H2.6 top features

Top feature 0 in H2.6: (feature 7253)

TOP ACTIVATIONS
MAX = 1.827

CA
TokenCA
Feature activation+0.751
grounded
Token grounded
Feature activation+1.176
the
Token the
Feature activation+1.416
pilot
Token pilot
Feature activation+1.434
on
Token on
Feature activation+1.063
Monday
Token Monday
Feature activation+1.827
afternoon
Token afternoon
Feature activation+1.476
.
Token.
Feature activation+0.859
By
Token By
Feature activation+0.381
then
Token then
Feature activation+0.388
,
Token,
Feature activation+0.429
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
street
Token street
Feature activation+0.000
."
Token."
Feature activation+0.000
I
TokenI
Feature activation+0.445
thought
Token thought
Feature activation+1.599
these
Token these
Feature activation+1.397
were
Token were
Feature activation+1.383
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
,"
Token,"
Feature activation+0.000
she
Token she
Feature activation+0.000
said
Token said
Feature activation+0.000
."
Token."
Feature activation+0.000
We
TokenWe
Feature activation+0.404
remain
Token remain
Feature activation+1.550
committed
Token committed
Feature activation+0.999
to
Token to
Feature activation+0.816
holding
Token holding
Feature activation+0.448
this
Token this
Feature activation+0.832
dialogue
Token dialogue
Feature activation+0.630
I
TokenI
Feature activation+0.492
don
Token don
Feature activation+1.253
't
Token't
Feature activation+1.146
know
Token know
Feature activation+1.453
what
Token what
Feature activation+1.438
the
Token the
Feature activation+1.545
answer
Token answer
Feature activation+1.194
is
Token is
Feature activation+1.067
,
Token,
Feature activation+1.452
or
Token or
Feature activation+0.843
how
Token how
Feature activation+1.014
grounded
Token grounded
Feature activation+1.176
the
Token the
Feature activation+1.416
pilot
Token pilot
Feature activation+1.434
on
Token on
Feature activation+1.063
Monday
Token Monday
Feature activation+1.827
afternoon
Token afternoon
Feature activation+1.476
.
Token.
Feature activation+0.859
By
Token By
Feature activation+0.381
then
Token then
Feature activation+0.388
,
Token,
Feature activation+0.429
he
Token he
Feature activation+0.367
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
the
Token the
Feature activation+1.070
ground
Token ground
Feature activation+0.771
,
Token,
Feature activation+1.468
but
Token but
Feature activation+1.320
apparently
Token apparently
Feature activation+0.922
they
Token they
Feature activation+1.105
're
Token're
Feature activation+0.599
not
Token not
Feature activation+0.176
department
Token department
Feature activation+0.000
."
Token."
Feature activation+0.000
I
TokenI
Feature activation+0.492
don
Token don
Feature activation+1.253
't
Token't
Feature activation+1.146
know
Token know
Feature activation+1.453
what
Token what
Feature activation+1.438
the
Token the
Feature activation+1.545
answer
Token answer
Feature activation+1.194
is
Token is
Feature activation+1.067
,
Token,
Feature activation+1.452
know
Token know
Feature activation+1.453
what
Token what
Feature activation+1.438
the
Token the
Feature activation+1.545
answer
Token answer
Feature activation+1.194
is
Token is
Feature activation+1.067
,
Token,
Feature activation+1.452
or
Token or
Feature activation+0.843
how
Token how
Feature activation+1.014
to
Token to
Feature activation+0.823
send
Token send
Feature activation+0.539
the
Token the
Feature activation+0.553
."
Token."
Feature activation+0.000
I
TokenI
Feature activation+0.492
don
Token don
Feature activation+1.253
't
Token't
Feature activation+1.146
know
Token know
Feature activation+1.453
what
Token what
Feature activation+1.438
the
Token the
Feature activation+1.545
answer
Token answer
Feature activation+1.194
is
Token is
Feature activation+1.067
,
Token,
Feature activation+1.452
or
Token or
Feature activation+0.843
The
TokenThe
Feature activation+0.465
DG
Token DG
Feature activation+1.336
CA
TokenCA
Feature activation+0.751
grounded
Token grounded
Feature activation+1.176
the
Token the
Feature activation+1.416
pilot
Token pilot
Feature activation+1.434
on
Token on
Feature activation+1.063
Monday
Token Monday
Feature activation+1.827
afternoon
Token afternoon
Feature activation+1.476
.
Token.
Feature activation+0.859
By
Token By
Feature activation+0.381
."
Token."
Feature activation+0.000
The
TokenThe
Feature activation+0.465
DG
Token DG
Feature activation+1.336
CA
TokenCA
Feature activation+0.751
grounded
Token grounded
Feature activation+1.176
the
Token the
Feature activation+1.416
pilot
Token pilot
Feature activation+1.434
on
Token on
Feature activation+1.063
Monday
Token Monday
Feature activation+1.827
afternoon
Token afternoon
Feature activation+1.476
.
Token.
Feature activation+0.859
the
Token the
Feature activation+0.000
street
Token street
Feature activation+0.000
."
Token."
Feature activation+0.000
I
TokenI
Feature activation+0.445
thought
Token thought
Feature activation+1.599
these
Token these
Feature activation+1.397
were
Token were
Feature activation+1.383
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
the
Token the
Feature activation+1.070
street
Token street
Feature activation+0.000
."
Token."
Feature activation+0.000
I
TokenI
Feature activation+0.445
thought
Token thought
Feature activation+1.599
these
Token these
Feature activation+1.397
were
Token were
Feature activation+1.383
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
the
Token the
Feature activation+1.070
ground
Token ground
Feature activation+0.771
Delhi
Token Delhi
Feature activation+0.000
on
Token on
Feature activation+0.000
Tuesday
Token Tuesday
Feature activation+0.000
."
Token."
Feature activation+0.000
The
TokenThe
Feature activation+0.465
DG
Token DG
Feature activation+1.336
CA
TokenCA
Feature activation+0.751
grounded
Token grounded
Feature activation+1.176
the
Token the
Feature activation+1.416
pilot
Token pilot
Feature activation+1.434
on
Token on
Feature activation+1.063
Thursday
Token Thursday
Feature activation+0.000
afternoon
Token afternoon
Feature activation+0.000
."
Token."
Feature activation+0.000
Our
TokenOur
Feature activation+0.453
suspect
Token suspect
Feature activation+1.317
completely
Token completely
Feature activation+1.329
confessed
Token confessed
Feature activation+0.960
to
Token to
Feature activation+1.004
what
Token what
Feature activation+0.894
he
Token he
Feature activation+1.268
did
Token did
Feature activation+1.223
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
the
Token the
Feature activation+1.070
ground
Token ground
Feature activation+0.771
,
Token,
Feature activation+1.468
but
Token but
Feature activation+1.320
apparently
Token apparently
Feature activation+0.922
they
Token they
Feature activation+1.105
're
Token're
Feature activation+0.599
not
Token not
Feature activation+0.176
.
Token.
Feature activation+0.328
late
Token late
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
afternoon
Token afternoon
Feature activation+0.000
."
Token."
Feature activation+0.000
Our
TokenOur
Feature activation+0.453
suspect
Token suspect
Feature activation+1.317
completely
Token completely
Feature activation+1.329
confessed
Token confessed
Feature activation+0.960
to
Token to
Feature activation+1.004
what
Token what
Feature activation+0.894
he
Token he
Feature activation+1.268
and
Token and
Feature activation+0.000
isolated
Token isolated
Feature activation+0.000
neighbour
Token neighbour
Feature activation+0.000
."
Token."
Feature activation+0.302
I
TokenI
Feature activation+0.488
am
Token am
Feature activation+1.305
asking
Token asking
Feature activation+1.245
foreign
Token foreign
Feature activation+1.355
and
Token and
Feature activation+1.238
security
Token security
Feature activation+1.205
ministries
Token ministries
Feature activation+1.075
suspect
Token suspect
Feature activation+1.317
completely
Token completely
Feature activation+1.329
confessed
Token confessed
Feature activation+0.960
to
Token to
Feature activation+1.004
what
Token what
Feature activation+0.894
he
Token he
Feature activation+1.268
did
Token did
Feature activation+1.223
.
Token.
Feature activation+0.975
We
Token We
Feature activation+0.634
âĢ
TokenâĢ
Feature activation+0.235
Ļ
TokenĻ
Feature activation+0.000
eight
Token eight
Feature activation+0.000
games
Token games
Feature activation+0.000
."
Token."
Feature activation+0.069
I
TokenI
Feature activation+0.089
thought
Token thought
Feature activation+0.943
he
Token he
Feature activation+1.254
was
Token was
Feature activation+1.133
buzzing
Token buzzing
Feature activation+0.806
,"
Token,"
Feature activation+0.976
Leafs
Token Leafs
Feature activation+0.480
forward
Token forward
Feature activation+0.387

Top DFA by src position
MAX = 2.020

return
Token return
Feature activation+0.006
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
Delhi
Token Delhi
Feature activation+0.021
Top resid features:
on
Token on
Feature activation+0.004
Top resid features:
Tuesday
Token Tuesday
Feature activation-0.005
Top resid features:
."
Token."
Feature activation+2.020
Top resid features:
The
TokenThe
Feature activation+0.126
Top resid features:
DG
Token DG
Feature activation+0.056
Top resid features:
CA
TokenCA
Feature activation+0.039
Top resid features:
grounded
Token grounded
Feature activation+0.077
Top resid features:
the
Token the
Feature activation+0.229
Top resid features:
swimming
Token swimming
Feature activation-0.001
Top resid features:
pool
Token pool
Feature activation+0.004
Top resid features:
across
Token across
Feature activation+0.023
Top resid features:
the
Token the
Feature activation+0.048
Top resid features:
street
Token street
Feature activation+0.008
Top resid features:
."
Token."
Feature activation+1.911
Top resid features:
I
TokenI
Feature activation+0.438
Top resid features:
thought
Token thought
Feature activation+0.282
Top resid features:
these
Token these
Feature activation+0.000
Top resid features:
were
Token were
Feature activation+0.000
Top resid features:
really
Token really
Feature activation+0.000
Top resid features:
do
Token do
Feature activation+0.009
Top resid features:
that
Token that
Feature activation+0.011
Top resid features:
,"
Token,"
Feature activation+0.081
Top resid features:
she
Token she
Feature activation+0.038
Top resid features:
said
Token said
Feature activation+0.090
Top resid features:
."
Token."
Feature activation+1.785
Top resid features:
We
TokenWe
Feature activation+0.534
Top resid features:
remain
Token remain
Feature activation+0.186
Top resid features:
committed
Token committed
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
holding
Token holding
Feature activation+0.000
Top resid features:
of
Token of
Feature activation-0.002
Top resid features:
Men
Token Men
Feature activation+0.023
Top resid features:
asha
Tokenasha
Feature activation+0.002
Top resid features:
police
Token police
Feature activation-0.004
Top resid features:
department
Token department
Feature activation+0.003
Top resid features:
."
Token."
Feature activation+1.646
Top resid features:
I
TokenI
Feature activation+0.204
Top resid features:
don
Token don
Feature activation+0.040
Top resid features:
't
Token't
Feature activation+0.048
Top resid features:
know
Token know
Feature activation+0.186
Top resid features:
what
Token what
Feature activation+0.411
Top resid features:
return
Token return
Feature activation+0.006
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
Delhi
Token Delhi
Feature activation+0.016
Top resid features:
on
Token on
Feature activation+0.001
Top resid features:
Tuesday
Token Tuesday
Feature activation-0.016
Top resid features:
."
Token."
Feature activation+1.724
Top resid features:
The
TokenThe
Feature activation+0.115
Top resid features:
DG
Token DG
Feature activation+0.045
Top resid features:
CA
TokenCA
Feature activation+0.033
Top resid features:
grounded
Token grounded
Feature activation+0.044
Top resid features:
the
Token the
Feature activation+0.185
Top resid features:
swimming
Token swimming
Feature activation-0.002
Top resid features:
pool
Token pool
Feature activation+0.000
Top resid features:
across
Token across
Feature activation+0.004
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
street
Token street
Feature activation-0.003
Top resid features:
."
Token."
Feature activation+1.840
Top resid features:
I
TokenI
Feature activation+0.153
Top resid features:
thought
Token thought
Feature activation+0.060
Top resid features:
these
Token these
Feature activation+0.048
Top resid features:
were
Token were
Feature activation+0.045
Top resid features:
really
Token really
Feature activation+0.061
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
Men
Token Men
Feature activation+0.020
Top resid features:
asha
Tokenasha
Feature activation+0.001
Top resid features:
police
Token police
Feature activation+0.003
Top resid features:
department
Token department
Feature activation+0.007
Top resid features:
."
Token."
Feature activation+1.578
Top resid features:
I
TokenI
Feature activation+0.499
Top resid features:
don
Token don
Feature activation+0.158
Top resid features:
't
Token't
Feature activation+0.153
Top resid features:
know
Token know
Feature activation+0.140
Top resid features:
what
Token what
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
Men
Token Men
Feature activation+0.011
Top resid features:
asha
Tokenasha
Feature activation-0.001
Top resid features:
police
Token police
Feature activation-0.004
Top resid features:
department
Token department
Feature activation+0.003
Top resid features:
."
Token."
Feature activation+1.924
Top resid features:
I
TokenI
Feature activation+0.126
Top resid features:
don
Token don
Feature activation+0.023
Top resid features:
't
Token't
Feature activation+0.041
Top resid features:
know
Token know
Feature activation+0.060
Top resid features:
what
Token what
Feature activation+0.147
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
Men
Token Men
Feature activation+0.019
Top resid features:
asha
Tokenasha
Feature activation+0.002
Top resid features:
police
Token police
Feature activation+0.021
Top resid features:
department
Token department
Feature activation+0.009
Top resid features:
."
Token."
Feature activation+1.566
Top resid features:
I
TokenI
Feature activation+0.231
Top resid features:
don
Token don
Feature activation+0.073
Top resid features:
't
Token't
Feature activation+0.112
Top resid features:
know
Token know
Feature activation+0.457
Top resid features:
what
Token what
Feature activation+0.149
Top resid features:
return
Token return
Feature activation+0.005
Top resid features:
to
Token to
Feature activation-0.004
Top resid features:
Delhi
Token Delhi
Feature activation+0.026
Top resid features:
on
Token on
Feature activation+0.003
Top resid features:
Tuesday
Token Tuesday
Feature activation-0.020
Top resid features:
."
Token."
Feature activation+1.621
Top resid features:
The
TokenThe
Feature activation+0.180
Top resid features:
DG
Token DG
Feature activation+0.104
Top resid features:
CA
TokenCA
Feature activation+0.059
Top resid features:
grounded
Token grounded
Feature activation+0.130
Top resid features:
the
Token the
Feature activation+0.407
Top resid features:
return
Token return
Feature activation+0.010
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
Delhi
Token Delhi
Feature activation+0.030
Top resid features:
on
Token on
Feature activation+0.015
Top resid features:
Tuesday
Token Tuesday
Feature activation-0.003
Top resid features:
."
Token."
Feature activation+1.817
Top resid features:
The
TokenThe
Feature activation+0.150
Top resid features:
DG
Token DG
Feature activation+0.109
Top resid features:
CA
TokenCA
Feature activation+0.070
Top resid features:
grounded
Token grounded
Feature activation+0.172
Top resid features:
the
Token the
Feature activation+0.204
Top resid features:
swimming
Token swimming
Feature activation-0.002
Top resid features:
pool
Token pool
Feature activation+0.001
Top resid features:
across
Token across
Feature activation+0.026
Top resid features:
the
Token the
Feature activation+0.041
Top resid features:
street
Token street
Feature activation-0.010
Top resid features:
."
Token."
Feature activation+1.614
Top resid features:
I
TokenI
Feature activation+0.350
Top resid features:
thought
Token thought
Feature activation+0.396
Top resid features:
these
Token these
Feature activation+0.157
Top resid features:
were
Token were
Feature activation+0.000
Top resid features:
really
Token really
Feature activation+0.000
Top resid features:
swimming
Token swimming
Feature activation-0.006
Top resid features:
pool
Token pool
Feature activation-0.001
Top resid features:
across
Token across
Feature activation+0.014
Top resid features:
the
Token the
Feature activation+0.016
Top resid features:
street
Token street
Feature activation-0.006
Top resid features:
."
Token."
Feature activation+1.560
Top resid features:
I
TokenI
Feature activation+0.291
Top resid features:
thought
Token thought
Feature activation+0.355
Top resid features:
these
Token these
Feature activation+0.361
Top resid features:
were
Token were
Feature activation+0.048
Top resid features:
really
Token really
Feature activation+0.000
Top resid features:
return
Token return
Feature activation+0.011
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
Delhi
Token Delhi
Feature activation+0.043
Top resid features:
on
Token on
Feature activation+0.041
Top resid features:
Tuesday
Token Tuesday
Feature activation-0.007
Top resid features:
."
Token."
Feature activation+1.534
Top resid features:
The
TokenThe
Feature activation+0.645
Top resid features:
DG
Token DG
Feature activation+0.166
Top resid features:
CA
TokenCA
Feature activation+0.000
Top resid features:
grounded
Token grounded
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
police
Token police
Feature activation-0.001
Top resid features:
station
Token station
Feature activation-0.009
Top resid features:
late
Token late
Feature activation-0.006
Top resid features:
Thursday
Token Thursday
Feature activation-0.003
Top resid features:
afternoon
Token afternoon
Feature activation-0.001
Top resid features:
."
Token."
Feature activation+1.485
Top resid features:
Our
TokenOur
Feature activation+0.449
Top resid features:
suspect
Token suspect
Feature activation+0.377
Top resid features:
completely
Token completely
Feature activation+0.132
Top resid features:
confessed
Token confessed
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
swimming
Token swimming
Feature activation-0.001
Top resid features:
pool
Token pool
Feature activation-0.000
Top resid features:
across
Token across
Feature activation+0.003
Top resid features:
the
Token the
Feature activation+0.005
Top resid features:
street
Token street
Feature activation-0.004
Top resid features:
."
Token."
Feature activation+1.514
Top resid features:
I
TokenI
Feature activation+0.216
Top resid features:
thought
Token thought
Feature activation+0.035
Top resid features:
these
Token these
Feature activation+0.033
Top resid features:
were
Token were
Feature activation+0.049
Top resid features:
really
Token really
Feature activation+0.048
Top resid features:
police
Token police
Feature activation+0.002
Top resid features:
station
Token station
Feature activation-0.003
Top resid features:
late
Token late
Feature activation+0.022
Top resid features:
Thursday
Token Thursday
Feature activation-0.006
Top resid features:
afternoon
Token afternoon
Feature activation+0.009
Top resid features:
."
Token."
Feature activation+1.676
Top resid features:
Our
TokenOur
Feature activation+0.591
Top resid features:
suspect
Token suspect
Feature activation+0.164
Top resid features:
completely
Token completely
Feature activation+0.000
Top resid features:
confessed
Token confessed
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
's
Token's
Feature activation+0.013
Top resid features:
poor
Token poor
Feature activation+0.011
Top resid features:
and
Token and
Feature activation+0.022
Top resid features:
isolated
Token isolated
Feature activation+0.007
Top resid features:
neighbour
Token neighbour
Feature activation+0.040
Top resid features:
."
Token."
Feature activation+1.811
Top resid features:
I
TokenI
Feature activation+0.342
Top resid features:
am
Token am
Feature activation+0.115
Top resid features:
asking
Token asking
Feature activation+0.000
Top resid features:
foreign
Token foreign
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
police
Token police
Feature activation-0.001
Top resid features:
station
Token station
Feature activation-0.002
Top resid features:
late
Token late
Feature activation-0.001
Top resid features:
Thursday
Token Thursday
Feature activation+0.001
Top resid features:
afternoon
Token afternoon
Feature activation-0.001
Top resid features:
."
Token."
Feature activation+1.531
Top resid features:
Our
TokenOur
Feature activation+0.233
Top resid features:
suspect
Token suspect
Feature activation+0.043
Top resid features:
completely
Token completely
Feature activation+0.036
Top resid features:
confessed
Token confessed
Feature activation+0.136
Top resid features:
to
Token to
Feature activation+0.045
Top resid features:
goal
Token goal
Feature activation-0.013
Top resid features:
drought
Token drought
Feature activation+0.015
Top resid features:
to
Token to
Feature activation+0.008
Top resid features:
eight
Token eight
Feature activation+0.002
Top resid features:
games
Token games
Feature activation+0.031
Top resid features:
."
Token."
Feature activation+1.437
Top resid features:
I
TokenI
Feature activation+0.303
Top resid features:
thought
Token thought
Feature activation+0.576
Top resid features:
he
Token he
Feature activation+0.157
Top resid features:
was
Token was
Feature activation+0.000
Top resid features:
buzzing
Token buzzing
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.08

Head 2: 0.04

Head 3: 0.05

Head 4: 0.04

Head 5: 0.06

Head 6: 0.35

Head 7: 0.06

Head 8: 0.05

Head 9: 0.06

Head 10: 0.09

Head 11: 0.04

Positive logits

%"3.40

..."2.93

)"2.84

)",2.80

"]2.60

;"2.56

}"2.55

…"2.44

"...2.39

"2.33

..."2.33

!"2.32

!",2.29

":2.25

")2.24

"],2.24

";2.23

,"2.21

",2.20

"'2.19

Negative logits

respectively-1.70

utterstock-1.60

.).-1.57

RPG-1.48

atively-1.47

astical-1.43

?).-1.40

arthed-1.36

AFP-1.33

rawled-1.30

).-1.30

-1.30

ivably-1.29

hetti-1.29

+.-1.28

gado-1.28

Cue-1.24

iably-1.23

ë-1.22

apiece-1.21

INTERVAL 1.645 - 1.827
CONTAINS 0.000%

CA
TokenCA
Feature activation+0.751
grounded
Token grounded
Feature activation+1.176
the
Token the
Feature activation+1.416
pilot
Token pilot
Feature activation+1.434
on
Token on
Feature activation+1.063
Monday
Token Monday
Feature activation+1.827
afternoon
Token afternoon
Feature activation+1.476
.
Token.
Feature activation+0.859
By
Token By
Feature activation+0.381
then
Token then
Feature activation+0.388
,
Token,
Feature activation+0.429

INTERVAL 1.462 - 1.645
CONTAINS 0.001%

I
TokenI
Feature activation+0.492
don
Token don
Feature activation+1.253
't
Token't
Feature activation+1.146
know
Token know
Feature activation+1.453
what
Token what
Feature activation+1.438
the
Token the
Feature activation+1.545
answer
Token answer
Feature activation+1.194
is
Token is
Feature activation+1.067
,
Token,
Feature activation+1.452
or
Token or
Feature activation+0.843
how
Token how
Feature activation+1.014
grounded
Token grounded
Feature activation+1.176
the
Token the
Feature activation+1.416
pilot
Token pilot
Feature activation+1.434
on
Token on
Feature activation+1.063
Monday
Token Monday
Feature activation+1.827
afternoon
Token afternoon
Feature activation+1.476
.
Token.
Feature activation+0.859
By
Token By
Feature activation+0.381
then
Token then
Feature activation+0.388
,
Token,
Feature activation+0.429
he
Token he
Feature activation+0.367
,"
Token,"
Feature activation+0.000
she
Token she
Feature activation+0.000
said
Token said
Feature activation+0.000
."
Token."
Feature activation+0.000
We
TokenWe
Feature activation+0.404
remain
Token remain
Feature activation+1.550
committed
Token committed
Feature activation+0.999
to
Token to
Feature activation+0.816
holding
Token holding
Feature activation+0.448
this
Token this
Feature activation+0.832
dialogue
Token dialogue
Feature activation+0.630
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
street
Token street
Feature activation+0.000
."
Token."
Feature activation+0.000
I
TokenI
Feature activation+0.445
thought
Token thought
Feature activation+1.599
these
Token these
Feature activation+1.397
were
Token were
Feature activation+1.383
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
the
Token the
Feature activation+1.070
ground
Token ground
Feature activation+0.771
,
Token,
Feature activation+1.468
but
Token but
Feature activation+1.320
apparently
Token apparently
Feature activation+0.922
they
Token they
Feature activation+1.105
're
Token're
Feature activation+0.599
not
Token not
Feature activation+0.176

INTERVAL 1.279 - 1.462
CONTAINS 0.002%

the
Token the
Feature activation+0.000
street
Token street
Feature activation+0.000
."
Token."
Feature activation+0.000
I
TokenI
Feature activation+0.445
thought
Token thought
Feature activation+1.599
these
Token these
Feature activation+1.397
were
Token were
Feature activation+1.383
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
the
Token the
Feature activation+1.070
Delhi
Token Delhi
Feature activation+0.000
on
Token on
Feature activation+0.000
Tuesday
Token Tuesday
Feature activation+0.000
."
Token."
Feature activation+0.000
The
TokenThe
Feature activation+0.465
DG
Token DG
Feature activation+1.336
CA
TokenCA
Feature activation+0.751
grounded
Token grounded
Feature activation+1.176
the
Token the
Feature activation+1.416
pilot
Token pilot
Feature activation+1.434
on
Token on
Feature activation+1.063
know
Token know
Feature activation+1.453
what
Token what
Feature activation+1.438
the
Token the
Feature activation+1.545
answer
Token answer
Feature activation+1.194
is
Token is
Feature activation+1.067
,
Token,
Feature activation+1.452
or
Token or
Feature activation+0.843
how
Token how
Feature activation+1.014
to
Token to
Feature activation+0.823
send
Token send
Feature activation+0.539
the
Token the
Feature activation+0.553
street
Token street
Feature activation+0.000
."
Token."
Feature activation+0.000
I
TokenI
Feature activation+0.445
thought
Token thought
Feature activation+1.599
these
Token these
Feature activation+1.397
were
Token were
Feature activation+1.383
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
the
Token the
Feature activation+1.070
ground
Token ground
Feature activation+0.771
and
Token and
Feature activation+0.000
isolated
Token isolated
Feature activation+0.000
neighbour
Token neighbour
Feature activation+0.000
."
Token."
Feature activation+0.302
I
TokenI
Feature activation+0.488
am
Token am
Feature activation+1.305
asking
Token asking
Feature activation+1.245
foreign
Token foreign
Feature activation+1.355
and
Token and
Feature activation+1.238
security
Token security
Feature activation+1.205
ministries
Token ministries
Feature activation+1.075

INTERVAL 1.096 - 1.279
CONTAINS 0.002%

to
Token to
Feature activation+0.816
holding
Token holding
Feature activation+0.448
this
Token this
Feature activation+0.832
dialogue
Token dialogue
Feature activation+0.630
and
Token and
Feature activation+0.778
we
Token we
Feature activation+1.126
'll
Token'll
Feature activation+0.557
look
Token look
Feature activation+0.606
for
Token for
Feature activation+0.522
a
Token a
Feature activation+0.299
time
Token time
Feature activation+0.174
an
Token an
Feature activation+0.000
unexpected
Token unexpected
Feature activation+0.000
appearance
Token appearance
Feature activation+0.000
."
Token."
Feature activation+0.082
We
TokenWe
Feature activation+0.539
never
Token never
Feature activation+1.179
once
Token once
Feature activation+1.223
imagined
Token imagined
Feature activation+1.016
that
Token that
Feature activation+0.916
this
Token this
Feature activation+1.156
would
Token would
Feature activation+0.898
We
TokenWe
Feature activation+0.539
never
Token never
Feature activation+1.179
once
Token once
Feature activation+1.223
imagined
Token imagined
Feature activation+1.016
that
Token that
Feature activation+0.916
this
Token this
Feature activation+1.156
would
Token would
Feature activation+0.898
touch
Token touch
Feature activation+0.535
anyone
Token anyone
Feature activation+0.701
high
Token high
Feature activation+0.549
up
Token up
Feature activation+0.360
don
Token don
Feature activation+1.253
't
Token't
Feature activation+1.146
know
Token know
Feature activation+1.453
what
Token what
Feature activation+1.438
the
Token the
Feature activation+1.545
answer
Token answer
Feature activation+1.194
is
Token is
Feature activation+1.067
,
Token,
Feature activation+1.452
or
Token or
Feature activation+0.843
how
Token how
Feature activation+1.014
to
Token to
Feature activation+0.823
the
Token the
Feature activation+1.070
ground
Token ground
Feature activation+0.771
,
Token,
Feature activation+1.468
but
Token but
Feature activation+1.320
apparently
Token apparently
Feature activation+0.922
they
Token they
Feature activation+1.105
're
Token're
Feature activation+0.599
not
Token not
Feature activation+0.176
.
Token.
Feature activation+0.328
It
Token It
Feature activation+0.000
must
Token must
Feature activation+0.000

INTERVAL 0.914 - 1.096
CONTAINS 0.002%

I
TokenI
Feature activation+0.445
thought
Token thought
Feature activation+1.599
these
Token these
Feature activation+1.397
were
Token were
Feature activation+1.383
really
Token really
Feature activation+1.050
planted
Token planted
Feature activation+0.921
in
Token in
Feature activation+0.841
the
Token the
Feature activation+1.070
ground
Token ground
Feature activation+0.771
,
Token,
Feature activation+1.468
but
Token but
Feature activation+1.320
appearance
Token appearance
Feature activation+0.000
."
Token."
Feature activation+0.082
We
TokenWe
Feature activation+0.539
never
Token never
Feature activation+1.179
once
Token once
Feature activation+1.223
imagined
Token imagined
Feature activation+1.016
that
Token that
Feature activation+0.916
this
Token this
Feature activation+1.156
would
Token would
Feature activation+0.898
touch
Token touch
Feature activation+0.535
anyone
Token anyone
Feature activation+0.701
on
Token on
Feature activation+0.633
the
Token the
Feature activation+0.743
tasting
Token tasting
Feature activation+0.369
menu
Token menu
Feature activation+0.276
,
Token,
Feature activation+0.733
(
Token (
Feature activation+1.085
Che
TokenChe
Feature activation+0.000
f
Tokenf
Feature activation+0.055
Jorge
Token Jorge
Feature activation+0.053
)
Token)
Feature activation+0.000
Val
Token Val
Feature activation+0.000
I
TokenI
Feature activation+0.089
thought
Token thought
Feature activation+0.943
he
Token he
Feature activation+1.254
was
Token was
Feature activation+1.133
buzzing
Token buzzing
Feature activation+0.806
,"
Token,"
Feature activation+0.976
Leafs
Token Leafs
Feature activation+0.480
forward
Token forward
Feature activation+0.387
Clarke
Token Clarke
Feature activation+0.772
MacArthur
Token MacArthur
Feature activation+0.623
said
Token said
Feature activation+0.824
afternoon
Token afternoon
Feature activation+0.000
."
Token."
Feature activation+0.000
Our
TokenOur
Feature activation+0.453
suspect
Token suspect
Feature activation+1.317
completely
Token completely
Feature activation+1.329
confessed
Token confessed
Feature activation+0.960
to
Token to
Feature activation+1.004
what
Token what
Feature activation+0.894
he
Token he
Feature activation+1.268
did
Token did
Feature activation+1.223
.
Token.
Feature activation+0.975

INTERVAL 0.731 - 0.914
CONTAINS 0.002%

Apple
Token Apple
Feature activation+0.471
we
Token we
Feature activation+0.983
are
Token are
Feature activation+0.702
not
Token not
Feature activation+0.642
just
Token just
Feature activation+0.752
enabling
Token enabling
Feature activation+0.849
others
Token others
Feature activation+0.504
to
Token to
Feature activation+0.354
speak
Token speak
Feature activation+0.478
up
Token up
Feature activation+0.236
,
Token,
Feature activation+0.721
was
Token was
Feature activation+1.133
buzzing
Token buzzing
Feature activation+0.806
,"
Token,"
Feature activation+0.976
Leafs
Token Leafs
Feature activation+0.480
forward
Token forward
Feature activation+0.387
Clarke
Token Clarke
Feature activation+0.772
MacArthur
Token MacArthur
Feature activation+0.623
said
Token said
Feature activation+0.824
of
Token of
Feature activation+0.334
K
Token K
Feature activation+0.216
essel
Tokenessel
Feature activation+0.259
night
Token night
Feature activation+0.000
."
Token."
Feature activation+0.029
He
TokenHe
Feature activation+0.048
was
Token was
Feature activation+0.884
not
Token not
Feature activation+0.620
happy
Token happy
Feature activation+0.824
with
Token with
Feature activation+0.604
the
Token the
Feature activation+0.651
marriage
Token marriage
Feature activation+0.722
.
Token.
Feature activation+0.239
He
Token He
Feature activation+0.125
,"
Token,"
Feature activation+0.976
Leafs
Token Leafs
Feature activation+0.480
forward
Token forward
Feature activation+0.387
Clarke
Token Clarke
Feature activation+0.772
MacArthur
Token MacArthur
Feature activation+0.623
said
Token said
Feature activation+0.824
of
Token of
Feature activation+0.334
K
Token K
Feature activation+0.216
essel
Tokenessel
Feature activation+0.259
.
Token.
Feature activation+0.051
"
Token "
Feature activation+0.121
At
TokenAt
Feature activation+0.000
Apple
Token Apple
Feature activation+0.471
we
Token we
Feature activation+0.983
are
Token are
Feature activation+0.702
not
Token not
Feature activation+0.642
just
Token just
Feature activation+0.752
enabling
Token enabling
Feature activation+0.849
others
Token others
Feature activation+0.504
to
Token to
Feature activation+0.354
speak
Token speak
Feature activation+0.478
up
Token up
Feature activation+0.236

INTERVAL 0.548 - 0.731
CONTAINS 0.005%

that
Token that
Feature activation+0.000
night
Token night
Feature activation+0.000
."
Token."
Feature activation+0.029
He
TokenHe
Feature activation+0.048
was
Token was
Feature activation+0.884
not
Token not
Feature activation+0.620
happy
Token happy
Feature activation+0.824
with
Token with
Feature activation+0.604
the
Token the
Feature activation+0.651
marriage
Token marriage
Feature activation+0.722
.
Token.
Feature activation+0.239
holding
Token holding
Feature activation+0.448
this
Token this
Feature activation+0.832
dialogue
Token dialogue
Feature activation+0.630
and
Token and
Feature activation+0.778
we
Token we
Feature activation+1.126
'll
Token'll
Feature activation+0.557
look
Token look
Feature activation+0.606
for
Token for
Feature activation+0.522
a
Token a
Feature activation+0.299
time
Token time
Feature activation+0.174
to
Token to
Feature activation+0.280
happiness
Token happiness
Feature activation+0.000
."
Token."
Feature activation+0.046
Charl
TokenCharl
Feature activation+0.000
ottesville
Tokenottesville
Feature activation+0.000
will
Token will
Feature activation+0.329
become
Token become
Feature activation+0.597
the
Token the
Feature activation+0.704
city
Token city
Feature activation+0.495
of
Token of
Feature activation+0.590
pot
Token pot
Feature activation+0.389
heads
Tokenheads
Feature activation+0.313
imagined
Token imagined
Feature activation+1.016
that
Token that
Feature activation+0.916
this
Token this
Feature activation+1.156
would
Token would
Feature activation+0.898
touch
Token touch
Feature activation+0.535
anyone
Token anyone
Feature activation+0.701
high
Token high
Feature activation+0.549
up
Token up
Feature activation+0.360
in
Token in
Feature activation+0.607
the
Token the
Feature activation+0.612
Papa
Token Papa
Feature activation+0.704
ground
Token ground
Feature activation+0.771
,
Token,
Feature activation+1.468
but
Token but
Feature activation+1.320
apparently
Token apparently
Feature activation+0.922
they
Token they
Feature activation+1.105
're
Token're
Feature activation+0.599
not
Token not
Feature activation+0.176
.
Token.
Feature activation+0.328
It
Token It
Feature activation+0.000
must
Token must
Feature activation+0.000
have
Token have
Feature activation+0.015

INTERVAL 0.365 - 0.548
CONTAINS 0.009%

that
Token that
Feature activation+0.388
you
Token you
Feature activation+0.570
shouldn
Token shouldn
Feature activation+0.587
't
Token't
Feature activation+0.157
drive
Token drive
Feature activation+0.000
,"
Token,"
Feature activation+0.532
he
Token he
Feature activation+0.516
said
Token said
Feature activation+0.438
.
Token.
Feature activation+0.000
He
TokenHe
Feature activation+0.000
says
Token says
Feature activation+0.215
ting
Tokenting
Feature activation+0.004
flights
Token flights
Feature activation+0.401
over
Token over
Feature activation+0.171
Ukraine
Token Ukraine
Feature activation+0.437
on
Token on
Feature activation+0.348
Thursday
Token Thursday
Feature activation+0.537
night
Token night
Feature activation+0.447
were
Token were
Feature activation+0.308
Italy
Token Italy
Feature activation+0.000
's
Token's
Feature activation+0.027
Al
Token Al
Feature activation+0.000
radical
Tokenradical
Feature activation+0.029
and
Token and
Feature activation+0.000
Islamic
Token Islamic
Feature activation+0.000
outfits
Token outfits
Feature activation+0.246
,
Token,
Feature activation+0.167
the
Token the
Feature activation+0.427
fist
Token fist
Feature activation+0.129
-
Token-
Feature activation+0.000
b
Tokenb
Feature activation+0.000
ump
Tokenump
Feature activation+0.000
,
Token,
Feature activation+0.000
afternoon
Token afternoon
Feature activation+1.476
.
Token.
Feature activation+0.859
By
Token By
Feature activation+0.381
then
Token then
Feature activation+0.388
,
Token,
Feature activation+0.429
he
Token he
Feature activation+0.367
had
Token had
Feature activation+0.250
already
Token already
Feature activation+0.000
command
Token command
Feature activation+0.000
e
Tokene
Feature activation+0.000
ered
Tokenered
Feature activation+0.000
how
Token how
Feature activation+1.014
to
Token to
Feature activation+0.823
send
Token send
Feature activation+0.539
the
Token the
Feature activation+0.553
message
Token message
Feature activation+0.217
that
Token that
Feature activation+0.388
you
Token you
Feature activation+0.570
shouldn
Token shouldn
Feature activation+0.587
't
Token't
Feature activation+0.157
drive
Token drive
Feature activation+0.000
,"
Token,"
Feature activation+0.532

INTERVAL 0.183 - 0.365
CONTAINS 0.016%

<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
'
Token'
Feature activation+0.000
The
TokenThe
Feature activation+0.000
men
Token men
Feature activation+0.000
,
Token,
Feature activation+0.048
over
Token over
Feature activation+0.250
a
Token a
Feature activation+0.005
prolonged
Token prolonged
Feature activation+0.000
period
Token period
Feature activation+0.000
of
Token of
Feature activation+0.000
time
Token time
Feature activation+0.000
that
Token that
Feature activation+0.000
rewards
Token rewards
Feature activation+0.000
value
Token value
Feature activation+0.000
."
Token."
Feature activation+0.000
Sarah
Token Sarah
Feature activation+0.009
Kl
Token Kl
Feature activation+0.341
iff
Tokeniff
Feature activation+0.221
at
Token at
Feature activation+0.463
Vox
Token Vox
Feature activation+0.489
.
Token.
Feature activation+0.244
Ċ
TokenĊ
Feature activation+0.000
profile
Token profile
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Davis
TokenDavis
Feature activation+0.000
,
Token,
Feature activation+0.204
who
Token who
Feature activation+0.263
won
Token won
Feature activation+0.230
the
Token the
Feature activation+0.000
supporting
Token supporting
Feature activation+0.000
actress
Token actress
Feature activation+0.000
Ronald
Token Ronald
Feature activation+0.000
Ko
Token Ko
Feature activation+0.000
eman
Tokeneman
Feature activation+0.070
already
Token already
Feature activation+0.322
seems
Token seems
Feature activation+0.348
to
Token to
Feature activation+0.277
have
Token have
Feature activation+0.255
tightened
Token tightened
Feature activation+0.037
up
Token up
Feature activation+0.000
the
Token the
Feature activation+0.193
Everton
Token Everton
Feature activation+0.012
New
Token New
Feature activation+0.030
Yorker
Token Yorker
Feature activation+0.165
may
Token may
Feature activation+0.389
think
Token think
Feature activation+0.373
,
Token,
Feature activation+0.360
as
Token as
Feature activation+0.305
one
Token one
Feature activation+0.012
of
Token of
Feature activation+0.000
their
Token their
Feature activation+0.000
staff
Token staff
Feature activation+0.000
explained
Token explained
Feature activation+0.000

INTERVAL 0.000 - 0.183
CONTAINS 99.962%

and
Token and
Feature activation+0.000
spices
Token spices
Feature activation+0.000
.
Token.
Feature activation+0.000
Heat
Token Heat
Feature activation+0.000
once
Token once
Feature activation+0.000
more
Token more
Feature activation+0.000
and
Token and
Feature activation+0.000
set
Token set
Feature activation+0.000
aside
Token aside
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+0.000
Ċ
TokenĊ
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
F
TokenF
Feature activation+0.000
ulf
Tokenulf
Feature activation+0.000
illing
Tokenilling
Feature activation+0.000
a
Token a
Feature activation+0.000
contract
Token contract
Feature activation+0.000
signed
Token signed
Feature activation+0.000
,
Token,
Feature activation+0.000
with
Token with
Feature activation+0.000
penalties
Token penalties
Feature activation+0.000
of
Token of
Feature activation+0.000
five
Token five
Feature activation+0.000
years
Token years
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
imprisonment
Token imprisonment
Feature activation+0.000
or
Token or
Feature activation+0.000
a
Token a
Feature activation+0.000
www
Token www
Feature activation+0.000
.
Token.
Feature activation+0.000
in
Tokenin
Feature activation+0.000
4
Token4
Feature activation+0.000
ins
Tokenins
Feature activation+0.000
.
Token.
Feature activation+0.000
com
Tokencom
Feature activation+0.000
for
Token for
Feature activation+0.000
additional
Token additional
Feature activation+0.000
details
Token details
Feature activation+0.000
about
Token about
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
M
Token M
Feature activation+0.000
4
Token4
Feature activation+0.000
shotgun
Token shotgun
Feature activation+0.000
is
Token is
Feature activation+0.000
sold
Token sold
Feature activation+0.000
in
Token in
Feature activation+0.000
three
Token three
Feature activation+0.000
configurations
Token configurations
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 1 in H2.6: (feature 6808)

TOP ACTIVATIONS
MAX = 2.545

in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
the
Token the
Feature activation+1.606
in
Token in
Feature activation+0.367
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
Div
Token Div
Feature activation+0.000
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.367
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.367
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
Div
Token Div
Feature activation+0.000
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
the
Token the
Feature activation+1.606
Spanish
Token Spanish
Feature activation+1.599
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
dialogue
Token dialogue
Feature activation+0.109
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
dialogue
Token dialogue
Feature activation+0.109
on
Token on
Feature activation+0.025
October
Token October
Feature activation+1.877
7
Token 7
Feature activation+1.117
in
Token in
Feature activation+0.733
Madrid
Token Madrid
Feature activation+1.360
.
Token.
Feature activation+1.015
Neither
Token Neither
Feature activation+0.174
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
dialogue
Token dialogue
Feature activation+0.109
on
Token on
Feature activation+0.025
October
Token October
Feature activation+1.877
7
Token 7
Feature activation+1.117
:
Token:
Feature activation+0.000
Div
Token Div
Feature activation+0.000
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
of
Token of
Feature activation+0.107
the
Token the
Feature activation+0.795
United
Token United
Feature activation+1.513
States
Token States
Feature activation+1.386
on
Token on
Feature activation+0.562
January
Token January
Feature activation+1.692
20
Token 20
Feature activation+1.152
,
Token,
Feature activation+1.300
2017
Token 2017
Feature activation+1.588
,
Token,
Feature activation+0.724
outlining
Token outlining
Feature activation+1.087
first
Token first
Feature activation+1.196
70
Token 70
Feature activation+0.973
days
Token days
Feature activation+0.752
Donald
Token Donald
Feature activation+0.984
Trump
Token Trump
Feature activation+1.143
is
Token is
Feature activation+1.678
sworn
Token sworn
Feature activation+1.333
in
Token in
Feature activation+0.228
as
Token as
Feature activation+0.743
president
Token president
Feature activation+0.782
of
Token of
Feature activation+0.107
:
Token:
Feature activation+0.000
Div
Token Div
Feature activation+0.000
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.367
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
dialogue
Token dialogue
Feature activation+0.109
on
Token on
Feature activation+0.025
October
Token October
Feature activation+1.877
7
Token 7
Feature activation+1.117
in
Token in
Feature activation+0.733
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
the
Token the
Feature activation+1.606
Spanish
Token Spanish
Feature activation+1.599
word
Token word
Feature activation+0.706
for
Token for
Feature activation+0.896
"
Token "
Feature activation+1.104
peace
Tokenpeace
Feature activation+0.610
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
the
Token the
Feature activation+1.606
Spanish
Token Spanish
Feature activation+1.599
word
Token word
Feature activation+0.706
for
Token for
Feature activation+0.896
"
Token "
Feature activation+1.104
peace
Tokenpeace
Feature activation+0.610
"
Token"
Feature activation+0.406
States
Token States
Feature activation+1.386
on
Token on
Feature activation+0.562
January
Token January
Feature activation+1.692
20
Token 20
Feature activation+1.152
,
Token,
Feature activation+1.300
2017
Token 2017
Feature activation+1.588
,
Token,
Feature activation+0.724
outlining
Token outlining
Feature activation+1.087
his
Token his
Feature activation+0.180
"
Token "
Feature activation+0.397
America
TokenAmerica
Feature activation+0.000

Top DFA by src position
MAX = 2.594

Hide
Token Hide
Feature activation+0.027
Top resid features:
Caption
Token Caption
Feature activation+1.006
Top resid features:
5
Token 5
Feature activation-0.008
Top resid features:
of
Token of
Feature activation+0.012
Top resid features:
37
Token 37
Feature activation+0.028
Top resid features:
Photos
Token Photos
Feature activation+1.784
Top resid features:
:
Token:
Feature activation+0.037
Top resid features:
Div
Token Div
Feature activation+0.016
Top resid features:
isions
Tokenisions
Feature activation+0.038
Top resid features:
in
Token in
Feature activation+0.078
Top resid features:
Spain
Token Spain
Feature activation+0.074
Top resid features:
Hide
Token Hide
Feature activation+0.072
Top resid features:
Caption
Token Caption
Feature activation+0.887
Top resid features:
6
Token 6
Feature activation-0.020
Top resid features:
of
Token of
Feature activation+0.007
Top resid features:
37
Token 37
Feature activation+0.050
Top resid features:
Photos
Token Photos
Feature activation+1.723
Top resid features:
:
Token:
Feature activation+0.036
Top resid features:
Div
Token Div
Feature activation+0.019
Top resid features:
isions
Tokenisions
Feature activation+0.056
Top resid features:
in
Token in
Feature activation+0.045
Top resid features:
Spain
Token Spain
Feature activation+0.063
Top resid features:
Hide
Token Hide
Feature activation+0.073
Top resid features:
Caption
Token Caption
Feature activation+0.932
Top resid features:
6
Token 6
Feature activation-0.023
Top resid features:
of
Token of
Feature activation+0.008
Top resid features:
37
Token 37
Feature activation+0.056
Top resid features:
Photos
Token Photos
Feature activation+1.424
Top resid features:
:
Token:
Feature activation+0.037
Top resid features:
Div
Token Div
Feature activation+0.029
Top resid features:
isions
Tokenisions
Feature activation+0.069
Top resid features:
in
Token in
Feature activation+0.067
Top resid features:
Spain
Token Spain
Feature activation+0.112
Top resid features:
Hide
Token Hide
Feature activation+0.115
Top resid features:
Caption
Token Caption
Feature activation+0.704
Top resid features:
6
Token 6
Feature activation-0.037
Top resid features:
of
Token of
Feature activation+0.014
Top resid features:
37
Token 37
Feature activation+0.059
Top resid features:
Photos
Token Photos
Feature activation+1.520
Top resid features:
:
Token:
Feature activation-0.029
Top resid features:
Div
Token Div
Feature activation+0.039
Top resid features:
isions
Tokenisions
Feature activation+0.090
Top resid features:
in
Token in
Feature activation+0.138
Top resid features:
Spain
Token Spain
Feature activation+0.118
Top resid features:
Hide
Token Hide
Feature activation+0.088
Top resid features:
Caption
Token Caption
Feature activation+0.538
Top resid features:
6
Token 6
Feature activation-0.028
Top resid features:
of
Token of
Feature activation+0.011
Top resid features:
37
Token 37
Feature activation+0.036
Top resid features:
Photos
Token Photos
Feature activation+1.395
Top resid features:
:
Token:
Feature activation-0.010
Top resid features:
Div
Token Div
Feature activation+0.024
Top resid features:
isions
Tokenisions
Feature activation+0.059
Top resid features:
in
Token in
Feature activation+0.099
Top resid features:
Spain
Token Spain
Feature activation+0.117
Top resid features:
Hide
Token Hide
Feature activation+0.043
Top resid features:
Caption
Token Caption
Feature activation+0.713
Top resid features:
5
Token 5
Feature activation-0.013
Top resid features:
of
Token of
Feature activation+0.016
Top resid features:
37
Token 37
Feature activation+0.048
Top resid features:
Photos
Token Photos
Feature activation+1.635
Top resid features:
:
Token:
Feature activation-0.024
Top resid features:
Div
Token Div
Feature activation+0.032
Top resid features:
isions
Tokenisions
Feature activation+0.070
Top resid features:
in
Token in
Feature activation+0.168
Top resid features:
Spain
Token Spain
Feature activation+0.109
Top resid features:
Hide
Token Hide
Feature activation+0.082
Top resid features:
Caption
Token Caption
Feature activation+0.639
Top resid features:
6
Token 6
Feature activation-0.013
Top resid features:
of
Token of
Feature activation+0.007
Top resid features:
37
Token 37
Feature activation+0.055
Top resid features:
Photos
Token Photos
Feature activation+1.153
Top resid features:
:
Token:
Feature activation+0.073
Top resid features:
Div
Token Div
Feature activation+0.015
Top resid features:
isions
Tokenisions
Feature activation+0.033
Top resid features:
in
Token in
Feature activation+0.017
Top resid features:
Spain
Token Spain
Feature activation+0.040
Top resid features:
Hide
Token Hide
Feature activation+0.048
Top resid features:
Caption
Token Caption
Feature activation+0.677
Top resid features:
5
Token 5
Feature activation+0.006
Top resid features:
of
Token of
Feature activation+0.007
Top resid features:
37
Token 37
Feature activation+0.066
Top resid features:
Photos
Token Photos
Feature activation+1.387
Top resid features:
:
Token:
Feature activation+0.075
Top resid features:
Div
Token Div
Feature activation+0.003
Top resid features:
isions
Tokenisions
Feature activation+0.013
Top resid features:
in
Token in
Feature activation+0.005
Top resid features:
Spain
Token Spain
Feature activation+0.030
Top resid features:
Hide
Token Hide
Feature activation+0.050
Top resid features:
Caption
Token Caption
Feature activation+0.678
Top resid features:
5
Token 5
Feature activation+0.011
Top resid features:
of
Token of
Feature activation+0.003
Top resid features:
37
Token 37
Feature activation+0.093
Top resid features:
Photos
Token Photos
Feature activation+1.379
Top resid features:
:
Token:
Feature activation+0.034
Top resid features:
Div
Token Div
Feature activation+0.002
Top resid features:
isions
Tokenisions
Feature activation+0.005
Top resid features:
in
Token in
Feature activation+0.002
Top resid features:
Spain
Token Spain
Feature activation+0.012
Top resid features:
Hide
Token Hide
Feature activation+0.034
Top resid features:
Caption
Token Caption
Feature activation+0.510
Top resid features:
5
Token 5
Feature activation-0.007
Top resid features:
of
Token of
Feature activation+0.014
Top resid features:
37
Token 37
Feature activation+0.027
Top resid features:
Photos
Token Photos
Feature activation+1.466
Top resid features:
:
Token:
Feature activation+0.001
Top resid features:
Div
Token Div
Feature activation+0.014
Top resid features:
isions
Tokenisions
Feature activation+0.038
Top resid features:
in
Token in
Feature activation+0.111
Top resid features:
Spain
Token Spain
Feature activation+0.116
Top resid features:
Hide
Token Hide
Feature activation+0.057
Top resid features:
Caption
Token Caption
Feature activation+0.755
Top resid features:
5
Token 5
Feature activation+0.004
Top resid features:
of
Token of
Feature activation+0.009
Top resid features:
37
Token 37
Feature activation+0.090
Top resid features:
Photos
Token Photos
Feature activation+1.347
Top resid features:
:
Token:
Feature activation+0.078
Top resid features:
Div
Token Div
Feature activation+0.007
Top resid features:
isions
Tokenisions
Feature activation+0.026
Top resid features:
in
Token in
Feature activation+0.044
Top resid features:
Spain
Token Spain
Feature activation+0.052
Top resid features:
Hide
Token Hide
Feature activation+0.047
Top resid features:
Caption
Token Caption
Feature activation+0.662
Top resid features:
5
Token 5
Feature activation+0.005
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
37
Token 37
Feature activation+0.055
Top resid features:
Photos
Token Photos
Feature activation+1.152
Top resid features:
:
Token:
Feature activation+0.061
Top resid features:
Div
Token Div
Feature activation+0.003
Top resid features:
isions
Tokenisions
Feature activation+0.009
Top resid features:
in
Token in
Feature activation+0.003
Top resid features:
Spain
Token Spain
Feature activation+0.014
Top resid features:
Hide
Token Hide
Feature activation+0.028
Top resid features:
Caption
Token Caption
Feature activation+0.700
Top resid features:
5
Token 5
Feature activation-0.013
Top resid features:
of
Token of
Feature activation+0.014
Top resid features:
37
Token 37
Feature activation+0.083
Top resid features:
Photos
Token Photos
Feature activation+1.488
Top resid features:
:
Token:
Feature activation-0.061
Top resid features:
Div
Token Div
Feature activation+0.050
Top resid features:
isions
Tokenisions
Feature activation+0.197
Top resid features:
in
Token in
Feature activation+0.268
Top resid features:
Spain
Token Spain
Feature activation+0.302
Top resid features:
's
Token's
Feature activation+0.001
Top resid features:
first
Token first
Feature activation+0.009
Top resid features:
70
Token 70
Feature activation+0.004
Top resid features:
days
Token days
Feature activation+0.008
Top resid features:
See
Token See
Feature activation+0.167
Top resid features:
Gallery
Token Gallery
Feature activation+1.833
Top resid features:
Inside
Token Inside
Feature activation+0.021
Top resid features:
President
Token President
Feature activation-0.007
Top resid features:
Trump
Token Trump
Feature activation-0.026
Top resid features:
's
Token's
Feature activation-0.002
Top resid features:
first
Token first
Feature activation+0.007
Top resid features:
's
Token's
Feature activation-0.004
Top resid features:
first
Token first
Feature activation+0.002
Top resid features:
70
Token 70
Feature activation-0.006
Top resid features:
days
Token days
Feature activation+0.008
Top resid features:
See
Token See
Feature activation+0.052
Top resid features:
Gallery
Token Gallery
Feature activation+2.594
Top resid features:
Inside
Token Inside
Feature activation+0.034
Top resid features:
President
Token President
Feature activation-0.013
Top resid features:
Trump
Token Trump
Feature activation-0.063
Top resid features:
's
Token's
Feature activation+0.012
Top resid features:
first
Token first
Feature activation+0.025
Top resid features:
Hide
Token Hide
Feature activation+0.073
Top resid features:
Caption
Token Caption
Feature activation+0.615
Top resid features:
6
Token 6
Feature activation-0.038
Top resid features:
of
Token of
Feature activation+0.015
Top resid features:
37
Token 37
Feature activation+0.059
Top resid features:
Photos
Token Photos
Feature activation+1.267
Top resid features:
:
Token:
Feature activation-0.056
Top resid features:
Div
Token Div
Feature activation+0.049
Top resid features:
isions
Tokenisions
Feature activation+0.204
Top resid features:
in
Token in
Feature activation+0.257
Top resid features:
Spain
Token Spain
Feature activation+0.290
Top resid features:
Hide
Token Hide
Feature activation+0.055
Top resid features:
Caption
Token Caption
Feature activation+0.616
Top resid features:
5
Token 5
Feature activation+0.006
Top resid features:
of
Token of
Feature activation+0.006
Top resid features:
37
Token 37
Feature activation+0.051
Top resid features:
Photos
Token Photos
Feature activation+1.354
Top resid features:
:
Token:
Feature activation+0.069
Top resid features:
Div
Token Div
Feature activation+0.001
Top resid features:
isions
Tokenisions
Feature activation+0.006
Top resid features:
in
Token in
Feature activation+0.002
Top resid features:
Spain
Token Spain
Feature activation+0.016
Top resid features:
Hide
Token Hide
Feature activation+0.065
Top resid features:
Caption
Token Caption
Feature activation+0.491
Top resid features:
6
Token 6
Feature activation-0.005
Top resid features:
of
Token of
Feature activation+0.005
Top resid features:
37
Token 37
Feature activation+0.034
Top resid features:
Photos
Token Photos
Feature activation+1.015
Top resid features:
:
Token:
Feature activation+0.066
Top resid features:
Div
Token Div
Feature activation+0.009
Top resid features:
isions
Tokenisions
Feature activation+0.018
Top resid features:
in
Token in
Feature activation+0.004
Top resid features:
Spain
Token Spain
Feature activation+0.023
Top resid features:
Hide
Token Hide
Feature activation+0.058
Top resid features:
Caption
Token Caption
Feature activation+0.434
Top resid features:
6
Token 6
Feature activation-0.005
Top resid features:
of
Token of
Feature activation+0.009
Top resid features:
37
Token 37
Feature activation+0.042
Top resid features:
Photos
Token Photos
Feature activation+1.161
Top resid features:
:
Token:
Feature activation+0.057
Top resid features:
Div
Token Div
Feature activation+0.010
Top resid features:
isions
Tokenisions
Feature activation+0.025
Top resid features:
in
Token in
Feature activation+0.005
Top resid features:
Spain
Token Spain
Feature activation+0.023
Top resid features:
's
Token's
Feature activation+0.001
Top resid features:
first
Token first
Feature activation+0.009
Top resid features:
70
Token 70
Feature activation+0.006
Top resid features:
days
Token days
Feature activation+0.008
Top resid features:
See
Token See
Feature activation+0.134
Top resid features:
Gallery
Token Gallery
Feature activation+1.885
Top resid features:
Inside
Token Inside
Feature activation+0.029
Top resid features:
President
Token President
Feature activation-0.004
Top resid features:
Trump
Token Trump
Feature activation-0.022
Top resid features:
's
Token's
Feature activation-0.001
Top resid features:
first
Token first
Feature activation+0.007
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.08

Head 2: 0.04

Head 3: 0.05

Head 4: 0.03

Head 5: 0.06

Head 6: 0.35

Head 7: 0.07

Head 8: 0.06

Head 9: 0.07

Head 10: 0.10

Head 11: 0.04

Positive logits

weep1.58

ideshow1.54

ription1.35

pictured1.34

caption1.30

Emma1.30

Nicholas1.29

itialized1.29

anson1.28

riet1.27

selfie1.27

?'1.27

recl1.27

photo1.24

MISS1.24

[/1.24

DAV1.21

cerpt1.21

ucha1.21

Lub1.21

Negative logits

ModLoader-2.04

Pwr-1.70

ETF-1.58

soType-1.58

Transmission-1.41

Marketable-1.40

xual-1.40

mods-1.38

sbm-1.37

chrom-1.29

constitu-1.27

Defin-1.25

Parties-1.25

TPP-1.25

Compliance-1.21

Protocol-1.21

volume-1.19

Logic-1.19

gin-1.19

Hash-1.19

INTERVAL 2.290 - 2.545
CONTAINS 0.000%

in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
in
Token in
Feature activation+0.367
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
the
Token the
Feature activation+1.606

INTERVAL 2.036 - 2.290
CONTAINS 0.000%

Div
Token Div
Feature activation+0.000
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
Div
Token Div
Feature activation+0.000
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.367
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.367
Spain
Token Spain
Feature activation+1.386
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107

INTERVAL 1.781 - 2.036
CONTAINS 0.000%

isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
dialogue
Token dialogue
Feature activation+0.109
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
dialogue
Token dialogue
Feature activation+0.109
on
Token on
Feature activation+0.025
October
Token October
Feature activation+1.877
7
Token 7
Feature activation+1.117
in
Token in
Feature activation+0.733
Madrid
Token Madrid
Feature activation+1.360
.
Token.
Feature activation+1.015
Neither
Token Neither
Feature activation+0.174
over
Token over
Feature activation+1.640
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
the
Token the
Feature activation+1.606
Spanish
Token Spanish
Feature activation+1.599

INTERVAL 1.527 - 1.781
CONTAINS 0.001%

Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
dialogue
Token dialogue
Feature activation+0.109
on
Token on
Feature activation+0.025
October
Token October
Feature activation+1.877
7
Token 7
Feature activation+1.117
in
Token in
Feature activation+0.733
States
Token States
Feature activation+1.386
on
Token on
Feature activation+0.562
January
Token January
Feature activation+1.692
20
Token 20
Feature activation+1.152
,
Token,
Feature activation+1.300
2017
Token 2017
Feature activation+1.588
,
Token,
Feature activation+0.724
outlining
Token outlining
Feature activation+1.087
his
Token his
Feature activation+0.180
"
Token "
Feature activation+0.397
America
TokenAmerica
Feature activation+0.000
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
the
Token the
Feature activation+1.606
Spanish
Token Spanish
Feature activation+1.599
word
Token word
Feature activation+0.706
for
Token for
Feature activation+0.896
"
Token "
Feature activation+1.104
peace
Tokenpeace
Feature activation+0.610
"
Token"
Feature activation+0.406
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
dialogue
Token dialogue
Feature activation+0.109
on
Token on
Feature activation+0.025

INTERVAL 1.272 - 1.527
CONTAINS 0.001%

70
Token 70
Feature activation+0.973
days
Token days
Feature activation+0.752
Donald
Token Donald
Feature activation+0.984
Trump
Token Trump
Feature activation+1.143
is
Token is
Feature activation+1.678
sworn
Token sworn
Feature activation+1.333
in
Token in
Feature activation+0.228
as
Token as
Feature activation+0.743
president
Token president
Feature activation+0.782
of
Token of
Feature activation+0.107
the
Token the
Feature activation+0.795
Catalonia
Token Catalonia
Feature activation+2.246
crisis
Token crisis
Feature activation+2.070
A
Token A
Feature activation+2.441
woman
Token woman
Feature activation+2.534
wears
Token wears
Feature activation+1.978
a
Token a
Feature activation+1.442
sticker
Token sticker
Feature activation+1.107
with
Token with
Feature activation+1.063
the
Token the
Feature activation+1.606
Spanish
Token Spanish
Feature activation+1.599
word
Token word
Feature activation+0.706
as
Token as
Feature activation+0.743
president
Token president
Feature activation+0.782
of
Token of
Feature activation+0.107
the
Token the
Feature activation+0.795
United
Token United
Feature activation+1.513
States
Token States
Feature activation+1.386
on
Token on
Feature activation+0.562
January
Token January
Feature activation+1.692
20
Token 20
Feature activation+1.152
,
Token,
Feature activation+1.300
2017
Token 2017
Feature activation+1.588
Photos
Token Photos
Feature activation+0.000
:
Token:
Feature activation+0.000
Div
Token Div
Feature activation+0.000
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
in
Token in
Feature activation+0.228
as
Token as
Feature activation+0.743
president
Token president
Feature activation+0.782
of
Token of
Feature activation+0.107
the
Token the
Feature activation+0.795
United
Token United
Feature activation+1.513
States
Token States
Feature activation+1.386
on
Token on
Feature activation+0.562
January
Token January
Feature activation+1.692
20
Token 20
Feature activation+1.152
,
Token,
Feature activation+1.300

INTERVAL 1.018 - 1.272
CONTAINS 0.002%

the
Token the
Feature activation+0.795
United
Token United
Feature activation+1.513
States
Token States
Feature activation+1.386
on
Token on
Feature activation+0.562
January
Token January
Feature activation+1.692
20
Token 20
Feature activation+1.152
,
Token,
Feature activation+1.300
2017
Token 2017
Feature activation+1.588
,
Token,
Feature activation+0.724
outlining
Token outlining
Feature activation+1.087
his
Token his
Feature activation+0.180
's
Token's
Feature activation+0.552
first
Token first
Feature activation+1.196
70
Token 70
Feature activation+0.973
days
Token days
Feature activation+0.752
Donald
Token Donald
Feature activation+0.984
Trump
Token Trump
Feature activation+1.143
is
Token is
Feature activation+1.678
sworn
Token sworn
Feature activation+1.333
in
Token in
Feature activation+0.228
as
Token as
Feature activation+0.743
president
Token president
Feature activation+0.782
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
"
Token "
Feature activation+1.104
peace
Tokenpeace
Feature activation+0.610
"
Token"
Feature activation+0.406
at
Token at
Feature activation+0.772
a
Token a
Feature activation+0.654
demonstration
Token demonstration
Feature activation+1.063
in
Token in
Feature activation+0.355
Madrid
Token Madrid
Feature activation+1.120
on
Token on
Feature activation+0.268
October
Token October
Feature activation+1.696
7
Token 7
Feature activation+0.901
6
Token 6
Feature activation+0.000
Photos
Token Photos
Feature activation+0.384
:
Token:
Feature activation+0.000
Who
Token Who
Feature activation+0.076
's
Token's
Feature activation+0.298
running
Token running
Feature activation+1.259
for
Token for
Feature activation+0.830
president
Token president
Feature activation+1.268
?
Token?
Feature activation+0.817
Ċ
TokenĊ
Feature activation+0.103
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 0.763 - 1.018
CONTAINS 0.003%

?
Token?
Feature activation+0.264
Ohio
Token Ohio
Feature activation+0.858
Gov
Token Gov
Feature activation+0.872
.
Token.
Feature activation+0.564
John
Token John
Feature activation+1.383
Kasich
Token Kasich
Feature activation+0.903
joined
Token joined
Feature activation+1.258
the
Token the
Feature activation+0.601
Republican
Token Republican
Feature activation+0.225
field
Token field
Feature activation+0.222
July
Token July
Feature activation+1.159
Offic
Token Offic
Feature activation+0.553
ially
Tokenially
Feature activation+0.480
a
Token a
Feature activation+0.756
U
Token U
Feature activation+0.991
.
Token.
Feature activation+0.964
S
TokenS
Feature activation+0.862
.
Token.
Feature activation+0.993
common
Token common
Feature activation+0.470
wealth
Tokenwealth
Feature activation+0.751
,
Token,
Feature activation+0.574
Puerto
Token Puerto
Feature activation+0.714
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
a
Token a
Feature activation+1.355
demonstration
Token demonstration
Feature activation+1.748
urging
Token urging
Feature activation+1.627
Inside
Token Inside
Feature activation+0.140
President
Token President
Feature activation+0.645
Trump
Token Trump
Feature activation+0.770
's
Token's
Feature activation+0.552
first
Token first
Feature activation+1.196
70
Token 70
Feature activation+0.973
days
Token days
Feature activation+0.752
Donald
Token Donald
Feature activation+0.984
Trump
Token Trump
Feature activation+1.143
is
Token is
Feature activation+1.678
sworn
Token sworn
Feature activation+1.333
6
Token 6
Feature activation+0.000
Photos
Token Photos
Feature activation+0.000
:
Token:
Feature activation+0.000
Who
Token Who
Feature activation+0.000
's
Token's
Feature activation+0.000
running
Token running
Feature activation+0.828
for
Token for
Feature activation+0.288
president
Token president
Feature activation+0.718
?
Token?
Feature activation+0.264
Ohio
Token Ohio
Feature activation+0.858
Gov
Token Gov
Feature activation+0.872

INTERVAL 0.509 - 0.763
CONTAINS 0.003%

Donald
Token Donald
Feature activation+0.984
Trump
Token Trump
Feature activation+1.143
is
Token is
Feature activation+1.678
sworn
Token sworn
Feature activation+1.333
in
Token in
Feature activation+0.228
as
Token as
Feature activation+0.743
president
Token president
Feature activation+0.782
of
Token of
Feature activation+0.107
the
Token the
Feature activation+0.795
United
Token United
Feature activation+1.513
States
Token States
Feature activation+1.386
)
Token)
Feature activation+0.000
Four
Token Four
Feature activation+0.036
million
Token million
Feature activation+0.000
people
Token people
Feature activation+0.483
around
Token around
Feature activation+0.406
the
Token the
Feature activation+0.569
world
Token world
Feature activation+0.386
,
Token,
Feature activation+0.290
including
Token including
Feature activation+0.062
500
Token 500
Feature activation+0.154
,
Token,
Feature activation+0.000
37
Token 37
Feature activation+0.000
Photos
Token Photos
Feature activation+0.000
:
Token:
Feature activation+0.000
Div
Token Div
Feature activation+0.000
isions
Tokenisions
Feature activation+0.000
in
Token in
Feature activation+0.544
Spain
Token Spain
Feature activation+1.491
over
Token over
Feature activation+1.702
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
Catalonia
Token Catalonia
Feature activation+2.065
crisis
Token crisis
Feature activation+1.850
People
Token People
Feature activation+2.545
show
Token show
Feature activation+1.773
their
Token their
Feature activation+1.025
hands
Token hands
Feature activation+0.662
painted
Token painted
Feature activation+1.140
in
Token in
Feature activation+0.939
white
Token white
Feature activation+1.892
during
Token during
Feature activation+1.582
a
Token a
Feature activation+1.355
the
Token the
Feature activation+1.606
Spanish
Token Spanish
Feature activation+1.599
word
Token word
Feature activation+0.706
for
Token for
Feature activation+0.896
"
Token "
Feature activation+1.104
peace
Tokenpeace
Feature activation+0.610
"
Token"
Feature activation+0.406
at
Token at
Feature activation+0.772
a
Token a
Feature activation+0.654
demonstration
Token demonstration
Feature activation+1.063
in
Token in
Feature activation+0.355

INTERVAL 0.254 - 0.509
CONTAINS 0.003%

round
Token round
Feature activation+0.000
of
Token of
Feature activation+0.000
spar
Token spar
Feature activation+0.331
ring
Tokenring
Feature activation+0.000
.
Token.
Feature activation+0.429
Mem
Token Mem
Feature activation+0.360
orable
Tokenorable
Feature activation+0.000
quotes
Token quotes
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
third
Token third
Feature activation+0.000
Who
Token Who
Feature activation+0.000
's
Token's
Feature activation+0.000
running
Token running
Feature activation+0.828
for
Token for
Feature activation+0.288
president
Token president
Feature activation+0.718
?
Token?
Feature activation+0.264
Ohio
Token Ohio
Feature activation+0.858
Gov
Token Gov
Feature activation+0.872
.
Token.
Feature activation+0.564
John
Token John
Feature activation+1.383
Kasich
Token Kasich
Feature activation+0.903
lost
Token lost
Feature activation+0.000
in
Token in
Feature activation+0.000
Puerto
Token Puerto
Feature activation+0.000
Rico
Token Rico
Feature activation+0.000
View
Token View
Feature activation+0.000
Photos
Token Photos
Feature activation+0.309
Offic
Token Offic
Feature activation+0.553
ially
Tokenially
Feature activation+0.480
a
Token a
Feature activation+0.756
U
Token U
Feature activation+0.991
.
Token.
Feature activation+0.964
Photos
Token Photos
Feature activation+0.000
:
Token:
Feature activation+0.000
Who
Token Who
Feature activation+0.000
's
Token's
Feature activation+0.000
running
Token running
Feature activation+0.828
for
Token for
Feature activation+0.288
president
Token president
Feature activation+0.718
?
Token?
Feature activation+0.264
Ohio
Token Ohio
Feature activation+0.858
Gov
Token Gov
Feature activation+0.872
.
Token.
Feature activation+0.564
together
Token together
Feature activation+0.000
in
Token in
Feature activation+0.108
Colorado
Token Colorado
Feature activation+0.956
for
Token for
Feature activation+0.497
the
Token the
Feature activation+0.803
latest
Token latest
Feature activation+0.459
round
Token round
Feature activation+0.000
of
Token of
Feature activation+0.000
spar
Token spar
Feature activation+0.196
ring
Tokenring
Feature activation+0.000
.
Token.
Feature activation+0.098

INTERVAL 0.000 - 0.254
CONTAINS 99.986%

that
Token that
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
informative
Token informative
Feature activation+0.000
popup
Token popup
Feature activation+0.000
at
Token at
Feature activation+0.000
all
Token all
Feature activation+0.000
(
Token (
Feature activation+0.000
is
Tokenis
Feature activation+0.000
it
Token it
Feature activation+0.000
presence
Token presence
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
paper
Token paper
Feature activation+0.000
;
Token;
Feature activation+0.000
her
Token her
Feature activation+0.000
colors
Token colors
Feature activation+0.000
are
Token are
Feature activation+0.000
bright
Token bright
Feature activation+0.000
,
Token,
Feature activation+0.000
cl
Token cl
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
lacks
Token lacks
Feature activation+0.000
support
Token support
Feature activation+0.000
for
Token for
Feature activation+0.000
SSH
Token SSH
Feature activation+0.000
2
Token2
Feature activation+0.000
servers
Token servers
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
it
Token it
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
drone
Token drone
Feature activation+0.000
equipped
Token equipped
Feature activation+0.000
with
Token with
Feature activation+0.000
a
Token a
Feature activation+0.000
small
Token small
Feature activation+0.000
video
Token video
Feature activation+0.000
camera
Token camera
Feature activation+0.000
takes
Token takes
Feature activation+0.000
flight
Token flight
Feature activation+0.000
them
Token them
Feature activation+0.000
in
Token in
Feature activation+0.000
any
Token any
Feature activation+0.000
direction
Token direction
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Enc
TokenEnc
Feature activation+0.000
ourage
Tokenourage
Feature activation+0.000
acceptance
Token acceptance
Feature activation+0.000
,
Token,
Feature activation+0.000
open
Token open
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 H2.6: (feature 16392)

TOP ACTIVATIONS
MAX = 2.849

hear
Token hear
Feature activation+0.000
the
Token the
Feature activation+0.000
full
Token full
Feature activation+0.000
story
Token story
Feature activation+0.000
,
Token,
Feature activation+0.778
click
Token click
Feature activation+2.849
the
Token the
Feature activation+0.823
audio
Token audio
Feature activation+0.337
labelled
Token labelled
Feature activation+0.143
:
Token:
Feature activation+0.000
Hardcore
Token Hardcore
Feature activation+0.000
multi
Token multi
Feature activation+0.122
channel
Token channel
Feature activation+0.000
fun
Token fun
Feature activation+0.000
nels
Tokennels
Feature activation+0.028
,
Token,
Feature activation+1.018
read
Token read
Feature activation+2.634
the
Token the
Feature activation+1.072
article
Token article
Feature activation+0.299
:
Token:
Feature activation+0.341
Google
Token Google
Feature activation+0.300
Analytics
Token Analytics
Feature activation+0.000
a
Token a
Feature activation+0.181
more
Token more
Feature activation+0.345
effective
Token effective
Feature activation+0.000
manner
Token manner
Feature activation+0.042
,
Token,
Feature activation+1.040
let
Token let
Feature activation+2.152
me
Token me
Feature activation+0.867
show
Token show
Feature activation+0.703
you
Token you
Feature activation+0.302
two
Token two
Feature activation+0.000
versions
Token versions
Feature activation+0.000
).
Token).
Feature activation+0.000
To
Token To
Feature activation+0.000
contact
Token contact
Feature activation+0.053
him
Token him
Feature activation+0.000
,
Token,
Feature activation+0.549
visit
Token visit
Feature activation+2.120
his
Token his
Feature activation+0.456
website
Token website
Feature activation+0.442
at
Token at
Feature activation+0.230
www
Token www
Feature activation+0.000
.
Token.
Feature activation+0.000
er
Token er
Feature activation+0.000
rat
Tokenrat
Feature activation+0.000
um
Tokenum
Feature activation+0.000
,
Token,
Feature activation+0.524
please
Token please
Feature activation+1.164
click
Token click
Feature activation+2.078
here
Token here
Feature activation+0.754
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
K
TokenK
Feature activation+0.000
crowd
Token crowd
Feature activation+0.000
-
Token-
Feature activation+0.000
funding
Tokenfunding
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
,
Token,
Feature activation+0.738
go
Token go
Feature activation+1.854
to
Token to
Feature activation+0.558
:
Token:
Feature activation+0.030
www
Token www
Feature activation+0.000
.
Token.
Feature activation+0.000
ind
Tokenind
Feature activation+0.000
X
TokenX
Feature activation+0.000
within
Token within
Feature activation+0.000
the
Token the
Feature activation+0.000
network
Token network
Feature activation+0.000
,
Token,
Feature activation+0.372
see
Token see
Feature activation+1.797
our
Token our
Feature activation+0.106
FAQ
Token FAQ
Feature activation+0.000
.
Token.
Feature activation+0.000
Alternatively
Token Alternatively
Feature activation+0.000
,
Token,
Feature activation+0.000
To
TokenTo
Feature activation+0.000
give
Token give
Feature activation+0.000
some
Token some
Feature activation+0.000
perspective
Token perspective
Feature activation+0.157
,
Token,
Feature activation+0.806
let
Token let
Feature activation+1.743
âĢ
TokenâĢ
Feature activation+0.364
Ļ
TokenĻ
Feature activation+0.073
s
Tokens
Feature activation+0.000
take
Token take
Feature activation+0.878
the
Token the
Feature activation+0.294
read
Token read
Feature activation+0.351
more
Token more
Feature activation+0.116
about
Token about
Feature activation+0.000
gardening
Token gardening
Feature activation+0.000
,
Token,
Feature activation+0.503
see
Token see
Feature activation+1.675
the
Token the
Feature activation+0.472
Monitor
Token Monitor
Feature activation+0.000
's
Token's
Feature activation+0.000
main
Token main
Feature activation+0.230
gardening
Token gardening
Feature activation+0.000
Model
Token Model
Feature activation+0.000
X
Token X
Feature activation+0.000
holiday
Token holiday
Feature activation+0.383
performance
Token performance
Feature activation+0.144
,
Token,
Feature activation+1.119
just
Token just
Feature activation+1.581
type
Token type
Feature activation+1.351
holidays
Token holidays
Feature activation+0.348
or
Token or
Feature activation+0.133
Model
Token Model
Feature activation+0.000
X
TokenX
Feature activation+0.000
.
Token.
Feature activation+0.000
To
Token To
Feature activation+0.000
do
Token do
Feature activation+0.000
so
Token so
Feature activation+0.008
,
Token,
Feature activation+0.816
start
Token start
Feature activation+1.554
by
Token by
Feature activation+0.324
searching
Token searching
Feature activation+0.358
for
Token for
Feature activation+0.104
a
Token a
Feature activation+0.203
location
Token location
Feature activation+0.005
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
begin
Token begin
Feature activation+0.000
,
Token,
Feature activation+0.107
let
Token let
Feature activation+1.541
âĢ
TokenâĢ
Feature activation+0.163
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
look
Token look
Feature activation+1.050
at
Token at
Feature activation+0.143
Out
Token Out
Feature activation+0.000
line
Tokenline
Feature activation+0.000
from
Token from
Feature activation+0.026
examples
Token examples
Feature activation+0.410
,
Token,
Feature activation+0.741
we
Token we
Feature activation+1.528
first
Token first
Feature activation+0.586
need
Token need
Feature activation+1.026
some
Token some
Feature activation+0.000
examples
Token examples
Feature activation+0.000
.
Token.
Feature activation+0.000
To
TokenTo
Feature activation+0.000
bypass
Token bypass
Feature activation+0.000
this
Token this
Feature activation+0.000
complexity
Token complexity
Feature activation+0.000
,
Token,
Feature activation+1.031
you
Token you
Feature activation+1.500
can
Token can
Feature activation+0.755
use
Token use
Feature activation+1.010
a
Token a
Feature activation+0.520
Make
Token Make
Feature activation+0.000
file
Tokenfile
Feature activation+0.000
understand
Token understand
Feature activation+0.748
this
Token this
Feature activation+0.000
last
Token last
Feature activation+0.000
one
Token one
Feature activation+0.076
,
Token,
Feature activation+0.565
consider
Token consider
Feature activation+1.498
that
Token that
Feature activation+0.478
:
Token:
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
serve
Token serve
Feature activation+0.000
,
Token,
Feature activation+0.305
make
Token make
Feature activation+1.489
a
Token a
Feature activation+1.038
nice
Token nice
Feature activation+0.729
,
Token,
Feature activation+0.627
generous
Token generous
Feature activation+0.844
bed
Token bed
Feature activation+0.263
content
Token content
Feature activation+0.000
on
Token on
Feature activation+0.139
this
Token this
Feature activation+0.296
page
Token page
Feature activation+0.000
,
Token,
Feature activation+0.897
please
Token please
Feature activation+1.488
rotate
Token rotate
Feature activation+0.993
your
Token your
Feature activation+0.450
device
Token device
Feature activation+0.141
to
Token to
Feature activation+0.388
the
Token the
Feature activation+0.000
in
Token in
Feature activation+0.177
your
Token your
Feature activation+0.606
marijuana
Token marijuana
Feature activation+0.043
garden
Token garden
Feature activation+0.138
,
Token,
Feature activation+0.980
adding
Token adding
Feature activation+1.473
gran
Token gran
Feature activation+0.127
ulated
Tokenulated
Feature activation+0.000
products
Token products
Feature activation+0.113
made
Token made
Feature activation+0.200
from
Token from
Feature activation+0.000
operate
Token operate
Feature activation+0.000
the
Token the
Feature activation+0.000
slide
Token slide
Feature activation+0.278
,
Token,
Feature activation+1.032
engineers
Token engineers
Feature activation+0.600
open
Token open
Feature activation+1.388
the
Token the
Feature activation+0.791
aircraft
Token aircraft
Feature activation+0.194
door
Token door
Feature activation+0.231
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.067
!
Token!
Feature activation+0.000
To
Token To
Feature activation+0.000
continue
Token continue
Feature activation+0.193
reading
Token reading
Feature activation+0.427
,
Token,
Feature activation+0.557
we
Token we
Feature activation+1.366
recommend
Token recommend
Feature activation+1.123
our
Token our
Feature activation+0.688
Read
Token Read
Feature activation+0.652
Now
Token Now
Feature activation+0.000
Pay
Token Pay
Feature activation+0.000

Top DFA by src position
MAX = 3.062

On
Token On
Feature activation+0.084
Top resid features:
The
Token The
Feature activation+0.014
Top resid features:
Coast
Token Coast
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.023
Top resid features:
Ċ
TokenĊ
Feature activation+0.021
Top resid features:
To
TokenTo
Feature activation+3.062
Top resid features:
hear
Token hear
Feature activation+0.112
Top resid features:
the
Token the
Feature activation+0.094
Top resid features:
full
Token full
Feature activation+0.072
Top resid features:
story
Token story
Feature activation+0.042
Top resid features:
,
Token,
Feature activation+0.503
Top resid features:
fun
Token fun
Feature activation+0.002
Top resid features:
nels
Tokennels
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.018
Top resid features:
Ċ
TokenĊ
Feature activation+0.021
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
To
TokenTo
Feature activation+2.811
Top resid features:
learn
Token learn
Feature activation+0.113
Top resid features:
more
Token more
Feature activation+0.046
Top resid features:
about
Token about
Feature activation+0.073
Top resid features:
multi
Token multi
Feature activation+0.026
Top resid features:
channel
Token channel
Feature activation+0.011
Top resid features:
programming
Token programming
Feature activation+0.059
Top resid features:
languages
Token languages
Feature activation+0.052
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
To
TokenTo
Feature activation+2.320
Top resid features:
demonstrate
Token demonstrate
Feature activation+0.112
Top resid features:
the
Token the
Feature activation+0.020
Top resid features:
concept
Token concept
Feature activation+0.058
Top resid features:
of
Token of
Feature activation+0.020
Top resid features:
early
Token early
Feature activation+0.015
Top resid features:
R
TokenR
Feature activation+0.002
Top resid features:
-
Token-
Feature activation-0.004
Top resid features:
Ber
TokenBer
Feature activation+0.005
Top resid features:
ks
Tokenks
Feature activation-0.004
Top resid features:
).
Token).
Feature activation+0.008
Top resid features:
To
Token To
Feature activation+2.361
Top resid features:
contact
Token contact
Feature activation+0.172
Top resid features:
him
Token him
Feature activation+0.099
Top resid features:
,
Token,
Feature activation+0.516
Top resid features:
visit
Token visit
Feature activation+0.124
Top resid features:
his
Token his
Feature activation+0.000
Top resid features:
published
Token published
Feature activation+0.014
Top resid features:
for
Token for
Feature activation+0.002
Top resid features:
this
Token this
Feature activation+0.002
Top resid features:
article
Token article
Feature activation-0.008
Top resid features:
.
Token.
Feature activation+0.030
Top resid features:
To
Token To
Feature activation+2.367
Top resid features:
view
Token view
Feature activation+0.117
Top resid features:
the
Token the
Feature activation+0.093
Top resid features:
er
Token er
Feature activation+0.035
Top resid features:
rat
Tokenrat
Feature activation+0.018
Top resid features:
um
Tokenum
Feature activation+0.021
Top resid features:
,
Token,
Feature activation+0.001
Top resid features:
230
Token230
Feature activation-0.000
Top resid features:
).
Token).
Feature activation+0.031
Top resid features:
Ċ
TokenĊ
Feature activation+0.017
Top resid features:
Ċ
TokenĊ
Feature activation+0.019
Top resid features:
To
TokenTo
Feature activation+2.148
Top resid features:
learn
Token learn
Feature activation+0.117
Top resid features:
more
Token more
Feature activation+0.059
Top resid features:
about
Token about
Feature activation+0.069
Top resid features:
the
Token the
Feature activation+0.050
Top resid features:
crowd
Token crowd
Feature activation+0.016
Top resid features:
early
Token early
Feature activation+0.006
Top resid features:
August
Token August
Feature activation+0.024
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
To
TokenTo
Feature activation+2.280
Top resid features:
learn
Token learn
Feature activation+0.077
Top resid features:
more
Token more
Feature activation+0.014
Top resid features:
about
Token about
Feature activation+0.017
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
technical
Token technical
Feature activation+0.016
Top resid features:
Fl
Token Fl
Feature activation+0.007
Top resid features:
ux
Tokenux
Feature activation-0.001
Top resid features:
?
Token?
Feature activation+0.045
Top resid features:
Ċ
TokenĊ
Feature activation+0.032
Top resid features:
Ċ
TokenĊ
Feature activation+0.026
Top resid features:
To
TokenTo
Feature activation+1.725
Top resid features:
give
Token give
Feature activation+0.129
Top resid features:
some
Token some
Feature activation+0.053
Top resid features:
perspective
Token perspective
Feature activation+0.140
Top resid features:
,
Token,
Feature activation+0.509
Top resid features:
let
Token let
Feature activation+0.145
Top resid features:
Ċ
TokenĊ
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.010
Top resid features:
-----
Token-----
Feature activation+0.038
Top resid features:
Ċ
TokenĊ
Feature activation+0.012
Top resid features:
Ċ
TokenĊ
Feature activation+0.012
Top resid features:
To
TokenTo
Feature activation+1.995
Top resid features:
read
Token read
Feature activation+0.069
Top resid features:
more
Token more
Feature activation+0.080
Top resid features:
about
Token about
Feature activation+0.062
Top resid features:
gardening
Token gardening
Feature activation+0.013
Top resid features:
,
Token,
Feature activation+0.525
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.009
Top resid features:
:
Token:
Feature activation+0.023
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
To
TokenTo
Feature activation+1.843
Top resid features:
activate
Token activate
Feature activation+0.135
Top resid features:
the
Token the
Feature activation+0.066
Top resid features:
Model
Token Model
Feature activation+0.043
Top resid features:
X
Token X
Feature activation+0.056
Top resid features:
holiday
Token holiday
Feature activation+0.025
Top resid features:
Wi
Token Wi
Feature activation+0.018
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
Fi
TokenFi
Feature activation+0.015
Top resid features:
connection
Token connection
Feature activation+0.036
Top resid features:
.
Token.
Feature activation+0.040
Top resid features:
To
Token To
Feature activation+1.604
Top resid features:
do
Token do
Feature activation+0.162
Top resid features:
so
Token so
Feature activation+0.145
Top resid features:
,
Token,
Feature activation+0.477
Top resid features:
start
Token start
Feature activation+0.128
Top resid features:
by
Token by
Feature activation+0.000
Top resid features:
any
Token any
Feature activation+0.009
Top resid features:
marketplace
Token marketplace
Feature activation+0.005
Top resid features:
.
Token.
Feature activation+0.021
Top resid features:
Ċ
TokenĊ
Feature activation+0.019
Top resid features:
Ċ
TokenĊ
Feature activation+0.027
Top resid features:
To
TokenTo
Feature activation+1.775
Top resid features:
begin
Token begin
Feature activation+0.066
Top resid features:
,
Token,
Feature activation+0.583
Top resid features:
let
Token let
Feature activation+0.167
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.015
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.013
Top resid features:
To
TokenTo
Feature activation+1.820
Top resid features:
derive
Token derive
Feature activation+0.111
Top resid features:
a
Token a
Feature activation+0.036
Top resid features:
Sc
Token Sc
Feature activation+0.041
Top resid features:
enario
Tokenenario
Feature activation+0.019
Top resid features:
Out
Token Out
Feature activation+0.039
Top resid features:
docker
Token docker
Feature activation+0.007
Top resid features:
user
Token user
Feature activation+0.005
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation+0.023
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
To
TokenTo
Feature activation+1.804
Top resid features:
bypass
Token bypass
Feature activation+0.100
Top resid features:
this
Token this
Feature activation+0.092
Top resid features:
complexity
Token complexity
Feature activation+0.055
Top resid features:
,
Token,
Feature activation+0.461
Top resid features:
you
Token you
Feature activation+0.121
Top resid features:
(
Token (
Feature activation+0.013
Top resid features:
Di
TokenDi
Feature activation+0.019
Top resid features:
agram
Tokenagram
Feature activation+0.018
Top resid features:
4
Token 4
Feature activation+0.001
Top resid features:
).
Token).
Feature activation+0.023
Top resid features:
To
Token To
Feature activation+1.884
Top resid features:
understand
Token understand
Feature activation+0.201
Top resid features:
this
Token this
Feature activation+0.057
Top resid features:
last
Token last
Feature activation-0.044
Top resid features:
one
Token one
Feature activation+0.023
Top resid features:
,
Token,
Feature activation+0.367
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.024
Top resid features:
Ċ
TokenĊ
Feature activation-0.007
Top resid features:
To
TokenTo
Feature activation+1.759
Top resid features:
serve
Token serve
Feature activation+0.174
Top resid features:
,
Token,
Feature activation+0.642
Top resid features:
make
Token make
Feature activation+0.191
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
nice
Token nice
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
About
TokenAbout
Feature activation+0.030
Top resid features:
And
Token And
Feature activation+0.001
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.017
Top resid features:
To
TokenTo
Feature activation+1.857
Top resid features:
best
Token best
Feature activation+0.052
Top resid features:
view
Token view
Feature activation+0.067
Top resid features:
the
Token the
Feature activation+0.033
Top resid features:
content
Token content
Feature activation+0.046
Top resid features:
on
Token on
Feature activation+0.039
Top resid features:
Cotton
Token Cotton
Feature activation+0.004
Top resid features:
seed
Tokenseed
Feature activation-0.001
Top resid features:
Meal
Token Meal
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.029
Top resid features:
Ċ
TokenĊ
Feature activation+0.021
Top resid features:
To
TokenTo
Feature activation+1.718
Top resid features:
correct
Token correct
Feature activation+0.133
Top resid features:
nitrogen
Token nitrogen
Feature activation+0.050
Top resid features:
deficiencies
Token deficiencies
Feature activation+0.024
Top resid features:
in
Token in
Feature activation+0.020
Top resid features:
your
Token your
Feature activation+0.052
Top resid features:
work
Token work
Feature activation+0.014
Top resid features:
properly
Token properly
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.034
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
Ċ
TokenĊ
Feature activation+0.019
Top resid features:
To
TokenTo
Feature activation+1.607
Top resid features:
operate
Token operate
Feature activation+0.107
Top resid features:
the
Token the
Feature activation+0.077
Top resid features:
slide
Token slide
Feature activation+0.034
Top resid features:
,
Token,
Feature activation+0.268
Top resid features:
engineers
Token engineers
Feature activation+0.135
Top resid features:
have
Token have
Feature activation+0.006
Top resid features:
enjoyed
Token enjoyed
Feature activation+0.008
Top resid features:
your
Token your
Feature activation+0.015
Top resid features:
trial
Token trial
Feature activation+0.011
Top resid features:
!
Token!
Feature activation+0.086
Top resid features:
To
Token To
Feature activation+1.560
Top resid features:
continue
Token continue
Feature activation+0.127
Top resid features:
reading
Token reading
Feature activation+0.078
Top resid features:
,
Token,
Feature activation+0.478
Top resid features:
we
Token we
Feature activation+0.125
Top resid features:
recommend
Token recommend
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.04

Head 2: 0.07

Head 3: 0.07

Head 4: 0.04

Head 5: 0.10

Head 6: 0.32

Head 7: 0.05

Head 8: 0.06

Head 9: 0.05

Head 10: 0.10

Head 11: 0.04

Positive logits

CLICK1.75

hower1.72

illac1.52

Customers1.41

click1.39

please1.38

Click1.38

DIT1.38

sand1.35

onym1.34

instead1.32

omez1.31

ablo1.31

Sym1.29

Browse1.29

fen1.28

PRESS1.28

stals1.27

amazon1.26

docker1.25

Negative logits

pse-1.34

;;;;;;;;;;;;-1.32

harb-1.28

Pol-1.20

drowned-1.19

eternity-1.19

Morning-1.18

handwriting-1.18

undermin-1.17

measles-1.16

Burk-1.15

berra-1.14

instrumental-1.10

iyah-1.10

CCC-1.10

electorate-1.09

ioch-1.09

emia-1.09

millenn-1.08

unaccount-1.07

INTERVAL 2.564 - 2.849
CONTAINS 0.000%

hear
Token hear
Feature activation+0.000
the
Token the
Feature activation+0.000
full
Token full
Feature activation+0.000
story
Token story
Feature activation+0.000
,
Token,
Feature activation+0.778
click
Token click
Feature activation+2.849
the
Token the
Feature activation+0.823
audio
Token audio
Feature activation+0.337
labelled
Token labelled
Feature activation+0.143
:
Token:
Feature activation+0.000
Hardcore
Token Hardcore
Feature activation+0.000
multi
Token multi
Feature activation+0.122
channel
Token channel
Feature activation+0.000
fun
Token fun
Feature activation+0.000
nels
Tokennels
Feature activation+0.028
,
Token,
Feature activation+1.018
read
Token read
Feature activation+2.634
the
Token the
Feature activation+1.072
article
Token article
Feature activation+0.299
:
Token:
Feature activation+0.341
Google
Token Google
Feature activation+0.300
Analytics
Token Analytics
Feature activation+0.000

INTERVAL 2.279 - 2.564
CONTAINS 0.000%

INTERVAL 1.994 - 2.279
CONTAINS 0.000%

er
Token er
Feature activation+0.000
rat
Tokenrat
Feature activation+0.000
um
Tokenum
Feature activation+0.000
,
Token,
Feature activation+0.524
please
Token please
Feature activation+1.164
click
Token click
Feature activation+2.078
here
Token here
Feature activation+0.754
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
K
TokenK
Feature activation+0.000
).
Token).
Feature activation+0.000
To
Token To
Feature activation+0.000
contact
Token contact
Feature activation+0.053
him
Token him
Feature activation+0.000
,
Token,
Feature activation+0.549
visit
Token visit
Feature activation+2.120
his
Token his
Feature activation+0.456
website
Token website
Feature activation+0.442
at
Token at
Feature activation+0.230
www
Token www
Feature activation+0.000
.
Token.
Feature activation+0.000
a
Token a
Feature activation+0.181
more
Token more
Feature activation+0.345
effective
Token effective
Feature activation+0.000
manner
Token manner
Feature activation+0.042
,
Token,
Feature activation+1.040
let
Token let
Feature activation+2.152
me
Token me
Feature activation+0.867
show
Token show
Feature activation+0.703
you
Token you
Feature activation+0.302
two
Token two
Feature activation+0.000
versions
Token versions
Feature activation+0.000

INTERVAL 1.709 - 1.994
CONTAINS 0.000%

X
TokenX
Feature activation+0.000
within
Token within
Feature activation+0.000
the
Token the
Feature activation+0.000
network
Token network
Feature activation+0.000
,
Token,
Feature activation+0.372
see
Token see
Feature activation+1.797
our
Token our
Feature activation+0.106
FAQ
Token FAQ
Feature activation+0.000
.
Token.
Feature activation+0.000
Alternatively
Token Alternatively
Feature activation+0.000
,
Token,
Feature activation+0.000
To
TokenTo
Feature activation+0.000
give
Token give
Feature activation+0.000
some
Token some
Feature activation+0.000
perspective
Token perspective
Feature activation+0.157
,
Token,
Feature activation+0.806
let
Token let
Feature activation+1.743
âĢ
TokenâĢ
Feature activation+0.364
Ļ
TokenĻ
Feature activation+0.073
s
Tokens
Feature activation+0.000
take
Token take
Feature activation+0.878
the
Token the
Feature activation+0.294
crowd
Token crowd
Feature activation+0.000
-
Token-
Feature activation+0.000
funding
Tokenfunding
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
,
Token,
Feature activation+0.738
go
Token go
Feature activation+1.854
to
Token to
Feature activation+0.558
:
Token:
Feature activation+0.030
www
Token www
Feature activation+0.000
.
Token.
Feature activation+0.000
ind
Tokenind
Feature activation+0.000

INTERVAL 1.424 - 1.709
CONTAINS 0.001%

.
Token.
Feature activation+0.000
To
Token To
Feature activation+0.000
do
Token do
Feature activation+0.000
so
Token so
Feature activation+0.008
,
Token,
Feature activation+0.816
start
Token start
Feature activation+1.554
by
Token by
Feature activation+0.324
searching
Token searching
Feature activation+0.358
for
Token for
Feature activation+0.104
a
Token a
Feature activation+0.203
location
Token location
Feature activation+0.005
Model
Token Model
Feature activation+0.000
X
Token X
Feature activation+0.000
holiday
Token holiday
Feature activation+0.383
performance
Token performance
Feature activation+0.144
,
Token,
Feature activation+1.119
just
Token just
Feature activation+1.581
type
Token type
Feature activation+1.351
holidays
Token holidays
Feature activation+0.348
or
Token or
Feature activation+0.133
Model
Token Model
Feature activation+0.000
X
TokenX
Feature activation+0.000
read
Token read
Feature activation+0.351
more
Token more
Feature activation+0.116
about
Token about
Feature activation+0.000
gardening
Token gardening
Feature activation+0.000
,
Token,
Feature activation+0.503
see
Token see
Feature activation+1.675
the
Token the
Feature activation+0.472
Monitor
Token Monitor
Feature activation+0.000
's
Token's
Feature activation+0.000
main
Token main
Feature activation+0.230
gardening
Token gardening
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
serve
Token serve
Feature activation+0.000
,
Token,
Feature activation+0.305
make
Token make
Feature activation+1.489
a
Token a
Feature activation+1.038
nice
Token nice
Feature activation+0.729
,
Token,
Feature activation+0.627
generous
Token generous
Feature activation+0.844
bed
Token bed
Feature activation+0.263
Out
Token Out
Feature activation+0.000
line
Tokenline
Feature activation+0.000
from
Token from
Feature activation+0.026
examples
Token examples
Feature activation+0.410
,
Token,
Feature activation+0.741
we
Token we
Feature activation+1.528
first
Token first
Feature activation+0.586
need
Token need
Feature activation+1.026
some
Token some
Feature activation+0.000
examples
Token examples
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 1.139 - 1.424
CONTAINS 0.001%

!
Token!
Feature activation+0.000
To
Token To
Feature activation+0.000
continue
Token continue
Feature activation+0.193
reading
Token reading
Feature activation+0.427
,
Token,
Feature activation+0.557
we
Token we
Feature activation+1.366
recommend
Token recommend
Feature activation+1.123
our
Token our
Feature activation+0.688
Read
Token Read
Feature activation+0.652
Now
Token Now
Feature activation+0.000
Pay
Token Pay
Feature activation+0.000
fuel
Token fuel
Feature activation+0.000
these
Token these
Feature activation+0.000
things
Token things
Feature activation+0.000
,
Token,
Feature activation+0.750
you
Token you
Feature activation+1.157
need
Token need
Feature activation+1.260
good
Token good
Feature activation+0.162
sleep
Token sleep
Feature activation+0.068
and
Token and
Feature activation+0.000
water
Token water
Feature activation+0.000
,
Token,
Feature activation+0.148
see
Token see
Feature activation+0.207
the
Token the
Feature activation+0.000
latest
Token latest
Feature activation+0.000
releases
Token releases
Feature activation+0.000
,
Token,
Feature activation+0.700
you
Token you
Feature activation+1.149
can
Token can
Feature activation+0.468
either
Token either
Feature activation+0.224
filter
Token filter
Feature activation+0.020
the
Token the
Feature activation+0.000
table
Token table
Feature activation+0.000
X
Token X
Feature activation+0.000
holiday
Token holiday
Feature activation+0.383
performance
Token performance
Feature activation+0.144
,
Token,
Feature activation+1.119
just
Token just
Feature activation+1.581
type
Token type
Feature activation+1.351
holidays
Token holidays
Feature activation+0.348
or
Token or
Feature activation+0.133
Model
Token Model
Feature activation+0.000
X
TokenX
Feature activation+0.000
mas
Tokenmas
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
complete
Token complete
Feature activation+0.325
the
Token the
Feature activation+0.085
project
Token project
Feature activation+0.062
,
Token,
Feature activation+1.275
P
Token P
Feature activation+0.527
edia
Tokenedia
Feature activation+0.219
Press
TokenPress
Feature activation+0.319
is
Token is
Feature activation+0.833
trying
Token trying
Feature activation+0.746

INTERVAL 0.855 - 1.139
CONTAINS 0.005%

in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.181
more
Token more
Feature activation+0.345
effective
Token effective
Feature activation+0.000
manner
Token manner
Feature activation+0.042
,
Token,
Feature activation+1.040
let
Token let
Feature activation+2.152
me
Token me
Feature activation+0.867
show
Token show
Feature activation+0.703
you
Token you
Feature activation+0.302
two
Token two
Feature activation+0.000
the
Token the
Feature activation+0.082
Model
Token Model
Feature activation+0.000
X
Token X
Feature activation+0.000
holiday
Token holiday
Feature activation+0.383
performance
Token performance
Feature activation+0.144
,
Token,
Feature activation+1.119
just
Token just
Feature activation+1.581
type
Token type
Feature activation+1.351
holidays
Token holidays
Feature activation+0.348
or
Token or
Feature activation+0.133
Model
Token Model
Feature activation+0.000
more
Token more
Feature activation+0.345
effective
Token effective
Feature activation+0.000
manner
Token manner
Feature activation+0.042
,
Token,
Feature activation+1.040
let
Token let
Feature activation+2.152
me
Token me
Feature activation+0.867
show
Token show
Feature activation+0.703
you
Token you
Feature activation+0.302
two
Token two
Feature activation+0.000
versions
Token versions
Feature activation+0.000
of
Token of
Feature activation+0.000
things
Token things
Feature activation+0.000
easier
Token easier
Feature activation+0.822
for
Token for
Feature activation+0.000
prospective
Token prospective
Feature activation+0.029
users
Token users
Feature activation+0.000
,
Token,
Feature activation+1.130
the
Token the
Feature activation+1.031
eve
Token eve
Feature activation+0.279
rest
Tokenrest
Feature activation+0.055
script
Token script
Feature activation+0.374
performs
Token performs
Feature activation+0.377
channel
Token channel
Feature activation+0.000
fun
Token fun
Feature activation+0.000
nels
Tokennels
Feature activation+0.028
,
Token,
Feature activation+1.018
read
Token read
Feature activation+2.634
the
Token the
Feature activation+1.072
article
Token article
Feature activation+0.299
:
Token:
Feature activation+0.341
Google
Token Google
Feature activation+0.300
Analytics
Token Analytics
Feature activation+0.000
Attribution
Token Attribution
Feature activation+0.000

INTERVAL 0.570 - 0.855
CONTAINS 0.011%

enario
Tokenenario
Feature activation+0.000
Out
Token Out
Feature activation+0.000
line
Tokenline
Feature activation+0.000
from
Token from
Feature activation+0.026
examples
Token examples
Feature activation+0.410
,
Token,
Feature activation+0.741
we
Token we
Feature activation+1.528
first
Token first
Feature activation+0.586
need
Token need
Feature activation+1.026
some
Token some
Feature activation+0.000
examples
Token examples
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
clear
Token clear
Feature activation+0.269
,
Token,
Feature activation+0.640
given
Token given
Feature activation+0.792
the
Token the
Feature activation+0.534
way
Token way
Feature activation+0.575
the
Token the
Feature activation+0.436
poll
Token poll
Feature activation+0.000
was
Token was
Feature activation+0.299
.
Token.
Feature activation+0.000
To
Token To
Feature activation+0.000
keep
Token keep
Feature activation+0.000
the
Token the
Feature activation+0.000
vision
Token vision
Feature activation+0.000
,
Token,
Feature activation+0.658
the
Token the
Feature activation+0.554
o
Token o
Feature activation+0.197
phthal
Tokenphthal
Feature activation+0.000
m
Tokenm
Feature activation+0.000
ologist
Tokenologist
Feature activation+0.094
To
Token To
Feature activation+0.000
an
Token an
Feature activation+0.000
extent
Token extent
Feature activation+0.000
,
Token,
Feature activation+0.933
the
Token the
Feature activation+1.039
analysis
Token analysis
Feature activation+0.576
will
Token will
Feature activation+0.622
illustrate
Token illustrate
Feature activation+0.998
why
Token why
Feature activation+0.684
it
Token it
Feature activation+0.105
has
Token has
Feature activation+0.000
set
Token set
Feature activation+0.548
of
Token of
Feature activation+0.205
e
Token e
Feature activation+0.000
igen
Tokenigen
Feature activation+0.000
faces
Tokenfaces
Feature activation+0.000
,
Token,
Feature activation+0.673
one
Token one
Feature activation+0.966
must
Token must
Feature activation+1.015
:
Token:
Feature activation+0.029
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 0.285 - 0.570
CONTAINS 0.023%

To
Token To
Feature activation+0.000
do
Token do
Feature activation+0.000
so
Token so
Feature activation+0.008
,
Token,
Feature activation+0.816
start
Token start
Feature activation+1.554
by
Token by
Feature activation+0.324
searching
Token searching
Feature activation+0.358
for
Token for
Feature activation+0.104
a
Token a
Feature activation+0.203
location
Token location
Feature activation+0.005
,
Token,
Feature activation+0.569
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
add
Token add
Feature activation+0.532
insult
Token insult
Feature activation+0.000
to
Token to
Feature activation+0.483
injury
Token injury
Feature activation+0.000
the
Token the
Feature activation+0.442
Boy
Token Boy
Feature activation+0.000
nton
Tokennton
Feature activation+0.000
Beach
Token Beach
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
Hogan
Token Hogan
Feature activation+0.000
,
Token,
Feature activation+0.000
this
Token this
Feature activation+0.366
race
Token race
Feature activation+0.024
looks
Token looks
Feature activation+0.715
a
Token a
Feature activation+0.352
lot
Token lot
Feature activation+0.000
like
Token like
Feature activation+0.000
truly
Token truly
Feature activation+0.000
understand
Token understand
Feature activation+0.806
it
Token it
Feature activation+0.000
,
Token,
Feature activation+0.649
I
Token I
Feature activation+0.856
had
Token had
Feature activation+0.402
to
Token to
Feature activation+0.046
build
Token build
Feature activation+0.000
it
Token it
Feature activation+0.000
from
Token from
Feature activation+0.000
scratch
Token scratch
Feature activation+0.000
studios
Token studios
Feature activation+0.080
have
Token have
Feature activation+0.871
been
Token been
Feature activation+0.332
willing
Token willing
Feature activation+0.202
to
Token to
Feature activation+0.000
add
Token add
Feature activation+0.396
Chinese
Token Chinese
Feature activation+0.000
actors
Token actors
Feature activation+0.000
,
Token,
Feature activation+0.230
locations
Token locations
Feature activation+0.000
and
Token and
Feature activation+0.000

INTERVAL 0.000 - 0.285
CONTAINS 99.957%

a
Token a
Feature activation+0.000
key
Token key
Feature activation+0.000
role
Token role
Feature activation+0.000
in
Token in
Feature activation+0.000
bringing
Token bringing
Feature activation+0.000
down
Token down
Feature activation+0.000
the
Token the
Feature activation+0.000
shield
Token shield
Feature activation+0.000
around
Token around
Feature activation+0.000
the
Token the
Feature activation+0.000
second
Token second
Feature activation+0.000
thousands
Token thousands
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
genetics
Token genetics
Feature activation+0.000
of
Token of
Feature activation+0.000
cognitive
Token cognitive
Feature activation+0.000
performance
Token performance
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
team
Token team
Feature activation+0.000
on
Token on
Feature activation+0.000
defeating
Token defeating
Feature activation+0.000
Sweden
Token Sweden
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
World
Token World
Feature activation+0.000
Championships
Token Championships
Feature activation+0.000
!
Token!
Feature activation+0.000
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
factor
Token factor
Feature activation+0.000
in
Token in
Feature activation+0.000
your
Token your
Feature activation+0.000
time
Token time
Feature activation+0.000
-
Token-
Feature activation+0.000
to
Tokento
Feature activation+0.000
-
Token-
Feature activation+0.000
sleep
Tokensleep
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
often
Token often
Feature activation+0.000
those
Token those
Feature activation+0.000
players
Token players
Feature activation+0.000
,
Token,
Feature activation+0.000
seven
Token seven
Feature activation+0.000
never
Token never
Feature activation+0.000
returned
Token returned
Feature activation+0.000
to
Token to
Feature activation+0.000
play
Token play
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
NBA
Token NBA
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 H2.6: (feature 14381)

TOP ACTIVATIONS
MAX = 2.765

MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
information
Token information
Feature activation+1.925
.
Token.
Feature activation+2.125
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
Ļ
TokenĻ
Feature activation+1.001
s
Tokens
Feature activation+1.015
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714
âĢ
TokenâĢ
Feature activation+0.773
Ļ
TokenĻ
Feature activation+1.001
s
Tokens
Feature activation+1.015
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+1.856
âĢ
TokenâĢ
Feature activation+0.773
Ļ
TokenĻ
Feature activation+1.001
s
Tokens
Feature activation+1.015
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
s
Tokens
Feature activation+1.015
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714
Alabama
Token Alabama
Feature activation+1.614
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
information
Token information
Feature activation+1.925
.
Token.
Feature activation+2.125
Ċ
TokenĊ
Feature activation+1.132
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
information
Token information
Feature activation+1.925
.
Token.
Feature activation+2.125
Ċ
TokenĊ
Feature activation+1.132
Ċ
TokenĊ
Feature activation+0.807
âĢ
TokenâĢ
Feature activation+0.366
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
information
Token information
Feature activation+1.925
.
Token.
Feature activation+2.125
Ċ
TokenĊ
Feature activation+1.132
Ċ
TokenĊ
Feature activation+0.807
âĢ
TokenâĢ
Feature activation+0.366
ľ
Tokenľ
Feature activation+0.000
All
TokenAll
Feature activation+0.000
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
information
Token information
Feature activation+1.925
.
Token.
Feature activation+2.125
Ċ
TokenĊ
Feature activation+1.132
Ċ
TokenĊ
Feature activation+0.807
it
Token it
Feature activation+0.000
will
Token will
Feature activation+0.000
kill
Token kill
Feature activation+0.000
you
Token you
Feature activation+0.000
'
Token'
Feature activation+0.000
MORE
Token MORE
Feature activation+2.093
(
Token (
Feature activation+0.970
R
TokenR
Feature activation+1.100
-
Token-
Feature activation+1.454
Utah
TokenUtah
Feature activation+1.684
)
Token)
Feature activation+1.665
in
Token in
Feature activation+0.000
prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
years
Token years
Feature activation+0.000
in
Token in
Feature activation+0.000
prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714
Alabama
Token Alabama
Feature activation+1.614
voters
Token voters
Feature activation+0.915
and
Token and
Feature activation+1.221
(
Token (
Feature activation+0.970
R
TokenR
Feature activation+1.100
-
Token-
Feature activation+1.454
Utah
TokenUtah
Feature activation+1.684
)
Token)
Feature activation+1.665
are
Token are
Feature activation+2.014
among
Token among
Feature activation+1.697
the
Token the
Feature activation+1.691
key
Token key
Feature activation+1.941
GOP
Token GOP
Feature activation+1.791
voices
Token voices
Feature activation+1.302
than
Token than
Feature activation+0.000
10
Token 10
Feature activation+0.000
years
Token years
Feature activation+0.000
in
Token in
Feature activation+0.000
prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714
Alabama
Token Alabama
Feature activation+1.614
voters
Token voters
Feature activation+0.915
and
Token and
Feature activation+1.221
what
Token what
Feature activation+1.532
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714
Alabama
Token Alabama
Feature activation+1.614
voters
Token voters
Feature activation+0.915

Top DFA by src position
MAX = 3.546

than
Token than
Feature activation+0.050
Top resid features:
10
Token 10
Feature activation+0.039
Top resid features:
years
Token years
Feature activation+0.093
Top resid features:
in
Token in
Feature activation+0.070
Top resid features:
prison
Token prison
Feature activation+0.126
Top resid features:
MORE
Token MORE
Feature activation+3.490
Top resid features:
for
Token for
Feature activation+0.211
Top resid features:
a
Token a
Feature activation+0.198
Top resid features:
tweet
Token tweet
Feature activation+0.396
Top resid features:
warning
Token warning
Feature activation+0.623
Top resid features:
former
Token former
Feature activation+0.202
Top resid features:
than
Token than
Feature activation+0.070
Top resid features:
10
Token 10
Feature activation+0.040
Top resid features:
years
Token years
Feature activation+0.113
Top resid features:
in
Token in
Feature activation+0.099
Top resid features:
prison
Token prison
Feature activation+0.143
Top resid features:
MORE
Token MORE
Feature activation+3.270
Top resid features:
for
Token for
Feature activation+0.283
Top resid features:
a
Token a
Feature activation+0.288
Top resid features:
tweet
Token tweet
Feature activation+0.434
Top resid features:
warning
Token warning
Feature activation+0.386
Top resid features:
former
Token former
Feature activation+0.000
Top resid features:
than
Token than
Feature activation+0.034
Top resid features:
10
Token 10
Feature activation+0.028
Top resid features:
years
Token years
Feature activation+0.048
Top resid features:
in
Token in
Feature activation+0.051
Top resid features:
prison
Token prison
Feature activation+0.061
Top resid features:
MORE
Token MORE
Feature activation+3.345
Top resid features:
for
Token for
Feature activation+0.168
Top resid features:
a
Token a
Feature activation+0.119
Top resid features:
tweet
Token tweet
Feature activation+0.193
Top resid features:
warning
Token warning
Feature activation+0.265
Top resid features:
former
Token former
Feature activation+0.072
Top resid features:
than
Token than
Feature activation+0.038
Top resid features:
10
Token 10
Feature activation+0.028
Top resid features:
years
Token years
Feature activation+0.070
Top resid features:
in
Token in
Feature activation+0.061
Top resid features:
prison
Token prison
Feature activation+0.099
Top resid features:
MORE
Token MORE
Feature activation+3.546
Top resid features:
for
Token for
Feature activation+0.182
Top resid features:
a
Token a
Feature activation+0.122
Top resid features:
tweet
Token tweet
Feature activation+0.308
Top resid features:
warning
Token warning
Feature activation+0.394
Top resid features:
former
Token former
Feature activation+0.122
Top resid features:
than
Token than
Feature activation+0.044
Top resid features:
10
Token 10
Feature activation+0.035
Top resid features:
years
Token years
Feature activation+0.092
Top resid features:
in
Token in
Feature activation+0.066
Top resid features:
prison
Token prison
Feature activation+0.129
Top resid features:
MORE
Token MORE
Feature activation+3.085
Top resid features:
âĢ
TokenâĢ
Feature activation+0.150
Top resid features:
Ļ
TokenĻ
Feature activation+0.155
Top resid features:
s
Tokens
Feature activation+0.129
Top resid features:
wall
Token wall
Feature activation+0.241
Top resid features:
on
Token on
Feature activation+0.419
Top resid features:
than
Token than
Feature activation+0.052
Top resid features:
10
Token 10
Feature activation+0.041
Top resid features:
years
Token years
Feature activation+0.104
Top resid features:
in
Token in
Feature activation+0.071
Top resid features:
prison
Token prison
Feature activation+0.145
Top resid features:
MORE
Token MORE
Feature activation+2.854
Top resid features:
âĢ
TokenâĢ
Feature activation+0.210
Top resid features:
Ļ
TokenĻ
Feature activation+0.186
Top resid features:
s
Tokens
Feature activation+0.178
Top resid features:
wall
Token wall
Feature activation+0.290
Top resid features:
on
Token on
Feature activation+0.537
Top resid features:
than
Token than
Feature activation+0.079
Top resid features:
10
Token 10
Feature activation+0.052
Top resid features:
years
Token years
Feature activation+0.115
Top resid features:
in
Token in
Feature activation+0.139
Top resid features:
prison
Token prison
Feature activation+0.166
Top resid features:
MORE
Token MORE
Feature activation+2.846
Top resid features:
âĢ
TokenâĢ
Feature activation+0.252
Top resid features:
Ļ
TokenĻ
Feature activation+0.214
Top resid features:
s
Tokens
Feature activation+0.570
Top resid features:
wall
Token wall
Feature activation+0.261
Top resid features:
on
Token on
Feature activation+0.000
Top resid features:
than
Token than
Feature activation+0.041
Top resid features:
10
Token 10
Feature activation+0.028
Top resid features:
years
Token years
Feature activation+0.072
Top resid features:
in
Token in
Feature activation+0.066
Top resid features:
prison
Token prison
Feature activation+0.112
Top resid features:
MORE
Token MORE
Feature activation+3.108
Top resid features:
âĢ
TokenâĢ
Feature activation+0.135
Top resid features:
Ļ
TokenĻ
Feature activation+0.114
Top resid features:
s
Tokens
Feature activation+0.092
Top resid features:
wall
Token wall
Feature activation+0.143
Top resid features:
on
Token on
Feature activation+0.421
Top resid features:
than
Token than
Feature activation+0.032
Top resid features:
10
Token 10
Feature activation+0.020
Top resid features:
years
Token years
Feature activation+0.042
Top resid features:
in
Token in
Feature activation+0.044
Top resid features:
prison
Token prison
Feature activation+0.057
Top resid features:
MORE
Token MORE
Feature activation+3.100
Top resid features:
for
Token for
Feature activation+0.142
Top resid features:
a
Token a
Feature activation+0.099
Top resid features:
tweet
Token tweet
Feature activation+0.159
Top resid features:
warning
Token warning
Feature activation+0.261
Top resid features:
former
Token former
Feature activation+0.036
Top resid features:
than
Token than
Feature activation+0.022
Top resid features:
10
Token 10
Feature activation+0.011
Top resid features:
years
Token years
Feature activation+0.031
Top resid features:
in
Token in
Feature activation+0.029
Top resid features:
prison
Token prison
Feature activation+0.037
Top resid features:
MORE
Token MORE
Feature activation+2.931
Top resid features:
for
Token for
Feature activation+0.089
Top resid features:
a
Token a
Feature activation+0.047
Top resid features:
tweet
Token tweet
Feature activation+0.117
Top resid features:
warning
Token warning
Feature activation+0.164
Top resid features:
former
Token former
Feature activation+0.050
Top resid features:
than
Token than
Feature activation+0.021
Top resid features:
10
Token 10
Feature activation+0.024
Top resid features:
years
Token years
Feature activation+0.046
Top resid features:
in
Token in
Feature activation+0.021
Top resid features:
prison
Token prison
Feature activation+0.032
Top resid features:
MORE
Token MORE
Feature activation+2.601
Top resid features:
for
Token for
Feature activation+0.076
Top resid features:
a
Token a
Feature activation+0.039
Top resid features:
tweet
Token tweet
Feature activation+0.122
Top resid features:
warning
Token warning
Feature activation+0.135
Top resid features:
former
Token former
Feature activation+0.054
Top resid features:
than
Token than
Feature activation+0.030
Top resid features:
10
Token 10
Feature activation+0.024
Top resid features:
years
Token years
Feature activation+0.047
Top resid features:
in
Token in
Feature activation+0.040
Top resid features:
prison
Token prison
Feature activation+0.049
Top resid features:
MORE
Token MORE
Feature activation+2.769
Top resid features:
for
Token for
Feature activation+0.134
Top resid features:
a
Token a
Feature activation+0.074
Top resid features:
tweet
Token tweet
Feature activation+0.162
Top resid features:
warning
Token warning
Feature activation+0.206
Top resid features:
former
Token former
Feature activation+0.080
Top resid features:
it
Token it
Feature activation+0.133
Top resid features:
will
Token will
Feature activation+0.139
Top resid features:
kill
Token kill
Feature activation+0.152
Top resid features:
you
Token you
Feature activation+0.208
Top resid features:
'
Token'
Feature activation+0.155
Top resid features:
MORE
Token MORE
Feature activation+3.154
Top resid features:
(
Token (
Feature activation+0.000
Top resid features:
R
TokenR
Feature activation+0.000
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
Utah
TokenUtah
Feature activation+0.000
Top resid features:
)
Token)
Feature activation+0.000
Top resid features:
than
Token than
Feature activation+0.072
Top resid features:
10
Token 10
Feature activation+0.047
Top resid features:
years
Token years
Feature activation+0.132
Top resid features:
in
Token in
Feature activation+0.103
Top resid features:
prison
Token prison
Feature activation+0.184
Top resid features:
MORE
Token MORE
Feature activation+3.068
Top resid features:
for
Token for
Feature activation+0.335
Top resid features:
a
Token a
Feature activation+0.308
Top resid features:
tweet
Token tweet
Feature activation+0.399
Top resid features:
warning
Token warning
Feature activation+0.000
Top resid features:
former
Token former
Feature activation+0.000
Top resid features:
than
Token than
Feature activation+0.069
Top resid features:
10
Token 10
Feature activation+0.067
Top resid features:
years
Token years
Feature activation+0.184
Top resid features:
in
Token in
Feature activation+0.129
Top resid features:
prison
Token prison
Feature activation+0.292
Top resid features:
MORE
Token MORE
Feature activation+3.307
Top resid features:
for
Token for
Feature activation+0.289
Top resid features:
a
Token a
Feature activation+0.218
Top resid features:
tweet
Token tweet
Feature activation+0.000
Top resid features:
warning
Token warning
Feature activation+0.000
Top resid features:
former
Token former
Feature activation+0.000
Top resid features:
than
Token than
Feature activation+0.033
Top resid features:
10
Token 10
Feature activation+0.025
Top resid features:
years
Token years
Feature activation+0.058
Top resid features:
in
Token in
Feature activation+0.043
Top resid features:
prison
Token prison
Feature activation+0.063
Top resid features:
MORE
Token MORE
Feature activation+2.928
Top resid features:
âĢ
TokenâĢ
Feature activation+0.121
Top resid features:
Ļ
TokenĻ
Feature activation+0.112
Top resid features:
s
Tokens
Feature activation+0.094
Top resid features:
wall
Token wall
Feature activation+0.106
Top resid features:
on
Token on
Feature activation+0.226
Top resid features:
it
Token it
Feature activation+0.057
Top resid features:
will
Token will
Feature activation+0.109
Top resid features:
kill
Token kill
Feature activation+0.130
Top resid features:
you
Token you
Feature activation+0.097
Top resid features:
'
Token'
Feature activation+0.102
Top resid features:
MORE
Token MORE
Feature activation+3.008
Top resid features:
(
Token (
Feature activation+0.058
Top resid features:
R
TokenR
Feature activation+0.115
Top resid features:
-
Token-
Feature activation+0.067
Top resid features:
Utah
TokenUtah
Feature activation+0.164
Top resid features:
)
Token)
Feature activation+0.369
Top resid features:
than
Token than
Feature activation+0.115
Top resid features:
10
Token 10
Feature activation+0.155
Top resid features:
years
Token years
Feature activation+0.175
Top resid features:
in
Token in
Feature activation+0.169
Top resid features:
prison
Token prison
Feature activation+0.264
Top resid features:
MORE
Token MORE
Feature activation+2.953
Top resid features:
for
Token for
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:
warning
Token warning
Feature activation+0.000
Top resid features:
former
Token former
Feature activation+0.000
Top resid features:
than
Token than
Feature activation+0.031
Top resid features:
10
Token 10
Feature activation+0.020
Top resid features:
years
Token years
Feature activation+0.068
Top resid features:
in
Token in
Feature activation+0.039
Top resid features:
prison
Token prison
Feature activation+0.057
Top resid features:
MORE
Token MORE
Feature activation+2.683
Top resid features:
âĢ
TokenâĢ
Feature activation+0.111
Top resid features:
Ļ
TokenĻ
Feature activation+0.084
Top resid features:
s
Tokens
Feature activation+0.056
Top resid features:
wall
Token wall
Feature activation+0.089
Top resid features:
on
Token on
Feature activation+0.226
Top resid features:
than
Token than
Feature activation+0.038
Top resid features:
10
Token 10
Feature activation+0.035
Top resid features:
years
Token years
Feature activation+0.061
Top resid features:
in
Token in
Feature activation+0.049
Top resid features:
prison
Token prison
Feature activation+0.068
Top resid features:
MORE
Token MORE
Feature activation+2.836
Top resid features:
âĢ
TokenâĢ
Feature activation+0.160
Top resid features:
Ļ
TokenĻ
Feature activation+0.118
Top resid features:
s
Tokens
Feature activation+0.095
Top resid features:
wall
Token wall
Feature activation+0.124
Top resid features:
on
Token on
Feature activation+0.322
Top resid features:

Decoder Weights Distribution

Head 0: 0.05

Head 1: 0.10

Head 2: 0.03

Head 3: 0.03

Head 4: 0.02

Head 5: 0.04

Head 6: 0.32

Head 7: 0.11

Head 8: 0.03

Head 9: 0.04

Head 10: 0.14

Head 11: 0.08

Positive logits

subcommittee1.75

ospons1.66

Subcommittee1.52

congressional1.52

isconsin1.48

Lobby1.47

"))1.42

taxpayer1.41

PAC1.36

ovember1.35

PAC1.34

Caucus1.33

caucus1.33

gov1.32

Vote1.31

Rasmussen1.30

DeVos1.30

lobbying1.30

lobby1.30

clinton1.29

Negative logits

});-1.71

here-1.54

Nietzsche-1.49

})-1.49

               -1.46

});-1.38

...)-1.37

 -1.37

…]-1.34

CLR-1.33

AMY-1.32

  -1.32

heres-1.30

......-1.30

emptiness-1.27

Shard-1.23

Pic-1.22

Mages-1.22

Thief-1.21

};-1.21

INTERVAL 2.489 - 2.765
CONTAINS 0.000%

MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109

INTERVAL 2.212 - 2.489
CONTAINS 0.001%

prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
information
Token information
Feature activation+1.925
.
Token.
Feature activation+2.125
âĢ
TokenâĢ
Feature activation+0.773
Ļ
TokenĻ
Feature activation+1.001
s
Tokens
Feature activation+1.015
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
Ļ
TokenĻ
Feature activation+1.001
s
Tokens
Feature activation+1.015
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714

INTERVAL 1.936 - 2.212
CONTAINS 0.002%

Utah
TokenUtah
Feature activation+1.684
)
Token)
Feature activation+1.665
are
Token are
Feature activation+2.014
among
Token among
Feature activation+1.697
the
Token the
Feature activation+1.691
key
Token key
Feature activation+1.941
GOP
Token GOP
Feature activation+1.791
voices
Token voices
Feature activation+1.302
who
Token who
Feature activation+1.224
have
Token have
Feature activation+1.789
called
Token called
Feature activation+1.795
years
Token years
Feature activation+0.000
in
Token in
Feature activation+0.000
prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Director
Token Director
Feature activation+1.810
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714
Alabama
Token Alabama
Feature activation+1.614
voters
Token voters
Feature activation+0.915
and
Token and
Feature activation+1.221
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714
Alabama
Token Alabama
Feature activation+1.614
voters
Token voters
Feature activation+0.915
prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+1.856
âĢ
TokenâĢ
Feature activation+0.773
Ļ
TokenĻ
Feature activation+1.001
s
Tokens
Feature activation+1.015
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954

INTERVAL 1.659 - 1.936
CONTAINS 0.001%

Director
Token Director
Feature activation+1.810
James
Token James
Feature activation+2.442
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
information
Token information
Feature activation+1.925
.
Token.
Feature activation+2.125
Ċ
TokenĊ
Feature activation+1.132
Ċ
TokenĊ
Feature activation+0.807
âĢ
TokenâĢ
Feature activation+0.366
ľ
Tokenľ
Feature activation+0.000
MORE
Token MORE
Feature activation+2.093
(
Token (
Feature activation+0.970
R
TokenR
Feature activation+1.100
-
Token-
Feature activation+1.454
Utah
TokenUtah
Feature activation+1.684
)
Token)
Feature activation+1.665
are
Token are
Feature activation+2.014
among
Token among
Feature activation+1.697
the
Token the
Feature activation+1.691
key
Token key
Feature activation+1.941
GOP
Token GOP
Feature activation+1.791
R
TokenR
Feature activation+1.100
-
Token-
Feature activation+1.454
Utah
TokenUtah
Feature activation+1.684
)
Token)
Feature activation+1.665
are
Token are
Feature activation+2.014
among
Token among
Feature activation+1.697
the
Token the
Feature activation+1.691
key
Token key
Feature activation+1.941
GOP
Token GOP
Feature activation+1.791
voices
Token voices
Feature activation+1.302
who
Token who
Feature activation+1.224
key
Token key
Feature activation+1.941
GOP
Token GOP
Feature activation+1.791
voices
Token voices
Feature activation+1.302
who
Token who
Feature activation+1.224
have
Token have
Feature activation+1.789
called
Token called
Feature activation+1.795
for
Token for
Feature activation+1.338
rec
Token rec
Feature activation+1.319
usal
Tokenusal
Feature activation+1.026
.
Token.
Feature activation+1.655
Ċ
TokenĊ
Feature activation+0.271
)
Token)
Feature activation+1.665
are
Token are
Feature activation+2.014
among
Token among
Feature activation+1.697
the
Token the
Feature activation+1.691
key
Token key
Feature activation+1.941
GOP
Token GOP
Feature activation+1.791
voices
Token voices
Feature activation+1.302
who
Token who
Feature activation+1.224
have
Token have
Feature activation+1.789
called
Token called
Feature activation+1.795
for
Token for
Feature activation+1.338

INTERVAL 1.383 - 1.659
CONTAINS 0.001%

blocking
Token blocking
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
's
Token's
Feature activation+0.000
emergency
Token emergency
Feature activation+0.000
declaration
Token declaration
Feature activation+0.000
MORE
Token MORE
Feature activation+1.481
(
Token (
Feature activation+0.236
D
TokenD
Feature activation+0.339
-
Token-
Feature activation+0.811
Mass
TokenMass
Feature activation+0.750
.)
Token.)
Feature activation+0.695
MORE
Token MORE
Feature activation+1.856
âĢ
TokenâĢ
Feature activation+0.773
Ļ
TokenĻ
Feature activation+1.001
s
Tokens
Feature activation+1.015
wall
Token wall
Feature activation+2.176
on
Token on
Feature activation+1.607
the
Token the
Feature activation+2.284
Mexican
Token Mexican
Feature activation+2.419
border
Token border
Feature activation+2.171
,
Token,
Feature activation+1.954
the
Token the
Feature activation+2.017
-
Token-
Feature activation+1.081
N
TokenN
Feature activation+0.462
.
Token.
Feature activation+0.993
Y
TokenY
Feature activation+0.919
.)
Token.)
Feature activation+0.903
and
Token and
Feature activation+1.464
Bill
Token Bill
Feature activation+1.527
Nelson
Token Nelson
Feature activation+1.221
Clarence
Token Clarence
Feature activation+1.628
(
Token (
Feature activation+1.199
Bill
TokenBill
Feature activation+0.647
N
TokenN
Feature activation+0.462
.
Token.
Feature activation+0.993
Y
TokenY
Feature activation+0.919
.)
Token.)
Feature activation+0.903
and
Token and
Feature activation+1.464
Bill
Token Bill
Feature activation+1.527
Nelson
Token Nelson
Feature activation+1.221
Clarence
Token Clarence
Feature activation+1.628
(
Token (
Feature activation+1.199
Bill
TokenBill
Feature activation+0.647
)
Token)
Feature activation+0.631
Y
TokenY
Feature activation+0.919
.)
Token.)
Feature activation+0.903
and
Token and
Feature activation+1.464
Bill
Token Bill
Feature activation+1.527
Nelson
Token Nelson
Feature activation+1.221
Clarence
Token Clarence
Feature activation+1.628
(
Token (
Feature activation+1.199
Bill
TokenBill
Feature activation+0.647
)
Token)
Feature activation+0.631
William
Token William
Feature activation+0.575
Nelson
Token Nelson
Feature activation+0.372

INTERVAL 1.106 - 1.383
CONTAINS 0.001%

voices
Token voices
Feature activation+1.302
who
Token who
Feature activation+1.224
have
Token have
Feature activation+1.789
called
Token called
Feature activation+1.795
for
Token for
Feature activation+1.338
rec
Token rec
Feature activation+1.319
usal
Tokenusal
Feature activation+1.026
.
Token.
Feature activation+1.655
Ċ
TokenĊ
Feature activation+0.271
Ċ
TokenĊ
Feature activation+0.000
S
TokenS
Feature activation+0.000
10
Token 10
Feature activation+0.000
years
Token years
Feature activation+0.000
in
Token in
Feature activation+0.000
prison
Token prison
Feature activation+0.000
MORE
Token MORE
Feature activation+2.008
for
Token for
Feature activation+1.360
a
Token a
Feature activation+2.070
tweet
Token tweet
Feature activation+2.078
warning
Token warning
Feature activation+2.466
former
Token former
Feature activation+2.765
FBI
Token FBI
Feature activation+2.440
Comey
Token Comey
Feature activation+2.166
about
Token about
Feature activation+2.109
leaking
Token leaking
Feature activation+2.128
information
Token information
Feature activation+1.925
.
Token.
Feature activation+2.125
Ċ
TokenĊ
Feature activation+1.132
Ċ
TokenĊ
Feature activation+0.807
âĢ
TokenâĢ
Feature activation+0.366
ľ
Tokenľ
Feature activation+0.000
All
TokenAll
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+2.017
issues
Token issues
Feature activation+1.956
impacting
Token impacting
Feature activation+1.714
Alabama
Token Alabama
Feature activation+1.614
voters
Token voters
Feature activation+0.915
and
Token and
Feature activation+1.221
what
Token what
Feature activation+1.532
characteristics
Token characteristics
Feature activation+0.595
he
Token he
Feature activation+0.752
thinks
Token thinks
Feature activation+1.086
make
Token make
Feature activation+0.985
(
Token (
Feature activation+1.340
R
TokenR
Feature activation+1.035
-
Token-
Feature activation+0.823
Ohio
TokenOhio
Feature activation+0.781
)
Token)
Feature activation+1.196
forced
Token forced
Feature activation+1.150
resignation
Token resignation
Feature activation+0.985
in
Token in
Feature activation+0.958
2015
Token 2015
Feature activation+1.060
set
Token set
Feature activation+0.619
off
Token off
Feature activation+0.424

INTERVAL 0.830 - 1.106
CONTAINS 0.002%

(
Token (
Feature activation+0.236
D
TokenD
Feature activation+0.339
-
Token-
Feature activation+0.811
Mass
TokenMass
Feature activation+0.750
.)
Token.)
Feature activation+0.695
for
Token for
Feature activation+0.918
staying
Token staying
Feature activation+0.696
neutral
Token neutral
Feature activation+0.972
in
Token in
Feature activation+0.882
the
Token the
Feature activation+1.168
Democratic
Token Democratic
Feature activation+1.202
neutral
Token neutral
Feature activation+0.972
in
Token in
Feature activation+0.882
the
Token the
Feature activation+1.168
Democratic
Token Democratic
Feature activation+1.202
primary
Token primary
Feature activation+0.814
.
Token.
Feature activation+0.911
Ċ
TokenĊ
Feature activation+0.145
Ċ
TokenĊ
Feature activation+0.000
Some
TokenSome
Feature activation+0.000
Sanders
Token Sanders
Feature activation+0.000
fans
Token fans
Feature activation+0.000
2015
Token 2015
Feature activation+1.060
set
Token set
Feature activation+0.619
off
Token off
Feature activation+0.424
a
Token a
Feature activation+0.770
leadership
Token leadership
Feature activation+0.823
scramble
Token scramble
Feature activation+0.902
.
Token.
Feature activation+0.761
McM
Token McM
Feature activation+0.358
or
Tokenor
Feature activation+0.000
ris
Tokenris
Feature activation+0.000
Rodgers
Token Rodgers
Feature activation+0.037
-
Token-
Feature activation+0.811
Mass
TokenMass
Feature activation+0.750
.)
Token.)
Feature activation+0.695
for
Token for
Feature activation+0.918
staying
Token staying
Feature activation+0.696
neutral
Token neutral
Feature activation+0.972
in
Token in
Feature activation+0.882
the
Token the
Feature activation+1.168
Democratic
Token Democratic
Feature activation+1.202
primary
Token primary
Feature activation+0.814
.
Token.
Feature activation+0.911
Ohio
TokenOhio
Feature activation+0.781
)
Token)
Feature activation+1.196
forced
Token forced
Feature activation+1.150
resignation
Token resignation
Feature activation+0.985
in
Token in
Feature activation+0.958
2015
Token 2015
Feature activation+1.060
set
Token set
Feature activation+0.619
off
Token off
Feature activation+0.424
a
Token a
Feature activation+0.770
leadership
Token leadership
Feature activation+0.823
scramble
Token scramble
Feature activation+0.902

INTERVAL 0.553 - 0.830
CONTAINS 0.002%

is
Token is
Feature activation+0.000
doing
Token doing
Feature activation+0.000
right
Token right
Feature activation+0.000
MORE
Token MORE
Feature activation+1.640
(
Token (
Feature activation+0.453
D
TokenD
Feature activation+0.564
-
Token-
Feature activation+1.081
N
TokenN
Feature activation+0.462
.
Token.
Feature activation+0.993
Y
TokenY
Feature activation+0.919
.)
Token.)
Feature activation+0.903
)
Token)
Feature activation+1.196
forced
Token forced
Feature activation+1.150
resignation
Token resignation
Feature activation+0.985
in
Token in
Feature activation+0.958
2015
Token 2015
Feature activation+1.060
set
Token set
Feature activation+0.619
off
Token off
Feature activation+0.424
a
Token a
Feature activation+0.770
leadership
Token leadership
Feature activation+0.823
scramble
Token scramble
Feature activation+0.902
.
Token.
Feature activation+0.761
D
TokenD
Feature activation+0.339
-
Token-
Feature activation+0.811
Mass
TokenMass
Feature activation+0.750
.)
Token.)
Feature activation+0.695
for
Token for
Feature activation+0.918
staying
Token staying
Feature activation+0.696
neutral
Token neutral
Feature activation+0.972
in
Token in
Feature activation+0.882
the
Token the
Feature activation+1.168
Democratic
Token Democratic
Feature activation+1.202
primary
Token primary
Feature activation+0.814
declaration
Token declaration
Feature activation+0.000
MORE
Token MORE
Feature activation+1.481
(
Token (
Feature activation+0.236
D
TokenD
Feature activation+0.339
-
Token-
Feature activation+0.811
Mass
TokenMass
Feature activation+0.750
.)
Token.)
Feature activation+0.695
for
Token for
Feature activation+0.918
staying
Token staying
Feature activation+0.696
neutral
Token neutral
Feature activation+0.972
in
Token in
Feature activation+0.882
Ļ
TokenĻ
Feature activation+0.887
s
Tokens
Feature activation+0.629
(
Token (
Feature activation+1.340
R
TokenR
Feature activation+1.035
-
Token-
Feature activation+0.823
Ohio
TokenOhio
Feature activation+0.781
)
Token)
Feature activation+1.196
forced
Token forced
Feature activation+1.150
resignation
Token resignation
Feature activation+0.985
in
Token in
Feature activation+0.958
2015
Token 2015
Feature activation+1.060

INTERVAL 0.277 - 0.553
CONTAINS 0.001%

forced
Token forced
Feature activation+1.150
resignation
Token resignation
Feature activation+0.985
in
Token in
Feature activation+0.958
2015
Token 2015
Feature activation+1.060
set
Token set
Feature activation+0.619
off
Token off
Feature activation+0.424
a
Token a
Feature activation+0.770
leadership
Token leadership
Feature activation+0.823
scramble
Token scramble
Feature activation+0.902
.
Token.
Feature activation+0.761
McM
Token McM
Feature activation+0.358
Clarence
Token Clarence
Feature activation+1.628
(
Token (
Feature activation+1.199
Bill
TokenBill
Feature activation+0.647
)
Token)
Feature activation+0.631
William
Token William
Feature activation+0.575
Nelson
Token Nelson
Feature activation+0.372
2020
Token2020
Feature activation+0.405
party
Token party
Feature activation+0.403
politics
Token politics
Feature activation+0.482
in
Token in
Feature activation+0.147
Puerto
Token Puerto
Feature activation+0.249
off
Token off
Feature activation+0.424
a
Token a
Feature activation+0.770
leadership
Token leadership
Feature activation+0.823
scramble
Token scramble
Feature activation+0.902
.
Token.
Feature activation+0.761
McM
Token McM
Feature activation+0.358
or
Tokenor
Feature activation+0.000
ris
Tokenris
Feature activation+0.000
Rodgers
Token Rodgers
Feature activation+0.037
,
Token,
Feature activation+0.022
Majority
Token Majority
Feature activation+0.068
Bill
TokenBill
Feature activation+0.647
)
Token)
Feature activation+0.631
William
Token William
Feature activation+0.575
Nelson
Token Nelson
Feature activation+0.372
2020
Token2020
Feature activation+0.405
party
Token party
Feature activation+0.403
politics
Token politics
Feature activation+0.482
in
Token in
Feature activation+0.147
Puerto
Token Puerto
Feature activation+0.249
Rico
Token Rico
Feature activation+0.238
There
Token There
Feature activation+0.155
Trump
Token Trump
Feature activation+0.000
is
Token is
Feature activation+0.000
doing
Token doing
Feature activation+0.000
right
Token right
Feature activation+0.000
MORE
Token MORE
Feature activation+1.640
(
Token (
Feature activation+0.453
D
TokenD
Feature activation+0.564
-
Token-
Feature activation+1.081
N
TokenN
Feature activation+0.462
.
Token.
Feature activation+0.993
Y
TokenY
Feature activation+0.919

INTERVAL 0.000 - 0.277
CONTAINS 99.989%

a
Token a
Feature activation+0.000
statement
Token statement
Feature activation+0.000
.
Token.
Feature activation+0.000
"
Token "
Feature activation+0.000
The
TokenThe
Feature activation+0.000
rel
Token rel
Feature activation+0.000
ocations
Tokenocations
Feature activation+0.000
within
Token within
Feature activation+0.000
the
Token the
Feature activation+0.000
mall
Token mall
Feature activation+0.000
by
Token by
Feature activation+0.000
's
Token's
Feature activation+0.000
Inter
Token Inter
Feature activation+0.000
continental
Tokencontinental
Feature activation+0.000
Hotel
Token Hotel
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
menu
Token menu
Feature activation+0.000
"
Token "
Feature activation+0.000
drawn
Tokendrawn
Feature activation+0.000
directly
Token directly
Feature activation+0.000
from
Token from
Feature activation+0.000
Republic
Token Republic
Feature activation+0.000
of
Token of
Feature activation+0.000
Armenia
Token Armenia
Feature activation+0.000
in
Token in
Feature activation+0.000
1994
Token 1994
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Internet
Token Internet
Feature activation+0.000
Society
Token Society
Feature activation+0.000
NGO
Token NGO
Feature activation+0.000
huge
Token huge
Feature activation+0.000
effort
Token effort
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
here
Token here
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
World
Token World
Feature activation+0.000
Cup
Token Cup
Feature activation+0.000
.
Token .
Feature activation+0.000
,
Token,
Feature activation+0.000
where
Token where
Feature activation+0.000
your
Token your
Feature activation+0.000
environmental
Token environmental
Feature activation+0.000
variables
Token variables
Feature activation+0.000
are
Token are
Feature activation+0.000
shown
Token shown
Feature activation+0.000
.
Token.
Feature activation+0.000
If
Token If
Feature activation+0.000
you
Token you
Feature activation+0.000
do
Token do
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 H2.6: (feature 17116)

TOP ACTIVATIONS
MAX = 2.373

-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
-
Token-
Feature activation+0.891
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
R
TokenR
Feature activation+0.939
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Nobody
TokenNobody
Feature activation+0.000
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
âĢ
TokenâĢ
Feature activation+0.000
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
introduced
Token introduced
Feature activation+1.163
D
TokenD
Feature activation+0.628
-
Token-
Feature activation+0.891
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
warnings
Token warnings
Feature activation+0.000
from
Token from
Feature activation+0.000
new
Token new
Feature activation+0.000
climate
Token climate
Feature activation+0.000
studies
Token studies
Feature activation+0.000
MORE
Token MORE
Feature activation+1.798
(
Token (
Feature activation+0.844
R
TokenR
Feature activation+0.939
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
(
Token (
Feature activation+0.844
R
TokenR
Feature activation+0.939
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
MORE
Token MORE
Feature activation+1.166
(
Token (
Feature activation+0.528
R
TokenR
Feature activation+1.131
-
Token-
Feature activation+1.310
Ari
TokenAri
Feature activation+1.393
z
Tokenz
Feature activation+1.637
.)
Token.)
Feature activation+0.974
in
Token in
Feature activation+1.264
a
Token a
Feature activation+1.587
new
Token new
Feature activation+2.074
interview
Token interview
Feature activation+1.465
studies
Token studies
Feature activation+0.000
MORE
Token MORE
Feature activation+1.798
(
Token (
Feature activation+0.844
R
TokenR
Feature activation+0.939
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
introduced
Token introduced
Feature activation+1.163
in
Token in
Feature activation+0.577
the
Token the
Feature activation+0.664
coming
Token coming
Feature activation+0.716
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
introduced
Token introduced
Feature activation+1.163
in
Token in
Feature activation+0.577
the
Token the
Feature activation+0.664

Top DFA by src position
MAX = 2.356

warnings
Token warnings
Feature activation+0.006
Top resid features:
from
Token from
Feature activation+0.008
Top resid features:
new
Token new
Feature activation+0.014
Top resid features:
climate
Token climate
Feature activation+0.008
Top resid features:
studies
Token studies
Feature activation+0.051
Top resid features:
MORE
Token MORE
Feature activation+1.989
Top resid features:
(
Token (
Feature activation-0.011
Top resid features:
R
TokenR
Feature activation+0.009
Top resid features:
-
Token-
Feature activation+0.037
Top resid features:
Ky
TokenKy
Feature activation+0.186
Top resid features:
.)
Token.)
Feature activation+0.406
Top resid features:
as
Token as
Feature activation-0.004
Top resid features:
long
Token long
Feature activation+0.006
Top resid features:
nomination
Token nomination
Feature activation-0.011
Top resid features:
contest
Token contest
Feature activation-0.020
Top resid features:
looms
Token looms
Feature activation-0.040
Top resid features:
MORE
Token MORE
Feature activation+1.620
Top resid features:
(
Token (
Feature activation-0.069
Top resid features:
D
TokenD
Feature activation+0.009
Top resid features:
-
Token-
Feature activation+0.049
Top resid features:
Fla
TokenFla
Feature activation+0.192
Top resid features:
.)
Token.)
Feature activation+0.262
Top resid features:
warnings
Token warnings
Feature activation+0.016
Top resid features:
from
Token from
Feature activation+0.017
Top resid features:
new
Token new
Feature activation+0.019
Top resid features:
climate
Token climate
Feature activation+0.009
Top resid features:
studies
Token studies
Feature activation+0.052
Top resid features:
MORE
Token MORE
Feature activation+1.741
Top resid features:
(
Token (
Feature activation-0.049
Top resid features:
R
TokenR
Feature activation+0.023
Top resid features:
-
Token-
Feature activation+0.037
Top resid features:
Ky
TokenKy
Feature activation+0.218
Top resid features:
.)
Token.)
Feature activation+0.396
Top resid features:
as
Token as
Feature activation-0.004
Top resid features:
long
Token long
Feature activation+0.004
Top resid features:
nomination
Token nomination
Feature activation-0.009
Top resid features:
contest
Token contest
Feature activation-0.013
Top resid features:
looms
Token looms
Feature activation-0.027
Top resid features:
MORE
Token MORE
Feature activation+1.362
Top resid features:
(
Token (
Feature activation-0.020
Top resid features:
D
TokenD
Feature activation-0.022
Top resid features:
-
Token-
Feature activation+0.017
Top resid features:
Fla
TokenFla
Feature activation+0.016
Top resid features:
.)
Token.)
Feature activation+0.220
Top resid features:
as
Token as
Feature activation-0.007
Top resid features:
long
Token long
Feature activation+0.005
Top resid features:
nomination
Token nomination
Feature activation-0.005
Top resid features:
contest
Token contest
Feature activation-0.020
Top resid features:
looms
Token looms
Feature activation-0.032
Top resid features:
MORE
Token MORE
Feature activation+1.068
Top resid features:
(
Token (
Feature activation-0.012
Top resid features:
D
TokenD
Feature activation-0.012
Top resid features:
-
Token-
Feature activation+0.025
Top resid features:
Fla
TokenFla
Feature activation+0.053
Top resid features:
.)
Token.)
Feature activation+0.233
Top resid features:
warnings
Token warnings
Feature activation+0.005
Top resid features:
from
Token from
Feature activation+0.009
Top resid features:
new
Token new
Feature activation+0.007
Top resid features:
climate
Token climate
Feature activation+0.001
Top resid features:
studies
Token studies
Feature activation+0.022
Top resid features:
MORE
Token MORE
Feature activation+1.400
Top resid features:
(
Token (
Feature activation-0.005
Top resid features:
R
TokenR
Feature activation-0.011
Top resid features:
-
Token-
Feature activation+0.018
Top resid features:
Ky
TokenKy
Feature activation+0.041
Top resid features:
.)
Token.)
Feature activation+0.258
Top resid features:
warnings
Token warnings
Feature activation+0.002
Top resid features:
from
Token from
Feature activation+0.008
Top resid features:
new
Token new
Feature activation+0.010
Top resid features:
climate
Token climate
Feature activation+0.002
Top resid features:
studies
Token studies
Feature activation+0.025
Top resid features:
MORE
Token MORE
Feature activation+1.512
Top resid features:
(
Token (
Feature activation-0.015
Top resid features:
R
TokenR
Feature activation+0.002
Top resid features:
-
Token-
Feature activation+0.022
Top resid features:
Ky
TokenKy
Feature activation+0.052
Top resid features:
.)
Token.)
Feature activation+0.396
Top resid features:
as
Token as
Feature activation-0.005
Top resid features:
long
Token long
Feature activation+0.006
Top resid features:
nomination
Token nomination
Feature activation-0.008
Top resid features:
contest
Token contest
Feature activation-0.013
Top resid features:
looms
Token looms
Feature activation-0.026
Top resid features:
MORE
Token MORE
Feature activation+1.167
Top resid features:
(
Token (
Feature activation-0.041
Top resid features:
D
TokenD
Feature activation-0.012
Top resid features:
-
Token-
Feature activation+0.036
Top resid features:
Fla
TokenFla
Feature activation+0.087
Top resid features:
.)
Token.)
Feature activation+0.239
Top resid features:
as
Token as
Feature activation-0.002
Top resid features:
long
Token long
Feature activation+0.002
Top resid features:
nomination
Token nomination
Feature activation-0.004
Top resid features:
contest
Token contest
Feature activation-0.010
Top resid features:
looms
Token looms
Feature activation-0.017
Top resid features:
MORE
Token MORE
Feature activation+1.393
Top resid features:
(
Token (
Feature activation-0.019
Top resid features:
D
TokenD
Feature activation-0.009
Top resid features:
-
Token-
Feature activation+0.010
Top resid features:
Fla
TokenFla
Feature activation+0.008
Top resid features:
.)
Token.)
Feature activation+0.161
Top resid features:
as
Token as
Feature activation-0.006
Top resid features:
long
Token long
Feature activation+0.006
Top resid features:
nomination
Token nomination
Feature activation-0.007
Top resid features:
contest
Token contest
Feature activation-0.018
Top resid features:
looms
Token looms
Feature activation-0.048
Top resid features:
MORE
Token MORE
Feature activation+1.540
Top resid features:
(
Token (
Feature activation-0.092
Top resid features:
D
TokenD
Feature activation+0.003
Top resid features:
-
Token-
Feature activation+0.053
Top resid features:
Fla
TokenFla
Feature activation+0.180
Top resid features:
.)
Token.)
Feature activation+0.260
Top resid features:
warnings
Token warnings
Feature activation+0.005
Top resid features:
from
Token from
Feature activation+0.006
Top resid features:
new
Token new
Feature activation+0.011
Top resid features:
climate
Token climate
Feature activation+0.004
Top resid features:
studies
Token studies
Feature activation+0.042
Top resid features:
MORE
Token MORE
Feature activation+1.653
Top resid features:
(
Token (
Feature activation-0.017
Top resid features:
R
TokenR
Feature activation+0.002
Top resid features:
-
Token-
Feature activation+0.027
Top resid features:
Ky
TokenKy
Feature activation+0.095
Top resid features:
.)
Token.)
Feature activation+0.382
Top resid features:
warnings
Token warnings
Feature activation+0.077
Top resid features:
from
Token from
Feature activation+0.145
Top resid features:
new
Token new
Feature activation+0.108
Top resid features:
climate
Token climate
Feature activation+0.029
Top resid features:
studies
Token studies
Feature activation+0.126
Top resid features:
MORE
Token MORE
Feature activation+2.111
Top resid features:
(
Token (
Feature activation+0.000
Top resid features:
R
TokenR
Feature activation+0.000
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
Ky
TokenKy
Feature activation+0.000
Top resid features:
.)
Token.)
Feature activation+0.000
Top resid features:
as
Token as
Feature activation-0.002
Top resid features:
long
Token long
Feature activation+0.003
Top resid features:
nomination
Token nomination
Feature activation-0.003
Top resid features:
contest
Token contest
Feature activation-0.010
Top resid features:
looms
Token looms
Feature activation-0.016
Top resid features:
MORE
Token MORE
Feature activation+1.487
Top resid features:
(
Token (
Feature activation-0.018
Top resid features:
D
TokenD
Feature activation-0.006
Top resid features:
-
Token-
Feature activation+0.012
Top resid features:
Fla
TokenFla
Feature activation+0.012
Top resid features:
.)
Token.)
Feature activation+0.173
Top resid features:
warnings
Token warnings
Feature activation+0.017
Top resid features:
from
Token from
Feature activation+0.031
Top resid features:
new
Token new
Feature activation+0.024
Top resid features:
climate
Token climate
Feature activation+0.006
Top resid features:
studies
Token studies
Feature activation+0.057
Top resid features:
MORE
Token MORE
Feature activation+1.678
Top resid features:
(
Token (
Feature activation-0.104
Top resid features:
R
TokenR
Feature activation+0.010
Top resid features:
-
Token-
Feature activation+0.045
Top resid features:
Ky
TokenKy
Feature activation+0.200
Top resid features:
.)
Token.)
Feature activation+0.418
Top resid features:
McCain
Token McCain
Feature activation-0.027
Top resid features:
after
Token after
Feature activation-0.009
Top resid features:
fight
Token fight
Feature activation-0.027
Top resid features:
over
Token over
Feature activation+0.023
Top resid features:
Assad
Token Assad
Feature activation+0.047
Top resid features:
MORE
Token MORE
Feature activation+2.356
Top resid features:
(
Token (
Feature activation-0.078
Top resid features:
R
TokenR
Feature activation-0.074
Top resid features:
-
Token-
Feature activation+0.176
Top resid features:
Ari
TokenAri
Feature activation+0.297
Top resid features:
z
Tokenz
Feature activation+0.120
Top resid features:
warnings
Token warnings
Feature activation+0.028
Top resid features:
from
Token from
Feature activation+0.029
Top resid features:
new
Token new
Feature activation+0.046
Top resid features:
climate
Token climate
Feature activation+0.013
Top resid features:
studies
Token studies
Feature activation+0.110
Top resid features:
MORE
Token MORE
Feature activation+2.299
Top resid features:
(
Token (
Feature activation-0.075
Top resid features:
R
TokenR
Feature activation-0.101
Top resid features:
-
Token-
Feature activation+0.131
Top resid features:
Ky
TokenKy
Feature activation+0.278
Top resid features:
.)
Token.)
Feature activation+0.000
Top resid features:
as
Token as
Feature activation-0.001
Top resid features:
long
Token long
Feature activation+0.003
Top resid features:
nomination
Token nomination
Feature activation-0.002
Top resid features:
contest
Token contest
Feature activation-0.007
Top resid features:
looms
Token looms
Feature activation-0.023
Top resid features:
MORE
Token MORE
Feature activation+1.305
Top resid features:
(
Token (
Feature activation-0.012
Top resid features:
D
TokenD
Feature activation-0.007
Top resid features:
-
Token-
Feature activation+0.005
Top resid features:
Fla
TokenFla
Feature activation+0.007
Top resid features:
.)
Token.)
Feature activation+0.116
Top resid features:
as
Token as
Feature activation-0.004
Top resid features:
long
Token long
Feature activation+0.004
Top resid features:
nomination
Token nomination
Feature activation-0.003
Top resid features:
contest
Token contest
Feature activation-0.009
Top resid features:
looms
Token looms
Feature activation-0.025
Top resid features:
MORE
Token MORE
Feature activation+1.198
Top resid features:
(
Token (
Feature activation-0.033
Top resid features:
D
TokenD
Feature activation-0.011
Top resid features:
-
Token-
Feature activation+0.013
Top resid features:
Fla
TokenFla
Feature activation+0.025
Top resid features:
.)
Token.)
Feature activation+0.164
Top resid features:
warnings
Token warnings
Feature activation+0.002
Top resid features:
from
Token from
Feature activation+0.012
Top resid features:
new
Token new
Feature activation+0.009
Top resid features:
climate
Token climate
Feature activation-0.000
Top resid features:
studies
Token studies
Feature activation+0.025
Top resid features:
MORE
Token MORE
Feature activation+1.032
Top resid features:
(
Token (
Feature activation-0.004
Top resid features:
R
TokenR
Feature activation-0.004
Top resid features:
-
Token-
Feature activation+0.020
Top resid features:
Ky
TokenKy
Feature activation+0.041
Top resid features:
.)
Token.)
Feature activation+0.334
Top resid features:
as
Token as
Feature activation-0.002
Top resid features:
long
Token long
Feature activation+0.003
Top resid features:
nomination
Token nomination
Feature activation-0.000
Top resid features:
contest
Token contest
Feature activation-0.007
Top resid features:
looms
Token looms
Feature activation-0.018
Top resid features:
MORE
Token MORE
Feature activation+1.133
Top resid features:
(
Token (
Feature activation-0.019
Top resid features:
D
TokenD
Feature activation-0.009
Top resid features:
-
Token-
Feature activation+0.007
Top resid features:
Fla
TokenFla
Feature activation+0.009
Top resid features:
.)
Token.)
Feature activation+0.161
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.09

Head 2: 0.04

Head 3: 0.05

Head 4: 0.03

Head 5: 0.05

Head 6: 0.31

Head 7: 0.07

Head 8: 0.05

Head 9: 0.06

Head 10: 0.12

Head 11: 0.06

Positive logits

congressional1.47

PAC1.43

subcommittee1.42

ospons1.40

"))1.40

ovember1.37

PAC1.36

isconsin1.30

Guantanamo1.22

constituency1.20

affer1.20

Subcommittee1.19

election1.18

ufact1.15

Hillary1.15

lobbying1.15

Prosecut1.15

taxpayer1.14

taboola1.12

Lago1.12

Negative logits

here-1.61

Shard-1.58

               -1.43

glim-1.39

Nietzsche-1.39

($)-1.33

  -1.31

});-1.31

Appearance-1.29

 -1.29

heres-1.29

Stones-1.28

norm-1.28

-->-1.28

Mages-1.27

CLR-1.24

»-1.23

Medium-1.23

AMY-1.21

-1.21

INTERVAL 2.136 - 2.373
CONTAINS 0.000%

-
Token-
Feature activation+0.891
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572

INTERVAL 1.899 - 2.136
CONTAINS 0.001%

.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
R
TokenR
Feature activation+0.939
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
âĢ
TokenâĢ
Feature activation+0.000
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Nobody
TokenNobody
Feature activation+0.000

INTERVAL 1.661 - 1.899
CONTAINS 0.001%

Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
introduced
Token introduced
Feature activation+1.163
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
warnings
Token warnings
Feature activation+0.000
from
Token from
Feature activation+0.000
new
Token new
Feature activation+0.000
climate
Token climate
Feature activation+0.000
studies
Token studies
Feature activation+0.000
MORE
Token MORE
Feature activation+1.798
(
Token (
Feature activation+0.844
R
TokenR
Feature activation+0.939
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886

INTERVAL 1.424 - 1.661
CONTAINS 0.001%

to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
introduced
Token introduced
Feature activation+1.163
in
Token in
Feature activation+0.577
the
Token the
Feature activation+0.664
(
Token (
Feature activation+0.337
D
TokenD
Feature activation+0.628
-
Token-
Feature activation+0.891
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
legislation
Token legislation
Feature activation+1.980
MORE
Token MORE
Feature activation+1.166
(
Token (
Feature activation+0.528
R
TokenR
Feature activation+1.131
-
Token-
Feature activation+1.310
Ari
TokenAri
Feature activation+1.393
z
Tokenz
Feature activation+1.637
.)
Token.)
Feature activation+0.974
in
Token in
Feature activation+1.264
a
Token a
Feature activation+1.587
new
Token new
Feature activation+2.074
interview
Token interview
Feature activation+1.465
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
(
Token (
Feature activation+0.844
R
TokenR
Feature activation+0.939
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499

INTERVAL 1.187 - 1.424
CONTAINS 0.001%

.)
Token.)
Feature activation+0.000
for
Token for
Feature activation+0.427
staying
Token staying
Feature activation+0.799
neutral
Token neutral
Feature activation+1.326
in
Token in
Feature activation+1.128
the
Token the
Feature activation+1.252
Democratic
Token Democratic
Feature activation+1.264
primary
Token primary
Feature activation+1.038
.
Token.
Feature activation+0.814
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
-
Token-
Feature activation+0.000
Mass
TokenMass
Feature activation+0.000
.)
Token.)
Feature activation+0.000
for
Token for
Feature activation+0.427
staying
Token staying
Feature activation+0.799
neutral
Token neutral
Feature activation+1.326
in
Token in
Feature activation+1.128
the
Token the
Feature activation+1.252
Democratic
Token Democratic
Feature activation+1.264
primary
Token primary
Feature activation+1.038
.
Token.
Feature activation+0.814
legislation
Token legislation
Feature activation+1.980
to
Token to
Feature activation+1.445
prohibit
Token prohibit
Feature activation+1.530
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
introduced
Token introduced
Feature activation+1.163
in
Token in
Feature activation+0.577
Assad
Token Assad
Feature activation+0.000
MORE
Token MORE
Feature activation+1.166
(
Token (
Feature activation+0.528
R
TokenR
Feature activation+1.131
-
Token-
Feature activation+1.310
Ari
TokenAri
Feature activation+1.393
z
Tokenz
Feature activation+1.637
.)
Token.)
Feature activation+0.974
in
Token in
Feature activation+1.264
a
Token a
Feature activation+1.587
new
Token new
Feature activation+2.074
over
Token over
Feature activation+0.000
Assad
Token Assad
Feature activation+0.000
MORE
Token MORE
Feature activation+1.166
(
Token (
Feature activation+0.528
R
TokenR
Feature activation+1.131
-
Token-
Feature activation+1.310
Ari
TokenAri
Feature activation+1.393
z
Tokenz
Feature activation+1.637
.)
Token.)
Feature activation+0.974
in
Token in
Feature activation+1.264
a
Token a
Feature activation+1.587

INTERVAL 0.949 - 1.187
CONTAINS 0.001%

staying
Token staying
Feature activation+0.799
neutral
Token neutral
Feature activation+1.326
in
Token in
Feature activation+1.128
the
Token the
Feature activation+1.252
Democratic
Token Democratic
Feature activation+1.264
primary
Token primary
Feature activation+1.038
.
Token.
Feature activation+0.814
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Some
TokenSome
Feature activation+0.000
Sanders
Token Sanders
Feature activation+0.000
fight
Token fight
Feature activation+0.000
over
Token over
Feature activation+0.000
Assad
Token Assad
Feature activation+0.000
MORE
Token MORE
Feature activation+1.166
(
Token (
Feature activation+0.528
R
TokenR
Feature activation+1.131
-
Token-
Feature activation+1.310
Ari
TokenAri
Feature activation+1.393
z
Tokenz
Feature activation+1.637
.)
Token.)
Feature activation+0.974
in
Token in
Feature activation+1.264
as
Token as
Feature activation+0.000
long
Token long
Feature activation+0.000
nomination
Token nomination
Feature activation+0.000
contest
Token contest
Feature activation+0.000
looms
Token looms
Feature activation+0.062
MORE
Token MORE
Feature activation+1.121
(
Token (
Feature activation+0.337
D
TokenD
Feature activation+0.628
-
Token-
Feature activation+0.891
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
MORE
Token MORE
Feature activation+1.121
(
Token (
Feature activation+0.337
D
TokenD
Feature activation+0.628
-
Token-
Feature activation+0.891
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146
on
Token on
Feature activation+1.876
similar
Token similar
Feature activation+1.976
plastic
Token plastic
Feature activation+1.781
guns
Token guns
Feature activation+1.847
,
Token,
Feature activation+1.411
which
Token which
Feature activation+1.487
could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
introduced
Token introduced
Feature activation+1.163
in
Token in
Feature activation+0.577
the
Token the
Feature activation+0.664
coming
Token coming
Feature activation+0.716
months
Token months
Feature activation+0.688

INTERVAL 0.712 - 0.949
CONTAINS 0.001%

could
Token could
Feature activation+1.544
be
Token be
Feature activation+0.983
introduced
Token introduced
Feature activation+1.163
in
Token in
Feature activation+0.577
the
Token the
Feature activation+0.664
coming
Token coming
Feature activation+0.716
months
Token months
Feature activation+0.688
.
Token.
Feature activation+0.777
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
While
TokenWhile
Feature activation+0.000
D
TokenD
Feature activation+0.000
-
Token-
Feature activation+0.000
Mass
TokenMass
Feature activation+0.000
.)
Token.)
Feature activation+0.000
for
Token for
Feature activation+0.427
staying
Token staying
Feature activation+0.799
neutral
Token neutral
Feature activation+1.326
in
Token in
Feature activation+1.128
the
Token the
Feature activation+1.252
Democratic
Token Democratic
Feature activation+1.264
primary
Token primary
Feature activation+1.038
new
Token new
Feature activation+0.000
climate
Token climate
Feature activation+0.000
studies
Token studies
Feature activation+0.000
MORE
Token MORE
Feature activation+1.798
(
Token (
Feature activation+0.844
R
TokenR
Feature activation+0.939
-
Token-
Feature activation+1.468
Ky
TokenKy
Feature activation+1.635
.)
Token.)
Feature activation+0.886
and
Token and
Feature activation+1.649
the
Token the
Feature activation+2.063
congressional
Token congressional
Feature activation+0.000
oversight
Token oversight
Feature activation+0.000
MORE
Token MORE
Feature activation+0.987
(
Token (
Feature activation+0.147
D
TokenD
Feature activation+0.463
-
Token-
Feature activation+0.728
N
TokenN
Feature activation+0.172
.
Token.
Feature activation+0.927
J
TokenJ
Feature activation+1.265
.),
Token.),
Feature activation+0.331
Jim
Token Jim
Feature activation+1.379
contest
Token contest
Feature activation+0.000
looms
Token looms
Feature activation+0.062
MORE
Token MORE
Feature activation+1.121
(
Token (
Feature activation+0.337
D
TokenD
Feature activation+0.628
-
Token-
Feature activation+0.891
Fla
TokenFla
Feature activation+1.130
.)
Token.)
Feature activation+0.956
are
Token are
Feature activation+1.442
also
Token also
Feature activation+1.822
working
Token working
Feature activation+2.146

INTERVAL 0.475 - 0.712
CONTAINS 0.002%

Republican
Token Republican
Feature activation+2.373
Senate
Token Senate
Feature activation+1.802
campaign
Token campaign
Feature activation+1.949
arm
Token arm
Feature activation+1.499
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.572
Ċ
TokenĊ
Feature activation+0.074
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Nobody
TokenNobody
Feature activation+0.000
wants
Token wants
Feature activation+0.147
GOP
Token GOP
Feature activation+0.267
voices
Token voices
Feature activation+0.000
who
Token who
Feature activation+0.000
have
Token have
Feature activation+0.513
called
Token called
Feature activation+0.582
for
Token for
Feature activation+0.559
rec
Token rec
Feature activation+0.239
usal
Tokenusal
Feature activation+0.275
.
Token.
Feature activation+0.248
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Y
TokenY
Feature activation+0.000
.)
Token.)
Feature activation+0.000
and
Token and
Feature activation+0.000
Bill
Token Bill
Feature activation+0.029
Nelson
Token Nelson
Feature activation+0.496
Clarence
Token Clarence
Feature activation+0.693
(
Token (
Feature activation+0.333
Bill
TokenBill
Feature activation+0.052
)
Token)
Feature activation+0.214
William
Token William
Feature activation+0.120
Nelson
Token Nelson
Feature activation+0.301
Nelson
Token Nelson
Feature activation+0.301
2020
Token2020
Feature activation+0.332
party
Token party
Feature activation+0.382
politics
Token politics
Feature activation+0.334
in
Token in
Feature activation+0.137
Puerto
Token Puerto
Feature activation+0.508
Rico
Token Rico
Feature activation+0.456
There
Token There
Feature activation+0.261
is
Token is
Feature activation+0.190
no
Token no
Feature activation+0.095
winning
Token winning
Feature activation+0.000
the
Token the
Feature activation+0.046
key
Token key
Feature activation+0.216
GOP
Token GOP
Feature activation+0.267
voices
Token voices
Feature activation+0.000
who
Token who
Feature activation+0.000
have
Token have
Feature activation+0.513
called
Token called
Feature activation+0.582
for
Token for
Feature activation+0.559
rec
Token rec
Feature activation+0.239
usal
Tokenusal
Feature activation+0.275
.
Token.
Feature activation+0.248

INTERVAL 0.237 - 0.475
CONTAINS 0.002%

(
Token (
Feature activation+0.333
Bill
TokenBill
Feature activation+0.052
)
Token)
Feature activation+0.214
William
Token William
Feature activation+0.120
Nelson
Token Nelson
Feature activation+0.301
2020
Token2020
Feature activation+0.332
party
Token party
Feature activation+0.382
politics
Token politics
Feature activation+0.334
in
Token in
Feature activation+0.137
Puerto
Token Puerto
Feature activation+0.508
Rico
Token Rico
Feature activation+0.456
Mexican
Token Mexican
Feature activation+0.122
border
Token border
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.220
issues
Token issues
Feature activation+0.301
impacting
Token impacting
Feature activation+0.292
Alabama
Token Alabama
Feature activation+0.112
voters
Token voters
Feature activation+0.000
and
Token and
Feature activation+0.000
what
Token what
Feature activation+0.163
characteristics
Token characteristics
Feature activation+0.000
)
Token)
Feature activation+0.214
William
Token William
Feature activation+0.120
Nelson
Token Nelson
Feature activation+0.301
2020
Token2020
Feature activation+0.332
party
Token party
Feature activation+0.382
politics
Token politics
Feature activation+0.334
in
Token in
Feature activation+0.137
Puerto
Token Puerto
Feature activation+0.508
Rico
Token Rico
Feature activation+0.456
There
Token There
Feature activation+0.261
is
Token is
Feature activation+0.190
and
Token and
Feature activation+0.000
what
Token what
Feature activation+0.163
characteristics
Token characteristics
Feature activation+0.000
he
Token he
Feature activation+0.000
thinks
Token thinks
Feature activation+0.376
make
Token make
Feature activation+0.353
a
Token a
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
really
Tokenreally
Feature activation+0.000
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
Warren
Token Warren
Feature activation+0.136
didn
Token didn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.034
t
Tokent
Feature activation+0.260
tilt
Token tilt
Feature activation+0.000
the
Token the
Feature activation+0.118
scales
Token scales
Feature activation+0.000
in
Token in
Feature activation+0.022
the
Token the
Feature activation+0.000

INTERVAL 0.000 - 0.237
CONTAINS 99.990%

about
Token about
Feature activation+0.000
Nation
Token Nation
Feature activation+0.000
Travels
Token Travels
Feature activation+0.000
destinations
Token destinations
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
explore
Token explore
Feature activation+0.000
the
Token the
Feature activation+0.000
world
Token world
Feature activation+0.000
with
Token with
Feature activation+0.000
kindred
Token kindred
Feature activation+0.000
if
Token if
Feature activation+0.000
the
Token the
Feature activation+0.000
show
Token show
Feature activation+0.000
is
Token is
Feature activation+0.000
ax
Token ax
Feature activation+0.000
ed
Tokened
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
it
Token it
Feature activation+0.000
's
Token's
Feature activation+0.000
highly
Token highly
Feature activation+0.000
Sky
Token Sky
Feature activation+0.000
and
Token and
Feature activation+0.000
Sean
Token Sean
Feature activation+0.000
did
Token did
Feature activation+0.000
,
Token,
Feature activation+0.000
too
Token too
Feature activation+0.000
,
Token,
Feature activation+0.000
so
Token so
Feature activation+0.000
much
Token much
Feature activation+0.000
so
Token so
Feature activation+0.000
that
Token that
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
green
Token green
Feature activation+0.000
bl
Token bl
Feature activation+0.000
azer
Tokenazer
Feature activation+0.000
waving
Token waving
Feature activation+0.000
a
Token a
Feature activation+0.000
fist
Token fist
Feature activation+0.000
ful
Tokenful
Feature activation+0.000
of
Token of
Feature activation+0.000
because
Token because
Feature activation+0.000
the
Token the
Feature activation+0.000
air
Token air
Feature activation+0.000
would
Token would
Feature activation+0.000
stop
Token stop
Feature activation+0.000
moving
Token moving
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
,
Token,
Feature activation+0.000
says
Token says
Feature activation+0.000
Ken
Token Ken
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 H2.6: (feature 6113)

TOP ACTIVATIONS
MAX = 1.231

With
TokenWith
Feature activation+0.000
the
Token the
Feature activation+0.000
two
Token two
Feature activation+0.000
cameras
Token cameras
Feature activation+0.000
,
Token,
Feature activation+0.493
you
Token you
Feature activation+1.231
can
Token can
Feature activation+0.701
achieve
Token achieve
Feature activation+0.209
an
Token an
Feature activation+0.217
outstanding
Token outstanding
Feature activation+0.032
240
Token 240
Feature activation+0.000
teams
Token teams
Feature activation+0.000
in
Token in
Feature activation+0.000
three
Token three
Feature activation+0.033
years
Token years
Feature activation+0.000
,
Token,
Feature activation+0.362
we
Token we
Feature activation+1.118
now
Token now
Feature activation+0.288
have
Token have
Feature activation+0.210
enough
Token enough
Feature activation+0.000
data
Token data
Feature activation+0.000
to
Token to
Feature activation+0.000
1
Token1
Feature activation+0.000
NCAA
Token NCAA
Feature activation+0.000
basketball
Token basketball
Feature activation+0.000
tournament
Token tournament
Feature activation+0.000
,
Token,
Feature activation+0.838
I
Token I
Feature activation+1.109
started
Token started
Feature activation+0.497
thinking
Token thinking
Feature activation+0.072
about
Token about
Feature activation+0.000
all
Token all
Feature activation+0.000
of
Token of
Feature activation+0.000
lege
Tokenlege
Feature activation+0.000
about
Token about
Feature activation+0.129
bit
Token bit
Feature activation+0.075
operations
Token operations
Feature activation+0.000
,
Token,
Feature activation+0.530
we
Token we
Feature activation+1.104
âĢ
TokenâĢ
Feature activation+0.319
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
make
Token make
Feature activation+0.190
an
Token an
Feature activation+0.000
back
Token back
Feature activation+0.000
in
Token in
Feature activation+0.000
full
Token full
Feature activation+0.000
swing
Token swing
Feature activation+0.000
,
Token,
Feature activation+0.766
it
Token it
Feature activation+1.066
is
Token is
Feature activation+0.738
a
Token a
Feature activation+0.124
matter
Token matter
Feature activation+0.000
of
Token of
Feature activation+0.000
time
Token time
Feature activation+0.000
without
Token without
Feature activation+0.000
the
Token the
Feature activation+0.000
baby
Token baby
Feature activation+0.000
boom
Token boom
Feature activation+0.000
,
Token,
Feature activation+0.708
we
Token we
Feature activation+1.056
âĢ
TokenâĢ
Feature activation+0.398
Ļ
TokenĻ
Feature activation+0.049
ll
Tokenll
Feature activation+0.000
still
Token still
Feature activation+0.243
see
Token see
Feature activation+0.042
in
Token in
Feature activation+0.000
neighbouring
Token neighbouring
Feature activation+0.000
Tamil
Token Tamil
Feature activation+0.000
Nadu
Token Nadu
Feature activation+0.000
,
Token,
Feature activation+0.490
it
Token it
Feature activation+1.035
is
Token is
Feature activation+0.447
now
Token now
Feature activation+0.000
the
Token the
Feature activation+0.000
turn
Token turn
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
base
Token base
Feature activation+0.000
frame
Token frame
Feature activation+0.000
completed
Token completed
Feature activation+0.000
,
Token,
Feature activation+0.544
it
Token it
Feature activation+0.999
was
Token was
Feature activation+0.703
time
Token time
Feature activation+0.071
to
Token to
Feature activation+0.000
install
Token install
Feature activation+0.000
the
Token the
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
way
Token way
Feature activation+0.000
,
Token,
Feature activation+0.238
let
Token let
Feature activation+0.970
me
Token me
Feature activation+0.229
get
Token get
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
meat
Token meat
Feature activation+0.000
expected
Token expected
Feature activation+0.000
to
Token to
Feature activation+0.000
impact
Token impact
Feature activation+0.000
Alabama
Token Alabama
Feature activation+0.000
,
Token,
Feature activation+0.485
we
Token we
Feature activation+0.961
are
Token are
Feature activation+0.350
taking
Token taking
Feature activation+0.164
precautions
Token precautions
Feature activation+0.000
to
Token to
Feature activation+0.000
protect
Token protect
Feature activation+0.000
of
Token of
Feature activation+0.000
machine
Token machine
Feature activation+0.000
learning
Token learning
Feature activation+0.000
,
Token,
Feature activation+0.568
bots
Token bots
Feature activation+0.437
will
Token will
Feature activation+0.942
begin
Token begin
Feature activation+0.508
to
Token to
Feature activation+0.000
understand
Token understand
Feature activation+0.000
and
Token and
Feature activation+0.000
remember
Token remember
Feature activation+0.000
the
Token the
Feature activation+0.000
top
Token top
Feature activation+0.000
layer
Token layer
Feature activation+0.000
active
Token active
Feature activation+0.000
,
Token,
Feature activation+0.240
click
Token click
Feature activation+0.936
the
Token the
Feature activation+0.298
Add
Token Add
Feature activation+0.000
Layer
Token Layer
Feature activation+0.000
Mask
Token Mask
Feature activation+0.000
icon
Token icon
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.064
MSG
Token MSG
Feature activation+0.000
Network
Token Network
Feature activation+0.000
,
Token,
Feature activation+0.618
there
Token there
Feature activation+0.922
is
Token is
Feature activation+0.504
a
Token a
Feature activation+0.097
lot
Token lot
Feature activation+0.000
of
Token of
Feature activation+0.000
quality
Token quality
Feature activation+0.000
the
Token the
Feature activation+0.000
Fire
Token Fire
Feature activation+0.000
Rapt
Token Rapt
Feature activation+0.000
or
Tokenor
Feature activation+0.000
,
Token,
Feature activation+0.572
it
Token it
Feature activation+0.898
can
Token can
Feature activation+0.386
target
Token target
Feature activation+0.000
a
Token a
Feature activation+0.000
different
Token different
Feature activation+0.000
enemy
Token enemy
Feature activation+0.000
ARY
TokenARY
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
With
Token With
Feature activation+0.000
oil
Token oil
Feature activation+0.000
prices
Token prices
Feature activation+0.000
remaining
Token remaining
Feature activation+0.868
below
Token below
Feature activation+0.352
$
Token $
Feature activation+0.486
50
Token50
Feature activation+0.036
a
Token a
Feature activation+0.382
barrel
Token barrel
Feature activation+0.165
upper
Token upper
Feature activation+0.000
-
Token-
Feature activation+0.000
90
Token90
Feature activation+0.000
s
Tokens
Feature activation+0.000
,
Token,
Feature activation+0.371
there
Token there
Feature activation+0.856
are
Token are
Feature activation+0.264
going
Token going
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
two
Token two
Feature activation+0.000
the
Token the
Feature activation+0.025
Model
Token Model
Feature activation+0.000
S
Token S
Feature activation+0.000
in
Token in
Feature activation+0.000
Oz
Token Oz
Feature activation+0.000
comes
Token comes
Feature activation+0.855
the
Token the
Feature activation+0.264
need
Token need
Feature activation+0.217
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
nationwide
Token nationwide
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
With
TokenWith
Feature activation+0.000
every
Token every
Feature activation+0.000
release
Token release
Feature activation+0.000
,
Token,
Feature activation+0.161
Apple
Token Apple
Feature activation+0.845
have
Token have
Feature activation+0.657
been
Token been
Feature activation+0.222
making
Token making
Feature activation+0.038
developer
Token developer
Feature activation+0.000
tools
Token tools
Feature activation+0.000
-
Token-
Feature activation+0.000
1
Token1
Feature activation+0.000
NCAA
Token NCAA
Feature activation+0.000
basketball
Token basketball
Feature activation+0.000
tournament
Token tournament
Feature activation+0.000
,
Token,
Feature activation+0.838
I
Token I
Feature activation+1.109
started
Token started
Feature activation+0.497
thinking
Token thinking
Feature activation+0.072
about
Token about
Feature activation+0.000
all
Token all
Feature activation+0.000
media
Token media
Feature activation+0.000
by
Token by
Feature activation+0.021
our
Token our
Feature activation+0.142
side
Token side
Feature activation+0.000
,
Token,
Feature activation+0.396
it
Token it
Feature activation+0.831
âĢ
TokenâĢ
Feature activation+0.318
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
possible
Token possible
Feature activation+0.039
now
Token now
Feature activation+0.000

Top DFA by src position
MAX = 1.567

added
Token added
Feature activation-0.045
Top resid features:
convenience
Token convenience
Feature activation+0.006
Top resid features:
.
Token.
Feature activation+0.013
Top resid features:
Ċ
TokenĊ
Feature activation-0.027
Top resid features:
Ċ
TokenĊ
Feature activation-0.029
Top resid features:
With
TokenWith
Feature activation+1.567
Top resid features:
the
Token the
Feature activation+0.022
Top resid features:
two
Token two
Feature activation-0.001
Top resid features:
cameras
Token cameras
Feature activation-0.023
Top resid features:
,
Token,
Feature activation+0.340
Top resid features:
you
Token you
Feature activation-0.014
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.036
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation-0.020
Top resid features:
Ċ
TokenĊ
Feature activation-0.021
Top resid features:
With
TokenWith
Feature activation+1.475
Top resid features:
12
Token 12
Feature activation-0.013
Top resid features:
teams
Token teams
Feature activation+0.009
Top resid features:
in
Token in
Feature activation-0.014
Top resid features:
three
Token three
Feature activation+0.006
Top resid features:
years
Token years
Feature activation-0.054
Top resid features:
don
Token don
Feature activation+0.001
Top resid features:
't
Token't
Feature activation+0.001
Top resid features:
talk
Token talk
Feature activation-0.003
Top resid features:
about
Token about
Feature activation+0.003
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.025
Top resid features:
With
TokenWith
Feature activation+1.529
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
recent
Token recent
Feature activation+0.017
Top resid features:
completion
Token completion
Feature activation+0.013
Top resid features:
of
Token of
Feature activation-0.004
Top resid features:
the
Token the
Feature activation-0.004
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Read
Token Read
Feature activation-0.020
Top resid features:
image
Token image
Feature activation-0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.001
Top resid features:
With
TokenWith
Feature activation+1.335
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
know
Token know
Feature activation-0.001
Top resid features:
lege
Tokenlege
Feature activation+0.005
Top resid features:
about
Token about
Feature activation+0.042
Top resid features:
bit
Token bit
Feature activation-0.007
Top resid features:
Cowboys
Token Cowboys
Feature activation+0.007
Top resid features:
merchandise
Token merchandise
Feature activation+0.007
Top resid features:
.
Token.
Feature activation+0.012
Top resid features:
Ċ
TokenĊ
Feature activation-0.014
Top resid features:
Ċ
TokenĊ
Feature activation-0.010
Top resid features:
With
TokenWith
Feature activation+1.401
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
NRL
Token NRL
Feature activation-0.013
Top resid features:
season
Token season
Feature activation-0.008
Top resid features:
back
Token back
Feature activation-0.008
Top resid features:
in
Token in
Feature activation-0.024
Top resid features:
the
Token the
Feature activation-0.006
Top resid features:
age
Token age
Feature activation-0.000
Top resid features:
of
Token of
Feature activation-0.007
Top resid features:
40
Token 40
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.030
Top resid features:
With
Token With
Feature activation+1.234
Top resid features:
or
Token or
Feature activation+0.013
Top resid features:
without
Token without
Feature activation+0.013
Top resid features:
the
Token the
Feature activation+0.012
Top resid features:
baby
Token baby
Feature activation-0.008
Top resid features:
boom
Token boom
Feature activation+0.071
Top resid features:
January
Token January
Feature activation-0.004
Top resid features:
26
Token 26
Feature activation-0.003
Top resid features:
.
Token.
Feature activation+0.015
Top resid features:
Ċ
TokenĊ
Feature activation-0.016
Top resid features:
Ċ
TokenĊ
Feature activation-0.013
Top resid features:
With
TokenWith
Feature activation+1.549
Top resid features:
the
Token the
Feature activation-0.005
Top resid features:
res
Token res
Feature activation-0.000
Top resid features:
ounding
Tokenounding
Feature activation+0.012
Top resid features:
success
Token success
Feature activation+0.011
Top resid features:
of
Token of
Feature activation-0.012
Top resid features:
Ċ
TokenĊ
Feature activation-0.019
Top resid features:
Bench
TokenBench
Feature activation+0.009
Top resid features:
top
Token top
Feature activation-0.018
Top resid features:
Ċ
TokenĊ
Feature activation-0.020
Top resid features:
Ċ
TokenĊ
Feature activation-0.020
Top resid features:
With
TokenWith
Feature activation+1.348
Top resid features:
the
Token the
Feature activation+0.005
Top resid features:
base
Token base
Feature activation-0.027
Top resid features:
frame
Token frame
Feature activation-0.010
Top resid features:
completed
Token completed
Feature activation-0.058
Top resid features:
,
Token,
Feature activation+0.333
Top resid features:
,
Token,
Feature activation+0.021
Top resid features:
guaranteed
Token guaranteed
Feature activation-0.006
Top resid features:
.
Token.
Feature activation-0.008
Top resid features:
Ċ
TokenĊ
Feature activation-0.038
Top resid features:
Ċ
TokenĊ
Feature activation-0.041
Top resid features:
With
TokenWith
Feature activation+1.348
Top resid features:
that
Token that
Feature activation+0.030
Top resid features:
out
Token out
Feature activation-0.023
Top resid features:
of
Token of
Feature activation-0.026
Top resid features:
the
Token the
Feature activation-0.019
Top resid features:
way
Token way
Feature activation+0.014
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation-0.015
Top resid features:
Ċ
TokenĊ
Feature activation-0.014
Top resid features:
âĢ
TokenâĢ
Feature activation+0.020
Top resid features:
ľ
Tokenľ
Feature activation+0.005
Top resid features:
With
TokenWith
Feature activation+1.314
Top resid features:
winter
Token winter
Feature activation-0.015
Top resid features:
weather
Token weather
Feature activation-0.011
Top resid features:
expected
Token expected
Feature activation-0.030
Top resid features:
to
Token to
Feature activation-0.017
Top resid features:
impact
Token impact
Feature activation-0.012
Top resid features:
Ċ
TokenĊ
Feature activation-0.014
Top resid features:
Deep
TokenDeep
Feature activation+0.010
Top resid features:
learners
Token learners
Feature activation+0.045
Top resid features:
Ċ
TokenĊ
Feature activation-0.021
Top resid features:
Ċ
TokenĊ
Feature activation-0.020
Top resid features:
With
TokenWith
Feature activation+1.314
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
help
Token help
Feature activation+0.008
Top resid features:
of
Token of
Feature activation-0.002
Top resid features:
machine
Token machine
Feature activation+0.012
Top resid features:
learning
Token learning
Feature activation+0.035
Top resid features:
mask
Token mask
Feature activation-0.001
Top resid features:
for
Token for
Feature activation-0.013
Top resid features:
this
Token this
Feature activation-0.007
Top resid features:
image
Token image
Feature activation-0.015
Top resid features:
.
Token.
Feature activation-0.028
Top resid features:
With
Token With
Feature activation+1.245
Top resid features:
the
Token the
Feature activation+0.007
Top resid features:
top
Token top
Feature activation-0.033
Top resid features:
layer
Token layer
Feature activation-0.007
Top resid features:
active
Token active
Feature activation-0.011
Top resid features:
,
Token,
Feature activation+0.328
Top resid features:
ie
Tokenie
Feature activation+0.000
Top resid features:
Rose
Token Rose
Feature activation-0.000
Top resid features:
twitter
Token twitter
Feature activation-0.005
Top resid features:
account
Token account
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.040
Top resid features:
With
Token With
Feature activation+1.228
Top resid features:
a
Token a
Feature activation-0.003
Top resid features:
star
Token star
Feature activation-0.004
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
st
Tokenst
Feature activation-0.003
Top resid features:
udd
Tokenudd
Feature activation+0.001
Top resid features:
little
Token little
Feature activation+0.009
Top resid features:
unfair
Token unfair
Feature activation+0.007
Top resid features:
.
Token.
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation-0.032
Top resid features:
Ċ
TokenĊ
Feature activation-0.026
Top resid features:
With
TokenWith
Feature activation+1.352
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
way
Token way
Feature activation+0.024
Top resid features:
the
Token the
Feature activation-0.004
Top resid features:
rules
Token rules
Feature activation-0.002
Top resid features:
are
Token are
Feature activation-0.027
Top resid features:
C
TokenC
Feature activation-0.014
Top resid features:
AL
TokenAL
Feature activation+0.001
Top resid features:
G
TokenG
Feature activation+0.006
Top resid features:
ARY
TokenARY
Feature activation+0.005
Top resid features:
âĢĵ
Token âĢĵ
Feature activation+0.128
Top resid features:
With
Token With
Feature activation+1.230
Top resid features:
oil
Token oil
Feature activation-0.022
Top resid features:
prices
Token prices
Feature activation+0.097
Top resid features:
remaining
Token remaining
Feature activation+0.064
Top resid features:
below
Token below
Feature activation+0.000
Top resid features:
$
Token $
Feature activation+0.000
Top resid features:
That
Token That
Feature activation+0.024
Top resid features:
's
Token's
Feature activation-0.000
Top resid features:
46
Token 46
Feature activation+0.004
Top resid features:
percent
Token percent
Feature activation+0.012
Top resid features:
.
Token.
Feature activation+0.034
Top resid features:
With
TokenWith
Feature activation+1.296
Top resid features:
any
Token any
Feature activation+0.005
Top resid features:
tall
Token tall
Feature activation-0.001
Top resid features:
,
Token,
Feature activation+0.037
Top resid features:
l
Token l
Feature activation-0.001
Top resid features:
anky
Tokenanky
Feature activation+0.013
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation-0.022
Top resid features:
Ċ
TokenĊ
Feature activation-0.023
Top resid features:
With
TokenWith
Feature activation+1.356
Top resid features:
the
Token the
Feature activation-0.003
Top resid features:
launch
Token launch
Feature activation-0.001
Top resid features:
of
Token of
Feature activation-0.006
Top resid features:
the
Token the
Feature activation-0.006
Top resid features:
Model
Token Model
Feature activation-0.029
Top resid features:
-
Token-
Feature activation-0.001
Top resid features:
C
TokenC
Feature activation-0.008
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.049
Top resid features:
Ċ
TokenĊ
Feature activation-0.052
Top resid features:
With
TokenWith
Feature activation+1.123
Top resid features:
every
Token every
Feature activation+0.003
Top resid features:
release
Token release
Feature activation-0.045
Top resid features:
,
Token,
Feature activation+0.358
Top resid features:
Apple
Token Apple
Feature activation-0.050
Top resid features:
have
Token have
Feature activation+0.000
Top resid features:
don
Token don
Feature activation+0.001
Top resid features:
't
Token't
Feature activation+0.002
Top resid features:
talk
Token talk
Feature activation-0.003
Top resid features:
about
Token about
Feature activation+0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.029
Top resid features:
With
TokenWith
Feature activation+1.225
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
recent
Token recent
Feature activation+0.016
Top resid features:
completion
Token completion
Feature activation+0.015
Top resid features:
of
Token of
Feature activation-0.005
Top resid features:
the
Token the
Feature activation-0.005
Top resid features:
ľ
Tokenľ
Feature activation-0.000
Top resid features:
PR
TokenPR
Feature activation+0.000
Top resid features:
!
Token!
Feature activation+0.024
Top resid features:
âĢ
TokenâĢ
Feature activation+0.010
Top resid features:
Ŀ
TokenĿ
Feature activation+0.012
Top resid features:
With
Token With
Feature activation+1.199
Top resid features:
the
Token the
Feature activation-0.006
Top resid features:
Internet
Token Internet
Feature activation-0.007
Top resid features:
and
Token and
Feature activation+0.006
Top resid features:
social
Token social
Feature activation-0.016
Top resid features:
media
Token media
Feature activation-0.017
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.05

Head 2: 0.09

Head 3: 0.08

Head 4: 0.04

Head 5: 0.12

Head 6: 0.28

Head 7: 0.03

Head 8: 0.08

Head 9: 0.05

Head 10: 0.08

Head 11: 0.03

Positive logits

xus1.67

isphere1.43

oubted1.43

illac1.43

oubtedly1.38

ync1.34

dfx1.28

automakers1.24

tempting1.23

ASUS1.23

bets1.22

icion1.22

agame1.22

raltar1.21

plementation1.20

ensibly1.19

CentOS1.19

slot1.17

wagen1.17

idious1.16

Negative logits

SPONSORED-1.81

.�-1.58

Interstitial-1.37

him-1.33

.]-1.32

dayName-1.30

vulgar-1.29

jee-1.19

}.-1.17

Contains-1.17

shit-1.17

]."-1.17

Mom-1.16

quin-1.15

Woman-1.15

].-1.15

��-1.15

PD-1.13

himself-1.11

jun-1.11

INTERVAL 1.108 - 1.231
CONTAINS 0.000%

teams
Token teams
Feature activation+0.000
in
Token in
Feature activation+0.000
three
Token three
Feature activation+0.033
years
Token years
Feature activation+0.000
,
Token,
Feature activation+0.362
we
Token we
Feature activation+1.118
now
Token now
Feature activation+0.288
have
Token have
Feature activation+0.210
enough
Token enough
Feature activation+0.000
data
Token data
Feature activation+0.000
to
Token to
Feature activation+0.000
1
Token1
Feature activation+0.000
NCAA
Token NCAA
Feature activation+0.000
basketball
Token basketball
Feature activation+0.000
tournament
Token tournament
Feature activation+0.000
,
Token,
Feature activation+0.838
I
Token I
Feature activation+1.109
started
Token started
Feature activation+0.497
thinking
Token thinking
Feature activation+0.072
about
Token about
Feature activation+0.000
all
Token all
Feature activation+0.000
of
Token of
Feature activation+0.000
With
TokenWith
Feature activation+0.000
the
Token the
Feature activation+0.000
two
Token two
Feature activation+0.000
cameras
Token cameras
Feature activation+0.000
,
Token,
Feature activation+0.493
you
Token you
Feature activation+1.231
can
Token can
Feature activation+0.701
achieve
Token achieve
Feature activation+0.209
an
Token an
Feature activation+0.217
outstanding
Token outstanding
Feature activation+0.032
240
Token 240
Feature activation+0.000

INTERVAL 0.985 - 1.108
CONTAINS 0.001%

in
Token in
Feature activation+0.000
neighbouring
Token neighbouring
Feature activation+0.000
Tamil
Token Tamil
Feature activation+0.000
Nadu
Token Nadu
Feature activation+0.000
,
Token,
Feature activation+0.490
it
Token it
Feature activation+1.035
is
Token is
Feature activation+0.447
now
Token now
Feature activation+0.000
the
Token the
Feature activation+0.000
turn
Token turn
Feature activation+0.000
of
Token of
Feature activation+0.000
without
Token without
Feature activation+0.000
the
Token the
Feature activation+0.000
baby
Token baby
Feature activation+0.000
boom
Token boom
Feature activation+0.000
,
Token,
Feature activation+0.708
we
Token we
Feature activation+1.056
âĢ
TokenâĢ
Feature activation+0.398
Ļ
TokenĻ
Feature activation+0.049
ll
Tokenll
Feature activation+0.000
still
Token still
Feature activation+0.243
see
Token see
Feature activation+0.042
back
Token back
Feature activation+0.000
in
Token in
Feature activation+0.000
full
Token full
Feature activation+0.000
swing
Token swing
Feature activation+0.000
,
Token,
Feature activation+0.766
it
Token it
Feature activation+1.066
is
Token is
Feature activation+0.738
a
Token a
Feature activation+0.124
matter
Token matter
Feature activation+0.000
of
Token of
Feature activation+0.000
time
Token time
Feature activation+0.000
the
Token the
Feature activation+0.000
base
Token base
Feature activation+0.000
frame
Token frame
Feature activation+0.000
completed
Token completed
Feature activation+0.000
,
Token,
Feature activation+0.544
it
Token it
Feature activation+0.999
was
Token was
Feature activation+0.703
time
Token time
Feature activation+0.071
to
Token to
Feature activation+0.000
install
Token install
Feature activation+0.000
the
Token the
Feature activation+0.000
lege
Tokenlege
Feature activation+0.000
about
Token about
Feature activation+0.129
bit
Token bit
Feature activation+0.075
operations
Token operations
Feature activation+0.000
,
Token,
Feature activation+0.530
we
Token we
Feature activation+1.104
âĢ
TokenâĢ
Feature activation+0.319
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
make
Token make
Feature activation+0.190
an
Token an
Feature activation+0.000

INTERVAL 0.862 - 0.985
CONTAINS 0.001%

ARY
TokenARY
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
With
Token With
Feature activation+0.000
oil
Token oil
Feature activation+0.000
prices
Token prices
Feature activation+0.000
remaining
Token remaining
Feature activation+0.868
below
Token below
Feature activation+0.352
$
Token $
Feature activation+0.486
50
Token50
Feature activation+0.036
a
Token a
Feature activation+0.382
barrel
Token barrel
Feature activation+0.165
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
,
Token,
Feature activation+0.238
let
Token let
Feature activation+0.970
me
Token me
Feature activation+0.229
get
Token get
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
meat
Token meat
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.064
MSG
Token MSG
Feature activation+0.000
Network
Token Network
Feature activation+0.000
,
Token,
Feature activation+0.618
there
Token there
Feature activation+0.922
is
Token is
Feature activation+0.504
a
Token a
Feature activation+0.097
lot
Token lot
Feature activation+0.000
of
Token of
Feature activation+0.000
quality
Token quality
Feature activation+0.000
expected
Token expected
Feature activation+0.000
to
Token to
Feature activation+0.000
impact
Token impact
Feature activation+0.000
Alabama
Token Alabama
Feature activation+0.000
,
Token,
Feature activation+0.485
we
Token we
Feature activation+0.961
are
Token are
Feature activation+0.350
taking
Token taking
Feature activation+0.164
precautions
Token precautions
Feature activation+0.000
to
Token to
Feature activation+0.000
protect
Token protect
Feature activation+0.000
the
Token the
Feature activation+0.000
Fire
Token Fire
Feature activation+0.000
Rapt
Token Rapt
Feature activation+0.000
or
Tokenor
Feature activation+0.000
,
Token,
Feature activation+0.572
it
Token it
Feature activation+0.898
can
Token can
Feature activation+0.386
target
Token target
Feature activation+0.000
a
Token a
Feature activation+0.000
different
Token different
Feature activation+0.000
enemy
Token enemy
Feature activation+0.000

INTERVAL 0.739 - 0.862
CONTAINS 0.002%

out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
,
Token,
Feature activation+0.326
we
Token we
Feature activation+0.830
would
Token would
Feature activation+0.552
like
Token like
Feature activation+0.108
to
Token to
Feature activation+0.000
give
Token give
Feature activation+0.000
players
Token players
Feature activation+0.000
the
Token the
Feature activation+0.025
Model
Token Model
Feature activation+0.000
S
Token S
Feature activation+0.000
in
Token in
Feature activation+0.000
Oz
Token Oz
Feature activation+0.000
comes
Token comes
Feature activation+0.855
the
Token the
Feature activation+0.264
need
Token need
Feature activation+0.217
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
nationwide
Token nationwide
Feature activation+0.000
media
Token media
Feature activation+0.000
by
Token by
Feature activation+0.021
our
Token our
Feature activation+0.142
side
Token side
Feature activation+0.000
,
Token,
Feature activation+0.396
it
Token it
Feature activation+0.831
âĢ
TokenâĢ
Feature activation+0.318
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
possible
Token possible
Feature activation+0.039
now
Token now
Feature activation+0.000
at
Token at
Feature activation+0.000
B
Token B
Feature activation+0.000
MO
TokenMO
Feature activation+0.000
Field
Token Field
Feature activation+0.000
,
Token,
Feature activation+0.295
will
Token will
Feature activation+0.757
T
Token T
Feature activation+0.000
FC
TokenFC
Feature activation+0.000
look
Token look
Feature activation+0.398
to
Token to
Feature activation+0.000
play
Token play
Feature activation+0.000
from
Token from
Feature activation+0.175
the
Token the
Feature activation+0.332
cycl
Token cycl
Feature activation+0.015
one
Tokenone
Feature activation+0.000
,
Token,
Feature activation+0.432
we
Token we
Feature activation+0.754
don
Token don
Feature activation+0.432
't
Token't
Feature activation+0.055
really
Token really
Feature activation+0.000
need
Token need
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 0.616 - 0.739
CONTAINS 0.003%

of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
stuff
Token stuff
Feature activation+0.000
,
Token,
Feature activation+0.572
Nissan
Token Nissan
Feature activation+0.000
is
Token is
Feature activation+0.623
teasing
Token teasing
Feature activation+0.000
us
Token us
Feature activation+0.000
all
Token all
Feature activation+0.000
again
Token again
Feature activation+0.000
with
Token with
Feature activation+0.000
base
Token base
Feature activation+0.000
frame
Token frame
Feature activation+0.000
completed
Token completed
Feature activation+0.000
,
Token,
Feature activation+0.544
it
Token it
Feature activation+0.999
was
Token was
Feature activation+0.703
time
Token time
Feature activation+0.071
to
Token to
Feature activation+0.000
install
Token install
Feature activation+0.000
the
Token the
Feature activation+0.000
bench
Token bench
Feature activation+0.000
a
Token a
Feature activation+0.187
little
Token little
Feature activation+0.223
way
Token way
Feature activation+0.024
below
Token below
Feature activation+0.064
,
Token,
Feature activation+0.517
we
Token we
Feature activation+0.637
can
Token can
Feature activation+0.249
see
Token see
Feature activation+0.008
and
Token and
Feature activation+0.000
hear
Token hear
Feature activation+0.000
the
Token the
Feature activation+0.000
an
Token an
Feature activation+0.000
incredibly
Token incredibly
Feature activation+0.000
intense
Token intense
Feature activation+0.000
live
Token live
Feature activation+0.000
show
Token show
Feature activation+0.000
,
Token,
Feature activation+0.690
and
Token and
Feature activation+0.500
one
Token one
Feature activation+0.282
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
tight
Token tight
Feature activation+0.000
Work
Token Work
Feature activation+0.000
day
Tokenday
Feature activation+0.000
Talent
Token Talent
Feature activation+0.000
Ins
Token Ins
Feature activation+0.000
ights
Tokenights
Feature activation+0.000
,
Token,
Feature activation+0.624
which
Token which
Feature activation+0.828
launched
Token launched
Feature activation+0.414
on
Token on
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
,
Token,
Feature activation+0.013

INTERVAL 0.492 - 0.616
CONTAINS 0.005%

by
Token by
Feature activation+0.000
a
Token a
Feature activation+0.187
little
Token little
Feature activation+0.223
way
Token way
Feature activation+0.024
below
Token below
Feature activation+0.064
,
Token,
Feature activation+0.517
we
Token we
Feature activation+0.637
can
Token can
Feature activation+0.249
see
Token see
Feature activation+0.008
and
Token and
Feature activation+0.000
hear
Token hear
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
,
Token,
Feature activation+0.326
we
Token we
Feature activation+0.830
would
Token would
Feature activation+0.552
like
Token like
Feature activation+0.108
to
Token to
Feature activation+0.000
give
Token give
Feature activation+0.000
players
Token players
Feature activation+0.000
an
Token an
Feature activation+0.000
little
Token little
Feature activation+0.000
bit
Token bit
Feature activation+0.000
of
Token of
Feature activation+0.000
recognition
Token recognition
Feature activation+0.000
,
Token,
Feature activation+0.534
the
Token the
Feature activation+0.562
pair
Token pair
Feature activation+0.137
has
Token has
Feature activation+0.526
managed
Token managed
Feature activation+0.061
to
Token to
Feature activation+0.000
sucker
Token sucker
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
other
Token other
Feature activation+0.000
side
Token side
Feature activation+0.000
,
Token,
Feature activation+0.297
I
Token I
Feature activation+0.529
can
Token can
Feature activation+0.196
understand
Token understand
Feature activation+0.051
âĢĶ
TokenâĢĶ
Feature activation+0.000
sort
Tokensort
Feature activation+0.000
of
Token of
Feature activation+0.000
retirement
Token retirement
Feature activation+0.000
,
Token,
Feature activation+0.279
John
Token John
Feature activation+0.364
Wall
Token Wall
Feature activation+0.000
,
Token,
Feature activation+0.227
who
Token who
Feature activation+0.584
has
Token has
Feature activation+0.085
been
Token been
Feature activation+0.000
instrumental
Token instrumental
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 0.369 - 0.492
CONTAINS 0.009%

With
TokenWith
Feature activation+0.000
each
Token each
Feature activation+0.000
passing
Token passing
Feature activation+0.000
day
Token day
Feature activation+0.000
,
Token,
Feature activation+0.323
China
Token China
Feature activation+0.422
âĢ
TokenâĢ
Feature activation+0.279
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
patent
Token patent
Feature activation+0.000
system
Token system
Feature activation+0.000
hardware
Token hardware
Feature activation+0.000
.
Token.
Feature activation+0.000
With
Token With
Feature activation+0.000
sinking
Token sinking
Feature activation+0.000
heart
Token heart
Feature activation+0.000
,
Token,
Feature activation+0.390
I
Token I
Feature activation+0.748
compiled
Token compiled
Feature activation+0.343
the
Token the
Feature activation+0.186
latest
Token latest
Feature activation+0.000
kernel
Token kernel
Feature activation+0.000
turbulent
Token turbulent
Feature activation+0.000
financial
Token financial
Feature activation+0.000
forecasts
Token forecasts
Feature activation+0.000
of
Token of
Feature activation+0.000
late
Token late
Feature activation+0.000
,
Token,
Feature activation+0.449
OMG
Token OMG
Feature activation+0.254
!
Token!
Feature activation+0.000
Ubuntu
Token Ubuntu
Feature activation+0.000
!
Token!
Feature activation+0.000
writer
Token writer
Feature activation+0.000
with
Token with
Feature activation+0.000
Russia
Token Russia
Feature activation+0.000
regarding
Token regarding
Feature activation+0.000
Syria
Token Syria
Feature activation+0.000
,
Token,
Feature activation+0.316
its
Token its
Feature activation+0.424
obvious
Token obvious
Feature activation+0.116
anti
Token anti
Feature activation+0.000
-
Token-
Feature activation+0.000
Iran
TokenIran
Feature activation+0.000
and
Token and
Feature activation+0.000
economy
Token economy
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.023
rise
Token rise
Feature activation+0.000
,
Token,
Feature activation+0.426
the
Token the
Feature activation+0.460
neighborhood
Token neighborhood
Feature activation+0.000
is
Token is
Feature activation+0.377
reinvent
Token reinvent
Feature activation+0.000
ing
Tokening
Feature activation+0.000
itself
Token itself
Feature activation+0.000

INTERVAL 0.246 - 0.369
CONTAINS 0.013%

to
Token to
Feature activation+0.000
gay
Token gay
Feature activation+0.000
rights
Token rights
Feature activation+0.000
,
Token,
Feature activation+0.325
it
Token it
Feature activation+0.653
âĢ
TokenâĢ
Feature activation+0.318
Ļ
TokenĻ
Feature activation+0.024
s
Tokens
Feature activation+0.015
wonderful
Token wonderful
Feature activation+0.000
that
Token that
Feature activation+0.049
marriage
Token marriage
Feature activation+0.000
intense
Token intense
Feature activation+0.000
live
Token live
Feature activation+0.000
show
Token show
Feature activation+0.000
,
Token,
Feature activation+0.690
and
Token and
Feature activation+0.500
one
Token one
Feature activation+0.282
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
tight
Token tight
Feature activation+0.000
est
Tokenest
Feature activation+0.000
drum
Token drum
Feature activation+0.000
yet
Token yet
Feature activation+0.000
to
Token to
Feature activation+0.000
see
Token see
Feature activation+0.000
a
Token a
Feature activation+0.000
release
Token release
Feature activation+0.000
,
Token,
Feature activation+0.340
it
Token it
Feature activation+0.649
âĢ
TokenâĢ
Feature activation+0.179
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
hard
Token hard
Feature activation+0.000
With
TokenWith
Feature activation+0.000
back
Token back
Feature activation+0.000
feeling
Token feeling
Feature activation+0.000
better
Token better
Feature activation+0.000
,
Token,
Feature activation+0.265
Heat
Token Heat
Feature activation+0.322
point
Token point
Feature activation+0.000
guard
Token guard
Feature activation+0.000
Ben
Token Ben
Feature activation+0.000
o
Tokeno
Feature activation+0.000
U
Token U
Feature activation+0.000
this
Token this
Feature activation+0.473
week
Token week
Feature activation+0.283
,
Token,
Feature activation+0.795
clubs
Token clubs
Feature activation+0.502
across
Token across
Feature activation+0.380
the
Token the
Feature activation+0.348
country
Token country
Feature activation+0.000
have
Token have
Feature activation+0.514
switched
Token switched
Feature activation+0.096
on
Token on
Feature activation+0.000
damage
Token damage
Feature activation+0.000

INTERVAL 0.123 - 0.246
CONTAINS 0.018%

With
TokenWith
Feature activation+0.000
that
Token that
Feature activation+0.000
said
Token said
Feature activation+0.000
,
Token,
Feature activation+0.080
I
Token I
Feature activation+0.681
âĢ
TokenâĢ
Feature activation+0.160
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+0.000
always
Token always
Feature activation+0.000
been
Token been
Feature activation+0.000
impressed
Token impressed
Feature activation+0.000
.
Token.
Feature activation+0.000
With
Token With
Feature activation+0.000
any
Token any
Feature activation+0.000
luck
Token luck
Feature activation+0.000
it
Token it
Feature activation+0.153
will
Token will
Feature activation+0.236
reach
Token reach
Feature activation+0.000
out
Token out
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
few
Token few
Feature activation+0.000
and
Token and
Feature activation+0.000
Recreation
Token Recreation
Feature activation+0.000
,
Token,
Feature activation+0.062
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
it
Token it
Feature activation+0.150
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
no
Token no
Feature activation+0.000
wonder
Token wonder
Feature activation+0.000
with
Token with
Feature activation+0.000
cost
Token cost
Feature activation+0.000
contracts
Token contracts
Feature activation+0.000
,
Token,
Feature activation+0.000
all
Token all
Feature activation+0.086
you
Token you
Feature activation+0.211
do
Token do
Feature activation+0.098
is
Token is
Feature activation+0.000
modify
Token modify
Feature activation+0.000
it
Token it
Feature activation+0.000
and
Token and
Feature activation+0.000
examined
Token examined
Feature activation+0.000
decades
Token decades
Feature activation+0.000
of
Token of
Feature activation+0.000
clinical
Token clinical
Feature activation+0.000
trials
Token trials
Feature activation+0.000
,
Token,
Feature activation+0.224
Mon
Token Mon
Feature activation+0.000
c
Tokenc
Feature activation+0.000
rie
Tokenrie
Feature activation+0.000
ff
Tokenff
Feature activation+0.000
's
Token's
Feature activation+0.000

INTERVAL 0.000 - 0.123
CONTAINS 99.948%

celebrations
Token celebrations
Feature activation+0.000
the
Token the
Feature activation+0.000
title
Token title
Feature activation+0.000
Paris
Token Paris
Feature activation+0.000
Se
Token Se
Feature activation+0.000
Lib
Token Lib
Feature activation+0.000
ere
Tokenere
Feature activation+0.000
!
Token!
Feature activation+0.000
-
Token -
Feature activation+0.000
Paris
Token Paris
Feature activation+0.000
Liber
Token Liber
Feature activation+0.000
and
Token and
Feature activation+0.000
references
Token references
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
subject
Token subject
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
in
Token in
Feature activation+0.000
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.000
K
Token K
Feature activation+0.000
AT
TokenAT
Feature activation+0.000
U
TokenU
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Police
TokenPolice
Feature activation+0.000
said
Token said
Feature activation+0.000
protests
Token protests
Feature activation+0.000
.
Token.
Feature activation+0.000
Think
Token Think
Feature activation+0.000
things
Token things
Feature activation+0.000
through
Token through
Feature activation+0.000
before
Token before
Feature activation+0.000
you
Token you
Feature activation+0.000
use
Token use
Feature activation+0.000
your
Token your
Feature activation+0.000
Roth
Token Roth
Feature activation+0.000
IRA
Token IRA
Feature activation+0.000
as
Token as
Feature activation+0.000
o
Tokeno
Feature activation+0.000
(
Token (
Feature activation+0.000
18
Token18
Feature activation+0.000
48
Token48
Feature activation+0.000
),
Token),
Feature activation+0.000
by
Token by
Feature activation+0.000
Karl
Token Karl
Feature activation+0.000
Marx
Token Marx
Feature activation+0.000
and
Token and
Feature activation+0.000
Friedrich
Token Friedrich
Feature activation+0.000
Engels
Token Engels
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 H2.6: (feature 7118)

TOP ACTIVATIONS
MAX = 1.785

sellers
Token sellers
Feature activation+0.620
pre
Token pre
Feature activation+0.480
New
Token New
Feature activation+0.026
52
Token 52
Feature activation+0.117
,
Token,
Feature activation+1.753
the
Token the
Feature activation+1.785
rest
Token rest
Feature activation+1.287
of
Token of
Feature activation+0.388
the
Token the
Feature activation+0.429
publishing
Token publishing
Feature activation+0.215
line
Token line
Feature activation+0.022
e
Tokene
Feature activation+0.000
's
Token's
Feature activation+0.000
career
Token career
Feature activation+0.000
was
Token was
Feature activation+0.778
short
Token short
Feature activation+0.914
,
Token,
Feature activation+1.773
he
Token he
Feature activation+1.417
end
Token end
Feature activation+0.534
ures
Tokenures
Feature activation+0.132
as
Token as
Feature activation+0.746
one
Token one
Feature activation+0.357
strong
Token strong
Feature activation+0.983
sellers
Token sellers
Feature activation+0.620
pre
Token pre
Feature activation+0.480
New
Token New
Feature activation+0.026
52
Token 52
Feature activation+0.117
,
Token,
Feature activation+1.753
the
Token the
Feature activation+1.785
rest
Token rest
Feature activation+1.287
of
Token of
Feature activation+0.388
the
Token the
Feature activation+0.429
publishing
Token publishing
Feature activation+0.215
user
Token user
Feature activation+0.220
-
Token-
Feature activation+0.000
visible
Tokenvisible
Feature activation+0.239
today
Token today
Feature activation+0.457
,
Token,
Feature activation+1.463
it
Token it
Feature activation+1.733
p
Token p
Feature activation+0.000
aves
Tokenaves
Feature activation+0.000
the
Token the
Feature activation+0.728
way
Token way
Feature activation+0.555
for
Token for
Feature activation+0.214
arsenic
Token arsenic
Feature activation+0.000
removal
Token removal
Feature activation+0.065
have
Token have
Feature activation+0.957
taken
Token taken
Feature activation+0.604
place
Token place
Feature activation+0.264
,
Token,
Feature activation+1.716
we
Token we
Feature activation+1.436
must
Token must
Feature activation+1.140
consider
Token consider
Feature activation+0.843
variations
Token variations
Feature activation+0.307
in
Token in
Feature activation+0.094
severity
Token severity
Feature activation+0.187
of
Token of
Feature activation+0.000
cluster
Token cluster
Feature activation+0.000
headaches
Token headaches
Feature activation+0.083
,
Token,
Feature activation+1.433
most
Token most
Feature activation+1.672
appreciate
Token appreciate
Feature activation+1.170
just
Token just
Feature activation+0.530
how
Token how
Feature activation+0.535
painful
Token painful
Feature activation+0.000
and
Token and
Feature activation+0.000
"
Token "
Feature activation+0.061
SN
TokenSN
Feature activation+0.000
L
TokenL
Feature activation+0.000
"
Token"
Feature activation+0.000
host
Token host
Feature activation+0.484
,
Token,
Feature activation+1.646
he
Token he
Feature activation+1.464
's
Token's
Feature activation+0.676
no
Token no
Feature activation+0.420
stranger
Token stranger
Feature activation+0.106
to
Token to
Feature activation+0.000
build
Token build
Feature activation+0.000
up
Token up
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.037
referendum
Token referendum
Feature activation+0.000
,
Token,
Feature activation+1.644
Cameron
Token Cameron
Feature activation+0.493
drew
Token drew
Feature activation+0.903
huge
Token huge
Feature activation+0.301
applause
Token applause
Feature activation+0.000
for
Token for
Feature activation+0.000
reach
Tokenreach
Feature activation+0.000
of
Token of
Feature activation+0.000
political
Token political
Feature activation+0.000
correctness
Token correctness
Feature activation+0.020
,
Token,
Feature activation+1.479
he
Token he
Feature activation+1.630
told
Token told
Feature activation+1.037
Feast
Token Feast
Feature activation+0.000
of
Token of
Feature activation+0.000
Fun
Token Fun
Feature activation+0.000
Monday
Token Monday
Feature activation+0.000
's
Token's
Feature activation+0.450
fourth
Token fourth
Feature activation+0.516
largest
Token largest
Feature activation+0.546
economy
Token economy
Feature activation+0.376
,
Token,
Feature activation+1.233
only
Token only
Feature activation+1.628
about
Token about
Feature activation+0.420
half
Token half
Feature activation+0.208
of
Token of
Feature activation+0.078
its
Token its
Feature activation+0.077
women
Token women
Feature activation+0.000
the
Token the
Feature activation+0.000
Smart
Token Smart
Feature activation+0.000
Track
TokenTrack
Feature activation+0.000
package
Token package
Feature activation+0.000
)
Token)
Feature activation+0.390
appears
Token appears
Feature activation+1.601
in
Token in
Feature activation+0.415
a
Token a
Feature activation+0.312
later
Token later
Feature activation+0.044
map
Token map
Feature activation+0.000
,
Token,
Feature activation+1.183
age
Token age
Feature activation+0.148
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.203
Earth
Token Earth
Feature activation+0.000
,
Token,
Feature activation+1.278
neither
Token neither
Feature activation+1.578
has
Token has
Feature activation+0.659
publicly
Token publicly
Feature activation+0.001
cited
Token cited
Feature activation+0.166
the
Token the
Feature activation+0.000
figures
Token figures
Feature activation+0.000
the
Token the
Feature activation+0.000
reported
Token reported
Feature activation+0.086
effect
Token effect
Feature activation+0.384
is
Token is
Feature activation+0.562
small
Token small
Feature activation+1.141
,
Token,
Feature activation+1.577
our
Token our
Feature activation+1.062
results
Token results
Feature activation+0.829
suggest
Token suggest
Feature activation+1.116
that
Token that
Feature activation+0.704
under
Token under
Feature activation+0.000
most
Token most
Feature activation+0.344
of
Token of
Feature activation+0.003
these
Token these
Feature activation+0.136
claims
Token claims
Feature activation+0.397
,
Token,
Feature activation+1.115
I
Token I
Feature activation+1.567
can
Token can
Feature activation+0.553
âĢ
TokenâĢ
Feature activation+0.263
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.066
support
Token support
Feature activation+0.197
Dem
Token Dem
Feature activation+0.000
enc
Tokenenc
Feature activation+0.000
io
Tokenio
Feature activation+0.000
are
Token are
Feature activation+0.437
related
Token related
Feature activation+0.368
,
Token,
Feature activation+1.549
it
Token it
Feature activation+1.505
is
Token is
Feature activation+1.107
a
Token a
Feature activation+0.652
tragedy
Token tragedy
Feature activation+0.283
that
Token that
Feature activation+0.988
with
Token with
Feature activation+0.357
a
Token a
Feature activation+0.559
general
Token general
Feature activation+1.360
terror
Token terror
Feature activation+0.246
threat
Token threat
Feature activation+0.517
,
Token,
Feature activation+1.537
the
Token the
Feature activation+1.369
nation
Token nation
Feature activation+0.394
has
Token has
Feature activation+0.974
been
Token been
Feature activation+0.288
down
Token down
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.172
CO
Token CO
Feature activation+0.000
2
Token 2
Feature activation+0.000
problem
Token problem
Feature activation+0.391
,
Token,
Feature activation+1.528
under
Token under
Feature activation+1.196
some
Token some
Feature activation+0.975
conditions
Token conditions
Feature activation+0.086
using
Token using
Feature activation+0.305
CO
Token CO
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.238
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.359
win
Token win
Feature activation+0.049
,
Token,
Feature activation+1.030
he
Token he
Feature activation+1.512
never
Token never
Feature activation+0.982
finished
Token finished
Feature activation+0.102
lower
Token lower
Feature activation+0.172
than
Token than
Feature activation+0.000
sixth
Token sixth
Feature activation+0.000
enc
Tokenenc
Feature activation+0.000
io
Tokenio
Feature activation+0.000
are
Token are
Feature activation+0.437
related
Token related
Feature activation+0.368
,
Token,
Feature activation+1.549
it
Token it
Feature activation+1.505
is
Token is
Feature activation+1.107
a
Token a
Feature activation+0.652
tragedy
Token tragedy
Feature activation+0.283
that
Token that
Feature activation+0.988
two
Token two
Feature activation+0.000
intellectual
Token intellectual
Feature activation+0.040
growth
Token growth
Feature activation+0.177
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.278
University
Token University
Feature activation+0.141
,
Token,
Feature activation+1.500
rather
Token rather
Feature activation+1.270
than
Token than
Feature activation+0.644
negatively
Token negatively
Feature activation+0.023
towards
Token towards
Feature activation+0.003
the
Token the
Feature activation+0.125

Top DFA by src position
MAX = 2.239

the
Token the
Feature activation+0.000
Top resid features:
last
Token last
Feature activation-0.001
Top resid features:
few
Token few
Feature activation+0.007
Top resid features:
years
Token years
Feature activation+0.010
Top resid features:
.
Token.
Feature activation+0.024
Top resid features:
While
Token While
Feature activation+2.009
Top resid features:
the
Token the
Feature activation+0.012
Top resid features:
Batman
Token Batman
Feature activation+0.028
Top resid features:
and
Token and
Feature activation+0.022
Top resid features:
Green
Token Green
Feature activation+0.020
Top resid features:
Lantern
Token Lantern
Feature activation+0.034
Top resid features:
Puerto
Token Puerto
Feature activation+0.021
Top resid features:
Rican
Token Rican
Feature activation+0.017
Top resid features:
B
Token B
Feature activation+0.024
Top resid features:
SN
TokenSN
Feature activation+0.018
Top resid features:
.
Token.
Feature activation+0.026
Top resid features:
Although
Token Although
Feature activation+2.030
Top resid features:
Pow
Token Pow
Feature activation+0.048
Top resid features:
e
Tokene
Feature activation+0.012
Top resid features:
's
Token's
Feature activation+0.055
Top resid features:
career
Token career
Feature activation+0.060
Top resid features:
was
Token was
Feature activation+0.159
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
last
Token last
Feature activation-0.000
Top resid features:
few
Token few
Feature activation+0.007
Top resid features:
years
Token years
Feature activation+0.009
Top resid features:
.
Token.
Feature activation+0.026
Top resid features:
While
Token While
Feature activation+1.903
Top resid features:
the
Token the
Feature activation+0.015
Top resid features:
Batman
Token Batman
Feature activation+0.031
Top resid features:
and
Token and
Feature activation+0.024
Top resid features:
Green
Token Green
Feature activation+0.022
Top resid features:
Lantern
Token Lantern
Feature activation+0.034
Top resid features:
new
Token new
Feature activation+0.000
Top resid features:
Rust
Token Rust
Feature activation+0.009
Top resid features:
M
Token M
Feature activation+0.006
Top resid features:
IR
TokenIR
Feature activation+0.010
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
While
Token While
Feature activation+2.239
Top resid features:
this
Token this
Feature activation+0.026
Top resid features:
feature
Token feature
Feature activation+0.054
Top resid features:
does
Token does
Feature activation+0.050
Top resid features:
not
Token not
Feature activation+0.049
Top resid features:
result
Token result
Feature activation+0.026
Top resid features:
investigate
Token investigate
Feature activation+0.016
Top resid features:
useful
Token useful
Feature activation+0.002
Top resid features:
treatment
Token treatment
Feature activation+0.021
Top resid features:
options
Token options
Feature activation+0.022
Top resid features:
.
Token.
Feature activation+0.036
Top resid features:
While
Token While
Feature activation+1.877
Top resid features:
a
Token a
Feature activation+0.025
Top resid features:
number
Token number
Feature activation+0.038
Top resid features:
of
Token of
Feature activation+0.036
Top resid features:
technological
Token technological
Feature activation+0.043
Top resid features:
developments
Token developments
Feature activation+0.061
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.114
Top resid features:
Ċ
TokenĊ
Feature activation+0.019
Top resid features:
Although
TokenAlthough
Feature activation+1.898
Top resid features:
medical
Token medical
Feature activation+0.018
Top resid features:
professionals
Token professionals
Feature activation+0.024
Top resid features:
once
Token once
Feature activation+0.043
Top resid features:
doubted
Token doubted
Feature activation+0.139
Top resid features:
the
Token the
Feature activation+0.072
Top resid features:
Saturday
Token Saturday
Feature activation+0.005
Top resid features:
broadcast
Token broadcast
Feature activation+0.021
Top resid features:
.
Token.
Feature activation+0.018
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
Ċ
TokenĊ
Feature activation+0.012
Top resid features:
While
TokenWhile
Feature activation+1.891
Top resid features:
the
Token the
Feature activation+0.014
Top resid features:
episode
Token episode
Feature activation+0.031
Top resid features:
will
Token will
Feature activation+0.016
Top resid features:
mark
Token mark
Feature activation+0.035
Top resid features:
C
Token C
Feature activation+0.015
Top resid features:
.
Token.
Feature activation+0.021
Top resid features:
âĢ
TokenâĢ
Feature activation+0.021
Top resid features:
Ŀ
TokenĿ
Feature activation+0.008
Top resid features:
Ċ
TokenĊ
Feature activation+0.016
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
While
TokenWhile
Feature activation+1.958
Top resid features:
his
Token his
Feature activation+0.028
Top resid features:
diary
Token diary
Feature activation+0.024
Top resid features:
was
Token was
Feature activation+0.047
Top resid features:
in
Token in
Feature activation+0.024
Top resid features:
a
Token a
Feature activation+0.016
Top resid features:
AD
TokenAD
Feature activation+0.013
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.022
Top resid features:
find
Token find
Feature activation+0.014
Top resid features:
derogatory
Token derogatory
Feature activation+0.015
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
Although
Token Although
Feature activation+2.048
Top resid features:
Hon
Token Hon
Feature activation+0.019
Top resid features:
ard
Tokenard
Feature activation+0.009
Top resid features:
's
Token's
Feature activation+0.018
Top resid features:
initial
Token initial
Feature activation+0.024
Top resid features:
video
Token video
Feature activation+0.034
Top resid features:
't
Token't
Feature activation+0.006
Top resid features:
use
Token use
Feature activation+0.019
Top resid features:
the
Token the
Feature activation+0.005
Top resid features:
help
Token help
Feature activation+0.021
Top resid features:
.
Token.
Feature activation+0.052
Top resid features:
Though
Token Though
Feature activation+1.839
Top resid features:
South
Token South
Feature activation+0.025
Top resid features:
Korea
Token Korea
Feature activation+0.034
Top resid features:
is
Token is
Feature activation+0.101
Top resid features:
Asia
Token Asia
Feature activation+0.033
Top resid features:
's
Token's
Feature activation+0.102
Top resid features:
irmed
Tokenirmed
Feature activation-0.004
Top resid features:
up
Token up
Feature activation+0.014
Top resid features:
.
Token.
Feature activation+0.022
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
Ċ
TokenĊ
Feature activation+0.023
Top resid features:
Although
TokenAlthough
Feature activation+1.976
Top resid features:
the
Token the
Feature activation+0.013
Top resid features:
E
Token E
Feature activation+0.014
Top resid features:
gl
Tokengl
Feature activation+0.015
Top resid features:
inton
Tokeninton
Feature activation+0.008
Top resid features:
West
Token West
Feature activation+0.022
Top resid features:
years
Token years
Feature activation+0.012
Top resid features:
old
Token old
Feature activation+0.014
Top resid features:
,
Token,
Feature activation+0.022
Top resid features:
respectively
Token respectively
Feature activation+0.014
Top resid features:
.
Token.
Feature activation+0.012
Top resid features:
While
Token While
Feature activation+1.855
Top resid features:
both
Token both
Feature activation+0.044
Top resid features:
men
Token men
Feature activation+0.026
Top resid features:
have
Token have
Feature activation+0.023
Top resid features:
made
Token made
Feature activation+0.028
Top resid features:
vague
Token vague
Feature activation+0.021
Top resid features:
unaffected
Token unaffected
Feature activation+0.023
Top resid features:
by
Token by
Feature activation+0.015
Top resid features:
scar
Token scar
Feature activation+0.013
Top resid features:
ring
Tokenring
Feature activation+0.018
Top resid features:
.
Token.
Feature activation+0.036
Top resid features:
Though
Token Though
Feature activation+1.482
Top resid features:
the
Token the
Feature activation+0.056
Top resid features:
reported
Token reported
Feature activation+0.190
Top resid features:
effect
Token effect
Feature activation+0.110
Top resid features:
is
Token is
Feature activation+0.214
Top resid features:
small
Token small
Feature activation+0.211
Top resid features:
Australia
Token Australia
Feature activation-0.000
Top resid features:
back
Token back
Feature activation+0.013
Top resid features:
fifty
Token fifty
Feature activation+0.002
Top resid features:
years
Token years
Feature activation+0.015
Top resid features:
.
Token.
Feature activation+0.051
Top resid features:
Although
Token Although
Feature activation+1.908
Top resid features:
I
Token I
Feature activation+0.010
Top resid features:
agree
Token agree
Feature activation+0.090
Top resid features:
with
Token with
Feature activation+0.056
Top resid features:
most
Token most
Feature activation+0.056
Top resid features:
of
Token of
Feature activation+0.053
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
âĢ
TokenâĢ
Feature activation+0.020
Top resid features:
ľ
Tokenľ
Feature activation+0.001
Top resid features:
While
TokenWhile
Feature activation+1.855
Top resid features:
I
Token I
Feature activation-0.001
Top resid features:
do
Token do
Feature activation+0.014
Top resid features:
not
Token not
Feature activation+0.039
Top resid features:
believe
Token believe
Feature activation+0.042
Top resid features:
the
Token the
Feature activation+0.020
Top resid features:
T
Token T
Feature activation+0.010
Top resid features:
ERROR
TokenERROR
Feature activation+0.020
Top resid features:
TH
Token TH
Feature activation+0.018
Top resid features:
RE
TokenRE
Feature activation+0.010
Top resid features:
AT
TokenAT
Feature activation+0.012
Top resid features:
While
Token While
Feature activation+1.808
Top resid features:
sunny
Token sunny
Feature activation+0.029
Top resid features:
Spain
Token Spain
Feature activation+0.014
Top resid features:
is
Token is
Feature activation+0.080
Top resid features:
dealing
Token dealing
Feature activation+0.046
Top resid features:
with
Token with
Feature activation+0.042
Top resid features:
It
Token It
Feature activation+0.010
Top resid features:
says
Token says
Feature activation+0.020
Top resid features:
:
Token:
Feature activation+0.049
Top resid features:
âĢ
Token âĢ
Feature activation+0.040
Top resid features:
ľ
Tokenľ
Feature activation+0.012
Top resid features:
Although
TokenAlthough
Feature activation+1.862
Top resid features:
only
Token only
Feature activation+0.052
Top resid features:
a
Token a
Feature activation+0.034
Top resid features:
partial
Token partial
Feature activation+0.044
Top resid features:
solution
Token solution
Feature activation+0.055
Top resid features:
to
Token to
Feature activation+0.046
Top resid features:
races
Token races
Feature activation+0.022
Top resid features:
at
Token at
Feature activation+0.015
Top resid features:
Dar
Token Dar
Feature activation+0.019
Top resid features:
lington
Tokenlington
Feature activation+0.009
Top resid features:
.
Token.
Feature activation+0.045
Top resid features:
While
Token While
Feature activation+1.886
Top resid features:
he
Token he
Feature activation+0.048
Top resid features:
didn
Token didn
Feature activation+0.054
Top resid features:
âĢ
TokenâĢ
Feature activation+0.066
Top resid features:
Ļ
TokenĻ
Feature activation+0.016
Top resid features:
t
Tokent
Feature activation+0.045
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.007
Top resid features:
âĢ
TokenâĢ
Feature activation+0.018
Top resid features:
ľ
Tokenľ
Feature activation-0.000
Top resid features:
While
TokenWhile
Feature activation+1.884
Top resid features:
I
Token I
Feature activation-0.008
Top resid features:
do
Token do
Feature activation+0.013
Top resid features:
not
Token not
Feature activation+0.029
Top resid features:
believe
Token believe
Feature activation+0.040
Top resid features:
the
Token the
Feature activation+0.016
Top resid features:
Folder
Token Folder
Feature activation+0.017
Top resid features:
13
Token 13
Feature activation+0.004
Top resid features:
)
Token)
Feature activation+0.025
Top resid features:
Ċ
TokenĊ
Feature activation+0.017
Top resid features:
Ċ
TokenĊ
Feature activation+0.014
Top resid features:
While
TokenWhile
Feature activation+1.826
Top resid features:
the
Token the
Feature activation+0.009
Top resid features:
report
Token report
Feature activation+0.043
Top resid features:
claimed
Token claimed
Feature activation+0.068
Top resid features:
to
Token to
Feature activation+0.016
Top resid features:
be
Token be
Feature activation+0.020
Top resid features:

Decoder Weights Distribution

Head 0: 0.05

Head 1: 0.05

Head 2: 0.06

Head 3: 0.05

Head 4: 0.03

Head 5: 0.08

Head 6: 0.28

Head 7: 0.09

Head 8: 0.04

Head 9: 0.04

Head 10: 0.16

Head 11: 0.08

Positive logits

nevertheless2.07

nonetheless1.75

notations1.57

fortunately1.46

undeniably1.46

etheless1.45

drawbacks1.41

technically1.41

anecd1.39

Nonetheless1.37

unavoid1.35

)1.32

anecdotal1.32

ć1.30

Nonetheless1.30

noticeably1.29

caveats1.28

Nevertheless1.28

hindsight1.27

intrinsically1.25

Negative logits

hai-1.69

shit-1.34

wom-1.33

Sabha-1.30

ansas-1.28

hs-1.27

ika-1.26

icer-1.24

atche-1.22

aryn-1.20

Sha-1.17

SourceFile-1.16

thia-1.16

chant-1.16

func-1.15

xf-1.14

ayne-1.14

uilding-1.13

ixt-1.13

cast-1.12

INTERVAL 1.607 - 1.785
CONTAINS 0.001%

"
Token "
Feature activation+0.061
SN
TokenSN
Feature activation+0.000
L
TokenL
Feature activation+0.000
"
Token"
Feature activation+0.000
host
Token host
Feature activation+0.484
,
Token,
Feature activation+1.646
he
Token he
Feature activation+1.464
's
Token's
Feature activation+0.676
no
Token no
Feature activation+0.420
stranger
Token stranger
Feature activation+0.106
to
Token to
Feature activation+0.000
e
Tokene
Feature activation+0.000
's
Token's
Feature activation+0.000
career
Token career
Feature activation+0.000
was
Token was
Feature activation+0.778
short
Token short
Feature activation+0.914
,
Token,
Feature activation+1.773
he
Token he
Feature activation+1.417
end
Token end
Feature activation+0.534
ures
Tokenures
Feature activation+0.132
as
Token as
Feature activation+0.746
one
Token one
Feature activation+0.357
build
Token build
Feature activation+0.000
up
Token up
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.037
referendum
Token referendum
Feature activation+0.000
,
Token,
Feature activation+1.644
Cameron
Token Cameron
Feature activation+0.493
drew
Token drew
Feature activation+0.903
huge
Token huge
Feature activation+0.301
applause
Token applause
Feature activation+0.000
for
Token for
Feature activation+0.000
arsenic
Token arsenic
Feature activation+0.000
removal
Token removal
Feature activation+0.065
have
Token have
Feature activation+0.957
taken
Token taken
Feature activation+0.604
place
Token place
Feature activation+0.264
,
Token,
Feature activation+1.716
we
Token we
Feature activation+1.436
must
Token must
Feature activation+1.140
consider
Token consider
Feature activation+0.843
variations
Token variations
Feature activation+0.307
in
Token in
Feature activation+0.094
reach
Tokenreach
Feature activation+0.000
of
Token of
Feature activation+0.000
political
Token political
Feature activation+0.000
correctness
Token correctness
Feature activation+0.020
,
Token,
Feature activation+1.479
he
Token he
Feature activation+1.630
told
Token told
Feature activation+1.037
Feast
Token Feast
Feature activation+0.000
of
Token of
Feature activation+0.000
Fun
Token Fun
Feature activation+0.000
Monday
Token Monday
Feature activation+0.000

INTERVAL 1.428 - 1.607
CONTAINS 0.002%

âĢ
TokenâĢ
Feature activation+0.238
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.359
win
Token win
Feature activation+0.049
,
Token,
Feature activation+1.030
he
Token he
Feature activation+1.512
never
Token never
Feature activation+0.982
finished
Token finished
Feature activation+0.102
lower
Token lower
Feature activation+0.172
than
Token than
Feature activation+0.000
sixth
Token sixth
Feature activation+0.000
age
Token age
Feature activation+0.148
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.203
Earth
Token Earth
Feature activation+0.000
,
Token,
Feature activation+1.278
neither
Token neither
Feature activation+1.578
has
Token has
Feature activation+0.659
publicly
Token publicly
Feature activation+0.001
cited
Token cited
Feature activation+0.166
the
Token the
Feature activation+0.000
figures
Token figures
Feature activation+0.000
Dem
Token Dem
Feature activation+0.000
enc
Tokenenc
Feature activation+0.000
io
Tokenio
Feature activation+0.000
are
Token are
Feature activation+0.437
related
Token related
Feature activation+0.368
,
Token,
Feature activation+1.549
it
Token it
Feature activation+1.505
is
Token is
Feature activation+1.107
a
Token a
Feature activation+0.652
tragedy
Token tragedy
Feature activation+0.283
that
Token that
Feature activation+0.988
t
Tokent
Feature activation+0.442
a
Token a
Feature activation+0.776
brand
Token brand
Feature activation+0.303
new
Token new
Feature activation+0.426
one
Token one
Feature activation+0.637
,
Token,
Feature activation+1.438
C
Token C
Feature activation+0.134
TA
TokenTA
Feature activation+0.000
is
Token is
Feature activation+1.018
releasing
Token releasing
Feature activation+0.193
a
Token a
Feature activation+0.182
class
Tokenclass
Feature activation+0.000
talents
Token talents
Feature activation+0.115
on
Token on
Feature activation+0.000
their
Token their
Feature activation+0.071
books
Token books
Feature activation+0.000
,
Token,
Feature activation+1.455
the
Token the
Feature activation+1.217
next
Token next
Feature activation+0.141
man
Token man
Feature activation+0.000
taking
Token taking
Feature activation+0.084
over
Token over
Feature activation+0.000

INTERVAL 1.250 - 1.428
CONTAINS 0.003%

to
Token to
Feature activation+0.050
endorse
Token endorse
Feature activation+0.273
Sanders
Token Sanders
Feature activation+0.107
'
Token'
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
,
Token,
Feature activation+1.253
and
Token and
Feature activation+0.969
scholarships
Token scholarships
Feature activation+0.000
were
Token were
Feature activation+0.488
created
Token created
Feature activation+0.000
.
Token.
Feature activation+0.000
bearer
Token bearer
Feature activation+0.000
of
Token of
Feature activation+0.000
bad
Token bad
Feature activation+0.081
news
Token news
Feature activation+0.247
,
Token,
Feature activation+1.160
I
Token I
Feature activation+1.385
felt
Token felt
Feature activation+0.676
obligated
Token obligated
Feature activation+0.195
to
Token to
Feature activation+0.030
clue
Token clue
Feature activation+0.079
him
Token him
Feature activation+0.000
ants
Tokenants
Feature activation+0.000
'
Token'
Feature activation+0.211
in
Token in
Feature activation+0.505
the
Token the
Feature activation+0.335
city
Token city
Feature activation+0.000
,
Token,
Feature activation+1.335
eagle
Token eagle
Feature activation+0.210
-
Token-
Feature activation+0.000
eyed
Tokeneyed
Feature activation+0.016
viewers
Token viewers
Feature activation+0.407
people
Token people
Feature activation+0.267
pre
Token pre
Feature activation+0.480
New
Token New
Feature activation+0.026
52
Token 52
Feature activation+0.117
,
Token,
Feature activation+1.753
the
Token the
Feature activation+1.785
rest
Token rest
Feature activation+1.287
of
Token of
Feature activation+0.388
the
Token the
Feature activation+0.429
publishing
Token publishing
Feature activation+0.215
line
Token line
Feature activation+0.022
had
Token had
Feature activation+0.953
the
Token the
Feature activation+0.125
eating
Token eating
Feature activation+0.000
clubs
Token clubs
Feature activation+0.000
,
Token,
Feature activation+0.820
Wilson
Token Wilson
Feature activation+0.457
did
Token did
Feature activation+1.284
make
Token make
Feature activation+0.258
his
Token his
Feature activation+0.041
feelings
Token feelings
Feature activation+0.000
clear
Token clear
Feature activation+0.000
:
Token:
Feature activation+0.136

INTERVAL 1.071 - 1.250
CONTAINS 0.005%

the
Token the
Feature activation+0.172
CO
Token CO
Feature activation+0.000
2
Token 2
Feature activation+0.000
problem
Token problem
Feature activation+0.391
,
Token,
Feature activation+1.528
under
Token under
Feature activation+1.196
some
Token some
Feature activation+0.975
conditions
Token conditions
Feature activation+0.086
using
Token using
Feature activation+0.305
CO
Token CO
Feature activation+0.000
2
Token 2
Feature activation+0.000
Asia
Token Asia
Feature activation+0.242
's
Token's
Feature activation+0.450
fourth
Token fourth
Feature activation+0.516
largest
Token largest
Feature activation+0.546
economy
Token economy
Feature activation+0.376
,
Token,
Feature activation+1.233
only
Token only
Feature activation+1.628
about
Token about
Feature activation+0.420
half
Token half
Feature activation+0.208
of
Token of
Feature activation+0.078
its
Token its
Feature activation+0.077
is
Token is
Feature activation+0.562
small
Token small
Feature activation+1.141
,
Token,
Feature activation+1.577
our
Token our
Feature activation+1.062
results
Token results
Feature activation+0.829
suggest
Token suggest
Feature activation+1.116
that
Token that
Feature activation+0.704
under
Token under
Feature activation+0.000
certain
Token certain
Feature activation+0.000
circumstances
Token circumstances
Feature activation+0.000
scars
Token scars
Feature activation+0.000
have
Token have
Feature activation+0.957
taken
Token taken
Feature activation+0.604
place
Token place
Feature activation+0.264
,
Token,
Feature activation+1.716
we
Token we
Feature activation+1.436
must
Token must
Feature activation+1.140
consider
Token consider
Feature activation+0.843
variations
Token variations
Feature activation+0.307
in
Token in
Feature activation+0.094
sources
Token sources
Feature activation+0.000
and
Token and
Feature activation+0.148
could
Token could
Feature activation+0.574
mean
Token mean
Feature activation+1.295
more
Token more
Feature activation+0.533
housing
Token housing
Feature activation+0.000
,
Token,
Feature activation+1.269
the
Token the
Feature activation+1.157
host
Token host
Feature activation+0.019
of
Token of
Feature activation+0.020
environmental
Token environmental
Feature activation+0.000
problems
Token problems
Feature activation+0.000
it
Token it
Feature activation+0.289

INTERVAL 0.893 - 1.071
CONTAINS 0.007%

:
Token:
Feature activation+0.000
The
Token The
Feature activation+0.000
Old
Token Old
Feature activation+0.000
Republic
Token Republic
Feature activation+0.000
,
Token,
Feature activation+0.402
I
Token I
Feature activation+0.933
have
Token have
Feature activation+0.103
to
Token to
Feature activation+0.000
honestly
Token honestly
Feature activation+0.000
admit
Token admit
Feature activation+0.000
the
Token the
Feature activation+0.000
s
Tokens
Feature activation+0.122
,
Token,
Feature activation+1.145
the
Token the
Feature activation+1.080
fact
Token fact
Feature activation+1.201
is
Token is
Feature activation+0.865
that
Token that
Feature activation+0.980
the
Token the
Feature activation+0.109
toxicity
Token toxicity
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
experience
Token experience
Feature activation+0.000
the
Token the
Feature activation+0.251
attention
Token attention
Feature activation+0.394
,
Token,
Feature activation+1.028
the
Token the
Feature activation+1.007
trolls
Token trolls
Feature activation+0.414
they
Token they
Feature activation+0.940
left
Token left
Feature activation+0.280
behind
Token behind
Feature activation+0.000
on
Token on
Feature activation+0.000
4
Token 4
Feature activation+0.000
chan
Tokenchan
Feature activation+0.000
with
Token with
Feature activation+0.000
Siri
Token Siri
Feature activation+0.000
is
Token is
Feature activation+0.325
relatively
Token relatively
Feature activation+0.196
short
Token short
Feature activation+0.000
,
Token,
Feature activation+0.974
Siri
Token Siri
Feature activation+0.058
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
usage
Token usage
Feature activation+0.000
was
Token was
Feature activation+0.000
performing
Token performing
Feature activation+0.000
newer
Token newer
Feature activation+0.000
songs
Token songs
Feature activation+0.000
,
Token,
Feature activation+0.518
some
Token some
Feature activation+0.901
attendees
Token attendees
Feature activation+0.000
chat
Token chat
Feature activation+0.000
tered
Tokentered
Feature activation+0.000
,
Token,
Feature activation+0.000
others
Token others
Feature activation+0.000

INTERVAL 0.714 - 0.893
CONTAINS 0.009%

theory
Token theory
Feature activation+0.047
of
Token of
Feature activation+0.050
consciousness
Token consciousness
Feature activation+0.000
,
Token,
Feature activation+0.830
there
Token there
Feature activation+1.113
is
Token is
Feature activation+0.785
a
Token a
Feature activation+0.440
widespread
Token widespread
Feature activation+0.373
,
Token,
Feature activation+0.476
if
Token if
Feature activation+1.007
less
Token less
Feature activation+0.000
the
Token the
Feature activation+0.377
classics
Token classics
Feature activation+0.378
,
Token,
Feature activation+1.316
J
Token J
Feature activation+0.369
ett
Tokenett
Feature activation+0.000
and
Token and
Feature activation+0.803
the
Token the
Feature activation+0.411
Black
Token Black
Feature activation+0.000
he
Tokenhe
Feature activation+0.000
arts
Tokenarts
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.296
short
Token short
Feature activation+0.914
,
Token,
Feature activation+1.773
he
Token he
Feature activation+1.417
end
Token end
Feature activation+0.534
ures
Tokenures
Feature activation+0.132
as
Token as
Feature activation+0.746
one
Token one
Feature activation+0.357
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
most
Token most
Feature activation+0.063
memorable
Token memorable
Feature activation+0.074
are
Token are
Feature activation+0.233
celebrating
Token celebrating
Feature activation+0.339
your
Token your
Feature activation+0.334
happy
Token happy
Feature activation+0.203
birthday
Token birthday
Feature activation+0.000
and
Token and
Feature activation+0.785
the
Token the
Feature activation+0.593
anniversary
Token anniversary
Feature activation+0.161
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.168
glorious
Token glorious
Feature activation+0.038
While
Token While
Feature activation+0.000
I
Token I
Feature activation+0.000
was
Token was
Feature activation+0.000
doing
Token doing
Feature activation+0.017
this
Token this
Feature activation+0.232
,
Token,
Feature activation+0.790
a
Token a
Feature activation+0.749
kind
Token kind
Feature activation+0.289
fellow
Token fellow
Feature activation+0.159
sk
Token sk
Feature activation+0.017
ier
Tokenier
Feature activation+0.117

INTERVAL 0.536 - 0.714
CONTAINS 0.012%

wealth
Token wealth
Feature activation+0.000
,
Token,
Feature activation+0.388
F
Token F
Feature activation+0.340
ogg
Tokenogg
Feature activation+0.174
lives
Token lives
Feature activation+0.465
a
Token a
Feature activation+0.633
modest
Token modest
Feature activation+0.458
life
Token life
Feature activation+0.233
with
Token with
Feature activation+0.292
habits
Token habits
Feature activation+0.015
carried
Token carried
Feature activation+0.238
all
Token all
Feature activation+0.437
of
Token of
Feature activation+0.065
the
Token the
Feature activation+0.143
"
Token "
Feature activation+0.471
love
Tokenlove
Feature activation+0.000
,
Token,
Feature activation+0.563
likes
Token likes
Feature activation+0.493
,
Token,
Feature activation+0.192
and
Token and
Feature activation+0.451
kind
Token kind
Feature activation+0.000
words
Token words
Feature activation+0.000
are
Token are
Feature activation+0.000
traditionally
Token traditionally
Feature activation+0.000
kept
Token kept
Feature activation+0.345
private
Token private
Feature activation+0.000
,
Token,
Feature activation+0.944
and
Token and
Feature activation+0.565
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
presidential
Token presidential
Feature activation+0.000
the
Token the
Feature activation+0.000
number
Token number
Feature activation+0.000
of
Token of
Feature activation+0.000
tax
Token tax
Feature activation+0.000
brackets
Token brackets
Feature activation+0.000
,
Token,
Feature activation+0.574
he
Token he
Feature activation+0.485
criticized
Token criticized
Feature activation+0.179
him
Token him
Feature activation+0.000
for
Token for
Feature activation+0.000
not
Token not
Feature activation+0.000
the
Token the
Feature activation+0.000
UN
Token UN
Feature activation+0.000
,
Token,
Feature activation+0.202
NATO
Token NATO
Feature activation+0.265
,
Token,
Feature activation+0.374
and
Token and
Feature activation+0.664
the
Token the
Feature activation+0.327
US
Token US
Feature activation+0.000
have
Token have
Feature activation+0.370
all
Token all
Feature activation+0.248
gone
Token gone
Feature activation+0.000

INTERVAL 0.357 - 0.536
CONTAINS 0.032%

the
Token the
Feature activation+0.000
report
Token report
Feature activation+0.000
claimed
Token claimed
Feature activation+0.799
to
Token to
Feature activation+0.078
be
Token be
Feature activation+0.242
focused
Token focused
Feature activation+0.372
positively
Token positively
Feature activation+0.512
towards
Token towards
Feature activation+0.397
the
Token the
Feature activation+0.597
intellectual
Token intellectual
Feature activation+0.040
growth
Token growth
Feature activation+0.177
implement
Token implement
Feature activation+0.000
such
Token such
Feature activation+0.000
a
Token a
Feature activation+0.000
program
Token program
Feature activation+0.000
,
Token,
Feature activation+0.606
any
Token any
Feature activation+0.409
materials
Token materials
Feature activation+0.000
developed
Token developed
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
program
Token program
Feature activation+0.000
a
Token a
Feature activation+0.000
steep
Token steep
Feature activation+0.061
precip
Token precip
Feature activation+0.000
ice
Tokenice
Feature activation+0.000
bounds
Token bounds
Feature activation+0.091
it
Token it
Feature activation+0.414
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.026
west
Token west
Feature activation+0.306
side
Token side
Feature activation+0.185
,
Token,
Feature activation+0.450
,
Token,
Feature activation+0.946
and
Token and
Feature activation+0.794
books
Token books
Feature activation+0.000
,
Token,
Feature activation+0.305
none
Token none
Feature activation+1.170
have
Token have
Feature activation+0.480
focused
Token focused
Feature activation+0.093
directly
Token directly
Feature activation+0.000
on
Token on
Feature activation+0.000
lowering
Token lowering
Feature activation+0.000
scores
Token scores
Feature activation+0.000
his
Token his
Feature activation+0.000
career
Token career
Feature activation+0.000
,
Token,
Feature activation+0.000
though
Token though
Feature activation+0.197
he
Token he
Feature activation+0.000
said
Token said
Feature activation+0.455
that
Token that
Feature activation+0.000
they
Token they
Feature activation+0.000
didn
Token didn
Feature activation+0.000
't
Token't
Feature activation+0.016
transport
Token transport
Feature activation+0.000

INTERVAL 0.179 - 0.357
CONTAINS 0.056%

lington
Tokenlington
Feature activation+0.000
.
Token.
Feature activation+0.000
While
Token While
Feature activation+0.174
he
Token he
Feature activation+0.223
didn
Token didn
Feature activation+0.452
âĢ
TokenâĢ
Feature activation+0.238
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.359
win
Token win
Feature activation+0.049
,
Token,
Feature activation+1.030
he
Token he
Feature activation+1.512
's
Token's
Feature activation+0.000
own
Token own
Feature activation+0.000
,
Token,
Feature activation+0.698
a
Token a
Feature activation+0.660
party
Token party
Feature activation+0.338
line
Token line
Feature activation+0.317
vote
Token vote
Feature activation+0.347
forced
Token forced
Feature activation+0.462
her
Token her
Feature activation+0.120
off
Token off
Feature activation+0.000
the
Token the
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
from
Token from
Feature activation+0.000
time
Token time
Feature activation+0.000
to
Token to
Feature activation+0.000
time
Token time
Feature activation+0.000
,
Token,
Feature activation+0.351
you
Token you
Feature activation+0.000
should
Token should
Feature activation+0.000
never
Token never
Feature activation+0.000
trim
Token trim
Feature activation+0.000
cat
Token cat
Feature activation+0.000
Germans
Token Germans
Feature activation+0.000
are
Token are
Feature activation+0.129
calling
Token calling
Feature activation+0.157
for
Token for
Feature activation+0.000
their
Token their
Feature activation+0.061
troops
Token troops
Feature activation+0.278
to
Token to
Feature activation+0.028
be
Token be
Feature activation+0.181
withdrawn
Token withdrawn
Feature activation+0.041
from
Token from
Feature activation+0.000
Afghanistan
Token Afghanistan
Feature activation+0.000
cl
Tokencl
Feature activation+0.000
am
Tokenam
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
,
Token,
Feature activation+0.075
incorporation
Token incorporation
Feature activation+0.000
does
Token does
Feature activation+0.271
not
Token not
Feature activation+0.000
imper
Token imper
Feature activation+0.000
il
Tokenil
Feature activation+0.000
every
Token every
Feature activation+0.000
law
Token law
Feature activation+0.000

INTERVAL 0.000 - 0.179
CONTAINS 99.874%

become
Token become
Feature activation+0.000
our
Token our
Feature activation+0.000
meat
Token meat
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
fat
Token fat
Feature activation+0.000
,
Token,
Feature activation+0.000
most
Token most
Feature activation+0.000
of
Token of
Feature activation+0.000
which
Token which
Feature activation+0.000
ends
Token ends
Feature activation+0.000
en
Tokenen
Feature activation+0.000
Ramsey
Token Ramsey
Feature activation+0.000
.
Token.
Feature activation+0.000
*
Token *
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Jaguars
Token Jaguars
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
second
Token second
Feature activation+0.000
-
Token-
Feature activation+0.000
year
Tokenyear
Feature activation+0.000
t
Tokent
Feature activation+0.000
be
Token be
Feature activation+0.000
as
Token as
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
ch
Tokench
Feature activation+0.000
arming
Tokenarming
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
in
Token in
Feature activation+0.000
this
Token this
Feature activation+0.000
linked
Token linked
Feature activation+0.000
to
Token to
Feature activation+0.000
this
Token this
Feature activation+0.000
issue
Token issue
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Topics
TokenTopics
Feature activation+0.000
:
Token:
Feature activation+0.000
federal
Token federal
Feature activation+0.000
-
Token-
Feature activation+0.000
,
Token,
Feature activation+0.000
having
Token having
Feature activation+0.000
particular
Token particular
Feature activation+0.000
appeal
Token appeal
Feature activation+0.000
to
Token to
Feature activation+0.000
men
Token men
Feature activation+0.000
.
Token.
Feature activation+0.000
Democratic
Token Democratic
Feature activation+0.000
strategist
Token strategist
Feature activation+0.000
James
Token James
Feature activation+0.000
Car
Token Car
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 H2.6: (feature 18725)

TOP ACTIVATIONS
MAX = 2.446

change
Token change
Feature activation+0.000
the
Token the
Feature activation+0.030
marijuana
Token marijuana
Feature activation+0.000
laws
Token laws
Feature activation+0.178
,
Token,
Feature activation+1.875
go
Token go
Feature activation+2.446
ahead
Token ahead
Feature activation+1.133
and
Token and
Feature activation+0.862
change
Token change
Feature activation+0.757
the
Token the
Feature activation+0.500
national
Token national
Feature activation+0.000
preferred
Token preferred
Feature activation+0.000
environment
Token environment
Feature activation+0.000
or
Token or
Feature activation+0.000
language
Token language
Feature activation+0.000
,
Token,
Feature activation+1.145
try
Token try
Feature activation+2.157
implementing
Token implementing
Feature activation+0.046
your
Token your
Feature activation+0.336
own
Token own
Feature activation+0.000
Common
Token Common
Feature activation+0.000
Mark
TokenMark
Feature activation+0.000
environmental
Token environmental
Feature activation+0.000
news
Token news
Feature activation+0.134
junk
Token junk
Feature activation+0.013
ie
Tokenie
Feature activation+0.000
,
Token,
Feature activation+1.349
you
Token you
Feature activation+2.007
already
Token already
Feature activation+0.441
know
Token know
Feature activation+0.566
that
Token that
Feature activation+0.653
the
Token the
Feature activation+0.000
past
Token past
Feature activation+0.000
and
Token and
Feature activation+0.517
rem
Token rem
Feature activation+0.015
ade
Tokenade
Feature activation+0.039
it
Token it
Feature activation+0.843
,
Token,
Feature activation+1.101
would
Token would
Feature activation+1.967
you
Token you
Feature activation+0.515
be
Token be
Feature activation+0.000
asking
Token asking
Feature activation+0.000
me
Token me
Feature activation+0.000
the
Token the
Feature activation+0.000
anti
Token anti
Feature activation+0.000
-
Token-
Feature activation+0.000
trust
Tokentrust
Feature activation+0.000
issue
Token issue
Feature activation+0.000
,
Token,
Feature activation+0.977
why
Token why
Feature activation+1.899
aren
Token aren
Feature activation+0.523
't
Token't
Feature activation+0.696
Amazon
Token Amazon
Feature activation+0.000
or
Token or
Feature activation+0.000
the
Token the
Feature activation+0.000
you
Token you
Feature activation+0.051
are
Token are
Feature activation+0.000
using
Token using
Feature activation+0.000
color
Token color
Feature activation+0.011
,
Token,
Feature activation+0.946
add
Token add
Feature activation+1.892
a
Token a
Feature activation+0.526
dash
Token dash
Feature activation+0.124
of
Token of
Feature activation+0.056
food
Token food
Feature activation+0.011
coloring
Token coloring
Feature activation+0.040
to
Token to
Feature activation+0.000
change
Token change
Feature activation+0.000
the
Token the
Feature activation+0.030
marijuana
Token marijuana
Feature activation+0.000
laws
Token laws
Feature activation+0.178
,
Token,
Feature activation+1.875
go
Token go
Feature activation+2.446
ahead
Token ahead
Feature activation+1.133
and
Token and
Feature activation+0.862
change
Token change
Feature activation+0.757
the
Token the
Feature activation+0.500
about
Token about
Feature activation+0.000
pand
Token pand
Feature activation+0.000
oc
Tokenoc
Feature activation+0.000
filters
Token filters
Feature activation+0.000
,
Token,
Feature activation+1.116
check
Token check
Feature activation+1.874
here
Token here
Feature activation+1.164
.
Token.
Feature activation+0.199
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Comments
TokenComments
Feature activation+0.000
fight
Token fight
Feature activation+0.000
each
Token each
Feature activation+0.000
other
Token other
Feature activation+0.000
,
Token,
Feature activation+0.576
Crawford
Token Crawford
Feature activation+0.015
would
Token would
Feature activation+1.859
favor
Token favor
Feature activation+0.378
Thur
Token Thur
Feature activation+0.000
man
Tokenman
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
F
Token F
Feature activation+0.000
oles
Tokenoles
Feature activation+0.000
was
Token was
Feature activation+0.000
there
Token there
Feature activation+0.074
,
Token,
Feature activation+0.862
who
Token who
Feature activation+1.786
was
Token was
Feature activation+0.374
lined
Token lined
Feature activation+0.000
up
Token up
Feature activation+0.000
as
Token as
Feature activation+0.108
quarterback
Token quarterback
Feature activation+0.000
date
Token date
Feature activation+0.000
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
rapist
Token rapist
Feature activation+0.000
,
Token,
Feature activation+1.136
you
Token you
Feature activation+1.770
shoot
Token shoot
Feature activation+0.680
them
Token them
Feature activation+0.886
,
Token,
Feature activation+0.670
âĢ
TokenâĢ
Feature activation+0.626
Ŀ
TokenĿ
Feature activation+0.000
looking
Token looking
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
friend
Token friend
Feature activation+0.090
,
Token,
Feature activation+1.026
you
Token you
Feature activation+1.767
âĢ
TokenâĢ
Feature activation+0.270
Ļ
TokenĻ
Feature activation+0.023
re
Tokenre
Feature activation+0.000
going
Token going
Feature activation+0.394
to
Token to
Feature activation+0.389
will
Token will
Feature activation+1.276
be
Token be
Feature activation+0.670
blown
Token blown
Feature activation+0.032
away
Token away
Feature activation+0.000
,
Token,
Feature activation+1.425
who
Token who
Feature activation+1.765
and
Token and
Feature activation+0.061
what
Token what
Feature activation+0.337
will
Token will
Feature activation+0.066
now
Token now
Feature activation+0.000
oppose
Token oppose
Feature activation+0.000
enjoy
Token enjoy
Feature activation+0.011
Trans
Token Trans
Feature activation+0.000
M
TokenM
Feature activation+0.000
issions
Tokenissions
Feature activation+0.000
,
Token,
Feature activation+0.733
please
Token please
Feature activation+1.758
rate
Token rate
Feature activation+0.444
us
Token us
Feature activation+0.108
on
Token on
Feature activation+0.000
iTunes
Token iTunes
Feature activation+0.000
and
Token and
Feature activation+0.000
models
Token models
Feature activation+0.000
or
Token or
Feature activation+0.000
other
Token other
Feature activation+0.033
clients
Token clients
Feature activation+0.000
,
Token,
Feature activation+1.123
make
Token make
Feature activation+1.731
sure
Token sure
Feature activation+1.020
they
Token they
Feature activation+0.472
understand
Token understand
Feature activation+0.126
your
Token your
Feature activation+0.057
terms
Token terms
Feature activation+0.000
out
Token out
Feature activation+0.092
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
true
Token true
Feature activation+0.000
,
Token,
Feature activation+0.803
then
Token then
Feature activation+1.725
it
Token it
Feature activation+0.697
seems
Token seems
Feature activation+0.692
likely
Token likely
Feature activation+0.305
that
Token that
Feature activation+0.189
the
Token the
Feature activation+0.000
think
Token think
Feature activation+0.002
we
Token we
Feature activation+0.067
might
Token might
Feature activation+0.233
enjoy
Token enjoy
Feature activation+0.151
,
Token,
Feature activation+0.924
please
Token please
Feature activation+1.721
send
Token send
Feature activation+0.774
them
Token them
Feature activation+0.440
in
Token in
Feature activation+0.083
.
Token.
Feature activation+0.053
All
Token All
Feature activation+0.000
origin
Token origin
Feature activation+0.192
story
Token story
Feature activation+0.601
sounds
Token sounds
Feature activation+1.523
familiar
Token familiar
Feature activation+1.169
,
Token,
Feature activation+1.282
that
Token that
Feature activation+1.720
âĢ
TokenâĢ
Feature activation+0.200
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
because
Token because
Feature activation+0.195
so
Token so
Feature activation+0.032
twin
Token twin
Feature activation+0.000
sun
Token sun
Feature activation+0.019
s
Tokens
Feature activation+0.000
,
Token,
Feature activation+1.365
they
Token they
Feature activation+1.316
would
Token would
Feature activation+1.709
have
Token have
Feature activation+0.203
to
Token to
Feature activation+0.304
be
Token be
Feature activation+0.248
extremely
Token extremely
Feature activation+0.000
adapt
Token adapt
Feature activation+0.000
've
Token've
Feature activation+0.000
found
Token found
Feature activation+0.000
an
Token an
Feature activation+0.000
error
Token error
Feature activation+0.000
,
Token,
Feature activation+0.612
please
Token please
Feature activation+1.698
check
Token check
Feature activation+0.706
the
Token the
Feature activation+0.020
wiki
Token wiki
Feature activation+0.000
and
Token and
Feature activation+0.000
/
Token/
Feature activation+0.000

Top DFA by src position
MAX = 2.333

to
Token to
Feature activation-0.004
Top resid features:
enforce
Token enforce
Feature activation+0.042
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
âĢ
TokenâĢ
Feature activation+0.037
Top resid features:
ľ
Tokenľ
Feature activation+0.007
Top resid features:
If
TokenIf
Feature activation+2.165
Top resid features:
you
Token you
Feature activation+0.076
Top resid features:
want
Token want
Feature activation+0.055
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
change
Token change
Feature activation+0.113
Top resid features:
the
Token the
Feature activation+0.053
Top resid features:
Provide
Token Provide
Feature activation+0.001
Top resid features:
feedback
Token feedback
Feature activation+0.013
Top resid features:
!
Token!
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.014
Top resid features:
Ċ
TokenĊ
Feature activation+0.020
Top resid features:
If
TokenIf
Feature activation+2.112
Top resid features:
a
Token a
Feature activation+0.013
Top resid features:
Common
Token Common
Feature activation+0.032
Top resid features:
Mark
TokenMark
Feature activation+0.019
Top resid features:
implementation
Token implementation
Feature activation+0.061
Top resid features:
does
Token does
Feature activation+0.053
Top resid features:
pushing
Token pushing
Feature activation+0.001
Top resid features:
for
Token for
Feature activation+0.011
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
Ċ
TokenĊ
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.019
Top resid features:
If
TokenIf
Feature activation+1.979
Top resid features:
you
Token you
Feature activation+0.062
Top resid features:
happen
Token happen
Feature activation+0.109
Top resid features:
to
Token to
Feature activation+0.001
Top resid features:
be
Token be
Feature activation+0.047
Top resid features:
an
Token an
Feature activation+0.026
Top resid features:
he
Token he
Feature activation+0.016
Top resid features:
responded
Token responded
Feature activation+0.007
Top resid features:
:
Token:
Feature activation+0.038
Top resid features:
âĢ
Token âĢ
Feature activation+0.019
Top resid features:
ľ
Tokenľ
Feature activation+0.006
Top resid features:
If
TokenIf
Feature activation+2.333
Top resid features:
you
Token you
Feature activation+0.038
Top resid features:
took
Token took
Feature activation+0.025
Top resid features:
that
Token that
Feature activation+0.024
Top resid features:
mug
Token mug
Feature activation+0.027
Top resid features:
and
Token and
Feature activation+0.009
Top resid features:
parts
Token parts
Feature activation+0.030
Top resid features:
of
Token of
Feature activation+0.013
Top resid features:
the
Token the
Feature activation+0.019
Top resid features:
US
Token US
Feature activation+0.022
Top resid features:
.
Token.
Feature activation-0.023
Top resid features:
If
Token If
Feature activation+1.944
Top resid features:
this
Token this
Feature activation+0.028
Top resid features:
really
Token really
Feature activation+0.006
Top resid features:
is
Token is
Feature activation+0.087
Top resid features:
an
Token an
Feature activation+0.017
Top resid features:
anti
Token anti
Feature activation-0.023
Top resid features:
Step
Token Step
Feature activation+0.026
Top resid features:
Three
Token Three
Feature activation+0.004
Top resid features:
|
Token |
Feature activation+0.015
Top resid features:
Add
Token Add
Feature activation+0.029
Top resid features:
Color
Token Color
Feature activation+0.021
Top resid features:
If
Token If
Feature activation+1.930
Top resid features:
you
Token you
Feature activation+0.058
Top resid features:
are
Token are
Feature activation+0.051
Top resid features:
using
Token using
Feature activation+0.067
Top resid features:
color
Token color
Feature activation+0.107
Top resid features:
,
Token,
Feature activation+0.631
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
enforce
Token enforce
Feature activation+0.039
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
âĢ
TokenâĢ
Feature activation+0.052
Top resid features:
ľ
Tokenľ
Feature activation-0.003
Top resid features:
If
TokenIf
Feature activation+1.680
Top resid features:
you
Token you
Feature activation+0.081
Top resid features:
want
Token want
Feature activation+0.057
Top resid features:
to
Token to
Feature activation-0.006
Top resid features:
change
Token change
Feature activation+0.094
Top resid features:
the
Token the
Feature activation+0.077
Top resid features:
the
Token the
Feature activation+0.011
Top resid features:
source
Token source
Feature activation+0.020
Top resid features:
code
Token code
Feature activation+0.022
Top resid features:
here
Token here
Feature activation+0.013
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
If
Token If
Feature activation+1.926
Top resid features:
you
Token you
Feature activation+0.060
Top resid features:
want
Token want
Feature activation+0.030
Top resid features:
to
Token to
Feature activation+0.009
Top resid features:
learn
Token learn
Feature activation+0.062
Top resid features:
more
Token more
Feature activation+0.042
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.027
Top resid features:
year
Token year
Feature activation+0.007
Top resid features:
.
Token.
Feature activation+0.007
Top resid features:
If
Token If
Feature activation+2.246
Top resid features:
they
Token they
Feature activation+0.001
Top resid features:
agree
Token agree
Feature activation+0.126
Top resid features:
to
Token to
Feature activation-0.005
Top resid features:
fight
Token fight
Feature activation+0.036
Top resid features:
each
Token each
Feature activation+0.014
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.006
Top resid features:
Video
Token Video
Feature activation+0.044
Top resid features:
Ċ
TokenĊ
Feature activation-0.020
Top resid features:
Ċ
TokenĊ
Feature activation-0.018
Top resid features:
If
TokenIf
Feature activation+2.259
Top resid features:
Nick
Token Nick
Feature activation+0.012
Top resid features:
F
Token F
Feature activation+0.021
Top resid features:
oles
Tokenoles
Feature activation+0.011
Top resid features:
was
Token was
Feature activation-0.020
Top resid features:
there
Token there
Feature activation+0.051
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.011
Top resid features:
If
TokenIf
Feature activation+1.775
Top resid features:
your
Token your
Feature activation+0.040
Top resid features:
date
Token date
Feature activation+0.096
Top resid features:
is
Token is
Feature activation+0.049
Top resid features:
a
Token a
Feature activation+0.020
Top resid features:
rapist
Token rapist
Feature activation+0.136
Top resid features:
a
Token a
Feature activation+0.004
Top resid features:
winner
Token winner
Feature activation+0.020
Top resid features:
.
Token.
Feature activation-0.020
Top resid features:
Ċ
TokenĊ
Feature activation-0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.006
Top resid features:
If
TokenIf
Feature activation+1.727
Top resid features:
you
Token you
Feature activation+0.096
Top resid features:
go
Token go
Feature activation+0.054
Top resid features:
looking
Token looking
Feature activation+0.071
Top resid features:
for
Token for
Feature activation+0.041
Top resid features:
a
Token a
Feature activation+0.037
Top resid features:
bel
Tokenbel
Feature activation+0.009
Top resid features:
wrote
Token wrote
Feature activation+0.024
Top resid features:
:
Token:
Feature activation+0.029
Top resid features:
âĢ
Token âĢ
Feature activation+0.013
Top resid features:
ĺ
Tokenĺ
Feature activation+0.005
Top resid features:
If
TokenIf
Feature activation+1.911
Top resid features:
the
Token the
Feature activation+0.012
Top resid features:
imm
Token imm
Feature activation+0.021
Top resid features:
em
Tokenem
Feature activation+0.014
Top resid features:
orial
Tokenorial
Feature activation+0.015
Top resid features:
taboo
Token taboo
Feature activation+0.063
Top resid features:
:
Token:
Feature activation+0.016
Top resid features:
11
Token11
Feature activation+0.012
Top resid features:
]
Token]
Feature activation+0.007
Top resid features:
Ċ
TokenĊ
Feature activation+0.018
Top resid features:
Ċ
TokenĊ
Feature activation+0.021
Top resid features:
If
TokenIf
Feature activation+1.947
Top resid features:
you
Token you
Feature activation+0.042
Top resid features:
enjoy
Token enjoy
Feature activation-0.043
Top resid features:
Trans
Token Trans
Feature activation+0.081
Top resid features:
M
TokenM
Feature activation+0.028
Top resid features:
issions
Tokenissions
Feature activation+0.011
Top resid features:
ivers
Tokenivers
Feature activation+0.002
Top resid features:
and
Token and
Feature activation-0.001
Top resid features:
Contracts
Token Contracts
Feature activation-0.021
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.006
Top resid features:
If
TokenIf
Feature activation+1.865
Top resid features:
you
Token you
Feature activation+0.090
Top resid features:
work
Token work
Feature activation+0.061
Top resid features:
with
Token with
Feature activation-0.000
Top resid features:
models
Token models
Feature activation+0.038
Top resid features:
or
Token or
Feature activation-0.017
Top resid features:
K
TokenK
Feature activation-0.003
Top resid features:
footage
Token footage
Feature activation+0.030
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.005
Top resid features:
If
TokenIf
Feature activation+1.950
Top resid features:
this
Token this
Feature activation+0.028
Top resid features:
report
Token report
Feature activation+0.053
Top resid features:
turns
Token turns
Feature activation+0.085
Top resid features:
out
Token out
Feature activation-0.009
Top resid features:
to
Token to
Feature activation-0.026
Top resid features:
.
Token.
Feature activation+0.008
Top resid features:
com
Tokencom
Feature activation+0.016
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Ċ
TokenĊ
Feature activation+0.015
Top resid features:
Ċ
TokenĊ
Feature activation+0.025
Top resid features:
If
TokenIf
Feature activation+1.817
Top resid features:
you
Token you
Feature activation+0.046
Top resid features:
have
Token have
Feature activation+0.041
Top resid features:
any
Token any
Feature activation+0.021
Top resid features:
problems
Token problems
Feature activation+0.056
Top resid features:
you
Token you
Feature activation+0.065
Top resid features:
Und
TokenUnd
Feature activation+0.010
Top resid features:
iscovered
Tokeniscovered
Feature activation+0.019
Top resid features:
Countries
Token Countries
Feature activation+0.035
Top resid features:
Ċ
TokenĊ
Feature activation+0.024
Top resid features:
Ċ
TokenĊ
Feature activation+0.026
Top resid features:
If
TokenIf
Feature activation+1.727
Top resid features:
the
Token the
Feature activation+0.030
Top resid features:
Alexa
Token Alexa
Feature activation+0.036
Top resid features:
skills
Token skills
Feature activation+0.049
Top resid features:
origin
Token origin
Feature activation+0.037
Top resid features:
story
Token story
Feature activation+0.065
Top resid features:
was
Token was
Feature activation-0.082
Top resid features:
nothing
Token nothing
Feature activation-0.002
Top resid features:
given
Token given
Feature activation+0.011
Top resid features:
size
Token size
Feature activation+0.016
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.076
Top resid features:
If
TokenIf
Feature activation+2.301
Top resid features:
creatures
Token creatures
Feature activation+0.047
Top resid features:
exist
Token exist
Feature activation+0.135
Top resid features:
on
Token on
Feature activation-0.001
Top resid features:
planets
Token planets
Feature activation+0.079
Top resid features:
with
Token with
Feature activation-0.001
Top resid features:
in
Token in
Feature activation+0.009
Top resid features:
day
Token day
Feature activation+0.012
Top resid features:
care
Token care
Feature activation+0.010
Top resid features:
.
Token.
Feature activation+0.006
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.084
Top resid features:
If
TokenIf
Feature activation+1.805
Top resid features:
you
Token you
Feature activation+0.028
Top resid features:
have
Token have
Feature activation+0.019
Top resid features:
suggestions
Token suggestions
Feature activation+0.018
Top resid features:
for
Token for
Feature activation+0.015
Top resid features:
improvements
Token improvements
Feature activation+0.035
Top resid features:

Decoder Weights Distribution

Head 0: 0.05

Head 1: 0.04

Head 2: 0.06

Head 3: 0.07

Head 4: 0.03

Head 5: 0.11

Head 6: 0.27

Head 7: 0.08

Head 8: 0.04

Head 9: 0.04

Head 10: 0.14

Head 11: 0.08

Positive logits

?'1.69

uld1.64

illac1.63

?",1.62

esson1.53

anymore1.48

odan1.47

?」1.47

?'"1.46

xus1.43

?:1.42

certific1.40

Otherwise1.38

lear1.38

instead1.37

ado1.37

heirs1.37

))1.36

please1.35

tek1.35

Negative logits

-1.37

EV-1.37

IV-1.36

subur-1.29

REM-1.29

Cosponsors-1.27

EStreamFrame-1.24

Trayvon-1.24

pit-1.23

Ig-1.23

ActionCode-1.20

Moroc-1.18

Haram-1.18

Gamb-1.18

Tas-1.18

-1.17

lip-1.17

Neh-1.16

AME-1.15

behavi-1.15

INTERVAL 2.201 - 2.446
CONTAINS 0.000%

change
Token change
Feature activation+0.000
the
Token the
Feature activation+0.030
marijuana
Token marijuana
Feature activation+0.000
laws
Token laws
Feature activation+0.178
,
Token,
Feature activation+1.875
go
Token go
Feature activation+2.446
ahead
Token ahead
Feature activation+1.133
and
Token and
Feature activation+0.862
change
Token change
Feature activation+0.757
the
Token the
Feature activation+0.500
national
Token national
Feature activation+0.000

INTERVAL 1.957 - 2.201
CONTAINS 0.000%

environmental
Token environmental
Feature activation+0.000
news
Token news
Feature activation+0.134
junk
Token junk
Feature activation+0.013
ie
Tokenie
Feature activation+0.000
,
Token,
Feature activation+1.349
you
Token you
Feature activation+2.007
already
Token already
Feature activation+0.441
know
Token know
Feature activation+0.566
that
Token that
Feature activation+0.653
the
Token the
Feature activation+0.000
past
Token past
Feature activation+0.000
and
Token and
Feature activation+0.517
rem
Token rem
Feature activation+0.015
ade
Tokenade
Feature activation+0.039
it
Token it
Feature activation+0.843
,
Token,
Feature activation+1.101
would
Token would
Feature activation+1.967
you
Token you
Feature activation+0.515
be
Token be
Feature activation+0.000
asking
Token asking
Feature activation+0.000
me
Token me
Feature activation+0.000
the
Token the
Feature activation+0.000
preferred
Token preferred
Feature activation+0.000
environment
Token environment
Feature activation+0.000
or
Token or
Feature activation+0.000
language
Token language
Feature activation+0.000
,
Token,
Feature activation+1.145
try
Token try
Feature activation+2.157
implementing
Token implementing
Feature activation+0.046
your
Token your
Feature activation+0.336
own
Token own
Feature activation+0.000
Common
Token Common
Feature activation+0.000
Mark
TokenMark
Feature activation+0.000

INTERVAL 1.712 - 1.957
CONTAINS 0.001%

looking
Token looking
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
friend
Token friend
Feature activation+0.090
,
Token,
Feature activation+1.026
you
Token you
Feature activation+1.767
âĢ
TokenâĢ
Feature activation+0.270
Ļ
TokenĻ
Feature activation+0.023
re
Tokenre
Feature activation+0.000
going
Token going
Feature activation+0.394
to
Token to
Feature activation+0.389
origin
Token origin
Feature activation+0.192
story
Token story
Feature activation+0.601
sounds
Token sounds
Feature activation+1.523
familiar
Token familiar
Feature activation+1.169
,
Token,
Feature activation+1.282
that
Token that
Feature activation+1.720
âĢ
TokenâĢ
Feature activation+0.200
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
because
Token because
Feature activation+0.195
so
Token so
Feature activation+0.032
F
Token F
Feature activation+0.000
oles
Tokenoles
Feature activation+0.000
was
Token was
Feature activation+0.000
there
Token there
Feature activation+0.074
,
Token,
Feature activation+0.862
who
Token who
Feature activation+1.786
was
Token was
Feature activation+0.374
lined
Token lined
Feature activation+0.000
up
Token up
Feature activation+0.000
as
Token as
Feature activation+0.108
quarterback
Token quarterback
Feature activation+0.000
think
Token think
Feature activation+0.002
we
Token we
Feature activation+0.067
might
Token might
Feature activation+0.233
enjoy
Token enjoy
Feature activation+0.151
,
Token,
Feature activation+0.924
please
Token please
Feature activation+1.721
send
Token send
Feature activation+0.774
them
Token them
Feature activation+0.440
in
Token in
Feature activation+0.083
.
Token.
Feature activation+0.053
All
Token All
Feature activation+0.000
enjoy
Token enjoy
Feature activation+0.011
Trans
Token Trans
Feature activation+0.000
M
TokenM
Feature activation+0.000
issions
Tokenissions
Feature activation+0.000
,
Token,
Feature activation+0.733
please
Token please
Feature activation+1.758
rate
Token rate
Feature activation+0.444
us
Token us
Feature activation+0.108
on
Token on
Feature activation+0.000
iTunes
Token iTunes
Feature activation+0.000
and
Token and
Feature activation+0.000

INTERVAL 1.468 - 1.712
CONTAINS 0.002%

es
Tokenes
Feature activation+0.000
the
Token the
Feature activation+0.000
exec
Token exec
Feature activation+0.000
s
Tokens
Feature activation+0.000
,
Token,
Feature activation+1.068
then
Token then
Feature activation+1.609
Revolution
Token Revolution
Feature activation+0.000
could
Token could
Feature activation+0.957
get
Token get
Feature activation+0.137
the
Token the
Feature activation+0.000
chop
Token chop
Feature activation+0.000
the
Token the
Feature activation+0.000
Alexa
Token Alexa
Feature activation+0.000
skills
Token skills
Feature activation+0.000
origin
Token origin
Feature activation+0.192
story
Token story
Feature activation+0.601
sounds
Token sounds
Feature activation+1.523
familiar
Token familiar
Feature activation+1.169
,
Token,
Feature activation+1.282
that
Token that
Feature activation+1.720
âĢ
TokenâĢ
Feature activation+0.200
Ļ
TokenĻ
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
traditional
Token traditional
Feature activation+0.000
sense
Token sense
Feature activation+0.000
,
Token,
Feature activation+0.931
you
Token you
Feature activation+1.632
âĢ
TokenâĢ
Feature activation+0.154
Ļ
TokenĻ
Feature activation+0.000
d
Tokend
Feature activation+0.000
go
Token go
Feature activation+0.704
to
Token to
Feature activation+0.272
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
woods
Token woods
Feature activation+0.000
,
Token,
Feature activation+1.027
you
Token you
Feature activation+1.541
can
Token can
Feature activation+0.991
âĢ
TokenâĢ
Feature activation+0.151
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
rely
Token rely
Feature activation+0.093
sightings
Token sightings
Feature activation+0.019
increase
Token increase
Feature activation+0.313
in
Token in
Feature activation+0.014
frequency
Token frequency
Feature activation+0.068
,
Token,
Feature activation+1.110
then
Token then
Feature activation+1.480
there
Token there
Feature activation+0.372
would
Token would
Feature activation+0.747
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
strong
Token strong
Feature activation+0.000

INTERVAL 1.223 - 1.468
CONTAINS 0.005%

a
Token a
Feature activation+0.000
light
Token light
Feature activation+0.000
with
Token with
Feature activation+0.000
this
Token this
Feature activation+0.000
sticker
Token sticker
Feature activation+0.206
,
Token,
Feature activation+1.257
you
Token you
Feature activation+1.696
can
Token can
Feature activation+0.872
be
Token be
Feature activation+0.441
sure
Token sure
Feature activation+0.302
that
Token that
Feature activation+0.328
use
Token use
Feature activation+0.000
it
Token it
Feature activation+0.088
too
Token too
Feature activation+0.000
much
Token much
Feature activation+0.000
,
Token,
Feature activation+0.481
it
Token it
Feature activation+1.361
becomes
Token becomes
Feature activation+0.613
more
Token more
Feature activation+0.096
difficult
Token difficult
Feature activation+0.168
to
Token to
Feature activation+0.000
think
Token think
Feature activation+0.011
in
Token in
Feature activation+0.000
group
Token group
Feature activation+0.098
cath
Token cath
Feature activation+0.017
ars
Tokenars
Feature activation+0.000
is
Tokenis
Feature activation+0.000
,
Token,
Feature activation+1.352
the
Token the
Feature activation+1.248
January
Token January
Feature activation+0.322
15
Token 15
Feature activation+0.187
th
Tokenth
Feature activation+0.000
Day
Token Day
Feature activation+0.000
they
Token they
Feature activation+0.000
train
Token train
Feature activation+0.000
on
Token on
Feature activation+0.000
it
Token it
Feature activation+0.058
,
Token,
Feature activation+0.519
maybe
Token maybe
Feature activation+1.378
they
Token they
Feature activation+1.017
get
Token get
Feature activation+0.274
an
Token an
Feature activation+0.000
advantage
Token advantage
Feature activation+0.000
,
Token,
Feature activation+0.755
they
Token they
Feature activation+1.017
get
Token get
Feature activation+0.274
an
Token an
Feature activation+0.000
advantage
Token advantage
Feature activation+0.000
,
Token,
Feature activation+0.755
but
Token but
Feature activation+1.374
the
Token the
Feature activation+0.000
game
Token game
Feature activation+0.000
comes
Token comes
Feature activation+0.104
down
Token down
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 0.978 - 1.223
CONTAINS 0.010%

not
Token not
Feature activation+0.000
completed
Token completed
Feature activation+0.000
on
Token on
Feature activation+0.000
time
Token time
Feature activation+0.056
,
Token,
Feature activation+0.847
the
Token the
Feature activation+0.988
offending
Token offending
Feature activation+0.105
member
Token member
Feature activation+0.113
will
Token will
Feature activation+1.017
be
Token be
Feature activation+0.522
subject
Token subject
Feature activation+0.000
sounds
Token sounds
Feature activation+0.000
far
Token far
Feature activation+0.000
-
Token-
Feature activation+0.000
fetched
Tokenfetched
Feature activation+0.000
,
Token,
Feature activation+0.632
it
Token it
Feature activation+1.176
âĢ
TokenâĢ
Feature activation+0.043
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
not
Token not
Feature activation+0.110
.
Token.
Feature activation+0.000
time
Token time
Feature activation+0.056
,
Token,
Feature activation+0.847
the
Token the
Feature activation+0.988
offending
Token offending
Feature activation+0.105
member
Token member
Feature activation+0.113
will
Token will
Feature activation+1.017
be
Token be
Feature activation+0.522
subject
Token subject
Feature activation+0.000
to
Token to
Feature activation+0.000
disqual
Token disqual
Feature activation+0.000
ification
Tokenification
Feature activation+0.000
t
Tokent
Feature activation+0.000
have
Token have
Feature activation+0.000
actors
Token actors
Feature activation+0.000
available
Token available
Feature activation+0.031
,
Token,
Feature activation+1.066
no
Token no
Feature activation+1.103
problem
Token problem
Feature activation+0.588
!
Token!
Feature activation+0.128
You
Token You
Feature activation+0.000
can
Token can
Feature activation+0.000
add
Token add
Feature activation+0.000
to
Token to
Feature activation+0.000
reduce
Token reduce
Feature activation+0.000
poverty
Token poverty
Feature activation+0.000
further
Token further
Feature activation+0.000
,
Token,
Feature activation+0.453
it
Token it
Feature activation+1.014
can
Token can
Feature activation+0.621
be
Token be
Feature activation+0.249
done
Token done
Feature activation+0.000
very
Token very
Feature activation+0.000
simply
Token simply
Feature activation+0.000

INTERVAL 0.734 - 0.978
CONTAINS 0.013%

reduces
Token reduces
Feature activation+0.000
the
Token the
Feature activation+0.000
air
Token air
Feature activation+0.000
pressure
Token pressure
Feature activation+0.000
,
Token,
Feature activation+0.922
lowers
Token lowers
Feature activation+0.736
the
Token the
Feature activation+0.449
air
Token air
Feature activation+0.000
resistance
Token resistance
Feature activation+0.000
,
Token,
Feature activation+0.444
in
Token in
Feature activation+0.219
marijuana
Token marijuana
Feature activation+0.000
laws
Token laws
Feature activation+0.178
,
Token,
Feature activation+1.875
go
Token go
Feature activation+2.446
ahead
Token ahead
Feature activation+1.133
and
Token and
Feature activation+0.862
change
Token change
Feature activation+0.757
the
Token the
Feature activation+0.500
national
Token national
Feature activation+0.000
marijuana
Token marijuana
Feature activation+0.000
laws
Token laws
Feature activation+0.203
If
Token If
Feature activation+0.000
anything
Token anything
Feature activation+0.000
,
Token,
Feature activation+0.290
what
Token what
Feature activation+0.819
Harden
Token Harden
Feature activation+0.047
does
Token does
Feature activation+0.917
should
Token should
Feature activation+0.499
be
Token be
Feature activation+0.300
called
Token called
Feature activation+0.000
defender
Token defender
Feature activation+0.000
-
Token-
Feature activation+0.000
fast
Token fast
Feature activation+0.000
with
Token with
Feature activation+0.000
your
Token your
Feature activation+0.000
fingers
Token fingers
Feature activation+0.000
,
Token,
Feature activation+0.802
there
Token there
Feature activation+0.813
will
Token will
Feature activation+1.021
now
Token now
Feature activation+0.077
be
Token be
Feature activation+0.450
times
Token times
Feature activation+0.070
when
Token when
Feature activation+0.431
or
Token or
Feature activation+0.264
murder
Token murder
Feature activation+0.000
you
Token you
Feature activation+0.355
,
Token,
Feature activation+0.088
I
Token I
Feature activation+0.673
would
Token would
Feature activation+0.819
likely
Token likely
Feature activation+0.020
assume
Token assume
Feature activation+0.000
they
Token they
Feature activation+0.000
would
Token would
Feature activation+0.000
also
Token also
Feature activation+0.000

INTERVAL 0.489 - 0.734
CONTAINS 0.026%

ll
Tokenll
Feature activation+0.000
recall
Token recall
Feature activation+0.000
from
Token from
Feature activation+0.057
this
Token this
Feature activation+0.071
review
Token review
Feature activation+0.142
,
Token,
Feature activation+0.664
our
Token our
Feature activation+0.286
Master
Token Master
Feature activation+0.000
System
Token System
Feature activation+0.000
actually
Token actually
Feature activation+0.270
belonged
Token belonged
Feature activation+0.000
of
Token of
Feature activation+0.000
most
Token most
Feature activation+0.000
Chinese
Token Chinese
Feature activation+0.000
exchanges
Token exchanges
Feature activation+0.000
,
Token,
Feature activation+0.102
it
Token it
Feature activation+0.622
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
heavily
Token heavily
Feature activation+0.000
retail
Token retail
Feature activation+0.000
taboo
Token taboo
Feature activation+0.000
of
Token of
Feature activation+0.000
heterosexual
Token heterosexual
Feature activation+0.000
marriage
Token marriage
Feature activation+0.000
will
Token will
Feature activation+1.276
be
Token be
Feature activation+0.670
blown
Token blown
Feature activation+0.032
away
Token away
Feature activation+0.000
,
Token,
Feature activation+1.425
who
Token who
Feature activation+1.765
and
Token and
Feature activation+0.061
the
Token the
Feature activation+0.754
State
Token State
Feature activation+0.000
of
Token of
Feature activation+0.061
Israel
Token Israel
Feature activation+0.000
,
Token,
Feature activation+0.425
being
Token being
Feature activation+0.727
led
Token led
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
IDF
Token IDF
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.088
too
Token too
Feature activation+0.000
much
Token much
Feature activation+0.000
,
Token,
Feature activation+0.481
it
Token it
Feature activation+1.361
becomes
Token becomes
Feature activation+0.613
more
Token more
Feature activation+0.096
difficult
Token difficult
Feature activation+0.168
to
Token to
Feature activation+0.000
think
Token think
Feature activation+0.011
coherent
Token coherent
Feature activation+0.000

INTERVAL 0.245 - 0.489
CONTAINS 0.042%

networks
Token networks
Feature activation+0.000
got
Token got
Feature activation+0.164
together
Token together
Feature activation+0.000
and
Token and
Feature activation+0.524
decided
Token decided
Feature activation+0.413
to
Token to
Feature activation+0.402
nationally
Token nationally
Feature activation+0.000
telev
Token telev
Feature activation+0.000
ise
Tokenise
Feature activation+0.000
a
Token a
Feature activation+0.000
presidential
Token presidential
Feature activation+0.000
in
Token in
Feature activation+0.000
another
Token another
Feature activation+0.000
project
Token project
Feature activation+0.000
,
Token,
Feature activation+0.171
now
Token now
Feature activation+0.366
you
Token you
Feature activation+0.392
have
Token have
Feature activation+0.000
no
Token no
Feature activation+0.000
choice
Token choice
Feature activation+0.000
other
Token other
Feature activation+0.000
than
Token than
Feature activation+0.000
you
Token you
Feature activation+0.000
have
Token have
Feature activation+0.000
a
Token a
Feature activation+0.000
M
Token M
Feature activation+0.000
WD
TokenWD
Feature activation+0.000
and
Token and
Feature activation+0.377
they
Token they
Feature activation+0.664
turn
Token turn
Feature activation+1.015
it
Token it
Feature activation+0.735
off
Token off
Feature activation+0.000
,
Token,
Feature activation+1.019
with
Token with
Feature activation+0.000
W
Token W
Feature activation+0.000
ires
Tokenires
Feature activation+0.000
h
Tokenh
Feature activation+0.000
ark
Tokenark
Feature activation+0.000
,
Token,
Feature activation+0.404
we
Token we
Feature activation+0.828
can
Token can
Feature activation+0.314
clearly
Token clearly
Feature activation+0.000
see
Token see
Feature activation+0.000
the
Token the
Feature activation+0.000
in
Token in
Feature activation+0.000
our
Token our
Feature activation+0.000
brother
Token brother
Feature activation+0.000
company
Token company
Feature activation+0.025
,
Token,
Feature activation+0.982
D
Token D
Feature activation+0.261
ump
Tokenump
Feature activation+0.000
ster
Tokenster
Feature activation+0.000
Daw
Token Daw
Feature activation+0.000
gs
Tokengs
Feature activation+0.000
,
Token,
Feature activation+0.188

INTERVAL 0.000 - 0.245
CONTAINS 99.900%

have
Token have
Feature activation+0.000
1
Token 1
Feature activation+0.000
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.000
tech
Token tech
Feature activation+0.000
trends
Token trends
Feature activation+0.000
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
might
Token might
Feature activation+0.000
have
Token have
Feature activation+0.000
a
Token a
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
report
Token report
Feature activation+0.000
,
Token,
Feature activation+0.000
federal
Token federal
Feature activation+0.000
authorities
Token authorities
Feature activation+0.000
announced
Token announced
Feature activation+0.000
they
Token they
Feature activation+0.000
would
Token would
Feature activation+0.000
suspend
Token suspend
Feature activation+0.000
Arpaio
Token Arpaio
Feature activation+0.000
Premier
Token Premier
Feature activation+0.000
League
Token League
Feature activation+0.000
sides
Token sides
Feature activation+0.000
but
Token but
Feature activation+0.000
will
Token will
Feature activation+0.000
sign
Token sign
Feature activation+0.000
for
Token for
Feature activation+0.000
Villa
Token Villa
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
season
Token season
Feature activation+0.000
those
Token those
Feature activation+0.000
they
Token they
Feature activation+0.000
consider
Token consider
Feature activation+0.000
less
Token less
Feature activation+0.000
strict
Token strict
Feature activation+0.000
in
Token in
Feature activation+0.000
their
Token their
Feature activation+0.000
religious
Token religious
Feature activation+0.000
observ
Token observ
Feature activation+0.000
ance
Tokenance
Feature activation+0.000
.
Token.
Feature activation+0.000
,
Token,
Feature activation+0.000
Abu
Token Abu
Feature activation+0.000
Zub
Token Zub
Feature activation+0.000
air
Tokenair
Feature activation+0.000
al
Token al
Feature activation+0.000
-
Token-
Feature activation+0.000
Mas
TokenMas
Feature activation+0.000
ri
Tokenri
Feature activation+0.000
,
Token,
Feature activation+0.000
was
Token was
Feature activation+0.000
among
Token among
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 H2.6: (feature 3732)

TOP ACTIVATIONS
MAX = 1.935

some
Token some
Feature activation+0.042
common
Token common
Feature activation+0.000
sense
Token sense
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
,
Token,
Feature activation+1.105
read
Token read
Feature activation+1.935
the
Token the
Feature activation+0.714
whole
Token whole
Feature activation+0.247
thing
Token thing
Feature activation+0.539
.
Token.
Feature activation+0.195
He
Token He
Feature activation+0.000
anyone
Token anyone
Feature activation+0.072
with
Token with
Feature activation+0.000
bipolar
Token bipolar
Feature activation+0.000
disorder
Token disorder
Feature activation+0.000
,
Token,
Feature activation+1.087
then
Token then
Feature activation+1.885
don
Token don
Feature activation+1.093
âĢ
TokenâĢ
Feature activation+0.467
Ļ
TokenĻ
Feature activation+0.071
t
Tokent
Feature activation+0.211
buy
Token buy
Feature activation+0.537
you
Token you
Feature activation+0.030
are
Token are
Feature activation+0.000
using
Token using
Feature activation+0.000
Bitcoin
Token Bitcoin
Feature activation+0.000
,
Token,
Feature activation+0.761
you
Token you
Feature activation+1.870
are
Token are
Feature activation+0.635
making
Token making
Feature activation+0.210
the
Token the
Feature activation+0.390
world
Token world
Feature activation+0.000
a
Token a
Feature activation+0.023
gy
Token gy
Feature activation+0.000
ro
Tokenro
Feature activation+0.000
scope
Tokenscope
Feature activation+0.000
sensor
Token sensor
Feature activation+0.000
,
Token,
Feature activation+0.974
you
Token you
Feature activation+1.820
can
Token can
Feature activation+1.023
download
Token download
Feature activation+0.592
the
Token the
Feature activation+0.548
application
Token application
Feature activation+0.109
from
Token from
Feature activation+0.144
man
Token man
Feature activation+0.000
sweat
Token sweat
Feature activation+0.000
on
Token on
Feature activation+0.000
you
Token you
Feature activation+0.446
,
Token,
Feature activation+0.164
then
Token then
Feature activation+1.810
you
Token you
Feature activation+0.596
needed
Token needed
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
near
Token near
Feature activation+0.000
of
Token of
Feature activation+0.014
countries
Token countries
Feature activation+0.000
to
Token to
Feature activation+0.082
visit
Token visit
Feature activation+0.576
,
Token,
Feature activation+1.060
you
Token you
Feature activation+1.801
should
Token should
Feature activation+1.104
definitely
Token definitely
Feature activation+0.337
give
Token give
Feature activation+0.071
it
Token it
Feature activation+0.227
a
Token a
Feature activation+0.000
best
Token best
Feature activation+0.158
choice
Token choice
Feature activation+0.000
for
Token for
Feature activation+0.000
you
Token you
Feature activation+0.237
,
Token,
Feature activation+0.499
here
Token here
Feature activation+1.774
are
Token are
Feature activation+0.673
top
Token top
Feature activation+0.161
10
Token 10
Feature activation+0.000
famous
Token famous
Feature activation+0.000
websites
Token websites
Feature activation+0.000
your
Token your
Feature activation+0.215
BlackBerry
Token BlackBerry
Feature activation+0.000
Play
Token Play
Feature activation+0.000
Book
TokenBook
Feature activation+0.000
,
Token,
Feature activation+0.844
you
Token you
Feature activation+1.754
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
probably
Token probably
Feature activation+0.305
wondering
Token wondering
Feature activation+0.099
the
Token the
Feature activation+0.026
republic
Token republic
Feature activation+0.000
of
Token of
Feature activation+0.000
Ireland
Token Ireland
Feature activation+0.000
,
Token,
Feature activation+1.092
you
Token you
Feature activation+1.746
may
Token may
Feature activation+0.956
be
Token be
Feature activation+0.385
astonished
Token astonished
Feature activation+0.033
to
Token to
Feature activation+0.092
see
Token see
Feature activation+0.368
of
Token of
Feature activation+0.000
catching
Token catching
Feature activation+0.000
a
Token a
Feature activation+0.000
Pokemon
Token Pokemon
Feature activation+0.000
,
Token,
Feature activation+0.981
you
Token you
Feature activation+1.703
need
Token need
Feature activation+0.907
to
Token to
Feature activation+0.137
wait
Token wait
Feature activation+0.533
until
Token until
Feature activation+0.241
the
Token the
Feature activation+0.000
people
Token people
Feature activation+0.000
experiencing
Token experiencing
Feature activation+0.000
touchscreen
Token touchscreen
Feature activation+0.000
issues
Token issues
Feature activation+0.158
,
Token,
Feature activation+1.010
please
Token please
Feature activation+1.695
let
Token let
Feature activation+0.696
us
Token us
Feature activation+0.007
know
Token know
Feature activation+0.101
how
Token how
Feature activation+0.260
it
Token it
Feature activation+0.000
in
Token in
Feature activation+0.000
spoon
Token spoon
Feature activation+0.126
-
Token-
Feature activation+0.000
feeding
Tokenfeeding
Feature activation+0.000
,
Token,
Feature activation+0.960
it
Token it
Feature activation+1.633
is
Token is
Feature activation+1.135
the
Token the
Feature activation+0.629
liberal
Token liberal
Feature activation+0.000
media
Token media
Feature activation+0.000
and
Token and
Feature activation+0.000
signing
Token signing
Feature activation+0.000
of
Token of
Feature activation+0.000
your
Token your
Feature activation+0.000
transactions
Token transactions
Feature activation+0.000
,
Token,
Feature activation+0.930
then
Token then
Feature activation+1.591
you
Token you
Feature activation+0.909
do
Token do
Feature activation+0.381
not
Token not
Feature activation+0.308
currently
Token currently
Feature activation+0.003
control
Token control
Feature activation+0.000
two
Token two
Feature activation+0.000
forward
Token forward
Feature activation+0.010
,
Token,
Feature activation+0.707
the
Token the
Feature activation+0.877
fish
Token fish
Feature activation+0.297
would
Token would
Feature activation+1.569
have
Token have
Feature activation+0.271
likely
Token likely
Feature activation+0.428
never
Token never
Feature activation+0.251
fed
Token fed
Feature activation+0.000
.
Token.
Feature activation+0.000
a
Token a
Feature activation+0.000
healthy
Token healthy
Feature activation+0.000
reading
Token reading
Feature activation+0.018
culture
Token culture
Feature activation+0.000
,
Token,
Feature activation+0.727
we
Token we
Feature activation+1.564
have
Token have
Feature activation+0.537
to
Token to
Feature activation+0.405
work
Token work
Feature activation+0.261
hard
Token hard
Feature activation+0.033
to
Token to
Feature activation+0.000
of
Token of
Feature activation+0.000
most
Token most
Feature activation+0.046
Chinese
Token Chinese
Feature activation+0.000
exchanges
Token exchanges
Feature activation+0.000
,
Token,
Feature activation+0.908
it
Token it
Feature activation+1.563
âĢ
TokenâĢ
Feature activation+0.137
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
heavily
Token heavily
Feature activation+0.000
retail
Token retail
Feature activation+0.000
get
Token get
Feature activation+0.000
an
Token an
Feature activation+0.000
answering
Token answering
Feature activation+0.235
machine
Token machine
Feature activation+0.000
,
Token,
Feature activation+1.121
leave
Token leave
Feature activation+1.549
a
Token a
Feature activation+0.809
message
Token message
Feature activation+1.527
,
Token,
Feature activation+0.965
we
Token we
Feature activation+1.192
are
Token are
Feature activation+0.581
changes
Token changes
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
routine
Token routine
Feature activation+0.000
,
Token,
Feature activation+0.956
let
Token let
Feature activation+1.542
the
Token the
Feature activation+0.515
child
Token child
Feature activation+0.111
know
Token know
Feature activation+0.157
in
Token in
Feature activation+0.000
advance
Token advance
Feature activation+0.000
answering
Token answering
Feature activation+0.235
machine
Token machine
Feature activation+0.000
,
Token,
Feature activation+1.121
leave
Token leave
Feature activation+1.549
a
Token a
Feature activation+0.809
message
Token message
Feature activation+1.527
,
Token,
Feature activation+0.965
we
Token we
Feature activation+1.192
are
Token are
Feature activation+0.581
in
Token in
Feature activation+0.141
the
Token the
Feature activation+0.141
interested
Token interested
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
conversion
Token conversion
Feature activation+0.000
,
Token,
Feature activation+0.640
you
Token you
Feature activation+1.502
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
need
Token need
Feature activation+0.414
to
Token to
Feature activation+0.086

Top DFA by src position
MAX = 2.131

perpet
Token perpet
Feature activation+0.008
Top resid features:
uity
Tokenuity
Feature activation+0.006
Top resid features:
.
Token.
Feature activation+0.027
Top resid features:
Ċ
TokenĊ
Feature activation+0.028
Top resid features:
Ċ
TokenĊ
Feature activation+0.026
Top resid features:
If
TokenIf
Feature activation+1.803
Top resid features:
you
Token you
Feature activation+0.069
Top resid features:
're
Token're
Feature activation+0.032
Top resid features:
in
Token in
Feature activation-0.004
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
mood
Token mood
Feature activation+0.049
Top resid features:
on
Token on
Feature activation+0.005
Top resid features:
Twitter
Token Twitter
Feature activation+0.011
Top resid features:
:
Token:
Feature activation+0.042
Top resid features:
âĢ
Token âĢ
Feature activation+0.017
Top resid features:
ľ
Tokenľ
Feature activation+0.028
Top resid features:
If
TokenIf
Feature activation+1.821
Top resid features:
you
Token you
Feature activation+0.037
Top resid features:
know
Token know
Feature activation+0.025
Top resid features:
anyone
Token anyone
Feature activation+0.017
Top resid features:
with
Token with
Feature activation-0.040
Top resid features:
bipolar
Token bipolar
Feature activation+0.042
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
say
Token say
Feature activation+0.010
Top resid features:
,
Token,
Feature activation+0.053
Top resid features:
âĢ
Token âĢ
Feature activation+0.024
Top resid features:
ľ
Tokenľ
Feature activation+0.015
Top resid features:
If
TokenIf
Feature activation+2.067
Top resid features:
you
Token you
Feature activation+0.031
Top resid features:
are
Token are
Feature activation-0.014
Top resid features:
using
Token using
Feature activation+0.013
Top resid features:
Bitcoin
Token Bitcoin
Feature activation-0.006
Top resid features:
,
Token,
Feature activation+0.348
Top resid features:
you
Token you
Feature activation+0.013
Top resid features:
how
Token how
Feature activation+0.014
Top resid features:
.
Token.
Feature activation+0.012
Top resid features:
Ċ
TokenĊ
Feature activation+0.014
Top resid features:
Ċ
TokenĊ
Feature activation+0.009
Top resid features:
If
TokenIf
Feature activation+1.945
Top resid features:
you
Token you
Feature activation+0.033
Top resid features:
have
Token have
Feature activation+0.008
Top resid features:
a
Token a
Feature activation+0.011
Top resid features:
device
Token device
Feature activation+0.050
Top resid features:
that
Token that
Feature activation+0.019
Top resid features:
Grove
Token Grove
Feature activation+0.005
Top resid features:
Comp
Token Comp
Feature activation+0.009
Top resid features:
ost
Tokenost
Feature activation+0.003
Top resid features:
ing
Tokening
Feature activation+0.005
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.075
Top resid features:
If
TokenIf
Feature activation+2.054
Top resid features:
your
Token your
Feature activation+0.022
Top resid features:
idea
Token idea
Feature activation+0.006
Top resid features:
of
Token of
Feature activation+0.005
Top resid features:
fun
Token fun
Feature activation+0.025
Top resid features:
is
Token is
Feature activation+0.012
Top resid features:
time
Tokentime
Feature activation+0.008
Top resid features:
.
Token.
Feature activation+0.011
Top resid features:
com
Tokencom
Feature activation+0.007
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.086
Top resid features:
If
TokenIf
Feature activation+1.899
Top resid features:
you
Token you
Feature activation+0.034
Top resid features:
never
Token never
Feature activation-0.001
Top resid features:
considered
Token considered
Feature activation-0.007
Top resid features:
putting
Token putting
Feature activation+0.016
Top resid features:
Laos
Token Laos
Feature activation+0.018
Top resid features:
-
Token-
Feature activation+0.006
Top resid features:
ski
Tokenski
Feature activation+0.006
Top resid features:
glow
Token glow
Feature activation+0.009
Top resid features:
.
Token.
Feature activation+0.018
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.102
Top resid features:
If
TokenIf
Feature activation+1.695
Top resid features:
you
Token you
Feature activation+0.059
Top resid features:
're
Token're
Feature activation+0.052
Top resid features:
still
Token still
Feature activation-0.004
Top resid features:
hes
Token hes
Feature activation-0.035
Top resid features:
itating
Tokenitating
Feature activation+0.004
Top resid features:
it
Token it
Feature activation+0.005
Top resid features:
here
Token here
Feature activation+0.011
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.010
Top resid features:
Ċ
TokenĊ
Feature activation+0.008
Top resid features:
If
TokenIf
Feature activation+2.023
Top resid features:
you
Token you
Feature activation+0.026
Top resid features:
installed
Token installed
Feature activation+0.017
Top resid features:
the
Token the
Feature activation+0.008
Top resid features:
recently
Token recently
Feature activation-0.013
Top resid features:
leaked
Token leaked
Feature activation+0.020
Top resid features:
it
Token it
Feature activation+0.009
Top resid features:
to
Token to
Feature activation+0.007
Top resid features:
me
Token me
Feature activation+0.010
Top resid features:
."
Token."
Feature activation+0.011
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.074
Top resid features:
If
TokenIf
Feature activation+2.016
Top resid features:
you
Token you
Feature activation+0.041
Top resid features:
happen
Token happen
Feature activation+0.033
Top resid features:
to
Token to
Feature activation-0.008
Top resid features:
be
Token be
Feature activation+0.013
Top resid features:
flying
Token flying
Feature activation+0.022
Top resid features:
-
Token-
Feature activation+0.006
Top resid features:
depth
Tokendepth
Feature activation+0.010
Top resid features:
:
Token:
Feature activation+0.060
Top resid features:
Ċ
TokenĊ
Feature activation+0.039
Top resid features:
Ċ
TokenĊ
Feature activation+0.043
Top resid features:
If
TokenIf
Feature activation+1.758
Top resid features:
you
Token you
Feature activation+0.045
Top resid features:
want
Token want
Feature activation+0.034
Top resid features:
the
Token the
Feature activation+0.007
Top resid features:
highest
Token highest
Feature activation+0.006
Top resid features:
chance
Token chance
Feature activation+0.047
Top resid features:
discussed
Token discussed
Feature activation+0.010
Top resid features:
in
Token in
Feature activation+0.008
Top resid features:
this
Token this
Feature activation+0.007
Top resid features:
thread
Token thread
Feature activation+0.008
Top resid features:
.
Token.
Feature activation+0.051
Top resid features:
If
Token If
Feature activation+1.797
Top resid features:
you
Token you
Feature activation+0.027
Top resid features:
âĢ
TokenâĢ
Feature activation+0.081
Top resid features:
Ļ
TokenĻ
Feature activation+0.014
Top resid features:
re
Tokenre
Feature activation+0.003
Top resid features:
one
Token one
Feature activation-0.010
Top resid features:
2014
Token 2014
Feature activation+0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.019
Top resid features:
Ċ
TokenĊ
Feature activation+0.018
Top resid features:
âĢ
TokenâĢ
Feature activation+0.021
Top resid features:
ľ
Tokenľ
Feature activation+0.030
Top resid features:
If
TokenIf
Feature activation+1.650
Top resid features:
anybody
Token anybody
Feature activation+0.028
Top resid features:
is
Token is
Feature activation+0.016
Top resid features:
an
Token an
Feature activation-0.004
Top resid features:
expert
Token expert
Feature activation+0.066
Top resid features:
in
Token in
Feature activation-0.038
Top resid features:
is
Token is
Feature activation+0.008
Top resid features:
imperative
Token imperative
Feature activation+0.016
Top resid features:
.
Token.
Feature activation+0.016
Top resid features:
Ċ
TokenĊ
Feature activation+0.021
Top resid features:
Ċ
TokenĊ
Feature activation+0.012
Top resid features:
If
TokenIf
Feature activation+1.691
Top resid features:
you
Token you
Feature activation+0.039
Top resid features:
don
Token don
Feature activation+0.028
Top resid features:
âĢ
TokenâĢ
Feature activation+0.053
Top resid features:
Ļ
TokenĻ
Feature activation+0.009
Top resid features:
t
Tokent
Feature activation+0.006
Top resid features:
reason
Token reason
Feature activation+0.012
Top resid features:
this
Token this
Feature activation+0.006
Top resid features:
fish
Token fish
Feature activation+0.014
Top resid features:
survived
Token survived
Feature activation+0.036
Top resid features:
.
Token.
Feature activation+0.041
Top resid features:
If
Token If
Feature activation+2.131
Top resid features:
it
Token it
Feature activation+0.015
Top resid features:
had
Token had
Feature activation-0.018
Top resid features:
been
Token been
Feature activation+0.007
Top resid features:
located
Token located
Feature activation+0.032
Top resid features:
an
Token an
Feature activation+0.004
Top resid features:
news
Token news
Feature activation+0.011
Top resid features:
sites
Token sites
Feature activation+0.013
Top resid features:
and
Token and
Feature activation-0.005
Top resid features:
more
Token more
Feature activation+0.012
Top resid features:
.
Token.
Feature activation+0.017
Top resid features:
If
Token If
Feature activation+1.641
Top resid features:
we
Token we
Feature activation+0.021
Top resid features:
want
Token want
Feature activation+0.022
Top resid features:
a
Token a
Feature activation-0.025
Top resid features:
healthy
Token healthy
Feature activation+0.053
Top resid features:
reading
Token reading
Feature activation+0.065
Top resid features:
may
Token may
Feature activation+0.012
Top resid features:
follow
Token follow
Feature activation+0.010
Top resid features:
.
Token.
Feature activation+0.034
Top resid features:
âĢ
Token âĢ
Feature activation+0.032
Top resid features:
ľ
Tokenľ
Feature activation+0.027
Top resid features:
If
TokenIf
Feature activation+1.824
Top resid features:
you
Token you
Feature activation+0.044
Top resid features:
look
Token look
Feature activation-0.010
Top resid features:
at
Token at
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
client
Token client
Feature activation+0.019
Top resid features:
99
Token99
Feature activation+0.007
Top resid features:
-
Token-
Feature activation+0.004
Top resid features:
8
Token8
Feature activation+0.005
Top resid features:
559
Token559
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.033
Top resid features:
If
Token If
Feature activation+1.463
Top resid features:
you
Token you
Feature activation+0.025
Top resid features:
get
Token get
Feature activation+0.023
Top resid features:
an
Token an
Feature activation-0.032
Top resid features:
answering
Token answering
Feature activation+0.028
Top resid features:
machine
Token machine
Feature activation+0.095
Top resid features:
this
Token this
Feature activation+0.008
Top resid features:
information
Token information
Feature activation+0.019
Top resid features:
.
Token.
Feature activation+0.025
Top resid features:
Ċ
TokenĊ
Feature activation+0.018
Top resid features:
Ċ
TokenĊ
Feature activation+0.013
Top resid features:
If
TokenIf
Feature activation+1.676
Top resid features:
there
Token there
Feature activation+0.009
Top resid features:
are
Token are
Feature activation+0.029
Top resid features:
going
Token going
Feature activation+0.017
Top resid features:
to
Token to
Feature activation-0.005
Top resid features:
be
Token be
Feature activation+0.031
Top resid features:
99
Token99
Feature activation+0.002
Top resid features:
-
Token-
Feature activation+0.003
Top resid features:
8
Token8
Feature activation+0.006
Top resid features:
559
Token559
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.033
Top resid features:
If
Token If
Feature activation+1.730
Top resid features:
you
Token you
Feature activation+0.036
Top resid features:
get
Token get
Feature activation+0.016
Top resid features:
an
Token an
Feature activation+0.006
Top resid features:
answering
Token answering
Feature activation+0.055
Top resid features:
machine
Token machine
Feature activation+0.077
Top resid features:
R
Token R
Feature activation+0.005
Top resid features:
MD
TokenMD
Feature activation+0.005
Top resid features:
.
Token.
Feature activation+0.012
Top resid features:
Ċ
TokenĊ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation-0.004
Top resid features:
If
TokenIf
Feature activation+1.788
Top resid features:
you
Token you
Feature activation+0.030
Top resid features:
are
Token are
Feature activation-0.009
Top resid features:
interested
Token interested
Feature activation-0.049
Top resid features:
in
Token in
Feature activation-0.047
Top resid features:
a
Token a
Feature activation-0.040
Top resid features:

Decoder Weights Distribution

Head 0: 0.05

Head 1: 0.04

Head 2: 0.08

Head 3: 0.05

Head 4: 0.03

Head 5: 0.12

Head 6: 0.27

Head 7: 0.08

Head 8: 0.04

Head 9: 0.04

Head 10: 0.15

Head 11: 0.07

Positive logits

xus1.58

?'1.53

uld1.49

esson1.48

anymore1.48

?",1.47

?」1.42

))1.42

racuse1.41

llor1.41

?'"1.39

illac1.38

ado1.38

lear1.36

tek1.36

gdala1.34

please1.33

odan1.33

certific1.31

oute1.30

Negative logits

-1.39

subur-1.29

EV-1.26

IV-1.25

aves-1.17

pit-1.14

manifest-1.11

PD-1.10

Victoria-1.10

envelope-1.08

REM-1.08

ang-1.08

-1.08

oha-1.08

verbal-1.07

pers-1.07

lip-1.07

accommodate-1.06

EStreamFrame-1.06

plac-1.06

INTERVAL 1.742 - 1.935
CONTAINS 0.001%

some
Token some
Feature activation+0.042
common
Token common
Feature activation+0.000
sense
Token sense
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
,
Token,
Feature activation+1.105
read
Token read
Feature activation+1.935
the
Token the
Feature activation+0.714
whole
Token whole
Feature activation+0.247
thing
Token thing
Feature activation+0.539
.
Token.
Feature activation+0.195
He
Token He
Feature activation+0.000
you
Token you
Feature activation+0.030
are
Token are
Feature activation+0.000
using
Token using
Feature activation+0.000
Bitcoin
Token Bitcoin
Feature activation+0.000
,
Token,
Feature activation+0.761
you
Token you
Feature activation+1.870
are
Token are
Feature activation+0.635
making
Token making
Feature activation+0.210
the
Token the
Feature activation+0.390
world
Token world
Feature activation+0.000
a
Token a
Feature activation+0.023
your
Token your
Feature activation+0.215
BlackBerry
Token BlackBerry
Feature activation+0.000
Play
Token Play
Feature activation+0.000
Book
TokenBook
Feature activation+0.000
,
Token,
Feature activation+0.844
you
Token you
Feature activation+1.754
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
probably
Token probably
Feature activation+0.305
wondering
Token wondering
Feature activation+0.099
best
Token best
Feature activation+0.158
choice
Token choice
Feature activation+0.000
for
Token for
Feature activation+0.000
you
Token you
Feature activation+0.237
,
Token,
Feature activation+0.499
here
Token here
Feature activation+1.774
are
Token are
Feature activation+0.673
top
Token top
Feature activation+0.161
10
Token 10
Feature activation+0.000
famous
Token famous
Feature activation+0.000
websites
Token websites
Feature activation+0.000
the
Token the
Feature activation+0.026
republic
Token republic
Feature activation+0.000
of
Token of
Feature activation+0.000
Ireland
Token Ireland
Feature activation+0.000
,
Token,
Feature activation+1.092
you
Token you
Feature activation+1.746
may
Token may
Feature activation+0.956
be
Token be
Feature activation+0.385
astonished
Token astonished
Feature activation+0.033
to
Token to
Feature activation+0.092
see
Token see
Feature activation+0.368

INTERVAL 1.548 - 1.742
CONTAINS 0.001%

get
Token get
Feature activation+0.000
an
Token an
Feature activation+0.000
answering
Token answering
Feature activation+0.235
machine
Token machine
Feature activation+0.000
,
Token,
Feature activation+1.121
leave
Token leave
Feature activation+1.549
a
Token a
Feature activation+0.809
message
Token message
Feature activation+1.527
,
Token,
Feature activation+0.965
we
Token we
Feature activation+1.192
are
Token are
Feature activation+0.581
in
Token in
Feature activation+0.000
spoon
Token spoon
Feature activation+0.126
-
Token-
Feature activation+0.000
feeding
Tokenfeeding
Feature activation+0.000
,
Token,
Feature activation+0.960
it
Token it
Feature activation+1.633
is
Token is
Feature activation+1.135
the
Token the
Feature activation+0.629
liberal
Token liberal
Feature activation+0.000
media
Token media
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
healthy
Token healthy
Feature activation+0.000
reading
Token reading
Feature activation+0.018
culture
Token culture
Feature activation+0.000
,
Token,
Feature activation+0.727
we
Token we
Feature activation+1.564
have
Token have
Feature activation+0.537
to
Token to
Feature activation+0.405
work
Token work
Feature activation+0.261
hard
Token hard
Feature activation+0.033
to
Token to
Feature activation+0.000
two
Token two
Feature activation+0.000
forward
Token forward
Feature activation+0.010
,
Token,
Feature activation+0.707
the
Token the
Feature activation+0.877
fish
Token fish
Feature activation+0.297
would
Token would
Feature activation+1.569
have
Token have
Feature activation+0.271
likely
Token likely
Feature activation+0.428
never
Token never
Feature activation+0.251
fed
Token fed
Feature activation+0.000
.
Token.
Feature activation+0.000
of
Token of
Feature activation+0.000
catching
Token catching
Feature activation+0.000
a
Token a
Feature activation+0.000
Pokemon
Token Pokemon
Feature activation+0.000
,
Token,
Feature activation+0.981
you
Token you
Feature activation+1.703
need
Token need
Feature activation+0.907
to
Token to
Feature activation+0.137
wait
Token wait
Feature activation+0.533
until
Token until
Feature activation+0.241
the
Token the
Feature activation+0.000

INTERVAL 1.355 - 1.548
CONTAINS 0.002%

fit
Token fit
Feature activation+0.000
me
Token me
Feature activation+0.000
,
Token,
Feature activation+0.393
my
Token my
Feature activation+0.631
brother
Token brother
Feature activation+0.041
would
Token would
Feature activation+1.494
be
Token be
Feature activation+0.582
getting
Token getting
Feature activation+0.092
something
Token something
Feature activation+0.082
else
Token else
Feature activation+0.000
for
Token for
Feature activation+0.000
garden
Token garden
Feature activation+0.000
and
Token and
Feature activation+0.129
a
Token a
Feature activation+0.000
library
Token library
Feature activation+0.000
,
Token,
Feature activation+0.454
you
Token you
Feature activation+1.400
have
Token have
Feature activation+0.387
everything
Token everything
Feature activation+0.319
you
Token you
Feature activation+0.105
need
Token need
Feature activation+0.000
."
Token."
Feature activation+0.000
suggest
Token suggest
Feature activation+0.199
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
class
Token class
Feature activation+0.000
,
Token,
Feature activation+0.683
feel
Token feel
Feature activation+1.396
free
Token free
Feature activation+0.389
to
Token to
Feature activation+0.368
call
Token call
Feature activation+0.384
us
Token us
Feature activation+0.000
at
Token at
Feature activation+0.000
Subway
Token Subway
Feature activation+0.000
Surf
Token Surf
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
II
Token II
Feature activation+0.000
,
Token,
Feature activation+0.774
we
Token we
Feature activation+1.484
âĢ
TokenâĢ
Feature activation+0.143
Ļ
TokenĻ
Feature activation+0.000
d
Tokend
Feature activation+0.000
be
Token be
Feature activation+0.539
splitting
Token splitting
Feature activation+0.000
cannot
Token cannot
Feature activation+0.000
produce
Token produce
Feature activation+0.000
prog
Token prog
Feature activation+0.000
eny
Tokeneny
Feature activation+0.000
,
Token,
Feature activation+0.858
then
Token then
Feature activation+1.480
something
Token something
Feature activation+0.770
must
Token must
Feature activation+0.947
be
Token be
Feature activation+0.127
done
Token done
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 1.161 - 1.355
CONTAINS 0.004%

are
Token are
Feature activation+0.000
thinking
Token thinking
Feature activation+0.000
about
Token about
Feature activation+0.000
it
Token it
Feature activation+0.020
,
Token,
Feature activation+0.531
you
Token you
Feature activation+1.347
should
Token should
Feature activation+0.544
start
Token start
Feature activation+0.107
early
Token early
Feature activation+0.000
.
Token.
Feature activation+0.011
Min
Token Min
Feature activation+0.000
-
Token-
Feature activation+0.000
style
Tokenstyle
Feature activation+0.000
sports
Token sports
Feature activation+0.000
bar
Token bar
Feature activation+0.000
,
Token,
Feature activation+1.423
this
Token this
Feature activation+1.225
might
Token might
Feature activation+1.296
be
Token be
Feature activation+0.607
the
Token the
Feature activation+0.205
result
Token result
Feature activation+0.000
.
Token.
Feature activation+0.000
look
Token look
Feature activation+0.763
on
Token on
Feature activation+0.036
the
Token the
Feature activation+0.000
surface
Token surface
Feature activation+0.000
,
Token,
Feature activation+0.693
who
Token who
Feature activation+1.302
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
the
Token the
Feature activation+0.063
real
Token real
Feature activation+0.000
manure
Token manure
Feature activation+0.000
from
Token from
Feature activation+0.036
a
Token a
Feature activation+0.000
garden
Token garden
Feature activation+0.000
center
Token center
Feature activation+0.000
,
Token,
Feature activation+1.163
this
Token this
Feature activation+1.055
is
Token is
Feature activation+0.996
the
Token the
Feature activation+0.451
best
Token best
Feature activation+0.533
kind
Token kind
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
divorced
Token divorced
Feature activation+0.000
family
Token family
Feature activation+0.000
,
Token,
Feature activation+1.197
the
Token the
Feature activation+1.285
eldest
Token eldest
Feature activation+0.000
daughter
Token daughter
Feature activation+0.010
was
Token was
Feature activation+0.621
more
Token more
Feature activation+0.263
likely
Token likely
Feature activation+0.220

INTERVAL 0.968 - 1.161
CONTAINS 0.007%

or
Token or
Feature activation+0.000
somewhere
Token somewhere
Feature activation+0.000
in
Token in
Feature activation+0.000
Europe
Token Europe
Feature activation+0.000
,
Token,
Feature activation+0.373
you
Token you
Feature activation+1.070
may
Token may
Feature activation+0.351
know
Token know
Feature activation+0.000
what
Token what
Feature activation+0.045
it
Token it
Feature activation+0.000
feels
Token feels
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
Kiw
Token Kiw
Feature activation+0.000
is
Tokenis
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.369
who
Tokenwho
Feature activation+1.030
will
Token will
Feature activation+0.084
have
Token have
Feature activation+0.000
Abby
Token Abby
Feature activation+0.000
Er
Token Er
Feature activation+0.000
ce
Tokence
Feature activation+0.000
that
Token that
Feature activation+0.000
line
Token line
Feature activation+0.000
of
Token of
Feature activation+0.000
thinking
Token thinking
Feature activation+0.110
,
Token,
Feature activation+0.623
consider
Token consider
Feature activation+1.126
the
Token the
Feature activation+0.324
2016
Token 2016
Feature activation+0.000
college
Token college
Feature activation+0.000
football
Token football
Feature activation+0.044
season
Token season
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
City
Token City
Feature activation+0.000
contract
Token contract
Feature activation+0.000
,
Token,
Feature activation+0.438
it
Token it
Feature activation+1.068
would
Token would
Feature activation+0.949
be
Token be
Feature activation+0.000
best
Token best
Feature activation+0.000
if
Token if
Feature activation+0.000
the
Token the
Feature activation+0.000
of
Token of
Feature activation+0.000
mouth
Token mouth
Feature activation+0.000
is
Token is
Feature activation+0.000
good
Token good
Feature activation+0.341
,
Token,
Feature activation+0.553
it
Token it
Feature activation+1.158
could
Token could
Feature activation+0.801
hold
Token hold
Feature activation+0.056
on
Token on
Feature activation+0.000
well
Token well
Feature activation+0.193
but
Token but
Feature activation+0.364

INTERVAL 0.774 - 0.968
CONTAINS 0.011%

If
Token If
Feature activation+0.000
enacted
Token enacted
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.349
laws
Token laws
Feature activation+0.000
would
Token would
Feature activation+0.881
represent
Token represent
Feature activation+0.000
the
Token the
Feature activation+0.076
first
Token first
Feature activation+0.000
roll
Token roll
Feature activation+0.000
back
Tokenback
Feature activation+0.000
trading
Token trading
Feature activation+0.000
under
Token under
Feature activation+0.000
WTO
Token WTO
Feature activation+0.000
rules
Token rules
Feature activation+0.000
is
Token is
Feature activation+0.000
so
Token so
Feature activation+0.942
bad
Token bad
Feature activation+0.168
,
Token,
Feature activation+0.872
how
Token how
Feature activation+1.322
does
Token does
Feature activation+0.688
the
Token the
Feature activation+0.131
or
Token or
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
polar
Token polar
Feature activation+0.000
region
Token region
Feature activation+0.000
,
Token,
Feature activation+0.777
then
Token then
Feature activation+1.308
the
Token the
Feature activation+0.533
natural
Token natural
Feature activation+0.000
state
Token state
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
executive
Token executive
Feature activation+0.000
pardon
Token pardon
Feature activation+0.000
,
Token,
Feature activation+0.774
it
Token it
Feature activation+1.209
can
Token can
Feature activation+0.794
be
Token be
Feature activation+0.346
granted
Token granted
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
offending
Token offending
Feature activation+0.000
had
Token had
Feature activation+0.000
dri
Token dri
Feature activation+0.000
er
Tokener
Feature activation+0.000
conditions
Token conditions
Feature activation+0.000
,
Token,
Feature activation+0.321
as
Token as
Feature activation+0.814
we
Token we
Feature activation+0.360
'll
Token'll
Feature activation+0.431
start
Token start
Feature activation+0.000
to
Token to
Feature activation+0.000
see
Token see
Feature activation+0.147

INTERVAL 0.581 - 0.774
CONTAINS 0.019%

common
Token common
Feature activation+0.000
sense
Token sense
Feature activation+0.000
abuse
Token abuse
Feature activation+0.000
,
Token,
Feature activation+1.105
read
Token read
Feature activation+1.935
the
Token the
Feature activation+0.714
whole
Token whole
Feature activation+0.247
thing
Token thing
Feature activation+0.539
.
Token.
Feature activation+0.195
He
Token He
Feature activation+0.000
'll
Token'll
Feature activation+0.000
group
Token group
Feature activation+0.000
of
Token of
Feature activation+0.000
g
Token g
Feature activation+0.000
eeks
Tokeneeks
Feature activation+0.000
,
Token,
Feature activation+0.444
have
Token have
Feature activation+0.650
a
Token a
Feature activation+0.156
meet
Token meet
Feature activation+0.000
up
Tokenup
Feature activation+0.000
plan
Token plan
Feature activation+0.000
and
Token and
Feature activation+0.000
don
Token don
Feature activation+0.000
't
Token't
Feature activation+0.000
reverse
Token reverse
Feature activation+0.000
the
Token the
Feature activation+0.000
decision
Token decision
Feature activation+0.000
,
Token,
Feature activation+0.602
they
Token they
Feature activation+0.753
should
Token should
Feature activation+0.420
be
Token be
Feature activation+0.073
removed
Token removed
Feature activation+0.000
from
Token from
Feature activation+0.000
it
Token it
Feature activation+0.000
is
Token is
Feature activation+0.000
inaccessible
Token inaccessible
Feature activation+0.000
,
Token,
Feature activation+0.289
they
Token they
Feature activation+0.693
will
Token will
Feature activation+0.590
usually
Token usually
Feature activation+0.313
stand
Token stand
Feature activation+0.000
near
Token near
Feature activation+0.000
where
Token where
Feature activation+0.072
they
Token they
Feature activation+0.000
want
Token want
Feature activation+0.000
a
Token a
Feature activation+0.000
healthy
Token healthy
Feature activation+0.000
reading
Token reading
Feature activation+0.018
culture
Token culture
Feature activation+0.000
,
Token,
Feature activation+0.727
we
Token we
Feature activation+1.564
have
Token have
Feature activation+0.537
to
Token to
Feature activation+0.405
work
Token work
Feature activation+0.261
hard
Token hard
Feature activation+0.033

INTERVAL 0.387 - 0.581
CONTAINS 0.028%

who
Token who
Feature activation+0.538
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
d
Tokend
Feature activation+0.000
prefer
Token prefer
Feature activation+0.577
to
Token to
Feature activation+0.406
stay
Token stay
Feature activation+0.239
far
Token far
Feature activation+0.000
away
Token away
Feature activation+0.000
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.000
fters
Tokenfters
Feature activation+0.000
at
Token at
Feature activation+0.000
Staples
Token Staples
Feature activation+0.000
Center
Token Center
Feature activation+0.000
,
Token,
Feature activation+0.461
few
Token few
Feature activation+0.528
will
Token will
Feature activation+0.542
remember
Token remember
Feature activation+0.300
or
Token or
Feature activation+0.000
care
Token care
Feature activation+0.000
how
Token how
Feature activation+0.000
,
Token,
Feature activation+0.774
we
Token we
Feature activation+1.484
âĢ
TokenâĢ
Feature activation+0.143
Ļ
TokenĻ
Feature activation+0.000
d
Tokend
Feature activation+0.000
be
Token be
Feature activation+0.539
splitting
Token splitting
Feature activation+0.000
our
Token our
Feature activation+0.051
player
Token player
Feature activation+0.000
base
Tokenbase
Feature activation+0.000
instead
Token instead
Feature activation+0.244
pass
Token pass
Feature activation+0.000
this
Token this
Feature activation+0.000
bad
Token bad
Feature activation+0.000
boy
Token boy
Feature activation+0.000
,
Token,
Feature activation+0.221
we
Token we
Feature activation+0.571
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
going
Token going
Feature activation+0.000
to
Token to
Feature activation+0.000
this
Token this
Feature activation+0.000
kind
Token kind
Feature activation+0.000
of
Token of
Feature activation+0.000
inequality
Token inequality
Feature activation+0.000
off
Token off
Feature activation+0.000
,
Token,
Feature activation+0.520
you
Token you
Feature activation+1.053
would
Token would
Feature activation+0.708
also
Token also
Feature activation+0.000
have
Token have
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 0.194 - 0.387
CONTAINS 0.047%

of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.050
brain
Token brain
Feature activation+0.000
,
Token,
Feature activation+0.818
even
Token even
Feature activation+0.835
if
Token if
Feature activation+0.196
the
Token the
Feature activation+0.000
skull
Token skull
Feature activation+0.000
and
Token and
Feature activation+0.000
outer
Token outer
Feature activation+0.081
layer
Token layer
Feature activation+0.000
year
Token year
Feature activation+0.000
it
Token it
Feature activation+0.000
was
Token was
Feature activation+0.000
,
Token,
Feature activation+0.000
we
Token we
Feature activation+0.136
would
Token would
Feature activation+0.306
probably
Token probably
Feature activation+0.000
politely
Token politely
Feature activation+0.000
answer
Token answer
Feature activation+0.000
that
Token that
Feature activation+0.000
it
Token it
Feature activation+0.000
separate
Token separate
Feature activation+0.000
dining
Token dining
Feature activation+0.000
room
Token room
Feature activation+0.000
,
Token,
Feature activation+0.678
rather
Token rather
Feature activation+0.614
than
Token than
Feature activation+0.327
a
Token a
Feature activation+0.000
rudimentary
Token rudimentary
Feature activation+0.000
"
Token "
Feature activation+0.000
d
Tokend
Feature activation+0.000
ining
Tokenining
Feature activation+0.000
,
Token,
Feature activation+1.294
âĢ
TokenâĢ
Feature activation+0.470
Ŀ
TokenĿ
Feature activation+0.160
he
Token he
Feature activation+0.758
says
Token says
Feature activation+0.116
,
Token,
Feature activation+0.292
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
I
TokenI
Feature activation+0.000
will
Token will
Feature activation+0.000
speak
Token speak
Feature activation+0.000
If
TokenIf
Feature activation+0.000
you
Token you
Feature activation+0.000
have
Token have
Feature activation+0.000
a
Token a
Feature activation+0.000
device
Token device
Feature activation+0.000
that
Token that
Feature activation+0.263
contains
Token contains
Feature activation+0.000
a
Token a
Feature activation+0.000
gy
Token gy
Feature activation+0.000
ro
Tokenro
Feature activation+0.000
scope
Tokenscope
Feature activation+0.000

INTERVAL 0.000 - 0.194
CONTAINS 99.880%

IT
TokenIT
Feature activation+0.000
ICAL
TokenICAL
Feature activation+0.000
PROV
Token PROV
Feature activation+0.000
OC
TokenOC
Feature activation+0.000
ATION
TokenATION
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
Ser
TokenSer
Feature activation+0.000
bian
Tokenbian
Feature activation+0.000
only
Token only
Feature activation+0.000
show
Token show
Feature activation+0.000
the
Token the
Feature activation+0.000
Name
Token Name
Feature activation+0.000
Your
Token Your
Feature activation+0.000
Price
Token Price
Feature activation+0.000
comp
Token comp
Feature activation+0.000
il
Tokenil
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
.
Token.
Feature activation+0.000
To
Token To
Feature activation+0.000
don
Token don
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
care
Token care
Feature activation+0.000
,
Token,
Feature activation+0.000
glory
Token glory
Feature activation+0.000
ing
Tokening
Feature activation+0.000
in
Token in
Feature activation+0.000
their
Token their
Feature activation+0.000
ignorance
Token ignorance
Feature activation+0.000
of
Token of
Feature activation+0.000
theoretical
Token theoretical
Feature activation+0.000
morphology
Token morphology
Feature activation+0.000
of
Token of
Feature activation+0.000
biological
Token biological
Feature activation+0.000
shapes
Token shapes
Feature activation+0.000
,
Token,
Feature activation+0.000
co
Token co
Feature activation+0.000
iling
Tokeniling
Feature activation+0.000
shells
Token shells
Feature activation+0.000
have
Token have
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Critics
TokenCritics
Feature activation+0.000
say
Token say
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
administration
Token administration
Feature activation+0.000
has
Token has
Feature activation+0.000
not
Token not
Feature activation+0.000
moved
Token moved
Feature activation+0.000
quickly
Token quickly
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 H2.6: (feature 22204)

TOP ACTIVATIONS
MAX = 2.916

:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729
Image
Token Image
Feature activation+0.000
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
Sheriff
Token Sheriff
Feature activation+2.486
's
Token's
Feature activation+1.903
Office
Token Office
Feature activation+1.973
Image
Token Image
Feature activation+0.496
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729
Image
Token Image
Feature activation+0.000
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
/
Token /
Feature activation+0.000
:
Token:
Feature activation+0.115
Mike
Token Mike
Feature activation+1.282
K
Token K
Feature activation+1.865
ep
Tokenep
Feature activation+0.816
ka
Tokenka
Feature activation+0.893
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
Sheriff
Token Sheriff
Feature activation+1.750
's
Token's
Feature activation+1.267
Office
Token Office
Feature activation+1.284
Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
Sheriff
Token Sheriff
Feature activation+2.486
's
Token's
Feature activation+1.903
Adam
Token Adam
Feature activation+1.027
Cohen
Token Cohen
Feature activation+1.121
Photo
Token Photo
Feature activation+1.771
:
Token:
Feature activation+1.207
Adam
Token Adam
Feature activation+2.233
Cohen
Token Cohen
Feature activation+2.342
Image
Token Image
Feature activation+0.918
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.091
/
Token /
Feature activation+0.531
4
Token 4
Feature activation+0.000
Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
Sheriff
Token Sheriff
Feature activation+2.486
's
Token's
Feature activation+1.903
Office
Token Office
Feature activation+1.973
Image
Token Image
Feature activation+0.496
1
Token 1
Feature activation+0.000
:
Token:
Feature activation+0.000
Adam
Token Adam
Feature activation+1.027
Cohen
Token Cohen
Feature activation+1.121
Photo
Token Photo
Feature activation+1.771
:
Token:
Feature activation+1.207
Adam
Token Adam
Feature activation+2.233
Cohen
Token Cohen
Feature activation+2.342
Image
Token Image
Feature activation+0.918
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.091
/
Token /
Feature activation+0.531
net
Tokennet
Feature activation+0.449
Photo
Token Photo
Feature activation+1.554
:
Token:
Feature activation+0.636
NE
Token NE
Feature activation+1.988
IL
TokenIL
Feature activation+1.454
BL
Token BL
Feature activation+2.145
A
TokenA
Feature activation+1.253
KE
TokenKE
Feature activation+1.263
|
Token |
Feature activation+1.714
N
Token N
Feature activation+1.335
bl
Tokenbl
Feature activation+1.007
n
Tokenn
Feature activation+0.143
.
Token.
Feature activation+0.631
net
Tokennet
Feature activation+0.449
Photo
Token Photo
Feature activation+1.554
:
Token:
Feature activation+0.636
NE
Token NE
Feature activation+1.988
IL
TokenIL
Feature activation+1.454
BL
Token BL
Feature activation+2.145
A
TokenA
Feature activation+1.253
KE
TokenKE
Feature activation+1.263
|
Token |
Feature activation+1.714
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
Sheriff
Token Sheriff
Feature activation+2.486
's
Token's
Feature activation+1.903
Office
Token Office
Feature activation+1.973
Image
Token Image
Feature activation+0.496
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
/
Token /
Feature activation+0.210
39
Token 39
Feature activation+0.000
Photo
Token Photo
Feature activation+0.000
:
Token:
Feature activation+0.000
Jeff
Token Jeff
Feature activation+0.665
Ch
Token Ch
Feature activation+1.033
iu
Tokeniu
Feature activation+0.700
/
Token /
Feature activation+1.931
Associated
Token Associated
Feature activation+1.671
Press
Token Press
Feature activation+0.960
Image
Token Image
Feature activation+0.000
10
Token 10
Feature activation+0.000
of
Token of
Feature activation+0.000
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729
Image
Token Image
Feature activation+0.000
1
Token 1
Feature activation+0.000
Mike
Token Mike
Feature activation+1.282
K
Token K
Feature activation+1.865
ep
Tokenep
Feature activation+0.816
ka
Tokenka
Feature activation+0.893
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
Sheriff
Token Sheriff
Feature activation+2.486
's
Token's
Feature activation+1.903
Office
Token Office
Feature activation+1.973
Image
Token Image
Feature activation+0.496
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
/
Token /
Feature activation+0.210
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
alf
Tokenalf
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+0.656
:
Token:
Feature activation+0.115
Mike
Token Mike
Feature activation+1.282
K
Token K
Feature activation+1.865
ep
Tokenep
Feature activation+0.816
ka
Tokenka
Feature activation+0.893
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+0.407
:
Token:
Feature activation+0.000
Adam
Token Adam
Feature activation+1.027
Cohen
Token Cohen
Feature activation+1.121
Photo
Token Photo
Feature activation+1.771
:
Token:
Feature activation+1.207
Adam
Token Adam
Feature activation+2.233
Cohen
Token Cohen
Feature activation+2.342
Image
Token Image
Feature activation+0.918
1
Token 1
Feature activation+0.000
ar
Tokenar
Feature activation+0.280
County
Token County
Feature activation+1.437
Sheriff
Token Sheriff
Feature activation+1.750
's
Token's
Feature activation+1.267
Office
Token Office
Feature activation+1.284
Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317

Top DFA by src position
MAX = 2.933

ep
Tokenep
Feature activation+0.009
Top resid features:
ka
Tokenka
Feature activation+0.015
Top resid features:
,
Token,
Feature activation+0.010
Top resid features:
The
Token The
Feature activation+0.086
Top resid features:
Chronicle
Token Chronicle
Feature activation+0.345
Top resid features:
Photo
Token Photo
Feature activation+1.893
Top resid features:
:
Token:
Feature activation+0.203
Top resid features:
Mike
Token Mike
Feature activation+0.286
Top resid features:
K
Token K
Feature activation+0.246
Top resid features:
ep
Tokenep
Feature activation+0.054
Top resid features:
ka
Tokenka
Feature activation+0.103
Top resid features:
ep
Tokenep
Feature activation+0.020
Top resid features:
ka
Tokenka
Feature activation+0.025
Top resid features:
,
Token,
Feature activation+0.042
Top resid features:
The
Token The
Feature activation+0.110
Top resid features:
Chronicle
Token Chronicle
Feature activation+0.378
Top resid features:
Photo
Token Photo
Feature activation+1.895
Top resid features:
:
Token:
Feature activation+0.280
Top resid features:
Mike
Token Mike
Feature activation+0.573
Top resid features:
K
Token K
Feature activation+0.332
Top resid features:
ep
Tokenep
Feature activation+0.000
Top resid features:
ka
Tokenka
Feature activation+0.000
Top resid features:
ar
Tokenar
Feature activation+0.014
Top resid features:
County
Token County
Feature activation+0.060
Top resid features:
Sheriff
Token Sheriff
Feature activation+0.072
Top resid features:
's
Token's
Feature activation+0.043
Top resid features:
Office
Token Office
Feature activation+0.038
Top resid features:
Photo
Token Photo
Feature activation+1.631
Top resid features:
:
Token:
Feature activation+0.239
Top resid features:
B
Token B
Feature activation+0.191
Top resid features:
ex
Tokenex
Feature activation+0.060
Top resid features:
ar
Tokenar
Feature activation+0.083
Top resid features:
County
Token County
Feature activation+0.343
Top resid features:
ep
Tokenep
Feature activation+0.009
Top resid features:
ka
Tokenka
Feature activation+0.014
Top resid features:
,
Token,
Feature activation+0.010
Top resid features:
The
Token The
Feature activation+0.034
Top resid features:
Chronicle
Token Chronicle
Feature activation+0.292
Top resid features:
Photo
Token Photo
Feature activation+1.878
Top resid features:
:
Token:
Feature activation+0.104
Top resid features:
Mike
Token Mike
Feature activation+0.241
Top resid features:
K
Token K
Feature activation+0.177
Top resid features:
ep
Tokenep
Feature activation+0.038
Top resid features:
ka
Tokenka
Feature activation+0.054
Top resid features:
San
Token San
Feature activation+0.015
Top resid features:
Francisco
Token Francisco
Feature activation-0.007
Top resid features:
C
Token C
Feature activation+0.043
Top resid features:
alf
Tokenalf
Feature activation+0.010
Top resid features:
.
Token.
Feature activation+0.081
Top resid features:
Photo
Token Photo
Feature activation+2.933
Top resid features:
:
Token:
Feature activation+0.230
Top resid features:
Mike
Token Mike
Feature activation+0.247
Top resid features:
K
Token K
Feature activation+0.217
Top resid features:
ep
Tokenep
Feature activation+0.011
Top resid features:
ka
Tokenka
Feature activation+0.004
Top resid features:
ar
Tokenar
Feature activation+0.033
Top resid features:
County
Token County
Feature activation+0.114
Top resid features:
Sheriff
Token Sheriff
Feature activation+0.116
Top resid features:
's
Token's
Feature activation+0.112
Top resid features:
Office
Token Office
Feature activation+0.073
Top resid features:
Photo
Token Photo
Feature activation+1.803
Top resid features:
:
Token:
Feature activation+0.307
Top resid features:
B
Token B
Feature activation+0.350
Top resid features:
ex
Tokenex
Feature activation+0.000
Top resid features:
ar
Tokenar
Feature activation+0.000
Top resid features:
County
Token County
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.108
Top resid features:
Photo
Token Photo
Feature activation+0.999
Top resid features:
:
Token:
Feature activation+0.071
Top resid features:
Adam
Token Adam
Feature activation+0.135
Top resid features:
Cohen
Token Cohen
Feature activation+0.133
Top resid features:
Photo
Token Photo
Feature activation+1.612
Top resid features:
:
Token:
Feature activation+0.295
Top resid features:
Adam
Token Adam
Feature activation+0.414
Top resid features:
Cohen
Token Cohen
Feature activation+0.376
Top resid features:
Image
Token Image
Feature activation+0.000
Top resid features:
1
Token 1
Feature activation+0.000
Top resid features:
ar
Tokenar
Feature activation+0.015
Top resid features:
County
Token County
Feature activation+0.072
Top resid features:
Sheriff
Token Sheriff
Feature activation+0.090
Top resid features:
's
Token's
Feature activation+0.058
Top resid features:
Office
Token Office
Feature activation+0.045
Top resid features:
Photo
Token Photo
Feature activation+1.663
Top resid features:
:
Token:
Feature activation+0.280
Top resid features:
B
Token B
Feature activation+0.227
Top resid features:
ex
Tokenex
Feature activation+0.068
Top resid features:
ar
Tokenar
Feature activation+0.100
Top resid features:
County
Token County
Feature activation+0.304
Top resid features:
.
Token.
Feature activation+0.112
Top resid features:
Photo
Token Photo
Feature activation+1.192
Top resid features:
:
Token:
Feature activation+0.072
Top resid features:
Adam
Token Adam
Feature activation+0.138
Top resid features:
Cohen
Token Cohen
Feature activation+0.163
Top resid features:
Photo
Token Photo
Feature activation+1.713
Top resid features:
:
Token:
Feature activation+0.248
Top resid features:
Adam
Token Adam
Feature activation+0.383
Top resid features:
Cohen
Token Cohen
Feature activation+0.000
Top resid features:
Image
Token Image
Feature activation+0.000
Top resid features:
1
Token 1
Feature activation+0.000
Top resid features:
@
Token@
Feature activation+0.025
Top resid features:
md
Tokenmd
Feature activation+0.034
Top resid features:
n
Tokenn
Feature activation+0.022
Top resid features:
.
Token.
Feature activation+0.071
Top resid features:
net
Tokennet
Feature activation+0.087
Top resid features:
Photo
Token Photo
Feature activation+1.981
Top resid features:
:
Token:
Feature activation+0.266
Top resid features:
NE
Token NE
Feature activation+0.243
Top resid features:
IL
TokenIL
Feature activation+0.122
Top resid features:
BL
Token BL
Feature activation+0.284
Top resid features:
A
TokenA
Feature activation+0.000
Top resid features:
@
Token@
Feature activation+0.041
Top resid features:
md
Tokenmd
Feature activation+0.050
Top resid features:
n
Tokenn
Feature activation+0.042
Top resid features:
.
Token.
Feature activation+0.072
Top resid features:
net
Tokennet
Feature activation+0.103
Top resid features:
Photo
Token Photo
Feature activation+2.133
Top resid features:
:
Token:
Feature activation+0.184
Top resid features:
NE
Token NE
Feature activation+0.276
Top resid features:
IL
TokenIL
Feature activation+0.000
Top resid features:
BL
Token BL
Feature activation+0.000
Top resid features:
A
TokenA
Feature activation+0.000
Top resid features:
ar
Tokenar
Feature activation+0.009
Top resid features:
County
Token County
Feature activation+0.052
Top resid features:
Sheriff
Token Sheriff
Feature activation+0.059
Top resid features:
's
Token's
Feature activation+0.022
Top resid features:
Office
Token Office
Feature activation+0.023
Top resid features:
Photo
Token Photo
Feature activation+1.332
Top resid features:
:
Token:
Feature activation+0.113
Top resid features:
B
Token B
Feature activation+0.111
Top resid features:
ex
Tokenex
Feature activation+0.037
Top resid features:
ar
Tokenar
Feature activation+0.042
Top resid features:
County
Token County
Feature activation+0.184
Top resid features:
to
Token to
Feature activation+0.020
Top resid features:
resume
Token resume
Feature activation+0.023
Top resid features:
his
Token his
Feature activation+0.050
Top resid features:
...
Token ...
Feature activation+0.074
Top resid features:
more
Token more
Feature activation+0.005
Top resid features:
Photo
Token Photo
Feature activation+2.744
Top resid features:
:
Token:
Feature activation+0.197
Top resid features:
Jeff
Token Jeff
Feature activation+0.183
Top resid features:
Ch
Token Ch
Feature activation+0.109
Top resid features:
iu
Tokeniu
Feature activation+0.019
Top resid features:
/
Token /
Feature activation+0.230
Top resid features:
ep
Tokenep
Feature activation+0.012
Top resid features:
ka
Tokenka
Feature activation+0.020
Top resid features:
,
Token,
Feature activation+0.048
Top resid features:
The
Token The
Feature activation+0.110
Top resid features:
Chronicle
Token Chronicle
Feature activation+0.298
Top resid features:
Photo
Token Photo
Feature activation+1.472
Top resid features:
:
Token:
Feature activation+0.207
Top resid features:
Mike
Token Mike
Feature activation+0.375
Top resid features:
K
Token K
Feature activation+0.324
Top resid features:
ep
Tokenep
Feature activation+0.111
Top resid features:
ka
Tokenka
Feature activation+0.091
Top resid features:
San
Token San
Feature activation+0.016
Top resid features:
Francisco
Token Francisco
Feature activation-0.005
Top resid features:
C
Token C
Feature activation+0.037
Top resid features:
alf
Tokenalf
Feature activation+0.009
Top resid features:
.
Token.
Feature activation+0.084
Top resid features:
Photo
Token Photo
Feature activation+2.393
Top resid features:
:
Token:
Feature activation+0.128
Top resid features:
Mike
Token Mike
Feature activation+0.218
Top resid features:
K
Token K
Feature activation+0.168
Top resid features:
ep
Tokenep
Feature activation+0.003
Top resid features:
ka
Tokenka
Feature activation-0.002
Top resid features:
ar
Tokenar
Feature activation+0.012
Top resid features:
County
Token County
Feature activation+0.058
Top resid features:
Sheriff
Token Sheriff
Feature activation+0.067
Top resid features:
's
Token's
Feature activation+0.020
Top resid features:
Office
Token Office
Feature activation+0.027
Top resid features:
Photo
Token Photo
Feature activation+1.295
Top resid features:
:
Token:
Feature activation+0.184
Top resid features:
B
Token B
Feature activation+0.151
Top resid features:
ex
Tokenex
Feature activation+0.040
Top resid features:
ar
Tokenar
Feature activation+0.065
Top resid features:
County
Token County
Feature activation+0.174
Top resid features:
ep
Tokenep
Feature activation+0.020
Top resid features:
ka
Tokenka
Feature activation+0.024
Top resid features:
,
Token,
Feature activation+0.055
Top resid features:
The
Token The
Feature activation+0.143
Top resid features:
Chronicle
Token Chronicle
Feature activation+0.284
Top resid features:
Photo
Token Photo
Feature activation+1.663
Top resid features:
:
Token:
Feature activation+0.250
Top resid features:
Mike
Token Mike
Feature activation+0.453
Top resid features:
K
Token K
Feature activation+0.000
Top resid features:
ep
Tokenep
Feature activation+0.000
Top resid features:
ka
Tokenka
Feature activation+0.000
Top resid features:
San
Token San
Feature activation+0.031
Top resid features:
Francisco
Token Francisco
Feature activation-0.008
Top resid features:
C
Token C
Feature activation+0.087
Top resid features:
alf
Tokenalf
Feature activation+0.020
Top resid features:
.
Token.
Feature activation+0.112
Top resid features:
Photo
Token Photo
Feature activation+2.394
Top resid features:
:
Token:
Feature activation+0.283
Top resid features:
Mike
Token Mike
Feature activation+0.367
Top resid features:
K
Token K
Feature activation+0.354
Top resid features:
ep
Tokenep
Feature activation+0.000
Top resid features:
ka
Tokenka
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.152
Top resid features:
Photo
Token Photo
Feature activation+1.041
Top resid features:
:
Token:
Feature activation+0.192
Top resid features:
Adam
Token Adam
Feature activation+0.266
Top resid features:
Cohen
Token Cohen
Feature activation+0.263
Top resid features:
Photo
Token Photo
Feature activation+1.546
Top resid features:
:
Token:
Feature activation+0.000
Top resid features:
Adam
Token Adam
Feature activation+0.000
Top resid features:
Cohen
Token Cohen
Feature activation+0.000
Top resid features:
Image
Token Image
Feature activation+0.000
Top resid features:
1
Token 1
Feature activation+0.000
Top resid features:
ar
Tokenar
Feature activation+0.037
Top resid features:
County
Token County
Feature activation+0.164
Top resid features:
Sheriff
Token Sheriff
Feature activation+0.112
Top resid features:
's
Token's
Feature activation+0.336
Top resid features:
Office
Token Office
Feature activation+0.115
Top resid features:
Photo
Token Photo
Feature activation+1.478
Top resid features:
:
Token:
Feature activation+0.000
Top resid features:
B
Token B
Feature activation+0.000
Top resid features:
ex
Tokenex
Feature activation+0.000
Top resid features:
ar
Tokenar
Feature activation+0.000
Top resid features:
County
Token County
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.07

Head 2: 0.05

Head 3: 0.08

Head 4: 0.03

Head 5: 0.07

Head 6: 0.27

Head 7: 0.06

Head 8: 0.09

Head 9: 0.06

Head 10: 0.11

Head 11: 0.05

Positive logits

(@1.68

]."1.64

)1.49

IMAGES1.44

)."1.33

]"1.32

?]1.32

url1.30

ubb1.29

.","1.28

..."1.28

.]1.26

]1.26

akers1.25

"],"1.25

});1.24

lyss1.23

Edison1.23

"]1.21

]}1.21

Negative logits

issance-1.60

tested-1.51

discrimination-1.50

ModLoader-1.41

psychiat-1.40

utical-1.34

Reincarn-1.33

proven-1.32

reincarn-1.32

bringing-1.31

gae-1.31

patient-1.31

Magicka-1.31

whichever-1.29

spir-1.29

utics-1.28

xual-1.28

hap-1.28

MMO-1.28

bucks-1.26

INTERVAL 2.624 - 2.916
CONTAINS 0.000%

:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729
Image
Token Image
Feature activation+0.000
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 2.332 - 2.624
CONTAINS 0.001%

:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
Sheriff
Token Sheriff
Feature activation+2.486
's
Token's
Feature activation+1.903
Office
Token Office
Feature activation+1.973
Image
Token Image
Feature activation+0.496
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
Adam
Token Adam
Feature activation+1.027
Cohen
Token Cohen
Feature activation+1.121
Photo
Token Photo
Feature activation+1.771
:
Token:
Feature activation+1.207
Adam
Token Adam
Feature activation+2.233
Cohen
Token Cohen
Feature activation+2.342
Image
Token Image
Feature activation+0.918
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.091
/
Token /
Feature activation+0.531
4
Token 4
Feature activation+0.000
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729
Image
Token Image
Feature activation+0.000
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
/
Token /
Feature activation+0.000
Sheriff
Token Sheriff
Feature activation+1.750
's
Token's
Feature activation+1.267
Office
Token Office
Feature activation+1.284
Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
Sheriff
Token Sheriff
Feature activation+2.486
's
Token's
Feature activation+1.903
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729

INTERVAL 2.041 - 2.332
CONTAINS 0.000%

Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
Sheriff
Token Sheriff
Feature activation+2.486
's
Token's
Feature activation+1.903
Office
Token Office
Feature activation+1.973
Image
Token Image
Feature activation+0.496
1
Token 1
Feature activation+0.000
net
Tokennet
Feature activation+0.449
Photo
Token Photo
Feature activation+1.554
:
Token:
Feature activation+0.636
NE
Token NE
Feature activation+1.988
IL
TokenIL
Feature activation+1.454
BL
Token BL
Feature activation+2.145
A
TokenA
Feature activation+1.253
KE
TokenKE
Feature activation+1.263
|
Token |
Feature activation+1.714
N
Token N
Feature activation+1.335
bl
Tokenbl
Feature activation+1.007
:
Token:
Feature activation+0.000
Adam
Token Adam
Feature activation+1.027
Cohen
Token Cohen
Feature activation+1.121
Photo
Token Photo
Feature activation+1.771
:
Token:
Feature activation+1.207
Adam
Token Adam
Feature activation+2.233
Cohen
Token Cohen
Feature activation+2.342
Image
Token Image
Feature activation+0.918
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.091
/
Token /
Feature activation+0.531

INTERVAL 1.749 - 2.041
CONTAINS 0.001%

Mike
Token Mike
Feature activation+1.282
K
Token K
Feature activation+1.865
ep
Tokenep
Feature activation+0.816
ka
Tokenka
Feature activation+0.893
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729
Image
Token Image
Feature activation+0.000
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
ar
Tokenar
Feature activation+0.280
County
Token County
Feature activation+1.437
Sheriff
Token Sheriff
Feature activation+1.750
's
Token's
Feature activation+1.267
Office
Token Office
Feature activation+1.284
Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
County
Token County
Feature activation+2.317
n
Tokenn
Feature activation+0.143
.
Token.
Feature activation+0.631
net
Tokennet
Feature activation+0.449
Photo
Token Photo
Feature activation+1.554
:
Token:
Feature activation+0.636
NE
Token NE
Feature activation+1.988
IL
TokenIL
Feature activation+1.454
BL
Token BL
Feature activation+2.145
A
TokenA
Feature activation+1.253
KE
TokenKE
Feature activation+1.263
|
Token |
Feature activation+1.714

INTERVAL 1.458 - 1.749
CONTAINS 0.001%

K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
,
Token,
Feature activation+2.657
The
Token The
Feature activation+2.469
Chronicle
Token Chronicle
Feature activation+1.729
Image
Token Image
Feature activation+0.000
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
/
Token /
Feature activation+0.000
1
Token 1
Feature activation+0.000
NE
Token NE
Feature activation+1.988
IL
TokenIL
Feature activation+1.454
BL
Token BL
Feature activation+2.145
A
TokenA
Feature activation+1.253
KE
TokenKE
Feature activation+1.263
|
Token |
Feature activation+1.714
N
Token N
Feature activation+1.335
bl
Tokenbl
Feature activation+1.007
ake
Tokenake
Feature activation+0.603
@
Token@
Feature activation+1.012
md
Tokenmd
Feature activation+0.653
ep
Tokenep
Feature activation+0.816
ka
Tokenka
Feature activation+0.893
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037
Mike
Token Mike
Feature activation+1.886
K
Token K
Feature activation+2.541
ep
Tokenep
Feature activation+1.755
ka
Tokenka
Feature activation+1.920
AN
TokenAN
Feature activation+0.000
Photo
Token Photo
Feature activation+0.792
/
Token/
Feature activation+1.031
Frank
Token Frank
Feature activation+1.217
Martinez
Token Martinez
Feature activation+1.105
.
Token.
Feature activation+1.704
Ċ
TokenĊ
Feature activation+0.051
Ċ
TokenĊ
Feature activation+0.000
Copyright
TokenCopyright
Feature activation+0.000
by
Token by
Feature activation+0.000
K
Token K
Feature activation+0.000
:
Token:
Feature activation+0.000
Jeff
Token Jeff
Feature activation+0.665
Ch
Token Ch
Feature activation+1.033
iu
Tokeniu
Feature activation+0.700
/
Token /
Feature activation+1.931
Associated
Token Associated
Feature activation+1.671
Press
Token Press
Feature activation+0.960
Image
Token Image
Feature activation+0.000
10
Token 10
Feature activation+0.000
of
Token of
Feature activation+0.000
13
Token 13
Feature activation+0.000

INTERVAL 1.166 - 1.458
CONTAINS 0.002%

ex
Tokenex
Feature activation+0.338
ar
Tokenar
Feature activation+0.280
County
Token County
Feature activation+1.437
Sheriff
Token Sheriff
Feature activation+1.750
's
Token's
Feature activation+1.267
Office
Token Office
Feature activation+1.284
Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
ar
Tokenar
Feature activation+1.415
K
Token K
Feature activation+0.000
X
TokenX
Feature activation+0.000
AN
TokenAN
Feature activation+0.000
Photo
Token Photo
Feature activation+0.792
/
Token/
Feature activation+1.031
Frank
Token Frank
Feature activation+1.217
Martinez
Token Martinez
Feature activation+1.105
.
Token.
Feature activation+1.704
Ċ
TokenĊ
Feature activation+0.051
Ċ
TokenĊ
Feature activation+0.000
Copyright
TokenCopyright
Feature activation+0.000
B
Token B
Feature activation+1.435
ex
Tokenex
Feature activation+0.338
ar
Tokenar
Feature activation+0.280
County
Token County
Feature activation+1.437
Sheriff
Token Sheriff
Feature activation+1.750
's
Token's
Feature activation+1.267
Office
Token Office
Feature activation+1.284
Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194
B
Token B
Feature activation+2.359
ex
Tokenex
Feature activation+1.580
C
Token C
Feature activation+0.000
alf
Tokenalf
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+0.656
:
Token:
Feature activation+0.115
Mike
Token Mike
Feature activation+1.282
K
Token K
Feature activation+1.865
ep
Tokenep
Feature activation+0.816
ka
Tokenka
Feature activation+0.893
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Photo
Token Photo
Feature activation+0.758
:
Token:
Feature activation+0.000
B
Token B
Feature activation+1.435
ex
Tokenex
Feature activation+0.338
ar
Tokenar
Feature activation+0.280
County
Token County
Feature activation+1.437
Sheriff
Token Sheriff
Feature activation+1.750
's
Token's
Feature activation+1.267
Office
Token Office
Feature activation+1.284
Photo
Token Photo
Feature activation+1.762
:
Token:
Feature activation+1.194

INTERVAL 0.875 - 1.166
CONTAINS 0.002%

Cohen
Token Cohen
Feature activation+1.121
Photo
Token Photo
Feature activation+1.771
:
Token:
Feature activation+1.207
Adam
Token Adam
Feature activation+2.233
Cohen
Token Cohen
Feature activation+2.342
Image
Token Image
Feature activation+0.918
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.091
/
Token /
Feature activation+0.531
4
Token 4
Feature activation+0.000
Caption
Token Caption
Feature activation+0.000
.
Token.
Feature activation+0.000
K
Token K
Feature activation+0.000
X
TokenX
Feature activation+0.000
AN
TokenAN
Feature activation+0.000
Photo
Token Photo
Feature activation+0.792
/
Token/
Feature activation+1.031
Frank
Token Frank
Feature activation+1.217
Martinez
Token Martinez
Feature activation+1.105
.
Token.
Feature activation+1.704
Ċ
TokenĊ
Feature activation+0.051
Ċ
TokenĊ
Feature activation+0.000
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.251
S
Token S
Feature activation+0.561
oren
Tokenoren
Feature activation+0.524
son
Tokenson
Feature activation+0.444
.
Token.
Feature activation+0.928
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Your
TokenYour
Feature activation+0.000
upper
Token upper
Feature activation+0.000
back
Token back
Feature activation+0.000
1962
Token 1962
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+0.516
:
Token:
Feature activation+0.000
CIA
Token CIA
Feature activation+0.702
.
Token.
Feature activation+1.044
gov
Tokengov
Feature activation+0.863
The
Token The
Feature activation+0.982
A
Token A
Feature activation+0.664
-
Token-
Feature activation+0.176
12
Token12
Feature activation+0.070
Photo
Token Photo
Feature activation+0.656
:
Token:
Feature activation+0.115
Mike
Token Mike
Feature activation+1.282
K
Token K
Feature activation+1.865
ep
Tokenep
Feature activation+0.816
ka
Tokenka
Feature activation+0.893
,
Token,
Feature activation+2.416
The
Token The
Feature activation+1.904
Chronicle
Token Chronicle
Feature activation+1.277
Photo
Token Photo
Feature activation+1.516
:
Token:
Feature activation+1.037

INTERVAL 0.583 - 0.875
CONTAINS 0.003%

ON
TokenON
Feature activation+0.000
TO
TokenTO
Feature activation+0.000
STAR
Token STAR
Feature activation+0.000
FILE
Token FILE
Feature activation+0.000
PHOTO
Token PHOTO
Feature activation+0.771
)
Token )
Feature activation+0.597
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
case
Token case
Feature activation+0.000
was
Token was
Feature activation+0.000
:
Token:
Feature activation+0.000
CIA
Token CIA
Feature activation+0.702
.
Token.
Feature activation+1.044
gov
Tokengov
Feature activation+0.863
The
Token The
Feature activation+0.982
A
Token A
Feature activation+0.664
-
Token-
Feature activation+0.176
12
Token12
Feature activation+0.070
reconnaissance
Token reconnaissance
Feature activation+0.000
aircraft
Token aircraft
Feature activation+0.000
was
Token was
Feature activation+0.000
KE
TokenKE
Feature activation+0.570
|
Token |
Feature activation+1.454
N
Token N
Feature activation+0.996
bl
Tokenbl
Feature activation+0.624
ake
Tokenake
Feature activation+0.267
@
Token@
Feature activation+0.803
md
Tokenmd
Feature activation+0.298
n
Tokenn
Feature activation+0.143
.
Token.
Feature activation+0.631
net
Tokennet
Feature activation+0.449
Photo
Token Photo
Feature activation+1.554
.
Token.
Feature activation+0.000
AP
Token AP
Feature activation+0.000
Photo
Token Photo
Feature activation+0.079
/
Token/
Feature activation+0.177
NASA
TokenNASA
Feature activation+0.307
/
Token/
Feature activation+0.649
J
TokenJ
Feature activation+0.304
PL
TokenPL
Feature activation+0.050
-
Token-
Feature activation+0.169
Cal
TokenCal
Feature activation+0.000
tech
Tokentech
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+0.000
courtesy
Token courtesy
Feature activation+0.061
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.258
Damascus
Token Damascus
Feature activation+0.729
Police
Token Police
Feature activation+1.033
Force
Token Force
Feature activation+0.402
.
Token.
Feature activation+0.332
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 0.292 - 0.583
CONTAINS 0.007%

:
Token:
Feature activation+0.000
Steve
Token Steve
Feature activation+0.000
Ru
Token Ru
Feature activation+0.000
ark
Tokenark
Feature activation+0.000
,
Token,
Feature activation+0.016
AP
Token AP
Feature activation+0.460
)
Token)
Feature activation+0.241
Story
Token Story
Feature activation+0.000
Highlights
Token Highlights
Feature activation+0.000
Death
Token Death
Feature activation+0.000
benefits
Token benefits
Feature activation+0.000
affidavit
Token affidavit
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+0.758
:
Token:
Feature activation+0.000
B
Token B
Feature activation+1.435
ex
Tokenex
Feature activation+0.338
ar
Tokenar
Feature activation+0.280
County
Token County
Feature activation+1.437
Sheriff
Token Sheriff
Feature activation+1.750
's
Token's
Feature activation+1.267
Office
Token Office
Feature activation+1.284
Ron
Token Ron
Feature activation+0.000
Schw
Token Schw
Feature activation+0.000
ane
Tokenane
Feature activation+0.000
,
Token,
Feature activation+0.303
Ron
Token Ron
Feature activation+0.185
Schw
Token Schw
Feature activation+0.422
ane
Tokenane
Feature activation+0.367
-
Token-
Feature activation+0.241
USA
TokenUSA
Feature activation+0.027
TODAY
Token TODAY
Feature activation+0.261
Sports
Token Sports
Feature activation+0.000
Robinson
Token Robinson
Feature activation+0.000
and
Token and
Feature activation+0.000
Photos
Token Photos
Feature activation+0.565
by
Token by
Feature activation+0.403
Cor
Token Cor
Feature activation+0.444
bis
Tokenbis
Feature activation+0.503
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
ez
Tokenez
Feature activation+0.000
er
Tokener
Feature activation+0.000
Photo
Token Photo
Feature activation+0.000
/
Token/
Feature activation+0.000
F
TokenF
Feature activation+0.000
res
Tokenres
Feature activation+0.000
no
Tokenno
Feature activation+0.000
Police
Token Police
Feature activation+0.310
Department
Token Department
Feature activation+0.000
)
Token)
Feature activation+0.387
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000

INTERVAL 0.000 - 0.292
CONTAINS 99.983%

.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
are
Token are
Feature activation+0.000
their
Token their
Feature activation+0.000
sub
Token sub
Feature activation+0.000
lim
Tokenlim
Feature activation+0.000
inally
Tokeninally
Feature activation+0.000
(
Token (
Feature activation+0.000
is
Tokenis
Feature activation+0.000
that
Token that
Feature activation+0.000
a
Token a
Feature activation+0.000
serve
Token serve
Feature activation+0.000
craft
Token craft
Feature activation+0.000
beer
Token beer
Feature activation+0.000
,
Token,
Feature activation+0.000
boutique
Token boutique
Feature activation+0.000
wine
Token wine
Feature activation+0.000
and
Token and
Feature activation+0.000
appet
Token appet
Feature activation+0.000
izers
Tokenizers
Feature activation+0.000
is
Token is
Feature activation+0.000
taking
Token taking
Feature activation+0.000
Be
TokenBe
Feature activation+0.000
prepared
Token prepared
Feature activation+0.000
for
Token for
Feature activation+0.000
all
Token all
Feature activation+0.000
types
Token types
Feature activation+0.000
of
Token of
Feature activation+0.000
weather
Token weather
Feature activation+0.000
!
Token!
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Slack
Token Slack
Feature activation+0.000
:
Token:
Feature activation+0.000
Street
Token Street
Feature activation+0.000
Fighting
Token Fighting
Feature activation+0.000
Ro
Token Ro
Feature activation+0.000
os
Tokenos
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Someone
TokenSomeone
Feature activation+0.000
Built
Token Built
Feature activation+0.000
a
Token a
Feature activation+0.000
knows
Token knows
Feature activation+0.000
Kara
Token Kara
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
secret
Token secret
Feature activation+0.000
,
Token,
Feature activation+0.000
gives
Token gives
Feature activation+0.000
her
Token her
Feature activation+0.000
some
Token some
Feature activation+0.000
great
Token great
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