H8.2 top features

Top feature 0 in H8.2: (feature 7648)

TOP ACTIVATIONS
MAX = 8.988

to
Token to
Feature activation+0.000
begin
Token begin
Feature activation+0.000
executing
Token executing
Feature activation+0.000
them
Token them
Feature activation+0.000
.
Token.
Feature activation+1.281
Ċ
TokenĊ
Feature activation+8.988
Ċ
TokenĊ
Feature activation+4.824
A
TokenA
Feature activation+3.583
keeper
Token keeper
Feature activation+0.000
toss
Token toss
Feature activation+0.000
es
Tokenes
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
support
Token support
Feature activation+0.000
.
Token.
Feature activation+0.367
Ċ
TokenĊ
Feature activation+8.774
Ċ
TokenĊ
Feature activation+4.504
U
TokenU
Feature activation+2.646
.
Token.
Feature activation+1.480
S
TokenS
Feature activation+5.226
.
Token.
Feature activation+5.406
S
TokenS
Feature activation+2.060
.
Token.
Feature activation+2.586
watchdog
Token watchdog
Feature activation+0.000
said
Token said
Feature activation+2.216
.
Token.
Feature activation+0.876
Ċ
TokenĊ
Feature activation+8.589
Ċ
TokenĊ
Feature activation+4.666
The
TokenThe
Feature activation+2.556
total
Token total
Feature activation+0.000
cost
Token cost
Feature activation+0.000
of
Token of
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
state
Token state
Feature activation+0.530
Senate
Token Senate
Feature activation+2.457
.
Token.
Feature activation+0.746
Ċ
TokenĊ
Feature activation+8.289
Ċ
TokenĊ
Feature activation+2.914
The
TokenThe
Feature activation+0.102
bill
Token bill
Feature activation+0.166
,
Token,
Feature activation+0.000
passed
Token passed
Feature activation+1.210
-
Token-
Feature activation+0.000
big
Tokenbig
Feature activation+0.000
gest
Tokengest
Feature activation+0.000
economy
Token economy
Feature activation+0.000
.
Token.
Feature activation+0.063
Ċ
TokenĊ
Feature activation+8.271
Ċ
TokenĊ
Feature activation+4.726
A
TokenA
Feature activation+5.287
man
Token man
Feature activation+4.350
looks
Token looks
Feature activation+3.147
at
Token at
Feature activation+1.993
.
Token.
Feature activation+5.406
Defense
Token Defense
Feature activation+0.903
Secretary
Token Secretary
Feature activation+0.861
Jim
Token Jim
Feature activation+0.000
Mattis
Token Mattis
Feature activation+3.991
(
Token (
Feature activation+8.257
L
TokenL
Feature activation+6.234
)
Token)
Feature activation+0.000
shakes
Token shakes
Feature activation+0.000
hands
Token hands
Feature activation+0.000
with
Token with
Feature activation+0.000
Times
Token Times
Feature activation+0.000
reported
Token reported
Feature activation+0.082
on
Token on
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.105
Ċ
TokenĊ
Feature activation+2.530
C
TokenC
Feature activation+0.694
iting
Tokeniting
Feature activation+0.000
a
Token a
Feature activation+0.000
State
Token State
Feature activation+0.686
LAND
TokenLAND
Feature activation+0.000
,
Token,
Feature activation+0.000
Oregon
Token Oregon
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+8.064
-
Token -
Feature activation+2.937
Wildlife
Token Wildlife
Feature activation+0.000
activists
Token activists
Feature activation+0.108
sued
Token sued
Feature activation+0.786
on
Token on
Feature activation+0.782
amid
Token amid
Feature activation+0.000
deep
Token deep
Feature activation+0.000
political
Token political
Feature activation+0.000
turbulence
Token turbulence
Feature activation+0.000
.
Token.
Feature activation+0.499
Ċ
TokenĊ
Feature activation+8.006
Ċ
TokenĊ
Feature activation+3.447
Brazil
TokenBrazil
Feature activation+0.000
's
Token's
Feature activation+3.427
interim
Token interim
Feature activation+0.000
President
Token President
Feature activation+1.174
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
report
Token report
Feature activation+0.000
.
Token.
Feature activation+0.751
Ċ
TokenĊ
Feature activation+8.003
Ċ
TokenĊ
Feature activation+4.031
Tal
TokenTal
Feature activation+0.000
iban
Tokeniban
Feature activation+1.015
fighters
Token fighters
Feature activation+0.772
pose
Token pose
Feature activation+2.720
to
Token to
Feature activation+0.000
end
Token end
Feature activation+0.000
their
Token their
Feature activation+0.000
lives
Token lives
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.947
Ċ
TokenĊ
Feature activation+1.546
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
This
TokenThis
Feature activation+2.181
end
Token end
Feature activation+0.000
they
Token they
Feature activation+0.000
would
Token would
Feature activation+0.000
never
Token never
Feature activation+0.000
succeed
Token succeed
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.932
Ċ
TokenĊ
Feature activation+3.425
Pro
TokenPro
Feature activation+0.000
-
Token-
Feature activation+0.000
independence
Tokenindependence
Feature activation+0.000
protesters
Token protesters
Feature activation+0.000
months
Token months
Feature activation+0.000
-
Token-
Feature activation+0.000
long
Tokenlong
Feature activation+0.000
conflict
Token conflict
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.866
Ċ
TokenĊ
Feature activation+2.236
M
TokenM
Feature activation+0.000
ilit
Tokenilit
Feature activation+0.000
ia
Tokenia
Feature activation+0.000
men
Token men
Feature activation+0.829
SE
TokenSE
Feature activation+0.000
OU
TokenOU
Feature activation+0.000
L
TokenL
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+7.662
-
Token -
Feature activation+2.314
U
Token U
Feature activation+0.238
.
Token.
Feature activation+0.410
S
TokenS
Feature activation+3.698
.
Token.
Feature activation+2.946
visit
Token visit
Feature activation+0.000
China
Token China
Feature activation+0.000
this
Token this
Feature activation+0.000
year
Token year
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.613
Ċ
TokenĊ
Feature activation+4.112
FILE
TokenFILE
Feature activation+1.504
PHOTO
Token PHOTO
Feature activation+1.189
:
Token:
Feature activation+1.726
Chinese
Token Chinese
Feature activation+0.727
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
AD
TokenAD
Feature activation+0.000
EN
TokenEN
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+7.547
-
Token -
Feature activation+2.415
An
Token An
Feature activation+0.919
unidentified
Token unidentified
Feature activation+0.000
war
Token war
Feature activation+0.000
plane
Tokenplane
Feature activation+0.067
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
NEW
TokenNEW
Feature activation+0.000
YORK
Token YORK
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+7.203
-
Token -
Feature activation+1.510
Warren
Token Warren
Feature activation+0.000
Buffett
Token Buffett
Feature activation+0.000
said
Token said
Feature activation+3.624
on
Token on
Feature activation+0.116
,
Token,
Feature activation+0.000
Calif
Token Calif
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+7.184
-
Token -
Feature activation+1.639
Phys
Token Phys
Feature activation+0.000
ician
Tokenician
Feature activation+0.000
-
Token-
Feature activation+0.000
assisted
Tokenassisted
Feature activation+0.000
unn
Tokenunn
Feature activation+0.000
'
Token'
Feature activation+0.000
Zoo
Token Zoo
Feature activation+1.262
in
Token in
Feature activation+1.213
Vienna
Token Vienna
Feature activation+2.513
,
Token,
Feature activation+7.151
January
Token January
Feature activation+1.214
10
Token 10
Feature activation+1.630
,
Token,
Feature activation+0.798
2012
Token 2012
Feature activation+1.655
.
Token.
Feature activation+2.622
/
Token/
Feature activation+0.000
L
TokenL
Feature activation+0.000
ONDON
TokenONDON
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+6.972
-
Token -
Feature activation+1.426
China
Token China
Feature activation+0.318
has
Token has
Feature activation+0.000
invited
Token invited
Feature activation+0.000
British
Token British
Feature activation+0.152

Top DFA by src position
MAX = 9.700

PORT
TokenPORT
Feature activation+0.005
Top resid features:
LAND
TokenLAND
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
Oregon
Token Oregon
Feature activation-0.009
Top resid features:
(
Token (
Feature activation+0.001
Top resid features:
Reuters
TokenReuters
Feature activation+9.700
Top resid features:
)
Token)
Feature activation+0.211
Top resid features:
-
Token -
Feature activation+0.020
Top resid features:
Wildlife
Token Wildlife
Feature activation+0.007
Top resid features:
activists
Token activists
Feature activation-0.005
Top resid features:
sued
Token sued
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.053
Top resid features:
SE
TokenSE
Feature activation+0.004
Top resid features:
OU
TokenOU
Feature activation+0.003
Top resid features:
L
TokenL
Feature activation+0.003
Top resid features:
(
Token (
Feature activation+0.004
Top resid features:
Reuters
TokenReuters
Feature activation+9.384
Top resid features:
)
Token)
Feature activation+0.205
Top resid features:
-
Token -
Feature activation+0.037
Top resid features:
U
Token U
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
S
TokenS
Feature activation+0.013
Top resid features:
-
Token-
Feature activation-0.000
Top resid features:
war
Tokenwar
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.044
Top resid features:
WASHINGTON
TokenWASHINGTON
Feature activation+0.015
Top resid features:
(
Token (
Feature activation+0.003
Top resid features:
Reuters
TokenReuters
Feature activation+9.190
Top resid features:
)
Token)
Feature activation+0.200
Top resid features:
-
Token -
Feature activation+0.031
Top resid features:
Ten
Token Ten
Feature activation+0.020
Top resid features:
months
Token months
Feature activation+0.002
Top resid features:
after
Token after
Feature activation+0.003
Top resid features:
O
TokenO
Feature activation-0.001
Top resid features:
,
Token,
Feature activation-0.000
Top resid features:
Calif
Token Calif
Feature activation-0.017
Top resid features:
.
Token.
Feature activation+0.009
Top resid features:
(
Token (
Feature activation+0.007
Top resid features:
Reuters
TokenReuters
Feature activation+8.806
Top resid features:
)
Token)
Feature activation+0.232
Top resid features:
-
Token -
Feature activation+0.036
Top resid features:
Phys
Token Phys
Feature activation+0.020
Top resid features:
ician
Tokenician
Feature activation+0.015
Top resid features:
-
Token-
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.434
Top resid features:
O
TokenO
Feature activation-0.003
Top resid features:
(
Token (
Feature activation-0.006
Top resid features:
Reuters
TokenReuters
Feature activation+8.966
Top resid features:
)
Token)
Feature activation+0.288
Top resid features:
-
Token -
Feature activation+0.039
Top resid features:
The
Token The
Feature activation+0.029
Top resid features:
yen
Token yen
Feature activation-0.001
Top resid features:
rocket
Token rocket
Feature activation+0.008
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.014
Top resid features:
SE
TokenSE
Feature activation+0.004
Top resid features:
OU
TokenOU
Feature activation+0.003
Top resid features:
L
TokenL
Feature activation-0.000
Top resid features:
(
Token (
Feature activation+0.000
Top resid features:
Reuters
TokenReuters
Feature activation+9.487
Top resid features:
)
Token)
Feature activation+0.046
Top resid features:
-
Token -
Feature activation+0.004
Top resid features:
U
Token U
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
S
TokenS
Feature activation+0.001
Top resid features:
Vanilla
Token Vanilla
Feature activation+0.001
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.043
Top resid features:
WASHINGTON
TokenWASHINGTON
Feature activation+0.021
Top resid features:
(
Token (
Feature activation+0.003
Top resid features:
Reuters
TokenReuters
Feature activation+8.691
Top resid features:
)
Token)
Feature activation+0.222
Top resid features:
-
Token -
Feature activation+0.031
Top resid features:
The
Token The
Feature activation+0.027
Top resid features:
United
Token United
Feature activation+0.006
Top resid features:
States
Token States
Feature activation+0.005
Top resid features:
PORT
TokenPORT
Feature activation+0.006
Top resid features:
LAND
TokenLAND
Feature activation+0.017
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
Oregon
Token Oregon
Feature activation+0.002
Top resid features:
(
Token (
Feature activation-0.055
Top resid features:
Reuters
TokenReuters
Feature activation+8.944
Top resid features:
)
Token)
Feature activation+0.057
Top resid features:
-
Token -
Feature activation+0.000
Top resid features:
Wildlife
Token Wildlife
Feature activation+0.000
Top resid features:
activists
Token activists
Feature activation+0.000
Top resid features:
sued
Token sued
Feature activation+0.000
Top resid features:
O
TokenO
Feature activation+0.001
Top resid features:
PA
Token PA
Feature activation+0.008
Top resid features:
UL
TokenUL
Feature activation+0.009
Top resid features:
O
TokenO
Feature activation+0.001
Top resid features:
(
Token (
Feature activation+0.004
Top resid features:
Reuters
TokenReuters
Feature activation+8.671
Top resid features:
)
Token)
Feature activation+0.183
Top resid features:
-
Token -
Feature activation+0.031
Top resid features:
A
Token A
Feature activation+0.014
Top resid features:
college
Token college
Feature activation+0.002
Top resid features:
drop
Token drop
Feature activation+0.002
Top resid features:
UN
TokenUN
Feature activation+0.012
Top resid features:
ITED
TokenITED
Feature activation+0.007
Top resid features:
N
Token N
Feature activation+0.001
Top resid features:
ATIONS
TokenATIONS
Feature activation+0.009
Top resid features:
(
Token (
Feature activation+0.001
Top resid features:
Reuters
TokenReuters
Feature activation+8.704
Top resid features:
)
Token)
Feature activation+0.171
Top resid features:
-
Token -
Feature activation+0.029
Top resid features:
The
Token The
Feature activation+0.024
Top resid features:
Taliban
Token Taliban
Feature activation+0.017
Top resid features:
âĢ
TokenâĢ
Feature activation+0.022
Top resid features:
O
TokenO
Feature activation+0.002
Top resid features:
,
Token,
Feature activation-0.001
Top resid features:
Calif
Token Calif
Feature activation-0.014
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
(
Token (
Feature activation+0.005
Top resid features:
Reuters
TokenReuters
Feature activation+8.756
Top resid features:
)
Token)
Feature activation+0.231
Top resid features:
-
Token -
Feature activation+0.041
Top resid features:
Phys
Token Phys
Feature activation+0.023
Top resid features:
ician
Tokenician
Feature activation+0.003
Top resid features:
-
Token-
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.052
Top resid features:
BE
TokenBE
Feature activation+0.003
Top resid features:
IJ
TokenIJ
Feature activation+0.004
Top resid features:
ING
TokenING
Feature activation+0.002
Top resid features:
(
Token (
Feature activation+0.003
Top resid features:
Reuters
TokenReuters
Feature activation+8.667
Top resid features:
)
Token)
Feature activation+0.202
Top resid features:
-
Token -
Feature activation+0.030
Top resid features:
China
Token China
Feature activation+0.014
Top resid features:
said
Token said
Feature activation+0.011
Top resid features:
on
Token on
Feature activation+0.011
Top resid features:
;
Token;
Feature activation+0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.006
Top resid features:
AD
TokenAD
Feature activation+0.007
Top resid features:
EN
TokenEN
Feature activation+0.007
Top resid features:
(
Token (
Feature activation+0.001
Top resid features:
Reuters
TokenReuters
Feature activation+8.478
Top resid features:
)
Token)
Feature activation+0.196
Top resid features:
-
Token -
Feature activation+0.027
Top resid features:
An
Token An
Feature activation+0.012
Top resid features:
unidentified
Token unidentified
Feature activation+0.002
Top resid features:
war
Token war
Feature activation-0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.031
Top resid features:
SE
TokenSE
Feature activation+0.034
Top resid features:
OU
TokenOU
Feature activation+0.012
Top resid features:
L
TokenL
Feature activation-0.024
Top resid features:
(
Token (
Feature activation-0.049
Top resid features:
Reuters
TokenReuters
Feature activation+8.501
Top resid features:
)
Token)
Feature activation+0.104
Top resid features:
-
Token -
Feature activation+0.000
Top resid features:
U
Token U
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
S
TokenS
Feature activation+0.000
Top resid features:
ING
TokenING
Feature activation-0.001
Top resid features:
/
Token/
Feature activation+0.007
Top resid features:
L
TokenL
Feature activation+0.001
Top resid features:
ONDON
TokenONDON
Feature activation+0.001
Top resid features:
(
Token (
Feature activation+0.002
Top resid features:
Reuters
TokenReuters
Feature activation+7.507
Top resid features:
)
Token)
Feature activation+0.162
Top resid features:
-
Token -
Feature activation+0.024
Top resid features:
China
Token China
Feature activation+0.012
Top resid features:
has
Token has
Feature activation+0.008
Top resid features:
invited
Token invited
Feature activation-0.005
Top resid features:
;
Token;
Feature activation+0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.006
Top resid features:
AD
TokenAD
Feature activation+0.002
Top resid features:
EN
TokenEN
Feature activation+0.076
Top resid features:
(
Token (
Feature activation-0.067
Top resid features:
Reuters
TokenReuters
Feature activation+8.306
Top resid features:
)
Token)
Feature activation+0.108
Top resid features:
-
Token -
Feature activation+0.000
Top resid features:
An
Token An
Feature activation+0.000
Top resid features:
unidentified
Token unidentified
Feature activation+0.000
Top resid features:
war
Token war
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.052
Top resid features:
NEW
TokenNEW
Feature activation+0.028
Top resid features:
YORK
Token YORK
Feature activation+0.089
Top resid features:
(
Token (
Feature activation-0.016
Top resid features:
Reuters
TokenReuters
Feature activation+7.864
Top resid features:
)
Token)
Feature activation+0.102
Top resid features:
-
Token -
Feature activation+0.000
Top resid features:
Warren
Token Warren
Feature activation+0.000
Top resid features:
Buffett
Token Buffett
Feature activation+0.000
Top resid features:
said
Token said
Feature activation+0.000
Top resid features:
O
TokenO
Feature activation-0.002
Top resid features:
,
Token,
Feature activation-0.002
Top resid features:
Calif
Token Calif
Feature activation-0.007
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
(
Token (
Feature activation-0.013
Top resid features:
Reuters
TokenReuters
Feature activation+8.013
Top resid features:
)
Token)
Feature activation+0.091
Top resid features:
-
Token -
Feature activation+0.000
Top resid features:
Phys
Token Phys
Feature activation+0.000
Top resid features:
ician
Tokenician
Feature activation+0.000
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
PORT
TokenPORT
Feature activation+0.008
Top resid features:
LAND
TokenLAND
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
Oregon
Token Oregon
Feature activation+0.048
Top resid features:
(
Token (
Feature activation+0.001
Top resid features:
Reuters
TokenReuters
Feature activation+8.966
Top resid features:
)
Token)
Feature activation+0.045
Top resid features:
-
Token -
Feature activation+0.021
Top resid features:
Wildlife
Token Wildlife
Feature activation+0.003
Top resid features:
activists
Token activists
Feature activation-0.001
Top resid features:
sued
Token sued
Feature activation-0.001
Top resid features:
ING
TokenING
Feature activation+0.012
Top resid features:
/
Token/
Feature activation+0.094
Top resid features:
L
TokenL
Feature activation-0.002
Top resid features:
ONDON
TokenONDON
Feature activation+0.121
Top resid features:
(
Token (
Feature activation-0.030
Top resid features:
Reuters
TokenReuters
Feature activation+7.580
Top resid features:
)
Token)
Feature activation+0.058
Top resid features:
-
Token -
Feature activation+0.000
Top resid features:
China
Token China
Feature activation+0.000
Top resid features:
has
Token has
Feature activation+0.000
Top resid features:
invited
Token invited
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.01

Head 2: 0.59

Head 3: 0.05

Head 4: 0.02

Head 5: 0.04

Head 6: 0.02

Head 7: 0.03

Head 8: 0.03

Head 9: 0.02

Head 10: 0.10

Head 11: 0.04

Positive logits

REUTERS3.26

Reuters2.44

Whilst2.43

)",2.34

Finish2.19

whilst2.16

Intellectual2.15

Tayyip2.15

US2.12

LIMITED2.12

)=(2.10

RTX2.07

GAM2.06

BIT2.06

'."2.05

MEN2.02

humour2.01

Electrical2.01

NET2.00

MLG1.99

Negative logits

Newsletter-3.66

-3.07

ukong-2.97

Philly-2.87

-2.85

docs-2.84

»-2.76

··-2.66

-2.60

-2.60

atown-2.57

-2.55

vets-2.54

ité-2.52

waukee-2.51

halla-2.46

-2.43

Champ-2.43

────────-2.42

-2.38

INTERVAL 8.089 - 8.988
CONTAINS 0.001%

Times
Token Times
Feature activation+0.000
reported
Token reported
Feature activation+0.082
on
Token on
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.105
Ċ
TokenĊ
Feature activation+2.530
C
TokenC
Feature activation+0.694
iting
Tokeniting
Feature activation+0.000
a
Token a
Feature activation+0.000
State
Token State
Feature activation+0.686
S
TokenS
Feature activation+2.060
.
Token.
Feature activation+2.586
watchdog
Token watchdog
Feature activation+0.000
said
Token said
Feature activation+2.216
.
Token.
Feature activation+0.876
Ċ
TokenĊ
Feature activation+8.589
Ċ
TokenĊ
Feature activation+4.666
The
TokenThe
Feature activation+2.556
total
Token total
Feature activation+0.000
cost
Token cost
Feature activation+0.000
of
Token of
Feature activation+0.000
-
Token-
Feature activation+0.000
big
Tokenbig
Feature activation+0.000
gest
Tokengest
Feature activation+0.000
economy
Token economy
Feature activation+0.000
.
Token.
Feature activation+0.063
Ċ
TokenĊ
Feature activation+8.271
Ċ
TokenĊ
Feature activation+4.726
A
TokenA
Feature activation+5.287
man
Token man
Feature activation+4.350
looks
Token looks
Feature activation+3.147
at
Token at
Feature activation+1.993
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
state
Token state
Feature activation+0.530
Senate
Token Senate
Feature activation+2.457
.
Token.
Feature activation+0.746
Ċ
TokenĊ
Feature activation+8.289
Ċ
TokenĊ
Feature activation+2.914
The
TokenThe
Feature activation+0.102
bill
Token bill
Feature activation+0.166
,
Token,
Feature activation+0.000
passed
Token passed
Feature activation+1.210
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
support
Token support
Feature activation+0.000
.
Token.
Feature activation+0.367
Ċ
TokenĊ
Feature activation+8.774
Ċ
TokenĊ
Feature activation+4.504
U
TokenU
Feature activation+2.646
.
Token.
Feature activation+1.480
S
TokenS
Feature activation+5.226
.
Token.
Feature activation+5.406

INTERVAL 7.190 - 8.089
CONTAINS 0.001%

they
Token they
Feature activation+0.000
would
Token would
Feature activation+0.000
never
Token never
Feature activation+0.000
succeed
Token succeed
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.932
Ċ
TokenĊ
Feature activation+3.425
Pro
TokenPro
Feature activation+0.000
-
Token-
Feature activation+0.000
independence
Tokenindependence
Feature activation+0.000
protesters
Token protesters
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
report
Token report
Feature activation+0.000
.
Token.
Feature activation+0.751
Ċ
TokenĊ
Feature activation+8.003
Ċ
TokenĊ
Feature activation+4.031
Tal
TokenTal
Feature activation+0.000
iban
Tokeniban
Feature activation+1.015
fighters
Token fighters
Feature activation+0.772
pose
Token pose
Feature activation+2.720
amid
Token amid
Feature activation+0.000
deep
Token deep
Feature activation+0.000
political
Token political
Feature activation+0.000
turbulence
Token turbulence
Feature activation+0.000
.
Token.
Feature activation+0.499
Ċ
TokenĊ
Feature activation+8.006
Ċ
TokenĊ
Feature activation+3.447
Brazil
TokenBrazil
Feature activation+0.000
's
Token's
Feature activation+3.427
interim
Token interim
Feature activation+0.000
President
Token President
Feature activation+1.174
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
NEW
TokenNEW
Feature activation+0.000
YORK
Token YORK
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+7.203
-
Token -
Feature activation+1.510
Warren
Token Warren
Feature activation+0.000
Buffett
Token Buffett
Feature activation+0.000
said
Token said
Feature activation+3.624
on
Token on
Feature activation+0.116
to
Token to
Feature activation+0.000
end
Token end
Feature activation+0.000
their
Token their
Feature activation+0.000
lives
Token lives
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.947
Ċ
TokenĊ
Feature activation+1.546
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
This
TokenThis
Feature activation+2.181
end
Token end
Feature activation+0.000

INTERVAL 6.291 - 7.190
CONTAINS 0.001%

unn
Tokenunn
Feature activation+0.000
'
Token'
Feature activation+0.000
Zoo
Token Zoo
Feature activation+1.262
in
Token in
Feature activation+1.213
Vienna
Token Vienna
Feature activation+2.513
,
Token,
Feature activation+7.151
January
Token January
Feature activation+1.214
10
Token 10
Feature activation+1.630
,
Token,
Feature activation+0.798
2012
Token 2012
Feature activation+1.655
.
Token.
Feature activation+2.622
H
Token H
Feature activation+0.000
AG
TokenAG
Feature activation+0.000
UE
TokenUE
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+6.738
-
Token -
Feature activation+2.020
An
Token An
Feature activation+0.703
international
Token international
Feature activation+0.000
arbitration
Token arbitration
Feature activation+0.000
court
Token court
Feature activation+0.000
/
Token/
Feature activation+0.000
L
TokenL
Feature activation+0.000
ONDON
TokenONDON
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+6.972
-
Token -
Feature activation+1.426
China
Token China
Feature activation+0.318
has
Token has
Feature activation+0.000
invited
Token invited
Feature activation+0.000
British
Token British
Feature activation+0.152
PA
Token PA
Feature activation+0.000
UL
TokenUL
Feature activation+0.000
O
TokenO
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+6.807
-
Token -
Feature activation+1.813
A
Token A
Feature activation+0.612
college
Token college
Feature activation+0.088
drop
Token drop
Feature activation+0.000
out
Tokenout
Feature activation+0.000
ITED
TokenITED
Feature activation+0.000
N
Token N
Feature activation+0.000
ATIONS
TokenATIONS
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+6.820
-
Token -
Feature activation+2.230
The
Token The
Feature activation+0.305
Taliban
Token Taliban
Feature activation+0.132
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000

INTERVAL 5.393 - 6.291
CONTAINS 0.000%

Ċ
TokenĊ
Feature activation+0.492
L
TokenL
Feature activation+0.149
ONDON
TokenONDON
Feature activation+1.534
(
Token (
Feature activation+3.422
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+5.439
-
Token -
Feature activation+1.673
Legislation
Token Legislation
Feature activation+0.563
to
Token to
Feature activation+0.310
begin
Token begin
Feature activation+0.000
the
Token the
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.774
Ċ
TokenĊ
Feature activation+4.504
U
TokenU
Feature activation+2.646
.
Token.
Feature activation+1.480
S
TokenS
Feature activation+5.226
.
Token.
Feature activation+5.406
Defense
Token Defense
Feature activation+0.903
Secretary
Token Secretary
Feature activation+0.861
Jim
Token Jim
Feature activation+0.000
Mattis
Token Mattis
Feature activation+3.991
(
Token (
Feature activation+8.257
Defense
Token Defense
Feature activation+0.903
Secretary
Token Secretary
Feature activation+0.861
Jim
Token Jim
Feature activation+0.000
Mattis
Token Mattis
Feature activation+3.991
(
Token (
Feature activation+8.257
L
TokenL
Feature activation+6.234
)
Token)
Feature activation+0.000
shakes
Token shakes
Feature activation+0.000
hands
Token hands
Feature activation+0.000
with
Token with
Feature activation+0.000
South
Token South
Feature activation+0.000

INTERVAL 4.494 - 5.393
CONTAINS 0.001%

gest
Tokengest
Feature activation+0.000
economy
Token economy
Feature activation+0.000
.
Token.
Feature activation+0.063
Ċ
TokenĊ
Feature activation+8.271
Ċ
TokenĊ
Feature activation+4.726
A
TokenA
Feature activation+5.287
man
Token man
Feature activation+4.350
looks
Token looks
Feature activation+3.147
at
Token at
Feature activation+1.993
at
Token at
Feature activation+0.849
an
Token an
Feature activation+0.000
.
Token.
Feature activation+0.367
Ċ
TokenĊ
Feature activation+8.774
Ċ
TokenĊ
Feature activation+4.504
U
TokenU
Feature activation+2.646
.
Token.
Feature activation+1.480
S
TokenS
Feature activation+5.226
.
Token.
Feature activation+5.406
Defense
Token Defense
Feature activation+0.903
Secretary
Token Secretary
Feature activation+0.861
Jim
Token Jim
Feature activation+0.000
Mattis
Token Mattis
Feature activation+3.991
begin
Token begin
Feature activation+0.000
executing
Token executing
Feature activation+0.000
them
Token them
Feature activation+0.000
.
Token.
Feature activation+1.281
Ċ
TokenĊ
Feature activation+8.988
Ċ
TokenĊ
Feature activation+4.824
A
TokenA
Feature activation+3.583
keeper
Token keeper
Feature activation+0.000
toss
Token toss
Feature activation+0.000
es
Tokenes
Feature activation+0.000
a
Token a
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
support
Token support
Feature activation+0.000
.
Token.
Feature activation+0.367
Ċ
TokenĊ
Feature activation+8.774
Ċ
TokenĊ
Feature activation+4.504
U
TokenU
Feature activation+2.646
.
Token.
Feature activation+1.480
S
TokenS
Feature activation+5.226
.
Token.
Feature activation+5.406
Defense
Token Defense
Feature activation+0.903
at
Token at
Feature activation+0.849
an
Token an
Feature activation+0.000
electronic
Token electronic
Feature activation+0.644
stock
Token stock
Feature activation+0.140
quotation
Token quotation
Feature activation+0.000
board
Token board
Feature activation+4.749
outside
Token outside
Feature activation+0.271
a
Token a
Feature activation+0.000
brokerage
Token brokerage
Feature activation+0.000
in
Token in
Feature activation+0.000
Tokyo
Token Tokyo
Feature activation+2.766

INTERVAL 3.595 - 4.494
CONTAINS 0.002%

Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+7.662
-
Token -
Feature activation+2.314
U
Token U
Feature activation+0.238
.
Token.
Feature activation+0.410
S
TokenS
Feature activation+3.698
.
Token.
Feature activation+2.946
President
Token President
Feature activation+4.332
Donald
Token Donald
Feature activation+0.559
Trump
Token Trump
Feature activation+0.031
âĢ
TokenâĢ
Feature activation+0.000
,
Token,
Feature activation+0.000
Oct
Token Oct
Feature activation+0.000
14
Token 14
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+4.371
-
Token -
Feature activation+1.149
Coal
Token Coal
Feature activation+0.125
will
Token will
Feature activation+0.000
surpass
Token surpass
Feature activation+0.000
oil
Token oil
Feature activation+0.000
REUTERS
Token REUTERS
Feature activation+0.000
/
Token/
Feature activation+0.000
Y
TokenY
Feature activation+0.000
ves
Tokenves
Feature activation+0.000
Herman
Token Herman
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.171
Ċ
TokenĊ
Feature activation+0.192
But
TokenBut
Feature activation+0.000
her
Token her
Feature activation+0.000
attempts
Token attempts
Feature activation+0.000
to
Token to
Feature activation+0.000
are
Tokenare
Feature activation+0.000
lli
Tokenlli
Feature activation+0.000
told
Token told
Feature activation+0.000
Reuters
Token Reuters
Feature activation+0.000
.
Token.
Feature activation+0.333
Ċ
TokenĊ
Feature activation+3.754
Ċ
TokenĊ
Feature activation+2.339
The
TokenThe
Feature activation+0.072
date
Token date
Feature activation+0.000
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
economy
Token economy
Feature activation+0.000
.
Token.
Feature activation+0.063
Ċ
TokenĊ
Feature activation+8.271
Ċ
TokenĊ
Feature activation+4.726
A
TokenA
Feature activation+5.287
man
Token man
Feature activation+4.350
looks
Token looks
Feature activation+3.147
at
Token at
Feature activation+1.993
at
Token at
Feature activation+0.849
an
Token an
Feature activation+0.000
electronic
Token electronic
Feature activation+0.644

INTERVAL 2.696 - 3.595
CONTAINS 0.003%

would
Token would
Feature activation+0.000
never
Token never
Feature activation+0.000
succeed
Token succeed
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.932
Ċ
TokenĊ
Feature activation+3.425
Pro
TokenPro
Feature activation+0.000
-
Token-
Feature activation+0.000
independence
Tokenindependence
Feature activation+0.000
protesters
Token protesters
Feature activation+0.000
carry
Token carry
Feature activation+0.000
rally
Token rally
Feature activation+0.000
,
Token,
Feature activation+0.000
told
Token told
Feature activation+0.000
Reuters
Token Reuters
Feature activation+0.000
.
Token.
Feature activation+0.343
Ċ
TokenĊ
Feature activation+2.852
Ċ
TokenĊ
Feature activation+1.754
An
TokenAn
Feature activation+0.574
a
Tokena
Feature activation+0.000
Mel
Token Mel
Feature activation+0.000
endez
Tokenendez
Feature activation+0.178
government
Token government
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
decision
Token decision
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.920
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
court
Token court
Feature activation+0.000
ordered
Token ordered
Feature activation+0.000
ambassador
Token ambassador
Feature activation+0.000
)
Token)
Feature activation+7.662
-
Token -
Feature activation+2.314
U
Token U
Feature activation+0.238
.
Token.
Feature activation+0.410
S
TokenS
Feature activation+3.698
.
Token.
Feature activation+2.946
President
Token President
Feature activation+4.332
Donald
Token Donald
Feature activation+0.559
Trump
Token Trump
Feature activation+0.031
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
)
Token)
Feature activation+0.000
shakes
Token shakes
Feature activation+0.000
hand
Token hand
Feature activation+0.000
with
Token with
Feature activation+0.191
British
Token British
Feature activation+0.222
Prime
Token Prime
Feature activation+3.009
Minister
Token Minister
Feature activation+0.491
Theresa
Token Theresa
Feature activation+0.000
May
Token May
Feature activation+0.000
before
Token before
Feature activation+0.000
their
Token their
Feature activation+0.265

INTERVAL 1.798 - 2.696
CONTAINS 0.006%

to
Tokento
Feature activation+0.000
Palace
Token Palace
Feature activation+0.976
in
Token in
Feature activation+0.000
Bras
Token Bras
Feature activation+0.000
ilia
Tokenilia
Feature activation+1.260
,
Token,
Feature activation+2.467
Brazil
Token Brazil
Feature activation+0.649
July
Token July
Feature activation+0.000
14
Token 14
Feature activation+1.323
,
Token,
Feature activation+0.637
2016
Token 2016
Feature activation+0.064
n
Tokenn
Feature activation+0.000
ins
Tokenins
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.814
Ċ
TokenĊ
Feature activation+0.475
BER
TokenBER
Feature activation+0.000
LIN
TokenLIN
Feature activation+1.863
(
Token (
Feature activation+4.232
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+5.166
-
Token -
Feature activation+1.934
German
Token German
Feature activation+0.310
-
Token -
Feature activation+1.426
China
Token China
Feature activation+0.318
has
Token has
Feature activation+0.000
invited
Token invited
Feature activation+0.000
British
Token British
Feature activation+0.152
Prime
Token Prime
Feature activation+2.356
Minister
Token Minister
Feature activation+1.263
Theresa
Token Theresa
Feature activation+0.273
May
Token May
Feature activation+0.000
to
Token to
Feature activation+0.000
attend
Token attend
Feature activation+0.000
Kh
TokenKh
Feature activation+0.000
aled
Tokenaled
Feature activation+0.000
Elf
Token Elf
Feature activation+0.000
iq
Tokeniq
Feature activation+0.000
i
Tokeni
Feature activation+1.017
/
Token/
Feature activation+2.081
Pool
TokenPool
Feature activation+0.741
/
Token/
Feature activation+2.622
File
Token File
Feature activation+1.953
Photo
Token Photo
Feature activation+1.989
Ċ
TokenĊ
Feature activation+3.537
(
Token (
Feature activation+0.000
S
TokenS
Feature activation+0.000
ANA
TokenANA
Feature activation+0.000
/
Token /
Feature activation+0.000
REUTERS
Token REUTERS
Feature activation+0.000
)
Token)
Feature activation+1.851
Ċ
TokenĊ
Feature activation+0.305
Ċ
TokenĊ
Feature activation+0.000
That
TokenThat
Feature activation+0.000
extremism
Token extremism
Feature activation+0.000
is
Token is
Feature activation+0.000

INTERVAL 0.899 - 1.798
CONTAINS 0.010%

Ċ
TokenĊ
Feature activation+8.289
Ċ
TokenĊ
Feature activation+2.914
The
TokenThe
Feature activation+0.102
bill
Token bill
Feature activation+0.166
,
Token,
Feature activation+0.000
passed
Token passed
Feature activation+1.210
by
Token by
Feature activation+2.156
the
Token the
Feature activation+0.213
Senate
Token Senate
Feature activation+3.420
Health
Token Health
Feature activation+0.008
Committee
Token Committee
Feature activation+1.482
Ŀ
TokenĿ
Feature activation+0.000
The
Token The
Feature activation+0.000
notice
Token notice
Feature activation+0.000
read
Token read
Feature activation+1.415
:
Token:
Feature activation+0.967
Ċ
TokenĊ
Feature activation+1.506
Ċ
TokenĊ
Feature activation+0.020
All
TokenAll
Feature activation+0.000
trading
Token trading
Feature activation+0.000
exchanges
Token exchanges
Feature activation+0.000
must
Token must
Feature activation+0.000
'
Token'
Feature activation+0.000
Zoo
Token Zoo
Feature activation+1.262
in
Token in
Feature activation+1.213
Vienna
Token Vienna
Feature activation+2.513
,
Token,
Feature activation+7.151
January
Token January
Feature activation+1.214
10
Token 10
Feature activation+1.630
,
Token,
Feature activation+0.798
2012
Token 2012
Feature activation+1.655
.
Token.
Feature activation+2.622
REUTERS
Token REUTERS
Feature activation+0.000
in
Token in
Feature activation+0.000
November
Token November
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
U
Token U
Feature activation+0.297
.
Token.
Feature activation+1.062
S
TokenS
Feature activation+2.060
.
Token.
Feature activation+2.586
watchdog
Token watchdog
Feature activation+0.000
said
Token said
Feature activation+2.216
.
Token.
Feature activation+0.876
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
The
Token The
Feature activation+0.000
notice
Token notice
Feature activation+0.000
read
Token read
Feature activation+1.415
:
Token:
Feature activation+0.967
Ċ
TokenĊ
Feature activation+1.506
Ċ
TokenĊ
Feature activation+0.020
All
TokenAll
Feature activation+0.000
trading
Token trading
Feature activation+0.000
exchanges
Token exchanges
Feature activation+0.000

INTERVAL 0.000 - 0.899
CONTAINS 99.974%

Harry
Token Harry
Feature activation+0.000
Reid
Token Reid
Feature activation+0.000
criticism
Token criticism
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
r
Token r
Feature activation+0.000
ips
Tokenips
Feature activation+0.000
Harry
Token Harry
Feature activation+0.000
Reid
Token Reid
Feature activation+0.000
for
Token for
Feature activation+0.000
'
Token '
Feature activation+0.000
failed
Tokenfailed
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
NEW
TokenNEW
Feature activation+0.000
YORK
Token YORK
Feature activation+0.000
--
Token --
Feature activation+0.000
Justin
Token Justin
Feature activation+0.000
Ver
Token Ver
Feature activation+0.000
lander
Tokenlander
Feature activation+0.000
bree
Token bree
Feature activation+0.000
zed
Tokenzed
Feature activation+0.000
Players
TokenPlayers
Feature activation+0.000
have
Token have
Feature activation+0.000
reported
Token reported
Feature activation+0.000
running
Token running
Feature activation+0.000
into
Token into
Feature activation+0.000
frame
Token frame
Feature activation+0.000
rate
Token rate
Feature activation+0.000
and
Token and
Feature activation+0.000
rum
Token rum
Feature activation+0.000
ble
Tokenble
Feature activation+0.000
issues
Token issues
Feature activation+0.000
a
Token a
Feature activation+0.000
hero
Token hero
Feature activation+0.000
,
Token,
Feature activation+0.000
saying
Token saying
Feature activation+0.000
he
Token he
Feature activation+0.000
put
Token put
Feature activation+0.000
his
Token his
Feature activation+0.000
own
Token own
Feature activation+0.000
life
Token life
Feature activation+0.000
at
Token at
Feature activation+0.000
risk
Token risk
Feature activation+0.000
ruled
Token ruled
Feature activation+0.000
State
Token State
Feature activation+0.000
of
Token of
Feature activation+0.000
Missouri
Token Missouri
Feature activation+0.000
whose
Token whose
Feature activation+0.000
property
Token property
Feature activation+0.000
tax
Token tax
Feature activation+0.000
rate
Token rate
Feature activation+0.000
is
Token is
Feature activation+0.000
one
Token one
Feature activation+0.000
of
Token of
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 H8.2: (feature 16181)

TOP ACTIVATIONS
MAX = 8.506

says
Token says
Feature activation+0.000
think
Token think
Feature activation+0.000
tank
Tokentank
Feature activation+0.000
Read
Token Read
Feature activation+4.479
more
Token more
Feature activation+1.863
Ċ
TokenĊ
Feature activation+8.506
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
simplistic
Token simplistic
Feature activation+0.000
as
Token as
Feature activation+0.000
it
Token it
Feature activation+0.000
happened
Token happened
Feature activation+0.000
Read
Token Read
Feature activation+5.405
more
Token more
Feature activation+0.441
Ċ
TokenĊ
Feature activation+8.383
Ċ
TokenĊ
Feature activation+0.000
Di
TokenDi
Feature activation+0.000
Nat
Token Nat
Feature activation+0.000
ale
Tokenale
Feature activation+0.000
was
Token was
Feature activation+0.000
Spurs
Token Spurs
Feature activation+0.000
midfield
Token midfield
Feature activation+0.000
role
Token role
Feature activation+0.000
Read
Token Read
Feature activation+7.914
more
Token more
Feature activation+1.732
Ċ
TokenĊ
Feature activation+8.179
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Following
TokenFollowing
Feature activation+0.000
last
Token last
Feature activation+0.000
settle
Token settle
Feature activation+0.000
the
Token the
Feature activation+0.000
matter
Token matter
Feature activation+0.000
promptly
Token promptly
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.130
Ċ
TokenĊ
Feature activation+0.000
Roberts
TokenRoberts
Feature activation+0.000
told
Token told
Feature activation+0.000
the
Token the
Feature activation+0.000
Senate
Token Senate
Feature activation+0.000
brother
Token brother
Feature activation+0.000
'
Token'
Feature activation+0.000
concerns
Token concerns
Feature activation+0.000
Read
Token Read
Feature activation+6.809
more
Token more
Feature activation+2.399
Ċ
TokenĊ
Feature activation+7.965
Ċ
TokenĊ
Feature activation+0.000
Nation
TokenNation
Feature activation+0.000
wide
Tokenwide
Feature activation+0.000
,
Token,
Feature activation+0.000
however
Token however
Feature activation+0.000
en
Tokenen
Feature activation+0.000
into
Token into
Feature activation+0.000
Spurs
Token Spurs
Feature activation+0.000
midfield
Token midfield
Feature activation+0.000
role
Token role
Feature activation+0.000
Read
Token Read
Feature activation+7.914
more
Token more
Feature activation+1.732
Ċ
TokenĊ
Feature activation+8.179
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
a
Token a
Feature activation+0.000
softer
Token softer
Feature activation+0.000
Brexit
Token Brexit
Feature activation+0.000
Read
Token Read
Feature activation+7.663
more
Token more
Feature activation+2.359
Ċ
TokenĊ
Feature activation+7.885
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
ruling
Token ruling
Feature activation+0.000
follows
Token follows
Feature activation+0.000
complaints
Token complaints
Feature activation+0.000
Hanson
Token Hanson
Feature activation+0.000
got
Token got
Feature activation+0.000
there
Token there
Feature activation+0.000
first
Token first
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.678
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Senate
Token Senate
Feature activation+0.000
president
Token president
Feature activation+0.000
,
Token,
Feature activation+0.000
May
Token May
Feature activation+0.000
towards
Token towards
Feature activation+0.000
a
Token a
Feature activation+0.000
softer
Token softer
Feature activation+0.000
Brexit
Token Brexit
Feature activation+0.000
Read
Token Read
Feature activation+7.663
more
Token more
Feature activation+2.359
Ċ
TokenĊ
Feature activation+7.885
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
ruling
Token ruling
Feature activation+0.000
child
Token child
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
operation
Token operation
Feature activation+0.000
Read
Token Read
Feature activation+6.042
more
Token more
Feature activation+3.203
Ċ
TokenĊ
Feature activation+7.363
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
police
Token police
Feature activation+0.000
chief
Token chief
Feature activation+0.000
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
security
Token security
Feature activation+0.000
rail
Token rail
Feature activation+0.000
ings
Tokenings
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.302
Ċ
TokenĊ
Feature activation+0.000
These
TokenThese
Feature activation+0.000
banners
Token banners
Feature activation+0.000
,
Token,
Feature activation+0.000
welcoming
Token welcoming
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
in
Token in
Feature activation+0.000
pictures
Token pictures
Feature activation+0.000
Read
Token Read
Feature activation+5.153
more
Token more
Feature activation+1.710
Ċ
TokenĊ
Feature activation+7.294
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
significant
Token significant
Feature activation+0.000
if
Token if
Feature activation+0.000
covert
Token covert
Feature activation+0.000
in
Token in
Feature activation+0.000
January
Token January
Feature activation+0.000
window
Token window
Feature activation+0.000
Read
Token Read
Feature activation+6.389
more
Token more
Feature activation+2.906
Ċ
TokenĊ
Feature activation+7.257
Ċ
TokenĊ
Feature activation+0.000
But
TokenBut
Feature activation+0.000
it
Token it
Feature activation+0.000
does
Token does
Feature activation+0.000
seem
Token seem
Feature activation+0.000
big
Token big
Feature activation+0.000
-
Token-
Feature activation+0.000
money
Tokenmoney
Feature activation+0.000
moves
Token moves
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.227
Ċ
TokenĊ
Feature activation+0.000
City
TokenCity
Feature activation+0.000
were
Token were
Feature activation+0.000
genuinely
Token genuinely
Feature activation+0.000
keen
Token keen
Feature activation+0.000
'
Token '
Feature activation+0.000
big
Tokenbig
Feature activation+0.000
brother
Token brother
Feature activation+0.000
'
Token'
Feature activation+0.000
concerns
Token concerns
Feature activation+0.000
Read
Token Read
Feature activation+6.809
more
Token more
Feature activation+2.399
Ċ
TokenĊ
Feature activation+7.965
Ċ
TokenĊ
Feature activation+0.000
Nation
TokenNation
Feature activation+0.000
wide
Tokenwide
Feature activation+0.000
at
Token at
Feature activation+0.000
Arsenal
Token Arsenal
Feature activation+0.000
in
Token in
Feature activation+0.000
January
Token January
Feature activation+0.000
window
Token window
Feature activation+0.000
Read
Token Read
Feature activation+6.389
more
Token more
Feature activation+2.906
Ċ
TokenĊ
Feature activation+7.257
Ċ
TokenĊ
Feature activation+0.000
But
TokenBut
Feature activation+0.000
it
Token it
Feature activation+0.000
Philadelphia
Token Philadelphia
Feature activation+0.000
correctional
Token correctional
Feature activation+0.000
facility
Token facility
Feature activation+0.000
-
Token -
Feature activation+0.000
live
Token live
Feature activation+0.000
Read
Token Read
Feature activation+6.346
more
Token more
Feature activation+2.805
Ċ
TokenĊ
Feature activation+7.236
Ċ
TokenĊ
Feature activation+0.000
Fr
TokenFr
Feature activation+0.000
Feder
Token Feder
Feature activation+0.000
Earth
Token Earth
Feature activation+0.000
rotation
Token rotation
Feature activation+0.000
slows
Token slows
Feature activation+0.000
Read
Token Read
Feature activation+5.082
more
Token more
Feature activation+1.110
Ċ
TokenĊ
Feature activation+6.152
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
US
Token US
Feature activation+0.000
Geological
Token Geological
Feature activation+0.000
Survey
Token Survey
Feature activation+0.000
wide
Tokenwide
Feature activation+0.000
online
Token online
Feature activation+0.000
child
Token child
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
operation
Token operation
Feature activation+0.000
Read
Token Read
Feature activation+6.042
more
Token more
Feature activation+3.203
Ċ
TokenĊ
Feature activation+7.363
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
police
Token police
Feature activation+0.000
matched
Token matched
Feature activation+0.000
by
Token by
Feature activation+0.000
actions
Token actions
Feature activation+0.000
Read
Token Read
Feature activation+2.129
more
Token more
Feature activation+1.431
Ċ
TokenĊ
Feature activation+5.775
Ċ
TokenĊ
Feature activation+0.000
But
TokenBut
Feature activation+0.000
Arsenal
Token Arsenal
Feature activation+0.000
are
Token are
Feature activation+0.000
making
Token making
Feature activation+0.000

Top DFA by src position
MAX = 10.183

option
Token option
Feature activation-0.007
Top resid features:
,
Token,
Feature activation-0.012
Top resid features:
says
Token says
Feature activation-0.014
Top resid features:
think
Token think
Feature activation-0.002
Top resid features:
tank
Tokentank
Feature activation-0.012
Top resid features:
Read
Token Read
Feature activation+10.183
Top resid features:
more
Token more
Feature activation-0.043
Top resid features:
Ċ
TokenĊ
Feature activation+0.098
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+0.000
Top resid features:
pants
Tokenpants
Feature activation-0.007
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.004
Top resid features:
as
Token as
Feature activation+0.004
Top resid features:
it
Token it
Feature activation-0.004
Top resid features:
happened
Token happened
Feature activation-0.001
Top resid features:
Read
Token Read
Feature activation+10.049
Top resid features:
more
Token more
Feature activation-0.017
Top resid features:
Ċ
TokenĊ
Feature activation+0.120
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Di
TokenDi
Feature activation+0.000
Top resid features:
Nat
Token Nat
Feature activation+0.000
Top resid features:
en
Tokenen
Feature activation+0.009
Top resid features:
into
Token into
Feature activation+0.003
Top resid features:
Spurs
Token Spurs
Feature activation+0.002
Top resid features:
midfield
Token midfield
Feature activation+0.004
Top resid features:
role
Token role
Feature activation+0.004
Top resid features:
Read
Token Read
Feature activation+9.822
Top resid features:
more
Token more
Feature activation-0.017
Top resid features:
Ċ
TokenĊ
Feature activation+0.125
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+0.000
Top resid features:
pants
Tokenpants
Feature activation-0.003
Top resid features:
âĢĵ
Token âĢĵ
Feature activation-0.001
Top resid features:
as
Token as
Feature activation+0.001
Top resid features:
it
Token it
Feature activation-0.004
Top resid features:
happened
Token happened
Feature activation-0.001
Top resid features:
Read
Token Read
Feature activation+10.100
Top resid features:
more
Token more
Feature activation-0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.052
Top resid features:
Ċ
TokenĊ
Feature activation+0.122
Top resid features:
Di
TokenDi
Feature activation+0.018
Top resid features:
Nat
Token Nat
Feature activation+0.004
Top resid features:
'
Token '
Feature activation-0.007
Top resid features:
big
Tokenbig
Feature activation-0.003
Top resid features:
brother
Token brother
Feature activation-0.003
Top resid features:
'
Token'
Feature activation-0.001
Top resid features:
concerns
Token concerns
Feature activation+0.003
Top resid features:
Read
Token Read
Feature activation+9.761
Top resid features:
more
Token more
Feature activation-0.031
Top resid features:
Ċ
TokenĊ
Feature activation+0.069
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Nation
TokenNation
Feature activation+0.000
Top resid features:
wide
Tokenwide
Feature activation+0.000
Top resid features:
en
Tokenen
Feature activation+0.006
Top resid features:
into
Token into
Feature activation-0.009
Top resid features:
Spurs
Token Spurs
Feature activation+0.000
Top resid features:
midfield
Token midfield
Feature activation+0.019
Top resid features:
role
Token role
Feature activation+0.071
Top resid features:
Read
Token Read
Feature activation+9.429
Top resid features:
more
Token more
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
ľ
Tokenľ
Feature activation+0.000
Top resid features:
May
Token May
Feature activation+0.059
Top resid features:
towards
Token towards
Feature activation-0.021
Top resid features:
a
Token a
Feature activation-0.024
Top resid features:
softer
Token softer
Feature activation-0.019
Top resid features:
Brexit
Token Brexit
Feature activation-0.002
Top resid features:
Read
Token Read
Feature activation+9.473
Top resid features:
more
Token more
Feature activation-0.022
Top resid features:
Ċ
TokenĊ
Feature activation+0.126
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
ruling
Token ruling
Feature activation+0.000
Top resid features:
pants
Tokenpants
Feature activation-0.008
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.002
Top resid features:
as
Token as
Feature activation+0.002
Top resid features:
it
Token it
Feature activation-0.010
Top resid features:
happened
Token happened
Feature activation-0.000
Top resid features:
Read
Token Read
Feature activation+9.637
Top resid features:
more
Token more
Feature activation-0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.084
Top resid features:
Ċ
TokenĊ
Feature activation+0.204
Top resid features:
Di
TokenDi
Feature activation+0.028
Top resid features:
Nat
Token Nat
Feature activation+0.013
Top resid features:
May
Token May
Feature activation+0.017
Top resid features:
towards
Token towards
Feature activation-0.001
Top resid features:
a
Token a
Feature activation-0.002
Top resid features:
softer
Token softer
Feature activation+0.014
Top resid features:
Brexit
Token Brexit
Feature activation+0.064
Top resid features:
Read
Token Read
Feature activation+9.183
Top resid features:
more
Token more
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
ruling
Token ruling
Feature activation+0.000
Top resid features:
wide
Tokenwide
Feature activation+0.003
Top resid features:
online
Token online
Feature activation+0.004
Top resid features:
child
Token child
Feature activation+0.002
Top resid features:
abuse
Token abuse
Feature activation+0.017
Top resid features:
operation
Token operation
Feature activation+0.005
Top resid features:
Read
Token Read
Feature activation+9.236
Top resid features:
more
Token more
Feature activation-0.021
Top resid features:
Ċ
TokenĊ
Feature activation+0.196
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
police
Token police
Feature activation+0.000
Top resid features:
in
Token in
Feature activation-0.004
Top resid features:
London
Token London
Feature activation+0.013
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.002
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
pictures
Token pictures
Feature activation-0.007
Top resid features:
Read
Token Read
Feature activation+9.056
Top resid features:
more
Token more
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.029
Top resid features:
Ċ
TokenĊ
Feature activation+0.133
Top resid features:
The
TokenThe
Feature activation+0.005
Top resid features:
significant
Token significant
Feature activation+0.001
Top resid features:
in
Token in
Feature activation-0.008
Top resid features:
London
Token London
Feature activation+0.053
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.009
Top resid features:
in
Token in
Feature activation-0.003
Top resid features:
pictures
Token pictures
Feature activation-0.003
Top resid features:
Read
Token Read
Feature activation+9.127
Top resid features:
more
Token more
Feature activation-0.017
Top resid features:
Ċ
TokenĊ
Feature activation+0.156
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
significant
Token significant
Feature activation+0.000
Top resid features:
at
Token at
Feature activation-0.015
Top resid features:
Arsenal
Token Arsenal
Feature activation-0.071
Top resid features:
in
Token in
Feature activation-0.017
Top resid features:
January
Token January
Feature activation-0.023
Top resid features:
window
Token window
Feature activation-0.025
Top resid features:
Read
Token Read
Feature activation+9.552
Top resid features:
more
Token more
Feature activation-0.035
Top resid features:
Ċ
TokenĊ
Feature activation+0.081
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
But
TokenBut
Feature activation+0.000
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
at
Token at
Feature activation-0.005
Top resid features:
Arsenal
Token Arsenal
Feature activation-0.023
Top resid features:
in
Token in
Feature activation-0.003
Top resid features:
January
Token January
Feature activation-0.002
Top resid features:
window
Token window
Feature activation-0.001
Top resid features:
Read
Token Read
Feature activation+9.703
Top resid features:
more
Token more
Feature activation-0.010
Top resid features:
Ċ
TokenĊ
Feature activation-0.006
Top resid features:
Ċ
TokenĊ
Feature activation+0.130
Top resid features:
But
TokenBut
Feature activation+0.016
Top resid features:
it
Token it
Feature activation+0.003
Top resid features:
'
Token '
Feature activation-0.009
Top resid features:
big
Tokenbig
Feature activation-0.023
Top resid features:
brother
Token brother
Feature activation-0.002
Top resid features:
'
Token'
Feature activation-0.007
Top resid features:
concerns
Token concerns
Feature activation+0.016
Top resid features:
Read
Token Read
Feature activation+8.467
Top resid features:
more
Token more
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Nation
TokenNation
Feature activation+0.000
Top resid features:
wide
Tokenwide
Feature activation+0.000
Top resid features:
at
Token at
Feature activation-0.001
Top resid features:
Arsenal
Token Arsenal
Feature activation-0.011
Top resid features:
in
Token in
Feature activation-0.004
Top resid features:
January
Token January
Feature activation-0.013
Top resid features:
window
Token window
Feature activation-0.002
Top resid features:
Read
Token Read
Feature activation+8.373
Top resid features:
more
Token more
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
But
TokenBut
Feature activation+0.000
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
Philadelphia
Token Philadelphia
Feature activation+0.010
Top resid features:
correctional
Token correctional
Feature activation-0.007
Top resid features:
facility
Token facility
Feature activation-0.003
Top resid features:
-
Token -
Feature activation+0.003
Top resid features:
live
Token live
Feature activation+0.127
Top resid features:
Read
Token Read
Feature activation+8.080
Top resid features:
more
Token more
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Fr
TokenFr
Feature activation+0.000
Top resid features:
Feder
Token Feder
Feature activation+0.000
Top resid features:
2018
Token 2018
Feature activation-0.002
Top resid features:
as
Token as
Feature activation+0.004
Top resid features:
Earth
Token Earth
Feature activation-0.008
Top resid features:
rotation
Token rotation
Feature activation-0.006
Top resid features:
slows
Token slows
Feature activation+0.003
Top resid features:
Read
Token Read
Feature activation+8.322
Top resid features:
more
Token more
Feature activation-0.030
Top resid features:
Ċ
TokenĊ
Feature activation+0.236
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
US
Token US
Feature activation+0.000
Top resid features:
wide
Tokenwide
Feature activation+0.005
Top resid features:
online
Token online
Feature activation+0.002
Top resid features:
child
Token child
Feature activation+0.005
Top resid features:
abuse
Token abuse
Feature activation+0.138
Top resid features:
operation
Token operation
Feature activation+0.045
Top resid features:
Read
Token Read
Feature activation+7.714
Top resid features:
more
Token more
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
police
Token police
Feature activation+0.000
Top resid features:
not
Token not
Feature activation+0.004
Top resid features:
yet
Token yet
Feature activation-0.001
Top resid features:
matched
Token matched
Feature activation-0.001
Top resid features:
by
Token by
Feature activation-0.008
Top resid features:
actions
Token actions
Feature activation-0.015
Top resid features:
Read
Token Read
Feature activation+7.531
Top resid features:
more
Token more
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.395
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
But
TokenBut
Feature activation+0.000
Top resid features:
Arsenal
Token Arsenal
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.04

Head 1: 0.01

Head 2: 0.59

Head 3: 0.03

Head 4: 0.03

Head 5: 0.05

Head 6: 0.03

Head 7: 0.04

Head 8: 0.04

Head 9: 0.04

Head 10: 0.05

Head 11: 0.04

Positive logits

Nato6.76

6.16

­5.78

apologised5.23

maths5.16

NHS5.09

emphas5.06

Nasa4.97

ageing4.93

Miliband4.92

programme4.89

criticised4.89

£4.85

Ukip4.84

organising4.78

Facebook4.73

apologise4.59

programmes4.55

behavioural4.52

flavours4.51

Negative logits

,)-4.82

.---4.33

.......-4.16

}}}-4.10

....-4.00

utilize-4.00

......-3.97

----3.94

....-3.92

......-3.90

.....-3.87

-3.87

utilized-3.82

alot-3.72

bilt-3.72

colored-3.71

,...-3.61

utilizing-3.61

"...-3.57

----3.55

INTERVAL 7.655 - 8.506
CONTAINS 0.001%

en
Tokenen
Feature activation+0.000
into
Token into
Feature activation+0.000
Spurs
Token Spurs
Feature activation+0.000
midfield
Token midfield
Feature activation+0.000
role
Token role
Feature activation+0.000
Read
Token Read
Feature activation+7.914
more
Token more
Feature activation+1.732
Ċ
TokenĊ
Feature activation+8.179
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Spurs
Token Spurs
Feature activation+0.000
midfield
Token midfield
Feature activation+0.000
role
Token role
Feature activation+0.000
Read
Token Read
Feature activation+7.914
more
Token more
Feature activation+1.732
Ċ
TokenĊ
Feature activation+8.179
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Following
TokenFollowing
Feature activation+0.000
last
Token last
Feature activation+0.000
settle
Token settle
Feature activation+0.000
the
Token the
Feature activation+0.000
matter
Token matter
Feature activation+0.000
promptly
Token promptly
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.130
Ċ
TokenĊ
Feature activation+0.000
Roberts
TokenRoberts
Feature activation+0.000
told
Token told
Feature activation+0.000
the
Token the
Feature activation+0.000
Senate
Token Senate
Feature activation+0.000
May
Token May
Feature activation+0.000
towards
Token towards
Feature activation+0.000
a
Token a
Feature activation+0.000
softer
Token softer
Feature activation+0.000
Brexit
Token Brexit
Feature activation+0.000
Read
Token Read
Feature activation+7.663
more
Token more
Feature activation+2.359
Ċ
TokenĊ
Feature activation+7.885
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
ruling
Token ruling
Feature activation+0.000
says
Token says
Feature activation+0.000
think
Token think
Feature activation+0.000
tank
Tokentank
Feature activation+0.000
Read
Token Read
Feature activation+4.479
more
Token more
Feature activation+1.863
Ċ
TokenĊ
Feature activation+8.506
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
simplistic
Token simplistic
Feature activation+0.000

INTERVAL 6.805 - 7.655
CONTAINS 0.001%

âĢĵ
Token âĢĵ
Feature activation+0.000
in
Token in
Feature activation+0.000
pictures
Token pictures
Feature activation+0.000
Read
Token Read
Feature activation+5.153
more
Token more
Feature activation+1.710
Ċ
TokenĊ
Feature activation+7.294
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
significant
Token significant
Feature activation+0.000
if
Token if
Feature activation+0.000
covert
Token covert
Feature activation+0.000
child
Token child
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
operation
Token operation
Feature activation+0.000
Read
Token Read
Feature activation+6.042
more
Token more
Feature activation+3.203
Ċ
TokenĊ
Feature activation+7.363
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
police
Token police
Feature activation+0.000
chief
Token chief
Feature activation+0.000
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
security
Token security
Feature activation+0.000
rail
Token rail
Feature activation+0.000
ings
Tokenings
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.302
Ċ
TokenĊ
Feature activation+0.000
These
TokenThese
Feature activation+0.000
banners
Token banners
Feature activation+0.000
,
Token,
Feature activation+0.000
welcoming
Token welcoming
Feature activation+0.000
in
Token in
Feature activation+0.000
January
Token January
Feature activation+0.000
window
Token window
Feature activation+0.000
Read
Token Read
Feature activation+6.389
more
Token more
Feature activation+2.906
Ċ
TokenĊ
Feature activation+7.257
Ċ
TokenĊ
Feature activation+0.000
But
TokenBut
Feature activation+0.000
it
Token it
Feature activation+0.000
does
Token does
Feature activation+0.000
seem
Token seem
Feature activation+0.000
big
Token big
Feature activation+0.000
-
Token-
Feature activation+0.000
money
Tokenmoney
Feature activation+0.000
moves
Token moves
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.227
Ċ
TokenĊ
Feature activation+0.000
City
TokenCity
Feature activation+0.000
were
Token were
Feature activation+0.000
genuinely
Token genuinely
Feature activation+0.000
keen
Token keen
Feature activation+0.000

INTERVAL 5.954 - 6.805
CONTAINS 0.000%

wide
Tokenwide
Feature activation+0.000
online
Token online
Feature activation+0.000
child
Token child
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
operation
Token operation
Feature activation+0.000
Read
Token Read
Feature activation+6.042
more
Token more
Feature activation+3.203
Ċ
TokenĊ
Feature activation+7.363
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
police
Token police
Feature activation+0.000
Earth
Token Earth
Feature activation+0.000
rotation
Token rotation
Feature activation+0.000
slows
Token slows
Feature activation+0.000
Read
Token Read
Feature activation+5.082
more
Token more
Feature activation+1.110
Ċ
TokenĊ
Feature activation+6.152
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
US
Token US
Feature activation+0.000
Geological
Token Geological
Feature activation+0.000
Survey
Token Survey
Feature activation+0.000
Philadelphia
Token Philadelphia
Feature activation+0.000
correctional
Token correctional
Feature activation+0.000
facility
Token facility
Feature activation+0.000
-
Token -
Feature activation+0.000
live
Token live
Feature activation+0.000
Read
Token Read
Feature activation+6.346
more
Token more
Feature activation+2.805
Ċ
TokenĊ
Feature activation+7.236
Ċ
TokenĊ
Feature activation+0.000
Fr
TokenFr
Feature activation+0.000
Feder
Token Feder
Feature activation+0.000
at
Token at
Feature activation+0.000
Arsenal
Token Arsenal
Feature activation+0.000
in
Token in
Feature activation+0.000
January
Token January
Feature activation+0.000
window
Token window
Feature activation+0.000
Read
Token Read
Feature activation+6.389
more
Token more
Feature activation+2.906
Ċ
TokenĊ
Feature activation+7.257
Ċ
TokenĊ
Feature activation+0.000
But
TokenBut
Feature activation+0.000
it
Token it
Feature activation+0.000

INTERVAL 5.103 - 5.954
CONTAINS 0.000%

in
Token in
Feature activation+0.000
London
Token London
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
in
Token in
Feature activation+0.000
pictures
Token pictures
Feature activation+0.000
Read
Token Read
Feature activation+5.153
more
Token more
Feature activation+1.710
Ċ
TokenĊ
Feature activation+7.294
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
significant
Token significant
Feature activation+0.000
hard
Token hard
Feature activation+0.000
to
Token to
Feature activation+0.000
govern
Token govern
Feature activation+0.000
Read
Token Read
Feature activation+1.652
more
Token more
Feature activation+2.217
Ċ
TokenĊ
Feature activation+5.374
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
data
Token data
Feature activation+0.000
was
Token was
Feature activation+0.000
posted
Token posted
Feature activation+0.000
matched
Token matched
Feature activation+0.000
by
Token by
Feature activation+0.000
actions
Token actions
Feature activation+0.000
Read
Token Read
Feature activation+2.129
more
Token more
Feature activation+1.431
Ċ
TokenĊ
Feature activation+5.775
Ċ
TokenĊ
Feature activation+0.000
But
TokenBut
Feature activation+0.000
Arsenal
Token Arsenal
Feature activation+0.000
are
Token are
Feature activation+0.000
making
Token making
Feature activation+0.000
pants
Tokenpants
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
as
Token as
Feature activation+0.000
it
Token it
Feature activation+0.000
happened
Token happened
Feature activation+0.000
Read
Token Read
Feature activation+5.405
more
Token more
Feature activation+0.441
Ċ
TokenĊ
Feature activation+8.383
Ċ
TokenĊ
Feature activation+0.000
Di
TokenDi
Feature activation+0.000
Nat
Token Nat
Feature activation+0.000

INTERVAL 4.253 - 5.103
CONTAINS 0.000%

option
Token option
Feature activation+0.000
,
Token,
Feature activation+0.000
says
Token says
Feature activation+0.000
think
Token think
Feature activation+0.000
tank
Tokentank
Feature activation+0.000
Read
Token Read
Feature activation+4.479
more
Token more
Feature activation+1.863
Ċ
TokenĊ
Feature activation+8.506
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
2018
Token 2018
Feature activation+0.000
as
Token as
Feature activation+0.000
Earth
Token Earth
Feature activation+0.000
rotation
Token rotation
Feature activation+0.000
slows
Token slows
Feature activation+0.000
Read
Token Read
Feature activation+5.082
more
Token more
Feature activation+1.110
Ċ
TokenĊ
Feature activation+6.152
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
US
Token US
Feature activation+0.000

INTERVAL 3.402 - 4.253
CONTAINS 0.000%

and
Token and
Feature activation+0.000
disinformation
Token disinformation
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.771
Ċ
TokenĊ
Feature activation+0.000
Int
TokenInt
Feature activation+0.000
elligence
Tokenelligence
Feature activation+0.000
agencies
Token agencies
Feature activation+0.000
in
Token in
Feature activation+0.000
players
Token players
Feature activation+0.000
until
Token until
Feature activation+0.000
January
Token January
Feature activation+0.000
2018
Token 2018
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.983
Ċ
TokenĊ
Feature activation+0.000
Arsenal
TokenArsenal
Feature activation+0.000
value
Token value
Feature activation+0.000
Lac
Token Lac
Feature activation+0.000
az
Tokenaz
Feature activation+0.000

INTERVAL 2.552 - 3.402
CONTAINS 0.001%

online
Token online
Feature activation+0.000
child
Token child
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
operation
Token operation
Feature activation+0.000
Read
Token Read
Feature activation+6.042
more
Token more
Feature activation+3.203
Ċ
TokenĊ
Feature activation+7.363
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
police
Token police
Feature activation+0.000
chief
Token chief
Feature activation+0.000
should
Token should
Feature activation+0.000
be
Token be
Feature activation+0.000
doing
Token doing
Feature activation+0.000
Read
Token Read
Feature activation+0.209
more
Token more
Feature activation+0.307
Ċ
TokenĊ
Feature activation+3.248
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
This
TokenThis
Feature activation+0.000
is
Token is
Feature activation+0.000
S
TokenS
Feature activation+0.000
AA
TokenAA
Feature activation+0.000
)
Token)
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.587
Ċ
TokenĊ
Feature activation+0.000
D
TokenD
Feature activation+0.000
ix
Tokenix
Feature activation+0.000
it
Tokenit
Feature activation+0.000
is
Token is
Feature activation+0.000
Arsenal
Token Arsenal
Feature activation+0.000
in
Token in
Feature activation+0.000
January
Token January
Feature activation+0.000
window
Token window
Feature activation+0.000
Read
Token Read
Feature activation+6.389
more
Token more
Feature activation+2.906
Ċ
TokenĊ
Feature activation+7.257
Ċ
TokenĊ
Feature activation+0.000
But
TokenBut
Feature activation+0.000
it
Token it
Feature activation+0.000
does
Token does
Feature activation+0.000

INTERVAL 1.701 - 2.552
CONTAINS 0.001%

big
Tokenbig
Feature activation+0.000
brother
Token brother
Feature activation+0.000
'
Token'
Feature activation+0.000
concerns
Token concerns
Feature activation+0.000
Read
Token Read
Feature activation+6.809
more
Token more
Feature activation+2.399
Ċ
TokenĊ
Feature activation+7.965
Ċ
TokenĊ
Feature activation+0.000
Nation
TokenNation
Feature activation+0.000
wide
Tokenwide
Feature activation+0.000
,
Token,
Feature activation+0.000
into
Token into
Feature activation+0.000
Spurs
Token Spurs
Feature activation+0.000
midfield
Token midfield
Feature activation+0.000
role
Token role
Feature activation+0.000
Read
Token Read
Feature activation+7.914
more
Token more
Feature activation+1.732
Ċ
TokenĊ
Feature activation+8.179
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Following
TokenFollowing
Feature activation+0.000
London
Token London
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
in
Token in
Feature activation+0.000
pictures
Token pictures
Feature activation+0.000
Read
Token Read
Feature activation+5.153
more
Token more
Feature activation+1.710
Ċ
TokenĊ
Feature activation+7.294
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
significant
Token significant
Feature activation+0.000
if
Token if
Feature activation+0.000
,
Token,
Feature activation+0.000
says
Token says
Feature activation+0.000
think
Token think
Feature activation+0.000
tank
Tokentank
Feature activation+0.000
Read
Token Read
Feature activation+4.479
more
Token more
Feature activation+1.863
Ċ
TokenĊ
Feature activation+8.506
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
towards
Token towards
Feature activation+0.000
a
Token a
Feature activation+0.000
softer
Token softer
Feature activation+0.000
Brexit
Token Brexit
Feature activation+0.000
Read
Token Read
Feature activation+7.663
more
Token more
Feature activation+2.359
Ċ
TokenĊ
Feature activation+7.885
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
ruling
Token ruling
Feature activation+0.000
follows
Token follows
Feature activation+0.000

INTERVAL 0.851 - 1.701
CONTAINS 0.001%

find
Token find
Feature activation+0.000
it
Token it
Feature activation+0.000
hard
Token hard
Feature activation+0.000
to
Token to
Feature activation+0.000
govern
Token govern
Feature activation+0.000
Read
Token Read
Feature activation+1.652
more
Token more
Feature activation+2.217
Ċ
TokenĊ
Feature activation+5.374
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
data
Token data
Feature activation+0.000
as
Token as
Feature activation+0.000
Earth
Token Earth
Feature activation+0.000
rotation
Token rotation
Feature activation+0.000
slows
Token slows
Feature activation+0.000
Read
Token Read
Feature activation+5.082
more
Token more
Feature activation+1.110
Ċ
TokenĊ
Feature activation+6.152
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
US
Token US
Feature activation+0.000
Geological
Token Geological
Feature activation+0.000
traffic
Token traffic
Feature activation+0.000
and
Token and
Feature activation+0.000
reducing
Token reducing
Feature activation+0.000
carbon
Token carbon
Feature activation+0.000
emissions
Token emissions
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.211
Ċ
TokenĊ
Feature activation+0.000
Energy
TokenEnergy
Feature activation+0.000
-
Token-
Feature activation+0.000
efficient
Tokenefficient
Feature activation+0.000
housing
Token housing
Feature activation+0.000
/
Token/
Feature activation+0.000
Rex
TokenRex
Feature activation+0.000
/
Token/
Feature activation+0.000
Sh
TokenSh
Feature activation+0.000
utterstock
Tokenutterstock
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.233
Ċ
TokenĊ
Feature activation+0.000
Chris
TokenChris
Feature activation+0.000
Rob
Token Rob
Feature activation+0.000
shaw
Tokenshaw
Feature activation+0.000
had
Token had
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Day
Token Day
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.154
Ċ
TokenĊ
Feature activation+0.000
D
TokenD
Feature activation+0.000
ix
Tokenix
Feature activation+0.000
it
Tokenit
Feature activation+0.000
,
Token,
Feature activation+0.000

INTERVAL 0.000 - 0.851
CONTAINS 99.995%

changes
Token changes
Feature activation+0.000
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
have
Token have
Feature activation+0.000
to
Token to
Feature activation+0.000
re
Token re
Feature activation+0.000
-
Token-
Feature activation+0.000
cert
Tokencert
Feature activation+0.000
ify
Tokenify
Feature activation+0.000
it
Token it
Feature activation+0.000
.
Token.
Feature activation+0.000
Online
Token Online
Feature activation+0.000
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
were
Token were
Feature activation+0.000
all
Token all
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
real
Token real
Feature activation+0.000
izations
Tokenizations
Feature activation+0.000
;
Token;
Feature activation+0.000
to
Token to
Feature activation+0.000
wit
Token wit
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
Apple
Token Apple
Feature activation+0.000
Power
Token Power
Feature activation+0.000
Book
TokenBook
Feature activation+0.000
logo
Token logo
Feature activation+0.000
houses
Token houses
Feature activation+0.000
more
Token more
Feature activation+0.000
children
Token children
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
proportion
Token proportion
Feature activation+0.000
than
Token than
Feature activation+0.000
social
Token social
Feature activation+0.000
housing
Token housing
Feature activation+0.000
does
Token does
Feature activation+0.000
.
Token.
Feature activation+0.000
theories
Token theories
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
very
Token very
Feature activation+0.000
Tolkien
Token Tolkien
Feature activation+0.000
ian
Tokenian
Feature activation+0.000
fashion
Token fashion
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 2 in H8.2: (feature 10534)

TOP ACTIVATIONS
MAX = 8.040

free
Token free
Feature activation+0.000
republic
Token republic
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.040
Ċ
TokenĊ
Feature activation+1.713
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.300
Ċ
TokenĊ
Feature activation+0.281
In
TokenIn
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.282
Democratic
Token Democratic
Feature activation+0.320
primary
Token primary
Feature activation+0.000
.
Token.
Feature activation+2.611
Ċ
TokenĊ
Feature activation+7.945
Ċ
TokenĊ
Feature activation+4.688
Some
TokenSome
Feature activation+0.496
Sanders
Token Sanders
Feature activation+0.846
fans
Token fans
Feature activation+0.000
are
Token are
Feature activation+0.083
warning
Token warning
Feature activation+0.000
former
Token former
Feature activation+0.000
FBI
Token FBI
Feature activation+0.499
Director
Token Director
Feature activation+0.000
James
Token James
Feature activation+0.000
Comey
Token Comey
Feature activation+7.812
about
Token about
Feature activation+0.000
leaking
Token leaking
Feature activation+0.000
information
Token information
Feature activation+0.000
.
Token.
Feature activation+5.481
Ċ
TokenĊ
Feature activation+7.033
of
Token of
Feature activation+0.000
Massachusetts
Token Massachusetts
Feature activation+0.000
,
Token,
Feature activation+0.000
where
Token where
Feature activation+0.000
Hillary
Token Hillary
Feature activation+2.971
Clinton
Token Clinton
Feature activation+7.564
Hillary
Token Hillary
Feature activation+0.000
Diane
Token Diane
Feature activation+0.000
Rodham
Token Rodham
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
Sanders
TokenSanders
Feature activation+0.000
in
Token in
Feature activation+0.151
the
Token the
Feature activation+1.542
coming
Token coming
Feature activation+0.374
months
Token months
Feature activation+0.901
.
Token.
Feature activation+4.433
Ċ
TokenĊ
Feature activation+7.481
Ċ
TokenĊ
Feature activation+4.252
While
TokenWhile
Feature activation+0.181
Republicans
Token Republicans
Feature activation+0.011
support
Token support
Feature activation+0.000
the
Token the
Feature activation+0.073
No
Token No
Feature activation+1.613
.
Token.
Feature activation+0.000
2
Token 2
Feature activation+0.000
post
Token post
Feature activation+0.000
.
Token.
Feature activation+1.449
Ċ
TokenĊ
Feature activation+7.237
Ċ
TokenĊ
Feature activation+3.916
But
TokenBut
Feature activation+0.000
the
Token the
Feature activation+0.000
episode
Token episode
Feature activation+0.000
proved
Token proved
Feature activation+0.000
Oversight
Token Oversight
Feature activation+0.000
Committee
Token Committee
Feature activation+0.274
ranking
Token ranking
Feature activation+0.000
member
Token member
Feature activation+0.048
Elijah
Token Elijah
Feature activation+0.000
Cummings
Token Cummings
Feature activation+7.093
(
Token (
Feature activation+2.763
D
TokenD
Feature activation+0.000
-
Token-
Feature activation+4.488
M
TokenM
Feature activation+0.000
d
Tokend
Feature activation+0.000
Comey
Token Comey
Feature activation+7.812
about
Token about
Feature activation+0.000
leaking
Token leaking
Feature activation+0.000
information
Token information
Feature activation+0.000
.
Token.
Feature activation+5.481
Ċ
TokenĊ
Feature activation+7.033
Ċ
TokenĊ
Feature activation+0.884
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
All
TokenAll
Feature activation+0.000
of
Token of
Feature activation+0.000
then
Token then
Feature activation+1.102
-
Token-
Feature activation+0.000
Rep
TokenRep
Feature activation+0.000
,
Token,
Feature activation+0.000
Tom
Token Tom
Feature activation+0.000
Price
Token Price
Feature activation+6.894
(
Token (
Feature activation+4.545
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+4.212
Ga
TokenGa
Feature activation+2.674
.)
Token.)
Feature activation+0.000
.
Token.
Feature activation+0.000
Y
TokenY
Feature activation+5.065
.)
Token.)
Feature activation+0.000
and
Token and
Feature activation+0.000
Bill
Token Bill
Feature activation+0.000
Nelson
Token Nelson
Feature activation+6.838
Clarence
Token Clarence
Feature activation+0.000
(
Token (
Feature activation+0.000
Bill
TokenBill
Feature activation+0.000
)
Token)
Feature activation+0.000
William
Token William
Feature activation+0.000
Republican
Token Republican
Feature activation+0.000
Senate
Token Senate
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
arm
Token arm
Feature activation+0.000
.
Token.
Feature activation+1.909
Ċ
TokenĊ
Feature activation+6.772
Ċ
TokenĊ
Feature activation+1.877
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Nobody
TokenNobody
Feature activation+0.000
wants
Token wants
Feature activation+0.000
,
Token,
Feature activation+0.687
Majority
Token Majority
Feature activation+0.000
Whip
Token Whip
Feature activation+0.000
Steve
Token Steve
Feature activation+0.000
Scal
Token Scal
Feature activation+0.000
ise
Tokenise
Feature activation+6.761
(
Token (
Feature activation+4.665
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+5.523
La
TokenLa
Feature activation+0.000
.)
Token.)
Feature activation+0.894
called
Token called
Feature activation+0.000
for
Token for
Feature activation+0.495
rec
Token rec
Feature activation+0.000
usal
Tokenusal
Feature activation+0.000
.
Token.
Feature activation+2.912
Ċ
TokenĊ
Feature activation+6.744
Ċ
TokenĊ
Feature activation+1.819
S
TokenS
Feature activation+0.000
essions
Tokenessions
Feature activation+0.228
said
Token said
Feature activation+0.580
early
Token early
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Moore
Token Moore
Feature activation+0.000
said
Token said
Feature activation+0.189
.
Token.
Feature activation+2.581
Ċ
TokenĊ
Feature activation+6.743
Ċ
TokenĊ
Feature activation+2.925
March
TokenMarch
Feature activation+0.000
said
Token said
Feature activation+0.000
after
Token after
Feature activation+0.000
the
Token the
Feature activation+0.000
"
Token "
Feature activation+0.000
appropriate
Tokenappropriate
Feature activation+0.000
.
Token.
Feature activation+0.068
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.149
Ċ
TokenĊ
Feature activation+6.677
Ċ
TokenĊ
Feature activation+2.453
Other
TokenOther
Feature activation+0.388
top
Token top
Feature activation+0.203
Democrats
Token Democrats
Feature activation+0.077
,
Token,
Feature activation+0.052
.)
Token.)
Feature activation+0.161
and
Token and
Feature activation+0.366
Sen
Token Sen
Feature activation+0.000
.
Token.
Feature activation+0.000
Elizabeth
Token Elizabeth
Feature activation+0.000
Warren
Token Warren
Feature activation+6.605
Elizabeth
Token Elizabeth
Feature activation+0.000
Ann
Token Ann
Feature activation+0.000
Warren
Token Warren
Feature activation+0.000
Sanders
TokenSanders
Feature activation+0.000
:
Token:
Feature activation+0.000
emergency
Token emergency
Feature activation+0.000
declaration
Token declaration
Feature activation+0.000
MORE
Token MORE
Feature activation+0.000
(
Token (
Feature activation+0.801
D
TokenD
Feature activation+0.000
-
Token-
Feature activation+6.539
Mass
TokenMass
Feature activation+0.000
.)
Token.)
Feature activation+0.820
for
Token for
Feature activation+0.000
staying
Token staying
Feature activation+0.054
neutral
Token neutral
Feature activation+0.029
you
Token you
Feature activation+0.000
'
Token'
Feature activation+0.000
MORE
Token MORE
Feature activation+0.000
(
Token (
Feature activation+2.033
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+6.457
Utah
TokenUtah
Feature activation+0.674
)
Token)
Feature activation+0.401
are
Token are
Feature activation+0.029
among
Token among
Feature activation+0.000
the
Token the
Feature activation+0.000
off
Token off
Feature activation+1.201
after
Token after
Feature activation+0.000
Majority
Token Majority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Kevin
Token Kevin
Feature activation+0.398
McCarthy
Token McCarthy
Feature activation+6.430
(
Token (
Feature activation+2.167
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+3.901
Calif
TokenCalif
Feature activation+0.796
.)
Token.)
Feature activation+0.000
Steve
Token Steve
Feature activation+0.000
Scal
Token Scal
Feature activation+0.000
ise
Tokenise
Feature activation+6.761
(
Token (
Feature activation+4.665
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+5.523
La
TokenLa
Feature activation+0.000
.)
Token.)
Feature activation+0.894
and
Token and
Feature activation+0.000
then
Token then
Feature activation+1.102
-
Token-
Feature activation+0.000

Top DFA by src position
MAX = 9.133

than
Token than
Feature activation+0.000
Top resid features:
10
Token 10
Feature activation-0.000
Top resid features:
years
Token years
Feature activation-0.002
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
prison
Token prison
Feature activation-0.001
Top resid features:
MORE
Token MORE
Feature activation+9.133
Top resid features:
for
Token for
Feature activation+0.002
Top resid features:
a
Token a
Feature activation-0.000
Top resid features:
tweet
Token tweet
Feature activation+0.009
Top resid features:
warning
Token warning
Feature activation+0.007
Top resid features:
former
Token former
Feature activation+0.003
Top resid features:
blocking
Token blocking
Feature activation-0.000
Top resid features:
Trump
Token Trump
Feature activation+0.000
Top resid features:
's
Token's
Feature activation-0.000
Top resid features:
emergency
Token emergency
Feature activation+0.000
Top resid features:
declaration
Token declaration
Feature activation-0.000
Top resid features:
MORE
Token MORE
Feature activation+8.414
Top resid features:
(
Token (
Feature activation+0.275
Top resid features:
D
TokenD
Feature activation+0.019
Top resid features:
-
Token-
Feature activation+0.060
Top resid features:
Mass
TokenMass
Feature activation-0.005
Top resid features:
.)
Token.)
Feature activation+0.143
Top resid features:
than
Token than
Feature activation+0.001
Top resid features:
10
Token 10
Feature activation+0.002
Top resid features:
years
Token years
Feature activation-0.003
Top resid features:
in
Token in
Feature activation+0.009
Top resid features:
prison
Token prison
Feature activation-0.008
Top resid features:
MORE
Token MORE
Feature activation+8.729
Top resid features:
for
Token for
Feature activation+0.018
Top resid features:
a
Token a
Feature activation-0.009
Top resid features:
tweet
Token tweet
Feature activation+0.031
Top resid features:
warning
Token warning
Feature activation+0.234
Top resid features:
former
Token former
Feature activation+0.057
Top resid features:
blocking
Token blocking
Feature activation-0.000
Top resid features:
Trump
Token Trump
Feature activation+0.001
Top resid features:
's
Token's
Feature activation+0.000
Top resid features:
emergency
Token emergency
Feature activation+0.000
Top resid features:
declaration
Token declaration
Feature activation+0.001
Top resid features:
MORE
Token MORE
Feature activation+8.599
Top resid features:
(
Token (
Feature activation+0.072
Top resid features:
D
TokenD
Feature activation+0.000
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
Mass
TokenMass
Feature activation-0.004
Top resid features:
.)
Token.)
Feature activation+0.070
Top resid features:
as
Token as
Feature activation+0.000
Top resid features:
long
Token long
Feature activation+0.001
Top resid features:
nomination
Token nomination
Feature activation+0.000
Top resid features:
contest
Token contest
Feature activation-0.001
Top resid features:
looms
Token looms
Feature activation-0.000
Top resid features:
MORE
Token MORE
Feature activation+4.377
Top resid features:
(
Token (
Feature activation+0.085
Top resid features:
D
TokenD
Feature activation+0.005
Top resid features:
-
Token-
Feature activation+0.016
Top resid features:
Fla
TokenFla
Feature activation-0.004
Top resid features:
.)
Token.)
Feature activation+0.051
Top resid features:
was
Token was
Feature activation-0.004
Top resid features:
right
Token right
Feature activation-0.002
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.007
Top resid features:
in
Token in
Feature activation-0.002
Top resid features:
2014
Token 2014
Feature activation-0.005
Top resid features:
MORE
Token MORE
Feature activation+8.376
Top resid features:
âĢ
TokenâĢ
Feature activation+0.160
Top resid features:
Ļ
TokenĻ
Feature activation+0.003
Top resid features:
s
Tokens
Feature activation-0.001
Top resid features:
(
Token (
Feature activation+0.004
Top resid features:
R
TokenR
Feature activation+0.002
Top resid features:
it
Token it
Feature activation-0.000
Top resid features:
will
Token will
Feature activation-0.000
Top resid features:
kill
Token kill
Feature activation+0.000
Top resid features:
you
Token you
Feature activation-0.000
Top resid features:
'
Token'
Feature activation+0.001
Top resid features:
MORE
Token MORE
Feature activation+8.555
Top resid features:
(
Token (
Feature activation+0.058
Top resid features:
R
TokenR
Feature activation+0.005
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
Utah
TokenUtah
Feature activation-0.002
Top resid features:
)
Token)
Feature activation+0.006
Top resid features:
than
Token than
Feature activation+0.001
Top resid features:
10
Token 10
Feature activation+0.001
Top resid features:
years
Token years
Feature activation-0.005
Top resid features:
in
Token in
Feature activation+0.002
Top resid features:
prison
Token prison
Feature activation-0.001
Top resid features:
MORE
Token MORE
Feature activation+8.226
Top resid features:
for
Token for
Feature activation+0.003
Top resid features:
a
Token a
Feature activation-0.002
Top resid features:
tweet
Token tweet
Feature activation+0.025
Top resid features:
warning
Token warning
Feature activation+0.007
Top resid features:
former
Token former
Feature activation+0.012
Top resid features:
was
Token was
Feature activation-0.005
Top resid features:
right
Token right
Feature activation-0.002
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.005
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
2014
Token 2014
Feature activation-0.003
Top resid features:
MORE
Token MORE
Feature activation+8.002
Top resid features:
âĢ
TokenâĢ
Feature activation+0.137
Top resid features:
Ļ
TokenĻ
Feature activation+0.006
Top resid features:
s
Tokens
Feature activation-0.002
Top resid features:
(
Token (
Feature activation+0.018
Top resid features:
R
TokenR
Feature activation+0.001
Top resid features:
'
Token'
Feature activation-0.001
Top resid features:
Trump
Token Trump
Feature activation+0.004
Top resid features:
is
Token is
Feature activation+0.003
Top resid features:
doing
Token doing
Feature activation-0.001
Top resid features:
right
Token right
Feature activation-0.004
Top resid features:
MORE
Token MORE
Feature activation+7.061
Top resid features:
(
Token (
Feature activation+0.589
Top resid features:
D
TokenD
Feature activation+0.021
Top resid features:
-
Token-
Feature activation+0.017
Top resid features:
N
TokenN
Feature activation-0.011
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
warnings
Token warnings
Feature activation-0.002
Top resid features:
from
Token from
Feature activation+0.000
Top resid features:
new
Token new
Feature activation+0.000
Top resid features:
climate
Token climate
Feature activation-0.000
Top resid features:
studies
Token studies
Feature activation-0.000
Top resid features:
MORE
Token MORE
Feature activation+7.905
Top resid features:
(
Token (
Feature activation+0.181
Top resid features:
R
TokenR
Feature activation+0.052
Top resid features:
-
Token-
Feature activation+0.026
Top resid features:
Ky
TokenKy
Feature activation-0.005
Top resid features:
.)
Token.)
Feature activation+0.039
Top resid features:
was
Token was
Feature activation-0.027
Top resid features:
right
Token right
Feature activation-0.003
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.013
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
2014
Token 2014
Feature activation-0.012
Top resid features:
MORE
Token MORE
Feature activation+7.906
Top resid features:
âĢ
TokenâĢ
Feature activation+0.310
Top resid features:
Ļ
TokenĻ
Feature activation+0.008
Top resid features:
s
Tokens
Feature activation-0.002
Top resid features:
(
Token (
Feature activation+0.036
Top resid features:
R
TokenR
Feature activation+0.003
Top resid features:
it
Token it
Feature activation-0.003
Top resid features:
will
Token will
Feature activation-0.002
Top resid features:
kill
Token kill
Feature activation-0.000
Top resid features:
you
Token you
Feature activation-0.001
Top resid features:
'
Token'
Feature activation+0.001
Top resid features:
MORE
Token MORE
Feature activation+7.368
Top resid features:
(
Token (
Feature activation+0.209
Top resid features:
R
TokenR
Feature activation+0.189
Top resid features:
-
Token-
Feature activation+0.048
Top resid features:
Utah
TokenUtah
Feature activation-0.009
Top resid features:
)
Token)
Feature activation+0.018
Top resid features:
than
Token than
Feature activation+0.000
Top resid features:
10
Token 10
Feature activation+0.000
Top resid features:
years
Token years
Feature activation-0.004
Top resid features:
in
Token in
Feature activation-0.000
Top resid features:
prison
Token prison
Feature activation-0.000
Top resid features:
MORE
Token MORE
Feature activation+7.631
Top resid features:
âĢ
TokenâĢ
Feature activation+0.147
Top resid features:
Ļ
TokenĻ
Feature activation+0.018
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
wall
Token wall
Feature activation+0.000
Top resid features:
on
Token on
Feature activation+0.000
Top resid features:
it
Token it
Feature activation-0.001
Top resid features:
will
Token will
Feature activation-0.001
Top resid features:
kill
Token kill
Feature activation-0.000
Top resid features:
you
Token you
Feature activation-0.001
Top resid features:
'
Token'
Feature activation+0.001
Top resid features:
MORE
Token MORE
Feature activation+7.373
Top resid features:
(
Token (
Feature activation+0.161
Top resid features:
R
TokenR
Feature activation+0.105
Top resid features:
-
Token-
Feature activation+0.037
Top resid features:
Utah
TokenUtah
Feature activation-0.007
Top resid features:
)
Token)
Feature activation+0.017
Top resid features:
it
Token it
Feature activation-0.000
Top resid features:
will
Token will
Feature activation-0.000
Top resid features:
kill
Token kill
Feature activation+0.000
Top resid features:
you
Token you
Feature activation-0.000
Top resid features:
'
Token'
Feature activation+0.001
Top resid features:
MORE
Token MORE
Feature activation+7.724
Top resid features:
(
Token (
Feature activation+0.084
Top resid features:
R
TokenR
Feature activation+0.011
Top resid features:
-
Token-
Feature activation+0.003
Top resid features:
Utah
TokenUtah
Feature activation-0.002
Top resid features:
)
Token)
Feature activation+0.004
Top resid features:
blocking
Token blocking
Feature activation-0.002
Top resid features:
Trump
Token Trump
Feature activation+0.000
Top resid features:
's
Token's
Feature activation-0.001
Top resid features:
emergency
Token emergency
Feature activation-0.000
Top resid features:
declaration
Token declaration
Feature activation+0.001
Top resid features:
MORE
Token MORE
Feature activation+7.937
Top resid features:
(
Token (
Feature activation-0.211
Top resid features:
D
TokenD
Feature activation-0.002
Top resid features:
-
Token-
Feature activation-0.022
Top resid features:
Mass
TokenMass
Feature activation+0.000
Top resid features:
.)
Token.)
Feature activation+0.000
Top resid features:
it
Token it
Feature activation-0.002
Top resid features:
will
Token will
Feature activation-0.001
Top resid features:
kill
Token kill
Feature activation-0.002
Top resid features:
you
Token you
Feature activation-0.002
Top resid features:
'
Token'
Feature activation+0.017
Top resid features:
MORE
Token MORE
Feature activation+7.712
Top resid features:
(
Token (
Feature activation+0.013
Top resid features:
R
TokenR
Feature activation-0.050
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
Utah
TokenUtah
Feature activation+0.000
Top resid features:
)
Token)
Feature activation+0.000
Top resid features:
was
Token was
Feature activation-0.002
Top resid features:
right
Token right
Feature activation-0.002
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.004
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
2014
Token 2014
Feature activation-0.006
Top resid features:
MORE
Token MORE
Feature activation+7.781
Top resid features:
âĢ
TokenâĢ
Feature activation+0.077
Top resid features:
Ļ
TokenĻ
Feature activation+0.007
Top resid features:
s
Tokens
Feature activation-0.000
Top resid features:
(
Token (
Feature activation+0.009
Top resid features:
R
TokenR
Feature activation+0.000
Top resid features:
was
Token was
Feature activation-0.001
Top resid features:
right
Token right
Feature activation-0.001
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.001
Top resid features:
in
Token in
Feature activation-0.001
Top resid features:
2014
Token 2014
Feature activation-0.003
Top resid features:
MORE
Token MORE
Feature activation+7.688
Top resid features:
âĢ
TokenâĢ
Feature activation+0.042
Top resid features:
Ļ
TokenĻ
Feature activation+0.014
Top resid features:
s
Tokens
Feature activation-0.003
Top resid features:
(
Token (
Feature activation+0.010
Top resid features:
R
TokenR
Feature activation-0.002
Top resid features:

Decoder Weights Distribution

Head 0: 0.04

Head 1: 0.02

Head 2: 0.58

Head 3: 0.03

Head 4: 0.02

Head 5: 0.04

Head 6: 0.02

Head 7: 0.03

Head 8: 0.03

Head 9: 0.04

Head 10: 0.06

Head 11: 0.08

Positive logits

ADVERTISEMENT3.96

­3.82

POLITICO3.40

Emanuel3.32

3.25

Betsy3.21

,"3.05

Barack3.01

lawmaker2.99

cellphone2.95

congressional2.92

Jeffrey2.90

Ivanka2.89

aide2.89

Nunes2.88

lawmakers2.88

ObamaCare2.87

JPMorgan2.86

CNN2.82

Hillary2.82

Negative logits

=================-3.64

Whilst-3.47

realise-3.36

////////////////////////////////-3.34

recognised-3.27

whilst-3.27

recognise-3.23

!!!-3.19

´-3.11

organise-3.09

:-)-3.09

...)-3.06

realised-2.94

!".-2.87

...?-2.85

Firstly-2.83

avour-2.82

edit-2.78

Conclusion-2.77

ynthesis-2.74

INTERVAL 7.236 - 8.040
CONTAINS 0.001%

warning
Token warning
Feature activation+0.000
former
Token former
Feature activation+0.000
FBI
Token FBI
Feature activation+0.499
Director
Token Director
Feature activation+0.000
James
Token James
Feature activation+0.000
Comey
Token Comey
Feature activation+7.812
about
Token about
Feature activation+0.000
leaking
Token leaking
Feature activation+0.000
information
Token information
Feature activation+0.000
.
Token.
Feature activation+5.481
Ċ
TokenĊ
Feature activation+7.033
of
Token of
Feature activation+0.000
Massachusetts
Token Massachusetts
Feature activation+0.000
,
Token,
Feature activation+0.000
where
Token where
Feature activation+0.000
Hillary
Token Hillary
Feature activation+2.971
Clinton
Token Clinton
Feature activation+7.564
Hillary
Token Hillary
Feature activation+0.000
Diane
Token Diane
Feature activation+0.000
Rodham
Token Rodham
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
Sanders
TokenSanders
Feature activation+0.000
No
Token No
Feature activation+1.613
.
Token.
Feature activation+0.000
2
Token 2
Feature activation+0.000
post
Token post
Feature activation+0.000
.
Token.
Feature activation+1.449
Ċ
TokenĊ
Feature activation+7.237
Ċ
TokenĊ
Feature activation+3.916
But
TokenBut
Feature activation+0.000
the
Token the
Feature activation+0.000
episode
Token episode
Feature activation+0.000
proved
Token proved
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.282
Democratic
Token Democratic
Feature activation+0.320
primary
Token primary
Feature activation+0.000
.
Token.
Feature activation+2.611
Ċ
TokenĊ
Feature activation+7.945
Ċ
TokenĊ
Feature activation+4.688
Some
TokenSome
Feature activation+0.496
Sanders
Token Sanders
Feature activation+0.846
fans
Token fans
Feature activation+0.000
are
Token are
Feature activation+0.083
free
Token free
Feature activation+0.000
republic
Token republic
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.040
Ċ
TokenĊ
Feature activation+1.713
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.300
Ċ
TokenĊ
Feature activation+0.281
In
TokenIn
Feature activation+0.000

INTERVAL 6.432 - 7.236
CONTAINS 0.001%

"
Token "
Feature activation+0.000
appropriate
Tokenappropriate
Feature activation+0.000
.
Token.
Feature activation+0.068
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.149
Ċ
TokenĊ
Feature activation+6.677
Ċ
TokenĊ
Feature activation+2.453
Other
TokenOther
Feature activation+0.388
top
Token top
Feature activation+0.203
Democrats
Token Democrats
Feature activation+0.077
,
Token,
Feature activation+0.052
you
Token you
Feature activation+0.000
'
Token'
Feature activation+0.000
MORE
Token MORE
Feature activation+0.000
(
Token (
Feature activation+2.033
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+6.457
Utah
TokenUtah
Feature activation+0.674
)
Token)
Feature activation+0.401
are
Token are
Feature activation+0.029
among
Token among
Feature activation+0.000
the
Token the
Feature activation+0.000
called
Token called
Feature activation+0.000
for
Token for
Feature activation+0.495
rec
Token rec
Feature activation+0.000
usal
Tokenusal
Feature activation+0.000
.
Token.
Feature activation+2.912
Ċ
TokenĊ
Feature activation+6.744
Ċ
TokenĊ
Feature activation+1.819
S
TokenS
Feature activation+0.000
essions
Tokenessions
Feature activation+0.228
said
Token said
Feature activation+0.580
early
Token early
Feature activation+0.000
then
Token then
Feature activation+1.102
-
Token-
Feature activation+0.000
Rep
TokenRep
Feature activation+0.000
,
Token,
Feature activation+0.000
Tom
Token Tom
Feature activation+0.000
Price
Token Price
Feature activation+6.894
(
Token (
Feature activation+4.545
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+4.212
Ga
TokenGa
Feature activation+2.674
.)
Token.)
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Moore
Token Moore
Feature activation+0.000
said
Token said
Feature activation+0.189
.
Token.
Feature activation+2.581
Ċ
TokenĊ
Feature activation+6.743
Ċ
TokenĊ
Feature activation+2.925
March
TokenMarch
Feature activation+0.000
said
Token said
Feature activation+0.000
after
Token after
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 5.628 - 6.432
CONTAINS 0.000%

off
Token off
Feature activation+1.201
after
Token after
Feature activation+0.000
Majority
Token Majority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Kevin
Token Kevin
Feature activation+0.398
McCarthy
Token McCarthy
Feature activation+6.430
(
Token (
Feature activation+2.167
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+3.901
Calif
TokenCalif
Feature activation+0.796
.)
Token.)
Feature activation+0.000

INTERVAL 4.824 - 5.628
CONTAINS 0.001%

(
Token (
Feature activation+1.427
D
TokenD
Feature activation+0.000
-
Token-
Feature activation+4.695
N
TokenN
Feature activation+0.000
.
Token.
Feature activation+0.000
Y
TokenY
Feature activation+5.065
.)
Token.)
Feature activation+0.000
and
Token and
Feature activation+0.000
Bill
Token Bill
Feature activation+0.000
Nelson
Token Nelson
Feature activation+6.838
Clarence
Token Clarence
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+1.188
s
Tokens
Feature activation+0.000
(
Token (
Feature activation+2.256
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+5.268
Ohio
TokenOhio
Feature activation+0.105
)
Token)
Feature activation+0.000
forced
Token forced
Feature activation+0.000
resignation
Token resignation
Feature activation+0.992
in
Token in
Feature activation+0.000
Steve
Token Steve
Feature activation+0.000
Scal
Token Scal
Feature activation+0.000
ise
Tokenise
Feature activation+6.761
(
Token (
Feature activation+4.665
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+5.523
La
TokenLa
Feature activation+0.000
.)
Token.)
Feature activation+0.894
and
Token and
Feature activation+0.000
then
Token then
Feature activation+1.102
-
Token-
Feature activation+0.000
James
Token James
Feature activation+0.000
Comey
Token Comey
Feature activation+7.812
about
Token about
Feature activation+0.000
leaking
Token leaking
Feature activation+0.000
information
Token information
Feature activation+0.000
.
Token.
Feature activation+5.481
Ċ
TokenĊ
Feature activation+7.033
Ċ
TokenĊ
Feature activation+0.884
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
All
TokenAll
Feature activation+0.000
good
Token good
Feature activation+0.000
senator
Token senator
Feature activation+0.000
.
Token.
Feature activation+2.261
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.147
Ċ
TokenĊ
Feature activation+5.454
Ċ
TokenĊ
Feature activation+1.552
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.013
Following
TokenFollowing
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 4.020 - 4.824
CONTAINS 0.001%

Majority
Token Majority
Feature activation+0.000
Whip
Token Whip
Feature activation+0.000
Steve
Token Steve
Feature activation+0.000
Scal
Token Scal
Feature activation+0.000
ise
Tokenise
Feature activation+6.761
(
Token (
Feature activation+4.665
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+5.523
La
TokenLa
Feature activation+0.000
.)
Token.)
Feature activation+0.894
and
Token and
Feature activation+0.000
over
Token over
Feature activation+0.000
Assad
Token Assad
Feature activation+0.134
MORE
Token MORE
Feature activation+1.130
(
Token (
Feature activation+1.857
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+4.708
Ari
TokenAri
Feature activation+0.000
z
Tokenz
Feature activation+2.275
.)
Token.)
Feature activation+0.135
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.508
-
Token-
Feature activation+0.000
Rep
TokenRep
Feature activation+0.000
,
Token,
Feature activation+0.000
Tom
Token Tom
Feature activation+0.000
Price
Token Price
Feature activation+6.894
(
Token (
Feature activation+4.545
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+4.212
Ga
TokenGa
Feature activation+2.674
.)
Token.)
Feature activation+0.000
v
Token v
Feature activation+0.000
congressional
Token congressional
Feature activation+0.000
oversight
Token oversight
Feature activation+0.000
MORE
Token MORE
Feature activation+0.000
(
Token (
Feature activation+1.556
D
TokenD
Feature activation+0.000
-
Token-
Feature activation+4.064
N
TokenN
Feature activation+0.000
.
Token.
Feature activation+0.000
J
TokenJ
Feature activation+2.252
.),
Token.),
Feature activation+0.000
Jim
Token Jim
Feature activation+0.000
doing
Token doing
Feature activation+0.000
right
Token right
Feature activation+0.000
MORE
Token MORE
Feature activation+0.145
(
Token (
Feature activation+1.427
D
TokenD
Feature activation+0.000
-
Token-
Feature activation+4.695
N
TokenN
Feature activation+0.000
.
Token.
Feature activation+0.000
Y
TokenY
Feature activation+5.065
.)
Token.)
Feature activation+0.000
and
Token and
Feature activation+0.000

INTERVAL 3.216 - 4.020
CONTAINS 0.000%

Leader
Token Leader
Feature activation+0.000
Kevin
Token Kevin
Feature activation+0.398
McCarthy
Token McCarthy
Feature activation+6.430
(
Token (
Feature activation+2.167
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+3.901
Calif
TokenCalif
Feature activation+0.796
.)
Token.)
Feature activation+0.000
abruptly
Token abruptly
Feature activation+0.000
abandoned
Token abandoned
Feature activation+0.000
his
Token his
Feature activation+0.000
.
Token.
Feature activation+0.000
2
Token 2
Feature activation+0.000
post
Token post
Feature activation+0.000
.
Token.
Feature activation+1.449
Ċ
TokenĊ
Feature activation+7.237
Ċ
TokenĊ
Feature activation+3.916
But
TokenBut
Feature activation+0.000
the
Token the
Feature activation+0.000
episode
Token episode
Feature activation+0.000
proved
Token proved
Feature activation+0.000
once
Token once
Feature activation+0.000

INTERVAL 2.412 - 3.216
CONTAINS 0.001%

appropriate
Tokenappropriate
Feature activation+0.000
.
Token.
Feature activation+0.068
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.149
Ċ
TokenĊ
Feature activation+6.677
Ċ
TokenĊ
Feature activation+2.453
Other
TokenOther
Feature activation+0.388
top
Token top
Feature activation+0.203
Democrats
Token Democrats
Feature activation+0.077
,
Token,
Feature activation+0.052
including
Token including
Feature activation+0.166
state
Token state
Feature activation+0.064
of
Token of
Feature activation+0.000
Massachusetts
Token Massachusetts
Feature activation+0.000
,
Token,
Feature activation+0.000
where
Token where
Feature activation+0.000
Hillary
Token Hillary
Feature activation+2.971
Clinton
Token Clinton
Feature activation+7.564
Hillary
Token Hillary
Feature activation+0.000
Diane
Token Diane
Feature activation+0.000
Rodham
Token Rodham
Feature activation+0.000
Clinton
Token Clinton
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Moore
Token Moore
Feature activation+0.000
said
Token said
Feature activation+0.189
.
Token.
Feature activation+2.581
Ċ
TokenĊ
Feature activation+6.743
Ċ
TokenĊ
Feature activation+2.925
March
TokenMarch
Feature activation+0.000
said
Token said
Feature activation+0.000
after
Token after
Feature activation+0.000
the
Token the
Feature activation+0.000
interview
Token interview
Feature activation+0.023
Committee
Token Committee
Feature activation+0.274
ranking
Token ranking
Feature activation+0.000
member
Token member
Feature activation+0.048
Elijah
Token Elijah
Feature activation+0.000
Cummings
Token Cummings
Feature activation+7.093
(
Token (
Feature activation+2.763
D
TokenD
Feature activation+0.000
-
Token-
Feature activation+4.488
M
TokenM
Feature activation+0.000
d
Tokend
Feature activation+0.000
.)
Token.)
Feature activation+0.161
neutral
Token neutral
Feature activation+0.029
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.282
Democratic
Token Democratic
Feature activation+0.320
primary
Token primary
Feature activation+0.000
.
Token.
Feature activation+2.611
Ċ
TokenĊ
Feature activation+7.945
Ċ
TokenĊ
Feature activation+4.688
Some
TokenSome
Feature activation+0.496
Sanders
Token Sanders
Feature activation+0.846
fans
Token fans
Feature activation+0.000

INTERVAL 1.608 - 2.412
CONTAINS 0.002%

after
Token after
Feature activation+0.000
fight
Token fight
Feature activation+0.000
over
Token over
Feature activation+0.000
Assad
Token Assad
Feature activation+0.134
MORE
Token MORE
Feature activation+1.130
(
Token (
Feature activation+1.857
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+4.708
Ari
TokenAri
Feature activation+0.000
z
Tokenz
Feature activation+2.275
.)
Token.)
Feature activation+0.135
scramble
Token scramble
Feature activation+0.312
.
Token.
Feature activation+1.687
McM
Token McM
Feature activation+0.000
or
Tokenor
Feature activation+0.000
ris
Tokenris
Feature activation+0.000
Rodgers
Token Rodgers
Feature activation+1.648
,
Token,
Feature activation+0.687
Majority
Token Majority
Feature activation+0.000
Whip
Token Whip
Feature activation+0.000
Steve
Token Steve
Feature activation+0.000
Scal
Token Scal
Feature activation+0.000
republic
Token republic
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.040
Ċ
TokenĊ
Feature activation+1.713
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.300
Ċ
TokenĊ
Feature activation+0.281
In
TokenIn
Feature activation+0.000
a
Token a
Feature activation+0.311
set
Token set
Feature activation+0.000
off
Token off
Feature activation+0.000
a
Token a
Feature activation+0.000
leadership
Token leadership
Feature activation+0.000
scramble
Token scramble
Feature activation+0.312
.
Token.
Feature activation+1.687
McM
Token McM
Feature activation+0.000
or
Tokenor
Feature activation+0.000
ris
Tokenris
Feature activation+0.000
Rodgers
Token Rodgers
Feature activation+1.648
,
Token,
Feature activation+0.687
after
Token after
Feature activation+0.000
Majority
Token Majority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Kevin
Token Kevin
Feature activation+0.398
McCarthy
Token McCarthy
Feature activation+6.430
(
Token (
Feature activation+2.167
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+3.901
Calif
TokenCalif
Feature activation+0.796
.)
Token.)
Feature activation+0.000
abruptly
Token abruptly
Feature activation+0.000

INTERVAL 0.804 - 1.608
CONTAINS 0.004%

are
Token are
Feature activation+0.335
also
Token also
Feature activation+0.125
working
Token working
Feature activation+0.260
on
Token on
Feature activation+0.941
similar
Token similar
Feature activation+1.117
legislation
Token legislation
Feature activation+1.460
to
Token to
Feature activation+0.166
prohibit
Token prohibit
Feature activation+0.398
plastic
Token plastic
Feature activation+0.000
guns
Token guns
Feature activation+0.592
,
Token,
Feature activation+0.146
Ċ
TokenĊ
Feature activation+1.819
S
TokenS
Feature activation+0.000
essions
Tokenessions
Feature activation+0.228
said
Token said
Feature activation+0.580
early
Token early
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.857
that
Token that
Feature activation+0.078
he
Token he
Feature activation+0.151
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
willing
Token willing
Feature activation+0.812
McCain
Token McCain
Feature activation+0.000
after
Token after
Feature activation+0.000
fight
Token fight
Feature activation+0.000
over
Token over
Feature activation+0.000
Assad
Token Assad
Feature activation+0.134
MORE
Token MORE
Feature activation+1.130
(
Token (
Feature activation+1.857
R
TokenR
Feature activation+0.000
-
Token-
Feature activation+4.708
Ari
TokenAri
Feature activation+0.000
z
Tokenz
Feature activation+2.275
their
Token their
Feature activation+0.000
feedback
Token feedback
Feature activation+0.000
last
Token last
Feature activation+0.000
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.928
Ċ
TokenĊ
Feature activation+0.311
Article
TokenArticle
Feature activation+0.043
Continued
Token Continued
Feature activation+0.000
Below
Token Below
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Some
TokenSome
Feature activation+0.496
Sanders
Token Sanders
Feature activation+0.846
fans
Token fans
Feature activation+0.000
are
Token are
Feature activation+0.083
particularly
Token particularly
Feature activation+0.000
upset
Token upset
Feature activation+1.216
that
Token that
Feature activation+0.412
Warren
Token Warren
Feature activation+0.016
didn
Token didn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000

INTERVAL 0.000 - 0.804
CONTAINS 99.989%

latest
Token latest
Feature activation+0.000
personal
Token personal
Feature activation+0.000
food
Token food
Feature activation+0.000
revolution
Token revolution
Feature activation+0.000
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
c
Token c
Feature activation+0.000
ic
Tokenic
Feature activation+0.000
at
Tokenat
Feature activation+0.000
ric
Tokenric
Feature activation+0.000
ial
Tokenial
Feature activation+0.000
get
Token get
Feature activation+0.000
to
Token to
Feature activation+0.000
take
Token take
Feature activation+0.000
a
Token a
Feature activation+0.000
look
Token look
Feature activation+0.000
under
Token under
Feature activation+0.000
the
Token the
Feature activation+0.000
hood
Token hood
Feature activation+0.000
thanks
Token thanks
Feature activation+0.000
to
Token to
Feature activation+0.000
publicly
Token publicly
Feature activation+0.000
see
Token see
Feature activation+0.000
the
Token the
Feature activation+0.000
same
Token same
Feature activation+0.000
trend
Token trend
Feature activation+0.000
in
Token in
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
,
Token,
Feature activation+0.000
where
Token where
Feature activation+0.000
new
Token new
Feature activation+0.000
home
Token home
Feature activation+0.000
sales
Token sales
Feature activation+0.000
ev
Tokenev
Feature activation+0.000
sky
Tokensky
Feature activation+0.000
himself
Token himself
Feature activation+0.000
reports
Token reports
Feature activation+0.000
that
Token that
Feature activation+0.000
his
Token his
Feature activation+0.000
imagination
Token imagination
Feature activation+0.000
was
Token was
Feature activation+0.000
brought
Token brought
Feature activation+0.000
alive
Token alive
Feature activation+0.000
by
Token by
Feature activation+0.000
overl
Token overl
Feature activation+0.000
aps
Tokenaps
Feature activation+0.000
of
Token of
Feature activation+0.000
language
Token language
Feature activation+0.000
seen
Token seen
Feature activation+0.000
in
Token in
Feature activation+0.000
this
Token this
Feature activation+0.000
CBS
Token CBS
Feature activation+0.000
News
Token News
Feature activation+0.000
report
Token report
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 3 in H8.2: (feature 17269)

TOP ACTIVATIONS
MAX = 3.177

Friday
Token Friday
Feature activation+0.000
,
Token,
Feature activation+0.000
13
Token 13
Feature activation+0.000
August
Token August
Feature activation+0.000
.
Token.
Feature activation+0.564
Ċ
TokenĊ
Feature activation+3.177
Ċ
TokenĊ
Feature activation+0.045
Ste
TokenSte
Feature activation+0.000
gan
Tokengan
Feature activation+0.000
ography
Tokenography
Feature activation+0.000
normally
Token normally
Feature activation+0.000
't
Token't
Feature activation+0.000
make
Token make
Feature activation+0.000
any
Token any
Feature activation+0.000
sense
Token sense
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.049
Ċ
TokenĊ
Feature activation+0.185
And
TokenAnd
Feature activation+0.000
unlike
Token unlike
Feature activation+0.000
Sa
Token Sa
Feature activation+0.000
eed
Tokeneed
Feature activation+0.000
halls
Token halls
Feature activation+0.000
,
Token,
Feature activation+0.000
brightly
Token brightly
Feature activation+0.000
coloured
Token coloured
Feature activation+0.000
carp
Token carp
Feature activation+0.000
ets
Tokenets
Feature activation+2.972
and
Token and
Feature activation+0.122
futuristic
Token futuristic
Feature activation+0.982
chairs
Token chairs
Feature activation+0.627
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.537
consumed
Token consumed
Feature activation+0.000
drinks
Token drinks
Feature activation+0.000
containing
Token containing
Feature activation+0.000
fat
Token fat
Feature activation+0.099
,
Token,
Feature activation+0.000
glucose
Token glucose
Feature activation+2.922
,
Token,
Feature activation+0.498
protein
Token protein
Feature activation+2.395
and
Token and
Feature activation+0.000
placebo
Token placebo
Feature activation+0.212
.
Token.
Feature activation+0.207
today
Token today
Feature activation+0.000
's
Token's
Feature activation+0.282
market
Token market
Feature activation+0.036
price
Token price
Feature activation+0.328
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.746
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
administration
Token administration
Feature activation+0.000
insisted
Token insisted
Feature activation+0.000
countries
Token countries
Feature activation+0.000
it
Token it
Feature activation+0.000
claimed
Token claimed
Feature activation+0.000
use
Token use
Feature activation+0.000
".
Token".
Feature activation+0.157
Ċ
TokenĊ
Feature activation+2.571
Ċ
TokenĊ
Feature activation+0.000
There
TokenThere
Feature activation+0.000
's
Token's
Feature activation+0.000
no
Token no
Feature activation+0.000
hint
Token hint
Feature activation+0.000
whilst
Token whilst
Feature activation+0.000
still
Token still
Feature activation+0.000
in
Token in
Feature activation+0.000
your
Token your
Feature activation+0.000
mouth
Token mouth
Feature activation+1.677
,
Token,
Feature activation+2.431
particularly
Token particularly
Feature activation+0.000
with
Token with
Feature activation+0.000
carbohydrates
Token carbohydrates
Feature activation+0.327
.
Token.
Feature activation+0.000
If
Token If
Feature activation+0.000
from
Token from
Feature activation+0.000
internet
Token internet
Feature activation+0.000
search
Token search
Feature activation+0.000
results
Token results
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.424
Ċ
TokenĊ
Feature activation+0.073
The
TokenThe
Feature activation+0.000
proposals
Token proposals
Feature activation+0.000
were
Token were
Feature activation+0.000
made
Token made
Feature activation+0.000
containing
Token containing
Feature activation+0.000
fat
Token fat
Feature activation+0.099
,
Token,
Feature activation+0.000
glucose
Token glucose
Feature activation+2.922
,
Token,
Feature activation+0.498
protein
Token protein
Feature activation+2.395
and
Token and
Feature activation+0.000
placebo
Token placebo
Feature activation+0.212
.
Token.
Feature activation+0.207
Cognitive
Token Cognitive
Feature activation+0.096
performance
Token performance
Feature activation+0.342
least
Token least
Feature activation+0.000
a
Token a
Feature activation+0.000
second
Token second
Feature activation+0.000
season
Token season
Feature activation+0.000
.
Token.
Feature activation+0.910
Ċ
TokenĊ
Feature activation+2.272
Ċ
TokenĊ
Feature activation+0.085
Show
TokenShow
Feature activation+0.000
runners
Tokenrunners
Feature activation+0.000
Maur
Token Maur
Feature activation+0.000
issa
Tokenissa
Feature activation+0.000
result
Token result
Feature activation+0.000
on
Token on
Feature activation+0.028
23
Token 23
Feature activation+0.000
June
Token June
Feature activation+0.132
.
Token.
Feature activation+1.169
Ċ
TokenĊ
Feature activation+2.183
Ċ
TokenĊ
Feature activation+0.179
He
TokenHe
Feature activation+0.075
accused
Token accused
Feature activation+0.274
the
Token the
Feature activation+0.233
British
Token British
Feature activation+0.000
or
Token or
Feature activation+0.081
23
Token 23
Feature activation+0.000
April
Token April
Feature activation+0.000
2011
Token 2011
Feature activation+0.000
.
Token.
Feature activation+0.165
Ċ
TokenĊ
Feature activation+2.151
Ċ
TokenĊ
Feature activation+0.000
Mr
TokenMr
Feature activation+0.345
Yar
Token Yar
Feature activation+0.000
'
Token'
Feature activation+0.000
A
TokenA
Feature activation+0.000
,
Token,
Feature activation+0.000
330
Token 330
Feature activation+0.000
m
Tokenm
Feature activation+0.000
citizens
Token citizens
Feature activation+0.132
.
Token.
Feature activation+1.053
Ċ
TokenĊ
Feature activation+2.129
Ċ
TokenĊ
Feature activation+0.601
But
TokenBut
Feature activation+0.225
if
Token if
Feature activation+0.437
you
Token you
Feature activation+0.000
are
Token are
Feature activation+0.365
regulators
Token regulators
Feature activation+1.172
,
Token,
Feature activation+0.209
entrepreneurs
Token entrepreneurs
Feature activation+1.179
,
Token,
Feature activation+0.129
policy
Token policy
Feature activation+0.131
makers
Token makers
Feature activation+2.126
and
Token and
Feature activation+0.000
academics
Token academics
Feature activation+0.868
to
Token to
Feature activation+0.080
explore
Token explore
Feature activation+0.000
blockchain
Token blockchain
Feature activation+0.223
public
Token public
Feature activation+0.000
monument
Token monument
Feature activation+0.022
and
Token and
Feature activation+0.000
park
Token park
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.057
Ċ
TokenĊ
Feature activation+0.000
Z
TokenZ
Feature activation+0.000
ab
Tokenab
Feature activation+0.000
inski
Tokeninski
Feature activation+0.000
re
Token re
Feature activation+0.000
experts
Token experts
Feature activation+1.186
,
Token,
Feature activation+0.562
startups
Token startups
Feature activation+1.147
,
Token,
Feature activation+0.393
corpor
Token corpor
Feature activation+0.083
ates
Tokenates
Feature activation+2.051
,
Token,
Feature activation+0.503
leading
Token leading
Feature activation+0.061
F
Token F
Feature activation+0.000
SI
TokenSI
Feature activation+0.400
executives
Token executives
Feature activation+1.029
surviving
Token surviving
Feature activation+0.058
Ken
Token Ken
Feature activation+0.000
y
Tokeny
Feature activation+0.000
ans
Tokenans
Feature activation+0.000
.
Token.
Feature activation+0.139
Ċ
TokenĊ
Feature activation+2.047
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
I
TokenI
Feature activation+0.000
think
Token think
Feature activation+0.000
they
Token they
Feature activation+0.000
other
Token other
Feature activation+0.000
parts
Token parts
Feature activation+0.000
of
Token of
Feature activation+0.000
Britain
Token Britain
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.028
Ċ
TokenĊ
Feature activation+0.000
These
TokenThese
Feature activation+0.000
steps
Token steps
Feature activation+0.000
would
Token would
Feature activation+0.000
go
Token go
Feature activation+0.000
adaptive
Token adaptive
Feature activation+0.000
and
Token and
Feature activation+0.000
Bo
Token Bo
Feature activation+0.000
L
TokenL
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.017
Ċ
TokenĊ
Feature activation+0.325
As
TokenAs
Feature activation+0.000
pointed
Token pointed
Feature activation+0.000
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
Qu
Token Qu
Feature activation+0.000
er
Tokener
Feature activation+0.000
ét
Tokenét
Feature activation+0.000
aro
Tokenaro
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.994
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
job
Token job
Feature activation+0.000
was
Token was
Feature activation+0.000
ret
Token ret
Feature activation+0.000

Top DFA by src position
MAX = 3.709

unveiled
Token unveiled
Feature activation+0.005
Top resid features:
on
Token on
Feature activation+0.004
Top resid features:
Friday
Token Friday
Feature activation-0.013
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
13
Token 13
Feature activation-0.007
Top resid features:
August
Token August
Feature activation+3.709
Top resid features:
.
Token.
Feature activation-0.004
Top resid features:
Ċ
TokenĊ
Feature activation-0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ste
TokenSte
Feature activation+0.000
Top resid features:
gan
Tokengan
Feature activation+0.000
Top resid features:
who
Token who
Feature activation-0.006
Top resid features:
dis
Token dis
Feature activation+0.000
Top resid features:
f
Tokenf
Feature activation+0.001
Top resid features:
avour
Tokenavour
Feature activation+0.495
Top resid features:
normal
Token normal
Feature activation+0.004
Top resid features:
isation
Tokenisation
Feature activation+3.037
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
relations
Token relations
Feature activation+0.002
Top resid features:
with
Token with
Feature activation-0.001
Top resid features:
Pakistan
Token Pakistan
Feature activation-0.010
Top resid features:
until
Token until
Feature activation-0.002
Top resid features:
the
Token the
Feature activation+0.015
Top resid features:
empty
Token empty
Feature activation+0.010
Top resid features:
halls
Token halls
Feature activation-0.005
Top resid features:
,
Token,
Feature activation-0.031
Top resid features:
brightly
Token brightly
Feature activation+0.017
Top resid features:
coloured
Token coloured
Feature activation+3.058
Top resid features:
carp
Token carp
Feature activation-0.004
Top resid features:
ets
Tokenets
Feature activation-0.002
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
futuristic
Token futuristic
Feature activation+0.000
Top resid features:
chairs
Token chairs
Feature activation+0.000
Top resid features:
placebo
Token placebo
Feature activation-0.000
Top resid features:
-
Token-
Feature activation-0.013
Top resid features:
controlled
Tokencontrolled
Feature activation-0.002
Top resid features:
,
Token,
Feature activation-0.022
Top resid features:
random
Token random
Feature activation+0.001
Top resid features:
ised
Tokenised
Feature activation+3.158
Top resid features:
crossover
Token crossover
Feature activation+0.001
Top resid features:
design
Token design
Feature activation-0.005
Top resid features:
,
Token,
Feature activation-0.009
Top resid features:
18
Token 18
Feature activation+0.023
Top resid features:
healthy
Token healthy
Feature activation-0.005
Top resid features:
who
Token who
Feature activation-0.001
Top resid features:
are
Token are
Feature activation-0.000
Top resid features:
owed
Token owed
Feature activation-0.000
Top resid features:
$
Token $
Feature activation-0.000
Top resid features:
27
Token27
Feature activation+0.001
Top resid features:
bn
Tokenbn
Feature activation+3.171
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.046
Top resid features:
a
Token a
Feature activation-0.001
Top resid features:
tough
Token tough
Feature activation+0.000
Top resid features:
proposition
Token proposition
Feature activation+0.002
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
bit
Tokenbit
Feature activation+0.000
Top resid features:
ant
Tokenant
Feature activation+0.000
Top resid features:
"
Token"
Feature activation+0.003
Top resid features:
$
Token $
Feature activation-0.000
Top resid features:
50
Token50
Feature activation+0.002
Top resid features:
m
Tokenm
Feature activation+2.115
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.000
Top resid features:
Gmail
Token Gmail
Feature activation-0.000
Top resid features:
trademark
Token trademark
Feature activation-0.000
Top resid features:
and
Token and
Feature activation-0.003
Top resid features:
must
Token must
Feature activation-0.085
Top resid features:
mix
Token mix
Feature activation+0.001
Top resid features:
with
Token with
Feature activation-0.007
Top resid features:
the
Token the
Feature activation-0.009
Top resid features:
food
Token food
Feature activation-0.007
Top resid features:
whilst
Token whilst
Feature activation+3.320
Top resid features:
still
Token still
Feature activation+0.015
Top resid features:
in
Token in
Feature activation-0.008
Top resid features:
your
Token your
Feature activation-0.000
Top resid features:
mouth
Token mouth
Feature activation-0.016
Top resid features:
,
Token,
Feature activation-0.023
Top resid features:
Group
Token Group
Feature activation+0.005
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
shared
Token shared
Feature activation+0.003
Top resid features:
with
Token with
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
Guardian
Token Guardian
Feature activation+3.067
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
urges
Token urges
Feature activation+0.004
Top resid features:
the
Token the
Feature activation-0.003
Top resid features:
government
Token government
Feature activation-0.002
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
placebo
Token placebo
Feature activation-0.000
Top resid features:
-
Token-
Feature activation-0.009
Top resid features:
controlled
Tokencontrolled
Feature activation-0.002
Top resid features:
,
Token,
Feature activation-0.027
Top resid features:
random
Token random
Feature activation+0.001
Top resid features:
ised
Tokenised
Feature activation+2.629
Top resid features:
crossover
Token crossover
Feature activation+0.001
Top resid features:
design
Token design
Feature activation-0.006
Top resid features:
,
Token,
Feature activation-0.005
Top resid features:
18
Token 18
Feature activation+0.014
Top resid features:
healthy
Token healthy
Feature activation-0.005
Top resid features:
dipped
Token dipped
Feature activation+0.001
Top resid features:
by
Token by
Feature activation+0.000
Top resid features:
more
Token more
Feature activation-0.000
Top resid features:
than
Token than
Feature activation-0.003
Top resid features:
6
Token 6
Feature activation-0.001
Top resid features:
m
Tokenm
Feature activation+2.543
Top resid features:
viewers
Token viewers
Feature activation+0.001
Top resid features:
since
Token since
Feature activation-0.005
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
pilot
Token pilot
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.007
Top resid features:
the
Token the
Feature activation-0.005
Top resid features:
referendum
Token referendum
Feature activation+0.010
Top resid features:
result
Token result
Feature activation+0.008
Top resid features:
on
Token on
Feature activation-0.008
Top resid features:
23
Token 23
Feature activation+0.011
Top resid features:
June
Token June
Feature activation+2.700
Top resid features:
.
Token.
Feature activation-0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.017
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
He
TokenHe
Feature activation+0.000
Top resid features:
accused
Token accused
Feature activation+0.000
Top resid features:
either
Token either
Feature activation-0.000
Top resid features:
be
Token be
Feature activation-0.000
Top resid features:
held
Token held
Feature activation+0.001
Top resid features:
on
Token on
Feature activation-0.001
Top resid features:
22
Token 22
Feature activation+0.004
Top resid features:
January
Token January
Feature activation+2.276
Top resid features:
or
Token or
Feature activation-0.004
Top resid features:
23
Token 23
Feature activation-0.000
Top resid features:
April
Token April
Feature activation+0.524
Top resid features:
2011
Token 2011
Feature activation+0.017
Top resid features:
.
Token.
Feature activation-0.013
Top resid features:
in
Token in
Feature activation+0.003
Top resid features:
this
Token this
Feature activation+0.009
Top resid features:
case
Token case
Feature activation+0.007
Top resid features:
,
Token,
Feature activation+0.011
Top resid features:
330
Token 330
Feature activation+0.006
Top resid features:
m
Tokenm
Feature activation+2.620
Top resid features:
citizens
Token citizens
Feature activation+0.018
Top resid features:
.
Token.
Feature activation-0.004
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
But
TokenBut
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.048
Top resid features:
two
Token two
Feature activation-0.001
Top resid features:
-
Token-
Feature activation-0.001
Top resid features:
day
Tokenday
Feature activation+0.006
Top resid features:
conference
Token conference
Feature activation-0.004
Top resid features:
organised
Token organised
Feature activation+2.671
Top resid features:
by
Token by
Feature activation+0.001
Top resid features:
F
Token F
Feature activation+0.006
Top resid features:
int
Tokenint
Feature activation+0.000
Top resid features:
ech
Tokenech
Feature activation-0.001
Top resid features:
Valley
Token Valley
Feature activation+0.001
Top resid features:
others
Token others
Feature activation+0.001
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
museum
Token museum
Feature activation+0.001
Top resid features:
and
Token and
Feature activation+0.032
Top resid features:
cultural
Token cultural
Feature activation+0.003
Top resid features:
centre
Token centre
Feature activation+2.458
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
and
Token and
Feature activation-0.000
Top resid features:
others
Token others
Feature activation+0.001
Top resid features:
still
Token still
Feature activation+0.001
Top resid features:
think
Token think
Feature activation+0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.056
Top resid features:
two
Token two
Feature activation-0.002
Top resid features:
-
Token-
Feature activation-0.002
Top resid features:
day
Tokenday
Feature activation+0.012
Top resid features:
conference
Token conference
Feature activation-0.005
Top resid features:
organised
Token organised
Feature activation+2.403
Top resid features:
by
Token by
Feature activation+0.002
Top resid features:
F
Token F
Feature activation+0.019
Top resid features:
int
Tokenint
Feature activation+0.001
Top resid features:
ech
Tokenech
Feature activation-0.001
Top resid features:
Valley
Token Valley
Feature activation+0.002
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
£
Token £
Feature activation+0.047
Top resid features:
19
Token19
Feature activation+0.008
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
9
Token9
Feature activation+0.000
Top resid features:
m
Tokenm
Feature activation+2.557
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
surviving
Token surviving
Feature activation+0.000
Top resid features:
Ken
Token Ken
Feature activation-0.005
Top resid features:
y
Tokeny
Feature activation+0.003
Top resid features:
for
Token for
Feature activation-0.001
Top resid features:
attracting
Token attracting
Feature activation+0.001
Top resid features:
migrants
Token migrants
Feature activation+0.002
Top resid features:
to
Token to
Feature activation-0.000
Top resid features:
revital
Token revital
Feature activation+0.001
Top resid features:
ise
Tokenise
Feature activation+2.188
Top resid features:
their
Token their
Feature activation+0.001
Top resid features:
economies
Token economies
Feature activation+0.005
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.131
Top resid features:
from
Token from
Feature activation+0.002
Top resid features:
targeting
Token targeting
Feature activation+0.002
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.010
Top resid features:
Several
TokenSeveral
Feature activation-0.006
Top resid features:
old
Token old
Feature activation-0.004
Top resid features:
armour
Token armour
Feature activation+2.563
Top resid features:
sets
Token sets
Feature activation-0.003
Top resid features:
from
Token from
Feature activation-0.000
Top resid features:
class
Token class
Feature activation+0.005
Top resid features:
quests
Token quests
Feature activation-0.002
Top resid features:
look
Token look
Feature activation+0.008
Top resid features:
project
Token project
Feature activation-0.001
Top resid features:
to
Token to
Feature activation-0.000
Top resid features:
build
Token build
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
220
Token 220
Feature activation+0.000
Top resid features:
km
Tokenkm
Feature activation+2.546
Top resid features:
line
Token line
Feature activation+0.000
Top resid features:
from
Token from
Feature activation-0.000
Top resid features:
Mexico
Token Mexico
Feature activation-0.005
Top resid features:
City
Token City
Feature activation-0.005
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:

Decoder Weights Distribution

Head 0: 0.05

Head 1: 0.02

Head 2: 0.56

Head 3: 0.03

Head 4: 0.03

Head 5: 0.06

Head 6: 0.02

Head 7: 0.04

Head 8: 0.04

Head 9: 0.04

Head 10: 0.05

Head 11: 0.06

Positive logits

recognise10.54

realise10.12

£9.87

organise9.87

recognised9.45

Whilst9.38

organised9.35

organising9.34

Nato9.34

analyse9.32

organis9.29

realised9.22

organisation9.17

analys9.16

emphas9.07

programme9.04

aluminium9.01

programmes8.91

organisations8.90

apologise8.85

Negative logits

colored-6.29

defense-5.18

caliber-5.07

traveled-4.91

honoring-4.82

honored-4.80

honor-4.79

avored-4.76

harbor-4.69

gotten-4.68

favors-4.67

demeanor-4.66

---4.62

favorably-4.61

canceled-4.61

center-4.60

nont-4.60

---4.55

behaviors-4.53

-4.50

INTERVAL 2.860 - 3.177
CONTAINS 0.000%

Friday
Token Friday
Feature activation+0.000
,
Token,
Feature activation+0.000
13
Token 13
Feature activation+0.000
August
Token August
Feature activation+0.000
.
Token.
Feature activation+0.564
Ċ
TokenĊ
Feature activation+3.177
Ċ
TokenĊ
Feature activation+0.045
Ste
TokenSte
Feature activation+0.000
gan
Tokengan
Feature activation+0.000
ography
Tokenography
Feature activation+0.000
normally
Token normally
Feature activation+0.000
't
Token't
Feature activation+0.000
make
Token make
Feature activation+0.000
any
Token any
Feature activation+0.000
sense
Token sense
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.049
Ċ
TokenĊ
Feature activation+0.185
And
TokenAnd
Feature activation+0.000
unlike
Token unlike
Feature activation+0.000
Sa
Token Sa
Feature activation+0.000
eed
Tokeneed
Feature activation+0.000
halls
Token halls
Feature activation+0.000
,
Token,
Feature activation+0.000
brightly
Token brightly
Feature activation+0.000
coloured
Token coloured
Feature activation+0.000
carp
Token carp
Feature activation+0.000
ets
Tokenets
Feature activation+2.972
and
Token and
Feature activation+0.122
futuristic
Token futuristic
Feature activation+0.982
chairs
Token chairs
Feature activation+0.627
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.537
consumed
Token consumed
Feature activation+0.000
drinks
Token drinks
Feature activation+0.000
containing
Token containing
Feature activation+0.000
fat
Token fat
Feature activation+0.099
,
Token,
Feature activation+0.000
glucose
Token glucose
Feature activation+2.922
,
Token,
Feature activation+0.498
protein
Token protein
Feature activation+2.395
and
Token and
Feature activation+0.000
placebo
Token placebo
Feature activation+0.212
.
Token.
Feature activation+0.207

INTERVAL 2.542 - 2.860
CONTAINS 0.000%

countries
Token countries
Feature activation+0.000
it
Token it
Feature activation+0.000
claimed
Token claimed
Feature activation+0.000
use
Token use
Feature activation+0.000
".
Token".
Feature activation+0.157
Ċ
TokenĊ
Feature activation+2.571
Ċ
TokenĊ
Feature activation+0.000
There
TokenThere
Feature activation+0.000
's
Token's
Feature activation+0.000
no
Token no
Feature activation+0.000
hint
Token hint
Feature activation+0.000
today
Token today
Feature activation+0.000
's
Token's
Feature activation+0.282
market
Token market
Feature activation+0.036
price
Token price
Feature activation+0.328
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.746
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
administration
Token administration
Feature activation+0.000
insisted
Token insisted
Feature activation+0.000

INTERVAL 2.224 - 2.542
CONTAINS 0.000%

from
Token from
Feature activation+0.000
internet
Token internet
Feature activation+0.000
search
Token search
Feature activation+0.000
results
Token results
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.424
Ċ
TokenĊ
Feature activation+0.073
The
TokenThe
Feature activation+0.000
proposals
Token proposals
Feature activation+0.000
were
Token were
Feature activation+0.000
made
Token made
Feature activation+0.000
whilst
Token whilst
Feature activation+0.000
still
Token still
Feature activation+0.000
in
Token in
Feature activation+0.000
your
Token your
Feature activation+0.000
mouth
Token mouth
Feature activation+1.677
,
Token,
Feature activation+2.431
particularly
Token particularly
Feature activation+0.000
with
Token with
Feature activation+0.000
carbohydrates
Token carbohydrates
Feature activation+0.327
.
Token.
Feature activation+0.000
If
Token If
Feature activation+0.000
least
Token least
Feature activation+0.000
a
Token a
Feature activation+0.000
second
Token second
Feature activation+0.000
season
Token season
Feature activation+0.000
.
Token.
Feature activation+0.910
Ċ
TokenĊ
Feature activation+2.272
Ċ
TokenĊ
Feature activation+0.085
Show
TokenShow
Feature activation+0.000
runners
Tokenrunners
Feature activation+0.000
Maur
Token Maur
Feature activation+0.000
issa
Tokenissa
Feature activation+0.000
containing
Token containing
Feature activation+0.000
fat
Token fat
Feature activation+0.099
,
Token,
Feature activation+0.000
glucose
Token glucose
Feature activation+2.922
,
Token,
Feature activation+0.498
protein
Token protein
Feature activation+2.395
and
Token and
Feature activation+0.000
placebo
Token placebo
Feature activation+0.212
.
Token.
Feature activation+0.207
Cognitive
Token Cognitive
Feature activation+0.096
performance
Token performance
Feature activation+0.342

INTERVAL 1.906 - 2.224
CONTAINS 0.002%

other
Token other
Feature activation+0.000
parts
Token parts
Feature activation+0.000
of
Token of
Feature activation+0.000
Britain
Token Britain
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.028
Ċ
TokenĊ
Feature activation+0.000
These
TokenThese
Feature activation+0.000
steps
Token steps
Feature activation+0.000
would
Token would
Feature activation+0.000
go
Token go
Feature activation+0.000
adaptive
Token adaptive
Feature activation+0.000
and
Token and
Feature activation+0.000
Bo
Token Bo
Feature activation+0.000
L
TokenL
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.017
Ċ
TokenĊ
Feature activation+0.325
As
TokenAs
Feature activation+0.000
pointed
Token pointed
Feature activation+0.000
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
regulators
Token regulators
Feature activation+1.172
,
Token,
Feature activation+0.209
entrepreneurs
Token entrepreneurs
Feature activation+1.179
,
Token,
Feature activation+0.129
policy
Token policy
Feature activation+0.131
makers
Token makers
Feature activation+2.126
and
Token and
Feature activation+0.000
academics
Token academics
Feature activation+0.868
to
Token to
Feature activation+0.080
explore
Token explore
Feature activation+0.000
blockchain
Token blockchain
Feature activation+0.223
surviving
Token surviving
Feature activation+0.058
Ken
Token Ken
Feature activation+0.000
y
Tokeny
Feature activation+0.000
ans
Tokenans
Feature activation+0.000
.
Token.
Feature activation+0.139
Ċ
TokenĊ
Feature activation+2.047
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
I
TokenI
Feature activation+0.000
think
Token think
Feature activation+0.000
they
Token they
Feature activation+0.000
resident
Token resident
Feature activation+0.000
citizens
Token citizens
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.909
Ċ
TokenĊ
Feature activation+0.000
T
TokenT
Feature activation+0.000
usk
Tokenusk
Feature activation+0.000
hit
Token hit
Feature activation+0.000
back
Token back
Feature activation+0.000

INTERVAL 1.589 - 1.906
CONTAINS 0.002%

on
Token on
Feature activation+0.386
national
Token national
Feature activation+0.000
security
Token security
Feature activation+0.017
grounds
Token grounds
Feature activation+0.135
.
Token.
Feature activation+0.266
Ċ
TokenĊ
Feature activation+1.707
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
broadcaster
Token broadcaster
Feature activation+0.000
was
Token was
Feature activation+0.000
given
Token given
Feature activation+0.000
still
Token still
Feature activation+0.000
has
Token has
Feature activation+0.000
$
Token $
Feature activation+0.000
970
Token970
Feature activation+1.346
.
Token.
Feature activation+0.152
2
Token2
Feature activation+1.590
m
Tokenm
Feature activation+0.000
in
Token in
Feature activation+0.000
cash
Token cash
Feature activation+0.000
,
Token,
Feature activation+0.000
equivalents
Token equivalents
Feature activation+0.102
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
FAR
Token FAR
Feature activation+0.000
J
TokenJ
Feature activation+0.038
.
Token.
Feature activation+0.108
Ċ
TokenĊ
Feature activation+1.756
Ċ
TokenĊ
Feature activation+0.027
The
TokenThe
Feature activation+0.000
PDF
Token PDF
Feature activation+0.000
can
Token can
Feature activation+0.000
be
Token be
Feature activation+0.016
ce
Tokence
Feature activation+0.000
?
Token?
Feature activation+0.000
Photograph
Token Photograph
Feature activation+0.087
:
Token:
Feature activation+0.000
PA
Token PA
Feature activation+0.243
Ċ
TokenĊ
Feature activation+1.591
Ċ
TokenĊ
Feature activation+0.619
Pres
TokenPres
Feature activation+0.043
umption
Tokenumption
Feature activation+0.000
replaces
Token replaces
Feature activation+0.000
routine
Token routine
Feature activation+0.000
food
Token food
Feature activation+0.000
whilst
Token whilst
Feature activation+0.000
still
Token still
Feature activation+0.000
in
Token in
Feature activation+0.000
your
Token your
Feature activation+0.000
mouth
Token mouth
Feature activation+1.677
,
Token,
Feature activation+2.431
particularly
Token particularly
Feature activation+0.000
with
Token with
Feature activation+0.000
carbohydrates
Token carbohydrates
Feature activation+0.327
.
Token.
Feature activation+0.000

INTERVAL 1.271 - 1.589
CONTAINS 0.005%

was
Token was
Feature activation+0.000
raised
Token raised
Feature activation+0.000
to
Token to
Feature activation+0.000
love
Token love
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.315
Ċ
TokenĊ
Feature activation+0.000
Even
TokenEven
Feature activation+0.000
so
Token so
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
lot
Token lot
Feature activation+0.000
about
Token about
Feature activation+0.000
prehistoric
Token prehistoric
Feature activation+0.000
life
Token life
Feature activation+0.000
.
Token.
Feature activation+0.145
Ċ
TokenĊ
Feature activation+1.311
Ċ
TokenĊ
Feature activation+0.000
Thanks
TokenThanks
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
latest
Token latest
Feature activation+0.000
awareness
Token awareness
Feature activation+0.010
of
Token of
Feature activation+0.000
human
Token human
Feature activation+0.075
rights
Token rights
Feature activation+0.061
.
Token.
Feature activation+0.012
Ċ
TokenĊ
Feature activation+1.420
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.033
publication
Token publication
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
spir
Token spir
Feature activation+0.637
alling
Tokenalling
Feature activation+0.371
private
Token private
Feature activation+0.353
rents
Token rents
Feature activation+0.362
.
Token.
Feature activation+0.520
Ċ
TokenĊ
Feature activation+1.580
Ċ
TokenĊ
Feature activation+0.508
Something
TokenSomething
Feature activation+0.337
else
Token else
Feature activation+0.269
that
Token that
Feature activation+0.237
is
Token is
Feature activation+0.427
400
Token400
Feature activation+0.000
m
Tokenm
Feature activation+0.259
but
Token but
Feature activation+0.000
serve
Token serve
Feature activation+0.000
between
Token between
Feature activation+0.000
10
Token 10
Feature activation+1.496
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.780
and
Token and
Feature activation+0.000
17
Token 17
Feature activation+0.380
,
Token,
Feature activation+0.000

INTERVAL 0.953 - 1.271
CONTAINS 0.011%

quick
Token quick
Feature activation+0.000
to
Token to
Feature activation+0.014
capital
Token capital
Feature activation+0.000
ise
Tokenise
Feature activation+0.000
:
Token:
Feature activation+0.250
Ċ
TokenĊ
Feature activation+0.990
Ċ
TokenĊ
Feature activation+0.000
Once
TokenOnce
Feature activation+0.000
word
Token word
Feature activation+0.399
got
Token got
Feature activation+0.445
around
Token around
Feature activation+0.000
almost
Token almost
Feature activation+0.000
400
Token 400
Feature activation+0.000
NHS
Token NHS
Feature activation+0.000
services
Token services
Feature activation+0.281
,
Token,
Feature activation+0.034
worth
Token worth
Feature activation+1.194
a
Token a
Feature activation+0.498
quarter
Token quarter
Feature activation+0.783
of
Token of
Feature activation+0.145
billion
Token billion
Feature activation+0.097
pounds
Token pounds
Feature activation+0.398
secure
Token secure
Feature activation+0.123
against
Token against
Feature activation+0.000
terrorism
Token terrorism
Feature activation+0.099
,
Token,
Feature activation+0.027
climate
Token climate
Feature activation+0.000
change
Token change
Feature activation+1.114
,
Token,
Feature activation+0.384
unfair
Token unfair
Feature activation+0.000
wage
Token wage
Feature activation+0.026
competition
Token competition
Feature activation+0.933
and
Token and
Feature activation+0.068
favourable
Token favourable
Feature activation+0.000
to
Token to
Feature activation+0.000
jelly
Token jelly
Feature activation+0.000
fish
Tokenfish
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.980
Ċ
TokenĊ
Feature activation+0.000
Though
TokenThough
Feature activation+0.000
some
Token some
Feature activation+0.000
species
Token species
Feature activation+0.000
are
Token are
Feature activation+0.000
case
Token case
Feature activation+0.000
,
Token,
Feature activation+0.000
330
Token 330
Feature activation+0.000
m
Tokenm
Feature activation+0.000
citizens
Token citizens
Feature activation+0.132
.
Token.
Feature activation+1.053
Ċ
TokenĊ
Feature activation+2.129
Ċ
TokenĊ
Feature activation+0.601
But
TokenBut
Feature activation+0.225
if
Token if
Feature activation+0.437
you
Token you
Feature activation+0.000

INTERVAL 0.635 - 0.953
CONTAINS 0.033%

Royal
Token Royal
Feature activation+0.000
Air
Token Air
Feature activation+0.000
Force
Token Force
Feature activation+0.000
have
Token have
Feature activation+0.181
now
Token now
Feature activation+0.586
confirmed
Token confirmed
Feature activation+0.730
to
Token to
Feature activation+0.594
the
Token the
Feature activation+0.520
UK
Token UK
Feature activation+0.072
Defence
Token Defence
Feature activation+0.000
Journal
Token Journal
Feature activation+0.611
I
Token I
Feature activation+0.000
hear
Token hear
Feature activation+0.000
'
Token '
Feature activation+0.158
sp
Tokensp
Feature activation+0.040
o
Tokeno
Feature activation+0.000
iled
Tokeniled
Feature activation+0.671
',
Token',
Feature activation+0.351
I
Token I
Feature activation+0.019
hear
Token hear
Feature activation+0.154
'
Token '
Feature activation+0.144
m
Tokenm
Feature activation+0.000
.
Token.
Feature activation+0.924
He
Token He
Feature activation+1.105
was
Token was
Feature activation+0.903
not
Token not
Feature activation+0.658
arrested
Token arrested
Feature activation+0.695
.
Token.
Feature activation+0.768
Ċ
TokenĊ
Feature activation+0.103
Ċ
TokenĊ
Feature activation+0.000
'
Token'
Feature activation+0.000
Operation
TokenOperation
Feature activation+0.052
Gold
Token Gold
Feature activation+0.000
the
Token the
Feature activation+0.000
bl
Token bl
Feature activation+0.000
oke
Tokenoke
Feature activation+0.220
sometimes
Token sometimes
Feature activation+0.238
,
Token,
Feature activation+1.052
but
Token but
Feature activation+0.866
as
Token as
Feature activation+0.809
a
Token a
Feature activation+0.433
farmer
Token farmer
Feature activation+0.174
or
Token or
Feature activation+0.750
a
Token a
Feature activation+0.354
Muslim
Token Muslim
Feature activation+0.000
population
Token population
Feature activation+0.000
is
Token is
Feature activation+0.000
greater
Token greater
Feature activation+0.000
than
Token than
Feature activation+0.000
250
Token 250
Feature activation+0.897
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.217
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
is
Token is
Feature activation+0.000

INTERVAL 0.318 - 0.635
CONTAINS 0.149%

a
Token a
Feature activation+0.000
vacuum
Token vacuum
Feature activation+0.000
cleaner
Token cleaner
Feature activation+0.000
salesman
Token salesman
Feature activation+0.000
:
Token:
Feature activation+0.133
Ċ
TokenĊ
Feature activation+0.332
Ċ
TokenĊ
Feature activation+0.071
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Imagine
TokenImagine
Feature activation+0.000
,
Token,
Feature activation+0.000
they
Token they
Feature activation+0.000
can
Token can
Feature activation+0.000
use
Token use
Feature activation+0.000
high
Token high
Feature activation+0.102
resolution
Token resolution
Feature activation+0.000
random
Token random
Feature activation+0.337
isation
Tokenisation
Feature activation+0.000
in
Token in
Feature activation+0.000
parts
Token parts
Feature activation+0.072
of
Token of
Feature activation+0.168
the
Token the
Feature activation+0.266
built
Token built
Feature activation+0.000
their
Token their
Feature activation+0.000
own
Token own
Feature activation+0.085
renewable
Token renewable
Feature activation+0.254
power
Token power
Feature activation+0.278
station
Token station
Feature activation+0.533
that
Token that
Feature activation+0.276
uses
Token uses
Feature activation+0.310
recycled
Token recycled
Feature activation+0.297
heat
Token heat
Feature activation+0.221
from
Token from
Feature activation+0.558
a
Token a
Feature activation+0.000
Europe
Token Europe
Feature activation+0.000
that
Token that
Feature activation+0.000
protects
Token protects
Feature activation+0.000
".
Token".
Feature activation+0.000
This
Token This
Feature activation+0.477
,
Token,
Feature activation+0.541
he
Token he
Feature activation+0.532
said
Token said
Feature activation+0.087
,
Token,
Feature activation+0.000
meant
Token meant
Feature activation+0.251
ansson
Tokenansson
Feature activation+0.271
,
Token,
Feature activation+0.000
Robert
Token Robert
Feature activation+0.000
Down
Token Down
Feature activation+0.000
ey
Tokeney
Feature activation+0.000
Jr
Token Jr
Feature activation+0.548
and
Token and
Feature activation+0.000
Samuel
Token Samuel
Feature activation+0.000
L
Token L
Feature activation+0.000
Jackson
Token Jackson
Feature activation+0.000
are
Token are
Feature activation+0.000

INTERVAL 0.000 - 0.318
CONTAINS 99.797%

the
Token the
Feature activation+0.000
aud
Token aud
Feature activation+0.000
acity
Tokenacity
Feature activation+0.000
not
Token not
Feature activation+0.000
to
Token to
Feature activation+0.000
feel
Token feel
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
represented
Tokenrepresented
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
made
Token made
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
is
Token is
Feature activation+0.000
stretched
Token stretched
Feature activation+0.000
out
Token out
Feature activation+0.000
to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
it
Token it
Feature activation+0.000
small
Token small
Feature activation+0.000
enough
Token enough
Feature activation+0.000
.
Token.
Feature activation+0.000
There
Token There
Feature activation+0.000
have
Token have
Feature activation+0.000
been
Token been
Feature activation+0.000
sour
Token sour
Feature activation+0.000
beers
Token beers
Feature activation+0.000
of
Token of
Feature activation+0.000
various
Token various
Feature activation+0.000
kinds
Token kinds
Feature activation+0.000
from
Token from
Feature activation+0.000
almost
Token almost
Feature activation+0.000
elsewhere
Token elsewhere
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
USA
Token USA
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ex
TokenEx
Feature activation+0.000
-
Token-
Feature activation+0.000
cand
Tokencand
Feature activation+0.000
idate
Tokenidate
Feature activation+0.000
"
Token "
Feature activation+0.000
in
Tokenin
Feature activation+0.000
flation
Tokenflation
Feature activation+0.000
will
Token will
Feature activation+0.000
continue
Token continue
Feature activation+0.000
to
Token to
Feature activation+0.000
increase
Token increase
Feature activation+0.000
significantly
Token significantly
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
U
Token U
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 H8.2: (feature 6375)

TOP ACTIVATIONS
MAX = 6.906

without
Token without
Feature activation+0.000
sanitation
Token sanitation
Feature activation+0.000
at
Token at
Feature activation+0.000
French
Token French
Feature activation+0.000
port
Token port
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.906
Ċ
TokenĊ
Feature activation+2.859
Facebook
TokenFacebook
Feature activation+3.002
Twitter
Token Twitter
Feature activation+0.000
Pinterest
Token Pinterest
Feature activation+0.000
Calais
Token Calais
Feature activation+0.000
burden
Token burden
Feature activation+0.000
of
Token of
Feature activation+0.000
eating
Token eating
Feature activation+0.000
real
Token real
Feature activation+0.000
food
Token food
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.815
Ċ
TokenĊ
Feature activation+0.740
It
TokenIt
Feature activation+0.563
is
Token is
Feature activation+0.000
late
Token late
Feature activation+0.000
evening
Token evening
Feature activation+0.000
place
Token place
Feature activation+0.000
between
Token between
Feature activation+0.000
December
Token December
Feature activation+0.000
and
Token and
Feature activation+0.000
February
Token February
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.276
Ċ
TokenĊ
Feature activation+1.483
Wal
TokenWal
Feature activation+0.000
ter
Tokenter
Feature activation+0.000
De
Token De
Feature activation+0.000
Greg
Token Greg
Feature activation+0.000
the
Token the
Feature activation+0.000
mid
Token mid
Feature activation+0.000
-
Token-
Feature activation+0.000
Atlantic
TokenAtlantic
Feature activation+0.000
region
Token region
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.972
Ċ
TokenĊ
Feature activation+0.903
An
TokenAn
Feature activation+0.890
earthquake
Token earthquake
Feature activation+0.000
has
Token has
Feature activation+1.571
j
Token j
Feature activation+0.000
Toyota
Token Toyota
Feature activation+0.000
Cor
Token Cor
Feature activation+0.000
olla
Tokenolla
Feature activation+0.000
while
Token while
Feature activation+0.000
fleeing
Token fleeing
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.395
Ċ
TokenĊ
Feature activation+0.216
Like
TokenLike
Feature activation+0.000
many
Token many
Feature activation+0.000
residents
Token residents
Feature activation+0.000
of
Token of
Feature activation+0.000
of
Token of
Feature activation+0.000
damage
Token damage
Feature activation+0.000
or
Token or
Feature activation+0.000
injuries
Token injuries
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.892
Ċ
TokenĊ
Feature activation+0.000
U
TokenU
Feature activation+0.000
ps
Tokenps
Feature activation+0.000
ur
Tokenur
Feature activation+0.000
ge
Tokenge
Feature activation+0.000
the
Token the
Feature activation+0.000
development
Token development
Feature activation+0.000
of
Token of
Feature activation+0.000
artificial
Token artificial
Feature activation+0.000
alternatives
Token alternatives
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.514
Ċ
TokenĊ
Feature activation+0.000
WASHINGTON
TokenWASHINGTON
Feature activation+0.995
,
Token,
Feature activation+0.750
D
Token D
Feature activation+0.000
.
Token.
Feature activation+0.000
the
Token the
Feature activation+0.000
movie
Token movie
Feature activation+0.000
opening
Token opening
Feature activation+0.000
Christmas
Token Christmas
Feature activation+0.000
day
Token day
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.486
Ċ
TokenĊ
Feature activation+0.000
V
TokenV
Feature activation+0.000
icious
Tokenicious
Feature activation+0.000
ly
Tokenly
Feature activation+0.000
m
Token m
Feature activation+0.000
murder
Token murder
Feature activation+0.000
of
Token of
Feature activation+0.000
Trayvon
Token Trayvon
Feature activation+0.000
Martin
Token Martin
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.198
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
police
Token police
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
could
Token could
Feature activation+0.000
not
Token not
Feature activation+0.000
see
Token see
Feature activation+0.000
the
Token the
Feature activation+0.000
salute
Token salute
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.093
Ċ
TokenĊ
Feature activation+0.035
M
TokenM
Feature activation+0.000
ilo
Tokenilo
Feature activation+0.000
Yiannopoulos
Token Yiannopoulos
Feature activation+0.000
has
Token has
Feature activation+0.343
at
Token at
Feature activation+0.000
French
Token French
Feature activation+0.000
port
Token port
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.906
Ċ
TokenĊ
Feature activation+2.859
Facebook
TokenFacebook
Feature activation+3.002
Twitter
Token Twitter
Feature activation+0.000
Pinterest
Token Pinterest
Feature activation+0.000
Calais
Token Calais
Feature activation+0.000
migrants
Token migrants
Feature activation+0.000
:
Token:
Feature activation+1.010
if
Token if
Feature activation+0.000
George
Token George
Feature activation+0.000
Zimmerman
Token Zimmerman
Feature activation+0.000
is
Token is
Feature activation+0.000
acquitted
Token acquitted
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.900
Ċ
TokenĊ
Feature activation+0.000
Police
TokenPolice
Feature activation+0.000
in
Token in
Feature activation+0.000
Sanford
Token Sanford
Feature activation+0.000
,
Token,
Feature activation+0.994
handling
Token handling
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
case
Token case
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.888
Ċ
TokenĊ
Feature activation+0.000
Now
TokenNow
Feature activation+0.000
they
Token they
Feature activation+0.000
are
Token are
Feature activation+0.000
taking
Token taking
Feature activation+0.000
sanitation
Token sanitation
Feature activation+0.000
at
Token at
Feature activation+0.000
French
Token French
Feature activation+0.000
port
Token port
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.906
Ċ
TokenĊ
Feature activation+2.859
Facebook
TokenFacebook
Feature activation+3.002
Twitter
Token Twitter
Feature activation+0.000
Pinterest
Token Pinterest
Feature activation+0.000
Calais
Token Calais
Feature activation+0.000
migrants
Token migrants
Feature activation+0.000
Roman
Token Roman
Feature activation+0.000
farm
Token farm
Feature activation+0.000
stead
Tokenstead
Feature activation+0.000
centuries
Token centuries
Feature activation+0.000
ago
Token ago
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.804
Ċ
TokenĊ
Feature activation+0.000
©
Token©
Feature activation+0.000
Wor
Token Wor
Feature activation+0.000
ces
Tokences
Feature activation+0.000
ters
Tokenters
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
House
Token House
Feature activation+0.000
of
Token of
Feature activation+0.000
Commons
Token Commons
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.743
Ċ
TokenĊ
Feature activation+0.000
Instead
TokenInstead
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
believes
Token believes
Feature activation+0.000
are
Token are
Feature activation+0.000
heard
Token heard
Feature activation+0.000
inside
Token inside
Feature activation+0.000
Westminster
Token Westminster
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.605
Ċ
TokenĊ
Feature activation+0.000
Mr
TokenMr
Feature activation+0.118
Carroll
Token Carroll
Feature activation+0.000
also
Token also
Feature activation+0.000
denied
Token denied
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
restaurant
Token restaurant
Feature activation+0.000
on
Token on
Feature activation+0.042
Christmas
Token Christmas
Feature activation+0.000
Day
Token Day
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.603
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.600
campaign
Token campaign
Feature activation+0.105
to
Token to
Feature activation+0.355
fly
Token fly
Feature activation+0.026
from
Token from
Feature activation+0.000
whistleblower
Token whistleblower
Feature activation+0.000
Edward
Token Edward
Feature activation+0.000
Snowden
Token Snowden
Feature activation+0.000
.
Token.
Feature activation+0.096
Ċ
TokenĊ
Feature activation+2.414
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
proposals
Token proposals
Feature activation+0.000
range
Token range
Feature activation+0.000
from
Token from
Feature activation+0.000
RS
TokenRS
Feature activation+0.000
before
Token before
Feature activation+0.000
fatal
Token fatal
Feature activation+0.000
heart
Token heart
Feature activation+0.000
attack
Token attack
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.359
Ċ
TokenĊ
Feature activation+0.000
Actor
TokenActor
Feature activation+0.000
,
Token,
Feature activation+0.000
51
Token 51
Feature activation+0.323
,
Token,
Feature activation+0.000

Top DFA by src position
MAX = 7.626

without
Token without
Feature activation-0.004
Top resid features:
sanitation
Token sanitation
Feature activation-0.002
Top resid features:
at
Token at
Feature activation+0.030
Top resid features:
French
Token French
Feature activation-0.015
Top resid features:
port
Token port
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+7.626
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Facebook
TokenFacebook
Feature activation+0.000
Top resid features:
Twitter
Token Twitter
Feature activation+0.000
Top resid features:
Pinterest
Token Pinterest
Feature activation+0.000
Top resid features:
Calais
Token Calais
Feature activation+0.000
Top resid features:
burden
Token burden
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
eating
Token eating
Feature activation-0.004
Top resid features:
real
Token real
Feature activation+0.000
Top resid features:
food
Token food
Feature activation-0.005
Top resid features:
Ċ
TokenĊ
Feature activation+6.593
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
It
TokenIt
Feature activation+0.000
Top resid features:
is
Token is
Feature activation+0.000
Top resid features:
late
Token late
Feature activation+0.000
Top resid features:
evening
Token evening
Feature activation+0.000
Top resid features:
place
Token place
Feature activation-0.005
Top resid features:
between
Token between
Feature activation-0.008
Top resid features:
December
Token December
Feature activation-0.003
Top resid features:
and
Token and
Feature activation-0.006
Top resid features:
February
Token February
Feature activation-0.008
Top resid features:
Ċ
TokenĊ
Feature activation+6.351
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Wal
TokenWal
Feature activation+0.000
Top resid features:
ter
Tokenter
Feature activation+0.000
Top resid features:
De
Token De
Feature activation+0.000
Top resid features:
Greg
Token Greg
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.006
Top resid features:
mid
Token mid
Feature activation+0.003
Top resid features:
-
Token-
Feature activation+0.002
Top resid features:
Atlantic
TokenAtlantic
Feature activation+0.014
Top resid features:
region
Token region
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+5.667
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
An
TokenAn
Feature activation+0.000
Top resid features:
earthquake
Token earthquake
Feature activation+0.000
Top resid features:
has
Token has
Feature activation+0.000
Top resid features:
j
Token j
Feature activation+0.000
Top resid features:
Toyota
Token Toyota
Feature activation-0.011
Top resid features:
Cor
Token Cor
Feature activation-0.002
Top resid features:
olla
Tokenolla
Feature activation-0.004
Top resid features:
while
Token while
Feature activation-0.009
Top resid features:
fleeing
Token fleeing
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+4.967
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Like
TokenLike
Feature activation+0.000
Top resid features:
many
Token many
Feature activation+0.000
Top resid features:
residents
Token residents
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
mid
Token mid
Feature activation+0.000
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
Atlantic
TokenAtlantic
Feature activation+0.002
Top resid features:
region
Token region
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+4.466
Top resid features:
Ċ
TokenĊ
Feature activation+0.344
Top resid features:
An
TokenAn
Feature activation-0.013
Top resid features:
earthquake
Token earthquake
Feature activation-0.057
Top resid features:
has
Token has
Feature activation-0.003
Top resid features:
j
Token j
Feature activation+0.003
Top resid features:
the
Token the
Feature activation-0.007
Top resid features:
development
Token development
Feature activation+0.003
Top resid features:
of
Token of
Feature activation-0.008
Top resid features:
artificial
Token artificial
Feature activation-0.009
Top resid features:
alternatives
Token alternatives
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+4.156
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
WASHINGTON
TokenWASHINGTON
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
D
Token D
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.006
Top resid features:
movie
Token movie
Feature activation-0.025
Top resid features:
opening
Token opening
Feature activation-0.008
Top resid features:
Christmas
Token Christmas
Feature activation-0.012
Top resid features:
day
Token day
Feature activation-0.004
Top resid features:
Ċ
TokenĊ
Feature activation+4.216
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
V
TokenV
Feature activation+0.000
Top resid features:
icious
Tokenicious
Feature activation+0.000
Top resid features:
ly
Tokenly
Feature activation+0.000
Top resid features:
m
Token m
Feature activation+0.000
Top resid features:
if
Token if
Feature activation+0.004
Top resid features:
George
Token George
Feature activation+0.003
Top resid features:
Zimmerman
Token Zimmerman
Feature activation-0.002
Top resid features:
is
Token is
Feature activation+0.005
Top resid features:
acquitted
Token acquitted
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+3.428
Top resid features:
Ċ
TokenĊ
Feature activation+0.755
Top resid features:
Police
TokenPolice
Feature activation-0.011
Top resid features:
in
Token in
Feature activation-0.015
Top resid features:
Sanford
Token Sanford
Feature activation-0.017
Top resid features:
,
Token,
Feature activation+0.010
Top resid features:
could
Token could
Feature activation+0.000
Top resid features:
not
Token not
Feature activation+0.002
Top resid features:
see
Token see
Feature activation-0.003
Top resid features:
the
Token the
Feature activation-0.007
Top resid features:
salute
Token salute
Feature activation-0.021
Top resid features:
Ċ
TokenĊ
Feature activation+3.747
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
M
TokenM
Feature activation+0.000
Top resid features:
ilo
Tokenilo
Feature activation+0.000
Top resid features:
Yiannopoulos
Token Yiannopoulos
Feature activation+0.000
Top resid features:
has
Token has
Feature activation+0.000
Top resid features:
without
Token without
Feature activation+0.005
Top resid features:
sanitation
Token sanitation
Feature activation+0.006
Top resid features:
at
Token at
Feature activation+0.006
Top resid features:
French
Token French
Feature activation+0.000
Top resid features:
port
Token port
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation+2.810
Top resid features:
Ċ
TokenĊ
Feature activation+0.944
Top resid features:
Facebook
TokenFacebook
Feature activation-0.035
Top resid features:
Twitter
Token Twitter
Feature activation+0.000
Top resid features:
Pinterest
Token Pinterest
Feature activation+0.000
Top resid features:
Calais
Token Calais
Feature activation+0.000
Top resid features:
if
Token if
Feature activation-0.001
Top resid features:
George
Token George
Feature activation-0.004
Top resid features:
Zimmerman
Token Zimmerman
Feature activation-0.009
Top resid features:
is
Token is
Feature activation+0.001
Top resid features:
acquitted
Token acquitted
Feature activation-0.006
Top resid features:
Ċ
TokenĊ
Feature activation+3.913
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Police
TokenPolice
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
Sanford
Token Sanford
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
if
Token if
Feature activation+0.003
Top resid features:
George
Token George
Feature activation+0.004
Top resid features:
Zimmerman
Token Zimmerman
Feature activation-0.001
Top resid features:
is
Token is
Feature activation+0.004
Top resid features:
acquitted
Token acquitted
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation+3.260
Top resid features:
Ċ
TokenĊ
Feature activation+0.746
Top resid features:
Police
TokenPolice
Feature activation-0.003
Top resid features:
in
Token in
Feature activation-0.006
Top resid features:
Sanford
Token Sanford
Feature activation-0.005
Top resid features:
,
Token,
Feature activation+0.004
Top resid features:
without
Token without
Feature activation-0.005
Top resid features:
sanitation
Token sanitation
Feature activation-0.024
Top resid features:
at
Token at
Feature activation+0.036
Top resid features:
French
Token French
Feature activation-0.063
Top resid features:
port
Token port
Feature activation-0.032
Top resid features:
Ċ
TokenĊ
Feature activation+3.625
Top resid features:
Ċ
TokenĊ
Feature activation+0.520
Top resid features:
Facebook
TokenFacebook
Feature activation+0.000
Top resid features:
Twitter
Token Twitter
Feature activation+0.000
Top resid features:
Pinterest
Token Pinterest
Feature activation+0.000
Top resid features:
Calais
Token Calais
Feature activation+0.000
Top resid features:
Roman
Token Roman
Feature activation+0.024
Top resid features:
farm
Token farm
Feature activation-0.003
Top resid features:
stead
Tokenstead
Feature activation-0.035
Top resid features:
centuries
Token centuries
Feature activation-0.046
Top resid features:
ago
Token ago
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+3.599
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
©
Token©
Feature activation+0.000
Top resid features:
Wor
Token Wor
Feature activation+0.000
Top resid features:
ces
Tokences
Feature activation+0.000
Top resid features:
ters
Tokenters
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.009
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
House
Token House
Feature activation-0.022
Top resid features:
of
Token of
Feature activation+0.019
Top resid features:
Commons
Token Commons
Feature activation-0.022
Top resid features:
Ċ
TokenĊ
Feature activation+3.868
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Instead
TokenInstead
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
he
Token he
Feature activation+0.000
Top resid features:
believes
Token believes
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
House
Token House
Feature activation-0.013
Top resid features:
of
Token of
Feature activation+0.012
Top resid features:
Commons
Token Commons
Feature activation-0.004
Top resid features:
Ċ
TokenĊ
Feature activation+3.270
Top resid features:
Ċ
TokenĊ
Feature activation+0.322
Top resid features:
Instead
TokenInstead
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
he
Token he
Feature activation+0.026
Top resid features:
believes
Token believes
Feature activation+0.002
Top resid features:
Ļ
TokenĻ
Feature activation-0.009
Top resid features:
restaurant
Token restaurant
Feature activation-0.023
Top resid features:
on
Token on
Feature activation-0.016
Top resid features:
Christmas
Token Christmas
Feature activation+0.010
Top resid features:
Day
Token Day
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+1.888
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
A
TokenA
Feature activation+0.000
Top resid features:
campaign
Token campaign
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
fly
Token fly
Feature activation+0.000
Top resid features:
powers
Token powers
Feature activation-0.001
Top resid features:
since
Token since
Feature activation+0.001
Top resid features:
9
Token 9
Feature activation+0.001
Top resid features:
/
Token/
Feature activation+0.001
Top resid features:
11
Token11
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+3.176
Top resid features:
Ċ
TokenĊ
Feature activation+0.131
Top resid features:
Members
TokenMembers
Feature activation-0.007
Top resid features:
of
Token of
Feature activation-0.002
Top resid features:
Congress
Token Congress
Feature activation-0.013
Top resid features:
are
Token are
Feature activation+0.004
Top resid features:
RS
TokenRS
Feature activation-0.032
Top resid features:
before
Token before
Feature activation-0.013
Top resid features:
fatal
Token fatal
Feature activation+0.011
Top resid features:
heart
Token heart
Feature activation-0.001
Top resid features:
attack
Token attack
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+3.276
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Actor
TokenActor
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
51
Token 51
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.01

Head 2: 0.55

Head 3: 0.04

Head 4: 0.02

Head 5: 0.05

Head 6: 0.03

Head 7: 0.03

Head 8: 0.04

Head 9: 0.04

Head 10: 0.07

Head 11: 0.04

Positive logits

4.77

Nato4.63

Facebook4.31

Nasa4.08

maths4.03

­4.00

organising3.99

Miliband3.99

ageing3.96

criticised3.82

Ukip3.73

apologised3.70

emphas3.70

scept3.64

NHS3.64

recognise3.59

analys3.58

organisers3.55

Corbyn3.50

£3.49

Negative logits

,)-3.96

......-3.76

.---3.57

.......-3.50

"...-3.44

Operator-3.42

......-3.40

.....-3.34

-3.28

utilize-3.23

honorable-3.18

----3.15

....-3.07

....-3.03

huh-3.01

alot-2.99

INAL-2.98

.?-2.97

COUNTY-2.94

somet-2.89

INTERVAL 6.215 - 6.906
CONTAINS 0.000%

without
Token without
Feature activation+0.000
sanitation
Token sanitation
Feature activation+0.000
at
Token at
Feature activation+0.000
French
Token French
Feature activation+0.000
port
Token port
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.906
Ċ
TokenĊ
Feature activation+2.859
Facebook
TokenFacebook
Feature activation+3.002
Twitter
Token Twitter
Feature activation+0.000
Pinterest
Token Pinterest
Feature activation+0.000
Calais
Token Calais
Feature activation+0.000

INTERVAL 5.524 - 6.215
CONTAINS 0.000%

burden
Token burden
Feature activation+0.000
of
Token of
Feature activation+0.000
eating
Token eating
Feature activation+0.000
real
Token real
Feature activation+0.000
food
Token food
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.815
Ċ
TokenĊ
Feature activation+0.740
It
TokenIt
Feature activation+0.563
is
Token is
Feature activation+0.000
late
Token late
Feature activation+0.000
evening
Token evening
Feature activation+0.000

INTERVAL 4.834 - 5.524
CONTAINS 0.000%

place
Token place
Feature activation+0.000
between
Token between
Feature activation+0.000
December
Token December
Feature activation+0.000
and
Token and
Feature activation+0.000
February
Token February
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.276
Ċ
TokenĊ
Feature activation+1.483
Wal
TokenWal
Feature activation+0.000
ter
Tokenter
Feature activation+0.000
De
Token De
Feature activation+0.000
Greg
Token Greg
Feature activation+0.000
the
Token the
Feature activation+0.000
mid
Token mid
Feature activation+0.000
-
Token-
Feature activation+0.000
Atlantic
TokenAtlantic
Feature activation+0.000
region
Token region
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.972
Ċ
TokenĊ
Feature activation+0.903
An
TokenAn
Feature activation+0.890
earthquake
Token earthquake
Feature activation+0.000
has
Token has
Feature activation+1.571
j
Token j
Feature activation+0.000

INTERVAL 4.143 - 4.834
CONTAINS 0.000%

Toyota
Token Toyota
Feature activation+0.000
Cor
Token Cor
Feature activation+0.000
olla
Tokenolla
Feature activation+0.000
while
Token while
Feature activation+0.000
fleeing
Token fleeing
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.395
Ċ
TokenĊ
Feature activation+0.216
Like
TokenLike
Feature activation+0.000
many
Token many
Feature activation+0.000
residents
Token residents
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 3.453 - 4.143
CONTAINS 0.000%

the
Token the
Feature activation+0.000
movie
Token movie
Feature activation+0.000
opening
Token opening
Feature activation+0.000
Christmas
Token Christmas
Feature activation+0.000
day
Token day
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.486
Ċ
TokenĊ
Feature activation+0.000
V
TokenV
Feature activation+0.000
icious
Tokenicious
Feature activation+0.000
ly
Tokenly
Feature activation+0.000
m
Token m
Feature activation+0.000
of
Token of
Feature activation+0.000
damage
Token damage
Feature activation+0.000
or
Token or
Feature activation+0.000
injuries
Token injuries
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.892
Ċ
TokenĊ
Feature activation+0.000
U
TokenU
Feature activation+0.000
ps
Tokenps
Feature activation+0.000
ur
Tokenur
Feature activation+0.000
ge
Tokenge
Feature activation+0.000
the
Token the
Feature activation+0.000
development
Token development
Feature activation+0.000
of
Token of
Feature activation+0.000
artificial
Token artificial
Feature activation+0.000
alternatives
Token alternatives
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.514
Ċ
TokenĊ
Feature activation+0.000
WASHINGTON
TokenWASHINGTON
Feature activation+0.995
,
Token,
Feature activation+0.750
D
Token D
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 2.762 - 3.453
CONTAINS 0.001%

Roman
Token Roman
Feature activation+0.000
farm
Token farm
Feature activation+0.000
stead
Tokenstead
Feature activation+0.000
centuries
Token centuries
Feature activation+0.000
ago
Token ago
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.804
Ċ
TokenĊ
Feature activation+0.000
©
Token©
Feature activation+0.000
Wor
Token Wor
Feature activation+0.000
ces
Tokences
Feature activation+0.000
ters
Tokenters
Feature activation+0.000
if
Token if
Feature activation+0.000
George
Token George
Feature activation+0.000
Zimmerman
Token Zimmerman
Feature activation+0.000
is
Token is
Feature activation+0.000
acquitted
Token acquitted
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.900
Ċ
TokenĊ
Feature activation+0.000
Police
TokenPolice
Feature activation+0.000
in
Token in
Feature activation+0.000
Sanford
Token Sanford
Feature activation+0.000
,
Token,
Feature activation+0.994
could
Token could
Feature activation+0.000
not
Token not
Feature activation+0.000
see
Token see
Feature activation+0.000
the
Token the
Feature activation+0.000
salute
Token salute
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.093
Ċ
TokenĊ
Feature activation+0.035
M
TokenM
Feature activation+0.000
ilo
Tokenilo
Feature activation+0.000
Yiannopoulos
Token Yiannopoulos
Feature activation+0.000
has
Token has
Feature activation+0.343
sanitation
Token sanitation
Feature activation+0.000
at
Token at
Feature activation+0.000
French
Token French
Feature activation+0.000
port
Token port
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.906
Ċ
TokenĊ
Feature activation+2.859
Facebook
TokenFacebook
Feature activation+3.002
Twitter
Token Twitter
Feature activation+0.000
Pinterest
Token Pinterest
Feature activation+0.000
Calais
Token Calais
Feature activation+0.000
migrants
Token migrants
Feature activation+0.000
handling
Token handling
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
case
Token case
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.888
Ċ
TokenĊ
Feature activation+0.000
Now
TokenNow
Feature activation+0.000
they
Token they
Feature activation+0.000
are
Token are
Feature activation+0.000
taking
Token taking
Feature activation+0.000

INTERVAL 2.072 - 2.762
CONTAINS 0.001%

are
Token are
Feature activation+0.000
heard
Token heard
Feature activation+0.000
inside
Token inside
Feature activation+0.000
Westminster
Token Westminster
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.605
Ċ
TokenĊ
Feature activation+0.000
Mr
TokenMr
Feature activation+0.118
Carroll
Token Carroll
Feature activation+0.000
also
Token also
Feature activation+0.000
denied
Token denied
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
House
Token House
Feature activation+0.000
of
Token of
Feature activation+0.000
Commons
Token Commons
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.743
Ċ
TokenĊ
Feature activation+0.000
Instead
TokenInstead
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
believes
Token believes
Feature activation+0.000
Wal
TokenWal
Feature activation+0.000
ter
Tokenter
Feature activation+0.000
De
Token De
Feature activation+0.000
Greg
Token Greg
Feature activation+0.000
orio
Tokenorio
Feature activation+1.957
has
Token has
Feature activation+2.328
left
Token left
Feature activation+0.960
his
Token his
Feature activation+0.000
post
Token post
Feature activation+0.901
as
Token as
Feature activation+0.222
Fifa
Token Fifa
Feature activation+0.153
RS
TokenRS
Feature activation+0.000
before
Token before
Feature activation+0.000
fatal
Token fatal
Feature activation+0.000
heart
Token heart
Feature activation+0.000
attack
Token attack
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.359
Ċ
TokenĊ
Feature activation+0.000
Actor
TokenActor
Feature activation+0.000
,
Token,
Feature activation+0.000
51
Token 51
Feature activation+0.323
,
Token,
Feature activation+0.000
from
Token from
Feature activation+0.000
whistleblower
Token whistleblower
Feature activation+0.000
Edward
Token Edward
Feature activation+0.000
Snowden
Token Snowden
Feature activation+0.000
.
Token.
Feature activation+0.096
Ċ
TokenĊ
Feature activation+2.414
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
proposals
Token proposals
Feature activation+0.000
range
Token range
Feature activation+0.000
from
Token from
Feature activation+0.000

INTERVAL 1.381 - 2.072
CONTAINS 0.003%

'
Token'
Feature activation+0.000
legal
Token legal
Feature activation+0.000
interests
Token interests
Feature activation+0.000
abroad
Token abroad
Feature activation+0.000
".
Token".
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.587
Ċ
TokenĊ
Feature activation+0.000
Mr
TokenMr
Feature activation+0.000
Lu
Token Lu
Feature activation+0.000
said
Token said
Feature activation+0.000
he
Token he
Feature activation+0.000
version
Token version
Feature activation+0.000
of
Token of
Feature activation+0.000
Open
Token Open
Feature activation+0.000
SSL
TokenSSL
Feature activation+0.000
software
Token software
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.804
Ċ
TokenĊ
Feature activation+0.000
F
TokenF
Feature activation+0.000
law
Tokenlaw
Feature activation+0.000
allows
Token allows
Feature activation+0.000
hackers
Token hackers
Feature activation+0.000
between
Token between
Feature activation+0.000
December
Token December
Feature activation+0.000
and
Token and
Feature activation+0.000
February
Token February
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.276
Ċ
TokenĊ
Feature activation+1.483
Wal
TokenWal
Feature activation+0.000
ter
Tokenter
Feature activation+0.000
De
Token De
Feature activation+0.000
Greg
Token Greg
Feature activation+0.000
orio
Tokenorio
Feature activation+1.957
used
Token used
Feature activation+0.000
the
Token the
Feature activation+0.000
strategy
Token strategy
Feature activation+0.000
twice
Token twice
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.779
Ċ
TokenĊ
Feature activation+0.000
China
TokenChina
Feature activation+0.000
National
Token National
Feature activation+0.000
Building
Token Building
Feature activation+0.000
Materials
Token Materials
Feature activation+0.000
enshr
Token enshr
Feature activation+0.000
ined
Tokenined
Feature activation+0.000
in
Token in
Feature activation+0.000
US
Token US
Feature activation+0.000
law
Token law
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.031
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
claim
Token claim
Feature activation+0.000
of
Token of
Feature activation+0.000
sovereign
Token sovereign
Feature activation+0.000

INTERVAL 0.691 - 1.381
CONTAINS 0.005%

Greg
Token Greg
Feature activation+0.000
orio
Tokenorio
Feature activation+1.957
has
Token has
Feature activation+2.328
left
Token left
Feature activation+0.960
his
Token his
Feature activation+0.000
post
Token post
Feature activation+0.901
as
Token as
Feature activation+0.222
Fifa
Token Fifa
Feature activation+0.153
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Facebook
TokenFacebook
Feature activation+3.002
Twitter
Token Twitter
Feature activation+0.000
Pinterest
Token Pinterest
Feature activation+0.000
Calais
Token Calais
Feature activation+0.000
migrants
Token migrants
Feature activation+0.000
:
Token:
Feature activation+1.010
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
Get
TokenGet
Feature activation+0.000
to
Token to
Feature activation+0.000
England
Token England
Feature activation+0.000
east
Token east
Feature activation+0.055
coast
Token coast
Feature activation+0.289
,
Token,
Feature activation+0.931
but
Token but
Feature activation+0.147
there
Token there
Feature activation+1.391
are
Token are
Feature activation+0.983
no
Token no
Feature activation+0.000
immediate
Token immediate
Feature activation+0.000
reports
Token reports
Feature activation+0.000
of
Token of
Feature activation+0.000
damage
Token damage
Feature activation+0.000
of
Token of
Feature activation+0.000
eating
Token eating
Feature activation+0.000
real
Token real
Feature activation+0.000
food
Token food
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.815
Ċ
TokenĊ
Feature activation+0.740
It
TokenIt
Feature activation+0.563
is
Token is
Feature activation+0.000
late
Token late
Feature activation+0.000
evening
Token evening
Feature activation+0.000
and
Token and
Feature activation+0.000
Earth
Token Earth
Feature activation+0.000
rotation
Token rotation
Feature activation+0.000
slows
Token slows
Feature activation+0.000
Read
Token Read
Feature activation+0.540
more
Token more
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.741
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
US
Token US
Feature activation+0.000
Geological
Token Geological
Feature activation+0.000
Survey
Token Survey
Feature activation+0.543

INTERVAL 0.000 - 0.691
CONTAINS 99.989%

arrived
Token arrived
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
Yas
Token Yas
Feature activation+0.000
Marina
Token Marina
Feature activation+0.000
Circuit
Token Circuit
Feature activation+0.000
in
Token in
Feature activation+0.000
control
Token control
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
own
Token own
Feature activation+0.000
travers
Token travers
Feature activation+0.000
ing
Tokening
Feature activation+0.000
the
Token the
Feature activation+0.000
world
Token world
Feature activation+0.000
in
Token in
Feature activation+0.000
search
Token search
Feature activation+0.000
of
Token of
Feature activation+0.000
some
Token some
Feature activation+0.000
dream
Token dream
Feature activation+0.000
career
Token career
Feature activation+0.000
resurrection
Token resurrection
Feature activation+0.000
,
Token,
Feature activation+0.000
right
Token right
Feature activation+0.000
now
Token now
Feature activation+0.000
,
Token,
Feature activation+0.000
missing
Token missing
Feature activation+0.000
the
Token the
Feature activation+0.000
only
Token only
Feature activation+0.000
opportunity
Token opportunity
Feature activation+0.000
for
Token for
Feature activation+0.000
grace
Token grace
Feature activation+0.000
that
Token that
Feature activation+0.000
half
Token half
Feature activation+0.000
-
Token-
Feature activation+0.000
million
Tokenmillion
Feature activation+0.000
-
Token-
Feature activation+0.000
dollar
Tokendollar
Feature activation+0.000
contribution
Token contribution
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
comes
Token comes
Feature activation+0.000
Among
TokenAmong
Feature activation+0.000
the
Token the
Feature activation+0.000
far
Token far
Feature activation+0.000
-
Token-
Feature activation+0.000
reaching
Tokenreaching
Feature activation+0.000
ambitions
Token ambitions
Feature activation+0.000
of
Token of
Feature activation+0.000
Mayor
Token Mayor
Feature activation+0.000
Bill
Token Bill
Feature activation+0.000
de
Token de
Feature activation+0.000
Blasio
Token Blasio
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 H8.2: (feature 21042)

TOP ACTIVATIONS
MAX = 8.796

final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+4.929
was
Token was
Feature activation+2.999
dra
Token dra
Feature activation+8.796
f
Tokenf
Feature activation+4.536
ÂŃ
TokenÂŃ
Feature activation+0.000
ted
Tokented
Feature activation+2.738
by
Token by
Feature activation+2.081
H
Token H
Feature activation+5.194
red
Token red
Feature activation+0.000
ac
Tokenac
Feature activation+0.000
ÂŃ
TokenÂŃ
Feature activation+0.000
tions
Tokentions
Feature activation+6.111
and
Token and
Feature activation+6.315
fin
Token fin
Feature activation+8.778
ÂŃ
TokenÂŃ
Feature activation+0.000
ess
Tokeness
Feature activation+3.992
ÂŃ
TokenÂŃ
Feature activation+0.851
ing
Tokening
Feature activation+7.304
,
Token,
Feature activation+7.762
sum
Token sum
Feature activation+2.720
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+0.728
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.721
ori
Token ori
Feature activation+8.485
ÂŃ
TokenÂŃ
Feature activation+0.365
gin
Tokengin
Feature activation+1.460
ÂŃ
TokenÂŃ
Feature activation+0.000
al
Tokenal
Feature activation+3.065
intel
Token intel
Feature activation+7.510
ats
Tokenats
Feature activation+1.702
spoke
Token spoke
Feature activation+0.462
with
Token with
Feature activation+0.452
Paul
Token Paul
Feature activation+0.148
,
Token,
Feature activation+0.817
com
Token com
Feature activation+8.449
ÂŃ
TokenÂŃ
Feature activation+0.000
pared
Tokenpared
Feature activation+1.674
with
Token with
Feature activation+0.757
just
Token just
Feature activation+1.881
three
Token three
Feature activation+0.614
ÂŃ
TokenÂŃ
Feature activation+0.000
pat
Tokenpat
Feature activation+0.885
ched
Tokenched
Feature activation+0.995
to
Token to
Feature activation+0.667
the
Token the
Feature activation+1.007
Sec
Token Sec
Feature activation+8.368
ÂŃ
TokenÂŃ
Feature activation+1.216
ret
Tokenret
Feature activation+2.891
ÂŃ
TokenÂŃ
Feature activation+0.452
ary
Tokenary
Feature activation+0.993
of
Token of
Feature activation+0.208
vis
Token vis
Feature activation+4.971
ÂŃ
TokenÂŃ
Feature activation+0.000
it
Tokenit
Feature activation+3.436
set
Token set
Feature activation+0.892
in
Token in
Feature activation+1.216
mo
Token mo
Feature activation+8.144
ÂŃ
TokenÂŃ
Feature activation+0.000
tion
Tokention
Feature activation+2.726
more
Token more
Feature activation+2.414
Demo
Token Demo
Feature activation+4.617
ÂŃ
TokenÂŃ
Feature activation+0.001
of
Token of
Feature activation+0.208
State
Token State
Feature activation+1.300
.
Token.
Feature activation+7.362
So
Token So
Feature activation+1.401
the
Token the
Feature activation+1.507
min
Token min
Feature activation+8.101
ÂŃ
TokenÂŃ
Feature activation+0.000
is
Tokenis
Feature activation+3.277
ÂŃ
TokenÂŃ
Feature activation+0.000
ter
Tokenter
Feature activation+1.050
was
Token was
Feature activation+1.088
ÂŃ
TokenÂŃ
Feature activation+0.000
ter
Tokenter
Feature activation+1.050
was
Token was
Feature activation+1.088
only
Token only
Feature activation+1.252
ever
Token ever
Feature activation+2.745
pres
Token pres
Feature activation+7.775
en
Tokenen
Feature activation+6.877
ÂŃ
TokenÂŃ
Feature activation+0.360
ted
Tokented
Feature activation+2.042
with
Token with
Feature activation+1.218
was
Token was
Feature activation+3.653
fin
Token fin
Feature activation+8.778
ÂŃ
TokenÂŃ
Feature activation+0.000
ess
Tokeness
Feature activation+3.992
ÂŃ
TokenÂŃ
Feature activation+0.851
ing
Tokening
Feature activation+7.304
,
Token,
Feature activation+7.762
before
Token before
Feature activation+5.200
a
Token a
Feature activation+4.997
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.006
Ċ
TokenĊ
Feature activation+2.103
8
Token8
Feature activation+4.275
.
Token.
Feature activation+1.294
Add
Token Add
Feature activation+2.468
i
Tokeni
Feature activation+7.608
ÂŃ
TokenÂŃ
Feature activation+0.000
tion
Tokention
Feature activation+1.574
ÂŃ
TokenÂŃ
Feature activation+0.289
ally
Tokenally
Feature activation+0.000
,
Token,
Feature activation+0.000
ori
Token ori
Feature activation+8.485
ÂŃ
TokenÂŃ
Feature activation+0.365
gin
Tokengin
Feature activation+1.460
ÂŃ
TokenÂŃ
Feature activation+0.000
al
Tokenal
Feature activation+3.065
intel
Token intel
Feature activation+7.510
ÂŃ
TokenÂŃ
Feature activation+0.000
li
Tokenli
Feature activation+3.308
ÂŃ
TokenÂŃ
Feature activation+0.000
gence
Tokengence
Feature activation+1.375
case
Token case
Feature activation+0.619
DD
Token DD
Feature activation+4.086
G
TokenG
Feature activation+4.196
,
Token,
Feature activation+1.991
and
Token and
Feature activation+2.303
then
Token then
Feature activation+1.421
dis
Token dis
Feature activation+7.450
ÂŃ
TokenÂŃ
Feature activation+0.000
pat
Tokenpat
Feature activation+0.885
ched
Tokenched
Feature activation+0.995
to
Token to
Feature activation+0.667
the
Token the
Feature activation+1.007
ret
Tokenret
Feature activation+2.891
ÂŃ
TokenÂŃ
Feature activation+0.452
ary
Tokenary
Feature activation+0.993
of
Token of
Feature activation+0.208
State
Token State
Feature activation+1.300
.
Token.
Feature activation+7.362
So
Token So
Feature activation+1.401
the
Token the
Feature activation+1.507
min
Token min
Feature activation+8.101
ÂŃ
TokenÂŃ
Feature activation+0.000
is
Tokenis
Feature activation+3.277
and
Token and
Feature activation+6.315
fin
Token fin
Feature activation+8.778
ÂŃ
TokenÂŃ
Feature activation+0.000
ess
Tokeness
Feature activation+3.992
ÂŃ
TokenÂŃ
Feature activation+0.851
ing
Tokening
Feature activation+7.304
,
Token,
Feature activation+7.762
before
Token before
Feature activation+5.200
a
Token a
Feature activation+4.997
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ist
Tokenist
Feature activation+2.033
ÂŃ
TokenÂŃ
Feature activation+0.000
ry
Tokenry
Feature activation+0.090
of
Token of
Feature activation+1.661
human
Token human
Feature activation+0.872
res
Token res
Feature activation+7.251
our
Tokenour
Feature activation+4.652
ÂŃ
TokenÂŃ
Feature activation+0.000
ce
Tokence
Feature activation+2.379
and
Token and
Feature activation+0.580
overse
Token overse
Feature activation+4.817
,
Token,
Feature activation+0.621
and
Token and
Feature activation+0.370
Con
Token Con
Feature activation+6.266
ÂŃ
TokenÂŃ
Feature activation+0.000
nect
Tokennect
Feature activation+1.712
i
Tokeni
Feature activation+7.071
ÂŃ
TokenÂŃ
Feature activation+0.216
c
Tokenc
Feature activation+0.346
ÂŃ
TokenÂŃ
Feature activation+0.000
ut
Tokenut
Feature activation+0.704
âĢ
TokenâĢ
Feature activation+0.011
ing
Tokening
Feature activation+7.304
,
Token,
Feature activation+7.762
before
Token before
Feature activation+5.200
a
Token a
Feature activation+4.997
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+4.929
was
Token was
Feature activation+2.999
dra
Token dra
Feature activation+8.796
f
Tokenf
Feature activation+4.536
ter
Tokenter
Feature activation+1.050
was
Token was
Feature activation+1.088
only
Token only
Feature activation+1.252
ever
Token ever
Feature activation+2.745
pres
Token pres
Feature activation+7.775
en
Tokenen
Feature activation+6.877
ÂŃ
TokenÂŃ
Feature activation+0.360
ted
Tokented
Feature activation+2.042
with
Token with
Feature activation+1.218
was
Token was
Feature activation+3.653
a
Token a
Feature activation+2.998
with
Token with
Feature activation+0.757
just
Token just
Feature activation+1.881
three
Token three
Feature activation+0.614
Re
Token Re
Feature activation+5.228
ÂŃ
TokenÂŃ
Feature activation+0.000
pub
Tokenpub
Feature activation+6.809
ÂŃ
TokenÂŃ
Feature activation+0.000
lic
Tokenlic
Feature activation+0.938
ÂŃ
TokenÂŃ
Feature activation+0.000
ans
Tokenans
Feature activation+1.916
.
Token.
Feature activation+3.993
ÂŃ
TokenÂŃ
Feature activation+0.001
cr
Tokencr
Feature activation+0.975
at
Tokenat
Feature activation+5.533
ÂŃ
TokenÂŃ
Feature activation+0.000
ic
Tokenic
Feature activation+1.734
Sen
Token Sen
Feature activation+6.732
ÂŃ
TokenÂŃ
Feature activation+0.000
ate
Tokenate
Feature activation+1.135
speeches
Token speeches
Feature activation+1.821
from
Token from
Feature activation+3.561
New
Token New
Feature activation+1.369

Top DFA by src position
MAX = 9.488

<|endoftext|>
Token<|endoftext|>
Feature activation+0.151
Top resid features:
red
Token red
Feature activation+0.001
Top resid features:
ac
Tokenac
Feature activation+0.003
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+5.178
Top resid features:
tions
Tokentions
Feature activation+0.012
Top resid features:
and
Token and
Feature activation-0.007
Top resid features:
fin
Token fin
Feature activation+0.005
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.981
Top resid features:
ess
Tokeness
Feature activation+0.009
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.173
Top resid features:
red
Token red
Feature activation+0.003
Top resid features:
ac
Tokenac
Feature activation+0.003
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+9.488
Top resid features:
tions
Tokentions
Feature activation+0.135
Top resid features:
and
Token and
Feature activation+0.028
Top resid features:
fin
Token fin
Feature activation+0.024
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+0.000
Top resid features:
ess
Tokeness
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.156
Top resid features:
red
Token red
Feature activation+0.000
Top resid features:
ac
Tokenac
Feature activation-0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+4.611
Top resid features:
tions
Tokentions
Feature activation+0.003
Top resid features:
and
Token and
Feature activation-0.004
Top resid features:
fin
Token fin
Feature activation+0.000
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+0.920
Top resid features:
ess
Tokeness
Feature activation+0.002
Top resid features:
from
Token from
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.003
Top resid features:
last
Token last
Feature activation-0.002
Top resid features:
.
Token.
Feature activation-0.009
Top resid features:
Sev
Token Sev
Feature activation-0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+6.994
Top resid features:
en
Tokenen
Feature activation+0.015
Top resid features:
Demo
Token Demo
Feature activation+0.002
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+2.342
Top resid features:
cr
Tokencr
Feature activation+0.001
Top resid features:
ats
Tokenats
Feature activation+0.021
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.138
Top resid features:
red
Token red
Feature activation+0.001
Top resid features:
ac
Tokenac
Feature activation+0.003
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+5.242
Top resid features:
tions
Tokentions
Feature activation+0.008
Top resid features:
and
Token and
Feature activation-0.011
Top resid features:
fin
Token fin
Feature activation-0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.342
Top resid features:
ess
Tokeness
Feature activation+0.004
Top resid features:
from
Token from
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
last
Token last
Feature activation-0.002
Top resid features:
.
Token.
Feature activation-0.016
Top resid features:
Sev
Token Sev
Feature activation-0.004
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+3.059
Top resid features:
en
Tokenen
Feature activation+0.004
Top resid features:
Demo
Token Demo
Feature activation+0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.610
Top resid features:
cr
Tokencr
Feature activation+0.000
Top resid features:
ats
Tokenats
Feature activation+0.008
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.157
Top resid features:
red
Token red
Feature activation+0.001
Top resid features:
ac
Tokenac
Feature activation+0.005
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+4.183
Top resid features:
tions
Tokentions
Feature activation+0.016
Top resid features:
and
Token and
Feature activation-0.009
Top resid features:
fin
Token fin
Feature activation+0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.392
Top resid features:
ess
Tokeness
Feature activation+0.006
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.143
Top resid features:
red
Token red
Feature activation+0.000
Top resid features:
ac
Tokenac
Feature activation+0.002
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+2.511
Top resid features:
tions
Tokentions
Feature activation+0.011
Top resid features:
and
Token and
Feature activation-0.005
Top resid features:
fin
Token fin
Feature activation+0.002
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.155
Top resid features:
ess
Tokeness
Feature activation+0.006
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.172
Top resid features:
red
Token red
Feature activation-0.002
Top resid features:
ac
Tokenac
Feature activation+0.008
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+4.882
Top resid features:
tions
Tokentions
Feature activation+0.116
Top resid features:
and
Token and
Feature activation-0.009
Top resid features:
fin
Token fin
Feature activation+0.021
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.569
Top resid features:
ess
Tokeness
Feature activation+0.078
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.177
Top resid features:
red
Token red
Feature activation-0.002
Top resid features:
ac
Tokenac
Feature activation+0.006
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+2.011
Top resid features:
tions
Tokentions
Feature activation+0.010
Top resid features:
and
Token and
Feature activation-0.004
Top resid features:
fin
Token fin
Feature activation+0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.433
Top resid features:
ess
Tokeness
Feature activation+0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.160
Top resid features:
red
Token red
Feature activation-0.001
Top resid features:
ac
Tokenac
Feature activation-0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+3.269
Top resid features:
tions
Tokentions
Feature activation+0.006
Top resid features:
and
Token and
Feature activation-0.006
Top resid features:
fin
Token fin
Feature activation+0.000
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+0.818
Top resid features:
ess
Tokeness
Feature activation+0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.129
Top resid features:
red
Token red
Feature activation-0.000
Top resid features:
ac
Tokenac
Feature activation+0.008
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+4.125
Top resid features:
tions
Tokentions
Feature activation+0.036
Top resid features:
and
Token and
Feature activation-0.009
Top resid features:
fin
Token fin
Feature activation+0.003
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.446
Top resid features:
ess
Tokeness
Feature activation+0.020
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.119
Top resid features:
red
Token red
Feature activation-0.023
Top resid features:
ac
Tokenac
Feature activation+0.022
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+4.310
Top resid features:
tions
Tokentions
Feature activation+0.057
Top resid features:
and
Token and
Feature activation-0.011
Top resid features:
fin
Token fin
Feature activation+0.003
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+0.721
Top resid features:
ess
Tokeness
Feature activation+0.007
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.165
Top resid features:
red
Token red
Feature activation-0.003
Top resid features:
ac
Tokenac
Feature activation-0.002
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+4.848
Top resid features:
tions
Tokentions
Feature activation+0.177
Top resid features:
and
Token and
Feature activation+0.012
Top resid features:
fin
Token fin
Feature activation+0.044
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.482
Top resid features:
ess
Tokeness
Feature activation+0.150
Top resid features:
charge
Token charge
Feature activation+0.006
Top resid features:
of
Token of
Feature activation+0.009
Top resid features:
the
Token the
Feature activation+0.031
Top resid features:
min
Token min
Feature activation+0.027
Top resid features:
ist
Tokenist
Feature activation+0.014
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+6.320
Top resid features:
ry
Tokenry
Feature activation+0.047
Top resid features:
of
Token of
Feature activation+0.068
Top resid features:
human
Token human
Feature activation+0.061
Top resid features:
res
Token res
Feature activation+0.027
Top resid features:
our
Tokenour
Feature activation+0.000
Top resid features:
from
Token from
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
last
Token last
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.005
Top resid features:
Sev
Token Sev
Feature activation-0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.609
Top resid features:
en
Tokenen
Feature activation+0.002
Top resid features:
Demo
Token Demo
Feature activation+0.004
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+0.766
Top resid features:
cr
Tokencr
Feature activation-0.002
Top resid features:
ats
Tokenats
Feature activation+0.008
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.152
Top resid features:
red
Token red
Feature activation+0.004
Top resid features:
ac
Tokenac
Feature activation+0.010
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+3.806
Top resid features:
tions
Tokentions
Feature activation+0.035
Top resid features:
and
Token and
Feature activation-0.006
Top resid features:
fin
Token fin
Feature activation+0.005
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.611
Top resid features:
ess
Tokeness
Feature activation+0.033
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.190
Top resid features:
red
Token red
Feature activation-0.001
Top resid features:
ac
Tokenac
Feature activation+0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.822
Top resid features:
tions
Tokentions
Feature activation+0.005
Top resid features:
and
Token and
Feature activation-0.005
Top resid features:
fin
Token fin
Feature activation+0.000
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+0.767
Top resid features:
ess
Tokeness
Feature activation+0.006
Top resid features:
from
Token from
Feature activation-0.000
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
last
Token last
Feature activation-0.003
Top resid features:
.
Token.
Feature activation-0.007
Top resid features:
Sev
Token Sev
Feature activation+0.000
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+3.219
Top resid features:
en
Tokenen
Feature activation+0.026
Top resid features:
Demo
Token Demo
Feature activation+0.007
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.493
Top resid features:
cr
Tokencr
Feature activation+0.006
Top resid features:
ats
Tokenats
Feature activation+0.012
Top resid features:
pared
Tokenpared
Feature activation+0.001
Top resid features:
with
Token with
Feature activation-0.002
Top resid features:
just
Token just
Feature activation+0.001
Top resid features:
three
Token three
Feature activation+0.005
Top resid features:
Re
Token Re
Feature activation+0.009
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.335
Top resid features:
pub
Tokenpub
Feature activation+0.003
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+1.278
Top resid features:
lic
Tokenlic
Feature activation+0.001
Top resid features:
ÂŃ
TokenÂŃ
Feature activation+0.162
Top resid features:
ans
Tokenans
Feature activation+0.018
Top resid features:

Decoder Weights Distribution

Head 0: 0.12

Head 1: 0.01

Head 2: 0.55

Head 3: 0.03

Head 4: 0.02

Head 5: 0.04

Head 6: 0.03

Head 7: 0.04

Head 8: 0.03

Head 9: 0.04

Head 10: 0.05

Head 11: 0.05

Positive logits

­6.62

­5.42

4.17

3.96

»3.84

.—3.69

.–3.16

«3.01

2.99

2.94

2.84

\<2.83

2.82

pec2.80

^2.78

§2.75

tha2.74

2.74

Â2.71

«2.71

Negative logits

"...-3.16

"...-3.07

......-2.69

BBC-2.68

:-)-2.65

)...-2.59

...?-2.45

--2.45

...-2.44

JUST-2.43

..."-2.42

BBC-2.39

...-2.39

....-2.31

yip-2.26

(?,-2.24

Calories-2.24

;)-2.23

NEVER-2.22

+/--2.16

INTERVAL 7.916 - 8.796
CONTAINS 0.001%

vis
Token vis
Feature activation+4.971
ÂŃ
TokenÂŃ
Feature activation+0.000
it
Tokenit
Feature activation+3.436
set
Token set
Feature activation+0.892
in
Token in
Feature activation+1.216
mo
Token mo
Feature activation+8.144
ÂŃ
TokenÂŃ
Feature activation+0.000
tion
Tokention
Feature activation+2.726
more
Token more
Feature activation+2.414
Demo
Token Demo
Feature activation+4.617
ÂŃ
TokenÂŃ
Feature activation+0.001
red
Token red
Feature activation+0.000
ac
Tokenac
Feature activation+0.000
ÂŃ
TokenÂŃ
Feature activation+0.000
tions
Tokentions
Feature activation+6.111
and
Token and
Feature activation+6.315
fin
Token fin
Feature activation+8.778
ÂŃ
TokenÂŃ
Feature activation+0.000
ess
Tokeness
Feature activation+3.992
ÂŃ
TokenÂŃ
Feature activation+0.851
ing
Tokening
Feature activation+7.304
,
Token,
Feature activation+7.762
sum
Token sum
Feature activation+2.720
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+0.728
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.721
ori
Token ori
Feature activation+8.485
ÂŃ
TokenÂŃ
Feature activation+0.365
gin
Tokengin
Feature activation+1.460
ÂŃ
TokenÂŃ
Feature activation+0.000
al
Tokenal
Feature activation+3.065
intel
Token intel
Feature activation+7.510
of
Token of
Feature activation+0.208
State
Token State
Feature activation+1.300
.
Token.
Feature activation+7.362
So
Token So
Feature activation+1.401
the
Token the
Feature activation+1.507
min
Token min
Feature activation+8.101
ÂŃ
TokenÂŃ
Feature activation+0.000
is
Tokenis
Feature activation+3.277
ÂŃ
TokenÂŃ
Feature activation+0.000
ter
Tokenter
Feature activation+1.050
was
Token was
Feature activation+1.088
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+4.929
was
Token was
Feature activation+2.999
dra
Token dra
Feature activation+8.796
f
Tokenf
Feature activation+4.536
ÂŃ
TokenÂŃ
Feature activation+0.000
ted
Tokented
Feature activation+2.738
by
Token by
Feature activation+2.081
H
Token H
Feature activation+5.194

INTERVAL 7.037 - 7.916
CONTAINS 0.001%

ÂŃ
TokenÂŃ
Feature activation+0.000
ter
Tokenter
Feature activation+1.050
was
Token was
Feature activation+1.088
only
Token only
Feature activation+1.252
ever
Token ever
Feature activation+2.745
pres
Token pres
Feature activation+7.775
en
Tokenen
Feature activation+6.877
ÂŃ
TokenÂŃ
Feature activation+0.360
ted
Tokented
Feature activation+2.042
with
Token with
Feature activation+1.218
was
Token was
Feature activation+3.653
DD
Token DD
Feature activation+4.086
G
TokenG
Feature activation+4.196
,
Token,
Feature activation+1.991
and
Token and
Feature activation+2.303
then
Token then
Feature activation+1.421
dis
Token dis
Feature activation+7.450
ÂŃ
TokenÂŃ
Feature activation+0.000
pat
Tokenpat
Feature activation+0.885
ched
Tokenched
Feature activation+0.995
to
Token to
Feature activation+0.667
the
Token the
Feature activation+1.007
fin
Token fin
Feature activation+8.778
ÂŃ
TokenÂŃ
Feature activation+0.000
ess
Tokeness
Feature activation+3.992
ÂŃ
TokenÂŃ
Feature activation+0.851
ing
Tokening
Feature activation+7.304
,
Token,
Feature activation+7.762
before
Token before
Feature activation+5.200
a
Token a
Feature activation+4.997
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.006
Ċ
TokenĊ
Feature activation+2.103
8
Token8
Feature activation+4.275
.
Token.
Feature activation+1.294
Add
Token Add
Feature activation+2.468
i
Tokeni
Feature activation+7.608
ÂŃ
TokenÂŃ
Feature activation+0.000
tion
Tokention
Feature activation+1.574
ÂŃ
TokenÂŃ
Feature activation+0.289
ally
Tokenally
Feature activation+0.000
,
Token,
Feature activation+0.000
ori
Token ori
Feature activation+8.485
ÂŃ
TokenÂŃ
Feature activation+0.365
gin
Tokengin
Feature activation+1.460
ÂŃ
TokenÂŃ
Feature activation+0.000
al
Tokenal
Feature activation+3.065
intel
Token intel
Feature activation+7.510
ÂŃ
TokenÂŃ
Feature activation+0.000
li
Tokenli
Feature activation+3.308
ÂŃ
TokenÂŃ
Feature activation+0.000
gence
Tokengence
Feature activation+1.375
case
Token case
Feature activation+0.619

INTERVAL 6.157 - 7.037
CONTAINS 0.001%

ÂŃ
TokenÂŃ
Feature activation+0.000
li
Tokenli
Feature activation+3.308
ÂŃ
TokenÂŃ
Feature activation+0.000
gence
Tokengence
Feature activation+1.375
case
Token case
Feature activation+0.619
.
Token.
Feature activation+6.238
Ċ
TokenĊ
Feature activation+3.006
Ċ
TokenĊ
Feature activation+2.103
8
Token8
Feature activation+4.275
.
Token.
Feature activation+1.294
Add
Token Add
Feature activation+2.468
ing
Tokening
Feature activation+7.304
,
Token,
Feature activation+7.762
before
Token before
Feature activation+5.200
a
Token a
Feature activation+4.997
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+4.929
was
Token was
Feature activation+2.999
dra
Token dra
Feature activation+8.796
f
Tokenf
Feature activation+4.536
ÂŃ
TokenÂŃ
Feature activation+0.000
ate
Tokenate
Feature activation+1.135
speeches
Token speeches
Feature activation+1.821
from
Token from
Feature activation+3.561
New
Token New
Feature activation+1.369
Mex
Token Mex
Feature activation+6.383
ÂŃ
TokenÂŃ
Feature activation+0.000
ico
Tokenico
Feature activation+2.533
âĢ
TokenâĢ
Feature activation+0.639
Ļ
TokenĻ
Feature activation+1.804
s
Tokens
Feature activation+0.765
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
red
Token red
Feature activation+0.000
ac
Tokenac
Feature activation+0.000
ÂŃ
TokenÂŃ
Feature activation+0.000
tions
Tokentions
Feature activation+6.111
and
Token and
Feature activation+6.315
fin
Token fin
Feature activation+8.778
ÂŃ
TokenÂŃ
Feature activation+0.000
ess
Tokeness
Feature activation+3.992
ÂŃ
TokenÂŃ
Feature activation+0.851
ing
Tokening
Feature activation+7.304
ÂŃ
TokenÂŃ
Feature activation+0.001
cr
Tokencr
Feature activation+0.975
at
Tokenat
Feature activation+5.533
ÂŃ
TokenÂŃ
Feature activation+0.000
ic
Tokenic
Feature activation+1.734
Sen
Token Sen
Feature activation+6.732
ÂŃ
TokenÂŃ
Feature activation+0.000
ate
Tokenate
Feature activation+1.135
speeches
Token speeches
Feature activation+1.821
from
Token from
Feature activation+3.561
New
Token New
Feature activation+1.369

INTERVAL 5.278 - 6.157
CONTAINS 0.001%

rib
Tokenrib
Feature activation+0.000
u
Tokenu
Feature activation+0.000
ÂŃ
TokenÂŃ
Feature activation+0.000
tor
Tokentor
Feature activation+6.987
to
Token to
Feature activation+6.517
the
Token the
Feature activation+6.084
Advocate
Token Advocate
Feature activation+2.956
suggests
Token suggests
Feature activation+2.808
that
Token that
Feature activation+2.067
the
Token the
Feature activation+2.993
editors
Token editors
Feature activation+0.508
tion
Tokention
Feature activation+2.726
more
Token more
Feature activation+2.414
Demo
Token Demo
Feature activation+4.617
ÂŃ
TokenÂŃ
Feature activation+0.001
cr
Tokencr
Feature activation+0.975
at
Tokenat
Feature activation+5.533
ÂŃ
TokenÂŃ
Feature activation+0.000
ic
Tokenic
Feature activation+1.734
Sen
Token Sen
Feature activation+6.732
ÂŃ
TokenÂŃ
Feature activation+0.000
ate
Tokenate
Feature activation+1.135
ori
Token ori
Feature activation+4.844
ÂŃ
TokenÂŃ
Feature activation+0.000
gin
Tokengin
Feature activation+1.336
ÂŃ
TokenÂŃ
Feature activation+0.177
al
Tokenal
Feature activation+1.058
intel
Token intel
Feature activation+5.461
ÂŃ
TokenÂŃ
Feature activation+0.000
li
Tokenli
Feature activation+2.521
ÂŃ
TokenÂŃ
Feature activation+0.000
gence
Tokengence
Feature activation+0.565
case
Token case
Feature activation+0.000

INTERVAL 4.398 - 5.278
CONTAINS 0.002%

before
Token before
Feature activation+5.200
a
Token a
Feature activation+4.997
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+4.929
was
Token was
Feature activation+2.999
dra
Token dra
Feature activation+8.796
f
Tokenf
Feature activation+4.536
ÂŃ
TokenÂŃ
Feature activation+0.000
ted
Tokented
Feature activation+2.738
ÂŃ
TokenÂŃ
Feature activation+0.000
ess
Tokeness
Feature activation+3.992
ÂŃ
TokenÂŃ
Feature activation+0.851
ing
Tokening
Feature activation+7.304
,
Token,
Feature activation+7.762
before
Token before
Feature activation+5.200
a
Token a
Feature activation+4.997
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+4.929
W
Token W
Feature activation+3.654
yden
Tokenyden
Feature activation+0.965
âĢ
TokenâĢ
Feature activation+3.499
Ļ
TokenĻ
Feature activation+2.283
s
Tokens
Feature activation+0.617
vis
Token vis
Feature activation+4.971
ÂŃ
TokenÂŃ
Feature activation+0.000
it
Tokenit
Feature activation+3.436
set
Token set
Feature activation+0.892
in
Token in
Feature activation+1.216
mo
Token mo
Feature activation+8.144
ÂŃ
TokenÂŃ
Feature activation+0.000
ce
Tokence
Feature activation+2.379
and
Token and
Feature activation+0.580
overse
Token overse
Feature activation+4.817
ÂŃ
TokenÂŃ
Feature activation+0.000
as
Tokenas
Feature activation+4.445
Pak
Token Pak
Feature activation+1.875
ist
Tokenist
Feature activation+1.559
ÂŃ
TokenÂŃ
Feature activation+0.000
an
Tokenan
Feature activation+0.683
is
Tokenis
Feature activation+3.009
tion
Tokention
Feature activation+1.574
ÂŃ
TokenÂŃ
Feature activation+0.289
ally
Tokenally
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.868
ori
Token ori
Feature activation+4.844
ÂŃ
TokenÂŃ
Feature activation+0.000
gin
Tokengin
Feature activation+1.336
ÂŃ
TokenÂŃ
Feature activation+0.177
al
Tokenal
Feature activation+1.058
intel
Token intel
Feature activation+5.461

INTERVAL 3.518 - 4.398
CONTAINS 0.002%

of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
John
Token John
Feature activation+0.000
Harvard
Token Harvard
Feature activation+0.000
statue
Token statue
Feature activation+0.012
."
Token."
Feature activation+4.044
[
Token [
Feature activation+0.000
29
Token29
Feature activation+0.071
]
Token]
Feature activation+0.907
1890
Token 1890
Feature activation+0.000
painting
Token painting
Feature activation+0.000
ÂŃ
TokenÂŃ
Feature activation+0.000
lic
Tokenlic
Feature activation+0.938
ÂŃ
TokenÂŃ
Feature activation+0.000
ans
Tokenans
Feature activation+1.916
.
Token.
Feature activation+3.993
W
Token W
Feature activation+3.654
yden
Tokenyden
Feature activation+0.965
âĢ
TokenâĢ
Feature activation+3.499
Ļ
TokenĻ
Feature activation+2.283
s
Tokens
Feature activation+0.617
vis
Token vis
Feature activation+4.971
ic
Tokenic
Feature activation+1.734
Sen
Token Sen
Feature activation+6.732
ÂŃ
TokenÂŃ
Feature activation+0.000
ate
Tokenate
Feature activation+1.135
speeches
Token speeches
Feature activation+1.821
from
Token from
Feature activation+3.561
New
Token New
Feature activation+1.369
Mex
Token Mex
Feature activation+6.383
ÂŃ
TokenÂŃ
Feature activation+0.000
ico
Tokenico
Feature activation+2.533
âĢ
TokenâĢ
Feature activation+0.639
pre
Tokenpre
Feature activation+0.000
ma
Tokenma
Feature activation+0.000
ÂŃ
TokenÂŃ
Feature activation+0.627
c
Tokenc
Feature activation+0.311
ist
Tokenist
Feature activation+3.505
cir
Token cir
Feature activation+4.140
ÂŃ
TokenÂŃ
Feature activation+0.000
cles
Tokencles
Feature activation+0.143
,
Token,
Feature activation+0.000
according
Token according
Feature activation+0.867
to
Token to
Feature activation+0.132
Branch
Token Branch
Feature activation+0.923
,
Token,
Feature activation+2.580
signed
Token signed
Feature activation+2.481
by
Token by
Feature activation+1.320
the
Token the
Feature activation+1.684
DD
Token DD
Feature activation+4.086
G
TokenG
Feature activation+4.196
,
Token,
Feature activation+1.991
and
Token and
Feature activation+2.303
then
Token then
Feature activation+1.421
dis
Token dis
Feature activation+7.450

INTERVAL 2.639 - 3.518
CONTAINS 0.002%

is
Tokenis
Feature activation+3.277
ÂŃ
TokenÂŃ
Feature activation+0.000
ter
Tokenter
Feature activation+1.050
was
Token was
Feature activation+1.088
only
Token only
Feature activation+1.252
ever
Token ever
Feature activation+2.745
pres
Token pres
Feature activation+7.775
en
Tokenen
Feature activation+6.877
ÂŃ
TokenÂŃ
Feature activation+0.360
ted
Tokented
Feature activation+2.042
with
Token with
Feature activation+1.218
more
Token more
Feature activation+0.000
than
Token than
Feature activation+0.000
two
Token two
Feature activation+0.000
decades
Token decades
Feature activation+0.000
,
Token,
Feature activation+0.000
re
Token re
Feature activation+2.668
ÂŃ
TokenÂŃ
Feature activation+0.000
count
Tokencount
Feature activation+0.554
ed
Tokened
Feature activation+0.000
what
Token what
Feature activation+0.000
had
Token had
Feature activation+0.000
of
Token of
Feature activation+0.779
the
Token the
Feature activation+0.984
college
Token college
Feature activation+0.417
papers
Token papers
Feature activation+0.246
fer
Token fer
Feature activation+6.469
ret
Tokenret
Feature activation+2.911
out
Token out
Feature activation+0.375
the
Token the
Feature activation+0.000
authors
Token authors
Feature activation+1.732
of
Token of
Feature activation+0.030
the
Token the
Feature activation+1.579
u
Tokenu
Feature activation+0.000
ÂŃ
TokenÂŃ
Feature activation+0.000
tor
Tokentor
Feature activation+6.987
to
Token to
Feature activation+6.517
the
Token the
Feature activation+6.084
Advocate
Token Advocate
Feature activation+2.956
suggests
Token suggests
Feature activation+2.808
that
Token that
Feature activation+2.067
the
Token the
Feature activation+2.993
editors
Token editors
Feature activation+0.508
of
Token of
Feature activation+0.779
a
Token a
Feature activation+4.997
final
Token final
Feature activation+3.481
sum
Token sum
Feature activation+7.020
ÂŃ
TokenÂŃ
Feature activation+0.000
mary
Tokenmary
Feature activation+4.929
was
Token was
Feature activation+2.999
dra
Token dra
Feature activation+8.796
f
Tokenf
Feature activation+4.536
ÂŃ
TokenÂŃ
Feature activation+0.000
ted
Tokented
Feature activation+2.738
by
Token by
Feature activation+2.081

INTERVAL 1.759 - 2.639
CONTAINS 0.004%

ÂŃ
TokenÂŃ
Feature activation+0.000
gence
Tokengence
Feature activation+1.375
case
Token case
Feature activation+0.619
.
Token.
Feature activation+6.238
Ċ
TokenĊ
Feature activation+3.006
Ċ
TokenĊ
Feature activation+2.103
8
Token8
Feature activation+4.275
.
Token.
Feature activation+1.294
Add
Token Add
Feature activation+2.468
i
Tokeni
Feature activation+7.608
ÂŃ
TokenÂŃ
Feature activation+0.000
ans
Tokenans
Feature activation+1.916
.
Token.
Feature activation+3.993
W
Token W
Feature activation+3.654
yden
Tokenyden
Feature activation+0.965
âĢ
TokenâĢ
Feature activation+3.499
Ļ
TokenĻ
Feature activation+2.283
s
Tokens
Feature activation+0.617
vis
Token vis
Feature activation+4.971
ÂŃ
TokenÂŃ
Feature activation+0.000
it
Tokenit
Feature activation+3.436
set
Token set
Feature activation+0.892
,
Token,
Feature activation+0.817
com
Token com
Feature activation+8.449
ÂŃ
TokenÂŃ
Feature activation+0.000
pared
Tokenpared
Feature activation+1.674
with
Token with
Feature activation+0.757
just
Token just
Feature activation+1.881
three
Token three
Feature activation+0.614
Re
Token Re
Feature activation+5.228
ÂŃ
TokenÂŃ
Feature activation+0.000
pub
Tokenpub
Feature activation+6.809
ÂŃ
TokenÂŃ
Feature activation+0.000
,
Token,
Feature activation+0.000
according
Token according
Feature activation+0.867
to
Token to
Feature activation+0.132
the
Token the
Feature activation+0.056
Anti
Token Anti
Feature activation+4.993
-
Token-
Feature activation+2.129
Def
TokenDef
Feature activation+2.619
amation
Tokenamation
Feature activation+0.000
League
Token League
Feature activation+0.000
.
Token.
Feature activation+2.297
Ċ
TokenĊ
Feature activation+2.578
to
Token to
Feature activation+0.260
get
Token get
Feature activation+0.877
charge
Token charge
Feature activation+2.308
of
Token of
Feature activation+0.642
the
Token the
Feature activation+0.446
min
Token min
Feature activation+2.336
ist
Tokenist
Feature activation+2.033
ÂŃ
TokenÂŃ
Feature activation+0.000
ry
Tokenry
Feature activation+0.090
of
Token of
Feature activation+1.661
human
Token human
Feature activation+0.872

INTERVAL 0.880 - 1.759
CONTAINS 0.010%

and
Token and
Feature activation+2.303
then
Token then
Feature activation+1.421
dis
Token dis
Feature activation+7.450
ÂŃ
TokenÂŃ
Feature activation+0.000
pat
Tokenpat
Feature activation+0.885
ched
Tokenched
Feature activation+0.995
to
Token to
Feature activation+0.667
the
Token the
Feature activation+1.007
Sec
Token Sec
Feature activation+8.368
ÂŃ
TokenÂŃ
Feature activation+1.216
ret
Tokenret
Feature activation+2.891
neo
Token neo
Feature activation+0.000
-
Token-
Feature activation+0.000
Nazi
TokenNazi
Feature activation+0.000
groups
Token groups
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.487
Ċ
TokenĊ
Feature activation+0.000
Know
TokenKnow
Feature activation+2.277
these
Token these
Feature activation+0.000
pun
Token pun
Feature activation+0.983
ks
Tokenks
Feature activation+0.000
well
Tokenwell
Feature activation+0.000
,
Token,
Feature activation+0.621
and
Token and
Feature activation+0.370
Con
Token Con
Feature activation+6.266
ÂŃ
TokenÂŃ
Feature activation+0.000
nect
Tokennect
Feature activation+1.712
i
Tokeni
Feature activation+7.071
ÂŃ
TokenÂŃ
Feature activation+0.216
c
Tokenc
Feature activation+0.346
ÂŃ
TokenÂŃ
Feature activation+0.000
ut
Tokenut
Feature activation+0.704
f
Tokenf
Feature activation+4.536
ÂŃ
TokenÂŃ
Feature activation+0.000
ted
Tokented
Feature activation+2.738
by
Token by
Feature activation+2.081
H
Token H
Feature activation+5.194
Branch
Token Branch
Feature activation+0.923
,
Token,
Feature activation+2.580
signed
Token signed
Feature activation+2.481
by
Token by
Feature activation+1.320
the
Token the
Feature activation+1.684
DD
Token DD
Feature activation+4.086
case
Token case
Feature activation+0.619
.
Token.
Feature activation+6.238
Ċ
TokenĊ
Feature activation+3.006
Ċ
TokenĊ
Feature activation+2.103
8
Token8
Feature activation+4.275
.
Token.
Feature activation+1.294
Add
Token Add
Feature activation+2.468
i
Tokeni
Feature activation+7.608
ÂŃ
TokenÂŃ
Feature activation+0.000
tion
Tokention
Feature activation+1.574
ÂŃ
TokenÂŃ
Feature activation+0.289

INTERVAL 0.000 - 0.880
CONTAINS 99.977%

including
Tokenincluding
Feature activation+0.000
D
Token D
Feature activation+0.000
CH
TokenCH
Feature activation+0.000
P
TokenP
Feature activation+0.000
/
Token/
Feature activation+0.000
NAT
TokenNAT
Feature activation+0.000
networks
Token networks
Feature activation+0.000
)
Token)
Feature activation+0.000
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
website
Token website
Feature activation+0.000
.
Token.
Feature activation+0.000
Including
Token Including
Feature activation+0.000
non
Token non
Feature activation+0.000
-
Token-
Feature activation+0.000
College
TokenCollege
Feature activation+0.000
students
Token students
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
devices
Token devices
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
tended
Token tended
Feature activation+0.000
to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
set
Token set
Feature activation+0.000
decoration
Token decoration
Feature activation+0.000
seem
Token seem
Feature activation+0.000
less
Token less
Feature activation+0.000
realistic
Token realistic
Feature activation+0.000
ards
Tokenards
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+0.000
It
Token It
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
something
Token something
Feature activation+0.000
you
Token you
Feature activation+0.000
live
Token live
Feature activation+0.000
by
Token by
Feature activation+0.000
.
Token.
Feature activation+0.000
be
Token be
Feature activation+0.000
dele
Token dele
Feature activation+0.000
ter
Tokenter
Feature activation+0.000
ious
Tokenious
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
wider
Token wider
Feature activation+0.000
region
Token region
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
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 H8.2: (feature 21409)

TOP ACTIVATIONS
MAX = 4.326

actually
Token actually
Feature activation+0.073
make
Token make
Feature activation+0.000
authors
Token authors
Feature activation+0.000
more
Token more
Feature activation+0.000
money
Token money
Feature activation+1.148
.
Token.
Feature activation+4.326
They
Token They
Feature activation+0.000
lay
Token lay
Feature activation+0.000
it
Token it
Feature activation+0.000
out
Token out
Feature activation+0.000
pretty
Token pretty
Feature activation+0.000
their
Token their
Feature activation+0.000
feedback
Token feedback
Feature activation+0.000
last
Token last
Feature activation+0.016
week
Token week
Feature activation+0.442
.
Token.
Feature activation+1.401
Ċ
TokenĊ
Feature activation+4.189
Ċ
TokenĊ
Feature activation+2.739
Article
TokenArticle
Feature activation+1.814
Continued
Token Continued
Feature activation+0.000
Below
Token Below
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
May
Token May
Feature activation+0.203
23
Token 23
Feature activation+1.015
,
Token,
Feature activation+0.117
1969
Token 1969
Feature activation+1.451
.
Token.
Feature activation+1.340
Ċ
TokenĊ
Feature activation+4.107
Ċ
TokenĊ
Feature activation+0.095
For
TokenFor
Feature activation+0.000
one
Token one
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
love
Token love
Feature activation+0.000
Dave
Token Dave
Feature activation+0.000
Gro
Token Gro
Feature activation+0.000
hl
Tokenhl
Feature activation+0.856
.
Token .
Feature activation+0.733
Ċ
TokenĊ
Feature activation+3.973
Ċ
TokenĊ
Feature activation+0.757
So
TokenSo
Feature activation+0.000
,
Token,
Feature activation+0.000
after
Token after
Feature activation+0.000
Y
Token Y
Feature activation+0.000
f
Tokenf
Feature activation+0.000
Again
Token Again
Feature activation+0.000
.
Token.
Feature activation+3.451
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.795
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
move
Token move
Feature activation+0.000
has
Token has
Feature activation+0.000
sparked
Token sparked
Feature activation+0.000
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
innovative
Token innovative
Feature activation+0.000
way
Token way
Feature activation+0.000
.
Token.
Feature activation+1.310
Ċ
TokenĊ
Feature activation+3.625
Ċ
TokenĊ
Feature activation+0.000
R
TokenR
Feature activation+0.000
ip
Tokenip
Feature activation+0.000
Cl
Token Cl
Feature activation+0.000
ips
Tokenips
Feature activation+0.000
to
Token to
Feature activation+0.000
dump
Token dump
Feature activation+0.000
your
Token your
Feature activation+0.000
doctor
Token doctor
Feature activation+0.000
.)
Token .)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.594
Ċ
TokenĊ
Feature activation+0.949
MORE
TokenMORE
Feature activation+0.014
:
Token:
Feature activation+0.270
7
Token 7
Feature activation+0.554
Weird
Token Weird
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Supp
Token Supp
Feature activation+0.000
lied
Tokenlied
Feature activation+0.000
)
Token )
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.589
Ċ
TokenĊ
Feature activation+0.112
T
TokenT
Feature activation+0.000
iny
Tokeniny
Feature activation+0.000
cap
Token cap
Feature activation+0.000
es
Tokenes
Feature activation+0.000
Donald
Token Donald
Feature activation+0.000
Dr
Token Dr
Feature activation+0.000
ump
Tokenump
Feature activation+0.000
f
Tokenf
Feature activation+0.000
Again
Token Again
Feature activation+0.000
.
Token.
Feature activation+3.451
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.795
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
the
Token the
Feature activation+0.000
P
Token P
Feature activation+0.000
GA
TokenGA
Feature activation+0.000
Tour
Token Tour
Feature activation+0.182
.
Token.
Feature activation+2.857
Ċ
TokenĊ
Feature activation+3.175
Ċ
TokenĊ
Feature activation+0.202
Let
TokenLet
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Baby
Token Baby
Feature activation+0.000
Metal
TokenMetal
Feature activation+0.740
.
Token.
Feature activation+0.231
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.174
Ċ
TokenĊ
Feature activation+0.000
Baby
TokenBaby
Feature activation+0.000
Metal
TokenMetal
Feature activation+0.729
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
-
Token-
Feature activation+0.000
Mobile
TokenMobile
Feature activation+0.207
AWS
Token AWS
Feature activation+0.000
is
Token is
Feature activation+0.000
included
Token included
Feature activation+0.250
.
Token.
Feature activation+3.137
Notice
Token Notice
Feature activation+0.409
ably
Tokenably
Feature activation+0.000
missing
Token missing
Feature activation+0.388
,
Token,
Feature activation+0.096
however
Token however
Feature activation+0.027
for
Token for
Feature activation+0.000
personal
Token personal
Feature activation+0.000
use
Token use
Feature activation+0.000
only
Token only
Feature activation+0.000
.]
Token.]
Feature activation+0.331
Ċ
TokenĊ
Feature activation+3.105
Ċ
TokenĊ
Feature activation+0.167
1
Token1
Feature activation+0.000
Set
Token Set
Feature activation+0.000
up
Token up
Feature activation+0.000
on
Token on
Feature activation+0.000
for
Token for
Feature activation+0.000
over
Token over
Feature activation+0.000
5
Token 5
Feature activation+0.140
years
Token years
Feature activation+1.108
.
Token.
Feature activation+2.338
Ċ
TokenĊ
Feature activation+3.093
Ċ
TokenĊ
Feature activation+0.129
I
TokenI
Feature activation+0.000
spent
Token spent
Feature activation+0.000
27
Token 27
Feature activation+0.289
months
Token months
Feature activation+0.000
$
Token $
Feature activation+0.247
1
Token1
Feature activation+0.000
.
Token.
Feature activation+0.103
7
Token7
Feature activation+0.000
million
Token million
Feature activation+1.565
.
Token.
Feature activation+3.087
Ċ
TokenĊ
Feature activation+2.638
Ċ
TokenĊ
Feature activation+0.000
Ell
TokenEll
Feature activation+0.000
is
Tokenis
Feature activation+0.000
is
Token is
Feature activation+0.000
of
Token of
Feature activation+0.000
whiskey
Token whiskey
Feature activation+0.000
,
Token ,
Feature activation+0.000
2
Token 2
Feature activation+0.000
beers
Token beers
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.915
Ċ
TokenĊ
Feature activation+0.000
11
Token11
Feature activation+0.000
:
Token:
Feature activation+0.000
22
Token22
Feature activation+0.000
AM
Token AM
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
P
Token P
Feature activation+0.000
GA
TokenGA
Feature activation+0.000
Tour
Token Tour
Feature activation+0.182
.
Token.
Feature activation+2.857
Ċ
TokenĊ
Feature activation+3.175
Ċ
TokenĊ
Feature activation+0.202
Let
TokenLet
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
and
Token and
Feature activation+0.000
Hiro
Token Hiro
Feature activation+0.000
ki
Tokenki
Feature activation+0.000
Har
Token Har
Feature activation+0.000
ada
Tokenada
Feature activation+0.000
(
Token (
Feature activation+2.832
R
TokenR
Feature activation+0.000
ome
Tokenome
Feature activation+0.000
o
Tokeno
Feature activation+0.000
ÃĹ
Token ÃĹ
Feature activation+0.000
Juliet
Token Juliet
Feature activation+0.000
.
Token .
Feature activation+0.000
You
Token You
Feature activation+0.000
can
Token can
Feature activation+0.000
even
Token even
Feature activation+0.000
Dark
Token Dark
Feature activation+0.000
Rebirth
Token Rebirth
Feature activation+2.822
the
Token the
Feature activation+0.000
sage
Token sage
Feature activation+0.000
for
Token for
Feature activation+0.000
double
Token double
Feature activation+0.000
the
Token the
Feature activation+0.000
feedback
Token feedback
Feature activation+0.000
last
Token last
Feature activation+0.016
week
Token week
Feature activation+0.442
.
Token.
Feature activation+1.401
Ċ
TokenĊ
Feature activation+4.189
Ċ
TokenĊ
Feature activation+2.739
Article
TokenArticle
Feature activation+1.814
Continued
Token Continued
Feature activation+0.000
Below
Token Below
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

Top DFA by src position
MAX = 5.296

what
Token what
Feature activation+0.001
Top resid features:
price
Token price
Feature activation+0.012
Top resid features:
elastic
Token elastic
Feature activation+0.020
Top resid features:
ity
Tokenity
Feature activation+0.013
Top resid features:
means
Token means
Feature activation+0.012
Top resid features:
,
Token ,
Feature activation+5.065
Top resid features:
and
Token and
Feature activation+0.037
Top resid features:
how
Token how
Feature activation+0.003
Top resid features:
Amazon
Token Amazon
Feature activation-0.009
Top resid features:
's
Token's
Feature activation+0.007
Top resid features:
plan
Token plan
Feature activation+0.006
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.160
Top resid features:
photo
Token photo
Feature activation-0.000
Top resid features:
)
Token )
Feature activation+5.296
Top resid features:
Ċ
TokenĊ
Feature activation-0.032
Top resid features:
Ċ
TokenĊ
Feature activation-0.053
Top resid features:
H
TokenH
Feature activation-0.011
Top resid features:
os
Tokenos
Feature activation+0.000
Top resid features:
kins
Tokenkins
Feature activation+0.005
Top resid features:
autobiography
Token autobiography
Feature activation+0.006
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
Who
Token Who
Feature activation-0.006
Top resid features:
I
Token I
Feature activation+0.001
Top resid features:
Am
Token Am
Feature activation+0.000
Top resid features:
,
Token ,
Feature activation+4.904
Top resid features:
Pete
Token Pete
Feature activation+0.001
Top resid features:
discussed
Token discussed
Feature activation+0.017
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
perfect
Token perfect
Feature activation+0.004
Top resid features:
storm
Token storm
Feature activation+0.002
Top resid features:
I
Token I
Feature activation+0.001
Top resid features:
love
Token love
Feature activation+0.001
Top resid features:
Dave
Token Dave
Feature activation+0.005
Top resid features:
Gro
Token Gro
Feature activation-0.001
Top resid features:
hl
Tokenhl
Feature activation+0.005
Top resid features:
.
Token .
Feature activation+3.081
Top resid features:
Ċ
TokenĊ
Feature activation+0.019
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
So
TokenSo
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
after
Token after
Feature activation+0.000
Top resid features:
New
Token New
Feature activation+0.018
Top resid features:
York
Token York
Feature activation+0.007
Top resid features:
Post
Token Post
Feature activation-0.011
Top resid features:
reported
Token reported
Feature activation+0.016
Top resid features:
Friday
Token Friday
Feature activation+0.066
Top resid features:
,
Token ,
Feature activation+4.772
Top resid features:
after
Token after
Feature activation-0.016
Top resid features:
he
Token he
Feature activation-0.000
Top resid features:
had
Token had
Feature activation+0.002
Top resid features:
his
Token his
Feature activation-0.001
Top resid features:
class
Token class
Feature activation-0.004
Top resid features:
at
Token at
Feature activation-0.001
Top resid features:
a
Token a
Feature activation-0.000
Top resid features:
magnetic
Token magnetic
Feature activation+0.000
Top resid features:
name
Token name
Feature activation+0.002
Top resid features:
tag
Token tag
Feature activation+0.001
Top resid features:
,
Token ,
Feature activation+4.509
Top resid features:
Nick
Token Nick
Feature activation-0.025
Top resid features:
and
Token and
Feature activation-0.003
Top resid features:
Chris
Token Chris
Feature activation-0.003
Top resid features:
knew
Token knew
Feature activation+0.000
Top resid features:
that
Token that
Feature activation-0.001
Top resid features:
time
Token time
Feature activation+0.002
Top resid features:
to
Token to
Feature activation-0.000
Top resid features:
dump
Token dump
Feature activation+0.002
Top resid features:
your
Token your
Feature activation+0.001
Top resid features:
doctor
Token doctor
Feature activation+0.002
Top resid features:
.)
Token .)
Feature activation+4.807
Top resid features:
Ċ
TokenĊ
Feature activation+0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
MORE
TokenMORE
Feature activation+0.000
Top resid features:
:
Token:
Feature activation+0.000
Top resid features:
7
Token 7
Feature activation+0.000
Top resid features:
underwear
Token underwear
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
(
Token (
Feature activation+0.008
Top resid features:
Supp
Token Supp
Feature activation+0.019
Top resid features:
lied
Tokenlied
Feature activation+0.153
Top resid features:
)
Token )
Feature activation+4.525
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
T
TokenT
Feature activation+0.000
Top resid features:
iny
Tokeniny
Feature activation+0.000
Top resid features:
cap
Token cap
Feature activation+0.000
Top resid features:
New
Token New
Feature activation+0.009
Top resid features:
York
Token York
Feature activation+0.005
Top resid features:
Post
Token Post
Feature activation-0.007
Top resid features:
reported
Token reported
Feature activation+0.019
Top resid features:
Friday
Token Friday
Feature activation+0.022
Top resid features:
,
Token ,
Feature activation+4.580
Top resid features:
after
Token after
Feature activation-0.014
Top resid features:
he
Token he
Feature activation-0.019
Top resid features:
had
Token had
Feature activation+0.004
Top resid features:
his
Token his
Feature activation-0.004
Top resid features:
class
Token class
Feature activation-0.015
Top resid features:
local
Token local
Feature activation+0.013
Top resid features:
Michigan
Token Michigan
Feature activation+0.001
Top resid features:
sports
Token sports
Feature activation+0.004
Top resid features:
radio
Token radio
Feature activation+0.011
Top resid features:
station
Token station
Feature activation+0.002
Top resid features:
.
Token .
Feature activation+4.122
Top resid features:
There
Token There
Feature activation+0.003
Top resid features:
,
Token,
Feature activation+0.005
Top resid features:
Olsen
Token Olsen
Feature activation+0.001
Top resid features:
claimed
Token claimed
Feature activation-0.001
Top resid features:
that
Token that
Feature activation+0.012
Top resid features:
mastermind
Token mastermind
Feature activation+0.015
Top resid features:
Kob
Token Kob
Feature activation+0.002
Top resid features:
am
Tokenam
Feature activation+0.002
Top resid features:
et
Tokenet
Feature activation+0.001
Top resid features:
al
Tokenal
Feature activation-0.002
Top resid features:
.
Token .
Feature activation+4.001
Top resid features:
"
Token "
Feature activation+0.023
Top resid features:
Through
TokenThrough
Feature activation+0.005
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
last
Token last
Feature activation-0.001
Top resid features:
tour
Token tour
Feature activation+0.000
Top resid features:
would
Token would
Feature activation+0.005
Top resid features:
be
Token be
Feature activation+0.001
Top resid features:
a
Token a
Feature activation-0.000
Top resid features:
Verizon
Token Verizon
Feature activation+0.002
Top resid features:
exclusive
Token exclusive
Feature activation+0.003
Top resid features:
,
Token ,
Feature activation+3.586
Top resid features:
we
Token we
Feature activation+0.015
Top resid features:
could
Token could
Feature activation+0.003
Top resid features:
be
Token be
Feature activation+0.002
Top resid features:
looking
Token looking
Feature activation+0.000
Top resid features:
at
Token at
Feature activation-0.002
Top resid features:
these
Token these
Feature activation-0.002
Top resid features:
techniques
Token techniques
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.011
Top resid features:
click
Token click
Feature activation-0.001
Top resid features:
here
Token here
Feature activation+0.012
Top resid features:
.
Token .
Feature activation+4.138
Top resid features:
All
Token All
Feature activation+0.002
Top resid features:
files
Token files
Feature activation+0.000
Top resid features:
are
Token are
Feature activation+0.004
Top resid features:
for
Token for
Feature activation-0.002
Top resid features:
personal
Token personal
Feature activation+0.000
Top resid features:
economic
Token economic
Feature activation+0.007
Top resid features:
development
Token development
Feature activation+0.016
Top resid features:
advisor
Token advisor
Feature activation+0.010
Top resid features:
in
Token in
Feature activation+0.006
Top resid features:
Vietnam
Token Vietnam
Feature activation+0.016
Top resid features:
,
Token ,
Feature activation+3.874
Top resid features:
and
Token and
Feature activation+0.012
Top resid features:
was
Token was
Feature activation+0.004
Top resid features:
captured
Token captured
Feature activation+0.003
Top resid features:
by
Token by
Feature activation+0.004
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
to
Token to
Feature activation-0.004
Top resid features:
his
Token his
Feature activation+0.019
Top resid features:
latest
Token latest
Feature activation-0.025
Top resid features:
fundraising
Token fundraising
Feature activation+0.033
Top resid features:
report
Token report
Feature activation+0.001
Top resid features:
,
Token ,
Feature activation+4.015
Top resid features:
and
Token and
Feature activation+0.025
Top resid features:
raised
Token raised
Feature activation+0.046
Top resid features:
a
Token a
Feature activation-0.006
Top resid features:
total
Token total
Feature activation+0.022
Top resid features:
of
Token of
Feature activation+0.008
Top resid features:
:
Token:
Feature activation-0.007
Top resid features:
3
Token 3
Feature activation-0.016
Top resid features:
shots
Token shots
Feature activation+0.009
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
whiskey
Token whiskey
Feature activation+0.004
Top resid features:
,
Token ,
Feature activation+3.928
Top resid features:
2
Token 2
Feature activation+0.001
Top resid features:
beers
Token beers
Feature activation+0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.030
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
11
Token11
Feature activation+0.000
Top resid features:
local
Token local
Feature activation+0.014
Top resid features:
Michigan
Token Michigan
Feature activation-0.002
Top resid features:
sports
Token sports
Feature activation+0.000
Top resid features:
radio
Token radio
Feature activation+0.014
Top resid features:
station
Token station
Feature activation+0.005
Top resid features:
.
Token .
Feature activation+3.778
Top resid features:
There
Token There
Feature activation+0.017
Top resid features:
,
Token,
Feature activation+0.008
Top resid features:
Olsen
Token Olsen
Feature activation+0.015
Top resid features:
claimed
Token claimed
Feature activation-0.006
Top resid features:
that
Token that
Feature activation+0.026
Top resid features:
Love
Token Love
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.002
Top resid features:
Election
Token Election
Feature activation+0.000
Top resid features:
&
Token &
Feature activation+0.010
Top resid features:
Chocolate
Token Chocolate
Feature activation-0.004
Top resid features:
)
Token )
Feature activation+3.707
Top resid features:
is
Token is
Feature activation+0.028
Top resid features:
in
Token in
Feature activation+0.004
Top resid features:
charge
Token charge
Feature activation+0.005
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
impact
Token impact
Feature activation+0.004
Top resid features:
would
Token would
Feature activation+0.007
Top resid features:
be
Token be
Feature activation-0.003
Top resid features:
Royal
Token Royal
Feature activation+0.006
Top resid features:
Sage
Token Sage
Feature activation+0.013
Top resid features:
.
Token .
Feature activation+3.752
Top resid features:
You
Token You
Feature activation+0.015
Top resid features:
can
Token can
Feature activation+0.094
Top resid features:
even
Token even
Feature activation+0.009
Top resid features:
Dark
Token Dark
Feature activation+0.037
Top resid features:
Rebirth
Token Rebirth
Feature activation+0.035
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.089
Top resid features:
photo
Token photo
Feature activation-0.002
Top resid features:
)
Token )
Feature activation+3.583
Top resid features:
Ċ
TokenĊ
Feature activation-0.020
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
H
TokenH
Feature activation+0.009
Top resid features:
os
Tokenos
Feature activation-0.004
Top resid features:
kins
Tokenkins
Feature activation+0.038
Top resid features:

Decoder Weights Distribution

Head 0: 0.05

Head 1: 0.03

Head 2: 0.55

Head 3: 0.04

Head 4: 0.03

Head 5: 0.05

Head 6: 0.03

Head 7: 0.03

Head 8: 0.04

Head 9: 0.05

Head 10: 0.07

Head 11: 0.04

Positive logits

,7.55

.)7.10

,"7.08

."7.00

.6.72

 6.67

).6.23

;6.23

),6.17

)6.00

  5.98

 5.96

:5.89

   5.89

 5.83

):5.78

  5.74

'.5.35

)))5.30

))5.26

Negative logits

++)-3.25

////////////////////////////////-3.01

="/-2.89

)-2.77

-2.68

////////////////-2.58

="#-2.54

="-2.52

×-2.48

oken-2.46

],"-2.44

],[-2.37

astical-2.34

["-2.33

><-2.32

("-2.29

['-2.24

,—-2.19

—"-2.18

["-2.17

INTERVAL 3.893 - 4.326
CONTAINS 0.000%

love
Token love
Feature activation+0.000
Dave
Token Dave
Feature activation+0.000
Gro
Token Gro
Feature activation+0.000
hl
Tokenhl
Feature activation+0.856
.
Token .
Feature activation+0.733
Ċ
TokenĊ
Feature activation+3.973
Ċ
TokenĊ
Feature activation+0.757
So
TokenSo
Feature activation+0.000
,
Token,
Feature activation+0.000
after
Token after
Feature activation+0.000
Y
Token Y
Feature activation+0.000
actually
Token actually
Feature activation+0.073
make
Token make
Feature activation+0.000
authors
Token authors
Feature activation+0.000
more
Token more
Feature activation+0.000
money
Token money
Feature activation+1.148
.
Token.
Feature activation+4.326
They
Token They
Feature activation+0.000
lay
Token lay
Feature activation+0.000
it
Token it
Feature activation+0.000
out
Token out
Feature activation+0.000
pretty
Token pretty
Feature activation+0.000
May
Token May
Feature activation+0.203
23
Token 23
Feature activation+1.015
,
Token,
Feature activation+0.117
1969
Token 1969
Feature activation+1.451
.
Token.
Feature activation+1.340
Ċ
TokenĊ
Feature activation+4.107
Ċ
TokenĊ
Feature activation+0.095
For
TokenFor
Feature activation+0.000
one
Token one
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
their
Token their
Feature activation+0.000
feedback
Token feedback
Feature activation+0.000
last
Token last
Feature activation+0.016
week
Token week
Feature activation+0.442
.
Token.
Feature activation+1.401
Ċ
TokenĊ
Feature activation+4.189
Ċ
TokenĊ
Feature activation+2.739
Article
TokenArticle
Feature activation+1.814
Continued
Token Continued
Feature activation+0.000
Below
Token Below
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 3.461 - 3.893
CONTAINS 0.001%

to
Token to
Feature activation+0.000
dump
Token dump
Feature activation+0.000
your
Token your
Feature activation+0.000
doctor
Token doctor
Feature activation+0.000
.)
Token .)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.594
Ċ
TokenĊ
Feature activation+0.949
MORE
TokenMORE
Feature activation+0.014
:
Token:
Feature activation+0.270
7
Token 7
Feature activation+0.554
Weird
Token Weird
Feature activation+0.000
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
innovative
Token innovative
Feature activation+0.000
way
Token way
Feature activation+0.000
.
Token.
Feature activation+1.310
Ċ
TokenĊ
Feature activation+3.625
Ċ
TokenĊ
Feature activation+0.000
R
TokenR
Feature activation+0.000
ip
Tokenip
Feature activation+0.000
Cl
Token Cl
Feature activation+0.000
ips
Tokenips
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Supp
Token Supp
Feature activation+0.000
lied
Tokenlied
Feature activation+0.000
)
Token )
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.589
Ċ
TokenĊ
Feature activation+0.112
T
TokenT
Feature activation+0.000
iny
Tokeniny
Feature activation+0.000
cap
Token cap
Feature activation+0.000
es
Tokenes
Feature activation+0.000
f
Tokenf
Feature activation+0.000
Again
Token Again
Feature activation+0.000
.
Token.
Feature activation+3.451
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.795
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
move
Token move
Feature activation+0.000
has
Token has
Feature activation+0.000
sparked
Token sparked
Feature activation+0.000

INTERVAL 3.028 - 3.461
CONTAINS 0.001%

for
Token for
Feature activation+0.000
personal
Token personal
Feature activation+0.000
use
Token use
Feature activation+0.000
only
Token only
Feature activation+0.000
.]
Token.]
Feature activation+0.331
Ċ
TokenĊ
Feature activation+3.105
Ċ
TokenĊ
Feature activation+0.167
1
Token1
Feature activation+0.000
Set
Token Set
Feature activation+0.000
up
Token up
Feature activation+0.000
on
Token on
Feature activation+0.000
Baby
Token Baby
Feature activation+0.000
Metal
TokenMetal
Feature activation+0.740
.
Token.
Feature activation+0.231
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.174
Ċ
TokenĊ
Feature activation+0.000
Baby
TokenBaby
Feature activation+0.000
Metal
TokenMetal
Feature activation+0.729
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
Donald
Token Donald
Feature activation+0.000
Dr
Token Dr
Feature activation+0.000
ump
Tokenump
Feature activation+0.000
f
Tokenf
Feature activation+0.000
Again
Token Again
Feature activation+0.000
.
Token.
Feature activation+3.451
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.795
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
the
Token the
Feature activation+0.000
P
Token P
Feature activation+0.000
GA
TokenGA
Feature activation+0.000
Tour
Token Tour
Feature activation+0.182
.
Token.
Feature activation+2.857
Ċ
TokenĊ
Feature activation+3.175
Ċ
TokenĊ
Feature activation+0.202
Let
TokenLet
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
$
Token $
Feature activation+0.247
1
Token1
Feature activation+0.000
.
Token.
Feature activation+0.103
7
Token7
Feature activation+0.000
million
Token million
Feature activation+1.565
.
Token.
Feature activation+3.087
Ċ
TokenĊ
Feature activation+2.638
Ċ
TokenĊ
Feature activation+0.000
Ell
TokenEll
Feature activation+0.000
is
Tokenis
Feature activation+0.000
is
Token is
Feature activation+0.000

INTERVAL 2.596 - 3.028
CONTAINS 0.001%

1
Token1
Feature activation+0.000
.
Token.
Feature activation+0.103
7
Token7
Feature activation+0.000
million
Token million
Feature activation+1.565
.
Token.
Feature activation+3.087
Ċ
TokenĊ
Feature activation+2.638
Ċ
TokenĊ
Feature activation+0.000
Ell
TokenEll
Feature activation+0.000
is
Tokenis
Feature activation+0.000
is
Token is
Feature activation+0.000
running
Token running
Feature activation+0.000
.
Token .
Feature activation+0.000
You
Token You
Feature activation+0.000
can
Token can
Feature activation+0.000
even
Token even
Feature activation+0.000
Dark
Token Dark
Feature activation+0.000
Rebirth
Token Rebirth
Feature activation+2.822
the
Token the
Feature activation+0.000
sage
Token sage
Feature activation+0.000
for
Token for
Feature activation+0.000
double
Token double
Feature activation+0.000
the
Token the
Feature activation+0.000
feedback
Token feedback
Feature activation+0.000
last
Token last
Feature activation+0.016
week
Token week
Feature activation+0.442
.
Token.
Feature activation+1.401
Ċ
TokenĊ
Feature activation+4.189
Ċ
TokenĊ
Feature activation+2.739
Article
TokenArticle
Feature activation+1.814
Continued
Token Continued
Feature activation+0.000
Below
Token Below
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
of
Token of
Feature activation+0.000
whiskey
Token whiskey
Feature activation+0.000
,
Token ,
Feature activation+0.000
2
Token 2
Feature activation+0.000
beers
Token beers
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.915
Ċ
TokenĊ
Feature activation+0.000
11
Token11
Feature activation+0.000
:
Token:
Feature activation+0.000
22
Token22
Feature activation+0.000
AM
Token AM
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
P
Token P
Feature activation+0.000
GA
TokenGA
Feature activation+0.000
Tour
Token Tour
Feature activation+0.182
.
Token.
Feature activation+2.857
Ċ
TokenĊ
Feature activation+3.175
Ċ
TokenĊ
Feature activation+0.202
Let
TokenLet
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000

INTERVAL 2.163 - 2.596
CONTAINS 0.002%

card
Token card
Feature activation+0.000
igan
Tokenigan
Feature activation+0.312
,
Token,
Feature activation+0.000
below
Token below
Feature activation+0.000
.
Token.
Feature activation+0.812
Ċ
TokenĊ
Feature activation+2.485
Ċ
TokenĊ
Feature activation+0.000
Even
TokenEven
Feature activation+0.000
without
Token without
Feature activation+0.000
step
Token step
Feature activation+0.000
-
Token-
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
ruling
Token ruling
Feature activation+0.281
yesterday
Token yesterday
Feature activation+0.470
(
Token (
Feature activation+2.303
Friday
TokenFriday
Feature activation+0.000
).
Token).
Feature activation+0.437
Ċ
TokenĊ
Feature activation+1.366
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
prisoner
Token prisoner
Feature activation+0.000
for
Token for
Feature activation+0.000
over
Token over
Feature activation+0.000
5
Token 5
Feature activation+0.140
years
Token years
Feature activation+1.108
.
Token.
Feature activation+2.338
Ċ
TokenĊ
Feature activation+3.093
Ċ
TokenĊ
Feature activation+0.129
I
TokenI
Feature activation+0.000
spent
Token spent
Feature activation+0.000
27
Token 27
Feature activation+0.289
using
Tokenusing
Feature activation+0.000
Ni
Token Ni
Feature activation+0.000
all
Tokenall
Feature activation+0.000
Hor
Token Hor
Feature activation+0.000
an
Tokenan
Feature activation+0.496
'
Token '
Feature activation+2.259
s
Tokens
Feature activation+0.032
tissues
Token tissues
Feature activation+0.061
with
Token with
Feature activation+0.045
Sw
Token Sw
Feature activation+0.000
arov
Tokenarov
Feature activation+0.000
French
TokenFrench
Feature activation+0.000
General
Token General
Feature activation+0.000
Charles
Token Charles
Feature activation+0.000
de
Token de
Feature activation+0.000
Gaul
Token Gaul
Feature activation+0.000
le
Tokenle
Feature activation+2.369
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
other
Token other
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000

INTERVAL 1.730 - 2.163
CONTAINS 0.002%

streaming
Token streaming
Feature activation+0.000
and
Token and
Feature activation+0.000
download
Token download
Feature activation+0.000
.
Token.
Feature activation+0.321
Click
Token Click
Feature activation+0.000
here
Token here
Feature activation+1.924
to
Token to
Feature activation+0.097
subscribe
Token subscribe
Feature activation+0.000
.
Token.
Feature activation+0.000
E
Token E
Feature activation+0.000
-
Token-
Feature activation+0.000
the
Token the
Feature activation+0.000
phrase
Token phrase
Feature activation+0.000
,
Token,
Feature activation+0.000
river
Token river
Feature activation+0.000
,
Token ,
Feature activation+0.000
Ċ
TokenĊ
Feature activation+2.098
Ċ
TokenĊ
Feature activation+0.000
SELECT
TokenSELECT
Feature activation+0.000
cart
Token cart
Feature activation+0.000
od
Tokenod
Feature activation+0.000
b
Tokenb
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
possible
Token possible
Feature activation+0.000
deport
Token deport
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
.
Token.
Feature activation+1.918
Ċ
TokenĊ
Feature activation+2.305
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
Of
TokenOf
Feature activation+0.000
course
Token course
Feature activation+0.000
bombs
Token bombs
Feature activation+0.000
and
Token and
Feature activation+0.000
tear
Token tear
Feature activation+0.000
gas
Token gas
Feature activation+0.607
can
Token can
Feature activation+0.000
isters
Tokenisters
Feature activation+1.887
at
Token at
Feature activation+0.000
them
Token them
Feature activation+0.958
.
Token.
Feature activation+1.079
As
Token As
Feature activation+0.000
a
Token a
Feature activation+0.000
watch
Token watch
Feature activation+0.000
the
Token the
Feature activation+0.000
clip
Token clip
Feature activation+0.000
from
Token from
Feature activation+0.000
John
Token John
Feature activation+0.000
Oliver
Token Oliver
Feature activation+2.008
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
show
Token show
Feature activation+0.878
,
Token,
Feature activation+0.000

INTERVAL 1.298 - 1.730
CONTAINS 0.004%

on
Token on
Feature activation+0.483
May
Token May
Feature activation+0.203
23
Token 23
Feature activation+1.015
,
Token,
Feature activation+0.117
1969
Token 1969
Feature activation+1.451
.
Token.
Feature activation+1.340
Ċ
TokenĊ
Feature activation+4.107
Ċ
TokenĊ
Feature activation+0.095
For
TokenFor
Feature activation+0.000
one
Token one
Feature activation+0.000
,
Token,
Feature activation+0.000
local
Token local
Feature activation+0.000
Michigan
Token Michigan
Feature activation+0.000
sports
Token sports
Feature activation+0.000
radio
Token radio
Feature activation+0.000
station
Token station
Feature activation+0.000
.
Token .
Feature activation+1.646
There
Token There
Feature activation+0.000
,
Token,
Feature activation+0.000
Olsen
Token Olsen
Feature activation+0.000
claimed
Token claimed
Feature activation+0.409
that
Token that
Feature activation+0.401
released
Token released
Feature activation+1.021
on
Token on
Feature activation+0.483
May
Token May
Feature activation+0.203
23
Token 23
Feature activation+1.015
,
Token,
Feature activation+0.117
1969
Token 1969
Feature activation+1.451
.
Token.
Feature activation+1.340
Ċ
TokenĊ
Feature activation+4.107
Ċ
TokenĊ
Feature activation+0.095
For
TokenFor
Feature activation+0.000
one
Token one
Feature activation+0.000
:
Token :
Feature activation+0.000
2
Token 2
Feature activation+0.000
X
Token X
Feature activation+0.691
:
Token :
Feature activation+0.000
4
Token 4
Feature activation+0.000
Sum
Token Sum
Feature activation+1.458
:
Token :
Feature activation+0.000
6
Token 6
Feature activation+0.000
X
Token X
Feature activation+1.214
:
Token :
Feature activation+0.000
8
Token 8
Feature activation+0.000
upstream
Token upstream
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
P
Token P
Feature activation+0.000
AM
TokenAM
Feature activation+0.000
(
Token (
Feature activation+1.397
J
TokenJ
Feature activation+0.000
ine
Tokenine
Feature activation+0.000
k
Tokenk
Feature activation+0.000
et
Token et
Feature activation+0.000
al
Token al
Feature activation+0.000

INTERVAL 0.865 - 1.298
CONTAINS 0.006%

s
Tokens
Feature activation+0.000
student
Token student
Feature activation+0.000
and
Token and
Feature activation+0.000
parent
Token parent
Feature activation+0.000
supporters
Token supporters
Feature activation+0.000
.
Token.
Feature activation+0.900
Ċ
TokenĊ
Feature activation+2.292
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
online
Token online
Feature activation+0.000
petition
Token petition
Feature activation+0.000
foo
Token foo
Feature activation+0.000
,
Token ,
Feature activation+0.000
bar
Token bar
Feature activation+1.121
,
Token ,
Feature activation+0.000
b
Token b
Feature activation+0.110
az
Tokenaz
Feature activation+0.912
and
Token and
Feature activation+0.000
qu
Token qu
Feature activation+0.000
ux
Tokenux
Feature activation+0.000
over
Token over
Feature activation+0.000
four
Token four
Feature activation+0.000
Set
Token Set
Feature activation+0.000
up
Token up
Feature activation+0.000
on
Token on
Feature activation+0.000
Trip
Token Trip
Feature activation+0.000
od
Tokenod
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.067
Ċ
TokenĊ
Feature activation+0.095
Before
TokenBefore
Feature activation+0.000
you
Token you
Feature activation+0.000
start
Token start
Feature activation+0.000
in
Token in
Feature activation+0.000
opera
Token opera
Feature activation+0.027
Tommy
Token Tommy
Feature activation+0.000
,
Token ,
Feature activation+0.579
which
Token which
Feature activation+0.043
was
Token was
Feature activation+0.000
released
Token released
Feature activation+1.021
on
Token on
Feature activation+0.483
May
Token May
Feature activation+0.203
23
Token 23
Feature activation+1.015
,
Token,
Feature activation+0.117
1969
Token 1969
Feature activation+1.451
which
Token which
Feature activation+0.043
was
Token was
Feature activation+0.000
released
Token released
Feature activation+1.021
on
Token on
Feature activation+0.483
May
Token May
Feature activation+0.203
23
Token 23
Feature activation+1.015
,
Token,
Feature activation+0.117
1969
Token 1969
Feature activation+1.451
.
Token.
Feature activation+1.340
Ċ
TokenĊ
Feature activation+4.107
Ċ
TokenĊ
Feature activation+0.095

INTERVAL 0.433 - 0.865
CONTAINS 0.016%

to
Token to
Feature activation+0.000
promote
Token promote
Feature activation+0.000
their
Token their
Feature activation+0.000
new
Token new
Feature activation+0.000
album
Token album
Feature activation+0.057
.
Token.
Feature activation+0.544
See
Token See
Feature activation+0.000
where
Token where
Feature activation+0.000
they
Token they
Feature activation+0.000
'll
Token'll
Feature activation+0.000
be
Token be
Feature activation+0.000
Correct
Token Correct
Feature activation+0.000
it
Token it
Feature activation+0.048
is
Token is
Feature activation+0.000
an
Token an
Feature activation+0.021
Ancient
Token Ancient
Feature activation+0.000
System
Token System
Feature activation+0.790
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
System
Token System
Feature activation+0.152
not
Token not
Feature activation+0.046
only
Token only
Feature activation+0.001
control
Token control
Feature activation+0.000
of
Token of
Feature activation+0.000
organized
Token organized
Feature activation+0.000
labor
Token labor
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.437
Ċ
TokenĊ
Feature activation+0.000
Brazil
TokenBrazil
Feature activation+0.000
underwent
Token underwent
Feature activation+0.000
the
Token the
Feature activation+0.000
full
Token full
Feature activation+0.000
",
Token",
Feature activation+0.000
see
Token see
Feature activation+0.008
Euph
Token Euph
Feature activation+0.000
or
Tokenor
Feature activation+0.000
bi
Tokenbi
Feature activation+0.000
aceae
Tokenaceae
Feature activation+0.574
Ċ
TokenĊ
Feature activation+0.147
Ċ
TokenĊ
Feature activation+0.000
E
TokenE
Feature activation+0.000
uph
Tokenuph
Feature activation+0.000
or
Tokenor
Feature activation+0.000
insects
Token insects
Feature activation+0.000
can
Token can
Feature activation+0.000
overcome
Token overcome
Feature activation+0.000
those
Token those
Feature activation+0.000
defenses
Token defenses
Feature activation+0.000
.
Token.
Feature activation+0.641
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
At
Token At
Feature activation+0.000
the
Token the
Feature activation+0.000
smallest
Token smallest
Feature activation+0.000

INTERVAL 0.000 - 0.433
CONTAINS 99.969%

grants
Token grants
Feature activation+0.000
offered
Token offered
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
Panc
Token Panc
Feature activation+0.000
reat
Tokenreat
Feature activation+0.000
ic
Tokenic
Feature activation+0.000
Cancer
Token Cancer
Feature activation+0.000
Action
Token Action
Feature activation+0.000
Network
Token Network
Feature activation+0.000
(
Token (
Feature activation+0.000
De
Token De
Feature activation+0.000
Sh
TokenSh
Feature activation+0.000
azo
Tokenazo
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
visiting
Token visiting
Feature activation+0.000
professor
Token professor
Feature activation+0.000
in
Token in
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
original
Token original
Feature activation+0.000
PS
Token PS
Feature activation+0.000
D
TokenD
Feature activation+0.000
,
Token,
Feature activation+0.000
brushes
Token brushes
Feature activation+0.000
,
Token,
Feature activation+0.000
step
Token step
Feature activation+0.000
by
Token by
Feature activation+0.000
step
Token step
Feature activation+0.000
,
Token,
Feature activation+0.000
Court
Token Court
Feature activation+0.000
in
Token in
Feature activation+0.000
Washington
Token Washington
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Gray
TokenGray
Feature activation+0.000
-
Token-
Feature activation+0.000
Bur
TokenBur
Feature activation+0.000
r
Tokenr
Feature activation+0.000
iss
Tokeniss
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ben
TokenBen
Feature activation+0.000
io
Tokenio
Feature activation+0.000
ff
Tokenff
Feature activation+0.000
used
Token used
Feature activation+0.000
the
Token the
Feature activation+0.000
hashtag
Token hashtag
Feature activation+0.000
#
Token #
Feature activation+0.000
App
TokenApp
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 H8.2: (feature 8313)

TOP ACTIVATIONS
MAX = 7.708

"
Token "
Feature activation+0.000
def
Tokendef
Feature activation+0.000
eat
Tokeneat
Feature activation+0.000
devices
Token devices
Feature activation+0.000
"
Token"
Feature activation+1.572
Ċ
TokenĊ
Feature activation+7.708
Ċ
TokenĊ
Feature activation+4.102
Car
TokenCar
Feature activation+0.000
firm
Token firm
Feature activation+0.000
Volkswagen
Token Volkswagen
Feature activation+0.565
(
Token (
Feature activation+0.390
Image
TokenImage
Feature activation+0.000
copyright
Token copyright
Feature activation+0.085
Tim
Token Tim
Feature activation+0.000
Pe
Token Pe
Feature activation+0.000
ake
Tokenake
Feature activation+1.444
Ċ
TokenĊ
Feature activation+7.586
Ċ
TokenĊ
Feature activation+3.071
A
TokenA
Feature activation+0.839
copy
Token copy
Feature activation+0.000
of
Token of
Feature activation+0.000
one
Token one
Feature activation+0.000
conference
Token conference
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.278
space
Token space
Feature activation+0.000
station
Token station
Feature activation+0.258
Ċ
TokenĊ
Feature activation+7.472
Ċ
TokenĊ
Feature activation+4.138
"
Token"
Feature activation+0.000
It
TokenIt
Feature activation+0.000
's
Token's
Feature activation+0.000
not
Token not
Feature activation+0.000
US
Token US
Feature activation+0.000
on
Token on
Feature activation+0.841
foreign
Token foreign
Feature activation+0.000
policy
Token policy
Feature activation+0.558
.
Token.
Feature activation+3.370
Ċ
TokenĊ
Feature activation+7.136
Ċ
TokenĊ
Feature activation+0.847
David
TokenDavid
Feature activation+0.000
regards
Token regards
Feature activation+0.000
the
Token the
Feature activation+0.000
substantive
Token substantive
Feature activation+0.000
copyright
Token copyright
Feature activation+0.000
AFP
Token AFP
Feature activation+0.630
/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.134
Ċ
TokenĊ
Feature activation+3.761
Her
TokenHer
Feature activation+0.000
fian
Token fian
Feature activation+0.000
cé
Tokencé
Feature activation+0.000
,
Token,
Feature activation+0.000
concepts
Token concepts
Feature activation+0.000
would
Token would
Feature activation+0.000
have
Token have
Feature activation+0.000
polar
Token polar
Feature activation+0.000
applications
Token applications
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.023
Ċ
TokenĊ
Feature activation+4.830
The
TokenThe
Feature activation+0.846
multi
Token multi
Feature activation+0.365
-
Token-
Feature activation+0.000
billion
Tokenbillion
Feature activation+0.000
agent
Tokenagent
Feature activation+0.000
prov
Token prov
Feature activation+0.000
ocate
Tokenocate
Feature activation+0.000
ur
Tokenur
Feature activation+0.000
"
Token"
Feature activation+1.310
Ċ
TokenĊ
Feature activation+7.002
Ċ
TokenĊ
Feature activation+2.396
Her
TokenHer
Feature activation+0.000
husband
Token husband
Feature activation+0.072
of
Token of
Feature activation+0.000
53
Token 53
Feature activation+1.259
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Image
TokenImage
Feature activation+0.000
copyright
Token copyright
Feature activation+0.014
Getty
Token Getty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.952
Ċ
TokenĊ
Feature activation+3.390
A
TokenA
Feature activation+1.149
new
Token new
Feature activation+0.553
study
Token study
Feature activation+0.982
suggests
Token suggests
Feature activation+0.648
the
Token the
Feature activation+2.072
BBC
Token BBC
Feature activation+0.000
Wales
Token Wales
Feature activation+0.000
leaders
Token leaders
Feature activation+0.000
debate
Token debate
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.702
Ċ
TokenĊ
Feature activation+1.450
A
TokenA
Feature activation+0.262
spokesman
Token spokesman
Feature activation+0.000
for
Token for
Feature activation+0.012
BBC
Token BBC
Feature activation+0.000
may
Token may
Feature activation+0.000
have
Token have
Feature activation+0.000
sparked
Token sparked
Feature activation+0.000
the
Token the
Feature activation+0.177
fire
Token fire
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.563
Ċ
TokenĊ
Feature activation+2.894
The
TokenThe
Feature activation+0.286
death
Token death
Feature activation+0.000
toll
Token toll
Feature activation+0.000
from
Token from
Feature activation+0.000
get
Tokenget
Feature activation+0.000
Stoke
Token Stoke
Feature activation+0.000
in
Token in
Feature activation+0.000
space
Token space
Feature activation+0.000
"
Token"
Feature activation+2.599
Ċ
TokenĊ
Feature activation+6.435
Ċ
TokenĊ
Feature activation+2.823
The
TokenThe
Feature activation+0.000
M
Token M
Feature activation+0.000
appa
Tokenappa
Feature activation+0.000
Mund
Token Mund
Feature activation+0.000
have
Token have
Feature activation+0.000
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
shadow
Token shadow
Feature activation+0.000
cabinet
Token cabinet
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.426
Ċ
TokenĊ
Feature activation+4.247
Following
TokenFollowing
Feature activation+0.760
his
Token his
Feature activation+0.849
speech
Token speech
Feature activation+0.524
to
Token to
Feature activation+1.165
ush
Tokenush
Feature activation+0.000
an
Tokenan
Feature activation+0.079
in
Token in
Feature activation+0.865
2005
Token 2005
Feature activation+3.192
.
Token.
Feature activation+0.835
Ċ
TokenĊ
Feature activation+6.200
Ċ
TokenĊ
Feature activation+1.756
M
TokenM
Feature activation+0.000
ang
Tokenang
Feature activation+0.000
esh
Tokenesh
Feature activation+0.000
kar
Tokenkar
Feature activation+0.000
taste
Token taste
Feature activation+0.000
of
Token of
Feature activation+0.128
her
Token her
Feature activation+0.049
daily
Token daily
Feature activation+0.241
commute
Token commute
Feature activation+1.512
Ċ
TokenĊ
Feature activation+6.118
Ċ
TokenĊ
Feature activation+3.515
For
TokenFor
Feature activation+0.979
C
Token C
Feature activation+0.000
resp
Tokenresp
Feature activation+0.000
o
Tokeno
Feature activation+0.532
a
Token a
Feature activation+0.000
prominent
Token prominent
Feature activation+0.000
businessman
Token businessman
Feature activation+0.000
was
Token was
Feature activation+0.000
murdered
Token murdered
Feature activation+0.161
Ċ
TokenĊ
Feature activation+6.054
Ċ
TokenĊ
Feature activation+2.038
Lead
TokenLead
Feature activation+0.009
ers
Tokeners
Feature activation+0.060
of
Token of
Feature activation+0.625
Ukraine
Token Ukraine
Feature activation+0.014
know
Token know
Feature activation+0.000
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
details
Token details
Feature activation+0.000
"
Token"
Feature activation+3.747
Ċ
TokenĊ
Feature activation+5.839
Ċ
TokenĊ
Feature activation+2.365
Syria
TokenSyria
Feature activation+0.000
's
Token's
Feature activation+1.071
President
Token President
Feature activation+0.000
Bashar
Token Bashar
Feature activation+0.000
lict
Tokenlict
Feature activation+0.000
-
Token-
Feature activation+0.000
free
Tokenfree
Feature activation+0.000
"
Token"
Feature activation+0.000
mines
Token mines
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.826
Ċ
TokenĊ
Feature activation+2.913
More
TokenMore
Feature activation+0.434
than
Token than
Feature activation+0.319
two
Token two
Feature activation+0.646
-
Token-
Feature activation+0.235
blocking
Token blocking
Feature activation+0.000
Sev
Token Sev
Feature activation+0.000
ast
Tokenast
Feature activation+0.000
opol
Tokenopol
Feature activation+0.000
airport
Token airport
Feature activation+1.144
Ċ
TokenĊ
Feature activation+5.815
Ċ
TokenĊ
Feature activation+4.697
And
TokenAnd
Feature activation+0.040
he
Token he
Feature activation+0.634
said
Token said
Feature activation+0.154
the
Token the
Feature activation+0.000
victims
Token victims
Feature activation+0.000
died
Token died
Feature activation+0.000
in
Token in
Feature activation+0.000
hospital
Token hospital
Feature activation+0.000
.
Token.
Feature activation+2.433
Ċ
TokenĊ
Feature activation+5.769
Ċ
TokenĊ
Feature activation+2.634
The
TokenThe
Feature activation+0.159
fire
Token fire
Feature activation+0.000
was
Token was
Feature activation+0.000
sparked
Token sparked
Feature activation+0.000
violence
Token violence
Feature activation+0.000
often
Token often
Feature activation+0.156
goes
Token goes
Feature activation+0.168
unre
Token unre
Feature activation+0.000
ported
Tokenported
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.721
Ċ
TokenĊ
Feature activation+2.176
The
TokenThe
Feature activation+0.000
Conservatives
Token Conservatives
Feature activation+0.000
have
Token have
Feature activation+0.000
pledged
Token pledged
Feature activation+0.000

Top DFA by src position
MAX = 8.181

<|endoftext|>
Token<|endoftext|>
Feature activation+0.061
Top resid features:
Image
TokenImage
Feature activation+0.171
Top resid features:
copyright
Token copyright
Feature activation+6.932
Top resid features:
AFP
Token AFP
Feature activation-0.030
Top resid features:
/
Token/
Feature activation+0.010
Top resid features:
Getty
TokenGetty
Feature activation-0.002
Top resid features:
Images
Token Images
Feature activation-0.005
Top resid features:
Image
Token Image
Feature activation+0.013
Top resid features:
room
Token room
Feature activation+0.000
Top resid features:
,
Token,
Feature activation-0.000
Top resid features:
dress
Token dress
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.159
Top resid features:
Image
TokenImage
Feature activation+0.368
Top resid features:
copyright
Token copyright
Feature activation+8.181
Top resid features:
Tim
Token Tim
Feature activation+0.012
Top resid features:
Pe
Token Pe
Feature activation-0.007
Top resid features:
ake
Tokenake
Feature activation-0.007
Top resid features:
Ċ
TokenĊ
Feature activation-0.201
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.048
Top resid features:
Ċ
TokenĊ
Feature activation-0.060
Top resid features:
Image
TokenImage
Feature activation+0.197
Top resid features:
copyright
Token copyright
Feature activation+6.797
Top resid features:
Es
Token Es
Feature activation-0.001
Top resid features:
a
Tokena
Feature activation-0.007
Top resid features:
Image
Token Image
Feature activation+0.012
Top resid features:
caption
Token caption
Feature activation+1.336
Top resid features:
Tim
Token Tim
Feature activation+0.003
Top resid features:
brothers
Token brothers
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+0.018
Top resid features:
Image
TokenImage
Feature activation+0.429
Top resid features:
caption
Token caption
Feature activation+7.127
Top resid features:
Ed
Token Ed
Feature activation+0.001
Top resid features:
Miliband
Token Miliband
Feature activation-0.003
Top resid features:
will
Token will
Feature activation+0.004
Top resid features:
have
Token have
Feature activation+0.001
Top resid features:
a
Token a
Feature activation+0.001
Top resid features:
emperor
Token emperor
Feature activation-0.008
Top resid features:
.
Token.
Feature activation-0.010
Top resid features:
Ċ
TokenĊ
Feature activation-0.024
Top resid features:
Ċ
TokenĊ
Feature activation-0.016
Top resid features:
Image
TokenImage
Feature activation+0.346
Top resid features:
copyright
Token copyright
Feature activation+7.679
Top resid features:
AFP
Token AFP
Feature activation-0.049
Top resid features:
/
Token/
Feature activation+0.016
Top resid features:
Getty
TokenGetty
Feature activation-0.008
Top resid features:
Images
Token Images
Feature activation-0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.021
Top resid features:
construction
Token construction
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.055
Top resid features:
Image
TokenImage
Feature activation+0.280
Top resid features:
copyright
Token copyright
Feature activation+5.914
Top resid features:
ESA
Token ESA
Feature activation-0.021
Top resid features:
Image
Token Image
Feature activation+0.015
Top resid features:
caption
Token caption
Feature activation+1.284
Top resid features:
Three
Token Three
Feature activation-0.005
Top resid features:
of
Token of
Feature activation+0.008
Top resid features:
website
Token website
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.008
Top resid features:
Ċ
TokenĊ
Feature activation-0.012
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
Image
TokenImage
Feature activation+0.443
Top resid features:
caption
Token caption
Feature activation+7.254
Top resid features:
Claire
Token Claire
Feature activation+0.000
Top resid features:
Ray
Token Ray
Feature activation-0.001
Top resid features:
ner
Tokenner
Feature activation+0.001
Top resid features:
's
Token's
Feature activation+0.004
Top resid features:
son
Token son
Feature activation+0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.075
Top resid features:
sky
Token sky
Feature activation-0.002
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.083
Top resid features:
Image
TokenImage
Feature activation+0.545
Top resid features:
copyright
Token copyright
Feature activation+7.347
Top resid features:
Getty
Token Getty
Feature activation-0.033
Top resid features:
Images
Token Images
Feature activation-0.032
Top resid features:
Ċ
TokenĊ
Feature activation-0.203
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
A
TokenA
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.062
Top resid features:
Ċ
TokenĊ
Feature activation-0.045
Top resid features:
Image
TokenImage
Feature activation+0.379
Top resid features:
caption
Token caption
Feature activation+6.995
Top resid features:
Mr
Token Mr
Feature activation+0.060
Top resid features:
Mill
Token Mill
Feature activation-0.000
Top resid features:
ar
Tokenar
Feature activation-0.006
Top resid features:
is
Token is
Feature activation+0.002
Top resid features:
one
Token one
Feature activation+0.003
Top resid features:
unsupported
Token unsupported
Feature activation-0.035
Top resid features:
on
Token on
Feature activation-0.003
Top resid features:
your
Token your
Feature activation-0.015
Top resid features:
device
Token device
Feature activation-0.003
Top resid features:
Media
Token Media
Feature activation+0.145
Top resid features:
caption
Token caption
Feature activation+7.224
Top resid features:
People
Token People
Feature activation-0.003
Top resid features:
smoking
Token smoking
Feature activation+0.020
Top resid features:
at
Token at
Feature activation-0.036
Top resid features:
the
Token the
Feature activation-0.009
Top resid features:
scene
Token scene
Feature activation+0.010
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation-0.024
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Image
TokenImage
Feature activation+0.338
Top resid features:
caption
Token caption
Feature activation+6.885
Top resid features:
Pe
Token Pe
Feature activation-0.005
Top resid features:
ake
Tokenake
Feature activation-0.012
Top resid features:
packed
Token packed
Feature activation-0.008
Top resid features:
a
Token a
Feature activation+0.005
Top resid features:
Stoke
Token Stoke
Feature activation-0.016
Top resid features:
brothers
Token brothers
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.050
Top resid features:
Image
TokenImage
Feature activation+0.873
Top resid features:
caption
Token caption
Feature activation+6.087
Top resid features:
Ed
Token Ed
Feature activation+0.004
Top resid features:
Miliband
Token Miliband
Feature activation-0.009
Top resid features:
will
Token will
Feature activation+0.011
Top resid features:
have
Token have
Feature activation+0.004
Top resid features:
a
Token a
Feature activation+0.007
Top resid features:
s
Tokens
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.018
Top resid features:
Ċ
TokenĊ
Feature activation+0.028
Top resid features:
Image
TokenImage
Feature activation+0.321
Top resid features:
caption
Token caption
Feature activation+6.430
Top resid features:
M
Token M
Feature activation-0.009
Top resid features:
anna
Tokenanna
Feature activation-0.001
Top resid features:
De
Token De
Feature activation+0.002
Top resid features:
y
Tokeny
Feature activation+0.004
Top resid features:
won
Token won
Feature activation+0.014
Top resid features:
unsupported
Token unsupported
Feature activation+0.053
Top resid features:
on
Token on
Feature activation-0.000
Top resid features:
your
Token your
Feature activation-0.002
Top resid features:
device
Token device
Feature activation+0.004
Top resid features:
Media
Token Media
Feature activation+0.158
Top resid features:
caption
Token caption
Feature activation+4.099
Top resid features:
Fab
Token Fab
Feature activation+0.002
Top resid features:
iana
Tokeniana
Feature activation-0.000
Top resid features:
C
Token C
Feature activation-0.004
Top resid features:
resp
Tokenresp
Feature activation-0.004
Top resid features:
o
Tokeno
Feature activation-0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.074
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.050
Top resid features:
Image
TokenImage
Feature activation+0.773
Top resid features:
copyright
Token copyright
Feature activation+4.835
Top resid features:
Getty
Token Getty
Feature activation-0.006
Top resid features:
Images
Token Images
Feature activation-0.004
Top resid features:
Image
Token Image
Feature activation+0.041
Top resid features:
caption
Token caption
Feature activation+1.411
Top resid features:
Some
Token Some
Feature activation-0.001
Top resid features:
unsupported
Token unsupported
Feature activation+0.039
Top resid features:
on
Token on
Feature activation+0.002
Top resid features:
your
Token your
Feature activation+0.000
Top resid features:
device
Token device
Feature activation+0.003
Top resid features:
Media
Token Media
Feature activation+0.139
Top resid features:
caption
Token caption
Feature activation+3.959
Top resid features:
President
Token President
Feature activation-0.005
Top resid features:
Bashar
Token Bashar
Feature activation-0.014
Top resid features:
al
Token al
Feature activation+0.003
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
Assad
TokenAssad
Feature activation-0.051
Top resid features:
to
Token to
Feature activation-0.002
Top resid features:
@
Token @
Feature activation-0.003
Top resid features:
Jewish
TokenJewish
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.124
Top resid features:
Image
TokenImage
Feature activation+0.900
Top resid features:
copyright
Token copyright
Feature activation+4.736
Top resid features:
AFP
Token AFP
Feature activation-0.064
Top resid features:
Image
Token Image
Feature activation+0.031
Top resid features:
caption
Token caption
Feature activation+1.243
Top resid features:
Many
Token Many
Feature activation+0.015
Top resid features:
electronic
Token electronic
Feature activation-0.006
Top resid features:
unsupported
Token unsupported
Feature activation+0.095
Top resid features:
on
Token on
Feature activation+0.002
Top resid features:
your
Token your
Feature activation-0.001
Top resid features:
device
Token device
Feature activation+0.015
Top resid features:
Media
Token Media
Feature activation+0.385
Top resid features:
caption
Token caption
Feature activation+3.864
Top resid features:
Christian
Token Christian
Feature activation-0.001
Top resid features:
Fraser
Token Fraser
Feature activation-0.000
Top resid features:
says
Token says
Feature activation-0.016
Top resid features:
barriers
Token barriers
Feature activation+0.006
Top resid features:
and
Token and
Feature activation-0.004
Top resid features:
unsupported
Token unsupported
Feature activation-0.023
Top resid features:
on
Token on
Feature activation+0.001
Top resid features:
your
Token your
Feature activation+0.002
Top resid features:
device
Token device
Feature activation-0.000
Top resid features:
Media
Token Media
Feature activation+0.259
Top resid features:
caption
Token caption
Feature activation+6.371
Top resid features:
People
Token People
Feature activation-0.012
Top resid features:
smoking
Token smoking
Feature activation+0.001
Top resid features:
at
Token at
Feature activation-0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
scene
Token scene
Feature activation-0.000
Top resid features:
border
Token border
Feature activation-0.003
Top resid features:
and
Token and
Feature activation-0.002
Top resid features:
av
Token av
Feature activation-0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.062
Top resid features:
Image
TokenImage
Feature activation+0.459
Top resid features:
copyright
Token copyright
Feature activation+4.846
Top resid features:
Think
Token Think
Feature activation-0.005
Top resid features:
stock
Tokenstock
Feature activation+0.022
Top resid features:
Image
Token Image
Feature activation+0.025
Top resid features:
caption
Token caption
Feature activation+1.341
Top resid features:
Domestic
Token Domestic
Feature activation-0.027
Top resid features:

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.01

Head 2: 0.54

Head 3: 0.04

Head 4: 0.03

Head 5: 0.04

Head 6: 0.02

Head 7: 0.03

Head 8: 0.03

Head 9: 0.04

Head 10: 0.08

Head 11: 0.05

Positive logits

Nato5.91

BBC5.45

recognise5.04

BBC4.96

Nasa4.63

apologised4.62

programme4.60

apologise4.60

4.58

maths4.57

civilisation4.56

scept4.47

recognised4.42

programmes4.25

realise4.23

organisers4.17

analys4.14

organise4.14

realised4.05

organising4.03

Negative logits

utilized-3.39

orgetown-3.32

ADVERTISEMENT-3.31

aukee-3.18

caliber-3.14

colored-3.13

utilize-3.11

utilizing-3.06

imize-2.99

advertisement-2.99

ngth-2.97

Coliseum-2.97

amily-2.83

totaling-2.77

LEASE-2.76

ichick-2.73

capitalize-2.70

-2.69

adena-2.68

-2.65

INTERVAL 6.937 - 7.708
CONTAINS 0.001%

copyright
Token copyright
Feature activation+0.000
AFP
Token AFP
Feature activation+0.630
/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.134
Ċ
TokenĊ
Feature activation+3.761
Her
TokenHer
Feature activation+0.000
fian
Token fian
Feature activation+0.000
cé
Tokencé
Feature activation+0.000
,
Token,
Feature activation+0.000
agent
Tokenagent
Feature activation+0.000
prov
Token prov
Feature activation+0.000
ocate
Tokenocate
Feature activation+0.000
ur
Tokenur
Feature activation+0.000
"
Token"
Feature activation+1.310
Ċ
TokenĊ
Feature activation+7.002
Ċ
TokenĊ
Feature activation+2.396
Her
TokenHer
Feature activation+0.000
husband
Token husband
Feature activation+0.072
of
Token of
Feature activation+0.000
53
Token 53
Feature activation+1.259
concepts
Token concepts
Feature activation+0.000
would
Token would
Feature activation+0.000
have
Token have
Feature activation+0.000
polar
Token polar
Feature activation+0.000
applications
Token applications
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.023
Ċ
TokenĊ
Feature activation+4.830
The
TokenThe
Feature activation+0.846
multi
Token multi
Feature activation+0.365
-
Token-
Feature activation+0.000
billion
Tokenbillion
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Image
TokenImage
Feature activation+0.000
copyright
Token copyright
Feature activation+0.014
Getty
Token Getty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.952
Ċ
TokenĊ
Feature activation+3.390
A
TokenA
Feature activation+1.149
new
Token new
Feature activation+0.553
study
Token study
Feature activation+0.982
suggests
Token suggests
Feature activation+0.648
conference
Token conference
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.278
space
Token space
Feature activation+0.000
station
Token station
Feature activation+0.258
Ċ
TokenĊ
Feature activation+7.472
Ċ
TokenĊ
Feature activation+4.138
"
Token"
Feature activation+0.000
It
TokenIt
Feature activation+0.000
's
Token's
Feature activation+0.000
not
Token not
Feature activation+0.000

INTERVAL 6.166 - 6.937
CONTAINS 0.001%

have
Token have
Feature activation+0.000
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
shadow
Token shadow
Feature activation+0.000
cabinet
Token cabinet
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.426
Ċ
TokenĊ
Feature activation+4.247
Following
TokenFollowing
Feature activation+0.760
his
Token his
Feature activation+0.849
speech
Token speech
Feature activation+0.524
to
Token to
Feature activation+1.165
the
Token the
Feature activation+2.072
BBC
Token BBC
Feature activation+0.000
Wales
Token Wales
Feature activation+0.000
leaders
Token leaders
Feature activation+0.000
debate
Token debate
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.702
Ċ
TokenĊ
Feature activation+1.450
A
TokenA
Feature activation+0.262
spokesman
Token spokesman
Feature activation+0.000
for
Token for
Feature activation+0.012
BBC
Token BBC
Feature activation+0.000
ush
Tokenush
Feature activation+0.000
an
Tokenan
Feature activation+0.079
in
Token in
Feature activation+0.865
2005
Token 2005
Feature activation+3.192
.
Token.
Feature activation+0.835
Ċ
TokenĊ
Feature activation+6.200
Ċ
TokenĊ
Feature activation+1.756
M
TokenM
Feature activation+0.000
ang
Tokenang
Feature activation+0.000
esh
Tokenesh
Feature activation+0.000
kar
Tokenkar
Feature activation+0.000
get
Tokenget
Feature activation+0.000
Stoke
Token Stoke
Feature activation+0.000
in
Token in
Feature activation+0.000
space
Token space
Feature activation+0.000
"
Token"
Feature activation+2.599
Ċ
TokenĊ
Feature activation+6.435
Ċ
TokenĊ
Feature activation+2.823
The
TokenThe
Feature activation+0.000
M
Token M
Feature activation+0.000
appa
Tokenappa
Feature activation+0.000
Mund
Token Mund
Feature activation+0.000
may
Token may
Feature activation+0.000
have
Token have
Feature activation+0.000
sparked
Token sparked
Feature activation+0.000
the
Token the
Feature activation+0.177
fire
Token fire
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.563
Ċ
TokenĊ
Feature activation+2.894
The
TokenThe
Feature activation+0.286
death
Token death
Feature activation+0.000
toll
Token toll
Feature activation+0.000
from
Token from
Feature activation+0.000

INTERVAL 5.396 - 6.166
CONTAINS 0.002%

immensely
Token immensely
Feature activation+0.000
proud
Token proud
Feature activation+0.000
of
Token of
Feature activation+0.000
her
Token her
Feature activation+0.000
."
Token."
Feature activation+2.633
Ċ
TokenĊ
Feature activation+5.707
Ċ
TokenĊ
Feature activation+0.762
Mr
TokenMr
Feature activation+0.363
Ray
Token Ray
Feature activation+0.000
ner
Tokenner
Feature activation+0.670
,
Token,
Feature activation+0.947
lead
Token lead
Feature activation+0.027
on
Token on
Feature activation+0.119
the
Token the
Feature activation+0.000
project
Token project
Feature activation+0.423
.
Token.
Feature activation+0.665
Ċ
TokenĊ
Feature activation+5.580
Ċ
TokenĊ
Feature activation+0.881
In
TokenIn
Feature activation+0.000
other
Token other
Feature activation+0.000
words
Token words
Feature activation+0.400
,
Token,
Feature activation+0.433
having
Token having
Feature activation+0.000
treatment
Token treatment
Feature activation+0.000
following
Token following
Feature activation+0.000
the
Token the
Feature activation+0.000
allegations
Token allegations
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.693
Ċ
TokenĊ
Feature activation+1.785
House
TokenHouse
Feature activation+0.000
of
Token of
Feature activation+0.000
Cards
Token Cards
Feature activation+0.000
could
Token could
Feature activation+0.000
many
Token many
Feature activation+0.066
court
Token court
Feature activation+0.092
users
Token users
Feature activation+0.632
,
Token,
Feature activation+3.546
the
Token the
Feature activation+3.231
report
Token report
Feature activation+5.677
suggested
Token suggested
Feature activation+2.756
Ċ
TokenĊ
Feature activation+5.009
Ċ
TokenĊ
Feature activation+0.919
Vict
TokenVict
Feature activation+0.000
ims
Tokenims
Feature activation+0.000
know
Token know
Feature activation+0.000
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
details
Token details
Feature activation+0.000
"
Token"
Feature activation+3.747
Ċ
TokenĊ
Feature activation+5.839
Ċ
TokenĊ
Feature activation+2.365
Syria
TokenSyria
Feature activation+0.000
's
Token's
Feature activation+1.071
President
Token President
Feature activation+0.000
Bashar
Token Bashar
Feature activation+0.000

INTERVAL 4.625 - 5.396
CONTAINS 0.001%

his
Token his
Feature activation+0.000
own
Token own
Feature activation+0.000
career
Token career
Feature activation+0.356
intentions
Token intentions
Feature activation+0.074
.
Token.
Feature activation+2.039
Ċ
TokenĊ
Feature activation+5.079
Ċ
TokenĊ
Feature activation+0.516
Brother
TokenBrother
Feature activation+0.000
ly
Tokenly
Feature activation+0.000
scrutiny
Token scrutiny
Feature activation+0.298
Ċ
TokenĊ
Feature activation+3.257
more
Token more
Feature activation+0.000
than
Token than
Feature activation+0.000
a
Token a
Feature activation+0.000
nuisance
Token nuisance
Feature activation+0.000
.
Token.
Feature activation+2.076
Ċ
TokenĊ
Feature activation+4.978
Ċ
TokenĊ
Feature activation+1.309
Heavy
TokenHeavy
Feature activation+0.000
traffic
Token traffic
Feature activation+0.000
is
Token is
Feature activation+0.000
an
Token an
Feature activation+0.000
Allen
Token Allen
Feature activation+0.000
homes
Token homes
Feature activation+0.878
in
Token in
Feature activation+0.538
North
Token North
Feature activation+0.094
Philadelphia
Token Philadelphia
Feature activation+0.506
Ċ
TokenĊ
Feature activation+4.701
Ċ
TokenĊ
Feature activation+1.203
Less
TokenLess
Feature activation+0.000
than
Token than
Feature activation+0.000
two
Token two
Feature activation+0.000
weeks
Token weeks
Feature activation+0.489
would
Token would
Feature activation+0.000
have
Token have
Feature activation+0.000
polar
Token polar
Feature activation+0.000
applications
Token applications
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.023
Ċ
TokenĊ
Feature activation+4.830
The
TokenThe
Feature activation+0.846
multi
Token multi
Feature activation+0.365
-
Token-
Feature activation+0.000
billion
Tokenbillion
Feature activation+0.000
-
Token-
Feature activation+0.002
,
Token,
Feature activation+0.527
B
Token B
Feature activation+1.457
aid
Tokenaid
Feature activation+0.012
u
Tokenu
Feature activation+4.774
said
Token said
Feature activation+1.741
Ċ
TokenĊ
Feature activation+5.234
Ċ
TokenĊ
Feature activation+2.202
Elect
TokenElect
Feature activation+0.000
ronic
Tokenronic
Feature activation+0.000
chop
Token chop
Feature activation+0.000
sticks
Tokensticks
Feature activation+0.745

INTERVAL 3.854 - 4.625
CONTAINS 0.002%

from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.278
space
Token space
Feature activation+0.000
station
Token station
Feature activation+0.258
Ċ
TokenĊ
Feature activation+7.472
Ċ
TokenĊ
Feature activation+4.138
"
Token"
Feature activation+0.000
It
TokenIt
Feature activation+0.000
's
Token's
Feature activation+0.000
not
Token not
Feature activation+0.000
an
Token an
Feature activation+0.000
Semitism
TokenSemitism
Feature activation+0.000
"
Token"
Feature activation+0.022
in
Token in
Feature activation+0.000
Ukraine
Token Ukraine
Feature activation+0.612
.
Token.
Feature activation+4.735
Ċ
TokenĊ
Feature activation+3.994
Ċ
TokenĊ
Feature activation+0.000
Russian
TokenRussian
Feature activation+0.000
media
Token media
Feature activation+0.000
and
Token and
Feature activation+0.000
officials
Token officials
Feature activation+0.821
US
Token US
Feature activation+0.000
-
Token-
Feature activation+0.000
based
Tokenbased
Feature activation+0.000
Enough
Token Enough
Feature activation+0.000
Project
Token Project
Feature activation+0.261
says
Token says
Feature activation+4.125
.
Token.
Feature activation+2.495
Ċ
TokenĊ
Feature activation+1.760
Ċ
TokenĊ
Feature activation+0.000
It
TokenIt
Feature activation+0.514
follows
Token follows
Feature activation+0.000
dis
Tokendis
Feature activation+0.000
gr
Tokengr
Feature activation+0.000
ace
Tokenace
Feature activation+0.000
ful
Tokenful
Feature activation+0.000
'
Token'
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.306
Ċ
TokenĊ
Feature activation+3.256
Mrs
TokenMrs
Feature activation+0.000
O
Token O
Feature activation+0.000
'
Token'
Feature activation+0.000
Neill
TokenNeill
Feature activation+0.593
Darren
Token Darren
Feature activation+0.000
Mill
Token Mill
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
AM
Token AM
Feature activation+0.000
."
Token."
Feature activation+0.458
Ċ
TokenĊ
Feature activation+4.169
Ċ
TokenĊ
Feature activation+0.872
First
TokenFirst
Feature activation+0.000
Minister
Token Minister
Feature activation+0.000
Car
Token Car
Feature activation+0.000
wyn
Tokenwyn
Feature activation+0.000

INTERVAL 3.083 - 3.854
CONTAINS 0.003%

at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
same
Token same
Feature activation+0.000
university
Token university
Feature activation+0.000
.
Token.
Feature activation+1.763
Ċ
TokenĊ
Feature activation+3.659
Ċ
TokenĊ
Feature activation+0.370
In
TokenIn
Feature activation+0.000
the
Token the
Feature activation+0.000
televised
Token televised
Feature activation+0.000
conference
Token conference
Feature activation+0.826
tech
Token tech
Feature activation+0.257
company
Token company
Feature activation+0.480
B
Token B
Feature activation+0.000
aid
Tokenaid
Feature activation+0.000
u
Tokenu
Feature activation+0.831
.
Token.
Feature activation+3.569
Ċ
TokenĊ
Feature activation+1.096
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
search
Token search
Feature activation+0.000
giant
Token giant
Feature activation+0.422
cha
Tokencha
Feature activation+0.000
otic
Tokenotic
Feature activation+0.000
",
Token",
Feature activation+3.136
a
Token a
Feature activation+1.183
report
Token report
Feature activation+2.735
by
Token by
Feature activation+3.256
the
Token the
Feature activation+1.209
Criminal
Token Criminal
Feature activation+0.000
Justice
Token Justice
Feature activation+0.000
Alliance
Token Alliance
Feature activation+1.757
has
Token has
Feature activation+2.489
in
Token in
Feature activation+0.000
its
Token its
Feature activation+0.000
conflict
Token conflict
Feature activation+0.000
with
Token with
Feature activation+0.000
Russia
Token Russia
Feature activation+0.563
,
Token,
Feature activation+3.684
rejecting
Token rejecting
Feature activation+0.889
Moscow
Token Moscow
Feature activation+0.000
's
Token's
Feature activation+0.000
accusations
Token accusations
Feature activation+0.386
that
Token that
Feature activation+0.000
coalition
Token coalition
Feature activation+0.145
of
Token of
Feature activation+2.412
90
Token 90
Feature activation+0.593
justice
Token justice
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
groups
Token groups
Feature activation+3.656
said
Token said
Feature activation+1.324
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.975
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000

INTERVAL 2.312 - 3.083
CONTAINS 0.005%

am
Token am
Feature activation+0.000
immensely
Token immensely
Feature activation+0.000
proud
Token proud
Feature activation+0.000
of
Token of
Feature activation+0.000
her
Token her
Feature activation+0.000
."
Token."
Feature activation+2.633
Ċ
TokenĊ
Feature activation+5.707
Ċ
TokenĊ
Feature activation+0.762
Mr
TokenMr
Feature activation+0.363
Ray
Token Ray
Feature activation+0.000
ner
Tokenner
Feature activation+0.670
died
Token died
Feature activation+0.000
in
Token in
Feature activation+0.000
hospital
Token hospital
Feature activation+0.000
.
Token.
Feature activation+2.433
Ċ
TokenĊ
Feature activation+5.769
Ċ
TokenĊ
Feature activation+2.634
The
TokenThe
Feature activation+0.159
fire
Token fire
Feature activation+0.000
was
Token was
Feature activation+0.000
sparked
Token sparked
Feature activation+0.000
by
Token by
Feature activation+0.000
a
Tokena
Feature activation+0.000
Image
Token Image
Feature activation+0.104
caption
Token caption
Feature activation+0.000
Tim
Token Tim
Feature activation+0.000
Pe
Token Pe
Feature activation+0.000
ake
Tokenake
Feature activation+2.809
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
tweeting
Token tweeting
Feature activation+0.660
and
Token and
Feature activation+0.000
giving
Token giving
Feature activation+0.246
by
Token by
Feature activation+3.256
the
Token the
Feature activation+1.209
Criminal
Token Criminal
Feature activation+0.000
Justice
Token Justice
Feature activation+0.000
Alliance
Token Alliance
Feature activation+1.757
has
Token has
Feature activation+2.489
said
Token said
Feature activation+1.883
.
Token.
Feature activation+1.306
Ċ
TokenĊ
Feature activation+1.131
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
war
Token war
Feature activation+0.000
lords
Tokenlords
Feature activation+0.000
,
Token,
Feature activation+4.039
a
Token a
Feature activation+2.180
report
Token report
Feature activation+4.434
by
Token by
Feature activation+2.452
US
Token US
Feature activation+0.000
-
Token-
Feature activation+0.000
based
Tokenbased
Feature activation+0.000
Enough
Token Enough
Feature activation+0.000
Project
Token Project
Feature activation+0.261

INTERVAL 1.542 - 2.312
CONTAINS 0.008%

common
Token common
Feature activation+0.000
concern
Token concern
Feature activation+0.000
in
Token in
Feature activation+0.371
the
Token the
Feature activation+0.331
country
Token country
Feature activation+0.457
.
Token.
Feature activation+2.172
Ċ
TokenĊ
Feature activation+0.690
Ċ
TokenĊ
Feature activation+0.000
At
TokenAt
Feature activation+0.176
its
Token its
Feature activation+0.611
annual
Token annual
Feature activation+0.298
were
Token were
Feature activation+0.000
conve
Token conve
Feature activation+0.000
ying
Tokenying
Feature activation+0.000
"
Token "
Feature activation+0.000
information
Tokeninformation
Feature activation+0.000
".
Token".
Feature activation+1.829
Ċ
TokenĊ
Feature activation+2.385
Ċ
TokenĊ
Feature activation+0.263
The
TokenThe
Feature activation+0.665
US
Token US
Feature activation+0.181
National
Token National
Feature activation+0.000
B
Token B
Feature activation+1.457
aid
Tokenaid
Feature activation+0.012
u
Tokenu
Feature activation+4.774
said
Token said
Feature activation+1.741
Ċ
TokenĊ
Feature activation+5.234
Ċ
TokenĊ
Feature activation+2.202
Elect
TokenElect
Feature activation+0.000
ronic
Tokenronic
Feature activation+0.000
chop
Token chop
Feature activation+0.000
sticks
Tokensticks
Feature activation+0.745
that
Token that
Feature activation+1.404
performance
Token performance
Feature activation+0.000
to
Token to
Feature activation+0.000
improve
Token improve
Feature activation+0.000
results
Token results
Feature activation+0.000
.
Token.
Feature activation+0.128
Ċ
TokenĊ
Feature activation+1.989
Ċ
TokenĊ
Feature activation+0.238
VW
TokenVW
Feature activation+0.315
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
cars
Token cars
Feature activation+0.000
an
Tokenan
Feature activation+0.079
in
Token in
Feature activation+0.865
2005
Token 2005
Feature activation+3.192
.
Token.
Feature activation+0.835
Ċ
TokenĊ
Feature activation+6.200
Ċ
TokenĊ
Feature activation+1.756
M
TokenM
Feature activation+0.000
ang
Tokenang
Feature activation+0.000
esh
Tokenesh
Feature activation+0.000
kar
Tokenkar
Feature activation+0.000
's
Token's
Feature activation+0.000

INTERVAL 0.771 - 1.542
CONTAINS 0.018%

Bill
Token Bill
Feature activation+0.000
Cosby
Token Cosby
Feature activation+0.000
still
Token still
Feature activation+0.000
stands
Token stands
Feature activation+0.000
near
Token near
Feature activation+0.115
the
Token the
Feature activation+1.158
Richard
Token Richard
Feature activation+0.000
Allen
Token Allen
Feature activation+0.000
homes
Token homes
Feature activation+0.878
in
Token in
Feature activation+0.538
North
Token North
Feature activation+0.094
"
Token "
Feature activation+0.000
g
Tokeng
Feature activation+0.000
raphic
Tokenraphic
Feature activation+0.000
"
Token"
Feature activation+0.000
images
Token images
Feature activation+0.324
show
Token show
Feature activation+1.527
some
Token some
Feature activation+0.333
of
Token of
Feature activation+1.276
the
Token the
Feature activation+0.418
thousands
Token thousands
Feature activation+0.229
of
Token of
Feature activation+0.450
right
Token right
Feature activation+0.000
to
Token to
Feature activation+0.000
hold
Token hold
Feature activation+0.541
a
Token a
Feature activation+0.000
referendum
Token referendum
Feature activation+0.194
on
Token on
Feature activation+1.057
independence
Token independence
Feature activation+0.306
.
Token.
Feature activation+3.453
Ċ
TokenĊ
Feature activation+2.192
Ċ
TokenĊ
Feature activation+0.000
Particip
TokenParticip
Feature activation+0.000
,
Token,
Feature activation+0.000
victims
Token victims
Feature activation+0.000
and
Token and
Feature activation+0.000
defendants
Token defendants
Feature activation+0.000
.
Token.
Feature activation+0.382
Ċ
TokenĊ
Feature activation+1.068
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Ministry
Token Ministry
Feature activation+0.000
of
Token of
Feature activation+0.000
Justice
Token Justice
Feature activation+0.000
strikes
Token strikes
Feature activation+0.000
began
Token began
Feature activation+0.000
in
Token in
Feature activation+0.000
Syria
Token Syria
Feature activation+0.136
in
Token in
Feature activation+0.076
September
Token September
Feature activation+1.145
.
Token.
Feature activation+0.640
Ċ
TokenĊ
Feature activation+2.385
Ċ
TokenĊ
Feature activation+0.000
But
TokenBut
Feature activation+0.000
third
Token third
Feature activation+0.132

INTERVAL 0.000 - 0.771
CONTAINS 99.961%

lus
Tokenlus
Feature activation+0.000
coni
Tokenconi
Feature activation+0.000
had
Token had
Feature activation+0.000
secured
Token secured
Feature activation+0.000
a
Token a
Feature activation+0.000
more
Token more
Feature activation+0.000
comfortable
Token comfortable
Feature activation+0.000
victory
Token victory
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
confidence
Token confidence
Feature activation+0.000
Police
Token Police
Feature activation+0.000
Public
Token Public
Feature activation+0.000
Relations
Token Relations
Feature activation+0.000
Officer
Token Officer
Feature activation+0.000
,
Token,
Feature activation+0.000
SP
Token SP
Feature activation+0.000
Dol
Token Dol
Feature activation+0.000
apo
Tokenapo
Feature activation+0.000
Bad
Token Bad
Feature activation+0.000
mos
Tokenmos
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
problem
Token problem
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
team
Token team
Feature activation+0.000
trying
Token trying
Feature activation+0.000
to
Token to
Feature activation+0.000
win
Token win
Feature activation+0.000
the
Token the
Feature activation+0.000
the
Token the
Feature activation+0.000
race
Token race
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
m
Tokenm
Feature activation+0.000
already
Token already
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
7
Token 7
Feature activation+0.000
.
Token.
Feature activation+0.000
56
Token56
Feature activation+0.000
to
Token to
Feature activation+0.000
7
Token 7
Feature activation+0.000
.
Token.
Feature activation+0.000
together
Token together
Feature activation+0.000
by
Token by
Feature activation+0.000
syn
Token syn
Feature activation+0.000
apses
Tokenapses
Feature activation+0.000
.
Token.
Feature activation+0.000
If
Token If
Feature activation+0.000
sufficient
Token sufficient
Feature activation+0.000
synaptic
Token synaptic
Feature activation+0.000
inputs
Token inputs
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
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 H8.2: (feature 14811)

TOP ACTIVATIONS
MAX = 7.472

Ċ
TokenĊ
Feature activation+0.000
House
TokenHouse
Feature activation+0.000
Minority
Token Minority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+7.472
(
Token (
Feature activation+2.011
D
TokenD
Feature activation+1.134
-
Token-
Feature activation+5.724
Calif
TokenCalif
Feature activation+1.820
.),
Token.),
Feature activation+0.000
coverage
Token coverage
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
administration
Token administration
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.167
Ċ
TokenĊ
Feature activation+1.041
Earlier
TokenEarlier
Feature activation+0.101
this
Token this
Feature activation+0.000
month
Token month
Feature activation+0.000
,
Token,
Feature activation+0.000
agents
Token agents
Feature activation+0.000
uncover
Token uncover
Feature activation+0.000
the
Token the
Feature activation+0.000
scheme
Token scheme
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.702
Ċ
TokenĊ
Feature activation+0.000
Trump
TokenTrump
Feature activation+0.824
frequently
Token frequently
Feature activation+0.000
targets
Token targets
Feature activation+0.000
the
Token the
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ham
Token Ham
Feature activation+0.000
mill
Tokenmill
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.364
Ċ
TokenĊ
Feature activation+0.201
P
TokenP
Feature activation+0.000
el
Tokenel
Feature activation+0.000
osi
Tokenosi
Feature activation+0.000
made
Token made
Feature activation+0.000
States
Token States
Feature activation+0.000
'
Token'
Feature activation+0.000
uranium
Token uranium
Feature activation+0.000
supply
Token supply
Feature activation+0.000
.
Token.
Feature activation+0.918
Ċ
TokenĊ
Feature activation+6.019
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Senate
Token Senate
Feature activation+0.000
Judiciary
Token Judiciary
Feature activation+0.000
Committee
Token Committee
Feature activation+0.000
I
Token I
Feature activation+0.000
know
Token know
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.933
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
an
Token an
Feature activation+0.000
interview
Token interview
Feature activation+0.000
in
Token in
Feature activation+0.000
online
Token online
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.919
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Ign
TokenIgn
Feature activation+0.000
oring
Tokenoring
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+7.472
(
Token (
Feature activation+2.011
D
TokenD
Feature activation+1.134
-
Token-
Feature activation+5.724
Calif
TokenCalif
Feature activation+1.820
.),
Token.),
Feature activation+0.000
character
Token character
Feature activation+0.000
izing
Tokenizing
Feature activation+0.236
the
Token the
Feature activation+0.000
urring
Tokenurring
Feature activation+0.000
so
Token so
Feature activation+0.000
much
Token much
Feature activation+0.000
devastation
Token devastation
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.408
Ċ
TokenĊ
Feature activation+0.326
On
TokenOn
Feature activation+0.000
the
Token the
Feature activation+0.000
show
Token show
Feature activation+0.000
,
Token,
Feature activation+0.000
plane
Token plane
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.358
Ċ
TokenĊ
Feature activation+0.877
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
[
Token[
Feature activation+0.000
New
TokenNew
Feature activation+0.000
statement
Token statement
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.292
Ċ
TokenĊ
Feature activation+0.674
"
Token"
Feature activation+0.000
Leader
TokenLeader
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+0.000
previously
Token previously
Feature activation+0.000
source
Token source
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.873
Ċ
TokenĊ
Feature activation+0.331
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
2015
Token 2015
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.728
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
Times
Token Times
Feature activation+1.766
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.504
Ċ
TokenĊ
Feature activation+0.000
House
TokenHouse
Feature activation+0.000
Minority
Token Minority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
captured
Token captured
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.501
Ċ
TokenĊ
Feature activation+0.430
Jamie
TokenJamie
Feature activation+0.000
tells
Token tells
Feature activation+0.000
M
Token M
Feature activation+0.000
urt
Tokenurt
Feature activation+0.000
,
Token,
Feature activation+0.000
2017
Token 2017
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.339
Ċ
TokenĊ
Feature activation+0.144
The
TokenThe
Feature activation+0.000
tweet
Token tweet
Feature activation+0.000
comes
Token comes
Feature activation+0.000
after
Token after
Feature activation+0.000
and
Token and
Feature activation+0.000
money
Token money
Feature activation+0.000
laundering
Token laundering
Feature activation+0.000
before
Token before
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+3.831
administration
Token administration
Feature activation+0.693
approved
Token approved
Feature activation+0.000
a
Token a
Feature activation+0.000
controversial
Token controversial
Feature activation+0.000
deal
Token deal
Feature activation+0.000
style
Token style
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.774
Ċ
TokenĊ
Feature activation+0.000
Although
TokenAlthough
Feature activation+0.000
Alfred
Token Alfred
Feature activation+0.000
Kin
Token Kin
Feature activation+0.000
sey
Tokensey
Feature activation+0.103
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.662
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
a
Token a
Feature activation+0.000
tweet
Token tweet
Feature activation+0.000
that
Token that
Feature activation+0.000
after
Token after
Feature activation+0.000
a
Token a
Feature activation+0.000
report
Token report
Feature activation+0.343
published
Token published
Feature activation+0.181
by
Token by
Feature activation+0.219
The
Token The
Feature activation+3.033
Hill
Token Hill
Feature activation+2.161
that
Token that
Feature activation+0.254
the
Token the
Feature activation+0.271
FBI
Token FBI
Feature activation+0.664
had
Token had
Feature activation+0.000

Top DFA by src position
MAX = 8.670

.
Token.
Feature activation+0.041
Top resid features:
âĢ
TokenâĢ
Feature activation+0.006
Top resid features:
Ŀ
TokenĿ
Feature activation-0.014
Top resid features:
Ċ
TokenĊ
Feature activation+0.024
Top resid features:
Ċ
TokenĊ
Feature activation+0.042
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+8.670
Top resid features:
Ċ
TokenĊ
Feature activation+0.069
Top resid features:
Ċ
TokenĊ
Feature activation+0.072
Top resid features:
House
TokenHouse
Feature activation-0.024
Top resid features:
Minority
Token Minority
Feature activation-0.035
Top resid features:
Leader
Token Leader
Feature activation+0.021
Top resid features:
19
Token 19
Feature activation-0.001
Top resid features:
,
Token,
Feature activation-0.000
Top resid features:
2017
Token 2017
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.065
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+8.187
Top resid features:
Ċ
TokenĊ
Feature activation+0.059
Top resid features:
Ċ
TokenĊ
Feature activation+0.087
Top resid features:
The
TokenThe
Feature activation+0.019
Top resid features:
tweet
Token tweet
Feature activation-0.001
Top resid features:
comes
Token comes
Feature activation-0.002
Top resid features:
19
Token 19
Feature activation-0.001
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
2017
Token 2017
Feature activation-0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.012
Top resid features:
Ċ
TokenĊ
Feature activation+0.061
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+8.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.054
Top resid features:
Ċ
TokenĊ
Feature activation+0.088
Top resid features:
The
TokenThe
Feature activation+0.027
Top resid features:
tweet
Token tweet
Feature activation-0.002
Top resid features:
comes
Token comes
Feature activation-0.003
Top resid features:
a
Token a
Feature activation+0.001
Top resid features:
statement
Token statement
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+7.780
Top resid features:
Ċ
TokenĊ
Feature activation+0.078
Top resid features:
Ċ
TokenĊ
Feature activation+0.089
Top resid features:
"
Token"
Feature activation+0.011
Top resid features:
Leader
TokenLeader
Feature activation+0.001
Top resid features:
Pelosi
Token Pelosi
Feature activation-0.001
Top resid features:
19
Token 19
Feature activation-0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
2017
Token 2017
Feature activation-0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.022
Top resid features:
Ċ
TokenĊ
Feature activation+0.079
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+7.301
Top resid features:
Ċ
TokenĊ
Feature activation+0.093
Top resid features:
Ċ
TokenĊ
Feature activation+0.123
Top resid features:
The
TokenThe
Feature activation+0.040
Top resid features:
tweet
Token tweet
Feature activation+0.003
Top resid features:
comes
Token comes
Feature activation-0.005
Top resid features:
a
Token a
Feature activation+0.001
Top resid features:
statement
Token statement
Feature activation-0.001
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation-0.029
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+7.424
Top resid features:
Ċ
TokenĊ
Feature activation+0.069
Top resid features:
Ċ
TokenĊ
Feature activation+0.069
Top resid features:
"
Token"
Feature activation+0.006
Top resid features:
Leader
TokenLeader
Feature activation+0.000
Top resid features:
Pelosi
Token Pelosi
Feature activation-0.002
Top resid features:
âĢĶ
Token âĢĶ
Feature activation-0.000
Top resid features:
online
Token online
Feature activation-0.004
Top resid features:
.
Token.
Feature activation+0.013
Top resid features:
Ċ
TokenĊ
Feature activation+0.034
Top resid features:
Ċ
TokenĊ
Feature activation+0.160
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+6.906
Top resid features:
Ċ
TokenĊ
Feature activation+0.212
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+0.000
Top resid features:
Ign
TokenIgn
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
âĢ
TokenâĢ
Feature activation+0.004
Top resid features:
Ŀ
TokenĿ
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.013
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+7.306
Top resid features:
Ċ
TokenĊ
Feature activation+0.096
Top resid features:
Ċ
TokenĊ
Feature activation+0.028
Top resid features:
House
TokenHouse
Feature activation-0.015
Top resid features:
Minority
Token Minority
Feature activation-0.009
Top resid features:
Leader
Token Leader
Feature activation-0.039
Top resid features:
men
Token men
Feature activation-0.002
Top resid features:
captured
Token captured
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.006
Top resid features:
Ċ
TokenĊ
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.067
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+6.780
Top resid features:
Ċ
TokenĊ
Feature activation+0.131
Top resid features:
Ċ
TokenĊ
Feature activation+0.145
Top resid features:
Jamie
TokenJamie
Feature activation-0.001
Top resid features:
tells
Token tells
Feature activation-0.019
Top resid features:
M
Token M
Feature activation+0.000
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
plane
Token plane
Feature activation-0.001
Top resid features:
."
Token."
Feature activation-0.015
Top resid features:
Ċ
TokenĊ
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.011
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+6.690
Top resid features:
Ċ
TokenĊ
Feature activation+0.230
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+0.000
Top resid features:
[
Token[
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.007
Top resid features:
statement
Token statement
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.033
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+6.782
Top resid features:
Ċ
TokenĊ
Feature activation+0.169
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
"
Token"
Feature activation+0.000
Top resid features:
Leader
TokenLeader
Feature activation+0.000
Top resid features:
Pelosi
Token Pelosi
Feature activation+0.000
Top resid features:
House
Token House
Feature activation+0.004
Top resid features:
source
Token source
Feature activation-0.014
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
Ċ
TokenĊ
Feature activation+0.042
Top resid features:
Ċ
TokenĊ
Feature activation+0.149
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+6.104
Top resid features:
Ċ
TokenĊ
Feature activation+0.202
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+0.000
Top resid features:
We
TokenWe
Feature activation+0.000
Top resid features:
and
Token and
Feature activation-0.005
Top resid features:
2015
Token 2015
Feature activation-0.002
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.063
Top resid features:
Ċ
TokenĊ
Feature activation+0.112
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+5.739
Top resid features:
Ċ
TokenĊ
Feature activation+0.324
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
New
Token New
Feature activation+0.000
Top resid features:
York
Token York
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
âĢ
TokenâĢ
Feature activation+0.015
Top resid features:
Ŀ
TokenĿ
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.022
Top resid features:
Ċ
TokenĊ
Feature activation+0.101
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+5.407
Top resid features:
Ċ
TokenĊ
Feature activation+0.221
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
House
TokenHouse
Feature activation+0.000
Top resid features:
Minority
Token Minority
Feature activation+0.000
Top resid features:
Leader
Token Leader
Feature activation+0.000
Top resid features:
men
Token men
Feature activation-0.005
Top resid features:
captured
Token captured
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.017
Top resid features:
Ċ
TokenĊ
Feature activation+0.035
Top resid features:
Ċ
TokenĊ
Feature activation+0.156
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+5.649
Top resid features:
Ċ
TokenĊ
Feature activation+0.327
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Jamie
TokenJamie
Feature activation+0.000
Top resid features:
tells
Token tells
Feature activation+0.000
Top resid features:
M
Token M
Feature activation+0.000
Top resid features:
19
Token 19
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.009
Top resid features:
2017
Token 2017
Feature activation-0.016
Top resid features:
Ċ
TokenĊ
Feature activation+0.030
Top resid features:
Ċ
TokenĊ
Feature activation+0.154
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+5.434
Top resid features:
Ċ
TokenĊ
Feature activation+0.250
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
The
TokenThe
Feature activation+0.000
Top resid features:
tweet
Token tweet
Feature activation+0.000
Top resid features:
comes
Token comes
Feature activation+0.000
Top resid features:
19
Token 19
Feature activation-0.000
Top resid features:
,
Token,
Feature activation+0.003
Top resid features:
2017
Token 2017
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.025
Top resid features:
Ċ
TokenĊ
Feature activation+0.043
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+4.617
Top resid features:
Ċ
TokenĊ
Feature activation+0.380
Top resid features:
Ċ
TokenĊ
Feature activation+0.085
Top resid features:
The
TokenThe
Feature activation+0.022
Top resid features:
tweet
Token tweet
Feature activation+0.001
Top resid features:
comes
Token comes
Feature activation+0.004
Top resid features:
of
Token of
Feature activation-0.007
Top resid features:
style
Token style
Feature activation-0.003
Top resid features:
.
Token.
Feature activation+0.049
Top resid features:
Ċ
TokenĊ
Feature activation+0.047
Top resid features:
Ċ
TokenĊ
Feature activation+0.211
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+5.046
Top resid features:
Ċ
TokenĊ
Feature activation+0.187
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Although
TokenAlthough
Feature activation+0.000
Top resid features:
Alfred
Token Alfred
Feature activation+0.000
Top resid features:
Kin
Token Kin
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.016
Top resid features:
âĢ
TokenâĢ
Feature activation+0.011
Top resid features:
Ŀ
TokenĿ
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.056
Top resid features:
Ċ
TokenĊ
Feature activation+0.122
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+4.632
Top resid features:
Ċ
TokenĊ
Feature activation+0.230
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
In
TokenIn
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
tweet
Token tweet
Feature activation+0.000
Top resid features:
19
Token 19
Feature activation+0.005
Top resid features:
,
Token,
Feature activation+0.016
Top resid features:
2017
Token 2017
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.036
Top resid features:
Ċ
TokenĊ
Feature activation+0.105
Top resid features:
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+3.671
Top resid features:
Ċ
TokenĊ
Feature activation+0.401
Top resid features:
Ċ
TokenĊ
Feature activation+0.140
Top resid features:
The
TokenThe
Feature activation+0.065
Top resid features:
tweet
Token tweet
Feature activation-0.007
Top resid features:
comes
Token comes
Feature activation+0.001
Top resid features:

Decoder Weights Distribution

Head 0: 0.10

Head 1: 0.01

Head 2: 0.54

Head 3: 0.03

Head 4: 0.02

Head 5: 0.04

Head 6: 0.02

Head 7: 0.04

Head 8: 0.04

Head 9: 0.03

Head 10: 0.06

Head 11: 0.07

Positive logits

POLITICO3.16

lawmaker3.01

lawmakers2.83

Tuls2.83

Biden2.81

Ivanka2.81

Seth2.78

Georgetown2.73

Politico2.71

Recep2.70

ADVERTISEMENT2.70

2.69

Bannon2.69

aide2.68

Kel2.63

Barack2.61

Sonia2.60

­2.59

Sen2.56

counterterrorism2.54

Negative logits

``-3.61

////////////////////////////////-3.51

++++++++++++++++-3.29

Reviewer-3.09

////////////////-3.07

recognise-3.06

colour-2.92

:-)-2.82

...]-2.80

\">-2.78

isations-2.78

BBC-2.76

"...-2.72

...?-2.72

=================-2.71

recogn-2.71

avour-2.61

"}-2.61

mys-2.61

////-2.59

INTERVAL 6.725 - 7.472
CONTAINS 0.000%

Ċ
TokenĊ
Feature activation+0.000
House
TokenHouse
Feature activation+0.000
Minority
Token Minority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+7.472
(
Token (
Feature activation+2.011
D
TokenD
Feature activation+1.134
-
Token-
Feature activation+5.724
Calif
TokenCalif
Feature activation+1.820
.),
Token.),
Feature activation+0.000
coverage
Token coverage
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
administration
Token administration
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.167
Ċ
TokenĊ
Feature activation+1.041
Earlier
TokenEarlier
Feature activation+0.101
this
Token this
Feature activation+0.000
month
Token month
Feature activation+0.000
,
Token,
Feature activation+0.000

INTERVAL 5.978 - 6.725
CONTAINS 0.000%

Ŀ
TokenĿ
Feature activation+0.000
Ham
Token Ham
Feature activation+0.000
mill
Tokenmill
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.364
Ċ
TokenĊ
Feature activation+0.201
P
TokenP
Feature activation+0.000
el
Tokenel
Feature activation+0.000
osi
Tokenosi
Feature activation+0.000
made
Token made
Feature activation+0.000
States
Token States
Feature activation+0.000
'
Token'
Feature activation+0.000
uranium
Token uranium
Feature activation+0.000
supply
Token supply
Feature activation+0.000
.
Token.
Feature activation+0.918
Ċ
TokenĊ
Feature activation+6.019
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Senate
Token Senate
Feature activation+0.000
Judiciary
Token Judiciary
Feature activation+0.000
Committee
Token Committee
Feature activation+0.000
agents
Token agents
Feature activation+0.000
uncover
Token uncover
Feature activation+0.000
the
Token the
Feature activation+0.000
scheme
Token scheme
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.702
Ċ
TokenĊ
Feature activation+0.000
Trump
TokenTrump
Feature activation+0.824
frequently
Token frequently
Feature activation+0.000
targets
Token targets
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 5.230 - 5.978
CONTAINS 0.001%

Leader
Token Leader
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+7.472
(
Token (
Feature activation+2.011
D
TokenD
Feature activation+1.134
-
Token-
Feature activation+5.724
Calif
TokenCalif
Feature activation+1.820
.),
Token.),
Feature activation+0.000
character
Token character
Feature activation+0.000
izing
Tokenizing
Feature activation+0.236
the
Token the
Feature activation+0.000
statement
Token statement
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.292
Ċ
TokenĊ
Feature activation+0.674
"
Token"
Feature activation+0.000
Leader
TokenLeader
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+0.000
previously
Token previously
Feature activation+0.000
I
Token I
Feature activation+0.000
know
Token know
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.933
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
an
Token an
Feature activation+0.000
interview
Token interview
Feature activation+0.000
in
Token in
Feature activation+0.000
plane
Token plane
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.358
Ċ
TokenĊ
Feature activation+0.877
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
[
Token[
Feature activation+0.000
New
TokenNew
Feature activation+0.000
online
Token online
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+5.919
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Ign
TokenIgn
Feature activation+0.000
oring
Tokenoring
Feature activation+0.000

INTERVAL 4.483 - 5.230
CONTAINS 0.000%

captured
Token captured
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.501
Ċ
TokenĊ
Feature activation+0.430
Jamie
TokenJamie
Feature activation+0.000
tells
Token tells
Feature activation+0.000
M
Token M
Feature activation+0.000
urt
Tokenurt
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.504
Ċ
TokenĊ
Feature activation+0.000
House
TokenHouse
Feature activation+0.000
Minority
Token Minority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
source
Token source
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.873
Ċ
TokenĊ
Feature activation+0.331
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
2015
Token 2015
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.728
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
Times
Token Times
Feature activation+1.766

INTERVAL 3.736 - 4.483
CONTAINS 0.000%

style
Token style
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.774
Ċ
TokenĊ
Feature activation+0.000
Although
TokenAlthough
Feature activation+0.000
Alfred
Token Alfred
Feature activation+0.000
Kin
Token Kin
Feature activation+0.000
sey
Tokensey
Feature activation+0.103
,
Token,
Feature activation+0.000
2017
Token 2017
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+4.339
Ċ
TokenĊ
Feature activation+0.144
The
TokenThe
Feature activation+0.000
tweet
Token tweet
Feature activation+0.000
comes
Token comes
Feature activation+0.000
after
Token after
Feature activation+0.000
and
Token and
Feature activation+0.000
money
Token money
Feature activation+0.000
laundering
Token laundering
Feature activation+0.000
before
Token before
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+3.831
administration
Token administration
Feature activation+0.693
approved
Token approved
Feature activation+0.000
a
Token a
Feature activation+0.000
controversial
Token controversial
Feature activation+0.000
deal
Token deal
Feature activation+0.000

INTERVAL 2.989 - 3.736
CONTAINS 0.000%

âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ADVERTISEMENT
TokenADVERTISEMENT
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.662
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
a
Token a
Feature activation+0.000
tweet
Token tweet
Feature activation+0.000
that
Token that
Feature activation+0.000
after
Token after
Feature activation+0.000
a
Token a
Feature activation+0.000
report
Token report
Feature activation+0.343
published
Token published
Feature activation+0.181
by
Token by
Feature activation+0.219
The
Token The
Feature activation+3.033
Hill
Token Hill
Feature activation+2.161
that
Token that
Feature activation+0.254
the
Token the
Feature activation+0.271
FBI
Token FBI
Feature activation+0.664
had
Token had
Feature activation+0.000

INTERVAL 2.242 - 2.989
CONTAINS 0.000%

INTERVAL 1.494 - 2.242
CONTAINS 0.001%

Ċ
TokenĊ
Feature activation+4.728
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
Times
Token Times
Feature activation+1.766
reports
Token reports
Feature activation+0.000
that
Token that
Feature activation+0.000
Ph
Token Ph
Feature activation+0.000
o
Tokeno
Feature activation+0.000
is
Token is
Feature activation+0.000
House
TokenHouse
Feature activation+0.000
Minority
Token Minority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+7.472
(
Token (
Feature activation+2.011
D
TokenD
Feature activation+1.134
-
Token-
Feature activation+5.724
Calif
TokenCalif
Feature activation+1.820
.),
Token.),
Feature activation+0.000
character
Token character
Feature activation+0.000
a
Token a
Feature activation+0.000
report
Token report
Feature activation+0.343
published
Token published
Feature activation+0.181
by
Token by
Feature activation+0.219
The
Token The
Feature activation+3.033
Hill
Token Hill
Feature activation+2.161
that
Token that
Feature activation+0.254
the
Token the
Feature activation+0.271
FBI
Token FBI
Feature activation+0.664
had
Token had
Feature activation+0.000
gathered
Token gathered
Feature activation+0.000
stunned
Token stunned
Feature activation+0.000
when
Token when
Feature activation+0.000
top
Token top
Feature activation+0.000
player
Token player
Feature activation+0.000
Venus
Token Venus
Feature activation+0.000
Williams
Token Williams
Feature activation+1.980
abruptly
Token abruptly
Feature activation+0.000
dropped
Token dropped
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+7.472
(
Token (
Feature activation+2.011
D
TokenD
Feature activation+1.134
-
Token-
Feature activation+5.724
Calif
TokenCalif
Feature activation+1.820
.),
Token.),
Feature activation+0.000
character
Token character
Feature activation+0.000
izing
Tokenizing
Feature activation+0.236
the
Token the
Feature activation+0.000
devastation
Token devastation
Feature activation+0.000

INTERVAL 0.747 - 1.494
CONTAINS 0.001%

arch
Tokenarch
Feature activation+0.000
/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
).
Token).
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.090
Ċ
TokenĊ
Feature activation+0.000
Ph
TokenPh
Feature activation+0.000
oe
Tokenoe
Feature activation+0.000
be
Tokenbe
Feature activation+0.000
Anne
Token Anne
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
the
Token the
Feature activation+0.000
source
Token source
Feature activation+0.000
added
Token added
Feature activation+1.652
.
Token.
Feature activation+0.914
Ċ
TokenĊ
Feature activation+6.549
Ċ
TokenĊ
Feature activation+1.783
One
TokenOne
Feature activation+0.000
point
Token point
Feature activation+0.000
of
Token of
Feature activation+0.000
el
Tokenel
Feature activation+0.000
osi
Tokenosi
Feature activation+0.000
made
Token made
Feature activation+0.000
headlines
Token headlines
Feature activation+0.000
on
Token on
Feature activation+0.000
Tuesday
Token Tuesday
Feature activation+1.195
when
Token when
Feature activation+0.000
she
Token she
Feature activation+0.201
told
Token told
Feature activation+0.159
CNN
Token CNN
Feature activation+0.621
Gingrich
Token Gingrich
Feature activation+1.460
Minority
Token Minority
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
Nancy
Token Nancy
Feature activation+0.000
Pelosi
Token Pelosi
Feature activation+7.472
(
Token (
Feature activation+2.011
D
TokenD
Feature activation+1.134
-
Token-
Feature activation+5.724
Calif
TokenCalif
Feature activation+1.820
.),
Token.),
Feature activation+0.000
character
Token character
Feature activation+0.000
izing
Tokenizing
Feature activation+0.236
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
administration
Token administration
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+7.167
Ċ
TokenĊ
Feature activation+1.041
Earlier
TokenEarlier
Feature activation+0.101
this
Token this
Feature activation+0.000
month
Token month
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000

INTERVAL 0.000 - 0.747
CONTAINS 99.996%

of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
larger
Token larger
Feature activation+0.000
than
Token than
Feature activation+0.000
life
Token life
Feature activation+0.000
diet
Token diet
Feature activation+0.000
on
Token on
Feature activation+0.000
these
Token these
Feature activation+0.000
women
Token women
Feature activation+0.000
.
Token.
Feature activation+0.000
Will
Token Will
Feature activation+0.000
The
Token The
Feature activation+0.000
Greeks
Token Greeks
Feature activation+0.000
also
Token also
Feature activation+0.000
used
Token used
Feature activation+0.000
this
Token this
Feature activation+0.000
word
Token word
Feature activation+0.000
to
Token to
Feature activation+0.000
refer
Token refer
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
power
Token power
Feature activation+0.000
is
Token is
Feature activation+0.000
absent
Token absent
Feature activation+0.000
from
Token from
Feature activation+0.000
its
Token its
Feature activation+0.000
position
Token position
Feature activation+0.000
,
Token,
Feature activation+0.000
then
Token then
Feature activation+0.000
there
Token there
Feature activation+0.000
is
Token is
Feature activation+0.000
an
Token an
Feature activation+0.000
implied
Token implied
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Yet
Token Yet
Feature activation+0.000
a
Token a
Feature activation+0.000
wiser
Token wiser
Feature activation+0.000
California
Token California
Feature activation+0.000
public
Token public
Feature activation+0.000
has
Token has
Feature activation+0.000
consistently
Token consistently
Feature activation+0.000
voted
Token voted
Feature activation+0.000
for
Token for
Feature activation+0.000
more
Token more
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
idea
Token idea
Feature activation+0.000
that
Token that
Feature activation+0.000
such
Token such
Feature activation+0.000
care
Token care
Feature activation+0.000
should
Token should
Feature activation+0.000
include
Token include
Feature activation+0.000
coverage
Token coverage
Feature activation+0.000
for
Token for
Feature activation+0.000
birth
Token birth
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 H8.2: (feature 10329)

TOP ACTIVATIONS
MAX = 8.405

Picture
TokenPicture
Feature activation+2.008
:
Token:
Feature activation+1.350
KC
Token KC
Feature activation+0.000
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
Kim
Token Kim
Feature activation+0.000
Jong
Token Jong
Feature activation+0.000
-
Token-
Feature activation+0.000
un
Tokenun
Feature activation+0.000
.
Token.
Feature activation+1.824
Ċ
TokenĊ
Feature activation+8.239
Ċ
TokenĊ
Feature activation+5.693
An
TokenAn
Feature activation+0.000
official
Token official
Feature activation+0.056
then
Token then
Feature activation+0.000
watches
Token watches
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
probably
Token probably
Feature activation+0.000
never
Token never
Feature activation+0.000
visit
Token visit
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.137
Ċ
TokenĊ
Feature activation+3.778
C
TokenC
Feature activation+1.125
itizens
Tokenitizens
Feature activation+0.384
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
governing
Token governing
Feature activation+0.000
coalition
Token coalition
Feature activation+0.000
.
Token.
Feature activation+1.234
Ċ
TokenĊ
Feature activation+8.056
Ċ
TokenĊ
Feature activation+4.104
When
TokenWhen
Feature activation+0.000
entering
Token entering
Feature activation+0.000
the
Token the
Feature activation+0.000
polling
Token polling
Feature activation+0.000
position
Token position
Feature activation+0.492
in
Token in
Feature activation+0.000
their
Token their
Feature activation+0.000
district
Token district
Feature activation+0.000
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
watches
Token watches
Feature activation+0.000
as
Token as
Feature activation+0.000
they
Token they
Feature activation+0.000
vote
Token vote
Feature activation+0.000
.
Token.
Feature activation+1.375
Ċ
TokenĊ
Feature activation+7.827
Ċ
TokenĊ
Feature activation+4.579
C
TokenC
Feature activation+0.208
itizens
Tokenitizens
Feature activation+0.000
are
Token are
Feature activation+0.000
free
Token free
Feature activation+0.035
KC
Token KC
Feature activation+0.000
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
their
Token their
Feature activation+0.000
district
Token district
Feature activation+0.000
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Ċ
TokenĊ
Feature activation+4.844
Ċ
TokenĊ
Feature activation+1.341
Picture
TokenPicture
Feature activation+0.612
:
Token:
Feature activation+0.357
W
Token W
Feature activation+0.000
ENN
TokenENN
Feature activation+0.509
)
Token)
Feature activation+1.261
Ċ
TokenĊ
Feature activation+7.408
Ċ
TokenĊ
Feature activation+2.284
This
TokenThis
Feature activation+1.319
has
Token has
Feature activation+0.000
arguably
Token arguably
Feature activation+0.000
been
Token been
Feature activation+0.000
(
Token (
Feature activation+0.000
Picture
TokenPicture
Feature activation+2.008
:
Token:
Feature activation+1.350
KC
Token KC
Feature activation+0.000
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
:
Token:
Feature activation+1.350
KC
Token KC
Feature activation+0.000
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
British
Token British
Feature activation+0.006
politics
Token politics
Feature activation+0.475
for
Token for
Feature activation+0.136
generations
Token generations
Feature activation+0.582
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.176
Ċ
TokenĊ
Feature activation+0.277
After
TokenAfter
Feature activation+0.000
taking
Token taking
Feature activation+0.000
a
Token a
Feature activation+0.000
massive
Token massive
Feature activation+0.000
district
Token district
Feature activation+0.000
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Ċ
TokenĊ
Feature activation+4.844
Ċ
TokenĊ
Feature activation+1.341
Cand
TokenCand
Feature activation+0.000
Jong
Token Jong
Feature activation+0.000
-
Token-
Feature activation+0.000
un
Tokenun
Feature activation+0.000
.
Token.
Feature activation+1.824
Ċ
TokenĊ
Feature activation+8.239
Ċ
TokenĊ
Feature activation+5.693
An
TokenAn
Feature activation+0.000
official
Token official
Feature activation+0.056
then
Token then
Feature activation+0.000
watches
Token watches
Feature activation+0.000
as
Token as
Feature activation+0.000
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Ċ
TokenĊ
Feature activation+4.844
Ċ
TokenĊ
Feature activation+1.341
Cand
TokenCand
Feature activation+0.000
idates
Tokenidates
Feature activation+0.000
are
Token are
Feature activation+0.089
been
Token been
Feature activation+0.000
the
Token the
Feature activation+0.000
governing
Token governing
Feature activation+0.000
coalition
Token coalition
Feature activation+0.000
.
Token.
Feature activation+1.234
Ċ
TokenĊ
Feature activation+8.056
Ċ
TokenĊ
Feature activation+4.104
When
TokenWhen
Feature activation+0.000
entering
Token entering
Feature activation+0.000
the
Token the
Feature activation+0.000
polling
Token polling
Feature activation+0.000
booth
Token booth
Feature activation+0.000
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
stunning
Token stunning
Feature activation+1.613
new
Token new
Feature activation+3.767
airport
Token airport
Feature activation+1.842
which
Token which
Feature activation+1.882
ll
Tokenll
Feature activation+0.000
probably
Token probably
Feature activation+0.000
never
Token never
Feature activation+0.000
visit
Token visit
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.137
Ċ
TokenĊ
Feature activation+3.778
C
TokenC
Feature activation+1.125
itizens
Tokenitizens
Feature activation+0.384
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
presented
Token presented
Feature activation+0.000
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
stunning
Token stunning
Feature activation+1.613
new
Token new
Feature activation+3.767
airport
Token airport
Feature activation+1.842
which
Token which
Feature activation+1.882
you
Token you
Feature activation+0.725
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000

Top DFA by src position
MAX = 10.027

will
Token will
Feature activation-0.010
Top resid features:
probably
Token probably
Feature activation-0.014
Top resid features:
be
Token be
Feature activation-0.002
Top resid features:
killed
Token killed
Feature activation-0.026
Top resid features:
(
Token (
Feature activation+0.042
Top resid features:
Picture
TokenPicture
Feature activation+10.027
Top resid features:
:
Token:
Feature activation+0.008
Top resid features:
KC
Token KC
Feature activation+0.001
Top resid features:
NA
TokenNA
Feature activation-0.101
Top resid features:
)
Token)
Feature activation-0.091
Top resid features:
Ċ
TokenĊ
Feature activation+0.153
Top resid features:
will
Token will
Feature activation-0.002
Top resid features:
probably
Token probably
Feature activation-0.002
Top resid features:
be
Token be
Feature activation-0.000
Top resid features:
killed
Token killed
Feature activation-0.001
Top resid features:
(
Token (
Feature activation+0.004
Top resid features:
Picture
TokenPicture
Feature activation+9.493
Top resid features:
:
Token:
Feature activation+0.004
Top resid features:
KC
Token KC
Feature activation-0.000
Top resid features:
NA
TokenNA
Feature activation-0.014
Top resid features:
)
Token)
Feature activation-0.015
Top resid features:
Ċ
TokenĊ
Feature activation+0.044
Top resid features:
will
Token will
Feature activation-0.007
Top resid features:
probably
Token probably
Feature activation-0.012
Top resid features:
be
Token be
Feature activation-0.001
Top resid features:
killed
Token killed
Feature activation-0.003
Top resid features:
(
Token (
Feature activation+0.018
Top resid features:
Picture
TokenPicture
Feature activation+9.613
Top resid features:
:
Token:
Feature activation+0.024
Top resid features:
KC
Token KC
Feature activation+0.001
Top resid features:
NA
TokenNA
Feature activation-0.009
Top resid features:
)
Token)
Feature activation-0.018
Top resid features:
Ċ
TokenĊ
Feature activation+0.053
Top resid features:
will
Token will
Feature activation-0.002
Top resid features:
probably
Token probably
Feature activation-0.002
Top resid features:
be
Token be
Feature activation-0.000
Top resid features:
killed
Token killed
Feature activation-0.001
Top resid features:
(
Token (
Feature activation+0.003
Top resid features:
Picture
TokenPicture
Feature activation+9.523
Top resid features:
:
Token:
Feature activation-0.018
Top resid features:
KC
Token KC
Feature activation-0.007
Top resid features:
NA
TokenNA
Feature activation-0.032
Top resid features:
)
Token)
Feature activation-0.057
Top resid features:
Ċ
TokenĊ
Feature activation+0.045
Top resid features:
will
Token will
Feature activation-0.004
Top resid features:
probably
Token probably
Feature activation-0.004
Top resid features:
be
Token be
Feature activation-0.001
Top resid features:
killed
Token killed
Feature activation-0.002
Top resid features:
(
Token (
Feature activation+0.008
Top resid features:
Picture
TokenPicture
Feature activation+9.201
Top resid features:
:
Token:
Feature activation+0.010
Top resid features:
KC
Token KC
Feature activation-0.005
Top resid features:
NA
TokenNA
Feature activation-0.031
Top resid features:
)
Token)
Feature activation-0.041
Top resid features:
Ċ
TokenĊ
Feature activation+0.062
Top resid features:
will
Token will
Feature activation-0.002
Top resid features:
probably
Token probably
Feature activation-0.001
Top resid features:
be
Token be
Feature activation+0.000
Top resid features:
killed
Token killed
Feature activation-0.001
Top resid features:
(
Token (
Feature activation+0.003
Top resid features:
Picture
TokenPicture
Feature activation+9.228
Top resid features:
:
Token:
Feature activation+0.004
Top resid features:
KC
Token KC
Feature activation-0.003
Top resid features:
NA
TokenNA
Feature activation-0.025
Top resid features:
)
Token)
Feature activation-0.023
Top resid features:
Ċ
TokenĊ
Feature activation+0.039
Top resid features:
will
Token will
Feature activation-0.013
Top resid features:
probably
Token probably
Feature activation-0.006
Top resid features:
be
Token be
Feature activation-0.002
Top resid features:
killed
Token killed
Feature activation-0.018
Top resid features:
(
Token (
Feature activation+0.016
Top resid features:
Picture
TokenPicture
Feature activation+8.709
Top resid features:
:
Token:
Feature activation+0.123
Top resid features:
KC
Token KC
Feature activation+0.003
Top resid features:
NA
TokenNA
Feature activation+0.003
Top resid features:
)
Token)
Feature activation+0.020
Top resid features:
Ċ
TokenĊ
Feature activation+0.132
Top resid features:
will
Token will
Feature activation-0.020
Top resid features:
probably
Token probably
Feature activation-0.004
Top resid features:
be
Token be
Feature activation+0.005
Top resid features:
killed
Token killed
Feature activation-0.015
Top resid features:
(
Token (
Feature activation+0.009
Top resid features:
Picture
TokenPicture
Feature activation+9.029
Top resid features:
:
Token:
Feature activation+0.042
Top resid features:
KC
Token KC
Feature activation-0.044
Top resid features:
NA
TokenNA
Feature activation+0.009
Top resid features:
)
Token)
Feature activation+0.046
Top resid features:
Ċ
TokenĊ
Feature activation+0.538
Top resid features:
will
Token will
Feature activation-0.010
Top resid features:
probably
Token probably
Feature activation-0.006
Top resid features:
be
Token be
Feature activation+0.000
Top resid features:
killed
Token killed
Feature activation-0.002
Top resid features:
(
Token (
Feature activation-0.000
Top resid features:
Picture
TokenPicture
Feature activation+8.036
Top resid features:
:
Token:
Feature activation+0.123
Top resid features:
KC
Token KC
Feature activation+0.003
Top resid features:
NA
TokenNA
Feature activation+0.011
Top resid features:
)
Token)
Feature activation+0.040
Top resid features:
Ċ
TokenĊ
Feature activation+0.169
Top resid features:
May
Token May
Feature activation+0.011
Top resid features:
outside
Token outside
Feature activation-0.010
Top resid features:
Downing
Token Downing
Feature activation+0.003
Top resid features:
St
Token St
Feature activation+0.005
Top resid features:
(
Token (
Feature activation+0.037
Top resid features:
Picture
TokenPicture
Feature activation+9.220
Top resid features:
:
Token:
Feature activation+0.102
Top resid features:
W
Token W
Feature activation+0.002
Top resid features:
ENN
TokenENN
Feature activation-0.006
Top resid features:
)
Token)
Feature activation-0.160
Top resid features:
Ċ
TokenĊ
Feature activation+0.054
Top resid features:
will
Token will
Feature activation-0.013
Top resid features:
probably
Token probably
Feature activation-0.011
Top resid features:
be
Token be
Feature activation+0.003
Top resid features:
killed
Token killed
Feature activation-0.012
Top resid features:
(
Token (
Feature activation+0.001
Top resid features:
Picture
TokenPicture
Feature activation+8.328
Top resid features:
:
Token:
Feature activation+0.226
Top resid features:
KC
Token KC
Feature activation+0.043
Top resid features:
NA
TokenNA
Feature activation-0.001
Top resid features:
)
Token)
Feature activation-0.025
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
will
Token will
Feature activation-0.035
Top resid features:
probably
Token probably
Feature activation-0.046
Top resid features:
be
Token be
Feature activation+0.020
Top resid features:
killed
Token killed
Feature activation-0.057
Top resid features:
(
Token (
Feature activation+0.009
Top resid features:
Picture
TokenPicture
Feature activation+8.247
Top resid features:
:
Token:
Feature activation+0.110
Top resid features:
KC
Token KC
Feature activation-0.026
Top resid features:
NA
TokenNA
Feature activation-0.000
Top resid features:
)
Token)
Feature activation+0.155
Top resid features:
Ċ
TokenĊ
Feature activation+0.264
Top resid features:
May
Token May
Feature activation+0.008
Top resid features:
outside
Token outside
Feature activation-0.001
Top resid features:
Downing
Token Downing
Feature activation+0.003
Top resid features:
St
Token St
Feature activation-0.001
Top resid features:
(
Token (
Feature activation+0.009
Top resid features:
Picture
TokenPicture
Feature activation+7.926
Top resid features:
:
Token:
Feature activation+0.196
Top resid features:
W
Token W
Feature activation+0.001
Top resid features:
ENN
TokenENN
Feature activation-0.004
Top resid features:
)
Token)
Feature activation-0.239
Top resid features:
Ċ
TokenĊ
Feature activation+0.045
Top resid features:
will
Token will
Feature activation-0.004
Top resid features:
probably
Token probably
Feature activation-0.001
Top resid features:
be
Token be
Feature activation-0.000
Top resid features:
killed
Token killed
Feature activation-0.001
Top resid features:
(
Token (
Feature activation+0.006
Top resid features:
Picture
TokenPicture
Feature activation+6.649
Top resid features:
:
Token:
Feature activation+0.090
Top resid features:
KC
Token KC
Feature activation-0.004
Top resid features:
NA
TokenNA
Feature activation-0.018
Top resid features:
)
Token)
Feature activation+0.113
Top resid features:
Ċ
TokenĊ
Feature activation+0.040
Top resid features:
will
Token will
Feature activation-0.006
Top resid features:
probably
Token probably
Feature activation+0.005
Top resid features:
be
Token be
Feature activation+0.003
Top resid features:
killed
Token killed
Feature activation-0.003
Top resid features:
(
Token (
Feature activation-0.002
Top resid features:
Picture
TokenPicture
Feature activation+6.608
Top resid features:
:
Token:
Feature activation+0.029
Top resid features:
KC
Token KC
Feature activation-0.002
Top resid features:
NA
TokenNA
Feature activation+0.007
Top resid features:
)
Token)
Feature activation+0.064
Top resid features:
Ċ
TokenĊ
Feature activation+0.037
Top resid features:
will
Token will
Feature activation-0.002
Top resid features:
probably
Token probably
Feature activation+0.001
Top resid features:
be
Token be
Feature activation+0.001
Top resid features:
killed
Token killed
Feature activation-0.000
Top resid features:
(
Token (
Feature activation+0.003
Top resid features:
Picture
TokenPicture
Feature activation+5.528
Top resid features:
:
Token:
Feature activation+0.075
Top resid features:
KC
Token KC
Feature activation-0.001
Top resid features:
NA
TokenNA
Feature activation-0.014
Top resid features:
)
Token)
Feature activation+0.060
Top resid features:
Ċ
TokenĊ
Feature activation+0.049
Top resid features:
will
Token will
Feature activation-0.006
Top resid features:
probably
Token probably
Feature activation+0.007
Top resid features:
be
Token be
Feature activation+0.003
Top resid features:
killed
Token killed
Feature activation-0.005
Top resid features:
(
Token (
Feature activation-0.003
Top resid features:
Picture
TokenPicture
Feature activation+5.104
Top resid features:
:
Token:
Feature activation+0.024
Top resid features:
KC
Token KC
Feature activation-0.003
Top resid features:
NA
TokenNA
Feature activation+0.014
Top resid features:
)
Token)
Feature activation+0.053
Top resid features:
Ċ
TokenĊ
Feature activation+0.040
Top resid features:
will
Token will
Feature activation-0.014
Top resid features:
probably
Token probably
Feature activation+0.000
Top resid features:
be
Token be
Feature activation-0.003
Top resid features:
killed
Token killed
Feature activation-0.009
Top resid features:
(
Token (
Feature activation+0.005
Top resid features:
Picture
TokenPicture
Feature activation+4.748
Top resid features:
:
Token:
Feature activation+0.111
Top resid features:
KC
Token KC
Feature activation+0.006
Top resid features:
NA
TokenNA
Feature activation+0.014
Top resid features:
)
Token)
Feature activation+0.076
Top resid features:
Ċ
TokenĊ
Feature activation+0.182
Top resid features:
will
Token will
Feature activation-0.028
Top resid features:
probably
Token probably
Feature activation-0.020
Top resid features:
be
Token be
Feature activation+0.015
Top resid features:
killed
Token killed
Feature activation-0.009
Top resid features:
(
Token (
Feature activation+0.003
Top resid features:
Picture
TokenPicture
Feature activation+4.725
Top resid features:
:
Token:
Feature activation+0.053
Top resid features:
KC
Token KC
Feature activation+0.001
Top resid features:
NA
TokenNA
Feature activation+0.001
Top resid features:
)
Token)
Feature activation+0.108
Top resid features:
Ċ
TokenĊ
Feature activation+0.134
Top resid features:
will
Token will
Feature activation-0.009
Top resid features:
probably
Token probably
Feature activation+0.003
Top resid features:
be
Token be
Feature activation-0.001
Top resid features:
killed
Token killed
Feature activation-0.005
Top resid features:
(
Token (
Feature activation-0.001
Top resid features:
Picture
TokenPicture
Feature activation+4.738
Top resid features:
:
Token:
Feature activation+0.118
Top resid features:
KC
Token KC
Feature activation+0.004
Top resid features:
NA
TokenNA
Feature activation+0.023
Top resid features:
)
Token)
Feature activation+0.054
Top resid features:
Ċ
TokenĊ
Feature activation+0.134
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.02

Head 2: 0.54

Head 3: 0.05

Head 4: 0.02

Head 5: 0.05

Head 6: 0.02

Head 7: 0.03

Head 8: 0.04

Head 9: 0.05

Head 10: 0.07

Head 11: 0.04

Positive logits

5.40

!'5.11

­4.89

?'4.73

').4.66

Nasa4.53

4.42

£4.33

'.4.32

');4.21

…)4.02

.'4.01

'3.93

':3.91

',3.90

')3.83

!'"3.79

',"3.77

maths3.76

3.72

Negative logits

---3.95

----3.45

.---3.39

``-3.26

---3.22

owship-3.09

Newsletter-3.00

JUST-2.99

endeavor-2.92

characterization-2.81

UNCLASSIFIED-2.76

avored-2.72

-2.69

organizational-2.69

-----2.67

endeavors-2.66

..............-2.66

characterize-2.65

NetMessage-2.65

decentralized-2.62

INTERVAL 7.565 - 8.405
CONTAINS 0.001%

position
Token position
Feature activation+0.492
in
Token in
Feature activation+0.000
their
Token their
Feature activation+0.000
district
Token district
Feature activation+0.000
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Kim
Token Kim
Feature activation+0.000
Jong
Token Jong
Feature activation+0.000
-
Token-
Feature activation+0.000
un
Tokenun
Feature activation+0.000
.
Token.
Feature activation+1.824
Ċ
TokenĊ
Feature activation+8.239
Ċ
TokenĊ
Feature activation+5.693
An
TokenAn
Feature activation+0.000
official
Token official
Feature activation+0.056
then
Token then
Feature activation+0.000
watches
Token watches
Feature activation+0.000
watches
Token watches
Feature activation+0.000
as
Token as
Feature activation+0.000
they
Token they
Feature activation+0.000
vote
Token vote
Feature activation+0.000
.
Token.
Feature activation+1.375
Ċ
TokenĊ
Feature activation+7.827
Ċ
TokenĊ
Feature activation+4.579
C
TokenC
Feature activation+0.208
itizens
Tokenitizens
Feature activation+0.000
are
Token are
Feature activation+0.000
free
Token free
Feature activation+0.035
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
governing
Token governing
Feature activation+0.000
coalition
Token coalition
Feature activation+0.000
.
Token.
Feature activation+1.234
Ċ
TokenĊ
Feature activation+8.056
Ċ
TokenĊ
Feature activation+4.104
When
TokenWhen
Feature activation+0.000
entering
Token entering
Feature activation+0.000
the
Token the
Feature activation+0.000
polling
Token polling
Feature activation+0.000
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859

INTERVAL 6.724 - 7.565
CONTAINS 0.000%

Picture
TokenPicture
Feature activation+0.612
:
Token:
Feature activation+0.357
W
Token W
Feature activation+0.000
ENN
TokenENN
Feature activation+0.509
)
Token)
Feature activation+1.261
Ċ
TokenĊ
Feature activation+7.408
Ċ
TokenĊ
Feature activation+2.284
This
TokenThis
Feature activation+1.319
has
Token has
Feature activation+0.000
arguably
Token arguably
Feature activation+0.000
been
Token been
Feature activation+0.000
(
Token (
Feature activation+0.000
Picture
TokenPicture
Feature activation+2.008
:
Token:
Feature activation+1.350
KC
Token KC
Feature activation+0.000
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
:
Token:
Feature activation+1.350
KC
Token KC
Feature activation+0.000
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
their
Token their
Feature activation+0.000
district
Token district
Feature activation+0.000
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Ċ
TokenĊ
Feature activation+4.844
Ċ
TokenĊ
Feature activation+1.341

INTERVAL 5.884 - 6.724
CONTAINS 0.000%

British
Token British
Feature activation+0.006
politics
Token politics
Feature activation+0.475
for
Token for
Feature activation+0.136
generations
Token generations
Feature activation+0.582
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+6.176
Ċ
TokenĊ
Feature activation+0.277
After
TokenAfter
Feature activation+0.000
taking
Token taking
Feature activation+0.000
a
Token a
Feature activation+0.000
massive
Token massive
Feature activation+0.000

INTERVAL 5.043 - 5.884
CONTAINS 0.000%

district
Token district
Feature activation+0.000
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Ċ
TokenĊ
Feature activation+4.844
Ċ
TokenĊ
Feature activation+1.341
Cand
TokenCand
Feature activation+0.000
Jong
Token Jong
Feature activation+0.000
-
Token-
Feature activation+0.000
un
Tokenun
Feature activation+0.000
.
Token.
Feature activation+1.824
Ċ
TokenĊ
Feature activation+8.239
Ċ
TokenĊ
Feature activation+5.693
An
TokenAn
Feature activation+0.000
official
Token official
Feature activation+0.056
then
Token then
Feature activation+0.000
watches
Token watches
Feature activation+0.000
as
Token as
Feature activation+0.000

INTERVAL 4.203 - 5.043
CONTAINS 0.000%

Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Ċ
TokenĊ
Feature activation+4.844
Ċ
TokenĊ
Feature activation+1.341
Cand
TokenCand
Feature activation+0.000
idates
Tokenidates
Feature activation+0.000
are
Token are
Feature activation+0.089
been
Token been
Feature activation+0.000

INTERVAL 3.362 - 4.203
CONTAINS 0.001%

in
Token in
Feature activation+0.000
their
Token their
Feature activation+0.000
district
Token district
Feature activation+0.000
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Ċ
TokenĊ
Feature activation+4.844
ll
Tokenll
Feature activation+0.000
probably
Token probably
Feature activation+0.000
never
Token never
Feature activation+0.000
visit
Token visit
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.137
Ċ
TokenĊ
Feature activation+3.778
C
TokenC
Feature activation+1.125
itizens
Tokenitizens
Feature activation+0.384
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
presented
Token presented
Feature activation+0.000
the
Token the
Feature activation+0.000
governing
Token governing
Feature activation+0.000
coalition
Token coalition
Feature activation+0.000
.
Token.
Feature activation+1.234
Ċ
TokenĊ
Feature activation+8.056
Ċ
TokenĊ
Feature activation+4.104
When
TokenWhen
Feature activation+0.000
entering
Token entering
Feature activation+0.000
the
Token the
Feature activation+0.000
polling
Token polling
Feature activation+0.000
booth
Token booth
Feature activation+0.000
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
stunning
Token stunning
Feature activation+1.613
new
Token new
Feature activation+3.767
airport
Token airport
Feature activation+1.842
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
stunning
Token stunning
Feature activation+1.613
new
Token new
Feature activation+3.767
airport
Token airport
Feature activation+1.842
which
Token which
Feature activation+1.882

INTERVAL 2.522 - 3.362
CONTAINS 0.000%

killed
Token killed
Feature activation+0.000
(
Token (
Feature activation+0.000
Picture
TokenPicture
Feature activation+2.008
:
Token:
Feature activation+1.350
KC
Token KC
Feature activation+0.000
NA
TokenNA
Feature activation+2.677
)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
Advertisement
TokenAdvertisement
Feature activation+0.954
Ċ
TokenĊ
Feature activation+4.844
Ċ
TokenĊ
Feature activation+1.341
Cand
TokenCand
Feature activation+0.000
idates
Tokenidates
Feature activation+0.000

INTERVAL 1.681 - 2.522
CONTAINS 0.001%

)
Token)
Feature activation+7.149
Ċ
TokenĊ
Feature activation+8.405
Ċ
TokenĊ
Feature activation+7.087
MORE
TokenMORE
Feature activation+7.681
:
Token:
Feature activation+7.589
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
of
Token of
Feature activation+0.000
Kim
Token Kim
Feature activation+0.000
Jong
Token Jong
Feature activation+0.000
-
Token-
Feature activation+0.000
un
Tokenun
Feature activation+0.000
.
Token.
Feature activation+1.824
Ċ
TokenĊ
Feature activation+8.239
Ċ
TokenĊ
Feature activation+5.693
An
TokenAn
Feature activation+0.000
official
Token official
Feature activation+0.056
then
Token then
Feature activation+0.000
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
stunning
Token stunning
Feature activation+1.613
new
Token new
Feature activation+3.767
airport
Token airport
Feature activation+1.842
which
Token which
Feature activation+1.882
you
Token you
Feature activation+0.725
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
probably
Token probably
Feature activation+0.000
each
Token each
Feature activation+0.000
position
Token position
Feature activation+0.492
in
Token in
Feature activation+0.000
their
Token their
Feature activation+0.000
district
Token district
Feature activation+0.000
.
Token.
Feature activation+2.462
Ċ
TokenĊ
Feature activation+7.829
Ċ
TokenĊ
Feature activation+3.648
Advertisement
TokenAdvertisement
Feature activation+7.454
Ċ
TokenĊ
Feature activation+5.771
Ċ
TokenĊ
Feature activation+2.854
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
stunning
Token stunning
Feature activation+1.613
new
Token new
Feature activation+3.767
airport
Token airport
Feature activation+1.842
which
Token which
Feature activation+1.882
you
Token you
Feature activation+0.725
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000

INTERVAL 0.841 - 1.681
CONTAINS 0.001%

probably
Token probably
Feature activation+0.000
never
Token never
Feature activation+0.000
visit
Token visit
Feature activation+0.000
Ċ
TokenĊ
Feature activation+8.137
Ċ
TokenĊ
Feature activation+3.778
C
TokenC
Feature activation+1.125
itizens
Tokenitizens
Feature activation+0.384
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
presented
Token presented
Feature activation+0.000
with
Token with
Feature activation+0.000
then
Token then
Feature activation+0.000
watches
Token watches
Feature activation+0.000
as
Token as
Feature activation+0.000
they
Token they
Feature activation+0.000
vote
Token vote
Feature activation+0.000
.
Token.
Feature activation+1.375
Ċ
TokenĊ
Feature activation+7.827
Ċ
TokenĊ
Feature activation+4.579
C
TokenC
Feature activation+0.208
itizens
Tokenitizens
Feature activation+0.000
are
Token are
Feature activation+0.000
W
Token W
Feature activation+0.000
ENN
TokenENN
Feature activation+0.509
)
Token)
Feature activation+1.261
Ċ
TokenĊ
Feature activation+7.408
Ċ
TokenĊ
Feature activation+2.284
This
TokenThis
Feature activation+1.319
has
Token has
Feature activation+0.000
arguably
Token arguably
Feature activation+0.000
been
Token been
Feature activation+0.000
one
Token one
Feature activation+0.000
of
Token of
Feature activation+0.000
North
Token North
Feature activation+2.407
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
stunning
Token stunning
Feature activation+1.613
new
Token new
Feature activation+3.767
airport
Token airport
Feature activation+1.842
which
Token which
Feature activation+1.882
you
Token you
Feature activation+0.725
Korea
Token Korea
Feature activation+1.203
just
Token just
Feature activation+0.146
built
Token built
Feature activation+3.681
a
Token a
Feature activation+3.859
really
Token really
Feature activation+1.355
stunning
Token stunning
Feature activation+1.613
new
Token new
Feature activation+3.767
airport
Token airport
Feature activation+1.842
which
Token which
Feature activation+1.882
you
Token you
Feature activation+0.725
âĢ
TokenâĢ
Feature activation+0.000

INTERVAL 0.000 - 0.841
CONTAINS 99.995%

,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
is
Token is
Feature activation+0.000
crystal
Token crystal
Feature activation+0.000
meth
Token meth
Feature activation+0.000
or
Token or
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
ice
Tokenice
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
David
TokenDavid
Feature activation+0.000
Friedman
Token Friedman
Feature activation+0.000
LIVE
Token LIVE
Feature activation+0.000
:
Token:
Feature activation+0.000
Har
Token Har
Feature activation+0.000
ald
Tokenald
Feature activation+0.000
&
Token &
Feature activation+0.000
Salam
Token Salam
Feature activation+0.000
ander
Tokenander
Feature activation+0.000
Lo
Token Lo
Feature activation+0.000
vers
Tokenvers
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
reason
Token reason
Feature activation+0.000
that
Token that
Feature activation+0.000
SQL
Token SQL
Feature activation+0.000
injection
Token injection
Feature activation+0.000
and
Token and
Feature activation+0.000
many
Token many
Feature activation+0.000
other
Token other
Feature activation+0.000
exploits
Token exploits
Feature activation+0.000
,
Token,
Feature activation+0.000
chips
Token chips
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
same
Token same
Feature activation+0.000
multi
Token multi
Feature activation+0.000
-
Token-
Feature activation+0.000
chip
Tokenchip
Feature activation+0.000
module
Token module
Feature activation+0.000
package
Token package
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
city
Token city
Feature activation+0.000
itself
Token itself
Feature activation+0.000
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.000
even
Token even
Feature activation+0.000
the
Token the
Feature activation+0.000
country
Token country
Feature activation+0.000
.
Token.
Feature activation+0.000
Thus
Token Thus
Feature activation+0.000
the
Token the
Feature activation+0.000
region
Token region
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