H9.2 top features

Top feature 0 in H9.2: (feature 21988)

TOP ACTIVATIONS
MAX = 5.081

Bureau
Token Bureau
Feature activation+0.106
of
Token of
Feature activation+0.329
Pr
Token Pr
Feature activation+0.000
isons
Tokenisons
Feature activation+1.354
announced
Token announced
Feature activation+2.463
the
Token the
Feature activation+5.081
death
Token death
Feature activation+0.000
the
Token the
Feature activation+0.000
following
Token following
Feature activation+0.000
day
Token day
Feature activation+0.058
,
Token,
Feature activation+0.287
and
Token and
Feature activation+0.760
our
Token our
Feature activation+1.729
thoughts
Token thoughts
Feature activation+0.031
are
Token are
Feature activation+0.340
with
Token with
Feature activation+4.298
the
Token the
Feature activation+4.509
cyclist
Token cyclist
Feature activation+2.454
's
Token's
Feature activation+3.706
family
Token family
Feature activation+0.132
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
accident
Token accident
Feature activation+0.007
and
Token and
Feature activation+0.760
our
Token our
Feature activation+1.729
thoughts
Token thoughts
Feature activation+0.031
are
Token are
Feature activation+0.340
with
Token with
Feature activation+4.298
the
Token the
Feature activation+4.509
cyclist
Token cyclist
Feature activation+2.454
's
Token's
Feature activation+3.706
family
Token family
Feature activation+0.132
."
Token."
Feature activation+0.000
held
Token held
Feature activation+0.000
accountable
Token accountable
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
death
Token death
Feature activation+0.345
he
Token he
Feature activation+4.124
caused
Token caused
Feature activation+1.329
by
Token by
Feature activation+0.439
negligence
Token negligence
Feature activation+0.259
as
Token as
Feature activation+0.311
Milwaukee
Token Milwaukee
Feature activation+1.079
evening
Token evening
Feature activation+0.000
.
Token.
Feature activation+1.128
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+2.708
The
TokenThe
Feature activation+3.876
police
Token police
Feature activation+0.563
are
Token are
Feature activation+0.729
investigating
Token investigating
Feature activation+1.548
the
Token the
Feature activation+3.720
accident
Token accident
Feature activation+0.007
last
Token last
Feature activation+0.258
20
Token 20
Feature activation+0.000
minutes
Token minutes
Feature activation+0.108
leading
Token leading
Feature activation+0.000
into
Token into
Feature activation+0.470
his
Token his
Feature activation+3.854
unt
Token unt
Feature activation+3.479
ime
Tokenime
Feature activation+0.241
ly
Tokenly
Feature activation+1.810
demise
Token demise
Feature activation+0.000
hoping
Token hoping
Feature activation+0.000
Marines
Token Marines
Feature activation+0.833
Reported
Token Reported
Feature activation+0.622
ly
Tokenly
Feature activation+0.467
Ur
Token Ur
Feature activation+0.392
inating
Tokeninating
Feature activation+1.521
On
Token On
Feature activation+3.761
Dead
Token Dead
Feature activation+0.851
Taliban
Token Taliban
Feature activation+0.233
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Secret
TokenSecret
Feature activation+0.270
beyond
Token beyond
Feature activation+2.929
categories
Token categories
Feature activation+0.000
such
Token such
Feature activation+0.446
as
Token as
Feature activation+3.514
"
Token "
Feature activation+2.964
natural
Tokennatural
Feature activation+3.739
cause
Token cause
Feature activation+0.813
,"
Token,"
Feature activation+1.023
"
Token "
Feature activation+1.642
hom
Tokenhom
Feature activation+2.270
icide
Tokenicide
Feature activation+0.168
"
Token"
Feature activation+2.708
The
TokenThe
Feature activation+3.876
police
Token police
Feature activation+0.563
are
Token are
Feature activation+0.729
investigating
Token investigating
Feature activation+1.548
the
Token the
Feature activation+3.720
accident
Token accident
Feature activation+0.007
and
Token and
Feature activation+0.760
our
Token our
Feature activation+1.729
thoughts
Token thoughts
Feature activation+0.031
are
Token are
Feature activation+0.340
have
Token have
Feature activation+0.531
reportedly
Token reportedly
Feature activation+0.576
died
Token died
Feature activation+0.000
.
Token.
Feature activation+0.241
Of
Token Of
Feature activation+0.000
those
Token those
Feature activation+3.706
,
Token,
Feature activation+0.971
15
Token 15
Feature activation+1.967
committed
Token committed
Feature activation+1.844
suicide
Token suicide
Feature activation+0.000
or
Token or
Feature activation+0.842
thoughts
Token thoughts
Feature activation+0.031
are
Token are
Feature activation+0.340
with
Token with
Feature activation+4.298
the
Token the
Feature activation+4.509
cyclist
Token cyclist
Feature activation+2.454
's
Token's
Feature activation+3.706
family
Token family
Feature activation+0.132
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.949
enough
Token enough
Feature activation+0.011
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.226
Before
TokenBefore
Feature activation+1.862
her
Token her
Feature activation+3.643
death
Token death
Feature activation+0.000
,
Token,
Feature activation+0.000
Ms
Token Ms
Feature activation+0.000
N
Token N
Feature activation+0.000
att
Tokenatt
Feature activation+0.000
repeated
Token repeated
Feature activation+0.000
.
Token.
Feature activation+0.031
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Of
TokenOf
Feature activation+0.507
the
Token the
Feature activation+3.625
information
Token information
Feature activation+0.000
concerning
Token concerning
Feature activation+1.770
the
Token the
Feature activation+3.163
deaths
Token deaths
Feature activation+0.290
of
Token of
Feature activation+1.132
certificates
Token certificates
Feature activation+0.186
typically
Token typically
Feature activation+1.423
contain
Token contain
Feature activation+1.315
little
Token little
Feature activation+0.027
information
Token information
Feature activation+0.000
about
Token about
Feature activation+3.552
manner
Token manner
Feature activation+0.573
of
Token of
Feature activation+3.096
death
Token death
Feature activation+0.354
beyond
Token beyond
Feature activation+2.929
categories
Token categories
Feature activation+0.000
now
Token now
Feature activation+0.000
dead
Token dead
Feature activation+0.000
(
Token (
Feature activation+0.008
if
Tokenif
Feature activation+0.000
not
Token not
Feature activation+2.516
quite
Token quite
Feature activation+3.533
buried
Token buried
Feature activation+0.000
).
Token).
Feature activation+0.176
Obama
Token Obama
Feature activation+0.926
and
Token and
Feature activation+0.000
Secretary
Token Secretary
Feature activation+0.000
ales
Tokenales
Feature activation+0.000
e
Tokene
Feature activation+1.305
teacher
Token teacher
Feature activation+0.617
in
Token in
Feature activation+0.000
Qatar
Token Qatar
Feature activation+0.000
,
Token,
Feature activation+3.532
who
Token who
Feature activation+1.763
was
Token was
Feature activation+1.648
imprisoned
Token imprisoned
Feature activation+0.000
for
Token for
Feature activation+0.000
12
Token 12
Feature activation+0.000
campus
Token campus
Feature activation+0.634
earlier
Token earlier
Feature activation+0.000
in
Token in
Feature activation+0.125
the
Token the
Feature activation+0.000
week
Token week
Feature activation+0.847
and
Token and
Feature activation+3.531
eventually
Token eventually
Feature activation+0.649
had
Token had
Feature activation+0.674
to
Token to
Feature activation+0.786
undergo
Token undergo
Feature activation+1.116
emergency
Token emergency
Feature activation+0.019
reaction
Token reaction
Feature activation+0.000
should
Token should
Feature activation+0.399
the
Token the
Feature activation+0.610
verdict
Token verdict
Feature activation+0.000
in
Token in
Feature activation+2.528
the
Token the
Feature activation+3.522
George
Token George
Feature activation+0.316
Zimmerman
Token Zimmerman
Feature activation+0.380
case
Token case
Feature activation+0.000
end
Token end
Feature activation+0.041
differently
Token differently
Feature activation+0.000
of
Token of
Feature activation+3.096
death
Token death
Feature activation+0.354
beyond
Token beyond
Feature activation+2.929
categories
Token categories
Feature activation+0.000
such
Token such
Feature activation+0.446
as
Token as
Feature activation+3.514
"
Token "
Feature activation+2.964
natural
Tokennatural
Feature activation+3.739
cause
Token cause
Feature activation+0.813
,"
Token,"
Feature activation+1.023
"
Token "
Feature activation+1.642
med
Token med
Feature activation+1.139
ics
Tokenics
Feature activation+1.030
could
Token could
Feature activation+1.578
transport
Token transport
Feature activation+2.647
him
Token him
Feature activation+2.784
to
Token to
Feature activation+3.500
a
Token a
Feature activation+2.404
hospital
Token hospital
Feature activation+0.564
.
Token.
Feature activation+0.000
Sch
TokenSch
Feature activation+0.000
affer
Tokenaffer
Feature activation+0.429

Top DFA by src position
MAX = 4.869

<|endoftext|>
Token<|endoftext|>
Feature activation+0.097
Top resid features:
found
Token found
Feature activation-0.001
Top resid features:
dead
Token dead
Feature activation+4.869
Top resid features:
in
Token in
Feature activation+0.336
Top resid features:
his
Token his
Feature activation+0.109
Top resid features:
cell
Token cell
Feature activation+0.051
Top resid features:
at
Token at
Feature activation+0.055
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
confirm
Token confirm
Feature activation+0.000
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
a
Token a
Feature activation+0.001
Top resid features:
cyclist
Token cyclist
Feature activation+0.010
Top resid features:
tragically
Token tragically
Feature activation+1.468
Top resid features:
died
Token died
Feature activation+3.371
Top resid features:
as
Token as
Feature activation+0.040
Top resid features:
a
Token a
Feature activation+0.006
Top resid features:
result
Token result
Feature activation+0.011
Top resid features:
of
Token of
Feature activation+0.026
Top resid features:
a
Token a
Feature activation+0.009
Top resid features:
confirm
Token confirm
Feature activation+0.000
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
a
Token a
Feature activation+0.001
Top resid features:
cyclist
Token cyclist
Feature activation-0.000
Top resid features:
tragically
Token tragically
Feature activation+2.212
Top resid features:
died
Token died
Feature activation+2.558
Top resid features:
as
Token as
Feature activation+0.016
Top resid features:
a
Token a
Feature activation+0.006
Top resid features:
result
Token result
Feature activation+0.013
Top resid features:
of
Token of
Feature activation+0.019
Top resid features:
a
Token a
Feature activation+0.010
Top resid features:
being
Token being
Feature activation+0.002
Top resid features:
held
Token held
Feature activation-0.000
Top resid features:
accountable
Token accountable
Feature activation-0.027
Top resid features:
for
Token for
Feature activation+0.007
Top resid features:
a
Token a
Feature activation+0.005
Top resid features:
death
Token death
Feature activation+4.352
Top resid features:
he
Token he
Feature activation+0.232
Top resid features:
caused
Token caused
Feature activation+0.000
Top resid features:
by
Token by
Feature activation+0.000
Top resid features:
negligence
Token negligence
Feature activation+0.000
Top resid features:
as
Token as
Feature activation+0.000
Top resid features:
confirm
Token confirm
Feature activation+0.000
Top resid features:
that
Token that
Feature activation-0.000
Top resid features:
a
Token a
Feature activation+0.003
Top resid features:
cyclist
Token cyclist
Feature activation+0.016
Top resid features:
tragically
Token tragically
Feature activation+0.591
Top resid features:
died
Token died
Feature activation+3.386
Top resid features:
as
Token as
Feature activation+0.041
Top resid features:
a
Token a
Feature activation+0.013
Top resid features:
result
Token result
Feature activation+0.031
Top resid features:
of
Token of
Feature activation+0.084
Top resid features:
a
Token a
Feature activation+0.019
Top resid features:
hopes
Token hopes
Feature activation-0.000
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
escape
Token escape
Feature activation-0.026
Top resid features:
,
Token,
Feature activation+0.011
Top resid features:
he
Token he
Feature activation+0.000
Top resid features:
dies
Token dies
Feature activation+3.397
Top resid features:
.
Token.
Feature activation+0.035
Top resid features:
A
Token A
Feature activation+0.003
Top resid features:
gr
Token gr
Feature activation-0.003
Top resid features:
is
Tokenis
Feature activation-0.000
Top resid features:
ly
Tokenly
Feature activation+0.037
Top resid features:
on
Token on
Feature activation+0.001
Top resid features:
what
Token what
Feature activation+0.020
Top resid features:
appear
Token appear
Feature activation-0.001
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
be
Token be
Feature activation+0.003
Top resid features:
dead
Token dead
Feature activation+4.114
Top resid features:
Afghans
Token Afghans
Feature activation-0.042
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
saying
Token saying
Feature activation+0.000
Top resid features:
she
Token she
Feature activation+0.000
Top resid features:
would
Token would
Feature activation+0.000
Top resid features:
little
Token little
Feature activation-0.000
Top resid features:
information
Token information
Feature activation-0.001
Top resid features:
about
Token about
Feature activation+0.002
Top resid features:
manner
Token manner
Feature activation+0.009
Top resid features:
of
Token of
Feature activation+0.014
Top resid features:
death
Token death
Feature activation+1.567
Top resid features:
beyond
Token beyond
Feature activation+0.005
Top resid features:
categories
Token categories
Feature activation+0.002
Top resid features:
such
Token such
Feature activation+0.016
Top resid features:
as
Token as
Feature activation+0.012
Top resid features:
"
Token "
Feature activation+0.053
Top resid features:
confirm
Token confirm
Feature activation+0.001
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
a
Token a
Feature activation+0.001
Top resid features:
cyclist
Token cyclist
Feature activation+0.007
Top resid features:
tragically
Token tragically
Feature activation+0.572
Top resid features:
died
Token died
Feature activation+3.075
Top resid features:
as
Token as
Feature activation+0.053
Top resid features:
a
Token a
Feature activation+0.015
Top resid features:
result
Token result
Feature activation+0.022
Top resid features:
of
Token of
Feature activation+0.093
Top resid features:
a
Token a
Feature activation+0.018
Top resid features:
77
Token77
Feature activation+0.043
Top resid features:
asylum
Token asylum
Feature activation-0.019
Top resid features:
seekers
Token seekers
Feature activation-0.017
Top resid features:
have
Token have
Feature activation+0.004
Top resid features:
reportedly
Token reportedly
Feature activation+0.012
Top resid features:
died
Token died
Feature activation+3.541
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
Of
Token Of
Feature activation-0.014
Top resid features:
those
Token those
Feature activation-0.009
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
15
Token 15
Feature activation+0.000
Top resid features:
confirm
Token confirm
Feature activation+0.001
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
cyclist
Token cyclist
Feature activation-0.001
Top resid features:
tragically
Token tragically
Feature activation+1.181
Top resid features:
died
Token died
Feature activation+2.796
Top resid features:
as
Token as
Feature activation+0.036
Top resid features:
a
Token a
Feature activation+0.016
Top resid features:
result
Token result
Feature activation+0.021
Top resid features:
of
Token of
Feature activation+0.030
Top resid features:
a
Token a
Feature activation+0.011
Top resid features:
Nick
Token Nick
Feature activation+0.000
Top resid features:
Xen
Token Xen
Feature activation+0.003
Top resid features:
ophon
Tokenophon
Feature activation+0.003
Top resid features:
and
Token and
Feature activation-0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
dead
Token dead
Feature activation+2.358
Top resid features:
woman
Token woman
Feature activation+0.021
Top resid features:
's
Token's
Feature activation+0.003
Top resid features:
mother
Token mother
Feature activation-0.012
Top resid features:
have
Token have
Feature activation+0.003
Top resid features:
criticised
Token criticised
Feature activation+0.001
Top resid features:
you
Token you
Feature activation-0.001
Top resid features:
care
Token care
Feature activation+0.003
Top resid features:
whether
Token whether
Feature activation-0.002
Top resid features:
four
Token four
Feature activation+0.004
Top resid features:
Americans
Token Americans
Feature activation+0.075
Top resid features:
died
Token died
Feature activation+3.870
Top resid features:
?
Token?
Feature activation+0.006
Top resid features:
The
Token The
Feature activation-0.000
Top resid features:
reasons
Token reasons
Feature activation-0.004
Top resid features:
for
Token for
Feature activation+0.007
Top resid features:
that
Token that
Feature activation+0.004
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
public
Token public
Feature activation-0.001
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
are
Token are
Feature activation-0.002
Top resid features:
actually
Token actually
Feature activation+0.003
Top resid features:
dead
Token dead
Feature activation+2.687
Top resid features:
,
Token,
Feature activation+0.019
Top resid features:
are
Token are
Feature activation-0.001
Top resid features:
legendary
Token legendary
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Death
Token Death
Feature activation+1.525
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.101
Top resid features:
is
Token is
Feature activation-0.003
Top resid features:
now
Token now
Feature activation-0.009
Top resid features:
dead
Token dead
Feature activation+3.857
Top resid features:
(
Token (
Feature activation+0.063
Top resid features:
if
Tokenif
Feature activation+0.052
Top resid features:
not
Token not
Feature activation+0.017
Top resid features:
quite
Token quite
Feature activation+0.043
Top resid features:
buried
Token buried
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.090
Top resid features:
died
Token died
Feature activation+2.940
Top resid features:
in
Token in
Feature activation+0.515
Top resid features:
the
Token the
Feature activation+0.036
Top resid features:
country
Token country
Feature activation+0.013
Top resid features:
in
Token in
Feature activation+0.040
Top resid features:
2012
Token 2012
Feature activation+0.044
Top resid features:
Reve
Token Reve
Feature activation-0.001
Top resid features:
ille
Tokenille
Feature activation-0.004
Top resid features:
VII
Token VII
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
has
Token has
Feature activation+0.001
Top resid features:
died
Token died
Feature activation+3.430
Top resid features:
,
Token,
Feature activation+0.491
Top resid features:
according
Token according
Feature activation+0.004
Top resid features:
to
Token to
Feature activation-0.004
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
university
Token university
Feature activation+0.012
Top resid features:
member
Token member
Feature activation+0.023
Top resid features:
be
Token be
Feature activation+0.002
Top resid features:
acquitted
Token acquitted
Feature activation+0.044
Top resid features:
in
Token in
Feature activation+0.009
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
murder
Token murder
Feature activation+3.712
Top resid features:
trial
Token trial
Feature activation+0.006
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Task
TokenTask
Feature activation+0.007
Top resid features:
little
Token little
Feature activation-0.001
Top resid features:
information
Token information
Feature activation-0.007
Top resid features:
about
Token about
Feature activation+0.004
Top resid features:
manner
Token manner
Feature activation+0.015
Top resid features:
of
Token of
Feature activation+0.034
Top resid features:
death
Token death
Feature activation+2.978
Top resid features:
beyond
Token beyond
Feature activation+0.001
Top resid features:
categories
Token categories
Feature activation+0.032
Top resid features:
such
Token such
Feature activation+0.012
Top resid features:
as
Token as
Feature activation-0.008
Top resid features:
"
Token "
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
suspects
Token suspects
Feature activation+0.014
Top resid features:
,
Token,
Feature activation+0.005
Top resid features:
and
Token and
Feature activation+0.030
Top resid features:
died
Token died
Feature activation+3.170
Top resid features:
before
Token before
Feature activation+0.014
Top resid features:
med
Token med
Feature activation+0.020
Top resid features:
ics
Tokenics
Feature activation+0.001
Top resid features:
could
Token could
Feature activation-0.003
Top resid features:
transport
Token transport
Feature activation+0.003
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.02

Head 2: 0.51

Head 3: 0.08

Head 4: 0.04

Head 5: 0.08

Head 6: 0.05

Head 7: 0.06

Head 8: 0.03

Head 9: 0.03

Head 10: 0.02

Head 11: 0.01

Positive logits

death7.00

funeral6.84

coroner6.73

deaths6.61

autopsy6.60

death6.48

deceased6.43

crem6.34

murder6.28

Deaths6.27

burial6.20

homicide6.16

murdered6.14

Death6.09

corpse6.07

murderer5.91

corpses5.77

killers5.75

Death5.72

killed5.72

Negative logits

NFC-2.99

Libre-2.94

issance-2.88

renegoti-2.76

Entreprene-2.73

Plugin-2.70

vouchers-2.65

ECO-2.64

dayName-2.64

itech-2.61

hops-2.57

borrowers-2.56

-2.55

voucher-2.52

endors-2.51

otiation-2.51

Parser-2.50

SEA-2.49

Integration-2.47

Expansion-2.47

INTERVAL 4.573 - 5.081
CONTAINS 0.000%

Bureau
Token Bureau
Feature activation+0.106
of
Token of
Feature activation+0.329
Pr
Token Pr
Feature activation+0.000
isons
Tokenisons
Feature activation+1.354
announced
Token announced
Feature activation+2.463
the
Token the
Feature activation+5.081
death
Token death
Feature activation+0.000
the
Token the
Feature activation+0.000
following
Token following
Feature activation+0.000
day
Token day
Feature activation+0.058
,
Token,
Feature activation+0.287

INTERVAL 4.065 - 4.573
CONTAINS 0.000%

accident
Token accident
Feature activation+0.007
and
Token and
Feature activation+0.760
our
Token our
Feature activation+1.729
thoughts
Token thoughts
Feature activation+0.031
are
Token are
Feature activation+0.340
with
Token with
Feature activation+4.298
the
Token the
Feature activation+4.509
cyclist
Token cyclist
Feature activation+2.454
's
Token's
Feature activation+3.706
family
Token family
Feature activation+0.132
."
Token."
Feature activation+0.000
and
Token and
Feature activation+0.760
our
Token our
Feature activation+1.729
thoughts
Token thoughts
Feature activation+0.031
are
Token are
Feature activation+0.340
with
Token with
Feature activation+4.298
the
Token the
Feature activation+4.509
cyclist
Token cyclist
Feature activation+2.454
's
Token's
Feature activation+3.706
family
Token family
Feature activation+0.132
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
held
Token held
Feature activation+0.000
accountable
Token accountable
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
death
Token death
Feature activation+0.345
he
Token he
Feature activation+4.124
caused
Token caused
Feature activation+1.329
by
Token by
Feature activation+0.439
negligence
Token negligence
Feature activation+0.259
as
Token as
Feature activation+0.311
Milwaukee
Token Milwaukee
Feature activation+1.079

INTERVAL 3.557 - 4.065
CONTAINS 0.001%

last
Token last
Feature activation+0.258
20
Token 20
Feature activation+0.000
minutes
Token minutes
Feature activation+0.108
leading
Token leading
Feature activation+0.000
into
Token into
Feature activation+0.470
his
Token his
Feature activation+3.854
unt
Token unt
Feature activation+3.479
ime
Tokenime
Feature activation+0.241
ly
Tokenly
Feature activation+1.810
demise
Token demise
Feature activation+0.000
hoping
Token hoping
Feature activation+0.000
have
Token have
Feature activation+0.531
reportedly
Token reportedly
Feature activation+0.576
died
Token died
Feature activation+0.000
.
Token.
Feature activation+0.241
Of
Token Of
Feature activation+0.000
those
Token those
Feature activation+3.706
,
Token,
Feature activation+0.971
15
Token 15
Feature activation+1.967
committed
Token committed
Feature activation+1.844
suicide
Token suicide
Feature activation+0.000
or
Token or
Feature activation+0.842
beyond
Token beyond
Feature activation+2.929
categories
Token categories
Feature activation+0.000
such
Token such
Feature activation+0.446
as
Token as
Feature activation+3.514
"
Token "
Feature activation+2.964
natural
Tokennatural
Feature activation+3.739
cause
Token cause
Feature activation+0.813
,"
Token,"
Feature activation+1.023
"
Token "
Feature activation+1.642
hom
Tokenhom
Feature activation+2.270
icide
Tokenicide
Feature activation+0.168
"
Token"
Feature activation+2.708
The
TokenThe
Feature activation+3.876
police
Token police
Feature activation+0.563
are
Token are
Feature activation+0.729
investigating
Token investigating
Feature activation+1.548
the
Token the
Feature activation+3.720
accident
Token accident
Feature activation+0.007
and
Token and
Feature activation+0.760
our
Token our
Feature activation+1.729
thoughts
Token thoughts
Feature activation+0.031
are
Token are
Feature activation+0.340
repeated
Token repeated
Feature activation+0.000
.
Token.
Feature activation+0.031
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Of
TokenOf
Feature activation+0.507
the
Token the
Feature activation+3.625
information
Token information
Feature activation+0.000
concerning
Token concerning
Feature activation+1.770
the
Token the
Feature activation+3.163
deaths
Token deaths
Feature activation+0.290
of
Token of
Feature activation+1.132

INTERVAL 3.049 - 3.557
CONTAINS 0.003%

ľ
Tokenľ
Feature activation+0.000
Of
TokenOf
Feature activation+0.507
the
Token the
Feature activation+3.625
information
Token information
Feature activation+0.000
concerning
Token concerning
Feature activation+1.770
the
Token the
Feature activation+3.163
deaths
Token deaths
Feature activation+0.290
of
Token of
Feature activation+1.132
four
Token four
Feature activation+1.646
brave
Token brave
Feature activation+0.741
Americans
Token Americans
Feature activation+0.871
maker
Tokenmaker
Feature activation+0.000
had
Token had
Feature activation+0.000
not
Token not
Feature activation+0.000
marked
Token marked
Feature activation+0.277
their
Token their
Feature activation+1.018
the
Token the
Feature activation+3.239
station
Token station
Feature activation+0.273
as
Token as
Feature activation+2.130
a
Token a
Feature activation+2.307
Pokemon
Token Pokemon
Feature activation+0.281
Go
Token Go
Feature activation+0.048
now
Token now
Feature activation+0.000
dead
Token dead
Feature activation+0.000
(
Token (
Feature activation+0.008
if
Tokenif
Feature activation+0.000
not
Token not
Feature activation+2.516
quite
Token quite
Feature activation+3.533
buried
Token buried
Feature activation+0.000
).
Token).
Feature activation+0.176
Obama
Token Obama
Feature activation+0.926
and
Token and
Feature activation+0.000
Secretary
Token Secretary
Feature activation+0.000
ales
Tokenales
Feature activation+0.000
e
Tokene
Feature activation+1.305
teacher
Token teacher
Feature activation+0.617
in
Token in
Feature activation+0.000
Qatar
Token Qatar
Feature activation+0.000
,
Token,
Feature activation+3.532
who
Token who
Feature activation+1.763
was
Token was
Feature activation+1.648
imprisoned
Token imprisoned
Feature activation+0.000
for
Token for
Feature activation+0.000
12
Token 12
Feature activation+0.000
were
Token were
Feature activation+1.522
unsuccessful
Token unsuccessful
Feature activation+0.000
in
Token in
Feature activation+2.465
their
Token their
Feature activation+0.125
attempts
Token attempts
Feature activation+0.066
to
Token to
Feature activation+3.065
revive
Token revive
Feature activation+1.223
Gil
Token Gil
Feature activation+0.000
lette
Tokenlette
Feature activation+0.141
from
Token from
Feature activation+1.498
a
Token a
Feature activation+0.863

INTERVAL 2.541 - 3.049
CONTAINS 0.007%

one
Token one
Feature activation+1.180
of
Token of
Feature activation+0.469
my
Token my
Feature activation+0.000
favorite
Token favorite
Feature activation+0.413
childhood
Token childhood
Feature activation+0.203
idols
Token idols
Feature activation+2.821
quite
Token quite
Feature activation+0.971
possibly
Token possibly
Feature activation+2.456
used
Token used
Feature activation+3.302
his
Token his
Feature activation+2.351
bisexual
Token bisexual
Feature activation+0.000
ac
Tokenac
Feature activation+0.000
Federal
Token Federal
Feature activation+0.000
Detention
Token Detention
Feature activation+0.000
Center
Token Center
Feature activation+0.000
.
Token.
Feature activation+0.878
The
Token The
Feature activation+2.561
U
Token U
Feature activation+0.932
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+1.037
Bureau
Token Bureau
Feature activation+0.106
city
Token city
Feature activation+1.299
medical
Token medical
Feature activation+0.000
examiner
Token examiner
Feature activation+0.000
's
Token's
Feature activation+0.339
office
Token office
Feature activation+0.564
;
Token;
Feature activation+2.727
authorities
Token authorities
Feature activation+0.402
hoped
Token hoped
Feature activation+2.116
to
Token to
Feature activation+2.217
use
Token use
Feature activation+2.575
the
Token the
Feature activation+2.903
Tuesday
Token Tuesday
Feature activation+0.000
morning
Token morning
Feature activation+0.000
.
Token.
Feature activation+0.073
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.510
Her
TokenHer
Feature activation+2.999
family
Token family
Feature activation+0.000
paid
Token paid
Feature activation+0.680
tribute
Token tribute
Feature activation+0.000
to
Token to
Feature activation+0.000
her
Token her
Feature activation+1.365
CBC
Token CBC
Feature activation+0.000
News
TokenNews
Feature activation+0.000
also
Token also
Feature activation+0.000
reports
Token reports
Feature activation+0.000
that
Token that
Feature activation+1.009
the
Token the
Feature activation+2.987
benign
Token benign
Feature activation+3.377
ly
Tokenly
Feature activation+0.374
besieged
Token besieged
Feature activation+1.950
occupants
Token occupants
Feature activation+0.796
of
Token of
Feature activation+0.932

INTERVAL 2.032 - 2.541
CONTAINS 0.013%

not
Token not
Feature activation+0.751
doctors
Token doctors
Feature activation+0.000
who
Token who
Feature activation+0.998
declared
Token declared
Feature activation+1.415
a
Token a
Feature activation+1.572
person
Token person
Feature activation+2.057
dead
Token dead
Feature activation+0.351
.
Token.
Feature activation+0.000
When
Token When
Feature activation+0.020
in
Token in
Feature activation+0.173
doubt
Token doubt
Feature activation+0.000
head
Token head
Feature activation+0.537
,
Token,
Feature activation+1.354
said
Token said
Feature activation+2.939
Ed
Token Ed
Feature activation+0.972
Winter
Token Winter
Feature activation+0.000
,
Token,
Feature activation+2.101
assistant
Token assistant
Feature activation+2.908
chief
Token chief
Feature activation+3.081
of
Token of
Feature activation+3.003
the
Token the
Feature activation+2.838
Los
Token Los
Feature activation+0.308
into
Token into
Feature activation+0.341
a
Token a
Feature activation+0.436
wood
Token wood
Feature activation+0.000
ch
Token ch
Feature activation+0.000
ipper
Tokenipper
Feature activation+1.076
and
Token and
Feature activation+2.313
was
Token was
Feature activation+2.290
killed
Token killed
Feature activation+0.111
Monday
Token Monday
Feature activation+0.281
in
Token in
Feature activation+0.303
Dav
Token Dav
Feature activation+0.000
took
Token took
Feature activation+0.000
her
Token her
Feature activation+0.000
own
Token own
Feature activation+0.000
life
Token life
Feature activation+0.000
.
Token.
Feature activation+0.000
She
Token She
Feature activation+2.111
killed
Token killed
Feature activation+0.000
herself
Token herself
Feature activation+0.032
.
Token.
Feature activation+0.213
A
Token A
Feature activation+1.597
community
Token community
Feature activation+1.060
death
Token death
Feature activation+0.000
Friday
Token Friday
Feature activation+0.000
morning
Token morning
Feature activation+0.119
was
Token was
Feature activation+3.255
confirmed
Token confirmed
Feature activation+0.961
by
Token by
Feature activation+2.198
police
Token police
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.045
Witness
TokenWitness
Feature activation+0.000

INTERVAL 1.524 - 2.032
CONTAINS 0.032%

ias
Tokenias
Feature activation+0.498
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
body
Token body
Feature activation+0.506
had
Token had
Feature activation+0.063
been
Token been
Feature activation+1.800
buried
Token buried
Feature activation+0.144
in
Token in
Feature activation+0.000
sand
Token sand
Feature activation+0.028
off
Token off
Feature activation+0.000
Cul
Token Cul
Feature activation+0.000
ĵ
Tokenĵ
Feature activation+0.000
If
TokenIf
Feature activation+0.000
we
Token we
Feature activation+0.000
die
Token die
Feature activation+0.033
,
Token,
Feature activation+0.599
we
Token we
Feature activation+1.962
want
Token want
Feature activation+1.185
people
Token people
Feature activation+0.080
to
Token to
Feature activation+1.022
accept
Token accept
Feature activation+0.192
it
Token it
Feature activation+0.189
reports
Token reports
Feature activation+0.000
that
Token that
Feature activation+1.009
the
Token the
Feature activation+2.987
benign
Token benign
Feature activation+3.377
ly
Tokenly
Feature activation+0.374
besieged
Token besieged
Feature activation+1.950
occupants
Token occupants
Feature activation+0.796
of
Token of
Feature activation+0.932
an
Token an
Feature activation+1.299
Australian
Token Australian
Feature activation+0.606
police
Token police
Feature activation+0.086
âĢĶ
Token âĢĶ
Feature activation+0.167
that
Token that
Feature activation+0.553
if
Token if
Feature activation+0.753
a
Token a
Feature activation+1.215
person
Token person
Feature activation+1.507
is
Token is
Feature activation+1.928
in
Token in
Feature activation+0.853
emergency
Token emergency
Feature activation+0.000
distress
Token distress
Feature activation+0.278
and
Token and
Feature activation+0.897
calls
Token calls
Feature activation+0.366
all
Token all
Feature activation+0.000
the
Token the
Feature activation+0.000
tra
Token tra
Feature activation+0.000
ppings
Tokenppings
Feature activation+0.000
of
Token of
Feature activation+0.475
a
Token a
Feature activation+1.627
murder
Token murder
Feature activation+0.316
mystery
Token mystery
Feature activation+0.000
,
Token,
Feature activation+0.135
albeit
Token albeit
Feature activation+0.798
one
Token one
Feature activation+1.078

INTERVAL 1.016 - 1.524
CONTAINS 0.081%

been
Token been
Feature activation+0.713
ruled
Token ruled
Feature activation+1.924
a
Token a
Feature activation+2.093
homicide
Token homicide
Feature activation+0.061
,
Token,
Feature activation+1.266
a
Token a
Feature activation+1.283
coroner
Token coroner
Feature activation+0.009
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.771
s
Tokens
Feature activation+0.460
official
Token official
Feature activation+0.329
non
Token non
Feature activation+0.000
-
Token-
Feature activation+0.000
dead
Tokendead
Feature activation+0.000
ly
Tokenly
Feature activation+0.261
force
Token force
Feature activation+0.000
against
Token against
Feature activation+1.021
a
Token a
Feature activation+0.872
child
Token child
Feature activation+0.474
by
Token by
Feature activation+0.079
a
Token a
Feature activation+0.093
parent
Token parent
Feature activation+0.000
clusive
Tokenclusive
Feature activation+0.000
and
Token and
Feature activation+0.890
awaits
Token awaits
Feature activation+1.221
the
Token the
Feature activation+1.271
results
Token results
Feature activation+0.000
of
Token of
Feature activation+1.209
toxic
Token toxic
Feature activation+0.186
ology
Tokenology
Feature activation+0.000
tests
Token tests
Feature activation+0.000
,
Token,
Feature activation+0.416
said
Token said
Feature activation+1.390
Department
Token Department
Feature activation+0.000
to
Token to
Feature activation+1.684
find
Token find
Feature activation+0.918
and
Token and
Feature activation+1.621
recover
Token recover
Feature activation+1.194
his
Token his
Feature activation+1.427
body
Token body
Feature activation+0.129
.
Token.
Feature activation+0.000
After
Token After
Feature activation+0.868
six
Token six
Feature activation+0.794
men
Token men
Feature activation+0.091
What
TokenWhat
Feature activation+0.473
a
Token a
Feature activation+0.368
tragedy
Token tragedy
Feature activation+0.000
!
Token!
Feature activation+0.900
Now
Token Now
Feature activation+0.000
his
Token his
Feature activation+1.192
kindness
Token kindness
Feature activation+0.000
is
Token is
Feature activation+0.000
gone
Token gone
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 0.508 - 1.016
CONTAINS 0.233%

require
Token require
Feature activation+0.231
a
Token a
Feature activation+0.725
deadly
Token deadly
Feature activation+0.017
weapon
Token weapon
Feature activation+0.003
to
Token to
Feature activation+1.190
kill
Token kill
Feature activation+0.587
an
Token an
Feature activation+0.865
innocent
Token innocent
Feature activation+0.400
victim
Token victim
Feature activation+0.002
.
Token.
Feature activation+0.672
Ċ
TokenĊ
Feature activation+0.000
iously
Tokeniously
Feature activation+0.507
dumped
Token dumped
Feature activation+0.129
with
Token with
Feature activation+0.625
nothing
Token nothing
Feature activation+0.000
more
Token more
Feature activation+0.000
than
Token than
Feature activation+0.636
a
Token a
Feature activation+0.088
flashlight
Token flashlight
Feature activation+0.000
(
Token (
Feature activation+0.000
one
Tokenone
Feature activation+0.000
of
Token of
Feature activation+0.000
he
Token he
Feature activation+0.000
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
This
TokenThis
Feature activation+0.930
sends
Token sends
Feature activation+0.000
a
Token a
Feature activation+0.000
disturbing
Token disturbing
Feature activation+0.000
message
Token message
Feature activation+0.000
that
Token that
Feature activation+0.256
photo
Token photo
Feature activation+0.000
of
Token of
Feature activation+0.686
Omar
Token Omar
Feature activation+0.000
Grant
Token Grant
Feature activation+0.011
,
Token,
Feature activation+0.182
who
Token who
Feature activation+0.803
led
Token led
Feature activation+0.000
his
Token his
Feature activation+0.000
blind
Token blind
Feature activation+0.000
colleague
Token colleague
Feature activation+0.051
off
Token off
Feature activation+0.000
agent
Token agent
Feature activation+0.220
promptly
Token promptly
Feature activation+0.761
sued
Token sued
Feature activation+0.106
the
Token the
Feature activation+0.762
City
Token City
Feature activation+0.147
claiming
Token claiming
Feature activation+0.667
âĢĶ
Token âĢĶ
Feature activation+0.000
rightfully
Token rightfully
Feature activation+0.000
so
Token so
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.167
that
Token that
Feature activation+0.553

INTERVAL 0.000 - 0.508
CONTAINS 99.630%

these
Token these
Feature activation+0.000
issues
Token issues
Feature activation+0.000
are
Token are
Feature activation+0.000
close
Token close
Feature activation+0.000
to
Token to
Feature activation+0.000
my
Token my
Feature activation+0.000
heart
Token heart
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
the
Token the
Feature activation+0.000
Latin
Token Latin
Feature activation+0.000
American
Token American
Feature activation+0.000
,
Token,
Feature activation+0.000
Latino
Token Latino
Feature activation+0.000
and
Token and
Feature activation+0.000
Caribbean
Token Caribbean
Feature activation+0.000
Studies
Token Studies
Feature activation+0.000
program
Token program
Feature activation+0.000
at
Token at
Feature activation+0.000
Dartmouth
Token Dartmouth
Feature activation+0.000
the
Token the
Feature activation+0.000
parties
Token parties
Feature activation+0.000
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
able
Token able
Feature activation+0.000
to
Token to
Feature activation+0.000
afford
Token afford
Feature activation+0.000
all
Token all
Feature activation+0.000
of
Token of
Feature activation+0.000
their
Token their
Feature activation+0.000
promises
Token promises
Feature activation+0.000
showed
Token showed
Feature activation+0.003
up
Token up
Feature activation+0.000
--
Token --
Feature activation+0.034
a
Token a
Feature activation+0.313
villain
Token villain
Feature activation+0.000
who
Token who
Feature activation+0.125
fe
Token fe
Feature activation+0.000
asted
Tokenasted
Feature activation+0.000
on
Token on
Feature activation+0.299
so
Token so
Feature activation+0.000
-
Token-
Feature activation+0.000
were
Token were
Feature activation+0.000
allowed
Token allowed
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
it
Token it
Feature activation+0.000
for
Token for
Feature activation+0.000
religious
Token religious
Feature activation+0.000
purposes
Token purposes
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
it
Token it
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 H9.2: (feature 2252)

TOP ACTIVATIONS
MAX = 2.583

Here
Token Here
Feature activation+0.030
,
Token,
Feature activation+0.737
a
Token a
Feature activation+0.763
brief
Token brief
Feature activation+0.494
guide
Token guide
Feature activation+0.000
to
Token to
Feature activation+2.583
this
Token this
Feature activation+1.889
condition
Token condition
Feature activation+0.000
:
Token:
Feature activation+1.295
Ċ
TokenĊ
Feature activation+0.225
Ċ
TokenĊ
Feature activation+2.079
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.331
more
Token more
Feature activation+0.482
commonly
Token commonly
Feature activation+0.231
known
Token known
Feature activation+0.098
,
Token,
Feature activation+2.390
Lou
Token Lou
Feature activation+0.901
Geh
Token Geh
Feature activation+0.000
rig
Tokenrig
Feature activation+2.328
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.278
commonly
Token commonly
Feature activation+0.231
known
Token known
Feature activation+0.098
,
Token,
Feature activation+2.390
Lou
Token Lou
Feature activation+0.901
Geh
Token Geh
Feature activation+0.000
rig
Tokenrig
Feature activation+2.328
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.278
s
Tokens
Feature activation+1.150
disease
Token disease
Feature activation+0.000
.
Token.
Feature activation+0.305
this
Token this
Feature activation+1.889
condition
Token condition
Feature activation+0.000
:
Token:
Feature activation+1.295
Ċ
TokenĊ
Feature activation+0.225
Ċ
TokenĊ
Feature activation+2.079
What
TokenWhat
Feature activation+2.124
is
Token is
Feature activation+1.114
S
Token S
Feature activation+0.000
j
Tokenj
Feature activation+0.000
ö
Tokenö
Feature activation+0.000
gren
Tokengren
Feature activation+0.000
syndrome
Token syndrome
Feature activation+0.000
,
Token,
Feature activation+0.450
a
Token a
Feature activation+1.692
little
Token little
Feature activation+0.000
-
Token-
Feature activation+0.000
known
Tokenknown
Feature activation+2.109
disease
Token disease
Feature activation+0.749
with
Token with
Feature activation+1.320
a
Token a
Feature activation+1.521
wide
Token wide
Feature activation+1.177
range
Token range
Feature activation+0.000
to
Token to
Feature activation+2.583
this
Token this
Feature activation+1.889
condition
Token condition
Feature activation+0.000
:
Token:
Feature activation+1.295
Ċ
TokenĊ
Feature activation+0.225
Ċ
TokenĊ
Feature activation+2.079
What
TokenWhat
Feature activation+2.124
is
Token is
Feature activation+1.114
S
Token S
Feature activation+0.000
j
Tokenj
Feature activation+0.000
ö
Tokenö
Feature activation+0.000
so
Token so
Feature activation+0.000
common
Token common
Feature activation+0.012
and
Token and
Feature activation+0.000
important
Token important
Feature activation+0.000
among
Token among
Feature activation+1.051
people
Token people
Feature activation+2.023
with
Token with
Feature activation+1.029
epilepsy
Token epilepsy
Feature activation+0.000
.
Token.
Feature activation+0.390
We
Token We
Feature activation+0.058
hope
Token hope
Feature activation+0.000
restrictions
Token restrictions
Feature activation+0.000
'
Token'
Feature activation+0.016
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.087
The
TokenThe
Feature activation+1.545
American
Token American
Feature activation+2.018
Civil
Token Civil
Feature activation+0.223
Liberties
Token Liberties
Feature activation+0.000
Union
Token Union
Feature activation+0.000
of
Token of
Feature activation+0.000
Ohio
Token Ohio
Feature activation+0.000
things
Token things
Feature activation+0.000
you
Token you
Feature activation+0.000
can
Token can
Feature activation+0.000
do
Token do
Feature activation+0.000
for
Token for
Feature activation+1.030
someone
Token someone
Feature activation+1.983
who
Token who
Feature activation+1.638
has
Token has
Feature activation+1.188
depression
Token depression
Feature activation+0.000
,
Token,
Feature activation+0.105
anxiety
Token anxiety
Feature activation+0.000
more
Token more
Feature activation+0.606
than
Token than
Feature activation+0.000
100
Token 100
Feature activation+0.753
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.788
people
Token people
Feature activation+1.970
,
Token,
Feature activation+0.026
including
Token including
Feature activation+0.792
those
Token those
Feature activation+1.758
affected
Token affected
Feature activation+0.241
and
Token and
Feature activation+0.871
12
Token 12
Feature activation+1.092
to
Token to
Feature activation+0.000
25
Token 25
Feature activation+1.746
%
Token%
Feature activation+1.195
of
Token of
Feature activation+1.261
people
Token people
Feature activation+1.932
in
Token in
Feature activation+0.597
the
Token the
Feature activation+0.633
United
Token United
Feature activation+0.000
States
Token States
Feature activation+1.728
have
Token have
Feature activation+0.987
terrible
Token terrible
Feature activation+0.000
disease
Token disease
Feature activation+0.000
and
Token and
Feature activation+0.665
you
Token you
Feature activation+0.171
had
Token had
Feature activation+0.372
to
Token to
Feature activation+1.926
pick
Token pick
Feature activation+1.678
a
Token a
Feature activation+1.201
doctor
Token doctor
Feature activation+0.000
for
Token for
Feature activation+0.541
Ċ
TokenĊ
Feature activation+0.000
34
Token 34
Feature activation+1.810
patients
Token patients
Feature activation+1.673
published
Token published
Feature activation+0.000
in
Token in
Feature activation+1.244
The
Token The
Feature activation+1.897
American
Token American
Feature activation+1.912
Journal
Token Journal
Feature activation+1.027
of
Token of
Feature activation+1.907
Gast
Token Gast
Feature activation+1.254
ro
Tokenro
Feature activation+0.225
enter
Tokenenter
Feature activation+0.126
Ċ
TokenĊ
Feature activation+0.525
About
TokenAbout
Feature activation+0.185
14
Token 14
Feature activation+0.681
%
Token%
Feature activation+1.060
of
Token of
Feature activation+1.276
adults
Token adults
Feature activation+1.908
are
Token are
Feature activation+1.725
affected
Token affected
Feature activation+0.318
by
Token by
Feature activation+0.066
the
Token the
Feature activation+0.522
condition
Token condition
Feature activation+0.000
published
Token published
Feature activation+0.000
in
Token in
Feature activation+1.244
The
Token The
Feature activation+1.897
American
Token American
Feature activation+1.912
Journal
Token Journal
Feature activation+1.027
of
Token of
Feature activation+1.907
Gast
Token Gast
Feature activation+1.254
ro
Tokenro
Feature activation+0.225
enter
Tokenenter
Feature activation+0.126
ology
Tokenology
Feature activation+0.241
,
Token,
Feature activation+0.525
of
Token of
Feature activation+0.883
34
Token 34
Feature activation+1.810
patients
Token patients
Feature activation+1.673
published
Token published
Feature activation+0.000
in
Token in
Feature activation+1.244
The
Token The
Feature activation+1.897
American
Token American
Feature activation+1.912
Journal
Token Journal
Feature activation+1.027
of
Token of
Feature activation+1.907
Gast
Token Gast
Feature activation+1.254
ro
Tokenro
Feature activation+0.225
,
Token,
Feature activation+0.737
a
Token a
Feature activation+0.763
brief
Token brief
Feature activation+0.494
guide
Token guide
Feature activation+0.000
to
Token to
Feature activation+2.583
this
Token this
Feature activation+1.889
condition
Token condition
Feature activation+0.000
:
Token:
Feature activation+1.295
Ċ
TokenĊ
Feature activation+0.225
Ċ
TokenĊ
Feature activation+2.079
What
TokenWhat
Feature activation+2.124
.
Token.
Feature activation+0.968
Robert
Token Robert
Feature activation+0.748
Sp
Token Sp
Feature activation+0.000
iera
Tokeniera
Feature activation+0.060
of
Token of
Feature activation+1.520
the
Token the
Feature activation+1.886
Hospital
Token Hospital
Feature activation+0.303
for
Token for
Feature activation+0.837
Special
Token Special
Feature activation+0.113
Surgery
Token Surgery
Feature activation+0.000
in
Token in
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.519
revelations
Token revelations
Feature activation+0.000
that
Token that
Feature activation+0.218
he
Token he
Feature activation+1.827
had
Token had
Feature activation+1.143
hired
Token hired
Feature activation+0.000
a
Token a
Feature activation+0.000
ghost
Token ghost
Feature activation+0.000
writer
Tokenwriter
Feature activation+0.000
.
Token.
Feature activation+0.555
In
Token In
Feature activation+0.210
a
Token a
Feature activation+0.901
study
Token study
Feature activation+0.004
of
Token of
Feature activation+0.883
34
Token 34
Feature activation+1.810
patients
Token patients
Feature activation+1.673
published
Token published
Feature activation+0.000
in
Token in
Feature activation+1.244
The
Token The
Feature activation+1.897
American
Token American
Feature activation+1.912

Top DFA by src position
MAX = 3.221

S
Token S
Feature activation-0.003
Top resid features:
j
Tokenj
Feature activation-0.001
Top resid features:
ö
Tokenö
Feature activation+0.001
Top resid features:
gren
Tokengren
Feature activation+0.133
Top resid features:
's
Token's
Feature activation+0.080
Top resid features:
syndrome
Token syndrome
Feature activation+1.038
Top resid features:
,
Token,
Feature activation+0.035
Top resid features:
a
Token a
Feature activation+0.015
Top resid features:
little
Token little
Feature activation-0.002
Top resid features:
-
Token-
Feature activation-0.000
Top resid features:
known
Tokenknown
Feature activation+0.008
Top resid features:
unfortunate
Token unfortunate
Feature activation+0.042
Top resid features:
people
Token people
Feature activation+0.009
Top resid features:
who
Token who
Feature activation-0.000
Top resid features:
suffer
Token suffer
Feature activation-0.042
Top resid features:
from
Token from
Feature activation+0.002
Top resid features:
ALS
Token ALS
Feature activation+3.221
Top resid features:
,
Token,
Feature activation+0.033
Top resid features:
or
Token or
Feature activation-0.036
Top resid features:
as
Token as
Feature activation+0.001
Top resid features:
it
Token it
Feature activation+0.003
Top resid features:
âĢ
TokenâĢ
Feature activation-0.001
Top resid features:
unfortunate
Token unfortunate
Feature activation+0.008
Top resid features:
people
Token people
Feature activation+0.001
Top resid features:
who
Token who
Feature activation+0.000
Top resid features:
suffer
Token suffer
Feature activation-0.004
Top resid features:
from
Token from
Feature activation-0.001
Top resid features:
ALS
Token ALS
Feature activation+2.533
Top resid features:
,
Token,
Feature activation+0.053
Top resid features:
or
Token or
Feature activation-0.005
Top resid features:
as
Token as
Feature activation+0.011
Top resid features:
it
Token it
Feature activation+0.003
Top resid features:
âĢ
TokenâĢ
Feature activation-0.001
Top resid features:
S
Token S
Feature activation+0.004
Top resid features:
j
Tokenj
Feature activation+0.022
Top resid features:
ö
Tokenö
Feature activation-0.007
Top resid features:
gren
Tokengren
Feature activation+0.049
Top resid features:
's
Token's
Feature activation+0.071
Top resid features:
syndrome
Token syndrome
Feature activation+0.690
Top resid features:
,
Token,
Feature activation+0.020
Top resid features:
a
Token a
Feature activation+0.004
Top resid features:
little
Token little
Feature activation+0.001
Top resid features:
-
Token-
Feature activation-0.000
Top resid features:
known
Tokenknown
Feature activation+0.004
Top resid features:
S
Token S
Feature activation+0.001
Top resid features:
j
Tokenj
Feature activation-0.003
Top resid features:
ö
Tokenö
Feature activation+0.007
Top resid features:
gren
Tokengren
Feature activation+0.111
Top resid features:
's
Token's
Feature activation+0.375
Top resid features:
syndrome
Token syndrome
Feature activation+2.277
Top resid features:
,
Token,
Feature activation-0.034
Top resid features:
a
Token a
Feature activation-0.152
Top resid features:
little
Token little
Feature activation-0.050
Top resid features:
-
Token-
Feature activation-0.001
Top resid features:
known
Tokenknown
Feature activation-0.032
Top resid features:
S
Token S
Feature activation-0.087
Top resid features:
j
Tokenj
Feature activation+0.005
Top resid features:
ö
Tokenö
Feature activation-0.005
Top resid features:
gren
Tokengren
Feature activation+0.075
Top resid features:
's
Token's
Feature activation+0.087
Top resid features:
syndrome
Token syndrome
Feature activation+0.674
Top resid features:
,
Token,
Feature activation+0.012
Top resid features:
a
Token a
Feature activation+0.018
Top resid features:
little
Token little
Feature activation+0.000
Top resid features:
-
Token-
Feature activation-0.000
Top resid features:
known
Tokenknown
Feature activation+0.014
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.117
Top resid features:
epilepsy
Token epilepsy
Feature activation+2.464
Top resid features:
on
Token on
Feature activation+0.030
Top resid features:
this
Token this
Feature activation-0.003
Top resid features:
kind
Token kind
Feature activation-0.004
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
memory
Token memory
Feature activation+0.003
Top resid features:
because
Token because
Feature activation+0.000
Top resid features:
they
Token they
Feature activation+0.003
Top resid features:
might
Token might
Feature activation-0.001
Top resid features:
have
Token have
Feature activation-0.000
Top resid features:
Down
Token Down
Feature activation+0.055
Top resid features:
syndrome
Token syndrome
Feature activation+2.260
Top resid features:
."
Token."
Feature activation-0.019
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.016
Top resid features:
'
Token'
Feature activation-0.004
Top resid features:
What
TokenWhat
Feature activation-0.001
Top resid features:
their
Token their
Feature activation-0.003
Top resid features:
peers
Token peers
Feature activation+0.003
Top resid features:
struggling
Token struggling
Feature activation+0.036
Top resid features:
with
Token with
Feature activation+0.014
Top resid features:
mental
Token mental
Feature activation+0.266
Top resid features:
illnesses
Token illnesses
Feature activation+1.463
Top resid features:
,"
Token,"
Feature activation-0.002
Top resid features:
writes
Token writes
Feature activation-0.001
Top resid features:
Ble
Token Ble
Feature activation+0.000
Top resid features:
her
Tokenher
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
linked
Token linked
Feature activation+0.001
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
onset
Token onset
Feature activation+0.062
Top resid features:
of
Token of
Feature activation+0.013
Top resid features:
migraine
Token migraine
Feature activation+1.187
Top resid features:
.
Token.
Feature activation-0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Stud
TokenStud
Feature activation-0.001
Top resid features:
ying
Tokenying
Feature activation-0.005
Top resid features:
leading
Token leading
Feature activation-0.001
Top resid features:
cause
Token cause
Feature activation+0.045
Top resid features:
of
Token of
Feature activation-0.013
Top resid features:
chronic
Token chronic
Feature activation+0.371
Top resid features:
liver
Token liver
Feature activation+0.032
Top resid features:
disease
Token disease
Feature activation+1.320
Top resid features:
as
Token as
Feature activation-0.001
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
2017
Token 2017
Feature activation-0.026
Top resid features:
.[
Token.[
Feature activation+0.009
Top resid features:
11
Token11
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.157
Top resid features:
had
Token had
Feature activation-0.013
Top resid features:
a
Token a
Feature activation-0.000
Top resid features:
terrible
Token terrible
Feature activation+0.024
Top resid features:
disease
Token disease
Feature activation+2.227
Top resid features:
and
Token and
Feature activation+0.162
Top resid features:
you
Token you
Feature activation+0.003
Top resid features:
had
Token had
Feature activation+0.000
Top resid features:
to
Token to
Feature activation-0.000
Top resid features:
pick
Token pick
Feature activation+0.000
Top resid features:
intolerance
Token intolerance
Feature activation+0.438
Top resid features:
in
Token in
Feature activation-0.009
Top resid features:
patients
Token patients
Feature activation+0.390
Top resid features:
with
Token with
Feature activation+0.022
Top resid features:
I
Token I
Feature activation+0.012
Top resid features:
BS
TokenBS
Feature activation+0.811
Top resid features:
.
Token.
Feature activation+0.032
Top resid features:
In
Token In
Feature activation-0.004
Top resid features:
a
Token a
Feature activation-0.003
Top resid features:
study
Token study
Feature activation+0.004
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
linked
Token linked
Feature activation-0.001
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
onset
Token onset
Feature activation+0.081
Top resid features:
of
Token of
Feature activation+0.015
Top resid features:
migraine
Token migraine
Feature activation+2.410
Top resid features:
.
Token.
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Stud
TokenStud
Feature activation-0.001
Top resid features:
ying
Tokenying
Feature activation-0.004
Top resid features:
intolerance
Token intolerance
Feature activation+0.327
Top resid features:
in
Token in
Feature activation+0.001
Top resid features:
patients
Token patients
Feature activation+0.347
Top resid features:
with
Token with
Feature activation+0.023
Top resid features:
I
Token I
Feature activation+0.006
Top resid features:
BS
TokenBS
Feature activation+0.785
Top resid features:
.
Token.
Feature activation+0.036
Top resid features:
In
Token In
Feature activation-0.003
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
study
Token study
Feature activation+0.003
Top resid features:
of
Token of
Feature activation-0.001
Top resid features:
-
Token-
Feature activation-0.002
Top resid features:
ce
Tokence
Feature activation+0.003
Top resid features:
li
Tokenli
Feature activation+0.021
Top resid features:
ac
Tokenac
Feature activation+0.007
Top resid features:
gluten
Token gluten
Feature activation+0.449
Top resid features:
intolerance
Token intolerance
Feature activation+0.792
Top resid features:
in
Token in
Feature activation-0.017
Top resid features:
patients
Token patients
Feature activation+0.260
Top resid features:
with
Token with
Feature activation+0.071
Top resid features:
I
Token I
Feature activation+0.008
Top resid features:
BS
TokenBS
Feature activation+0.558
Top resid features:
S
Token S
Feature activation+0.007
Top resid features:
j
Tokenj
Feature activation-0.001
Top resid features:
ö
Tokenö
Feature activation-0.003
Top resid features:
gren
Tokengren
Feature activation+0.095
Top resid features:
's
Token's
Feature activation+0.232
Top resid features:
syndrome
Token syndrome
Feature activation+0.890
Top resid features:
,
Token,
Feature activation+0.060
Top resid features:
a
Token a
Feature activation+0.014
Top resid features:
little
Token little
Feature activation-0.059
Top resid features:
-
Token-
Feature activation-0.003
Top resid features:
known
Tokenknown
Feature activation+0.015
Top resid features:
's
Token's
Feature activation+0.074
Top resid features:
.
Token.
Feature activation+0.034
Top resid features:
"
Token "
Feature activation-0.000
Top resid features:
How
TokenHow
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
disease
Token disease
Feature activation+0.662
Top resid features:
will
Token will
Feature activation+0.012
Top resid features:
play
Token play
Feature activation+0.005
Top resid features:
out
Token out
Feature activation+0.011
Top resid features:
for
Token for
Feature activation+0.003
Top resid features:
Williams
Token Williams
Feature activation+0.001
Top resid features:
perhaps
Token perhaps
Feature activation-0.004
Top resid features:
even
Token even
Feature activation+0.006
Top resid features:
faked
Token faked
Feature activation-0.008
Top resid features:
his
Token his
Feature activation+0.004
Top resid features:
hearing
Token hearing
Feature activation+0.259
Top resid features:
disability
Token disability
Feature activation+2.226
Top resid features:
.
Token.
Feature activation+0.019
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.021
Top resid features:
Across
TokenAcross
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
intolerance
Token intolerance
Feature activation+0.241
Top resid features:
in
Token in
Feature activation+0.005
Top resid features:
patients
Token patients
Feature activation+0.618
Top resid features:
with
Token with
Feature activation+0.074
Top resid features:
I
Token I
Feature activation+0.022
Top resid features:
BS
TokenBS
Feature activation+1.174
Top resid features:
.
Token.
Feature activation+0.059
Top resid features:
In
Token In
Feature activation-0.001
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
study
Token study
Feature activation-0.009
Top resid features:
of
Token of
Feature activation+0.055
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.02

Head 2: 0.50

Head 3: 0.09

Head 4: 0.03

Head 5: 0.07

Head 6: 0.03

Head 7: 0.05

Head 8: 0.05

Head 9: 0.03

Head 10: 0.04

Head 11: 0.02

Positive logits

diagnosis6.55

diagnosed5.93

Diagn5.90

diagnoses5.81

diagnose5.60

Symptoms5.57

diagn5.57

disease5.48

symptoms5.38

neurolog5.34

illness5.27

diagn5.11

cure5.11

Disease5.02

Neurolog5.02

Cure4.95

epilepsy4.95

symptom4.94

Sym4.89

illnesses4.87

Negative logits

selage-2.85

Reviewer-2.84

Lago-2.60

ilitary-2.50

grab-2.42

asant-2.39

raltar-2.39

hops-2.38

ionage-2.36

abba-2.35

-+-+-2.35

Kislyak-2.32

ieval-2.31

material-2.31

Vote-2.29

ラン-2.27

EStream-2.27

ejected-2.26

exchanged-2.26

skirts-2.23

INTERVAL 2.324 - 2.583
CONTAINS 0.000%

Here
Token Here
Feature activation+0.030
,
Token,
Feature activation+0.737
a
Token a
Feature activation+0.763
brief
Token brief
Feature activation+0.494
guide
Token guide
Feature activation+0.000
to
Token to
Feature activation+2.583
this
Token this
Feature activation+1.889
condition
Token condition
Feature activation+0.000
:
Token:
Feature activation+1.295
Ċ
TokenĊ
Feature activation+0.225
Ċ
TokenĊ
Feature activation+2.079
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.331
more
Token more
Feature activation+0.482
commonly
Token commonly
Feature activation+0.231
known
Token known
Feature activation+0.098
,
Token,
Feature activation+2.390
Lou
Token Lou
Feature activation+0.901
Geh
Token Geh
Feature activation+0.000
rig
Tokenrig
Feature activation+2.328
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.278
commonly
Token commonly
Feature activation+0.231
known
Token known
Feature activation+0.098
,
Token,
Feature activation+2.390
Lou
Token Lou
Feature activation+0.901
Geh
Token Geh
Feature activation+0.000
rig
Tokenrig
Feature activation+2.328
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.278
s
Tokens
Feature activation+1.150
disease
Token disease
Feature activation+0.000
.
Token.
Feature activation+0.305

INTERVAL 2.066 - 2.324
CONTAINS 0.000%

syndrome
Token syndrome
Feature activation+0.000
,
Token,
Feature activation+0.450
a
Token a
Feature activation+1.692
little
Token little
Feature activation+0.000
-
Token-
Feature activation+0.000
known
Tokenknown
Feature activation+2.109
disease
Token disease
Feature activation+0.749
with
Token with
Feature activation+1.320
a
Token a
Feature activation+1.521
wide
Token wide
Feature activation+1.177
range
Token range
Feature activation+0.000
to
Token to
Feature activation+2.583
this
Token this
Feature activation+1.889
condition
Token condition
Feature activation+0.000
:
Token:
Feature activation+1.295
Ċ
TokenĊ
Feature activation+0.225
Ċ
TokenĊ
Feature activation+2.079
What
TokenWhat
Feature activation+2.124
is
Token is
Feature activation+1.114
S
Token S
Feature activation+0.000
j
Tokenj
Feature activation+0.000
ö
Tokenö
Feature activation+0.000
this
Token this
Feature activation+1.889
condition
Token condition
Feature activation+0.000
:
Token:
Feature activation+1.295
Ċ
TokenĊ
Feature activation+0.225
Ċ
TokenĊ
Feature activation+2.079
What
TokenWhat
Feature activation+2.124
is
Token is
Feature activation+1.114
S
Token S
Feature activation+0.000
j
Tokenj
Feature activation+0.000
ö
Tokenö
Feature activation+0.000
gren
Tokengren
Feature activation+0.000

INTERVAL 1.808 - 2.066
CONTAINS 0.001%

more
Token more
Feature activation+0.606
than
Token than
Feature activation+0.000
100
Token 100
Feature activation+0.753
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.788
people
Token people
Feature activation+1.970
,
Token,
Feature activation+0.026
including
Token including
Feature activation+0.792
those
Token those
Feature activation+1.758
affected
Token affected
Feature activation+0.241
and
Token and
Feature activation+0.871
12
Token 12
Feature activation+1.092
to
Token to
Feature activation+0.000
25
Token 25
Feature activation+1.746
%
Token%
Feature activation+1.195
of
Token of
Feature activation+1.261
people
Token people
Feature activation+1.932
in
Token in
Feature activation+0.597
the
Token the
Feature activation+0.633
United
Token United
Feature activation+0.000
States
Token States
Feature activation+1.728
have
Token have
Feature activation+0.987
,
Token,
Feature activation+0.737
a
Token a
Feature activation+0.763
brief
Token brief
Feature activation+0.494
guide
Token guide
Feature activation+0.000
to
Token to
Feature activation+2.583
this
Token this
Feature activation+1.889
condition
Token condition
Feature activation+0.000
:
Token:
Feature activation+1.295
Ċ
TokenĊ
Feature activation+0.225
Ċ
TokenĊ
Feature activation+2.079
What
TokenWhat
Feature activation+2.124
things
Token things
Feature activation+0.000
you
Token you
Feature activation+0.000
can
Token can
Feature activation+0.000
do
Token do
Feature activation+0.000
for
Token for
Feature activation+1.030
someone
Token someone
Feature activation+1.983
who
Token who
Feature activation+1.638
has
Token has
Feature activation+1.188
depression
Token depression
Feature activation+0.000
,
Token,
Feature activation+0.105
anxiety
Token anxiety
Feature activation+0.000
34
Token 34
Feature activation+1.810
patients
Token patients
Feature activation+1.673
published
Token published
Feature activation+0.000
in
Token in
Feature activation+1.244
The
Token The
Feature activation+1.897
American
Token American
Feature activation+1.912
Journal
Token Journal
Feature activation+1.027
of
Token of
Feature activation+1.907
Gast
Token Gast
Feature activation+1.254
ro
Tokenro
Feature activation+0.225
enter
Tokenenter
Feature activation+0.126

INTERVAL 1.550 - 1.808
CONTAINS 0.003%

âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.278
s
Tokens
Feature activation+1.150
disease
Token disease
Feature activation+0.000
.
Token.
Feature activation+0.305
His
Token His
Feature activation+1.576
story
Token story
Feature activation+0.000
of
Token of
Feature activation+0.773
Bitcoin
Token Bitcoin
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
of
Token of
Feature activation+1.261
people
Token people
Feature activation+1.932
in
Token in
Feature activation+0.597
the
Token the
Feature activation+0.633
United
Token United
Feature activation+0.000
States
Token States
Feature activation+1.728
have
Token have
Feature activation+0.987
NA
Token NA
Feature activation+0.000
FL
TokenFL
Feature activation+0.000
D
TokenD
Feature activation+0.000
,[
Token,[
Feature activation+0.000
his
Token his
Feature activation+0.000
condition
Token condition
Feature activation+0.000
.
Token.
Feature activation+0.035
Lou
Token Lou
Feature activation+0.000
Geh
Token Geh
Feature activation+0.000
rig
Tokenrig
Feature activation+1.564
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.275
s
Tokens
Feature activation+1.043
disease
Token disease
Feature activation+0.000
causes
Token causes
Feature activation+0.000
%
Token%
Feature activation+0.518
of
Token of
Feature activation+0.460
those
Token those
Feature activation+0.565
that
Token that
Feature activation+0.410
enroll
Token enroll
Feature activation+0.000
students
Token students
Feature activation+1.603
with
Token with
Feature activation+1.102
learning
Token learning
Feature activation+0.149
disabilities
Token disabilities
Feature activation+0.000
,
Token,
Feature activation+0.000
only
Token only
Feature activation+0.000
disease
Token disease
Feature activation+0.000
and
Token and
Feature activation+0.665
you
Token you
Feature activation+0.171
had
Token had
Feature activation+0.372
to
Token to
Feature activation+1.926
pick
Token pick
Feature activation+1.678
a
Token a
Feature activation+1.201
doctor
Token doctor
Feature activation+0.000
for
Token for
Feature activation+0.541
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.207

INTERVAL 1.291 - 1.550
CONTAINS 0.004%

,"
Token,"
Feature activation+0.561
says
Token says
Feature activation+0.888
Scott
Token Scott
Feature activation+0.000
Be
Token Be
Feature activation+0.000
a
Tokena
Feature activation+0.000
,
Token,
Feature activation+1.456
a
Token a
Feature activation+1.749
psychologist
Token psychologist
Feature activation+0.011
at
Token at
Feature activation+0.327
the
Token the
Feature activation+0.241
Cleveland
Token Cleveland
Feature activation+0.000
for
Token for
Feature activation+1.233
which
Token which
Feature activation+1.462
home
Token home
Feature activation+0.149
opathy
Tokenopathy
Feature activation+0.245
had
Token had
Feature activation+0.610
no
Token no
Feature activation+1.331
effect
Token effect
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
let
Token let
Feature activation+0.000
out
Token out
Feature activation+0.000
disability
Token disability
Feature activation+0.006
to
Token to
Feature activation+0.243
den
Token den
Feature activation+0.000
igrate
Tokenigrate
Feature activation+1.495
or
Token or
Feature activation+0.000
insult
Token insult
Feature activation+1.524
in
Token in
Feature activation+0.000
our
Token our
Feature activation+0.000
conversations
Token conversations
Feature activation+0.000
and
Token and
Feature activation+0.000
organizing
Token organizing
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.233
Now
TokenNow
Feature activation+1.105
My
Token My
Feature activation+0.057
kay
Tokenkay
Feature activation+0.000
la
Tokenla
Feature activation+0.780
is
Token is
Feature activation+1.320
in
Token in
Feature activation+1.218
the
Token the
Feature activation+0.814
middle
Token middle
Feature activation+1.125
of
Token of
Feature activation+1.394
another
Token another
Feature activation+1.264
G
TokenG
Feature activation+0.139
,
Token,
Feature activation+0.135
an
Token an
Feature activation+0.657
evidence
Token evidence
Feature activation+0.115
-
Token-
Feature activation+0.000
based
Tokenbased
Feature activation+1.454
and
Token and
Feature activation+0.465
effective
Token effective
Feature activation+1.218
program
Token program
Feature activation+0.190
for
Token for
Feature activation+0.969
those
Token those
Feature activation+1.709

INTERVAL 1.033 - 1.291
CONTAINS 0.010%

kay
Tokenkay
Feature activation+0.000
la
Tokenla
Feature activation+0.780
is
Token is
Feature activation+1.320
in
Token in
Feature activation+1.218
the
Token the
Feature activation+0.814
middle
Token middle
Feature activation+1.125
of
Token of
Feature activation+1.394
another
Token another
Feature activation+1.264
battle
Token battle
Feature activation+0.770
âĢĶ
TokenâĢĶ
Feature activation+0.555
and
Tokenand
Feature activation+0.409
His
TokenHis
Feature activation+0.375
parents
Token parents
Feature activation+0.511
were
Token were
Feature activation+0.335
told
Token told
Feature activation+0.682
he
Token he
Feature activation+0.633
would
Token would
Feature activation+1.041
either
Token either
Feature activation+0.916
be
Token be
Feature activation+0.384
still
Token still
Feature activation+0.000
born
Tokenborn
Feature activation+0.000
or
Token or
Feature activation+0.130
can
Token can
Feature activation+0.000
do
Token do
Feature activation+0.000
for
Token for
Feature activation+1.030
someone
Token someone
Feature activation+1.983
who
Token who
Feature activation+1.638
has
Token has
Feature activation+1.188
depression
Token depression
Feature activation+0.000
,
Token,
Feature activation+0.105
anxiety
Token anxiety
Feature activation+0.000
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
12
Token 12
Feature activation+0.657
%
Token%
Feature activation+0.295
.[
Token.[
Feature activation+0.000
12
Token12
Feature activation+0.000
]
Token]
Feature activation+1.246
The
Token The
Feature activation+1.273
annual
Token annual
Feature activation+1.260
economic
Token economic
Feature activation+0.017
burden
Token burden
Feature activation+0.000
was
Token was
Feature activation+0.000
estimated
Token estimated
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.265
Hal
TokenHal
Feature activation+0.529
didn
Token didn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.055
think
Token think
Feature activation+0.677
much
Token much
Feature activation+0.000
about
Token about
Feature activation+1.187
his
Token his
Feature activation+1.357
Bitcoin
Token Bitcoin
Feature activation+0.000

INTERVAL 0.775 - 1.033
CONTAINS 0.021%

and
Token and
Feature activation+0.000
neck
Token neck
Feature activation+0.000
cancers
Token cancers
Feature activation+0.000
,
Token,
Feature activation+0.238
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.782
two
Token two
Feature activation+1.014
new
Token new
Feature activation+1.417
studies
Token studies
Feature activation+0.000
published
Token published
Feature activation+0.000
in
Token in
Feature activation+0.868
lens
Token lens
Feature activation+0.000
through
Token through
Feature activation+0.000
which
Token which
Feature activation+0.000
we
Token we
Feature activation+0.000
see
Token see
Feature activation+0.444
women
Token women
Feature activation+0.934
in
Token in
Feature activation+0.338
music
Token music
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
author
Token author
Feature activation+0.000
and
Token and
Feature activation+1.027
post
Token post
Feature activation+0.000
doctoral
Tokendoctoral
Feature activation+0.000
researcher
Token researcher
Feature activation+0.000
at
Token at
Feature activation+0.894
the
Token the
Feature activation+0.951
S
Token S
Feature activation+0.254
alk
Tokenalk
Feature activation+0.000
Institute
Token Institute
Feature activation+0.000
.
Token.
Feature activation+0.000
the
Token the
Feature activation+0.084
prestigious
Token prestigious
Feature activation+0.356
British
Token British
Feature activation+0.246
medical
Token medical
Feature activation+0.000
journal
Token journal
Feature activation+1.288
The
Token The
Feature activation+1.031
Lancet
Token Lancet
Feature activation+0.029
(
Token (
Feature activation+0.000
in
Tokenin
Feature activation+0.000
its
Token its
Feature activation+0.000
edition
Token edition
Feature activation+0.000
My
Token My
Feature activation+0.057
kay
Tokenkay
Feature activation+0.000
la
Tokenla
Feature activation+0.780
is
Token is
Feature activation+1.320
in
Token in
Feature activation+1.218
the
Token the
Feature activation+0.814
middle
Token middle
Feature activation+1.125
of
Token of
Feature activation+1.394
another
Token another
Feature activation+1.264
battle
Token battle
Feature activation+0.770
âĢĶ
TokenâĢĶ
Feature activation+0.555

INTERVAL 0.517 - 0.775
CONTAINS 0.045%

those
Token those
Feature activation+1.709
with
Token with
Feature activation+0.520
PD
Token PD
Feature activation+0.056
seeking
Token seeking
Feature activation+0.420
to
Token to
Feature activation+0.327
improve
Token improve
Feature activation+0.517
movement
Token movement
Feature activation+0.356
,
Token,
Feature activation+0.000
balance
Token balance
Feature activation+0.000
,
Token,
Feature activation+0.000
walking
Token walking
Feature activation+0.000
.
Token.
Feature activation+0.261
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.047
METHOD
TokenMETHOD
Feature activation+0.000
:
Token:
Feature activation+0.363
Ċ
TokenĊ
Feature activation+0.706
Ċ
TokenĊ
Feature activation+0.570
Se
TokenSe
Feature activation+0.536
vent
Tokenvent
Feature activation+0.292
y
Tokeny
Feature activation+0.161
seven
Token seven
Feature activation+0.089
charity
Token charity
Feature activation+0.000
working
Token working
Feature activation+0.000
to
Token to
Feature activation+0.400
combat
Token combat
Feature activation+0.084
a
Token a
Feature activation+0.000
deadly
Token deadly
Feature activation+0.569
epidemic
Token epidemic
Feature activation+0.000
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
a
Token a
Feature activation+0.000
charity
Token charity
Feature activation+0.000
the
Token the
Feature activation+0.800
nonprofit
Token nonprofit
Feature activation+1.304
Global
Token Global
Feature activation+0.878
C
Token C
Feature activation+1.016
ures
Tokenures
Feature activation+0.071
to
Token to
Feature activation+0.677
promote
Token promote
Feature activation+0.000
research
Token research
Feature activation+0.000
of
Token of
Feature activation+0.000
cost
Token cost
Feature activation+0.000
-
Token-
Feature activation+0.000
need
Token need
Feature activation+0.000
anyone
Token anyone
Feature activation+0.000
to
Token to
Feature activation+0.090
tell
Token tell
Feature activation+0.000
him
Token him
Feature activation+0.000
his
Token his
Feature activation+0.732
prog
Token prog
Feature activation+0.607
nosis
Tokennosis
Feature activation+0.000
.
Token.
Feature activation+0.730
Although
Token Although
Feature activation+1.214
trained
Token trained
Feature activation+0.428

INTERVAL 0.258 - 0.517
CONTAINS 0.106%

tack
Token tack
Feature activation+0.000
and
Token and
Feature activation+0.000
diet
Token diet
Feature activation+0.000
to
Token to
Feature activation+0.212
save
Token save
Feature activation+0.000
their
Token their
Feature activation+0.461
health
Token health
Feature activation+0.000
?
Token?
Feature activation+0.000
Or
Token Or
Feature activation+0.000
will
Token will
Feature activation+0.000
they
Token they
Feature activation+0.169
information
Token information
Feature activation+0.000
to
Token to
Feature activation+0.134
more
Token more
Feature activation+0.236
quickly
Token quickly
Feature activation+0.346
identify
Token identify
Feature activation+0.011
and
Token and
Feature activation+0.374
solve
Token solve
Feature activation+0.000
this
Token this
Feature activation+0.000
problem
Token problem
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
hyper
Tokenhyper
Feature activation+0.000
kin
Tokenkin
Feature activation+0.000
etic
Tokenetic
Feature activation+0.076
children
Token children
Feature activation+0.701
,
Token,
Feature activation+0.467
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.977
or
Token or
Feature activation+0.821
children
Token children
Feature activation+0.750
diagnosed
Token diagnosed
Feature activation+0.102
platform
Token platform
Feature activation+0.000
world
Token world
Feature activation+0.000
between
Token between
Feature activation+0.246
the
Token the
Feature activation+0.317
whole
Token whole
Feature activation+0.311
ex
Token ex
Feature activation+0.275
ome
Tokenome
Feature activation+0.000
/
Token /
Feature activation+0.000
genome
Token genome
Feature activation+0.000
camp
Token camp
Feature activation+0.000
(
Token (
Feature activation+0.000
m
Token m
Feature activation+0.000
oping
Tokenoping
Feature activation+0.000
around
Token around
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.322
could
Token could
Feature activation+0.432
be
Token be
Feature activation+0.011
a
Token a
Feature activation+0.234
sign
Token sign
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000

INTERVAL 0.000 - 0.258
CONTAINS 99.809%

warm
Token warm
Feature activation+0.000
from
Token from
Feature activation+0.000
her
Token her
Feature activation+0.000
ass
Token ass
Feature activation+0.000
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
the
Token the
Feature activation+0.000
next
Token next
Feature activation+0.000
break
Token break
Feature activation+0.000
,
Token,
Feature activation+0.000
G
Token G
Feature activation+0.000
in
Token in
Feature activation+0.000
deck
Token deck
Feature activation+0.000
ed
Tokened
Feature activation+0.000
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
trucks
Token trucks
Feature activation+0.000
,
Token,
Feature activation+0.000
je
Token je
Feature activation+0.000
eps
Tokeneps
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
this
Token this
Feature activation+0.000
to
Token to
Feature activation+0.000
an
Token an
Feature activation+0.000
Eval
Token Eval
Feature activation+0.000
and
Token and
Feature activation+0.000
it
Token it
Feature activation+0.000
will
Token will
Feature activation+0.000
return
Token return
Feature activation+0.000
to
Token to
Feature activation+0.000
you
Token you
Feature activation+0.000
a
Token a
Feature activation+0.000
22
Token 22
Feature activation+0.000
centuries
Token centuries
Feature activation+0.000
,
Token,
Feature activation+0.000
with
Token with
Feature activation+0.000
its
Token its
Feature activation+0.000
quasi
Token quasi
Feature activation+0.000
âĪ¼
Token âĪ¼
Feature activation+0.000
100
Token100
Feature activation+0.000
,
Token,
Feature activation+0.000
50
Token 50
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
makes
Token makes
Feature activation+0.000
it
Token it
Feature activation+0.000
really
Token really
Feature activation+0.000
easy
Token easy
Feature activation+0.000
to
Token to
Feature activation+0.000
see
Token see
Feature activation+0.000
where
Token where
Feature activation+0.000
all
Token all
Feature activation+0.000
the
Token the
Feature activation+0.000
money
Token money
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 H9.2: (feature 773)

TOP ACTIVATIONS
MAX = 7.060

discrimination
Token discrimination
Feature activation+0.539
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107
based
Token based
Feature activation+0.901
on
Token on
Feature activation+7.060
sexual
Token sexual
Feature activation+4.236
orientation
Token orientation
Feature activation+0.000
and
Token and
Feature activation+4.703
gender
Token gender
Feature activation+0.231
identity
Token identity
Feature activation+0.392
from
Token from
Feature activation+0.000
your
Token your
Feature activation+0.000
job
Token job
Feature activation+0.000
just
Token just
Feature activation+1.080
for
Token for
Feature activation+6.711
being
Token being
Feature activation+7.002
gay
Token gay
Feature activation+0.000
in
Token in
Feature activation+1.417
29
Token 29
Feature activation+0.132
states
Token states
Feature activation+0.821
.
Token.
Feature activation+0.916
fired
Token fired
Feature activation+0.000
from
Token from
Feature activation+0.000
your
Token your
Feature activation+0.000
job
Token job
Feature activation+0.000
just
Token just
Feature activation+1.080
for
Token for
Feature activation+6.711
being
Token being
Feature activation+7.002
gay
Token gay
Feature activation+0.000
in
Token in
Feature activation+1.417
29
Token 29
Feature activation+0.132
states
Token states
Feature activation+0.821
script
Token script
Feature activation+0.075
on
Token on
Feature activation+4.529
their
Token their
Feature activation+0.827
desks
Token desks
Feature activation+1.471
contain
Token contain
Feature activation+4.916
a
Token a
Feature activation+6.626
slur
Token slur
Feature activation+0.000
,
Token,
Feature activation+0.512
reports
Token reports
Feature activation+0.000
Deadline
Token Deadline
Feature activation+0.000
.
Token.
Feature activation+0.000
ale
Tokenale
Feature activation+0.000
Est
Token Est
Feature activation+0.000
ates
Tokenates
Feature activation+0.332
)
Token)
Feature activation+0.000
to
Token to
Feature activation+6.116
prohibit
Token prohibit
Feature activation+6.504
discrimination
Token discrimination
Feature activation+0.539
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107
based
Token based
Feature activation+0.901
Christian
Token Christian
Feature activation+0.000
fans
Token fans
Feature activation+0.000
about
Token about
Feature activation+4.525
her
Token her
Feature activation+5.636
support
Token support
Feature activation+0.995
of
Token of
Feature activation+6.403
LGBTQ
Token LGBTQ
Feature activation+0.086
people
Token people
Feature activation+0.650
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.031
,
Token,
Feature activation+0.000
people
Token people
Feature activation+3.387
will
Token will
Feature activation+0.948
love
Token love
Feature activation+1.644
them
Token them
Feature activation+1.068
regardless
Token regardless
Feature activation+1.662
of
Token of
Feature activation+6.374
what
Token what
Feature activation+4.572
happens
Token happens
Feature activation+1.494
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
being
Token being
Feature activation+1.680
fired
Token fired
Feature activation+0.000
or
Token or
Feature activation+0.239
denied
Token denied
Feature activation+1.142
services
Token services
Feature activation+0.609
for
Token for
Feature activation+6.196
who
Token who
Feature activation+3.993
they
Token they
Feature activation+4.743
are
Token are
Feature activation+3.275
or
Token or
Feature activation+0.432
who
Token who
Feature activation+2.996
Zimbabwe
Token Zimbabwe
Feature activation+0.106
had
Token had
Feature activation+1.244
some
Token some
Feature activation+0.808
form
Token form
Feature activation+1.749
of
Token of
Feature activation+2.680
legal
Token legal
Feature activation+6.116
persecution
Token persecution
Feature activation+0.335
against
Token against
Feature activation+5.170
homosexuals
Token homosexuals
Feature activation+0.334
.
Token.
Feature activation+1.365
Reading
Token Reading
Feature activation+0.189
ond
Tokenond
Feature activation+0.000
ale
Tokenale
Feature activation+0.000
Est
Token Est
Feature activation+0.000
ates
Tokenates
Feature activation+0.332
)
Token)
Feature activation+0.000
to
Token to
Feature activation+6.116
prohibit
Token prohibit
Feature activation+6.504
discrimination
Token discrimination
Feature activation+0.539
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107
discrimination
Token discrimination
Feature activation+0.000
against
Token against
Feature activation+4.071
anyone
Token anyone
Feature activation+3.933
based
Token based
Feature activation+0.436
on
Token on
Feature activation+5.691
their
Token their
Feature activation+5.920
"
Token "
Feature activation+4.770
rel
Tokenrel
Feature activation+3.250
igion
Tokenigion
Feature activation+0.000
or
Token or
Feature activation+3.314
their
Token their
Feature activation+3.900
no
Token no
Feature activation+1.229
discrimination
Token discrimination
Feature activation+0.000
against
Token against
Feature activation+4.071
anyone
Token anyone
Feature activation+3.933
based
Token based
Feature activation+0.436
on
Token on
Feature activation+5.691
their
Token their
Feature activation+5.920
"
Token "
Feature activation+4.770
rel
Tokenrel
Feature activation+3.250
igion
Tokenigion
Feature activation+0.000
or
Token or
Feature activation+3.314
back
Tokenback
Feature activation+0.000
from
Token from
Feature activation+0.073
Christian
Token Christian
Feature activation+0.000
fans
Token fans
Feature activation+0.000
about
Token about
Feature activation+4.525
her
Token her
Feature activation+5.636
support
Token support
Feature activation+0.995
of
Token of
Feature activation+6.403
LGBTQ
Token LGBTQ
Feature activation+0.086
people
Token people
Feature activation+0.650
.
Token.
Feature activation+0.000
ates
Tokenates
Feature activation+0.332
)
Token)
Feature activation+0.000
to
Token to
Feature activation+6.116
prohibit
Token prohibit
Feature activation+6.504
discrimination
Token discrimination
Feature activation+0.539
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107
based
Token based
Feature activation+0.901
on
Token on
Feature activation+7.060
sexual
Token sexual
Feature activation+4.236
currently
Token currently
Feature activation+0.265
lobbying
Token lobbying
Feature activation+0.224
for
Token for
Feature activation+2.746
passage
Token passage
Feature activation+0.075
of
Token of
Feature activation+4.899
a
Token a
Feature activation+5.189
bill
Token bill
Feature activation+2.260
sponsored
Token sponsored
Feature activation+0.000
by
Token by
Feature activation+1.030
state
Token state
Feature activation+0.900
Rep
Token Rep
Feature activation+1.367
to
Token to
Feature activation+2.224
Hong
Token Hong
Feature activation+1.488
Kong
Token Kong
Feature activation+1.044
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.130
s
Tokens
Feature activation+5.143
immigration
Token immigration
Feature activation+0.666
policy
Token policy
Feature activation+0.000
,
Token,
Feature activation+0.000
besides
Token besides
Feature activation+0.276
being
Token being
Feature activation+0.061
media
Token media
Feature activation+0.370
has
Token has
Feature activation+0.000
on
Token on
Feature activation+3.458
the
Token the
Feature activation+4.131
way
Token way
Feature activation+2.691
people
Token people
Feature activation+4.936
learn
Token learn
Feature activation+1.071
to
Token to
Feature activation+0.910
perceive
Token perceive
Feature activation+3.831
the
Token the
Feature activation+3.786
world
Token world
Feature activation+0.749
being
Token being
Feature activation+3.625
a
Token a
Feature activation+3.793
gay
Token gay
Feature activation+0.000
Gordon
Token Gordon
Feature activation+0.394
he
Token he
Feature activation+2.668
was
Token was
Feature activation+4.933
a
Token a
Feature activation+3.150
passionate
Token passionate
Feature activation+0.753
woman
Token woman
Feature activation+0.000
iser
Tokeniser
Feature activation+0.000
,
Token,
Feature activation+0.184
a
Token a
Feature activation+4.821
script
Token script
Feature activation+0.075
on
Token on
Feature activation+4.529
their
Token their
Feature activation+0.827
desks
Token desks
Feature activation+1.471
contain
Token contain
Feature activation+4.916
a
Token a
Feature activation+6.626
slur
Token slur
Feature activation+0.000
,
Token,
Feature activation+0.512
reports
Token reports
Feature activation+0.000
Deadline
Token Deadline
Feature activation+0.000
is
Token is
Feature activation+0.479
currently
Token currently
Feature activation+0.265
lobbying
Token lobbying
Feature activation+0.224
for
Token for
Feature activation+2.746
passage
Token passage
Feature activation+0.075
of
Token of
Feature activation+4.899
a
Token a
Feature activation+5.189
bill
Token bill
Feature activation+2.260
sponsored
Token sponsored
Feature activation+0.000
by
Token by
Feature activation+1.030
state
Token state
Feature activation+0.900

Top DFA by src position
MAX = 7.864

state
Token state
Feature activation+0.002
Top resid features:
âĢ
TokenâĢ
Feature activation+0.016
Top resid features:
Ļ
TokenĻ
Feature activation+0.002
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
largest
Token largest
Feature activation+0.001
Top resid features:
LGBT
Token LGBT
Feature activation+7.864
Top resid features:
advocacy
Token advocacy
Feature activation-0.004
Top resid features:
group
Token group
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Georgia
Token Georgia
Feature activation-0.001
Top resid features:
Equality
Token Equality
Feature activation+0.001
Top resid features:
AS
TokenAS
Feature activation+0.002
Top resid features:
:
Token:
Feature activation-0.001
Top resid features:
With
Token With
Feature activation+0.000
Top resid features:
regards
Token regards
Feature activation-0.001
Top resid features:
to
Token to
Feature activation-0.000
Top resid features:
gay
Token gay
Feature activation+7.479
Top resid features:
rights
Token rights
Feature activation+0.051
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
it
Token it
Feature activation-0.001
Top resid features:
âĢ
TokenâĢ
Feature activation+0.006
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
AS
TokenAS
Feature activation+0.001
Top resid features:
:
Token:
Feature activation-0.001
Top resid features:
With
Token With
Feature activation-0.000
Top resid features:
regards
Token regards
Feature activation-0.001
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
gay
Token gay
Feature activation+6.965
Top resid features:
rights
Token rights
Feature activation+0.083
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
it
Token it
Feature activation-0.001
Top resid features:
âĢ
TokenâĢ
Feature activation+0.007
Top resid features:
Ļ
TokenĻ
Feature activation+0.001
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
s
Tokens
Feature activation+0.001
Top resid features:
stereotypical
Token stereotypical
Feature activation-0.020
Top resid features:
portrayal
Token portrayal
Feature activation-0.070
Top resid features:
of
Token of
Feature activation-0.007
Top resid features:
LGBT
Token LGBT
Feature activation+7.011
Top resid features:
people
Token people
Feature activation+0.198
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.011
Top resid features:
Sony
TokenSony
Feature activation+0.008
Top resid features:
state
Token state
Feature activation+0.021
Top resid features:
âĢ
TokenâĢ
Feature activation+0.045
Top resid features:
Ļ
TokenĻ
Feature activation+0.003
Top resid features:
s
Tokens
Feature activation+0.001
Top resid features:
largest
Token largest
Feature activation+0.003
Top resid features:
LGBT
Token LGBT
Feature activation+6.917
Top resid features:
advocacy
Token advocacy
Feature activation+0.005
Top resid features:
group
Token group
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.012
Top resid features:
Georgia
Token Georgia
Feature activation-0.016
Top resid features:
Equality
Token Equality
Feature activation+0.003
Top resid features:
reiterated
Token reiterated
Feature activation-0.000
Top resid features:
her
Token her
Feature activation+0.003
Top resid features:
love
Token love
Feature activation-0.002
Top resid features:
for
Token for
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
gay
Token gay
Feature activation+6.997
Top resid features:
community
Token community
Feature activation+0.176
Top resid features:
in
Token in
Feature activation+0.003
Top resid features:
an
Token an
Feature activation+0.000
Top resid features:
interview
Token interview
Feature activation-0.001
Top resid features:
on
Token on
Feature activation-0.000
Top resid features:
ly
Tokenly
Feature activation-0.000
Top resid features:
love
Token love
Feature activation-0.077
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
many
Token many
Feature activation-0.001
Top resid features:
LGBT
Token LGBT
Feature activation+4.158
Top resid features:
kids
Token kids
Feature activation+0.023
Top resid features:
need
Token need
Feature activation+0.003
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
understand
Token understand
Feature activation+0.009
Top resid features:
that
Token that
Feature activation+0.043
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.205
Top resid features:
that
Token that
Feature activation-0.006
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
while
Token while
Feature activation+0.001
Top resid features:
LGBT
Token LGBT
Feature activation+6.998
Top resid features:
Americans
Token Americans
Feature activation+0.007
Top resid features:
can
Token can
Feature activation-0.010
Top resid features:
legally
Token legally
Feature activation-0.019
Top resid features:
get
Token get
Feature activation-0.006
Top resid features:
married
Token married
Feature activation-0.008
Top resid features:
extremes
Token extremes
Feature activation+0.004
Top resid features:
in
Token in
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
protection
Token protection
Feature activation-0.017
Top resid features:
of
Token of
Feature activation+0.010
Top resid features:
gay
Token gay
Feature activation+4.773
Top resid features:
rights
Token rights
Feature activation+0.032
Top resid features:
.
Token.
Feature activation+0.009
Top resid features:
In
Token In
Feature activation-0.000
Top resid features:
ten
Token ten
Feature activation-0.001
Top resid features:
countries
Token countries
Feature activation+0.002
Top resid features:
state
Token state
Feature activation+0.018
Top resid features:
âĢ
TokenâĢ
Feature activation+0.025
Top resid features:
Ļ
TokenĻ
Feature activation+0.003
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
largest
Token largest
Feature activation-0.001
Top resid features:
LGBT
Token LGBT
Feature activation+6.597
Top resid features:
advocacy
Token advocacy
Feature activation+0.001
Top resid features:
group
Token group
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.019
Top resid features:
Georgia
Token Georgia
Feature activation-0.036
Top resid features:
Equality
Token Equality
Feature activation+0.004
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
people
Token people
Feature activation-0.001
Top resid features:
with
Token with
Feature activation-0.008
Top resid features:
different
Token different
Feature activation-0.008
Top resid features:
sexual
Token sexual
Feature activation+0.009
Top resid features:
orientation
Token orientation
Feature activation+6.522
Top resid features:
are
Token are
Feature activation-0.001
Top resid features:
welcome
Token welcome
Feature activation+0.002
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
compete
Token compete
Feature activation+0.001
Top resid features:
and
Token and
Feature activation-0.000
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
people
Token people
Feature activation-0.003
Top resid features:
with
Token with
Feature activation-0.007
Top resid features:
different
Token different
Feature activation+0.008
Top resid features:
sexual
Token sexual
Feature activation+0.071
Top resid features:
orientation
Token orientation
Feature activation+6.262
Top resid features:
are
Token are
Feature activation-0.004
Top resid features:
welcome
Token welcome
Feature activation+0.002
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
compete
Token compete
Feature activation-0.000
Top resid features:
and
Token and
Feature activation-0.000
Top resid features:
reiterated
Token reiterated
Feature activation+0.001
Top resid features:
her
Token her
Feature activation+0.000
Top resid features:
love
Token love
Feature activation+0.006
Top resid features:
for
Token for
Feature activation+0.002
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
gay
Token gay
Feature activation+6.055
Top resid features:
community
Token community
Feature activation+0.139
Top resid features:
in
Token in
Feature activation+0.005
Top resid features:
an
Token an
Feature activation+0.000
Top resid features:
interview
Token interview
Feature activation-0.003
Top resid features:
on
Token on
Feature activation-0.001
Top resid features:
state
Token state
Feature activation+0.007
Top resid features:
âĢ
TokenâĢ
Feature activation+0.024
Top resid features:
Ļ
TokenĻ
Feature activation+0.002
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
largest
Token largest
Feature activation+0.002
Top resid features:
LGBT
Token LGBT
Feature activation+5.631
Top resid features:
advocacy
Token advocacy
Feature activation-0.005
Top resid features:
group
Token group
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.008
Top resid features:
Georgia
Token Georgia
Feature activation-0.006
Top resid features:
Equality
Token Equality
Feature activation+0.004
Top resid features:
state
Token state
Feature activation+0.035
Top resid features:
âĢ
TokenâĢ
Feature activation+0.035
Top resid features:
Ļ
TokenĻ
Feature activation+0.003
Top resid features:
s
Tokens
Feature activation+0.002
Top resid features:
largest
Token largest
Feature activation+0.002
Top resid features:
LGBT
Token LGBT
Feature activation+5.409
Top resid features:
advocacy
Token advocacy
Feature activation+0.006
Top resid features:
group
Token group
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.013
Top resid features:
Georgia
Token Georgia
Feature activation-0.050
Top resid features:
Equality
Token Equality
Feature activation+0.011
Top resid features:
change
Token change
Feature activation-0.002
Top resid features:
its
Token its
Feature activation-0.003
Top resid features:
attitudes
Token attitudes
Feature activation+0.001
Top resid features:
towards
Token towards
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
LGBT
Token LGBT
Feature activation+4.226
Top resid features:
community
Token community
Feature activation+0.034
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
âĢ
TokenâĢ
Feature activation+0.003
Top resid features:
Ļ
TokenĻ
Feature activation-0.000
Top resid features:
s
Tokens
Feature activation-0.002
Top resid features:
stereotypical
Token stereotypical
Feature activation-0.002
Top resid features:
portrayal
Token portrayal
Feature activation+0.008
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
LGBT
Token LGBT
Feature activation+5.229
Top resid features:
people
Token people
Feature activation+0.120
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Sony
TokenSony
Feature activation+0.003
Top resid features:
that
Token that
Feature activation+0.019
Top resid features:
far
Token far
Feature activation+0.010
Top resid features:
from
Token from
Feature activation+0.019
Top resid features:
being
Token being
Feature activation+0.022
Top resid features:
a
Token a
Feature activation+0.037
Top resid features:
gay
Token gay
Feature activation+3.713
Top resid features:
Gordon
Token Gordon
Feature activation+0.008
Top resid features:
he
Token he
Feature activation+0.132
Top resid features:
was
Token was
Feature activation+0.047
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
passionate
Token passionate
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation-0.001
Top resid features:
s
Tokens
Feature activation+0.001
Top resid features:
stereotypical
Token stereotypical
Feature activation-0.026
Top resid features:
portrayal
Token portrayal
Feature activation-0.017
Top resid features:
of
Token of
Feature activation+0.011
Top resid features:
LGBT
Token LGBT
Feature activation+5.300
Top resid features:
people
Token people
Feature activation+0.261
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
Sony
TokenSony
Feature activation+0.011
Top resid features:
state
Token state
Feature activation+0.026
Top resid features:
âĢ
TokenâĢ
Feature activation+0.057
Top resid features:
Ļ
TokenĻ
Feature activation+0.005
Top resid features:
s
Tokens
Feature activation-0.001
Top resid features:
largest
Token largest
Feature activation+0.001
Top resid features:
LGBT
Token LGBT
Feature activation+5.304
Top resid features:
advocacy
Token advocacy
Feature activation+0.006
Top resid features:
group
Token group
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.022
Top resid features:
Georgia
Token Georgia
Feature activation-0.070
Top resid features:
Equality
Token Equality
Feature activation-0.001
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.01

Head 2: 0.49

Head 3: 0.07

Head 4: 0.02

Head 5: 0.09

Head 6: 0.04

Head 7: 0.04

Head 8: 0.08

Head 9: 0.03

Head 10: 0.02

Head 11: 0.03

Positive logits

LGBT7.02

homophobia6.99

gay6.97

gays6.75

homosexuality6.61

LGBTQ6.60

homophobic6.60

Gay6.60

LGBT6.59

homosexuals6.57

homosexual6.54

Lesbian6.41

lesbian6.27

lesbians6.15

bisexual6.11

gay6.10

heterosexual6.01

Gay5.94

Transgender5.90

queer5.85

Negative logits

resp-3.00

ebted-2.79

Beg-2.78

grain-2.77

weap-2.68

retake-2.61

thora-2.59

Logged-2.49

Export-2.49

scaven-2.47

guiActiveUn-2.44

conom-2.42

assic-2.40

efficient-2.39

export-2.39

Costs-2.35

-2.33

akings-2.29

efficiency-2.29

budget-2.26

INTERVAL 6.354 - 7.060
CONTAINS 0.001%

from
Token from
Feature activation+0.000
your
Token your
Feature activation+0.000
job
Token job
Feature activation+0.000
just
Token just
Feature activation+1.080
for
Token for
Feature activation+6.711
being
Token being
Feature activation+7.002
gay
Token gay
Feature activation+0.000
in
Token in
Feature activation+1.417
29
Token 29
Feature activation+0.132
states
Token states
Feature activation+0.821
.
Token.
Feature activation+0.916
discrimination
Token discrimination
Feature activation+0.539
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107
based
Token based
Feature activation+0.901
on
Token on
Feature activation+7.060
sexual
Token sexual
Feature activation+4.236
orientation
Token orientation
Feature activation+0.000
and
Token and
Feature activation+4.703
gender
Token gender
Feature activation+0.231
identity
Token identity
Feature activation+0.392
Christian
Token Christian
Feature activation+0.000
fans
Token fans
Feature activation+0.000
about
Token about
Feature activation+4.525
her
Token her
Feature activation+5.636
support
Token support
Feature activation+0.995
of
Token of
Feature activation+6.403
LGBTQ
Token LGBTQ
Feature activation+0.086
people
Token people
Feature activation+0.650
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.031
,
Token,
Feature activation+0.000
fired
Token fired
Feature activation+0.000
from
Token from
Feature activation+0.000
your
Token your
Feature activation+0.000
job
Token job
Feature activation+0.000
just
Token just
Feature activation+1.080
for
Token for
Feature activation+6.711
being
Token being
Feature activation+7.002
gay
Token gay
Feature activation+0.000
in
Token in
Feature activation+1.417
29
Token 29
Feature activation+0.132
states
Token states
Feature activation+0.821
people
Token people
Feature activation+3.387
will
Token will
Feature activation+0.948
love
Token love
Feature activation+1.644
them
Token them
Feature activation+1.068
regardless
Token regardless
Feature activation+1.662
of
Token of
Feature activation+6.374
what
Token what
Feature activation+4.572
happens
Token happens
Feature activation+1.494
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 5.648 - 6.354
CONTAINS 0.001%

being
Token being
Feature activation+1.680
fired
Token fired
Feature activation+0.000
or
Token or
Feature activation+0.239
denied
Token denied
Feature activation+1.142
services
Token services
Feature activation+0.609
for
Token for
Feature activation+6.196
who
Token who
Feature activation+3.993
they
Token they
Feature activation+4.743
are
Token are
Feature activation+3.275
or
Token or
Feature activation+0.432
who
Token who
Feature activation+2.996
no
Token no
Feature activation+1.229
discrimination
Token discrimination
Feature activation+0.000
against
Token against
Feature activation+4.071
anyone
Token anyone
Feature activation+3.933
based
Token based
Feature activation+0.436
on
Token on
Feature activation+5.691
their
Token their
Feature activation+5.920
"
Token "
Feature activation+4.770
rel
Tokenrel
Feature activation+3.250
igion
Tokenigion
Feature activation+0.000
or
Token or
Feature activation+3.314
discrimination
Token discrimination
Feature activation+0.000
against
Token against
Feature activation+4.071
anyone
Token anyone
Feature activation+3.933
based
Token based
Feature activation+0.436
on
Token on
Feature activation+5.691
their
Token their
Feature activation+5.920
"
Token "
Feature activation+4.770
rel
Tokenrel
Feature activation+3.250
igion
Tokenigion
Feature activation+0.000
or
Token or
Feature activation+3.314
their
Token their
Feature activation+3.900
Zimbabwe
Token Zimbabwe
Feature activation+0.106
had
Token had
Feature activation+1.244
some
Token some
Feature activation+0.808
form
Token form
Feature activation+1.749
of
Token of
Feature activation+2.680
legal
Token legal
Feature activation+6.116
persecution
Token persecution
Feature activation+0.335
against
Token against
Feature activation+5.170
homosexuals
Token homosexuals
Feature activation+0.334
.
Token.
Feature activation+1.365
Reading
Token Reading
Feature activation+0.189
ond
Tokenond
Feature activation+0.000
ale
Tokenale
Feature activation+0.000
Est
Token Est
Feature activation+0.000
ates
Tokenates
Feature activation+0.332
)
Token)
Feature activation+0.000
to
Token to
Feature activation+6.116
prohibit
Token prohibit
Feature activation+6.504
discrimination
Token discrimination
Feature activation+0.539
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107

INTERVAL 4.942 - 5.648
CONTAINS 0.001%

to
Token to
Feature activation+2.224
Hong
Token Hong
Feature activation+1.488
Kong
Token Kong
Feature activation+1.044
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.130
s
Tokens
Feature activation+5.143
immigration
Token immigration
Feature activation+0.666
policy
Token policy
Feature activation+0.000
,
Token,
Feature activation+0.000
besides
Token besides
Feature activation+0.276
being
Token being
Feature activation+0.061
back
Tokenback
Feature activation+0.000
from
Token from
Feature activation+0.073
Christian
Token Christian
Feature activation+0.000
fans
Token fans
Feature activation+0.000
about
Token about
Feature activation+4.525
her
Token her
Feature activation+5.636
support
Token support
Feature activation+0.995
of
Token of
Feature activation+6.403
LGBTQ
Token LGBTQ
Feature activation+0.086
people
Token people
Feature activation+0.650
.
Token.
Feature activation+0.000
currently
Token currently
Feature activation+0.265
lobbying
Token lobbying
Feature activation+0.224
for
Token for
Feature activation+2.746
passage
Token passage
Feature activation+0.075
of
Token of
Feature activation+4.899
a
Token a
Feature activation+5.189
bill
Token bill
Feature activation+2.260
sponsored
Token sponsored
Feature activation+0.000
by
Token by
Feature activation+1.030
state
Token state
Feature activation+0.900
Rep
Token Rep
Feature activation+1.367
ates
Tokenates
Feature activation+0.332
)
Token)
Feature activation+0.000
to
Token to
Feature activation+6.116
prohibit
Token prohibit
Feature activation+6.504
discrimination
Token discrimination
Feature activation+0.539
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107
based
Token based
Feature activation+0.901
on
Token on
Feature activation+7.060
sexual
Token sexual
Feature activation+4.236

INTERVAL 4.236 - 4.942
CONTAINS 0.003%

will
Token will
Feature activation+0.948
love
Token love
Feature activation+1.644
them
Token them
Feature activation+1.068
regardless
Token regardless
Feature activation+1.662
of
Token of
Feature activation+6.374
what
Token what
Feature activation+4.572
happens
Token happens
Feature activation+1.494
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.136
a
Token a
Feature activation+4.821
script
Token script
Feature activation+0.075
on
Token on
Feature activation+4.529
their
Token their
Feature activation+0.827
desks
Token desks
Feature activation+1.471
contain
Token contain
Feature activation+4.916
a
Token a
Feature activation+6.626
slur
Token slur
Feature activation+0.000
,
Token,
Feature activation+0.512
reports
Token reports
Feature activation+0.000
Deadline
Token Deadline
Feature activation+0.000
to
Token to
Feature activation+3.689
simply
Token simply
Feature activation+2.548
'
Token '
Feature activation+1.197
cross
Tokencross
Feature activation+4.139
out
Token out
Feature activation+4.276
'
Token'
Feature activation+4.568
slurs
Token slurs
Feature activation+0.410
and
Token and
Feature activation+1.616
negative
Token negative
Feature activation+0.766
portray
Token portray
Feature activation+0.000
als
Tokenals
Feature activation+0.219
to
Token to
Feature activation+1.091
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
sim
Tokensim
Feature activation+0.000
ply
Tokenply
Feature activation+3.294
cross
Token cross
Feature activation+4.635
it
Token it
Feature activation+0.000
out
Token out
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.148
when
Token when
Feature activation+2.098
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107
based
Token based
Feature activation+0.901
on
Token on
Feature activation+7.060
sexual
Token sexual
Feature activation+4.236
orientation
Token orientation
Feature activation+0.000
and
Token and
Feature activation+4.703
gender
Token gender
Feature activation+0.231
identity
Token identity
Feature activation+0.392
.
Token.
Feature activation+0.275

INTERVAL 3.530 - 4.236
CONTAINS 0.004%

while
Token while
Feature activation+0.000
LGBT
Token LGBT
Feature activation+0.000
Americans
Token Americans
Feature activation+0.120
can
Token can
Feature activation+0.689
legally
Token legally
Feature activation+4.452
get
Token get
Feature activation+3.870
married
Token married
Feature activation+0.921
,
Token,
Feature activation+1.442
millions
Token millions
Feature activation+0.000
remain
Token remain
Feature activation+1.065
at
Token at
Feature activation+0.628
data
Token data
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.721
Though
TokenThough
Feature activation+2.452
debates
Token debates
Feature activation+0.000
around
Token around
Feature activation+3.793
gay
Token gay
Feature activation+0.439
marriage
Token marriage
Feature activation+0.000
in
Token in
Feature activation+0.566
Europe
Token Europe
Feature activation+0.000
and
Token and
Feature activation+0.655
coverage
Token coverage
Feature activation+1.498
for
Token for
Feature activation+0.893
certain
Token certain
Feature activation+1.463
forms
Token forms
Feature activation+0.087
of
Token of
Feature activation+2.206
birth
Token birth
Feature activation+3.568
control
Token control
Feature activation+0.000
.
Token.
Feature activation+0.255
The
Token The
Feature activation+0.000
link
Token link
Feature activation+0.000
between
Token between
Feature activation+0.000
to
Token to
Feature activation+0.550
show
Token show
Feature activation+0.944
that
Token that
Feature activation+0.704
far
Token far
Feature activation+1.307
from
Token from
Feature activation+3.478
being
Token being
Feature activation+3.625
a
Token a
Feature activation+3.793
gay
Token gay
Feature activation+0.000
Gordon
Token Gordon
Feature activation+0.394
he
Token he
Feature activation+2.668
was
Token was
Feature activation+4.933
fired
Token fired
Feature activation+0.000
or
Token or
Feature activation+0.239
denied
Token denied
Feature activation+1.142
services
Token services
Feature activation+0.609
for
Token for
Feature activation+6.196
who
Token who
Feature activation+3.993
they
Token they
Feature activation+4.743
are
Token are
Feature activation+3.275
or
Token or
Feature activation+0.432
who
Token who
Feature activation+2.996
they
Token they
Feature activation+3.903

INTERVAL 2.824 - 3.530
CONTAINS 0.008%

before
Token before
Feature activation+0.511
a
Token a
Feature activation+0.677
person
Token person
Feature activation+1.588
settles
Token settles
Feature activation+1.096
on
Token on
Feature activation+2.576
their
Token their
Feature activation+3.053
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+1.216
true
Tokentrue
Feature activation+2.581
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+2.706
on
Token on
Feature activation+0.000
gay
Token gay
Feature activation+0.039
sex
Token sex
Feature activation+0.337
and
Token and
Feature activation+3.177
Australia
Token Australia
Feature activation+0.242
's
Token's
Feature activation+3.115
high
Token high
Feature activation+0.348
court
Token court
Feature activation+1.164
has
Token has
Feature activation+0.078
struck
Token struck
Feature activation+0.000
down
Token down
Feature activation+1.817
countries
Token countries
Feature activation+0.000
whether
Token whether
Feature activation+0.815
you
Token you
Feature activation+0.943
are
Token are
Feature activation+1.028
gay
Token gay
Feature activation+0.000
or
Token or
Feature activation+3.268
straight
Token straight
Feature activation+0.043
."
Token."
Feature activation+0.118
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.050
K
TokenK
Feature activation+0.000
alone
Token alone
Feature activation+0.000
.
Token.
Feature activation+0.106
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.014
Same
TokenSame
Feature activation+1.456
-
Token-
Feature activation+3.012
sex
Tokensex
Feature activation+1.925
marriage
Token marriage
Feature activation+0.000
still
Token still
Feature activation+0.000
isn
Token isn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
referring
Token referring
Feature activation+0.405
to
Token to
Feature activation+4.761
a
Token a
Feature activation+3.609
Pride
Token Pride
Feature activation+0.570
event
Token event
Feature activation+0.897
as
Token as
Feature activation+3.106
an
Token an
Feature activation+2.875
âĢ
Token âĢ
Feature activation+0.613
ľ
Tokenľ
Feature activation+1.110
ab
Tokenab
Feature activation+2.827
omination
Tokenomination
Feature activation+0.014

INTERVAL 2.118 - 2.824
CONTAINS 0.017%

use
Token use
Feature activation+2.542
the
Token the
Feature activation+1.988
bathroom
Token bathroom
Feature activation+0.354
and
Token and
Feature activation+0.000
changing
Token changing
Feature activation+0.142
facilities
Token facilities
Feature activation+2.614
assigned
Token assigned
Feature activation+1.885
to
Token to
Feature activation+1.306
persons
Token persons
Feature activation+2.766
of
Token of
Feature activation+1.667
their
Token their
Feature activation+2.208
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.879
World
TokenWorld
Feature activation+0.888
leaders
Token leaders
Feature activation+0.892
condemned
Token condemned
Feature activation+4.477
the
Token the
Feature activation+2.598
proposal
Token proposal
Feature activation+0.000
,
Token,
Feature activation+0.272
prompting
Token prompting
Feature activation+2.331
some
Token some
Feature activation+1.171
nations
Token nations
Feature activation+0.395
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.415
new
Token new
Feature activation+0.000
organization
Token organization
Feature activation+0.165
,
Token,
Feature activation+1.871
The
Token The
Feature activation+2.390
Family
Token Family
Feature activation+0.939
Leader
Token Leader
Feature activation+0.000
,
Token,
Feature activation+0.000
recently
Token recently
Feature activation+0.000
blasted
Token blasted
Feature activation+1.123
underscore
Token underscore
Feature activation+0.000
the
Token the
Feature activation+0.000
persecution
Token persecution
Feature activation+0.000
of
Token of
Feature activation+0.599
gays
Token gays
Feature activation+0.727
in
Token in
Feature activation+2.228
our
Token our
Feature activation+1.113
country
Token country
Feature activation+0.000
.
Token.
Feature activation+0.003
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.132
)
Token)
Feature activation+0.000
to
Token to
Feature activation+6.116
prohibit
Token prohibit
Feature activation+6.504
discrimination
Token discrimination
Feature activation+0.539
against
Token against
Feature activation+5.379
state
Token state
Feature activation+2.336
employees
Token employees
Feature activation+4.107
based
Token based
Feature activation+0.901
on
Token on
Feature activation+7.060
sexual
Token sexual
Feature activation+4.236
orientation
Token orientation
Feature activation+0.000

INTERVAL 1.412 - 2.118
CONTAINS 0.036%

rating
Tokenrating
Feature activation+0.237
ly
Tokenly
Feature activation+0.000
,
Token,
Feature activation+1.207
even
Token even
Feature activation+0.526
within
Token within
Feature activation+1.298
the
Token the
Feature activation+1.417
most
Token most
Feature activation+1.471
LGBT
Token LGBT
Feature activation+0.000
-
Token-
Feature activation+1.202
friendly
Tokenfriendly
Feature activation+1.869
circles
Token circles
Feature activation+0.000
gender
Token gender
Feature activation+0.499
as
Token as
Feature activation+0.117
a
Token a
Feature activation+0.209
basis
Token basis
Feature activation+0.000
for
Token for
Feature activation+0.148
nond
Token nond
Feature activation+1.993
isc
Tokenisc
Feature activation+0.314
rimination
Tokenrimination
Feature activation+0.000
protections
Token protections
Feature activation+0.000
,
Token,
Feature activation+0.087
while
Token while
Feature activation+0.101
53
Token 53
Feature activation+0.000
people
Token people
Feature activation+0.000
at
Token at
Feature activation+0.000
Pulse
Token Pulse
Feature activation+0.000
,
Token,
Feature activation+0.691
a
Token a
Feature activation+1.617
gay
Token gay
Feature activation+0.518
night
Token night
Feature activation+0.227
club
Token club
Feature activation+0.824
,
Token,
Feature activation+0.287
on
Token on
Feature activation+1.240
when
Token when
Feature activation+2.098
a
Token a
Feature activation+4.821
script
Token script
Feature activation+0.075
on
Token on
Feature activation+4.529
their
Token their
Feature activation+0.827
desks
Token desks
Feature activation+1.471
contain
Token contain
Feature activation+4.916
a
Token a
Feature activation+6.626
slur
Token slur
Feature activation+0.000
,
Token,
Feature activation+0.512
reports
Token reports
Feature activation+0.000
within
Token within
Feature activation+1.298
the
Token the
Feature activation+1.417
most
Token most
Feature activation+1.471
LGBT
Token LGBT
Feature activation+0.000
-
Token-
Feature activation+1.202
friendly
Tokenfriendly
Feature activation+1.869
circles
Token circles
Feature activation+0.000
,
Token,
Feature activation+1.273
you
Token you
Feature activation+0.000
encounter
Token encounter
Feature activation+0.811
the
Token the
Feature activation+0.095

INTERVAL 0.706 - 1.412
CONTAINS 0.084%

month
Token month
Feature activation+0.846
.
Token.
Feature activation+0.008
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.808
"
Token"
Feature activation+2.363
You
TokenYou
Feature activation+0.866
couldn
Token couldn
Feature activation+0.000
't
Token't
Feature activation+0.271
hear
Token hear
Feature activation+1.684
yourself
Token yourself
Feature activation+1.576
think
Token think
Feature activation+1.067
Disney
TokenDisney
Feature activation+0.000
's
Token's
Feature activation+1.578
Frozen
Token Frozen
Feature activation+0.000
may
Token may
Feature activation+0.210
not
Token not
Feature activation+0.465
have
Token have
Feature activation+1.091
intended
Token intended
Feature activation+1.191
for
Token for
Feature activation+2.759
one
Token one
Feature activation+0.644
of
Token of
Feature activation+0.224
the
Token the
Feature activation+1.686
risk
Token risk
Feature activation+0.000
of
Token of
Feature activation+1.367
being
Token being
Feature activation+1.680
fired
Token fired
Feature activation+0.000
or
Token or
Feature activation+0.239
denied
Token denied
Feature activation+1.142
services
Token services
Feature activation+0.609
for
Token for
Feature activation+6.196
who
Token who
Feature activation+3.993
they
Token they
Feature activation+4.743
are
Token are
Feature activation+3.275
Ŀ
TokenĿ
Feature activation+0.000
Vietnam
Token Vietnam
Feature activation+0.886
âĢĶ
Token âĢĶ
Feature activation+0.000
unlike
Token unlike
Feature activation+0.787
much
Token much
Feature activation+0.000
of
Token of
Feature activation+0.753
the
Token the
Feature activation+0.942
United
Token United
Feature activation+0.000
States
Token States
Feature activation+0.000
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.000
The
TokenThe
Feature activation+1.081
proposed
Token proposed
Feature activation+1.359
change
Token change
Feature activation+0.029
to
Token to
Feature activation+2.224
Hong
Token Hong
Feature activation+1.488
Kong
Token Kong
Feature activation+1.044
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.130
s
Tokens
Feature activation+5.143
immigration
Token immigration
Feature activation+0.666
policy
Token policy
Feature activation+0.000

INTERVAL 0.000 - 0.706
CONTAINS 99.846%

L
Token L
Feature activation+0.000
unk
Tokenunk
Feature activation+0.000
Yet
Token Yet
Feature activation+0.000
is
Tokenis
Feature activation+0.000
Are
Token Are
Feature activation+0.000
Too
Token Too
Feature activation+0.000
Chill
Token Chill
Feature activation+0.000
Platform
Token Platform
Feature activation+0.000
ing
Tokening
Feature activation+0.000
at
Token at
Feature activation+0.000
Its
Token Its
Feature activation+0.000
await
Token await
Feature activation+0.000
official
Token official
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
Metal
TokenMetal
Feature activation+0.000
Gear
Token Gear
Feature activation+0.000
Solid
Token Solid
Feature activation+0.000
V
Token V
Feature activation+0.000
:
Token:
Feature activation+0.000
include
Token include
Feature activation+0.000
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
number
Token number
Feature activation+0.000
6
Token 6
Feature activation+0.000
28
Token28
Feature activation+0.000
315
Token315
Feature activation+0.000
.
Token.
Feature activation+0.000
P
Token P
Feature activation+0.000
asting
Tokenasting
Feature activation+0.000
the
Token the
Feature activation+0.000
energy
Token energy
Feature activation+0.000
solution
Token solution
Feature activation+0.000
since
Token since
Feature activation+0.000
Jimmy
Token Jimmy
Feature activation+0.000
Carter
Token Carter
Feature activation+0.000
first
Token first
Feature activation+0.000
installed
Token installed
Feature activation+0.000
phot
Token phot
Feature activation+0.000
ov
Tokenov
Feature activation+0.000
olt
Tokenolt
Feature activation+0.000
aic
Tokenaic
Feature activation+0.000
dismissed
Token dismissed
Feature activation+0.000
the
Token the
Feature activation+0.000
request
Token request
Feature activation+0.000
and
Token and
Feature activation+0.000
said
Token said
Feature activation+0.000
"
Token "
Feature activation+0.000
Instead
TokenInstead
Feature activation+0.000
of
Token of
Feature activation+0.000
apolog
Token apolog
Feature activation+0.000
ising
Tokenising
Feature activation+0.000
to
Token to
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 H9.2: (feature 11365)

TOP ACTIVATIONS
MAX = 3.749

bris
Tokenbris
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
actual
Token actual
Feature activation+0.000
problem
Token problem
Feature activation+0.000
we
Token we
Feature activation+3.653
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+3.282
facing
Token facing
Feature activation+0.000
is
Token is
Feature activation+0.000
actual
Token actual
Feature activation+0.000
problem
Token problem
Feature activation+0.000
we
Token we
Feature activation+3.653
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+3.282
facing
Token facing
Feature activation+0.000
is
Token is
Feature activation+0.000
to
Token to
Feature activation+0.000
get
Token get
Feature activation+0.000
applications
Token applications
Feature activation+0.000
fixes
Token fixes
Feature activation+0.000
a
Token a
Feature activation+0.000
problem
Token problem
Feature activation+0.000
too
Token too
Feature activation+0.831
many
Token many
Feature activation+0.212
people
Token people
Feature activation+3.255
experience
Token experience
Feature activation+0.393
,
Token,
Feature activation+0.173
in
Token in
Feature activation+0.043
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
Over
Token Over
Feature activation+0.000
cooked
Tokencooked
Feature activation+0.000
and
Token and
Feature activation+0.824
are
Token are
Feature activation+1.972
working
Token working
Feature activation+1.952
to
Token to
Feature activation+3.194
resolve
Token resolve
Feature activation+0.934
those
Token those
Feature activation+1.155
issues
Token issues
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
to
Token to
Feature activation+0.000
understand
Token understand
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
problems
Token problems
Feature activation+0.153
we
Token we
Feature activation+3.057
face
Token face
Feature activation+0.000
now
Token now
Feature activation+0.000
,
Token,
Feature activation+0.000
like
Token like
Feature activation+0.337
climate
Token climate
Feature activation+0.000
information
Token information
Feature activation+0.000
to
Token to
Feature activation+2.107
more
Token more
Feature activation+2.184
quickly
Token quickly
Feature activation+2.567
identify
Token identify
Feature activation+2.323
and
Token and
Feature activation+3.025
solve
Token solve
Feature activation+0.994
this
Token this
Feature activation+1.248
problem
Token problem
Feature activation+0.000
.
Token.
Feature activation+0.038
But
Token But
Feature activation+0.000
those
Token those
Feature activation+0.000
wicked
Token wicked
Feature activation+0.000
problems
Token problems
Feature activation+0.000
we
Token we
Feature activation+1.557
simply
Token simply
Feature activation+1.243
cannot
Token cannot
Feature activation+2.912
arrest
Token arrest
Feature activation+0.303
our
Token our
Feature activation+0.000
way
Token way
Feature activation+0.257
out
Token out
Feature activation+0.083
of
Token of
Feature activation+0.802
problem
Token problem
Feature activation+0.000
,
Token,
Feature activation+0.000
there
Token there
Feature activation+0.000
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.963
another
Token another
Feature activation+2.883
one
Token one
Feature activation+0.711
that
Token that
Feature activation+1.982
comes
Token comes
Feature activation+0.296
up
Token up
Feature activation+0.000
.
Token.
Feature activation+0.079
on
Token on
Feature activation+0.000
problems
Token problems
Feature activation+0.000
.
Token.
Feature activation+0.080
Most
Token Most
Feature activation+0.841
of
Token of
Feature activation+0.437
those
Token those
Feature activation+2.804
glitches
Token glitches
Feature activation+0.000
have
Token have
Feature activation+1.029
since
Token since
Feature activation+1.359
been
Token been
Feature activation+1.775
corrected
Token corrected
Feature activation+0.000
and
Token and
Feature activation+0.000
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
problem
Token problem
Feature activation+0.000
could
Token could
Feature activation+0.860
be
Token be
Feature activation+2.771
solved
Token solved
Feature activation+0.043
by
Token by
Feature activation+0.000
Theresa
Token Theresa
Feature activation+0.000
May
Token May
Feature activation+0.000
triggering
Token triggering
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
back
Token back
Feature activation+0.000
problem
Token problem
Feature activation+0.000
might
Token might
Feature activation+2.193
be
Token be
Feature activation+2.769
a
Token a
Feature activation+1.521
long
Token long
Feature activation+0.201
-
Token-
Feature activation+0.921
term
Tokenterm
Feature activation+0.723
issue
Token issue
Feature activation+0.000
an
Token an
Feature activation+1.557
investigation
Token investigation
Feature activation+0.000
.
Token.
Feature activation+0.406
Why
Token Why
Feature activation+0.597
do
Token do
Feature activation+0.716
some
Token some
Feature activation+2.681
built
Token built
Feature activation+0.000
forms
Token forms
Feature activation+0.264
resemble
Token resemble
Feature activation+0.090
biological
Token biological
Feature activation+0.000
forms
Token forms
Feature activation+0.000
few
Token few
Feature activation+0.000
issues
Token issues
Feature activation+0.000
that
Token that
Feature activation+0.109
we
Token we
Feature activation+1.828
wanted
Token wanted
Feature activation+1.841
to
Token to
Feature activation+2.664
clean
Token clean
Feature activation+0.007
up
Token up
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
problem
Token problem
Feature activation+0.167
...
Token...
Feature activation+0.306
but
Token but
Feature activation+0.270
no
Token no
Feature activation+2.204
willingness
Token willingness
Feature activation+0.000
to
Token to
Feature activation+2.651
commit
Token commit
Feature activation+0.232
to
Token to
Feature activation+2.555
either
Token either
Feature activation+1.358
a
Token a
Feature activation+1.299
strong
Token strong
Feature activation+1.107
actively
Token actively
Feature activation+2.199
working
Token working
Feature activation+2.342
with
Token with
Feature activation+0.809
relevant
Token relevant
Feature activation+0.661
parties
Token parties
Feature activation+0.146
to
Token to
Feature activation+2.623
fix
Token fix
Feature activation+0.846
this
Token this
Feature activation+0.356
problem
Token problem
Feature activation+0.000
.
Token.
Feature activation+0.114
However
Token However
Feature activation+0.000
,
Token,
Feature activation+2.054
insists
Token insists
Feature activation+1.643
it
Token it
Feature activation+1.887
is
Token is
Feature activation+2.573
close
Token close
Feature activation+0.161
to
Token to
Feature activation+2.588
iron
Token iron
Feature activation+0.000
ing
Tokening
Feature activation+0.484
out
Token out
Feature activation+0.444
.
Token.
Feature activation+0.250
Ċ
TokenĊ
Feature activation+0.000
biggest
Token biggest
Feature activation+0.000
problem
Token problem
Feature activation+0.000
we
Token we
Feature activation+1.321
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+2.576
got
Token got
Feature activation+0.000
is
Token is
Feature activation+0.000
physically
Token physically
Feature activation+0.000
being
Token being
Feature activation+0.000
able
Token able
Feature activation+0.000
transit
Token transit
Feature activation+0.044
agency
Token agency
Feature activation+0.403
,
Token,
Feature activation+2.054
insists
Token insists
Feature activation+1.643
it
Token it
Feature activation+1.887
is
Token is
Feature activation+2.573
close
Token close
Feature activation+0.161
to
Token to
Feature activation+2.588
iron
Token iron
Feature activation+0.000
ing
Tokening
Feature activation+0.484
out
Token out
Feature activation+0.444
some
Token some
Feature activation+0.775
additional
Token additional
Feature activation+1.048
information
Token information
Feature activation+0.000
to
Token to
Feature activation+2.107
more
Token more
Feature activation+2.184
quickly
Token quickly
Feature activation+2.567
identify
Token identify
Feature activation+2.323
and
Token and
Feature activation+3.025
solve
Token solve
Feature activation+0.994
this
Token this
Feature activation+1.248
problem
Token problem
Feature activation+0.000
but
Token but
Feature activation+0.270
no
Token no
Feature activation+2.204
willingness
Token willingness
Feature activation+0.000
to
Token to
Feature activation+2.651
commit
Token commit
Feature activation+0.232
to
Token to
Feature activation+2.555
either
Token either
Feature activation+1.358
a
Token a
Feature activation+1.299
strong
Token strong
Feature activation+1.107
message
Token message
Feature activation+0.000
or
Token or
Feature activation+0.000

Top DFA by src position
MAX = 3.566

hy
Tokenhy
Feature activation+0.002
Top resid features:
bris
Tokenbris
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.012
Top resid features:
The
Token The
Feature activation+0.007
Top resid features:
actual
Token actual
Feature activation+0.003
Top resid features:
problem
Token problem
Feature activation+3.566
Top resid features:
we
Token we
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:
re
Tokenre
Feature activation+0.000
Top resid features:
facing
Token facing
Feature activation+0.000
Top resid features:
hy
Tokenhy
Feature activation+0.001
Top resid features:
bris
Tokenbris
Feature activation+0.011
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
The
Token The
Feature activation+0.008
Top resid features:
actual
Token actual
Feature activation+0.001
Top resid features:
problem
Token problem
Feature activation+3.013
Top resid features:
we
Token we
Feature activation-0.086
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation+0.001
Top resid features:
re
Tokenre
Feature activation-0.023
Top resid features:
facing
Token facing
Feature activation+0.000
Top resid features:
project
Token project
Feature activation-0.002
Top resid features:
because
Token because
Feature activation+0.015
Top resid features:
it
Token it
Feature activation+0.004
Top resid features:
fixes
Token fixes
Feature activation+0.045
Top resid features:
a
Token a
Feature activation+0.021
Top resid features:
problem
Token problem
Feature activation+3.460
Top resid features:
too
Token too
Feature activation+0.033
Top resid features:
many
Token many
Feature activation+0.083
Top resid features:
people
Token people
Feature activation+0.108
Top resid features:
experience
Token experience
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
Games
Token Games
Feature activation+0.007
Top resid features:
is
Token is
Feature activation+0.005
Top resid features:
aware
Token aware
Feature activation+0.009
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
technical
Token technical
Feature activation+0.023
Top resid features:
issues
Token issues
Feature activation+3.440
Top resid features:
players
Token players
Feature activation+0.009
Top resid features:
have
Token have
Feature activation+0.022
Top resid features:
faced
Token faced
Feature activation+0.183
Top resid features:
on
Token on
Feature activation+0.005
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
beginning
Token beginning
Feature activation+0.025
Top resid features:
to
Token to
Feature activation+0.014
Top resid features:
understand
Token understand
Feature activation+0.030
Top resid features:
that
Token that
Feature activation+0.039
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
problems
Token problems
Feature activation+3.165
Top resid features:
we
Token we
Feature activation+0.059
Top resid features:
face
Token face
Feature activation+0.000
Top resid features:
now
Token now
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
like
Token like
Feature activation+0.000
Top resid features:
relevant
Token relevant
Feature activation+0.017
Top resid features:
parties
Token parties
Feature activation+0.017
Top resid features:
to
Token to
Feature activation-0.010
Top resid features:
fix
Token fix
Feature activation+0.176
Top resid features:
this
Token this
Feature activation+0.567
Top resid features:
problem
Token problem
Feature activation+1.906
Top resid features:
.
Token.
Feature activation+0.071
Top resid features:
However
Token However
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.015
Top resid features:
we
Token we
Feature activation-0.003
Top resid features:
could
Token could
Feature activation-0.002
Top resid features:
is
Token is
Feature activation+0.001
Top resid features:
one
Token one
Feature activation+0.008
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
those
Token those
Feature activation+0.004
Top resid features:
wicked
Token wicked
Feature activation+0.006
Top resid features:
problems
Token problems
Feature activation+3.467
Top resid features:
we
Token we
Feature activation-0.033
Top resid features:
simply
Token simply
Feature activation+0.016
Top resid features:
cannot
Token cannot
Feature activation+0.021
Top resid features:
arrest
Token arrest
Feature activation+0.000
Top resid features:
our
Token our
Feature activation+0.000
Top resid features:
âĢ
TokenâĢ
Feature activation+0.010
Top resid features:
Ļ
TokenĻ
Feature activation+0.006
Top resid features:
ve
Tokenve
Feature activation+0.009
Top resid features:
solved
Token solved
Feature activation+0.666
Top resid features:
one
Token one
Feature activation+0.096
Top resid features:
problem
Token problem
Feature activation+2.635
Top resid features:
,
Token,
Feature activation+0.063
Top resid features:
there
Token there
Feature activation+0.023
Top resid features:
will
Token will
Feature activation+0.032
Top resid features:
be
Token be
Feature activation+0.009
Top resid features:
another
Token another
Feature activation-0.015
Top resid features:
Nevada
Token Nevada
Feature activation+0.004
Top resid features:
residents
Token residents
Feature activation+0.032
Top resid features:
encountered
Token encountered
Feature activation+0.194
Top resid features:
log
Token log
Feature activation-0.023
Top resid features:
on
Token on
Feature activation-0.072
Top resid features:
problems
Token problems
Feature activation+3.179
Top resid features:
.
Token.
Feature activation+0.067
Top resid features:
Most
Token Most
Feature activation+0.023
Top resid features:
of
Token of
Feature activation+0.022
Top resid features:
those
Token those
Feature activation+0.020
Top resid features:
glitches
Token glitches
Feature activation+0.000
Top resid features:
uncertainty
Token uncertainty
Feature activation+0.037
Top resid features:
,
Token,
Feature activation+0.011
Top resid features:
and
Token and
Feature activation+0.008
Top resid features:
said
Token said
Feature activation+0.060
Top resid features:
the
Token the
Feature activation+0.005
Top resid features:
problem
Token problem
Feature activation+2.988
Top resid features:
could
Token could
Feature activation-0.026
Top resid features:
be
Token be
Feature activation+0.146
Top resid features:
solved
Token solved
Feature activation+0.000
Top resid features:
by
Token by
Feature activation+0.000
Top resid features:
Theresa
Token Theresa
Feature activation+0.000
Top resid features:
ek
Tokenek
Feature activation+0.008
Top resid features:
âĢ
TokenâĢ
Feature activation+0.003
Top resid features:
Ļ
TokenĻ
Feature activation+0.007
Top resid features:
s
Tokens
Feature activation+0.005
Top resid features:
back
Token back
Feature activation-0.018
Top resid features:
problem
Token problem
Feature activation+2.883
Top resid features:
might
Token might
Feature activation+0.162
Top resid features:
be
Token be
Feature activation+0.081
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
long
Token long
Feature activation+0.000
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
some
Token some
Feature activation+0.003
Top resid features:
overall
Token overall
Feature activation+0.018
Top resid features:
map
Token map
Feature activation-0.063
Top resid features:
of
Token of
Feature activation+0.020
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
problem
Token problem
Feature activation+2.996
Top resid features:
.
Token.
Feature activation+0.011
Top resid features:
Because
Token Because
Feature activation+0.003
Top resid features:
this
Token this
Feature activation-0.000
Top resid features:
is
Token is
Feature activation+0.001
Top resid features:
a
Token a
Feature activation+0.003
Top resid features:
but
Token but
Feature activation+0.016
Top resid features:
it
Token it
Feature activation+0.010
Top resid features:
had
Token had
Feature activation+0.003
Top resid features:
a
Token a
Feature activation+0.007
Top resid features:
few
Token few
Feature activation+0.023
Top resid features:
issues
Token issues
Feature activation+1.860
Top resid features:
that
Token that
Feature activation+1.341
Top resid features:
we
Token we
Feature activation+0.022
Top resid features:
wanted
Token wanted
Feature activation+0.042
Top resid features:
to
Token to
Feature activation+0.007
Top resid features:
clean
Token clean
Feature activation+0.000
Top resid features:
sense
Token sense
Feature activation+0.013
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
there
Token there
Feature activation+0.010
Top resid features:
is
Token is
Feature activation+0.005
Top resid features:
a
Token a
Feature activation+0.007
Top resid features:
problem
Token problem
Feature activation+2.689
Top resid features:
...
Token...
Feature activation+0.020
Top resid features:
but
Token but
Feature activation+0.035
Top resid features:
no
Token no
Feature activation+0.022
Top resid features:
willingness
Token willingness
Feature activation+0.134
Top resid features:
to
Token to
Feature activation+0.036
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.118
Top resid features:
are
Token are
Feature activation+0.004
Top resid features:
having
Token having
Feature activation+0.008
Top resid features:
issues
Token issues
Feature activation+2.883
Top resid features:
logging
Token logging
Feature activation-0.003
Top resid features:
in
Token in
Feature activation-0.031
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
League
Token League
Feature activation-0.001
Top resid features:
of
Token of
Feature activation-0.001
Top resid features:
machines
Token machines
Feature activation+0.004
Top resid features:
across
Token across
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
city
Token city
Feature activation-0.001
Top resid features:
,
Token,
Feature activation+0.031
Top resid features:
problems
Token problems
Feature activation+2.033
Top resid features:
that
Token that
Feature activation+0.449
Top resid features:
Metro
Token Metro
Feature activation-0.003
Top resid features:
lin
Tokenlin
Feature activation-0.006
Top resid features:
x
Tokenx
Feature activation+0.021
Top resid features:
,
Token,
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
âĢ
TokenâĢ
Feature activation+0.001
Top resid features:
ľ
Tokenľ
Feature activation+0.002
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
biggest
Token biggest
Feature activation+0.001
Top resid features:
problem
Token problem
Feature activation+3.385
Top resid features:
we
Token we
Feature activation-0.084
Top resid features:
âĢ
TokenâĢ
Feature activation-0.032
Top resid features:
Ļ
TokenĻ
Feature activation+0.001
Top resid features:
ve
Tokenve
Feature activation-0.036
Top resid features:
got
Token got
Feature activation+0.000
Top resid features:
machines
Token machines
Feature activation+0.006
Top resid features:
across
Token across
Feature activation-0.001
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
city
Token city
Feature activation-0.002
Top resid features:
,
Token,
Feature activation+0.013
Top resid features:
problems
Token problems
Feature activation+1.724
Top resid features:
that
Token that
Feature activation+0.725
Top resid features:
Metro
Token Metro
Feature activation+0.007
Top resid features:
lin
Tokenlin
Feature activation-0.001
Top resid features:
x
Tokenx
Feature activation+0.038
Top resid features:
,
Token,
Feature activation+0.008
Top resid features:
relevant
Token relevant
Feature activation+0.022
Top resid features:
parties
Token parties
Feature activation+0.017
Top resid features:
to
Token to
Feature activation-0.008
Top resid features:
fix
Token fix
Feature activation+0.150
Top resid features:
this
Token this
Feature activation+0.469
Top resid features:
problem
Token problem
Feature activation+1.861
Top resid features:
.
Token.
Feature activation+0.155
Top resid features:
However
Token However
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.025
Top resid features:
we
Token we
Feature activation-0.014
Top resid features:
could
Token could
Feature activation-0.001
Top resid features:
sense
Token sense
Feature activation+0.007
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
there
Token there
Feature activation+0.007
Top resid features:
is
Token is
Feature activation+0.006
Top resid features:
a
Token a
Feature activation+0.008
Top resid features:
problem
Token problem
Feature activation+2.505
Top resid features:
...
Token...
Feature activation+0.014
Top resid features:
but
Token but
Feature activation+0.024
Top resid features:
no
Token no
Feature activation+0.011
Top resid features:
willingness
Token willingness
Feature activation+0.084
Top resid features:
to
Token to
Feature activation+0.041
Top resid features:

Decoder Weights Distribution

Head 0: 0.10

Head 1: 0.03

Head 2: 0.48

Head 3: 0.09

Head 4: 0.02

Head 5: 0.07

Head 6: 0.03

Head 7: 0.06

Head 8: 0.03

Head 9: 0.04

Head 10: 0.05

Head 11: 0.02

Positive logits

Solution5.62

solutions4.99

fix4.95

Problem4.87

solved4.87

solve4.70

fixes4.69

Solution4.59

solution4.59

Fix4.58

troubles4.56

solving4.56

remed4.54

problems4.54

Fix4.52

remedy4.50

solves4.48

fixing4.36

fix4.34

remedies4.32

Negative logits

Oath-2.80

earned-2.54

ftime-2.49

ceremonies-2.49

celebr-2.48

emetery-2.48

symbolism-2.47

ceremonial-2.46

ceremony-2.43

oath-2.42

orgetown-2.39

procession-2.32

ceremon-2.32

EGA-2.30

Lago-2.29

hon-2.27

consec-2.26

pilgr-2.26

nods-2.25

antry-2.25

INTERVAL 3.374 - 3.749
CONTAINS 0.000%

bris
Tokenbris
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
actual
Token actual
Feature activation+0.000
problem
Token problem
Feature activation+0.000
we
Token we
Feature activation+3.653
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+3.282
facing
Token facing
Feature activation+0.000
is
Token is
Feature activation+0.000

INTERVAL 2.999 - 3.374
CONTAINS 0.001%

to
Token to
Feature activation+0.000
understand
Token understand
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
problems
Token problems
Feature activation+0.153
we
Token we
Feature activation+3.057
face
Token face
Feature activation+0.000
now
Token now
Feature activation+0.000
,
Token,
Feature activation+0.000
like
Token like
Feature activation+0.337
climate
Token climate
Feature activation+0.000
fixes
Token fixes
Feature activation+0.000
a
Token a
Feature activation+0.000
problem
Token problem
Feature activation+0.000
too
Token too
Feature activation+0.831
many
Token many
Feature activation+0.212
people
Token people
Feature activation+3.255
experience
Token experience
Feature activation+0.393
,
Token,
Feature activation+0.173
in
Token in
Feature activation+0.043
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
information
Token information
Feature activation+0.000
to
Token to
Feature activation+2.107
more
Token more
Feature activation+2.184
quickly
Token quickly
Feature activation+2.567
identify
Token identify
Feature activation+2.323
and
Token and
Feature activation+3.025
solve
Token solve
Feature activation+0.994
this
Token this
Feature activation+1.248
problem
Token problem
Feature activation+0.000
.
Token.
Feature activation+0.038
But
Token But
Feature activation+0.000
Over
Token Over
Feature activation+0.000
cooked
Tokencooked
Feature activation+0.000
and
Token and
Feature activation+0.824
are
Token are
Feature activation+1.972
working
Token working
Feature activation+1.952
to
Token to
Feature activation+3.194
resolve
Token resolve
Feature activation+0.934
those
Token those
Feature activation+1.155
issues
Token issues
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
actual
Token actual
Feature activation+0.000
problem
Token problem
Feature activation+0.000
we
Token we
Feature activation+3.653
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+3.282
facing
Token facing
Feature activation+0.000
is
Token is
Feature activation+0.000
to
Token to
Feature activation+0.000
get
Token get
Feature activation+0.000
applications
Token applications
Feature activation+0.000

INTERVAL 2.624 - 2.999
CONTAINS 0.001%

few
Token few
Feature activation+0.000
issues
Token issues
Feature activation+0.000
that
Token that
Feature activation+0.109
we
Token we
Feature activation+1.828
wanted
Token wanted
Feature activation+1.841
to
Token to
Feature activation+2.664
clean
Token clean
Feature activation+0.007
up
Token up
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
back
Token back
Feature activation+0.000
problem
Token problem
Feature activation+0.000
might
Token might
Feature activation+2.193
be
Token be
Feature activation+2.769
a
Token a
Feature activation+1.521
long
Token long
Feature activation+0.201
-
Token-
Feature activation+0.921
term
Tokenterm
Feature activation+0.723
issue
Token issue
Feature activation+0.000
and
Token and
Feature activation+0.000
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
problem
Token problem
Feature activation+0.000
could
Token could
Feature activation+0.860
be
Token be
Feature activation+2.771
solved
Token solved
Feature activation+0.043
by
Token by
Feature activation+0.000
Theresa
Token Theresa
Feature activation+0.000
May
Token May
Feature activation+0.000
triggering
Token triggering
Feature activation+0.000
problem
Token problem
Feature activation+0.167
...
Token...
Feature activation+0.306
but
Token but
Feature activation+0.270
no
Token no
Feature activation+2.204
willingness
Token willingness
Feature activation+0.000
to
Token to
Feature activation+2.651
commit
Token commit
Feature activation+0.232
to
Token to
Feature activation+2.555
either
Token either
Feature activation+1.358
a
Token a
Feature activation+1.299
strong
Token strong
Feature activation+1.107
those
Token those
Feature activation+0.000
wicked
Token wicked
Feature activation+0.000
problems
Token problems
Feature activation+0.000
we
Token we
Feature activation+1.557
simply
Token simply
Feature activation+1.243
cannot
Token cannot
Feature activation+2.912
arrest
Token arrest
Feature activation+0.303
our
Token our
Feature activation+0.000
way
Token way
Feature activation+0.257
out
Token out
Feature activation+0.083
of
Token of
Feature activation+0.802

INTERVAL 2.249 - 2.624
CONTAINS 0.003%

In
TokenIn
Feature activation+0.226
total
Token total
Feature activation+0.294
,
Token,
Feature activation+0.577
the
Token the
Feature activation+1.218
TTC
Token TTC
Feature activation+1.371
has
Token has
Feature activation+2.410
logged
Token logged
Feature activation+2.039
some
Token some
Feature activation+1.662
620
Token 620
Feature activation+1.969
Prest
Token Prest
Feature activation+0.000
o
Tokeno
Feature activation+0.118
these
Token these
Feature activation+0.300
problems
Token problems
Feature activation+0.000
seem
Token seem
Feature activation+1.367
relatively
Token relatively
Feature activation+2.332
simple
Token simple
Feature activation+0.000
to
Token to
Feature activation+2.333
address
Token address
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
not
Token not
Feature activation+0.314
least
Token least
Feature activation+0.000
because
Token because
Feature activation+0.000
,
Token,
Feature activation+0.182
but
Token but
Feature activation+0.451
again
Token again
Feature activation+0.011
if
Token if
Feature activation+0.323
it
Token it
Feature activation+1.223
was
Token was
Feature activation+2.500
legal
Token legal
Feature activation+0.000
ised
Tokenised
Feature activation+0.000
we
Token we
Feature activation+0.000
wouldn
Token wouldn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
but
Token but
Feature activation+0.270
no
Token no
Feature activation+2.204
willingness
Token willingness
Feature activation+0.000
to
Token to
Feature activation+2.651
commit
Token commit
Feature activation+0.232
to
Token to
Feature activation+2.555
either
Token either
Feature activation+1.358
a
Token a
Feature activation+1.299
strong
Token strong
Feature activation+1.107
message
Token message
Feature activation+0.000
or
Token or
Feature activation+0.000
got
Token got
Feature activation+0.485
to
Token to
Feature activation+2.274
be
Token be
Feature activation+2.337
able
Token able
Feature activation+2.445
to
Token to
Feature activation+2.353
work
Token work
Feature activation+2.434
together
Token together
Feature activation+1.321
on
Token on
Feature activation+0.058
,
Token,
Feature activation+0.000
like
Token like
Feature activation+0.000
regulations
Token regulations
Feature activation+0.000

INTERVAL 1.874 - 2.249
CONTAINS 0.004%

--
Token--
Feature activation+0.000
including
Tokenincluding
Feature activation+0.012
I
Token I
Feature activation+0.000
ve
Tokenve
Feature activation+0.000
--
Token--
Feature activation+0.000
to
Tokento
Feature activation+2.221
fix
Token fix
Feature activation+0.238
these
Token these
Feature activation+0.342
problems
Token problems
Feature activation+0.000
on
Token on
Feature activation+0.000
OS
Token OS
Feature activation+0.000
,
Token,
Feature activation+0.442
problems
Token problems
Feature activation+0.000
that
Token that
Feature activation+1.047
Metro
Token Metro
Feature activation+0.388
lin
Tokenlin
Feature activation+0.000
x
Tokenx
Feature activation+2.042
,
Token,
Feature activation+0.206
the
Token the
Feature activation+0.103
provincial
Token provincial
Feature activation+0.053
transit
Token transit
Feature activation+0.044
agency
Token agency
Feature activation+0.403
could
Token could
Feature activation+0.724
use
Token use
Feature activation+0.642
some
Token some
Feature activation+0.775
additional
Token additional
Feature activation+1.048
information
Token information
Feature activation+0.000
to
Token to
Feature activation+2.107
more
Token more
Feature activation+2.184
quickly
Token quickly
Feature activation+2.567
identify
Token identify
Feature activation+2.323
and
Token and
Feature activation+3.025
solve
Token solve
Feature activation+0.994
plagued
Token plagued
Feature activation+0.152
card
Token card
Feature activation+0.000
load
Token load
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
can
Token can
Feature activation+1.877
't
Token't
Feature activation+2.241
be
Token be
Feature activation+2.455
worked
Token worked
Feature activation+1.611
out
Token out
Feature activation+1.170
,
Token,
Feature activation+0.360
Metro
Token Metro
Feature activation+0.000
host
Token host
Feature activation+0.000
of
Token of
Feature activation+0.000
environmental
Token environmental
Feature activation+0.000
problems
Token problems
Feature activation+0.000
it
Token it
Feature activation+1.196
would
Token would
Feature activation+1.971
bring
Token bring
Feature activation+0.000
leads
Token leads
Feature activation+0.000
us
Token us
Feature activation+0.000
to
Token to
Feature activation+0.000
believe
Token believe
Feature activation+0.000

INTERVAL 1.500 - 1.874
CONTAINS 0.007%

Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.831
We
TokenWe
Feature activation+1.654
're
Token're
Feature activation+1.792
currently
Token currently
Feature activation+1.991
working
Token working
Feature activation+1.771
on
Token on
Feature activation+1.870
an
Token an
Feature activation+1.902
update
Token update
Feature activation+0.000
to
Token to
Feature activation+0.000
improve
Token improve
Feature activation+0.374
one
Token one
Feature activation+0.000
of
Token of
Feature activation+0.000
those
Token those
Feature activation+0.000
wicked
Token wicked
Feature activation+0.000
problems
Token problems
Feature activation+0.000
we
Token we
Feature activation+1.557
simply
Token simply
Feature activation+1.243
cannot
Token cannot
Feature activation+2.912
arrest
Token arrest
Feature activation+0.303
our
Token our
Feature activation+0.000
way
Token way
Feature activation+0.257
of
Token of
Feature activation+0.000
problems
Token problems
Feature activation+0.000
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
they
Token they
Feature activation+0.818
're
Token're
Feature activation+1.645
as
Token as
Feature activation+1.421
divided
Token divided
Feature activation+0.000
as
Token as
Feature activation+0.000
any
Token any
Feature activation+0.000
other
Token other
Feature activation+0.000
Metro
Token Metro
Feature activation+0.000
lin
Tokenlin
Feature activation+0.000
x
Tokenx
Feature activation+0.818
says
Token says
Feature activation+0.939
it
Token it
Feature activation+1.276
has
Token has
Feature activation+1.813
a
Token a
Feature activation+1.077
brand
Token brand
Feature activation+0.615
new
Token new
Feature activation+1.593
generation
Token generation
Feature activation+0.557
of
Token of
Feature activation+1.379
a
Token a
Feature activation+1.048
few
Token few
Feature activation+0.567
lines
Token lines
Feature activation+0.000
of
Token of
Feature activation+1.140
code
Token code
Feature activation+0.000
to
Token to
Feature activation+1.568
fix
Token fix
Feature activation+0.333
,
Token,
Feature activation+0.238
whereas
Token whereas
Feature activation+0.537
in
Token in
Feature activation+0.120
a
Token a
Feature activation+0.477

INTERVAL 1.125 - 1.500
CONTAINS 0.012%

only
Token only
Feature activation+0.105
the
Token the
Feature activation+0.000
current
Token current
Feature activation+0.063
version
Token version
Feature activation+0.000
that
Token that
Feature activation+1.298
is
Token is
Feature activation+1.489
affected
Token affected
Feature activation+0.000
but
Token but
Feature activation+0.108
,
Token,
Feature activation+0.000
given
Token given
Feature activation+0.000
that
Token that
Feature activation+0.250
ers
Tokeners
Feature activation+0.000
can
Token can
Feature activation+1.877
't
Token't
Feature activation+2.241
be
Token be
Feature activation+2.455
worked
Token worked
Feature activation+1.611
out
Token out
Feature activation+1.170
,
Token,
Feature activation+0.360
Metro
Token Metro
Feature activation+0.000
lin
Tokenlin
Feature activation+0.000
x
Tokenx
Feature activation+0.818
says
Token says
Feature activation+0.939
now
Token now
Feature activation+0.000
that
Token that
Feature activation+0.505
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.449
s
Tokens
Feature activation+0.452
not
Token not
Feature activation+1.480
what
Token what
Feature activation+0.000
we
Token we
Feature activation+0.200
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.491
re
Tokenre
Feature activation+0.206
2011
Token 2011
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.145
most
Token most
Feature activation+0.332
companies
Token companies
Feature activation+0.302
don
Token don
Feature activation+0.000
't
Token't
Feature activation+1.265
see
Token see
Feature activation+0.525
changes
Token changes
Feature activation+0.000
in
Token in
Feature activation+0.261
consumer
Token consumer
Feature activation+0.145
purchasing
Token purchasing
Feature activation+0.000
think
Token think
Feature activation+1.155
we
Token we
Feature activation+0.362
might
Token might
Feature activation+1.025
enjoy
Token enjoy
Feature activation+0.000
,
Token,
Feature activation+1.286
please
Token please
Feature activation+1.320
send
Token send
Feature activation+1.639
them
Token them
Feature activation+0.169
in
Token in
Feature activation+0.040
.
Token.
Feature activation+0.101
All
Token All
Feature activation+0.258

INTERVAL 0.750 - 1.125
CONTAINS 0.028%

Update
TokenUpdate
Feature activation+0.000
:
Token:
Feature activation+0.000
It
Token It
Feature activation+0.080
should
Token should
Feature activation+0.407
also
Token also
Feature activation+0.727
be
Token be
Feature activation+0.870
noted
Token noted
Feature activation+0.000
that
Token that
Feature activation+0.000
New
Token New
Feature activation+0.000
Leaf
Token Leaf
Feature activation+0.000
has
Token has
Feature activation+0.000
complaining
Token complaining
Feature activation+0.000
about
Token about
Feature activation+0.000
,
Token,
Feature activation+0.270
and
Token and
Feature activation+0.718
they
Token they
Feature activation+0.547
're
Token're
Feature activation+0.923
typically
Token typically
Feature activation+0.341
older
Token older
Feature activation+0.056
people
Token people
Feature activation+0.000
with
Token with
Feature activation+0.100
serious
Token serious
Feature activation+0.000
government
Token government
Feature activation+0.658
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.133
responsibility
Token responsibility
Feature activation+0.000
to
Token to
Feature activation+0.784
clarify
Token clarify
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
And
TokenAnd
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.179
sure
Token sure
Feature activation+0.000
there
Token there
Feature activation+0.000
will
Token will
Feature activation+0.000
be
Token be
Feature activation+1.068
no
Token no
Feature activation+0.000
further
Token further
Feature activation+0.000
escalation
Token escalation
Feature activation+0.000
of
Token of
Feature activation+0.000
what
Token what
Feature activation+0.000
,"
Token,"
Feature activation+0.000
a
Token a
Feature activation+0.000
representative
Token representative
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
"
Token "
Feature activation+0.899
We
TokenWe
Feature activation+1.250
are
Token are
Feature activation+2.157
working
Token working
Feature activation+1.377
to
Token to
Feature activation+2.134
restore
Token restore
Feature activation+0.432

INTERVAL 0.375 - 0.750
CONTAINS 0.101%

for
Token for
Feature activation+0.410
an
Token an
Feature activation+1.557
investigation
Token investigation
Feature activation+0.000
.
Token.
Feature activation+0.406
Why
Token Why
Feature activation+0.597
do
Token do
Feature activation+0.716
some
Token some
Feature activation+2.681
built
Token built
Feature activation+0.000
forms
Token forms
Feature activation+0.264
resemble
Token resemble
Feature activation+0.090
biological
Token biological
Feature activation+0.000
the
Token the
Feature activation+0.000
product
Token product
Feature activation+0.000
.
Token.
Feature activation+0.000
"
Token "
Feature activation+0.063
We
TokenWe
Feature activation+0.990
're
Token're
Feature activation+0.744
in
Token in
Feature activation+0.186
the
Token the
Feature activation+0.022
process
Token process
Feature activation+0.182
of
Token of
Feature activation+1.337
fixing
Token fixing
Feature activation+0.000
sea
Token sea
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
It
TokenIt
Feature activation+0.347
is
Token is
Feature activation+0.471
unlikely
Token unlikely
Feature activation+0.000
now
Token now
Feature activation+0.000
to
Token to
Feature activation+0.200
persuade
Token persuade
Feature activation+0.000
those
Token those
Feature activation+0.000
hotly
Token hotly
Feature activation+0.000
anticipated
Token anticipated
Feature activation+0.000
.
Token.
Feature activation+0.000
Not
Token Not
Feature activation+0.000
only
Token only
Feature activation+0.595
will
Token will
Feature activation+0.408
1
Token 1
Feature activation+0.000
.
Token.
Feature activation+0.000
4
Token4
Feature activation+0.398
bring
Token bring
Feature activation+0.292
bug
Token bug
Feature activation+0.000
Or
Token Or
Feature activation+0.218
Apple
Token Apple
Feature activation+0.370
says
Token says
Feature activation+0.300
only
Token only
Feature activation+0.706
they
Token they
Feature activation+0.847
should
Token should
Feature activation+0.576
be
Token be
Feature activation+0.322
able
Token able
Feature activation+0.266
to
Token to
Feature activation+0.585
fix
Token fix
Feature activation+0.267
your
Token your
Feature activation+0.000

INTERVAL 0.000 - 0.375
CONTAINS 99.844%

fire
Token fire
Feature activation+0.000
on
Token on
Feature activation+0.000
them
Token them
Feature activation+0.000
with
Token with
Feature activation+0.000
impunity
Token impunity
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
There
TokenThere
Feature activation+0.000
a
Token a
Feature activation+0.000
be
Token be
Feature activation+0.000
eline
Tokeneline
Feature activation+0.000
straight
Token straight
Feature activation+0.000
to
Token to
Feature activation+0.000
Tough
Token Tough
Feature activation+0.000
ie
Tokenie
Feature activation+0.000
's
Token's
Feature activation+0.000
small
Token small
Feature activation+0.000
home
Token home
Feature activation+0.000
.
Token.
Feature activation+0.000
other
Token other
Feature activation+0.000
common
Token common
Feature activation+0.000
alk
Token alk
Feature activation+0.000
ali
Tokenali
Feature activation+0.000
metals
Token metals
Feature activation+0.000
(
Token (
Feature activation+0.000
s
Tokens
Feature activation+0.000
odium
Tokenodium
Feature activation+0.000
and
Token and
Feature activation+0.000
potassium
Token potassium
Feature activation+0.000
)
Token)
Feature activation+0.000
Keep
Token Keep
Feature activation+0.000
on
Token on
Feature activation+0.000
digging
Token digging
Feature activation+0.000
my
Token my
Feature activation+0.000
Foss
Token Foss
Feature activation+0.000
il
Tokenil
Feature activation+0.000
Fun
Token Fun
Feature activation+0.000
atics
Tokenatics
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
be
Token be
Feature activation+0.000
to
Token to
Feature activation+0.000
ADS
Token ADS
Feature activation+0.000
L
TokenL
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
that
Token that
Feature activation+0.000
sort
Token sort
Feature activation+0.000
of
Token of
Feature activation+0.000
future
Token future
Feature activation+0.000
-
Token-
Feature activation+0.000
proof
Tokenproof
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 H9.2: (feature 11431)

TOP ACTIVATIONS
MAX = 5.032

been
Token been
Feature activation+3.644
a
Token a
Feature activation+3.143
lot
Token lot
Feature activation+1.439
of
Token of
Feature activation+3.856
talk
Token talk
Feature activation+0.127
about
Token about
Feature activation+5.032
women
Token women
Feature activation+0.000
and
Token and
Feature activation+0.667
women
Token women
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
while
Token while
Feature activation+1.127
we
Token we
Feature activation+0.000
were
Token were
Feature activation+0.000
saddened
Token saddened
Feature activation+0.023
to
Token to
Feature activation+0.606
see
Token see
Feature activation+5.030
Loc
Token Loc
Feature activation+0.000
ust
Tokenust
Feature activation+0.000
Moon
Token Moon
Feature activation+0.761
close
Token close
Feature activation+0.130
shop
Token shop
Feature activation+0.000
Fund
Token Fund
Feature activation+0.000
for
Token for
Feature activation+0.000
Women
Token Women
Feature activation+0.000
President
Token President
Feature activation+4.999
and
Token and
Feature activation+0.561
CEO
Token CEO
Feature activation+5.025
Mus
Token Mus
Feature activation+0.377
im
Tokenim
Feature activation+0.000
ni
Tokenni
Feature activation+0.000
K
Token K
Feature activation+0.300
any
Tokenany
Feature activation+0.000
was
Token was
Feature activation+0.000
Global
Token Global
Feature activation+0.000
Fund
Token Fund
Feature activation+0.000
for
Token for
Feature activation+0.000
Women
Token Women
Feature activation+0.000
President
Token President
Feature activation+4.999
and
Token and
Feature activation+0.561
CEO
Token CEO
Feature activation+5.025
Mus
Token Mus
Feature activation+0.377
im
Tokenim
Feature activation+0.000
ni
Tokenni
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.674
b
Token b
Feature activation+0.000
antam
Tokenantam
Feature activation+0.000
weight
Tokenweight
Feature activation+1.452
champion
Token champion
Feature activation+4.703
,
Token,
Feature activation+3.664
R
Token R
Feature activation+2.412
onda
Tokenonda
Feature activation+0.000
Rouse
Token Rouse
Feature activation+0.000
y
Tokeny
Feature activation+0.000
Vegas
Token Vegas
Feature activation+0.000
.
Token.
Feature activation+0.280
Her
Token Her
Feature activation+0.000
legendary
Token legendary
Feature activation+0.000
rivalry
Token rivalry
Feature activation+0.000
with
Token with
Feature activation+4.421
UFC
Token UFC
Feature activation+1.461
women
Token women
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.674
Also
TokenAlso
Feature activation+0.000
Read
Token Read
Feature activation+0.442
:
Token:
Feature activation+2.873
Navy
Token Navy
Feature activation+2.101
pitches
Token pitches
Feature activation+2.125
for
Token for
Feature activation+4.109
women
Token women
Feature activation+0.340
pilots
Token pilots
Feature activation+0.000
but
Token but
Feature activation+1.919
no
Token no
Feature activation+3.027
combat
Token combat
Feature activation+1.399
there
Tokenthere
Feature activation+2.273
has
Token has
Feature activation+0.997
been
Token been
Feature activation+3.644
a
Token a
Feature activation+3.143
lot
Token lot
Feature activation+1.439
of
Token of
Feature activation+3.856
talk
Token talk
Feature activation+0.127
about
Token about
Feature activation+5.032
women
Token women
Feature activation+0.000
and
Token and
Feature activation+0.667
women
Token women
Feature activation+0.000
White
Token White
Feature activation+0.000
,
Token,
Feature activation+1.981
Jordan
Token Jordan
Feature activation+0.825
Nob
Token Nob
Feature activation+0.000
bs
Tokenbs
Feature activation+0.000
and
Token and
Feature activation+3.828
T
Token T
Feature activation+1.902
oni
Tokenoni
Feature activation+0.000
Dug
Token Dug
Feature activation+0.000
gan
Tokengan
Feature activation+0.000
were
Token were
Feature activation+0.000
,
Token,
Feature activation+0.091
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
a
Token a
Feature activation+0.793
nod
Token nod
Feature activation+0.000
to
Token to
Feature activation+3.787
the
Token the
Feature activation+2.339
emergence
Token emergence
Feature activation+0.425
of
Token of
Feature activation+1.864
contraceptive
Token contraceptive
Feature activation+0.678
rights
Token rights
Feature activation+0.000
Here
TokenHere
Feature activation+0.239
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.753
The
Token The
Feature activation+2.441
New
Token New
Feature activation+3.690
York
Token York
Feature activation+0.876
Times
Token Times
Feature activation+0.655
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
dest
Tokendest
Feature activation+0.351
s
Tokens
Feature activation+0.674
b
Token b
Feature activation+0.000
antam
Tokenantam
Feature activation+0.000
weight
Tokenweight
Feature activation+1.452
champion
Token champion
Feature activation+4.703
,
Token,
Feature activation+3.664
R
Token R
Feature activation+2.412
onda
Tokenonda
Feature activation+0.000
Rouse
Token Rouse
Feature activation+0.000
y
Tokeny
Feature activation+0.000
,
Token,
Feature activation+0.000
noted
Token noted
Feature activation+1.310
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
there
Tokenthere
Feature activation+2.273
has
Token has
Feature activation+0.997
been
Token been
Feature activation+3.644
a
Token a
Feature activation+3.143
lot
Token lot
Feature activation+1.439
of
Token of
Feature activation+3.856
talk
Token talk
Feature activation+0.127
about
Token about
Feature activation+5.032
to
Token to
Feature activation+1.493
women
Token women
Feature activation+0.000
around
Token around
Feature activation+0.061
the
Token the
Feature activation+0.077
state
Token state
Feature activation+0.000
about
Token about
Feature activation+3.600
the
Token the
Feature activation+2.648
things
Token things
Feature activation+0.357
that
Token that
Feature activation+0.294
matter
Token matter
Feature activation+0.004
most
Token most
Feature activation+0.000
was
Token was
Feature activation+0.000
published
Token published
Feature activation+0.000
in
Token in
Feature activation+1.596
the
Token the
Feature activation+0.682
prestigious
Token prestigious
Feature activation+0.255
journal
Token journal
Feature activation+3.519
Science
Token Science
Feature activation+0.109
in
Token in
Feature activation+0.436
2010
Token 2010
Feature activation+0.000
.
Token.
Feature activation+0.963
Ċ
TokenĊ
Feature activation+0.000
three
Token three
Feature activation+0.113
Service
Token Service
Feature activation+0.000
Chiefs
Token Chiefs
Feature activation+0.000
recently
Token recently
Feature activation+0.000
to
Token to
Feature activation+0.000
discuss
Token discuss
Feature activation+3.495
the
Token the
Feature activation+2.312
issue
Token issue
Feature activation+0.000
.
Token.
Feature activation+0.858
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.363
Well
TokenWell
Feature activation+0.488
,
Token,
Feature activation+2.153
we
Token we
Feature activation+0.000
don
Token don
Feature activation+0.000
't
Token't
Feature activation+0.454
want
Token want
Feature activation+3.471
to
Token to
Feature activation+1.787
publish
Token publish
Feature activation+2.186
it
Token it
Feature activation+1.326
because
Token because
Feature activation+2.994
that
Token that
Feature activation+1.758
-
Token-
Feature activation+0.000
by
Tokenby
Feature activation+0.000
-
Token-
Feature activation+0.000
play
Tokenplay
Feature activation+0.000
)
Token)
Feature activation+0.000
and
Token and
Feature activation+3.417
1976
Token 1976
Feature activation+0.000
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
still
Token still
Feature activation+1.622
has
Token has
Feature activation+2.090
work
Token work
Feature activation+0.064
to
Token to
Feature activation+0.023
do
Token do
Feature activation+0.236
to
Token to
Feature activation+3.234
reach
Token reach
Feature activation+1.826
gender
Token gender
Feature activation+1.056
equality
Token equality
Feature activation+0.128
.
Token.
Feature activation+1.168
âĢ
Token âĢ
Feature activation+0.000
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
Olymp
Token Olymp
Feature activation+0.000
ian
Tokenian
Feature activation+3.228
Ann
Token Ann
Feature activation+0.122
Me
Token Me
Feature activation+0.000
yers
Tokenyers
Feature activation+0.000
(
Token (
Feature activation+0.000
an
Tokenan
Feature activation+0.000

Top DFA by src position
MAX = 5.493

a
Token a
Feature activation+0.001
Top resid features:
White
Token White
Feature activation-0.003
Top resid features:
House
Token House
Feature activation+0.004
Top resid features:
event
Token event
Feature activation+0.002
Top resid features:
on
Token on
Feature activation+0.003
Top resid features:
women
Token women
Feature activation+5.493
Top resid features:
and
Token and
Feature activation+0.036
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
economy
Token economy
Feature activation+0.036
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
Obama
Token Obama
Feature activation+0.005
Top resid features:
in
Token in
Feature activation+0.008
Top resid features:
comics
Token comics
Feature activation+0.037
Top resid features:
in
Token in
Feature activation+0.006
Top resid features:
2015
Token 2015
Feature activation-0.005
Top resid features:
:
Token:
Feature activation+0.002
Top resid features:
women
Token women
Feature activation+5.055
Top resid features:
in
Token in
Feature activation+0.087
Top resid features:
comics
Token comics
Feature activation+0.040
Top resid features:
as
Token as
Feature activation+0.004
Top resid features:
characters
Token characters
Feature activation-0.004
Top resid features:
,
Token,
Feature activation+0.010
Top resid features:
panel
Token panel
Feature activation+0.001
Top resid features:
was
Token was
Feature activation+0.002
Top resid features:
Global
Token Global
Feature activation+0.002
Top resid features:
Fund
Token Fund
Feature activation+0.001
Top resid features:
for
Token for
Feature activation+0.019
Top resid features:
Women
Token Women
Feature activation+5.405
Top resid features:
President
Token President
Feature activation+0.049
Top resid features:
and
Token and
Feature activation+0.016
Top resid features:
CEO
Token CEO
Feature activation+0.008
Top resid features:
Mus
Token Mus
Feature activation+0.000
Top resid features:
im
Tokenim
Feature activation+0.000
Top resid features:
panel
Token panel
Feature activation+0.001
Top resid features:
was
Token was
Feature activation+0.002
Top resid features:
Global
Token Global
Feature activation+0.003
Top resid features:
Fund
Token Fund
Feature activation+0.008
Top resid features:
for
Token for
Feature activation+0.025
Top resid features:
Women
Token Women
Feature activation+5.384
Top resid features:
President
Token President
Feature activation+0.025
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
CEO
Token CEO
Feature activation+0.000
Top resid features:
Mus
Token Mus
Feature activation+0.000
Top resid features:
im
Tokenim
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.131
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
UFC
Token UFC
Feature activation-0.130
Top resid features:
women
Token women
Feature activation+2.204
Top resid features:
âĢ
TokenâĢ
Feature activation-0.004
Top resid features:
Ļ
TokenĻ
Feature activation+0.001
Top resid features:
s
Tokens
Feature activation+0.004
Top resid features:
b
Token b
Feature activation+0.000
Top resid features:
antam
Tokenantam
Feature activation-0.001
Top resid features:
,
Token,
Feature activation+0.008
Top resid features:
a
Token a
Feature activation+0.003
Top resid features:
former
Token former
Feature activation+0.007
Top resid features:
Strike
Token Strike
Feature activation+0.005
Top resid features:
force
Tokenforce
Feature activation-0.013
Top resid features:
Women
Token Women
Feature activation+2.781
Top resid features:
âĢ
TokenâĢ
Feature activation+0.004
Top resid features:
Ļ
TokenĻ
Feature activation+0.001
Top resid features:
s
Tokens
Feature activation+0.002
Top resid features:
b
Token b
Feature activation+0.001
Top resid features:
antam
Tokenantam
Feature activation-0.000
Top resid features:
ways
Token ways
Feature activation+0.005
Top resid features:
in
Token in
Feature activation-0.000
Top resid features:
which
Token which
Feature activation+0.000
Top resid features:
role
Token role
Feature activation+0.027
Top resid features:
of
Token of
Feature activation+0.028
Top resid features:
women
Token women
Feature activation+4.123
Top resid features:
can
Token can
Feature activation+0.001
Top resid features:
be
Token be
Feature activation+0.001
Top resid features:
increased
Token increased
Feature activation+0.047
Top resid features:
in
Token in
Feature activation+0.071
Top resid features:
armed
Token armed
Feature activation+0.060
Top resid features:
a
Token a
Feature activation+0.002
Top resid features:
White
Token White
Feature activation-0.000
Top resid features:
House
Token House
Feature activation-0.000
Top resid features:
event
Token event
Feature activation+0.001
Top resid features:
on
Token on
Feature activation+0.002
Top resid features:
women
Token women
Feature activation+4.045
Top resid features:
and
Token and
Feature activation+0.124
Top resid features:
the
Token the
Feature activation+0.018
Top resid features:
economy
Token economy
Feature activation+0.036
Top resid features:
,
Token,
Feature activation-0.001
Top resid features:
Obama
Token Obama
Feature activation-0.004
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
tr
Tokentr
Feature activation-0.000
Top resid features:
ick
Tokenick
Feature activation-0.010
Top resid features:
,
Token,
Feature activation-0.014
Top resid features:
while
Token while
Feature activation-0.019
Top resid features:
Ellen
Token Ellen
Feature activation+2.481
Top resid features:
White
Token White
Feature activation-0.007
Top resid features:
,
Token,
Feature activation-0.043
Top resid features:
Jordan
Token Jordan
Feature activation+0.014
Top resid features:
Nob
Token Nob
Feature activation-0.008
Top resid features:
bs
Tokenbs
Feature activation-0.001
Top resid features:
a
Token a
Feature activation+0.004
Top resid features:
lot
Token lot
Feature activation+0.005
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
talk
Token talk
Feature activation-0.012
Top resid features:
about
Token about
Feature activation+0.003
Top resid features:
women
Token women
Feature activation+2.365
Top resid features:
and
Token and
Feature activation+0.023
Top resid features:
women
Token women
Feature activation+0.177
Top resid features:
âĢ
TokenâĢ
Feature activation+0.002
Top resid features:
Ļ
TokenĻ
Feature activation+0.006
Top resid features:
s
Tokens
Feature activation+0.058
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.145
Top resid features:
plenty
Token plenty
Feature activation-0.011
Top resid features:
of
Token of
Feature activation-0.010
Top resid features:
female
Token female
Feature activation+2.567
Top resid features:
empowerment
Token empowerment
Feature activation+0.482
Top resid features:
chatter
Token chatter
Feature activation+0.002
Top resid features:
through
Token through
Feature activation-0.003
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
through
Token through
Feature activation-0.001
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
a
Token a
Feature activation+0.001
Top resid features:
former
Token former
Feature activation+0.004
Top resid features:
Strike
Token Strike
Feature activation-0.004
Top resid features:
force
Tokenforce
Feature activation-0.022
Top resid features:
Women
Token Women
Feature activation+2.186
Top resid features:
âĢ
TokenâĢ
Feature activation-0.003
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
s
Tokens
Feature activation-0.001
Top resid features:
b
Token b
Feature activation+0.000
Top resid features:
antam
Tokenantam
Feature activation-0.000
Top resid features:
a
Token a
Feature activation+0.004
Top resid features:
White
Token White
Feature activation-0.000
Top resid features:
House
Token House
Feature activation+0.001
Top resid features:
event
Token event
Feature activation+0.002
Top resid features:
on
Token on
Feature activation+0.002
Top resid features:
women
Token women
Feature activation+3.977
Top resid features:
and
Token and
Feature activation+0.093
Top resid features:
the
Token the
Feature activation+0.012
Top resid features:
economy
Token economy
Feature activation+0.015
Top resid features:
,
Token,
Feature activation-0.001
Top resid features:
Obama
Token Obama
Feature activation+0.002
Top resid features:
they
Token they
Feature activation-0.001
Top resid features:
understand
Token understand
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
needs
Token needs
Feature activation-0.003
Top resid features:
of
Token of
Feature activation+0.004
Top resid features:
women
Token women
Feature activation+2.235
Top resid features:
and
Token and
Feature activation+0.016
Top resid features:
middle
Token middle
Feature activation+0.019
Top resid features:
-
Token-
Feature activation+0.002
Top resid features:
class
Tokenclass
Feature activation+0.001
Top resid features:
families
Token families
Feature activation-0.001
Top resid features:
correlated
Token correlated
Feature activation-0.005
Top resid features:
with
Token with
Feature activation-0.002
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
number
Token number
Feature activation+0.056
Top resid features:
of
Token of
Feature activation-0.004
Top resid features:
women
Token women
Feature activation+3.754
Top resid features:
on
Token on
Feature activation+0.009
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
team
Token team
Feature activation+0.040
Top resid features:
.
Token.
Feature activation+0.064
Top resid features:
This
Token This
Feature activation-0.004
Top resid features:
Read
Token Read
Feature activation+0.000
Top resid features:
:
Token:
Feature activation+0.001
Top resid features:
Navy
Token Navy
Feature activation+0.005
Top resid features:
pitches
Token pitches
Feature activation+0.010
Top resid features:
for
Token for
Feature activation+0.011
Top resid features:
women
Token women
Feature activation+1.743
Top resid features:
pilots
Token pilots
Feature activation+0.036
Top resid features:
but
Token but
Feature activation+0.003
Top resid features:
no
Token no
Feature activation+0.015
Top resid features:
combat
Token combat
Feature activation+0.001
Top resid features:
role
Token role
Feature activation+0.029
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
game
Token game
Feature activation+0.011
Top resid features:
due
Token due
Feature activation+0.001
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
its
Token its
Feature activation+0.003
Top resid features:
female
Token female
Feature activation+3.194
Top resid features:
protagonist
Token protagonist
Feature activation+0.126
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Coverage
Token Coverage
Feature activation-0.151
Top resid features:
of
Token of
Feature activation+0.005
Top resid features:
Team
Token Team
Feature activation+0.004
Top resid features:
USA
Token USA
Feature activation-0.006
Top resid features:
women
Token women
Feature activation+4.364
Top resid features:
âĢ
TokenâĢ
Feature activation+0.011
Top resid features:
Ļ
TokenĻ
Feature activation+0.008
Top resid features:
s
Tokens
Feature activation+0.013
Top resid features:
basketball
Token basketball
Feature activation+0.004
Top resid features:
will
Token will
Feature activation+0.001
Top resid features:
re
Tokenre
Feature activation+0.003
Top resid features:
providing
Token providing
Feature activation+0.002
Top resid features:
adequate
Token adequate
Feature activation+0.111
Top resid features:
support
Token support
Feature activation+0.045
Top resid features:
for
Token for
Feature activation+0.029
Top resid features:
women
Token women
Feature activation+1.979
Top resid features:
?
Token?
Feature activation+0.006
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
Ŀ
TokenĿ
Feature activation+0.002
Top resid features:
Gill
Token Gill
Feature activation+0.000
Top resid features:
iland
Tokeniland
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Coverage
Token Coverage
Feature activation-0.023
Top resid features:
of
Token of
Feature activation+0.002
Top resid features:
Team
Token Team
Feature activation+0.009
Top resid features:
USA
Token USA
Feature activation+0.009
Top resid features:
women
Token women
Feature activation+3.809
Top resid features:
âĢ
TokenâĢ
Feature activation+0.041
Top resid features:
Ļ
TokenĻ
Feature activation+0.007
Top resid features:
s
Tokens
Feature activation+0.007
Top resid features:
basketball
Token basketball
Feature activation+0.004
Top resid features:
will
Token will
Feature activation+0.002
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.02

Head 2: 0.48

Head 3: 0.08

Head 4: 0.04

Head 5: 0.07

Head 6: 0.07

Head 7: 0.06

Head 8: 0.06

Head 9: 0.03

Head 10: 0.03

Head 11: 0.02

Positive logits

Women8.12

Femin7.87

feminists7.77

women7.76

women7.71

feminism7.67

femin7.61

feminist7.60

femin7.41

Feminist7.38

female7.38

female7.36

gender7.36

sexism7.32

Female7.30

Women7.27

Female7.20

gender7.11

sexist7.10

Gender6.96

Negative logits

Reload-3.48

reload-3.28

dilig-2.85

Reviewer-2.83

cryptoc-2.81

seizures-2.68

veland-2.67

dstg-2.64

ovych-2.62

combust-2.58

FEMA-2.54

reper-2.54

seizure-2.50

Moody-2.50

retri-2.45

ipment-2.43

oard-2.40

evac-2.39

coma-2.39

iseum-2.35

INTERVAL 4.529 - 5.032
CONTAINS 0.001%

while
Token while
Feature activation+1.127
we
Token we
Feature activation+0.000
were
Token were
Feature activation+0.000
saddened
Token saddened
Feature activation+0.023
to
Token to
Feature activation+0.606
see
Token see
Feature activation+5.030
Loc
Token Loc
Feature activation+0.000
ust
Tokenust
Feature activation+0.000
Moon
Token Moon
Feature activation+0.761
close
Token close
Feature activation+0.130
shop
Token shop
Feature activation+0.000
been
Token been
Feature activation+3.644
a
Token a
Feature activation+3.143
lot
Token lot
Feature activation+1.439
of
Token of
Feature activation+3.856
talk
Token talk
Feature activation+0.127
about
Token about
Feature activation+5.032
women
Token women
Feature activation+0.000
and
Token and
Feature activation+0.667
women
Token women
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
was
Token was
Feature activation+0.000
Global
Token Global
Feature activation+0.000
Fund
Token Fund
Feature activation+0.000
for
Token for
Feature activation+0.000
Women
Token Women
Feature activation+0.000
President
Token President
Feature activation+4.999
and
Token and
Feature activation+0.561
CEO
Token CEO
Feature activation+5.025
Mus
Token Mus
Feature activation+0.377
im
Tokenim
Feature activation+0.000
ni
Tokenni
Feature activation+0.000
Fund
Token Fund
Feature activation+0.000
for
Token for
Feature activation+0.000
Women
Token Women
Feature activation+0.000
President
Token President
Feature activation+4.999
and
Token and
Feature activation+0.561
CEO
Token CEO
Feature activation+5.025
Mus
Token Mus
Feature activation+0.377
im
Tokenim
Feature activation+0.000
ni
Tokenni
Feature activation+0.000
K
Token K
Feature activation+0.300
any
Tokenany
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.674
b
Token b
Feature activation+0.000
antam
Tokenantam
Feature activation+0.000
weight
Tokenweight
Feature activation+1.452
champion
Token champion
Feature activation+4.703
,
Token,
Feature activation+3.664
R
Token R
Feature activation+2.412
onda
Tokenonda
Feature activation+0.000
Rouse
Token Rouse
Feature activation+0.000
y
Tokeny
Feature activation+0.000

INTERVAL 4.026 - 4.529
CONTAINS 0.000%

Also
TokenAlso
Feature activation+0.000
Read
Token Read
Feature activation+0.442
:
Token:
Feature activation+2.873
Navy
Token Navy
Feature activation+2.101
pitches
Token pitches
Feature activation+2.125
for
Token for
Feature activation+4.109
women
Token women
Feature activation+0.340
pilots
Token pilots
Feature activation+0.000
but
Token but
Feature activation+1.919
no
Token no
Feature activation+3.027
combat
Token combat
Feature activation+1.399
Vegas
Token Vegas
Feature activation+0.000
.
Token.
Feature activation+0.280
Her
Token Her
Feature activation+0.000
legendary
Token legendary
Feature activation+0.000
rivalry
Token rivalry
Feature activation+0.000
with
Token with
Feature activation+4.421
UFC
Token UFC
Feature activation+1.461
women
Token women
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.674

INTERVAL 3.523 - 4.026
CONTAINS 0.001%

White
Token White
Feature activation+0.000
,
Token,
Feature activation+1.981
Jordan
Token Jordan
Feature activation+0.825
Nob
Token Nob
Feature activation+0.000
bs
Tokenbs
Feature activation+0.000
and
Token and
Feature activation+3.828
T
Token T
Feature activation+1.902
oni
Tokenoni
Feature activation+0.000
Dug
Token Dug
Feature activation+0.000
gan
Tokengan
Feature activation+0.000
were
Token were
Feature activation+0.000
s
Tokens
Feature activation+0.674
b
Token b
Feature activation+0.000
antam
Tokenantam
Feature activation+0.000
weight
Tokenweight
Feature activation+1.452
champion
Token champion
Feature activation+4.703
,
Token,
Feature activation+3.664
R
Token R
Feature activation+2.412
onda
Tokenonda
Feature activation+0.000
Rouse
Token Rouse
Feature activation+0.000
y
Tokeny
Feature activation+0.000
,
Token,
Feature activation+0.000
Here
TokenHere
Feature activation+0.239
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.753
The
Token The
Feature activation+2.441
New
Token New
Feature activation+3.690
York
Token York
Feature activation+0.876
Times
Token Times
Feature activation+0.655
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
dest
Tokendest
Feature activation+0.351
there
Tokenthere
Feature activation+2.273
has
Token has
Feature activation+0.997
been
Token been
Feature activation+3.644
a
Token a
Feature activation+3.143
lot
Token lot
Feature activation+1.439
of
Token of
Feature activation+3.856
talk
Token talk
Feature activation+0.127
about
Token about
Feature activation+5.032
women
Token women
Feature activation+0.000
and
Token and
Feature activation+0.667
women
Token women
Feature activation+0.000
,
Token,
Feature activation+0.091
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
a
Token a
Feature activation+0.793
nod
Token nod
Feature activation+0.000
to
Token to
Feature activation+3.787
the
Token the
Feature activation+2.339
emergence
Token emergence
Feature activation+0.425
of
Token of
Feature activation+1.864
contraceptive
Token contraceptive
Feature activation+0.678
rights
Token rights
Feature activation+0.000

INTERVAL 3.019 - 3.523
CONTAINS 0.001%

pitches
Token pitches
Feature activation+2.125
for
Token for
Feature activation+4.109
women
Token women
Feature activation+0.340
pilots
Token pilots
Feature activation+0.000
but
Token but
Feature activation+1.919
no
Token no
Feature activation+3.027
combat
Token combat
Feature activation+1.399
role
Token role
Feature activation+0.647
yet
Token yet
Feature activation+0.769
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.306
is
Token is
Feature activation+0.152
broadly
Token broadly
Feature activation+0.208
credited
Token credited
Feature activation+0.000
with
Token with
Feature activation+1.654
kicking
Token kicking
Feature activation+1.295
off
Token off
Feature activation+3.145
the
Token the
Feature activation+2.887
Second
Token Second
Feature activation+1.935
Wave
Token Wave
Feature activation+2.305
feminist
Token feminist
Feature activation+1.611
movement
Token movement
Feature activation+0.410
-
Token-
Feature activation+0.000
by
Tokenby
Feature activation+0.000
-
Token-
Feature activation+0.000
play
Tokenplay
Feature activation+0.000
)
Token)
Feature activation+0.000
and
Token and
Feature activation+3.417
1976
Token 1976
Feature activation+0.000
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
there
Tokenthere
Feature activation+2.273
has
Token has
Feature activation+0.997
been
Token been
Feature activation+3.644
a
Token a
Feature activation+3.143
lot
Token lot
Feature activation+1.439
of
Token of
Feature activation+3.856
talk
Token talk
Feature activation+0.127
about
Token about
Feature activation+5.032
women
Token women
Feature activation+0.000
Well
TokenWell
Feature activation+0.488
,
Token,
Feature activation+2.153
we
Token we
Feature activation+0.000
don
Token don
Feature activation+0.000
't
Token't
Feature activation+0.454
want
Token want
Feature activation+3.471
to
Token to
Feature activation+1.787
publish
Token publish
Feature activation+2.186
it
Token it
Feature activation+1.326
because
Token because
Feature activation+2.994
that
Token that
Feature activation+1.758

INTERVAL 2.516 - 3.019
CONTAINS 0.003%

hat
Token hat
Feature activation+0.000
-
Token-
Feature activation+0.000
tr
Tokentr
Feature activation+0.000
ick
Tokenick
Feature activation+0.000
,
Token,
Feature activation+0.530
while
Token while
Feature activation+2.813
Ellen
Token Ellen
Feature activation+0.376
White
Token White
Feature activation+0.000
,
Token,
Feature activation+1.981
Jordan
Token Jordan
Feature activation+0.825
Nob
Token Nob
Feature activation+0.000
Samp
Token Samp
Feature activation+0.000
son
Tokenson
Feature activation+0.000
.
Token.
Feature activation+0.062
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.627
J
TokenJ
Feature activation+2.531
od
Tokenod
Feature activation+0.000
ie
Tokenie
Feature activation+0.000
Taylor
Token Taylor
Feature activation+0.000
scored
Token scored
Feature activation+0.080
a
Token a
Feature activation+0.000
multimedia
Token multimedia
Feature activation+0.026
essay
Token essay
Feature activation+0.434
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.008
Represent
TokenRepresent
Feature activation+0.754
ing
Tokening
Feature activation+2.850
Muslims
Token Muslims
Feature activation+0.000
and
Token and
Feature activation+1.292
Islam
Token Islam
Feature activation+0.000
in
Token in
Feature activation+0.001
the
Token the
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
dest
Tokendest
Feature activation+0.351
roying
Tokenroying
Feature activation+1.860
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.115
the
Token the
Feature activation+2.785
film
Token film
Feature activation+0.667
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.575
s
Tokens
Feature activation+2.558
lack
Token lack
Feature activation+0.677
million
Token million
Feature activation+0.000
women
Token women
Feature activation+0.705
experience
Token experience
Feature activation+2.843
at
Token at
Feature activation+0.687
least
Token least
Feature activation+1.890
one
Token one
Feature activation+2.599
incident
Token incident
Feature activation+0.000
of
Token of
Feature activation+1.791
domestic
Token domestic
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
every
Token every
Feature activation+0.000

INTERVAL 2.013 - 2.516
CONTAINS 0.009%

the
Token the
Feature activation+0.708
French
Token French
Feature activation+0.017
Revolution
Token Revolution
Feature activation+0.000
and
Token and
Feature activation+0.000
campaigned
Token campaigned
Feature activation+0.000
for
Token for
Feature activation+2.205
the
Token the
Feature activation+1.073
abolition
Token abolition
Feature activation+0.000
of
Token of
Feature activation+0.444
slavery
Token slavery
Feature activation+0.000
,
Token,
Feature activation+0.000
am
Tokenam
Feature activation+0.000
will
Token will
Feature activation+0.000
carry
Token carry
Feature activation+0.000
the
Token the
Feature activation+0.000
torch
Token torch
Feature activation+0.000
for
Token for
Feature activation+2.105
indie
Token indie
Feature activation+0.011
comics
Token comics
Feature activation+0.000
.
Token.
Feature activation+0.000
Also
Token Also
Feature activation+0.000
the
Token the
Feature activation+0.041
tourism
Token tourism
Feature activation+0.000
destination
Token destination
Feature activation+0.000
,
Token,
Feature activation+1.146
the
Token the
Feature activation+1.189
issue
Token issue
Feature activation+0.000
of
Token of
Feature activation+2.444
wealthy
Token wealthy
Feature activation+0.423
foreigners
Token foreigners
Feature activation+0.000
paying
Token paying
Feature activation+0.953
poor
Token poor
Feature activation+0.134
Indians
Token Indians
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
said
Token said
Feature activation+2.004
the
Token the
Feature activation+2.019
party
Token party
Feature activation+0.141
should
Token should
Feature activation+0.612
introduce
Token introduce
Feature activation+2.383
mandatory
Token mandatory
Feature activation+2.830
quotas
Token quotas
Feature activation+0.137
to
Token to
Feature activation+0.961
boost
Token boost
Feature activation+1.892
the
Token the
Feature activation+1.515
,
Token,
Feature activation+1.960
there
Token there
Feature activation+0.636
will
Token will
Feature activation+0.602
be
Token be
Feature activation+1.653
a
Token a
Feature activation+1.146
million
Token million
Feature activation+2.440
unf
Token unf
Feature activation+1.217
illed
Tokenilled
Feature activation+0.556
computer
Token computer
Feature activation+0.000
science
Token science
Feature activation+0.188
âĢĵ
TokenâĢĵ
Feature activation+0.024

INTERVAL 1.510 - 2.013
CONTAINS 0.019%

az
Tokenaz
Feature activation+0.000
ova
Tokenova
Feature activation+0.000
of
Token of
Feature activation+0.000
Russia
Token Russia
Feature activation+0.000
and
Token and
Feature activation+0.302
Mart
Token Mart
Feature activation+1.712
ina
Tokenina
Feature activation+0.000
Bat
Token Bat
Feature activation+0.000
ini
Tokenini
Feature activation+0.000
of
Token of
Feature activation+0.000
Italy
Token Italy
Feature activation+0.000
when
Token when
Feature activation+0.866
she
Token she
Feature activation+0.241
attended
Token attended
Feature activation+0.732
the
Token the
Feature activation+1.662
State
Token State
Feature activation+0.779
of
Token of
Feature activation+1.669
the
Token the
Feature activation+0.845
Union
Token Union
Feature activation+0.000
address
Token address
Feature activation+0.379
she
Token she
Feature activation+0.000
got
Token got
Feature activation+0.581
to
Token to
Feature activation+0.116
make
Token make
Feature activation+0.255
the
Token the
Feature activation+0.000
career
Token career
Feature activation+0.000
work
Token work
Feature activation+0.000
,
Token,
Feature activation+1.688
there
Token there
Feature activation+0.304
have
Token have
Feature activation+0.073
been
Token been
Feature activation+0.719
the
Token the
Feature activation+0.290
complications
Token complications
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.692
Much
TokenMuch
Feature activation+0.048
like
Token like
Feature activation+0.330
B
Token B
Feature activation+0.000
umble
Tokenumble
Feature activation+0.320
's
Token's
Feature activation+1.678
B
Token B
Feature activation+0.116
FF
TokenFF
Feature activation+0.409
feature
Token feature
Feature activation+0.245
,
Token,
Feature activation+0.873
users
Token users
Feature activation+0.644
)
Token)
Feature activation+1.454
started
Token started
Feature activation+2.025
the
Token the
Feature activation+1.750
group
Token group
Feature activation+0.238
,
Token,
Feature activation+0.792
and
Token and
Feature activation+1.744
Sens
Token Sens
Feature activation+0.018
.
Token.
Feature activation+0.000
Lisa
Token Lisa
Feature activation+0.000
Mur
Token Mur
Feature activation+0.000
kowski
Tokenkowski
Feature activation+0.000

INTERVAL 1.006 - 1.510
CONTAINS 0.063%

IAL
TokenIAL
Feature activation+0.480
S
TokenS
Feature activation+0.000
EX
Token EX
Feature activation+0.000
PL
TokenPL
Feature activation+0.000
AIN
TokenAIN
Feature activation+0.054
WHY
Token WHY
Feature activation+1.198
B
Token B
Feature activation+0.000
ISON
TokenISON
Feature activation+0.000
CAL
Token CAL
Feature activation+0.000
F
TokenF
Feature activation+0.000
WAS
Token WAS
Feature activation+0.012
),
Token),
Feature activation+0.241
but
Token but
Feature activation+0.531
recognizes
Token recognizes
Feature activation+0.747
that
Token that
Feature activation+1.748
the
Token the
Feature activation+1.435
Hutch
Token Hutch
Feature activation+1.193
âĢĶ
Token âĢĶ
Feature activation+0.033
and
Token and
Feature activation+0.856
academic
Token academic
Feature activation+0.538
science
Token science
Feature activation+0.198
overall
Token overall
Feature activation+0.458
who
Token who
Feature activation+0.000
uniformly
Token uniformly
Feature activation+0.000
support
Token support
Feature activation+0.000
women
Token women
Feature activation+0.000
's
Token's
Feature activation+0.559
right
Token right
Feature activation+1.124
to
Token to
Feature activation+0.439
choose
Token choose
Feature activation+0.000
,
Token,
Feature activation+0.000
joined
Token joined
Feature activation+0.000
with
Token with
Feature activation+0.562
choose
Token choose
Feature activation+0.000
,
Token,
Feature activation+0.000
joined
Token joined
Feature activation+0.000
with
Token with
Feature activation+0.562
anti
Token anti
Feature activation+0.025
-
Token-
Feature activation+1.245
abortion
Tokenabortion
Feature activation+0.000
ists
Tokenists
Feature activation+0.000
in
Token in
Feature activation+0.000
rejecting
Token rejecting
Feature activation+2.494
the
Token the
Feature activation+1.855
at
Token at
Feature activation+0.561
any
Token any
Feature activation+0.248
institution
Token institution
Feature activation+0.000
.
Token.
Feature activation+1.795
How
Token How
Feature activation+0.856
do
Token do
Feature activation+1.131
you
Token you
Feature activation+0.869
make
Token make
Feature activation+1.104
sure
Token sure
Feature activation+1.822
you
Token you
Feature activation+0.937
âĢ
TokenâĢ
Feature activation+0.000

INTERVAL 0.503 - 1.006
CONTAINS 0.165%

Yes
TokenYes
Feature activation+1.284
,
Token,
Feature activation+1.947
Ge
Token Ge
Feature activation+0.481
ena
Tokenena
Feature activation+0.000
Davis
Token Davis
Feature activation+0.084
is
Token is
Feature activation+0.714
one
Token one
Feature activation+0.616
of
Token of
Feature activation+1.027
the
Token the
Feature activation+1.706
top
Token top
Feature activation+1.210
32
Token 32
Feature activation+0.884
no
Token no
Feature activation+0.000
woman
Token woman
Feature activation+0.000
who
Token who
Feature activation+0.000
does
Token does
Feature activation+0.000
not
Token not
Feature activation+0.000
support
Token support
Feature activation+0.553
any
Token any
Feature activation+0.413
individuals
Token individuals
Feature activation+0.000
right
Token right
Feature activation+0.000
to
Token to
Feature activation+0.618
live
Token live
Feature activation+0.449
who
Token who
Feature activation+0.000
he
Token he
Feature activation+0.134
would
Token would
Feature activation+0.611
'
Token '
Feature activation+0.000
absolutely
Tokenabsolutely
Feature activation+0.328
'
Token'
Feature activation+0.951
support
Token support
Feature activation+0.000
in
Token in
Feature activation+0.000
2016
Token 2016
Feature activation+0.000
while
Token while
Feature activation+0.000
speaking
Token speaking
Feature activation+0.000
,
Token,
Feature activation+0.080
but
Token but
Feature activation+0.054
the
Token the
Feature activation+0.090
storylines
Token storylines
Feature activation+0.000
too
Token too
Feature activation+0.000
.
Token.
Feature activation+0.938
The
Token The
Feature activation+0.199
first
Token first
Feature activation+0.329
few
Token few
Feature activation+0.235
episodes
Token episodes
Feature activation+0.000
of
Token of
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.605
entirely
Token entirely
Feature activation+0.593
optional
Token optional
Feature activation+0.000
.
Token.
Feature activation+0.671
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.998
As
TokenAs
Feature activation+0.866
many
Token many
Feature activation+0.347
of
Token of
Feature activation+0.185
these
Token these
Feature activation+0.632
women
Token women
Feature activation+0.096

INTERVAL 0.000 - 0.503
CONTAINS 99.737%

and
Token and
Feature activation+0.000
W
Token W
Feature activation+0.000
yden
Tokenyden
Feature activation+0.000
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
fact
Token fact
Feature activation+0.000
sheet
Token sheet
Feature activation+0.000
seems
Token seems
Feature activation+0.000
to
Token to
Feature activation+0.000
contradict
Token contradict
Feature activation+0.000
earlier
Token earlier
Feature activation+0.000
a
Token a
Feature activation+0.101
little
Token little
Feature activation+0.000
biased
Token biased
Feature activation+0.000
because
Token because
Feature activation+0.218
I
Token I
Feature activation+0.000
've
Token've
Feature activation+0.070
known
Token known
Feature activation+0.000
her
Token her
Feature activation+0.000
for
Token for
Feature activation+0.000
years
Token years
Feature activation+0.000
.
Token.
Feature activation+0.000
Fi
TokenFi
Feature activation+0.000
:
Token:
Feature activation+0.000
802
Token 802
Feature activation+0.000
.
Token.
Feature activation+0.000
11
Token11
Feature activation+0.000
a
Token a
Feature activation+0.000
/
Token/
Feature activation+0.000
b
Tokenb
Feature activation+0.000
/
Token/
Feature activation+0.000
g
Tokeng
Feature activation+0.000
/
Token/
Feature activation+0.000
ia
Tokenia
Feature activation+0.000
di
Token di
Feature activation+0.000
o
Token o
Feature activation+0.000
ss
Tokenss
Feature activation+0.000
erv
Tokenerv
Feature activation+0.000
az
Tokenaz
Feature activation+0.000
ion
Tokenion
Feature activation+0.000
i
Tokeni
Feature activation+0.000
te
Token te
Feature activation+0.000
oric
Tokenoric
Feature activation+0.000
he
Tokenhe
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
Saturday
Token Saturday
Feature activation+0.000
's
Token's
Feature activation+0.000
not
Token not
Feature activation+0.000
about
Token about
Feature activation+0.000
me
Token me
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
's
Token's
Feature activation+0.000
about
Token about
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 H9.2: (feature 17913)

TOP ACTIVATIONS
MAX = 3.271

issue
Token issue
Feature activation+0.000
that
Token that
Feature activation+0.731
blows
Token blows
Feature activation+0.425
up
Token up
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+1.113
like
Token like
Feature activation+3.271
when
Token when
Feature activation+2.463
the
Token the
Feature activation+2.250
roof
Token roof
Feature activation+0.408
tore
Token tore
Feature activation+0.000
off
Token off
Feature activation+0.261
will
Token will
Feature activation+0.135
invest
Token invest
Feature activation+0.091
more
Token more
Feature activation+0.571
effort
Token effort
Feature activation+1.056
in
Token in
Feature activation+2.981
improving
Token improving
Feature activation+3.258
the
Token the
Feature activation+2.216
safety
Token safety
Feature activation+0.000
of
Token of
Feature activation+0.904
their
Token their
Feature activation+1.320
fleets
Token fleets
Feature activation+0.121
to
Token to
Feature activation+0.000
increase
Token increase
Feature activation+0.000
safety
Token safety
Feature activation+0.000
,
Token,
Feature activation+0.770
like
Token like
Feature activation+2.025
installing
Token installing
Feature activation+3.118
flashing
Token flashing
Feature activation+0.628
stop
Token stop
Feature activation+0.000
signs
Token signs
Feature activation+0.181
or
Token or
Feature activation+0.585
electronic
Token electronic
Feature activation+0.467
of
Token of
Feature activation+0.000
Air
Token Air
Feature activation+0.000
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
care
Token care
Feature activation+0.000
and
Token and
Feature activation+3.103
safe
Token safe
Feature activation+0.000
handling
Token handling
Feature activation+0.000
of
Token of
Feature activation+0.000
your
Token your
Feature activation+0.000
air
Token air
Feature activation+0.649
they
Token they
Feature activation+0.000
will
Token will
Feature activation+0.135
invest
Token invest
Feature activation+0.091
more
Token more
Feature activation+0.571
effort
Token effort
Feature activation+1.056
in
Token in
Feature activation+2.981
improving
Token improving
Feature activation+3.258
the
Token the
Feature activation+2.216
safety
Token safety
Feature activation+0.000
of
Token of
Feature activation+0.904
their
Token their
Feature activation+1.320
,
Token,
Feature activation+0.000
assume
Token assume
Feature activation+0.000
more
Token more
Feature activation+0.000
risk
Token risk
Feature activation+0.000
(
Token (
Feature activation+0.346
since
Tokensince
Feature activation+2.812
they
Token they
Feature activation+1.799
could
Token could
Feature activation+0.447
get
Token get
Feature activation+0.969
caught
Token caught
Feature activation+0.000
trying
Token trying
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
the
Token the
Feature activation+0.000
annual
Token annual
Feature activation+0.000
safety
Token safety
Feature activation+0.000
lecture
Token lecture
Feature activation+0.000
on
Token on
Feature activation+2.656
BBC
Token BBC
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+1.997
âĢ
Token âĢ
Feature activation+0.000
first
Token first
Feature activation+0.000
and
Token and
Feature activation+0.000
foremost
Token foremost
Feature activation+1.263
-
Token -
Feature activation+0.957
if
Token if
Feature activation+2.291
a
Token a
Feature activation+2.653
player
Token player
Feature activation+1.467
gets
Token gets
Feature activation+2.504
an
Token an
Feature activation+2.020
injury
Token injury
Feature activation+0.000
,
Token,
Feature activation+0.371
that
Token that
Feature activation+0.675
there
Token there
Feature activation+0.252
was
Token was
Feature activation+1.567
likely
Token likely
Feature activation+1.371
to
Token to
Feature activation+0.845
be
Token be
Feature activation+2.592
a
Token a
Feature activation+2.148
pick
Token pick
Feature activation+0.000
et
Tokenet
Feature activation+0.000
with
Token with
Feature activation+0.398
lots
Token lots
Feature activation+0.000
a
Token a
Feature activation+0.000
daily
Token daily
Feature activation+0.000
basis
Token basis
Feature activation+0.000
by
Token by
Feature activation+1.144
highly
Token highly
Feature activation+1.091
trained
Token trained
Feature activation+2.547
engineers
Token engineers
Feature activation+0.000
to
Token to
Feature activation+0.249
ensure
Token ensure
Feature activation+1.330
that
Token that
Feature activation+1.112
they
Token they
Feature activation+1.529
foremost
Token foremost
Feature activation+1.263
-
Token -
Feature activation+0.957
if
Token if
Feature activation+2.291
a
Token a
Feature activation+2.653
player
Token player
Feature activation+1.467
gets
Token gets
Feature activation+2.504
an
Token an
Feature activation+2.020
injury
Token injury
Feature activation+0.000
,
Token,
Feature activation+0.371
it
Token it
Feature activation+0.264
's
Token's
Feature activation+0.603
that
Token that
Feature activation+0.731
blows
Token blows
Feature activation+0.425
up
Token up
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+1.113
like
Token like
Feature activation+3.271
when
Token when
Feature activation+2.463
the
Token the
Feature activation+2.250
roof
Token roof
Feature activation+0.408
tore
Token tore
Feature activation+0.000
off
Token off
Feature activation+0.261
a
Token a
Feature activation+0.833
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
legal
Token legal
Feature activation+0.000
risks
Token risks
Feature activation+0.000
contractors
Token contractors
Feature activation+0.543
could
Token could
Feature activation+2.324
face
Token face
Feature activation+0.000
from
Token from
Feature activation+0.000
Canada
Token Canada
Feature activation+0.000
,
Token,
Feature activation+0.000
Eck
Token Eck
Feature activation+0.000
,
Token,
Feature activation+0.794
it
Token it
Feature activation+0.811
isn
Token isn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+2.300
.
Token.
Feature activation+0.425
On
Token On
Feature activation+0.710
the
Token the
Feature activation+0.376
same
Token same
Feature activation+0.000
note
Token note
Feature activation+0.000
is
Token is
Feature activation+0.081
first
Token first
Feature activation+0.000
and
Token and
Feature activation+0.000
foremost
Token foremost
Feature activation+1.263
-
Token -
Feature activation+0.957
if
Token if
Feature activation+2.291
a
Token a
Feature activation+2.653
player
Token player
Feature activation+1.467
gets
Token gets
Feature activation+2.504
an
Token an
Feature activation+2.020
injury
Token injury
Feature activation+0.000
the
Token the
Feature activation+0.752
tools
Token tools
Feature activation+0.000
we
Token we
Feature activation+0.000
have
Token have
Feature activation+0.000
will
Token will
Feature activation+0.736
not
Token not
Feature activation+2.268
be
Token be
Feature activation+0.942
effective
Token effective
Feature activation+1.290
unless
Token unless
Feature activation+0.701
drivers
Token drivers
Feature activation+0.563
adjust
Token adjust
Feature activation+0.474
blows
Token blows
Feature activation+0.425
up
Token up
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+1.113
like
Token like
Feature activation+3.271
when
Token when
Feature activation+2.463
the
Token the
Feature activation+2.250
roof
Token roof
Feature activation+0.408
tore
Token tore
Feature activation+0.000
off
Token off
Feature activation+0.261
a
Token a
Feature activation+0.833
Southwest
Token Southwest
Feature activation+0.000
became
Token became
Feature activation+0.000
Chairman
Token Chairman
Feature activation+0.000
,
Token,
Feature activation+1.345
I
Token I
Feature activation+0.420
observed
Token observed
Feature activation+0.908
that
Token that
Feature activation+2.249
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.256
[
Token[
Feature activation+0.188
y
Tokeny
Feature activation+0.000
]
Token]
Feature activation+0.000
of
Token of
Feature activation+1.177
the
Token the
Feature activation+1.087
bank
Token bank
Feature activation+0.461
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+2.223
v
Token v
Feature activation+0.053
aults
Tokenaults
Feature activation+0.900
near
Token near
Feature activation+0.000
Kh
Token Kh
Feature activation+0.000
air
Tokenair
Feature activation+0.000
invest
Token invest
Feature activation+0.091
more
Token more
Feature activation+0.571
effort
Token effort
Feature activation+1.056
in
Token in
Feature activation+2.981
improving
Token improving
Feature activation+3.258
the
Token the
Feature activation+2.216
safety
Token safety
Feature activation+0.000
of
Token of
Feature activation+0.904
their
Token their
Feature activation+1.320
fleets
Token fleets
Feature activation+0.121
,
Token,
Feature activation+0.714

Top DFA by src position
MAX = 3.788

Unless
TokenUnless
Feature activation+0.005
Top resid features:
it
Token it
Feature activation+0.011
Top resid features:
's
Token's
Feature activation+0.003
Top resid features:
a
Token a
Feature activation-0.003
Top resid features:
real
Token real
Feature activation-0.001
Top resid features:
safety
Token safety
Feature activation+3.788
Top resid features:
issue
Token issue
Feature activation+0.047
Top resid features:
that
Token that
Feature activation+0.015
Top resid features:
blows
Token blows
Feature activation-0.003
Top resid features:
up
Token up
Feature activation+0.025
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.071
Top resid features:
they
Token they
Feature activation-0.015
Top resid features:
are
Token are
Feature activation-0.013
Top resid features:
being
Token being
Feature activation-0.010
Top resid features:
evaluated
Token evaluated
Feature activation+0.207
Top resid features:
for
Token for
Feature activation+0.023
Top resid features:
safety
Token safety
Feature activation+3.510
Top resid features:
,
Token,
Feature activation+0.017
Top resid features:
they
Token they
Feature activation+0.018
Top resid features:
will
Token will
Feature activation+0.012
Top resid features:
invest
Token invest
Feature activation+0.014
Top resid features:
more
Token more
Feature activation+0.010
Top resid features:
"
Token"
Feature activation+0.009
Top resid features:
of
Token of
Feature activation-0.003
Top resid features:
methods
Token methods
Feature activation+0.005
Top resid features:
to
Token to
Feature activation-0.004
Top resid features:
increase
Token increase
Feature activation+0.007
Top resid features:
safety
Token safety
Feature activation+3.727
Top resid features:
,
Token,
Feature activation+0.011
Top resid features:
like
Token like
Feature activation-0.067
Top resid features:
installing
Token installing
Feature activation-0.009
Top resid features:
flashing
Token flashing
Feature activation+0.000
Top resid features:
stop
Token stop
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
For
Token For
Feature activation-0.003
Top resid features:
extra
Token extra
Feature activation-0.003
Top resid features:
care
Token care
Feature activation+0.054
Top resid features:
and
Token and
Feature activation+0.015
Top resid features:
safe
Token safe
Feature activation+3.205
Top resid features:
handling
Token handling
Feature activation+0.276
Top resid features:
of
Token of
Feature activation+0.046
Top resid features:
your
Token your
Feature activation+0.005
Top resid features:
air
Token air
Feature activation-0.004
Top resid features:
,
Token,
Feature activation+0.019
Top resid features:
they
Token they
Feature activation-0.014
Top resid features:
are
Token are
Feature activation-0.012
Top resid features:
being
Token being
Feature activation-0.013
Top resid features:
evaluated
Token evaluated
Feature activation+0.354
Top resid features:
for
Token for
Feature activation+0.026
Top resid features:
safety
Token safety
Feature activation+3.038
Top resid features:
,
Token,
Feature activation+0.019
Top resid features:
they
Token they
Feature activation+0.025
Top resid features:
will
Token will
Feature activation+0.022
Top resid features:
invest
Token invest
Feature activation+0.024
Top resid features:
more
Token more
Feature activation+0.016
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.103
Top resid features:
,
Token,
Feature activation+0.022
Top resid features:
assume
Token assume
Feature activation+0.016
Top resid features:
more
Token more
Feature activation+0.015
Top resid features:
risk
Token risk
Feature activation+3.166
Top resid features:
(
Token (
Feature activation+0.119
Top resid features:
since
Tokensince
Feature activation+0.026
Top resid features:
they
Token they
Feature activation+0.000
Top resid features:
could
Token could
Feature activation+0.000
Top resid features:
get
Token get
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.084
Top resid features:
the
Token the
Feature activation+0.017
Top resid features:
annual
Token annual
Feature activation-0.008
Top resid features:
safety
Token safety
Feature activation+3.187
Top resid features:
lecture
Token lecture
Feature activation-0.041
Top resid features:
on
Token on
Feature activation+0.075
Top resid features:
BBC
Token BBC
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:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.070
Top resid features:
A
Token A
Feature activation+0.002
Top resid features:
player
Token player
Feature activation-0.019
Top resid features:
's
Token's
Feature activation+0.002
Top resid features:
safety
Token safety
Feature activation+3.037
Top resid features:
is
Token is
Feature activation+0.016
Top resid features:
first
Token first
Feature activation+0.012
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
foremost
Token foremost
Feature activation+0.007
Top resid features:
-
Token -
Feature activation+0.034
Top resid features:
"
Token "
Feature activation-0.000
Top resid features:
could
Tokencould
Feature activation+0.008
Top resid features:
not
Token not
Feature activation-0.041
Top resid features:
guarantee
Token guarantee
Feature activation+0.123
Top resid features:
the
Token the
Feature activation+0.044
Top resid features:
safety
Token safety
Feature activation+2.499
Top resid features:
of
Token of
Feature activation+0.226
Top resid features:
students
Token students
Feature activation+0.063
Top resid features:
".
Token".
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
with
Token with
Feature activation+0.000
Top resid features:
eight
Token eight
Feature activation-0.003
Top resid features:
slides
Token slides
Feature activation+0.011
Top resid features:
for
Token for
Feature activation+0.008
Top resid features:
maximum
Token maximum
Feature activation+0.003
Top resid features:
safety
Token safety
Feature activation+2.529
Top resid features:
and
Token and
Feature activation+0.015
Top resid features:
efficiency
Token efficiency
Feature activation+0.018
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.008
Top resid features:
and
Token and
Feature activation+0.010
Top resid features:
these
Token these
Feature activation+0.007
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.093
Top resid features:
A
Token A
Feature activation+0.011
Top resid features:
player
Token player
Feature activation-0.064
Top resid features:
's
Token's
Feature activation+0.001
Top resid features:
safety
Token safety
Feature activation+2.675
Top resid features:
is
Token is
Feature activation+0.005
Top resid features:
first
Token first
Feature activation+0.010
Top resid features:
and
Token and
Feature activation+0.002
Top resid features:
foremost
Token foremost
Feature activation+0.004
Top resid features:
-
Token -
Feature activation+0.017
Top resid features:
Unless
TokenUnless
Feature activation+0.015
Top resid features:
it
Token it
Feature activation+0.005
Top resid features:
's
Token's
Feature activation+0.003
Top resid features:
a
Token a
Feature activation+0.003
Top resid features:
real
Token real
Feature activation+0.008
Top resid features:
safety
Token safety
Feature activation+2.934
Top resid features:
issue
Token issue
Feature activation+0.043
Top resid features:
that
Token that
Feature activation+0.012
Top resid features:
blows
Token blows
Feature activation+0.001
Top resid features:
up
Token up
Feature activation+0.023
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.086
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
Asked
TokenAsked
Feature activation-0.002
Top resid features:
about
Token about
Feature activation+0.004
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
legal
Token legal
Feature activation+0.006
Top resid features:
risks
Token risks
Feature activation+2.821
Top resid features:
contractors
Token contractors
Feature activation+0.043
Top resid features:
could
Token could
Feature activation+0.064
Top resid features:
face
Token face
Feature activation+0.000
Top resid features:
from
Token from
Feature activation+0.000
Top resid features:
Canada
Token Canada
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.038
Top resid features:
Is
Token Is
Feature activation+0.008
Top resid features:
your
Token your
Feature activation+0.002
Top resid features:
robot
Token robot
Feature activation+0.007
Top resid features:
still
Token still
Feature activation+0.048
Top resid features:
safe
Token safe
Feature activation+2.413
Top resid features:
?
Token?
Feature activation+0.064
Top resid features:
No
Token No
Feature activation+0.018
Top resid features:
,
Token,
Feature activation+0.092
Top resid features:
it
Token it
Feature activation+0.057
Top resid features:
isn
Token isn
Feature activation+0.084
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.125
Top resid features:
A
Token A
Feature activation+0.003
Top resid features:
player
Token player
Feature activation-0.028
Top resid features:
's
Token's
Feature activation-0.031
Top resid features:
safety
Token safety
Feature activation+2.769
Top resid features:
is
Token is
Feature activation+0.011
Top resid features:
first
Token first
Feature activation+0.018
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
foremost
Token foremost
Feature activation+0.006
Top resid features:
-
Token -
Feature activation+0.061
Top resid features:
't
Token't
Feature activation+0.009
Top resid features:
be
Token be
Feature activation-0.001
Top resid features:
enough
Token enough
Feature activation-0.009
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
guarantee
Token guarantee
Feature activation+0.321
Top resid features:
safety
Token safety
Feature activation+2.036
Top resid features:
.
Token.
Feature activation+0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
"
Token"
Feature activation+0.032
Top resid features:
All
TokenAll
Feature activation+0.005
Top resid features:
Unless
TokenUnless
Feature activation+0.010
Top resid features:
it
Token it
Feature activation+0.003
Top resid features:
's
Token's
Feature activation+0.002
Top resid features:
a
Token a
Feature activation+0.003
Top resid features:
real
Token real
Feature activation+0.012
Top resid features:
safety
Token safety
Feature activation+2.715
Top resid features:
issue
Token issue
Feature activation+0.030
Top resid features:
that
Token that
Feature activation+0.016
Top resid features:
blows
Token blows
Feature activation+0.001
Top resid features:
up
Token up
Feature activation+0.014
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.023
Top resid features:
specifically
Token specifically
Feature activation+0.000
Top resid features:
pointed
Token pointed
Feature activation+0.009
Top resid features:
out
Token out
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
public
Token public
Feature activation-0.009
Top resid features:
safety
Token safety
Feature activation+2.545
Top resid features:
benefits
Token benefits
Feature activation+0.028
Top resid features:
of
Token of
Feature activation+0.023
Top resid features:
doing
Token doing
Feature activation-0.013
Top resid features:
so
Token so
Feature activation-0.003
Top resid features:
.
Token.
Feature activation+0.019
Top resid features:
Ļ
TokenĻ
Feature activation+0.001
Top resid features:
t
Tokent
Feature activation+0.001
Top resid features:
know
Token know
Feature activation+0.007
Top resid features:
about
Token about
Feature activation+0.009
Top resid features:
the
Token the
Feature activation+0.020
Top resid features:
safety
Token safety
Feature activation+1.962
Top resid features:
of
Token of
Feature activation+0.310
Top resid features:
the
Token the
Feature activation+0.096
Top resid features:
bank
Token bank
Feature activation+0.183
Top resid features:
âĢ
TokenâĢ
Feature activation+0.003
Top resid features:
Ļ
TokenĻ
Feature activation+0.013
Top resid features:
they
Token they
Feature activation-0.006
Top resid features:
are
Token are
Feature activation-0.006
Top resid features:
being
Token being
Feature activation-0.009
Top resid features:
evaluated
Token evaluated
Feature activation+0.233
Top resid features:
for
Token for
Feature activation+0.027
Top resid features:
safety
Token safety
Feature activation+2.402
Top resid features:
,
Token,
Feature activation+0.009
Top resid features:
they
Token they
Feature activation+0.012
Top resid features:
will
Token will
Feature activation+0.013
Top resid features:
invest
Token invest
Feature activation+0.024
Top resid features:
more
Token more
Feature activation+0.005
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.02

Head 2: 0.48

Head 3: 0.09

Head 4: 0.03

Head 5: 0.07

Head 6: 0.06

Head 7: 0.04

Head 8: 0.04

Head 9: 0.06

Head 10: 0.03

Head 11: 0.01

Positive logits

Safety6.64

safety6.51

safety6.46

risks6.18

safer6.08

Safety6.06

hazards6.04

safest6.04

hazard6.01

afety5.96

dangers5.88

risk5.79

unsafe5.57

precautions5.52

precaution5.41

Risk5.31

risk5.07

danger4.97

hazard4.94

danger4.89

Negative logits

�醒-2.82

ithing-2.74

Abram-2.69

Transcript-2.64

phia-2.57

colonial-2.55

Bliss-2.51

utenberg-2.48

rans-2.48

Glory-2.47

resent-2.44

phabet-2.44

izzle-2.42

isan-2.41

ieg-2.41

Latin-2.39

issance-2.39

orate-2.38

ocrat-2.36

Marx-2.36

INTERVAL 2.944 - 3.271
CONTAINS 0.001%

to
Token to
Feature activation+0.000
increase
Token increase
Feature activation+0.000
safety
Token safety
Feature activation+0.000
,
Token,
Feature activation+0.770
like
Token like
Feature activation+2.025
installing
Token installing
Feature activation+3.118
flashing
Token flashing
Feature activation+0.628
stop
Token stop
Feature activation+0.000
signs
Token signs
Feature activation+0.181
or
Token or
Feature activation+0.585
electronic
Token electronic
Feature activation+0.467
issue
Token issue
Feature activation+0.000
that
Token that
Feature activation+0.731
blows
Token blows
Feature activation+0.425
up
Token up
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+1.113
like
Token like
Feature activation+3.271
when
Token when
Feature activation+2.463
the
Token the
Feature activation+2.250
roof
Token roof
Feature activation+0.408
tore
Token tore
Feature activation+0.000
off
Token off
Feature activation+0.261
they
Token they
Feature activation+0.000
will
Token will
Feature activation+0.135
invest
Token invest
Feature activation+0.091
more
Token more
Feature activation+0.571
effort
Token effort
Feature activation+1.056
in
Token in
Feature activation+2.981
improving
Token improving
Feature activation+3.258
the
Token the
Feature activation+2.216
safety
Token safety
Feature activation+0.000
of
Token of
Feature activation+0.904
their
Token their
Feature activation+1.320
will
Token will
Feature activation+0.135
invest
Token invest
Feature activation+0.091
more
Token more
Feature activation+0.571
effort
Token effort
Feature activation+1.056
in
Token in
Feature activation+2.981
improving
Token improving
Feature activation+3.258
the
Token the
Feature activation+2.216
safety
Token safety
Feature activation+0.000
of
Token of
Feature activation+0.904
their
Token their
Feature activation+1.320
fleets
Token fleets
Feature activation+0.121
of
Token of
Feature activation+0.000
Air
Token Air
Feature activation+0.000
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
care
Token care
Feature activation+0.000
and
Token and
Feature activation+3.103
safe
Token safe
Feature activation+0.000
handling
Token handling
Feature activation+0.000
of
Token of
Feature activation+0.000
your
Token your
Feature activation+0.000
air
Token air
Feature activation+0.649

INTERVAL 2.617 - 2.944
CONTAINS 0.000%

,
Token,
Feature activation+0.000
assume
Token assume
Feature activation+0.000
more
Token more
Feature activation+0.000
risk
Token risk
Feature activation+0.000
(
Token (
Feature activation+0.346
since
Tokensince
Feature activation+2.812
they
Token they
Feature activation+1.799
could
Token could
Feature activation+0.447
get
Token get
Feature activation+0.969
caught
Token caught
Feature activation+0.000
trying
Token trying
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
the
Token the
Feature activation+0.000
annual
Token annual
Feature activation+0.000
safety
Token safety
Feature activation+0.000
lecture
Token lecture
Feature activation+0.000
on
Token on
Feature activation+2.656
BBC
Token BBC
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+1.997
âĢ
Token âĢ
Feature activation+0.000
first
Token first
Feature activation+0.000
and
Token and
Feature activation+0.000
foremost
Token foremost
Feature activation+1.263
-
Token -
Feature activation+0.957
if
Token if
Feature activation+2.291
a
Token a
Feature activation+2.653
player
Token player
Feature activation+1.467
gets
Token gets
Feature activation+2.504
an
Token an
Feature activation+2.020
injury
Token injury
Feature activation+0.000
,
Token,
Feature activation+0.371

INTERVAL 2.290 - 2.617
CONTAINS 0.001%

that
Token that
Feature activation+0.675
there
Token there
Feature activation+0.252
was
Token was
Feature activation+1.567
likely
Token likely
Feature activation+1.371
to
Token to
Feature activation+0.845
be
Token be
Feature activation+2.592
a
Token a
Feature activation+2.148
pick
Token pick
Feature activation+0.000
et
Tokenet
Feature activation+0.000
with
Token with
Feature activation+0.398
lots
Token lots
Feature activation+0.000
,
Token,
Feature activation+0.794
it
Token it
Feature activation+0.811
isn
Token isn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+2.300
.
Token.
Feature activation+0.425
On
Token On
Feature activation+0.710
the
Token the
Feature activation+0.376
same
Token same
Feature activation+0.000
note
Token note
Feature activation+0.000
foremost
Token foremost
Feature activation+1.263
-
Token -
Feature activation+0.957
if
Token if
Feature activation+2.291
a
Token a
Feature activation+2.653
player
Token player
Feature activation+1.467
gets
Token gets
Feature activation+2.504
an
Token an
Feature activation+2.020
injury
Token injury
Feature activation+0.000
,
Token,
Feature activation+0.371
it
Token it
Feature activation+0.264
's
Token's
Feature activation+0.603
a
Token a
Feature activation+0.000
daily
Token daily
Feature activation+0.000
basis
Token basis
Feature activation+0.000
by
Token by
Feature activation+1.144
highly
Token highly
Feature activation+1.091
trained
Token trained
Feature activation+2.547
engineers
Token engineers
Feature activation+0.000
to
Token to
Feature activation+0.249
ensure
Token ensure
Feature activation+1.330
that
Token that
Feature activation+1.112
they
Token they
Feature activation+1.529
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
legal
Token legal
Feature activation+0.000
risks
Token risks
Feature activation+0.000
contractors
Token contractors
Feature activation+0.543
could
Token could
Feature activation+2.324
face
Token face
Feature activation+0.000
from
Token from
Feature activation+0.000
Canada
Token Canada
Feature activation+0.000
,
Token,
Feature activation+0.000
Eck
Token Eck
Feature activation+0.000

INTERVAL 1.963 - 2.290
CONTAINS 0.002%

their
Token their
Feature activation+1.320
fleets
Token fleets
Feature activation+0.121
,
Token,
Feature activation+0.714
which
Token which
Feature activation+0.608
will
Token will
Feature activation+0.766
help
Token help
Feature activation+2.063
them
Token them
Feature activation+1.172
attract
Token attract
Feature activation+0.941
more
Token more
Feature activation+1.046
business
Token business
Feature activation+0.000
."
Token."
Feature activation+0.551
became
Token became
Feature activation+0.000
Chairman
Token Chairman
Feature activation+0.000
,
Token,
Feature activation+1.345
I
Token I
Feature activation+0.420
observed
Token observed
Feature activation+0.908
that
Token that
Feature activation+2.249
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.256
[
Token[
Feature activation+0.188
y
Tokeny
Feature activation+0.000
]
Token]
Feature activation+0.000
of
Token of
Feature activation+1.177
the
Token the
Feature activation+1.087
bank
Token bank
Feature activation+0.461
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+2.223
v
Token v
Feature activation+0.053
aults
Tokenaults
Feature activation+0.900
near
Token near
Feature activation+0.000
Kh
Token Kh
Feature activation+0.000
air
Tokenair
Feature activation+0.000
the
Token the
Feature activation+0.504
manufacturer
Token manufacturer
Feature activation+0.000
's
Token's
Feature activation+0.484
literature
Token literature
Feature activation+0.000
and
Token and
Feature activation+1.051
be
Token be
Feature activation+2.136
careful
Token careful
Feature activation+0.418
to
Token to
Feature activation+0.648
watch
Token watch
Feature activation+0.467
out
Token out
Feature activation+0.000
for
Token for
Feature activation+0.474
invest
Token invest
Feature activation+0.091
more
Token more
Feature activation+0.571
effort
Token effort
Feature activation+1.056
in
Token in
Feature activation+2.981
improving
Token improving
Feature activation+3.258
the
Token the
Feature activation+2.216
safety
Token safety
Feature activation+0.000
of
Token of
Feature activation+0.904
their
Token their
Feature activation+1.320
fleets
Token fleets
Feature activation+0.121
,
Token,
Feature activation+0.714

INTERVAL 1.636 - 1.963
CONTAINS 0.002%

assume
Token assume
Feature activation+0.000
more
Token more
Feature activation+0.000
risk
Token risk
Feature activation+0.000
(
Token (
Feature activation+0.346
since
Tokensince
Feature activation+2.812
they
Token they
Feature activation+1.799
could
Token could
Feature activation+0.447
get
Token get
Feature activation+0.969
caught
Token caught
Feature activation+0.000
trying
Token trying
Feature activation+0.000
to
Token to
Feature activation+0.037
risks
Token risks
Feature activation+0.000
.
Token.
Feature activation+0.250
While
Token While
Feature activation+0.410
the
Token the
Feature activation+0.437
concept
Token concept
Feature activation+0.000
of
Token of
Feature activation+1.879
induction
Token induction
Feature activation+0.000
charg
Token charg
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
isn
Token isn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
the
Token the
Feature activation+0.000
order
Token order
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
day
Token day
Feature activation+0.000
.
Token.
Feature activation+1.663
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.298
F
TokenF
Feature activation+0.234
am
Tokenam
Feature activation+0.000
ilies
Tokenilies
Feature activation+0.702
smartphones
Token smartphones
Feature activation+0.424
with
Token with
Feature activation+0.598
activated
Token activated
Feature activation+0.211
FM
Token FM
Feature activation+0.000
chips
Token chips
Feature activation+0.145
can
Token can
Feature activation+1.711
allow
Token allow
Feature activation+1.496
Americans
Token Americans
Feature activation+0.524
to
Token to
Feature activation+1.002
get
Token get
Feature activation+1.195
vital
Token vital
Feature activation+0.804
possible
Token possible
Feature activation+0.000
lawsuits
Token lawsuits
Feature activation+0.000
should
Token should
Feature activation+0.000
an
Token an
Feature activation+0.000
accident
Token accident
Feature activation+0.552
actually
Token actually
Feature activation+1.932
take
Token take
Feature activation+0.235
place
Token place
Feature activation+0.000
.
Token.
Feature activation+0.271
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.128

INTERVAL 1.308 - 1.636
CONTAINS 0.005%

the
Token the
Feature activation+0.000
tall
Token tall
Feature activation+0.000
p
Token p
Feature activation+0.000
ines
Tokenines
Feature activation+0.000
grow
Token grow
Feature activation+0.000
are
Token are
Feature activation+1.622
more
Token more
Feature activation+0.000
likely
Token likely
Feature activation+0.000
to
Token to
Feature activation+0.000
disappear
Token disappear
Feature activation+0.000
into
Token into
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.015
city
Token city
Feature activation+0.000
attorney
Token attorney
Feature activation+0.000
.
Token.
Feature activation+0.000
Similar
Token Similar
Feature activation+1.440
laws
Token laws
Feature activation+0.000
have
Token have
Feature activation+0.000
applied
Token applied
Feature activation+0.000
in
Token in
Feature activation+0.000
other
Token other
Feature activation+0.000
take
Token take
Feature activation+1.545
other
Token other
Feature activation+0.844
important
Token important
Feature activation+0.521
steps
Token steps
Feature activation+0.391
to
Token to
Feature activation+1.360
ensure
Token ensure
Feature activation+1.371
that
Token that
Feature activation+0.581
your
Token your
Feature activation+0.505
house
Token house
Feature activation+0.710
is
Token is
Feature activation+1.631
safe
Token safe
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.715
a
Token a
Feature activation+0.352
misconception
Token misconception
Feature activation+0.000
of
Token of
Feature activation+1.376
what
Token what
Feature activation+1.225
a
Token a
Feature activation+0.836
collaborative
Token collaborative
Feature activation+0.000
robot
Token robot
Feature activation+0.415
is
Token is
Feature activation+0.070
trained
Token trained
Feature activation+2.547
engineers
Token engineers
Feature activation+0.000
to
Token to
Feature activation+0.249
ensure
Token ensure
Feature activation+1.330
that
Token that
Feature activation+1.112
they
Token they
Feature activation+1.529
work
Token work
Feature activation+0.488
properly
Token properly
Feature activation+0.000
.
Token.
Feature activation+0.044
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.409

INTERVAL 0.981 - 1.308
CONTAINS 0.015%

to
Token to
Feature activation+0.000
children
Token children
Feature activation+0.261
suggesting
Token suggesting
Feature activation+1.108
that
Token that
Feature activation+0.620
they
Token they
Feature activation+1.615
are
Token are
Feature activation+1.178
100
Token 100
Feature activation+0.038
%
Token%
Feature activation+1.734
effective
Token effective
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
mainly
Token mainly
Feature activation+0.000
during
Token during
Feature activation+0.368
these
Token these
Feature activation+0.178
inspections
Token inspections
Feature activation+0.000
to
Token to
Feature activation+0.548
identify
Token identify
Feature activation+1.083
motor
Token motor
Feature activation+0.112
carriers
Token carriers
Feature activation+0.758
that
Token that
Feature activation+0.885
are
Token are
Feature activation+1.299
operating
Token operating
Feature activation+0.742
a
Token a
Feature activation+0.000
week
Token week
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+1.373
sh
Tokensh
Feature activation+0.000
owed
Tokenowed
Feature activation+1.078
similar
Token similar
Feature activation+1.200
patterns
Token patterns
Feature activation+0.000
of
Token of
Feature activation+0.290
higher
Token higher
Feature activation+0.383
mortality
Token mortality
Feature activation+0.000
.
Token.
Feature activation+0.000
I
Token I
Feature activation+0.000
went
Token went
Feature activation+0.000
through
Token through
Feature activation+0.000
all
Token all
Feature activation+0.000
the
Token the
Feature activation+0.717
different
Token different
Feature activation+1.067
force
Token force
Feature activation+0.382
,
Token,
Feature activation+0.384
pressure
Token pressure
Feature activation+0.020
and
Token and
Feature activation+0.000
energy
Token energy
Feature activation+0.365
deadly
Token deadly
Feature activation+0.000
toxic
Token toxic
Feature activation+0.000
fumes
Token fumes
Feature activation+0.000
.
Token.
Feature activation+0.266
A
Token A
Feature activation+0.661
final
Token final
Feature activation+1.284
theory
Token theory
Feature activation+0.000
on
Token on
Feature activation+0.541
the
Token the
Feature activation+0.154
per
Token per
Feature activation+0.000
ils
Tokenils
Feature activation+0.000

INTERVAL 0.654 - 0.981
CONTAINS 0.037%

bank
Token bank
Feature activation+0.461
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+2.223
v
Token v
Feature activation+0.053
aults
Tokenaults
Feature activation+0.900
near
Token near
Feature activation+0.000
Kh
Token Kh
Feature activation+0.000
air
Tokenair
Feature activation+0.000
at
Tokenat
Feature activation+0.000
abad
Tokenabad
Feature activation+0.000
undoubtedly
Token undoubtedly
Feature activation+0.281
,
Token,
Feature activation+0.614
there
Token there
Feature activation+0.605
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.659
more
Token more
Feature activation+0.801
fires
Token fires
Feature activation+0.311
to
Token to
Feature activation+0.152
come
Token come
Feature activation+0.121
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.087
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.185
threat
Token threat
Feature activation+0.000
,
Token,
Feature activation+0.132
and
Token and
Feature activation+0.192
many
Token many
Feature activation+0.712
ground
Token ground
Feature activation+0.000
-
Token-
Feature activation+0.000
based
Tokenbased
Feature activation+0.539
weapons
Token weapons
Feature activation+0.000
would
Token would
Feature activation+0.501
do
Token do
Feature activation+0.346
that
Token that
Feature activation+0.428
,
Token,
Feature activation+0.687
you
Token you
Feature activation+0.525
do
Token do
Feature activation+0.543
that
Token that
Feature activation+0.745
only
Token only
Feature activation+0.430
when
Token when
Feature activation+0.748
you
Token you
Feature activation+0.362
are
Token are
Feature activation+0.623
sure
Token sure
Feature activation+0.373
.
Token.
Feature activation+0.589
It
Token It
Feature activation+0.550
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.556
completely
Token completely
Feature activation+0.816
random
Token random
Feature activation+0.575
when
Token when
Feature activation+0.420
it
Token it
Feature activation+0.327
happens
Token happens
Feature activation+0.334
.
Token.
Feature activation+0.547

INTERVAL 0.327 - 0.654
CONTAINS 0.136%

also
Token also
Feature activation+0.000
admitted
Token admitted
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
agency
Token agency
Feature activation+0.174
did
Token did
Feature activation+0.357
not
Token not
Feature activation+0.501
have
Token have
Feature activation+0.226
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
access
Tokenaccess
Feature activation+0.000
result
Token result
Feature activation+0.011
of
Token of
Feature activation+0.587
a
Token a
Feature activation+0.573
collision
Token collision
Feature activation+0.000
with
Token with
Feature activation+0.276
a
Token a
Feature activation+0.351
bus
Token bus
Feature activation+0.000
carrying
Token carrying
Feature activation+0.000
media
Token media
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
new
Token new
Feature activation+0.000
quick
Token quick
Feature activation+0.000
-
Token-
Feature activation+0.000
lock
Tokenlock
Feature activation+0.000
mechanism
Token mechanism
Feature activation+0.000
that
Token that
Feature activation+0.560
maintains
Token maintains
Feature activation+0.312
the
Token the
Feature activation+0.000
right
Token right
Feature activation+0.000
shape
Token shape
Feature activation+0.000
for
Token for
Feature activation+0.079
also
Token also
Feature activation+0.000
be
Token be
Feature activation+0.000
disturbing
Token disturbing
Feature activation+0.000
,
Token,
Feature activation+0.000
so
Token so
Feature activation+0.257
consider
Token consider
Feature activation+0.341
feeding
Token feeding
Feature activation+0.051
your
Token your
Feature activation+0.000
cat
Token cat
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
of
Token of
Feature activation+0.213
damage
Token damage
Feature activation+0.000
would
Token would
Feature activation+0.118
be
Token be
Feature activation+0.022
even
Token even
Feature activation+0.097
more
Token more
Feature activation+0.651
reassuring
Token reassuring
Feature activation+0.051
.
Token.
Feature activation+0.128
Back
Token Back
Feature activation+0.033
down
Token down
Feature activation+0.000
here
Token here
Feature activation+0.000

INTERVAL 0.000 - 0.327
CONTAINS 99.801%

to
Token to
Feature activation+0.000
explore
Token explore
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
dirt
Token dirt
Feature activation+0.000
road
Token road
Feature activation+0.000
leading
Token leading
Feature activation+0.000
up
Token up
Feature activation+0.000
to
Token to
Feature activation+0.000
my
Token my
Feature activation+0.000
modest
Token modest
Feature activation+0.000
much
Token much
Feature activation+0.000
appreciated
Token appreciated
Feature activation+0.000
.
Token.
Feature activation+0.000
Edit
Token Edit
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Try
TokenTry
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
the
Token the
Feature activation+0.000
categories
Token categories
Feature activation+0.000
fixes
Token fixes
Feature activation+0.000
and
Token and
Feature activation+0.000
optimizations
Token optimizations
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
also
Token also
Feature activation+0.000
sees
Token sees
Feature activation+0.000
the
Token the
Feature activation+0.000
release
Token release
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
try
Token try
Feature activation+0.000
to
Token to
Feature activation+0.000
start
Token start
Feature activation+0.000
thinking
Token thinking
Feature activation+0.000
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
future
Token future
Feature activation+0.000
.
Token.
Feature activation+0.000
I
Token I
Feature activation+0.000
tried
Token tried
Feature activation+0.000
to
Token to
Feature activation+0.000
'
Token '
Feature activation+0.000
m
Tokenm
Feature activation+0.000
ilit
Tokenilit
Feature activation+0.000
ants
Tokenants
Feature activation+0.000
'
Token'
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
city
Token city
Feature activation+0.000
,
Token,
Feature activation+0.000
eagle
Token eagle
Feature activation+0.000
-
Token-
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 6 in H9.2: (feature 6234)

TOP ACTIVATIONS
MAX = 3.892

.
Token.
Feature activation+0.000
The
Token The
Feature activation+2.812
reason
Token reason
Feature activation+0.000
for
Token for
Feature activation+2.781
her
Token her
Feature activation+3.836
sudden
Token sudden
Feature activation+3.892
withdrawal
Token withdrawal
Feature activation+0.000
was
Token was
Feature activation+0.160
a
Token a
Feature activation+1.402
diagnosis
Token diagnosis
Feature activation+0.000
of
Token of
Feature activation+0.155
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+2.812
reason
Token reason
Feature activation+0.000
for
Token for
Feature activation+2.781
her
Token her
Feature activation+3.836
sudden
Token sudden
Feature activation+3.892
withdrawal
Token withdrawal
Feature activation+0.000
was
Token was
Feature activation+0.160
a
Token a
Feature activation+1.402
diagnosis
Token diagnosis
Feature activation+0.000
report
Token report
Feature activation+1.530
that
Token that
Feature activation+1.020
Ch
Token Ch
Feature activation+0.000
ah
Tokenah
Feature activation+0.000
al
Tokenal
Feature activation+3.112
was
Token was
Feature activation+3.358
out
Token out
Feature activation+2.072
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.108
Bill
TokenBill
Feature activation+0.000
.
Token.
Feature activation+0.830
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.515
Some
TokenSome
Feature activation+1.768
of
Token of
Feature activation+1.494
the
Token the
Feature activation+3.327
cuts
Token cuts
Feature activation+0.198
will
Token will
Feature activation+1.039
und
Token und
Feature activation+0.578
oubt
Tokenoubt
Feature activation+0.000
ably
Tokenably
Feature activation+1.264
in
Token in
Feature activation+0.000
2001
Token 2001
Feature activation+0.000
,
Token,
Feature activation+0.285
tend
Token tend
Feature activation+0.000
ered
Tokenered
Feature activation+1.516
his
Token his
Feature activation+3.292
resignation
Token resignation
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
state
Token state
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
to
Token to
Feature activation+1.054
report
Token report
Feature activation+1.530
that
Token that
Feature activation+1.020
Ch
Token Ch
Feature activation+0.000
ah
Tokenah
Feature activation+0.000
al
Tokenal
Feature activation+3.112
was
Token was
Feature activation+3.358
out
Token out
Feature activation+2.072
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.108
Smith
TokenSmith
Feature activation+0.364
clarified
Token clarified
Feature activation+0.000
that
Token that
Feature activation+0.597
while
Token while
Feature activation+0.780
he
Token he
Feature activation+1.526
is
Token is
Feature activation+3.042
closing
Token closing
Feature activation+0.158
the
Token the
Feature activation+0.488
curtain
Token curtain
Feature activation+0.000
on
Token on
Feature activation+0.239
theatrical
Token theatrical
Feature activation+0.000
on
Token on
Feature activation+0.000
Oak
Token Oak
Feature activation+0.000
l
Tokenl
Feature activation+0.000
awn
Tokenawn
Feature activation+0.041
Avenue
Token Avenue
Feature activation+2.721
will
Token will
Feature activation+3.019
shut
Token shut
Feature activation+0.000
its
Token its
Feature activation+0.000
doors
Token doors
Feature activation+0.117
after
Token after
Feature activation+0.613
Nov
Token Nov
Feature activation+0.278
thought
Token thought
Feature activation+0.000
to
Token to
Feature activation+0.075
be
Token be
Feature activation+0.323
be
Token be
Feature activation+0.408
undecided
Token undecided
Feature activation+0.000
about
Token about
Feature activation+3.007
the
Token the
Feature activation+1.892
move
Token move
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
and
Tokenand
Feature activation+0.000
ow
Tokenow
Feature activation+0.179
ki
Tokenki
Feature activation+1.632
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.062
s
Tokens
Feature activation+2.988
firing
Token firing
Feature activation+0.000
.
Token.
Feature activation+0.000
Here
Token Here
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
leaves
Token leaves
Feature activation+0.000
office
Token office
Feature activation+0.000
is
Token is
Feature activation+0.000
this
Token this
Feature activation+0.000
:
Token:
Feature activation+0.788
who
Token who
Feature activation+2.840
will
Token will
Feature activation+2.156
replace
Token replace
Feature activation+0.000
them
Token them
Feature activation+0.000
?
Token?
Feature activation+0.852
Often
Token Often
Feature activation+0.274
.
Token.
Feature activation+0.000
Open
Token Open
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+2.812
reason
Token reason
Feature activation+0.000
for
Token for
Feature activation+2.781
her
Token her
Feature activation+3.836
sudden
Token sudden
Feature activation+3.892
withdrawal
Token withdrawal
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+2.812
reason
Token reason
Feature activation+0.000
for
Token for
Feature activation+2.781
her
Token her
Feature activation+3.836
sudden
Token sudden
Feature activation+3.892
withdrawal
Token withdrawal
Feature activation+0.000
was
Token was
Feature activation+0.160
a
Token a
Feature activation+1.402
or
Token or
Feature activation+0.411
developer
Token developer
Feature activation+0.000
Play
Token Play
Feature activation+0.000
dom
Tokendom
Feature activation+0.719
on
Token on
Feature activation+2.279
the
Token the
Feature activation+2.773
matter
Token matter
Feature activation+0.261
,
Token,
Feature activation+1.300
players
Token players
Feature activation+0.252
are
Token are
Feature activation+0.856
now
Token now
Feature activation+0.855
been
Token been
Feature activation+0.000
fired
Token fired
Feature activation+0.000
.
Token.
Feature activation+0.265
He
Token He
Feature activation+0.913
appealed
Token appealed
Feature activation+0.644
the
Token the
Feature activation+2.727
dismissal
Token dismissal
Feature activation+0.000
to
Token to
Feature activation+0.072
the
Token the
Feature activation+0.000
Board
Token Board
Feature activation+0.000
of
Token of
Feature activation+0.000
Club
Token Club
Feature activation+0.423
on
Token on
Feature activation+0.000
Oak
Token Oak
Feature activation+0.000
l
Tokenl
Feature activation+0.000
awn
Tokenawn
Feature activation+0.041
Avenue
Token Avenue
Feature activation+2.721
will
Token will
Feature activation+3.019
shut
Token shut
Feature activation+0.000
its
Token its
Feature activation+0.000
doors
Token doors
Feature activation+0.117
after
Token after
Feature activation+0.613
gave
Token gave
Feature activation+0.442
no
Token no
Feature activation+1.285
indication
Token indication
Feature activation+0.048
that
Token that
Feature activation+0.711
he
Token he
Feature activation+1.398
would
Token would
Feature activation+2.701
consider
Token consider
Feature activation+2.527
the
Token the
Feature activation+0.803
next
Token next
Feature activation+0.046
necessary
Token necessary
Feature activation+0.585
step
Token step
Feature activation+0.000
the
Token the
Feature activation+1.323
nationality
Token nationality
Feature activation+0.443
of
Token of
Feature activation+1.512
the
Token the
Feature activation+2.298
people
Token people
Feature activation+2.322
that
Token that
Feature activation+2.684
left
Token left
Feature activation+0.207
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
141
Token141
Feature activation+0.062
G
Token G
Feature activation+0.000
att
Tokenatt
Feature activation+0.000
is
Tokenis
Feature activation+0.000
said
Token said
Feature activation+0.690
that
Token that
Feature activation+1.464
his
Token his
Feature activation+2.668
decision
Token decision
Feature activation+0.712
came
Token came
Feature activation+0.000
down
Token down
Feature activation+0.000
to
Token to
Feature activation+0.201
two
Token two
Feature activation+0.000
pressure
Token pressure
Feature activation+0.000
and
Token and
Feature activation+0.000
resigned
Token resigned
Feature activation+0.000
before
Token before
Feature activation+0.035
Christmas
Token Christmas
Feature activation+0.000
,
Token,
Feature activation+2.559
but
Token but
Feature activation+0.000
the
Token the
Feature activation+0.262
capit
Token capit
Feature activation+0.000
ulation
Tokenulation
Feature activation+0.008
sparked
Token sparked
Feature activation+0.000

Top DFA by src position
MAX = 3.748

player
Token player
Feature activation+0.002
Top resid features:
Venus
Token Venus
Feature activation-0.000
Top resid features:
Williams
Token Williams
Feature activation+0.002
Top resid features:
abruptly
Token abruptly
Feature activation+0.286
Top resid features:
dropped
Token dropped
Feature activation+0.505
Top resid features:
out
Token out
Feature activation+2.596
Top resid features:
of
Token of
Feature activation+0.653
Top resid features:
the
Token the
Feature activation+0.017
Top resid features:
U
Token U
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
S
TokenS
Feature activation+0.011
Top resid features:
player
Token player
Feature activation+0.000
Top resid features:
Venus
Token Venus
Feature activation-0.000
Top resid features:
Williams
Token Williams
Feature activation+0.003
Top resid features:
abruptly
Token abruptly
Feature activation+0.616
Top resid features:
dropped
Token dropped
Feature activation+0.484
Top resid features:
out
Token out
Feature activation+2.367
Top resid features:
of
Token of
Feature activation+0.509
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
U
Token U
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
S
TokenS
Feature activation+0.009
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
Rad
Token Rad
Feature activation+0.003
Top resid features:
ium
Tokenium
Feature activation+0.000
Top resid features:
One
TokenOne
Feature activation+0.001
Top resid features:
has
Token has
Feature activation+0.004
Top resid features:
fired
Token fired
Feature activation+3.061
Top resid features:
its
Token its
Feature activation+0.068
Top resid features:
CEO
Token CEO
Feature activation+0.139
Top resid features:
G
Token G
Feature activation+0.000
Top resid features:
urb
Tokenurb
Feature activation-0.001
Top resid features:
ak
Tokenak
Feature activation+0.001
Top resid features:
AOL
Token AOL
Feature activation-0.021
Top resid features:
is
Token is
Feature activation+0.003
Top resid features:
planning
Token planning
Feature activation+0.003
Top resid features:
on
Token on
Feature activation+0.014
Top resid features:
laying
Token laying
Feature activation+0.011
Top resid features:
off
Token off
Feature activation+3.336
Top resid features:
hundreds
Token hundreds
Feature activation+0.121
Top resid features:
of
Token of
Feature activation+0.055
Top resid features:
employees
Token employees
Feature activation+0.132
Top resid features:
next
Token next
Feature activation+0.003
Top resid features:
week
Token week
Feature activation+0.025
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.095
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
step
Token step
Feature activation+0.004
Top resid features:
down
Token down
Feature activation+3.748
Top resid features:
after
Token after
Feature activation+0.055
Top resid features:
being
Token being
Feature activation-0.004
Top resid features:
accused
Token accused
Feature activation-0.006
Top resid features:
of
Token of
Feature activation-0.003
Top resid features:
using
Token using
Feature activation+0.002
Top resid features:
of
Token of
Feature activation+0.002
Top resid features:
Rad
Token Rad
Feature activation+0.001
Top resid features:
ium
Tokenium
Feature activation+0.000
Top resid features:
One
TokenOne
Feature activation+0.001
Top resid features:
has
Token has
Feature activation+0.004
Top resid features:
fired
Token fired
Feature activation+2.529
Top resid features:
its
Token its
Feature activation+0.028
Top resid features:
CEO
Token CEO
Feature activation+0.243
Top resid features:
G
Token G
Feature activation+0.000
Top resid features:
urb
Tokenurb
Feature activation-0.000
Top resid features:
ak
Tokenak
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
he
Token he
Feature activation+0.009
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+0.001
Top resid features:
leaving
Token leaving
Feature activation+2.740
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
Ŀ
TokenĿ
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Rhode
Token Rhode
Feature activation+0.000
Top resid features:
Island
Token Island
Feature activation+0.000
Top resid features:
night
Token night
Feature activation+0.000
Top resid features:
life
Tokenlife
Feature activation+0.009
Top resid features:
is
Token is
Feature activation+0.008
Top resid features:
closing
Token closing
Feature activation+3.124
Top resid features:
.
Token.
Feature activation+0.152
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
M
TokenM
Feature activation+0.000
Top resid features:
ardi
Tokenardi
Feature activation-0.000
Top resid features:
Ran
Token Ran
Feature activation-0.000
Top resid features:
oc
Tokenoc
Feature activation+0.001
Top resid features:
ch
Tokench
Feature activation-0.000
Top resid features:
ia
Tokenia
Feature activation+0.003
Top resid features:
to
Token to
Feature activation+0.007
Top resid features:
leave
Token leave
Feature activation+2.808
Top resid features:
but
Token but
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
centre
Token centre
Feature activation-0.003
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
half
Tokenhalf
Feature activation-0.018
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.073
Top resid features:
s
Tokens
Feature activation+0.002
Top resid features:
abrupt
Token abrupt
Feature activation+0.081
Top resid features:
departure
Token departure
Feature activation+3.170
Top resid features:
caught
Token caught
Feature activation+0.002
Top resid features:
many
Token many
Feature activation+0.001
Top resid features:
Trump
Token Trump
Feature activation-0.011
Top resid features:
staffers
Token staffers
Feature activation+0.003
Top resid features:
by
Token by
Feature activation-0.003
Top resid features:
ask
Token ask
Feature activation-0.007
Top resid features:
if
Token if
Feature activation+0.001
Top resid features:
or
Token or
Feature activation-0.003
Top resid features:
when
Token when
Feature activation+0.004
Top resid features:
someone
Token someone
Feature activation-0.019
Top resid features:
leaves
Token leaves
Feature activation+3.162
Top resid features:
office
Token office
Feature activation+0.111
Top resid features:
is
Token is
Feature activation-0.005
Top resid features:
this
Token this
Feature activation+0.002
Top resid features:
:
Token:
Feature activation+0.046
Top resid features:
who
Token who
Feature activation-0.002
Top resid features:
when
Token when
Feature activation+0.004
Top resid features:
top
Token top
Feature activation-0.010
Top resid features:
player
Token player
Feature activation+0.033
Top resid features:
Venus
Token Venus
Feature activation-0.010
Top resid features:
Williams
Token Williams
Feature activation+0.026
Top resid features:
abruptly
Token abruptly
Feature activation+1.054
Top resid features:
dropped
Token dropped
Feature activation+0.585
Top resid features:
out
Token out
Feature activation+0.973
Top resid features:
of
Token of
Feature activation+0.339
Top resid features:
the
Token the
Feature activation+0.011
Top resid features:
U
Token U
Feature activation+0.004
Top resid features:
player
Token player
Feature activation+0.001
Top resid features:
Venus
Token Venus
Feature activation-0.015
Top resid features:
Williams
Token Williams
Feature activation+0.006
Top resid features:
abruptly
Token abruptly
Feature activation+0.611
Top resid features:
dropped
Token dropped
Feature activation+0.687
Top resid features:
out
Token out
Feature activation+1.219
Top resid features:
of
Token of
Feature activation+0.276
Top resid features:
the
Token the
Feature activation+0.014
Top resid features:
U
Token U
Feature activation+0.002
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
S
TokenS
Feature activation+0.006
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
2014
Token 2014
Feature activation+0.001
Top resid features:
will
Token will
Feature activation+0.082
Top resid features:
be
Token be
Feature activation+0.026
Top resid features:
shut
Token shut
Feature activation+0.540
Top resid features:
down
Token down
Feature activation+1.895
Top resid features:
on
Token on
Feature activation+0.048
Top resid features:
22
Token 22
Feature activation+0.012
Top resid features:
nd
Tokennd
Feature activation-0.000
Top resid features:
October
Token October
Feature activation+0.010
Top resid features:
2014
Token 2014
Feature activation+0.028
Top resid features:
notice
Token notice
Feature activation-0.071
Top resid features:
that
Token that
Feature activation+0.001
Top resid features:
he
Token he
Feature activation+0.011
Top resid features:
had
Token had
Feature activation+0.005
Top resid features:
been
Token been
Feature activation+0.000
Top resid features:
fired
Token fired
Feature activation+2.925
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
He
Token He
Feature activation+0.002
Top resid features:
appealed
Token appealed
Feature activation-0.033
Top resid features:
the
Token the
Feature activation-0.041
Top resid features:
dismissal
Token dismissal
Feature activation+0.000
Top resid features:
Rhode
Token Rhode
Feature activation+0.005
Top resid features:
Island
Token Island
Feature activation+0.007
Top resid features:
night
Token night
Feature activation-0.003
Top resid features:
life
Tokenlife
Feature activation+0.002
Top resid features:
is
Token is
Feature activation+0.010
Top resid features:
closing
Token closing
Feature activation+2.345
Top resid features:
.
Token.
Feature activation+0.763
Top resid features:
Ċ
TokenĊ
Feature activation+0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.031
Top resid features:
M
TokenM
Feature activation+0.002
Top resid features:
ardi
Tokenardi
Feature activation+0.000
Top resid features:
those
Token those
Feature activation+0.000
Top resid features:
lies
Token lies
Feature activation-0.003
Top resid features:
and
Token and
Feature activation+0.001
Top resid features:
that
Token that
Feature activation+0.003
Top resid features:
he
Token he
Feature activation+0.010
Top resid features:
resign
Token resign
Feature activation+1.599
Top resid features:
as
Token as
Feature activation+0.077
Top resid features:
the
Token the
Feature activation+0.031
Top resid features:
nation
Token nation
Feature activation+0.000
Top resid features:
âĢ
TokenâĢ
Feature activation+0.003
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
that
Token that
Feature activation+0.001
Top resid features:
320
Token 320
Feature activation-0.004
Top resid features:
,
Token,
Feature activation-0.000
Top resid features:
000
Token000
Feature activation-0.002
Top resid features:
people
Token people
Feature activation-0.003
Top resid features:
left
Token left
Feature activation+2.504
Top resid features:
the
Token the
Feature activation+0.023
Top resid features:
UK
Token UK
Feature activation+0.055
Top resid features:
in
Token in
Feature activation-0.002
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
year
Token year
Feature activation-0.004
Top resid features:
.
Token.
Feature activation+0.008
Top resid features:
So
Token So
Feature activation+0.000
Top resid features:
why
Token why
Feature activation+0.009
Top resid features:
would
Token would
Feature activation-0.000
Top resid features:
he
Token he
Feature activation+0.022
Top resid features:
leave
Token leave
Feature activation+2.297
Top resid features:
the
Token the
Feature activation+0.060
Top resid features:
company
Token company
Feature activation+0.337
Top resid features:
right
Token right
Feature activation+0.008
Top resid features:
before
Token before
Feature activation+0.013
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
i
Tokeni
Feature activation+0.004
Top resid features:
bowed
Token bowed
Feature activation+0.055
Top resid features:
to
Token to
Feature activation+0.016
Top resid features:
pressure
Token pressure
Feature activation-0.003
Top resid features:
and
Token and
Feature activation+0.013
Top resid features:
resigned
Token resigned
Feature activation+2.709
Top resid features:
before
Token before
Feature activation+0.022
Top resid features:
Christmas
Token Christmas
Feature activation+0.035
Top resid features:
,
Token,
Feature activation+0.099
Top resid features:
but
Token but
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.09

Head 1: 0.02

Head 2: 0.47

Head 3: 0.08

Head 4: 0.03

Head 5: 0.08

Head 6: 0.04

Head 7: 0.06

Head 8: 0.03

Head 9: 0.03

Head 10: 0.05

Head 11: 0.01

Positive logits

disband4.97

departure4.88

withdrawal4.88

quitting4.79

cancellation4.79

disbanded4.72

exodus4.72

quit4.59

farewell4.55

resignation4.54

departures4.53

termination4.48

resign4.48

layoffs4.45

discontin4.43

reinstated4.34

disapp4.33

goodbye4.27

exit4.26

exit4.26

Negative logits

innate-2.38

proficiency-2.37

amplification-2.37

Expert-2.37

benchmarks-2.36

Powerful-2.32

measurement-2.27

doi-2.26

Precision-2.25

measuring-2.22

Parables-2.19

crit-2.18

Medium-2.17

inea-2.16

measurements-2.15

Prec-2.14

Measure-2.13

Amp-2.13

Reference-2.13

Technique-2.11

INTERVAL 3.503 - 3.892
CONTAINS 0.000%

week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+2.812
reason
Token reason
Feature activation+0.000
for
Token for
Feature activation+2.781
her
Token her
Feature activation+3.836
sudden
Token sudden
Feature activation+3.892
withdrawal
Token withdrawal
Feature activation+0.000
was
Token was
Feature activation+0.160
a
Token a
Feature activation+1.402
diagnosis
Token diagnosis
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+2.812
reason
Token reason
Feature activation+0.000
for
Token for
Feature activation+2.781
her
Token her
Feature activation+3.836
sudden
Token sudden
Feature activation+3.892
withdrawal
Token withdrawal
Feature activation+0.000
was
Token was
Feature activation+0.160
a
Token a
Feature activation+1.402
diagnosis
Token diagnosis
Feature activation+0.000
of
Token of
Feature activation+0.155

INTERVAL 3.114 - 3.503
CONTAINS 0.000%

report
Token report
Feature activation+1.530
that
Token that
Feature activation+1.020
Ch
Token Ch
Feature activation+0.000
ah
Tokenah
Feature activation+0.000
al
Tokenal
Feature activation+3.112
was
Token was
Feature activation+3.358
out
Token out
Feature activation+2.072
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.108
Bill
TokenBill
Feature activation+0.000
.
Token.
Feature activation+0.830
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.515
Some
TokenSome
Feature activation+1.768
of
Token of
Feature activation+1.494
the
Token the
Feature activation+3.327
cuts
Token cuts
Feature activation+0.198
will
Token will
Feature activation+1.039
und
Token und
Feature activation+0.578
oubt
Tokenoubt
Feature activation+0.000
ably
Tokenably
Feature activation+1.264
in
Token in
Feature activation+0.000
2001
Token 2001
Feature activation+0.000
,
Token,
Feature activation+0.285
tend
Token tend
Feature activation+0.000
ered
Tokenered
Feature activation+1.516
his
Token his
Feature activation+3.292
resignation
Token resignation
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
state
Token state
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000

INTERVAL 2.725 - 3.114
CONTAINS 0.001%

thought
Token thought
Feature activation+0.000
to
Token to
Feature activation+0.075
be
Token be
Feature activation+0.323
be
Token be
Feature activation+0.408
undecided
Token undecided
Feature activation+0.000
about
Token about
Feature activation+3.007
the
Token the
Feature activation+1.892
move
Token move
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
fired
Token fired
Feature activation+0.000
.
Token.
Feature activation+0.265
He
Token He
Feature activation+0.913
appealed
Token appealed
Feature activation+0.644
the
Token the
Feature activation+2.727
dismissal
Token dismissal
Feature activation+0.000
to
Token to
Feature activation+0.072
the
Token the
Feature activation+0.000
Board
Token Board
Feature activation+0.000
of
Token of
Feature activation+0.000
to
Token to
Feature activation+1.054
report
Token report
Feature activation+1.530
that
Token that
Feature activation+1.020
Ch
Token Ch
Feature activation+0.000
ah
Tokenah
Feature activation+0.000
al
Tokenal
Feature activation+3.112
was
Token was
Feature activation+3.358
out
Token out
Feature activation+2.072
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.108
.
Token.
Feature activation+0.000
Open
Token Open
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+2.812
reason
Token reason
Feature activation+0.000
for
Token for
Feature activation+2.781
her
Token her
Feature activation+3.836
sudden
Token sudden
Feature activation+3.892
withdrawal
Token withdrawal
Feature activation+0.000
or
Token or
Feature activation+0.411
developer
Token developer
Feature activation+0.000
Play
Token Play
Feature activation+0.000
dom
Tokendom
Feature activation+0.719
on
Token on
Feature activation+2.279
the
Token the
Feature activation+2.773
matter
Token matter
Feature activation+0.261
,
Token,
Feature activation+1.300
players
Token players
Feature activation+0.252
are
Token are
Feature activation+0.856
now
Token now
Feature activation+0.855

INTERVAL 2.335 - 2.725
CONTAINS 0.001%

G
Token G
Feature activation+0.000
att
Tokenatt
Feature activation+0.000
is
Tokenis
Feature activation+0.000
said
Token said
Feature activation+0.690
that
Token that
Feature activation+1.464
his
Token his
Feature activation+2.668
decision
Token decision
Feature activation+0.712
came
Token came
Feature activation+0.000
down
Token down
Feature activation+0.000
to
Token to
Feature activation+0.201
two
Token two
Feature activation+0.000
couldn
Token couldn
Feature activation+0.000
't
Token't
Feature activation+0.704
disclose
Token disclose
Feature activation+0.893
his
Token his
Feature activation+1.509
reason
Token reason
Feature activation+0.015
for
Token for
Feature activation+2.491
resignation
Token resignation
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Re
TokenRe
Feature activation+0.209
You
TokenYou
Feature activation+0.000
are
Token are
Feature activation+0.000
logged
Token logged
Feature activation+0.000
out
Token out
Feature activation+0.000
.
Token.
Feature activation+0.000
Login
Token Login
Feature activation+2.354
|
Token |
Feature activation+1.912
Sign
Token Sign
Feature activation+0.500
up
Token up
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
s
Tokens
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Such
TokenSuch
Feature activation+1.589
a
Token a
Feature activation+2.479
move
Token move
Feature activation+0.777
would
Token would
Feature activation+1.083
be
Token be
Feature activation+0.487
a
Token a
Feature activation+1.117
worrying
Token worrying
Feature activation+0.346
council
Token council
Feature activation+0.530
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.814
vote
Token vote
Feature activation+0.454
to
Token to
Feature activation+2.464
cut
Token cut
Feature activation+0.000
his
Token his
Feature activation+0.064
assistant
Token assistant
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 1.946 - 2.335
CONTAINS 0.003%

but
Token but
Feature activation+0.333
it
Token it
Feature activation+0.388
,
Token,
Feature activation+0.000
too
Token too
Feature activation+0.000
,
Token,
Feature activation+1.235
is
Token is
Feature activation+1.971
closing
Token closing
Feature activation+0.000
235
Token 235
Feature activation+0.000
under
Token under
Feature activation+0.302
-
Token-
Feature activation+0.163
performing
Tokenperforming
Feature activation+0.000
,
Token,
Feature activation+0.000
resigned
Token resigned
Feature activation+0.000
after
Token after
Feature activation+0.008
two
Token two
Feature activation+0.000
games
Token games
Feature activation+0.000
,
Token,
Feature activation+2.084
and
Token and
Feature activation+0.000
his
Token his
Feature activation+0.000
replacement
Token replacement
Feature activation+0.000
,
Token,
Feature activation+0.000
Tom
Token Tom
Feature activation+0.000
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
revealed
Token revealed
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
temporary
Token temporary
Feature activation+1.998
venue
Token venue
Feature activation+0.000
being
Token being
Feature activation+0.284
lined
Token lined
Feature activation+0.000
cannot
Token cannot
Feature activation+0.000
be
Token be
Feature activation+0.060
Francisco
Token Francisco
Feature activation+0.000
49
Token 49
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
,
Token,
Feature activation+1.606
was
Token was
Feature activation+1.445
given
Token given
Feature activation+2.185
a
Token a
Feature activation+1.971
contract
Token contract
Feature activation+0.000
extension
Token extension
Feature activation+0.000
Sunday
Token Sunday
Feature activation+0.000
.)
Token.)
Feature activation+0.000
broadcast
Token broadcast
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
concert
Token concert
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+2.098
cancellation
Token cancellation
Feature activation+0.255
is
Token is
Feature activation+1.012
possibly
Token possibly
Feature activation+0.626
due
Token due
Feature activation+0.000
to
Token to
Feature activation+0.962

INTERVAL 1.557 - 1.946
CONTAINS 0.007%

2015
Token 2015
Feature activation+0.000
.
Token.
Feature activation+0.012
CAP
Token CAP
Feature activation+0.000
P
TokenP
Feature activation+0.591
said
Token said
Feature activation+0.876
those
Token those
Feature activation+1.782
cuts
Token cuts
Feature activation+0.292
include
Token include
Feature activation+1.197
25
Token 25
Feature activation+0.519
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.326
.
Token.
Feature activation+0.000
However
Token However
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+1.439
only
Token only
Feature activation+0.736
businesses
Token businesses
Feature activation+1.829
affected
Token affected
Feature activation+0.092
are
Token are
Feature activation+0.000
those
Token those
Feature activation+0.000
relating
Token relating
Feature activation+0.000
to
Token to
Feature activation+0.000
that
Token that
Feature activation+0.734
he
Token he
Feature activation+0.559
might
Token might
Feature activation+0.694
speed
Token speed
Feature activation+1.200
up
Token up
Feature activation+1.397
the
Token the
Feature activation+1.870
process
Token process
Feature activation+0.027
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.038
US
TokenUS
Feature activation+0.227
next
Token next
Feature activation+0.861
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.830
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.515
Some
TokenSome
Feature activation+1.768
of
Token of
Feature activation+1.494
the
Token the
Feature activation+3.327
cuts
Token cuts
Feature activation+0.198
will
Token will
Feature activation+1.039
und
Token und
Feature activation+0.578
thousands
Token thousands
Feature activation+1.244
of
Token of
Feature activation+0.570
other
Token other
Feature activation+1.580
people
Token people
Feature activation+1.574
who
Token who
Feature activation+1.288
have
Token have
Feature activation+1.604
lost
Token lost
Feature activation+0.039
their
Token their
Feature activation+0.000
jobs
Token jobs
Feature activation+0.000
as
Token as
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 1.168 - 1.557
CONTAINS 0.023%

About
Token About
Feature activation+0.256
300
Token 300
Feature activation+0.470
Hert
Token Hert
Feature activation+0.000
z
Tokenz
Feature activation+0.233
IT
Token IT
Feature activation+0.000
employees
Token employees
Feature activation+1.447
,
Token,
Feature activation+0.034
most
Token most
Feature activation+0.005
located
Token located
Feature activation+0.000
in
Token in
Feature activation+0.000
Oklahoma
Token Oklahoma
Feature activation+0.000
to
Token to
Feature activation+0.000
June
Token June
Feature activation+0.000
2012
Token 2012
Feature activation+0.000
there
Token there
Feature activation+0.067
were
Token were
Feature activation+0.016
282
Token 282
Feature activation+1.217
,
Token,
Feature activation+0.000
833
Token833
Feature activation+1.176
student
Token student
Feature activation+0.267
visas
Token visas
Feature activation+0.000
issued
Token issued
Feature activation+0.000
in
Token in
Feature activation+0.000
my
Token my
Feature activation+0.000
position
Token position
Feature activation+0.000
,
Token,
Feature activation+0.258
I
Token I
Feature activation+1.105
will
Token will
Feature activation+1.297
step
Token step
Feature activation+0.674
down
Token down
Feature activation+0.000
.
Token.
Feature activation+0.127
And
Token And
Feature activation+0.323
I
Token I
Feature activation+0.330
.
Token.
Feature activation+0.183
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.363
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+1.276
Fifa
Token Fifa
Feature activation+0.307
general
Token general
Feature activation+0.000
secretary
Token secretary
Feature activation+0.000
,
Token,
Feature activation+0.000
J
Token J
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.189
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.032
C
TokenC
Feature activation+0.000
ox
Tokenox
Feature activation+1.313
,
Token,
Feature activation+0.000
31
Token 31
Feature activation+0.000
,
Token,
Feature activation+0.540
was
Token was
Feature activation+0.568
also
Token also
Feature activation+0.338

INTERVAL 0.778 - 1.168
CONTAINS 0.057%

accordingly
Token accordingly
Feature activation+0.000
:
Token:
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Regarding
TokenRegarding
Feature activation+0.804
the
Token the
Feature activation+0.833
USA
Token USA
Feature activation+0.139
-
Token-
Feature activation+0.000
Mexico
TokenMexico
Feature activation+0.000
field
Token field
Feature activation+0.000
conditions
Token conditions
Feature activation+0.000
ident
Tokenident
Feature activation+0.000
Evil
Token Evil
Feature activation+0.000
4
Token 4
Feature activation+0.000
."
Token."
Feature activation+0.000
I
Token I
Feature activation+0.000
say
Token say
Feature activation+0.962
this
Token this
Feature activation+0.000
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
nice
Token nice
Feature activation+0.004
feature
Token feature
Feature activation+0.000
suspended
Token suspended
Feature activation+0.000
that
Token that
Feature activation+0.000
program
Token program
Feature activation+0.000
this
Token this
Feature activation+0.000
month
Token month
Feature activation+0.000
,
Token,
Feature activation+1.033
giving
Token giving
Feature activation+0.000
Congress
Token Congress
Feature activation+0.166
six
Token six
Feature activation+0.000
months
Token months
Feature activation+0.068
to
Token to
Feature activation+0.468
of
Token of
Feature activation+0.000
Silver
Token Silver
Feature activation+0.000
King
Token King
Feature activation+0.000
,
Token,
Feature activation+0.000
in
Token in
Feature activation+0.459
a
Token a
Feature activation+0.983
deal
Token deal
Feature activation+0.172
backed
Token backed
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
company
Token company
Feature activation+0.000
-
Token-
Feature activation+0.000
to
Tokento
Feature activation+0.000
-
Token-
Feature activation+0.000
cook
Tokencook
Feature activation+0.000
meal
Token meal
Feature activation+0.054
would
Token would
Feature activation+0.899
be
Token be
Feature activation+1.178
back
Token back
Feature activation+0.000
soon
Token soon
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 0.389 - 0.778
CONTAINS 0.197%

âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
say
Token say
Feature activation+0.046
I
Token I
Feature activation+0.525
was
Token was
Feature activation+0.503
looking
Token looking
Feature activation+0.228
forward
Token forward
Feature activation+0.000
to
Token to
Feature activation+0.000
fighting
Token fighting
Feature activation+0.000
in
Token in
Feature activation+0.000
weigh
Token weigh
Feature activation+0.000
the
Token the
Feature activation+0.053
cost
Token cost
Feature activation+0.000
of
Token of
Feature activation+0.043
each
Token each
Feature activation+0.369
individual
Token individual
Feature activation+0.404
ride
Token ride
Feature activation+0.000
may
Token may
Feature activation+0.000
encourage
Token encourage
Feature activation+0.425
people
Token people
Feature activation+0.380
to
Token to
Feature activation+0.202
a
Token a
Feature activation+0.057
student
Token student
Feature activation+0.000
mob
Token mob
Feature activation+0.000
that
Token that
Feature activation+0.357
viewed
Token viewed
Feature activation+0.015
his
Token his
Feature activation+0.543
remarks
Token remarks
Feature activation+0.000
as
Token as
Feature activation+0.000
racist
Token racist
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
he
Token he
Feature activation+0.000
said
Token said
Feature activation+0.000
"
Token "
Feature activation+0.000
that
Tokenthat
Feature activation+0.290
Paris
Token Paris
Feature activation+0.065
was
Token was
Feature activation+0.697
liberated
Token liberated
Feature activation+0.000
in
Token in
Feature activation+0.000
August
Token August
Feature activation+0.000
.
Token.
Feature activation+0.000
I
Token I
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Dev
TokenDev
Feature activation+0.000
al
Tokenal
Feature activation+0.000
uation
Tokenuation
Feature activation+0.000
and
Token and
Feature activation+0.556
In
Token In
Feature activation+0.084
flation
Tokenflation
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Although
TokenAlthough
Feature activation+0.000

INTERVAL 0.000 - 0.389
CONTAINS 99.711%

since
Token since
Feature activation+0.000
then
Token then
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
DA
TokenDA
Feature activation+0.000
V
TokenV
Feature activation+0.000
IES
TokenIES
Feature activation+0.000
:
Token:
Feature activation+0.000
(
Token (
Feature activation+0.000
L
TokenL
Feature activation+0.000
familiar
Token familiar
Feature activation+0.000
with
Token with
Feature activation+0.000
Twitter
Token Twitter
Feature activation+0.000
's
Token's
Feature activation+0.000
testimony
Token testimony
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
"
Token "
Feature activation+0.000
Our
TokenOur
Feature activation+0.000
review
Token review
Feature activation+0.000
was
Token was
Feature activation+0.000
B
Token B
Feature activation+0.000
ex
Tokenex
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
County
Token County
Feature activation+0.000
District
Token District
Feature activation+0.000
Attorney
Token Attorney
Feature activation+0.000
-
Token-
Feature activation+0.000
elect
Tokenelect
Feature activation+0.000
Nico
Token Nico
Feature activation+0.000
La
Token La
Feature activation+0.000
H
TokenH
Feature activation+0.000
conversation
Token conversation
Feature activation+0.000
about
Token about
Feature activation+0.000
Uruguay
Token Uruguay
Feature activation+0.000
's
Token's
Feature activation+0.000
marine
Token marine
Feature activation+0.000
ecosystem
Token ecosystem
Feature activation+0.000
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
can
Token can
Feature activation+0.000
follow
Token follow
Feature activation+0.000
the
Token the
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
city
Token city
Feature activation+0.000
of
Token of
Feature activation+0.000
Victoria
Token Victoria
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
G
TokenG
Feature activation+0.000
roups
Tokenroups
Feature activation+0.000
ranging
Token ranging
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 H9.2: (feature 21074)

TOP ACTIVATIONS
MAX = 2.428

Australians
Token Australians
Feature activation+0.317
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.076
People
TokenPeople
Feature activation+2.428
do
Token do
Feature activation+0.000
need
Token need
Feature activation+0.000
to
Token to
Feature activation+0.143
realise
Token realise
Feature activation+0.060
that
Token that
Feature activation+1.082
the
Token the
Feature activation+0.606
only
Token only
Feature activation+1.326
factors
Token factors
Feature activation+0.994
in
Token in
Feature activation+1.713
this
Token this
Feature activation+1.976
chemical
Token chemical
Feature activation+2.186
aging
Token aging
Feature activation+0.637
process
Token process
Feature activation+0.055
.
Token.
Feature activation+1.097
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.754
Ċ
TokenĊ
Feature activation+0.754
Device
TokenDevice
Feature activation+1.283
use
Token use
Feature activation+0.869
also
Token also
Feature activation+1.523
affects
Token affects
Feature activation+1.617
the
Token the
Feature activation+2.072
performance
Token performance
Feature activation+0.177
of
Token of
Feature activation+0.452
a
Token a
Feature activation+0.406
battery
Token battery
Feature activation+0.071
over
Token over
Feature activation+0.233
that
Token that
Feature activation+0.000
he
Token he
Feature activation+0.000
would
Token would
Feature activation+0.293
like
Token like
Feature activation+0.718
to
Token to
Feature activation+1.494
begin
Token begin
Feature activation+2.045
digging
Token digging
Feature activation+0.000
into
Token into
Feature activation+0.035
the
Token the
Feature activation+0.645
administration
Token administration
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
mandatory
Token mandatory
Feature activation+0.000
retirement
Token retirement
Feature activation+0.395
age
Token age
Feature activation+0.733
of
Token of
Feature activation+1.998
50
Token 50
Feature activation+1.889
for
Token for
Feature activation+1.775
state
Token state
Feature activation+1.175
police
Token police
Feature activation+0.193
officers
Token officers
Feature activation+0.297
not
Token not
Feature activation+0.753
the
Token the
Feature activation+0.606
only
Token only
Feature activation+1.326
factors
Token factors
Feature activation+0.994
in
Token in
Feature activation+1.713
this
Token this
Feature activation+1.976
chemical
Token chemical
Feature activation+2.186
aging
Token aging
Feature activation+0.637
process
Token process
Feature activation+0.055
.
Token.
Feature activation+1.097
Ċ
TokenĊ
Feature activation+0.000
s
Tokens
Feature activation+0.000
mandatory
Token mandatory
Feature activation+0.000
retirement
Token retirement
Feature activation+0.395
age
Token age
Feature activation+0.733
of
Token of
Feature activation+1.998
50
Token 50
Feature activation+1.889
for
Token for
Feature activation+1.775
state
Token state
Feature activation+1.175
police
Token police
Feature activation+0.193
officers
Token officers
Feature activation+0.297
.
Token.
Feature activation+0.000
.
Token.
Feature activation+0.208
These
Token These
Feature activation+0.993
are
Token are
Feature activation+0.334
characteristics
Token characteristics
Feature activation+0.297
of
Token of
Feature activation+1.479
battery
Token battery
Feature activation+1.777
chemistry
Token chemistry
Feature activation+0.217
,
Token,
Feature activation+0.247
common
Token common
Feature activation+0.269
to
Token to
Feature activation+0.344
lithium
Token lithium
Feature activation+0.000
mandatory
Token mandatory
Feature activation+0.000
retirement
Token retirement
Feature activation+0.395
age
Token age
Feature activation+0.733
of
Token of
Feature activation+1.998
50
Token 50
Feature activation+1.889
for
Token for
Feature activation+1.775
state
Token state
Feature activation+1.175
police
Token police
Feature activation+0.193
officers
Token officers
Feature activation+0.297
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
budget
Token budget
Feature activation+0.000
.
Token.
Feature activation+0.236
That
Token That
Feature activation+1.274
is
Token is
Feature activation+0.191
way
Token way
Feature activation+0.343
too
Token too
Feature activation+1.755
short
Token short
Feature activation+0.000
sight
Token sight
Feature activation+0.000
ed
Tokened
Feature activation+0.000
.
Token.
Feature activation+0.155
Any
Token Any
Feature activation+0.176
are
Token are
Feature activation+0.557
not
Token not
Feature activation+0.753
the
Token the
Feature activation+0.606
only
Token only
Feature activation+1.326
factors
Token factors
Feature activation+0.994
in
Token in
Feature activation+1.713
this
Token this
Feature activation+1.976
chemical
Token chemical
Feature activation+2.186
aging
Token aging
Feature activation+0.637
process
Token process
Feature activation+0.055
.
Token.
Feature activation+1.097
not
Token not
Feature activation+0.303
to
Token to
Feature activation+0.312
worry
Token worry
Feature activation+0.000
about
Token about
Feature activation+1.607
what
Token what
Feature activation+0.675
you
Token you
Feature activation+1.680
remember
Token remember
Feature activation+0.000
and
Token and
Feature activation+0.091
what
Token what
Feature activation+0.592
you
Token you
Feature activation+0.716
don
Token don
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.754
Device
TokenDevice
Feature activation+1.283
use
Token use
Feature activation+0.869
also
Token also
Feature activation+1.523
affects
Token affects
Feature activation+1.617
the
Token the
Feature activation+2.072
performance
Token performance
Feature activation+0.177
of
Token of
Feature activation+0.452
a
Token a
Feature activation+0.406
battery
Token battery
Feature activation+0.071
told
Token told
Feature activation+0.000
him
Token him
Feature activation+0.990
not
Token not
Feature activation+0.303
to
Token to
Feature activation+0.312
worry
Token worry
Feature activation+0.000
about
Token about
Feature activation+1.607
what
Token what
Feature activation+0.675
you
Token you
Feature activation+1.680
remember
Token remember
Feature activation+0.000
and
Token and
Feature activation+0.091
what
Token what
Feature activation+0.592
want
Token want
Feature activation+0.375
to
Token to
Feature activation+0.437
partner
Token partner
Feature activation+0.193
up
Token up
Feature activation+0.308
with
Token with
Feature activation+0.871
someone
Token someone
Feature activation+1.588
younger
Token younger
Feature activation+0.000
or
Token or
Feature activation+0.000
have
Token have
Feature activation+0.064
children
Token children
Feature activation+0.000
,
Token,
Feature activation+0.000
she
Token she
Feature activation+0.000
had
Token had
Feature activation+0.000
to
Token to
Feature activation+0.000
look
Token look
Feature activation+0.000
old
Token old
Feature activation+0.000
,
Token,
Feature activation+1.534
except
Token except
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
brief
Token brief
Feature activation+0.000
flashback
Token flashback
Feature activation+0.000
is
Token is
Feature activation+0.000
30
Token 30
Feature activation+0.000
(
Token (
Feature activation+0.000
he
Tokenhe
Feature activation+0.295
'll
Token'll
Feature activation+1.464
be
Token be
Feature activation+1.529
31
Token 31
Feature activation+0.000
before
Token before
Feature activation+0.000
the
Token the
Feature activation+0.000
'
Token '
Feature activation+0.000
10
Token10
Feature activation+0.000
.
Token.
Feature activation+1.097
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.754
Device
TokenDevice
Feature activation+1.283
use
Token use
Feature activation+0.869
also
Token also
Feature activation+1.523
affects
Token affects
Feature activation+1.617
the
Token the
Feature activation+2.072
performance
Token performance
Feature activation+0.177
of
Token of
Feature activation+0.452
a
Token a
Feature activation+0.406
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.301
.
Token.
Feature activation+0.104
If
Token If
Feature activation+0.190
we
Token we
Feature activation+1.042
are
Token are
Feature activation+1.520
it
Token it
Feature activation+0.479
would
Token would
Feature activation+0.212
not
Token not
Feature activation+0.208
be
Token be
Feature activation+0.095
by
Token by
Feature activation+0.000
and
Token and
Feature activation+0.543
,
Token,
Feature activation+0.329
as
Token as
Feature activation+0.937
such
Token such
Feature activation+0.716
,
Token,
Feature activation+0.836
they
Token they
Feature activation+1.505
live
Token live
Feature activation+0.850
until
Token until
Feature activation+0.689
they
Token they
Feature activation+1.096
are
Token are
Feature activation+1.330
killed
Token killed
Feature activation+0.000

Top DFA by src position
MAX = 2.861

cost
Token cost
Feature activation-0.001
Top resid features:
of
Token of
Feature activation+0.002
Top resid features:
providing
Token providing
Feature activation-0.004
Top resid features:
care
Token care
Feature activation-0.021
Top resid features:
to
Token to
Feature activation-0.000
Top resid features:
older
Token older
Feature activation+2.861
Top resid features:
Australians
Token Australians
Feature activation+0.027
Top resid features:
.
Token.
Feature activation+0.016
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.010
Top resid features:
"
Token"
Feature activation+0.071
Top resid features:
less
Token less
Feature activation+0.123
Top resid features:
effective
Token effective
Feature activation+0.144
Top resid features:
as
Token as
Feature activation+0.027
Top resid features:
they
Token they
Feature activation+0.004
Top resid features:
chemically
Token chemically
Feature activation+0.004
Top resid features:
age
Token age
Feature activation+0.842
Top resid features:
and
Token and
Feature activation+0.011
Top resid features:
their
Token their
Feature activation+0.005
Top resid features:
ability
Token ability
Feature activation+0.039
Top resid features:
to
Token to
Feature activation+0.012
Top resid features:
hold
Token hold
Feature activation+0.023
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
How
TokenHow
Feature activation-0.001
Top resid features:
batteries
Token batteries
Feature activation-0.012
Top resid features:
age
Token age
Feature activation+1.158
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
All
TokenAll
Feature activation+0.001
Top resid features:
recharge
Token recharge
Feature activation-0.017
Top resid features:
able
Tokenable
Feature activation+0.001
Top resid features:
powers
Token powers
Feature activation-0.010
Top resid features:
that
Token that
Feature activation+0.004
Top resid features:
are
Token are
Feature activation+0.001
Top resid features:
set
Token set
Feature activation+0.001
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
expire
Token expire
Feature activation+2.455
Top resid features:
in
Token in
Feature activation+0.026
Top resid features:
November
Token November
Feature activation-0.003
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Massachusetts
Token Massachusetts
Feature activation-0.026
Top resid features:
âĢ
TokenâĢ
Feature activation+0.006
Top resid features:
Ļ
TokenĻ
Feature activation+0.009
Top resid features:
s
Tokens
Feature activation+0.002
Top resid features:
mandatory
Token mandatory
Feature activation-0.012
Top resid features:
retirement
Token retirement
Feature activation+2.033
Top resid features:
age
Token age
Feature activation+0.518
Top resid features:
of
Token of
Feature activation-0.007
Top resid features:
50
Token 50
Feature activation+0.000
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
state
Token state
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
How
TokenHow
Feature activation-0.000
Top resid features:
batteries
Token batteries
Feature activation-0.004
Top resid features:
age
Token age
Feature activation+0.760
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
All
TokenAll
Feature activation-0.000
Top resid features:
recharge
Token recharge
Feature activation-0.005
Top resid features:
able
Tokenable
Feature activation-0.000
Top resid features:
Massachusetts
Token Massachusetts
Feature activation-0.019
Top resid features:
âĢ
TokenâĢ
Feature activation+0.022
Top resid features:
Ļ
TokenĻ
Feature activation+0.010
Top resid features:
s
Tokens
Feature activation+0.002
Top resid features:
mandatory
Token mandatory
Feature activation-0.007
Top resid features:
retirement
Token retirement
Feature activation+1.672
Top resid features:
age
Token age
Feature activation+0.584
Top resid features:
of
Token of
Feature activation+0.026
Top resid features:
50
Token 50
Feature activation+0.061
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
state
Token state
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:
How
TokenHow
Feature activation-0.000
Top resid features:
batteries
Token batteries
Feature activation-0.014
Top resid features:
age
Token age
Feature activation+0.681
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
All
TokenAll
Feature activation+0.000
Top resid features:
recharge
Token recharge
Feature activation-0.010
Top resid features:
able
Tokenable
Feature activation+0.001
Top resid features:
Massachusetts
Token Massachusetts
Feature activation-0.043
Top resid features:
âĢ
TokenâĢ
Feature activation+0.034
Top resid features:
Ļ
TokenĻ
Feature activation+0.012
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
mandatory
Token mandatory
Feature activation-0.009
Top resid features:
retirement
Token retirement
Feature activation+2.069
Top resid features:
age
Token age
Feature activation+0.203
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
50
Token 50
Feature activation+0.014
Top resid features:
for
Token for
Feature activation+0.030
Top resid features:
state
Token state
Feature activation+0.000
Top resid features:
Australian
Token Australian
Feature activation+0.001
Top resid features:
society
Token society
Feature activation+0.016
Top resid features:
will
Token will
Feature activation+0.099
Top resid features:
regress
Token regress
Feature activation+0.017
Top resid features:
50
Token 50
Feature activation+0.099
Top resid features:
years
Token years
Feature activation+1.799
Top resid features:
with
Token with
Feature activation+0.131
Top resid features:
the
Token the
Feature activation+0.024
Top resid features:
impact
Token impact
Feature activation-0.051
Top resid features:
of
Token of
Feature activation+0.011
Top resid features:
this
Token this
Feature activation+0.011
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
How
TokenHow
Feature activation-0.002
Top resid features:
batteries
Token batteries
Feature activation+0.015
Top resid features:
age
Token age
Feature activation+0.892
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
All
TokenAll
Feature activation+0.002
Top resid features:
recharge
Token recharge
Feature activation-0.025
Top resid features:
able
Tokenable
Feature activation+0.001
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
worry
Token worry
Feature activation-0.007
Top resid features:
about
Token about
Feature activation+0.005
Top resid features:
remembering
Token remembering
Feature activation+0.421
Top resid features:
old
Token old
Feature activation+0.602
Top resid features:
times
Token times
Feature activation+0.891
Top resid features:
.
Token.
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
ľ
Tokenľ
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
How
TokenHow
Feature activation-0.002
Top resid features:
batteries
Token batteries
Feature activation+0.043
Top resid features:
age
Token age
Feature activation+0.693
Top resid features:
Ċ
TokenĊ
Feature activation-0.004
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
All
TokenAll
Feature activation+0.002
Top resid features:
recharge
Token recharge
Feature activation-0.004
Top resid features:
able
Tokenable
Feature activation+0.003
Top resid features:
to
Token to
Feature activation-0.007
Top resid features:
worry
Token worry
Feature activation-0.013
Top resid features:
about
Token about
Feature activation+0.003
Top resid features:
remembering
Token remembering
Feature activation+0.752
Top resid features:
old
Token old
Feature activation+0.495
Top resid features:
times
Token times
Feature activation+0.871
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
ľ
Tokenľ
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:
have
Token have
Feature activation+0.000
Top resid features:
far
Token far
Feature activation+0.000
Top resid features:
more
Token more
Feature activation-0.002
Top resid features:
older
Token older
Feature activation+0.601
Top resid features:
guys
Token guys
Feature activation+0.011
Top resid features:
out
Token out
Feature activation+0.001
Top resid features:
there
Token there
Feature activation+0.003
Top resid features:
as
Token as
Feature activation+0.003
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
film
Token film
Feature activation+0.008
Top resid features:
she
Token she
Feature activation+0.018
Top resid features:
had
Token had
Feature activation+0.017
Top resid features:
to
Token to
Feature activation+0.006
Top resid features:
look
Token look
Feature activation+0.000
Top resid features:
old
Token old
Feature activation+1.787
Top resid features:
,
Token,
Feature activation+0.296
Top resid features:
except
Token except
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
brief
Token brief
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.004
Top resid features:
Now
TokenNow
Feature activation+0.009
Top resid features:
Westbrook
Token Westbrook
Feature activation-0.016
Top resid features:
is
Token is
Feature activation-0.002
Top resid features:
30
Token 30
Feature activation+1.843
Top resid features:
(
Token (
Feature activation+0.077
Top resid features:
he
Tokenhe
Feature activation+0.066
Top resid features:
'll
Token'll
Feature activation+0.064
Top resid features:
be
Token be
Feature activation+0.036
Top resid features:
31
Token 31
Feature activation+0.000
Top resid features:
only
Token only
Feature activation-0.004
Top resid features:
factors
Token factors
Feature activation-0.002
Top resid features:
in
Token in
Feature activation+0.011
Top resid features:
this
Token this
Feature activation+0.013
Top resid features:
chemical
Token chemical
Feature activation+0.003
Top resid features:
aging
Token aging
Feature activation+0.803
Top resid features:
process
Token process
Feature activation+0.021
Top resid features:
.
Token.
Feature activation-0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation+0.016
Top resid features:
Device
TokenDevice
Feature activation+0.025
Top resid features:
if
Token if
Feature activation+0.040
Top resid features:
we
Token we
Feature activation+0.013
Top resid features:
are
Token are
Feature activation+0.004
Top resid features:
âĢ
Token âĢ
Feature activation+0.135
Top resid features:
ľ
Tokenľ
Feature activation-0.003
Top resid features:
done
Tokendone
Feature activation+1.780
Top resid features:
âĢ
TokenâĢ
Feature activation-0.001
Top resid features:
Ŀ
TokenĿ
Feature activation+0.009
Top resid features:
.
Token.
Feature activation+0.050
Top resid features:
If
Token If
Feature activation+0.020
Top resid features:
we
Token we
Feature activation+0.017
Top resid features:
by
Token by
Feature activation+0.000
Top resid features:
1
Token 1
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
Age
TokenAge
Feature activation+1.724
Top resid features:
.
Token.
Feature activation+0.039
Top resid features:
Un
Token Un
Feature activation+0.001
Top resid features:
made
Tokenmade
Feature activation+0.012
Top resid features:
were
Token were
Feature activation+0.020
Top resid features:
not
Token not
Feature activation-0.001
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.02

Head 2: 0.46

Head 3: 0.08

Head 4: 0.03

Head 5: 0.07

Head 6: 0.06

Head 7: 0.07

Head 8: 0.03

Head 9: 0.05

Head 10: 0.04

Head 11: 0.03

Positive logits

retirement5.98

aging5.96

retire5.95

longevity5.86

Aging5.69

ageing5.61

retiring5.55

Retirement5.34

immortality5.26

ongevity5.19

retirees5.12

sen5.07

dementia4.95

lifespan4.74

Alzheimer4.74

rejuven4.69

decay4.48

retired4.44

funeral4.31

aged4.11

Negative logits

coord-3.19

furt-2.74

grab-2.71

homophobia-2.66

anooga-2.62

Somali-2.62

ussian-2.56

homophobic-2.52

coord-2.51

Traff-2.38

Freak-2.37

arrests-2.33

Lesbian-2.32

Hook-2.30

Sniper-2.27

裏覚醒-2.27

フォ-2.26

ARGET-2.25

uggle-2.25

Chimera-2.23

INTERVAL 2.185 - 2.428
CONTAINS 0.000%

the
Token the
Feature activation+0.606
only
Token only
Feature activation+1.326
factors
Token factors
Feature activation+0.994
in
Token in
Feature activation+1.713
this
Token this
Feature activation+1.976
chemical
Token chemical
Feature activation+2.186
aging
Token aging
Feature activation+0.637
process
Token process
Feature activation+0.055
.
Token.
Feature activation+1.097
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.754
Australians
Token Australians
Feature activation+0.317
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.076
People
TokenPeople
Feature activation+2.428
do
Token do
Feature activation+0.000
need
Token need
Feature activation+0.000
to
Token to
Feature activation+0.143
realise
Token realise
Feature activation+0.060
that
Token that
Feature activation+1.082

INTERVAL 1.942 - 2.185
CONTAINS 0.000%

that
Token that
Feature activation+0.000
he
Token he
Feature activation+0.000
would
Token would
Feature activation+0.293
like
Token like
Feature activation+0.718
to
Token to
Feature activation+1.494
begin
Token begin
Feature activation+2.045
digging
Token digging
Feature activation+0.000
into
Token into
Feature activation+0.035
the
Token the
Feature activation+0.645
administration
Token administration
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
mandatory
Token mandatory
Feature activation+0.000
retirement
Token retirement
Feature activation+0.395
age
Token age
Feature activation+0.733
of
Token of
Feature activation+1.998
50
Token 50
Feature activation+1.889
for
Token for
Feature activation+1.775
state
Token state
Feature activation+1.175
police
Token police
Feature activation+0.193
officers
Token officers
Feature activation+0.297
not
Token not
Feature activation+0.753
the
Token the
Feature activation+0.606
only
Token only
Feature activation+1.326
factors
Token factors
Feature activation+0.994
in
Token in
Feature activation+1.713
this
Token this
Feature activation+1.976
chemical
Token chemical
Feature activation+2.186
aging
Token aging
Feature activation+0.637
process
Token process
Feature activation+0.055
.
Token.
Feature activation+1.097
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.754
Device
TokenDevice
Feature activation+1.283
use
Token use
Feature activation+0.869
also
Token also
Feature activation+1.523
affects
Token affects
Feature activation+1.617
the
Token the
Feature activation+2.072
performance
Token performance
Feature activation+0.177
of
Token of
Feature activation+0.452
a
Token a
Feature activation+0.406
battery
Token battery
Feature activation+0.071
over
Token over
Feature activation+0.233

INTERVAL 1.700 - 1.942
CONTAINS 0.001%

s
Tokens
Feature activation+0.000
mandatory
Token mandatory
Feature activation+0.000
retirement
Token retirement
Feature activation+0.395
age
Token age
Feature activation+0.733
of
Token of
Feature activation+1.998
50
Token 50
Feature activation+1.889
for
Token for
Feature activation+1.775
state
Token state
Feature activation+1.175
police
Token police
Feature activation+0.193
officers
Token officers
Feature activation+0.297
.
Token.
Feature activation+0.000
mandatory
Token mandatory
Feature activation+0.000
retirement
Token retirement
Feature activation+0.395
age
Token age
Feature activation+0.733
of
Token of
Feature activation+1.998
50
Token 50
Feature activation+1.889
for
Token for
Feature activation+1.775
state
Token state
Feature activation+1.175
police
Token police
Feature activation+0.193
officers
Token officers
Feature activation+0.297
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
budget
Token budget
Feature activation+0.000
.
Token.
Feature activation+0.236
That
Token That
Feature activation+1.274
is
Token is
Feature activation+0.191
way
Token way
Feature activation+0.343
too
Token too
Feature activation+1.755
short
Token short
Feature activation+0.000
sight
Token sight
Feature activation+0.000
ed
Tokened
Feature activation+0.000
.
Token.
Feature activation+0.155
Any
Token Any
Feature activation+0.176
are
Token are
Feature activation+0.557
not
Token not
Feature activation+0.753
the
Token the
Feature activation+0.606
only
Token only
Feature activation+1.326
factors
Token factors
Feature activation+0.994
in
Token in
Feature activation+1.713
this
Token this
Feature activation+1.976
chemical
Token chemical
Feature activation+2.186
aging
Token aging
Feature activation+0.637
process
Token process
Feature activation+0.055
.
Token.
Feature activation+1.097
.
Token.
Feature activation+0.208
These
Token These
Feature activation+0.993
are
Token are
Feature activation+0.334
characteristics
Token characteristics
Feature activation+0.297
of
Token of
Feature activation+1.479
battery
Token battery
Feature activation+1.777
chemistry
Token chemistry
Feature activation+0.217
,
Token,
Feature activation+0.247
common
Token common
Feature activation+0.269
to
Token to
Feature activation+0.344
lithium
Token lithium
Feature activation+0.000

INTERVAL 1.457 - 1.700
CONTAINS 0.002%

want
Token want
Feature activation+0.375
to
Token to
Feature activation+0.437
partner
Token partner
Feature activation+0.193
up
Token up
Feature activation+0.308
with
Token with
Feature activation+0.871
someone
Token someone
Feature activation+1.588
younger
Token younger
Feature activation+0.000
or
Token or
Feature activation+0.000
have
Token have
Feature activation+0.064
children
Token children
Feature activation+0.000
,
Token,
Feature activation+0.000
.
Token.
Feature activation+1.097
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.754
Device
TokenDevice
Feature activation+1.283
use
Token use
Feature activation+0.869
also
Token also
Feature activation+1.523
affects
Token affects
Feature activation+1.617
the
Token the
Feature activation+2.072
performance
Token performance
Feature activation+0.177
of
Token of
Feature activation+0.452
a
Token a
Feature activation+0.406
âĢĵ
Token âĢĵ
Feature activation+0.768
that
Token that
Feature activation+0.628
is
Token is
Feature activation+0.089
,
Token,
Feature activation+0.532
we
Token we
Feature activation+0.133
will
Token will
Feature activation+1.487
no
Token no
Feature activation+0.000
longer
Token longer
Feature activation+0.304
be
Token be
Feature activation+0.475
alive
Token alive
Feature activation+0.000
.
Token.
Feature activation+0.945
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.301
.
Token.
Feature activation+0.104
If
Token If
Feature activation+0.190
we
Token we
Feature activation+1.042
are
Token are
Feature activation+1.520
it
Token it
Feature activation+0.479
would
Token would
Feature activation+0.212
not
Token not
Feature activation+0.208
be
Token be
Feature activation+0.095
by
Token by
Feature activation+0.000
told
Token told
Feature activation+0.000
him
Token him
Feature activation+0.990
not
Token not
Feature activation+0.303
to
Token to
Feature activation+0.312
worry
Token worry
Feature activation+0.000
about
Token about
Feature activation+1.607
what
Token what
Feature activation+0.675
you
Token you
Feature activation+1.680
remember
Token remember
Feature activation+0.000
and
Token and
Feature activation+0.091
what
Token what
Feature activation+0.592

INTERVAL 1.214 - 1.457
CONTAINS 0.003%

so
Token so
Feature activation+0.000
we
Token we
Feature activation+0.000
picked
Token picked
Feature activation+0.243
actors
Token actors
Feature activation+0.847
of
Token of
Feature activation+1.023
all
Token all
Feature activation+1.287
ages
Token ages
Feature activation+0.000
.
Token.
Feature activation+0.000
Looking
Token Looking
Feature activation+0.000
exactly
Token exactly
Feature activation+0.177
like
Token like
Feature activation+0.000
existing
Token existing
Feature activation+0.560
pension
Token pension
Feature activation+0.000
ers
Tokeners
Feature activation+0.539
to
Token to
Feature activation+0.684
swap
Token swap
Feature activation+1.049
their
Token their
Feature activation+1.220
fixed
Token fixed
Feature activation+0.426
annual
Token annual
Feature activation+0.388
payments
Token payments
Feature activation+0.000
for
Token for
Feature activation+0.367
cash
Token cash
Feature activation+0.628
that
Token that
Feature activation+0.000
things
Token things
Feature activation+0.000
worked
Token worked
Feature activation+0.000
fine
Token fine
Feature activation+0.000
under
Token under
Feature activation+0.000
that
Token that
Feature activation+1.217
framework
Token framework
Feature activation+0.000
,
Token,
Feature activation+0.000
ISPs
Token ISPs
Feature activation+0.000
like
Token like
Feature activation+0.000
Comcast
Token Comcast
Feature activation+0.000
now
Token now
Feature activation+0.000
dead
Token dead
Feature activation+0.000
(
Token (
Feature activation+0.000
if
Tokenif
Feature activation+0.000
not
Token not
Feature activation+0.922
quite
Token quite
Feature activation+1.334
buried
Token buried
Feature activation+0.000
).
Token).
Feature activation+0.000
Obama
Token Obama
Feature activation+0.201
and
Token and
Feature activation+0.000
Secretary
Token Secretary
Feature activation+0.000
certificates
Token certificates
Feature activation+0.000
typically
Token typically
Feature activation+0.373
contain
Token contain
Feature activation+0.320
little
Token little
Feature activation+0.000
information
Token information
Feature activation+0.000
about
Token about
Feature activation+1.401
manner
Token manner
Feature activation+0.090
of
Token of
Feature activation+1.301
death
Token death
Feature activation+0.000
beyond
Token beyond
Feature activation+0.531
categories
Token categories
Feature activation+0.000

INTERVAL 0.971 - 1.214
CONTAINS 0.007%

in
Token in
Feature activation+0.000
May
Token May
Feature activation+0.000
but
Token but
Feature activation+0.187
he
Token he
Feature activation+0.452
is
Token is
Feature activation+0.961
not
Token not
Feature activation+1.176
expected
Token expected
Feature activation+0.283
to
Token to
Feature activation+1.057
stand
Token stand
Feature activation+0.073
again
Token again
Feature activation+0.000
.
Token.
Feature activation+0.000
we
Token we
Feature activation+0.388
are
Token are
Feature activation+0.443
not
Token not
Feature activation+0.489
completely
Token completely
Feature activation+0.331
certain
Token certain
Feature activation+0.347
what
Token what
Feature activation+1.181
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
being
Tokenbeing
Feature activation+1.170
dead
Token dead
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
indeed
Token indeed
Feature activation+0.000
points
Token points
Feature activation+0.000
towards
Token towards
Feature activation+0.144
the
Token the
Feature activation+0.041
erroneous
Token erroneous
Feature activation+0.000
post
Token post
Feature activation+1.043
hum
Tokenhum
Feature activation+0.000
ous
Tokenous
Feature activation+0.001
tra
Token tra
Feature activation+0.000
ppings
Tokenppings
Feature activation+0.000
of
Token of
Feature activation+0.467
is
Token is
Feature activation+0.036
no
Token no
Feature activation+0.000
time
Token time
Feature activation+0.000
limit
Token limit
Feature activation+0.000
on
Token on
Feature activation+0.647
how
Token how
Feature activation+0.998
long
Token long
Feature activation+0.094
a
Token a
Feature activation+0.122
dog
Token dog
Feature activation+0.000
or
Token or
Feature activation+0.000
cat
Token cat
Feature activation+0.203
Stanford
Token Stanford
Feature activation+0.000
Center
Token Center
Feature activation+0.000
on
Token on
Feature activation+0.000
L
Token L
Feature activation+0.000
ongevity
Tokenongevity
Feature activation+0.000
,
Token,
Feature activation+1.029
which
Token which
Feature activation+0.000
surveyed
Token surveyed
Feature activation+0.173
9
Token 9
Feature activation+0.000
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.248

INTERVAL 0.728 - 0.971
CONTAINS 0.014%

Ċ
TokenĊ
Feature activation+0.000
Ret
TokenRet
Feature activation+0.000
irement
Tokenirement
Feature activation+0.000
decision
Token decision
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.895
The
TokenThe
Feature activation+0.980
damage
Token damage
Feature activation+0.000
Im
Token Im
Feature activation+0.000
es
Tokenes
Feature activation+0.000
took
Token took
Feature activation+0.000
cer
Tokencer
Feature activation+0.000
a
Tokena
Feature activation+0.305
,
Token,
Feature activation+0.253
and
Token and
Feature activation+0.543
,
Token,
Feature activation+0.329
as
Token as
Feature activation+0.937
such
Token such
Feature activation+0.716
,
Token,
Feature activation+0.836
they
Token they
Feature activation+1.505
live
Token live
Feature activation+0.850
until
Token until
Feature activation+0.689
chemical
Token chemical
Feature activation+2.186
aging
Token aging
Feature activation+0.637
process
Token process
Feature activation+0.055
.
Token.
Feature activation+1.097
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.754
Device
TokenDevice
Feature activation+1.283
use
Token use
Feature activation+0.869
also
Token also
Feature activation+1.523
affects
Token affects
Feature activation+1.617
the
Token the
Feature activation+2.072
risk
Token risk
Feature activation+0.000
of
Token of
Feature activation+0.000
crown
Token crown
Feature activation+0.000
fire
Token fire
Feature activation+0.000
decreases
Token decreases
Feature activation+0.000
with
Token with
Feature activation+0.732
age
Token age
Feature activation+0.094
.
Token.
Feature activation+0.855
Another
Token Another
Feature activation+0.560
study
Token study
Feature activation+0.161
also
Token also
Feature activation+0.190
or
Token or
Feature activation+0.000
higher
Token higher
Feature activation+0.000
and
Token and
Feature activation+0.000
life
Token life
Feature activation+0.000
expectancy
Token expectancy
Feature activation+0.000
is
Token is
Feature activation+0.822
13
Token 13
Feature activation+1.113
years
Token years
Feature activation+0.027
longer
Token longer
Feature activation+0.000
than
Token than
Feature activation+0.000
in
Token in
Feature activation+0.000

INTERVAL 0.486 - 0.728
CONTAINS 0.042%

,
Token,
Feature activation+0.000
Pett
Token Pett
Feature activation+0.000
itte
Tokenitte
Feature activation+0.490
can
Token can
Feature activation+0.256
take
Token take
Feature activation+0.139
his
Token his
Feature activation+0.674
time
Token time
Feature activation+0.189
in
Token in
Feature activation+0.416
getting
Token getting
Feature activation+0.586
prepared
Token prepared
Feature activation+0.000
to
Token to
Feature activation+0.114
of
Token of
Feature activation+0.338
two
Token two
Feature activation+0.000
factors
Token factors
Feature activation+0.000
:
Token:
Feature activation+0.101
a
Token a
Feature activation+0.340
normal
Token normal
Feature activation+0.492
,
Token,
Feature activation+0.292
temporary
Token temporary
Feature activation+0.241
performance
Token performance
Feature activation+0.063
impact
Token impact
Feature activation+0.000
when
Token when
Feature activation+1.082
batteries
Token batteries
Feature activation+0.805
are
Token are
Feature activation+1.171
consum
Token consum
Feature activation+0.000
able
Tokenable
Feature activation+0.081
components
Token components
Feature activation+0.000
that
Token that
Feature activation+0.673
become
Token become
Feature activation+0.711
less
Token less
Feature activation+0.608
effective
Token effective
Feature activation+0.000
as
Token as
Feature activation+0.280
they
Token they
Feature activation+0.527
soon
Token soon
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.488
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.717
probably
Token probably
Feature activation+0.419
likely
Token likely
Feature activation+0.205
that
Token that
Feature activation+0.000
it
Token it
Feature activation+0.006
âĢ
TokenâĢ
Feature activation+0.000
started
Token started
Feature activation+0.609
changing
Token changing
Feature activation+0.811
the
Token the
Feature activation+0.288
business
Token business
Feature activation+0.000
when
Token when
Feature activation+0.000
he
Token he
Feature activation+0.656
was
Token was
Feature activation+0.561
still
Token still
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
Feature activation+0.000
20
Token 20
Feature activation+0.000

INTERVAL 0.243 - 0.486
CONTAINS 0.106%

ľ
Tokenľ
Feature activation+0.125
dos
Tokendos
Feature activation+0.000
age
Tokenage
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
to
Token to
Feature activation+0.271
their
Token their
Feature activation+0.055
r
Token r
Feature activation+0.000
ums
Tokenums
Feature activation+0.000
.
Token.
Feature activation+0.000
Plant
Token Plant
Feature activation+0.000
than
Token than
Feature activation+0.000
three
Token three
Feature activation+0.000
times
Token times
Feature activation+0.000
that
Token that
Feature activation+0.000
of
Token of
Feature activation+0.334
the
Token the
Feature activation+0.287
latter
Token latter
Feature activation+0.208
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
Republican
Token Republican
Feature activation+0.000
bill
Token bill
Feature activation+0.000
bill
Token bill
Feature activation+0.000
in
Token in
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
.
Token.
Feature activation+0.000
Typically
Token Typically
Feature activation+0.000
those
Token those
Feature activation+0.332
laws
Token laws
Feature activation+0.000
required
Token required
Feature activation+0.000
a
Token a
Feature activation+0.000
permit
Token permit
Feature activation+0.000
and
Token and
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
You
TokenYou
Feature activation+0.000
have
Token have
Feature activation+0.000
to
Token to
Feature activation+0.280
either
Token either
Feature activation+0.040
have
Token have
Feature activation+0.032
to
Token to
Feature activation+0.164
have
Token have
Feature activation+0.000
a
Token a
Feature activation+0.000
deal
Token deal
Feature activation+0.000
with
Token with
Feature activation+0.000
that
Token that
Feature activation+0.000
when
Token when
Feature activation+0.000
I
Token I
Feature activation+0.274
get
Token get
Feature activation+0.342
out
Token out
Feature activation+0.184
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000

INTERVAL 0.000 - 0.243
CONTAINS 99.827%

Challenger
Token Challenger
Feature activation+0.000
and
Token and
Feature activation+0.000
Charg
Token Charg
Feature activation+0.000
er
Tokener
Feature activation+0.000
Hell
Token Hell
Feature activation+0.000
cats
Tokencats
Feature activation+0.000
,
Token,
Feature activation+0.000
from
Token from
Feature activation+0.000
boost
Token boost
Feature activation+0.000
pressure
Token pressure
Feature activation+0.000
(
Token (
Feature activation+0.000
pan
Token pan
Feature activation+0.000
ics
Tokenics
Feature activation+0.000
instead
Token instead
Feature activation+0.000
of
Token of
Feature activation+0.000
considering
Token considering
Feature activation+0.000
ideas
Token ideas
Feature activation+0.000
in
Token in
Feature activation+0.000
full
Token full
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
party
Token party
Feature activation+0.000
,
Token,
Feature activation+0.000
unity
Token unity
Feature activation+0.000
is
Token is
Feature activation+0.000
needed
Token needed
Feature activation+0.000
so
Token so
Feature activation+0.000
the
Token the
Feature activation+0.000
big
Token big
Feature activation+0.000
things
Token things
Feature activation+0.000
can
Token can
Feature activation+0.000
ings
Tokenings
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
poor
Token poor
Feature activation+0.000
are
Token are
Feature activation+0.000
any
Token any
Feature activation+0.000
sign
Token sign
Feature activation+0.000
of
Token of
Feature activation+0.000
how
Token how
Feature activation+0.000
liberal
Token liberal
Feature activation+0.000
policymakers
Token policymakers
Feature activation+0.000
C
TokenC
Feature activation+0.000
ALL
TokenALL
Feature activation+0.000
ER
TokenER
Feature activation+0.000
:
Token:
Feature activation+0.000
When
Token When
Feature activation+0.000
you
Token you
Feature activation+0.000
go
Token go
Feature activation+0.000
back
Token back
Feature activation+0.000
and
Token and
Feature activation+0.000
study
Token study
Feature activation+0.000
this
Token this
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 H9.2: (feature 13660)

TOP ACTIVATIONS
MAX = 3.621

only
Token only
Feature activation+0.182
needed
Token needed
Feature activation+1.299
five
Token five
Feature activation+0.427
more
Token more
Feature activation+1.452
games
Token games
Feature activation+1.493
to
Token to
Feature activation+3.621
achieve
Token achieve
Feature activation+0.407
the
Token the
Feature activation+0.000
dreamed
Token dreamed
Feature activation+0.000
-
Token-
Feature activation+0.000
of
Tokenof
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.152
When
TokenWhen
Feature activation+1.041
you
Token you
Feature activation+2.643
see
Token see
Feature activation+2.092
a
Token a
Feature activation+1.956
guy
Token guy
Feature activation+3.485
down
Token down
Feature activation+0.146
,
Token,
Feature activation+0.107
you
Token you
Feature activation+0.333
don
Token don
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Beng
Token Beng
Feature activation+0.000
ts
Tokents
Feature activation+0.000
on
Tokenon
Feature activation+1.705
confirmed
Token confirmed
Feature activation+0.572
he
Token he
Feature activation+2.725
was
Token was
Feature activation+3.124
fine
Token fine
Feature activation+0.000
during
Token during
Feature activation+0.428
a
Token a
Feature activation+0.729
post
Token post
Feature activation+0.450
game
Tokengame
Feature activation+0.822
with
Token with
Feature activation+0.000
blood
Token blood
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.053
wound
Token wound
Feature activation+0.000
so
Token so
Feature activation+2.950
bad
Token bad
Feature activation+0.000
that
Token that
Feature activation+1.204
it
Token it
Feature activation+0.113
prompted
Token prompted
Feature activation+0.614
a
Token a
Feature activation+0.873
the
Token the
Feature activation+1.011
rest
Token rest
Feature activation+0.000
of
Token of
Feature activation+0.137
the
Token the
Feature activation+0.122
session
Token session
Feature activation+0.940
and
Token and
Feature activation+2.935
later
Token later
Feature activation+2.830
was
Token was
Feature activation+2.564
driven
Token driven
Feature activation+1.254
off
Token off
Feature activation+0.042
the
Token the
Feature activation+0.000
Of
Token Of
Feature activation+0.000
those
Token those
Feature activation+1.516
players
Token players
Feature activation+0.803
,
Token,
Feature activation+0.266
seven
Token seven
Feature activation+1.710
never
Token never
Feature activation+2.845
returned
Token returned
Feature activation+1.143
to
Token to
Feature activation+0.239
play
Token play
Feature activation+1.240
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.183
rest
Token rest
Feature activation+0.000
of
Token of
Feature activation+0.137
the
Token the
Feature activation+0.122
session
Token session
Feature activation+0.940
and
Token and
Feature activation+2.935
later
Token later
Feature activation+2.830
was
Token was
Feature activation+2.564
driven
Token driven
Feature activation+1.254
off
Token off
Feature activation+0.042
the
Token the
Feature activation+0.000
field
Token field
Feature activation+0.374
its
Token its
Feature activation+0.000
playoff
Token playoff
Feature activation+0.000
drive
Token drive
Feature activation+0.000
.
Token.
Feature activation+0.230
He
Token He
Feature activation+2.088
posted
Token posted
Feature activation+2.812
a
Token a
Feature activation+2.088
1
Token 1
Feature activation+0.041
.
Token.
Feature activation+0.000
79
Token79
Feature activation+0.000
goals
Token goals
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
That
TokenThat
Feature activation+2.425
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+2.202
s
Tokens
Feature activation+1.955
what
Token what
Feature activation+2.786
we
Token we
Feature activation+1.209
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.588
trying
Token trying
Feature activation+0.000
as
Token as
Feature activation+0.511
Beng
Token Beng
Feature activation+0.000
ts
Tokents
Feature activation+0.000
on
Tokenon
Feature activation+1.705
confirmed
Token confirmed
Feature activation+0.572
he
Token he
Feature activation+2.725
was
Token was
Feature activation+3.124
fine
Token fine
Feature activation+0.000
during
Token during
Feature activation+0.428
a
Token a
Feature activation+0.729
post
Token post
Feature activation+0.450
in
Token in
Feature activation+0.201
the
Token the
Feature activation+0.030
first
Token first
Feature activation+0.131
year
Token year
Feature activation+0.660
back
Token back
Feature activation+1.258
from
Token from
Feature activation+2.693
injury
Token injury
Feature activation+0.000
,
Token,
Feature activation+0.432
players
Token players
Feature activation+0.412
played
Token played
Feature activation+0.555
5
Token 5
Feature activation+0.000
said
Token said
Feature activation+0.368
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.152
When
TokenWhen
Feature activation+1.041
you
Token you
Feature activation+2.643
see
Token see
Feature activation+2.092
a
Token a
Feature activation+1.956
guy
Token guy
Feature activation+3.485
down
Token down
Feature activation+0.146
,
Token,
Feature activation+0.107
of
Token of
Feature activation+0.137
the
Token the
Feature activation+0.122
session
Token session
Feature activation+0.940
and
Token and
Feature activation+2.935
later
Token later
Feature activation+2.830
was
Token was
Feature activation+2.564
driven
Token driven
Feature activation+1.254
off
Token off
Feature activation+0.042
the
Token the
Feature activation+0.000
field
Token field
Feature activation+0.374
in
Token in
Feature activation+0.735
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
team
Token team
Feature activation+0.000
,
Token,
Feature activation+1.532
said
Token said
Feature activation+1.270
he
Token he
Feature activation+2.489
first
Token first
Feature activation+1.338
read
Token read
Feature activation+0.572
about
Token about
Feature activation+0.812
Pit
Token Pit
Feature activation+0.000
re
Tokenre
Feature activation+1.017
the
Token the
Feature activation+0.956
mandated
Token mandated
Feature activation+1.120
protocol
Token protocol
Feature activation+0.307
when
Token when
Feature activation+0.748
a
Token a
Feature activation+1.178
player
Token player
Feature activation+2.452
suffers
Token suffers
Feature activation+0.803
a
Token a
Feature activation+0.969
possible
Token possible
Feature activation+1.459
concussion
Token concussion
Feature activation+0.000
before
Token before
Feature activation+0.080
s
Tokens
Feature activation+0.000
injury
Token injury
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
That
TokenThat
Feature activation+2.425
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+2.202
s
Tokens
Feature activation+1.955
what
Token what
Feature activation+2.786
we
Token we
Feature activation+1.209
spokesman
Token spokesman
Feature activation+0.000
for
Token for
Feature activation+0.139
the
Token the
Feature activation+0.682
club
Token club
Feature activation+0.019
.
Token.
Feature activation+0.000
He
Token He
Feature activation+2.245
sat
Token sat
Feature activation+1.521
on
Token on
Feature activation+1.155
ble
Token ble
Feature activation+0.250
achers
Tokenachers
Feature activation+0.608
off
Token off
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
That
TokenThat
Feature activation+2.425
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+2.202
s
Tokens
Feature activation+1.955
what
Token what
Feature activation+2.786
we
Token we
Feature activation+1.209
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
apparent
Token apparent
Feature activation+0.000
injury
Token injury
Feature activation+0.000
.
Token.
Feature activation+0.129
Luckily
Token Luckily
Feature activation+0.161
his
Token his
Feature activation+2.185
apprehension
Token apprehension
Feature activation+0.000
quickly
Token quickly
Feature activation+0.366
turned
Token turned
Feature activation+0.153
to
Token to
Feature activation+0.867
relief
Token relief
Feature activation+0.000
ack
Tokenack
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Eh
TokenEh
Feature activation+0.000
mer
Tokenmer
Feature activation+2.185
was
Token was
Feature activation+1.865
stret
Token stret
Feature activation+0.000
che
Tokenche
Feature activation+0.000
red
Tokenred
Feature activation+0.000
off
Token off
Feature activation+0.000

Top DFA by src position
MAX = 3.249

together
Token together
Feature activation-0.000
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
When
Token When
Feature activation+0.001
Top resid features:
Bob
Token Bob
Feature activation+0.001
Top resid features:
seriously
Token seriously
Feature activation+0.010
Top resid features:
injured
Token injured
Feature activation+2.170
Top resid features:
a
Token a
Feature activation+0.107
Top resid features:
knee
Token knee
Feature activation+1.845
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
dying
Token dying
Feature activation+0.002
Top resid features:
show
Token show
Feature activation+0.004
Top resid features:
what
Token what
Feature activation-0.000
Top resid features:
happened
Token happened
Feature activation+0.028
Top resid features:
on
Token on
Feature activation-0.001
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
injury
Token injury
Feature activation+2.123
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
Ŀ
TokenĿ
Feature activation+0.002
Top resid features:
He
Token He
Feature activation+0.004
Top resid features:
aps
Tokenaps
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.162
Top resid features:
apparent
Token apparent
Feature activation+0.001
Top resid features:
injury
Token injury
Feature activation+3.249
Top resid features:
.
Token.
Feature activation+0.164
Top resid features:
Luckily
Token Luckily
Feature activation+0.011
Top resid features:
his
Token his
Feature activation+0.094
Top resid features:
apprehension
Token apprehension
Feature activation+0.001
Top resid features:
quickly
Token quickly
Feature activation+0.008
Top resid features:
covered
Token covered
Feature activation+0.013
Top resid features:
with
Token with
Feature activation-0.001
Top resid features:
blood
Token blood
Feature activation+0.136
Top resid features:
,
Token,
Feature activation+0.041
Top resid features:
a
Token a
Feature activation+0.040
Top resid features:
wound
Token wound
Feature activation+3.123
Top resid features:
so
Token so
Feature activation+0.064
Top resid features:
bad
Token bad
Feature activation+0.000
Top resid features:
that
Token that
Feature activation+0.000
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
prompted
Token prompted
Feature activation+0.000
Top resid features:
or
Tokenor
Feature activation+0.000
Top resid features:
appeared
Token appeared
Feature activation+0.019
Top resid features:
to
Token to
Feature activation+0.020
Top resid features:
injure
Token injure
Feature activation+1.172
Top resid features:
his
Token his
Feature activation+0.027
Top resid features:
knee
Token knee
Feature activation+1.810
Top resid features:
during
Token during
Feature activation+0.004
Top resid features:
a
Token a
Feature activation+0.006
Top resid features:
routine
Token routine
Feature activation-0.011
Top resid features:
drill
Token drill
Feature activation-0.000
Top resid features:
,
Token,
Feature activation+0.051
Top resid features:
who
Token who
Feature activation+0.012
Top resid features:
'd
Token'd
Feature activation+0.008
Top resid features:
sustained
Token sustained
Feature activation+0.090
Top resid features:
rupt
Token rupt
Feature activation+0.310
Top resid features:
ured
Tokenured
Feature activation+0.285
Top resid features:
Achilles
Token Achilles
Feature activation+1.513
Top resid features:
tend
Token tend
Feature activation+0.030
Top resid features:
ons
Tokenons
Feature activation+0.907
Top resid features:
between
Token between
Feature activation+0.006
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
years
Token years
Feature activation+0.002
Top resid features:
or
Tokenor
Feature activation+0.000
Top resid features:
appeared
Token appeared
Feature activation+0.010
Top resid features:
to
Token to
Feature activation+0.024
Top resid features:
injure
Token injure
Feature activation+1.197
Top resid features:
his
Token his
Feature activation+0.025
Top resid features:
knee
Token knee
Feature activation+1.672
Top resid features:
during
Token during
Feature activation+0.008
Top resid features:
a
Token a
Feature activation+0.006
Top resid features:
routine
Token routine
Feature activation-0.007
Top resid features:
drill
Token drill
Feature activation-0.017
Top resid features:
,
Token,
Feature activation+0.045
Top resid features:
two
Token two
Feature activation+0.003
Top resid features:
leading
Token leading
Feature activation-0.001
Top resid features:
goal
Token goal
Feature activation-0.049
Top resid features:
ies
Tokenies
Feature activation+0.031
Top resid features:
were
Token were
Feature activation+0.008
Top resid features:
injured
Token injured
Feature activation+3.010
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.015
Top resid features:
back
Token back
Feature activation+0.005
Top resid features:
st
Tokenst
Feature activation-0.000
Top resid features:
opped
Tokenopped
Feature activation+0.013
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
about
Token about
Feature activation+0.027
Top resid features:
Richardson
Token Richardson
Feature activation+0.007
Top resid features:
âĢ
TokenâĢ
Feature activation+0.004
Top resid features:
Ļ
TokenĻ
Feature activation+0.014
Top resid features:
s
Tokens
Feature activation+0.011
Top resid features:
injury
Token injury
Feature activation+2.885
Top resid features:
.
Token.
Feature activation+0.008
Top resid features:
âĢ
Token âĢ
Feature activation-0.004
Top resid features:
ľ
Tokenľ
Feature activation+0.011
Top resid features:
That
TokenThat
Feature activation+0.028
Top resid features:
âĢ
TokenâĢ
Feature activation+0.107
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.150
Top resid features:
apparent
Token apparent
Feature activation+0.005
Top resid features:
injury
Token injury
Feature activation+2.831
Top resid features:
.
Token.
Feature activation+0.224
Top resid features:
Luckily
Token Luckily
Feature activation+0.017
Top resid features:
his
Token his
Feature activation+0.070
Top resid features:
apprehension
Token apprehension
Feature activation+0.005
Top resid features:
quickly
Token quickly
Feature activation+0.019
Top resid features:
sustained
Token sustained
Feature activation+0.024
Top resid features:
rupt
Token rupt
Feature activation+0.084
Top resid features:
ured
Tokenured
Feature activation+0.444
Top resid features:
Achilles
Token Achilles
Feature activation+0.643
Top resid features:
tend
Token tend
Feature activation+0.033
Top resid features:
ons
Tokenons
Feature activation+0.928
Top resid features:
between
Token between
Feature activation+0.002
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
years
Token years
Feature activation+0.001
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
1992
Token 1992
Feature activation-0.018
Top resid features:
show
Token show
Feature activation+0.073
Top resid features:
what
Token what
Feature activation+0.003
Top resid features:
happened
Token happened
Feature activation+0.071
Top resid features:
on
Token on
Feature activation+0.016
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
injury
Token injury
Feature activation+2.369
Top resid features:
,
Token,
Feature activation+0.014
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
Ŀ
TokenĿ
Feature activation+0.003
Top resid features:
He
Token He
Feature activation+0.033
Top resid features:
aps
Tokenaps
Feature activation+0.003
Top resid features:
or
Tokenor
Feature activation+0.002
Top resid features:
appeared
Token appeared
Feature activation+0.009
Top resid features:
to
Token to
Feature activation+0.017
Top resid features:
injure
Token injure
Feature activation+0.969
Top resid features:
his
Token his
Feature activation+0.032
Top resid features:
knee
Token knee
Feature activation+1.136
Top resid features:
during
Token during
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.004
Top resid features:
routine
Token routine
Feature activation-0.008
Top resid features:
drill
Token drill
Feature activation-0.013
Top resid features:
,
Token,
Feature activation+0.063
Top resid features:
two
Token two
Feature activation+0.000
Top resid features:
leading
Token leading
Feature activation-0.000
Top resid features:
goal
Token goal
Feature activation-0.003
Top resid features:
ies
Tokenies
Feature activation-0.005
Top resid features:
were
Token were
Feature activation+0.006
Top resid features:
injured
Token injured
Feature activation+2.554
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.012
Top resid features:
back
Token back
Feature activation+0.034
Top resid features:
st
Tokenst
Feature activation-0.001
Top resid features:
opped
Tokenopped
Feature activation+0.009
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
at
Token at
Feature activation-0.001
Top resid features:
Cincinnati
Token Cincinnati
Feature activation-0.001
Top resid features:
after
Token after
Feature activation+0.017
Top resid features:
sustaining
Token sustaining
Feature activation+0.423
Top resid features:
a
Token a
Feature activation+0.019
Top resid features:
concussion
Token concussion
Feature activation+2.235
Top resid features:
on
Token on
Feature activation+0.010
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
opening
Token opening
Feature activation-0.003
Top resid features:
kickoff
Token kickoff
Feature activation-0.019
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
about
Token about
Feature activation+0.023
Top resid features:
Richardson
Token Richardson
Feature activation+0.019
Top resid features:
âĢ
TokenâĢ
Feature activation+0.041
Top resid features:
Ļ
TokenĻ
Feature activation+0.012
Top resid features:
s
Tokens
Feature activation+0.019
Top resid features:
injury
Token injury
Feature activation+2.596
Top resid features:
.
Token.
Feature activation+0.027
Top resid features:
âĢ
Token âĢ
Feature activation+0.005
Top resid features:
ľ
Tokenľ
Feature activation+0.010
Top resid features:
That
TokenThat
Feature activation+0.037
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
or
Tokenor
Feature activation+0.004
Top resid features:
appeared
Token appeared
Feature activation+0.024
Top resid features:
to
Token to
Feature activation+0.015
Top resid features:
injure
Token injure
Feature activation+0.539
Top resid features:
his
Token his
Feature activation+0.061
Top resid features:
knee
Token knee
Feature activation+1.187
Top resid features:
during
Token during
Feature activation+0.008
Top resid features:
a
Token a
Feature activation+0.010
Top resid features:
routine
Token routine
Feature activation+0.001
Top resid features:
drill
Token drill
Feature activation-0.011
Top resid features:
,
Token,
Feature activation+0.061
Top resid features:
about
Token about
Feature activation+0.009
Top resid features:
Richardson
Token Richardson
Feature activation+0.017
Top resid features:
âĢ
TokenâĢ
Feature activation+0.006
Top resid features:
Ļ
TokenĻ
Feature activation+0.007
Top resid features:
s
Tokens
Feature activation+0.008
Top resid features:
injury
Token injury
Feature activation+2.491
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
âĢ
Token âĢ
Feature activation-0.006
Top resid features:
ľ
Tokenľ
Feature activation+0.004
Top resid features:
That
TokenThat
Feature activation+0.172
Top resid features:
âĢ
TokenâĢ
Feature activation-0.018
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.208
Top resid features:
apparent
Token apparent
Feature activation+0.019
Top resid features:
injury
Token injury
Feature activation+2.008
Top resid features:
.
Token.
Feature activation+0.268
Top resid features:
Luckily
Token Luckily
Feature activation+0.144
Top resid features:
his
Token his
Feature activation+0.208
Top resid features:
apprehension
Token apprehension
Feature activation+0.000
Top resid features:
quickly
Token quickly
Feature activation+0.000
Top resid features:
more
Token more
Feature activation-0.001
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
concern
Token concern
Feature activation+0.001
Top resid features:
was
Token was
Feature activation+0.011
Top resid features:
injuries
Token injuries
Feature activation+1.893
Top resid features:
to
Token to
Feature activation+0.165
Top resid features:
Max
Token Max
Feature activation+0.002
Top resid features:
Eh
Token Eh
Feature activation+0.013
Top resid features:
mer
Tokenmer
Feature activation+0.001
Top resid features:
and
Token and
Feature activation+0.039
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.02

Head 2: 0.46

Head 3: 0.10

Head 4: 0.03

Head 5: 0.11

Head 6: 0.03

Head 7: 0.06

Head 8: 0.03

Head 9: 0.03

Head 10: 0.04

Head 11: 0.01

Positive logits

amput6.13

injury5.75

injuries5.72

healed5.61

injured5.61

surgery5.29

surgeons5.28

stret5.21

Surgery5.13

juries5.13

wounds5.10

Injury5.08

Doctors5.04

surgeries4.94

surgeon4.93

bruises4.91

heals4.89

healing4.85

fractures4.84

fracture4.82

Negative logits

galitarian-3.04

quotas-2.94

wcsstore-2.90

ECO-2.75

arcity-2.70

Agenda-2.57

soDeliveryDate-2.57

ahime-2.45

vertisement-2.44

endas-2.43

Meet-2.41

Diversity-2.41

abund-2.41

marketers-2.40

Princ-2.39

equality-2.38

Climate-2.38

Millenn-2.37

isSpecialOrderable-2.34

Charisma-2.34

INTERVAL 3.259 - 3.621
CONTAINS 0.000%

only
Token only
Feature activation+0.182
needed
Token needed
Feature activation+1.299
five
Token five
Feature activation+0.427
more
Token more
Feature activation+1.452
games
Token games
Feature activation+1.493
to
Token to
Feature activation+3.621
achieve
Token achieve
Feature activation+0.407
the
Token the
Feature activation+0.000
dreamed
Token dreamed
Feature activation+0.000
-
Token-
Feature activation+0.000
of
Tokenof
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.152
When
TokenWhen
Feature activation+1.041
you
Token you
Feature activation+2.643
see
Token see
Feature activation+2.092
a
Token a
Feature activation+1.956
guy
Token guy
Feature activation+3.485
down
Token down
Feature activation+0.146
,
Token,
Feature activation+0.107
you
Token you
Feature activation+0.333
don
Token don
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000

INTERVAL 2.897 - 3.259
CONTAINS 0.000%

the
Token the
Feature activation+1.011
rest
Token rest
Feature activation+0.000
of
Token of
Feature activation+0.137
the
Token the
Feature activation+0.122
session
Token session
Feature activation+0.940
and
Token and
Feature activation+2.935
later
Token later
Feature activation+2.830
was
Token was
Feature activation+2.564
driven
Token driven
Feature activation+1.254
off
Token off
Feature activation+0.042
the
Token the
Feature activation+0.000
with
Token with
Feature activation+0.000
blood
Token blood
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.053
wound
Token wound
Feature activation+0.000
so
Token so
Feature activation+2.950
bad
Token bad
Feature activation+0.000
that
Token that
Feature activation+1.204
it
Token it
Feature activation+0.113
prompted
Token prompted
Feature activation+0.614
a
Token a
Feature activation+0.873
Beng
Token Beng
Feature activation+0.000
ts
Tokents
Feature activation+0.000
on
Tokenon
Feature activation+1.705
confirmed
Token confirmed
Feature activation+0.572
he
Token he
Feature activation+2.725
was
Token was
Feature activation+3.124
fine
Token fine
Feature activation+0.000
during
Token during
Feature activation+0.428
a
Token a
Feature activation+0.729
post
Token post
Feature activation+0.450
game
Tokengame
Feature activation+0.822

INTERVAL 2.535 - 2.897
CONTAINS 0.001%

ľ
Tokenľ
Feature activation+0.000
That
TokenThat
Feature activation+2.425
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+2.202
s
Tokens
Feature activation+1.955
what
Token what
Feature activation+2.786
we
Token we
Feature activation+1.209
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.588
trying
Token trying
Feature activation+0.000
of
Token of
Feature activation+0.137
the
Token the
Feature activation+0.122
session
Token session
Feature activation+0.940
and
Token and
Feature activation+2.935
later
Token later
Feature activation+2.830
was
Token was
Feature activation+2.564
driven
Token driven
Feature activation+1.254
off
Token off
Feature activation+0.042
the
Token the
Feature activation+0.000
field
Token field
Feature activation+0.374
in
Token in
Feature activation+0.735
Of
Token Of
Feature activation+0.000
those
Token those
Feature activation+1.516
players
Token players
Feature activation+0.803
,
Token,
Feature activation+0.266
seven
Token seven
Feature activation+1.710
never
Token never
Feature activation+2.845
returned
Token returned
Feature activation+1.143
to
Token to
Feature activation+0.239
play
Token play
Feature activation+1.240
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.183
rest
Token rest
Feature activation+0.000
of
Token of
Feature activation+0.137
the
Token the
Feature activation+0.122
session
Token session
Feature activation+0.940
and
Token and
Feature activation+2.935
later
Token later
Feature activation+2.830
was
Token was
Feature activation+2.564
driven
Token driven
Feature activation+1.254
off
Token off
Feature activation+0.042
the
Token the
Feature activation+0.000
field
Token field
Feature activation+0.374
its
Token its
Feature activation+0.000
playoff
Token playoff
Feature activation+0.000
drive
Token drive
Feature activation+0.000
.
Token.
Feature activation+0.230
He
Token He
Feature activation+2.088
posted
Token posted
Feature activation+2.812
a
Token a
Feature activation+2.088
1
Token 1
Feature activation+0.041
.
Token.
Feature activation+0.000
79
Token79
Feature activation+0.000
goals
Token goals
Feature activation+0.000

INTERVAL 2.173 - 2.535
CONTAINS 0.001%

.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
That
TokenThat
Feature activation+2.425
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+2.202
s
Tokens
Feature activation+1.955
what
Token what
Feature activation+2.786
we
Token we
Feature activation+1.209
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
spokesman
Token spokesman
Feature activation+0.000
for
Token for
Feature activation+0.139
the
Token the
Feature activation+0.682
club
Token club
Feature activation+0.019
.
Token.
Feature activation+0.000
He
Token He
Feature activation+2.245
sat
Token sat
Feature activation+1.521
on
Token on
Feature activation+1.155
ble
Token ble
Feature activation+0.250
achers
Tokenachers
Feature activation+0.608
off
Token off
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
team
Token team
Feature activation+0.000
,
Token,
Feature activation+1.532
said
Token said
Feature activation+1.270
he
Token he
Feature activation+2.489
first
Token first
Feature activation+1.338
read
Token read
Feature activation+0.572
about
Token about
Feature activation+0.812
Pit
Token Pit
Feature activation+0.000
re
Tokenre
Feature activation+1.017
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
apparent
Token apparent
Feature activation+0.000
injury
Token injury
Feature activation+0.000
.
Token.
Feature activation+0.129
Luckily
Token Luckily
Feature activation+0.161
his
Token his
Feature activation+2.185
apprehension
Token apprehension
Feature activation+0.000
quickly
Token quickly
Feature activation+0.366
turned
Token turned
Feature activation+0.153
to
Token to
Feature activation+0.867
relief
Token relief
Feature activation+0.000
the
Token the
Feature activation+0.956
mandated
Token mandated
Feature activation+1.120
protocol
Token protocol
Feature activation+0.307
when
Token when
Feature activation+0.748
a
Token a
Feature activation+1.178
player
Token player
Feature activation+2.452
suffers
Token suffers
Feature activation+0.803
a
Token a
Feature activation+0.969
possible
Token possible
Feature activation+1.459
concussion
Token concussion
Feature activation+0.000
before
Token before
Feature activation+0.080

INTERVAL 1.811 - 2.173
CONTAINS 0.003%

in
Token in
Feature activation+0.000
his
Token his
Feature activation+0.000
playing
Token playing
Feature activation+0.000
career
Token career
Feature activation+0.000
.
Token.
Feature activation+0.003
He
Token He
Feature activation+2.007
broke
Token broke
Feature activation+0.217
the
Token the
Feature activation+0.000
thumb
Token thumb
Feature activation+0.000
on
Token on
Feature activation+0.000
his
Token his
Feature activation+0.000
we
Token we
Feature activation+0.465
were
Token were
Feature activation+0.317
doing
Token doing
Feature activation+0.999
at
Token at
Feature activation+1.617
home
Token home
Feature activation+0.866
before
Token before
Feature activation+1.830
we
Token we
Feature activation+0.725
were
Token were
Feature activation+1.092
interrupted
Token interrupted
Feature activation+0.000
,
Token,
Feature activation+0.357
âĢ
TokenâĢ
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
While
TokenWhile
Feature activation+1.070
the
Token the
Feature activation+1.927
surgery
Token surgery
Feature activation+0.000
to
Token to
Feature activation+1.321
improve
Token improve
Feature activation+0.619
Stewart
Token Stewart
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
games
Token games
Feature activation+0.000
last
Token last
Feature activation+0.000
season
Token season
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.088
the
Token the
Feature activation+0.659
primary
Token primary
Feature activation+1.932
reason
Token reason
Feature activation+0.000
Dallas
Token Dallas
Feature activation+0.000
slumped
Token slumped
Feature activation+0.000
from
Token from
Feature activation+0.000
12
Token 12
Feature activation+0.000
rehab
Token rehab
Feature activation+0.134
work
Token work
Feature activation+0.063
he
Token he
Feature activation+1.386
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+1.942
currently
Token currently
Feature activation+1.333
putting
Token putting
Feature activation+0.000
in
Token in
Feature activation+0.000
on
Token on
Feature activation+0.007
a
Token a
Feature activation+1.304

INTERVAL 1.448 - 1.811
CONTAINS 0.009%

Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.240
The
TokenThe
Feature activation+0.438
average
Token average
Feature activation+1.567
age
Token age
Feature activation+0.352
for
Token for
Feature activation+1.513
injured
Token injured
Feature activation+0.367
players
Token players
Feature activation+0.894
was
Token was
Feature activation+0.584
29
Token 29
Feature activation+0.945
.
Token.
Feature activation+0.000
worst
Token worst
Feature activation+0.000
possible
Token possible
Feature activation+0.000
trauma
Token trauma
Feature activation+0.000
any
Token any
Feature activation+1.192
parent
Token parent
Feature activation+1.328
can
Token can
Feature activation+1.788
hope
Token hope
Feature activation+0.644
to
Token to
Feature activation+1.609
endure
Token endure
Feature activation+0.992
,
Token,
Feature activation+0.042
the
Token the
Feature activation+0.575
.
Token.
Feature activation+0.000
He
Token He
Feature activation+0.300
tweeted
Token tweeted
Feature activation+0.796
that
Token that
Feature activation+0.501
he
Token he
Feature activation+1.516
had
Token had
Feature activation+1.536
his
Token his
Feature activation+1.318
six
Token six
Feature activation+0.000
-
Token-
Feature activation+0.306
month
Tokenmonth
Feature activation+0.480
MRI
Token MRI
Feature activation+0.000
care
Token care
Feature activation+0.000
for
Token for
Feature activation+0.300
one
Token one
Feature activation+0.318
son
Token son
Feature activation+1.338
who
Token who
Feature activation+1.061
was
Token was
Feature activation+1.470
in
Token in
Feature activation+0.989
a
Token a
Feature activation+0.876
coma
Token coma
Feature activation+0.000
and
Token and
Feature activation+0.246
is
Token is
Feature activation+1.164
had
Token had
Feature activation+0.816
to
Token to
Feature activation+0.502
deal
Token deal
Feature activation+0.000
with
Token with
Feature activation+1.400
plenty
Token plenty
Feature activation+1.704
of
Token of
Feature activation+1.572
difficult
Token difficult
Feature activation+0.580
situations
Token situations
Feature activation+0.000
and
Token and
Feature activation+0.301
we
Token we
Feature activation+0.000
will
Token will
Feature activation+0.284

INTERVAL 1.086 - 1.448
CONTAINS 0.023%

i
Tokeni
Feature activation+1.793
had
Token had
Feature activation+0.000
surgery
Token surgery
Feature activation+0.000
earlier
Token earlier
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+1.420
,
Token,
Feature activation+1.019
which
Token which
Feature activation+0.386
was
Token was
Feature activation+0.000
considered
Token considered
Feature activation+1.147
successful
Token successful
Feature activation+0.211
Spec
Token Spec
Feature activation+0.000
ifies
Tokenifies
Feature activation+0.355
the
Token the
Feature activation+0.937
scaling
Token scaling
Feature activation+0.000
factor
Token factor
Feature activation+0.000
for
Token for
Feature activation+1.103
the
Token the
Feature activation+1.204
damage
Token damage
Feature activation+0.509
players
Token players
Feature activation+0.840
deal
Token deal
Feature activation+0.000
with
Token with
Feature activation+0.173
.
Token.
Feature activation+0.561
But
Token But
Feature activation+1.044
we
Token we
Feature activation+0.856
spoke
Token spoke
Feature activation+0.463
with
Token with
Feature activation+1.593
him
Token him
Feature activation+1.331
last
Token last
Feature activation+0.248
night
Token night
Feature activation+0.288
and
Token and
Feature activation+0.461
everyone
Token everyone
Feature activation+0.477
felt
Token felt
Feature activation+0.860
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.456
he
Token he
Feature activation+0.378
will
Token will
Feature activation+0.519
have
Token have
Feature activation+0.465
to
Token to
Feature activation+1.350
recover
Token recover
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.268
concussion
Token concussion
Feature activation+0.166
on
Token on
Feature activation+0.000
difference
Token difference
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
says
Token says
Feature activation+1.044
the
Token the
Feature activation+1.240
Rep
Token Rep
Feature activation+0.068
sol
Tokensol
Feature activation+0.000
Honda
Token Honda
Feature activation+0.076
rider
Token rider
Feature activation+0.000
as
Token as
Feature activation+0.319

INTERVAL 0.724 - 1.086
CONTAINS 0.048%

.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Spect
TokenSpect
Feature activation+0.000
or
Tokenor
Feature activation+1.813
appeared
Token appeared
Feature activation+0.725
to
Token to
Feature activation+1.236
injure
Token injure
Feature activation+0.000
his
Token his
Feature activation+0.663
knee
Token knee
Feature activation+0.000
during
Token during
Feature activation+0.000
,
Token,
Feature activation+0.000
with
Token with
Feature activation+0.000
bruised
Token bruised
Feature activation+0.000
and
Token and
Feature activation+1.222
bright
Token bright
Feature activation+0.000
red
Token red
Feature activation+0.804
sp
Token sp
Feature activation+0.000
anked
Tokenanked
Feature activation+0.000
buttocks
Token buttocks
Feature activation+0.000
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
Type
TokenType
Feature activation+1.006
of
Token of
Feature activation+1.877
amput
Token amput
Feature activation+0.000
ation
Tokenation
Feature activation+0.895
for
Token for
Feature activation+1.088
an
Token an
Feature activation+0.994
A
Token A
Feature activation+0.000
1
Token1
Feature activation+0.470
classified
Token classified
Feature activation+0.102
sports
Token sports
Feature activation+0.439
person
Tokenperson
Feature activation+0.425
land
Tokenland
Feature activation+0.000
has
Token has
Feature activation+0.000
limited
Token limited
Feature activation+0.000
damage
Token damage
Feature activation+0.000
throughout
Token throughout
Feature activation+0.229
his
Token his
Feature activation+0.783
career
Token career
Feature activation+0.000
,
Token,
Feature activation+0.353
evidenced
Token evidenced
Feature activation+0.000
by
Token by
Feature activation+0.343
a
Token a
Feature activation+0.705
is
Token is
Feature activation+0.000
crucial
Token crucial
Feature activation+0.000
to
Token to
Feature activation+0.000
limiting
Token limiting
Feature activation+0.000
injuries
Token injuries
Feature activation+0.000
.
Token.
Feature activation+0.968
But
Token But
Feature activation+0.019
wait
Token wait
Feature activation+0.000
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
say
Token say
Feature activation+0.000

INTERVAL 0.362 - 0.724
CONTAINS 0.134%

Ċ
TokenĊ
Feature activation+0.237
Of
TokenOf
Feature activation+0.000
the
Token the
Feature activation+1.420
other
Token other
Feature activation+1.357
two
Token two
Feature activation+1.295
dogs
Token dogs
Feature activation+0.714
,
Token,
Feature activation+0.151
only
Token only
Feature activation+0.040
one
Token one
Feature activation+1.264
ended
Token ended
Feature activation+0.849
up
Token up
Feature activation+1.374
suffered
Token suffered
Feature activation+0.324
from
Token from
Feature activation+0.381
double
Token double
Feature activation+0.222
vision
Token vision
Feature activation+0.000
(
Token (
Feature activation+0.000
which
Tokenwhich
Feature activation+0.475
permanently
Token permanently
Feature activation+0.642
ended
Token ended
Feature activation+0.000
his
Token his
Feature activation+0.000
flying
Token flying
Feature activation+0.000
career
Token career
Feature activation+0.000
dead
Token dead
Feature activation+0.000
and
Token and
Feature activation+0.000
over
Token over
Feature activation+0.000
30
Token 30
Feature activation+0.635
injured
Token injured
Feature activation+0.000
in
Token in
Feature activation+0.378
yet
Token yet
Feature activation+0.087
another
Token another
Feature activation+0.348
attack
Token attack
Feature activation+0.000
on
Token on
Feature activation+0.000
an
Token an
Feature activation+0.016
dad
Token dad
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
ass
Token ass
Feature activation+0.000
and
Token and
Feature activation+0.588
then
Token then
Feature activation+0.379
scam
Token scam
Feature activation+0.000
pered
Tokenpered
Feature activation+0.000
away
Token away
Feature activation+0.000
.
Token.
Feature activation+0.000
.
Token.
Feature activation+0.362
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.733
S
TokenS
Feature activation+0.000
oooo
Tokenoooo
Feature activation+0.107
yeah
Token yeah
Feature activation+0.633
.
Token.
Feature activation+0.942
Another
Token Another
Feature activation+1.652
possible
Token possible
Feature activation+1.411
prompt
Token prompt
Feature activation+0.001
or
Token or
Feature activation+0.156

INTERVAL 0.000 - 0.362
CONTAINS 99.780%

Cl
Token Cl
Feature activation+0.000
erc
Tokenerc
Feature activation+0.000
,
Token,
Feature activation+0.000
13
Token 13
Feature activation+0.000
-
Token-
Feature activation+0.000
Rou
Token Rou
Feature activation+0.000
ger
Tokenger
Feature activation+0.000
ie
Tokenie
Feature activation+0.000
,
Token,
Feature activation+0.000
12
Token 12
Feature activation+0.000
-
Token-
Feature activation+0.000
needs
Token needs
Feature activation+0.000
to
Token to
Feature activation+0.000
publicly
Token publicly
Feature activation+0.000
support
Token support
Feature activation+0.000
Akin
Token Akin
Feature activation+0.000
.
Token.
Feature activation+0.000
"
Token "
Feature activation+0.000
I
TokenI
Feature activation+0.000
'm
Token'm
Feature activation+0.000
hoping
Token hoping
Feature activation+0.000
everybody
Token everybody
Feature activation+0.000
this
Token this
Feature activation+0.000
gets
Token gets
Feature activation+0.000
more
Token more
Feature activation+0.000
parents
Token parents
Feature activation+0.000
involved
Token involved
Feature activation+0.000
in
Token in
Feature activation+0.000
what
Token what
Feature activation+0.000
their
Token their
Feature activation+0.000
children
Token children
Feature activation+0.000
do
Token do
Feature activation+0.000
online
Token online
Feature activation+0.000
-
Token-
Feature activation+0.000
209
Token209
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
You
TokenYou
Feature activation+0.000
have
Token have
Feature activation+0.000
10
Token 10
Feature activation+0.000
seconds
Token seconds
Feature activation+0.000
to
Token to
Feature activation+0.000
appeared
Token appeared
Feature activation+0.000
,
Token,
Feature activation+0.000
take
Token take
Feature activation+0.000
him
Token him
Feature activation+0.000
down
Token down
Feature activation+0.000
,
Token,
Feature activation+0.000
get
Token get
Feature activation+0.000
a
Token a
Feature activation+0.000
champ
Token champ
Feature activation+0.000
skull
Token skull
Feature activation+0.000
(
Token (
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 9 in H9.2: (feature 9918)

TOP ACTIVATIONS
MAX = 4.148

for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
big
Token big
Feature activation+1.464
part
Token part
Feature activation+0.000
of
Token of
Feature activation+1.179
the
Token the
Feature activation+4.148
decline
Token decline
Feature activation+1.366
,
Token,
Feature activation+1.949
as
Token as
Feature activation+0.000
big
Token big
Feature activation+0.000
-
Token-
Feature activation+0.000
-
Token-
Feature activation+0.000
million
Tokenmillion
Feature activation+0.000
and
Token and
Feature activation+0.000
digital
Token digital
Feature activation+0.000
sales
Token sales
Feature activation+1.630
also
Token also
Feature activation+3.906
dropped
Token dropped
Feature activation+1.261
0
Token 0
Feature activation+0.000
.
Token.
Feature activation+0.000
7
Token7
Feature activation+0.000
%
Token%
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
We
TokenWe
Feature activation+0.183
are
Token are
Feature activation+1.213
seeing
Token seeing
Feature activation+1.030
similar
Token similar
Feature activation+3.820
effects
Token effects
Feature activation+0.000
on
Token on
Feature activation+0.000
print
Token print
Feature activation+0.000
advertising
Token advertising
Feature activation+0.000
revenues
Token revenues
Feature activation+0.000
The
Token The
Feature activation+2.444
graphs
Token graphs
Feature activation+0.000
below
Token below
Feature activation+0.088
will
Token will
Feature activation+0.000
show
Token show
Feature activation+1.949
the
Token the
Feature activation+3.615
dec
Token dec
Feature activation+1.248
adal
Tokenadal
Feature activation+1.107
changes
Token changes
Feature activation+0.000
between
Token between
Feature activation+0.000
18
Token 18
Feature activation+0.000
It
Token It
Feature activation+1.624
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.514
s
Tokens
Feature activation+1.698
the
Token the
Feature activation+2.245
fourth
Token fourth
Feature activation+3.196
-
Token-
Feature activation+1.407
low
Tokenlow
Feature activation+0.000
est
Tokenest
Feature activation+1.409
level
Token level
Feature activation+1.053
nationwide
Token nationwide
Feature activation+0.745
24
Token24
Feature activation+0.000
million
Token million
Feature activation+0.000
.
Token.
Feature activation+0.000
That
Token That
Feature activation+1.481
's
Token's
Feature activation+2.186
a
Token a
Feature activation+3.131
new
Token new
Feature activation+1.054
low
Token low
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
non
Token non
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
.
Token.
Feature activation+0.000
That
Token That
Feature activation+0.676
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+1.806
s
Tokens
Feature activation+2.897
down
Token down
Feature activation+0.000
from
Token from
Feature activation+0.000
15
Token 15
Feature activation+0.000
.
Token.
Feature activation+0.000
4
Token4
Feature activation+0.000
percent
Token percent
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
summer
Token summer
Feature activation+0.000
quarter
Token quarter
Feature activation+0.000
,
Token,
Feature activation+2.807
second
Token second
Feature activation+1.401
worst
Token worst
Feature activation+0.511
among
Token among
Feature activation+0.000
75
Token 75
Feature activation+0.000
markets
Token markets
Feature activation+0.000
Demand
TokenDemand
Feature activation+1.975
for
Token for
Feature activation+0.000
gasoline
Token gasoline
Feature activation+0.000
last
Token last
Feature activation+1.071
week
Token week
Feature activation+1.168
was
Token was
Feature activation+2.697
about
Token about
Feature activation+0.624
9
Token 9
Feature activation+0.406
.
Token.
Feature activation+0.000
5
Token5
Feature activation+0.000
million
Token million
Feature activation+0.000
decline
Token decline
Feature activation+0.059
has
Token has
Feature activation+1.767
been
Token been
Feature activation+2.242
less
Token less
Feature activation+1.009
pronounced
Token pronounced
Feature activation+0.000
,
Token,
Feature activation+2.607
Interior
Token Interior
Feature activation+0.000
Ministry
Token Ministry
Feature activation+0.476
data
Token data
Feature activation+0.008
show
Token show
Feature activation+0.000
.
Token.
Feature activation+0.334
terrorists
Token terrorists
Feature activation+0.000
are
Token are
Feature activation+0.000
winning
Token winning
Feature activation+0.000
the
Token the
Feature activation+0.000
war
Token war
Feature activation+0.000
,
Token,
Feature activation+2.601
up
Token up
Feature activation+0.275
from
Token from
Feature activation+0.019
33
Token 33
Feature activation+0.000
%
Token%
Feature activation+0.000
in
Token in
Feature activation+0.000
decade
Token decade
Feature activation+0.000
since
Token since
Feature activation+0.000
18
Token 18
Feature activation+0.000
72
Token72
Feature activation+0.000
.
Token.
Feature activation+0.134
The
Token The
Feature activation+2.444
graphs
Token graphs
Feature activation+0.000
below
Token below
Feature activation+0.088
will
Token will
Feature activation+0.000
show
Token show
Feature activation+1.949
the
Token the
Feature activation+3.615
2013
Token 2013
Feature activation+0.000
which
Token which
Feature activation+0.451
was
Token was
Feature activation+1.043
more
Token more
Feature activation+2.198
than
Token than
Feature activation+1.327
the
Token the
Feature activation+2.375
14
Token 14
Feature activation+1.514
,
Token,
Feature activation+0.000
000
Token000
Feature activation+1.952
fall
Token fall
Feature activation+0.000
in
Token in
Feature activation+0.000
The
TokenThe
Feature activation+1.808
country
Token country
Feature activation+0.174
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
biggest
Token biggest
Feature activation+2.350
police
Token police
Feature activation+0.197
departments
Token departments
Feature activation+1.595
are
Token are
Feature activation+0.000
responsible
Token responsible
Feature activation+0.000
for
Token for
Feature activation+0.000
penalties
Token penalties
Feature activation+0.000
took
Token took
Feature activation+0.000
effect
Token effect
Feature activation+0.000
in
Token in
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
,
Token,
Feature activation+2.344
but
Token but
Feature activation+0.000
the
Token the
Feature activation+1.956
J
Token J
Feature activation+0.000
AMA
TokenAMA
Feature activation+0.000
researchers
Token researchers
Feature activation+0.000
percent
Token percent
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
second
Token second
Feature activation+0.000
quarter
Token quarter
Feature activation+0.000
and
Token and
Feature activation+2.269
up
Token up
Feature activation+0.000
from
Token from
Feature activation+0.702
53
Token 53
Feature activation+0.000
.
Token.
Feature activation+0.000
2
Token2
Feature activation+0.000
.
Token.
Feature activation+0.000
8
Token8
Feature activation+0.000
million
Token million
Feature activation+0.000
subscribers
Token subscribers
Feature activation+0.000
.
Token.
Feature activation+0.000
That
Token That
Feature activation+2.253
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+1.302
nearly
Token nearly
Feature activation+1.082
half
Token half
Feature activation+0.512
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Much
TokenMuch
Feature activation+0.028
of
Token of
Feature activation+0.833
the
Token the
Feature activation+2.250
blame
Token blame
Feature activation+0.000
for
Token for
Feature activation+1.983
this
Token this
Feature activation+1.966
can
Token can
Feature activation+0.000
be
Token be
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+1.624
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.514
s
Tokens
Feature activation+1.698
the
Token the
Feature activation+2.245
fourth
Token fourth
Feature activation+3.196
-
Token-
Feature activation+1.407
low
Tokenlow
Feature activation+0.000
est
Tokenest
Feature activation+1.409
level
Token level
Feature activation+1.053
this
Token this
Feature activation+0.000
year
Token year
Feature activation+0.000
the
Token the
Feature activation+0.000
decline
Token decline
Feature activation+0.059
has
Token has
Feature activation+1.767
been
Token been
Feature activation+2.242
less
Token less
Feature activation+1.009
pronounced
Token pronounced
Feature activation+0.000
,
Token,
Feature activation+2.607
Interior
Token Interior
Feature activation+0.000
Ministry
Token Ministry
Feature activation+0.476

Top DFA by src position
MAX = 4.120

<|endoftext|>
Token<|endoftext|>
Feature activation+0.235
Top resid features:
per
Token per
Feature activation-0.007
Top resid features:
person
Token person
Feature activation-0.001
Top resid features:
is
Token is
Feature activation-0.001
Top resid features:
falling
Token falling
Feature activation+2.734
Top resid features:
nationally
Token nationally
Feature activation+0.040
Top resid features:
,
Token,
Feature activation+0.025
Top resid features:
along
Token along
Feature activation+0.002
Top resid features:
with
Token with
Feature activation+0.010
Top resid features:
reported
Token reported
Feature activation+0.025
Top resid features:
revenue
Token revenue
Feature activation-0.008
Top resid features:
from
Token from
Feature activation-0.002
Top resid features:
print
Token print
Feature activation+0.000
Top resid features:
advertising
Token advertising
Feature activation+0.000
Top resid features:
was
Token was
Feature activation-0.004
Top resid features:
down
Token down
Feature activation+2.142
Top resid features:
14
Token 14
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
7
Token7
Feature activation+0.079
Top resid features:
%
Token%
Feature activation+0.105
Top resid features:
to
Token to
Feature activation+0.043
Top resid features:
revenue
Token revenue
Feature activation-0.020
Top resid features:
from
Token from
Feature activation-0.001
Top resid features:
print
Token print
Feature activation+0.006
Top resid features:
advertising
Token advertising
Feature activation+0.002
Top resid features:
was
Token was
Feature activation+0.000
Top resid features:
down
Token down
Feature activation+1.934
Top resid features:
14
Token 14
Feature activation+0.023
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
7
Token7
Feature activation+0.113
Top resid features:
%
Token%
Feature activation+0.097
Top resid features:
to
Token to
Feature activation+0.013
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.249
Top resid features:
ic
Tokenic
Feature activation-0.044
Top resid features:
population
Token population
Feature activation+0.148
Top resid features:
has
Token has
Feature activation-0.014
Top resid features:
fallen
Token fallen
Feature activation+4.120
Top resid features:
each
Token each
Feature activation+0.041
Top resid features:
and
Token and
Feature activation-0.002
Top resid features:
every
Token every
Feature activation-0.002
Top resid features:
decade
Token decade
Feature activation+0.071
Top resid features:
since
Token since
Feature activation+0.011
Top resid features:
in
Token in
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
third
Token third
Feature activation-0.001
Top resid features:
quarter
Token quarter
Feature activation+0.020
Top resid features:
,
Token,
Feature activation+0.017
Top resid features:
down
Token down
Feature activation+3.306
Top resid features:
from
Token from
Feature activation+0.046
Top resid features:
53
Token 53
Feature activation+0.009
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
8
Token8
Feature activation+0.038
Top resid features:
percent
Token percent
Feature activation+0.018
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
17
Token17
Feature activation+0.086
Top resid features:
million
Token million
Feature activation-0.023
Top resid features:
viewers
Token viewers
Feature activation-0.082
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
down
Token down
Feature activation+3.457
Top resid features:
from
Token from
Feature activation-0.008
Top resid features:
last
Token last
Feature activation+0.003
Top resid features:
week
Token week
Feature activation-0.004
Top resid features:
's
Token's
Feature activation-0.003
Top resid features:
3
Token 3
Feature activation-0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.441
Top resid features:
per
Token per
Feature activation-0.003
Top resid features:
person
Token person
Feature activation-0.005
Top resid features:
is
Token is
Feature activation-0.004
Top resid features:
falling
Token falling
Feature activation+3.269
Top resid features:
nationally
Token nationally
Feature activation+0.071
Top resid features:
,
Token,
Feature activation+0.014
Top resid features:
along
Token along
Feature activation+0.003
Top resid features:
with
Token with
Feature activation+0.008
Top resid features:
reported
Token reported
Feature activation+0.007
Top resid features:
Los
Token Los
Feature activation+0.003
Top resid features:
Angeles
Token Angeles
Feature activation+0.004
Top resid features:
and
Token and
Feature activation-0.017
Top resid features:
Orange
Token Orange
Feature activation-0.003
Top resid features:
counties
Token counties
Feature activation-0.006
Top resid features:
fell
Token fell
Feature activation+2.387
Top resid features:
to
Token to
Feature activation+0.293
Top resid features:
46
Token 46
Feature activation-0.073
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
6
Token6
Feature activation-0.004
Top resid features:
percent
Token percent
Feature activation+0.045
Top resid features:
S
TokenS
Feature activation+0.001
Top resid features:
.
Token.
Feature activation-0.001
Top resid features:
gasoline
Token gasoline
Feature activation+0.005
Top resid features:
demand
Token demand
Feature activation+0.047
Top resid features:
:
Token:
Feature activation+0.011
Top resid features:
Falling
Token Falling
Feature activation+2.539
Top resid features:
demand
Token demand
Feature activation+0.047
Top resid features:
for
Token for
Feature activation+0.007
Top resid features:
petroleum
Token petroleum
Feature activation+0.001
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
based
Tokenbased
Feature activation+0.001
Top resid features:
worsened
Token worsened
Feature activation+0.040
Top resid features:
,
Token,
Feature activation-0.003
Top resid features:
this
Token this
Feature activation-0.020
Top resid features:
year
Token year
Feature activation+0.013
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
decline
Token decline
Feature activation+3.106
Top resid features:
has
Token has
Feature activation-0.071
Top resid features:
been
Token been
Feature activation-0.003
Top resid features:
less
Token less
Feature activation+0.070
Top resid features:
pronounced
Token pronounced
Feature activation+0.112
Top resid features:
,
Token,
Feature activation+0.048
Top resid features:
.
Token.
Feature activation+0.031
Top resid features:
That
Token That
Feature activation-0.005
Top resid features:
âĢ
TokenâĢ
Feature activation+0.013
Top resid features:
Ļ
TokenĻ
Feature activation+0.004
Top resid features:
s
Tokens
Feature activation-0.003
Top resid features:
down
Token down
Feature activation+2.934
Top resid features:
from
Token from
Feature activation+0.006
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
33
Token 33
Feature activation+0.006
Top resid features:
%
Token%
Feature activation+0.008
Top resid features:
measured
Token measured
Feature activation-0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.312
Top resid features:
ic
Tokenic
Feature activation-0.070
Top resid features:
population
Token population
Feature activation+0.138
Top resid features:
has
Token has
Feature activation-0.038
Top resid features:
fallen
Token fallen
Feature activation+2.829
Top resid features:
each
Token each
Feature activation+0.023
Top resid features:
and
Token and
Feature activation-0.003
Top resid features:
every
Token every
Feature activation-0.000
Top resid features:
decade
Token decade
Feature activation+0.093
Top resid features:
since
Token since
Feature activation-0.014
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.434
Top resid features:
Ċ
TokenĊ
Feature activation-0.015
Top resid features:
Em
TokenEm
Feature activation-0.009
Top resid features:
igration
Tokenigration
Feature activation-0.069
Top resid features:
fell
Token fell
Feature activation+1.937
Top resid features:
by
Token by
Feature activation+0.879
Top resid features:
29
Token 29
Feature activation+0.112
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
000
Token000
Feature activation+0.056
Top resid features:
between
Token between
Feature activation+0.018
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.373
Top resid features:
per
Token per
Feature activation-0.012
Top resid features:
person
Token person
Feature activation-0.004
Top resid features:
is
Token is
Feature activation+0.001
Top resid features:
falling
Token falling
Feature activation+1.581
Top resid features:
nationally
Token nationally
Feature activation+0.036
Top resid features:
,
Token,
Feature activation+0.022
Top resid features:
along
Token along
Feature activation+0.003
Top resid features:
with
Token with
Feature activation+0.023
Top resid features:
reported
Token reported
Feature activation+0.014
Top resid features:
showing
Token showing
Feature activation+0.023
Top resid features:
that
Token that
Feature activation+0.033
Top resid features:
read
Token read
Feature activation-0.010
Top resid features:
missions
Tokenmissions
Feature activation+0.036
Top resid features:
have
Token have
Feature activation+0.061
Top resid features:
fallen
Token fallen
Feature activation+2.809
Top resid features:
since
Token since
Feature activation+0.073
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
penalties
Token penalties
Feature activation-0.004
Top resid features:
took
Token took
Feature activation+0.001
Top resid features:
effect
Token effect
Feature activation+0.012
Top resid features:
in
Token in
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
third
Token third
Feature activation-0.001
Top resid features:
quarter
Token quarter
Feature activation-0.000
Top resid features:
,
Token,
Feature activation-0.011
Top resid features:
down
Token down
Feature activation+2.408
Top resid features:
from
Token from
Feature activation+0.283
Top resid features:
53
Token 53
Feature activation+0.012
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
8
Token8
Feature activation+0.153
Top resid features:
percent
Token percent
Feature activation+0.047
Top resid features:
but
Token but
Feature activation+0.127
Top resid features:
has
Token has
Feature activation+0.053
Top resid features:
seen
Token seen
Feature activation+0.039
Top resid features:
a
Token a
Feature activation+0.021
Top resid features:
steady
Token steady
Feature activation+0.026
Top resid features:
decline
Token decline
Feature activation+1.827
Top resid features:
since
Token since
Feature activation+0.012
Top resid features:
2011
Token 2011
Feature activation+0.007
Top resid features:
and
Token and
Feature activation+0.014
Top resid features:
now
Token now
Feature activation+0.101
Top resid features:
only
Token only
Feature activation+0.168
Top resid features:
,
Token,
Feature activation-0.001
Top resid features:
with
Token with
Feature activation+0.003
Top resid features:
last
Token last
Feature activation-0.003
Top resid features:
year
Token year
Feature activation+0.010
Top resid features:
seeing
Token seeing
Feature activation+0.041
Top resid features:
declines
Token declines
Feature activation+1.658
Top resid features:
of
Token of
Feature activation+0.104
Top resid features:
6
Token 6
Feature activation+0.011
Top resid features:
%
Token%
Feature activation+0.025
Top resid features:
in
Token in
Feature activation+0.053
Top resid features:
French
Token French
Feature activation-0.009
Top resid features:
in
Token in
Feature activation+0.002
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
third
Token third
Feature activation+0.004
Top resid features:
quarter
Token quarter
Feature activation+0.059
Top resid features:
,
Token,
Feature activation+0.031
Top resid features:
down
Token down
Feature activation+2.073
Top resid features:
from
Token from
Feature activation+0.035
Top resid features:
53
Token 53
Feature activation+0.038
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
8
Token8
Feature activation+0.073
Top resid features:
percent
Token percent
Feature activation+0.030
Top resid features:
worsened
Token worsened
Feature activation+0.038
Top resid features:
,
Token,
Feature activation+0.004
Top resid features:
this
Token this
Feature activation-0.000
Top resid features:
year
Token year
Feature activation+0.006
Top resid features:
the
Token the
Feature activation+0.030
Top resid features:
decline
Token decline
Feature activation+2.885
Top resid features:
has
Token has
Feature activation-0.022
Top resid features:
been
Token been
Feature activation+0.009
Top resid features:
less
Token less
Feature activation+0.000
Top resid features:
pronounced
Token pronounced
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.03

Head 2: 0.46

Head 3: 0.06

Head 4: 0.04

Head 5: 0.06

Head 6: 0.05

Head 7: 0.05

Head 8: 0.03

Head 9: 0.06

Head 10: 0.06

Head 11: 0.02

Positive logits

decline5.38

decrease4.94

declines4.92

plummet4.56

Decre4.53

slowdown4.48

declining4.42

slump4.25

decreases4.23

reductions4.15

decreased4.14

reduction4.10

decreasing4.03

deterioration4.01

diminishing4.00

weakening4.00

rebound3.99

stagnation3.97

decay3.90

shrinking3.88

Negative logits

intent-2.81

oath-2.67

Guests-2.64

extrad-2.57

oa-2.51

Oath-2.47

handlers-2.44

wrap-2.42

oshenko-2.40

agents-2.39

Authorization-2.38

authorizing-2.31

handler-2.29

tattoos-2.26

Extra-2.23

Guant-2.22

Hearth-2.21

ichick-2.18

suits-2.18

sworn-2.17

INTERVAL 3.733 - 4.148
CONTAINS 0.000%

for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
big
Token big
Feature activation+1.464
part
Token part
Feature activation+0.000
of
Token of
Feature activation+1.179
the
Token the
Feature activation+4.148
decline
Token decline
Feature activation+1.366
,
Token,
Feature activation+1.949
as
Token as
Feature activation+0.000
big
Token big
Feature activation+0.000
-
Token-
Feature activation+0.000
-
Token-
Feature activation+0.000
million
Tokenmillion
Feature activation+0.000
and
Token and
Feature activation+0.000
digital
Token digital
Feature activation+0.000
sales
Token sales
Feature activation+1.630
also
Token also
Feature activation+3.906
dropped
Token dropped
Feature activation+1.261
0
Token 0
Feature activation+0.000
.
Token.
Feature activation+0.000
7
Token7
Feature activation+0.000
%
Token%
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
We
TokenWe
Feature activation+0.183
are
Token are
Feature activation+1.213
seeing
Token seeing
Feature activation+1.030
similar
Token similar
Feature activation+3.820
effects
Token effects
Feature activation+0.000
on
Token on
Feature activation+0.000
print
Token print
Feature activation+0.000
advertising
Token advertising
Feature activation+0.000
revenues
Token revenues
Feature activation+0.000

INTERVAL 3.318 - 3.733
CONTAINS 0.000%

The
Token The
Feature activation+2.444
graphs
Token graphs
Feature activation+0.000
below
Token below
Feature activation+0.088
will
Token will
Feature activation+0.000
show
Token show
Feature activation+1.949
the
Token the
Feature activation+3.615
dec
Token dec
Feature activation+1.248
adal
Tokenadal
Feature activation+1.107
changes
Token changes
Feature activation+0.000
between
Token between
Feature activation+0.000
18
Token 18
Feature activation+0.000

INTERVAL 2.904 - 3.318
CONTAINS 0.000%

24
Token24
Feature activation+0.000
million
Token million
Feature activation+0.000
.
Token.
Feature activation+0.000
That
Token That
Feature activation+1.481
's
Token's
Feature activation+2.186
a
Token a
Feature activation+3.131
new
Token new
Feature activation+1.054
low
Token low
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
non
Token non
Feature activation+0.000
It
Token It
Feature activation+1.624
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.514
s
Tokens
Feature activation+1.698
the
Token the
Feature activation+2.245
fourth
Token fourth
Feature activation+3.196
-
Token-
Feature activation+1.407
low
Tokenlow
Feature activation+0.000
est
Tokenest
Feature activation+1.409
level
Token level
Feature activation+1.053
nationwide
Token nationwide
Feature activation+0.745

INTERVAL 2.489 - 2.904
CONTAINS 0.001%

terrorists
Token terrorists
Feature activation+0.000
are
Token are
Feature activation+0.000
winning
Token winning
Feature activation+0.000
the
Token the
Feature activation+0.000
war
Token war
Feature activation+0.000
,
Token,
Feature activation+2.601
up
Token up
Feature activation+0.275
from
Token from
Feature activation+0.019
33
Token 33
Feature activation+0.000
%
Token%
Feature activation+0.000
in
Token in
Feature activation+0.000
Demand
TokenDemand
Feature activation+1.975
for
Token for
Feature activation+0.000
gasoline
Token gasoline
Feature activation+0.000
last
Token last
Feature activation+1.071
week
Token week
Feature activation+1.168
was
Token was
Feature activation+2.697
about
Token about
Feature activation+0.624
9
Token 9
Feature activation+0.406
.
Token.
Feature activation+0.000
5
Token5
Feature activation+0.000
million
Token million
Feature activation+0.000
decline
Token decline
Feature activation+0.059
has
Token has
Feature activation+1.767
been
Token been
Feature activation+2.242
less
Token less
Feature activation+1.009
pronounced
Token pronounced
Feature activation+0.000
,
Token,
Feature activation+2.607
Interior
Token Interior
Feature activation+0.000
Ministry
Token Ministry
Feature activation+0.476
data
Token data
Feature activation+0.008
show
Token show
Feature activation+0.000
.
Token.
Feature activation+0.334
percent
Token percent
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
summer
Token summer
Feature activation+0.000
quarter
Token quarter
Feature activation+0.000
,
Token,
Feature activation+2.807
second
Token second
Feature activation+1.401
worst
Token worst
Feature activation+0.511
among
Token among
Feature activation+0.000
75
Token 75
Feature activation+0.000
markets
Token markets
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
.
Token.
Feature activation+0.000
That
Token That
Feature activation+0.676
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+1.806
s
Tokens
Feature activation+2.897
down
Token down
Feature activation+0.000
from
Token from
Feature activation+0.000
15
Token 15
Feature activation+0.000
.
Token.
Feature activation+0.000
4
Token4
Feature activation+0.000

INTERVAL 2.074 - 2.489
CONTAINS 0.001%

.
Token.
Feature activation+0.000
24
Token24
Feature activation+0.000
million
Token million
Feature activation+0.000
.
Token.
Feature activation+0.000
That
Token That
Feature activation+1.481
's
Token's
Feature activation+2.186
a
Token a
Feature activation+3.131
new
Token new
Feature activation+1.054
low
Token low
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
decade
Token decade
Feature activation+0.000
since
Token since
Feature activation+0.000
18
Token 18
Feature activation+0.000
72
Token72
Feature activation+0.000
.
Token.
Feature activation+0.134
The
Token The
Feature activation+2.444
graphs
Token graphs
Feature activation+0.000
below
Token below
Feature activation+0.088
will
Token will
Feature activation+0.000
show
Token show
Feature activation+1.949
the
Token the
Feature activation+3.615
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Much
TokenMuch
Feature activation+0.028
of
Token of
Feature activation+0.833
the
Token the
Feature activation+2.250
blame
Token blame
Feature activation+0.000
for
Token for
Feature activation+1.983
this
Token this
Feature activation+1.966
can
Token can
Feature activation+0.000
be
Token be
Feature activation+0.000
The
TokenThe
Feature activation+1.808
country
Token country
Feature activation+0.174
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
biggest
Token biggest
Feature activation+2.350
police
Token police
Feature activation+0.197
departments
Token departments
Feature activation+1.595
are
Token are
Feature activation+0.000
responsible
Token responsible
Feature activation+0.000
for
Token for
Feature activation+0.000
point
Token point
Feature activation+0.000
of
Token of
Feature activation+0.000
2004
Token 2004
Feature activation+0.000
.
Token.
Feature activation+0.000
Despite
Token Despite
Feature activation+0.478
this
Token this
Feature activation+2.146
decrease
Token decrease
Feature activation+0.442
,
Token,
Feature activation+0.000
there
Token there
Feature activation+0.000
are
Token are
Feature activation+0.380
a
Token a
Feature activation+0.141

INTERVAL 1.659 - 2.074
CONTAINS 0.002%

-
Token-
Feature activation+0.000
July
TokenJuly
Feature activation+0.000
.
Token.
Feature activation+0.341
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.294
Demand
TokenDemand
Feature activation+1.975
for
Token for
Feature activation+0.000
gasoline
Token gasoline
Feature activation+0.000
last
Token last
Feature activation+1.071
week
Token week
Feature activation+1.168
was
Token was
Feature activation+2.697
Much
TokenMuch
Feature activation+0.028
of
Token of
Feature activation+0.833
the
Token the
Feature activation+2.250
blame
Token blame
Feature activation+0.000
for
Token for
Feature activation+1.983
this
Token this
Feature activation+1.966
can
Token can
Feature activation+0.000
be
Token be
Feature activation+0.000
placed
Token placed
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
June
Token June
Feature activation+0.000
2012
Token 2012
Feature activation+0.000
).
Token).
Feature activation+0.000
A
Token A
Feature activation+0.263
total
Token total
Feature activation+0.894
that
Token that
Feature activation+1.885
compounded
Token compounded
Feature activation+0.019
on
Token on
Feature activation+0.000
their
Token their
Feature activation+0.075
$
Token $
Feature activation+0.000
4
Token4
Feature activation+0.000
,
Token,
Feature activation+0.000
this
Token this
Feature activation+0.000
year
Token year
Feature activation+0.000
the
Token the
Feature activation+0.000
decline
Token decline
Feature activation+0.059
has
Token has
Feature activation+1.767
been
Token been
Feature activation+2.242
less
Token less
Feature activation+1.009
pronounced
Token pronounced
Feature activation+0.000
,
Token,
Feature activation+2.607
Interior
Token Interior
Feature activation+0.000
in
Token in
Feature activation+0.387
activity
Token activity
Feature activation+0.123
,
Token,
Feature activation+0.296
after
Token after
Feature activation+0.625
showing
Token showing
Feature activation+1.518
some
Token some
Feature activation+1.912
signs
Token signs
Feature activation+0.020
of
Token of
Feature activation+2.001
things
Token things
Feature activation+0.360
picking
Token picking
Feature activation+0.727
up
Token up
Feature activation+0.449

INTERVAL 1.244 - 1.659
CONTAINS 0.006%

the
Token the
Feature activation+0.000
country
Token country
Feature activation+0.000
has
Token has
Feature activation+1.249
dropped
Token dropped
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+1.516
next
Token next
Feature activation+0.748
immediate
Token immediate
Feature activation+1.345
step
Token step
Feature activation+0.000
,
Token,
Feature activation+0.000
they
Token they
Feature activation+0.000
in
Token in
Feature activation+0.000
Los
Token Los
Feature activation+0.000
Angeles
Token Angeles
Feature activation+0.274
and
Token and
Feature activation+0.000
Orange
Token Orange
Feature activation+0.000
counties
Token counties
Feature activation+1.479
fell
Token fell
Feature activation+0.554
to
Token to
Feature activation+0.545
46
Token 46
Feature activation+0.000
.
Token.
Feature activation+0.000
6
Token6
Feature activation+0.000
to
Token to
Feature activation+0.000
finish
Token finish
Feature activation+0.000
his
Token his
Feature activation+0.000
talk
Token talk
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+1.342
shares
Token shares
Feature activation+0.000
recovered
Token recovered
Feature activation+0.000
shortly
Token shortly
Feature activation+0.000
thereafter
Token thereafter
Feature activation+0.000
.
Token.
Feature activation+0.000
down
Token down
Feature activation+0.359
in
Token in
Feature activation+0.387
activity
Token activity
Feature activation+0.123
,
Token,
Feature activation+0.296
after
Token after
Feature activation+0.625
showing
Token showing
Feature activation+1.518
some
Token some
Feature activation+1.912
signs
Token signs
Feature activation+0.020
of
Token of
Feature activation+2.001
things
Token things
Feature activation+0.360
picking
Token picking
Feature activation+0.727
there
Token there
Feature activation+0.000
are
Token are
Feature activation+0.380
a
Token a
Feature activation+0.141
number
Token number
Feature activation+0.114
of
Token of
Feature activation+0.717
other
Token other
Feature activation+1.540
metrics
Token metrics
Feature activation+0.516
that
Token that
Feature activation+0.529
point
Token point
Feature activation+0.044
to
Token to
Feature activation+0.978
increases
Token increases
Feature activation+0.080

INTERVAL 0.830 - 1.244
CONTAINS 0.012%

12
Token 12
Feature activation+0.000
months
Token months
Feature activation+0.000
to
Token to
Feature activation+0.000
October
Token October
Feature activation+0.000
2014
Token 2014
Feature activation+0.000
,
Token,
Feature activation+0.937
the
Token the
Feature activation+0.000
last
Token last
Feature activation+0.000
year
Token year
Feature activation+0.000
the
Token the
Feature activation+0.000
system
Token system
Feature activation+0.000
1995
Token 1995
Feature activation+0.000
,
Token,
Feature activation+0.828
though
Token though
Feature activation+0.123
the
Token the
Feature activation+1.160
murder
Token murder
Feature activation+0.000
rate
Token rate
Feature activation+0.962
followed
Token followed
Feature activation+0.000
only
Token only
Feature activation+0.074
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
mid
Token mid
Feature activation+0.000
and
Token and
Feature activation+0.000
ecstasy
Token ecstasy
Feature activation+0.000
,
Token,
Feature activation+0.966
with
Token with
Feature activation+0.388
other
Token other
Feature activation+0.093
drugs
Token drugs
Feature activation+1.120
,
Token,
Feature activation+0.000
including
Token including
Feature activation+0.000
amp
Token amp
Feature activation+0.000
het
Tokenhet
Feature activation+0.000
amines
Tokenamines
Feature activation+0.370
think
Token think
Feature activation+0.000
we
Token we
Feature activation+0.116
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+0.647
come
Token come
Feature activation+0.924
a
Token a
Feature activation+0.000
long
Token long
Feature activation+0.009
way
Token way
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Energy
Token Energy
Feature activation+0.000
Information
Token Information
Feature activation+0.000
Administration
Token Administration
Feature activation+0.000
,
Token,
Feature activation+0.884
a
Token a
Feature activation+0.496
government
Token government
Feature activation+0.000
agency
Token agency
Feature activation+0.000
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000

INTERVAL 0.415 - 0.830
CONTAINS 0.039%

was
Token was
Feature activation+0.565
now
Token now
Feature activation+0.124
seeking
Token seeking
Feature activation+0.000
another
Token another
Feature activation+0.000
shift
Token shift
Feature activation+0.000
:
Token:
Feature activation+0.529
Ri
Token Ri
Feature activation+0.000
Äį
TokenÄį
Feature activation+0.000
ard
Tokenard
Feature activation+0.000
as
Tokenas
Feature activation+0.000
took
Token took
Feature activation+0.000
live
Token live
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
0
Token0
Feature activation+0.000
:
Token:
Feature activation+0.432
42
Token42
Feature activation+0.000
And
Token And
Feature activation+0.000
Project
Token Project
Feature activation+0.000
Glass
Token Glass
Feature activation+0.000
is
Token is
Feature activation+0.000
countries
Token countries
Feature activation+0.000
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
America
Token America
Feature activation+0.000
the
Token the
Feature activation+1.221
fall
Token fall
Feature activation+0.528
began
Token began
Feature activation+0.426
around
Token around
Feature activation+0.000
1991
Token 1991
Feature activation+0.000
;
Token;
Feature activation+0.878
in
Token in
Feature activation+0.000
half
Token half
Feature activation+0.000
of
Token of
Feature activation+0.000
Cornwall
Token Cornwall
Feature activation+0.000
had
Token had
Feature activation+0.000
disappeared
Token disappeared
Feature activation+0.000
,
Token,
Feature activation+0.420
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
me
Token me
Feature activation+0.000
greater
Token greater
Feature activation+0.000
government
Token government
Feature activation+0.000
activism
Token activism
Feature activation+0.000
to
Token to
Feature activation+0.000
address
Token address
Feature activation+0.000
these
Token these
Feature activation+0.606
and
Token and
Feature activation+0.000
other
Token other
Feature activation+0.386
problems
Token problems
Feature activation+0.000
,
Token,
Feature activation+0.033
like
Token like
Feature activation+0.000

INTERVAL 0.000 - 0.415
CONTAINS 99.938%

his
Token his
Feature activation+0.000
first
Token first
Feature activation+0.000
kick
Token kick
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
Sp
Token Sp
Feature activation+0.000
edd
Tokenedd
Feature activation+0.000
ing
Tokening
Feature activation+0.000
hit
Token hit
Feature activation+0.000
the
Token the
Feature activation+0.000
upright
Token upright
Feature activation+0.000
Warriors
Token Warriors
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
I
Token I
Feature activation+0.000
really
Token really
Feature activation+0.000
love
Token love
Feature activation+0.000
The
Token The
Feature activation+0.000
New
Token New
Feature activation+0.000
Warriors
Token Warriors
Feature activation+0.000
.
Token.
Feature activation+0.000
I
Token I
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
promotion
Token promotion
Feature activation+0.000
extends
Token extends
Feature activation+0.000
to
Token to
Feature activation+0.000
owners
Token owners
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Lumia
Token Lumia
Feature activation+0.000
625
Token 625
Feature activation+0.000
,
Token,
Feature activation+0.000
were
Token were
Feature activation+0.000
suspended
Token suspended
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
State
Token State
Feature activation+0.000
Athletic
Token Athletic
Feature activation+0.000
Commission
Token Commission
Feature activation+0.000
on
Token on
Feature activation+0.000
Monday
Token Monday
Feature activation+0.000
t
Tokent
Feature activation+0.000
be
Token be
Feature activation+0.000
able
Token able
Feature activation+0.000
to
Token to
Feature activation+0.000
complete
Token complete
Feature activation+0.000
their
Token their
Feature activation+0.000
work
Token work
Feature activation+0.000
in
Token in
Feature activation+0.000
time
Token time
Feature activation+0.000
for
Token for
Feature activation+0.000
Need
Token Need
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