activation+0.000
enshr
Token enshr
Feature activation+0.000
ined
Tokenined
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Constitution
Token Constitution
Feature activation+0.489
,
Token,
Feature activation+0.000
is
Token is
Feature activation+0.000
taken
Token taken
Feature activation+0.000
in
Token in
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.047

INTERVAL 0.000 - 0.325
CONTAINS 99.927%

speech
Token speech
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
Braz
Token Braz
Feature activation+0.000
au
Tokenau
Feature activation+0.000
had
Token had
Feature activation+0.000
argued
Token argued
Feature activation+0.000
,
Token,
Feature activation+0.000
C
Token C
Feature activation+0.000
lements
Tokenlements
Feature activation+0.000
said
Token said
Feature activation+0.000
one
Token one
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
thinking
Token thinking
Feature activation+0.000
,
Token,
Feature activation+0.000
one
Token one
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
self
Token self
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
1
Token1
Feature activation+0.000
Set
Token Set
Feature activation+0.000
up
Token up
Feature activation+0.000
on
Token on
Feature activation+0.000
Trip
Token Trip
Feature activation+0.000
od
Tokenod
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Before
TokenBefore
Feature activation+0.000
you
Token you
Feature activation+0.000
by
Token by
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Station
TokenStation
Feature activation+0.000
ed
Tokened
Feature activation+0.000
around
Token around
Feature activation+0.000
the
Token the
Feature activation+0.000
coll
Token coll
Feature activation+0.000
ider
Tokenider
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
September
Token September
Feature activation+0.000
11
Token 11
Feature activation+0.000
attacks
Token attacks
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
in
Token in
Feature activation+0.000
2001
Token 2001
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 H3.4: (feature 20415)

TOP ACTIVATIONS
MAX = 2.024

Israel
Token Israel
Feature activation+0.000
quietly
Token quietly
Feature activation+0.000
worked
Token worked
Feature activation+0.000
to
Token to
Feature activation+0.000
upgrade
Token upgrade
Feature activation+0.000
China
Token China
Feature activation+2.024
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
military
Token military
Feature activation+0.000
arsenal
Token arsenal
Feature activation+0.000
Now
TokenNow
Feature activation+0.000
celebrating
Token celebrating
Feature activation+0.000
25
Token 25
Feature activation+0.000
years
Token years
Feature activation+0.000
,
Token,
Feature activation+0.000
India
Token India
Feature activation+2.014
and
Token and
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
must
Token must
Feature activation+0.000
make
Token make
Feature activation+0.000
connecting
Token connecting
Feature activation+0.000
flights
Token flights
Feature activation+0.000
in
Token in
Feature activation+0.000
Hong
Token Hong
Feature activation+1.823
Kong
Token Kong
Feature activation+1.365
or
Token or
Feature activation+0.000
Seoul
Token Seoul
Feature activation+0.289
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ful
Tokenful
Feature activation+0.000
filled
Tokenfilled
Feature activation+0.000
needs
Token needs
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Indian
Token Indian
Feature activation+1.715
military
Token military
Feature activation+0.000
for
Token for
Feature activation+0.000
modern
Token modern
Feature activation+0.000
,
Token,
Feature activation+0.000
quality
Token quality
Feature activation+0.000
there
Token there
Feature activation+0.000
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
sense
Token sense
Feature activation+0.000
that
Token that
Feature activation+0.000
Bush
Token Bush
Feature activation+1.665
was
Token was
Feature activation+0.000
making
Token making
Feature activation+0.000
an
Token an
Feature activation+0.000
effort
Token effort
Feature activation+0.000
with
Token with
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
that
Token that
Feature activation+0.000
therefore
Token therefore
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+1.583
could
Token could
Feature activation+0.000
just
Token just
Feature activation+0.000
as
Token as
Feature activation+0.000
easily
Token easily
Feature activation+0.000
start
Token start
Feature activation+0.000
-
Token-
Feature activation+0.000
related
Tokenrelated
Feature activation+0.000
scandal
Token scandal
Feature activation+0.000
tied
Token tied
Feature activation+0.000
to
Token to
Feature activation+0.000
Georgia
Token Georgia
Feature activation+1.566
Democrats
Token Democrats
Feature activation+0.062
.
Token.
Feature activation+0.000
Over
Token Over
Feature activation+0.000
the
Token the
Feature activation+0.000
weekend
Token weekend
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
mouth
Token mouth
Feature activation+0.000
piece
Tokenpiece
Feature activation+0.000
of
Token of
Feature activation+0.000
US
Token US
Feature activation+1.509
establishment
Token establishment
Feature activation+0.000
liberalism
Token liberalism
Feature activation+0.000
,
Token,
Feature activation+0.000
weighed
Token weighed
Feature activation+0.000
in
Token in
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
state
Token state
Feature activation+0.046
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Trump
TokenTrump
Feature activation+1.499
âĢ
TokenâĢ
Feature activation+0.001
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
self
Token self
Feature activation+0.000
-
Token-
Feature activation+0.000
proposed
Token proposed
Feature activation+0.000
sale
Token sale
Feature activation+0.000
from
Token from
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
to
Token to
Feature activation+0.000
China
Token China
Feature activation+1.485
of
Token of
Feature activation+0.000
an
Token an
Feature activation+0.000
advanced
Token advanced
Feature activation+0.000
radar
Token radar
Feature activation+0.000
system
Token system
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
very
Token very
Feature activation+0.000
same
Token same
Feature activation+0.000
Lockheed
Token Lockheed
Feature activation+1.419
Martin
Token Martin
Feature activation+0.318
who
Token who
Feature activation+0.000
were
Token were
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+0.000
and
Token and
Feature activation+0.000
if
Token if
Feature activation+0.000
he
Token he
Feature activation+0.000
thinks
Token thinks
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+1.381
won
Token won
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
support
Token support
Feature activation+0.000
make
Token make
Feature activation+0.000
connecting
Token connecting
Feature activation+0.000
flights
Token flights
Feature activation+0.000
in
Token in
Feature activation+0.000
Hong
Token Hong
Feature activation+1.823
Kong
Token Kong
Feature activation+1.365
or
Token or
Feature activation+0.000
Seoul
Token Seoul
Feature activation+0.289
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
pro
Token pro
Feature activation+0.000
-
Token-
Feature activation+0.000
Israel
TokenIsrael
Feature activation+0.000
circles
Token circles
Feature activation+0.000
for
Token for
Feature activation+0.000
Hillary
Token Hillary
Feature activation+1.348
Clinton
Token Clinton
Feature activation+1.030
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
vice
Token vice
Feature activation+0.000
right
Token right
Feature activation+0.000
to
Token to
Feature activation+0.000
defend
Token defend
Feature activation+0.000
themselves
Token themselves
Feature activation+0.000
,"
Token,"
Feature activation+0.000
Clinton
Token Clinton
Feature activation+1.342
said
Token said
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
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ab
TokenAb
Feature activation+0.000
bas
Tokenbas
Feature activation+0.231
urged
Token urged
Feature activation+0.000
Bush
Token Bush
Feature activation+1.334
to
Token to
Feature activation+0.000
press
Token press
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
to
Token to
Feature activation+0.000
halt
Token halt
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Ambassador
Token Ambassador
Feature activation+0.000
to
Token to
Feature activation+0.000
India
Token India
Feature activation+1.299
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
Now
TokenNow
Feature activation+0.000
.[
Token.[
Feature activation+0.000
6
Token6
Feature activation+0.000
]
Token]
Feature activation+0.000
U
Token U
Feature activation+0.298
.
Token.
Feature activation+0.282
S
TokenS
Feature activation+1.295
.
Token.
Feature activation+0.418
intelligence
Token intelligence
Feature activation+0.000
believed
Token believed
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
Israel
TokenIsrael
Feature activation+0.000
's
Token's
Feature activation+0.000
Netanyahu
Token Netanyahu
Feature activation+0.000
says
Token says
Feature activation+0.000
admitting
Token admitting
Feature activation+0.000
Syrian
Token Syrian
Feature activation+1.281
refugees
Token refugees
Feature activation+0.000
will
Token will
Feature activation+0.000
allow
Token allow
Feature activation+0.000
terrorists
Token terrorists
Feature activation+0.174
in
Token in
Feature activation+0.000
adopt
Token adopt
Feature activation+0.000
the
Token the
Feature activation+0.000
most
Token most
Feature activation+0.000
anti
Token anti
Feature activation+0.000
-
Token-
Feature activation+0.000
US
TokenUS
Feature activation+1.267
rhetoric
Token rhetoric
Feature activation+0.000
if
Token if
Feature activation+0.000
he
Token he
Feature activation+0.000
thought
Token thought
Feature activation+0.000
that
Token that
Feature activation+0.000

Top DFA by src position
MAX = 2.543

China
Token China
Feature activation-0.038
Top resid features:
establish
Token establish
Feature activation+0.001
Top resid features:
diplomatic
Token diplomatic
Feature activation+0.002
Top resid features:
relations
Token relations
Feature activation+0.000
Top resid features:
with
Token with
Feature activation+0.002
Top resid features:
Israel
Token Israel
Feature activation+1.374
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
two
Token two
Feature activation+0.003
Top resid features:
countries
Token countries
Feature activation-0.002
Top resid features:
have
Token have
Feature activation+0.001
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
improved
Token improved
Feature activation-0.001
Top resid features:
advanced
Token advanced
Feature activation-0.004
Top resid features:
tools
Token tools
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Israel
Token Israel
Feature activation+1.402
Top resid features:
and
Token and
Feature activation+0.020
Top resid features:
India
Token India
Feature activation-0.119
Top resid features:
consider
Token consider
Feature activation-0.002
Top resid features:
their
Token their
Feature activation-0.003
Top resid features:
defense
Token defense
Feature activation+0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.005
Top resid features:
between
Token between
Feature activation-0.001
Top resid features:
Israel
Token Israel
Feature activation+2.322
Top resid features:
and
Token and
Feature activation+0.010
Top resid features:
Japan
Token Japan
Feature activation+0.012
Top resid features:
currently
Token currently
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.005
Top resid features:
and
Token and
Feature activation+0.011
Top resid features:
ji
Tokenji
Feature activation-0.002
Top resid features:
Lloyd
Token Lloyd
Feature activation+0.032
Top resid features:
are
Token are
Feature activation+0.003
Top resid features:
partnering
Token partnering
Feature activation+0.001
Top resid features:
with
Token with
Feature activation+0.001
Top resid features:
Israeli
Token Israeli
Feature activation+2.543
Top resid features:
Weapons
Token Weapons
Feature activation+0.008
Top resid features:
Industries
Token Industries
Feature activation+0.003
Top resid features:
to
Token to
Feature activation+0.004
Top resid features:
open
Token open
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
calling
Token calling
Feature activation-0.004
Top resid features:
for
Token for
Feature activation+0.002
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
removal
Token removal
Feature activation-0.001
Top resid features:
of
Token of
Feature activation+0.002
Top resid features:
Israeli
Token Israeli
Feature activation+1.811
Top resid features:
settlements
Token settlements
Feature activation+0.051
Top resid features:
and
Token and
Feature activation+0.007
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
right
Token right
Feature activation+0.006
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.000
Top resid features:
Israel
Token Israel
Feature activation+2.098
Top resid features:
is
Token is
Feature activation+0.001
Top resid features:
able
Token able
Feature activation+0.005
Top resid features:
to
Token to
Feature activation+0.006
Top resid features:
attack
Token attack
Feature activation+0.001
Top resid features:
Syrian
Token Syrian
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
This
TokenThis
Feature activation-0.007
Top resid features:
is
Token is
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.006
Top resid features:
second
Token second
Feature activation-0.008
Top resid features:
Israel
Token Israel
Feature activation+2.193
Top resid features:
-
Token-
Feature activation+0.002
Top resid features:
related
Tokenrelated
Feature activation+0.003
Top resid features:
scandal
Token scandal
Feature activation-0.024
Top resid features:
tied
Token tied
Feature activation-0.048
Top resid features:
to
Token to
Feature activation-0.061
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
fourth
Token fourth
Feature activation+0.005
Top resid features:
day
Token day
Feature activation+0.031
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
Israeli
Token Israeli
Feature activation+1.728
Top resid features:
aerial
Token aerial
Feature activation+0.008
Top resid features:
blitz
Token blitz
Feature activation+0.003
Top resid features:
against
Token against
Feature activation+0.003
Top resid features:
the
Token the
Feature activation-0.003
Top resid features:
population
Token population
Feature activation+0.001
Top resid features:
underpin
Token underpin
Feature activation-0.008
Top resid features:
ned
Tokenned
Feature activation-0.000
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
aspirations
Token aspirations
Feature activation-0.006
Top resid features:
of
Token of
Feature activation+0.005
Top resid features:
Israelis
Token Israelis
Feature activation+0.992
Top resid features:
and
Token and
Feature activation+0.012
Top resid features:
Palestinians
Token Palestinians
Feature activation+0.105
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
more
Token more
Feature activation-0.004
Top resid features:
than
Token than
Feature activation+0.000
Top resid features:
vetoed
Token vetoed
Feature activation+0.008
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
proposed
Token proposed
Feature activation-0.003
Top resid features:
sale
Token sale
Feature activation-0.105
Top resid features:
from
Token from
Feature activation+0.012
Top resid features:
Israel
Token Israel
Feature activation+0.822
Top resid features:
to
Token to
Feature activation-0.059
Top resid features:
China
Token China
Feature activation-0.101
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
an
Token an
Feature activation+0.000
Top resid features:
advanced
Token advanced
Feature activation+0.000
Top resid features:
provided
Token provided
Feature activation-0.003
Top resid features:
by
Token by
Feature activation+0.009
Top resid features:
Lockheed
Token Lockheed
Feature activation+0.064
Top resid features:
Martin
Token Martin
Feature activation+0.004
Top resid features:
to
Token to
Feature activation+0.010
Top resid features:
Israel
Token Israel
Feature activation+0.864
Top resid features:
.
Token.
Feature activation-0.020
Top resid features:
The
Token The
Feature activation-0.062
Top resid features:
very
Token very
Feature activation-0.038
Top resid features:
same
Token same
Feature activation-0.091
Top resid features:
Lockheed
Token Lockheed
Feature activation+0.039
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.046
Top resid features:
state
Token state
Feature activation-0.012
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
Israel
Token Israel
Feature activation+2.128
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
Ultimately
Token Ultimately
Feature activation-0.005
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
Mubarak
Token Mubarak
Feature activation-0.068
Top resid features:
will
Token will
Feature activation-0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.044
Top resid features:
between
Token between
Feature activation-0.000
Top resid features:
Israel
Token Israel
Feature activation+1.903
Top resid features:
and
Token and
Feature activation+0.010
Top resid features:
Japan
Token Japan
Feature activation+0.014
Top resid features:
currently
Token currently
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.004
Top resid features:
and
Token and
Feature activation+0.014
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
This
TokenThis
Feature activation+0.002
Top resid features:
is
Token is
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
second
Token second
Feature activation+0.000
Top resid features:
Israel
Token Israel
Feature activation+0.568
Top resid features:
-
Token-
Feature activation+0.006
Top resid features:
related
Tokenrelated
Feature activation+0.001
Top resid features:
scandal
Token scandal
Feature activation-0.003
Top resid features:
tied
Token tied
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.004
Top resid features:
provoke
Token provoke
Feature activation-0.003
Top resid features:
action
Token action
Feature activation-0.000
Top resid features:
by
Token by
Feature activation+0.002
Top resid features:
entering
Token entering
Feature activation+0.006
Top resid features:
into
Token into
Feature activation+0.001
Top resid features:
Israeli
Token Israeli
Feature activation+1.162
Top resid features:
waters
Token waters
Feature activation-0.002
Top resid features:
and
Token and
Feature activation+0.005
Top resid features:
creating
Token creating
Feature activation+0.001
Top resid features:
a
Token a
Feature activation-0.001
Top resid features:
situation
Token situation
Feature activation+0.002
Top resid features:
-
Token-
Feature activation+0.002
Top resid features:
controlled
Tokencontrolled
Feature activation+0.001
Top resid features:
areas
Token areas
Feature activation-0.000
Top resid features:
divided
Token divided
Feature activation-0.001
Top resid features:
by
Token by
Feature activation+0.002
Top resid features:
Israeli
Token Israeli
Feature activation+0.878
Top resid features:
checkpoints
Token checkpoints
Feature activation+0.001
Top resid features:
and
Token and
Feature activation+0.003
Top resid features:
Jewish
Token Jewish
Feature activation+0.021
Top resid features:
settlements
Token settlements
Feature activation+0.004
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
improved
Token improved
Feature activation-0.001
Top resid features:
advanced
Token advanced
Feature activation-0.005
Top resid features:
tools
Token tools
Feature activation-0.003
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Israel
Token Israel
Feature activation+1.251
Top resid features:
and
Token and
Feature activation+0.018
Top resid features:
India
Token India
Feature activation-0.102
Top resid features:
consider
Token consider
Feature activation-0.002
Top resid features:
their
Token their
Feature activation-0.002
Top resid features:
defense
Token defense
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
The
Token The
Feature activation-0.005
Top resid features:
other
Token other
Feature activation+0.002
Top resid features:
states
Token states
Feature activation-0.014
Top resid features:
were
Token were
Feature activation-0.019
Top resid features:
Israel
Token Israel
Feature activation+1.723
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
South
Token South
Feature activation-0.005
Top resid features:
Africa
Token Africa
Feature activation-0.015
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
South
Token South
Feature activation-0.005
Top resid features:
to
Token to
Feature activation+0.004
Top resid features:
terrorists
Token terrorists
Feature activation+0.053
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.004
Top resid features:
Israel
TokenIsrael
Feature activation+1.485
Top resid features:
's
Token's
Feature activation+0.012
Top resid features:
Netanyahu
Token Netanyahu
Feature activation+0.397
Top resid features:
says
Token says
Feature activation-0.001
Top resid features:
admitting
Token admitting
Feature activation-0.000
Top resid features:
Syrian
Token Syrian
Feature activation-0.205
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.012
Top resid features:
state
Token state
Feature activation-0.015
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
Israel
Token Israel
Feature activation+1.948
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
Ultimately
Token Ultimately
Feature activation-0.002
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
Mubarak
Token Mubarak
Feature activation-0.042
Top resid features:
will
Token will
Feature activation-0.001
Top resid features:

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.03

Head 2: 0.03

Head 3: 0.03

Head 4: 0.48

Head 5: 0.05

Head 6: 0.10

Head 7: 0.05

Head 8: 0.04

Head 9: 0.03

Head 10: 0.06

Head 11: 0.04

Positive logits

204392.40

SSL2.32

ESPN2.28

ULTS2.15

thumbnails2.12

YES2.09

ELY2.07

tops2.06

SQL2.04

UES2.04

IGN2.03

oen2.03

POS2.01

OS1.97

assies1.94

Reviewer1.91

arily1.89

Featured1.86

UGH1.86

Represent1.85

Negative logits

countryside-2.41

rural-2.17

の�-2.11

subsistence-2.09

vernment-2.09

discharged-2.08

stray-2.07

fleet-2.07

villages-2.05

uyomi-2.04

regiment-2.00

Muk-1.98

chuk-1.98

Mong-1.98

rongh-1.97

Ku-1.97

village-1.97

outhern-1.96

Kaiser-1.93

Haz-1.92

INTERVAL 1.822 - 2.024
CONTAINS 0.000%

Now
TokenNow
Feature activation+0.000
celebrating
Token celebrating
Feature activation+0.000
25
Token 25
Feature activation+0.000
years
Token years
Feature activation+0.000
,
Token,
Feature activation+0.000
India
Token India
Feature activation+2.014
and
Token and
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
quietly
Token quietly
Feature activation+0.000
worked
Token worked
Feature activation+0.000
to
Token to
Feature activation+0.000
upgrade
Token upgrade
Feature activation+0.000
China
Token China
Feature activation+2.024
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
military
Token military
Feature activation+0.000
arsenal
Token arsenal
Feature activation+0.000
must
Token must
Feature activation+0.000
make
Token make
Feature activation+0.000
connecting
Token connecting
Feature activation+0.000
flights
Token flights
Feature activation+0.000
in
Token in
Feature activation+0.000
Hong
Token Hong
Feature activation+1.823
Kong
Token Kong
Feature activation+1.365
or
Token or
Feature activation+0.000
Seoul
Token Seoul
Feature activation+0.289
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 1.619 - 1.822
CONTAINS 0.000%

ful
Tokenful
Feature activation+0.000
filled
Tokenfilled
Feature activation+0.000
needs
Token needs
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Indian
Token Indian
Feature activation+1.715
military
Token military
Feature activation+0.000
for
Token for
Feature activation+0.000
modern
Token modern
Feature activation+0.000
,
Token,
Feature activation+0.000
quality
Token quality
Feature activation+0.000
there
Token there
Feature activation+0.000
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
sense
Token sense
Feature activation+0.000
that
Token that
Feature activation+0.000
Bush
Token Bush
Feature activation+1.665
was
Token was
Feature activation+0.000
making
Token making
Feature activation+0.000
an
Token an
Feature activation+0.000
effort
Token effort
Feature activation+0.000
with
Token with
Feature activation+0.000

INTERVAL 1.417 - 1.619
CONTAINS 0.001%

Israel
Token Israel
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
very
Token very
Feature activation+0.000
same
Token same
Feature activation+0.000
Lockheed
Token Lockheed
Feature activation+1.419
Martin
Token Martin
Feature activation+0.318
who
Token who
Feature activation+0.000
were
Token were
Feature activation+0.000
given
Token given
Feature activation+0.000
the
Token the
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
state
Token state
Feature activation+0.046
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Trump
TokenTrump
Feature activation+1.499
âĢ
TokenâĢ
Feature activation+0.001
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
self
Token self
Feature activation+0.000
-
Token-
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
mouth
Token mouth
Feature activation+0.000
piece
Tokenpiece
Feature activation+0.000
of
Token of
Feature activation+0.000
US
Token US
Feature activation+1.509
establishment
Token establishment
Feature activation+0.000
liberalism
Token liberalism
Feature activation+0.000
,
Token,
Feature activation+0.000
weighed
Token weighed
Feature activation+0.000
in
Token in
Feature activation+0.000
-
Token-
Feature activation+0.000
related
Tokenrelated
Feature activation+0.000
scandal
Token scandal
Feature activation+0.000
tied
Token tied
Feature activation+0.000
to
Token to
Feature activation+0.000
Georgia
Token Georgia
Feature activation+1.566
Democrats
Token Democrats
Feature activation+0.062
.
Token.
Feature activation+0.000
Over
Token Over
Feature activation+0.000
the
Token the
Feature activation+0.000
weekend
Token weekend
Feature activation+0.000
proposed
Token proposed
Feature activation+0.000
sale
Token sale
Feature activation+0.000
from
Token from
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
to
Token to
Feature activation+0.000
China
Token China
Feature activation+1.485
of
Token of
Feature activation+0.000
an
Token an
Feature activation+0.000
advanced
Token advanced
Feature activation+0.000
radar
Token radar
Feature activation+0.000
system
Token system
Feature activation+0.000

INTERVAL 1.214 - 1.417
CONTAINS 0.001%

pro
Token pro
Feature activation+0.000
-
Token-
Feature activation+0.000
Israel
TokenIsrael
Feature activation+0.000
circles
Token circles
Feature activation+0.000
for
Token for
Feature activation+0.000
Hillary
Token Hillary
Feature activation+1.348
Clinton
Token Clinton
Feature activation+1.030
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
vice
Token vice
Feature activation+0.000
adopt
Token adopt
Feature activation+0.000
the
Token the
Feature activation+0.000
most
Token most
Feature activation+0.000
anti
Token anti
Feature activation+0.000
-
Token-
Feature activation+0.000
US
TokenUS
Feature activation+1.267
rhetoric
Token rhetoric
Feature activation+0.000
if
Token if
Feature activation+0.000
he
Token he
Feature activation+0.000
thought
Token thought
Feature activation+0.000
that
Token that
Feature activation+0.000
make
Token make
Feature activation+0.000
connecting
Token connecting
Feature activation+0.000
flights
Token flights
Feature activation+0.000
in
Token in
Feature activation+0.000
Hong
Token Hong
Feature activation+1.823
Kong
Token Kong
Feature activation+1.365
or
Token or
Feature activation+0.000
Seoul
Token Seoul
Feature activation+0.289
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
and
Token and
Feature activation+0.000
if
Token if
Feature activation+0.000
he
Token he
Feature activation+0.000
thinks
Token thinks
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+1.381
won
Token won
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
support
Token support
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Ambassador
Token Ambassador
Feature activation+0.000
to
Token to
Feature activation+0.000
India
Token India
Feature activation+1.299
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
Now
TokenNow
Feature activation+0.000

INTERVAL 1.012 - 1.214
CONTAINS 0.001%

people
Token people
Feature activation+0.000
of
Token of
Feature activation+0.000
color
Token color
Feature activation+0.000
here
Token here
Feature activation+0.000
in
Token in
Feature activation+0.000
Durham
Token Durham
Feature activation+1.094
in
Token in
Feature activation+0.000
mind
Token mind
Feature activation+0.000
.
Token.
Feature activation+0.000
If
Token If
Feature activation+0.000
Ferguson
Token Ferguson
Feature activation+0.527
on
Token on
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Syrian
TokenSyrian
Feature activation+1.118
state
Token state
Feature activation+0.000
TV
Token TV
Feature activation+0.000
quoted
Token quoted
Feature activation+0.000
a
Token a
Feature activation+0.000
military
Token military
Feature activation+0.000
murdered
Token murdered
Feature activation+0.000
by
Token by
Feature activation+0.000
arms
Token arms
Feature activation+0.000
provided
Token provided
Feature activation+0.000
by
Token by
Feature activation+0.000
Lockheed
Token Lockheed
Feature activation+1.020
Martin
Token Martin
Feature activation+0.321
to
Token to
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
of
Token of
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
.
Token.
Feature activation+0.000
Ultimately
Token Ultimately
Feature activation+0.000
,
Token,
Feature activation+0.000
Mubarak
Token Mubarak
Feature activation+1.017
will
Token will
Feature activation+0.000
do
Token do
Feature activation+0.000
whatever
Token whatever
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
a
Token a
Feature activation+0.000
lawmaker
Token lawmaker
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
centrist
Token centrist
Feature activation+0.000
Labor
Token Labor
Feature activation+1.072
Party
Token Party
Feature activation+0.000
.
Token.
Feature activation+0.000
She
Token She
Feature activation+0.000
said
Token said
Feature activation+0.000
if
Token if
Feature activation+0.000

INTERVAL 0.810 - 1.012
CONTAINS 0.002%

ahead
Token ahead
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
demonstrations
Token demonstrations
Feature activation+0.000
against
Token against
Feature activation+0.000
Trump
Token Trump
Feature activation+0.834
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
proclamation
Token proclamation
Feature activation+0.000
.
Token.
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
Jewish
Token Jewish
Feature activation+0.000
state
Token state
Feature activation+0.032
needs
Token needs
Feature activation+0.000
US
Token US
Feature activation+0.942
support
Token support
Feature activation+0.000
to
Token to
Feature activation+0.000
continue
Token continue
Feature activation+0.000
defending
Token defending
Feature activation+0.000
itself
Token itself
Feature activation+0.000
the
Token the
Feature activation+0.000
start
Token start
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
first
Token first
Feature activation+0.000
Iraq
Token Iraq
Feature activation+0.868
war
Token war
Feature activation+0.000
in
Token in
Feature activation+0.000
1991
Token 1991
Feature activation+0.000
the
Token the
Feature activation+0.000
general
Token general
Feature activation+0.000
morning
Token morning
Feature activation+0.000
,
Token,
Feature activation+0.000
when
Token when
Feature activation+0.000
the
Token the
Feature activation+0.000
White
Token White
Feature activation+0.552
House
Token House
Feature activation+0.928
issued
Token issued
Feature activation+0.000
a
Token a
Feature activation+0.000
letter
Token letter
Feature activation+0.000
(.
Token (.
Feature activation+0.059
pdf
Tokenpdf
Feature activation+0.000
ventures
Token ventures
Feature activation+0.000
,
Token,
Feature activation+0.000
such
Token such
Feature activation+0.000
as
Token as
Feature activation+0.000
between
Token between
Feature activation+0.000
Pun
Token Pun
Feature activation+0.958
j
Tokenj
Feature activation+0.136
Lloyd
Token Lloyd
Feature activation+0.000
and
Token and
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
Weapons
Token Weapons
Feature activation+0.000

INTERVAL 0.607 - 0.810
CONTAINS 0.003%

emer
Tokenemer
Feature activation+0.000
ged
Tokenged
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Clinton
Token Clinton
Feature activation+1.007
administration
Token administration
Feature activation+0.767
and
Token and
Feature activation+0.000
engineered
Token engineered
Feature activation+0.000
the
Token the
Feature activation+0.000
breakup
Token breakup
Feature activation+0.000
of
Token of
Feature activation+0.000
endure
Token endure
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
UN
Token UN
Feature activation+0.719
H
TokenH
Feature activation+0.000
RC
TokenRC
Feature activation+0.000
backed
Token backed
Feature activation+0.000
a
Token a
Feature activation+0.000
resolution
Token resolution
Feature activation+0.001
I
TokenI
Feature activation+0.000
expressed
Token expressed
Feature activation+0.000
the
Token the
Feature activation+0.000
fear
Token fear
Feature activation+0.000
after
Token after
Feature activation+0.000
Sanders
Token Sanders
Feature activation+0.688
won
Token won
Feature activation+0.000
Wisconsin
Token Wisconsin
Feature activation+0.118
that
Token that
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.401
might
Token might
Feature activation+0.000
generate
Token generate
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
Obama
Token Obama
Feature activation+0.642
administration
Token administration
Feature activation+0.638
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
nuclear
Token nuclear
Feature activation+0.182
deal
Token deal
Feature activation+0.000
and
Token and
Feature activation+0.000
engineered
Token engineered
Feature activation+0.000
the
Token the
Feature activation+0.000
breakup
Token breakup
Feature activation+0.000
of
Token of
Feature activation+0.000
Yugoslavia
Token Yugoslavia
Feature activation+0.757
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
war
Token war
Feature activation+0.000
against
Token against
Feature activation+0.000
Serbia
Token Serbia
Feature activation+0.964

INTERVAL 0.405 - 0.607
CONTAINS 0.006%

detention
Token detention
Feature activation+0.000
facilities
Token facilities
Feature activation+0.000
.
Token.
Feature activation+0.000
Y
Token Y
Feature activation+0.000
ish
Tokenish
Feature activation+0.000
ai
Tokenai
Feature activation+0.452
said
Token said
Feature activation+0.000
that
Token that
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
is
Token is
Feature activation+0.000
willing
Token willing
Feature activation+0.000
planning
Token planning
Feature activation+0.000
on
Token on
Feature activation+0.000
trans
Token trans
Feature activation+0.000
iting
Tokeniting
Feature activation+0.000
through
Token through
Feature activation+0.000
international
Token international
Feature activation+0.518
waters
Token waters
Feature activation+0.000
to
Token to
Feature activation+0.000
Gaza
Token Gaza
Feature activation+0.000
's
Token's
Feature activation+0.000
territorial
Token territorial
Feature activation+0.000
stockp
Token stockp
Feature activation+0.000
iles
Tokeniles
Feature activation+0.000
belonging
Token belonging
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Assad
Token Assad
Feature activation+0.442
regime
Token regime
Feature activation+0.352
,
Token,
Feature activation+0.000
such
Token such
Feature activation+0.000
a
Token a
Feature activation+0.000
development
Token development
Feature activation+0.000
II
Token II
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
relationship
Token relationship
Feature activation+0.000
between
Token between
Feature activation+0.000
Catholicism
Token Catholicism
Feature activation+0.422
and
Token and
Feature activation+0.000
what
Token what
Feature activation+0.000
Pope
Token Pope
Feature activation+0.000
John
Token John
Feature activation+0.000
Paul
Token Paul
Feature activation+0.000
ter
Tokenter
Feature activation+0.000
dam
Tokendam
Feature activation+0.027
has
Token has
Feature activation+0.000
requested
Token requested
Feature activation+0.000
that
Token that
Feature activation+0.000
Turkish
Token Turkish
Feature activation+0.441
organizations
Token organizations
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Netherlands
Token Netherlands
Feature activation+0.353
report
Token report
Feature activation+0.000

INTERVAL 0.202 - 0.405
CONTAINS 0.010%

Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
positions
Token positions
Feature activation+0.000
treat
Token treat
Feature activation+0.000
our
Token our
Feature activation+0.000
constitutional
Token constitutional
Feature activation+0.211
law
Token law
Feature activation+0.000
as
Token as
Feature activation+0.000
if
Token if
Feature activation+0.000
it
Token it
Feature activation+0.000
were
Token were
Feature activation+0.000
voters
Token voters
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
While
TokenWhile
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.341
has
Token has
Feature activation+0.000
made
Token made
Feature activation+0.000
no
Token no
Feature activation+0.000
secret
Token secret
Feature activation+0.000
in
Token in
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
Sw
TokenSw
Feature activation+0.527
iss
Tokeniss
Feature activation+0.358
cheese
Token cheese
Feature activation+0.000
isn
Token isn
Feature activation+0.000
't
Token't
Feature activation+0.000
going
Token going
Feature activation+0.000
to
Token to
Feature activation+0.000
In
TokenIn
Feature activation+0.000
February
Token February
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.353
S
TokenS
Feature activation+0.065
.
Token.
Feature activation+0.146
State
Token State
Feature activation+0.000
Department
Token Department
Feature activation+0.000
told
Token told
Feature activation+0.000
forces
Token forces
Feature activation+0.000
moved
Token moved
Feature activation+0.000
into
Token into
Feature activation+0.000
Be
Token Be
Feature activation+0.032
it
Tokenit
Feature activation+0.000
Li
Token Li
Feature activation+0.268
q
Tokenq
Feature activation+0.000
ya
Tokenya
Feature activation+0.319
village
Token village
Feature activation+0.000
,
Token,
Feature activation+0.000
west
Token west
Feature activation+0.000

INTERVAL 0.000 - 0.202
CONTAINS 99.976%

ing
Tokening
Feature activation+0.000
URI
TokenURI
Feature activation+0.000
()
Token()
Feature activation+0.000
method
Token method
Feature activation+0.000
comes
Token comes
Feature activation+0.000
in
Token in
Feature activation+0.000
handy
Token handy
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
they
Token they
Feature activation+0.000
pay
Token pay
Feature activation+0.000
a
Token a
Feature activation+0.000
lower
Token lower
Feature activation+0.000
capital
Token capital
Feature activation+0.000
gains
Token gains
Feature activation+0.000
rate
Token rate
Feature activation+0.000
on
Token on
Feature activation+0.000
most
Token most
Feature activation+0.000
other
Token other
Feature activation+0.000
income
Token income
Feature activation+0.000
s
Tokens
Feature activation+0.000
surface
Token surface
Feature activation+0.000
,
Token,
Feature activation+0.000
becoming
Token becoming
Feature activation+0.000
the
Token the
Feature activation+0.000
biggest
Token biggest
Feature activation+0.000
bloom
Token bloom
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
lake
Token lake
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
role
Token role
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
recent
Token recent
Feature activation+0.000
Snowden
Token Snowden
Feature activation+0.000
leaks
Token leaks
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
not
Token not
Feature activation+0.000
Washington
Token Washington
Feature activation+0.000
Post
Token Post
Feature activation+0.000
as
Token as
Feature activation+0.000
Cl
Token Cl
Feature activation+0.000
othes
Tokenothes
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Man
Token Man
Feature activation+0.000
by
Token by
Feature activation+0.000
Sydney
Token Sydney
Feature activation+0.000
D
Token D
Feature activation+0.000
.
Token.
Feature activation+0.000
Barney
Token Barney
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 H3.4: (feature 24359)

TOP ACTIVATIONS
MAX = 4.116

Trump
Token Trump
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
Obama
Token Obama
Feature activation+4.116
administration
Token administration
Feature activation+0.050
sp
Token sp
Feature activation+0.000
ied
Tokenied
Feature activation+0.000
on
Token on
Feature activation+0.000
and
Token and
Feature activation+0.000
General
TokenGeneral
Feature activation+0.704
Michael
Token Michael
Feature activation+0.300
Flynn
Token Flynn
Feature activation+0.000
resigned
Token resigned
Feature activation+0.000
after
Token after
Feature activation+0.000
Obama
Token Obama
Feature activation+3.588
officials
Token officials
Feature activation+0.000
leaked
Token leaked
Feature activation+0.000
his
Token his
Feature activation+0.000
discussions
Token discussions
Feature activation+0.049
with
Token with
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
President
TokenPresident
Feature activation+0.766
Obama
Token Obama
Feature activation+3.500
is
Token is
Feature activation+0.000
heart
Token heart
Feature activation+0.000
ened
Tokenened
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
ther
Tokenther
Feature activation+0.000
"
Token"
Feature activation+0.014
for
Token for
Feature activation+0.000
much
Token much
Feature activation+0.000
of
Token of
Feature activation+0.000
2011
Token 2011
Feature activation+2.935
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
continued
Token continued
Feature activation+0.000
to
Token to
Feature activation+0.000
perpetuate
Token perpetuate
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
Ċ
TokenĊ
Feature activation+0.000
Obama
TokenObama
Feature activation+2.889
and
Token and
Feature activation+0.000
his
Token his
Feature activation+0.000
team
Token team
Feature activation+0.000
are
Token are
Feature activation+0.000
not
Token not
Feature activation+0.000
George
Token George
Feature activation+0.398
W
Token W
Feature activation+0.109
.
Token.
Feature activation+0.000
Bush
Token Bush
Feature activation+0.770
in
Token in
Feature activation+0.000
2008
Token 2008
Feature activation+2.836
âĢĶ
Token âĢĶ
Feature activation+0.000
after
Token after
Feature activation+0.000
W
Token W
Feature activation+0.000
had
Token had
Feature activation+0.000
led
Token led
Feature activation+0.000
Part
TokenPart
Feature activation+0.000
of
Token of
Feature activation+0.000
it
Token it
Feature activation+0.000
,"
Token,"
Feature activation+0.000
said
Token said
Feature activation+0.079
Obama
Token Obama
Feature activation+2.803
,
Token,
Feature activation+0.000
"
Token "
Feature activation+0.000
is
Tokenis
Feature activation+0.000
Fox
Token Fox
Feature activation+0.196
News
Token News
Feature activation+0.492
What
TokenWhat
Feature activation+0.000
IL
TokenIL
Feature activation+0.000
earned
Tokenearned
Feature activation+0.000
Today
TokenToday
Feature activation+0.000
unlike
Token unlike
Feature activation+0.000
Obama
Token Obama
Feature activation+2.720
President
Token President
Feature activation+0.339
Trump
Token Trump
Feature activation+0.383
is
Token is
Feature activation+0.000
being
Token being
Feature activation+0.000
treated
Token treated
Feature activation+0.000
a
Token a
Feature activation+0.000
citizen
Token citizen
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Obama
TokenObama
Feature activation+2.579
has
Token has
Feature activation+0.000
released
Token released
Feature activation+0.000
his
Token his
Feature activation+0.000
so
Token so
Feature activation+0.000
-
Token-
Feature activation+0.000
considered
Token considered
Feature activation+0.000
a
Token a
Feature activation+0.000
presidential
Token presidential
Feature activation+0.000
campaign
Token campaign
Feature activation+0.196
against
Token against
Feature activation+0.000
Obama
Token Obama
Feature activation+2.566
in
Token in
Feature activation+0.000
2011
Token 2011
Feature activation+1.776
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
gained
Token gained
Feature activation+0.000
and
Token and
Feature activation+0.000
money
Token money
Feature activation+0.000
laundering
Token laundering
Feature activation+0.000
before
Token before
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+2.528
administration
Token administration
Feature activation+0.008
approved
Token approved
Feature activation+0.000
a
Token a
Feature activation+0.000
controversial
Token controversial
Feature activation+0.000
deal
Token deal
Feature activation+0.000
that
Token that
Feature activation+0.000
he
Token he
Feature activation+0.000
would
Token would
Feature activation+0.000
rescind
Token rescind
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+2.465
-
Token-
Feature activation+0.000
era
Tokenera
Feature activation+0.000
program
Token program
Feature activation+0.000
,
Token,
Feature activation+0.000
giving
Token giving
Feature activation+0.000
a
Token a
Feature activation+0.000
shot
Token shot
Feature activation+0.000
at
Token at
Feature activation+0.000
President
Token President
Feature activation+0.043
Barack
Token Barack
Feature activation+1.462
Obama
Token Obama
Feature activation+2.463
on
Token on
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
for
Token for
Feature activation+0.000
not
Token not
Feature activation+0.000
attending
Token attending
Feature activation+0.000
Democrats
Token Democrats
Feature activation+0.417
have
Token have
Feature activation+0.000
,
Token,
Feature activation+0.000
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.000
Obama
Token Obama
Feature activation+2.445
,
Token,
Feature activation+0.000
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
that
Token that
Feature activation+0.000
they
Token they
Feature activation+0.000
resolution
Token resolution
Feature activation+0.000
disappro
Token disappro
Feature activation+0.000
ving
Tokenving
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+2.440
administration
Token administration
Feature activation+0.000
's
Token's
Feature activation+0.000
decision
Token decision
Feature activation+0.000
to
Token to
Feature activation+0.000
allow
Token allow
Feature activation+0.000
conspiracy
Token conspiracy
Feature activation+0.000
theory
Token theory
Feature activation+0.000
,
Token,
Feature activation+0.000
saying
Token saying
Feature activation+0.000
that
Token that
Feature activation+0.000
Obama
Token Obama
Feature activation+2.411
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Harvard
Token Harvard
Feature activation+0.130
transcripts
Token transcripts
Feature activation+0.000
candidate
Token candidate
Feature activation+0.433
whether
Token whether
Feature activation+0.000
he
Token he
Feature activation+0.000
now
Token now
Feature activation+0.000
believed
Token believed
Feature activation+0.000
Obama
Token Obama
Feature activation+2.306
was
Token was
Feature activation+0.000
born
Token born
Feature activation+0.000
in
Token in
Feature activation+0.000
Hawaii
Token Hawaii
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
signature
Token signature
Feature activation+0.000
achievements
Token achievements
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+2.257
administration
Token administration
Feature activation+0.000
seem
Token seem
Feature activation+0.000
very
Token very
Feature activation+0.000
much
Token much
Feature activation+0.000
alive
Token alive
Feature activation+0.000
by
Token by
Feature activation+0.000
House
Token House
Feature activation+0.099
Republicans
Token Republicans
Feature activation+0.000
against
Token against
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+2.227
administration
Token administration
Feature activation+0.000
.
Token.
Feature activation+0.000
House
Token House
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
won
Token won
Feature activation+0.000
in
Token in
Feature activation+0.000
related
Token related
Feature activation+0.000
false
Token false
Feature activation+0.000
claims
Token claims
Feature activation+0.000
that
Token that
Feature activation+0.000
Obama
Token Obama
Feature activation+2.208
was
Token was
Feature activation+0.000
born
Token born
Feature activation+0.000
outside
Token outside
Feature activation+0.000
the
Token the
Feature activation+0.000
United
Token United
Feature activation+0.000

Top DFA by src position
MAX = 4.404

<|endoftext|>
Token<|endoftext|>
Feature activation-0.055
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
President
Token President
Feature activation-0.015
Top resid features:
Trump
Token Trump
Feature activation+4.404
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.021
Top resid features:
Ċ
TokenĊ
Feature activation-0.034
Top resid features:
The
TokenThe
Feature activation+0.029
Top resid features:
Obama
Token Obama
Feature activation-0.148
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.067
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
President
Token President
Feature activation-0.003
Top resid features:
Trump
Token Trump
Feature activation+2.462
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
The
TokenThe
Feature activation+0.001
Top resid features:
Obama
Token Obama
Feature activation-0.055
Top resid features:
and
Token and
Feature activation-0.000
Top resid features:
it
Token it
Feature activation-0.000
Top resid features:
did
Token did
Feature activation-0.001
Top resid features:
not
Token not
Feature activation-0.001
Top resid features:
attack
Token attack
Feature activation+0.004
Top resid features:
Trump
Token Trump
Feature activation+3.802
Top resid features:
directly
Token directly
Feature activation-0.000
Top resid features:
.
Token.
Feature activation-0.003
Top resid features:
Rather
Token Rather
Feature activation-0.000
Top resid features:
it
Token it
Feature activation-0.002
Top resid features:
appealed
Token appealed
Feature activation-0.003
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.002
Top resid features:
and
Token and
Feature activation+0.002
Top resid features:
yet
Token yet
Feature activation+0.002
Top resid features:
it
Token it
Feature activation-0.000
Top resid features:
seems
Token seems
Feature activation+0.000
Top resid features:
Trump
Token Trump
Feature activation+2.820
Top resid features:
âĢ
TokenâĢ
Feature activation-0.001
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
embrace
Token embrace
Feature activation+0.003
Top resid features:
of
Token of
Feature activation+0.009
Top resid features:
and
Token and
Feature activation-0.000
Top resid features:
it
Token it
Feature activation-0.001
Top resid features:
did
Token did
Feature activation-0.000
Top resid features:
not
Token not
Feature activation-0.000
Top resid features:
attack
Token attack
Feature activation+0.003
Top resid features:
Trump
Token Trump
Feature activation+3.418
Top resid features:
directly
Token directly
Feature activation+0.002
Top resid features:
.
Token.
Feature activation-0.001
Top resid features:
Rather
Token Rather
Feature activation+0.000
Top resid features:
it
Token it
Feature activation-0.001
Top resid features:
appealed
Token appealed
Feature activation-0.000
Top resid features:
generally
Token generally
Feature activation-0.001
Top resid features:
mad
Token mad
Feature activation-0.000
Top resid features:
AF
Token AF
Feature activation-0.001
Top resid features:
at
Token at
Feature activation+0.003
Top resid features:
President
Token President
Feature activation+0.008
Top resid features:
Trump
Token Trump
Feature activation+2.607
Top resid features:
and
Token and
Feature activation+0.003
Top resid features:
ready
Token ready
Feature activation-0.001
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
cl
Token cl
Feature activation-0.000
Top resid features:
ap
Tokenap
Feature activation+0.000
Top resid features:
that
Token that
Feature activation+0.001
Top resid features:
supposedly
Token supposedly
Feature activation+0.002
Top resid features:
accounted
Token accounted
Feature activation-0.001
Top resid features:
for
Token for
Feature activation+0.001
Top resid features:
Donald
Token Donald
Feature activation+0.586
Top resid features:
Trump
Token Trump
Feature activation+2.181
Top resid features:
's
Token's
Feature activation+0.014
Top resid features:
victory
Token victory
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.006
Top resid features:
Absolutely
Token Absolutely
Feature activation-0.005
Top resid features:
not
Token not
Feature activation-0.001
Top resid features:
Forbidden
Token Forbidden
Feature activation-0.004
Top resid features:
City
Token City
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
President
TokenPresident
Feature activation-0.008
Top resid features:
Trump
Token Trump
Feature activation+1.315
Top resid features:
arrived
Token arrived
Feature activation-0.001
Top resid features:
in
Token in
Feature activation+0.003
Top resid features:
China
Token China
Feature activation-0.010
Top resid features:
on
Token on
Feature activation+0.002
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
presidential
Token presidential
Feature activation+0.003
Top resid features:
bid
Token bid
Feature activation+0.007
Top resid features:
last
Token last
Feature activation+0.001
Top resid features:
year
Token year
Feature activation-0.000
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
Trump
Token Trump
Feature activation+2.740
Top resid features:
himself
Token himself
Feature activation+0.005
Top resid features:
traff
Token traff
Feature activation-0.002
Top resid features:
icked
Tokenicked
Feature activation-0.000
Top resid features:
in
Token in
Feature activation+0.003
Top resid features:
related
Token related
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
What
TokenWhat
Feature activation+0.002
Top resid features:
was
Token was
Feature activation-0.001
Top resid features:
Trump
Token Trump
Feature activation+2.027
Top resid features:
âĢ
TokenâĢ
Feature activation-0.005
Top resid features:
Ļ
TokenĻ
Feature activation+0.001
Top resid features:
s
Tokens
Feature activation+0.001
Top resid features:
role
Token role
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.008
Top resid features:
!
Token!
Feature activation-0.002
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.000
Top resid features:
Donald
Token Donald
Feature activation+0.689
Top resid features:
J
Token J
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Trump
Token Trump
Feature activation+1.876
Top resid features:
(@
Token (@
Feature activation+0.003
Top resid features:
realDonaldTrump
TokenrealDonaldTrump
Feature activation+0.093
Top resid features:
)
Token)
Feature activation+0.002
Top resid features:
October
Token October
Feature activation-0.000
Top resid features:
19
Token 19
Feature activation+0.001
Top resid features:
reelection
Token reelection
Feature activation-0.006
Top resid features:
.
Token.
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
Ċ
TokenĊ
Feature activation-0.004
Top resid features:
President
TokenPresident
Feature activation-0.001
Top resid features:
Trump
Token Trump
Feature activation+2.731
Top resid features:
announced
Token announced
Feature activation+0.001
Top resid features:
in
Token in
Feature activation+0.012
Top resid features:
September
Token September
Feature activation-0.001
Top resid features:
that
Token that
Feature activation+0.010
Top resid features:
he
Token he
Feature activation-0.000
Top resid features:
,
Token,
Feature activation-0.000
Top resid features:
stations
Token stations
Feature activation-0.001
Top resid features:
and
Token and
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.045
Top resid features:
Donald
TokenDonald
Feature activation+0.405
Top resid features:
Trump
Token Trump
Feature activation+2.007
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
REUTERS
Token REUTERS
Feature activation+0.000
Top resid features:
/
Token/
Feature activation-0.002
Top resid features:
Brian
TokenBrian
Feature activation-0.003
Top resid features:
Snyder
Token Snyder
Feature activation-0.009
Top resid features:
that
Token that
Feature activation+0.001
Top resid features:
supposedly
Token supposedly
Feature activation+0.003
Top resid features:
accounted
Token accounted
Feature activation-0.001
Top resid features:
for
Token for
Feature activation+0.001
Top resid features:
Donald
Token Donald
Feature activation+0.492
Top resid features:
Trump
Token Trump
Feature activation+1.949
Top resid features:
's
Token's
Feature activation+0.011
Top resid features:
victory
Token victory
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.003
Top resid features:
Absolutely
Token Absolutely
Feature activation-0.001
Top resid features:
not
Token not
Feature activation+0.001
Top resid features:
GOP
Token GOP
Feature activation+0.025
Top resid features:
concerns
Token concerns
Feature activation-0.000
Top resid features:
about
Token about
Feature activation+0.001
Top resid features:
Mr
Token Mr
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Trump
Token Trump
Feature activation+1.910
Top resid features:
on
Token on
Feature activation+0.005
Top resid features:
foreign
Token foreign
Feature activation-0.001
Top resid features:
policy
Token policy
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
which
Token which
Feature activation-0.000
Top resid features:
presidential
Token presidential
Feature activation+0.004
Top resid features:
bid
Token bid
Feature activation+0.008
Top resid features:
last
Token last
Feature activation+0.000
Top resid features:
year
Token year
Feature activation-0.000
Top resid features:
,
Token,
Feature activation+0.004
Top resid features:
Trump
Token Trump
Feature activation+2.438
Top resid features:
himself
Token himself
Feature activation+0.006
Top resid features:
traff
Token traff
Feature activation-0.001
Top resid features:
icked
Tokenicked
Feature activation-0.000
Top resid features:
in
Token in
Feature activation+0.004
Top resid features:
related
Token related
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.003
Top resid features:
In
Token In
Feature activation+0.002
Top resid features:
an
Token an
Feature activation+0.005
Top resid features:
interview
Token interview
Feature activation+0.001
Top resid features:
with
Token with
Feature activation+0.002
Top resid features:
Trump
Token Trump
Feature activation+1.375
Top resid features:
on
Token on
Feature activation+0.004
Top resid features:
Wednesday
Token Wednesday
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
Costa
Token Costa
Feature activation-0.004
Top resid features:
asked
Token asked
Feature activation-0.004
Top resid features:
pact
Token pact
Feature activation-0.010
Top resid features:
.
Token.
Feature activation-0.001
Top resid features:
In
Token In
Feature activation+0.001
Top resid features:
June
Token June
Feature activation-0.002
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
Trump
Token Trump
Feature activation+2.524
Top resid features:
flew
Token flew
Feature activation-0.001
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
Miami
Token Miami
Feature activation+0.000
Top resid features:
for
Token for
Feature activation+0.001
Top resid features:
a
Token a
Feature activation+0.003
Top resid features:
answer
Token answer
Feature activation+0.000
Top resid features:
from
Token from
Feature activation+0.001
Top resid features:
Congress
Token Congress
Feature activation+0.012
Top resid features:
and
Token and
Feature activation-0.001
Top resid features:
President
Token President
Feature activation-0.001
Top resid features:
Trump
Token Trump
Feature activation+2.488
Top resid features:
's
Token's
Feature activation+0.008
Top resid features:
White
Token White
Feature activation+0.001
Top resid features:
House
Token House
Feature activation-0.002
Top resid features:
on
Token on
Feature activation+0.006
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
presidential
Token presidential
Feature activation+0.004
Top resid features:
bid
Token bid
Feature activation+0.011
Top resid features:
last
Token last
Feature activation+0.001
Top resid features:
year
Token year
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
Trump
Token Trump
Feature activation+2.376
Top resid features:
himself
Token himself
Feature activation+0.015
Top resid features:
traff
Token traff
Feature activation-0.000
Top resid features:
icked
Tokenicked
Feature activation-0.009
Top resid features:
in
Token in
Feature activation+0.009
Top resid features:
related
Token related
Feature activation-0.003
Top resid features:

Decoder Weights Distribution

Head 0: 0.11

Head 1: 0.04

Head 2: 0.05

Head 3: 0.03

Head 4: 0.45

Head 5: 0.06

Head 6: 0.06

Head 7: 0.04

Head 8: 0.03

Head 9: 0.03

Head 10: 0.07

Head 11: 0.04

Positive logits

Assy1.84

Schumer1.63

Pound1.59

ohl1.55

Morty1.55

actionDate1.54

Invalid1.50

���1.47

Nikki1.47

Kimmel1.47

oooooooo1.47

Alicia1.46

Oval1.45

acus1.45

Canter1.44

Nickel1.44

Tune1.44

Degree1.43

nces1.41

staking1.41

Negative logits

odder-2.13

ococ-1.89

rongh-1.87

arty-1.80

raft-1.78

elight-1.71

RR-1.67

isky-1.66

dark-1.63

retrospect-1.62

GW-1.62

eworthy-1.62

resy-1.62

haar-1.61

oir-1.57

Krug-1.56

gui-1.56

ispers-1.55

campaign-1.54

achine-1.54

INTERVAL 3.705 - 4.116
CONTAINS 0.000%

Trump
Token Trump
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
Obama
Token Obama
Feature activation+4.116
administration
Token administration
Feature activation+0.050
sp
Token sp
Feature activation+0.000
ied
Tokenied
Feature activation+0.000
on
Token on
Feature activation+0.000
and
Token and
Feature activation+0.000

INTERVAL 3.293 - 3.705
CONTAINS 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
President
TokenPresident
Feature activation+0.766
Obama
Token Obama
Feature activation+3.500
is
Token is
Feature activation+0.000
heart
Token heart
Feature activation+0.000
ened
Tokenened
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
General
TokenGeneral
Feature activation+0.704
Michael
Token Michael
Feature activation+0.300
Flynn
Token Flynn
Feature activation+0.000
resigned
Token resigned
Feature activation+0.000
after
Token after
Feature activation+0.000
Obama
Token Obama
Feature activation+3.588
officials
Token officials
Feature activation+0.000
leaked
Token leaked
Feature activation+0.000
his
Token his
Feature activation+0.000
discussions
Token discussions
Feature activation+0.049
with
Token with
Feature activation+0.000

INTERVAL 2.881 - 3.293
CONTAINS 0.000%

ther
Tokenther
Feature activation+0.000
"
Token"
Feature activation+0.014
for
Token for
Feature activation+0.000
much
Token much
Feature activation+0.000
of
Token of
Feature activation+0.000
2011
Token 2011
Feature activation+2.935
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
continued
Token continued
Feature activation+0.000
to
Token to
Feature activation+0.000
perpetuate
Token perpetuate
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
Ċ
TokenĊ
Feature activation+0.000
Obama
TokenObama
Feature activation+2.889
and
Token and
Feature activation+0.000
his
Token his
Feature activation+0.000
team
Token team
Feature activation+0.000
are
Token are
Feature activation+0.000
not
Token not
Feature activation+0.000

INTERVAL 2.470 - 2.881
CONTAINS 0.001%

What
TokenWhat
Feature activation+0.000
IL
TokenIL
Feature activation+0.000
earned
Tokenearned
Feature activation+0.000
Today
TokenToday
Feature activation+0.000
unlike
Token unlike
Feature activation+0.000
Obama
Token Obama
Feature activation+2.720
President
Token President
Feature activation+0.339
Trump
Token Trump
Feature activation+0.383
is
Token is
Feature activation+0.000
being
Token being
Feature activation+0.000
treated
Token treated
Feature activation+0.000
and
Token and
Feature activation+0.000
money
Token money
Feature activation+0.000
laundering
Token laundering
Feature activation+0.000
before
Token before
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+2.528
administration
Token administration
Feature activation+0.008
approved
Token approved
Feature activation+0.000
a
Token a
Feature activation+0.000
controversial
Token controversial
Feature activation+0.000
deal
Token deal
Feature activation+0.000
George
Token George
Feature activation+0.398
W
Token W
Feature activation+0.109
.
Token.
Feature activation+0.000
Bush
Token Bush
Feature activation+0.770
in
Token in
Feature activation+0.000
2008
Token 2008
Feature activation+2.836
âĢĶ
Token âĢĶ
Feature activation+0.000
after
Token after
Feature activation+0.000
W
Token W
Feature activation+0.000
had
Token had
Feature activation+0.000
led
Token led
Feature activation+0.000
considered
Token considered
Feature activation+0.000
a
Token a
Feature activation+0.000
presidential
Token presidential
Feature activation+0.000
campaign
Token campaign
Feature activation+0.196
against
Token against
Feature activation+0.000
Obama
Token Obama
Feature activation+2.566
in
Token in
Feature activation+0.000
2011
Token 2011
Feature activation+1.776
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
gained
Token gained
Feature activation+0.000
Part
TokenPart
Feature activation+0.000
of
Token of
Feature activation+0.000
it
Token it
Feature activation+0.000
,"
Token,"
Feature activation+0.000
said
Token said
Feature activation+0.079
Obama
Token Obama
Feature activation+2.803
,
Token,
Feature activation+0.000
"
Token "
Feature activation+0.000
is
Tokenis
Feature activation+0.000
Fox
Token Fox
Feature activation+0.196
News
Token News
Feature activation+0.492

INTERVAL 2.058 - 2.470
CONTAINS 0.001%

Democrats
Token Democrats
Feature activation+0.417
have
Token have
Feature activation+0.000
,
Token,
Feature activation+0.000
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.000
Obama
Token Obama
Feature activation+2.445
,
Token,
Feature activation+0.000
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
that
Token that
Feature activation+0.000
they
Token they
Feature activation+0.000
from
Token from
Feature activation+0.000
Canadian
Token Canadian
Feature activation+0.915
Prime
Token Prime
Feature activation+0.554
Minister
Token Minister
Feature activation+0.854
Justin
Token Justin
Feature activation+2.128
Trudeau
Token Trudeau
Feature activation+2.072
and
Token and
Feature activation+0.000
Mexican
Token Mexican
Feature activation+0.000
President
Token President
Feature activation+0.000
Enrique
Token Enrique
Feature activation+0.000
Pe
Token Pe
Feature activation+0.000
the
Token the
Feature activation+0.000
signature
Token signature
Feature activation+0.000
achievements
Token achievements
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+2.257
administration
Token administration
Feature activation+0.000
seem
Token seem
Feature activation+0.000
very
Token very
Feature activation+0.000
much
Token much
Feature activation+0.000
alive
Token alive
Feature activation+0.000
that
Token that
Feature activation+0.000
he
Token he
Feature activation+0.000
would
Token would
Feature activation+0.000
rescind
Token rescind
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+2.465
-
Token-
Feature activation+0.000
era
Tokenera
Feature activation+0.000
program
Token program
Feature activation+0.000
,
Token,
Feature activation+0.000
giving
Token giving
Feature activation+0.000
a
Token a
Feature activation+0.000
shot
Token shot
Feature activation+0.000
at
Token at
Feature activation+0.000
President
Token President
Feature activation+0.043
Barack
Token Barack
Feature activation+1.462
Obama
Token Obama
Feature activation+2.463
on
Token on
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
for
Token for
Feature activation+0.000
not
Token not
Feature activation+0.000
attending
Token attending
Feature activation+0.000

INTERVAL 1.647 - 2.058
CONTAINS 0.001%

Ċ
TokenĊ
Feature activation+0.000
Asked
TokenAsked
Feature activation+0.000
whether
Token whether
Feature activation+0.000
he
Token he
Feature activation+0.122
believes
Token believes
Feature activation+0.000
Obama
Token Obama
Feature activation+1.770
illegally
Token illegally
Feature activation+0.000
surve
Token surve
Feature activation+0.000
illed
Tokenilled
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
Tower
Token Tower
Feature activation+0.000
-
Token-
Feature activation+0.000
profile
Tokenprofile
Feature activation+0.000
announcement
Token announcement
Feature activation+0.000
rolling
Token rolling
Feature activation+0.000
back
Token back
Feature activation+0.000
Obama
Token Obama
Feature activation+1.900
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
historic
Token historic
Feature activation+0.000
opening
Token opening
Feature activation+0.000
be
Token be
Feature activation+0.000
true
Token true
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Bill
TokenBill
Feature activation+1.811
Clinton
Token Clinton
Feature activation+0.172
and
Token and
Feature activation+0.000
Juan
Token Juan
Feature activation+0.000
ita
Tokenita
Feature activation+0.000
Bro
Token Bro
Feature activation+0.000
result
Token result
Feature activation+0.222
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Mrs
TokenMrs
Feature activation+0.000
Obama
Token Obama
Feature activation+1.976
's
Token's
Feature activation+0.000
remarks
Token remarks
Feature activation+0.000
come
Token come
Feature activation+0.000
within
Token within
Feature activation+0.000
hours
Token hours
Feature activation+0.000
presidential
Token presidential
Feature activation+0.000
campaign
Token campaign
Feature activation+0.196
against
Token against
Feature activation+0.000
Obama
Token Obama
Feature activation+2.566
in
Token in
Feature activation+0.000
2011
Token 2011
Feature activation+1.776
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
gained
Token gained
Feature activation+0.000
media
Token media
Feature activation+0.000
attention
Token attention
Feature activation+0.000

INTERVAL 1.235 - 1.647
CONTAINS 0.003%

and
Token and
Feature activation+0.000
replace
Token replace
Feature activation+0.000
former
Token former
Feature activation+0.000
Democratic
Token Democratic
Feature activation+0.000
President
Token President
Feature activation+0.000
Barack
Token Barack
Feature activation+1.431
Obama
Token Obama
Feature activation+1.791
's
Token's
Feature activation+0.000
signature
Token signature
Feature activation+0.000
domestic
Token domestic
Feature activation+0.000
policy
Token policy
Feature activation+0.000
,
Token,
Feature activation+0.000
had
Token had
Feature activation+0.000
apologised
Token apologised
Feature activation+0.000
to
Token to
Feature activation+0.000
Ms
Token Ms
Feature activation+0.000
Mal
Token Mal
Feature activation+1.397
hot
Tokenhot
Feature activation+0.000
ra
Tokenra
Feature activation+0.000
's
Token's
Feature activation+0.000
team
Token team
Feature activation+0.000
and
Token and
Feature activation+0.000
answer
Token answer
Feature activation+0.000
until
Token until
Feature activation+0.000
he
Token he
Feature activation+0.000
met
Token met
Feature activation+0.000
with
Token with
Feature activation+0.000
Netanyahu
Token Netanyahu
Feature activation+1.280
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
planned
Token planned
Feature activation+0.000
trip
Token trip
Feature activation+0.000
to
Token to
Feature activation+0.000
do
Token do
Feature activation+0.000
a
Token a
Feature activation+0.000
special
Token special
Feature activation+0.000
dance
Token dance
Feature activation+0.000
with
Token with
Feature activation+0.000
Bennett
Token Bennett
Feature activation+1.331
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
character
Token character
Feature activation+0.000
before
Token before
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
and
Token and
Feature activation+0.000
yet
Token yet
Feature activation+0.000
it
Token it
Feature activation+0.000
seems
Token seems
Feature activation+0.000
Trump
Token Trump
Feature activation+1.397
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
embrace
Token embrace
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 0.823 - 1.235
CONTAINS 0.008%

Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
former
Token former
Feature activation+0.000
reality
Token reality
Feature activation+0.000
TV
Token TV
Feature activation+0.000
star
Token star
Feature activation+0.945
's
Token's
Feature activation+0.119
95
Token 95
Feature activation+0.000
percent
Token percent
Feature activation+0.000
pledge
Token pledge
Feature activation+0.000
was
Token was
Feature activation+0.000
0
Token0
Feature activation+0.000
Uh
TokenUh
Feature activation+0.000
D
TokenD
Feature activation+0.000
V
TokenV
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
Ali
Token Ali
Feature activation+0.990
Ab
Token Ab
Feature activation+0.762
un
Tokenun
Feature activation+0.032
im
Tokenim
Feature activation+0.205
ah
Tokenah
Feature activation+0.314
(@
Token (@
Feature activation+0.000
Mayor
Token Mayor
Feature activation+0.428
Bill
Token Bill
Feature activation+0.612
de
Token de
Feature activation+0.130
Blasio
Token Blasio
Feature activation+0.800
in
Token in
Feature activation+0.000
2017
Token 2017
Feature activation+0.883
since
Token since
Feature activation+0.000
his
Token his
Feature activation+0.178
well
Token well
Feature activation+0.000
-
Token-
Feature activation+0.000
received
Tokenreceived
Feature activation+0.000
friendly
Token friendly
Feature activation+0.000
calls
Token calls
Feature activation+0.000
from
Token from
Feature activation+0.000
Canadian
Token Canadian
Feature activation+0.915
Prime
Token Prime
Feature activation+0.554
Minister
Token Minister
Feature activation+0.854
Justin
Token Justin
Feature activation+2.128
Trudeau
Token Trudeau
Feature activation+2.072
and
Token and
Feature activation+0.000
Mexican
Token Mexican
Feature activation+0.000
President
Token President
Feature activation+0.000
did
Token did
Feature activation+0.000
not
Token not
Feature activation+0.000
turn
Token turn
Feature activation+0.000
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
2016
Token 2016
Feature activation+0.958
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
level
Token level
Feature activation+0.000
they
Token they
Feature activation+0.000
did
Token did
Feature activation+0.000

INTERVAL 0.412 - 0.823
CONTAINS 0.031%

Ċ
TokenĊ
Feature activation+0.000
Like
TokenLike
Feature activation+0.000
Sanders
Token Sanders
Feature activation+0.000
himself
Token himself
Feature activation+0.000
,
Token,
Feature activation+0.000
Bond
Token Bond
Feature activation+0.769
and
Token and
Feature activation+0.000
Ex
Token Ex
Feature activation+0.000
ley
Tokenley
Feature activation+0.000
aren
Token aren
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
minutes
Token minutes
Feature activation+0.000
of
Token of
Feature activation+0.000
coverage
Token coverage
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
Bernie
Token Bernie
Feature activation+0.418
Sanders
Token Sanders
Feature activation+0.149
campaign
Token campaign
Feature activation+0.000
has
Token has
Feature activation+0.000
received
Token received
Feature activation+0.000
a
Token a
Feature activation+0.000
According
TokenAccording
Feature activation+0.000
to
Token to
Feature activation+0.000
Media
Token Media
Feature activation+0.000
ite
Tokenite
Feature activation+0.000
,
Token,
Feature activation+0.000
Flowers
Token Flowers
Feature activation+0.719
was
Token was
Feature activation+0.121
thrust
Token thrust
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
national
Token national
Feature activation+0.000
amount
Token amount
Feature activation+0.000
of
Token of
Feature activation+0.000
time
Token time
Feature activation+0.000
that
Token that
Feature activation+0.000
Mitt
Token Mitt
Feature activation+0.635
Romney
Token Romney
Feature activation+0.792
got
Token got
Feature activation+0.000
,
Token,
Feature activation+0.000
when
Token when
Feature activation+0.000
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
More
Token More
Feature activation+0.000
Videos
Token Videos
Feature activation+0.000
...
Token ...
Feature activation+0.000
MUST
Token MUST
Feature activation+0.000
WATCH
Token WATCH
Feature activation+0.000
Booker
Token Booker
Feature activation+0.436
on
Token on
Feature activation+0.000
GOP
Token GOP
Feature activation+0.000
health
Token health
Feature activation+0.000
strategy
Token strategy
Feature activation+0.000
:
Token:
Feature activation+0.000

INTERVAL 0.000 - 0.412
CONTAINS 99.954%

District
Token District
Feature activation+0.000
of
Token of
Feature activation+0.000
Columbia
Token Columbia
Feature activation+0.000
.
Token.
Feature activation+0.000
Christ
TokenChrist
Feature activation+0.000
ie
Tokenie
Feature activation+0.000
currently
Token currently
Feature activation+0.000
sits
Token sits
Feature activation+0.000
ninth
Token ninth
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
of
Token of
Feature activation+0.000
Niger
Token Niger
Feature activation+0.000
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
just
Token just
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Ben
TokenBen
Feature activation+0.000
ghazi
Tokenghazi
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
chicks
Token chicks
Feature activation+0.000
pack
Token pack
Feature activation+0.000
a
Token a
Feature activation+0.000
bone
Token bone
Feature activation+0.000
-
Token-
Feature activation+0.000
cr
Tokencr
Feature activation+0.000
ushing
Tokenushing
Feature activation+0.000
punch
Token punch
Feature activation+0.000
!
Token!
Feature activation+0.000
Well
Token Well
Feature activation+0.000
,
Token,
Feature activation+0.000
routine
Token routine
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
cal
Token cal
Feature activation+0.000
mer
Tokenmer
Feature activation+0.000
you
Token you
Feature activation+0.000
are
Token are
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
easier
Token easier
Feature activation+0.000
it
Token it
Feature activation+0.000
information
Token information
Feature activation+0.000
security
Token security
Feature activation+0.000
demonstrates
Token demonstrates
Feature activation+0.000
one
Token one
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
most
Token most
Feature activation+0.000
per
Token per
Feature activation+0.000
n
Tokenn
Feature activation+0.000
icious
Tokenicious
Feature activation+0.000
challenges
Token challenges
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 H3.4: (feature 22609)

TOP ACTIVATIONS
MAX = 2.704

Ċ
TokenĊ
Feature activation+0.000
India
TokenIndia
Feature activation+0.000
had
Token had
Feature activation+0.000
summoned
Token summoned
Feature activation+0.000
the
Token the
Feature activation+0.000
Italian
Token Italian
Feature activation+2.704
ambassador
Token ambassador
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+0.000
to
Token to
Feature activation+0.000
protest
Token protest
Feature activation+0.000
of
Token of
Feature activation+0.000
refining
Token refining
Feature activation+0.000
capacity
Token capacity
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+2.698
.
Token.
Feature activation+0.000
Pet
TokenPet
Feature activation+0.000
rol
Tokenrol
Feature activation+0.000
price
Token price
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Supreme
Token Supreme
Feature activation+0.000
Court
Token Court
Feature activation+0.000
against
Token against
Feature activation+0.000
the
Token the
Feature activation+0.000
Italian
Token Italian
Feature activation+2.576
government
Token government
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
decision
Token decision
Feature activation+0.000
.
Token.
Feature activation+0.000
HI
TokenHI
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
As
Token As
Feature activation+0.000
expected
Token expected
Feature activation+0.000
,
Token,
Feature activation+0.000
Russia
Token Russia
Feature activation+1.939
and
Token and
Feature activation+0.000
India
Token India
Feature activation+0.000
,
Token,
Feature activation+0.000
longtime
Token longtime
Feature activation+0.000
allies
Token allies
Feature activation+0.000
of
Token of
Feature activation+0.000
Delhi
Token Delhi
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
two
Token two
Feature activation+0.000
Chinese
Token Chinese
Feature activation+1.615
G
TokenG
Feature activation+0.015
and
Tokenand
Feature activation+0.000
his
Tokenhis
Feature activation+0.000
are
Token are
Feature activation+0.000
doing
Token doing
Feature activation+0.000
(
Token (
Feature activation+0.000
the
Tokenthe
Feature activation+0.000
popular
Token popular
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.058
Italian
TokenItalian
Feature activation+1.571
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
dictator
Token dictator
Feature activation+0.309
)
Token)
Feature activation+0.000
like
Token like
Feature activation+0.000
was
Token was
Feature activation+0.000
held
Token held
Feature activation+0.000
by
Token by
Feature activation+0.000
a
Token a
Feature activation+0.000
South
Token South
Feature activation+0.193
African
Token African
Feature activation+1.524
woman
Token woman
Feature activation+0.000
who
Token who
Feature activation+0.000
ate
Token ate
Feature activation+0.000
eight
Token eight
Feature activation+0.000
j
Token j
Feature activation+0.000
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
February
Token February
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
Japanese
Token Japanese
Feature activation+1.496
woman
Token woman
Feature activation+0.000
accused
Token accused
Feature activation+0.000
a
Token a
Feature activation+0.000
tour
Token tour
Feature activation+0.000
guide
Token guide
Feature activation+0.000
strictly
Token strictly
Feature activation+0.000
conform
Token conform
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
Muss
Token Muss
Feature activation+1.296
olini
Tokenolini
Feature activation+1.438
(
Token (
Feature activation+0.000
the
Tokenthe
Feature activation+0.000
popular
Token popular
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.058
India
Token India
Feature activation+0.000
is
Token is
Feature activation+0.000
in
Token in
Feature activation+0.000
talks
Token talks
Feature activation+0.000
with
Token with
Feature activation+0.000
Canada
Token Canada
Feature activation+1.424
's
Token's
Feature activation+0.000
Te
Token Te
Feature activation+0.000
ck
Tokenck
Feature activation+0.000
Resources
Token Resources
Feature activation+0.000
Ltd
Token Ltd
Feature activation+0.000
ban
Token ban
Feature activation+0.000
on
Token on
Feature activation+0.000
gay
Token gay
Feature activation+0.225
sex
Token sex
Feature activation+0.000
and
Token and
Feature activation+0.000
Australia
Token Australia
Feature activation+1.385
's
Token's
Feature activation+0.000
high
Token high
Feature activation+0.000
court
Token court
Feature activation+0.000
has
Token has
Feature activation+0.000
struck
Token struck
Feature activation+0.000
women
Token women
Feature activation+0.000
visiting
Token visiting
Feature activation+0.000
India
Token India
Feature activation+0.000
.
Token.
Feature activation+0.000
An
Token An
Feature activation+0.000
American
Token American
Feature activation+1.357
tourist
Token tourist
Feature activation+0.000
accused
Token accused
Feature activation+0.000
two
Token two
Feature activation+0.000
men
Token men
Feature activation+0.000
of
Token of
Feature activation+0.000
to
Token to
Feature activation+0.000
strictly
Token strictly
Feature activation+0.000
conform
Token conform
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
Muss
Token Muss
Feature activation+1.296
olini
Tokenolini
Feature activation+1.438
(
Token (
Feature activation+0.000
the
Tokenthe
Feature activation+0.000
popular
Token popular
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
sources
Token sources
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
China
TokenChina
Feature activation+1.243
,
Token,
Feature activation+0.000
Japan
Token Japan
Feature activation+0.614
and
Token and
Feature activation+0.000
India
Token India
Feature activation+0.000
were
Token were
Feature activation+0.000
ji
Tokenji
Feature activation+0.000
Lloyd
Token Lloyd
Feature activation+0.000
are
Token are
Feature activation+0.000
partnering
Token partnering
Feature activation+0.000
with
Token with
Feature activation+0.000
Israeli
Token Israeli
Feature activation+1.210
Weapons
Token Weapons
Feature activation+0.000
Industries
Token Industries
Feature activation+0.000
to
Token to
Feature activation+0.000
open
Token open
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
court
Token court
Feature activation+0.000
ordered
Token ordered
Feature activation+0.000
ambassador
Token ambassador
Feature activation+0.000
Dan
Token Dan
Feature activation+1.206
ie
Tokenie
Feature activation+0.000
le
Tokenle
Feature activation+0.000
M
Token M
Feature activation+0.000
anc
Tokenanc
Feature activation+0.000
ini
Tokenini
Feature activation+0.000
Indian
Token Indian
Feature activation+0.000
,
Token,
Feature activation+0.000
because
Token because
Feature activation+0.000
I
Token I
Feature activation+0.000
am
Token am
Feature activation+0.000
Mexican
Token Mexican
Feature activation+1.202
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Says
Token Says
Feature activation+0.000
its
Token its
Feature activation+0.000
true
Token true
Feature activation+0.000
move
Token move
Feature activation+0.000
a
Token a
Feature activation+0.000
proposal
Token proposal
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
Bon
Token Bon
Feature activation+1.191
n
Tokenn
Feature activation+0.000
climate
Token climate
Feature activation+0.000
talks
Token talks
Feature activation+0.000
starting
Token starting
Feature activation+0.000
from
Token from
Feature activation+0.000
government
Token government
Feature activation+0.000
should
Token should
Feature activation+0.000
do
Token do
Feature activation+0.000
"
Token "
Feature activation+0.000
what
Tokenwhat
Feature activation+0.000
China
Token China
Feature activation+1.184
did
Token did
Feature activation+0.000
15
Token 15
Feature activation+0.000
years
Token years
Feature activation+0.000
ago
Token ago
Feature activation+0.000
and
Token and
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+0.000
to
Token to
Feature activation+0.000
protest
Token protest
Feature activation+0.000
against
Token against
Feature activation+0.000
Rome
Token Rome
Feature activation+1.182
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
decision
Token decision
Feature activation+0.000
not
Token not
Feature activation+0.000

Top DFA by src position
MAX = 3.907

iv
Tokeniv
Feature activation+0.011
Top resid features:
aram
Tokenaram
Feature activation+0.045
Top resid features:
V
Token V
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
India
TokenIndia
Feature activation+3.907
Top resid features:
had
Token had
Feature activation-0.001
Top resid features:
summoned
Token summoned
Feature activation-0.161
Top resid features:
the
Token the
Feature activation+0.030
Top resid features:
Italian
Token Italian
Feature activation-0.123
Top resid features:
ambassador
Token ambassador
Feature activation+0.000
Top resid features:
ov
Tokenov
Feature activation-0.000
Top resid features:
Hik
Token Hik
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.010
Top resid features:
NEW
TokenNEW
Feature activation+0.003
Top resid features:
DEL
Token DEL
Feature activation+0.201
Top resid features:
HI
TokenHI
Feature activation+3.152
Top resid features:
:
Token:
Feature activation+0.011
Top resid features:
Pet
Token Pet
Feature activation+0.007
Top resid features:
rol
Tokenrol
Feature activation+0.001
Top resid features:
and
Token and
Feature activation+0.010
Top resid features:
diesel
Token diesel
Feature activation+0.001
Top resid features:
iv
Tokeniv
Feature activation+0.004
Top resid features:
aram
Tokenaram
Feature activation+0.019
Top resid features:
V
Token V
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
India
TokenIndia
Feature activation+2.287
Top resid features:
had
Token had
Feature activation+0.001
Top resid features:
summoned
Token summoned
Feature activation-0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
Italian
Token Italian
Feature activation+0.017
Top resid features:
ambassador
Token ambassador
Feature activation-0.002
Top resid features:
Click
Token Click
Feature activation-0.002
Top resid features:
Here
Token Here
Feature activation-0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.010
Top resid features:
NEW
TokenNEW
Feature activation-0.007
Top resid features:
DEL
Token DEL
Feature activation+0.314
Top resid features:
HI
TokenHI
Feature activation+2.695
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.015
Top resid features:
As
Token As
Feature activation-0.094
Top resid features:
expected
Token expected
Feature activation-0.083
Top resid features:
,
Token,
Feature activation+0.054
Top resid features:
Russia
Token Russia
Feature activation-0.086
Top resid features:
is
Token is
Feature activation+0.002
Top resid features:
now
Token now
Feature activation-0.002
Top resid features:
in
Token in
Feature activation+0.005
Top resid features:
power
Token power
Feature activation+0.005
Top resid features:
in
Token in
Feature activation+0.005
Top resid features:
Delhi
Token Delhi
Feature activation+1.453
Top resid features:
;
Token;
Feature activation+0.005
Top resid features:
imagine
Token imagine
Feature activation-0.008
Top resid features:
when
Token when
Feature activation+0.002
Top resid features:
they
Token they
Feature activation+0.002
Top resid features:
âĢ
TokenâĢ
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.005
Top resid features:
fascist
Token fascist
Feature activation+0.001
Top resid features:
ideology
Token ideology
Feature activation-0.003
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
Indian
Token Indian
Feature activation+1.304
Top resid features:
National
Token National
Feature activation+0.009
Top resid features:
Congress
Token Congress
Feature activation+0.040
Top resid features:
where
Token where
Feature activation+0.001
Top resid features:
an
Token an
Feature activation+0.005
Top resid features:
average
Token average
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
Ass
Token Ass
Feature activation-0.001
Top resid features:
am
Tokenam
Feature activation+0.040
Top resid features:
region
Token region
Feature activation+0.010
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
India
Token India
Feature activation+2.436
Top resid features:
where
Token where
Feature activation+0.002
Top resid features:
Mrs
Token Mrs
Feature activation-0.001
Top resid features:
Tam
Token Tam
Feature activation-0.003
Top resid features:
uly
Tokenuly
Feature activation+0.002
Top resid features:
lives
Token lives
Feature activation+0.001
Top resid features:
reported
Token reported
Feature activation-0.006
Top resid features:
by
Token by
Feature activation+0.004
Top resid features:
foreign
Token foreign
Feature activation+0.005
Top resid features:
women
Token women
Feature activation-0.003
Top resid features:
visiting
Token visiting
Feature activation-0.004
Top resid features:
India
Token India
Feature activation+2.413
Top resid features:
.
Token.
Feature activation+0.035
Top resid features:
An
Token An
Feature activation+0.016
Top resid features:
American
Token American
Feature activation+0.009
Top resid features:
tourist
Token tourist
Feature activation-0.011
Top resid features:
accused
Token accused
Feature activation-0.020
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.119
Top resid features:
Vice
Token Vice
Feature activation+0.001
Top resid features:
President
Token President
Feature activation+0.001
Top resid features:
Rahul
Token Rahul
Feature activation+0.470
Top resid features:
Gandhi
Token Gandhi
Feature activation+0.880
Top resid features:
.
Token.
Feature activation+0.039
Top resid features:
This
Token This
Feature activation+0.001
Top resid features:
marks
Token marks
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.009
Top resid features:
official
Token official
Feature activation-0.002
Top resid features:
pta
Tokenpta
Feature activation+0.018
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
NEW
TokenNEW
Feature activation+0.005
Top resid features:
DEL
Token DEL
Feature activation+0.072
Top resid features:
HI
TokenHI
Feature activation+0.925
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
July
Token July
Feature activation-0.004
Top resid features:
8
Token 8
Feature activation-0.004
Top resid features:
(
Token (
Feature activation-0.004
Top resid features:
Reuters
TokenReuters
Feature activation+0.006
Top resid features:
island
Token island
Feature activation+0.001
Top resid features:
who
Token who
Feature activation-0.003
Top resid features:
's
Token's
Feature activation+0.004
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.014
Top resid features:
India
TokenIndia
Feature activation+2.295
Top resid features:
has
Token has
Feature activation+0.028
Top resid features:
reinstated
Token reinstated
Feature activation-0.021
Top resid features:
a
Token a
Feature activation+0.018
Top resid features:
ban
Token ban
Feature activation-0.008
Top resid features:
on
Token on
Feature activation+0.006
Top resid features:
reported
Token reported
Feature activation-0.012
Top resid features:
by
Token by
Feature activation+0.010
Top resid features:
foreign
Token foreign
Feature activation+0.036
Top resid features:
women
Token women
Feature activation-0.015
Top resid features:
visiting
Token visiting
Feature activation-0.025
Top resid features:
India
Token India
Feature activation+2.239
Top resid features:
.
Token.
Feature activation+0.056
Top resid features:
An
Token An
Feature activation-0.116
Top resid features:
American
Token American
Feature activation-0.021
Top resid features:
tourist
Token tourist
Feature activation+0.000
Top resid features:
accused
Token accused
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.010
Top resid features:
fascist
Token fascist
Feature activation+0.009
Top resid features:
ideology
Token ideology
Feature activation-0.001
Top resid features:
of
Token of
Feature activation+0.008
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
Indian
Token Indian
Feature activation+0.905
Top resid features:
National
Token National
Feature activation+0.014
Top resid features:
Congress
Token Congress
Feature activation+0.052
Top resid features:
where
Token where
Feature activation+0.004
Top resid features:
an
Token an
Feature activation+0.019
Top resid features:
average
Token average
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.012
Top resid features:
first
Token first
Feature activation-0.000
Top resid features:
formal
Token formal
Feature activation-0.002
Top resid features:
notification
Token notification
Feature activation-0.002
Top resid features:
that
Token that
Feature activation+0.002
Top resid features:
India
Token India
Feature activation+2.129
Top resid features:
had
Token had
Feature activation+0.023
Top resid features:
come
Token come
Feature activation-0.005
Top resid features:
up
Token up
Feature activation-0.001
Top resid features:
empty
Token empty
Feature activation-0.004
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
revolver
Token revolver
Feature activation-0.001
Top resid features:
designed
Token designed
Feature activation-0.001
Top resid features:
for
Token for
Feature activation+0.001
Top resid features:
sale
Token sale
Feature activation-0.001
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
Indian
Token Indian
Feature activation+1.195
Top resid features:
civilian
Token civilian
Feature activation+0.001
Top resid features:
women
Token women
Feature activation-0.006
Top resid features:
.
Token.
Feature activation+0.015
Top resid features:
Now
Token Now
Feature activation-0.008
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
iv
Tokeniv
Feature activation+0.011
Top resid features:
aram
Tokenaram
Feature activation+0.060
Top resid features:
V
Token V
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
India
TokenIndia
Feature activation+1.109
Top resid features:
had
Token had
Feature activation+0.006
Top resid features:
summoned
Token summoned
Feature activation-0.001
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
Italian
Token Italian
Feature activation+0.014
Top resid features:
ambassador
Token ambassador
Feature activation+0.004
Top resid features:
what
Token what
Feature activation+0.011
Top resid features:
it
Token it
Feature activation+0.004
Top resid features:
means
Token means
Feature activation-0.006
Top resid features:
to
Token to
Feature activation+0.016
Top resid features:
be
Token be
Feature activation-0.001
Top resid features:
Indian
Token Indian
Feature activation+0.976
Top resid features:
,
Token,
Feature activation+0.018
Top resid features:
because
Token because
Feature activation+0.011
Top resid features:
I
Token I
Feature activation-0.062
Top resid features:
am
Token am
Feature activation+0.026
Top resid features:
Mexican
Token Mexican
Feature activation-0.085
Top resid features:
hottest
Token hottest
Feature activation+0.000
Top resid features:
year
Token year
Feature activation+0.008
Top resid features:
on
Token on
Feature activation+0.003
Top resid features:
record
Token record
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.019
Top resid features:
India
Token India
Feature activation+1.667
Top resid features:
and
Token and
Feature activation+0.020
Top resid features:
China
Token China
Feature activation+0.009
Top resid features:
will
Token will
Feature activation+0.003
Top resid features:
negotiate
Token negotiate
Feature activation-0.000
Top resid features:
for
Token for
Feature activation+0.003
Top resid features:
art
Tokenart
Feature activation+0.060
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
suggested
Token suggested
Feature activation-0.005
Top resid features:
that
Token that
Feature activation-0.003
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
Indian
Token Indian
Feature activation+0.903
Top resid features:
government
Token government
Feature activation+0.001
Top resid features:
should
Token should
Feature activation-0.020
Top resid features:
do
Token do
Feature activation-0.033
Top resid features:
"
Token "
Feature activation+0.055
Top resid features:
what
Tokenwhat
Feature activation-0.032
Top resid features:
iv
Tokeniv
Feature activation+0.010
Top resid features:
aram
Tokenaram
Feature activation+0.031
Top resid features:
V
Token V
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.006
Top resid features:
India
TokenIndia
Feature activation+2.071
Top resid features:
had
Token had
Feature activation+0.012
Top resid features:
summoned
Token summoned
Feature activation-0.011
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
Italian
Token Italian
Feature activation+0.000
Top resid features:
ambassador
Token ambassador
Feature activation-0.030
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.07

Head 2: 0.05

Head 3: 0.03

Head 4: 0.45

Head 5: 0.05

Head 6: 0.05

Head 7: 0.05

Head 8: 0.05

Head 9: 0.03

Head 10: 0.07

Head 11: 0.05

Positive logits

worm2.11

versions2.07

insert2.05

worms2.04

asso2.02

ynthesis1.97

pload1.94

gged1.93

ufact1.92

roup1.92

Morty1.91

igree1.91

Reviewer1.90

tern1.87

odes1.85

anches1.83

acet1.82

iazep1.81

peanuts1.81

iments1.81

Negative logits

Nay-2.01

FI-1.90

Metropolitan-1.90

Gw-1.84

grim-1.78

condem-1.77

mourning-1.77

nec-1.75

watershed-1.75

Donetsk-1.74

Directorate-1.74

Prohibition-1.70

rebellion-1.68

police-1.68

Gloria-1.67

Rebellion-1.67

literacy-1.66

forb-1.66

attorney-1.64

RELE-1.63

INTERVAL 2.434 - 2.704
CONTAINS 0.000%

of
Token of
Feature activation+0.000
refining
Token refining
Feature activation+0.000
capacity
Token capacity
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+2.698
.
Token.
Feature activation+0.000
Pet
TokenPet
Feature activation+0.000
rol
Tokenrol
Feature activation+0.000
price
Token price
Feature activation+0.000
in
Token in
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
India
TokenIndia
Feature activation+0.000
had
Token had
Feature activation+0.000
summoned
Token summoned
Feature activation+0.000
the
Token the
Feature activation+0.000
Italian
Token Italian
Feature activation+2.704
ambassador
Token ambassador
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+0.000
to
Token to
Feature activation+0.000
protest
Token protest
Feature activation+0.000
the
Token the
Feature activation+0.000
Supreme
Token Supreme
Feature activation+0.000
Court
Token Court
Feature activation+0.000
against
Token against
Feature activation+0.000
the
Token the
Feature activation+0.000
Italian
Token Italian
Feature activation+2.576
government
Token government
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
decision
Token decision
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 2.163 - 2.434
CONTAINS 0.000%

INTERVAL 1.893 - 2.163
CONTAINS 0.000%

HI
TokenHI
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
As
Token As
Feature activation+0.000
expected
Token expected
Feature activation+0.000
,
Token,
Feature activation+0.000
Russia
Token Russia
Feature activation+1.939
and
Token and
Feature activation+0.000
India
Token India
Feature activation+0.000
,
Token,
Feature activation+0.000
longtime
Token longtime
Feature activation+0.000
allies
Token allies
Feature activation+0.000

INTERVAL 1.622 - 1.893
CONTAINS 0.000%

INTERVAL 1.352 - 1.622
CONTAINS 0.001%

(
Token (
Feature activation+0.000
the
Tokenthe
Feature activation+0.000
popular
Token popular
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.058
Italian
TokenItalian
Feature activation+1.571
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
dictator
Token dictator
Feature activation+0.309
)
Token)
Feature activation+0.000
like
Token like
Feature activation+0.000
of
Token of
Feature activation+0.000
Delhi
Token Delhi
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
two
Token two
Feature activation+0.000
Chinese
Token Chinese
Feature activation+1.615
G
TokenG
Feature activation+0.015
and
Tokenand
Feature activation+0.000
his
Tokenhis
Feature activation+0.000
are
Token are
Feature activation+0.000
doing
Token doing
Feature activation+0.000
women
Token women
Feature activation+0.000
visiting
Token visiting
Feature activation+0.000
India
Token India
Feature activation+0.000
.
Token.
Feature activation+0.000
An
Token An
Feature activation+0.000
American
Token American
Feature activation+1.357
tourist
Token tourist
Feature activation+0.000
accused
Token accused
Feature activation+0.000
two
Token two
Feature activation+0.000
men
Token men
Feature activation+0.000
of
Token of
Feature activation+0.000
strictly
Token strictly
Feature activation+0.000
conform
Token conform
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
Muss
Token Muss
Feature activation+1.296
olini
Tokenolini
Feature activation+1.438
(
Token (
Feature activation+0.000
the
Tokenthe
Feature activation+0.000
popular
Token popular
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.058
was
Token was
Feature activation+0.000
held
Token held
Feature activation+0.000
by
Token by
Feature activation+0.000
a
Token a
Feature activation+0.000
South
Token South
Feature activation+0.193
African
Token African
Feature activation+1.524
woman
Token woman
Feature activation+0.000
who
Token who
Feature activation+0.000
ate
Token ate
Feature activation+0.000
eight
Token eight
Feature activation+0.000
j
Token j
Feature activation+0.000

INTERVAL 1.082 - 1.352
CONTAINS 0.001%

-
Token-
Feature activation+0.000
long
Tokenlong
Feature activation+0.000
partnership
Token partnership
Feature activation+0.000
between
Token between
Feature activation+0.000
the
Token the
Feature activation+0.000
British
Token British
Feature activation+1.113
Council
Token Council
Feature activation+0.066
,
Token,
Feature activation+0.000
BBC
Token BBC
Feature activation+0.267
World
Token World
Feature activation+0.000
Service
Token Service
Feature activation+0.000
Indian
Token Indian
Feature activation+0.000
,
Token,
Feature activation+0.000
because
Token because
Feature activation+0.000
I
Token I
Feature activation+0.000
am
Token am
Feature activation+0.000
Mexican
Token Mexican
Feature activation+1.202
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Says
Token Says
Feature activation+0.000
its
Token its
Feature activation+0.000
true
Token true
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+0.000
to
Token to
Feature activation+0.000
protest
Token protest
Feature activation+0.000
against
Token against
Feature activation+0.000
Rome
Token Rome
Feature activation+1.182
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
decision
Token decision
Feature activation+0.000
not
Token not
Feature activation+0.000
move
Token move
Feature activation+0.000
a
Token a
Feature activation+0.000
proposal
Token proposal
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
Bon
Token Bon
Feature activation+1.191
n
Tokenn
Feature activation+0.000
climate
Token climate
Feature activation+0.000
talks
Token talks
Feature activation+0.000
starting
Token starting
Feature activation+0.000
from
Token from
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
court
Token court
Feature activation+0.000
ordered
Token ordered
Feature activation+0.000
ambassador
Token ambassador
Feature activation+0.000
Dan
Token Dan
Feature activation+1.206
ie
Tokenie
Feature activation+0.000
le
Tokenle
Feature activation+0.000
M
Token M
Feature activation+0.000
anc
Tokenanc
Feature activation+0.000
ini
Tokenini
Feature activation+0.000

INTERVAL 0.811 - 1.082
CONTAINS 0.002%

India
Token India
Feature activation+0.000
seeks
Token seeks
Feature activation+0.000
alternative
Token alternative
Feature activation+0.000
supplies
Token supplies
Feature activation+0.000
after
Token after
Feature activation+0.000
Australia
Token Australia
Feature activation+0.977
cycl
Token cycl
Feature activation+0.000
one
Tokenone
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
*
Token*
Feature activation+0.000
had
Token had
Feature activation+0.000
been
Token been
Feature activation+0.000
studying
Token studying
Feature activation+0.000
journalism
Token journalism
Feature activation+0.000
in
Token in
Feature activation+0.000
Islamic
Token Islamic
Feature activation+0.965
University
Token University
Feature activation+0.000
of
Token of
Feature activation+0.000
Science
Token Science
Feature activation+0.000
and
Token and
Feature activation+0.000
Technology
Token Technology
Feature activation+0.000
ban
Tokenban
Feature activation+0.299
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
In
Token In
Feature activation+0.000
West
Token West
Feature activation+0.833
Bengal
Token Bengal
Feature activation+0.091
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
senior
Token senior
Feature activation+0.000
minister
Token minister
Feature activation+0.000
most
Token most
Feature activation+0.000
dreaded
Token dreaded
Feature activation+0.000
terrorist
Token terrorist
Feature activation+0.420
organisation
Token organisation
Feature activation+0.000
Islamic
Token Islamic
Feature activation+1.018
State
Token State
Feature activation+0.966
of
Token of
Feature activation+0.000
Iraq
Token Iraq
Feature activation+0.551
and
Token and
Feature activation+0.000
Syria
Token Syria
Feature activation+0.835
has
Token has
Feature activation+0.000
of
Token of
Feature activation+0.000
even
Token even
Feature activation+0.000
thinking
Token thinking
Feature activation+0.000
against
Token against
Feature activation+0.000
the
Token the
Feature activation+0.000
Muss
Token Muss
Feature activation+1.045
ol
Tokenol
Feature activation+0.000
inis
Tokeninis
Feature activation+0.121
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
article
Token article
Feature activation+0.000

INTERVAL 0.541 - 0.811
CONTAINS 0.003%

in
Token in
Feature activation+0.000
Washington
Token Washington
Feature activation+0.000
in
Token in
Feature activation+0.000
September
Token September
Feature activation+0.000
when
Token when
Feature activation+0.000
Obama
Token Obama
Feature activation+0.572
took
Token took
Feature activation+0.000
Modi
Token Modi
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
memorial
Token memorial
Feature activation+0.000
terrorist
Token terrorist
Feature activation+0.420
organisation
Token organisation
Feature activation+0.000
Islamic
Token Islamic
Feature activation+1.018
State
Token State
Feature activation+0.966
of
Token of
Feature activation+0.000
Iraq
Token Iraq
Feature activation+0.551
and
Token and
Feature activation+0.000
Syria
Token Syria
Feature activation+0.835
has
Token has
Feature activation+0.000
reportedly
Token reportedly
Feature activation+0.000
exh
Token exh
Feature activation+0.000
Ad
Token Ad
Feature activation+0.000
ity
Tokenity
Feature activation+0.000
a
Tokena
Feature activation+0.000
B
Token B
Feature activation+0.000
irl
Tokenirl
Feature activation+0.000
a
Tokena
Feature activation+0.589
Group
Token Group
Feature activation+0.135
and
Token and
Feature activation+0.000
74
Token 74
Feature activation+0.000
%
Token%
Feature activation+0.000
for
Token for
Feature activation+0.000
,
Token,
Feature activation+0.000
poured
Token poured
Feature activation+0.000
a
Token a
Feature activation+0.000
cup
Token cup
Feature activation+0.000
for
Token for
Feature activation+0.000
Obama
Token Obama
Feature activation+0.635
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
more
Token more
Feature activation+0.000
planes
Token planes
Feature activation+0.000
and
Token and
Feature activation+0.000
Australian
Token Australian
Feature activation+1.033
and
Token and
Feature activation+0.000
Chinese
Token Chinese
Feature activation+0.581
navy
Token navy
Feature activation+0.000
vessels
Token vessels
Feature activation+0.000
were
Token were
Feature activation+0.000
also
Token also
Feature activation+0.000
ste
Token ste
Feature activation+0.000

INTERVAL 0.270 - 0.541
CONTAINS 0.007%

e
Tokene
Feature activation+0.000
ered
Tokenered
Feature activation+0.000
a
Token a
Feature activation+0.000
flight
Token flight
Feature activation+0.000
to
Token to
Feature activation+0.000
Bangkok
Token Bangkok
Feature activation+0.521
,"
Token,"
Feature activation+0.000
said
Token said
Feature activation+0.000
an
Token an
Feature activation+0.000
AI
Token AI
Feature activation+0.000
official
Token official
Feature activation+0.000
thought
Token thought
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
members
Token members
Feature activation+0.000
of
Token of
Feature activation+0.000
Pakistan
Token Pakistan
Feature activation+0.426
-
Token-
Feature activation+0.000
based
Tokenbased
Feature activation+0.000
Lash
Token Lash
Feature activation+0.000
kar
Tokenkar
Feature activation+0.000
-
Token-
Feature activation+0.000
a
Token a
Feature activation+0.000
paramilitary
Token paramilitary
Feature activation+0.000
force
Token force
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
Crimean
Token Crimean
Feature activation+0.518
Peninsula
Token Peninsula
Feature activation+0.000
early
Token early
Feature activation+0.000
this
Token this
Feature activation+0.000
year
Token year
Feature activation+0.000
as
Token as
Feature activation+0.000
says
Token says
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
India
Token India
Feature activation+0.000
-
Token-
Feature activation+0.000
Japan
TokenJapan
Feature activation+0.276
relationship
Token relationship
Feature activation+0.000
has
Token has
Feature activation+0.000
a
Token a
Feature activation+0.000
long
Token long
Feature activation+0.000
history
Token history
Feature activation+0.000
order
Token order
Feature activation+0.000
to
Token to
Feature activation+0.000
protect
Token protect
Feature activation+0.000
their
Token their
Feature activation+0.000
misunderstood
Token misunderstood
Feature activation+0.000
py
Token py
Feature activation+0.351
ro
Tokenro
Feature activation+0.000
,
Token,
Feature activation+0.000
P
Token P
Feature activation+0.000
ia
Tokenia
Feature activation+0.000
and
Token and
Feature activation+0.000

INTERVAL 0.000 - 0.270
CONTAINS 99.987%

have
Token have
Feature activation+0.000
had
Token had
Feature activation+0.000
no
Token no
Feature activation+0.000
deterrent
Token deterrent
Feature activation+0.000
effect
Token effect
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
past
Token past
Feature activation+0.000
years
Token years
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
http
Token http
Feature activation+0.000
://
Token://
Feature activation+0.000
fire
Tokenfire
Feature activation+0.000
arms
Tokenarms
Feature activation+0.000
policy
Tokenpolicy
Feature activation+0.000
.
Token.
Feature activation+0.000
org
Tokenorg
Feature activation+0.000
/
Token/
Feature activation+0.000
contact
Tokencontact
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
advocates
Token advocates
Feature activation+0.000
from
Token from
Feature activation+0.000
around
Token around
Feature activation+0.000
Dallas
Token Dallas
Feature activation+0.000
spent
Token spent
Feature activation+0.000
one
Token one
Feature activation+0.000
night
Token night
Feature activation+0.000
last
Token last
Feature activation+0.000
week
Token week
Feature activation+0.000
making
Token making
Feature activation+0.000
fake
Token fake
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
states
Token states
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Mem
Token Mem
Feature activation+0.000
o
Tokeno
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
activists
Token activists
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Following
TokenFollowing
Feature activation+0.000
his
Token his
Feature activation+0.000
comments
Token comments
Feature activation+0.000
,
Token,
Feature activation+0.000
Mr
Token Mr
Feature activation+0.000
Netanyahu
Token Netanyahu
Feature activation+0.000
announced
Token announced
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 H3.4: (feature 22316)

TOP ACTIVATIONS
MAX = 2.979

same
Token same
Feature activation+0.000
-
Token-
Feature activation+0.000
sex
Tokensex
Feature activation+0.000
relationship
Token relationship
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+2.979
was
Token was
Feature activation+0.000
not
Token not
Feature activation+0.000
allowed
Token allowed
Feature activation+0.000
to
Token to
Feature activation+0.000
do
Token do
Feature activation+0.000
instantly
Token instantly
Feature activation+0.000
healed
Token healed
Feature activation+0.000
.
Token.
Feature activation+0.000
48
Token 48
Feature activation+0.000
Then
Token Then
Feature activation+0.000
he
Token he
Feature activation+2.916
said
Token said
Feature activation+0.000
to
Token to
Feature activation+0.000
her
Token her
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
8
Token8
Feature activation+0.000
.
Token.
Feature activation+0.000
2
Token2
Feature activation+0.000
million
Token million
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+2.782
own
Token own
Feature activation+0.000
to
Token to
Feature activation+0.000
oust
Token oust
Feature activation+0.000
Sen
Token Sen
Feature activation+0.000
.
Token.
Feature activation+0.000
and
Token and
Feature activation+0.000
continues
Token continues
Feature activation+0.000
to
Token to
Feature activation+0.000
struggle
Token struggle
Feature activation+0.000
with
Token with
Feature activation+0.000
his
Token his
Feature activation+2.723
actions
Token actions
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
past
Token past
Feature activation+0.000
seasons
Token seasons
Feature activation+0.000
also
Token also
Feature activation+0.000
supported
Token supported
Feature activation+0.000
her
Token her
Feature activation+0.400
partner
Token partner
Feature activation+0.088
while
Token while
Feature activation+0.000
he
Token he
Feature activation+2.651
was
Token was
Feature activation+0.025
in
Token in
Feature activation+0.000
jail
Token jail
Feature activation+0.000
.
Token.
Feature activation+0.000
She
Token She
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
question
Token question
Feature activation+0.000
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
his
Token his
Feature activation+2.618
connections
Token connections
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Open
Token Open
Feature activation+0.000
Phil
Token Phil
Feature activation+0.000
.
Token.
Feature activation+0.000
Just
Token Just
Feature activation+0.000
like
Token like
Feature activation+0.000
her
Token her
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+2.547
has
Token has
Feature activation+0.000
regrets
Token regrets
Feature activation+0.000
and
Token and
Feature activation+0.000
continues
Token continues
Feature activation+0.000
to
Token to
Feature activation+0.000
N
TokenN
Feature activation+0.000
oor
Tokenoor
Feature activation+0.000
did
Token did
Feature activation+0.000
not
Token not
Feature activation+0.000
activate
Token activate
Feature activation+0.000
his
Token his
Feature activation+2.515
body
Token body
Feature activation+0.000
camera
Token camera
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
as
Token as
Feature activation+0.000
required
Token required
Feature activation+0.000
aven
Tokenaven
Feature activation+0.000
legs
Token legs
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
He
TokenHe
Feature activation+2.504
has
Token has
Feature activation+0.092
no
Token no
Feature activation+0.070
choice
Token choice
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
in
Token in
Feature activation+0.000
jail
Token jail
Feature activation+0.000
.
Token.
Feature activation+0.000
She
Token She
Feature activation+0.000
sent
Token sent
Feature activation+0.000
him
Token him
Feature activation+2.453
$
Token $
Feature activation+0.000
50
Token50
Feature activation+0.000
a
Token a
Feature activation+0.000
week
Token week
Feature activation+0.000
for
Token for
Feature activation+0.000
gave
Token gave
Feature activation+0.000
her
Token her
Feature activation+0.000
a
Token a
Feature activation+0.000
copy
Token copy
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+2.384
documentary
Token documentary
Feature activation+0.000
"
Token "
Feature activation+0.000
Before
TokenBefore
Feature activation+0.000
the
Token the
Feature activation+0.000
Flood
Token Flood
Feature activation+0.000
chief
Token chief
Feature activation+0.000
at
Token at
Feature activation+0.000
times
Token times
Feature activation+0.000
says
Token says
Feature activation+0.000
that
Token that
Feature activation+0.000
he
Token he
Feature activation+2.374
was
Token was
Feature activation+0.000
made
Token made
Feature activation+0.000
the
Token the
Feature activation+0.000
offer
Token offer
Feature activation+0.000
by
Token by
Feature activation+0.000
such
Token such
Feature activation+0.000
violence
Token violence
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
he
Token he
Feature activation+2.356
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
During
TokenDuring
Feature activation+0.000
to
Token to
Feature activation+0.000
her
Token her
Feature activation+0.000
boyfriend
Token boyfriend
Feature activation+0.000
actually
Token actually
Feature activation+0.000
knew
Token knew
Feature activation+0.000
he
Token he
Feature activation+2.350
was
Token was
Feature activation+0.185
driving
Token driving
Feature activation+0.000
,
Token,
Feature activation+0.000
so
Token so
Feature activation+0.000
she
Token she
Feature activation+0.000
-
Token-
Feature activation+0.000
38
Token38
Feature activation+0.000
).
Token).
Feature activation+0.000
Then
Token Then
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+2.279
visits
Token visits
Feature activation+0.000
Joseph
Token Joseph
Feature activation+0.000
to
Token to
Feature activation+0.000
ensure
Token ensure
Feature activation+0.000
him
Token him
Feature activation+1.461
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
for
Token for
Feature activation+0.000
some
Token some
Feature activation+0.000
reason
Token reason
Feature activation+0.000
he
Token he
Feature activation+2.264
gets
Token gets
Feature activation+0.000
to
Token to
Feature activation+0.000
wear
Token wear
Feature activation+0.000
a
Token a
Feature activation+0.000
lot
Token lot
Feature activation+0.000
$
Token $
Feature activation+0.000
1
Token1
Feature activation+0.000
/
Token/
Feature activation+0.000
minute
Tokenminute
Feature activation+0.000
for
Token for
Feature activation+0.000
him
Token him
Feature activation+2.233
to
Token to
Feature activation+0.000
call
Token call
Feature activation+0.000
home
Token home
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Bill
TokenBill
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
's
Token's
Feature activation+0.000
defense
Token defense
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+2.210
wife
Token wife
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
lack
Token lack
Feature activation+0.000
,
Token,
Feature activation+0.000
her
Token her
Feature activation+0.000
tail
Token tail
Feature activation+0.000
and
Token and
Feature activation+0.000
as
Token as
Feature activation+0.000
he
Token he
Feature activation+2.158
was
Token was
Feature activation+0.091
rubbing
Token rubbing
Feature activation+0.000
her
Token her
Feature activation+0.000
and
Token and
Feature activation+0.000
moving
Token moving
Feature activation+0.000
bombing
Token bombing
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
He
TokenHe
Feature activation+2.133
's
Token's
Feature activation+0.000
seen
Token seen
Feature activation+0.000
me
Token me
Feature activation+0.652
on
Token on
Feature activation+0.000
my
Token my
Feature activation+0.486

Top DFA by src position
MAX = 2.972

Mic
Token Mic
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
told
Token told
Feature activation-0.000
Top resid features:
VICE
Token VICE
Feature activation-0.000
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
she
Token she
Feature activation+1.862
Top resid features:
was
Token was
Feature activation-0.001
Top resid features:
once
Token once
Feature activation+0.000
Top resid features:
banned
Token banned
Feature activation-0.000
Top resid features:
from
Token from
Feature activation-0.000
Top resid features:
giving
Token giving
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.001
Top resid features:
why
Token why
Feature activation-0.013
Top resid features:
she
Token she
Feature activation+1.720
Top resid features:
had
Token had
Feature activation+0.009
Top resid features:
touched
Token touched
Feature activation+0.005
Top resid features:
him
Token him
Feature activation+0.029
Top resid features:
and
Token and
Feature activation+0.004
Top resid features:
how
Token how
Feature activation-0.006
Top resid features:
than
Token than
Feature activation-0.002
Top resid features:
$
Token $
Feature activation-0.001
Top resid features:
142
Token142
Feature activation+0.001
Top resid features:
million
Token million
Feature activation-0.003
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
her
Token her
Feature activation+2.972
Top resid features:
own
Token own
Feature activation+0.000
Top resid features:
money
Token money
Feature activation-0.001
Top resid features:
in
Token in
Feature activation-0.000
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
California
Token California
Feature activation-0.001
Top resid features:
by
Token by
Feature activation-0.001
Top resid features:
her
Token her
Feature activation+1.012
Top resid features:
side
Token side
Feature activation-0.001
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
help
Token help
Feature activation-0.002
Top resid features:
her
Token her
Feature activation+1.289
Top resid features:
lead
Token lead
Feature activation-0.005
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
they
Token they
Feature activation+0.044
Top resid features:
work
Token work
Feature activation-0.000
Top resid features:
better
Token better
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.012
Top resid features:
herself
Token herself
Feature activation+0.369
Top resid features:
and
Token and
Feature activation+0.013
Top resid features:
her
Token her
Feature activation+0.909
Top resid features:
daughter
Token daughter
Feature activation+0.099
Top resid features:
but
Token but
Feature activation+0.010
Top resid features:
could
Token could
Feature activation+0.004
Top resid features:
not
Token not
Feature activation+0.001
Top resid features:
get
Token get
Feature activation+0.001
Top resid features:
is
Tokenis
Feature activation+0.002
Top resid features:
Ay
Token Ay
Feature activation+0.002
Top resid features:
ala
Tokenala
Feature activation+0.002
Top resid features:
based
Token based
Feature activation+0.001
Top resid features:
on
Token on
Feature activation+0.002
Top resid features:
her
Token her
Feature activation+1.969
Top resid features:
agreeing
Token agreeing
Feature activation-0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
not
Token not
Feature activation+0.002
Top resid features:
pursue
Token pursue
Feature activation+0.001
Top resid features:
death
Token death
Feature activation-0.007
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.004
Top resid features:
ley
Tokenley
Feature activation-0.009
Top resid features:
)
Token)
Feature activation-0.003
Top resid features:
by
Token by
Feature activation-0.002
Top resid features:
her
Token her
Feature activation+1.213
Top resid features:
side
Token side
Feature activation-0.004
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
help
Token help
Feature activation-0.001
Top resid features:
her
Token her
Feature activation+1.148
Top resid features:
lead
Token lead
Feature activation-0.002
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
abdomen
Token abdomen
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
ultimately
Token ultimately
Feature activation+0.004
Top resid features:
killing
Token killing
Feature activation-0.002
Top resid features:
her
Token her
Feature activation+1.894
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
When
TokenWhen
Feature activation+0.000
Top resid features:
she
Token she
Feature activation+0.675
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.010
Top resid features:
.
Token.
Feature activation+0.009
Top resid features:
She
Token She
Feature activation+0.587
Top resid features:
says
Token says
Feature activation+0.002
Top resid features:
she
Token she
Feature activation+0.516
Top resid features:
waited
Token waited
Feature activation+0.000
Top resid features:
weeks
Token weeks
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.002
Top resid features:
herself
Token herself
Feature activation+0.469
Top resid features:
and
Token and
Feature activation+0.007
Top resid features:
her
Token her
Feature activation+0.529
Top resid features:
daughter
Token daughter
Feature activation+0.074
Top resid features:
but
Token but
Feature activation+0.004
Top resid features:
could
Token could
Feature activation+0.005
Top resid features:
not
Token not
Feature activation+0.002
Top resid features:
get
Token get
Feature activation+0.001
Top resid features:
with
Token with
Feature activation-0.005
Top resid features:
Ivanka
Token Ivanka
Feature activation+0.076
Top resid features:
Trump
Token Trump
Feature activation+0.007
Top resid features:
and
Token and
Feature activation-0.001
Top resid features:
gave
Token gave
Feature activation-0.030
Top resid features:
her
Token her
Feature activation+2.599
Top resid features:
a
Token a
Feature activation+0.007
Top resid features:
copy
Token copy
Feature activation-0.053
Top resid features:
of
Token of
Feature activation-0.059
Top resid features:
his
Token his
Feature activation+0.128
Top resid features:
documentary
Token documentary
Feature activation+0.000
Top resid features:
hereafter
Token hereafter
Feature activation-0.001
Top resid features:
".
Token".
Feature activation-0.012
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
In
TokenIn
Feature activation-0.000
Top resid features:
her
Token her
Feature activation+1.796
Top resid features:
tweet
Token tweet
Feature activation-0.005
Top resid features:
,
Token,
Feature activation+0.005
Top resid features:
she
Token she
Feature activation+0.973
Top resid features:
noted
Token noted
Feature activation-0.007
Top resid features:
that
Token that
Feature activation-0.012
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
funeral
Token funeral
Feature activation-0.000
Top resid features:
procession
Token procession
Feature activation-0.002
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
her
Token her
Feature activation+1.173
Top resid features:
father
Token father
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
Za
Token Za
Feature activation-0.000
Top resid features:
her
Tokenher
Feature activation+0.003
Top resid features:
N
Token N
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.002
Top resid features:
saying
Token saying
Feature activation+0.000
Top resid features:
that
Token that
Feature activation+0.001
Top resid features:
she
Token she
Feature activation+0.803
Top resid features:
knew
Token knew
Feature activation-0.003
Top resid features:
she
Token she
Feature activation+0.587
Top resid features:
was
Token was
Feature activation-0.001
Top resid features:
texting
Token texting
Feature activation+0.000
Top resid features:
someone
Token someone
Feature activation+0.004
Top resid features:
Gabriel
Token Gabriel
Feature activation+0.018
Top resid features:
visits
Token visits
Feature activation+0.029
Top resid features:
Mary
Token Mary
Feature activation+0.039
Top resid features:
to
Token to
Feature activation+0.006
Top resid features:
tell
Token tell
Feature activation+0.007
Top resid features:
her
Token her
Feature activation+1.491
Top resid features:
that
Token that
Feature activation+0.007
Top resid features:
she
Token she
Feature activation+1.024
Top resid features:
will
Token will
Feature activation+0.005
Top resid features:
give
Token give
Feature activation+0.001
Top resid features:
a
Token a
Feature activation-0.000
Top resid features:
be
Token be
Feature activation-0.002
Top resid features:
exposed
Token exposed
Feature activation-0.000
Top resid features:
in
Token in
Feature activation-0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
way
Token way
Feature activation-0.000
Top resid features:
she
Token she
Feature activation+1.083
Top resid features:
is
Token is
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
She
Token She
Feature activation+0.929
Top resid features:
has
Token has
Feature activation-0.003
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.013
Top resid features:
herself
Token herself
Feature activation+0.340
Top resid features:
and
Token and
Feature activation+0.005
Top resid features:
her
Token her
Feature activation+0.419
Top resid features:
daughter
Token daughter
Feature activation+0.169
Top resid features:
but
Token but
Feature activation+0.006
Top resid features:
could
Token could
Feature activation+0.002
Top resid features:
not
Token not
Feature activation+0.001
Top resid features:
get
Token get
Feature activation+0.001
Top resid features:
make
Token make
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
when
Token when
Feature activation+0.002
Top resid features:
she
Token she
Feature activation+1.656
Top resid features:
is
Token is
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
person
Token person
Feature activation+0.016
Top resid features:
that
Token that
Feature activation+0.000
Top resid features:
made
Token made
Feature activation-0.001
Top resid features:
that
Token that
Feature activation+0.001
Top resid features:
it
Token it
Feature activation+0.029
Top resid features:
only
Token only
Feature activation+0.001
Top resid features:
happened
Token happened
Feature activation-0.001
Top resid features:
because
Token because
Feature activation+0.000
Top resid features:
she
Token she
Feature activation+1.009
Top resid features:
"
Token "
Feature activation+0.000
Top resid features:
sed
Token sed
Feature activation+0.000
Top resid features:
uce
Tokenuce
Feature activation+0.000
Top resid features:
d
Token d
Feature activation-0.001
Top resid features:
"
Token"
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.002
Top resid features:
explosion
Token explosion
Feature activation-0.023
Top resid features:
caused
Token caused
Feature activation-0.001
Top resid features:
her
Token her
Feature activation+1.236
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
lose
Token lose
Feature activation-0.000
Top resid features:
a
Token a
Feature activation+0.001
Top resid features:
part
Token part
Feature activation-0.000
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:

Decoder Weights Distribution

Head 0: 0.11

Head 1: 0.03

Head 2: 0.03

Head 3: 0.02

Head 4: 0.44

Head 5: 0.09

Head 6: 0.09

Head 7: 0.03

Head 8: 0.03

Head 9: 0.03

Head 10: 0.06

Head 11: 0.03

Positive logits

asso2.11

Flavoring2.05

etooth1.98

todd1.84

icult1.83

itive1.83

rigged1.79

lass1.76

iculty1.76

rell1.74

inates1.74

glim1.72

rust1.71

WATCHED1.70

reet1.69

ilan1.68

ija1.68

odiac1.66

aldo1.65

hest1.64

Negative logits

utical-2.01

uterus-1.80

Disk-1.77

capacity-1.77

UCS-1.75

Aust-1.73

Carib-1.72

BuyableInstoreAndOnline-1.69

Cent-1.65

experien-1.65

gap-1.65

maternity-1.64

campuses-1.62

Bie-1.61

disag-1.60

wide-1.59

DPRK-1.58

Eag-1.57

North-1.55

Ear-1.53

INTERVAL 2.681 - 2.979
CONTAINS 0.000%

and
Token and
Feature activation+0.000
continues
Token continues
Feature activation+0.000
to
Token to
Feature activation+0.000
struggle
Token struggle
Feature activation+0.000
with
Token with
Feature activation+0.000
his
Token his
Feature activation+2.723
actions
Token actions
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
past
Token past
Feature activation+0.000
seasons
Token seasons
Feature activation+0.000
8
Token8
Feature activation+0.000
.
Token.
Feature activation+0.000
2
Token2
Feature activation+0.000
million
Token million
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+2.782
own
Token own
Feature activation+0.000
to
Token to
Feature activation+0.000
oust
Token oust
Feature activation+0.000
Sen
Token Sen
Feature activation+0.000
.
Token.
Feature activation+0.000
instantly
Token instantly
Feature activation+0.000
healed
Token healed
Feature activation+0.000
.
Token.
Feature activation+0.000
48
Token 48
Feature activation+0.000
Then
Token Then
Feature activation+0.000
he
Token he
Feature activation+2.916
said
Token said
Feature activation+0.000
to
Token to
Feature activation+0.000
her
Token her
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
same
Token same
Feature activation+0.000
-
Token-
Feature activation+0.000
sex
Tokensex
Feature activation+0.000
relationship
Token relationship
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+2.979
was
Token was
Feature activation+0.000
not
Token not
Feature activation+0.000
allowed
Token allowed
Feature activation+0.000
to
Token to
Feature activation+0.000
do
Token do
Feature activation+0.000

INTERVAL 2.383 - 2.681
CONTAINS 0.001%

to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
question
Token question
Feature activation+0.000
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
his
Token his
Feature activation+2.618
connections
Token connections
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Open
Token Open
Feature activation+0.000
Phil
Token Phil
Feature activation+0.000
gave
Token gave
Feature activation+0.000
her
Token her
Feature activation+0.000
a
Token a
Feature activation+0.000
copy
Token copy
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+2.384
documentary
Token documentary
Feature activation+0.000
"
Token "
Feature activation+0.000
Before
TokenBefore
Feature activation+0.000
the
Token the
Feature activation+0.000
Flood
Token Flood
Feature activation+0.000
.
Token.
Feature activation+0.000
Just
Token Just
Feature activation+0.000
like
Token like
Feature activation+0.000
her
Token her
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+2.547
has
Token has
Feature activation+0.000
regrets
Token regrets
Feature activation+0.000
and
Token and
Feature activation+0.000
continues
Token continues
Feature activation+0.000
to
Token to
Feature activation+0.000
also
Token also
Feature activation+0.000
supported
Token supported
Feature activation+0.000
her
Token her
Feature activation+0.400
partner
Token partner
Feature activation+0.088
while
Token while
Feature activation+0.000
he
Token he
Feature activation+2.651
was
Token was
Feature activation+0.025
in
Token in
Feature activation+0.000
jail
Token jail
Feature activation+0.000
.
Token.
Feature activation+0.000
She
Token She
Feature activation+0.000
N
TokenN
Feature activation+0.000
oor
Tokenoor
Feature activation+0.000
did
Token did
Feature activation+0.000
not
Token not
Feature activation+0.000
activate
Token activate
Feature activation+0.000
his
Token his
Feature activation+2.515
body
Token body
Feature activation+0.000
camera
Token camera
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
as
Token as
Feature activation+0.000
required
Token required
Feature activation+0.000

INTERVAL 2.086 - 2.383
CONTAINS 0.001%

,
Token,
Feature activation+0.000
her
Token her
Feature activation+0.000
tail
Token tail
Feature activation+0.000
and
Token and
Feature activation+0.000
as
Token as
Feature activation+0.000
he
Token he
Feature activation+2.158
was
Token was
Feature activation+0.091
rubbing
Token rubbing
Feature activation+0.000
her
Token her
Feature activation+0.000
and
Token and
Feature activation+0.000
moving
Token moving
Feature activation+0.000
chief
Token chief
Feature activation+0.000
at
Token at
Feature activation+0.000
times
Token times
Feature activation+0.000
says
Token says
Feature activation+0.000
that
Token that
Feature activation+0.000
he
Token he
Feature activation+2.374
was
Token was
Feature activation+0.000
made
Token made
Feature activation+0.000
the
Token the
Feature activation+0.000
offer
Token offer
Feature activation+0.000
by
Token by
Feature activation+0.000
at
Token at
Feature activation+0.000
worst
Token worst
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
he
Token he
Feature activation+2.098
said
Token said
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
to
Token to
Feature activation+0.000
her
Token her
Feature activation+0.000
boyfriend
Token boyfriend
Feature activation+0.000
actually
Token actually
Feature activation+0.000
knew
Token knew
Feature activation+0.000
he
Token he
Feature activation+2.350
was
Token was
Feature activation+0.185
driving
Token driving
Feature activation+0.000
,
Token,
Feature activation+0.000
so
Token so
Feature activation+0.000
she
Token she
Feature activation+0.000
-
Token-
Feature activation+0.000
38
Token38
Feature activation+0.000
).
Token).
Feature activation+0.000
Then
Token Then
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+2.279
visits
Token visits
Feature activation+0.000
Joseph
Token Joseph
Feature activation+0.000
to
Token to
Feature activation+0.000
ensure
Token ensure
Feature activation+0.000
him
Token him
Feature activation+1.461

INTERVAL 1.788 - 2.086
CONTAINS 0.003%

âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
learned
Token learned
Feature activation+0.000
from
Token from
Feature activation+0.000
his
Token his
Feature activation+2.025
past
Token past
Feature activation+0.000
and
Token and
Feature activation+0.000
uses
Token uses
Feature activation+0.000
it
Token it
Feature activation+0.108
to
Token to
Feature activation+0.000
handling
Token handling
Feature activation+0.000
venom
Token venom
Feature activation+0.000
ous
Tokenous
Feature activation+0.000
snakes
Token snakes
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+1.983
did
Token did
Feature activation+0.000
his
Token his
Feature activation+0.681
best
Token best
Feature activation+0.000
to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
She
TokenShe
Feature activation+0.000
is
Token is
Feature activation+0.000
also
Token also
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+1.898
said
Token said
Feature activation+0.000
,
Token,
Feature activation+0.000
"
Token "
Feature activation+0.000
a
Tokena
Feature activation+0.000
world
Token world
Feature activation+0.000
came
Token came
Feature activation+0.000
to
Token to
Feature activation+0.000
her
Token her
Feature activation+0.307
apartment
Token apartment
Feature activation+0.000
,
Token,
Feature activation+0.000
his
Token his
Feature activation+1.808
two
Token two
Feature activation+0.000
kids
Token kids
Feature activation+0.000
in
Token in
Feature activation+0.000
tow
Token tow
Feature activation+0.000
.
Token.
Feature activation+0.000
-
Token-
Feature activation+0.000
year
Tokenyear
Feature activation+0.000
relationship
Token relationship
Feature activation+0.000
.
Token.
Feature activation+0.000
Through
Token Through
Feature activation+0.000
him
Token him
Feature activation+1.835
she
Token she
Feature activation+0.000
met
Token met
Feature activation+0.000
crucial
Token crucial
Feature activation+0.000
players
Token players
Feature activation+0.000
like
Token like
Feature activation+0.000

INTERVAL 1.490 - 1.788
CONTAINS 0.003%

time
Token time
Feature activation+0.000
I
Token I
Feature activation+0.088
met
Token met
Feature activation+0.000
him
Token him
Feature activation+1.226
,
Token,
Feature activation+0.000
he
Token he
Feature activation+1.628
looked
Token looked
Feature activation+0.000
at
Token at
Feature activation+0.000
my
Token my
Feature activation+0.283
chest
Token chest
Feature activation+0.000
immediately
Token immediately
Feature activation+0.000
than
Token than
Feature activation+0.000
Quiet
Token Quiet
Feature activation+0.000
does
Token does
Feature activation+0.000
.
Token.
Feature activation+0.000
Probably
Token Probably
Feature activation+0.000
his
Token his
Feature activation+1.492
lack
Token lack
Feature activation+0.000
of
Token of
Feature activation+0.000
display
Token display
Feature activation+0.000
able
Tokenable
Feature activation+0.000
tits
Token tits
Feature activation+0.000
description
Token description
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
hopeless
Token hopeless
Feature activation+0.000
climate
Token climate
Feature activation+0.000
his
Token his
Feature activation+1.515
election
Token election
Feature activation+0.000
win
Token win
Feature activation+0.000
has
Token has
Feature activation+0.000
created
Token created
Feature activation+0.000
.
Token.
Feature activation+0.000
2002
Token 2002
Feature activation+0.000
Royal
Token Royal
Feature activation+0.000
Rumble
Token Rumble
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
his
Token his
Feature activation+1.724
eventual
Token eventual
Feature activation+0.000
departure
Token departure
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
company
Token company
Feature activation+0.000
how
Token how
Feature activation+0.000
years
Token years
Feature activation+0.000
after
Token after
Feature activation+0.000
their
Token their
Feature activation+0.138
divorce
Token divorce
Feature activation+0.000
his
Token his
Feature activation+1.642
painful
Token painful
Feature activation+0.000
words
Token words
Feature activation+0.000
led
Token led
Feature activation+0.000
her
Token her
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 1.192 - 1.490
CONTAINS 0.005%

ving
Tokenving
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
presidency
Token presidency
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+1.348
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
not
Token not
Feature activation+0.000
only
Token only
Feature activation+0.000
a
Token a
Feature activation+0.000
hero
Token hero
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
His
TokenHis
Feature activation+1.368
story
Token story
Feature activation+0.000
is
Token is
Feature activation+0.000
hugely
Token hugely
Feature activation+0.000
inspiring
Token inspiring
Feature activation+0.000
:
Token:
Feature activation+0.000
going
Token going
Feature activation+0.000
to
Token to
Feature activation+0.000
have
Token have
Feature activation+0.000
to
Token to
Feature activation+0.000
endorse
Token endorse
Feature activation+0.000
him
Token him
Feature activation+1.419
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
of
Token of
Feature activation+0.000
jail
Token jail
Feature activation+0.000
,
Token,
Feature activation+0.000
referring
Token referring
Feature activation+0.000
to
Token to
Feature activation+0.000
himself
Token himself
Feature activation+1.346
as
Token as
Feature activation+0.000
the
Token the
Feature activation+0.000
"
Token "
Feature activation+0.000
No
TokenNo
Feature activation+0.000
.
Token.
Feature activation+0.000
the
Token the
Feature activation+0.000
child
Token child
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
case
Token case
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+1.470
declined
Token declined
Feature activation+0.000
to
Token to
Feature activation+0.000
discuss
Token discuss
Feature activation+0.000
it
Token it
Feature activation+0.000
,
Token,
Feature activation+0.000

INTERVAL 0.894 - 1.192
CONTAINS 0.007%

language
Token language
Feature activation+0.000
when
Token when
Feature activation+0.000
talking
Token talking
Feature activation+0.000
about
Token about
Feature activation+0.000
how
Token how
Feature activation+0.000
he
Token he
Feature activation+1.075
violated
Token violated
Feature activation+0.000
Ber
Token Ber
Feature activation+0.000
tha
Tokentha
Feature activation+0.000
J
Token J
Feature activation+0.000
ork
Tokenork
Feature activation+0.000
back
Token back
Feature activation+0.000
in
Token in
Feature activation+0.000
reality
Token reality
Feature activation+0.000
.
Token.
Feature activation+0.000
During
Token During
Feature activation+0.000
his
Token his
Feature activation+1.028
return
Token return
Feature activation+0.000
to
Token to
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
in
Token in
Feature activation+0.000
time
Token time
Feature activation+0.000
she
Token she
Feature activation+0.000
presumed
Token presumed
Feature activation+0.000
he
Token he
Feature activation+0.970
had
Token had
Feature activation+0.000
died
Token died
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
oria
Tokenoria
Feature activation+0.000
,
Token,
Feature activation+0.000
after
Token after
Feature activation+0.000
her
Token her
Feature activation+0.000
teenage
Token teenage
Feature activation+0.000
son
Token son
Feature activation+1.032
,
Token,
Feature activation+0.000
R
Token R
Feature activation+0.000
olf
Tokenolf
Feature activation+0.000
,
Token,
Feature activation+0.000
had
Token had
Feature activation+0.000
affairs
Token affairs
Feature activation+0.000
under
Token under
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
he
Token he
Feature activation+1.050
did
Token did
Feature activation+0.000
not
Token not
Feature activation+0.000
make
Token make
Feature activation+0.000
himself
Token himself
Feature activation+0.548
available
Token available
Feature activation+0.000

INTERVAL 0.596 - 0.894
CONTAINS 0.012%

,
Token,
Feature activation+0.000
I
Token I
Feature activation+0.541
tell
Token tell
Feature activation+0.000
you
Token you
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.721
cannot
Token cannot
Feature activation+0.000
be
Token be
Feature activation+0.000
done
Token done
Feature activation+0.000
by
Token by
Feature activation+0.000
a
Token a
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
EPA
Token EPA
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
his
Token his
Feature activation+0.728
nomination
Token nomination
Feature activation+0.000
drew
Token drew
Feature activation+0.000
swift
Token swift
Feature activation+0.000
criticism
Token criticism
Feature activation+0.000
from
Token from
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
was
Token was
Feature activation+0.000
like
Token like
Feature activation+0.000
âĢ¦
Token âĢ¦
Feature activation+0.000
he
Token he
Feature activation+0.858
just
Token just
Feature activation+0.000
,
Token,
Feature activation+0.000
I
Token I
Feature activation+0.000
don
Token don
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
shocked
Token shocked
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
news
Token news
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.747
arrest
Token arrest
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
for
Token for
Feature activation+0.000
someone
Token someone
Feature activation+0.000
who
Token who
Feature activation+0.000
would
Token would
Feature activation+0.000
remind
Token remind
Feature activation+0.000
him
Token him
Feature activation+0.672
of
Token of
Feature activation+0.000
all
Token all
Feature activation+0.000
his
Token his
Feature activation+0.392
pain
Token pain
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 0.298 - 0.596
CONTAINS 0.022%

her
Token her
Feature activation+0.000
in
Token in
Feature activation+0.000
relationship
Token relationship
Feature activation+0.000
limbo
Token limbo
Feature activation+0.000
.
Token.
Feature activation+0.000
My
Token My
Feature activation+0.356
father
Token father
Feature activation+0.477
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
sto
Token sto
Feature activation+0.000
ic
Tokenic
Feature activation+0.000
week
Token week
Feature activation+0.000
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+0.324
with
Token with
Feature activation+0.000
my
Token my
Feature activation+0.000
in
Token in
Feature activation+0.000
-
Token-
Feature activation+0.000
laws
Tokenlaws
Feature activation+0.000
I
TokenI
Feature activation+0.000
remember
Token remember
Feature activation+0.000
distinctly
Token distinctly
Feature activation+0.000
her
Token her
Feature activation+0.000
saying
Token saying
Feature activation+0.000
you
Token you
Feature activation+0.312
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
do
Token do
Feature activation+0.000
overwhelmed
Token overwhelmed
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
so
Token so
Feature activation+0.000
was
Token was
Feature activation+0.000
my
Token my
Feature activation+0.344
wife
Token wife
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
he
Token he
Feature activation+1.954
.
Token.
Feature activation+0.000
Once
Token Once
Feature activation+0.000
he
Token he
Feature activation+0.000
finished
Token finished
Feature activation+0.000
,
Token,
Feature activation+0.000
I
Token I
Feature activation+0.391
offered
Token offered
Feature activation+0.000
my
Token my
Feature activation+0.000
part
Token part
Feature activation+0.000
.
Token.
Feature activation+0.000
I
Token I
Feature activation+0.000

INTERVAL 0.000 - 0.298
CONTAINS 99.945%

and
Token and
Feature activation+0.000
Territory
Token Territory
Feature activation+0.000
discrimination
Token discrimination
Feature activation+0.000
legislation
Token legislation
Feature activation+0.000
,
Token,
Feature activation+0.000
authorised
Token authorised
Feature activation+0.000
celebr
Token celebr
Feature activation+0.000
ants
Tokenants
Feature activation+0.000
who
Token who
Feature activation+0.000
are
Token are
Feature activation+0.000
not
Token not
Feature activation+0.000
Ad
Token Ad
Feature activation+0.000
elson
Tokenelson
Feature activation+0.000
and
Token and
Feature activation+0.000
Goldman
Token Goldman
Feature activation+0.000
Sachs
Token Sachs
Feature activation+0.000
,
Token,
Feature activation+0.000
fell
Token fell
Feature activation+0.000
apart
Token apart
Feature activation+0.000
toward
Token toward
Feature activation+0.000
the
Token the
Feature activation+0.000
end
Token end
Feature activation+0.000
Colorado
TokenColorado
Feature activation+0.000
State
Token State
Feature activation+0.000
University
Token University
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
College
Token College
Feature activation+0.000
of
Token of
Feature activation+0.000
Veterinary
Token Veterinary
Feature activation+0.000
Medicine
Token Medicine
Feature activation+0.000
&
Token &
Feature activation+0.000
suit
Token suit
Feature activation+0.000
as
Token as
Feature activation+0.000
homes
Token homes
Feature activation+0.000
under
Token under
Feature activation+0.000
construction
Token construction
Feature activation+0.000
reach
Token reach
Feature activation+0.000
completion
Token completion
Feature activation+0.000
and
Token and
Feature activation+0.000
more
Token more
Feature activation+0.000
supply
Token supply
Feature activation+0.000
comes
Token comes
Feature activation+0.000
he
Token he
Feature activation+0.000
definitely
Token definitely
Feature activation+0.000
was
Token was
Feature activation+0.000
,
Token,
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
certain
Token certain
Feature activation+0.000
period
Token period
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
first
Token first
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 H3.4: (feature 12435)

TOP ACTIVATIONS
MAX = 2.510

investigating
Token investigating
Feature activation+0.000
its
Token its
Feature activation+0.000
potential
Token potential
Feature activation+0.000
to
Token to
Feature activation+0.000
help
Token help
Feature activation+0.025
animals
Token animals
Feature activation+2.510
.
Token.
Feature activation+0.017
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Colorado
TokenColorado
Feature activation+0.000
State
Token State
Feature activation+0.000
legal
Token legal
Feature activation+0.000
to
Token to
Feature activation+0.000
prescribe
Token prescribe
Feature activation+0.229
it
Token it
Feature activation+0.000
to
Token to
Feature activation+0.000
animals
Token animals
Feature activation+2.235
,
Token,
Feature activation+0.026
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
he
Token he
Feature activation+0.000
told
Token told
Feature activation+0.000
voters
Token voters
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
While
TokenWhile
Feature activation+0.000
Clinton
Token Clinton
Feature activation+1.803
has
Token has
Feature activation+0.000
made
Token made
Feature activation+0.000
no
Token no
Feature activation+0.000
secret
Token secret
Feature activation+0.000
in
Token in
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Introdu
TokenIntrodu
Feature activation+0.000
cing
Tokencing
Feature activation+0.000
tam
Token tam
Feature activation+1.755
per
Tokenper
Feature activation+0.525
-
Token-
Feature activation+0.000
proof
Tokenproof
Feature activation+0.327
ing
Tokening
Feature activation+0.331
was
Token was
Feature activation+0.000
canceled
Token canceled
Feature activation+0.000
the
Token the
Feature activation+0.000
vote
Token vote
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
border
Token border
Feature activation+1.745
security
Token security
Feature activation+0.429
measure
Token measure
Feature activation+0.252
,
Token,
Feature activation+0.000
no
Token no
Feature activation+0.000
doubt
Token doubt
Feature activation+0.000
have
Token have
Feature activation+0.000
strict
Token strict
Feature activation+0.000
rules
Token rules
Feature activation+0.000
about
Token about
Feature activation+0.000
free
Token free
Feature activation+0.000
sexual
Token sexual
Feature activation+1.685
expression
Token expression
Feature activation+0.102
,
Token,
Feature activation+0.000
to
Token to
Feature activation+0.000
protect
Token protect
Feature activation+0.000
your
Token your
Feature activation+0.000
users
Token users
Feature activation+0.471
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
is
Token is
Feature activation+0.000
crystal
Token crystal
Feature activation+0.324
meth
Token meth
Feature activation+1.619
or
Token or
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
ice
Tokenice
Feature activation+0.704
âĢ
TokenâĢ
Feature activation+0.000
House
Token House
Feature activation+0.000
approved
Token approved
Feature activation+0.000
his
Token his
Feature activation+0.000
legislation
Token legislation
Feature activation+0.000
for
Token for
Feature activation+0.000
border
Token border
Feature activation+1.490
security
Token security
Feature activation+0.301
funding
Token funding
Feature activation+0.291
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
seemed
Token seemed
Feature activation+0.000
have
Token have
Feature activation+0.529
formal
Token formal
Feature activation+0.000
affiliation
Token affiliation
Feature activation+0.031
with
Token with
Feature activation+0.020
a
Token a
Feature activation+0.032
hospital
Token hospital
Feature activation+1.435
within
Token within
Feature activation+0.000
30
Token 30
Feature activation+0.000
miles
Token miles
Feature activation+0.000
(
Token (
Feature activation+0.000
48
Token48
Feature activation+0.000
-
Token-
Feature activation+0.000
le
Tokenle
Feature activation+0.000
aked
Tokenaked
Feature activation+0.000
documents
Token documents
Feature activation+0.000
showing
Token showing
Feature activation+0.000
Clinton
Token Clinton
Feature activation+1.433
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
strong
Token strong
Feature activation+0.000
opposition
Token opposition
Feature activation+0.000
.
Token.
Feature activation+0.000
government
Token government
Feature activation+0.000
listed
Token listed
Feature activation+0.000
d
Token d
Feature activation+0.769
iox
Tokeniox
Feature activation+0.921
in
Tokenin
Feature activation+1.420
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
"
Token "
Feature activation+0.000
known
Tokenknown
Feature activation+0.000
human
Token human
Feature activation+0.387
In
TokenIn
Feature activation+0.000
a
Token a
Feature activation+0.000
letter
Token letter
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Alcohol
Token Alcohol
Feature activation+1.391
and
Token and
Feature activation+0.000
Tobacco
Token Tobacco
Feature activation+0.336
Tax
Token Tax
Feature activation+0.000
and
Token and
Feature activation+0.000
Trade
Token Trade
Feature activation+0.000
task
Token task
Feature activation+0.000
force
Tokenforce
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+1.386
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
form
Token form
Feature activation+0.000
of
Token of
Feature activation+0.000
methamphetamine
Token methamphetamine
Feature activation+0.307
To
TokenTo
Feature activation+0.000
see
Token see
Feature activation+0.000
what
Token what
Feature activation+0.000
else
Token else
Feature activation+0.000
Hillary
Token Hillary
Feature activation+1.129
Clinton
Token Clinton
Feature activation+1.375
has
Token has
Feature activation+0.000
said
Token said
Feature activation+0.000
about
Token about
Feature activation+0.000
cannabis
Token cannabis
Feature activation+0.000
law
Token law
Feature activation+0.000
and
Token and
Feature activation+0.000
no
Token no
Feature activation+0.000
clear
Token clear
Feature activation+0.000
association
Token association
Feature activation+0.000
with
Token with
Feature activation+0.000
cancer
Token cancer
Feature activation+1.355
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Australian
Token Australian
Feature activation+0.000
added
Token added
Feature activation+0.000
sense
Token sense
Feature activation+0.000
of
Token of
Feature activation+0.000
urgency
Token urgency
Feature activation+0.000
for
Token for
Feature activation+0.000
Clinton
Token Clinton
Feature activation+1.346
to
Token to
Feature activation+0.000
evolve
Token evolve
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
question
Token question
Feature activation+0.000
,
Token,
Feature activation+0.000
subject
Token subject
Feature activation+0.000
to
Token to
Feature activation+0.000
getting
Token getting
Feature activation+0.000
massive
Token massive
Feature activation+0.154
border
Token border
Feature activation+1.340
controls
Token controls
Feature activation+0.679
.
Token.
Feature activation+0.000
We
Token We
Feature activation+0.000
're
Token're
Feature activation+0.000
working
Token working
Feature activation+0.000
to
Token to
Feature activation+0.000
pass
Token pass
Feature activation+0.000
it
Token it
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
Rs
Token Rs
Feature activation+1.323
then
Token then
Feature activation+0.000
announced
Token announced
Feature activation+0.000
they
Token they
Feature activation+0.000
were
Token were
Feature activation+0.000
postp
Token postp
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
overall
Token overall
Feature activation+0.000
methamphetamine
Token methamphetamine
Feature activation+0.131
use
Token use
Feature activation+1.245
hasn
Token hasn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
increased
Token increased
Feature activation+0.000
Indeed
TokenIndeed
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
use
Token use
Feature activation+0.599
of
Token of
Feature activation+0.000
meth
Token meth
Feature activation+1.193
/
Token/
Feature activation+0.000
amphetamine
Tokenamphetamine
Feature activation+0.679
is
Token is
Feature activation+0.000
lower
Token lower
Feature activation+0.000
when
Token when
Feature activation+0.000

Top DFA by src position
MAX = 1.911

states
Token states
Feature activation-0.004
Top resid features:
allow
Token allow
Feature activation+0.002
Top resid features:
medical
Token medical
Feature activation+0.041
Top resid features:
or
Token or
Feature activation+0.005
Top resid features:
recreational
Token recreational
Feature activation+0.026
Top resid features:
marijuana
Token marijuana
Feature activation+1.090
Top resid features:
use
Token use
Feature activation+0.075
Top resid features:
by
Token by
Feature activation+0.009
Top resid features:
humans
Token humans
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.014
Top resid features:
those
Token those
Feature activation+0.005
Top resid features:
states
Token states
Feature activation-0.005
Top resid features:
allow
Token allow
Feature activation+0.003
Top resid features:
medical
Token medical
Feature activation+0.031
Top resid features:
or
Token or
Feature activation+0.007
Top resid features:
recreational
Token recreational
Feature activation+0.028
Top resid features:
marijuana
Token marijuana
Feature activation+1.343
Top resid features:
use
Token use
Feature activation+0.102
Top resid features:
by
Token by
Feature activation+0.011
Top resid features:
humans
Token humans
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.018
Top resid features:
those
Token those
Feature activation+0.005
Top resid features:
understanding
Token understanding
Feature activation-0.002
Top resid features:
by
Token by
Feature activation+0.002
Top resid features:
political
Token political
Feature activation-0.010
Top resid features:
operatives
Token operatives
Feature activation-0.003
Top resid features:
that
Token that
Feature activation+0.003
Top resid features:
marijuana
Token marijuana
Feature activation+1.753
Top resid features:
law
Token law
Feature activation+0.023
Top resid features:
reform
Token reform
Feature activation+0.007
Top resid features:
is
Token is
Feature activation+0.004
Top resid features:
now
Token now
Feature activation-0.003
Top resid features:
a
Token a
Feature activation+0.002
Top resid features:
began
Token began
Feature activation-0.000
Top resid features:
releasing
Token releasing
Feature activation-0.001
Top resid features:
versions
Token versions
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
their
Token their
Feature activation+0.000
Top resid features:
opioid
Token opioid
Feature activation+0.982
Top resid features:
drugs
Token drugs
Feature activation+0.056
Top resid features:
that
Token that
Feature activation+0.007
Top resid features:
users
Token users
Feature activation+0.007
Top resid features:
could
Token could
Feature activation+0.001
Top resid features:
not
Token not
Feature activation+0.003
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.112
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
bill
Token bill
Feature activation+0.000
Top resid features:
rev
Token rev
Feature activation-0.001
Top resid features:
oking
Tokenoking
Feature activation+0.004
Top resid features:
DACA
Token DACA
Feature activation+1.695
Top resid features:
after
Token after
Feature activation+0.004
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
House
Token House
Feature activation-0.016
Top resid features:
approved
Token approved
Feature activation-0.000
Top resid features:
his
Token his
Feature activation-0.000
Top resid features:
obnoxious
Token obnoxious
Feature activation-0.004
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
allow
Token allow
Feature activation+0.002
Top resid features:
deadly
Token deadly
Feature activation+0.592
Top resid features:
concealed
Token concealed
Feature activation+0.189
Top resid features:
weapons
Token weapons
Feature activation+0.887
Top resid features:
in
Token in
Feature activation+0.015
Top resid features:
classrooms
Token classrooms
Feature activation-0.008
Top resid features:
,
Token,
Feature activation+0.008
Top resid features:
however
Token however
Feature activation+0.004
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
use
Token use
Feature activation+0.009
Top resid features:
of
Token of
Feature activation+0.002
Top resid features:
meth
Token meth
Feature activation+0.045
Top resid features:
/
Token/
Feature activation+0.012
Top resid features:
amphetamine
Tokenamphetamine
Feature activation+0.429
Top resid features:
is
Token is
Feature activation+0.002
Top resid features:
lower
Token lower
Feature activation-0.003
Top resid features:
when
Token when
Feature activation-0.001
Top resid features:
compared
Token compared
Feature activation-0.000
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.180
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
bill
Token bill
Feature activation+0.002
Top resid features:
rev
Token rev
Feature activation+0.002
Top resid features:
oking
Tokenoking
Feature activation+0.011
Top resid features:
DACA
Token DACA
Feature activation+1.911
Top resid features:
after
Token after
Feature activation+0.006
Top resid features:
the
Token the
Feature activation+0.009
Top resid features:
House
Token House
Feature activation-0.006
Top resid features:
approved
Token approved
Feature activation-0.003
Top resid features:
his
Token his
Feature activation+0.021
Top resid features:
,
Token,
Feature activation+0.006
Top resid features:
requiring
Token requiring
Feature activation+0.006
Top resid features:
doctors
Token doctors
Feature activation+0.051
Top resid features:
who
Token who
Feature activation-0.005
Top resid features:
perform
Token perform
Feature activation+0.042
Top resid features:
abortions
Token abortions
Feature activation+0.943
Top resid features:
to
Token to
Feature activation+0.016
Top resid features:
have
Token have
Feature activation+0.024
Top resid features:
formal
Token formal
Feature activation+0.010
Top resid features:
affiliation
Token affiliation
Feature activation+0.023
Top resid features:
with
Token with
Feature activation+0.034
Top resid features:
have
Token have
Feature activation+0.001
Top resid features:
made
Token made
Feature activation+0.000
Top resid features:
support
Token support
Feature activation-0.001
Top resid features:
for
Token for
Feature activation-0.000
Top resid features:
ending
Token ending
Feature activation+0.003
Top resid features:
cannabis
Token cannabis
Feature activation+1.876
Top resid features:
prohibition
Token prohibition
Feature activation+0.097
Top resid features:
center
Token center
Feature activation-0.003
Top resid features:
pieces
Tokenpieces
Feature activation-0.001
Top resid features:
of
Token of
Feature activation+0.002
Top resid features:
their
Token their
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
government
Token government
Feature activation+0.004
Top resid features:
listed
Token listed
Feature activation+0.089
Top resid features:
d
Token d
Feature activation+0.094
Top resid features:
iox
Tokeniox
Feature activation+0.107
Top resid features:
in
Tokenin
Feature activation+0.396
Top resid features:
as
Token as
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
"
Token "
Feature activation+0.000
Top resid features:
known
Tokenknown
Feature activation+0.000
Top resid features:
human
Token human
Feature activation+0.000
Top resid features:
Ŀ
TokenĿ
Feature activation-0.003
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.000
Top resid features:
another
Token another
Feature activation+0.000
Top resid features:
name
Token name
Feature activation-0.001
Top resid features:
for
Token for
Feature activation-0.000
Top resid features:
marijuana
Token marijuana
Feature activation+1.136
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.000
Top resid features:
to
Token to
Feature activation+0.006
Top resid features:
promote
Token promote
Feature activation-0.001
Top resid features:
their
Token their
Feature activation-0.000
Top resid features:
Bud
Token Bud
Feature activation+0.028
Top resid features:
recent
Token recent
Feature activation+0.001
Top resid features:
statistics
Token statistics
Feature activation+0.002
Top resid features:
show
Token show
Feature activation-0.001
Top resid features:
that
Token that
Feature activation+0.003
Top resid features:
overall
Token overall
Feature activation-0.001
Top resid features:
methamphetamine
Token methamphetamine
Feature activation+1.708
Top resid features:
use
Token use
Feature activation+0.131
Top resid features:
hasn
Token hasn
Feature activation+0.001
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation-0.000
Top resid features:
t
Tokent
Feature activation+0.000
Top resid features:
have
Token have
Feature activation+0.001
Top resid features:
made
Token made
Feature activation-0.000
Top resid features:
support
Token support
Feature activation-0.000
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
ending
Token ending
Feature activation+0.002
Top resid features:
cannabis
Token cannabis
Feature activation+0.838
Top resid features:
prohibition
Token prohibition
Feature activation+0.086
Top resid features:
center
Token center
Feature activation-0.003
Top resid features:
pieces
Tokenpieces
Feature activation-0.000
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
their
Token their
Feature activation-0.000
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
Medical
Token Medical
Feature activation-0.000
Top resid features:
Research
Token Research
Feature activation+0.000
Top resid features:
Council
Token Council
Feature activation-0.001
Top resid features:
said
Token said
Feature activation-0.001
Top resid features:
fluoride
Token fluoride
Feature activation+0.976
Top resid features:
at
Token at
Feature activation+0.001
Top resid features:
levels
Token levels
Feature activation+0.004
Top resid features:
for
Token for
Feature activation+0.004
Top resid features:
prevention
Token prevention
Feature activation+0.010
Top resid features:
of
Token of
Feature activation+0.014
Top resid features:
have
Token have
Feature activation+0.000
Top resid features:
made
Token made
Feature activation-0.000
Top resid features:
support
Token support
Feature activation-0.000
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
ending
Token ending
Feature activation+0.003
Top resid features:
cannabis
Token cannabis
Feature activation+1.015
Top resid features:
prohibition
Token prohibition
Feature activation+0.076
Top resid features:
center
Token center
Feature activation-0.003
Top resid features:
pieces
Tokenpieces
Feature activation-0.000
Top resid features:
of
Token of
Feature activation+0.002
Top resid features:
their
Token their
Feature activation-0.000
Top resid features:
-
Token-
Feature activation-0.000
Top resid features:
era
Tokenera
Feature activation-0.002
Top resid features:
program
Token program
Feature activation-0.001
Top resid features:
known
Token known
Feature activation-0.000
Top resid features:
as
Token as
Feature activation+0.001
Top resid features:
DACA
Token DACA
Feature activation+0.420
Top resid features:
,
Token,
Feature activation+0.006
Top resid features:
or
Token or
Feature activation+0.001
Top resid features:
Def
Token Def
Feature activation+0.009
Top resid features:
erred
Tokenerred
Feature activation+0.022
Top resid features:
Action
Token Action
Feature activation+0.008
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.113
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
bill
Token bill
Feature activation+0.002
Top resid features:
rev
Token rev
Feature activation+0.000
Top resid features:
oking
Tokenoking
Feature activation+0.009
Top resid features:
DACA
Token DACA
Feature activation+1.318
Top resid features:
after
Token after
Feature activation+0.008
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
House
Token House
Feature activation-0.007
Top resid features:
approved
Token approved
Feature activation-0.001
Top resid features:
his
Token his
Feature activation-0.002
Top resid features:
recent
Token recent
Feature activation+0.003
Top resid features:
statistics
Token statistics
Feature activation+0.001
Top resid features:
show
Token show
Feature activation+0.009
Top resid features:
that
Token that
Feature activation-0.017
Top resid features:
overall
Token overall
Feature activation-0.073
Top resid features:
methamphetamine
Token methamphetamine
Feature activation+1.684
Top resid features:
use
Token use
Feature activation+0.107
Top resid features:
hasn
Token hasn
Feature activation+0.000
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
t
Tokent
Feature activation+0.000
Top resid features:
2013
Token 2013
Feature activation+0.002
Top resid features:
was
Token was
Feature activation-0.001
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
misuse
Token misuse
Feature activation+0.021
Top resid features:
of
Token of
Feature activation+0.009
Top resid features:
pharmaceutical
Token pharmaceutical
Feature activation+0.611
Top resid features:
s
Tokens
Feature activation+0.014
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
There
Token There
Feature activation-0.001
Top resid features:
were
Token were
Feature activation-0.001
Top resid features:
declines
Token declines
Feature activation-0.068
Top resid features:

Decoder Weights Distribution

Head 0: 0.12

Head 1: 0.03

Head 2: 0.04

Head 3: 0.03

Head 4: 0.44

Head 5: 0.08

Head 6: 0.05

Head 7: 0.04

Head 8: 0.04

Head 9: 0.03

Head 10: 0.06

Head 11: 0.03

Positive logits

IPM1.98

imag1.69

hypothetical1.64

Democr1.61

exclusive1.60

Berman1.55

Cuomo1.52

Letter1.52

MLB1.50

Homeland1.46

Written1.46

Moral1.45

Stern1.44

lawmakers1.44

Schumer1.43

perate1.43

quoted1.41

Senators1.40

.>>1.40

exclus1.39

Negative logits

fleet-2.33

cised-2.18

-2.13

dra-1.97

uyomi-1.91

apons-1.90

essel-1.85

dark-1.85

Trails-1.80

Forge-1.78

chwitz-1.76

rail-1.76

achu-1.74

kai-1.74

selage-1.72

ghost-1.70

matter-1.70

roma-1.69

glim-1.68

coh-1.68

INTERVAL 2.259 - 2.510
CONTAINS 0.000%

investigating
Token investigating
Feature activation+0.000
its
Token its
Feature activation+0.000
potential
Token potential
Feature activation+0.000
to
Token to
Feature activation+0.000
help
Token help
Feature activation+0.025
animals
Token animals
Feature activation+2.510
.
Token.
Feature activation+0.017
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Colorado
TokenColorado
Feature activation+0.000
State
Token State
Feature activation+0.000

INTERVAL 2.008 - 2.259
CONTAINS 0.000%

legal
Token legal
Feature activation+0.000
to
Token to
Feature activation+0.000
prescribe
Token prescribe
Feature activation+0.229
it
Token it
Feature activation+0.000
to
Token to
Feature activation+0.000
animals
Token animals
Feature activation+2.235
,
Token,
Feature activation+0.026
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
he
Token he
Feature activation+0.000
told
Token told
Feature activation+0.000

INTERVAL 1.757 - 2.008
CONTAINS 0.000%

voters
Token voters
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
While
TokenWhile
Feature activation+0.000
Clinton
Token Clinton
Feature activation+1.803
has
Token has
Feature activation+0.000
made
Token made
Feature activation+0.000
no
Token no
Feature activation+0.000
secret
Token secret
Feature activation+0.000
in
Token in
Feature activation+0.000

INTERVAL 1.506 - 1.757
CONTAINS 0.000%

.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Introdu
TokenIntrodu
Feature activation+0.000
cing
Tokencing
Feature activation+0.000
tam
Token tam
Feature activation+1.755
per
Tokenper
Feature activation+0.525
-
Token-
Feature activation+0.000
proof
Tokenproof
Feature activation+0.327
ing
Tokening
Feature activation+0.331
was
Token was
Feature activation+0.000
users
Token users
Feature activation+0.471
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
is
Token is
Feature activation+0.000
crystal
Token crystal
Feature activation+0.324
meth
Token meth
Feature activation+1.619
or
Token or
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
ice
Tokenice
Feature activation+0.704
âĢ
TokenâĢ
Feature activation+0.000
canceled
Token canceled
Feature activation+0.000
the
Token the
Feature activation+0.000
vote
Token vote
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
border
Token border
Feature activation+1.745
security
Token security
Feature activation+0.429
measure
Token measure
Feature activation+0.252
,
Token,
Feature activation+0.000
no
Token no
Feature activation+0.000
doubt
Token doubt
Feature activation+0.000
have
Token have
Feature activation+0.000
strict
Token strict
Feature activation+0.000
rules
Token rules
Feature activation+0.000
about
Token about
Feature activation+0.000
free
Token free
Feature activation+0.000
sexual
Token sexual
Feature activation+1.685
expression
Token expression
Feature activation+0.102
,
Token,
Feature activation+0.000
to
Token to
Feature activation+0.000
protect
Token protect
Feature activation+0.000
your
Token your
Feature activation+0.000

INTERVAL 1.255 - 1.506
CONTAINS 0.001%

In
TokenIn
Feature activation+0.000
a
Token a
Feature activation+0.000
letter
Token letter
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Alcohol
Token Alcohol
Feature activation+1.391
and
Token and
Feature activation+0.000
Tobacco
Token Tobacco
Feature activation+0.336
Tax
Token Tax
Feature activation+0.000
and
Token and
Feature activation+0.000
Trade
Token Trade
Feature activation+0.000
and
Token and
Feature activation+0.000
no
Token no
Feature activation+0.000
clear
Token clear
Feature activation+0.000
association
Token association
Feature activation+0.000
with
Token with
Feature activation+0.000
cancer
Token cancer
Feature activation+1.355
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Australian
Token Australian
Feature activation+0.000
,
Token,
Feature activation+0.000
subject
Token subject
Feature activation+0.000
to
Token to
Feature activation+0.000
getting
Token getting
Feature activation+0.000
massive
Token massive
Feature activation+0.154
border
Token border
Feature activation+1.340
controls
Token controls
Feature activation+0.679
.
Token.
Feature activation+0.000
We
Token We
Feature activation+0.000
're
Token're
Feature activation+0.000
working
Token working
Feature activation+0.000
.
Token.
Feature activation+0.000
government
Token government
Feature activation+0.000
listed
Token listed
Feature activation+0.000
d
Token d
Feature activation+0.769
iox
Tokeniox
Feature activation+0.921
in
Tokenin
Feature activation+1.420
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
"
Token "
Feature activation+0.000
known
Tokenknown
Feature activation+0.000
human
Token human
Feature activation+0.387
added
Token added
Feature activation+0.000
sense
Token sense
Feature activation+0.000
of
Token of
Feature activation+0.000
urgency
Token urgency
Feature activation+0.000
for
Token for
Feature activation+0.000
Clinton
Token Clinton
Feature activation+1.346
to
Token to
Feature activation+0.000
evolve
Token evolve
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
question
Token question
Feature activation+0.000

INTERVAL 1.004 - 1.255
CONTAINS 0.002%

levels
Token levels
Feature activation+0.225
,
Token,
Feature activation+0.000
in
Token in
Feature activation+0.000
March
Token March
Feature activation+0.000
.
Token.
Feature activation+0.000
Rivers
Token Rivers
Feature activation+1.099
said
Token said
Feature activation+0.000
Tr
Token Tr
Feature activation+0.109
ul
Tokenul
Feature activation+0.000
ieve
Tokenieve
Feature activation+0.041
expects
Token expects
Feature activation+0.000
Libyan
Token Libyan
Feature activation+0.000
chemical
Token chemical
Feature activation+0.000
,
Token,
Feature activation+0.000
biological
Token biological
Feature activation+0.271
and
Token and
Feature activation+0.000
nuclear
Token nuclear
Feature activation+1.122
scientists
Token scientists
Feature activation+0.144
from
Token from
Feature activation+0.000
working
Token working
Feature activation+0.000
for
Token for
Feature activation+0.000
terror
Token terror
Feature activation+0.226
most
Token most
Feature activation+0.000
comprehensive
Token comprehensive
Feature activation+0.000
surveys
Token surveys
Feature activation+0.000
of
Token of
Feature activation+0.000
drug
Token drug
Feature activation+0.660
use
Token use
Feature activation+1.046
in
Token in
Feature activation+0.000
Australia
Token Australia
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
National
Token National
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
overall
Token overall
Feature activation+0.000
methamphetamine
Token methamphetamine
Feature activation+0.131
use
Token use
Feature activation+1.245
hasn
Token hasn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
increased
Token increased
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
College
Token College
Feature activation+0.000
of
Token of
Feature activation+0.000
Veterinary
Token Veterinary
Feature activation+1.090
Medicine
Token Medicine
Feature activation+0.000
&
Token &
Feature activation+0.000
Bi
Token Bi
Feature activation+0.070
omedical
Tokenomedical
Feature activation+0.000
Sciences
Token Sciences
Feature activation+0.000

INTERVAL 0.753 - 1.004
CONTAINS 0.007%

graffiti
Token graffiti
Feature activation+0.305
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
boys
Token boys
Feature activation+0.000
'
Token'
Feature activation+0.000
bathrooms
Token bathrooms
Feature activation+0.770
and
Token and
Feature activation+0.000
possession
Token possession
Feature activation+0.000
and
Token and
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
that
Token that
Feature activation+0.000
have
Token have
Feature activation+0.000
been
Token been
Feature activation+0.000
linked
Token linked
Feature activation+0.125
to
Token to
Feature activation+0.000
heart
Token heart
Feature activation+0.937
and
Token and
Feature activation+0.000
liver
Token liver
Feature activation+0.164
disease
Token disease
Feature activation+0.549
,
Token,
Feature activation+0.000
human
Token human
Feature activation+0.210
by
Token by
Feature activation+0.000
banning
Token banning
Feature activation+0.000
F
Token F
Feature activation+0.244
GM
TokenGM
Feature activation+0.195
/
Token/
Feature activation+0.457
C
TokenC
Feature activation+0.961
.
Token.
Feature activation+0.000
You
Token You
Feature activation+0.000
can
Token can
Feature activation+0.000
support
Token support
Feature activation+0.000
us
Token us
Feature activation+0.000
-
Token-
Feature activation+0.000
old
Tokenold
Feature activation+0.000
girl
Token girl
Feature activation+0.000
,
Token,
Feature activation+0.000
Donald
Token Donald
Feature activation+0.291
Trump
Token Trump
Feature activation+0.776
called
Token called
Feature activation+0.000
on
Token on
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.016
to
Token to
Feature activation+0.000
return
Token return
Feature activation+0.000
the
Token the
Feature activation+0.000
well
Token well
Feature activation+0.221
-
Token-
Feature activation+0.000
being
Tokenbeing
Feature activation+0.000
of
Token of
Feature activation+0.000
young
Token young
Feature activation+0.908
boys
Token boys
Feature activation+0.492
under
Token under
Feature activation+0.062
his
Token his
Feature activation+0.000
charge
Token charge
Feature activation+0.000
in
Token in
Feature activation+0.000

INTERVAL 0.502 - 0.753
CONTAINS 0.017%

of
Token of
Feature activation+0.000
ending
Token ending
Feature activation+0.030
prohibition
Token prohibition
Feature activation+0.000
prior
Token prior
Feature activation+0.043
to
Token to
Feature activation+0.000
Election
Token Election
Feature activation+0.666
Day
Token Day
Feature activation+0.026
.
Token.
Feature activation+0.007
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
shark
Token shark
Feature activation+0.000
is
Token is
Feature activation+0.000
detected
Token detected
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
drones
Token drones
Feature activation+0.526
will
Token will
Feature activation+0.000
alert
Token alert
Feature activation+0.000
sw
Token sw
Feature activation+0.000
immers
Tokenimmers
Feature activation+0.095
through
Token through
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Indeed
TokenIndeed
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
use
Token use
Feature activation+0.599
of
Token of
Feature activation+0.000
meth
Token meth
Feature activation+1.193
/
Token/
Feature activation+0.000
amphetamine
Tokenamphetamine
Feature activation+0.679
is
Token is
Feature activation+0.000
According
TokenAccording
Feature activation+0.000
to
Token to
Feature activation+0.000
Reuters
Token Reuters
Feature activation+0.000
,
Token,
Feature activation+0.000
capt
Token capt
Feature activation+0.384
agon
Tokenagon
Feature activation+0.730
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
gener
Tokengener
Feature activation+0.000
ates
Tokenates
Feature activation+0.000
hundreds
Token hundreds
Feature activation+0.000
to
Token to
Feature activation+0.000
start
Token start
Feature activation+0.000
searching
Token searching
Feature activation+0.000
for
Token for
Feature activation+0.000
new
Token new
Feature activation+0.000
compounds
Token compounds
Feature activation+0.599
again
Token again
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Te
TokenTe
Feature activation+0.000

INTERVAL 0.251 - 0.502
CONTAINS 0.063%

two
Token two
Feature activation+0.000
specifications
Token specifications
Feature activation+0.000
of
Token of
Feature activation+0.000
attempted
Token attempted
Feature activation+0.000
sexual
Token sexual
Feature activation+0.647
abuse
Token abuse
Feature activation+0.330
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
child
Token child
Feature activation+0.006
and
Token and
Feature activation+0.000
one
Token one
Feature activation+0.000
able
Token able
Feature activation+0.045
to
Token to
Feature activation+0.000
sell
Token sell
Feature activation+0.000
marijuana
Token marijuana
Feature activation+0.505
with
Token with
Feature activation+0.004
high
Token high
Feature activation+0.262
-
Token-
Feature activation+0.268
TH
TokenTH
Feature activation+0.814
C
TokenC
Feature activation+0.617
levels
Token levels
Feature activation+0.670
as
Token as
Feature activation+0.000
.
Token.
Feature activation+0.000
Among
Token Among
Feature activation+0.000
those
Token those
Feature activation+0.000
who
Token who
Feature activation+0.000
reported
Token reported
Feature activation+0.000
getting
Token getting
Feature activation+0.388
the
Token the
Feature activation+0.000
recommended
Token recommended
Feature activation+0.130
amounts
Token amounts
Feature activation+0.034
of
Token of
Feature activation+0.000
activity
Token activity
Feature activation+0.000
by
Token by
Feature activation+0.000
a
Token a
Feature activation+0.000
mix
Token mix
Feature activation+0.000
of
Token of
Feature activation+0.000
extreme
Token extreme
Feature activation+0.000
poverty
Token poverty
Feature activation+0.371
,
Token,
Feature activation+0.000
gangs
Token gangs
Feature activation+0.000
and
Token and
Feature activation+0.000
drug
Token drug
Feature activation+0.000
violence
Token violence
Feature activation+0.000
analyzing
Token analyzing
Feature activation+0.000
aerial
Token aerial
Feature activation+0.213
imagery
Token imagery
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
drone
Token drone
Feature activation+0.439
-
Token-
Feature activation+0.000
based
Tokenbased
Feature activation+0.000
system
Token system
Feature activation+0.000
can
Token can
Feature activation+0.000
detect
Token detect
Feature activation+0.000

INTERVAL 0.000 - 0.251
CONTAINS 99.909%

.
Token.
Feature activation+0.000
You
Token You
Feature activation+0.000
're
Token're
Feature activation+0.000
in
Token in
Feature activation+0.000
there
Token there
Feature activation+0.000
eating
Token eating
Feature activation+0.000
with
Token with
Feature activation+0.000
your
Token your
Feature activation+0.000
team
Token team
Feature activation+0.000
and
Token and
Feature activation+0.000
then
Token then
Feature activation+0.000
Ohio
Token Ohio
Feature activation+0.000
ans
Tokenans
Feature activation+0.000
to
Token to
Feature activation+0.000
End
Token End
Feature activation+0.000
Prohibition
Token Prohibition
Feature activation+0.000
,
Token,
Feature activation+0.000
said
Token said
Feature activation+0.000
P
Token P
Feature activation+0.000
ard
Tokenard
Feature activation+0.000
ee
Tokenee
Feature activation+0.000
must
Token must
Feature activation+0.000
despite
Token despite
Feature activation+0.000
wishing
Token wishing
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
left
Token left
Feature activation+0.000
alone
Token alone
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
accum
Token accum
Feature activation+0.000
ulates
Tokenulates
Feature activation+0.000
a
Token a
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
That
TokenThat
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
all
Token all
Feature activation+0.000
concerning
Token concerning
Feature activation+0.000
PAR
Token PAR
Feature activation+0.000
PG
TokenPG
Feature activation+0.000
for
Token for
Feature activation+0.000
today
Token today
Feature activation+0.000
Test
Token Test
Feature activation+0.000
scores
Token scores
Feature activation+0.000
and
Token and
Feature activation+0.000
undergraduate
Token undergraduate
Feature activation+0.000
Grade
Token Grade
Feature activation+0.000
Point
Token Point
Feature activation+0.000
A
Token A
Feature activation+0.000
verages
Tokenverages
Feature activation+0.000
are
Token are
Feature activation+0.000
the
Token the
Feature activation+0.000
measures
Token measures
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 H3.4: (feature 12213)

TOP ACTIVATIONS
MAX = 3.900

.
Token.
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
area
Token area
Feature activation+0.000
of
Token of
Feature activation+0.000
Camden
Token Camden
Feature activation+3.900
and
Token and
Feature activation+0.000
Union
Token Union
Feature activation+0.416
avenues
Token avenues
Feature activation+0.223
,
Token,
Feature activation+0.000
where
Token where
Feature activation+0.000
is
Token is
Feature activation+0.000
safe
Token safe
Feature activation+0.000
at
Token at
Feature activation+0.000
home
Token home
Feature activation+0.000
in
Token in
Feature activation+0.000
New
Token New
Feature activation+3.092
Jersey
Token Jersey
Feature activation+2.738
with
Token with
Feature activation+0.000
his
Token his
Feature activation+0.000
family
Token family
Feature activation+0.000
,
Token,
Feature activation+0.000
week
Token week
Feature activation+0.000
days
Tokendays
Feature activation+0.000
working
Token working
Feature activation+0.000
in
Token in
Feature activation+0.000
Johannes
Token Johannes
Feature activation+1.311
burg
Tokenburg
Feature activation+2.989
before
Token before
Feature activation+0.000
retreating
Token retreating
Feature activation+0.000
to
Token to
Feature activation+0.000
his
Token his
Feature activation+0.000
beloved
Token beloved
Feature activation+0.000
the
Token the
Feature activation+0.000
attorney
Token attorney
Feature activation+0.000
for
Token for
Feature activation+0.000
itiner
Token itiner
Feature activation+0.000
ant
Tokenant
Feature activation+0.000
Portland
Token Portland
Feature activation+2.984
musician
Token musician
Feature activation+0.000
Brandon
Token Brandon
Feature activation+0.541
Carpenter
Token Carpenter
Feature activation+0.267
,
Token,
Feature activation+0.000
said
Token said
Feature activation+0.000
compared
Token compared
Feature activation+0.000
Calgary
Token Calgary
Feature activation+0.000
to
Token to
Feature activation+0.000
places
Token places
Feature activation+0.000
like
Token like
Feature activation+0.000
Paris
Token Paris
Feature activation+2.917
,
Token,
Feature activation+0.000
Lyon
Token Lyon
Feature activation+1.595
,
Token,
Feature activation+0.000
Frankfurt
Token Frankfurt
Feature activation+1.832
,
Token,
Feature activation+0.000
unemployment
Token unemployment
Feature activation+0.000
--
Token --
Feature activation+0.000
in
Token in
Feature activation+0.000
places
Token places
Feature activation+0.000
like
Token like
Feature activation+0.000
Cincinnati
Token Cincinnati
Feature activation+2.911
,
Token,
Feature activation+0.000
Newark
Token Newark
Feature activation+2.059
(
Token (
Feature activation+0.000
N
TokenN
Feature activation+0.661
.
Token.
Feature activation+0.000
specialist
Token specialist
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
U
Token U
Feature activation+0.609
.
Token.
Feature activation+1.984
S
TokenS
Feature activation+2.855
.,
Token.,
Feature activation+0.000
and
Token and
Feature activation+0.000
she
Token she
Feature activation+0.000
'll
Token'll
Feature activation+0.000
soon
Token soon
Feature activation+0.000
shepherd
Token shepherd
Feature activation+0.000
when
Token when
Feature activation+0.000
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
at
Token at
Feature activation+0.000
Georgia
Token Georgia
Feature activation+2.824
Tech
Token Tech
Feature activation+0.000
in
Token in
Feature activation+0.000
Atlanta
Token Atlanta
Feature activation+2.093
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
newly
Token newly
Feature activation+0.000
appointed
Token appointed
Feature activation+0.000
head
Token head
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Atlanta
Token Atlanta
Feature activation+2.812
Federal
Token Federal
Feature activation+0.000
Reserve
Token Reserve
Feature activation+0.000
.
Token.
Feature activation+0.000
B
Token B
Feature activation+0.000
ostic
Tokenostic
Feature activation+0.000
backdrop
Token backdrop
Feature activation+0.000
of
Token of
Feature activation+0.000
John
Token John
Feature activation+0.036
stown
Tokenstown
Feature activation+0.207
,
Token,
Feature activation+0.000
Ont
Token Ont
Feature activation+2.802
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
group
Token group
Feature activation+0.000
of
Token of
Feature activation+0.000
Colombian
Token Colombian
Feature activation+1.199
safe
Token safe
Feature activation+0.000
at
Token at
Feature activation+0.000
home
Token home
Feature activation+0.000
in
Token in
Feature activation+0.000
New
Token New
Feature activation+3.092
Jersey
Token Jersey
Feature activation+2.738
with
Token with
Feature activation+0.000
his
Token his
Feature activation+0.000
family
Token family
Feature activation+0.000
,
Token,
Feature activation+0.000
his
Token his
Feature activation+0.000
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
Hawai
Token Hawai
Feature activation+2.698
âĢ
TokenâĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
i
Tokeni
Feature activation+1.421
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
An
Token An
Feature activation+0.000
extreme
Token extreme
Feature activation+0.000
example
Token example
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
London
Token London
Feature activation+2.624
metropolitan
Token metropolitan
Feature activation+1.542
area
Token area
Feature activation+0.378
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
vast
Token vast
Feature activation+0.000
live
Token live
Feature activation+0.000
-
Token -
Feature activation+0.000
Paris
Token Paris
Feature activation+1.926
,
Token,
Feature activation+0.000
New
Token New
Feature activation+0.943
York
Token York
Feature activation+2.476
,
Token,
Feature activation+0.000
Chicago
Token Chicago
Feature activation+0.235
,
Token,
Feature activation+0.000
Montreal
Token Montreal
Feature activation+0.000
or
Token or
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
woman
Token woman
Feature activation+0.000
in
Token in
Feature activation+0.000
New
Token New
Feature activation+1.937
York
Token York
Feature activation+2.407
City
Token City
Feature activation+0.162
recorded
Token recorded
Feature activation+0.000
the
Token the
Feature activation+0.000
man
Token man
Feature activation+0.000
she
Token she
Feature activation+0.000
nearly
Token nearly
Feature activation+0.000
80
Token 80
Feature activation+0.000
percent
Token percent
Feature activation+0.000
;
Token;
Feature activation+0.000
in
Token in
Feature activation+0.000
Chicago
Token Chicago
Feature activation+2.396
,
Token,
Feature activation+0.000
almost
Token almost
Feature activation+0.000
65
Token 65
Feature activation+0.009
percent
Token percent
Feature activation+0.000
.
Token.
Feature activation+0.000
18
Token 18
Feature activation+0.000
-
Token-
Feature activation+0.000
DEC
TokenDEC
Feature activation+0.000
-
Token-
Feature activation+0.000
2015
Token2015
Feature activation+0.000
London
Token London
Feature activation+2.349
,
Token,
Feature activation+0.000
United
Token United
Feature activation+0.491
Kingdom
Token Kingdom
Feature activation+1.290
Is
Token Is
Feature activation+0.000
lington
Tokenlington
Feature activation+0.652
Ċ
TokenĊ
Feature activation+0.000
April
TokenApril
Feature activation+0.000
29
Token 29
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
New
Token New
Feature activation+0.748
York
Token York
Feature activation+2.348
,
Token,
Feature activation+0.000
N
Token N
Feature activation+0.000
.
Token.
Feature activation+0.000
Y
TokenY
Feature activation+0.012
.,
Token.,
Feature activation+0.000
.),
Token.),
Feature activation+0.000
Kee
Token Kee
Feature activation+0.030
gan
Tokengan
Feature activation+0.000
Kelly
Token Kelly
Feature activation+0.000
(
Token (
Feature activation+0.000
Baltimore
TokenBaltimore
Feature activation+2.240
Celtic
Token Celtic
Feature activation+0.376
;
Token;
Feature activation+0.000
Baltimore
Token Baltimore
Feature activation+1.669
,
Token,
Feature activation+0.000
Md
Token Md
Feature activation+2.224
Also
Token Also
Feature activation+0.000
,
Token,
Feature activation+0.000
Tuesday
Token Tuesday
Feature activation+0.000
morning
Token morning
Feature activation+0.000
a
Token a
Feature activation+0.000
Jacksonville
Token Jacksonville
Feature activation+2.238
police
Token police
Feature activation+0.000
officer
Token officer
Feature activation+0.000
was
Token was
Feature activation+0.000
shot
Token shot
Feature activation+0.000
at
Token at
Feature activation+0.000

Top DFA by src position
MAX = 3.608

Drive
Token Drive
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.006
Top resid features:
in
Token in
Feature activation-0.002
Top resid features:
South
Token South
Feature activation+0.007
Top resid features:
San
Token San
Feature activation+0.240
Top resid features:
Jose
Token Jose
Feature activation+2.674
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
Two
TokenTwo
Feature activation-0.000
Top resid features:
officers
Token officers
Feature activation-0.001
Top resid features:
"
Token "
Feature activation-0.000
Top resid features:
felt
Tokenfelt
Feature activation-0.000
Top resid features:
treatment
Token treatment
Feature activation-0.001
Top resid features:
and
Token and
Feature activation+0.004
Top resid features:
education
Token education
Feature activation+0.001
Top resid features:
center
Token center
Feature activation-0.005
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
Utah
Token Utah
Feature activation+3.608
Top resid features:
is
Token is
Feature activation-0.006
Top resid features:
safe
Token safe
Feature activation-0.001
Top resid features:
at
Token at
Feature activation-0.014
Top resid features:
home
Token home
Feature activation+0.042
Top resid features:
in
Token in
Feature activation-0.089
Top resid features:
in
Token in
Feature activation-0.002
Top resid features:
South
Token South
Feature activation+0.004
Top resid features:
gate
Tokengate
Feature activation+0.028
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
north
Token north
Feature activation+0.007
Top resid features:
London
Token London
Feature activation+3.182
Top resid features:
and
Token and
Feature activation+0.005
Top resid features:
found
Token found
Feature activation-0.000
Top resid features:
his
Token his
Feature activation-0.001
Top resid features:
dream
Token dream
Feature activation+0.001
Top resid features:
on
Token on
Feature activation-0.002
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
Sul
Token Sul
Feature activation+0.013
Top resid features:
ph
Tokenph
Feature activation+0.002
Top resid features:
ur
Tokenur
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.008
Top resid features:
Louisiana
Token Louisiana
Feature activation+3.009
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
according
Token according
Feature activation-0.000
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
Holly
Token Holly
Feature activation+0.007
Top resid features:
Carter
Token Carter
Feature activation+0.014
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.011
Top resid features:
In
TokenIn
Feature activation-0.001
Top resid features:
May
Token May
Feature activation+0.003
Top resid features:
Tourism
Token Tourism
Feature activation+0.003
Top resid features:
Calgary
Token Calgary
Feature activation+2.086
Top resid features:
hosted
Token hosted
Feature activation+0.002
Top resid features:
their
Token their
Feature activation-0.001
Top resid features:
annual
Token annual
Feature activation+0.001
Top resid features:
"
Token "
Feature activation-0.003
Top resid features:
White
TokenWhite
Feature activation+0.003
Top resid features:
Some
TokenSome
Feature activation-0.000
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
violence
Token violence
Feature activation-0.000
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
Nebraska
Token Nebraska
Feature activation+1.341
Top resid features:
is
Token is
Feature activation-0.002
Top resid features:
found
Token found
Feature activation-0.000
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
predominantly
Token predominantly
Feature activation+0.001
Top resid features:
f
Tokenf
Feature activation+0.000
Top resid features:
ou
Tokenou
Feature activation-0.003
Top resid features:
z
Tokenz
Feature activation-0.000
Top resid features:
told
Token told
Feature activation+0.000
Top resid features:
CBC
Token CBC
Feature activation+0.416
Top resid features:
Toronto
Token Toronto
Feature activation+2.997
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
Like
TokenLike
Feature activation-0.002
Top resid features:
Bin
Token Bin
Feature activation+0.000
Top resid features:
free
Token free
Feature activation-0.001
Top resid features:
digital
Token digital
Feature activation-0.002
Top resid features:
access
Token access
Feature activation-0.001
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
The
Token The
Feature activation+0.002
Top resid features:
Sacramento
Token Sacramento
Feature activation+3.164
Top resid features:
Bee
Token Bee
Feature activation+0.027
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Last
TokenLast
Feature activation+0.000
Top resid features:
week
Token week
Feature activation-0.000
Top resid features:
ent
Tokenent
Feature activation-0.000
Top resid features:
from
Token from
Feature activation+0.000
Top resid features:
entering
Token entering
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
Los
Token Los
Feature activation+0.329
Top resid features:
Angeles
Token Angeles
Feature activation+2.396
Top resid features:
job
Token job
Feature activation-0.001
Top resid features:
market
Token market
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
and
Token and
Feature activation-0.001
Top resid features:
leading
Token leading
Feature activation-0.000
Top resid features:
Lake
Token Lake
Feature activation+0.002
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
state
Token state
Feature activation+0.012
Top resid features:
of
Token of
Feature activation+0.005
Top resid features:
Oregon
Token Oregon
Feature activation+3.227
Top resid features:
to
Token to
Feature activation-0.000
Top resid features:
illustrate
Token illustrate
Feature activation+0.001
Top resid features:
its
Token its
Feature activation-0.001
Top resid features:
commitment
Token commitment
Feature activation+0.001
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
treatment
Token treatment
Feature activation-0.001
Top resid features:
and
Token and
Feature activation+0.007
Top resid features:
education
Token education
Feature activation+0.000
Top resid features:
center
Token center
Feature activation-0.008
Top resid features:
in
Token in
Feature activation-0.003
Top resid features:
Utah
Token Utah
Feature activation+3.342
Top resid features:
is
Token is
Feature activation-0.011
Top resid features:
safe
Token safe
Feature activation-0.002
Top resid features:
at
Token at
Feature activation-0.005
Top resid features:
home
Token home
Feature activation+0.040
Top resid features:
in
Token in
Feature activation-0.095
Top resid features:
nearly
Token nearly
Feature activation+0.001
Top resid features:
as
Token as
Feature activation+0.000
Top resid features:
winding
Token winding
Feature activation+0.000
Top resid features:
as
Token as
Feature activation+0.000
Top resid features:
San
Token San
Feature activation+0.104
Top resid features:
Francisco
Token Francisco
Feature activation+2.876
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
s
Tokens
Feature activation-0.000
Top resid features:
famous
Token famous
Feature activation-0.000
Top resid features:
Lomb
Token Lomb
Feature activation+0.117
Top resid features:
train
Token train
Feature activation-0.001
Top resid features:
currently
Token currently
Feature activation+0.002
Top resid features:
under
Token under
Feature activation-0.000
Top resid features:
construction
Token construction
Feature activation+0.000
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
California
Token California
Feature activation+0.589
Top resid features:
is
Token is
Feature activation-0.001
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
connect
Token connect
Feature activation-0.000
Top resid features:
low
Token low
Feature activation-0.000
Top resid features:
wage
Token wage
Feature activation-0.001
Top resid features:
have
Token have
Feature activation-0.000
Top resid features:
perhaps
Token perhaps
Feature activation+0.000
Top resid features:
focused
Token focused
Feature activation+0.000
Top resid features:
more
Token more
Feature activation-0.000
Top resid features:
on
Token on
Feature activation-0.000
Top resid features:
Calgary
Token Calgary
Feature activation+3.128
Top resid features:
from
Token from
Feature activation-0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
tourist
Token tourist
Feature activation+0.001
Top resid features:
perspective
Token perspective
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
This
Token This
Feature activation-0.001
Top resid features:
summer
Token summer
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
a
Token a
Feature activation-0.001
Top resid features:
Minneapolis
Token Minneapolis
Feature activation+2.758
Top resid features:
woman
Token woman
Feature activation-0.004
Top resid features:
began
Token began
Feature activation-0.002
Top resid features:
recording
Token recording
Feature activation-0.002
Top resid features:
and
Token and
Feature activation+0.006
Top resid features:
uploading
Token uploading
Feature activation+0.000
Top resid features:
than
Token than
Feature activation-0.003
Top resid features:
176
Token 176
Feature activation+0.006
Top resid features:
percent
Token percent
Feature activation-0.009
Top resid features:
.
Token.
Feature activation-0.004
Top resid features:
In
Token In
Feature activation-0.001
Top resid features:
Oklahoma
Token Oklahoma
Feature activation+2.682
Top resid features:
City
Token City
Feature activation+0.126
Top resid features:
,
Token,
Feature activation+0.017
Top resid features:
prices
Token prices
Feature activation-0.009
Top resid features:
are
Token are
Feature activation-0.003
Top resid features:
up
Token up
Feature activation-0.019
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.036
Top resid features:
2015
Token2015
Feature activation-0.002
Top resid features:
Sacramento
Token Sacramento
Feature activation+1.165
Top resid features:
,
Token,
Feature activation+0.019
Top resid features:
California
Token California
Feature activation+0.629
Top resid features:
Ace
Token Ace
Feature activation+0.002
Top resid features:
of
Token of
Feature activation+0.014
Top resid features:
Sp
Token Sp
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.007
Top resid features:
April
TokenApril
Feature activation+0.003
Top resid features:
19
Token 19
Feature activation-0.002
Top resid features:
âĢĵ
Token âĢĵ
Feature activation-0.008
Top resid features:
Nashville
Token Nashville
Feature activation+1.919
Top resid features:
,
Token,
Feature activation+0.014
Top resid features:
Tenn
Token Tenn
Feature activation+0.902
Top resid features:
.,
Token.,
Feature activation+0.034
Top resid features:
Ry
Token Ry
Feature activation+0.001
Top resid features:
man
Tokenman
Feature activation+0.001
Top resid features:
;
Token;
Feature activation+0.002
Top resid features:
Pal
Token Pal
Feature activation+0.011
Top resid features:
md
Tokenmd
Feature activation+0.003
Top resid features:
ale
Tokenale
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.009
Top resid features:
Calif
Token Calif
Feature activation+0.716
Top resid features:
.),
Token.),
Feature activation-0.003
Top resid features:
Luc
Token Luc
Feature activation+0.000
Top resid features:
a
Tokena
Feature activation+0.001
Top resid features:
de
Token de
Feature activation+0.004
Top resid features:
la
Token la
Feature activation+0.008
Top resid features:
ambush
Token ambush
Feature activation-0.003
Top resid features:
deaths
Token deaths
Feature activation+0.001
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
two
Token two
Feature activation+0.000
Top resid features:
New
Token New
Feature activation+0.044
Top resid features:
York
Token York
Feature activation+2.493
Top resid features:
Police
Token Police
Feature activation-0.001
Top resid features:
Department
Token Department
Feature activation-0.000
Top resid features:
officers
Token officers
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.11

Head 1: 0.03

Head 2: 0.03

Head 3: 0.02

Head 4: 0.44

Head 5: 0.03

Head 6: 0.09

Head 7: 0.04

Head 8: 0.03

Head 9: 0.03

Head 10: 0.09

Head 11: 0.04

Positive logits

asso2.18

bol2.03

rious1.91

CLASSIFIED1.87

iazep1.86

VIDIA1.82

eson1.80

eval1.78

SCP1.75

rett1.75

acers1.75

enne1.74

avascript1.74

imus1.73

ilateral1.72

kies1.72

Rex1.71

iqu1.71

hement1.69

gee1.69

Negative logits

merce-1.86

conduc-1.83

Downloadha-1.74

ageing-1.73

cabinet-1.67

judiciary-1.64

MLA-1.63

backlog-1.63

estate-1.62

UD-1.61

commission-1.60

municipality-1.55

unemployment-1.55

SOFTWARE-1.54

mourning-1.53

-1.53

CAM-1.53

deduction-1.52

PD-1.51

Metropolitan-1.50

INTERVAL 3.510 - 3.900
CONTAINS 0.000%

.
Token.
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
area
Token area
Feature activation+0.000
of
Token of
Feature activation+0.000
Camden
Token Camden
Feature activation+3.900
and
Token and
Feature activation+0.000
Union
Token Union
Feature activation+0.416
avenues
Token avenues
Feature activation+0.223
,
Token,
Feature activation+0.000
where
Token where
Feature activation+0.000

INTERVAL 3.120 - 3.510
CONTAINS 0.000%

INTERVAL 2.730 - 3.120
CONTAINS 0.001%

week
Token week
Feature activation+0.000
days
Tokendays
Feature activation+0.000
working
Token working
Feature activation+0.000
in
Token in
Feature activation+0.000
Johannes
Token Johannes
Feature activation+1.311
burg
Tokenburg
Feature activation+2.989
before
Token before
Feature activation+0.000
retreating
Token retreating
Feature activation+0.000
to
Token to
Feature activation+0.000
his
Token his
Feature activation+0.000
beloved
Token beloved
Feature activation+0.000
safe
Token safe
Feature activation+0.000
at
Token at
Feature activation+0.000
home
Token home
Feature activation+0.000
in
Token in
Feature activation+0.000
New
Token New
Feature activation+3.092
Jersey
Token Jersey
Feature activation+2.738
with
Token with
Feature activation+0.000
his
Token his
Feature activation+0.000
family
Token family
Feature activation+0.000
,
Token,
Feature activation+0.000
his
Token his
Feature activation+0.000
unemployment
Token unemployment
Feature activation+0.000
--
Token --
Feature activation+0.000
in
Token in
Feature activation+0.000
places
Token places
Feature activation+0.000
like
Token like
Feature activation+0.000
Cincinnati
Token Cincinnati
Feature activation+2.911
,
Token,
Feature activation+0.000
Newark
Token Newark
Feature activation+2.059
(
Token (
Feature activation+0.000
N
TokenN
Feature activation+0.661
.
Token.
Feature activation+0.000
is
Token is
Feature activation+0.000
safe
Token safe
Feature activation+0.000
at
Token at
Feature activation+0.000
home
Token home
Feature activation+0.000
in
Token in
Feature activation+0.000
New
Token New
Feature activation+3.092
Jersey
Token Jersey
Feature activation+2.738
with
Token with
Feature activation+0.000
his
Token his
Feature activation+0.000
family
Token family
Feature activation+0.000
,
Token,
Feature activation+0.000
compared
Token compared
Feature activation+0.000
Calgary
Token Calgary
Feature activation+0.000
to
Token to
Feature activation+0.000
places
Token places
Feature activation+0.000
like
Token like
Feature activation+0.000
Paris
Token Paris
Feature activation+2.917
,
Token,
Feature activation+0.000
Lyon
Token Lyon
Feature activation+1.595
,
Token,
Feature activation+0.000
Frankfurt
Token Frankfurt
Feature activation+1.832
,
Token,
Feature activation+0.000

INTERVAL 2.340 - 2.730
CONTAINS 0.001%

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
Hawai
Token Hawai
Feature activation+2.698
âĢ
TokenâĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
i
Tokeni
Feature activation+1.421
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
An
Token An
Feature activation+0.000
extreme
Token extreme
Feature activation+0.000
example
Token example
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
London
Token London
Feature activation+2.624
metropolitan
Token metropolitan
Feature activation+1.542
area
Token area
Feature activation+0.378
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
vast
Token vast
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
woman
Token woman
Feature activation+0.000
in
Token in
Feature activation+0.000
New
Token New
Feature activation+1.937
York
Token York
Feature activation+2.407
City
Token City
Feature activation+0.162
recorded
Token recorded
Feature activation+0.000
the
Token the
Feature activation+0.000
man
Token man
Feature activation+0.000
she
Token she
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
April
TokenApril
Feature activation+0.000
29
Token 29
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
New
Token New
Feature activation+0.748
York
Token York
Feature activation+2.348
,
Token,
Feature activation+0.000
N
Token N
Feature activation+0.000
.
Token.
Feature activation+0.000
Y
TokenY
Feature activation+0.012
.,
Token.,
Feature activation+0.000
18
Token 18
Feature activation+0.000
-
Token-
Feature activation+0.000
DEC
TokenDEC
Feature activation+0.000
-
Token-
Feature activation+0.000
2015
Token2015
Feature activation+0.000
London
Token London
Feature activation+2.349
,
Token,
Feature activation+0.000
United
Token United
Feature activation+0.491
Kingdom
Token Kingdom
Feature activation+1.290
Is
Token Is
Feature activation+0.000
lington
Tokenlington
Feature activation+0.652

INTERVAL 1.950 - 2.340
CONTAINS 0.002%

Also
Token Also
Feature activation+0.000
,
Token,
Feature activation+0.000
Tuesday
Token Tuesday
Feature activation+0.000
morning
Token morning
Feature activation+0.000
a
Token a
Feature activation+0.000
Jacksonville
Token Jacksonville
Feature activation+2.238
police
Token police
Feature activation+0.000
officer
Token officer
Feature activation+0.000
was
Token was
Feature activation+0.000
shot
Token shot
Feature activation+0.000
at
Token at
Feature activation+0.000
13
Token 13
Feature activation+0.000
years
Token years
Feature activation+0.000
longer
Token longer
Feature activation+0.000
than
Token than
Feature activation+0.000
in
Token in
Feature activation+0.000
Washington
Token Washington
Feature activation+1.961
Park
Token Park
Feature activation+0.041
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
west
Token west
Feature activation+0.000
.
Token.
Feature activation+0.000
talked
Token talked
Feature activation+0.000
about
Token about
Feature activation+0.000
ship
Token ship
Feature activation+0.000
building
Tokenbuilding
Feature activation+0.000
in
Token in
Feature activation+0.000
Halifax
Token Halifax
Feature activation+2.060
,
Token,
Feature activation+0.000
N
Token N
Feature activation+0.474
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.129
Schools
Token Schools
Feature activation+0.000
,
Token,
Feature activation+0.000
goes
Token goes
Feature activation+0.000
out
Token out
Feature activation+0.000
to
Token to
Feature activation+0.000
Ky
Token Ky
Feature activation+2.120
ron
Tokenron
Feature activation+0.000
H
Token H
Feature activation+0.059
orman
Tokenorman
Feature activation+0.000
and
Token and
Feature activation+0.000
his
Token his
Feature activation+0.000
Louis
Token Louis
Feature activation+0.000
,
Token,
Feature activation+0.000
Missouri
Token Missouri
Feature activation+0.000
and
Token and
Feature activation+0.000
Fort
Token Fort
Feature activation+0.492
Worth
Token Worth
Feature activation+2.103
,
Token,
Feature activation+0.000
Texas
Token Texas
Feature activation+1.261
have
Token have
Feature activation+0.000
stated
Token stated
Feature activation+0.000
their
Token their
Feature activation+0.000

INTERVAL 1.560 - 1.950
CONTAINS 0.004%

E
Token E
Feature activation+0.000
lish
Tokenlish
Feature activation+0.000
a
Tokena
Feature activation+0.000
Cap
Token Cap
Feature activation+0.000
rac
Tokenrac
Feature activation+0.000
ota
Tokenota
Feature activation+1.597
,
Token,
Feature activation+0.000
29
Token 29
Feature activation+0.000
,
Token,
Feature activation+0.000
unlawful
Token unlawful
Feature activation+0.000
conduct
Token conduct
Feature activation+0.000
-
Token-
Feature activation+0.000
day
Tokenday
Feature activation+0.000
walk
Token walk
Feature activation+0.000
outs
Tokenouts
Feature activation+0.000
in
Token in
Feature activation+0.000
Chicago
Token Chicago
Feature activation+1.720
,
Token,
Feature activation+0.000
St
Token St
Feature activation+0.000
.
Token.
Feature activation+0.016
Louis
Token Louis
Feature activation+1.825
,
Token,
Feature activation+0.000
rey
Tokenrey
Feature activation+0.000
said
Token said
Feature activation+0.000
at
Token at
Feature activation+0.000
a
Token a
Feature activation+0.000
2014
Token 2014
Feature activation+0.000
London
Token London
Feature activation+1.669
press
Token press
Feature activation+0.000
conference
Token conference
Feature activation+0.000
announcing
Token announcing
Feature activation+0.000
this
Token this
Feature activation+0.000
year
Token year
Feature activation+0.000
Kelly
Token Kelly
Feature activation+0.000
(
Token (
Feature activation+0.000
Baltimore
TokenBaltimore
Feature activation+2.240
Celtic
Token Celtic
Feature activation+0.376
;
Token;
Feature activation+0.000
Baltimore
Token Baltimore
Feature activation+1.669
,
Token,
Feature activation+0.000
Md
Token Md
Feature activation+2.224
.),
Token.),
Feature activation+0.000
Thomas
Token Thomas
Feature activation+0.000
McCabe
Token McCabe
Feature activation+0.102
quite
Token quite
Feature activation+0.000
a
Token a
Feature activation+0.000
distance
Token distance
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
Tex
Token Tex
Feature activation+1.718
an
Tokenan
Feature activation+0.000
8
Token 8
Feature activation+0.000
percent
Token percent
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 1.170 - 1.560
CONTAINS 0.012%

|
Token |
Feature activation+0.000
N
Token N
Feature activation+0.027
bl
Tokenbl
Feature activation+0.000
ake
Tokenake
Feature activation+0.000
@
Token@
Feature activation+0.000
md
Tokenmd
Feature activation+1.252
n
Tokenn
Feature activation+0.396
.
Token.
Feature activation+0.000
net
Tokennet
Feature activation+0.000
Photo
Token Photo
Feature activation+0.000
:
Token:
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
might
Token might
Feature activation+0.000
try
Token try
Feature activation+0.000
to
Token to
Feature activation+0.000
turn
Token turn
Feature activation+0.000
New
Token New
Feature activation+1.220
York
Token York
Feature activation+0.831
into
Token into
Feature activation+0.000
an
Token an
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
referendum
Token referendum
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
coming
Token coming
Feature activation+0.000
to
Token to
Feature activation+0.000
Boston
Token Boston
Feature activation+1.336
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
will
Token will
Feature activation+0.000
instead
Token instead
Feature activation+0.000
move
Token move
Feature activation+0.000
diplomatic
Token diplomatic
Feature activation+0.000
and
Token and
Feature activation+0.000
military
Token military
Feature activation+0.000
support
Token support
Feature activation+0.000
for
Token for
Feature activation+0.000
US
Token US
Feature activation+1.237
and
Token and
Feature activation+0.000
Israel
Token Israel
Feature activation+0.000
policy
Token policy
Feature activation+0.000
and
Token and
Feature activation+0.000
actions
Token actions
Feature activation+0.000
Bon
Token Bon
Feature activation+0.111
ne
Tokenne
Feature activation+0.000
ville
Tokenville
Feature activation+0.383
County
Token County
Feature activation+0.911
,
Token,
Feature activation+0.000
Idaho
Token Idaho
Feature activation+1.398
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
payments
Token payments
Feature activation+0.000
are
Token are
Feature activation+0.000
made
Token made
Feature activation+0.000

INTERVAL 0.780 - 1.170
CONTAINS 0.023%

illustrate
Token illustrate
Feature activation+0.000
its
Token its
Feature activation+0.000
commitment
Token commitment
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Canadian
Token Canadian
Feature activation+0.920
great
Token great
Feature activation+0.000
outdoors
Token outdoors
Feature activation+0.000
.
Token.
Feature activation+0.000
Earlier
Token Earlier
Feature activation+0.000
in
Token in
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
,
Token,
Feature activation+0.000
Connecticut
Token Connecticut
Feature activation+0.000
and
Token and
Feature activation+0.000
Texas
Token Texas
Feature activation+0.864
,
Token,
Feature activation+0.000
relief
Token relief
Feature activation+0.000
proposals
Token proposals
Feature activation+0.000
have
Token have
Feature activation+0.000
been
Token been
Feature activation+0.000
miles
Token miles
Feature activation+0.000
east
Token east
Feature activation+0.000
of
Token of
Feature activation+0.000
Minneapolis
Token Minneapolis
Feature activation+1.103
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.074
SCP
TokenSCP
Feature activation+0.000
-
Token-
Feature activation+0.000
20
Token20
Feature activation+0.000
88
Token88
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Pictures
Token Pictures
Feature activation+0.000
:
Token:
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.106
by
Tokenby
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
CIA
TokenCIA
Feature activation+0.000
Director
Token Director
Feature activation+0.000
blood
Token blood
Feature activation+0.000
sample
Token sample
Feature activation+0.000
was
Token was
Feature activation+0.000
taken
Token taken
Feature activation+0.000
,
Token,
Feature activation+0.000
Beau
Token Beau
Feature activation+1.016
pre
Tokenpre
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 0.390 - 0.780
CONTAINS 0.071%

Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Miami
TokenMiami
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Mont
TokenMont
Feature activation+0.526
real
Tokenreal
Feature activation+1.187
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Sac
TokenSac
Feature activation+1.587
rament
Tokenrament
Feature activation+1.922
her
Token her
Feature activation+0.000
tenure
Token tenure
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
New
Token New
Feature activation+0.792
York
Token York
Feature activation+0.594
senator
Token senator
Feature activation+0.000
with
Token with
Feature activation+0.000
relatively
Token relatively
Feature activation+0.000
moderate
Token moderate
Feature activation+0.000
views
Token views
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Return
TokenReturn
Feature activation+0.000
to
Token to
Feature activation+0.000
az
Token az
Feature activation+0.000
central
Tokencentral
Feature activation+0.493
.
Token.
Feature activation+0.000
com
Tokencom
Feature activation+0.000
for
Token for
Feature activation+0.000
more
Token more
Feature activation+0.000
details
Token details
Feature activation+0.000
plain
Token plain
Feature activation+0.000
packaging
Token packaging
Feature activation+0.000
legislation
Token legislation
Feature activation+0.000
has
Token has
Feature activation+0.000
in
Token in
Feature activation+0.000
Australia
Token Australia
Feature activation+0.636
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
hasn
Token hasn
Feature activation+0.000
't
Token't
Feature activation+0.000
ruled
Token ruled
Feature activation+0.000
on
Token on
Feature activation+0.000
Racial
Token Racial
Feature activation+0.000
Equality
Token Equality
Feature activation+0.000
(
Token (
Feature activation+0.000
C
TokenC
Feature activation+0.078
ORE
TokenORE
Feature activation+0.618
)
Token)
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Student
Token Student
Feature activation+0.000
Non
Token Non
Feature activation+0.000

INTERVAL 0.000 - 0.390
CONTAINS 99.886%

principles
Token principles
Feature activation+0.000
and
Token and
Feature activation+0.000
facts
Token facts
Feature activation+0.000
,
Token,
Feature activation+0.000
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
vast
Token vast
Feature activation+0.000
and
Token and
Feature activation+0.000
noble
Token noble
Feature activation+0.000
science
Token science
Feature activation+0.000
,
Token,
Feature activation+0.000
way
Token way
Feature activation+0.000
ward
Tokenward
Feature activation+0.000
millennial
Token millennial
Feature activation+0.000
voters
Token voters
Feature activation+0.000
who
Token who
Feature activation+0.000
are
Token are
Feature activation+0.000
supporting
Token supporting
Feature activation+0.000
Libertarian
Token Libertarian
Feature activation+0.000
Party
Token Party
Feature activation+0.000
candidate
Token candidate
Feature activation+0.000
Gary
Token Gary
Feature activation+0.000
manifesto
Token manifesto
Feature activation+0.000
.
Token.
Feature activation+0.000
UKIP
Token UKIP
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
reaction
Token reaction
Feature activation+0.000
is
Token is
Feature activation+0.000
to
Token to
Feature activation+0.000
cry
Token cry
Feature activation+0.000
Lib
Token Lib
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
That
TokenThat
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
not
Token not
Feature activation+0.000
a
Token a
Feature activation+0.000
lot
Token lot
Feature activation+0.000
to
Token to
Feature activation+0.000
Post
Token Post
Feature activation+0.000
by
Token by
Feature activation+0.000
Willis
Token Willis
Feature activation+0.000
Esc
Token Esc
Feature activation+0.000
hen
Tokenhen
Feature activation+0.000
bach
Tokenbach
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Science
TokenScience
Feature activation+0.000
Magazine
Token Magazine
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 9 in H3.4: (feature 6490)

TOP ACTIVATIONS
MAX = 2.824

s
Tokens
Feature activation+0.000
the
Token the
Feature activation+0.000
cold
Token cold
Feature activation+0.000
hard
Token hard
Feature activation+0.000
truth
Token truth
Feature activation+0.000
...
Token...
Feature activation+2.824
We
TokenWe
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.290
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
on
Token on
Feature activation+0.000
scar
Token scar
Feature activation+0.000
red
Tokenred
Feature activation+0.000
next
Token next
Feature activation+0.000
.
Token.
Feature activation+0.000
Again
Token Again
Feature activation+0.000
âĢĵ
TokenâĢĵ
Feature activation+2.492
No
Token No
Feature activation+0.000
one
Token one
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.598
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.000
just
Tokenjust
Feature activation+0.000
like
Token like
Feature activation+0.000
good
Token good
Feature activation+0.000
old
Token old
Feature activation+0.000
-
Token -
Feature activation+2.475
w
Tokenw
Feature activation+0.000
in
Token in
Feature activation+0.000
Perl
Token Perl
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
until
Token until
Feature activation+0.000
they
Token they
Feature activation+0.000
are
Token are
Feature activation+0.000
in
Token in
Feature activation+0.000
charge
Token charge
Feature activation+0.000
-
Token -
Feature activation+2.182
Eric
TokenEric
Feature activation+0.000
MacDonald
Token MacDonald
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
if
Token if
Feature activation+0.000
you
Token you
Feature activation+0.000
have
Token have
Feature activation+0.000
any
Token any
Feature activation+0.000
questions
Token questions
Feature activation+0.000
-
Token -
Feature activation+1.945
getting
Token getting
Feature activation+0.000
good
Token good
Feature activation+0.000
information
Token information
Feature activation+0.000
in
Token in
Feature activation+0.000
front
Token front
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
Real
Token Real
Feature activation+0.000
Madrid
Token Madrid
Feature activation+0.000
and
Token and
Feature activation+0.000
Barcelona
Token Barcelona
Feature activation+0.000
-
Token -
Feature activation+1.772
siph
Token siph
Feature activation+0.000
on
Tokenon
Feature activation+0.000
off
Token off
Feature activation+0.000
so
Token so
Feature activation+0.000
much
Token much
Feature activation+0.000
That
TokenThat
Feature activation+0.000
means
Token means
Feature activation+0.000
a
Token a
Feature activation+0.000
terrorism
Token terrorism
Feature activation+0.000
suspect
Token suspect
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.662
Ļ
TokenĻ
Feature activation+0.053
s
Tokens
Feature activation+0.000
friends
Token friends
Feature activation+0.000
,
Token,
Feature activation+0.000
family
Token family
Feature activation+0.000
friends
Token friends
Feature activation+0.000
and
Token and
Feature activation+0.000
told
Token told
Feature activation+0.000
people
Token people
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.500
Ŀ
TokenĿ
Feature activation+0.000
Hill
Token Hill
Feature activation+0.000
told
Token told
Feature activation+0.000
the
Token the
Feature activation+0.000
NL
Token NL
Feature activation+0.000
group
Token group
Feature activation+0.000
central
Token central
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
occupation
Token occupation
Feature activation+0.000
-
Token -
Feature activation+1.433
which
Token which
Feature activation+0.000
seems
Token seems
Feature activation+0.000
to
Token to
Feature activation+0.000
have
Token have
Feature activation+0.000
sprung
Token sprung
Feature activation+0.000
were
Token were
Feature activation+0.000
baffled
Token baffled
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
authorities
Token authorities
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.411
Ļ
TokenĻ
Feature activation+0.000
conclusion
Token conclusion
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.119
ľ
Tokenľ
Feature activation+0.000
,
Token,
Feature activation+0.000
their
Token their
Feature activation+0.000
c
Token c
Feature activation+0.000
udd
Tokenudd
Feature activation+0.000
les
Tokenles
Feature activation+0.000
-
Token -
Feature activation+1.345
and
Token and
Feature activation+0.000
you
Token you
Feature activation+0.000
can
Token can
Feature activation+0.000
bet
Token bet
Feature activation+0.000
questions
Token questions
Feature activation+0.000
the
Token the
Feature activation+0.000
Great
Token Great
Feature activation+0.000
State
Token State
Feature activation+0.000
of
Token of
Feature activation+0.000
Texas
Token Texas
Feature activation+0.000
-
Token -
Feature activation+1.326
TEAM
Token TEAM
Feature activation+0.000
WORK
TokenWORK
Feature activation+0.000
!
Token!
Feature activation+0.000
Record
Token Record
Feature activation+0.000
setting
Token setting
Feature activation+0.000
But
Token But
Feature activation+0.000
somehow
Token somehow
Feature activation+0.000
,
Token,
Feature activation+0.000
things
Token things
Feature activation+0.000
seem
Token seem
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+1.264
off
Token off
Feature activation+0.000
.
Token.
Feature activation+0.000
Especially
Token Especially
Feature activation+0.000
that
Token that
Feature activation+0.000
too
Token too
Feature activation+0.000
be
Token be
Feature activation+0.000
forced
Token forced
Feature activation+0.000
to
Token to
Feature activation+0.000
register
Token register
Feature activation+0.000
themselves
Token themselves
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+1.263
or
Token or
Feature activation+0.000
face
Token face
Feature activation+0.000
a
Token a
Feature activation+0.000
felony
Token felony
Feature activation+0.000
charge
Token charge
Feature activation+0.000
million
Token million
Feature activation+0.000
unf
Token unf
Feature activation+0.000
illed
Tokenilled
Feature activation+0.000
computer
Token computer
Feature activation+0.000
science
Token science
Feature activation+0.000
âĢĵ
TokenâĢĵ
Feature activation+1.202
related
Tokenrelated
Feature activation+0.000
jobs
Token jobs
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
t
Tokent
Feature activation+0.000
baff
Token baff
Feature activation+0.000
les
Tokenles
Feature activation+0.000
me
Token me
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.187
Ŀ
TokenĿ
Feature activation+0.000
he
Token he
Feature activation+0.000
says
Token says
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
s
Tokens
Feature activation+0.000
essentially
Token essentially
Feature activation+0.000
a
Token a
Feature activation+0.000
TV
Token TV
Feature activation+0.000
show
Token show
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.181
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
repeats
Token repeats
Feature activation+0.000
on
Token on
Feature activation+0.000
Super
Token Super
Feature activation+0.000
are
Token are
Feature activation+0.000
central
Token central
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
nation
Token nation
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.173
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
identity
Token identity
Feature activation+0.000
:
Token:
Feature activation+0.000
the
Token the
Feature activation+0.000
DIS
TokenDIS
Feature activation+0.000
PLAY
TokenPLAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.216
âĢĵ
TokenâĢĵ
Feature activation+1.168
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Color
TokenColor
Feature activation+0.000
Gam
Token Gam
Feature activation+0.000
ut
Tokenut
Feature activation+0.000
a
Token a
Feature activation+0.000
person
Token person
Feature activation+0.000
settles
Token settles
Feature activation+0.000
on
Token on
Feature activation+0.000
their
Token their
Feature activation+0.000
âĢ
Token âĢ
Feature activation+1.147
ľ
Tokenľ
Feature activation+0.000
true
Tokentrue
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.548
Ŀ
TokenĿ
Feature activation+0.000
identity
Token identity
Feature activation+0.000

Top DFA by src position
MAX = 3.005

âĢ
TokenâĢ
Feature activation-0.004
Top resid features:
Ŀ
TokenĿ
Feature activation-0.000
Top resid features:
did
Token did
Feature activation-0.000
Top resid features:
just
Token just
Feature activation+0.000
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+1.512
Top resid features:
it
Token it
Feature activation-0.002
Top resid features:
bl
Token bl
Feature activation-0.000
Top resid features:
azed
Tokenazed
Feature activation+0.000
Top resid features:
up
Token up
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.004
Top resid features:
where
Token where
Feature activation-0.005
Top resid features:
our
Token our
Feature activation+0.003
Top resid features:
military
Token military
Feature activation-0.000
Top resid features:
personnel
Token personnel
Feature activation+0.000
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+1.899
Top resid features:
people
Token people
Feature activation-0.001
Top resid features:
like
Token like
Feature activation+0.002
Top resid features:
my
Token my
Feature activation-0.000
Top resid features:
Marine
Token Marine
Feature activation-0.000
Top resid features:
son
Token son
Feature activation-0.003
Top resid features:
pointing
Token pointing
Feature activation-0.006
Top resid features:
out
Token out
Feature activation-0.001
Top resid features:
questionable
Token questionable
Feature activation-0.001
Top resid features:
coding
Token coding
Feature activation-0.003
Top resid features:
practices
Token practices
Feature activation-0.007
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+3.005
Top resid features:
just
Tokenjust
Feature activation-0.005
Top resid features:
like
Token like
Feature activation+0.001
Top resid features:
good
Token good
Feature activation+0.021
Top resid features:
old
Token old
Feature activation+0.031
Top resid features:
-
Token -
Feature activation-0.082
Top resid features:
ions
Tokenions
Feature activation+0.001
Top resid features:
will
Token will
Feature activation-0.003
Top resid features:
never
Token never
Feature activation+0.001
Top resid features:
be
Token be
Feature activation-0.001
Top resid features:
satisfied
Token satisfied
Feature activation-0.003
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+2.878
Top resid features:
they
Token they
Feature activation-0.026
Top resid features:
will
Token will
Feature activation-0.011
Top resid features:
always
Token always
Feature activation+0.011
Top resid features:
up
Token up
Feature activation+0.004
Top resid features:
the
Token the
Feature activation-0.005
Top resid features:
make
Token make
Feature activation-0.002
Top resid features:
it
Token it
Feature activation-0.005
Top resid features:
easier
Token easier
Feature activation-0.000
Top resid features:
on
Token on
Feature activation+0.000
Top resid features:
you
Token you
Feature activation-0.000
Top resid features:
--
Token --
Feature activation+2.688
Top resid features:
a
Token a
Feature activation+0.006
Top resid features:
recent
Token recent
Feature activation+0.003
Top resid features:
email
Token email
Feature activation-0.004
Top resid features:
copied
Token copied
Feature activation+0.004
Top resid features:
below
Token below
Feature activation+0.004
Top resid features:
which
Token which
Feature activation-0.001
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
top
Token top
Feature activation-0.016
Top resid features:
two
Token two
Feature activation+0.012
Top resid features:
teams
Token teams
Feature activation+0.009
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+1.122
Top resid features:
Real
Token Real
Feature activation+0.004
Top resid features:
Madrid
Token Madrid
Feature activation+0.001
Top resid features:
and
Token and
Feature activation-0.055
Top resid features:
Barcelona
Token Barcelona
Feature activation+0.000
Top resid features:
-
Token -
Feature activation+0.038
Top resid features:
âĢ
TokenâĢ
Feature activation+0.010
Top resid features:
Ļ
TokenĻ
Feature activation-0.001
Top resid features:
re
Tokenre
Feature activation+0.000
Top resid features:
a
Token a
Feature activation-0.000
Top resid features:
terrorist
Token terrorist
Feature activation-0.000
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+1.639
Top resid features:
without
Tokenwithout
Feature activation+0.001
Top resid features:
providing
Token providing
Feature activation+0.001
Top resid features:
any
Token any
Feature activation+0.001
Top resid features:
concrete
Token concrete
Feature activation-0.000
Top resid features:
proof
Token proof
Feature activation-0.000
Top resid features:
that
Token that
Feature activation-0.003
Top resid features:
was
Token was
Feature activation-0.000
Top resid features:
directed
Token directed
Feature activation-0.001
Top resid features:
at
Token at
Feature activation-0.000
Top resid features:
me
Token me
Feature activation-0.007
Top resid features:
âĢķ
Token âĢķ
Feature activation+0.981
Top resid features:
is
Token is
Feature activation+0.002
Top resid features:
undermined
Token undermined
Feature activation+0.001
Top resid features:
by
Token by
Feature activation+0.003
Top resid features:
the
Token the
Feature activation-0.006
Top resid features:
fact
Token fact
Feature activation+0.000
Top resid features:
members
Token members
Feature activation+0.001
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
Home
Token Home
Feature activation-0.002
Top resid features:
Sweet
Token Sweet
Feature activation-0.002
Top resid features:
Home
Token Home
Feature activation-0.001
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+1.291
Top resid features:
the
Token the
Feature activation-0.014
Top resid features:
group
Token group
Feature activation-0.036
Top resid features:
central
Token central
Feature activation-0.030
Top resid features:
to
Token to
Feature activation+0.027
Top resid features:
the
Token the
Feature activation-0.009
Top resid features:
vicious
Token vicious
Feature activation-0.001
Top resid features:
ness
Tokenness
Feature activation-0.000
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
killing
Token killing
Feature activation-0.002
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+1.021
Top resid features:
T
TokenT
Feature activation-0.002
Top resid features:
ev
Tokenev
Feature activation+0.001
Top resid features:
lin
Tokenlin
Feature activation-0.001
Top resid features:
had
Token had
Feature activation+0.003
Top resid features:
been
Token been
Feature activation+0.001
Top resid features:
I
Token I
Feature activation-0.008
Top resid features:
would
Token would
Feature activation-0.002
Top resid features:
personally
Token personally
Feature activation-0.002
Top resid features:
never
Token never
Feature activation+0.001
Top resid features:
make
Token make
Feature activation-0.001
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+1.810
Top resid features:
I
Token I
Feature activation-0.007
Top resid features:
wouldn
Token wouldn
Feature activation+0.004
Top resid features:
âĢ
TokenâĢ
Feature activation+0.068
Top resid features:
Ļ
TokenĻ
Feature activation-0.006
Top resid features:
t
Tokent
Feature activation-0.005
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
working
Token working
Feature activation-0.000
Top resid features:
great
Token great
Feature activation-0.001
Top resid features:
together
Token together
Feature activation-0.000
Top resid features:
!
Token!
Feature activation+0.009
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+1.650
Top resid features:
Donald
Token Donald
Feature activation+0.000
Top resid features:
J
Token J
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
Trump
Token Trump
Feature activation-0.001
Top resid features:
(@
Token (@
Feature activation-0.008
Top resid features:
It
Token It
Feature activation-0.009
Top resid features:
âĢ
TokenâĢ
Feature activation+0.082
Top resid features:
Ļ
TokenĻ
Feature activation-0.003
Top resid features:
s
Tokens
Feature activation+0.001
Top resid features:
funny
Token funny
Feature activation-0.010
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.631
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
jokes
Token jokes
Feature activation-0.004
Top resid features:
land
Token land
Feature activation+0.004
Top resid features:
,
Token,
Feature activation+0.009
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
âĢ
TokenâĢ
Feature activation+0.036
Top resid features:
Ļ
TokenĻ
Feature activation-0.000
Top resid features:
re
Tokenre
Feature activation+0.000
Top resid features:
a
Token a
Feature activation-0.000
Top resid features:
terrorist
Token terrorist
Feature activation-0.000
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+0.954
Top resid features:
without
Tokenwithout
Feature activation+0.001
Top resid features:
providing
Token providing
Feature activation+0.001
Top resid features:
any
Token any
Feature activation+0.000
Top resid features:
concrete
Token concrete
Feature activation+0.000
Top resid features:
proof
Token proof
Feature activation-0.000
Top resid features:
fields
Token fields
Feature activation+0.000
Top resid features:
in
Token in
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
United
Token United
Feature activation-0.000
Top resid features:
States
Token States
Feature activation-0.000
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+1.502
Top resid features:
and
Token and
Feature activation+0.005
Top resid features:
not
Token not
Feature activation-0.000
Top resid features:
for
Token for
Feature activation-0.000
Top resid features:
lack
Token lack
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
volume
Token volume
Feature activation+0.000
Top resid features:
by
Token by
Feature activation+0.003
Top resid features:
selling
Token selling
Feature activation-0.000
Top resid features:
smaller
Token smaller
Feature activation+0.000
Top resid features:
amounts
Token amounts
Feature activation-0.000
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.611
Top resid features:
quarter
Token quarter
Feature activation-0.003
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
and
Token and
Feature activation-0.004
Top resid features:
eighth
Token eighth
Feature activation-0.001
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
going
Token going
Feature activation+0.009
Top resid features:
to
Token to
Feature activation-0.004
Top resid features:
easily
Token easily
Feature activation+0.000
Top resid features:
exceed
Token exceed
Feature activation+0.002
Top resid features:
that
Token that
Feature activation-0.002
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+0.776
Top resid features:
which
Tokenwhich
Feature activation+0.023
Top resid features:
is
Token is
Feature activation-0.004
Top resid features:
rather
Token rather
Feature activation+0.003
Top resid features:
impressive
Token impressive
Feature activation-0.003
Top resid features:
for
Token for
Feature activation+0.001
Top resid features:
lives
Token lives
Feature activation-0.001
Top resid features:
deeply
Token deeply
Feature activation+0.001
Top resid features:
all
Token all
Feature activation-0.002
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
time
Token time
Feature activation-0.000
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+0.722
Top resid features:
if
Tokenif
Feature activation-0.003
Top resid features:
I
Token I
Feature activation-0.005
Top resid features:
felt
Token felt
Feature activation-0.000
Top resid features:
it
Token it
Feature activation-0.004
Top resid features:
didn
Token didn
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.008
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
DIS
TokenDIS
Feature activation+0.007
Top resid features:
PLAY
TokenPLAY
Feature activation-0.004
Top resid features:
S
TokenS
Feature activation+0.002
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+1.070
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+0.420
Top resid features:
âĢĵ
TokenâĢĵ
Feature activation+0.055
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Color
TokenColor
Feature activation+0.000
Top resid features:
state
Token state
Feature activation-0.004
Top resid features:
of
Token of
Feature activation-0.001
Top resid features:
experimentation
Token experimentation
Feature activation-0.004
Top resid features:
or
Token or
Feature activation-0.013
Top resid features:
confusion
Token confusion
Feature activation-0.003
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+0.600
Top resid features:
st
Tokenst
Feature activation+0.000
Top resid features:
ere
Tokenere
Feature activation+0.001
Top resid features:
otyp
Tokenotyp
Feature activation+0.000
Top resid features:
ically
Tokenically
Feature activation-0.000
Top resid features:
experienced
Token experienced
Feature activation+0.001
Top resid features:

Decoder Weights Distribution

Head 0: 0.14

Head 1: 0.04

Head 2: 0.04

Head 3: 0.03

Head 4: 0.43

Head 5: 0.07

Head 6: 0.04

Head 7: 0.04

Head 8: 0.05

Head 9: 0.04

Head 10: 0.07

Head 11: 0.03

Positive logits

ope1.97

raq1.82

atz1.80

unchecked1.74

opes1.72

umbs1.69

-'1.67

'-1.63

heed1.62

osity1.60

alter1.58

'?1.52

grave1.52

1.50

ivot1.49

ims1.48

Pg1.48

emonic1.48

1.47

ombo1.47

Negative logits

anwhile-2.04

-2.00

Interstitial-1.90

)",-1.81

Learns-1.74

[|-1.71

ppo-1.67

-1.66

catentry-1.63

DragonMagazine-1.60

theless-1.60

hobbies-1.59

Reincarn-1.59

Pay-1.55

ItemImage-1.53

Recomm-1.53

Flavoring-1.53

MpServer-1.52

EStreamFrame-1.52

enegger-1.51

INTERVAL 2.542 - 2.824
CONTAINS 0.000%

s
Tokens
Feature activation+0.000
the
Token the
Feature activation+0.000
cold
Token cold
Feature activation+0.000
hard
Token hard
Feature activation+0.000
truth
Token truth
Feature activation+0.000
...
Token...
Feature activation+2.824
We
TokenWe
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.290
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
on
Token on
Feature activation+0.000

INTERVAL 2.259 - 2.542
CONTAINS 0.000%

scar
Token scar
Feature activation+0.000
red
Tokenred
Feature activation+0.000
next
Token next
Feature activation+0.000
.
Token.
Feature activation+0.000
Again
Token Again
Feature activation+0.000
âĢĵ
TokenâĢĵ
Feature activation+2.492
No
Token No
Feature activation+0.000
one
Token one
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.598
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.000
just
Tokenjust
Feature activation+0.000
like
Token like
Feature activation+0.000
good
Token good
Feature activation+0.000
old
Token old
Feature activation+0.000
-
Token -
Feature activation+2.475
w
Tokenw
Feature activation+0.000
in
Token in
Feature activation+0.000
Perl
Token Perl
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 1.977 - 2.259
CONTAINS 0.000%

until
Token until
Feature activation+0.000
they
Token they
Feature activation+0.000
are
Token are
Feature activation+0.000
in
Token in
Feature activation+0.000
charge
Token charge
Feature activation+0.000
-
Token -
Feature activation+2.182
Eric
TokenEric
Feature activation+0.000
MacDonald
Token MacDonald
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 1.694 - 1.977
CONTAINS 0.000%

if
Token if
Feature activation+0.000
you
Token you
Feature activation+0.000
have
Token have
Feature activation+0.000
any
Token any
Feature activation+0.000
questions
Token questions
Feature activation+0.000
-
Token -
Feature activation+1.945
getting
Token getting
Feature activation+0.000
good
Token good
Feature activation+0.000
information
Token information
Feature activation+0.000
in
Token in
Feature activation+0.000
front
Token front
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
Real
Token Real
Feature activation+0.000
Madrid
Token Madrid
Feature activation+0.000
and
Token and
Feature activation+0.000
Barcelona
Token Barcelona
Feature activation+0.000
-
Token -
Feature activation+1.772
siph
Token siph
Feature activation+0.000
on
Tokenon
Feature activation+0.000
off
Token off
Feature activation+0.000
so
Token so
Feature activation+0.000
much
Token much
Feature activation+0.000

INTERVAL 1.412 - 1.694
CONTAINS 0.000%

That
TokenThat
Feature activation+0.000
means
Token means
Feature activation+0.000
a
Token a
Feature activation+0.000
terrorism
Token terrorism
Feature activation+0.000
suspect
Token suspect
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.662
Ļ
TokenĻ
Feature activation+0.053
s
Tokens
Feature activation+0.000
friends
Token friends
Feature activation+0.000
,
Token,
Feature activation+0.000
family
Token family
Feature activation+0.000
group
Token group
Feature activation+0.000
central
Token central
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
occupation
Token occupation
Feature activation+0.000
-
Token -
Feature activation+1.433
which
Token which
Feature activation+0.000
seems
Token seems
Feature activation+0.000
to
Token to
Feature activation+0.000
have
Token have
Feature activation+0.000
sprung
Token sprung
Feature activation+0.000
friends
Token friends
Feature activation+0.000
and
Token and
Feature activation+0.000
told
Token told
Feature activation+0.000
people
Token people
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.500
Ŀ
TokenĿ
Feature activation+0.000
Hill
Token Hill
Feature activation+0.000
told
Token told
Feature activation+0.000
the
Token the
Feature activation+0.000
NL
Token NL
Feature activation+0.000

INTERVAL 1.130 - 1.412
CONTAINS 0.001%

are
Token are
Feature activation+0.000
central
Token central
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
nation
Token nation
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+1.173
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
identity
Token identity
Feature activation+0.000
:
Token:
Feature activation+0.000
the
Token the
Feature activation+0.000
But
Token But
Feature activation+0.000
somehow
Token somehow
Feature activation+0.000
,
Token,
Feature activation+0.000
things
Token things
Feature activation+0.000
seem
Token seem
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+1.264
off
Token off
Feature activation+0.000
.
Token.
Feature activation+0.000
Especially
Token Especially
Feature activation+0.000
that
Token that
Feature activation+0.000
too
Token too
Feature activation+0.000
a
Token a
Feature activation+0.000
person
Token person
Feature activation+0.000
settles
Token settles
Feature activation+0.000
on
Token on
Feature activation+0.000
their
Token their
Feature activation+0.000
âĢ
Token âĢ
Feature activation+1.147
ľ
Tokenľ
Feature activation+0.000
true
Tokentrue
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.548
Ŀ
TokenĿ
Feature activation+0.000
identity
Token identity
Feature activation+0.000
be
Token be
Feature activation+0.000
forced
Token forced
Feature activation+0.000
to
Token to
Feature activation+0.000
register
Token register
Feature activation+0.000
themselves
Token themselves
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+1.263
or
Token or
Feature activation+0.000
face
Token face
Feature activation+0.000
a
Token a
Feature activation+0.000
felony
Token felony
Feature activation+0.000
charge
Token charge
Feature activation+0.000
the
Token the
Feature activation+0.000
Great
Token Great
Feature activation+0.000
State
Token State
Feature activation+0.000
of
Token of
Feature activation+0.000
Texas
Token Texas
Feature activation+0.000
-
Token -
Feature activation+1.326
TEAM
Token TEAM
Feature activation+0.000
WORK
TokenWORK
Feature activation+0.000
!
Token!
Feature activation+0.000
Record
Token Record
Feature activation+0.000
setting
Token setting
Feature activation+0.000

INTERVAL 0.847 - 1.130
CONTAINS 0.004%

the
Token the
Feature activation+0.000
NL
Token NL
Feature activation+0.000
J
TokenJ
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.907
ľ
Tokenľ
Feature activation+0.000
So
TokenSo
Feature activation+0.000
this
Token this
Feature activation+0.000
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
for
Token for
Feature activation+0.000
,
Token,
Feature activation+0.000
have
Token have
Feature activation+0.000
their
Token their
Feature activation+0.000
melancholy
Token melancholy
Feature activation+0.000
;
Token;
Feature activation+0.889
for
Token for
Feature activation+0.000
what
Token what
Feature activation+0.000
we
Token we
Feature activation+0.000
leave
Token leave
Feature activation+0.000
behind
Token behind
Feature activation+0.000
el
Token el
Feature activation+0.000
it
Tokenit
Feature activation+0.000
ist
Tokenist
Feature activation+0.000
utopian
Token utopian
Feature activation+0.000
schemes
Token schemes
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.875
Ŀ
TokenĿ
Feature activation+0.000
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
The
Token The
Feature activation+0.000
Cron
Token Cron
Feature activation+0.000
be
Token be
Feature activation+0.000
the
Token the
Feature activation+0.000
best
Token best
Feature activation+0.000
thing
Token thing
Feature activation+0.000
about
Token about
Feature activation+0.000
âĢ
Token âĢ
Feature activation+1.013
ľ
Tokenľ
Feature activation+0.000
R
TokenR
Feature activation+0.000
apt
Tokenapt
Feature activation+0.000
ure
Tokenure
Feature activation+0.000
.
Token.
Feature activation+0.000
care
Token care
Feature activation+0.000
protocols
Token protocols
Feature activation+0.000
.
Token.
Feature activation+0.000
Read
Token Read
Feature activation+0.000
mission
Tokenmission
Feature activation+0.000
-
Token-
Feature activation+0.894
related
Tokenrelated
Feature activation+0.000
costs
Token costs
Feature activation+0.000
have
Token have
Feature activation+0.000
become
Token become
Feature activation+0.000
so
Token so
Feature activation+0.000

INTERVAL 0.565 - 0.847
CONTAINS 0.011%

all
Token all
Feature activation+0.000
of
Token of
Feature activation+0.000
which
Token which
Feature activation+0.000
I
Token I
Feature activation+0.000
condemn
Token condemn
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.651
and
Token and
Feature activation+0.000
clashes
Token clashes
Feature activation+0.000
between
Token between
Feature activation+0.000
Palestinians
Token Palestinians
Feature activation+0.000
and
Token and
Feature activation+0.000
what
Token what
Feature activation+0.000
?
Token?
Feature activation+0.000
The
Token The
Feature activation+0.000
issue
Token issue
Feature activation+0.000
isn
Token isn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.732
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
whether
Token whether
Feature activation+0.000
Sessions
Token Sessions
Feature activation+0.000
spoke
Token spoke
Feature activation+0.000
filmed
Token filmed
Feature activation+0.000
them
Token them
Feature activation+0.000
feeding
Token feeding
Feature activation+0.000
on
Token on
Feature activation+0.000
algae
Token algae
Feature activation+0.000
-
Token-
Feature activation+0.599
covered
Tokencovered
Feature activation+0.000
glass
Token glass
Feature activation+0.000
and
Token and
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.000
stim
Tokenstim
Feature activation+0.000
because
Token because
Feature activation+0.000
they
Token they
Feature activation+0.000
are
Token are
Feature activation+0.000
extraordinarily
Token extraordinarily
Feature activation+0.000
risk
Token risk
Feature activation+0.000
-
Token-
Feature activation+0.633
a
Tokena
Feature activation+0.000
verse
Tokenverse
Feature activation+0.000
.)
Token.)
Feature activation+0.000
But
Token But
Feature activation+0.000
Oy
Token Oy
Feature activation+0.000
literally
Token literally
Feature activation+0.000
refer
Token refer
Feature activation+0.000
to
Token to
Feature activation+0.000
them
Token them
Feature activation+0.000
as
Token as
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.820
ľ
Tokenľ
Feature activation+0.000
band
Tokenband
Feature activation+0.000
its
Tokenits
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.390
Ŀ
TokenĿ
Feature activation+0.000

INTERVAL 0.282 - 0.565
CONTAINS 0.028%

supportive
Token supportive
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
occupation
Token occupation
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.397
Ŀ
TokenĿ
Feature activation+0.000
Sarah
Token Sarah
Feature activation+0.000
Water
Token Water
Feature activation+0.000
feld
Tokenfeld
Feature activation+0.000
,
Token,
Feature activation+0.000
ist
Tokenist
Feature activation+0.000
in
Token in
Feature activation+0.000
Your
Token Your
Feature activation+0.000
Town
Token Town
Feature activation+0.000
bonus
Token bonus
Feature activation+0.000
-
Token -
Feature activation+0.320
mont
Token mont
Feature activation+0.000
age
Tokenage
Feature activation+0.000
-
Token -
Feature activation+0.026
St
Token St
Feature activation+0.000
eely
Tokeneely
Feature activation+0.000
.
Token.
Feature activation+0.000
Upon
Token Upon
Feature activation+0.000
graduation
Token graduation
Feature activation+0.000
,
Token,
Feature activation+0.000
they
Token they
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.439
Ļ
TokenĻ
Feature activation+0.000
d
Tokend
Feature activation+0.000
re
Token re
Feature activation+0.000
join
Tokenjoin
Feature activation+0.000
their
Token their
Feature activation+0.000
or
Token or
Feature activation+0.000
face
Token face
Feature activation+0.000
a
Token a
Feature activation+0.000
felony
Token felony
Feature activation+0.000
charge
Token charge
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.427
and
Token and
Feature activation+0.000
then
Token then
Feature activation+0.000
be
Token be
Feature activation+0.000
ske
Token ske
Feature activation+0.000
wered
Tokenwered
Feature activation+0.000
Washington
Token Washington
Feature activation+0.000
is
Token is
Feature activation+0.000
also
Token also
Feature activation+0.000
giving
Token giving
Feature activation+0.000
communities
Token communities
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.465
ľ
Tokenľ
Feature activation+0.000
liv
Tokenliv
Feature activation+0.000
ability
Tokenability
Feature activation+0.000
grants
Token grants
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.089

INTERVAL 0.000 - 0.282
CONTAINS 99.956%

viewed
Token viewed
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Create
TokenCreate
Feature activation+0.000
your
Token your
Feature activation+0.000
own
Token own
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
lamp
Token lamp
Feature activation+0.000
puts
Token puts
Feature activation+0.000
ch
Tokench
Feature activation+0.000
in
Token in
Feature activation+0.000
Kiev
Token Kiev
Feature activation+0.000
in
Token in
Feature activation+0.000
February
Token February
Feature activation+0.000
2014
Token 2014
Feature activation+0.000
,
Token,
Feature activation+0.000
received
Token received
Feature activation+0.000
promises
Token promises
Feature activation+0.000
of
Token of
Feature activation+0.000
Is
TokenIs
Feature activation+0.000
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
they
Token they
Feature activation+0.000
were
Token were
Feature activation+0.000
very
Token very
Feature activation+0.000
unlikely
Token unlikely
Feature activation+0.000
to
Token to
Feature activation+0.000
pass
Token pass
Feature activation+0.000
on
Token on
Feature activation+0.000
HIV
Token HIV
Feature activation+0.000
IDF
Token IDF
Feature activation+0.000
was
Token was
Feature activation+0.000
transporting
Token transporting
Feature activation+0.000
al
Token al
Feature activation+0.000
-
Token-
Feature activation+0.000
Nusra
TokenNusra
Feature activation+0.000
members
Token members
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
ambulance
Token ambulance
Feature activation+0.000
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
the
Token the
Feature activation+0.000
2010
Token 2010
Feature activation+0.000
-
Token-
Feature activation+0.000
11
Token11
Feature activation+0.000
National
Token National
Feature activation+0.000
Hockey
Token Hockey
Feature activation+0.000
League
Token League
Feature activation+0.000
(
Token (
Feature activation+0.000
N
TokenN
Feature activation+0.000
HL
TokenHL
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