H8.9 top features

Top feature 0 in H8.9: (feature 966)

TOP ACTIVATIONS
MAX = 7.341

Shot
Token Shot
Feature activation+0.000
Show
Token Show
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
Industry
Token Industry
Feature activation+6.960
Dinner
Token Dinner
Feature activation+0.000
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
the
Token the
Feature activation+0.000
letter
Token letter
Feature activation+0.000
attended
Token attended
Feature activation+0.000
the
Token the
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
Union
Token Union
Feature activation+6.485
address
Token address
Feature activation+0.000
she
Token she
Feature activation+0.000
got
Token got
Feature activation+0.000
uncomfortable
Token uncomfortable
Feature activation+0.000
as
Token as
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
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
Union
Token Union
Feature activation+5.683
address
Token address
Feature activation+0.000
.
Token.
Feature activation+0.000
Even
Token Even
Feature activation+0.000
folks
Token folks
Feature activation+0.000
like
Token like
Feature activation+0.000
-
Token-
Feature activation+0.000
of
Tokenof
Feature activation+0.000
-
Token-
Feature activation+0.000
the
Tokenthe
Feature activation+0.000
-
Token-
Feature activation+0.000
science
Tokenscience
Feature activation+2.985
synthesis
Token synthesis
Feature activation+0.000
and
Token and
Feature activation+0.000
recommendations
Token recommendations
Feature activation+0.000
prepared
Token prepared
Feature activation+0.000
for
Token for
Feature activation+0.000
's
Token's
Feature activation+0.000
"
Token "
Feature activation+0.000
State
TokenState
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Union
Token Union
Feature activation+2.878
"
Token"
Feature activation+0.000
program
Token program
Feature activation+0.000
that
Token that
Feature activation+0.000
48
Token 48
Feature activation+0.000
,
Token,
Feature activation+0.000
At
TokenAt
Feature activation+0.000
the
Token the
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
Union
Token Union
Feature activation+2.658
address
Token address
Feature activation+0.000
âĢ¦
Token âĢ¦
Feature activation+0.000
when
Token when
Feature activation+0.000
you
Token you
Feature activation+0.000
are
Token are
Feature activation+0.000
the
Token the
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
world
Token world
Feature activation+0.491
is
Token is
Feature activation+1.541
showing
Token showing
Feature activation+0.000
some
Token some
Feature activation+0.000
eerie
Token eerie
Feature activation+0.000
similarities
Token similarities
Feature activation+0.000
to
Token to
Feature activation+0.000
but
Token but
Feature activation+0.000
the
Token the
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
world
Token world
Feature activation+0.491
is
Token is
Feature activation+1.541
showing
Token showing
Feature activation+0.000
some
Token some
Feature activation+0.000
eerie
Token eerie
Feature activation+0.000
similarities
Token similarities
Feature activation+0.000
you
Token you
Feature activation+0.000
to
Token to
Feature activation+0.000
higher
Token higher
Feature activation+0.000
state
Token state
Feature activation+0.000
of
Token of
Feature activation+0.000
consciousness
Token consciousness
Feature activation+0.423
,
Token,
Feature activation+0.000
why
Token why
Feature activation+0.000
not
Token not
Feature activation+0.000
to
Token to
Feature activation+0.000
take
Token take
Feature activation+0.000
the
Token the
Feature activation+0.000
current
Token current
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
economy
Token economy
Feature activation+0.195
.
Token.
Feature activation+0.000
By
Token By
Feature activation+0.000
locating
Token locating
Feature activation+0.000
such
Token such
Feature activation+0.000
episodes
Token episodes
Feature activation+0.000
little
Token little
Feature activation+0.000
over
Token over
Feature activation+0.000
a
Token a
Feature activation+0.000
month
Token month
Feature activation+0.000
--
Token --
Feature activation+0.000
is
Token is
Feature activation+0.180
an
Token an
Feature activation+0.000
apt
Token apt
Feature activation+0.000
metaphor
Token metaphor
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
Terrorism
Token Terrorism
Feature activation+0.000
,
Token,
Feature activation+0.000
Weapons
Token Weapons
Feature activation+0.000
of
Token of
Feature activation+0.000
Mass
Token Mass
Feature activation+0.000
Destruction
Token Destruction
Feature activation+0.021
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
US
Token US
Feature activation+0.000
Strategy
Token Strategy
Feature activation+0.000
at
Token at
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
defence
Token defence
Feature activation+0.000
ministry
Token ministry
Feature activation+0.000
committee
Token committee
Feature activation+0.000
headed
Token headed
Feature activation+0.000
by
Token by
Feature activation+0.000
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
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

Top DFA by src position
MAX = 5.835

2017
Token 2017
Feature activation+0.200
Top resid features:
Shot
Token Shot
Feature activation-0.086
Top resid features:
Show
Token Show
Feature activation-0.086
Top resid features:
State
Token State
Feature activation+0.192
Top resid features:
of
Token of
Feature activation+4.236
Top resid features:
the
Token the
Feature activation+5.744
Top resid features:
Industry
Token Industry
Feature activation+0.126
Top resid features:
Dinner
Token Dinner
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
In
Token In
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
she
Token she
Feature activation+0.062
Top resid features:
attended
Token attended
Feature activation+0.228
Top resid features:
the
Token the
Feature activation+0.406
Top resid features:
State
Token State
Feature activation+0.089
Top resid features:
of
Token of
Feature activation+2.648
Top resid features:
the
Token the
Feature activation+5.805
Top resid features:
Union
Token Union
Feature activation+0.179
Top resid features:
address
Token address
Feature activation+0.000
Top resid features:
she
Token she
Feature activation+0.000
Top resid features:
got
Token got
Feature activation+0.000
Top resid features:
uncomfortable
Token uncomfortable
Feature activation+0.000
Top resid features:
later
Token later
Feature activation-0.082
Top resid features:
in
Token in
Feature activation+0.312
Top resid features:
his
Token his
Feature activation+1.404
Top resid features:
State
Token State
Feature activation+0.086
Top resid features:
of
Token of
Feature activation+1.153
Top resid features:
the
Token the
Feature activation+5.835
Top resid features:
Union
Token Union
Feature activation+0.145
Top resid features:
address
Token address
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
Even
Token Even
Feature activation+0.000
Top resid features:
folks
Token folks
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.381
Top resid features:
state
Token state
Feature activation+0.264
Top resid features:
-
Token-
Feature activation+0.294
Top resid features:
of
Tokenof
Feature activation+0.960
Top resid features:
-
Token-
Feature activation+1.331
Top resid features:
the
Tokenthe
Feature activation+1.644
Top resid features:
-
Token-
Feature activation+0.277
Top resid features:
science
Tokenscience
Feature activation+0.515
Top resid features:
synthesis
Token synthesis
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
recommendations
Token recommendations
Feature activation+0.000
Top resid features:
CNN
Token CNN
Feature activation-0.126
Top resid features:
's
Token's
Feature activation+0.814
Top resid features:
"
Token "
Feature activation+0.348
Top resid features:
State
TokenState
Feature activation+0.072
Top resid features:
of
Token of
Feature activation+1.611
Top resid features:
the
Token the
Feature activation+2.491
Top resid features:
Union
Token Union
Feature activation+0.295
Top resid features:
"
Token"
Feature activation+0.000
Top resid features:
program
Token program
Feature activation+0.000
Top resid features:
that
Token that
Feature activation+0.000
Top resid features:
48
Token 48
Feature activation+0.000
Top resid features:
she
Token she
Feature activation-0.004
Top resid features:
attended
Token attended
Feature activation-0.039
Top resid features:
the
Token the
Feature activation-0.016
Top resid features:
State
Token State
Feature activation+0.021
Top resid features:
of
Token of
Feature activation+1.294
Top resid features:
the
Token the
Feature activation+1.368
Top resid features:
Union
Token Union
Feature activation+0.189
Top resid features:
address
Token address
Feature activation+0.570
Top resid features:
she
Token she
Feature activation+0.011
Top resid features:
got
Token got
Feature activation-0.008
Top resid features:
uncomfortable
Token uncomfortable
Feature activation+0.002
Top resid features:
gets
Token gets
Feature activation-0.002
Top resid features:
,
Token,
Feature activation+0.051
Top resid features:
but
Token but
Feature activation+0.264
Top resid features:
the
Token the
Feature activation+0.078
Top resid features:
state
Token state
Feature activation+0.485
Top resid features:
of
Token of
Feature activation+1.422
Top resid features:
the
Token the
Feature activation+1.209
Top resid features:
world
Token world
Feature activation+1.211
Top resid features:
is
Token is
Feature activation+0.388
Top resid features:
showing
Token showing
Feature activation+0.000
Top resid features:
some
Token some
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.044
Top resid features:
but
Token but
Feature activation+0.220
Top resid features:
the
Token the
Feature activation+0.165
Top resid features:
state
Token state
Feature activation+0.334
Top resid features:
of
Token of
Feature activation+0.918
Top resid features:
the
Token the
Feature activation+0.983
Top resid features:
world
Token world
Feature activation+0.937
Top resid features:
is
Token is
Feature activation+0.000
Top resid features:
showing
Token showing
Feature activation+0.000
Top resid features:
some
Token some
Feature activation+0.000
Top resid features:
eerie
Token eerie
Feature activation+0.000
Top resid features:
brings
Token brings
Feature activation+0.061
Top resid features:
you
Token you
Feature activation+0.007
Top resid features:
to
Token to
Feature activation+0.107
Top resid features:
higher
Token higher
Feature activation+0.106
Top resid features:
state
Token state
Feature activation+0.278
Top resid features:
of
Token of
Feature activation+2.522
Top resid features:
consciousness
Token consciousness
Feature activation+0.165
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
why
Token why
Feature activation+0.000
Top resid features:
not
Token not
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
call
Token call
Feature activation-0.002
Top resid features:
"
Token "
Feature activation+0.005
Top resid features:
natural
Tokennatural
Feature activation+0.008
Top resid features:
experiments
Token experiments
Feature activation+0.019
Top resid features:
"
Token"
Feature activation+0.020
Top resid features:
--
Token--
Feature activation+1.139
Top resid features:
in
Tokenin
Feature activation+0.058
Top resid features:
this
Token this
Feature activation+0.000
Top resid features:
context
Token context
Feature activation+0.027
Top resid features:
,
Token,
Feature activation+0.030
Top resid features:
episodes
Token episodes
Feature activation+0.037
Top resid features:
.
Token.
Feature activation-0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.023
Top resid features:
Ċ
TokenĊ
Feature activation+0.081
Top resid features:
The
TokenThe
Feature activation+0.058
Top resid features:
state
Token state
Feature activation+0.157
Top resid features:
of
Token of
Feature activation+2.894
Top resid features:
Fior
Token Fior
Feature activation-0.022
Top resid features:
an
Tokenan
Feature activation+0.000
Top resid features:
elli
Tokenelli
Feature activation+0.088
Top resid features:
's
Token's
Feature activation+0.013
Top resid features:
office
Token office
Feature activation+0.002
Top resid features:
Conference
Token Conference
Feature activation+0.083
Top resid features:
on
Token on
Feature activation+0.503
Top resid features:
Terrorism
Token Terrorism
Feature activation+0.131
Top resid features:
,
Token,
Feature activation+0.310
Top resid features:
Weapons
Token Weapons
Feature activation+0.074
Top resid features:
of
Token of
Feature activation+0.906
Top resid features:
Mass
Token Mass
Feature activation+0.355
Top resid features:
Destruction
Token Destruction
Feature activation+0.085
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
US
Token US
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+1.125
Top resid features:
rik
Tokenrik
Feature activation-0.010
Top resid features:
ar
Tokenar
Feature activation+0.006
Top resid features:
said
Token said
Feature activation-0.046
Top resid features:
Saturday
Token Saturday
Feature activation-0.055
Top resid features:
.
Token.
Feature activation+0.019
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+1.121
Top resid features:
rik
Tokenrik
Feature activation+0.011
Top resid features:
ar
Tokenar
Feature activation+0.004
Top resid features:
said
Token said
Feature activation-0.056
Top resid features:
Saturday
Token Saturday
Feature activation-0.055
Top resid features:
.
Token.
Feature activation+0.033
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+1.231
Top resid features:
rik
Tokenrik
Feature activation-0.013
Top resid features:
ar
Tokenar
Feature activation+0.001
Top resid features:
said
Token said
Feature activation-0.110
Top resid features:
Saturday
Token Saturday
Feature activation-0.040
Top resid features:
.
Token.
Feature activation+0.019
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+1.072
Top resid features:
rik
Tokenrik
Feature activation-0.001
Top resid features:
ar
Tokenar
Feature activation+0.004
Top resid features:
said
Token said
Feature activation-0.140
Top resid features:
Saturday
Token Saturday
Feature activation-0.095
Top resid features:
.
Token.
Feature activation+0.042
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+1.276
Top resid features:
rik
Tokenrik
Feature activation-0.012
Top resid features:
ar
Tokenar
Feature activation+0.001
Top resid features:
said
Token said
Feature activation-0.135
Top resid features:
Saturday
Token Saturday
Feature activation-0.108
Top resid features:
.
Token.
Feature activation+0.005
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+1.185
Top resid features:
rik
Tokenrik
Feature activation-0.050
Top resid features:
ar
Tokenar
Feature activation+0.015
Top resid features:
said
Token said
Feature activation-0.442
Top resid features:
Saturday
Token Saturday
Feature activation-0.205
Top resid features:
.
Token.
Feature activation+0.020
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+1.035
Top resid features:
rik
Tokenrik
Feature activation-0.395
Top resid features:
ar
Tokenar
Feature activation-0.062
Top resid features:
said
Token said
Feature activation-0.509
Top resid features:
Saturday
Token Saturday
Feature activation-0.090
Top resid features:
.
Token.
Feature activation+0.071
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+1.025
Top resid features:
rik
Tokenrik
Feature activation-0.135
Top resid features:
ar
Tokenar
Feature activation-0.020
Top resid features:
said
Token said
Feature activation-0.412
Top resid features:
Saturday
Token Saturday
Feature activation-0.051
Top resid features:
.
Token.
Feature activation+0.056
Top resid features:

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.02

Head 2: 0.05

Head 3: 0.04

Head 4: 0.06

Head 5: 0.06

Head 6: 0.03

Head 7: 0.08

Head 8: 0.05

Head 9: 0.47

Head 10: 0.07

Head 11: 0.04

Positive logits

Ctrl3.38

summ3.25

Address3.25

concise3.14

Dialogue3.12

PRESIDENT3.08

briefing2.99

speeches2.93

Dialog2.90

speech2.89

Message2.88

address2.87

reconc2.84

keynote2.83

summary2.82

Address2.82

countdown2.80

rebuke2.80

sermon2.80

address2.78

Negative logits

cultures-3.12

speculated-3.12

speculate-3.08

grown-3.04

domest-3.04

evolved-2.96

experimenting-2.94

hybrids-2.93

experimented-2.88

speculation-2.84

ogenesis-2.82

fungi-2.82

genetically-2.81

prototypes-2.72

Invention-2.72

developed-2.67

possibilities-2.65

breeds-2.64

biologically-2.60

intrigued-2.60

INTERVAL 6.607 - 7.341
CONTAINS 0.000%

Shot
Token Shot
Feature activation+0.000
Show
Token Show
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
Industry
Token Industry
Feature activation+6.960
Dinner
Token Dinner
Feature activation+0.000
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
the
Token the
Feature activation+0.000
letter
Token letter
Feature activation+0.000

INTERVAL 5.873 - 6.607
CONTAINS 0.000%

attended
Token attended
Feature activation+0.000
the
Token the
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
Union
Token Union
Feature activation+6.485
address
Token address
Feature activation+0.000
she
Token she
Feature activation+0.000
got
Token got
Feature activation+0.000
uncomfortable
Token uncomfortable
Feature activation+0.000
as
Token as
Feature activation+0.000

INTERVAL 5.139 - 5.873
CONTAINS 0.000%

in
Token in
Feature activation+0.000
his
Token his
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
Union
Token Union
Feature activation+5.683
address
Token address
Feature activation+0.000
.
Token.
Feature activation+0.000
Even
Token Even
Feature activation+0.000
folks
Token folks
Feature activation+0.000
like
Token like
Feature activation+0.000

INTERVAL 4.405 - 5.139
CONTAINS 0.000%

INTERVAL 3.671 - 4.405
CONTAINS 0.000%

INTERVAL 2.936 - 3.671
CONTAINS 0.000%

-
Token-
Feature activation+0.000
of
Tokenof
Feature activation+0.000
-
Token-
Feature activation+0.000
the
Tokenthe
Feature activation+0.000
-
Token-
Feature activation+0.000
science
Tokenscience
Feature activation+2.985
synthesis
Token synthesis
Feature activation+0.000
and
Token and
Feature activation+0.000
recommendations
Token recommendations
Feature activation+0.000
prepared
Token prepared
Feature activation+0.000
for
Token for
Feature activation+0.000

INTERVAL 2.202 - 2.936
CONTAINS 0.000%

's
Token's
Feature activation+0.000
"
Token "
Feature activation+0.000
State
TokenState
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Union
Token Union
Feature activation+2.878
"
Token"
Feature activation+0.000
program
Token program
Feature activation+0.000
that
Token that
Feature activation+0.000
48
Token 48
Feature activation+0.000
,
Token,
Feature activation+0.000
At
TokenAt
Feature activation+0.000
the
Token the
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
Union
Token Union
Feature activation+2.658
address
Token address
Feature activation+0.000
âĢ¦
Token âĢ¦
Feature activation+0.000
when
Token when
Feature activation+0.000
you
Token you
Feature activation+0.000
are
Token are
Feature activation+0.000

INTERVAL 1.468 - 2.202
CONTAINS 0.000%

the
Token the
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
world
Token world
Feature activation+0.491
is
Token is
Feature activation+1.541
showing
Token showing
Feature activation+0.000
some
Token some
Feature activation+0.000
eerie
Token eerie
Feature activation+0.000
similarities
Token similarities
Feature activation+0.000
to
Token to
Feature activation+0.000

INTERVAL 0.734 - 1.468
CONTAINS 0.000%

INTERVAL 0.000 - 0.734
CONTAINS 99.999%

four
Token four
Feature activation+0.000
months
Token months
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
police
Token police
Feature activation+0.000
department
Token department
Feature activation+0.000
tweeted
Token tweeted
Feature activation+0.000
they
Token they
Feature activation+0.000
placed
Token placed
Feature activation+0.000
mind
Token mind
Feature activation+0.000
.
Token.
Feature activation+0.000
Despite
Token Despite
Feature activation+0.000
the
Token the
Feature activation+0.000
lack
Token lack
Feature activation+0.000
of
Token of
Feature activation+0.000
any
Token any
Feature activation+0.000
agreed
Token agreed
Feature activation+0.000
upon
Token upon
Feature activation+0.000
theory
Token theory
Feature activation+0.000
of
Token of
Feature activation+0.000
,
Token,
Feature activation+0.000
seven
Token seven
Feature activation+0.000
were
Token were
Feature activation+0.000
problematic
Token problematic
Feature activation+0.000
having
Token having
Feature activation+0.000
at
Token at
Feature activation+0.000
least
Token least
Feature activation+0.000
a
Token a
Feature activation+0.000
5
Token 5
Feature activation+0.000
percent
Token percent
Feature activation+0.000
variance
Token variance
Feature activation+0.000
with
Token with
Feature activation+0.000
North
Token North
Feature activation+0.000
Korea
Token Korea
Feature activation+0.000
"
Token "
Feature activation+0.000
is
Tokenis
Feature activation+0.000
over
Token over
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
To
TokenTo
Feature activation+0.000
those
Token those
Feature activation+0.000
alone
Token alone
Feature activation+0.000
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
meta
Token meta
Feature activation+0.000
-
Token-
Feature activation+0.000
example
Tokenexample
Feature activation+0.000
,
Token,
Feature activation+0.000
since
Token since
Feature activation+0.000
its
Token its
Feature activation+0.000
entire
Token entire
Feature activation+0.000
purpose
Token purpose
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 1 in H8.9: (feature 23568)

TOP ACTIVATIONS
MAX = 4.349

minority
Token minority
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+3.545
they
Token they
Feature activation+4.349
ask
Token ask
Feature activation+2.738
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
color
Token color
Feature activation+0.000
political
Token political
Feature activation+0.000
vision
Token vision
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+3.875
exactly
Token exactly
Feature activation+0.026
?
Token?
Feature activation+0.896
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.333
V
TokenV
Feature activation+0.000
income
Tokenincome
Feature activation+0.000
minority
Token minority
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+3.545
they
Token they
Feature activation+4.349
ask
Token ask
Feature activation+2.738
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
And
TokenAnd
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+3.123
pray
Token pray
Feature activation+0.538
tell
Token tell
Feature activation+0.396
,
Token,
Feature activation+0.000
did
Token did
Feature activation+0.000
they
Token they
Feature activation+0.000
from
Token from
Feature activation+0.000
ruining
Token ruining
Feature activation+0.000
another
Token another
Feature activation+0.000
Cowboys
Token Cowboys
Feature activation+0.000
season
Token season
Feature activation+0.000
remains
Token remains
Feature activation+2.872
a
Token a
Feature activation+2.481
mystery
Token mystery
Feature activation+0.964
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+3.545
they
Token they
Feature activation+4.349
ask
Token ask
Feature activation+2.738
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
color
Token color
Feature activation+0.000
or
Token or
Feature activation+0.000
ruining
Token ruining
Feature activation+0.000
another
Token another
Feature activation+0.000
Cowboys
Token Cowboys
Feature activation+0.000
season
Token season
Feature activation+0.000
remains
Token remains
Feature activation+2.872
a
Token a
Feature activation+2.481
mystery
Token mystery
Feature activation+0.964
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Big
TokenBig
Feature activation+0.000
Spect
Token Spect
Feature activation+0.000
er
Tokener
Feature activation+0.000
of
Token of
Feature activation+0.000
Capital
Token Capital
Feature activation+0.000
,
Token,
Feature activation+0.570
answers
Token answers
Feature activation+2.139
this
Token this
Feature activation+1.379
question
Token question
Feature activation+0.074
here
Token here
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
best
Token best
Feature activation+0.000
to
Token to
Feature activation+0.000
find
Token find
Feature activation+0.694
out
Token out
Feature activation+2.074
-
Token -
Feature activation+0.000
Lucky
Token Lucky
Feature activation+0.000
Peach
Token Peach
Feature activation+0.000
,
Token,
Feature activation+0.000
Put
Token Put
Feature activation+0.000
means
Token means
Feature activation+0.000
?
Token?
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+1.046
"
Token"
Feature activation+0.229
he
Token he
Feature activation+2.008
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Because
TokenBecause
Feature activation+0.000
les
Tokenles
Feature activation+0.353
me
Token me
Feature activation+0.061
,
Token,
Feature activation+1.299
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
he
Token he
Feature activation+1.863
says
Token says
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
As
TokenAs
Feature activation+0.000
What
Token What
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
beautiful
Token beautiful
Feature activation+0.000
,
Token,
Feature activation+1.414
he
Token he
Feature activation+1.744
said
Token said
Feature activation+1.089
,
Token,
Feature activation+0.802
is
Token is
Feature activation+0.000
if
Token if
Feature activation+0.000
someone
Token someone
Feature activation+0.000
been
Token been
Feature activation+0.000
between
Token between
Feature activation+0.000
March
Token March
Feature activation+0.000
and
Token and
Feature activation+0.000
May
Token May
Feature activation+0.000
is
Token is
Feature activation+1.639
unknown
Token unknown
Feature activation+0.388
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
it
Token it
Feature activation+0.000
is
Token is
Feature activation+0.000
plant
Token plant
Feature activation+0.000
producing
Token producing
Feature activation+0.000
seemingly
Token seemingly
Feature activation+0.000
useless
Token useless
Feature activation+0.000
traps
Token traps
Feature activation+0.000
?
Token?
Feature activation+1.616
Then
Token Then
Feature activation+0.172
one
Token one
Feature activation+0.428
day
Token day
Feature activation+0.000
they
Token they
Feature activation+0.000
found
Token found
Feature activation+0.000
seen
Token seen
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
rugby
Token rugby
Feature activation+0.000
pitch
Token pitch
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+1.567
goes
Token goes
Feature activation+1.041
far
Token far
Feature activation+0.076
deeper
Token deeper
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
m
Tokenm
Feature activation+0.000
not
Token not
Feature activation+0.000
telling
Token telling
Feature activation+0.505
you
Token you
Feature activation+1.561
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
she
Token she
Feature activation+0.175
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
One
TokenOne
Feature activation+0.000
possible
Token possible
Feature activation+1.456
reason
Token reason
Feature activation+1.502
is
Token is
Feature activation+0.000
that
Token that
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
oppose
Token oppose
Feature activation+0.000
taxpayer
Token taxpayer
Feature activation+0.000
why
Token why
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
One
TokenOne
Feature activation+0.000
possible
Token possible
Feature activation+1.456
reason
Token reason
Feature activation+1.502
is
Token is
Feature activation+0.000
that
Token that
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
oppose
Token oppose
Feature activation+0.000
.
Token.
Feature activation+0.000
What
Token What
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
beautiful
Token beautiful
Feature activation+0.000
,
Token,
Feature activation+1.414
he
Token he
Feature activation+1.744
said
Token said
Feature activation+1.089
,
Token,
Feature activation+0.802
is
Token is
Feature activation+0.000
if
Token if
Feature activation+0.000
play
Token play
Feature activation+0.000
Pokémon
Token Pokémon
Feature activation+0.000
Go
Token Go
Feature activation+0.000
safely
Token safely
Feature activation+0.000
is
Token is
Feature activation+0.655
another
Token another
Feature activation+1.403
post
Token post
Feature activation+0.892
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Here
TokenHere
Feature activation+0.000

Top DFA by src position
MAX = 3.167

-
Token-
Feature activation+0.003
Top resid features:
income
Tokenincome
Feature activation+0.001
Top resid features:
minority
Token minority
Feature activation-0.004
Top resid features:
youth
Token youth
Feature activation-0.015
Top resid features:
.
Token.
Feature activation-0.092
Top resid features:
Why
Token Why
Feature activation+2.661
Top resid features:
,
Token,
Feature activation+2.400
Top resid features:
they
Token they
Feature activation+0.054
Top resid features:
ask
Token ask
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
should
Token should
Feature activation+0.000
Top resid features:
their
Token their
Feature activation-0.002
Top resid features:
political
Token political
Feature activation+0.005
Top resid features:
vision
Token vision
Feature activation-0.002
Top resid features:
.
Token.
Feature activation-0.042
Top resid features:
But
Token But
Feature activation+0.053
Top resid features:
why
Token why
Feature activation+2.894
Top resid features:
,
Token,
Feature activation+1.505
Top resid features:
exactly
Token exactly
Feature activation+0.000
Top resid features:
?
Token?
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
income
Tokenincome
Feature activation-0.010
Top resid features:
minority
Token minority
Feature activation-0.014
Top resid features:
youth
Token youth
Feature activation-0.031
Top resid features:
.
Token.
Feature activation-0.042
Top resid features:
Why
Token Why
Feature activation+3.167
Top resid features:
,
Token,
Feature activation+1.038
Top resid features:
they
Token they
Feature activation+0.000
Top resid features:
ask
Token ask
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
should
Token should
Feature activation+0.000
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
Ŀ
TokenĿ
Feature activation-0.005
Top resid features:
Ċ
TokenĊ
Feature activation-0.012
Top resid features:
Ċ
TokenĊ
Feature activation-0.012
Top resid features:
And
TokenAnd
Feature activation+0.059
Top resid features:
why
Token why
Feature activation+3.153
Top resid features:
,
Token,
Feature activation+0.744
Top resid features:
pray
Token pray
Feature activation+0.000
Top resid features:
tell
Token tell
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
did
Token did
Feature activation+0.000
Top resid features:
RB
Token RB
Feature activation-0.057
Top resid features:
from
Token from
Feature activation-0.001
Top resid features:
Ohio
Token Ohio
Feature activation-0.001
Top resid features:
State
Token State
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.014
Top resid features:
Why
Token Why
Feature activation+2.265
Top resid features:
Jones
Token Jones
Feature activation+0.019
Top resid features:
didn
Token didn
Feature activation+1.282
Top resid features:
âĢ
TokenâĢ
Feature activation-0.000
Top resid features:
Ļ
TokenĻ
Feature activation-0.004
Top resid features:
t
Tokent
Feature activation+0.037
Top resid features:
income
Tokenincome
Feature activation-0.015
Top resid features:
minority
Token minority
Feature activation-0.008
Top resid features:
youth
Token youth
Feature activation-0.033
Top resid features:
.
Token.
Feature activation-0.080
Top resid features:
Why
Token Why
Feature activation+0.632
Top resid features:
,
Token,
Feature activation+2.770
Top resid features:
they
Token they
Feature activation+0.069
Top resid features:
ask
Token ask
Feature activation+0.011
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
should
Token should
Feature activation+0.000
Top resid features:
skin
Token skin
Feature activation+0.000
Top resid features:
RB
Token RB
Feature activation-0.060
Top resid features:
from
Token from
Feature activation-0.000
Top resid features:
Ohio
Token Ohio
Feature activation-0.000
Top resid features:
State
Token State
Feature activation-0.000
Top resid features:
.
Token.
Feature activation-0.008
Top resid features:
Why
Token Why
Feature activation+2.271
Top resid features:
Jones
Token Jones
Feature activation+0.025
Top resid features:
didn
Token didn
Feature activation+0.857
Top resid features:
âĢ
TokenâĢ
Feature activation-0.003
Top resid features:
Ļ
TokenĻ
Feature activation-0.006
Top resid features:
t
Tokent
Feature activation+0.035
Top resid features:
.
Token.
Feature activation-0.011
Top resid features:
But
Token But
Feature activation+0.026
Top resid features:
why
Token why
Feature activation+0.831
Top resid features:
,
Token,
Feature activation+0.655
Top resid features:
exactly
Token exactly
Feature activation+0.153
Top resid features:
?
Token?
Feature activation+1.222
Top resid features:
Ċ
TokenĊ
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
V
TokenV
Feature activation-0.001
Top resid features:
ive
Tokenive
Feature activation-0.002
Top resid features:
k
Tokenk
Feature activation-0.001
Top resid features:
Ŀ
TokenĿ
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.000
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.018
Top resid features:
How
TokenHow
Feature activation+0.796
Top resid features:
do
Token do
Feature activation+1.810
Top resid features:
the
Token the
Feature activation+0.020
Top resid features:
contemporary
Token contemporary
Feature activation-0.007
Top resid features:
food
Token food
Feature activation-0.005
Top resid features:
magazines
Token magazines
Feature activation-0.020
Top resid features:
use
Token use
Feature activation+0.071
Top resid features:
criticized
Token criticized
Feature activation-0.006
Top resid features:
me
Token me
Feature activation+0.045
Top resid features:
,
Token,
Feature activation+0.036
Top resid features:
âĢ
Token âĢ
Feature activation+0.001
Top resid features:
ĺ
Tokenĺ
Feature activation+0.024
Top resid features:
Why
TokenWhy
Feature activation+1.771
Top resid features:
can
Token can
Feature activation+0.610
Top resid features:
âĢ
TokenâĢ
Feature activation+0.004
Top resid features:
Ļ
TokenĻ
Feature activation+0.008
Top resid features:
t
Tokent
Feature activation+0.144
Top resid features:
we
Token we
Feature activation+0.044
Top resid features:
just
Token just
Feature activation-0.003
Top resid features:
about
Token about
Feature activation+0.001
Top resid features:
any
Token any
Feature activation-0.001
Top resid features:
reason
Token reason
Feature activation-0.008
Top resid features:
.
Token.
Feature activation-0.074
Top resid features:
Why
Token Why
Feature activation+1.574
Top resid features:
they
Token they
Feature activation+0.105
Top resid features:
don
Token don
Feature activation+0.028
Top resid features:
âĢ
TokenâĢ
Feature activation+0.010
Top resid features:
Ļ
TokenĻ
Feature activation+0.003
Top resid features:
t
Tokent
Feature activation+0.015
Top resid features:
.
Token.
Feature activation-0.026
Top resid features:
What
Token What
Feature activation+0.368
Top resid features:
would
Token would
Feature activation+0.668
Top resid features:
be
Token be
Feature activation+0.123
Top resid features:
beautiful
Token beautiful
Feature activation+0.047
Top resid features:
,
Token,
Feature activation+1.628
Top resid features:
he
Token he
Feature activation+0.013
Top resid features:
said
Token said
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
is
Token is
Feature activation+0.000
Top resid features:
if
Token if
Feature activation+0.000
Top resid features:
White
Token White
Feature activation+0.002
Top resid features:
horse
Tokenhorse
Feature activation+0.007
Top resid features:
Motors
Token Motors
Feature activation+0.003
Top resid features:
.
Token.
Feature activation-0.046
Top resid features:
Just
Token Just
Feature activation-0.037
Top resid features:
where
Token where
Feature activation+2.010
Top resid features:
this
Token this
Feature activation+0.020
Top resid features:
Mustang
Token Mustang
Feature activation-0.010
Top resid features:
had
Token had
Feature activation+0.267
Top resid features:
been
Token been
Feature activation+0.099
Top resid features:
between
Token between
Feature activation-0.007
Top resid features:
confused
Token confused
Feature activation-0.035
Top resid features:
researchers
Token researchers
Feature activation-0.004
Top resid features:
for
Token for
Feature activation-0.007
Top resid features:
months
Token months
Feature activation+0.008
Top resid features:
.
Token.
Feature activation-0.044
Top resid features:
Why
Token Why
Feature activation+1.481
Top resid features:
was
Token was
Feature activation+0.819
Top resid features:
the
Token the
Feature activation+0.135
Top resid features:
plant
Token plant
Feature activation+0.017
Top resid features:
producing
Token producing
Feature activation+0.063
Top resid features:
seemingly
Token seemingly
Feature activation+0.016
Top resid features:
Ċ
TokenĊ
Feature activation-0.002
Top resid features:
Ċ
TokenĊ
Feature activation-0.005
Top resid features:
But
TokenBut
Feature activation-0.011
Top resid features:
how
Token how
Feature activation+0.768
Top resid features:
we
Token we
Feature activation+0.130
Top resid features:
got
Token got
Feature activation+1.113
Top resid features:
here
Token here
Feature activation+0.142
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.014
Top resid features:
with
Token with
Feature activation+0.025
Top resid features:
the
Token the
Feature activation+0.002
Top resid features:
United
Token United
Feature activation+0.001
Top resid features:
had
Token had
Feature activation+0.002
Top resid features:
.
Token.
Feature activation-0.004
Top resid features:
âĢ
Token âĢ
Feature activation-0.008
Top resid features:
ľ
Tokenľ
Feature activation-0.002
Top resid features:
Where
TokenWhere
Feature activation+0.426
Top resid features:
is
Token is
Feature activation+1.275
Top resid features:
it
Token it
Feature activation+0.234
Top resid features:
?
Token?
Feature activation+0.028
Top resid features:
âĢ
TokenâĢ
Feature activation+0.002
Top resid features:
Ŀ
TokenĿ
Feature activation-0.011
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
I
TokenI
Feature activation-0.000
Top resid features:
).
Token).
Feature activation-0.034
Top resid features:
The
Token The
Feature activation+0.004
Top resid features:
puzzle
Token puzzle
Feature activation-0.001
Top resid features:
is
Token is
Feature activation-0.019
Top resid features:
why
Token why
Feature activation+2.171
Top resid features:
.
Token.
Feature activation-0.043
Top resid features:
Ċ
TokenĊ
Feature activation-0.007
Top resid features:
Ċ
TokenĊ
Feature activation-0.029
Top resid features:
One
TokenOne
Feature activation+0.006
Top resid features:
possible
Token possible
Feature activation+0.012
Top resid features:
I
TokenI
Feature activation-0.000
Top resid features:
).
Token).
Feature activation-0.040
Top resid features:
The
Token The
Feature activation+0.006
Top resid features:
puzzle
Token puzzle
Feature activation-0.021
Top resid features:
is
Token is
Feature activation-0.058
Top resid features:
why
Token why
Feature activation+2.521
Top resid features:
.
Token.
Feature activation-0.063
Top resid features:
Ċ
TokenĊ
Feature activation-0.009
Top resid features:
Ċ
TokenĊ
Feature activation-0.039
Top resid features:
One
TokenOne
Feature activation+0.015
Top resid features:
possible
Token possible
Feature activation-0.085
Top resid features:
they
Token they
Feature activation+0.005
Top resid features:
had
Token had
Feature activation+0.003
Top resid features:
to
Token to
Feature activation-0.001
Top resid features:
clear
Token clear
Feature activation-0.002
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
What
Token What
Feature activation+0.996
Top resid features:
would
Token would
Feature activation+0.268
Top resid features:
be
Token be
Feature activation+0.238
Top resid features:
beautiful
Token beautiful
Feature activation+0.163
Top resid features:
,
Token,
Feature activation+0.674
Top resid features:
he
Token he
Feature activation+0.000
Top resid features:
one
Token one
Feature activation+0.004
Top resid features:
ever
Token ever
Feature activation+0.001
Top resid features:
was
Token was
Feature activation+0.001
Top resid features:
.
Token.
Feature activation-0.011
Top resid features:
But
Token But
Feature activation-0.031
Top resid features:
how
Token how
Feature activation+1.279
Top resid features:
to
Token to
Feature activation+0.761
Top resid features:
play
Token play
Feature activation+0.099
Top resid features:
Pokémon
Token Pokémon
Feature activation-0.007
Top resid features:
Go
Token Go
Feature activation-0.005
Top resid features:
safely
Token safely
Feature activation+0.053
Top resid features:

Decoder Weights Distribution

Head 0: 0.05

Head 1: 0.03

Head 2: 0.06

Head 3: 0.05

Head 4: 0.04

Head 5: 0.06

Head 6: 0.04

Head 7: 0.04

Head 8: 0.07

Head 9: 0.45

Head 10: 0.07

Head 11: 0.04

Positive logits

answers5.22

Answer5.13

swers5.07

Answer4.96

questions4.94

question4.62

answer4.55

answer4.51

Question4.45

Answers4.42

question4.34

enqu4.20

answering4.17

inquired4.10

inquiry4.08

asks4.06

questioning4.05

mystery4.02

answ4.02

Questions4.00

Negative logits

victories-2.97

positives-2.93

endorsements-2.82

parity-2.73

policies-2.73

moderation-2.72

endors-2.71

Stead-2.63

balanced-2.62

shipments-2.59

boosted-2.57

strengths-2.57

excess-2.54

incre-2.54

momentum-2.51

successes-2.49

dominance-2.47

boosts-2.45

effectiveness-2.45

Berman-2.43

INTERVAL 3.914 - 4.349
CONTAINS 0.000%

minority
Token minority
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+3.545
they
Token they
Feature activation+4.349
ask
Token ask
Feature activation+2.738
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
color
Token color
Feature activation+0.000

INTERVAL 3.479 - 3.914
CONTAINS 0.000%

political
Token political
Feature activation+0.000
vision
Token vision
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+3.875
exactly
Token exactly
Feature activation+0.026
?
Token?
Feature activation+0.896
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.333
V
TokenV
Feature activation+0.000
income
Tokenincome
Feature activation+0.000
minority
Token minority
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+3.545
they
Token they
Feature activation+4.349
ask
Token ask
Feature activation+2.738
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000

INTERVAL 3.044 - 3.479
CONTAINS 0.000%

Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
And
TokenAnd
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+3.123
pray
Token pray
Feature activation+0.538
tell
Token tell
Feature activation+0.396
,
Token,
Feature activation+0.000
did
Token did
Feature activation+0.000
they
Token they
Feature activation+0.000

INTERVAL 2.609 - 3.044
CONTAINS 0.000%

from
Token from
Feature activation+0.000
ruining
Token ruining
Feature activation+0.000
another
Token another
Feature activation+0.000
Cowboys
Token Cowboys
Feature activation+0.000
season
Token season
Feature activation+0.000
remains
Token remains
Feature activation+2.872
a
Token a
Feature activation+2.481
mystery
Token mystery
Feature activation+0.964
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+3.545
they
Token they
Feature activation+4.349
ask
Token ask
Feature activation+2.738
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
color
Token color
Feature activation+0.000
or
Token or
Feature activation+0.000

INTERVAL 2.174 - 2.609
CONTAINS 0.000%

ruining
Token ruining
Feature activation+0.000
another
Token another
Feature activation+0.000
Cowboys
Token Cowboys
Feature activation+0.000
season
Token season
Feature activation+0.000
remains
Token remains
Feature activation+2.872
a
Token a
Feature activation+2.481
mystery
Token mystery
Feature activation+0.964
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Big
TokenBig
Feature activation+0.000

INTERVAL 1.739 - 2.174
CONTAINS 0.001%

Spect
Token Spect
Feature activation+0.000
er
Tokener
Feature activation+0.000
of
Token of
Feature activation+0.000
Capital
Token Capital
Feature activation+0.000
,
Token,
Feature activation+0.570
answers
Token answers
Feature activation+2.139
this
Token this
Feature activation+1.379
question
Token question
Feature activation+0.074
here
Token here
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
What
Token What
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
beautiful
Token beautiful
Feature activation+0.000
,
Token,
Feature activation+1.414
he
Token he
Feature activation+1.744
said
Token said
Feature activation+1.089
,
Token,
Feature activation+0.802
is
Token is
Feature activation+0.000
if
Token if
Feature activation+0.000
someone
Token someone
Feature activation+0.000
means
Token means
Feature activation+0.000
?
Token?
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+1.046
"
Token"
Feature activation+0.229
he
Token he
Feature activation+2.008
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Because
TokenBecause
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
best
Token best
Feature activation+0.000
to
Token to
Feature activation+0.000
find
Token find
Feature activation+0.694
out
Token out
Feature activation+2.074
-
Token -
Feature activation+0.000
Lucky
Token Lucky
Feature activation+0.000
Peach
Token Peach
Feature activation+0.000
,
Token,
Feature activation+0.000
Put
Token Put
Feature activation+0.000
les
Tokenles
Feature activation+0.353
me
Token me
Feature activation+0.061
,
Token,
Feature activation+1.299
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
he
Token he
Feature activation+1.863
says
Token says
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
As
TokenAs
Feature activation+0.000

INTERVAL 1.305 - 1.739
CONTAINS 0.001%

I
Token I
Feature activation+0.101
would
Token would
Feature activation+0.000
really
Token really
Feature activation+0.000
like
Token like
Feature activation+0.122
to
Token to
Feature activation+0.295
know
Token know
Feature activation+1.373
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĨ
TokenâĨ
Feature activation+0.000
IJ
TokenIJ
Feature activation+0.000
about
Token about
Feature activation+0.000
Tiger
Token Tiger
Feature activation+0.000
Woods
Token Woods
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+1.326
exactly
Token exactly
Feature activation+0.000
?
Token?
Feature activation+0.360
Because
Token Because
Feature activation+0.000
when
Token when
Feature activation+0.000
it
Token it
Feature activation+0.000
.
Token.
Feature activation+0.000
What
Token What
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
beautiful
Token beautiful
Feature activation+0.000
,
Token,
Feature activation+1.414
he
Token he
Feature activation+1.744
said
Token said
Feature activation+1.089
,
Token,
Feature activation+0.802
is
Token is
Feature activation+0.000
if
Token if
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
m
Tokenm
Feature activation+0.000
not
Token not
Feature activation+0.000
telling
Token telling
Feature activation+0.505
you
Token you
Feature activation+1.561
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
she
Token she
Feature activation+0.175
said
Token said
Feature activation+0.000
expect
Token expect
Feature activation+0.000
?
Token?
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
the
Token the
Feature activation+1.194
doctor
Token doctor
Feature activation+1.391
replied
Token replied
Feature activation+1.129
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
C
TokenC
Feature activation+0.000

INTERVAL 0.870 - 1.305
CONTAINS 0.003%

d
Tokend
Feature activation+0.000
like
Token like
Feature activation+0.000
to
Token to
Feature activation+0.000
know
Token know
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+0.918
after
Token after
Feature activation+0.000
the
Token the
Feature activation+0.000
better
Token better
Feature activation+0.000
part
Token part
Feature activation+0.000
of
Token of
Feature activation+0.000
order
Tokenorder
Feature activation+0.000
"...
Token"...
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.134
That
TokenThat
Feature activation+0.665
's
Token's
Feature activation+0.952
an
Token an
Feature activation+1.046
easy
Token easy
Feature activation+1.049
one
Token one
Feature activation+0.000
since
Token since
Feature activation+0.000
you
Token you
Feature activation+0.000
t
Tokent
Feature activation+0.000
I
Token I
Feature activation+0.000
get
Token get
Feature activation+0.000
stack
Token stack
Feature activation+0.000
traces
Token traces
Feature activation+0.000
?
Token?
Feature activation+0.928
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
presentation
Token presentation
Feature activation+0.000
).
Token).
Feature activation+0.000
Output
Token Output
Feature activation+0.000
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
One
TokenOne
Feature activation+0.000
potential
Token potential
Feature activation+0.397
answer
Token answer
Feature activation+1.211
is
Token is
Feature activation+0.000
phot
Token phot
Feature activation+0.000
onic
Tokenonic
Feature activation+0.000
propulsion
Token propulsion
Feature activation+0.000
.
Token.
Feature activation+0.000
our
Token our
Feature activation+0.000
star
Token star
Feature activation+0.000
?
Token?
Feature activation+0.178
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.013
,
Token,
Feature activation+1.020
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.109
Our
TokenOur
Feature activation+0.000
star
Token star
Feature activation+0.000
shouldn
Token shouldn
Feature activation+0.000

INTERVAL 0.435 - 0.870
CONTAINS 0.010%

who
Token who
Feature activation+0.000
exactly
Token exactly
Feature activation+0.000
workers
Token workers
Feature activation+0.000
are
Token are
Feature activation+0.000
,
Token,
Feature activation+1.133
the
Token the
Feature activation+0.500
prec
Token prec
Feature activation+0.000
arity
Tokenarity
Feature activation+0.000
of
Token of
Feature activation+0.000
work
Token work
Feature activation+0.000
today
Token today
Feature activation+0.067
five
Token five
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
best
Token best
Feature activation+0.000
to
Token to
Feature activation+0.000
find
Token find
Feature activation+0.694
out
Token out
Feature activation+2.074
-
Token -
Feature activation+0.000
Lucky
Token Lucky
Feature activation+0.000
Peach
Token Peach
Feature activation+0.000
,
Token,
Feature activation+0.000
Why
TokenWhy
Feature activation+0.000
Prince
Token Prince
Feature activation+0.000
Philip
Token Philip
Feature activation+0.000
?
Token?
Feature activation+0.052
Bay
Token Bay
Feature activation+0.000
lis
Tokenlis
Feature activation+0.487
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
prince
Token prince
Feature activation+0.000
's
Token's
Feature activation+0.000
visit
Token visit
Feature activation+0.000
here
Token here
Feature activation+0.000
?
Token?
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Dill
Token Dill
Feature activation+0.000
mann
Tokenmann
Feature activation+0.591
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
They
TokenThey
Feature activation+0.000
.
Token.
Feature activation+0.000
A
TokenA
Feature activation+0.000
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
how
Token how
Feature activation+0.000
,
Token,
Feature activation+0.504
after
Token after
Feature activation+0.000
a
Token a
Feature activation+0.000
month
Token month
Feature activation+0.000
,
Token,
Feature activation+0.000
she
Token she
Feature activation+0.000

INTERVAL 0.000 - 0.435
CONTAINS 99.984%

Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
investigation
Token investigation
Feature activation+0.000
of
Token of
Feature activation+0.000
Ber
Token Ber
Feature activation+0.000
gd
Tokengd
Feature activation+0.000
ahl
Tokenahl
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
disappearance
Token disappearance
Feature activation+0.000
backgrounds
Token backgrounds
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
to
Token to
Feature activation+0.000
represent
Token represent
Feature activation+0.000
the
Token the
Feature activation+0.000
people
Token people
Feature activation+0.000
who
Token who
Feature activation+0.000
died
Token died
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Next
Token Next
Feature activation+0.000
step
Token step
Feature activation+0.000
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Greek
Token Greek
Feature activation+0.000
experiment
Token experiment
Feature activation+0.000
:
Token:
Feature activation+0.000
the
Token the
Feature activation+0.000
destruction
Token destruction
Feature activation+0.000
of
Token of
Feature activation+0.000
Dro
Token Dro
Feature activation+0.000
pp
Tokenpp
Feature activation+0.000
.
Token.
Feature activation+0.000
"
Token "
Feature activation+0.000
That
TokenThat
Feature activation+0.000
being
Token being
Feature activation+0.000
said
Token said
Feature activation+0.000
,
Token,
Feature activation+0.000
most
Token most
Feature activation+0.000
voters
Token voters
Feature activation+0.000
still
Token still
Feature activation+0.000
canvas
Token canvas
Feature activation+0.000
element
Token element
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
luckily
Token luckily
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
framework
Token framework
Feature activation+0.000
can
Token can
Feature activation+0.000
take
Token take
Feature activation+0.000
advantage
Token advantage
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 2 in H8.9: (feature 9118)

TOP ACTIVATIONS
MAX = 4.203

the
Token the
Feature activation+0.000
next
Token next
Feature activation+0.000
level
Token level
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+2.587
None
Token None
Feature activation+3.211
of
Token of
Feature activation+4.203
that
Token that
Feature activation+2.978
.
Token.
Feature activation+1.463
If
Token If
Feature activation+0.247
you
Token you
Feature activation+0.306
were
Token were
Feature activation+0.284
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
next
Token next
Feature activation+0.000
level
Token level
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+2.587
None
Token None
Feature activation+3.211
of
Token of
Feature activation+4.203
that
Token that
Feature activation+2.978
.
Token.
Feature activation+1.463
If
Token If
Feature activation+0.247
you
Token you
Feature activation+0.306
least
Token least
Feature activation+0.000
one
Token one
Feature activation+0.000
or
Token or
Feature activation+0.000
two
Token two
Feature activation+0.000
of
Token of
Feature activation+0.677
those
Token those
Feature activation+3.120
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.114
we
Tokenwe
Feature activation+0.000
just
Token just
Feature activation+0.000
weren
Token weren
Feature activation+0.000
and
Token and
Feature activation+0.000
creativity
Token creativity
Feature activation+0.000
.
Token.
Feature activation+0.774
To
Token To
Feature activation+0.688
fuel
Token fuel
Feature activation+0.439
these
Token these
Feature activation+3.119
things
Token things
Feature activation+0.345
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
need
Token need
Feature activation+0.022
good
Token good
Feature activation+0.054
has
Token has
Feature activation+0.000
reversed
Token reversed
Feature activation+0.000
course
Token course
Feature activation+0.048
on
Token on
Feature activation+1.282
all
Token all
Feature activation+2.582
of
Token of
Feature activation+2.979
them
Token them
Feature activation+0.595
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
is
Token is
Feature activation+0.000
even
Token even
Feature activation+0.000
next
Token next
Feature activation+0.000
level
Token level
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+2.587
None
Token None
Feature activation+3.211
of
Token of
Feature activation+4.203
that
Token that
Feature activation+2.978
.
Token.
Feature activation+1.463
If
Token If
Feature activation+0.247
you
Token you
Feature activation+0.306
were
Token were
Feature activation+0.284
to
Token to
Feature activation+0.566
trump
Token trump
Feature activation+0.000
card
Token card
Feature activation+0.142
that
Token that
Feature activation+0.000
can
Token can
Feature activation+0.370
overcome
Token overcome
Feature activation+2.165
everything
Token everything
Feature activation+2.970
else
Token else
Feature activation+0.611
.
Token.
Feature activation+0.293
It
Token It
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ization
Tokenization
Feature activation+0.000
process
Token process
Feature activation+0.000
affect
Token affect
Feature activation+0.936
each
Token each
Feature activation+1.337
of
Token of
Feature activation+2.469
these
Token these
Feature activation+2.945
?
Token?
Feature activation+0.227
One
Token One
Feature activation+0.000
pathway
Token pathway
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
privatization
Token privatization
Feature activation+0.000
of
Token of
Feature activation+0.000
publicly
Token publicly
Feature activation+0.000
owned
Token owned
Feature activation+0.000
companies
Token companies
Feature activation+0.000
.
Token.
Feature activation+2.891
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.270
The
TokenThe
Feature activation+0.874
banking
Token banking
Feature activation+0.000
and
Token and
Feature activation+0.000
.
Token.
Feature activation+0.387
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Each
TokenEach
Feature activation+0.723
of
Token of
Feature activation+0.877
these
Token these
Feature activation+2.842
well
Token well
Feature activation+0.000
-
Token-
Feature activation+0.000
documented
Tokendocumented
Feature activation+0.950
social
Token social
Feature activation+0.000
class
Token class
Feature activation+0.000
out
Tokenout
Feature activation+0.000
let
Tokenlet
Feature activation+0.000
surgery
Token surgery
Feature activation+0.299
.
Token.
Feature activation+0.346
Despite
Token Despite
Feature activation+1.054
those
Token those
Feature activation+2.704
setbacks
Token setbacks
Feature activation+0.106
âĢĶ
Token âĢĶ
Feature activation+0.000
or
Token or
Feature activation+0.000
perhaps
Token perhaps
Feature activation+0.100
to
Token to
Feature activation+0.000
housing
Token housing
Feature activation+0.000
and
Token and
Feature activation+0.000
unstable
Token unstable
Feature activation+0.000
families
Token families
Feature activation+0.828
âĢĶ
Token âĢĶ
Feature activation+2.362
all
Token all
Feature activation+2.684
are
Token are
Feature activation+1.913
factors
Token factors
Feature activation+0.629
that
Token that
Feature activation+0.405
contribute
Token contribute
Feature activation+0.151
to
Token to
Feature activation+0.120
it
Token it
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
next
Token next
Feature activation+0.000
level
Token level
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+2.587
None
Token None
Feature activation+3.211
of
Token of
Feature activation+4.203
that
Token that
Feature activation+2.978
.
Token.
Feature activation+1.463
If
Token If
Feature activation+0.247
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
reversed
Token reversed
Feature activation+0.000
course
Token course
Feature activation+0.048
on
Token on
Feature activation+1.282
all
Token all
Feature activation+2.582
of
Token of
Feature activation+2.979
them
Token them
Feature activation+0.595
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
is
Token is
Feature activation+0.000
urban
Token urban
Feature activation+0.000
ization
Tokenization
Feature activation+0.000
process
Token process
Feature activation+0.000
affect
Token affect
Feature activation+0.936
each
Token each
Feature activation+1.337
of
Token of
Feature activation+2.469
these
Token these
Feature activation+2.945
?
Token?
Feature activation+0.227
One
Token One
Feature activation+0.000
pathway
Token pathway
Feature activation+0.000
would
Token would
Feature activation+0.000
responsibility
Token responsibility
Feature activation+0.197
and
Token and
Feature activation+0.347
freedom
Token freedom
Feature activation+0.221
,
Token,
Feature activation+1.395
among
Token among
Feature activation+2.451
others
Token others
Feature activation+2.468
âĢĶ
Token âĢĶ
Feature activation+1.922
are
Token are
Feature activation+2.218
not
Token not
Feature activation+1.554
unique
Token unique
Feature activation+0.660
to
Token to
Feature activation+0.728
is
Token is
Feature activation+0.488
an
Token an
Feature activation+0.407
attempt
Token attempt
Feature activation+0.000
to
Token to
Feature activation+0.940
put
Token put
Feature activation+0.652
all
Token all
Feature activation+2.468
the
Token the
Feature activation+1.919
resources
Token resources
Feature activation+0.299
in
Token in
Feature activation+0.102
one
Token one
Feature activation+0.000
place
Token place
Feature activation+0.000
personal
Token personal
Feature activation+0.000
responsibility
Token responsibility
Feature activation+0.197
and
Token and
Feature activation+0.347
freedom
Token freedom
Feature activation+0.221
,
Token,
Feature activation+1.395
among
Token among
Feature activation+2.451
others
Token others
Feature activation+2.468
âĢĶ
Token âĢĶ
Feature activation+1.922
are
Token are
Feature activation+2.218
not
Token not
Feature activation+1.554
unique
Token unique
Feature activation+0.660
standard
Tokenstandard
Feature activation+0.000
housing
Token housing
Feature activation+0.000
and
Token and
Feature activation+0.000
unstable
Token unstable
Feature activation+0.000
families
Token families
Feature activation+0.828
âĢĶ
Token âĢĶ
Feature activation+2.362
all
Token all
Feature activation+2.684
are
Token are
Feature activation+1.913
factors
Token factors
Feature activation+0.629
that
Token that
Feature activation+0.405
contribute
Token contribute
Feature activation+0.151
brain
Token brain
Feature activation+0.000
has
Token has
Feature activation+0.000
to
Token to
Feature activation+0.000
combine
Token combine
Feature activation+1.317
all
Token all
Feature activation+2.049
of
Token of
Feature activation+2.356
this
Token this
Feature activation+0.535
information
Token information
Feature activation+0.002
to
Token to
Feature activation+0.000
influence
Token influence
Feature activation+0.000
the
Token the
Feature activation+0.000

Top DFA by src position
MAX = 2.606

to
Token to
Feature activation+0.009
Top resid features:
destroy
Token destroy
Feature activation-0.001
Top resid features:
the
Token the
Feature activation-0.001
Top resid features:
world
Token world
Feature activation+0.005
Top resid features:
,
Token,
Feature activation+0.239
Top resid features:
no
Token no
Feature activation+2.037
Top resid features:
super
Token super
Feature activation+0.000
Top resid features:
villains
Token villains
Feature activation+0.022
Top resid features:
,
Token,
Feature activation+0.395
Top resid features:
no
Token no
Feature activation+0.632
Top resid features:
armies
Token armies
Feature activation+0.018
Top resid features:
to
Token to
Feature activation+0.018
Top resid features:
destroy
Token destroy
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.001
Top resid features:
world
Token world
Feature activation+0.009
Top resid features:
,
Token,
Feature activation+0.205
Top resid features:
no
Token no
Feature activation+1.469
Top resid features:
super
Token super
Feature activation-0.000
Top resid features:
villains
Token villains
Feature activation+0.034
Top resid features:
,
Token,
Feature activation+0.260
Top resid features:
no
Token no
Feature activation+0.543
Top resid features:
armies
Token armies
Feature activation+0.035
Top resid features:
cold
Token cold
Feature activation+0.013
Top resid features:
,
Token,
Feature activation+0.714
Top resid features:
poor
Token poor
Feature activation+0.167
Top resid features:
special
Token special
Feature activation+0.002
Top resid features:
teams
Token teams
Feature activation+0.069
Top resid features:
,
Token,
Feature activation+1.222
Top resid features:
or
Token or
Feature activation+0.427
Top resid features:
just
Token just
Feature activation+0.060
Top resid features:
plain
Token plain
Feature activation+0.006
Top resid features:
bad
Token bad
Feature activation+0.050
Top resid features:
luck
Token luck
Feature activation+0.084
Top resid features:
stamina
Token stamina
Feature activation+0.052
Top resid features:
,
Token,
Feature activation+0.374
Top resid features:
a
Token a
Feature activation+0.256
Top resid features:
clear
Token clear
Feature activation+0.019
Top resid features:
mind
Token mind
Feature activation+0.049
Top resid features:
and
Token and
Feature activation+2.606
Top resid features:
creativity
Token creativity
Feature activation+0.227
Top resid features:
.
Token.
Feature activation+0.082
Top resid features:
To
Token To
Feature activation+0.002
Top resid features:
fuel
Token fuel
Feature activation-0.021
Top resid features:
these
Token these
Feature activation-0.037
Top resid features:
illegal
Token illegal
Feature activation+0.005
Top resid features:
guns
Token guns
Feature activation+0.009
Top resid features:
,
Token,
Feature activation+1.254
Top resid features:
abortion
Token abortion
Feature activation+0.012
Top resid features:
rights
Token rights
Feature activation+0.024
Top resid features:
and
Token and
Feature activation+1.514
Top resid features:
health
Token health
Feature activation+0.012
Top resid features:
care
Token care
Feature activation+0.019
Top resid features:
.
Token.
Feature activation+0.147
Top resid features:
But
Token But
Feature activation+0.033
Top resid features:
he
Token he
Feature activation+0.013
Top resid features:
to
Token to
Feature activation+0.019
Top resid features:
destroy
Token destroy
Feature activation+0.006
Top resid features:
the
Token the
Feature activation+0.003
Top resid features:
world
Token world
Feature activation+0.012
Top resid features:
,
Token,
Feature activation+0.354
Top resid features:
no
Token no
Feature activation+1.345
Top resid features:
super
Token super
Feature activation-0.000
Top resid features:
villains
Token villains
Feature activation+0.038
Top resid features:
,
Token,
Feature activation+0.264
Top resid features:
no
Token no
Feature activation+0.495
Top resid features:
armies
Token armies
Feature activation+0.034
Top resid features:
,
Token,
Feature activation+0.134
Top resid features:
an
Token an
Feature activation+0.334
Top resid features:
offense
Token offense
Feature activation+0.020
Top resid features:
gone
Token gone
Feature activation+0.004
Top resid features:
cold
Token cold
Feature activation+0.020
Top resid features:
,
Token,
Feature activation+0.994
Top resid features:
poor
Token poor
Feature activation+0.107
Top resid features:
special
Token special
Feature activation+0.001
Top resid features:
teams
Token teams
Feature activation+0.017
Top resid features:
,
Token,
Feature activation+0.747
Top resid features:
or
Token or
Feature activation+0.799
Top resid features:
Health
Token Health
Feature activation+0.054
Top resid features:
care
Token care
Feature activation+0.024
Top resid features:
,
Token,
Feature activation+0.190
Top resid features:
good
Token good
Feature activation+0.181
Top resid features:
nutrition
Token nutrition
Feature activation+0.134
Top resid features:
,
Token,
Feature activation+0.761
Top resid features:
employment
Token employment
Feature activation+0.085
Top resid features:
,
Token,
Feature activation+0.679
Top resid features:
education
Token education
Feature activation+0.206
Top resid features:
for
Token for
Feature activation+0.045
Top resid features:
their
Token their
Feature activation-0.003
Top resid features:
around
Token around
Feature activation+0.003
Top resid features:
7
Token 7
Feature activation-0.000
Top resid features:
.
Token.
Feature activation-0.000
Top resid features:
5
Token5
Feature activation-0.000
Top resid features:
percent
Token percent
Feature activation-0.000
Top resid features:
),
Token),
Feature activation+1.073
Top resid features:
guaranteed
Token guaranteed
Feature activation+0.013
Top resid features:
pensions
Token pensions
Feature activation+0.003
Top resid features:
,
Token,
Feature activation+0.280
Top resid features:
and
Token and
Feature activation+0.430
Top resid features:
an
Token an
Feature activation+0.166
Top resid features:
nutrition
Token nutrition
Feature activation+0.018
Top resid features:
,
Token,
Feature activation+0.069
Top resid features:
inadequate
Token inadequate
Feature activation+0.479
Top resid features:
health
Token health
Feature activation+0.062
Top resid features:
care
Token care
Feature activation+0.032
Top resid features:
,
Token,
Feature activation+0.792
Top resid features:
sub
Token sub
Feature activation-0.001
Top resid features:
standard
Tokenstandard
Feature activation+0.550
Top resid features:
housing
Token housing
Feature activation+0.043
Top resid features:
and
Token and
Feature activation+0.545
Top resid features:
unstable
Token unstable
Feature activation+0.526
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
c
Tokenc
Feature activation+0.001
Top resid features:
uff
Tokenuff
Feature activation+0.005
Top resid features:
surgery
Token surgery
Feature activation+0.306
Top resid features:
,
Token,
Feature activation+1.188
Top resid features:
and
Token and
Feature activation+1.607
Top resid features:
thor
Token thor
Feature activation+0.002
Top resid features:
ac
Tokenac
Feature activation+0.001
Top resid features:
ic
Tokenic
Feature activation+0.007
Top resid features:
-
Token-
Feature activation+0.002
Top resid features:
out
Tokenout
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.651
Top resid features:
sub
Token sub
Feature activation-0.001
Top resid features:
standard
Tokenstandard
Feature activation+0.571
Top resid features:
housing
Token housing
Feature activation+0.063
Top resid features:
and
Token and
Feature activation+0.512
Top resid features:
unstable
Token unstable
Feature activation+0.759
Top resid features:
families
Token families
Feature activation+0.106
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.098
Top resid features:
all
Token all
Feature activation+0.015
Top resid features:
are
Token are
Feature activation+0.000
Top resid features:
factors
Token factors
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.014
Top resid features:
destroy
Token destroy
Feature activation+0.001
Top resid features:
the
Token the
Feature activation-0.002
Top resid features:
world
Token world
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.214
Top resid features:
no
Token no
Feature activation+1.017
Top resid features:
super
Token super
Feature activation+0.000
Top resid features:
villains
Token villains
Feature activation+0.036
Top resid features:
,
Token,
Feature activation+0.504
Top resid features:
no
Token no
Feature activation+0.282
Top resid features:
armies
Token armies
Feature activation+0.015
Top resid features:
illegal
Token illegal
Feature activation+0.007
Top resid features:
guns
Token guns
Feature activation+0.015
Top resid features:
,
Token,
Feature activation+1.128
Top resid features:
abortion
Token abortion
Feature activation+0.028
Top resid features:
rights
Token rights
Feature activation+0.041
Top resid features:
and
Token and
Feature activation+1.371
Top resid features:
health
Token health
Feature activation+0.026
Top resid features:
care
Token care
Feature activation+0.033
Top resid features:
.
Token.
Feature activation+0.086
Top resid features:
But
Token But
Feature activation+0.026
Top resid features:
he
Token he
Feature activation+0.006
Top resid features:
their
Token their
Feature activation-0.002
Top resid features:
kids
Token kids
Feature activation+0.012
Top resid features:
?
Token?
Feature activation+0.536
Top resid features:
Street
Token Street
Feature activation+0.062
Top resid features:
safety
Token safety
Feature activation+0.139
Top resid features:
?
Token?
Feature activation+0.722
Top resid features:
How
Token How
Feature activation+0.036
Top resid features:
does
Token does
Feature activation+0.015
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
urban
Token urban
Feature activation+0.033
Top resid features:
ization
Tokenization
Feature activation+0.004
Top resid features:
Republican
TokenRepublican
Feature activation-0.001
Top resid features:
values
Token values
Feature activation+0.006
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.187
Top resid features:
strong
Token strong
Feature activation+0.145
Top resid features:
families
Token families
Feature activation+0.061
Top resid features:
,
Token,
Feature activation+0.816
Top resid features:
faith
Token faith
Feature activation+0.036
Top resid features:
,
Token,
Feature activation+0.531
Top resid features:
personal
Token personal
Feature activation+0.032
Top resid features:
responsibility
Token responsibility
Feature activation+0.145
Top resid features:
and
Token and
Feature activation+0.351
Top resid features:
Kung
Token Kung
Feature activation-0.002
Top resid features:
Fu
Token Fu
Feature activation+0.001
Top resid features:
mailing
Token mailing
Feature activation-0.002
Top resid features:
list
Token list
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.099
Top resid features:
Ċ
TokenĊ
Feature activation+0.462
Top resid features:
This
TokenThis
Feature activation+0.006
Top resid features:
blog
Token blog
Feature activation+0.001
Top resid features:
post
Tokenpost
Feature activation+0.005
Top resid features:
is
Token is
Feature activation-0.000
Top resid features:
an
Token an
Feature activation+0.006
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.178
Top resid features:
strong
Token strong
Feature activation+0.086
Top resid features:
families
Token families
Feature activation+0.044
Top resid features:
,
Token,
Feature activation+0.652
Top resid features:
faith
Token faith
Feature activation+0.034
Top resid features:
,
Token,
Feature activation+0.837
Top resid features:
personal
Token personal
Feature activation+0.077
Top resid features:
responsibility
Token responsibility
Feature activation+0.100
Top resid features:
and
Token and
Feature activation+0.350
Top resid features:
freedom
Token freedom
Feature activation+0.197
Top resid features:
,
Token,
Feature activation+0.475
Top resid features:
nutrition
Token nutrition
Feature activation+0.007
Top resid features:
,
Token,
Feature activation+0.190
Top resid features:
inadequate
Token inadequate
Feature activation+0.220
Top resid features:
health
Token health
Feature activation+0.041
Top resid features:
care
Token care
Feature activation+0.024
Top resid features:
,
Token,
Feature activation+1.152
Top resid features:
sub
Token sub
Feature activation-0.001
Top resid features:
standard
Tokenstandard
Feature activation+0.280
Top resid features:
housing
Token housing
Feature activation+0.024
Top resid features:
and
Token and
Feature activation+0.708
Top resid features:
unstable
Token unstable
Feature activation+0.200
Top resid features:
does
Tokendoes
Feature activation+0.017
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
female
Token female
Feature activation-0.001
Top resid features:
look
Token look
Feature activation-0.000
Top resid features:
interested
Token interested
Feature activation-0.000
Top resid features:
?),
Token?),
Feature activation+1.463
Top resid features:
and
Token and
Feature activation+0.653
Top resid features:
touch
Token touch
Feature activation+0.051
Top resid features:
(
Token (
Feature activation+0.125
Top resid features:
am
Tokenam
Feature activation+0.001
Top resid features:
I
Token I
Feature activation+0.001
Top resid features:

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.03

Head 2: 0.05

Head 3: 0.10

Head 4: 0.03

Head 5: 0.05

Head 6: 0.03

Head 7: 0.05

Head 8: 0.05

Head 9: 0.43

Head 10: 0.11

Head 11: 0.04

Positive logits

these4.07

These3.93

None3.51

these3.38

These3.34

things3.25

Those3.12

everything3.04

none3.03

essentials3.02

THESE2.98

totality2.93

nutshell2.87

none2.85

negatives2.74

whatever2.70

those2.63

everything2.63

anything2.63

latter2.61

Negative logits

fficient-2.25

legraph-2.19

bra-2.15

hran-2.10

Correction-2.07

izont-2.05

awa-2.04

berth-2.02

iott-1.97

kiss-1.96

osponsors-1.95

Rouge-1.95

Penalty-1.95

Flam-1.94

\":-1.91

vati-1.91

banning-1.90

Tx-1.90

intercept-1.89

veland-1.89

INTERVAL 3.783 - 4.203
CONTAINS 0.000%

the
Token the
Feature activation+0.000
next
Token next
Feature activation+0.000
level
Token level
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+2.587
None
Token None
Feature activation+3.211
of
Token of
Feature activation+4.203
that
Token that
Feature activation+2.978
.
Token.
Feature activation+1.463
If
Token If
Feature activation+0.247
you
Token you
Feature activation+0.306
were
Token were
Feature activation+0.284

INTERVAL 3.362 - 3.783
CONTAINS 0.000%

INTERVAL 2.942 - 3.362
CONTAINS 0.001%

trump
Token trump
Feature activation+0.000
card
Token card
Feature activation+0.142
that
Token that
Feature activation+0.000
can
Token can
Feature activation+0.370
overcome
Token overcome
Feature activation+2.165
everything
Token everything
Feature activation+2.970
else
Token else
Feature activation+0.611
.
Token.
Feature activation+0.293
It
Token It
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
has
Token has
Feature activation+0.000
reversed
Token reversed
Feature activation+0.000
course
Token course
Feature activation+0.048
on
Token on
Feature activation+1.282
all
Token all
Feature activation+2.582
of
Token of
Feature activation+2.979
them
Token them
Feature activation+0.595
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
is
Token is
Feature activation+0.000
even
Token even
Feature activation+0.000
next
Token next
Feature activation+0.000
level
Token level
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+2.587
None
Token None
Feature activation+3.211
of
Token of
Feature activation+4.203
that
Token that
Feature activation+2.978
.
Token.
Feature activation+1.463
If
Token If
Feature activation+0.247
you
Token you
Feature activation+0.306
were
Token were
Feature activation+0.284
to
Token to
Feature activation+0.566
least
Token least
Feature activation+0.000
one
Token one
Feature activation+0.000
or
Token or
Feature activation+0.000
two
Token two
Feature activation+0.000
of
Token of
Feature activation+0.677
those
Token those
Feature activation+3.120
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.114
we
Tokenwe
Feature activation+0.000
just
Token just
Feature activation+0.000
weren
Token weren
Feature activation+0.000
and
Token and
Feature activation+0.000
creativity
Token creativity
Feature activation+0.000
.
Token.
Feature activation+0.774
To
Token To
Feature activation+0.688
fuel
Token fuel
Feature activation+0.439
these
Token these
Feature activation+3.119
things
Token things
Feature activation+0.345
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
need
Token need
Feature activation+0.022
good
Token good
Feature activation+0.054

INTERVAL 2.522 - 2.942
CONTAINS 0.001%

privatization
Token privatization
Feature activation+0.000
of
Token of
Feature activation+0.000
publicly
Token publicly
Feature activation+0.000
owned
Token owned
Feature activation+0.000
companies
Token companies
Feature activation+0.000
.
Token.
Feature activation+2.891
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.270
The
TokenThe
Feature activation+0.874
banking
Token banking
Feature activation+0.000
and
Token and
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
reversed
Token reversed
Feature activation+0.000
course
Token course
Feature activation+0.048
on
Token on
Feature activation+1.282
all
Token all
Feature activation+2.582
of
Token of
Feature activation+2.979
them
Token them
Feature activation+0.595
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
is
Token is
Feature activation+0.000
.
Token.
Feature activation+0.387
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Each
TokenEach
Feature activation+0.723
of
Token of
Feature activation+0.877
these
Token these
Feature activation+2.842
well
Token well
Feature activation+0.000
-
Token-
Feature activation+0.000
documented
Tokendocumented
Feature activation+0.950
social
Token social
Feature activation+0.000
class
Token class
Feature activation+0.000
out
Tokenout
Feature activation+0.000
let
Tokenlet
Feature activation+0.000
surgery
Token surgery
Feature activation+0.299
.
Token.
Feature activation+0.346
Despite
Token Despite
Feature activation+1.054
those
Token those
Feature activation+2.704
setbacks
Token setbacks
Feature activation+0.106
âĢĶ
Token âĢĶ
Feature activation+0.000
or
Token or
Feature activation+0.000
perhaps
Token perhaps
Feature activation+0.100
to
Token to
Feature activation+0.000
housing
Token housing
Feature activation+0.000
and
Token and
Feature activation+0.000
unstable
Token unstable
Feature activation+0.000
families
Token families
Feature activation+0.828
âĢĶ
Token âĢĶ
Feature activation+2.362
all
Token all
Feature activation+2.684
are
Token are
Feature activation+1.913
factors
Token factors
Feature activation+0.629
that
Token that
Feature activation+0.405
contribute
Token contribute
Feature activation+0.151
to
Token to
Feature activation+0.120

INTERVAL 2.101 - 2.522
CONTAINS 0.002%

personal
Token personal
Feature activation+0.000
responsibility
Token responsibility
Feature activation+0.197
and
Token and
Feature activation+0.347
freedom
Token freedom
Feature activation+0.221
,
Token,
Feature activation+1.395
among
Token among
Feature activation+2.451
others
Token others
Feature activation+2.468
âĢĶ
Token âĢĶ
Feature activation+1.922
are
Token are
Feature activation+2.218
not
Token not
Feature activation+1.554
unique
Token unique
Feature activation+0.660
standard
Tokenstandard
Feature activation+0.000
housing
Token housing
Feature activation+0.000
and
Token and
Feature activation+0.000
unstable
Token unstable
Feature activation+0.000
families
Token families
Feature activation+0.828
âĢĶ
Token âĢĶ
Feature activation+2.362
all
Token all
Feature activation+2.684
are
Token are
Feature activation+1.913
factors
Token factors
Feature activation+0.629
that
Token that
Feature activation+0.405
contribute
Token contribute
Feature activation+0.151
urban
Token urban
Feature activation+0.000
ization
Tokenization
Feature activation+0.000
process
Token process
Feature activation+0.000
affect
Token affect
Feature activation+0.936
each
Token each
Feature activation+1.337
of
Token of
Feature activation+2.469
these
Token these
Feature activation+2.945
?
Token?
Feature activation+0.227
One
Token One
Feature activation+0.000
pathway
Token pathway
Feature activation+0.000
would
Token would
Feature activation+0.000
freedom
Token freedom
Feature activation+0.221
,
Token,
Feature activation+1.395
among
Token among
Feature activation+2.451
others
Token others
Feature activation+2.468
âĢĶ
Token âĢĶ
Feature activation+1.922
are
Token are
Feature activation+2.218
not
Token not
Feature activation+1.554
unique
Token unique
Feature activation+0.660
to
Token to
Feature activation+0.728
specific
Token specific
Feature activation+0.914
subs
Token subs
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.313
The
TokenThe
Feature activation+0.021
influence
Token influence
Feature activation+0.182
of
Token of
Feature activation+0.791
all
Token all
Feature activation+2.212
these
Token these
Feature activation+1.650
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.377
climate
Tokenclimate
Feature activation+0.254
drivers
Token drivers
Feature activation+0.000

INTERVAL 1.681 - 2.101
CONTAINS 0.004%

to
Token to
Feature activation+0.000
particular
Token particular
Feature activation+0.225
industries
Token industries
Feature activation+0.000
or
Token or
Feature activation+0.000
companies
Token companies
Feature activation+0.000
.
Token.
Feature activation+1.819
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.638
Ċ
TokenĊ
Feature activation+0.401
Ċ
TokenĊ
Feature activation+0.394
In
TokenIn
Feature activation+0.950
a
Token a
Feature activation+0.000
well
Token well
Feature activation+0.000
-
Token-
Feature activation+0.000
told
Tokentold
Feature activation+0.000
tale
Token tale
Feature activation+0.000
...
Token...
Feature activation+1.803
the
Token the
Feature activation+0.765
ph
Token ph
Feature activation+0.000
at
Tokenat
Feature activation+0.000
loot
Token loot
Feature activation+0.000
.
Token.
Feature activation+1.511
money
Token money
Feature activation+0.000
,
Token,
Feature activation+0.799
growing
Token growing
Feature activation+0.102
ratings
Token ratings
Feature activation+0.000
.
Token.
Feature activation+1.873
They
Token They
Feature activation+1.883
are
Token are
Feature activation+1.971
all
Token all
Feature activation+1.752
post
Token post
Feature activation+0.847
-
Token-
Feature activation+0.847
Beck
TokenBeck
Feature activation+0.000
Americans
Token Americans
Feature activation+0.000
by
Token by
Feature activation+0.000
talking
Token talking
Feature activation+0.000
about
Token about
Feature activation+0.000
ISIS
Token ISIS
Feature activation+0.000
.
Token.
Feature activation+1.749
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.914
I
TokenI
Feature activation+0.150
was
Token was
Feature activation+0.143
stunned
Token stunned
Feature activation+0.278
he
Tokenhe
Feature activation+0.000
most
Token most
Feature activation+0.000
certainly
Token certainly
Feature activation+0.000
is
Token is
Feature activation+0.000
all
Token all
Feature activation+0.669
those
Token those
Feature activation+2.001
things
Token things
Feature activation+0.524
âĢĶ
TokenâĢĶ
Feature activation+0.000
he
Tokenhe
Feature activation+0.000
's
Token's
Feature activation+0.000
just
Token just
Feature activation+0.000

INTERVAL 1.261 - 1.681
CONTAINS 0.010%

government
Token government
Feature activation+0.000
communications
Token communications
Feature activation+0.000
,
Token,
Feature activation+0.146
as
Token as
Feature activation+0.000
saying
Token saying
Feature activation+0.149
the
Token the
Feature activation+1.492
13
Token 13
Feature activation+0.000
demands
Token demands
Feature activation+0.626
bore
Token bore
Feature activation+0.021
no
Token no
Feature activation+0.000
relation
Token relation
Feature activation+0.000
cur
Token cur
Feature activation+0.000
bs
Tokenbs
Feature activation+0.000
on
Token on
Feature activation+0.000
arms
Token arms
Feature activation+0.000
exports
Token exports
Feature activation+0.000
.
Token.
Feature activation+1.306
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.778
R
TokenR
Feature activation+0.000
IL
TokenIL
Feature activation+0.000
ING
TokenING
Feature activation+0.000
study
Token study
Feature activation+0.000
measures
Token measures
Feature activation+0.482
satisfaction
Token satisfaction
Feature activation+0.000
in
Token in
Feature activation+0.000
those
Token those
Feature activation+0.414
same
Token same
Feature activation+1.322
five
Token five
Feature activation+1.832
factors
Token factors
Feature activation+0.174
plus
Token plus
Feature activation+0.000
a
Token a
Feature activation+0.000
sixth
Token sixth
Feature activation+0.000
ert
Tokenert
Feature activation+0.000
said
Token said
Feature activation+0.000
,
Token,
Feature activation+0.000
adding
Token adding
Feature activation+0.000
that
Token that
Feature activation+0.190
these
Token these
Feature activation+1.269
will
Token will
Feature activation+0.191
be
Token be
Feature activation+0.171
the
Token the
Feature activation+0.260
first
Token first
Feature activation+0.129
steps
Token steps
Feature activation+0.000
...
Token...
Feature activation+1.803
the
Token the
Feature activation+0.765
ph
Token ph
Feature activation+0.000
at
Tokenat
Feature activation+0.000
loot
Token loot
Feature activation+0.000
.
Token.
Feature activation+1.511
While
TokenWhile
Feature activation+0.352
certainly
Token certainly
Feature activation+0.301
not
Token not
Feature activation+0.223
universally
Token universally
Feature activation+0.033
applicable
Token applicable
Feature activation+0.184

INTERVAL 0.841 - 1.261
CONTAINS 0.024%

comp
Tokencomp
Feature activation+0.178
ared
Tokenared
Feature activation+0.000
to
Token to
Feature activation+0.324
the
Token the
Feature activation+0.198
west
Token west
Feature activation+0.000
)
Token)
Feature activation+0.862
and
Token and
Feature activation+0.510
the
Token the
Feature activation+0.476
huge
Token huge
Feature activation+0.000
surge
Token surge
Feature activation+0.005
of
Token of
Feature activation+0.000
,
Token,
Feature activation+0.000
rough
Token rough
Feature activation+0.000
and
Token and
Feature activation+0.000
ready
Token ready
Feature activation+0.000
,
Token,
Feature activation+0.000
all
Token all
Feature activation+0.892
-
Token-
Feature activation+0.106
action
Tokenaction
Feature activation+0.000
guys
Token guys
Feature activation+0.000
and
Token and
Feature activation+0.000
that
Token that
Feature activation+1.195
it
Token it
Feature activation+0.000
currently
Token currently
Feature activation+0.000
is
Token is
Feature activation+0.000
.
Token.
Feature activation+0.188
All
Token All
Feature activation+0.908
of
Token of
Feature activation+1.235
which
Token which
Feature activation+0.645
are
Token are
Feature activation+0.927
things
Token things
Feature activation+0.529
to
Token to
Feature activation+0.000
consider
Token consider
Feature activation+0.278
,
Token,
Feature activation+0.000
climate
Token climate
Feature activation+0.000
change
Token change
Feature activation+0.000
and
Token and
Feature activation+0.040
any
Token any
Feature activation+0.791
other
Token other
Feature activation+1.031
urgent
Token urgent
Feature activation+0.744
,
Token,
Feature activation+0.019
hugely
Token hugely
Feature activation+0.011
important
Token important
Feature activation+0.910
issue
Token issue
Feature activation+0.103
users
Token users
Feature activation+0.000
.
Token.
Feature activation+0.020
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Each
TokenEach
Feature activation+0.451
of
Token of
Feature activation+0.904
these
Token these
Feature activation+0.460
methods
Token methods
Feature activation+0.000
has
Token has
Feature activation+0.000
its
Token its
Feature activation+0.000
own
Token own
Feature activation+0.000

INTERVAL 0.420 - 0.841
CONTAINS 0.093%

ecological
Token ecological
Feature activation+0.000
damage
Token damage
Feature activation+0.000
and
Token and
Feature activation+0.000
social
Token social
Feature activation+0.000
unrest
Token unrest
Feature activation+0.000
unless
Token unless
Feature activation+0.718
countries
Token countries
Feature activation+0.000
become
Token become
Feature activation+0.000
sustainable
Token sustainable
Feature activation+0.000
,
Token,
Feature activation+0.314
the
Token the
Feature activation+0.456
Ċ
TokenĊ
Feature activation+0.279
Ċ
TokenĊ
Feature activation+0.349
Mir
TokenMir
Feature activation+0.000
ac
Tokenac
Feature activation+0.000
ast
Tokenast
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.453
Ċ
TokenĊ
Feature activation+0.601
Google
TokenGoogle
Feature activation+0.000
Cast
Token Cast
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.192
a
Token a
Feature activation+0.006
hell
Token hell
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
lot
Token lot
Feature activation+0.325
of
Token of
Feature activation+0.450
death
Token death
Feature activation+0.000
).
Token).
Feature activation+0.226
Every
Token Every
Feature activation+0.000
moment
Token moment
Feature activation+0.000
women
Token women
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.246
I
TokenI
Feature activation+0.004
think
Token think
Feature activation+0.000
I
Token I
Feature activation+0.000
played
Token played
Feature activation+0.721
it
Token it
Feature activation+0.089
one
Token one
Feature activation+0.149
year
Token year
Feature activation+0.047
âĢĶ
Token âĢĶ
Feature activation+0.000
I
Token I
Feature activation+0.000
golf
Token golf
Feature activation+0.000
equipment
Token equipment
Feature activation+0.000
,
Token,
Feature activation+0.000
tennis
Token tennis
Feature activation+0.000
equipment
Token equipment
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.693
you
Token you
Feature activation+0.658
name
Token name
Feature activation+0.000
it
Token it
Feature activation+0.268
âĢĵ
Token âĢĵ
Feature activation+0.299
simply
Token simply
Feature activation+0.141

INTERVAL 0.000 - 0.420
CONTAINS 99.866%

garn
Token garn
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
any
Token any
Feature activation+0.000
interest
Token interest
Feature activation+0.000
.
Token.
Feature activation+0.000
Usually
Token Usually
Feature activation+0.000
,
Token,
Feature activation+0.000
whenever
Token whenever
Feature activation+0.000
I
Token I
Feature activation+0.000
post
Token post
Feature activation+0.000
about
Token about
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Here
TokenHere
Feature activation+0.000
are
Token are
Feature activation+0.000
two
Token two
Feature activation+0.000
quotes
Token quotes
Feature activation+0.000
with
Token with
Feature activation+0.000
which
Token which
Feature activation+0.000
I
Token I
Feature activation+0.000
tend
Token tend
Feature activation+0.000
to
Token to
Feature activation+0.000
agree
Token agree
Feature activation+0.000
look
Token look
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
university
Token university
Feature activation+0.000
fee
Token fee
Feature activation+0.000
issue
Token issue
Feature activation+0.000
I
Token I
Feature activation+0.000
'm
Token'm
Feature activation+0.000
seeing
Token seeing
Feature activation+0.000
people
Token people
Feature activation+0.000
who
Token who
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
1
Token 1
Feature activation+0.000
inch
Token inch
Feature activation+0.000
punch
Token punch
Feature activation+0.000
or
Token or
Feature activation+0.000
something
Token something
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
here
Token here
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
He
TokenHe
Feature activation+0.000
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
closely
Token closely
Feature activation+0.000
monitored
Token monitored
Feature activation+0.000
by
Token by
Feature activation+0.000
medical
Token medical
Feature activation+0.000
staff
Token staff
Feature activation+0.000
this
Token this
Feature activation+0.000
week
Token week
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 3 in H8.9: (feature 8531)

TOP ACTIVATIONS
MAX = 3.865

it
Token it
Feature activation+0.418
was
Token was
Feature activation+0.000
genius
Token genius
Feature activation+0.000
.
Token.
Feature activation+3.500
We
Token We
Feature activation+3.227
'll
Token'll
Feature activation+3.865
see
Token see
Feature activation+3.790
how
Token how
Feature activation+0.482
it
Token it
Feature activation+0.000
plays
Token plays
Feature activation+0.647
going
Token going
Feature activation+0.000
was
Token was
Feature activation+0.000
genius
Token genius
Feature activation+0.000
.
Token.
Feature activation+3.500
We
Token We
Feature activation+3.227
'll
Token'll
Feature activation+3.865
see
Token see
Feature activation+3.790
how
Token how
Feature activation+0.482
it
Token it
Feature activation+0.000
plays
Token plays
Feature activation+0.647
going
Token going
Feature activation+0.000
forward
Token forward
Feature activation+0.822
Or
Token Or
Feature activation+0.000
maybe
Token maybe
Feature activation+0.000
it
Token it
Feature activation+0.418
was
Token was
Feature activation+0.000
genius
Token genius
Feature activation+0.000
.
Token.
Feature activation+3.500
We
Token We
Feature activation+3.227
'll
Token'll
Feature activation+3.865
see
Token see
Feature activation+3.790
how
Token how
Feature activation+0.482
it
Token it
Feature activation+0.000
maybe
Token maybe
Feature activation+0.000
it
Token it
Feature activation+0.418
was
Token was
Feature activation+0.000
genius
Token genius
Feature activation+0.000
.
Token.
Feature activation+3.500
We
Token We
Feature activation+3.227
'll
Token'll
Feature activation+3.865
see
Token see
Feature activation+3.790
how
Token how
Feature activation+0.482
it
Token it
Feature activation+0.000
plays
Token plays
Feature activation+0.647
or
Token or
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
disorderly
Token disorderly
Feature activation+0.000
one
Token one
Feature activation+0.000
,
Token,
Feature activation+3.089
for
Token for
Feature activation+2.316
that
Token that
Feature activation+1.920
matter
Token matter
Feature activation+0.234
âĢĶ
Token âĢĶ
Feature activation+0.049
some
Token some
Feature activation+0.000
I
Token I
Feature activation+1.567
don
Token don
Feature activation+1.921
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.151
know
Token know
Feature activation+3.078
,
Token,
Feature activation+0.864
it
Token it
Feature activation+0.809
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
hundred
Token hundred
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
was
Token was
Feature activation+0.000
sink
Token sink
Feature activation+0.000
or
Token or
Feature activation+0.000
swim
Token swim
Feature activation+2.642
before
Token before
Feature activation+0.000
Captain
Token Captain
Feature activation+0.000
America
Token America
Feature activation+0.000
and
Token and
Feature activation+0.000
then
Token then
Feature activation+0.000
change
Token change
Feature activation+0.000
and
Token and
Feature activation+0.000
exploding
Token exploding
Feature activation+0.000
oil
Token oil
Feature activation+0.000
trains
Token trains
Feature activation+0.000
.
Token.
Feature activation+2.494
She
Token She
Feature activation+1.608
chose
Token chose
Feature activation+0.349
to
Token to
Feature activation+0.000
resist
Token resist
Feature activation+0.000
because
Token because
Feature activation+0.000
gamble
Token gamble
Feature activation+0.000
,
Token,
Feature activation+0.000
rather
Token rather
Feature activation+0.000
or
Token or
Feature activation+0.000
not
Token not
Feature activation+0.000
you
Token you
Feature activation+2.453
getting
Token getting
Feature activation+0.000
something
Token something
Feature activation+0.000
good
Token good
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
life
Token life
Feature activation+0.000
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.151
,
Token,
Feature activation+1.314
perhaps
Token perhaps
Feature activation+0.170
more
Token more
Feature activation+2.401
likely
Token likely
Feature activation+0.208
,
Token,
Feature activation+0.000
they
Token they
Feature activation+0.000
simply
Token simply
Feature activation+0.000
don
Token don
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
)
Token)
Feature activation+0.000
or
Token or
Feature activation+0.000
snow
Token snow
Feature activation+0.000
monsters
Token monsters
Feature activation+0.000
,
Token,
Feature activation+2.377
for
Token for
Feature activation+1.379
the
Token the
Feature activation+0.148
way
Token way
Feature activation+0.000
the
Token the
Feature activation+0.000
cl
Token cl
Feature activation+0.000
the
Token the
Feature activation+0.000
entire
Token entire
Feature activation+0.000
exchange
Token exchange
Feature activation+0.000
was
Token was
Feature activation+0.000
compromised
Token compromised
Feature activation+0.000
.
Token.
Feature activation+2.343
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
It
TokenIt
Feature activation+0.230
could
Token could
Feature activation+0.123
get
Token get
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
disorderly
Token disorderly
Feature activation+0.000
one
Token one
Feature activation+0.000
,
Token,
Feature activation+3.089
for
Token for
Feature activation+2.316
that
Token that
Feature activation+1.920
matter
Token matter
Feature activation+0.234
âĢĶ
Token âĢĶ
Feature activation+0.049
some
Token some
Feature activation+0.000
few
Token few
Feature activation+0.000
games
Token games
Feature activation+0.000
in
Token in
Feature activation+0.000
stock
Token stock
Feature activation+0.000
that
Token that
Feature activation+0.000
day
Token day
Feature activation+0.000
.
Token.
Feature activation+2.303
I
Token I
Feature activation+1.567
don
Token don
Feature activation+1.921
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.151
sideways
Token sideways
Feature activation+0.000
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.206
backward
Token backward
Feature activation+0.027
(
Token (
Feature activation+0.941
or
Tokenor
Feature activation+2.292
if
Token if
Feature activation+0.125
we
Token we
Feature activation+0.149
really
Token really
Feature activation+0.032
even
Token even
Feature activation+0.000
care
Token care
Feature activation+0.415
felt
Token felt
Feature activation+0.000
she
Token she
Feature activation+0.000
deserved
Token deserved
Feature activation+0.000
it
Token it
Feature activation+0.000
or
Token or
Feature activation+0.000
not
Token not
Feature activation+2.112
,
Token,
Feature activation+0.390
we
Token we
Feature activation+0.506
can
Token can
Feature activation+0.693
all
Token all
Feature activation+0.370
agree
Token agree
Feature activation+0.000
"
Token"
Feature activation+0.000
to
Token to
Feature activation+0.000
exagger
Token exagger
Feature activation+0.000
ate
Tokenate
Feature activation+0.102
contact
Token contact
Feature activation+0.000
.
Token.
Feature activation+2.088
Or
Token Or
Feature activation+0.153
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.548
can
Token can
Feature activation+0.000
say
Token say
Feature activation+0.246
superhero
Token superhero
Feature activation+0.000
team
Token team
Feature activation+0.000
-
Token-
Feature activation+0.000
up
Tokenup
Feature activation+0.000
film
Token film
Feature activation+0.000
.
Token.
Feature activation+1.976
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.002
Indeed
TokenIndeed
Feature activation+0.000
,
Token,
Feature activation+0.000
while
Token while
Feature activation+0.000
the
Token the
Feature activation+0.000
person
Token person
Feature activation+0.000
who
Token who
Feature activation+0.000
framed
Token framed
Feature activation+0.000
you
Token you
Feature activation+0.000
.
Token.
Feature activation+1.976
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Col
TokenCol
Feature activation+0.000
anton
Tokenanton
Feature activation+0.000
io
Tokenio
Feature activation+0.000
stock
Token stock
Feature activation+0.000
that
Token that
Feature activation+0.000
day
Token day
Feature activation+0.000
.
Token.
Feature activation+2.303
I
Token I
Feature activation+1.567
don
Token don
Feature activation+1.921
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.151
know
Token know
Feature activation+3.078
,
Token,
Feature activation+0.864

Top DFA by src position
MAX = 3.246

appeal
Token appeal
Feature activation-0.003
Top resid features:
for
Token for
Feature activation+0.002
Top resid features:
unity
Token unity
Feature activation-0.002
Top resid features:
.
Token.
Feature activation+0.042
Top resid features:
Or
Token Or
Feature activation+0.993
Top resid features:
maybe
Token maybe
Feature activation+2.245
Top resid features:
it
Token it
Feature activation+0.494
Top resid features:
was
Token was
Feature activation+0.375
Top resid features:
genius
Token genius
Feature activation+0.012
Top resid features:
.
Token.
Feature activation+0.149
Top resid features:
We
Token We
Feature activation-0.002
Top resid features:
appeal
Token appeal
Feature activation-0.003
Top resid features:
for
Token for
Feature activation+0.001
Top resid features:
unity
Token unity
Feature activation+0.003
Top resid features:
.
Token.
Feature activation+0.015
Top resid features:
Or
Token Or
Feature activation+0.751
Top resid features:
maybe
Token maybe
Feature activation+2.425
Top resid features:
it
Token it
Feature activation+0.483
Top resid features:
was
Token was
Feature activation+0.514
Top resid features:
genius
Token genius
Feature activation+0.012
Top resid features:
.
Token.
Feature activation+0.098
Top resid features:
We
Token We
Feature activation-0.001
Top resid features:
appeal
Token appeal
Feature activation-0.004
Top resid features:
for
Token for
Feature activation-0.002
Top resid features:
unity
Token unity
Feature activation-0.007
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
Or
Token Or
Feature activation+1.017
Top resid features:
maybe
Token maybe
Feature activation+2.090
Top resid features:
it
Token it
Feature activation+0.515
Top resid features:
was
Token was
Feature activation+0.289
Top resid features:
genius
Token genius
Feature activation-0.003
Top resid features:
.
Token.
Feature activation+0.106
Top resid features:
We
Token We
Feature activation+0.000
Top resid features:
appeal
Token appeal
Feature activation-0.011
Top resid features:
for
Token for
Feature activation-0.002
Top resid features:
unity
Token unity
Feature activation-0.008
Top resid features:
.
Token.
Feature activation+0.057
Top resid features:
Or
Token Or
Feature activation+1.209
Top resid features:
maybe
Token maybe
Feature activation+1.680
Top resid features:
it
Token it
Feature activation+0.268
Top resid features:
was
Token was
Feature activation+0.282
Top resid features:
genius
Token genius
Feature activation+0.010
Top resid features:
.
Token.
Feature activation+0.207
Top resid features:
We
Token We
Feature activation-0.000
Top resid features:
of
Token of
Feature activation+0.001
Top resid features:
that
Token that
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.015
Top resid features:
For
Token For
Feature activation-0.002
Top resid features:
better
Token better
Feature activation+0.008
Top resid features:
or
Token or
Feature activation+1.629
Top resid features:
worse
Token worse
Feature activation+0.251
Top resid features:
,
Token,
Feature activation+0.031
Top resid features:
he
Token he
Feature activation-0.008
Top resid features:
takes
Token takes
Feature activation-0.014
Top resid features:
it
Token it
Feature activation-0.001
Top resid features:
resist
Token resist
Feature activation-0.006
Top resid features:
this
Token this
Feature activation-0.009
Top resid features:
one
Token one
Feature activation-0.003
Top resid features:
.
Token.
Feature activation-0.013
Top resid features:
Or
Token Or
Feature activation+0.437
Top resid features:
maybe
Token maybe
Feature activation+2.906
Top resid features:
the
Token the
Feature activation+0.157
Top resid features:
store
Token store
Feature activation-0.003
Top resid features:
just
Token just
Feature activation+0.180
Top resid features:
didn
Token didn
Feature activation+0.032
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
three
Token three
Feature activation-0.004
Top resid features:
,
Token,
Feature activation+0.006
Top resid features:
it
Token it
Feature activation+0.006
Top resid features:
was
Token was
Feature activation+0.020
Top resid features:
sink
Token sink
Feature activation-0.015
Top resid features:
or
Token or
Feature activation+3.246
Top resid features:
swim
Token swim
Feature activation+0.055
Top resid features:
before
Token before
Feature activation+0.000
Top resid features:
Captain
Token Captain
Feature activation+0.000
Top resid features:
America
Token America
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
jeopard
Token jeopard
Feature activation-0.001
Top resid features:
ized
Tokenized
Feature activation+0.013
Top resid features:
by
Token by
Feature activation+0.007
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
police
Token police
Feature activation-0.007
Top resid features:
or
Token or
Feature activation+2.315
Top resid features:
by
Token by
Feature activation+0.171
Top resid features:
catastrophic
Token catastrophic
Feature activation+0.007
Top resid features:
climate
Token climate
Feature activation+0.006
Top resid features:
change
Token change
Feature activation+0.008
Top resid features:
and
Token and
Feature activation+0.067
Top resid features:
's
Token's
Feature activation-0.007
Top resid features:
a
Token a
Feature activation-0.002
Top resid features:
gamble
Token gamble
Feature activation-0.022
Top resid features:
,
Token,
Feature activation-0.004
Top resid features:
rather
Token rather
Feature activation+0.046
Top resid features:
or
Token or
Feature activation+2.259
Top resid features:
not
Token not
Feature activation+0.729
Top resid features:
you
Token you
Feature activation+0.038
Top resid features:
getting
Token getting
Feature activation+0.000
Top resid features:
something
Token something
Feature activation+0.000
Top resid features:
good
Token good
Feature activation+0.000
Top resid features:
essential
Token essential
Feature activation-0.004
Top resid features:
part
Token part
Feature activation+0.008
Top resid features:
of
Token of
Feature activation+0.005
Top resid features:
life
Token life
Feature activation-0.002
Top resid features:
,
Token,
Feature activation+0.022
Top resid features:
or
Token or
Feature activation+1.524
Top resid features:
,
Token,
Feature activation+0.570
Top resid features:
perhaps
Token perhaps
Feature activation+0.534
Top resid features:
more
Token more
Feature activation+0.011
Top resid features:
likely
Token likely
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
ice
Tokenice
Feature activation+0.008
Top resid features:
trees
Token trees
Feature activation+0.006
Top resid features:
âĢ
TokenâĢ
Feature activation+0.013
Top resid features:
Ŀ
TokenĿ
Feature activation+0.028
Top resid features:
)
Token)
Feature activation+0.010
Top resid features:
or
Token or
Feature activation+2.234
Top resid features:
snow
Token snow
Feature activation+0.002
Top resid features:
monsters
Token monsters
Feature activation+0.032
Top resid features:
,
Token,
Feature activation+0.036
Top resid features:
for
Token for
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
affected
Token affected
Feature activation+0.001
Top resid features:
your
Token your
Feature activation-0.024
Top resid features:
account
Token account
Feature activation+0.005
Top resid features:
,
Token,
Feature activation+0.029
Top resid features:
or
Token or
Feature activation+0.489
Top resid features:
maybe
Token maybe
Feature activation+1.962
Top resid features:
the
Token the
Feature activation+0.042
Top resid features:
entire
Token entire
Feature activation+0.007
Top resid features:
exchange
Token exchange
Feature activation-0.010
Top resid features:
was
Token was
Feature activation+0.007
Top resid features:
compromised
Token compromised
Feature activation+0.003
Top resid features:
well
Token well
Feature activation+0.036
Top resid features:
-
Token-
Feature activation+0.014
Top resid features:
ordered
Tokenordered
Feature activation+0.002
Top resid features:
state
Token state
Feature activation+0.008
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.124
Top resid features:
or
Token or
Feature activation+1.413
Top resid features:
in
Token in
Feature activation+0.086
Top resid features:
a
Token a
Feature activation+0.054
Top resid features:
disorderly
Token disorderly
Feature activation+0.020
Top resid features:
one
Token one
Feature activation+0.096
Top resid features:
,
Token,
Feature activation+0.328
Top resid features:
resist
Token resist
Feature activation-0.009
Top resid features:
this
Token this
Feature activation-0.003
Top resid features:
one
Token one
Feature activation+0.012
Top resid features:
.
Token.
Feature activation+0.042
Top resid features:
Or
Token Or
Feature activation+0.530
Top resid features:
maybe
Token maybe
Feature activation+2.069
Top resid features:
the
Token the
Feature activation+0.155
Top resid features:
store
Token store
Feature activation-0.015
Top resid features:
just
Token just
Feature activation+0.083
Top resid features:
didn
Token didn
Feature activation+0.009
Top resid features:
âĢ
TokenâĢ
Feature activation-0.001
Top resid features:
step
Token step
Feature activation-0.001
Top resid features:
forward
Token forward
Feature activation+0.059
Top resid features:
,
Token,
Feature activation+0.092
Top resid features:
sideways
Token sideways
Feature activation+0.064
Top resid features:
,
Token,
Feature activation+0.152
Top resid features:
or
Token or
Feature activation+1.672
Top resid features:
backward
Token backward
Feature activation+0.127
Top resid features:
(
Token (
Feature activation+0.156
Top resid features:
or
Tokenor
Feature activation+0.540
Top resid features:
if
Token if
Feature activation+0.000
Top resid features:
we
Token we
Feature activation+0.000
Top resid features:
you
Token you
Feature activation+0.035
Top resid features:
felt
Token felt
Feature activation+0.365
Top resid features:
she
Token she
Feature activation+0.069
Top resid features:
deserved
Token deserved
Feature activation+0.063
Top resid features:
it
Token it
Feature activation+0.042
Top resid features:
or
Token or
Feature activation+1.651
Top resid features:
not
Token not
Feature activation+0.115
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
we
Token we
Feature activation+0.000
Top resid features:
can
Token can
Feature activation+0.000
Top resid features:
all
Token all
Feature activation+0.000
Top resid features:
his
Token his
Feature activation-0.000
Top resid features:
head
Token head
Feature activation-0.001
Top resid features:
back
Token back
Feature activation+0.003
Top resid features:
âĢ
TokenâĢ
Feature activation+0.008
Top resid features:
Ŀ
TokenĿ
Feature activation+0.043
Top resid features:
or
Token or
Feature activation+2.494
Top resid features:
"
Token "
Feature activation+0.043
Top resid features:
fl
Tokenfl
Feature activation+0.002
Top resid features:
ails
Tokenails
Feature activation+0.008
Top resid features:
about
Token about
Feature activation+0.000
Top resid features:
"
Token"
Feature activation+0.052
Top resid features:
s
Tokens
Feature activation+0.034
Top resid features:
own
Token own
Feature activation+0.019
Top resid features:
drawings
Token drawings
Feature activation+0.003
Top resid features:
.
Token.
Feature activation-0.012
Top resid features:
Or
Token Or
Feature activation+0.435
Top resid features:
maybe
Token maybe
Feature activation+1.298
Top resid features:
it
Token it
Feature activation+0.300
Top resid features:
âĢ
TokenâĢ
Feature activation-0.018
Top resid features:
Ļ
TokenĻ
Feature activation-0.002
Top resid features:
s
Tokens
Feature activation+0.112
Top resid features:
simply
Token simply
Feature activation+0.187
Top resid features:
to
Token to
Feature activation+0.003
Top resid features:
its
Token its
Feature activation-0.001
Top resid features:
former
Token former
Feature activation-0.000
Top resid features:
state
Token state
Feature activation+0.005
Top resid features:
,
Token,
Feature activation+0.030
Top resid features:
or
Token or
Feature activation+1.657
Top resid features:
to
Token to
Feature activation+0.694
Top resid features:
get
Token get
Feature activation+0.043
Top resid features:
revenge
Token revenge
Feature activation+0.023
Top resid features:
on
Token on
Feature activation+0.021
Top resid features:
the
Token the
Feature activation+0.005
Top resid features:
resist
Token resist
Feature activation-0.012
Top resid features:
this
Token this
Feature activation-0.006
Top resid features:
one
Token one
Feature activation-0.002
Top resid features:
.
Token.
Feature activation-0.006
Top resid features:
Or
Token Or
Feature activation+0.500
Top resid features:
maybe
Token maybe
Feature activation+1.845
Top resid features:
the
Token the
Feature activation+0.085
Top resid features:
store
Token store
Feature activation-0.002
Top resid features:
just
Token just
Feature activation+0.045
Top resid features:
didn
Token didn
Feature activation+0.013
Top resid features:
âĢ
TokenâĢ
Feature activation-0.004
Top resid features:

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.06

Head 2: 0.05

Head 3: 0.06

Head 4: 0.04

Head 5: 0.05

Head 6: 0.04

Head 7: 0.03

Head 8: 0.09

Head 9: 0.43

Head 10: 0.11

Head 11: 0.03

Positive logits

depending5.95

whichever5.74

depending5.11

ichever4.88

Either4.45

depends4.28

Either4.26

Regardless3.74

undecided3.69

Whatever3.68

Whatever3.65

dunno3.59

Regardless3.40

Depending3.25

chooses3.21

Unsure3.18

determines3.18

preference3.17

varies3.15

Depending3.15

Negative logits

SourceFile-2.33

rehensive-2.27

impressive-2.22

abol-2.22

anded-2.20

deliveries-2.15

introdu-2.14

Assembly-2.14

Completed-2.14

itan-2.13

predecessor-2.12

inet-2.12

eva-2.04

aed-2.04

emn-2.04

NCT-2.04

incorporated-2.03

ibi-2.02

rote-2.02

accomplished-2.02

INTERVAL 3.479 - 3.865
CONTAINS 0.000%

Or
Token Or
Feature activation+0.000
maybe
Token maybe
Feature activation+0.000
it
Token it
Feature activation+0.418
was
Token was
Feature activation+0.000
genius
Token genius
Feature activation+0.000
.
Token.
Feature activation+3.500
We
Token We
Feature activation+3.227
'll
Token'll
Feature activation+3.865
see
Token see
Feature activation+3.790
how
Token how
Feature activation+0.482
it
Token it
Feature activation+0.000
it
Token it
Feature activation+0.418
was
Token was
Feature activation+0.000
genius
Token genius
Feature activation+0.000
.
Token.
Feature activation+3.500
We
Token We
Feature activation+3.227
'll
Token'll
Feature activation+3.865
see
Token see
Feature activation+3.790
how
Token how
Feature activation+0.482
it
Token it
Feature activation+0.000
plays
Token plays
Feature activation+0.647
going
Token going
Feature activation+0.000
was
Token was
Feature activation+0.000
genius
Token genius
Feature activation+0.000
.
Token.
Feature activation+3.500
We
Token We
Feature activation+3.227
'll
Token'll
Feature activation+3.865
see
Token see
Feature activation+3.790
how
Token how
Feature activation+0.482
it
Token it
Feature activation+0.000
plays
Token plays
Feature activation+0.647
going
Token going
Feature activation+0.000
forward
Token forward
Feature activation+0.822

INTERVAL 3.092 - 3.479
CONTAINS 0.000%

maybe
Token maybe
Feature activation+0.000
it
Token it
Feature activation+0.418
was
Token was
Feature activation+0.000
genius
Token genius
Feature activation+0.000
.
Token.
Feature activation+3.500
We
Token We
Feature activation+3.227
'll
Token'll
Feature activation+3.865
see
Token see
Feature activation+3.790
how
Token how
Feature activation+0.482
it
Token it
Feature activation+0.000
plays
Token plays
Feature activation+0.647

INTERVAL 2.706 - 3.092
CONTAINS 0.000%

or
Token or
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
disorderly
Token disorderly
Feature activation+0.000
one
Token one
Feature activation+0.000
,
Token,
Feature activation+3.089
for
Token for
Feature activation+2.316
that
Token that
Feature activation+1.920
matter
Token matter
Feature activation+0.234
âĢĶ
Token âĢĶ
Feature activation+0.049
some
Token some
Feature activation+0.000
I
Token I
Feature activation+1.567
don
Token don
Feature activation+1.921
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.151
know
Token know
Feature activation+3.078
,
Token,
Feature activation+0.864
it
Token it
Feature activation+0.809
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
hundred
Token hundred
Feature activation+0.000

INTERVAL 2.319 - 2.706
CONTAINS 0.001%

life
Token life
Feature activation+0.000
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.151
,
Token,
Feature activation+1.314
perhaps
Token perhaps
Feature activation+0.170
more
Token more
Feature activation+2.401
likely
Token likely
Feature activation+0.208
,
Token,
Feature activation+0.000
they
Token they
Feature activation+0.000
simply
Token simply
Feature activation+0.000
don
Token don
Feature activation+0.000
gamble
Token gamble
Feature activation+0.000
,
Token,
Feature activation+0.000
rather
Token rather
Feature activation+0.000
or
Token or
Feature activation+0.000
not
Token not
Feature activation+0.000
you
Token you
Feature activation+2.453
getting
Token getting
Feature activation+0.000
something
Token something
Feature activation+0.000
good
Token good
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
was
Token was
Feature activation+0.000
sink
Token sink
Feature activation+0.000
or
Token or
Feature activation+0.000
swim
Token swim
Feature activation+2.642
before
Token before
Feature activation+0.000
Captain
Token Captain
Feature activation+0.000
America
Token America
Feature activation+0.000
and
Token and
Feature activation+0.000
then
Token then
Feature activation+0.000
change
Token change
Feature activation+0.000
and
Token and
Feature activation+0.000
exploding
Token exploding
Feature activation+0.000
oil
Token oil
Feature activation+0.000
trains
Token trains
Feature activation+0.000
.
Token.
Feature activation+2.494
She
Token She
Feature activation+1.608
chose
Token chose
Feature activation+0.349
to
Token to
Feature activation+0.000
resist
Token resist
Feature activation+0.000
because
Token because
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
)
Token)
Feature activation+0.000
or
Token or
Feature activation+0.000
snow
Token snow
Feature activation+0.000
monsters
Token monsters
Feature activation+0.000
,
Token,
Feature activation+2.377
for
Token for
Feature activation+1.379
the
Token the
Feature activation+0.148
way
Token way
Feature activation+0.000
the
Token the
Feature activation+0.000
cl
Token cl
Feature activation+0.000

INTERVAL 1.933 - 2.319
CONTAINS 0.001%

the
Token the
Feature activation+0.000
person
Token person
Feature activation+0.000
who
Token who
Feature activation+0.000
framed
Token framed
Feature activation+0.000
you
Token you
Feature activation+0.000
.
Token.
Feature activation+1.976
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Col
TokenCol
Feature activation+0.000
anton
Tokenanton
Feature activation+0.000
io
Tokenio
Feature activation+0.000
felt
Token felt
Feature activation+0.000
she
Token she
Feature activation+0.000
deserved
Token deserved
Feature activation+0.000
it
Token it
Feature activation+0.000
or
Token or
Feature activation+0.000
not
Token not
Feature activation+2.112
,
Token,
Feature activation+0.390
we
Token we
Feature activation+0.506
can
Token can
Feature activation+0.693
all
Token all
Feature activation+0.370
agree
Token agree
Feature activation+0.000
games
Token games
Feature activation+0.000
in
Token in
Feature activation+0.000
stock
Token stock
Feature activation+0.000
that
Token that
Feature activation+0.000
day
Token day
Feature activation+0.000
.
Token.
Feature activation+2.303
I
Token I
Feature activation+1.567
don
Token don
Feature activation+1.921
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.151
sideways
Token sideways
Feature activation+0.000
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.206
backward
Token backward
Feature activation+0.027
(
Token (
Feature activation+0.941
or
Tokenor
Feature activation+2.292
if
Token if
Feature activation+0.125
we
Token we
Feature activation+0.149
really
Token really
Feature activation+0.032
even
Token even
Feature activation+0.000
care
Token care
Feature activation+0.415
superhero
Token superhero
Feature activation+0.000
team
Token team
Feature activation+0.000
-
Token-
Feature activation+0.000
up
Tokenup
Feature activation+0.000
film
Token film
Feature activation+0.000
.
Token.
Feature activation+1.976
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.002
Indeed
TokenIndeed
Feature activation+0.000
,
Token,
Feature activation+0.000
while
Token while
Feature activation+0.000

INTERVAL 1.546 - 1.933
CONTAINS 0.003%

,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.000
Buzz
Token Buzz
Feature activation+0.000
Killing
Token Killing
Feature activation+0.000
ton
Tokenton
Feature activation+0.000
:
Token:
Feature activation+1.665
Oh
Token Oh
Feature activation+0.000
t
Tokent
Feature activation+0.000
ani
Tokenani
Feature activation+0.000
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
and
Token and
Feature activation+0.000
apps
Token apps
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.491
or
Token or
Feature activation+0.224
maybe
Token maybe
Feature activation+0.031
more
Token more
Feature activation+1.776
directly
Token directly
Feature activation+0.106
through
Token through
Feature activation+0.000
athletics
Token athletics
Feature activation+0.000
.
Token.
Feature activation+0.680
Regardless
Token Regardless
Feature activation+1.680
child
Token child
Feature activation+0.000
or
Token or
Feature activation+0.000
leave
Token leave
Feature activation+0.000
Canada
Token Canada
Feature activation+0.000
.
Token.
Feature activation+0.000
Neither
Token Neither
Feature activation+1.703
of
Token of
Feature activation+0.660
which
Token which
Feature activation+0.634
is
Token is
Feature activation+0.406
a
Token a
Feature activation+0.276
good
Token good
Feature activation+0.000
,
Token,
Feature activation+0.000
CD
Token CD
Feature activation+0.000
quality
Token quality
Feature activation+0.000
or
Token or
Feature activation+0.082
higher
Token higher
Feature activation+0.230
,
Token,
Feature activation+1.718
the
Token the
Feature activation+0.891
way
Token way
Feature activation+0.000
the
Token the
Feature activation+0.000
artist
Token artist
Feature activation+0.000
made
Token made
Feature activation+0.000
cover
Token cover
Feature activation+0.000
for
Token for
Feature activation+0.000
prejudice
Token prejudice
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.000
or
Tokenor
Feature activation+0.000
at
Token at
Feature activation+1.759
least
Token least
Feature activation+0.000
disregard
Token disregard
Feature activation+0.000
for
Token for
Feature activation+0.000
people
Token people
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 1.160 - 1.546
CONTAINS 0.004%

true
Tokentrue
Feature activation+0.000
)
Token)
Feature activation+0.000
or
Token or
Feature activation+0.000
not
Token not
Feature activation+0.000
(
Token (
Feature activation+0.000
false
Tokenfalse
Feature activation+1.300
).
Token).
Feature activation+0.000
We
Token We
Feature activation+0.000
start
Token start
Feature activation+0.000
off
Token off
Feature activation+0.000
with
Token with
Feature activation+0.000
part
Token part
Feature activation+0.000
of
Token of
Feature activation+0.000
life
Token life
Feature activation+0.000
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.151
,
Token,
Feature activation+1.314
perhaps
Token perhaps
Feature activation+0.170
more
Token more
Feature activation+2.401
likely
Token likely
Feature activation+0.208
,
Token,
Feature activation+0.000
they
Token they
Feature activation+0.000
beings
Token beings
Feature activation+0.000
or
Token or
Feature activation+0.000
long
Token long
Feature activation+0.000
term
Token term
Feature activation+0.000
vision
Token vision
Feature activation+0.000
,
Token,
Feature activation+1.219
but
Token but
Feature activation+0.716
I
Token I
Feature activation+0.000
find
Token find
Feature activation+0.000
that
Token that
Feature activation+0.000
especially
Token especially
Feature activation+0.000
Marvel
Token Marvel
Feature activation+0.000
ous
Tokenous
Feature activation+0.000
Cou
Token Cou
Feature activation+0.000
der
Tokender
Feature activation+0.000
c
Tokenc
Feature activation+0.000
.
Token.
Feature activation+1.178
Wikipedia
Token Wikipedia
Feature activation+0.000
says
Token says
Feature activation+0.000
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
)
Token)
Feature activation+0.000
or
Token or
Feature activation+0.000
snow
Token snow
Feature activation+0.000
monsters
Token monsters
Feature activation+0.000
,
Token,
Feature activation+2.377
for
Token for
Feature activation+1.379
the
Token the
Feature activation+0.148
way
Token way
Feature activation+0.000
the
Token the
Feature activation+0.000
cl
Token cl
Feature activation+0.000
umps
Tokenumps
Feature activation+0.000

INTERVAL 0.773 - 1.160
CONTAINS 0.011%

don
Token don
Feature activation+1.921
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.151
know
Token know
Feature activation+3.078
,
Token,
Feature activation+0.864
it
Token it
Feature activation+0.809
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
hundred
Token hundred
Feature activation+0.000
years
Token years
Feature activation+0.000
soccer
Token soccer
Feature activation+0.000
........
Token ........
Feature activation+0.000
or
Tokenor
Feature activation+0.000
is
Token is
Feature activation+0.000
it
Token it
Feature activation+0.000
?
Token?
Feature activation+0.859
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.019
Last
TokenLast
Feature activation+0.000
week
Token week
Feature activation+0.000
there
Token there
Feature activation+0.000
indication
Token indication
Feature activation+0.000
of
Token of
Feature activation+0.000
its
Token its
Feature activation+0.000
wide
Token wide
Feature activation+0.000
release
Token release
Feature activation+0.000
is
Token is
Feature activation+1.005
yet
Token yet
Feature activation+0.793
to
Token to
Feature activation+0.523
be
Token be
Feature activation+0.528
seen
Token seen
Feature activation+0.579
.
Token.
Feature activation+0.000
of
Token of
Feature activation+0.000
its
Token its
Feature activation+0.000
wide
Token wide
Feature activation+0.000
release
Token release
Feature activation+0.000
is
Token is
Feature activation+1.005
yet
Token yet
Feature activation+0.793
to
Token to
Feature activation+0.523
be
Token be
Feature activation+0.528
seen
Token seen
Feature activation+0.579
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
to
Token to
Feature activation+0.000
pull
Token pull
Feature activation+0.000
last
Token last
Feature activation+0.000
second
Token second
Feature activation+0.000
maneuvers
Token maneuvers
Feature activation+0.000
,
Token,
Feature activation+0.795
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.351
the
Token the
Feature activation+0.388
post
Token post
Feature activation+0.000
read
Token read
Feature activation+0.000

INTERVAL 0.387 - 0.773
CONTAINS 0.031%

then
Token then
Feature activation+0.000
it
Token it
Feature activation+0.000
represents
Token represents
Feature activation+0.000
a
Token a
Feature activation+0.000
1
Token 1
Feature activation+0.000
.
Token.
Feature activation+0.412
As
Token As
Feature activation+0.000
we
Token we
Feature activation+0.000
know
Token know
Feature activation+0.000
the
Token the
Feature activation+0.000
order
Token order
Feature activation+0.000
,
Token,
Feature activation+0.412
or
Token or
Feature activation+0.417
for
Token for
Feature activation+0.270
any
Token any
Feature activation+0.801
other
Token other
Feature activation+0.909
reason
Token reason
Feature activation+0.521
,
Token,
Feature activation+1.068
PLEASE
Token PLEASE
Feature activation+0.310
let
Token let
Feature activation+0.540
me
Token me
Feature activation+0.000
know
Token know
Feature activation+0.437
in
Token in
Feature activation+0.000
its
Token its
Feature activation+0.000
designated
Token designated
Feature activation+0.000
orbit
Token orbit
Feature activation+0.000
-
Token -
Feature activation+0.775
this
Token this
Feature activation+0.620
is
Token is
Feature activation+0.139
called
Token called
Feature activation+0.000
orbit
Token orbit
Feature activation+0.000
control
Token control
Feature activation+0.000
.
Token.
Feature activation+0.000
31
Token 31
Feature activation+0.000
or
Token or
Feature activation+0.000
any
Token any
Feature activation+0.025
time
Token time
Feature activation+0.000
since
Token since
Feature activation+0.000
,
Token,
Feature activation+0.750
is
Token is
Feature activation+0.313
asked
Token asked
Feature activation+0.000
to
Token to
Feature activation+0.000
contact
Token contact
Feature activation+0.000
the
Token the
Feature activation+0.000
certain
Token certain
Feature activation+0.000
system
Token system
Feature activation+0.000
of
Token of
Feature activation+0.000
belief
Token belief
Feature activation+0.000
does
Token does
Feature activation+0.397
not
Token not
Feature activation+0.692
represent
Token represent
Feature activation+0.000
the
Token the
Feature activation+0.000
essence
Token essence
Feature activation+0.000
and
Token and
Feature activation+0.000
purpose
Token purpose
Feature activation+0.000

INTERVAL 0.000 - 0.387
CONTAINS 99.949%

IA
TokenIA
Feature activation+0.000
)
Token)
Feature activation+0.000
should
Token should
Feature activation+0.000
register
Token register
Feature activation+0.000
FIR
Token FIR
Feature activation+0.000
s
Tokens
Feature activation+0.000
against
Token against
Feature activation+0.000
the
Token the
Feature activation+0.000
sl
Token sl
Feature activation+0.000
um
Tokenum
Feature activation+0.000
dwell
Token dwell
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
(
Token(
Feature activation+0.000
Vis
TokenVis
Feature activation+0.000
ited
Tokenited
Feature activation+0.000
1
Token 1
Feature activation+0.000
,
Token,
Feature activation+0.000
170
Token170
Feature activation+0.000
times
Token times
Feature activation+0.000
,
Token,
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
C
TokenC
Feature activation+0.000
umm
Tokenumm
Feature activation+0.000
ings
Tokenings
Feature activation+0.000
says
Token says
Feature activation+0.000
he
Token he
Feature activation+0.000
is
Token is
Feature activation+0.000
contacting
Token contacting
Feature activation+0.000
organizations
Token organizations
Feature activation+0.000
independently
Token independently
Feature activation+0.000
to
Token to
Feature activation+0.000
7
Token7
Feature activation+0.000
and
Token and
Feature activation+0.000
A
Token A
Feature activation+0.000
38
Token38
Feature activation+0.000
meet
Token meet
Feature activation+0.000
here
Token here
Feature activation+0.000
).
Token).
Feature activation+0.000
Yes
Token Yes
Feature activation+0.000
,
Token,
Feature activation+0.000
Google
Token Google
Feature activation+0.000
Maps
Token Maps
Feature activation+0.000
flag
Token flag
Feature activation+0.000
to
Token to
Feature activation+0.000
when
Token when
Feature activation+0.000
something
Token something
Feature activation+0.000
important
Token important
Feature activation+0.000
is
Token is
Feature activation+0.000
going
Token going
Feature activation+0.000
on
Token on
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
player
Token player
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 4 in H8.9: (feature 15005)

TOP ACTIVATIONS
MAX = 9.770

to
Token to
Feature activation+0.000
identify
Token identify
Feature activation+0.346
the
Token the
Feature activation+0.101
policeman
Token policeman
Feature activation+0.049
responsible
Token responsible
Feature activation+0.000
for
Token for
Feature activation+9.770
shooting
Token shooting
Feature activation+0.301
and
Token and
Feature activation+0.028
injuring
Token injuring
Feature activation+0.000
Mah
Token Mah
Feature activation+0.000
am
Tokenam
Feature activation+0.000
arrest
Token arrest
Feature activation+1.106
and
Token and
Feature activation+0.000
weapons
Token weapons
Feature activation+0.000
violations
Token violations
Feature activation+0.000
charges
Token charges
Feature activation+0.082
.
Token.
Feature activation+9.570
Image
Token Image
Feature activation+0.334
8
Token 8
Feature activation+0.000
of
Token of
Feature activation+0.000
9
Token 9
Feature activation+3.467
Paul
Token Paul
Feature activation+0.000
@
Token @
Feature activation+1.046
ed
Tokened
Feature activation+0.000
ward
Tokenward
Feature activation+0.000
ian
Tokenian
Feature activation+0.035
_
Token_
Feature activation+0.000
era
Tokenera
Feature activation+6.925
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
UPDATE
TokenUPDATE
Feature activation+0.000
:
Token:
Feature activation+0.000
'
Token '
Feature activation+0.000
Black
TokenBlack
Feature activation+0.000
defeated
Token defeated
Feature activation+0.194
Greenwich
Token Greenwich
Feature activation+2.334
6
Token 6
Feature activation+0.000
-
Token-
Feature activation+0.000
5
Token5
Feature activation+0.930
.
Token.
Feature activation+6.615
Photo
Token Photo
Feature activation+0.229
:
Token:
Feature activation+0.102
Bob
Token Bob
Feature activation+0.000
Luc
Token Luc
Feature activation+0.000
key
Tokenkey
Feature activation+0.000
Activity
TokenActivity
Feature activation+0.000
:
Token:
Feature activation+0.000
448
Token 448
Feature activation+0.214
Merit
TokenMerit
Feature activation+0.000
:
Token:
Feature activation+0.000
250
Token 250
Feature activation+6.085
Re
Token Re
Feature activation+0.000
:
Token:
Feature activation+0.000
[
Token [
Feature activation+0.000
ANN
TokenANN
Feature activation+0.000
][
Token][
Feature activation+0.000
land
Tokenland
Feature activation+0.000
across
Token across
Feature activation+0.000
from
Token from
Feature activation+0.867
Dow
Token Dow
Feature activation+0.311
Diamond
Token Diamond
Feature activation+0.038
.
Token.
Feature activation+5.737
Photo
Token Photo
Feature activation+1.837
:
Token:
Feature activation+0.565
NE
Token NE
Feature activation+0.000
IL
TokenIL
Feature activation+0.000
BL
Token BL
Feature activation+0.000
ake
Tokenake
Feature activation+0.262
@
Token@
Feature activation+0.000
md
Tokenmd
Feature activation+0.000
n
Tokenn
Feature activation+0.000
.
Token.
Feature activation+0.000
net
Tokennet
Feature activation+5.154
Image
Token Image
Feature activation+0.292
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
/
Token /
Feature activation+0.000
3
Token 3
Feature activation+0.000
killed
Token killed
Feature activation+0.000
in
Token in
Feature activation+0.326
Greater
Token Greater
Feature activation+0.338
No
Token No
Feature activation+0.000
ida
Tokenida
Feature activation+0.054
.
Token.
Feature activation+5.017
Ċ
TokenĊ
Feature activation+0.169
Ċ
TokenĊ
Feature activation+0.412
It
TokenIt
Feature activation+0.000
has
Token has
Feature activation+0.000
all
Token all
Feature activation+0.000
Activity
TokenActivity
Feature activation+0.000
:
Token:
Feature activation+0.000
3
Token 3
Feature activation+2.167
Merit
TokenMerit
Feature activation+0.000
:
Token:
Feature activation+0.000
0
Token 0
Feature activation+4.995
To
Token To
Feature activation+0.000
Hal
Token Hal
Feature activation+0.000
Fin
Token Fin
Feature activation+0.000
ney
Tokenney
Feature activation+0.000
from
Token from
Feature activation+0.000
Activity
TokenActivity
Feature activation+0.000
:
Token:
Feature activation+0.000
126
Token 126
Feature activation+0.061
Merit
TokenMerit
Feature activation+0.000
:
Token:
Feature activation+0.000
100
Token 100
Feature activation+4.405
Re
Token Re
Feature activation+0.000
:
Token:
Feature activation+0.000
[
Token [
Feature activation+0.000
ANN
TokenANN
Feature activation+0.000
]
Token]
Feature activation+0.000
the
Token the
Feature activation+0.000
dog
Token dog
Feature activation+0.000
?
Token?
Feature activation+0.478
You
Token You
Feature activation+0.000
decide
Token decide
Feature activation+0.000
.
Token.
Feature activation+4.385
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Commun
TokenCommun
Feature activation+0.000
ist
Tokenist
Feature activation+0.000
Research
Token Research
Feature activation+0.000
Cluster
Token Cluster
Feature activation+0.000
arrest
Token arrest
Feature activation+0.149
and
Token and
Feature activation+0.000
weapons
Token weapons
Feature activation+0.000
violations
Token violations
Feature activation+0.000
charges
Token charges
Feature activation+0.000
.
Token.
Feature activation+4.147
Image
Token Image
Feature activation+0.803
9
Token 9
Feature activation+0.000
of
Token of
Feature activation+0.000
9
Token 9
Feature activation+0.723
C
Token C
Feature activation+0.000
a
Token a
Feature activation+0.000
right
Token right
Feature activation+0.000
to
Token to
Feature activation+0.000
your
Token your
Feature activation+0.000
help
Token help
Feature activation+1.145
"
Token"
Feature activation+4.072
Mar
Token Mar
Feature activation+0.000
at
Tokenat
Feature activation+0.599
's
Token's
Feature activation+0.507
figure
Token figure
Feature activation+0.000
is
Token is
Feature activation+0.000
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
issues
Token issues
Feature activation+0.000
that
Token that
Feature activation+0.000
matter
Token matter
Feature activation+0.264
.
Token.
Feature activation+3.959
Ċ
TokenĊ
Feature activation+0.513
Ċ
TokenĊ
Feature activation+0.203
Fight
TokenFight
Feature activation+0.000
Back
Token Back
Feature activation+0.000
!
Token!
Feature activation+0.000
ort
Tokenort
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
and
Token and
Feature activation+0.000
Det
Token Det
Feature activation+0.000
ainer
Tokenainer
Feature activation+0.000
Usage
Token Usage
Feature activation+3.953
Det
Token Det
Feature activation+0.000
ainer
Tokenainer
Feature activation+0.000
-
Token-
Feature activation+0.000
Connect
TokenConnect
Feature activation+0.000
ed
Tokened
Feature activation+0.000
and
Token and
Feature activation+0.000
you
Token you
Feature activation+0.000
are
Token are
Feature activation+0.000
barely
Token barely
Feature activation+0.000
warm
Token warm
Feature activation+0.000
.
Token.
Feature activation+3.867
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
As
TokenAs
Feature activation+0.000
Sony
Token Sony
Feature activation+0.000
's
Token's
Feature activation+0.000
release
Token release
Feature activation+0.000
/
Token/
Feature activation+0.000
NPR
TokenNPR
Feature activation+0.000
Noah
Token Noah
Feature activation+0.000
Adams
Token Adams
Feature activation+0.000
/
Token/
Feature activation+0.000
NPR
TokenNPR
Feature activation+3.758
En
Token En
Feature activation+0.000
large
Tokenlarge
Feature activation+0.000
this
Token this
Feature activation+0.000
image
Token image
Feature activation+0.000
toggle
Token toggle
Feature activation+0.000
charges
Token charges
Feature activation+0.082
.
Token.
Feature activation+9.570
Image
Token Image
Feature activation+0.334
8
Token 8
Feature activation+0.000
of
Token of
Feature activation+0.000
9
Token 9
Feature activation+3.467
Paul
Token Paul
Feature activation+0.000
Ant
Token Ant
Feature activation+0.000
oine
Tokenoine
Feature activation+0.000
,
Token,
Feature activation+0.000
28
Token 28
Feature activation+0.000
arrested
Token arrested
Feature activation+0.124
Sunday
Token Sunday
Feature activation+0.155
in
Token in
Feature activation+0.188
San
Token San
Feature activation+0.000
Francisco
Token Francisco
Feature activation+0.000
's
Token's
Feature activation+3.407
So
Token So
Feature activation+0.000
Ma
TokenMa
Feature activation+0.000
neighborhood
Token neighborhood
Feature activation+0.000
on
Token on
Feature activation+0.000
suspicion
Token suspicion
Feature activation+0.000
at
Token at
Feature activation+0.284
10
Token 10
Feature activation+0.000
:
Token:
Feature activation+0.000
38
Token38
Feature activation+0.000
AM
TokenAM
Feature activation+1.541
by
Token by
Feature activation+3.039
Tyler
Token Tyler
Feature activation+0.000
Thompson
Token Thompson
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Matt
TokenMatt
Feature activation+0.000

Top DFA by src position
MAX = 4.021

closed
Token closed
Feature activation+1.133
Top resid features:
the
Token the
Feature activation+0.130
Top resid features:
case
Token case
Feature activation+0.122
Top resid features:
because
Token because
Feature activation+0.929
Top resid features:
they
Token they
Feature activation+0.604
Top resid features:
claimed
Token claimed
Feature activation+3.323
Top resid features:
they
Token they
Feature activation+0.137
Top resid features:
were
Token were
Feature activation+0.383
Top resid features:
unable
Token unable
Feature activation+0.089
Top resid features:
to
Token to
Feature activation+0.093
Top resid features:
identify
Token identify
Feature activation+0.681
Top resid features:
arrest
Token arrest
Feature activation+0.124
Top resid features:
and
Token and
Feature activation+0.023
Top resid features:
weapons
Token weapons
Feature activation+0.012
Top resid features:
violations
Token violations
Feature activation+0.130
Top resid features:
charges
Token charges
Feature activation+0.309
Top resid features:
.
Token.
Feature activation+4.021
Top resid features:
Image
Token Image
Feature activation+0.000
Top resid features:
8
Token 8
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
9
Token 9
Feature activation+0.000
Top resid features:
Paul
Token Paul
Feature activation+0.000
Top resid features:
@
Token @
Feature activation+0.187
Top resid features:
ed
Tokened
Feature activation+0.051
Top resid features:
ward
Tokenward
Feature activation+0.022
Top resid features:
ian
Tokenian
Feature activation+0.049
Top resid features:
_
Token_
Feature activation+0.047
Top resid features:
era
Tokenera
Feature activation+2.288
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Top resid features:
UPDATE
TokenUPDATE
Feature activation+0.000
Top resid features:
:
Token:
Feature activation+0.000
Top resid features:
'
Token '
Feature activation+0.000
Top resid features:
Black
TokenBlack
Feature activation+0.000
Top resid features:
defeated
Token defeated
Feature activation+0.061
Top resid features:
Greenwich
Token Greenwich
Feature activation+0.043
Top resid features:
6
Token 6
Feature activation+0.016
Top resid features:
-
Token-
Feature activation+0.003
Top resid features:
5
Token5
Feature activation+0.110
Top resid features:
.
Token.
Feature activation+2.258
Top resid features:
Photo
Token Photo
Feature activation+0.000
Top resid features:
:
Token:
Feature activation+0.000
Top resid features:
Bob
Token Bob
Feature activation+0.000
Top resid features:
Luc
Token Luc
Feature activation+0.000
Top resid features:
key
Tokenkey
Feature activation+0.000
Top resid features:
Member
Token Member
Feature activation-0.030
Top resid features:
Activity
TokenActivity
Feature activation+0.159
Top resid features:
:
Token:
Feature activation+0.358
Top resid features:
448
Token 448
Feature activation+0.801
Top resid features:
Merit
TokenMerit
Feature activation-0.080
Top resid features:
:
Token:
Feature activation+3.295
Top resid features:
250
Token 250
Feature activation+2.917
Top resid features:
Re
Token Re
Feature activation+0.000
Top resid features:
:
Token:
Feature activation+0.000
Top resid features:
[
Token [
Feature activation+0.000
Top resid features:
ANN
TokenANN
Feature activation+0.000
Top resid features:
land
Tokenland
Feature activation+0.037
Top resid features:
across
Token across
Feature activation+0.046
Top resid features:
from
Token from
Feature activation+0.103
Top resid features:
Dow
Token Dow
Feature activation+0.024
Top resid features:
Diamond
Token Diamond
Feature activation+0.072
Top resid features:
.
Token.
Feature activation+2.525
Top resid features:
Photo
Token Photo
Feature activation+0.000
Top resid features:
:
Token:
Feature activation+0.000
Top resid features:
NE
Token NE
Feature activation+0.000
Top resid features:
IL
TokenIL
Feature activation+0.000
Top resid features:
BL
Token BL
Feature activation+0.000
Top resid features:
of
Token of
Feature activation+0.090
Top resid features:
the
Token the
Feature activation+0.047
Top resid features:
Lansing
Token Lansing
Feature activation+0.004
Top resid features:
Lug
Token Lug
Feature activation+0.006
Top resid features:
nuts
Tokennuts
Feature activation+0.170
Top resid features:
.
Token.
Feature activation+1.553
Top resid features:
The
Token The
Feature activation+0.204
Top resid features:
Gillespie
Token Gillespie
Feature activation+0.012
Top resid features:
Group
Token Group
Feature activation+0.056
Top resid features:
announced
Token announced
Feature activation+0.155
Top resid features:
a
Token a
Feature activation+0.179
Top resid features:
killed
Token killed
Feature activation+0.092
Top resid features:
in
Token in
Feature activation+0.121
Top resid features:
Greater
Token Greater
Feature activation+0.035
Top resid features:
No
Token No
Feature activation+0.063
Top resid features:
ida
Tokenida
Feature activation+0.631
Top resid features:
.
Token.
Feature activation+2.804
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
It
TokenIt
Feature activation+0.000
Top resid features:
has
Token has
Feature activation+0.000
Top resid features:
all
Token all
Feature activation+0.000
Top resid features:
bie
Tokenbie
Feature activation+0.013
Top resid features:
Activity
TokenActivity
Feature activation+0.178
Top resid features:
:
Token:
Feature activation+0.147
Top resid features:
3
Token 3
Feature activation+1.563
Top resid features:
Merit
TokenMerit
Feature activation-0.088
Top resid features:
:
Token:
Feature activation+2.747
Top resid features:
0
Token 0
Feature activation+1.432
Top resid features:
To
Token To
Feature activation+0.000
Top resid features:
Hal
Token Hal
Feature activation+0.000
Top resid features:
Fin
Token Fin
Feature activation+0.000
Top resid features:
ney
Tokenney
Feature activation+0.000
Top resid features:
Member
Token Member
Feature activation-0.062
Top resid features:
Activity
TokenActivity
Feature activation+0.090
Top resid features:
:
Token:
Feature activation+0.286
Top resid features:
126
Token 126
Feature activation+1.519
Top resid features:
Merit
TokenMerit
Feature activation-0.101
Top resid features:
:
Token:
Feature activation+2.622
Top resid features:
100
Token 100
Feature activation+1.269
Top resid features:
Re
Token Re
Feature activation+0.000
Top resid features:
:
Token:
Feature activation+0.000
Top resid features:
[
Token [
Feature activation+0.000
Top resid features:
ANN
TokenANN
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.006
Top resid features:
dog
Token dog
Feature activation+0.007
Top resid features:
?
Token?
Feature activation+0.221
Top resid features:
You
Token You
Feature activation+0.425
Top resid features:
decide
Token decide
Feature activation+0.174
Top resid features:
.
Token.
Feature activation+3.310
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Top resid features:
Commun
TokenCommun
Feature activation+0.000
Top resid features:
ist
Tokenist
Feature activation+0.000
Top resid features:
Research
Token Research
Feature activation+0.000
Top resid features:
Cluster
Token Cluster
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.342
Top resid features:
Paul
Token Paul
Feature activation+0.147
Top resid features:
Ant
Token Ant
Feature activation+0.070
Top resid features:
oine
Tokenoine
Feature activation+0.125
Top resid features:
,
Token,
Feature activation+0.283
Top resid features:
28
Token 28
Feature activation+0.732
Top resid features:
,
Token,
Feature activation+0.310
Top resid features:
was
Token was
Feature activation+0.119
Top resid features:
arrested
Token arrested
Feature activation+0.179
Top resid features:
Sunday
Token Sunday
Feature activation+0.178
Top resid features:
in
Token in
Feature activation+0.069
Top resid features:
a
Token a
Feature activation+0.012
Top resid features:
right
Token right
Feature activation+0.020
Top resid features:
to
Token to
Feature activation+0.008
Top resid features:
your
Token your
Feature activation+0.005
Top resid features:
help
Token help
Feature activation+0.045
Top resid features:
"
Token"
Feature activation+1.451
Top resid features:
Mar
Token Mar
Feature activation+0.000
Top resid features:
at
Tokenat
Feature activation+0.000
Top resid features:
's
Token's
Feature activation+0.000
Top resid features:
figure
Token figure
Feature activation+0.000
Top resid features:
is
Token is
Feature activation+0.000
Top resid features:
write
Token write
Feature activation+0.167
Top resid features:
about
Token about
Feature activation+0.325
Top resid features:
the
Token the
Feature activation+0.500
Top resid features:
issues
Token issues
Feature activation+0.424
Top resid features:
that
Token that
Feature activation+0.453
Top resid features:
matter
Token matter
Feature activation+0.866
Top resid features:
.
Token.
Feature activation+0.472
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Fight
TokenFight
Feature activation+0.000
Top resid features:
Back
Token Back
Feature activation+0.000
Top resid features:
ainer
Tokenainer
Feature activation+0.004
Top resid features:
Usage
Token Usage
Feature activation+0.078
Top resid features:
Figure
Token Figure
Feature activation+0.389
Top resid features:
3
Token 3
Feature activation+0.387
Top resid features:
.
Token.
Feature activation+0.521
Top resid features:
All
Token All
Feature activation+1.907
Top resid features:
ICE
Token ICE
Feature activation+0.317
Top resid features:
Dep
Token Dep
Feature activation+0.015
Top resid features:
ort
Tokenort
Feature activation+0.008
Top resid features:
ations
Tokenations
Feature activation+0.196
Top resid features:
and
Token and
Feature activation+0.090
Top resid features:
and
Token and
Feature activation+0.048
Top resid features:
you
Token you
Feature activation+0.045
Top resid features:
are
Token are
Feature activation+0.091
Top resid features:
barely
Token barely
Feature activation+0.043
Top resid features:
warm
Token warm
Feature activation+0.163
Top resid features:
.
Token.
Feature activation+3.446
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Top resid features:
As
TokenAs
Feature activation+0.000
Top resid features:
Sony
Token Sony
Feature activation+0.000
Top resid features:
's
Token's
Feature activation+0.000
Top resid features:
release
Token release
Feature activation+0.000
Top resid features:
Noah
Token Noah
Feature activation-0.046
Top resid features:
Adams
Token Adams
Feature activation-0.013
Top resid features:
/
Token/
Feature activation-0.107
Top resid features:
NPR
TokenNPR
Feature activation-0.077
Top resid features:
Noah
Token Noah
Feature activation+1.361
Top resid features:
Adams
Token Adams
Feature activation+3.659
Top resid features:
/
Token/
Feature activation+0.804
Top resid features:
NPR
TokenNPR
Feature activation+0.271
Top resid features:
En
Token En
Feature activation+0.000
Top resid features:
large
Tokenlarge
Feature activation+0.000
Top resid features:
this
Token this
Feature activation+0.000
Top resid features:
arrest
Token arrest
Feature activation+0.032
Top resid features:
and
Token and
Feature activation+0.022
Top resid features:
weapons
Token weapons
Feature activation+0.008
Top resid features:
violations
Token violations
Feature activation+0.034
Top resid features:
charges
Token charges
Feature activation+0.076
Top resid features:
.
Token.
Feature activation+1.640
Top resid features:
Image
Token Image
Feature activation+0.213
Top resid features:
8
Token 8
Feature activation-0.020
Top resid features:
of
Token of
Feature activation+0.024
Top resid features:
9
Token 9
Feature activation+0.026
Top resid features:
Paul
Token Paul
Feature activation+0.000
Top resid features:
28
Token 28
Feature activation+0.237
Top resid features:
,
Token,
Feature activation+0.206
Top resid features:
was
Token was
Feature activation+0.131
Top resid features:
arrested
Token arrested
Feature activation+0.257
Top resid features:
Sunday
Token Sunday
Feature activation+0.229
Top resid features:
in
Token in
Feature activation+0.448
Top resid features:
San
Token San
Feature activation+0.173
Top resid features:
Francisco
Token Francisco
Feature activation+0.158
Top resid features:
's
Token's
Feature activation+0.200
Top resid features:
So
Token So
Feature activation+0.000
Top resid features:
Ma
TokenMa
Feature activation+0.000
Top resid features:
2010
Token 2010
Feature activation+0.373
Top resid features:
at
Token at
Feature activation+0.194
Top resid features:
10
Token 10
Feature activation+0.146
Top resid features:
:
Token:
Feature activation+0.208
Top resid features:
38
Token38
Feature activation+0.063
Top resid features:
AM
TokenAM
Feature activation+1.331
Top resid features:
by
Token by
Feature activation+0.993
Top resid features:
Tyler
Token Tyler
Feature activation+0.000
Top resid features:
Thompson
Token Thompson
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.05

Head 2: 0.07

Head 3: 0.04

Head 4: 0.04

Head 5: 0.09

Head 6: 0.05

Head 7: 0.05

Head 8: 0.03

Head 9: 0.41

Head 10: 0.10

Head 11: 0.02

Positive logits

204392.10

lease2.09

2.02

blaze1.99

DNA1.98

reperto1.93

��1.85

ochet1.84

chic1.83

��1.80

tamp1.78

Saras1.78

Pixie1.78

murd1.78

ッド1.76

odder1.76

kay1.74

embr1.74

Nile1.74

loudspe1.73

Negative logits

Lear-2.15

FIG-2.09

:{-2.00

Option-1.92

staking-1.92

UTH-1.90

ild-1.90

Benef-1.89

rarily-1.88

weights-1.88

Ranked-1.88

Example-1.88

Bottom-1.87

Total-1.86

Rank-1.86

midt-1.84

Parent-1.81

-1.81

imir-1.80

imus-1.80

INTERVAL 8.793 - 9.770
CONTAINS 0.000%

to
Token to
Feature activation+0.000
identify
Token identify
Feature activation+0.346
the
Token the
Feature activation+0.101
policeman
Token policeman
Feature activation+0.049
responsible
Token responsible
Feature activation+0.000
for
Token for
Feature activation+9.770
shooting
Token shooting
Feature activation+0.301
and
Token and
Feature activation+0.028
injuring
Token injuring
Feature activation+0.000
Mah
Token Mah
Feature activation+0.000
am
Tokenam
Feature activation+0.000
arrest
Token arrest
Feature activation+1.106
and
Token and
Feature activation+0.000
weapons
Token weapons
Feature activation+0.000
violations
Token violations
Feature activation+0.000
charges
Token charges
Feature activation+0.082
.
Token.
Feature activation+9.570
Image
Token Image
Feature activation+0.334
8
Token 8
Feature activation+0.000
of
Token of
Feature activation+0.000
9
Token 9
Feature activation+3.467
Paul
Token Paul
Feature activation+0.000

INTERVAL 7.816 - 8.793
CONTAINS 0.000%

INTERVAL 6.839 - 7.816
CONTAINS 0.000%

@
Token @
Feature activation+1.046
ed
Tokened
Feature activation+0.000
ward
Tokenward
Feature activation+0.000
ian
Tokenian
Feature activation+0.035
_
Token_
Feature activation+0.000
era
Tokenera
Feature activation+6.925
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
UPDATE
TokenUPDATE
Feature activation+0.000
:
Token:
Feature activation+0.000
'
Token '
Feature activation+0.000
Black
TokenBlack
Feature activation+0.000

INTERVAL 5.862 - 6.839
CONTAINS 0.000%

defeated
Token defeated
Feature activation+0.194
Greenwich
Token Greenwich
Feature activation+2.334
6
Token 6
Feature activation+0.000
-
Token-
Feature activation+0.000
5
Token5
Feature activation+0.930
.
Token.
Feature activation+6.615
Photo
Token Photo
Feature activation+0.229
:
Token:
Feature activation+0.102
Bob
Token Bob
Feature activation+0.000
Luc
Token Luc
Feature activation+0.000
key
Tokenkey
Feature activation+0.000
Activity
TokenActivity
Feature activation+0.000
:
Token:
Feature activation+0.000
448
Token 448
Feature activation+0.214
Merit
TokenMerit
Feature activation+0.000
:
Token:
Feature activation+0.000
250
Token 250
Feature activation+6.085
Re
Token Re
Feature activation+0.000
:
Token:
Feature activation+0.000
[
Token [
Feature activation+0.000
ANN
TokenANN
Feature activation+0.000
][
Token][
Feature activation+0.000

INTERVAL 4.885 - 5.862
CONTAINS 0.000%

land
Tokenland
Feature activation+0.000
across
Token across
Feature activation+0.000
from
Token from
Feature activation+0.867
Dow
Token Dow
Feature activation+0.311
Diamond
Token Diamond
Feature activation+0.038
.
Token.
Feature activation+5.737
Photo
Token Photo
Feature activation+1.837
:
Token:
Feature activation+0.565
NE
Token NE
Feature activation+0.000
IL
TokenIL
Feature activation+0.000
BL
Token BL
Feature activation+0.000
ake
Tokenake
Feature activation+0.262
@
Token@
Feature activation+0.000
md
Tokenmd
Feature activation+0.000
n
Tokenn
Feature activation+0.000
.
Token.
Feature activation+0.000
net
Tokennet
Feature activation+5.154
Image
Token Image
Feature activation+0.292
1
Token 1
Feature activation+0.000
of
Token of
Feature activation+0.000
/
Token /
Feature activation+0.000
3
Token 3
Feature activation+0.000
killed
Token killed
Feature activation+0.000
in
Token in
Feature activation+0.326
Greater
Token Greater
Feature activation+0.338
No
Token No
Feature activation+0.000
ida
Tokenida
Feature activation+0.054
.
Token.
Feature activation+5.017
Ċ
TokenĊ
Feature activation+0.169
Ċ
TokenĊ
Feature activation+0.412
It
TokenIt
Feature activation+0.000
has
Token has
Feature activation+0.000
all
Token all
Feature activation+0.000
Activity
TokenActivity
Feature activation+0.000
:
Token:
Feature activation+0.000
3
Token 3
Feature activation+2.167
Merit
TokenMerit
Feature activation+0.000
:
Token:
Feature activation+0.000
0
Token 0
Feature activation+4.995
To
Token To
Feature activation+0.000
Hal
Token Hal
Feature activation+0.000
Fin
Token Fin
Feature activation+0.000
ney
Tokenney
Feature activation+0.000
from
Token from
Feature activation+0.000

INTERVAL 3.908 - 4.885
CONTAINS 0.001%

arrest
Token arrest
Feature activation+0.149
and
Token and
Feature activation+0.000
weapons
Token weapons
Feature activation+0.000
violations
Token violations
Feature activation+0.000
charges
Token charges
Feature activation+0.000
.
Token.
Feature activation+4.147
Image
Token Image
Feature activation+0.803
9
Token 9
Feature activation+0.000
of
Token of
Feature activation+0.000
9
Token 9
Feature activation+0.723
C
Token C
Feature activation+0.000
a
Token a
Feature activation+0.000
right
Token right
Feature activation+0.000
to
Token to
Feature activation+0.000
your
Token your
Feature activation+0.000
help
Token help
Feature activation+1.145
"
Token"
Feature activation+4.072
Mar
Token Mar
Feature activation+0.000
at
Tokenat
Feature activation+0.599
's
Token's
Feature activation+0.507
figure
Token figure
Feature activation+0.000
is
Token is
Feature activation+0.000
Activity
TokenActivity
Feature activation+0.000
:
Token:
Feature activation+0.000
126
Token 126
Feature activation+0.061
Merit
TokenMerit
Feature activation+0.000
:
Token:
Feature activation+0.000
100
Token 100
Feature activation+4.405
Re
Token Re
Feature activation+0.000
:
Token:
Feature activation+0.000
[
Token [
Feature activation+0.000
ANN
TokenANN
Feature activation+0.000
]
Token]
Feature activation+0.000
ort
Tokenort
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
and
Token and
Feature activation+0.000
Det
Token Det
Feature activation+0.000
ainer
Tokenainer
Feature activation+0.000
Usage
Token Usage
Feature activation+3.953
Det
Token Det
Feature activation+0.000
ainer
Tokenainer
Feature activation+0.000
-
Token-
Feature activation+0.000
Connect
TokenConnect
Feature activation+0.000
ed
Tokened
Feature activation+0.000
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
issues
Token issues
Feature activation+0.000
that
Token that
Feature activation+0.000
matter
Token matter
Feature activation+0.264
.
Token.
Feature activation+3.959
Ċ
TokenĊ
Feature activation+0.513
Ċ
TokenĊ
Feature activation+0.203
Fight
TokenFight
Feature activation+0.000
Back
Token Back
Feature activation+0.000
!
Token!
Feature activation+0.000

INTERVAL 2.931 - 3.908
CONTAINS 0.001%

arrested
Token arrested
Feature activation+0.124
Sunday
Token Sunday
Feature activation+0.155
in
Token in
Feature activation+0.188
San
Token San
Feature activation+0.000
Francisco
Token Francisco
Feature activation+0.000
's
Token's
Feature activation+3.407
So
Token So
Feature activation+0.000
Ma
TokenMa
Feature activation+0.000
neighborhood
Token neighborhood
Feature activation+0.000
on
Token on
Feature activation+0.000
suspicion
Token suspicion
Feature activation+0.000
,
Token,
Feature activation+0.000
2014
Token 2014
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Patch
Token Patch
Feature activation+0.000
work
Tokenwork
Feature activation+0.000
Girl
Token Girl
Feature activation+2.964
Ċ
TokenĊ
Feature activation+0.523
Ċ
TokenĊ
Feature activation+0.000
Reviewer
TokenReviewer
Feature activation+0.000
:
Token:
Feature activation+0.000
jud
Token jud
Feature activation+0.000
/
Token/
Feature activation+0.000
NPR
TokenNPR
Feature activation+0.000
Noah
Token Noah
Feature activation+0.000
Adams
Token Adams
Feature activation+0.000
/
Token/
Feature activation+0.000
NPR
TokenNPR
Feature activation+3.758
En
Token En
Feature activation+0.000
large
Tokenlarge
Feature activation+0.000
this
Token this
Feature activation+0.000
image
Token image
Feature activation+0.000
toggle
Token toggle
Feature activation+0.000
night
Token night
Feature activation+0.000
!
Token!
Feature activation+0.097
Follow
Token Follow
Feature activation+1.187
me
Token me
Feature activation+0.000
on
Token on
Feature activation+0.000
Twitter
Token Twitter
Feature activation+2.975
@
Token @
Feature activation+1.046
ed
Tokened
Feature activation+0.000
ward
Tokenward
Feature activation+0.000
ian
Tokenian
Feature activation+0.035
_
Token_
Feature activation+0.000
and
Token and
Feature activation+0.000
you
Token you
Feature activation+0.000
are
Token are
Feature activation+0.000
barely
Token barely
Feature activation+0.000
warm
Token warm
Feature activation+0.000
.
Token.
Feature activation+3.867
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
As
TokenAs
Feature activation+0.000
Sony
Token Sony
Feature activation+0.000
's
Token's
Feature activation+0.000
release
Token release
Feature activation+0.000

INTERVAL 1.954 - 2.931
CONTAINS 0.003%

Ċ
TokenĊ
Feature activation+0.000
New
TokenNew
Feature activation+0.000
bie
Tokenbie
Feature activation+0.000
Activity
TokenActivity
Feature activation+0.000
:
Token:
Feature activation+0.000
3
Token 3
Feature activation+2.167
Merit
TokenMerit
Feature activation+0.000
:
Token:
Feature activation+0.000
0
Token 0
Feature activation+4.995
To
Token To
Feature activation+0.000
Hal
Token Hal
Feature activation+0.000
and
Token and
Feature activation+0.000
create
Token create
Feature activation+0.000
beautiful
Token beautiful
Feature activation+0.000
light
Token light
Feature activation+0.104
stre
Token stre
Feature activation+0.000
aked
Tokenaked
Feature activation+2.391
inst
Token inst
Feature activation+0.000
ants
Tokenants
Feature activation+0.000
!
Token!
Feature activation+0.000
Color
Token Color
Feature activation+0.000
G
Token G
Feature activation+0.000
your
Token your
Feature activation+0.000
Graveyard
Token Graveyard
Feature activation+0.000
then
Token then
Feature activation+0.000
fuse
Token fuse
Feature activation+0.000
them
Token them
Feature activation+0.000
into
Token into
Feature activation+2.756
.
Token .
Feature activation+0.000
Pull
Token Pull
Feature activation+0.000
back
Token back
Feature activation+0.000
a
Token a
Feature activation+0.000
Tun
Token Tun
Feature activation+0.000
[
Token [
Feature activation+0.000
6
Token6
Feature activation+0.000
]
Token]
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.029
or
Tokenor
Feature activation+2.095
,
Token ,
Feature activation+0.000
in
Token in
Feature activation+0.453
the
Token the
Feature activation+1.039
7
Token 7
Feature activation+0.000
th
Tokenth
Feature activation+0.611
/
Token/
Feature activation+0.000
NPR
TokenNPR
Feature activation+2.039
Noah
Token Noah
Feature activation+0.000
Adams
Token Adams
Feature activation+0.000
/
Token/
Feature activation+0.000
NPR
TokenNPR
Feature activation+1.955
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Tem
TokenTem
Feature activation+0.000
ple
Tokenple
Feature activation+0.000
ton
Tokenton
Feature activation+0.000

INTERVAL 0.977 - 1.954
CONTAINS 0.007%

E
Token E
Feature activation+0.000
.
Token.
Feature activation+0.000
g
Tokeng
Feature activation+0.000
.
Token.
Feature activation+1.161
help
Token help
Feature activation+2.541
cd
Token cd
Feature activation+1.770
Also
Token Also
Feature activation+0.000
,
Token,
Feature activation+0.000
your
Token your
Feature activation+0.000
favorite
Token favorite
Feature activation+0.000
search
Token search
Feature activation+0.000
modules
Tokenmodules
Feature activation+0.000
/
Token/
Feature activation+0.000
article
Tokenarticle
Feature activation+0.000
/
Token/
Feature activation+0.246
blocks
Tokenblocks
Feature activation+0.104
/
Token/
Feature activation+1.066
block
Tokenblock
Feature activation+0.416
_
Token_
Feature activation+0.917
comm
Tokencomm
Feature activation+0.000
end
Tokenend
Feature activation+0.000
.
Token.
Feature activation+0.000
!
Token!
Feature activation+0.097
Follow
Token Follow
Feature activation+1.187
me
Token me
Feature activation+0.000
on
Token on
Feature activation+0.000
Twitter
Token Twitter
Feature activation+2.975
@
Token @
Feature activation+1.046
ed
Tokened
Feature activation+0.000
ward
Tokenward
Feature activation+0.000
ian
Tokenian
Feature activation+0.035
_
Token_
Feature activation+0.000
era
Tokenera
Feature activation+6.925
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.029
or
Tokenor
Feature activation+2.095
,
Token ,
Feature activation+0.000
in
Token in
Feature activation+0.453
the
Token the
Feature activation+1.039
7
Token 7
Feature activation+0.000
th
Tokenth
Feature activation+0.611
month
Token month
Feature activation+1.747
(
Token (
Feature activation+0.553
11
Token11
Feature activation+0.000
until
Token until
Feature activation+0.000
next
Token next
Feature activation+0.000
time
Token time
Feature activation+0.000
,
Token,
Feature activation+0.359
Amy
Token Amy
Feature activation+0.000
Lee
Token Lee
Feature activation+1.679
Ċ
TokenĊ
Feature activation+1.293
Ċ
TokenĊ
Feature activation+0.502
The
TokenThe
Feature activation+0.000
Lac
Token Lac
Feature activation+0.000
quer
Tokenquer
Feature activation+0.000

INTERVAL 0.000 - 0.977
CONTAINS 99.988%

.
Token.
Feature activation+0.000
Rather
Token Rather
Feature activation+0.000
,
Token,
Feature activation+0.000
they
Token they
Feature activation+0.000
can
Token can
Feature activation+0.000
be
Token be
Feature activation+0.000
programmed
Token programmed
Feature activation+0.000
to
Token to
Feature activation+0.000
respond
Token respond
Feature activation+0.000
to
Token to
Feature activation+0.000
their
Token their
Feature activation+0.000
brook
Tokenbrook
Feature activation+0.000
,
Token,
Feature activation+0.000
former
Token former
Feature activation+0.000
head
Token head
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
National
Token National
Feature activation+0.000
Highway
Token Highway
Feature activation+0.000
Traffic
Token Traffic
Feature activation+0.000
Safety
Token Safety
Feature activation+0.000
Administration
Token Administration
Feature activation+0.000
believe
Token believe
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
what
Token what
Feature activation+0.000
sets
Token sets
Feature activation+0.000
each
Token each
Feature activation+0.000
belt
Token belt
Feature activation+0.000
level
Token level
Feature activation+0.000
apart
Token apart
Feature activation+0.000
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
his
Token his
Feature activation+0.000
life
Token life
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
gone
Token gone
Feature activation+0.000
from
Token from
Feature activation+0.000
liberal
Token liberal
Feature activation+0.000
consumer
Token consumer
Feature activation+0.000
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.000
anti
Token anti
Feature activation+0.000
-
Token-
Feature activation+0.000
abortion
Tokenabortion
Feature activation+0.000
votes
Token votes
Feature activation+0.000
up
Token up
Feature activation+0.000
for
Token for
Feature activation+0.000
grabs
Token grabs
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 5 in H8.9: (feature 13655)

TOP ACTIVATIONS
MAX = 5.302

are
Token are
Feature activation+0.000
not
Token not
Feature activation+0.000
as
Token as
Feature activation+0.000
bullish
Token bullish
Feature activation+0.000
as
Token as
Feature activation+3.705
people
Token people
Feature activation+5.302
thought
Token thought
Feature activation+3.554
and
Token and
Feature activation+0.047
they
Token they
Feature activation+0.000
bought
Token bought
Feature activation+0.000
back
Token back
Feature activation+0.000
quite
Token quite
Feature activation+0.000
as
Token as
Feature activation+0.000
unlikely
Token unlikely
Feature activation+0.000
as
Token as
Feature activation+3.369
it
Token it
Feature activation+3.101
seems
Token seems
Feature activation+4.736
.
Token.
Feature activation+0.000
From
Token From
Feature activation+0.000
the
Token the
Feature activation+0.000
Battle
Token Battle
Feature activation+0.000
of
Token of
Feature activation+0.000
much
Token much
Feature activation+0.000
more
Token more
Feature activation+0.000
complicated
Token complicated
Feature activation+0.000
than
Token than
Feature activation+1.069
it
Token it
Feature activation+2.410
appears
Token appears
Feature activation+4.499
,
Token,
Feature activation+0.364
and
Token and
Feature activation+0.000
it
Token it
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
skulls
Token skulls
Feature activation+0.000
is
Token is
Feature activation+0.000
even
Token even
Feature activation+0.000
more
Token more
Feature activation+0.000
complicated
Token complicated
Feature activation+0.000
than
Token than
Feature activation+4.450
it
Token it
Feature activation+3.427
looks
Token looks
Feature activation+4.299
.
Token.
Feature activation+0.000
Thanks
Token Thanks
Feature activation+0.000
to
Token to
Feature activation+0.000
even
Token even
Feature activation+0.000
more
Token more
Feature activation+0.000
complicated
Token complicated
Feature activation+0.000
than
Token than
Feature activation+4.450
it
Token it
Feature activation+3.427
looks
Token looks
Feature activation+4.299
.
Token.
Feature activation+0.000
Thanks
Token Thanks
Feature activation+0.000
to
Token to
Feature activation+0.000
C
Token C
Feature activation+0.000
aj
Tokenaj
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+0.000
childhood
Token childhood
Feature activation+0.000
mortality
Token mortality
Feature activation+0.000
than
Token than
Feature activation+1.538
previously
Token previously
Feature activation+3.949
thought
Token thought
Feature activation+3.686
.
Token.
Feature activation+0.000
If
Token If
Feature activation+0.000
correct
Token correct
Feature activation+0.000
,
Token,
Feature activation+0.000
be
Token be
Feature activation+0.000
as
Token as
Feature activation+0.000
crazy
Token crazy
Feature activation+0.000
as
Token as
Feature activation+2.994
it
Token it
Feature activation+3.193
sounds
Token sounds
Feature activation+3.858
,
Token,
Feature activation+0.000
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.000
F
Token F
Feature activation+0.000
OSS
TokenOSS
Feature activation+0.000
work
Token work
Feature activation+0.000
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+1.600
they
Token they
Feature activation+3.801
had
Token had
Feature activation+3.431
wished
Token wished
Feature activation+1.657
,
Token,
Feature activation+0.168
and
Token and
Feature activation+0.000
left
Token left
Feature activation+0.000
bigger
Token bigger
Feature activation+0.000
than
Token than
Feature activation+0.764
the
Token the
Feature activation+0.524
New
Token New
Feature activation+0.000
Republic
Token Republic
Feature activation+1.857
hopes
Token hopes
Feature activation+3.743
it
Token it
Feature activation+1.139
is
Token is
Feature activation+1.741
.
Token.
Feature activation+0.000
H
Token H
Feature activation+0.000
iding
Tokeniding
Feature activation+0.000
numbers
Token numbers
Feature activation+0.000
are
Token are
Feature activation+0.000
not
Token not
Feature activation+0.000
as
Token as
Feature activation+0.000
bullish
Token bullish
Feature activation+0.000
as
Token as
Feature activation+3.705
people
Token people
Feature activation+5.302
thought
Token thought
Feature activation+3.554
and
Token and
Feature activation+0.047
they
Token they
Feature activation+0.000
bought
Token bought
Feature activation+0.000
the
Token the
Feature activation+0.000
box
Token box
Feature activation+0.000
office
Token office
Feature activation+0.000
as
Token as
Feature activation+1.436
its
Token its
Feature activation+0.331
producers
Token producers
Feature activation+3.695
hoped
Token hoped
Feature activation+3.126
,
Token,
Feature activation+0.682
but
Token but
Feature activation+0.000
instead
Token instead
Feature activation+0.000
of
Token of
Feature activation+0.000
of
Token of
Feature activation+0.000
childhood
Token childhood
Feature activation+0.000
mortality
Token mortality
Feature activation+0.000
than
Token than
Feature activation+1.538
previously
Token previously
Feature activation+3.949
thought
Token thought
Feature activation+3.686
.
Token.
Feature activation+0.000
If
Token If
Feature activation+0.000
correct
Token correct
Feature activation+0.000
,
Token,
Feature activation+0.000
this
Token this
Feature activation+0.000
a
Token a
Feature activation+0.000
paed
Token paed
Feature activation+0.000
ophile
Tokenophile
Feature activation+0.000
as
Token as
Feature activation+2.699
Virginia
Token Virginia
Feature activation+0.295
Roberts
Token Roberts
Feature activation+3.672
was
Token was
Feature activation+1.751
seventeen
Token seventeen
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
time
Token time
Feature activation+0.000
not
Token not
Feature activation+0.000
as
Token as
Feature activation+0.000
bullish
Token bullish
Feature activation+0.000
as
Token as
Feature activation+3.705
people
Token people
Feature activation+5.302
thought
Token thought
Feature activation+3.554
and
Token and
Feature activation+0.047
they
Token they
Feature activation+0.000
bought
Token bought
Feature activation+0.000
back
Token back
Feature activation+0.000
really
Token really
Feature activation+0.000
illa
Tokenilla
Feature activation+0.000
's
Token's
Feature activation+0.000
as
Token as
Feature activation+0.000
good
Token good
Feature activation+0.000
as
Token as
Feature activation+0.498
people
Token people
Feature activation+3.547
say
Token say
Feature activation+2.024
he
Token he
Feature activation+0.000
is
Token is
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+1.600
they
Token they
Feature activation+3.801
had
Token had
Feature activation+3.431
wished
Token wished
Feature activation+1.657
,
Token,
Feature activation+0.168
and
Token and
Feature activation+0.000
left
Token left
Feature activation+0.000
the
Token the
Feature activation+0.000
is
Token is
Feature activation+0.000
even
Token even
Feature activation+0.000
more
Token more
Feature activation+0.000
complicated
Token complicated
Feature activation+0.000
than
Token than
Feature activation+4.450
it
Token it
Feature activation+3.427
looks
Token looks
Feature activation+4.299
.
Token.
Feature activation+0.000
Thanks
Token Thanks
Feature activation+0.000
to
Token to
Feature activation+0.000
C
Token C
Feature activation+0.000
maintained
Token maintained
Feature activation+0.000
she
Token she
Feature activation+0.000
wasn
Token wasn
Feature activation+0.000
't
Token't
Feature activation+0.000
who
Token who
Feature activation+0.037
she
Token she
Feature activation+3.416
claimed
Token claimed
Feature activation+2.605
to
Token to
Feature activation+1.200
be
Token be
Feature activation+1.923
,
Token,
Feature activation+0.000
even
Token even
Feature activation+0.000
quite
Token quite
Feature activation+0.000
as
Token as
Feature activation+0.000
short
Token short
Feature activation+0.000
as
Token as
Feature activation+0.786
most
Token most
Feature activation+1.808
people
Token people
Feature activation+3.414
make
Token make
Feature activation+2.874
it
Token it
Feature activation+1.397
out
Token out
Feature activation+1.673
to
Token to
Feature activation+2.362
be
Token be
Feature activation+2.024
actually
Token actually
Feature activation+0.000
not
Token not
Feature activation+0.000
as
Token as
Feature activation+0.000
crazy
Token crazy
Feature activation+0.000
as
Token as
Feature activation+2.101
you
Token you
Feature activation+3.390
might
Token might
Feature activation+3.235
think
Token think
Feature activation+2.461
âĢĶ
Token âĢĶ
Feature activation+0.000
it
Token it
Feature activation+0.000
is
Token is
Feature activation+0.000

Top DFA by src position
MAX = 4.941

numbers
Token numbers
Feature activation-0.006
Top resid features:
are
Token are
Feature activation+0.012
Top resid features:
not
Token not
Feature activation+0.090
Top resid features:
as
Token as
Feature activation+1.094
Top resid features:
bullish
Token bullish
Feature activation+0.086
Top resid features:
as
Token as
Feature activation+4.622
Top resid features:
people
Token people
Feature activation+0.097
Top resid features:
thought
Token thought
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
they
Token they
Feature activation+0.000
Top resid features:
bought
Token bought
Feature activation+0.000
Top resid features:
is
Token is
Feature activation+0.019
Top resid features:
not
Token not
Feature activation+0.140
Top resid features:
quite
Token quite
Feature activation+0.173
Top resid features:
as
Token as
Feature activation+1.531
Top resid features:
unlikely
Token unlikely
Feature activation+0.213
Top resid features:
as
Token as
Feature activation+3.339
Top resid features:
it
Token it
Feature activation+0.138
Top resid features:
seems
Token seems
Feature activation+0.116
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
From
Token From
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
death
Token death
Feature activation-0.003
Top resid features:
is
Token is
Feature activation+0.037
Top resid features:
much
Token much
Feature activation+0.020
Top resid features:
more
Token more
Feature activation+0.130
Top resid features:
complicated
Token complicated
Feature activation+0.128
Top resid features:
than
Token than
Feature activation+4.941
Top resid features:
it
Token it
Feature activation+0.107
Top resid features:
appears
Token appears
Feature activation+0.067
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
skulls
Token skulls
Feature activation+0.025
Top resid features:
is
Token is
Feature activation+0.076
Top resid features:
even
Token even
Feature activation+0.028
Top resid features:
more
Token more
Feature activation+0.059
Top resid features:
complicated
Token complicated
Feature activation+0.091
Top resid features:
than
Token than
Feature activation+4.897
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
looks
Token looks
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
Thanks
Token Thanks
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
skulls
Token skulls
Feature activation+0.015
Top resid features:
is
Token is
Feature activation+0.031
Top resid features:
even
Token even
Feature activation+0.020
Top resid features:
more
Token more
Feature activation+0.058
Top resid features:
complicated
Token complicated
Feature activation+0.110
Top resid features:
than
Token than
Feature activation+4.549
Top resid features:
it
Token it
Feature activation+0.213
Top resid features:
looks
Token looks
Feature activation+0.054
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
Thanks
Token Thanks
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
important
Token important
Feature activation+0.087
Top resid features:
cause
Token cause
Feature activation+0.058
Top resid features:
of
Token of
Feature activation+0.011
Top resid features:
childhood
Token childhood
Feature activation-0.009
Top resid features:
mortality
Token mortality
Feature activation-0.012
Top resid features:
than
Token than
Feature activation+4.233
Top resid features:
previously
Token previously
Feature activation+0.052
Top resid features:
thought
Token thought
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
If
Token If
Feature activation+0.000
Top resid features:
correct
Token correct
Feature activation+0.000
Top resid features:
may
Token may
Feature activation+0.052
Top resid features:
not
Token not
Feature activation+0.061
Top resid features:
be
Token be
Feature activation+0.109
Top resid features:
as
Token as
Feature activation+1.127
Top resid features:
crazy
Token crazy
Feature activation+0.188
Top resid features:
as
Token as
Feature activation+2.907
Top resid features:
it
Token it
Feature activation+0.105
Top resid features:
sounds
Token sounds
Feature activation+0.065
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
according
Token according
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
not
Token not
Feature activation+0.079
Top resid features:
work
Token work
Feature activation+0.094
Top resid features:
out
Token out
Feature activation+0.397
Top resid features:
in
Token in
Feature activation+0.156
Top resid features:
the
Token the
Feature activation+0.053
Top resid features:
way
Token way
Feature activation+3.463
Top resid features:
they
Token they
Feature activation+0.105
Top resid features:
had
Token had
Feature activation+0.000
Top resid features:
wished
Token wished
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
hand
Token hand
Feature activation+0.003
Top resid features:
,
Token,
Feature activation+0.015
Top resid features:
is
Token is
Feature activation+0.025
Top resid features:
far
Token far
Feature activation+0.016
Top resid features:
bigger
Token bigger
Feature activation+0.210
Top resid features:
than
Token than
Feature activation+3.839
Top resid features:
the
Token the
Feature activation+0.109
Top resid features:
New
Token New
Feature activation+0.009
Top resid features:
Republic
Token Republic
Feature activation+0.052
Top resid features:
hopes
Token hopes
Feature activation+0.000
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
numbers
Token numbers
Feature activation-0.018
Top resid features:
are
Token are
Feature activation+0.006
Top resid features:
not
Token not
Feature activation+0.080
Top resid features:
as
Token as
Feature activation+1.839
Top resid features:
bullish
Token bullish
Feature activation+0.108
Top resid features:
as
Token as
Feature activation+2.386
Top resid features:
people
Token people
Feature activation+0.000
Top resid features:
thought
Token thought
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
they
Token they
Feature activation+0.000
Top resid features:
bought
Token bought
Feature activation+0.000
Top resid features:
well
Token well
Feature activation+0.039
Top resid features:
at
Token at
Feature activation+0.005
Top resid features:
the
Token the
Feature activation+0.009
Top resid features:
box
Token box
Feature activation-0.000
Top resid features:
office
Token office
Feature activation+0.005
Top resid features:
as
Token as
Feature activation+2.181
Top resid features:
its
Token its
Feature activation+0.028
Top resid features:
producers
Token producers
Feature activation+0.062
Top resid features:
hoped
Token hoped
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
but
Token but
Feature activation+0.000
Top resid features:
important
Token important
Feature activation+0.069
Top resid features:
cause
Token cause
Feature activation+0.058
Top resid features:
of
Token of
Feature activation+0.013
Top resid features:
childhood
Token childhood
Feature activation-0.008
Top resid features:
mortality
Token mortality
Feature activation-0.008
Top resid features:
than
Token than
Feature activation+3.881
Top resid features:
previously
Token previously
Feature activation+0.208
Top resid features:
thought
Token thought
Feature activation+0.087
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
If
Token If
Feature activation+0.000
Top resid features:
correct
Token correct
Feature activation+0.000
Top resid features:
is
Token is
Feature activation+0.017
Top resid features:
not
Token not
Feature activation+0.054
Top resid features:
a
Token a
Feature activation+0.076
Top resid features:
paed
Token paed
Feature activation-0.005
Top resid features:
ophile
Tokenophile
Feature activation+0.035
Top resid features:
as
Token as
Feature activation+4.183
Top resid features:
Virginia
Token Virginia
Feature activation-0.020
Top resid features:
Roberts
Token Roberts
Feature activation-0.007
Top resid features:
was
Token was
Feature activation+0.000
Top resid features:
seventeen
Token seventeen
Feature activation+0.000
Top resid features:
at
Token at
Feature activation+0.000
Top resid features:
numbers
Token numbers
Feature activation+0.013
Top resid features:
are
Token are
Feature activation+0.022
Top resid features:
not
Token not
Feature activation+0.093
Top resid features:
as
Token as
Feature activation+1.557
Top resid features:
bullish
Token bullish
Feature activation+0.068
Top resid features:
as
Token as
Feature activation+2.401
Top resid features:
people
Token people
Feature activation+0.065
Top resid features:
thought
Token thought
Feature activation+0.069
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
they
Token they
Feature activation+0.000
Top resid features:
bought
Token bought
Feature activation+0.000
Top resid features:
D
Token D
Feature activation+0.006
Top resid features:
illa
Tokenilla
Feature activation+0.016
Top resid features:
's
Token's
Feature activation+0.026
Top resid features:
as
Token as
Feature activation+0.430
Top resid features:
good
Token good
Feature activation+0.161
Top resid features:
as
Token as
Feature activation+3.635
Top resid features:
people
Token people
Feature activation+0.102
Top resid features:
say
Token say
Feature activation+0.000
Top resid features:
he
Token he
Feature activation+0.000
Top resid features:
is
Token is
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
not
Token not
Feature activation+0.066
Top resid features:
work
Token work
Feature activation+0.072
Top resid features:
out
Token out
Feature activation+0.348
Top resid features:
in
Token in
Feature activation+0.196
Top resid features:
the
Token the
Feature activation+0.100
Top resid features:
way
Token way
Feature activation+3.009
Top resid features:
they
Token they
Feature activation+0.116
Top resid features:
had
Token had
Feature activation+0.047
Top resid features:
wished
Token wished
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
skulls
Token skulls
Feature activation+0.022
Top resid features:
is
Token is
Feature activation+0.024
Top resid features:
even
Token even
Feature activation+0.019
Top resid features:
more
Token more
Feature activation+0.077
Top resid features:
complicated
Token complicated
Feature activation+0.163
Top resid features:
than
Token than
Feature activation+3.612
Top resid features:
it
Token it
Feature activation+0.258
Top resid features:
looks
Token looks
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
Thanks
Token Thanks
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
Officials
TokenOfficials
Feature activation+0.002
Top resid features:
maintained
Token maintained
Feature activation-0.002
Top resid features:
she
Token she
Feature activation+0.005
Top resid features:
wasn
Token wasn
Feature activation-0.006
Top resid features:
't
Token't
Feature activation+0.067
Top resid features:
who
Token who
Feature activation+3.617
Top resid features:
she
Token she
Feature activation+0.547
Top resid features:
claimed
Token claimed
Feature activation+0.000
Top resid features:
to
Token to
Feature activation+0.000
Top resid features:
be
Token be
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation-0.008
Top resid features:
t
Tokent
Feature activation+0.147
Top resid features:
quite
Token quite
Feature activation+0.197
Top resid features:
as
Token as
Feature activation+1.187
Top resid features:
short
Token short
Feature activation+0.231
Top resid features:
as
Token as
Feature activation+2.449
Top resid features:
most
Token most
Feature activation+0.033
Top resid features:
people
Token people
Feature activation+0.057
Top resid features:
make
Token make
Feature activation+0.000
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:
out
Token out
Feature activation+0.000
Top resid features:
is
Token is
Feature activation+0.019
Top resid features:
actually
Token actually
Feature activation+0.037
Top resid features:
not
Token not
Feature activation+0.343
Top resid features:
as
Token as
Feature activation+1.049
Top resid features:
crazy
Token crazy
Feature activation+0.348
Top resid features:
as
Token as
Feature activation+1.799
Top resid features:
you
Token you
Feature activation+0.163
Top resid features:
might
Token might
Feature activation+0.000
Top resid features:
think
Token think
Feature activation+0.000
Top resid features:
âĢĶ
Token âĢĶ
Feature activation+0.000
Top resid features:
it
Token it
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.02

Head 2: 0.04

Head 3: 0.07

Head 4: 0.13

Head 5: 0.06

Head 6: 0.02

Head 7: 0.08

Head 8: 0.06

Head 9: 0.40

Head 10: 0.07

Head 11: 0.02

Positive logits

doubted3.90

doubts3.41

essim3.31

imply3.27

conject3.26

claimed3.26

rumor3.25

rumours3.25

implied3.25

feared3.21

Rum3.21

errone3.18

uggest3.13

conjecture3.08

implying3.08

speculated3.06

rumors3.05

falsely3.02

doubt2.99

misled2.98

Negative logits

ZIP-2.68

collect-2.54

collisions-2.50

coordination-2.47

overlap-2.44

assemb-2.36

ACTIONS-2.30

teamwork-2.29

jurisdiction-2.28

coordinate-2.28

vector-2.25

pairs-2.25

stride-2.24

workflow-2.24

colleg-2.23

coordinating-2.23

comb-2.21

subset-2.18

collecting-2.17

vectors-2.16

INTERVAL 4.772 - 5.302
CONTAINS 0.000%

are
Token are
Feature activation+0.000
not
Token not
Feature activation+0.000
as
Token as
Feature activation+0.000
bullish
Token bullish
Feature activation+0.000
as
Token as
Feature activation+3.705
people
Token people
Feature activation+5.302
thought
Token thought
Feature activation+3.554
and
Token and
Feature activation+0.047
they
Token they
Feature activation+0.000
bought
Token bought
Feature activation+0.000
back
Token back
Feature activation+0.000

INTERVAL 4.242 - 4.772
CONTAINS 0.000%

skulls
Token skulls
Feature activation+0.000
is
Token is
Feature activation+0.000
even
Token even
Feature activation+0.000
more
Token more
Feature activation+0.000
complicated
Token complicated
Feature activation+0.000
than
Token than
Feature activation+4.450
it
Token it
Feature activation+3.427
looks
Token looks
Feature activation+4.299
.
Token.
Feature activation+0.000
Thanks
Token Thanks
Feature activation+0.000
to
Token to
Feature activation+0.000
even
Token even
Feature activation+0.000
more
Token more
Feature activation+0.000
complicated
Token complicated
Feature activation+0.000
than
Token than
Feature activation+4.450
it
Token it
Feature activation+3.427
looks
Token looks
Feature activation+4.299
.
Token.
Feature activation+0.000
Thanks
Token Thanks
Feature activation+0.000
to
Token to
Feature activation+0.000
C
Token C
Feature activation+0.000
aj
Tokenaj
Feature activation+0.000
much
Token much
Feature activation+0.000
more
Token more
Feature activation+0.000
complicated
Token complicated
Feature activation+0.000
than
Token than
Feature activation+1.069
it
Token it
Feature activation+2.410
appears
Token appears
Feature activation+4.499
,
Token,
Feature activation+0.364
and
Token and
Feature activation+0.000
it
Token it
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
quite
Token quite
Feature activation+0.000
as
Token as
Feature activation+0.000
unlikely
Token unlikely
Feature activation+0.000
as
Token as
Feature activation+3.369
it
Token it
Feature activation+3.101
seems
Token seems
Feature activation+4.736
.
Token.
Feature activation+0.000
From
Token From
Feature activation+0.000
the
Token the
Feature activation+0.000
Battle
Token Battle
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 3.712 - 4.242
CONTAINS 0.000%

bigger
Token bigger
Feature activation+0.000
than
Token than
Feature activation+0.764
the
Token the
Feature activation+0.524
New
Token New
Feature activation+0.000
Republic
Token Republic
Feature activation+1.857
hopes
Token hopes
Feature activation+3.743
it
Token it
Feature activation+1.139
is
Token is
Feature activation+1.741
.
Token.
Feature activation+0.000
H
Token H
Feature activation+0.000
iding
Tokeniding
Feature activation+0.000
work
Token work
Feature activation+0.000
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+1.600
they
Token they
Feature activation+3.801
had
Token had
Feature activation+3.431
wished
Token wished
Feature activation+1.657
,
Token,
Feature activation+0.168
and
Token and
Feature activation+0.000
left
Token left
Feature activation+0.000
be
Token be
Feature activation+0.000
as
Token as
Feature activation+0.000
crazy
Token crazy
Feature activation+0.000
as
Token as
Feature activation+2.994
it
Token it
Feature activation+3.193
sounds
Token sounds
Feature activation+3.858
,
Token,
Feature activation+0.000
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.000
F
Token F
Feature activation+0.000
OSS
TokenOSS
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+0.000
childhood
Token childhood
Feature activation+0.000
mortality
Token mortality
Feature activation+0.000
than
Token than
Feature activation+1.538
previously
Token previously
Feature activation+3.949
thought
Token thought
Feature activation+3.686
.
Token.
Feature activation+0.000
If
Token If
Feature activation+0.000
correct
Token correct
Feature activation+0.000
,
Token,
Feature activation+0.000

INTERVAL 3.181 - 3.712
CONTAINS 0.002%

to
Token to
Feature activation+0.000
adult
Token adult
Feature activation+0.000
health
Token health
Feature activation+0.000
than
Token than
Feature activation+1.584
we
Token we
Feature activation+2.957
had
Token had
Feature activation+3.260
previously
Token previously
Feature activation+3.196
imagined
Token imagined
Feature activation+2.779
.
Token.
Feature activation+0.000
Again
Token Again
Feature activation+0.000
,
Token,
Feature activation+0.000
maintained
Token maintained
Feature activation+0.000
she
Token she
Feature activation+0.000
wasn
Token wasn
Feature activation+0.000
't
Token't
Feature activation+0.000
who
Token who
Feature activation+0.037
she
Token she
Feature activation+3.416
claimed
Token claimed
Feature activation+2.605
to
Token to
Feature activation+1.200
be
Token be
Feature activation+1.923
,
Token,
Feature activation+0.000
even
Token even
Feature activation+0.000
as
Token as
Feature activation+0.711
lawyers
Token lawyers
Feature activation+0.512
and
Token and
Feature activation+0.000
victim
Token victim
Feature activation+0.000
advocates
Token advocates
Feature activation+1.499
wanted
Token wanted
Feature activation+3.290
him
Token him
Feature activation+1.374
to
Token to
Feature activation+1.504
be
Token be
Feature activation+0.626
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
out
Token out
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+1.600
they
Token they
Feature activation+3.801
had
Token had
Feature activation+3.431
wished
Token wished
Feature activation+1.657
,
Token,
Feature activation+0.168
and
Token and
Feature activation+0.000
left
Token left
Feature activation+0.000
the
Token the
Feature activation+0.000
not
Token not
Feature activation+0.000
as
Token as
Feature activation+0.000
crazy
Token crazy
Feature activation+0.000
as
Token as
Feature activation+2.101
you
Token you
Feature activation+3.390
might
Token might
Feature activation+3.235
think
Token think
Feature activation+2.461
âĢĶ
Token âĢĶ
Feature activation+0.000
it
Token it
Feature activation+0.000
is
Token is
Feature activation+0.000
really
Token really
Feature activation+0.000

INTERVAL 2.651 - 3.181
CONTAINS 0.002%

Ċ
TokenĊ
Feature activation+0.000
Less
TokenLess
Feature activation+0.000
hard
Token hard
Feature activation+0.000
than
Token than
Feature activation+0.064
you
Token you
Feature activation+3.203
think
Token think
Feature activation+2.751
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
lot
Token lot
Feature activation+0.000
of
Token of
Feature activation+0.000
complete
Token complete
Feature activation+0.000
or
Token or
Feature activation+0.000
comprehensive
Token comprehensive
Feature activation+0.000
as
Token as
Feature activation+1.064
they
Token they
Feature activation+2.182
appear
Token appear
Feature activation+3.002
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
the
Token the
Feature activation+0.000
ideas
Token ideas
Feature activation+0.000
that
Token that
Feature activation+0.000
box
Token box
Feature activation+0.000
office
Token office
Feature activation+0.000
as
Token as
Feature activation+1.436
its
Token its
Feature activation+0.331
producers
Token producers
Feature activation+3.695
hoped
Token hoped
Feature activation+3.126
,
Token,
Feature activation+0.682
but
Token but
Feature activation+0.000
instead
Token instead
Feature activation+0.000
of
Token of
Feature activation+0.000
fading
Token fading
Feature activation+0.000
as
Token as
Feature activation+0.000
short
Token short
Feature activation+0.000
as
Token as
Feature activation+0.786
most
Token most
Feature activation+1.808
people
Token people
Feature activation+3.414
make
Token make
Feature activation+2.874
it
Token it
Feature activation+1.397
out
Token out
Feature activation+1.673
to
Token to
Feature activation+2.362
be
Token be
Feature activation+2.024
(
Token (
Feature activation+0.000
not
Token not
Feature activation+0.235
a
Token a
Feature activation+0.030
continuous
Token continuous
Feature activation+0.000
network
Token network
Feature activation+0.000
like
Token like
Feature activation+2.831
the
Token the
Feature activation+2.690
blood
Token blood
Feature activation+0.000
vessels
Token vessels
Feature activation+1.629
of
Token of
Feature activation+0.489
the
Token the
Feature activation+0.277
body
Token body
Feature activation+1.917

INTERVAL 2.121 - 2.651
CONTAINS 0.003%

as
Token as
Feature activation+0.000
many
Token many
Feature activation+0.000
b
Token b
Feature activation+0.000
erg
Tokenerg
Feature activation+0.000
s
Tokens
Feature activation+0.219
as
Token as
Feature activation+2.314
San
Token San
Feature activation+0.000
Rafael
Token Rafael
Feature activation+0.523
,
Token,
Feature activation+0.662
but
Token but
Feature activation+0.000
it
Token it
Feature activation+0.000
as
Token as
Feature activation+0.000
in
Token in
Feature activation+0.000
-
Token-
Feature activation+0.000
depth
Tokendepth
Feature activation+0.000
as
Token as
Feature activation+0.034
I
Token I
Feature activation+2.477
would
Token would
Feature activation+1.860
like
Token like
Feature activation+1.900
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
might
Token might
Feature activation+0.000
not
Token not
Feature activation+0.000
be
Token be
Feature activation+0.000
as
Token as
Feature activation+0.000
rough
Token rough
Feature activation+0.000
as
Token as
Feature activation+2.175
Reynolds
Token Reynolds
Feature activation+0.556
'
Token'
Feature activation+0.255
ple
Token ple
Feature activation+0.000
be
Tokenbe
Feature activation+0.000
summer
Token summer
Feature activation+0.000
s
Tokens
Feature activation+0.000
not
Token not
Feature activation+0.000
as
Token as
Feature activation+0.000
bad
Token bad
Feature activation+0.000
as
Token as
Feature activation+1.231
it
Token it
Feature activation+2.209
was
Token was
Feature activation+1.283
last
Token last
Feature activation+0.000
year
Token year
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
she
Token she
Feature activation+0.000
wasn
Token wasn
Feature activation+0.000
't
Token't
Feature activation+0.000
who
Token who
Feature activation+0.037
she
Token she
Feature activation+3.416
claimed
Token claimed
Feature activation+2.605
to
Token to
Feature activation+1.200
be
Token be
Feature activation+1.923
,
Token,
Feature activation+0.000
even
Token even
Feature activation+0.000
though
Token though
Feature activation+0.000

INTERVAL 1.591 - 2.121
CONTAINS 0.004%

has
Token has
Feature activation+0.000
changed
Token changed
Feature activation+0.000
more
Token more
Feature activation+0.000
times
Token times
Feature activation+0.000
than
Token than
Feature activation+0.623
I
Token I
Feature activation+2.082
've
Token've
Feature activation+0.000
changed
Token changed
Feature activation+0.000
under
Token under
Feature activation+0.000
pants
Tokenpants
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
big
Token big
Feature activation+0.000
a
Token a
Feature activation+0.000
problem
Token problem
Feature activation+0.000
as
Token as
Feature activation+1.568
they
Token they
Feature activation+2.237
were
Token were
Feature activation+1.723
at
Token at
Feature activation+0.338
launch
Token launch
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
as
Token as
Feature activation+0.000
well
Token well
Feature activation+0.000
as
Token as
Feature activation+0.511
we
Token we
Feature activation+1.545
would
Token would
Feature activation+1.417
like
Token like
Feature activation+1.937
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
police
Token police
Feature activation+0.000
were
Token were
Feature activation+0.000
not
Token not
Feature activation+0.000
as
Token as
Feature activation+0.000
important
Token important
Feature activation+0.000
as
Token as
Feature activation+2.019
the
Token the
Feature activation+0.383
charge
Token charge
Feature activation+0.000
that
Token that
Feature activation+0.213
they
Token they
Feature activation+0.000
were
Token were
Feature activation+0.000
beliefs
Token beliefs
Feature activation+0.000
or
Token or
Feature activation+0.000
gods
Token gods
Feature activation+0.000
randomly
Token randomly
Feature activation+0.000
as
Token as
Feature activation+1.194
it
Token it
Feature activation+1.612
serves
Token serves
Feature activation+0.000
no
Token no
Feature activation+0.000
inherent
Token inherent
Feature activation+0.000
purpose
Token purpose
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 1.060 - 1.591
CONTAINS 0.008%

compassionate
Token compassionate
Feature activation+0.000
than
Token than
Feature activation+0.634
it
Token it
Feature activation+0.866
would
Token would
Feature activation+1.074
have
Token have
Feature activation+1.125
been
Token been
Feature activation+1.091
with
Token with
Feature activation+0.136
you
Token you
Feature activation+0.000
guiding
Token guiding
Feature activation+0.000
it
Token it
Feature activation+0.000
.
Token.
Feature activation+0.000
hypothesis
Token hypothesis
Feature activation+0.000
as
Token as
Feature activation+0.830
the
Token the
Feature activation+0.647
Catholic
Token Catholic
Feature activation+0.506
Church
Token Church
Feature activation+1.355
demanded
Token demanded
Feature activation+1.156
.
Token.
Feature activation+0.000
Still
Token Still
Feature activation+0.000
,
Token,
Feature activation+0.000
historians
Token historians
Feature activation+0.000
have
Token have
Feature activation+0.000
hands
Token hands
Feature activation+0.000
has
Token has
Feature activation+0.000
continued
Token continued
Feature activation+0.000
,
Token,
Feature activation+0.000
faster
Token faster
Feature activation+0.000
than
Token than
Feature activation+1.079
the
Token the
Feature activation+0.000
rate
Token rate
Feature activation+0.494
of
Token of
Feature activation+0.000
overall
Token overall
Feature activation+0.000
economic
Token economic
Feature activation+0.000
ating
Tokenating
Feature activation+0.000
as
Token as
Feature activation+0.000
usual
Token usual
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.264
is
Token is
Feature activation+1.210
the
Token the
Feature activation+0.350
land
Token land
Feature activation+0.000
er
Tokener
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
up
Token up
Feature activation+0.000
rather
Token rather
Feature activation+0.000
than
Token than
Feature activation+0.000
scaled
Token scaled
Feature activation+0.000
back
Token back
Feature activation+0.000
as
Token as
Feature activation+1.400
part
Token part
Feature activation+0.049
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
government
Token government
Feature activation+0.197
âĢ
TokenâĢ
Feature activation+0.000

INTERVAL 0.530 - 1.060
CONTAINS 0.016%

more
Token more
Feature activation+0.000
intertwined
Token intertwined
Feature activation+0.000
than
Token than
Feature activation+1.056
at
Token at
Feature activation+1.655
any
Token any
Feature activation+0.759
time
Token time
Feature activation+0.664
in
Token in
Feature activation+0.132
human
Token human
Feature activation+0.000
history
Token history
Feature activation+0.000
,"
Token,"
Feature activation+0.000
Mr
Token Mr
Feature activation+0.000
it
Token it
Feature activation+0.000
is
Token is
Feature activation+0.000
skill
Token skill
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.459
I
Token I
Feature activation+0.674
think
Token think
Feature activation+0.000
most
Token most
Feature activation+0.000
North
Token North
Feature activation+0.000
American
Token American
Feature activation+0.000
players
Token players
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
doesn
Token doesn
Feature activation+0.000
't
Token't
Feature activation+0.000
quite
Token quite
Feature activation+0.199
matter
Token matter
Feature activation+0.584
.
Token.
Feature activation+0.000
David
Token David
Feature activation+0.000
Roberts
Token Roberts
Feature activation+0.000
cast
Token cast
Feature activation+0.000
it
Token it
Feature activation+0.000
wicked
Token wicked
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
in
Token in
Feature activation+0.000
humane
Tokenhumane
Feature activation+0.000
as
Token as
Feature activation+0.871
any
Token any
Feature activation+0.485
of
Token of
Feature activation+0.853
history
Token history
Feature activation+0.484
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
requirement
Token requirement
Feature activation+0.000
of
Token of
Feature activation+0.000
standard
Token standard
Feature activation+0.000
building
Token building
Feature activation+0.000
codes
Token codes
Feature activation+0.000
,
Token,
Feature activation+0.596
stating
Token stating
Feature activation+0.000
that
Token that
Feature activation+0.000
"
Token "
Feature activation+0.000
live
Tokenlive
Feature activation+0.000
loads
Token loads
Feature activation+0.000

INTERVAL 0.000 - 0.530
CONTAINS 99.965%

well
Token well
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
National
Token National
Feature activation+0.000
Hockey
Token Hockey
Feature activation+0.000
League
Token League
Feature activation+0.000
awards
Token awards
Feature activation+0.000
three
Token three
Feature activation+0.000
stars
Token stars
Feature activation+0.000
during
Token during
Feature activation+0.000
every
Token every
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
an
Token an
Feature activation+0.000
opportunity
Token opportunity
Feature activation+0.000
that
Token that
Feature activation+0.000
Spotify
Token Spotify
Feature activation+0.000
itself
Token itself
Feature activation+0.000
has
Token has
Feature activation+0.000
not
Token not
Feature activation+0.000
tapped
Token tapped
Feature activation+0.000
,
Token,
Feature activation+0.000
international
Token international
Feature activation+0.000
development
Token development
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
global
Token global
Feature activation+0.000
vacuum
Token vacuum
Feature activation+0.000
of
Token of
Feature activation+0.000
accountability
Token accountability
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
new
Token new
Feature activation+0.000
creating
Token creating
Feature activation+0.000
variables
Token variables
Feature activation+0.000
and
Token and
Feature activation+0.000
logging
Token logging
Feature activation+0.000
it
Token it
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
console
Token console
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
digital
Token digital
Feature activation+0.000
rights
Token rights
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
E
Token E
Feature activation+0.000
PL
TokenPL
Feature activation+0.000
for
Token for
Feature activation+0.000
three
Token three
Feature activation+0.000
years
Token years
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 6 in H8.9: (feature 17751)

TOP ACTIVATIONS
MAX = 5.175

political
Token political
Feature activation+0.000
vision
Token vision
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+5.175
exactly
Token exactly
Feature activation+0.308
?
Token?
Feature activation+2.124
Ċ
TokenĊ
Feature activation+0.071
Ċ
TokenĊ
Feature activation+1.449
V
TokenV
Feature activation+0.000
minority
Token minority
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+4.137
they
Token they
Feature activation+4.937
ask
Token ask
Feature activation+3.111
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
color
Token color
Feature activation+0.000
Spect
Token Spect
Feature activation+0.000
er
Tokener
Feature activation+0.000
of
Token of
Feature activation+0.000
Capital
Token Capital
Feature activation+0.000
,
Token,
Feature activation+1.932
answers
Token answers
Feature activation+4.352
this
Token this
Feature activation+3.100
question
Token question
Feature activation+1.269
here
Token here
Feature activation+0.000
,
Token,
Feature activation+0.129
as
Token as
Feature activation+0.000
income
Tokenincome
Feature activation+0.000
minority
Token minority
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+4.137
they
Token they
Feature activation+4.937
ask
Token ask
Feature activation+3.111
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
?
Token?
Feature activation+1.781
Our
Token Our
Feature activation+0.979
second
Token second
Feature activation+0.981
experiment
Token experiment
Feature activation+0.043
answered
Token answered
Feature activation+2.883
these
Token these
Feature activation+4.029
questions
Token questions
Feature activation+1.556
.
Token.
Feature activation+0.175
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
What
Token What
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
beautiful
Token beautiful
Feature activation+0.000
,
Token,
Feature activation+2.820
he
Token he
Feature activation+3.632
said
Token said
Feature activation+1.550
,
Token,
Feature activation+1.262
is
Token is
Feature activation+0.417
if
Token if
Feature activation+0.000
someone
Token someone
Feature activation+0.000
of
Token of
Feature activation+0.038
the
Token the
Feature activation+0.041
best
Token best
Feature activation+0.018
to
Token to
Feature activation+0.406
find
Token find
Feature activation+1.722
out
Token out
Feature activation+3.498
-
Token -
Feature activation+0.161
Lucky
Token Lucky
Feature activation+0.000
Peach
Token Peach
Feature activation+0.000
,
Token,
Feature activation+0.000
Put
Token Put
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
And
TokenAnd
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+3.466
pray
Token pray
Feature activation+0.738
tell
Token tell
Feature activation+0.627
,
Token,
Feature activation+0.000
did
Token did
Feature activation+0.000
they
Token they
Feature activation+0.000
And
Token And
Feature activation+0.903
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
president
Token president
Feature activation+0.000
won
Token won
Feature activation+0.000
?
Token?
Feature activation+3.368
Ċ
TokenĊ
Feature activation+0.063
Ċ
TokenĊ
Feature activation+2.093
It
TokenIt
Feature activation+1.297
will
Token will
Feature activation+1.234
not
Token not
Feature activation+1.070
I
Token I
Feature activation+1.437
would
Token would
Feature activation+0.502
really
Token really
Feature activation+0.428
like
Token like
Feature activation+0.976
to
Token to
Feature activation+1.275
know
Token know
Feature activation+3.324
.
Token.
Feature activation+0.378
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.308
âĨ
TokenâĨ
Feature activation+0.000
IJ
TokenIJ
Feature activation+0.000
?
Token?
Feature activation+1.416
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.087
One
TokenOne
Feature activation+1.067
potential
Token potential
Feature activation+2.087
answer
Token answer
Feature activation+3.278
is
Token is
Feature activation+0.059
phot
Token phot
Feature activation+0.000
onic
Tokenonic
Feature activation+0.000
propulsion
Token propulsion
Feature activation+0.000
.
Token.
Feature activation+0.000
it
Token it
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
championship
Token championship
Feature activation+0.000
round
Token round
Feature activation+0.000
is
Token is
Feature activation+3.188
truly
Token truly
Feature activation+2.253
up
Token up
Feature activation+1.384
in
Token in
Feature activation+0.550
the
Token the
Feature activation+0.000
air
Token air
Feature activation+1.157
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+4.137
they
Token they
Feature activation+4.937
ask
Token ask
Feature activation+3.111
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
color
Token color
Feature activation+0.000
or
Token or
Feature activation+0.000
er
Tokener
Feature activation+0.000
of
Token of
Feature activation+0.000
Capital
Token Capital
Feature activation+0.000
,
Token,
Feature activation+1.932
answers
Token answers
Feature activation+4.352
this
Token this
Feature activation+3.100
question
Token question
Feature activation+1.269
here
Token here
Feature activation+0.000
,
Token,
Feature activation+0.129
as
Token as
Feature activation+0.000
well
Token well
Feature activation+0.012
t
Tokent
Feature activation+1.184
take
Token take
Feature activation+0.210
much
Token much
Feature activation+0.556
to
Token to
Feature activation+0.303
answer
Token answer
Feature activation+2.969
these
Token these
Feature activation+3.097
questions
Token questions
Feature activation+1.310
.
Token.
Feature activation+0.231
All
Token All
Feature activation+0.272
you
Token you
Feature activation+0.236
need
Token need
Feature activation+0.206
les
Tokenles
Feature activation+0.755
me
Token me
Feature activation+0.575
,
Token,
Feature activation+1.851
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.005
he
Token he
Feature activation+3.086
says
Token says
Feature activation+0.551
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.095
As
TokenAs
Feature activation+0.101
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
a
Token a
Feature activation+0.000
bad
Token bad
Feature activation+0.000
person
Token person
Feature activation+0.000
?
Token?
Feature activation+3.030
Even
Token Even
Feature activation+0.417
if
Token if
Feature activation+0.164
I
Token I
Feature activation+0.000
believed
Token believed
Feature activation+0.000
deep
Token deep
Feature activation+0.000
,
Token,
Feature activation+0.253
or
Token or
Feature activation+0.128
even
Token even
Feature activation+0.000
someone
Token someone
Feature activation+0.000
taller
Token taller
Feature activation+0.000
?
Token?
Feature activation+2.985
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.346
Advertisement
TokenAdvertisement
Feature activation+0.135
Ċ
TokenĊ
Feature activation+0.151
Ċ
TokenĊ
Feature activation+0.808
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.184
take
Token take
Feature activation+0.210
much
Token much
Feature activation+0.556
to
Token to
Feature activation+0.303
answer
Token answer
Feature activation+2.969
these
Token these
Feature activation+3.097
questions
Token questions
Feature activation+1.310
.
Token.
Feature activation+0.231
All
Token All
Feature activation+0.272
you
Token you
Feature activation+0.236
difference
Token difference
Feature activation+0.541
?
Token?
Feature activation+1.781
Our
Token Our
Feature activation+0.979
second
Token second
Feature activation+0.981
experiment
Token experiment
Feature activation+0.043
answered
Token answered
Feature activation+2.883
these
Token these
Feature activation+4.029
questions
Token questions
Feature activation+1.556
.
Token.
Feature activation+0.175
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

Top DFA by src position
MAX = 3.626

their
Token their
Feature activation+0.003
Top resid features:
political
Token political
Feature activation-0.003
Top resid features:
vision
Token vision
Feature activation+0.004
Top resid features:
.
Token.
Feature activation-0.007
Top resid features:
But
Token But
Feature activation+0.161
Top resid features:
why
Token why
Feature activation+3.626
Top resid features:
,
Token,
Feature activation+1.928
Top resid features:
exactly
Token exactly
Feature activation+0.000
Top resid features:
?
Token?
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
income
Tokenincome
Feature activation+0.003
Top resid features:
minority
Token minority
Feature activation+0.000
Top resid features:
youth
Token youth
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Why
Token Why
Feature activation+2.833
Top resid features:
,
Token,
Feature activation+2.499
Top resid features:
they
Token they
Feature activation+0.018
Top resid features:
ask
Token ask
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
should
Token should
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.002
Top resid features:
But
Token But
Feature activation+0.088
Top resid features:
why
Token why
Feature activation+1.110
Top resid features:
,
Token,
Feature activation+0.831
Top resid features:
exactly
Token exactly
Feature activation+0.232
Top resid features:
?
Token?
Feature activation+2.404
Top resid features:
Ċ
TokenĊ
Feature activation+0.024
Top resid features:
Ċ
TokenĊ
Feature activation+0.056
Top resid features:
V
TokenV
Feature activation+0.003
Top resid features:
ive
Tokenive
Feature activation-0.000
Top resid features:
k
Tokenk
Feature activation+0.001
Top resid features:
-
Token-
Feature activation+0.000
Top resid features:
income
Tokenincome
Feature activation-0.003
Top resid features:
minority
Token minority
Feature activation-0.005
Top resid features:
youth
Token youth
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.018
Top resid features:
Why
Token Why
Feature activation+3.397
Top resid features:
,
Token,
Feature activation+1.115
Top resid features:
they
Token they
Feature activation+0.000
Top resid features:
ask
Token ask
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
should
Token should
Feature activation+0.000
Top resid features:
aut
Tokenaut
Feature activation+0.000
Top resid features:
istic
Tokenistic
Feature activation+0.000
Top resid features:
people
Token people
Feature activation+0.002
Top resid features:
?
Token?
Feature activation+0.373
Top resid features:
And
Token And
Feature activation+0.171
Top resid features:
does
Token does
Feature activation+2.518
Top resid features:
it
Token it
Feature activation+0.060
Top resid features:
make
Token make
Feature activation+0.054
Top resid features:
a
Token a
Feature activation+0.013
Top resid features:
difference
Token difference
Feature activation+0.073
Top resid features:
?
Token?
Feature activation+0.253
Top resid features:
.
Token.
Feature activation+0.001
Top resid features:
What
Token What
Feature activation+0.418
Top resid features:
would
Token would
Feature activation+0.985
Top resid features:
be
Token be
Feature activation+0.183
Top resid features:
beautiful
Token beautiful
Feature activation+0.044
Top resid features:
,
Token,
Feature activation+2.614
Top resid features:
he
Token he
Feature activation+0.002
Top resid features:
said
Token said
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
is
Token is
Feature activation+0.000
Top resid features:
if
Token if
Feature activation+0.000
Top resid features:
Ŀ
TokenĿ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation+0.002
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.032
Top resid features:
How
TokenHow
Feature activation+0.674
Top resid features:
do
Token do
Feature activation+2.325
Top resid features:
the
Token the
Feature activation+0.030
Top resid features:
contemporary
Token contemporary
Feature activation+0.003
Top resid features:
food
Token food
Feature activation+0.001
Top resid features:
magazines
Token magazines
Feature activation+0.002
Top resid features:
use
Token use
Feature activation+0.159
Top resid features:
âĢ
TokenâĢ
Feature activation+0.002
Top resid features:
Ŀ
TokenĿ
Feature activation+0.001
Top resid features:
Ċ
TokenĊ
Feature activation+0.037
Top resid features:
Ċ
TokenĊ
Feature activation+0.031
Top resid features:
And
TokenAnd
Feature activation-0.004
Top resid features:
why
Token why
Feature activation+3.088
Top resid features:
,
Token,
Feature activation+0.756
Top resid features:
pray
Token pray
Feature activation+0.000
Top resid features:
tell
Token tell
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
did
Token did
Feature activation+0.000
Top resid features:
coming
Token coming
Feature activation+0.044
Top resid features:
to
Token to
Feature activation+0.006
Top resid features:
an
Token an
Feature activation+0.002
Top resid features:
end
Token end
Feature activation+0.020
Top resid features:
?
Token?
Feature activation+0.511
Top resid features:
And
Token And
Feature activation+1.820
Top resid features:
has
Token has
Feature activation+0.547
Top resid features:
the
Token the
Feature activation+0.024
Top resid features:
president
Token president
Feature activation+0.186
Top resid features:
won
Token won
Feature activation+0.052
Top resid features:
?
Token?
Feature activation+0.284
Top resid features:
isn
Token isn
Feature activation+0.065
Top resid features:
't
Token't
Feature activation+0.057
Top resid features:
?
Token?
Feature activation+0.081
Top resid features:
Ċ
TokenĊ
Feature activation+0.041
Top resid features:
Ċ
TokenĊ
Feature activation+0.038
Top resid features:
What
TokenWhat
Feature activation+0.939
Top resid features:
wouldn
Token wouldn
Feature activation+0.309
Top resid features:
't
Token't
Feature activation+0.155
Top resid features:
you
Token you
Feature activation+0.047
Top resid features:
do
Token do
Feature activation+0.091
Top resid features:
for
Token for
Feature activation+0.024
Top resid features:
system
Token system
Feature activation-0.001
Top resid features:
.
Token.
Feature activation-0.003
Top resid features:
But
Token But
Feature activation+0.026
Top resid features:
,
Token,
Feature activation+0.036
Top resid features:
how
Token how
Feature activation+0.922
Top resid features:
do
Token do
Feature activation+1.307
Top resid features:
we
Token we
Feature activation+1.077
Top resid features:
take
Token take
Feature activation+0.081
Top resid features:
that
Token that
Feature activation+0.125
Top resid features:
next
Token next
Feature activation+0.005
Top resid features:
step
Token step
Feature activation+0.024
Top resid features:
Coast
Token Coast
Feature activation+0.003
Top resid features:
lumin
Token lumin
Feature activation+0.001
Top resid features:
aries
Tokenaries
Feature activation+0.005
Top resid features:
.
Token.
Feature activation+0.014
Top resid features:
Who
Token Who
Feature activation+1.120
Top resid features:
will
Token will
Feature activation+2.155
Top resid features:
make
Token make
Feature activation+0.210
Top resid features:
it
Token it
Feature activation+0.023
Top resid features:
to
Token to
Feature activation+0.002
Top resid features:
the
Token the
Feature activation+0.009
Top resid features:
championship
Token championship
Feature activation+0.000
Top resid features:
income
Tokenincome
Feature activation-0.004
Top resid features:
minority
Token minority
Feature activation-0.002
Top resid features:
youth
Token youth
Feature activation+0.006
Top resid features:
.
Token.
Feature activation+0.002
Top resid features:
Why
Token Why
Feature activation+0.667
Top resid features:
,
Token,
Feature activation+2.850
Top resid features:
they
Token they
Feature activation+0.026
Top resid features:
ask
Token ask
Feature activation+0.007
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
should
Token should
Feature activation+0.000
Top resid features:
skin
Token skin
Feature activation+0.000
Top resid features:
.
Token.
Feature activation-0.014
Top resid features:
But
Token But
Feature activation+0.143
Top resid features:
why
Token why
Feature activation+1.231
Top resid features:
,
Token,
Feature activation+0.214
Top resid features:
exactly
Token exactly
Feature activation+0.173
Top resid features:
?
Token?
Feature activation+1.764
Top resid features:
Ċ
TokenĊ
Feature activation+0.027
Top resid features:
Ċ
TokenĊ
Feature activation+0.045
Top resid features:
V
TokenV
Feature activation-0.001
Top resid features:
ive
Tokenive
Feature activation+0.000
Top resid features:
k
Tokenk
Feature activation+0.001
Top resid features:
erase
Token erase
Feature activation+0.013
Top resid features:
a
Token a
Feature activation+0.003
Top resid features:
previous
Token previous
Feature activation-0.001
Top resid features:
language
Token language
Feature activation+0.002
Top resid features:
?
Token?
Feature activation+0.062
Top resid features:
Can
Token Can
Feature activation+0.567
Top resid features:
I
Token I
Feature activation+0.028
Top resid features:
borrow
Token borrow
Feature activation+0.001
Top resid features:
structures
Token structures
Feature activation-0.001
Top resid features:
without
Token without
Feature activation+0.001
Top resid features:
fatal
Token fatal
Feature activation+0.004
Top resid features:
just
Token just
Feature activation-0.004
Top resid features:
about
Token about
Feature activation-0.005
Top resid features:
any
Token any
Feature activation+0.000
Top resid features:
reason
Token reason
Feature activation-0.010
Top resid features:
.
Token.
Feature activation-0.016
Top resid features:
Why
Token Why
Feature activation+1.693
Top resid features:
they
Token they
Feature activation+0.051
Top resid features:
don
Token don
Feature activation+0.015
Top resid features:
âĢ
TokenâĢ
Feature activation+0.005
Top resid features:
Ļ
TokenĻ
Feature activation+0.004
Top resid features:
t
Tokent
Feature activation+0.008
Top resid features:
bad
Token bad
Feature activation-0.002
Top resid features:
person
Token person
Feature activation-0.004
Top resid features:
,
Token,
Feature activation+0.006
Top resid features:
does
Token does
Feature activation+0.042
Top resid features:
that
Token that
Feature activation+0.490
Top resid features:
mean
Token mean
Feature activation+2.429
Top resid features:
you
Token you
Feature activation+0.045
Top resid features:
âĢ
TokenâĢ
Feature activation+0.038
Top resid features:
Ļ
TokenĻ
Feature activation+0.192
Top resid features:
re
Tokenre
Feature activation+0.034
Top resid features:
a
Token a
Feature activation+0.002
Top resid features:
else
Token else
Feature activation-0.008
Top resid features:
look
Token look
Feature activation-0.005
Top resid features:
short
Token short
Feature activation+0.001
Top resid features:
.
Token.
Feature activation+0.012
Top resid features:
Could
Token Could
Feature activation+0.288
Top resid features:
I
Token I
Feature activation+1.222
Top resid features:
look
Token look
Feature activation+0.004
Top resid features:
taller
Token taller
Feature activation+0.011
Top resid features:
than
Token than
Feature activation+0.003
Top resid features:
someone
Token someone
Feature activation+0.016
Top resid features:
my
Token my
Feature activation-0.001
Top resid features:
All
Token All
Feature activation+0.025
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
which
Token which
Feature activation+0.117
Top resid features:
answer
Token answer
Feature activation+0.015
Top resid features:
:
Token:
Feature activation+0.044
Top resid features:
How
Token How
Feature activation+0.772
Top resid features:
difficult
Token difficult
Feature activation+0.040
Top resid features:
will
Token will
Feature activation+0.281
Top resid features:
it
Token it
Feature activation+0.048
Top resid features:
be
Token be
Feature activation+0.062
Top resid features:
,
Token,
Feature activation+0.142
Top resid features:
aut
Tokenaut
Feature activation+0.000
Top resid features:
istic
Tokenistic
Feature activation-0.000
Top resid features:
people
Token people
Feature activation+0.002
Top resid features:
?
Token?
Feature activation+0.414
Top resid features:
And
Token And
Feature activation+0.192
Top resid features:
does
Token does
Feature activation+1.206
Top resid features:
it
Token it
Feature activation+0.247
Top resid features:
make
Token make
Feature activation+0.062
Top resid features:
a
Token a
Feature activation+0.013
Top resid features:
difference
Token difference
Feature activation+0.112
Top resid features:
?
Token?
Feature activation+0.329
Top resid features:

Decoder Weights Distribution

Head 0: 0.04

Head 1: 0.02

Head 2: 0.06

Head 3: 0.07

Head 4: 0.04

Head 5: 0.07

Head 6: 0.03

Head 7: 0.04

Head 8: 0.08

Head 9: 0.39

Head 10: 0.12

Head 11: 0.03

Positive logits

answers5.76

Answer5.76

Answer5.71

answer5.33

answered5.12

answer5.07

Answers5.05

Question4.63

questions4.63

question4.57

answ4.57

Questions4.44

swers4.25

answering4.24

question4.01

answered3.97

Question3.80

asks3.73

replies3.64

QUEST3.63

Negative logits

"$:/-2.82

INST-2.33

ortment-2.26

repe-2.23

Notable-2.14

ugal-2.14

withdrawing-2.11

endors-2.10

ilan-2.09

counterpart-2.08

endorsements-2.07

withdrawal-2.06

accomplishment-2.04

cumbers-2.01

blot-2.00

oreal-1.99

Glacier-1.96

acc-1.95

unequ-1.95

estones-1.93

INTERVAL 4.658 - 5.175
CONTAINS 0.000%

political
Token political
Feature activation+0.000
vision
Token vision
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+5.175
exactly
Token exactly
Feature activation+0.308
?
Token?
Feature activation+2.124
Ċ
TokenĊ
Feature activation+0.071
Ċ
TokenĊ
Feature activation+1.449
V
TokenV
Feature activation+0.000
minority
Token minority
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+4.137
they
Token they
Feature activation+4.937
ask
Token ask
Feature activation+3.111
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
color
Token color
Feature activation+0.000

INTERVAL 4.140 - 4.658
CONTAINS 0.000%

Spect
Token Spect
Feature activation+0.000
er
Tokener
Feature activation+0.000
of
Token of
Feature activation+0.000
Capital
Token Capital
Feature activation+0.000
,
Token,
Feature activation+1.932
answers
Token answers
Feature activation+4.352
this
Token this
Feature activation+3.100
question
Token question
Feature activation+1.269
here
Token here
Feature activation+0.000
,
Token,
Feature activation+0.129
as
Token as
Feature activation+0.000

INTERVAL 3.623 - 4.140
CONTAINS 0.000%

What
Token What
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
beautiful
Token beautiful
Feature activation+0.000
,
Token,
Feature activation+2.820
he
Token he
Feature activation+3.632
said
Token said
Feature activation+1.550
,
Token,
Feature activation+1.262
is
Token is
Feature activation+0.417
if
Token if
Feature activation+0.000
someone
Token someone
Feature activation+0.000
?
Token?
Feature activation+1.781
Our
Token Our
Feature activation+0.979
second
Token second
Feature activation+0.981
experiment
Token experiment
Feature activation+0.043
answered
Token answered
Feature activation+2.883
these
Token these
Feature activation+4.029
questions
Token questions
Feature activation+1.556
.
Token.
Feature activation+0.175
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
income
Tokenincome
Feature activation+0.000
minority
Token minority
Feature activation+0.000
youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+4.137
they
Token they
Feature activation+4.937
ask
Token ask
Feature activation+3.111
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000

INTERVAL 3.105 - 3.623
CONTAINS 0.001%

youth
Token youth
Feature activation+0.000
.
Token.
Feature activation+0.000
Why
Token Why
Feature activation+0.000
,
Token,
Feature activation+4.137
they
Token they
Feature activation+4.937
ask
Token ask
Feature activation+3.111
,
Token,
Feature activation+0.000
should
Token should
Feature activation+0.000
skin
Token skin
Feature activation+0.000
color
Token color
Feature activation+0.000
or
Token or
Feature activation+0.000
And
Token And
Feature activation+0.903
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
president
Token president
Feature activation+0.000
won
Token won
Feature activation+0.000
?
Token?
Feature activation+3.368
Ċ
TokenĊ
Feature activation+0.063
Ċ
TokenĊ
Feature activation+2.093
It
TokenIt
Feature activation+1.297
will
Token will
Feature activation+1.234
not
Token not
Feature activation+1.070
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
And
TokenAnd
Feature activation+0.000
why
Token why
Feature activation+0.000
,
Token,
Feature activation+3.466
pray
Token pray
Feature activation+0.738
tell
Token tell
Feature activation+0.627
,
Token,
Feature activation+0.000
did
Token did
Feature activation+0.000
they
Token they
Feature activation+0.000
?
Token?
Feature activation+1.416
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.087
One
TokenOne
Feature activation+1.067
potential
Token potential
Feature activation+2.087
answer
Token answer
Feature activation+3.278
is
Token is
Feature activation+0.059
phot
Token phot
Feature activation+0.000
onic
Tokenonic
Feature activation+0.000
propulsion
Token propulsion
Feature activation+0.000
.
Token.
Feature activation+0.000
it
Token it
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
championship
Token championship
Feature activation+0.000
round
Token round
Feature activation+0.000
is
Token is
Feature activation+3.188
truly
Token truly
Feature activation+2.253
up
Token up
Feature activation+1.384
in
Token in
Feature activation+0.550
the
Token the
Feature activation+0.000
air
Token air
Feature activation+1.157

INTERVAL 2.588 - 3.105
CONTAINS 0.002%

.
Token.
Feature activation+0.000
What
Token What
Feature activation+0.000
would
Token would
Feature activation+0.000
be
Token be
Feature activation+0.000
beautiful
Token beautiful
Feature activation+0.000
,
Token,
Feature activation+2.820
he
Token he
Feature activation+3.632
said
Token said
Feature activation+1.550
,
Token,
Feature activation+1.262
is
Token is
Feature activation+0.417
if
Token if
Feature activation+0.000
difference
Token difference
Feature activation+0.541
?
Token?
Feature activation+1.781
Our
Token Our
Feature activation+0.979
second
Token second
Feature activation+0.981
experiment
Token experiment
Feature activation+0.043
answered
Token answered
Feature activation+2.883
these
Token these
Feature activation+4.029
questions
Token questions
Feature activation+1.556
.
Token.
Feature activation+0.175
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
les
Tokenles
Feature activation+0.755
me
Token me
Feature activation+0.575
,
Token,
Feature activation+1.851
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.005
he
Token he
Feature activation+3.086
says
Token says
Feature activation+0.551
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.095
As
TokenAs
Feature activation+0.101
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+1.184
take
Token take
Feature activation+0.210
much
Token much
Feature activation+0.556
to
Token to
Feature activation+0.303
answer
Token answer
Feature activation+2.969
these
Token these
Feature activation+3.097
questions
Token questions
Feature activation+1.310
.
Token.
Feature activation+0.231
All
Token All
Feature activation+0.272
you
Token you
Feature activation+0.236
expect
Token expect
Feature activation+0.000
?
Token?
Feature activation+0.557
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.218
the
Token the
Feature activation+2.181
doctor
Token doctor
Feature activation+2.599
replied
Token replied
Feature activation+2.554
.
Token.
Feature activation+0.263
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.211
C
TokenC
Feature activation+0.000

INTERVAL 2.070 - 2.588
CONTAINS 0.004%

to
Token to
Feature activation+0.000
recreate
Token recreate
Feature activation+0.000
an
Token an
Feature activation+0.000
object
Token object
Feature activation+0.000
,
Token,
Feature activation+0.188
he
Token he
Feature activation+2.333
said
Token said
Feature activation+1.634
,
Token,
Feature activation+0.234
it
Token it
Feature activation+0.508
doesn
Token doesn
Feature activation+0.140
âĢ
TokenâĢ
Feature activation+0.000
the
Token the
Feature activation+0.023
right
Token right
Feature activation+0.000
kind
Token kind
Feature activation+0.000
of
Token of
Feature activation+0.000
action
Token action
Feature activation+0.000
?
Token?
Feature activation+2.418
Extrem
Token Extrem
Feature activation+0.000
ism
Tokenism
Feature activation+0.000
researchers
Token researchers
Feature activation+1.046
said
Token said
Feature activation+1.178
that
Token that
Feature activation+0.639
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
m
Tokenm
Feature activation+0.000
not
Token not
Feature activation+0.063
telling
Token telling
Feature activation+1.081
you
Token you
Feature activation+2.527
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
she
Token she
Feature activation+1.205
said
Token said
Feature activation+0.000
period
Token period
Feature activation+0.000
coming
Token coming
Feature activation+0.000
to
Token to
Feature activation+0.000
an
Token an
Feature activation+0.000
end
Token end
Feature activation+0.000
?
Token?
Feature activation+2.194
And
Token And
Feature activation+0.903
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
president
Token president
Feature activation+0.000
won
Token won
Feature activation+0.000
ruining
Token ruining
Feature activation+0.000
another
Token another
Feature activation+0.000
Cowboys
Token Cowboys
Feature activation+0.000
season
Token season
Feature activation+0.000
remains
Token remains
Feature activation+2.624
a
Token a
Feature activation+2.383
mystery
Token mystery
Feature activation+1.186
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Big
TokenBig
Feature activation+0.000

INTERVAL 1.553 - 2.070
CONTAINS 0.017%

women
Token women
Feature activation+0.000
?
Token?
Feature activation+1.292
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.313
Gill
Token Gill
Feature activation+0.361
iland
Tokeniland
Feature activation+1.678
said
Token said
Feature activation+0.737
.
Token.
Feature activation+0.206
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.257
G
TokenG
Feature activation+0.000
which
Token which
Feature activation+0.000
would
Token would
Feature activation+0.020
you
Token you
Feature activation+0.000
pick
Token pick
Feature activation+0.000
?
Token?
Feature activation+1.500
More
Token More
Feature activation+1.889
than
Token than
Feature activation+1.430
likely
Token likely
Feature activation+1.102
,
Token,
Feature activation+0.576
you
Token you
Feature activation+0.541
would
Token would
Feature activation+0.274
does
Token does
Feature activation+0.084
it
Token it
Feature activation+0.215
make
Token make
Feature activation+0.201
a
Token a
Feature activation+0.050
difference
Token difference
Feature activation+0.541
?
Token?
Feature activation+1.781
Our
Token Our
Feature activation+0.979
second
Token second
Feature activation+0.981
experiment
Token experiment
Feature activation+0.043
answered
Token answered
Feature activation+2.883
these
Token these
Feature activation+4.029
that
Token that
Feature activation+0.643
you
Token you
Feature activation+0.511
would
Token would
Feature activation+0.145
not
Token not
Feature activation+0.000
obey
Token obey
Feature activation+0.000
?
Token?
Feature activation+1.942
After
Token After
Feature activation+1.159
all
Token all
Feature activation+0.546
,
Token,
Feature activation+0.431
once
Token once
Feature activation+0.055
genocide
Token genocide
Feature activation+0.000
truth
Tokentruth
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
actually
Token actually
Feature activation+0.000
is
Token is
Feature activation+0.030
?
Token?
Feature activation+1.585
Ċ
TokenĊ
Feature activation+0.885
Ċ
TokenĊ
Feature activation+1.480
I
TokenI
Feature activation+1.004
am
Token am
Feature activation+1.080
a
Token a
Feature activation+0.520

INTERVAL 1.035 - 1.553
CONTAINS 0.047%

the
Token the
Feature activation+0.000
Cold
Token Cold
Feature activation+0.000
War
Token War
Feature activation+0.000
?
Token?
Feature activation+1.172
That
Token That
Feature activation+1.517
does
Token does
Feature activation+1.391
not
Token not
Feature activation+1.318
mean
Token mean
Feature activation+0.600
we
Token we
Feature activation+0.000
do
Token do
Feature activation+0.000
not
Token not
Feature activation+0.000
complicated
Token complicated
Feature activation+0.000
to
Token to
Feature activation+0.000
a
Token a
Feature activation+0.000
child
Token child
Feature activation+0.000
?
Token?
Feature activation+0.720
That
Token That
Feature activation+1.319
's
Token's
Feature activation+1.520
what
Token what
Feature activation+0.763
it
Token it
Feature activation+1.084
involves
Token involves
Feature activation+0.243
,"
Token,"
Feature activation+0.768
March
Token March
Feature activation+0.000
?"
Token?"
Feature activation+0.628
even
Token even
Feature activation+0.000
exist
Token exist
Feature activation+0.000
(
Token (
Feature activation+0.334
and
Tokenand
Feature activation+1.185
there
Token there
Feature activation+0.232
are
Token are
Feature activation+0.000
other
Token other
Feature activation+0.329
versions
Token versions
Feature activation+0.274
of
Token of
Feature activation+0.230
we
Token we
Feature activation+0.000
take
Token take
Feature activation+0.000
that
Token that
Feature activation+0.000
next
Token next
Feature activation+0.000
step
Token step
Feature activation+0.000
?
Token?
Feature activation+1.416
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.087
One
TokenOne
Feature activation+1.067
potential
Token potential
Feature activation+2.087
answer
Token answer
Feature activation+3.278
made
Token made
Feature activation+0.000
?
Token?
Feature activation+0.917
Ċ
TokenĊ
Feature activation+0.717
Ċ
TokenĊ
Feature activation+1.077
While
TokenWhile
Feature activation+0.709
all
Token all
Feature activation+1.163
of
Token of
Feature activation+1.375
those
Token those
Feature activation+2.130
are
Token are
Feature activation+1.393
valid
Token valid
Feature activation+1.598
concerns
Token concerns
Feature activation+0.500

INTERVAL 0.518 - 1.035
CONTAINS 0.122%

need
Token need
Feature activation+0.000
,
Token,
Feature activation+1.152
and
Token and
Feature activation+1.804
can
Token can
Feature activation+0.221
we
Token we
Feature activation+0.000
express
Token express
Feature activation+0.930
it
Token it
Feature activation+0.000
in
Token in
Feature activation+0.000
a
Token a
Feature activation+0.000
straightforward
Token straightforward
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation+2.071
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.731
Our
TokenOur
Feature activation+1.002
star
Token star
Feature activation+0.031
shouldn
Token shouldn
Feature activation+0.734
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
have
Token have
Feature activation+0.795
left
Token left
Feature activation+0.067
a
Token a
Feature activation+0.000
bar
Token bar
Feature activation+0.000
in
Token in
Feature activation+0.000
Brooklyn
Token Brooklyn
Feature activation+0.000
.
Token.
Feature activation+0.125
This
Token This
Feature activation+0.662
is
Token is
Feature activation+0.888
what
Token what
Feature activation+0.667
I
Token I
Feature activation+0.847
learned
Token learned
Feature activation+0.574
last
Token last
Feature activation+0.252
ľ
Tokenľ
Feature activation+0.000
I
TokenI
Feature activation+0.484
wanted
Token wanted
Feature activation+0.197
them
Token them
Feature activation+0.016
to
Token to
Feature activation+0.123
know
Token know
Feature activation+0.532
,
Token,
Feature activation+0.000
'
Token '
Feature activation+0.008
Gu
TokenGu
Feature activation+0.000
ys
Tokenys
Feature activation+0.000
,
Token,
Feature activation+0.000
would
Token would
Feature activation+0.145
not
Token not
Feature activation+0.000
obey
Token obey
Feature activation+0.000
?
Token?
Feature activation+1.942
After
Token After
Feature activation+1.159
all
Token all
Feature activation+0.546
,
Token,
Feature activation+0.431
once
Token once
Feature activation+0.055
genocide
Token genocide
Feature activation+0.000
is
Token is
Feature activation+0.000
acceptable
Token acceptable
Feature activation+0.000

INTERVAL 0.000 - 0.518
CONTAINS 99.806%

impacting
Token impacting
Feature activation+0.000
nearly
Token nearly
Feature activation+0.000
one
Token one
Feature activation+0.000
billion
Token billion
Feature activation+0.000
users
Token users
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
was
Token was
Feature activation+0.000
reported
Token reported
Feature activation+0.000
in
Token in
Feature activation+0.000
December
Token December
Feature activation+0.000
statement
Token statement
Feature activation+0.000
claiming
Token claiming
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
sy
Tokensy
Feature activation+0.000
mp
Tokenmp
Feature activation+0.000
ath
Tokenath
Feature activation+0.000
ized
Tokenized
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
less
Tokenless
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
West
TokenWest
Feature activation+0.000
Region
Token Region
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
#
Token#
Feature activation+0.000
1
Token1
Feature activation+0.000
)
Token)
Feature activation+0.000
Wisconsin
Token Wisconsin
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
he
Tokenhe
Feature activation+0.000
might
Token might
Feature activation+0.000
scratch
Token scratch
Feature activation+0.000
my
Token my
Feature activation+0.000
black
Token black
Feature activation+0.000
Mercedes
Token Mercedes
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
against
Token against
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
odds
Token odds
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
asteroid
Token asteroid
Feature activation+0.000
hitting
Token hitting
Feature activation+0.000
Earth
Token Earth
Feature activation+0.000
were
Token were
Feature activation+0.000
first
Token first
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 7 in H8.9: (feature 7352)

TOP ACTIVATIONS
MAX = 3.411

a
Token a
Feature activation+0.320
get
Token get
Feature activation+0.000
-
Token-
Feature activation+0.000
rich
Tokenrich
Feature activation+0.811
-
Token-
Feature activation+0.000
quick
Tokenquick
Feature activation+3.411
sentiment
Token sentiment
Feature activation+0.555
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
is
Token is
Feature activation+0.000
causing
Token causing
Feature activation+0.000
s
Tokens
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
Back
TokenBack
Feature activation+0.000
To
Token To
Feature activation+0.000
School
Token School
Feature activation+3.230
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.384
event
Token event
Feature activation+0.000
this
Token this
Feature activation+0.000
morning
Token morning
Feature activation+0.000
Friday
Token Friday
Feature activation+0.000
'
Token '
Feature activation+0.000
Make
TokenMake
Feature activation+0.000
America
Token America
Feature activation+0.000
Great
Token Great
Feature activation+0.806
Again
Token Again
Feature activation+3.042
'
Token'
Feature activation+3.029
Merch
Token Merch
Feature activation+0.000
andise
Tokenandise
Feature activation+0.000
Deal
Token Deal
Feature activation+0.000
On
Token On
Feature activation+0.000
'
Token '
Feature activation+0.000
Make
TokenMake
Feature activation+0.000
America
Token America
Feature activation+0.000
Great
Token Great
Feature activation+0.806
Again
Token Again
Feature activation+3.042
'
Token'
Feature activation+3.029
Merch
Token Merch
Feature activation+0.000
andise
Tokenandise
Feature activation+0.000
Deal
Token Deal
Feature activation+0.000
On
Token On
Feature activation+0.000
Facebook
Token Facebook
Feature activation+0.000
Top
Token Top
Feature activation+2.409
,
Token,
Feature activation+0.529
No
Token No
Feature activation+0.000
Child
Token Child
Feature activation+0.000
Left
Token Left
Feature activation+0.000
Behind
Token Behind
Feature activation+2.648
,
Token,
Feature activation+1.216
these
Token these
Feature activation+0.651
benchmarks
Token benchmarks
Feature activation+0.831
that
Token that
Feature activation+0.000
have
Token have
Feature activation+0.118
"
Token "
Feature activation+0.000
Path
TokenPath
Feature activation+0.000
to
Token to
Feature activation+0.000
Prosper
Token Prosper
Feature activation+0.000
ity
Tokenity
Feature activation+1.952
"
Token"
Feature activation+2.599
and
Token and
Feature activation+0.000
passed
Token passed
Feature activation+0.291
by
Token by
Feature activation+0.000
House
Token House
Feature activation+0.030
Republicans
Token Republicans
Feature activation+0.112
es
Tokenes
Feature activation+0.000
co
Tokenco
Feature activation+0.000
Memory
Token Memory
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
World
Token World
Feature activation+2.587
International
Token International
Feature activation+0.704
Register
Token Register
Feature activation+0.830
in
Token in
Feature activation+0.499
2007
Token 2007
Feature activation+0.025
.
Token.
Feature activation+0.000
figure
Token figure
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Stop
Token Stop
Feature activation+0.000
the
Token the
Feature activation+0.000
War
Token War
Feature activation+2.526
Coalition
Token Coalition
Feature activation+0.463
,
Token,
Feature activation+0.000
hit
Token hit
Feature activation+0.000
out
Token out
Feature activation+0.000
at
Token at
Feature activation+0.000
that
Token that
Feature activation+0.000
this
Token this
Feature activation+0.000
Race
Token Race
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Top
Token Top
Feature activation+2.409
,
Token,
Feature activation+0.529
No
Token No
Feature activation+0.000
Child
Token Child
Feature activation+0.000
Left
Token Left
Feature activation+0.000
Behind
Token Behind
Feature activation+2.648
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
Cookies
Token Cookies
Feature activation+0.000
To
Token To
Feature activation+0.000
Head
Token Head
Feature activation+0.000
ers
Tokeners
Feature activation+2.371
X
Token X
Feature activation+0.000
SR
TokenSR
Feature activation+0.000
F
TokenF
Feature activation+1.181
protection
Token protection
Feature activation+0.000
technique
Token technique
Feature activation+0.000
a
Token a
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Lean
TokenLean
Feature activation+0.000
In
Token In
Feature activation+1.842
Circle
Token Circle
Feature activation+2.219
,
Token,
Feature activation+0.567
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.462
which
Token which
Feature activation+0.070
is
Token is
Feature activation+0.273
up
Token up
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
Stop
Token Stop
Feature activation+0.000
Acid
Token Acid
Feature activation+0.000
Attacks
Token Attacks
Feature activation+2.208
(
Token (
Feature activation+0.847
S
TokenS
Feature activation+0.175
AA
TokenAA
Feature activation+0.454
)
Token)
Feature activation+1.476
campaign
Token campaign
Feature activation+0.350
Shot
Token Shot
Feature activation+0.000
Show
Token Show
Feature activation+0.215
State
Token State
Feature activation+0.081
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.025
Industry
Token Industry
Feature activation+2.169
Dinner
Token Dinner
Feature activation+0.455
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
the
Token the
Feature activation+0.000
letter
Token letter
Feature activation+0.000
the
Token the
Feature activation+0.000
Out
Token Out
Feature activation+0.000
wit
Tokenwit
Feature activation+0.866
,
Token,
Feature activation+0.922
Out
Token Out
Feature activation+0.000
last
Tokenlast
Feature activation+2.148
,
Token,
Feature activation+0.434
Out
Token Out
Feature activation+0.000
play
Tokenplay
Feature activation+2.120
phenomenon
Token phenomenon
Feature activation+1.414
.
Token.
Feature activation+0.000
,
Token,
Feature activation+0.922
Out
Token Out
Feature activation+0.000
last
Tokenlast
Feature activation+2.148
,
Token,
Feature activation+0.434
Out
Token Out
Feature activation+0.000
play
Tokenplay
Feature activation+2.120
phenomenon
Token phenomenon
Feature activation+1.414
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.724
30
Token 30
Feature activation+0.000
-
Token-
Feature activation+0.000
their
Token their
Feature activation+0.000
current
Token current
Feature activation+0.000
On
Token On
Feature activation+0.000
The
Token The
Feature activation+0.000
Road
Token Road
Feature activation+0.761
Again
Token Again
Feature activation+2.072
Tour
Token Tour
Feature activation+0.283
.
Token .
Feature activation+0.000
Respons
Token Respons
Feature activation+0.000
ibilities
Tokenibilities
Feature activation+0.000
don
Token don
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
One
TokenOne
Feature activation+0.000
Belt
Token Belt
Feature activation+0.000
,
Token,
Feature activation+0.000
One
Token One
Feature activation+0.000
Road
Token Road
Feature activation+2.012
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.746
initiative
Token initiative
Feature activation+0.000
to
Token to
Feature activation+0.000
build
Token build
Feature activation+0.000
outlining
Token outlining
Feature activation+0.000
his
Token his
Feature activation+0.000
"
Token "
Feature activation+0.000
America
TokenAmerica
Feature activation+0.000
first
Token first
Feature activation+0.674
"
Token"
Feature activation+2.010
vision
Token vision
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
Feature activation+0.000
inaugural
Token inaugural
Feature activation+0.000
address
Token address
Feature activation+0.000
attended
Token attended
Feature activation+0.000
the
Token the
Feature activation+0.000
State
Token State
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.024
Union
Token Union
Feature activation+1.973
address
Token address
Feature activation+0.000
she
Token she
Feature activation+0.000
got
Token got
Feature activation+0.000
uncomfortable
Token uncomfortable
Feature activation+0.000
as
Token as
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
expl
Tokenexpl
Feature activation+0.000
ore
Tokenore
Feature activation+0.099
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+1.969
tab
Token tab
Feature activation+0.166
in
Token in
Feature activation+0.010
Apple
Token Apple
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000

Top DFA by src position
MAX = 3.480

people
Token people
Feature activation-0.024
Top resid features:
with
Token with
Feature activation+0.034
Top resid features:
a
Token a
Feature activation+0.209
Top resid features:
get
Token get
Feature activation+0.086
Top resid features:
-
Token-
Feature activation-0.388
Top resid features:
rich
Tokenrich
Feature activation+3.480
Top resid features:
-
Token-
Feature activation+0.395
Top resid features:
quick
Tokenquick
Feature activation+0.440
Top resid features:
sentiment
Token sentiment
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
This
Token This
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation+0.053
Top resid features:
s
Tokens
Feature activation+0.135
Top resid features:
âĢ
Token âĢ
Feature activation+0.000
Top resid features:
ĺ
Tokenĺ
Feature activation+0.045
Top resid features:
Back
TokenBack
Feature activation+0.325
Top resid features:
To
Token To
Feature activation+2.842
Top resid features:
School
Token School
Feature activation+0.427
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
event
Token event
Feature activation+0.000
Top resid features:
this
Token this
Feature activation+0.000
Top resid features:
Friday
Token Friday
Feature activation+0.062
Top resid features:
'
Token '
Feature activation+0.082
Top resid features:
Make
TokenMake
Feature activation+0.482
Top resid features:
America
Token America
Feature activation+0.313
Top resid features:
Great
Token Great
Feature activation+0.806
Top resid features:
Again
Token Again
Feature activation+1.900
Top resid features:
'
Token'
Feature activation+0.000
Top resid features:
Merch
Token Merch
Feature activation+0.000
Top resid features:
andise
Tokenandise
Feature activation+0.000
Top resid features:
Deal
Token Deal
Feature activation+0.000
Top resid features:
On
Token On
Feature activation+0.000
Top resid features:
Friday
Token Friday
Feature activation+0.091
Top resid features:
'
Token '
Feature activation+0.006
Top resid features:
Make
TokenMake
Feature activation+0.566
Top resid features:
America
Token America
Feature activation+0.133
Top resid features:
Great
Token Great
Feature activation+0.496
Top resid features:
Again
Token Again
Feature activation+2.195
Top resid features:
'
Token'
Feature activation+0.056
Top resid features:
Merch
Token Merch
Feature activation+0.000
Top resid features:
andise
Tokenandise
Feature activation+0.000
Top resid features:
Deal
Token Deal
Feature activation+0.000
Top resid features:
On
Token On
Feature activation+0.000
Top resid features:
This
Token This
Feature activation+0.021
Top resid features:
idea
Token idea
Feature activation-0.018
Top resid features:
that
Token that
Feature activation+0.004
Top resid features:
this
Token this
Feature activation+0.241
Top resid features:
Race
Token Race
Feature activation+0.091
Top resid features:
to
Token to
Feature activation+1.397
Top resid features:
the
Token the
Feature activation+0.090
Top resid features:
Top
Token Top
Feature activation+0.153
Top resid features:
,
Token,
Feature activation+0.112
Top resid features:
No
Token No
Feature activation+0.571
Top resid features:
Child
Token Child
Feature activation+0.436
Top resid features:
also
Token also
Feature activation+0.039
Top resid features:
called
Token called
Feature activation+0.037
Top resid features:
the
Token the
Feature activation+0.155
Top resid features:
"
Token "
Feature activation+0.002
Top resid features:
Path
TokenPath
Feature activation+0.316
Top resid features:
to
Token to
Feature activation+2.113
Top resid features:
Prosper
Token Prosper
Feature activation+0.093
Top resid features:
ity
Tokenity
Feature activation+0.263
Top resid features:
"
Token"
Feature activation+0.080
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
passed
Token passed
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.226
Top resid features:
Un
Token Un
Feature activation+0.054
Top resid features:
es
Tokenes
Feature activation+0.014
Top resid features:
co
Tokenco
Feature activation+0.007
Top resid features:
Memory
Token Memory
Feature activation+0.289
Top resid features:
of
Token of
Feature activation+2.062
Top resid features:
the
Token the
Feature activation+0.361
Top resid features:
World
Token World
Feature activation+0.191
Top resid features:
International
Token International
Feature activation+0.000
Top resid features:
Register
Token Register
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
figure
Token figure
Feature activation-0.004
Top resid features:
in
Token in
Feature activation+0.013
Top resid features:
the
Token the
Feature activation+0.083
Top resid features:
Stop
Token Stop
Feature activation+0.757
Top resid features:
the
Token the
Feature activation+1.158
Top resid features:
War
Token War
Feature activation+1.373
Top resid features:
Coalition
Token Coalition
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
hit
Token hit
Feature activation+0.000
Top resid features:
out
Token out
Feature activation+0.000
Top resid features:
at
Token at
Feature activation+0.000
Top resid features:
This
Token This
Feature activation+0.031
Top resid features:
idea
Token idea
Feature activation-0.009
Top resid features:
that
Token that
Feature activation+0.037
Top resid features:
this
Token this
Feature activation+0.258
Top resid features:
Race
Token Race
Feature activation+0.143
Top resid features:
to
Token to
Feature activation+2.142
Top resid features:
the
Token the
Feature activation+0.140
Top resid features:
Top
Token Top
Feature activation+0.408
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
No
Token No
Feature activation+0.000
Top resid features:
Child
Token Child
Feature activation+0.000
Top resid features:
in
Tokenin
Feature activation+0.019
Top resid features:
support
Token support
Feature activation+0.003
Top resid features:
for
Token for
Feature activation+0.007
Top resid features:
the
Token the
Feature activation+0.130
Top resid features:
Cookies
Token Cookies
Feature activation+0.145
Top resid features:
To
Token To
Feature activation+2.834
Top resid features:
Head
Token Head
Feature activation+0.119
Top resid features:
ers
Tokeners
Feature activation+0.089
Top resid features:
X
Token X
Feature activation+0.000
Top resid features:
SR
TokenSR
Feature activation+0.000
Top resid features:
F
TokenF
Feature activation+0.000
Top resid features:
join
Token join
Feature activation+0.007
Top resid features:
a
Token a
Feature activation+0.072
Top resid features:
âĢ
Token âĢ
Feature activation-0.056
Top resid features:
ľ
Tokenľ
Feature activation+0.064
Top resid features:
Lean
TokenLean
Feature activation+0.622
Top resid features:
In
Token In
Feature activation+1.562
Top resid features:
Circle
Token Circle
Feature activation+0.794
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
Ŀ
TokenĿ
Feature activation+0.000
Top resid features:
which
Token which
Feature activation+0.000
Top resid features:
ra
Tokenra
Feature activation-0.023
Top resid features:
set
Token set
Feature activation+0.015
Top resid features:
up
Token up
Feature activation+0.004
Top resid features:
by
Token by
Feature activation+0.067
Top resid features:
the
Token the
Feature activation+0.132
Top resid features:
Stop
Token Stop
Feature activation+1.470
Top resid features:
Acid
Token Acid
Feature activation+0.716
Top resid features:
Attacks
Token Attacks
Feature activation+0.687
Top resid features:
(
Token (
Feature activation+0.000
Top resid features:
S
TokenS
Feature activation+0.000
Top resid features:
AA
TokenAA
Feature activation+0.000
Top resid features:
2017
Token 2017
Feature activation+0.225
Top resid features:
Shot
Token Shot
Feature activation+0.045
Top resid features:
Show
Token Show
Feature activation+0.082
Top resid features:
State
Token State
Feature activation+0.067
Top resid features:
of
Token of
Feature activation+0.900
Top resid features:
the
Token the
Feature activation+1.457
Top resid features:
Industry
Token Industry
Feature activation+0.080
Top resid features:
Dinner
Token Dinner
Feature activation+0.000
Top resid features:
.
Token.
Feature activation+0.000
Top resid features:
In
Token In
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
vers
Tokenvers
Feature activation-0.005
Top resid features:
ed
Tokened
Feature activation+0.005
Top resid features:
in
Token in
Feature activation+0.068
Top resid features:
the
Token the
Feature activation+0.209
Top resid features:
Out
Token Out
Feature activation+0.208
Top resid features:
wit
Tokenwit
Feature activation+1.602
Top resid features:
,
Token,
Feature activation+0.416
Top resid features:
Out
Token Out
Feature activation+0.103
Top resid features:
last
Tokenlast
Feature activation+0.266
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
Out
Token Out
Feature activation+0.000
Top resid features:
vers
Tokenvers
Feature activation-0.008
Top resid features:
ed
Tokened
Feature activation+0.002
Top resid features:
in
Token in
Feature activation+0.043
Top resid features:
the
Token the
Feature activation+0.243
Top resid features:
Out
Token Out
Feature activation+0.157
Top resid features:
wit
Tokenwit
Feature activation+1.524
Top resid features:
,
Token,
Feature activation+0.318
Top resid features:
Out
Token Out
Feature activation+0.132
Top resid features:
last
Tokenlast
Feature activation+0.274
Top resid features:
,
Token,
Feature activation+0.007
Top resid features:
Out
Token Out
Feature activation+0.086
Top resid features:
their
Token their
Feature activation+0.232
Top resid features:
current
Token current
Feature activation+0.074
Top resid features:
On
Token On
Feature activation+0.274
Top resid features:
The
Token The
Feature activation+0.169
Top resid features:
Road
Token Road
Feature activation+0.428
Top resid features:
Again
Token Again
Feature activation+1.653
Top resid features:
Tour
Token Tour
Feature activation+0.000
Top resid features:
.
Token .
Feature activation+0.000
Top resid features:
Respons
Token Respons
Feature activation+0.000
Top resid features:
ibilities
Tokenibilities
Feature activation+0.000
Top resid features:
don
Token don
Feature activation+0.000
Top resid features:
its
Token its
Feature activation+0.070
Top resid features:
âĢ
Token âĢ
Feature activation-0.002
Top resid features:
ľ
Tokenľ
Feature activation+0.037
Top resid features:
One
TokenOne
Feature activation+0.475
Top resid features:
Belt
Token Belt
Feature activation+0.230
Top resid features:
,
Token,
Feature activation+0.972
Top resid features:
One
Token One
Feature activation+0.689
Top resid features:
Road
Token Road
Feature activation+0.442
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
Ŀ
TokenĿ
Feature activation+0.000
Top resid features:
initiative
Token initiative
Feature activation+0.000
Top resid features:
,
Token,
Feature activation-0.002
Top resid features:
outlining
Token outlining
Feature activation+0.093
Top resid features:
his
Token his
Feature activation+0.043
Top resid features:
"
Token "
Feature activation+0.014
Top resid features:
America
TokenAmerica
Feature activation+0.298
Top resid features:
first
Token first
Feature activation+2.407
Top resid features:
"
Token"
Feature activation+0.100
Top resid features:
vision
Token vision
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
his
Token his
Feature activation+0.000
Top resid features:
inaugural
Token inaugural
Feature activation+0.000
Top resid features:
she
Token she
Feature activation-0.000
Top resid features:
attended
Token attended
Feature activation+0.060
Top resid features:
the
Token the
Feature activation+0.198
Top resid features:
State
Token State
Feature activation+0.027
Top resid features:
of
Token of
Feature activation+0.571
Top resid features:
the
Token the
Feature activation+1.802
Top resid features:
Union
Token Union
Feature activation+0.089
Top resid features:
address
Token address
Feature activation+0.000
Top resid features:
she
Token she
Feature activation+0.000
Top resid features:
got
Token got
Feature activation+0.000
Top resid features:
uncomfortable
Token uncomfortable
Feature activation+0.000
Top resid features:
powers
Token powers
Feature activation+0.171
Top resid features:
the
Token the
Feature activation+0.125
Top resid features:
âĢ
Token âĢ
Feature activation+0.016
Top resid features:
ľ
Tokenľ
Feature activation+0.011
Top resid features:
expl
Tokenexpl
Feature activation+0.036
Top resid features:
ore
Tokenore
Feature activation+2.214
Top resid features:
âĢ
TokenâĢ
Feature activation+0.046
Top resid features:
Ŀ
TokenĿ
Feature activation+0.026
Top resid features:
tab
Token tab
Feature activation+0.000
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
Apple
Token Apple
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.02

Head 2: 0.04

Head 3: 0.04

Head 4: 0.05

Head 5: 0.11

Head 6: 0.04

Head 7: 0.11

Head 8: 0.05

Head 9: 0.39

Head 10: 0.10

Head 11: 0.03

Positive logits

pamphlet3.54

Pledge3.48

manifesto3.36

booklet3.26

playbook3.23

slogan3.23

mantra3.11

pledge3.05

chant3.05

banner3.00

motto2.99

campaign2.95

message2.92

Prayer2.87

Meditation2.83

playlist2.78

theme2.75

2.72

rallying2.70

revival2.69

Negative logits

ources-2.95

adequ-2.94

icably-2.77

edom-2.76

inarily-2.68

agree-2.67

Yugoslav-2.61

Yugoslavia-2.58

ogeneity-2.53

entimes-2.53

ogenous-2.47

accidents-2.46

hap-2.46

Neither-2.43

usual-2.41

ivably-2.41

extrad-2.37

Nevertheless-2.35

esides-2.34

subcontract-2.28

INTERVAL 3.070 - 3.411
CONTAINS 0.000%

a
Token a
Feature activation+0.320
get
Token get
Feature activation+0.000
-
Token-
Feature activation+0.000
rich
Tokenrich
Feature activation+0.811
-
Token-
Feature activation+0.000
quick
Tokenquick
Feature activation+3.411
sentiment
Token sentiment
Feature activation+0.555
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
is
Token is
Feature activation+0.000
causing
Token causing
Feature activation+0.000
s
Tokens
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
Back
TokenBack
Feature activation+0.000
To
Token To
Feature activation+0.000
School
Token School
Feature activation+3.230
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.384
event
Token event
Feature activation+0.000
this
Token this
Feature activation+0.000
morning
Token morning
Feature activation+0.000

INTERVAL 2.729 - 3.070
CONTAINS 0.000%

Friday
Token Friday
Feature activation+0.000
'
Token '
Feature activation+0.000
Make
TokenMake
Feature activation+0.000
America
Token America
Feature activation+0.000
Great
Token Great
Feature activation+0.806
Again
Token Again
Feature activation+3.042
'
Token'
Feature activation+3.029
Merch
Token Merch
Feature activation+0.000
andise
Tokenandise
Feature activation+0.000
Deal
Token Deal
Feature activation+0.000
On
Token On
Feature activation+0.000
'
Token '
Feature activation+0.000
Make
TokenMake
Feature activation+0.000
America
Token America
Feature activation+0.000
Great
Token Great
Feature activation+0.806
Again
Token Again
Feature activation+3.042
'
Token'
Feature activation+3.029
Merch
Token Merch
Feature activation+0.000
andise
Tokenandise
Feature activation+0.000
Deal
Token Deal
Feature activation+0.000
On
Token On
Feature activation+0.000
Facebook
Token Facebook
Feature activation+0.000

INTERVAL 2.388 - 2.729
CONTAINS 0.001%

es
Tokenes
Feature activation+0.000
co
Tokenco
Feature activation+0.000
Memory
Token Memory
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
World
Token World
Feature activation+2.587
International
Token International
Feature activation+0.704
Register
Token Register
Feature activation+0.830
in
Token in
Feature activation+0.499
2007
Token 2007
Feature activation+0.025
.
Token.
Feature activation+0.000
that
Token that
Feature activation+0.000
this
Token this
Feature activation+0.000
Race
Token Race
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Top
Token Top
Feature activation+2.409
,
Token,
Feature activation+0.529
No
Token No
Feature activation+0.000
Child
Token Child
Feature activation+0.000
Left
Token Left
Feature activation+0.000
Behind
Token Behind
Feature activation+2.648
Top
Token Top
Feature activation+2.409
,
Token,
Feature activation+0.529
No
Token No
Feature activation+0.000
Child
Token Child
Feature activation+0.000
Left
Token Left
Feature activation+0.000
Behind
Token Behind
Feature activation+2.648
,
Token,
Feature activation+1.216
these
Token these
Feature activation+0.651
benchmarks
Token benchmarks
Feature activation+0.831
that
Token that
Feature activation+0.000
have
Token have
Feature activation+0.118
"
Token "
Feature activation+0.000
Path
TokenPath
Feature activation+0.000
to
Token to
Feature activation+0.000
Prosper
Token Prosper
Feature activation+0.000
ity
Tokenity
Feature activation+1.952
"
Token"
Feature activation+2.599
and
Token and
Feature activation+0.000
passed
Token passed
Feature activation+0.291
by
Token by
Feature activation+0.000
House
Token House
Feature activation+0.030
Republicans
Token Republicans
Feature activation+0.112
figure
Token figure
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Stop
Token Stop
Feature activation+0.000
the
Token the
Feature activation+0.000
War
Token War
Feature activation+2.526
Coalition
Token Coalition
Feature activation+0.463
,
Token,
Feature activation+0.000
hit
Token hit
Feature activation+0.000
out
Token out
Feature activation+0.000
at
Token at
Feature activation+0.000

INTERVAL 2.047 - 2.388
CONTAINS 0.001%

up
Token up
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
Stop
Token Stop
Feature activation+0.000
Acid
Token Acid
Feature activation+0.000
Attacks
Token Attacks
Feature activation+2.208
(
Token (
Feature activation+0.847
S
TokenS
Feature activation+0.175
AA
TokenAA
Feature activation+0.454
)
Token)
Feature activation+1.476
campaign
Token campaign
Feature activation+0.350
,
Token,
Feature activation+0.922
Out
Token Out
Feature activation+0.000
last
Tokenlast
Feature activation+2.148
,
Token,
Feature activation+0.434
Out
Token Out
Feature activation+0.000
play
Tokenplay
Feature activation+2.120
phenomenon
Token phenomenon
Feature activation+1.414
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.724
30
Token 30
Feature activation+0.000
-
Token-
Feature activation+0.000
a
Token a
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Lean
TokenLean
Feature activation+0.000
In
Token In
Feature activation+1.842
Circle
Token Circle
Feature activation+2.219
,
Token,
Feature activation+0.567
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.462
which
Token which
Feature activation+0.070
is
Token is
Feature activation+0.273
the
Token the
Feature activation+0.000
Out
Token Out
Feature activation+0.000
wit
Tokenwit
Feature activation+0.866
,
Token,
Feature activation+0.922
Out
Token Out
Feature activation+0.000
last
Tokenlast
Feature activation+2.148
,
Token,
Feature activation+0.434
Out
Token Out
Feature activation+0.000
play
Tokenplay
Feature activation+2.120
phenomenon
Token phenomenon
Feature activation+1.414
.
Token.
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
Cookies
Token Cookies
Feature activation+0.000
To
Token To
Feature activation+0.000
Head
Token Head
Feature activation+0.000
ers
Tokeners
Feature activation+2.371
X
Token X
Feature activation+0.000
SR
TokenSR
Feature activation+0.000
F
TokenF
Feature activation+1.181
protection
Token protection
Feature activation+0.000
technique
Token technique
Feature activation+0.000

INTERVAL 1.706 - 2.047
CONTAINS 0.002%

with
Token with
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Ready
TokenReady
Feature activation+0.000
for
Token for
Feature activation+0.000
Hillary
Token Hillary
Feature activation+1.895
,
Token,
Feature activation+0.581
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
and
Token and
Feature activation+0.000
she
Token she
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
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
Union
Token Union
Feature activation+1.821
address
Token address
Feature activation+0.000
.
Token.
Feature activation+0.000
Even
Token Even
Feature activation+0.000
folks
Token folks
Feature activation+0.000
like
Token like
Feature activation+0.000
Our
Token Our
Feature activation+0.000
Business
Token Business
Feature activation+1.385
Start
Token Start
Feature activation+0.000
-
Token-
Feature activation+0.027
U
TokenU
Feature activation+0.000
ps
Tokenps
Feature activation+1.854
Act
Token Act
Feature activation+0.000
"
Token"
Feature activation+0.000
(
Token (
Feature activation+0.000
JO
TokenJO
Feature activation+0.000
BS
TokenBS
Feature activation+0.000
the
Token the
Feature activation+0.000
Preservation
Token Preservation
Feature activation+0.221
of
Token of
Feature activation+0.000
Firearms
Token Firearms
Feature activation+0.133
Owners
Token Owners
Feature activation+0.000
hip
Tokenhip
Feature activation+1.743
recently
Token recently
Feature activation+0.778
criticized
Token criticized
Feature activation+0.000
him
Token him
Feature activation+0.029
for
Token for
Feature activation+0.000
this
Token this
Feature activation+0.478
join
Token join
Feature activation+0.000
a
Token a
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Lean
TokenLean
Feature activation+0.000
In
Token In
Feature activation+1.842
Circle
Token Circle
Feature activation+2.219
,
Token,
Feature activation+0.567
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.462
which
Token which
Feature activation+0.070

INTERVAL 1.365 - 1.706
CONTAINS 0.002%

C
TokenC
Feature activation+0.000
.
Token.
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
Move
Token Move
Feature activation+0.072
,
Token,
Feature activation+1.625
a
Token a
Feature activation+1.690
10
Token 10
Feature activation+0.000
-
Token-
Feature activation+0.175
year
Tokenyear
Feature activation+0.929
transportation
Token transportation
Feature activation+0.263
]
Token]
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Read
Token Read
Feature activation+0.000
Free
Token Free
Feature activation+0.643
!
Token!
Feature activation+1.635
Kind
Token Kind
Feature activation+0.000
ness
Tokenness
Feature activation+0.353
Offensive
Token Offensive
Feature activation+0.000
.[
Token.[
Feature activation+0.000
8
Token8
Feature activation+0.000
Time
Token Time
Feature activation+0.000
We
Token We
Feature activation+0.000
Were
Token Were
Feature activation+0.000
Not
Token Not
Feature activation+0.086
In
Token In
Feature activation+0.111
Love
Token Love
Feature activation+1.629
took
Token took
Feature activation+0.000
a
Token a
Feature activation+0.000
break
Token break
Feature activation+0.000
from
Token from
Feature activation+0.130
filming
Token filming
Feature activation+0.000
Out
Token Out
Feature activation+0.000
last
Tokenlast
Feature activation+2.148
,
Token,
Feature activation+0.434
Out
Token Out
Feature activation+0.000
play
Tokenplay
Feature activation+2.120
phenomenon
Token phenomenon
Feature activation+1.414
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.724
30
Token 30
Feature activation+0.000
-
Token-
Feature activation+0.000
year
Tokenyear
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
#
Token #
Feature activation+0.000
Design
TokenDesign
Feature activation+0.000
In
TokenIn
Feature activation+0.000
Tech
TokenTech
Feature activation+1.368
report
Token report
Feature activation+0.000
.
Token.
Feature activation+0.000
Before
Token Before
Feature activation+0.000
2010
Token 2010
Feature activation+0.000
,
Token,
Feature activation+0.000

INTERVAL 1.023 - 1.365
CONTAINS 0.005%

to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Open
Token Open
Feature activation+0.000
Phil
Token Phil
Feature activation+0.000
anthrop
Tokenanthrop
Feature activation+0.000
y
Tokeny
Feature activation+1.081
Project
Token Project
Feature activation+0.080
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
Soros
Token Soros
Feature activation+0.000
spokesman
Token spokesman
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
"
Token "
Feature activation+0.000
san
Tokensan
Feature activation+0.000
ctuary
Tokenctuary
Feature activation+0.008
cities
Token cities
Feature activation+1.143
"
Token"
Feature activation+1.849
ban
Token ban
Feature activation+0.293
Senate
Token Senate
Feature activation+0.000
Bill
Token Bill
Feature activation+0.000
4
Token 4
Feature activation+0.000
Sub
Token Sub
Feature activation+0.000
t
Tokent
Feature activation+0.000
ract
Tokenract
Feature activation+0.000
From
Token From
Feature activation+0.000
Shape
Token Shape
Feature activation+1.123
Area
Token Area
Feature activation+1.050
button
Token button
Feature activation+0.000
.
Token.
Feature activation+0.000
Obviously
Token Obviously
Feature activation+0.000
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
to
Token to
Feature activation+0.000
call
Token call
Feature activation+0.000
the
Token the
Feature activation+0.000
Arizona
Token Arizona
Feature activation+0.000
Clean
Token Clean
Feature activation+0.000
Elections
Token Elections
Feature activation+1.234
public
Token public
Feature activation+0.520
financing
Token financing
Feature activation+0.689
law
Token law
Feature activation+0.068
a
Token a
Feature activation+0.000
national
Token national
Feature activation+0.436
B
Token B
Feature activation+0.000
isexual
Tokenisexual
Feature activation+0.000
and
Token and
Feature activation+0.000
Transgender
Token Transgender
Feature activation+0.000
Equality
Token Equality
Feature activation+0.166
:
Token:
Feature activation+1.062
Moving
Token Moving
Feature activation+0.000
Forward
Token Forward
Feature activation+0.581
.
Token.
Feature activation+0.607
Was
Token Was
Feature activation+0.000
published
Token published
Feature activation+0.000

INTERVAL 0.682 - 1.023
CONTAINS 0.011%

Move
Token Move
Feature activation+0.072
,
Token,
Feature activation+1.625
a
Token a
Feature activation+1.690
10
Token 10
Feature activation+0.000
-
Token-
Feature activation+0.175
year
Tokenyear
Feature activation+0.929
transportation
Token transportation
Feature activation+0.263
plan
Token plan
Feature activation+0.094
that
Token that
Feature activation+0.000
will
Token will
Feature activation+0.000
cost
Token cost
Feature activation+0.000
Health
Token Health
Feature activation+0.521
(
Token (
Feature activation+0.006
SU
TokenSU
Feature activation+0.000
FM
TokenFM
Feature activation+0.000
H
TokenH
Feature activation+0.018
)
Token)
Feature activation+0.729
comedy
Token comedy
Feature activation+0.000
show
Token show
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
Modern
Token Modern
Feature activation+0.000
Preservation
Token Preservation
Feature activation+0.221
of
Token of
Feature activation+0.000
Firearms
Token Firearms
Feature activation+0.133
Owners
Token Owners
Feature activation+0.000
hip
Tokenhip
Feature activation+1.743
recently
Token recently
Feature activation+0.778
criticized
Token criticized
Feature activation+0.000
him
Token him
Feature activation+0.029
for
Token for
Feature activation+0.000
this
Token this
Feature activation+0.478
week
Token week
Feature activation+0.000
case
Token case
Feature activation+0.000
of
Token of
Feature activation+0.000
The
Token The
Feature activation+0.000
Emperor
Token Emperor
Feature activation+0.000
is
Token is
Feature activation+0.000
Dead
Token Dead
Feature activation+0.824
,
Token,
Feature activation+0.000
long
Token long
Feature activation+0.000
live
Token live
Feature activation+0.182
the
Token the
Feature activation+0.175
Emperor
Token Emperor
Feature activation+0.000
I
Token I
Feature activation+0.000
wrote
Token wrote
Feature activation+0.000
Pract
Token Pract
Feature activation+0.000
ical
Tokenical
Feature activation+0.000
Creative
Token Creative
Feature activation+0.000
Writing
Token Writing
Feature activation+0.809
Ex
Token Ex
Feature activation+0.000
erc
Tokenerc
Feature activation+0.000
ises
Tokenises
Feature activation+0.359
.
Token.
Feature activation+0.000
I
Token I
Feature activation+0.000

INTERVAL 0.341 - 0.682
CONTAINS 0.044%

city
Token city
Feature activation+0.000
's
Token's
Feature activation+0.000
Nuclear
Token Nuclear
Feature activation+0.000
Weapons
Token Weapons
Feature activation+0.000
Free
Token Free
Feature activation+0.000
Zone
Token Zone
Feature activation+0.556
Ord
Token Ord
Feature activation+0.000
inance
Tokeninance
Feature activation+0.108
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
ordinance
Token ordinance
Feature activation+0.163
âĢĵ
Token âĢĵ
Feature activation+0.000
Dark
Token Dark
Feature activation+0.000
Matter
Token Matter
Feature activation+0.288
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.494
TWO
Token TWO
Feature activation+0.154
,
Token,
Feature activation+0.000
Melissa
Token Melissa
Feature activation+0.000
O
Token O
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
a
Token a
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Digital
TokenDigital
Feature activation+0.000
Nom
Token Nom
Feature activation+0.000
ad
Tokenad
Feature activation+0.508
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.268
visa
Token visa
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
call
Token call
Feature activation+0.000
the
Token the
Feature activation+0.000
Arizona
Token Arizona
Feature activation+0.000
Clean
Token Clean
Feature activation+0.000
Elections
Token Elections
Feature activation+1.234
public
Token public
Feature activation+0.520
financing
Token financing
Feature activation+0.689
law
Token law
Feature activation+0.068
a
Token a
Feature activation+0.000
national
Token national
Feature activation+0.436
model
Token model
Feature activation+0.241
Build
Token Build
Feature activation+0.000
it
Token it
Feature activation+0.000
They
Token They
Feature activation+0.000
Will
Token Will
Feature activation+0.000
Come
Token Come
Feature activation+0.000
,
Token,
Feature activation+0.350
in
Token in
Feature activation+0.139
which
Token which
Feature activation+0.000
he
Token he
Feature activation+0.000
participated
Token participated
Feature activation+0.000
,
Token,
Feature activation+0.000

INTERVAL 0.000 - 0.341
CONTAINS 99.935%

Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Warren
TokenWarren
Feature activation+0.000
Buffett
Token Buffett
Feature activation+0.000
,
Token,
Feature activation+0.000
CEO
Token CEO
Feature activation+0.000
of
Token of
Feature activation+0.000
Berkshire
Token Berkshire
Feature activation+0.000
Hath
Token Hath
Feature activation+0.000
away
Tokenaway
Feature activation+0.000
,
Token,
Feature activation+0.000
senator
Token senator
Feature activation+0.000
could
Token could
Feature activation+0.000
have
Token have
Feature activation+0.000
beaten
Token beaten
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
had
Token had
Feature activation+0.000
he
Token he
Feature activation+0.000
been
Token been
Feature activation+0.000
the
Token the
Feature activation+0.000
Democratic
Token Democratic
Feature activation+0.000
nominee
Token nominee
Feature activation+0.000
ocratic
Tokenocratic
Feature activation+0.000
and
Token and
Feature activation+0.000
unaccount
Token unaccount
Feature activation+0.000
able
Tokenable
Feature activation+0.000
fashion
Token fashion
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
report
Token report
Feature activation+0.000
,
Token,
Feature activation+0.000
conducted
Token conducted
Feature activation+0.000
by
Token by
Feature activation+0.000
Synt
Token Synt
Feature activation+0.000
ax
Tokenax
Feature activation+0.000
Error
TokenError
Feature activation+0.000
:
Token:
Feature activation+0.000
Cannot
Token Cannot
Feature activation+0.000
use
Token use
Feature activation+0.000
the
Token the
Feature activation+0.000
key
Token key
Feature activation+0.000
word
Token word
Feature activation+0.000
'
Token '
Feature activation+0.000
class
Tokenclass
Feature activation+0.000
Changing
Token Changing
Feature activation+0.000
gau
Token gau
Feature activation+0.000
ges
Tokenges
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
border
Token border
Feature activation+0.000
takes
Token takes
Feature activation+0.000
days
Token days
Feature activation+0.000
for
Token for
Feature activation+0.000
cargo
Token cargo
Feature activation+0.000
and
Token and
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 8 in H8.9: (feature 294)

TOP ACTIVATIONS
MAX = 2.450

there
Token there
Feature activation+0.000
is
Token is
Feature activation+0.000
no
Token no
Feature activation+0.000
God
Token God
Feature activation+0.000
)
Token)
Feature activation+1.896
is
Token is
Feature activation+2.450
daunting
Token daunting
Feature activation+1.311
:
Token:
Feature activation+0.082
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.104
What
TokenWhat
Feature activation+0.000
Las
Token Las
Feature activation+0.000
Vegas
Token Vegas
Feature activation+0.736
now
Token now
Feature activation+0.583
seems
Token seems
Feature activation+0.935
all
Token all
Feature activation+0.211
but
Token but
Feature activation+1.990
certain
Token certain
Feature activation+0.687
to
Token to
Feature activation+0.620
happen
Token happen
Feature activation+0.482
.
Token.
Feature activation+0.000
After
Token After
Feature activation+0.000
believe
Token believe
Feature activation+0.000
there
Token there
Feature activation+0.000
is
Token is
Feature activation+0.000
no
Token no
Feature activation+0.000
God
Token God
Feature activation+0.000
)
Token)
Feature activation+1.896
is
Token is
Feature activation+2.450
daunting
Token daunting
Feature activation+1.311
:
Token:
Feature activation+0.082
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.104
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
demise
Token demise
Feature activation+0.000
of
Token of
Feature activation+0.000
Pakistan
Token Pakistan
Feature activation+0.000
is
Token is
Feature activation+1.841
"
Token "
Feature activation+0.581
ine
Tokenine
Feature activation+0.000
v
Tokenv
Feature activation+0.000
itable
Tokenitable
Feature activation+0.417
"
Token"
Feature activation+0.000
's
Token's
Feature activation+0.000
exit
Token exit
Feature activation+0.781
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
European
Token European
Feature activation+0.000
Union
Token Union
Feature activation+1.807
(
Token (
Feature activation+0.000
EU
TokenEU
Feature activation+0.000
)
Token)
Feature activation+1.276
was
Token was
Feature activation+1.157
a
Token a
Feature activation+1.320
.
Token.
Feature activation+0.000
His
Token His
Feature activation+0.000
move
Token move
Feature activation+0.000
toward
Token toward
Feature activation+0.000
libertarian
Token libertarian
Feature activation+0.000
ism
Tokenism
Feature activation+1.621
coincided
Token coincided
Feature activation+0.047
with
Token with
Feature activation+0.432
a
Token a
Feature activation+0.203
larger
Token larger
Feature activation+0.033
role
Token role
Feature activation+0.000
daughter
Token daughter
Feature activation+0.000
becoming
Token becoming
Feature activation+0.000
an
Token an
Feature activation+0.000
Internet
Token Internet
Feature activation+0.000
sensation
Token sensation
Feature activation+0.785
is
Token is
Feature activation+1.438
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
quite
Tokenquite
Feature activation+1.061
exciting
Token exciting
Feature activation+0.406
.
Token.
Feature activation+0.000
or
Token or
Feature activation+0.000
coming
Token coming
Feature activation+0.000
-
Token-
Feature activation+0.000
of
Tokenof
Feature activation+0.000
-
Token-
Feature activation+0.000
age
Tokenage
Feature activation+1.405
story
Token story
Feature activation+0.000
,
Token,
Feature activation+0.000
focused
Token focused
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
the
Token the
Feature activation+0.000
removal
Token removal
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
coalition
Token coalition
Feature activation+0.000
was
Token was
Feature activation+1.402
pending
Token pending
Feature activation+0.000
a
Token a
Feature activation+0.000
review
Token review
Feature activation+0.000
of
Token of
Feature activation+0.000
child
Token child
Feature activation+0.000
Union
Token Union
Feature activation+1.807
(
Token (
Feature activation+0.000
EU
TokenEU
Feature activation+0.000
)
Token)
Feature activation+1.276
was
Token was
Feature activation+1.157
a
Token a
Feature activation+1.320
strategic
Token strategic
Feature activation+0.430
error
Token error
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
is
Token is
Feature activation+0.000
no
Token no
Feature activation+0.000
God
Token God
Feature activation+0.000
)
Token)
Feature activation+1.896
is
Token is
Feature activation+2.450
daunting
Token daunting
Feature activation+1.311
:
Token:
Feature activation+0.082
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.104
What
TokenWhat
Feature activation+0.000
will
Token will
Feature activation+0.000
Travel
Token Travel
Feature activation+0.000
ing
Tokening
Feature activation+0.000
Wil
Token Wil
Feature activation+0.000
bur
Tokenbur
Feature activation+0.000
ys
Tokenys
Feature activation+0.189
was
Token was
Feature activation+1.280
a
Token a
Feature activation+1.080
happy
Token happy
Feature activation+0.398
accident
Token accident
Feature activation+0.000
.
Token.
Feature activation+0.000
Warner
Token Warner
Feature activation+0.000
the
Token the
Feature activation+0.000
European
Token European
Feature activation+0.000
Union
Token Union
Feature activation+1.807
(
Token (
Feature activation+0.000
EU
TokenEU
Feature activation+0.000
)
Token)
Feature activation+1.276
was
Token was
Feature activation+1.157
a
Token a
Feature activation+1.320
strategic
Token strategic
Feature activation+0.430
error
Token error
Feature activation+0.000
.
Token.
Feature activation+0.000
could
Token could
Feature activation+0.000
that
Token that
Feature activation+0.000
its
Token its
Feature activation+0.000
temporary
Token temporary
Feature activation+0.000
closure
Token closure
Feature activation+0.275
was
Token was
Feature activation+1.247
the
Token the
Feature activation+0.542
most
Token most
Feature activation+0.374
dire
Token dire
Feature activation+0.000
effect
Token effect
Feature activation+0.000
of
Token of
Feature activation+0.000
to
Token to
Feature activation+0.365
fib
Token fib
Feature activation+0.000
rosis
Tokenrosis
Feature activation+0.622
or
Token or
Feature activation+0.000
tumor
Token tumor
Feature activation+0.000
is
Token is
Feature activation+1.222
a
Token a
Feature activation+0.963
common
Token common
Feature activation+0.951
therapy
Token therapy
Feature activation+0.125
that
Token that
Feature activation+0.164
utilizes
Token utilizes
Feature activation+0.131
Where
Token Where
Feature activation+0.000
it
Token it
Feature activation+0.000
goes
Token goes
Feature activation+0.000
from
Token from
Feature activation+0.000
here
Token here
Feature activation+0.207
will
Token will
Feature activation+1.212
help
Token help
Feature activation+0.212
define
Token define
Feature activation+0.000
the
Token the
Feature activation+0.000
next
Token next
Feature activation+0.000
generation
Token generation
Feature activation+0.000
two
Token two
Feature activation+0.000
countries
Token countries
Feature activation+0.000
must
Token must
Feature activation+0.000
make
Token make
Feature activation+0.000
connecting
Token connecting
Feature activation+0.000
flights
Token flights
Feature activation+1.176
in
Token in
Feature activation+0.258
Hong
Token Hong
Feature activation+0.000
Kong
Token Kong
Feature activation+0.371
or
Token or
Feature activation+0.000
Seoul
Token Seoul
Feature activation+0.385
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
country
Token country
Feature activation+0.000
of
Token of
Feature activation+0.000
Niger
Token Niger
Feature activation+0.000
is
Token is
Feature activation+1.173
not
Token not
Feature activation+0.898
just
Token just
Feature activation+0.757
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Ben
TokenBen
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
a
Token a
Feature activation+0.000
coming
Token coming
Feature activation+0.000
of
Token of
Feature activation+0.000
age
Token age
Feature activation+1.168
story
Token story
Feature activation+0.000
that
Token that
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
re
Token re
Feature activation+0.000
class
Tokenclass
Feature activation+0.000
ification
Tokenification
Feature activation+1.168
gave
Token gave
Feature activation+0.000
the
Token the
Feature activation+0.000
FCC
Token FCC
Feature activation+0.000
the
Token the
Feature activation+0.000
power
Token power
Feature activation+0.000

Top DFA by src position
MAX = 2.038

converts
Token converts
Feature activation+0.044
Top resid features:
.
Token.
Feature activation+0.010
Top resid features:
The
Token The
Feature activation+0.023
Top resid features:
process
Token process
Feature activation+0.008
Top resid features:
of
Token of
Feature activation+0.085
Top resid features:
transitioning
Token transitioning
Feature activation+1.538
Top resid features:
into
Token into
Feature activation+0.478
Top resid features:
Islam
Token Islam
Feature activation+0.100
Top resid features:
from
Token from
Feature activation+0.386
Top resid features:
a
Token a
Feature activation+0.074
Top resid features:
previous
Token previous
Feature activation+0.029
Top resid features:
Ļ
TokenĻ
Feature activation+0.014
Top resid features:
long
Token long
Feature activation-0.002
Top resid features:
-
Token-
Feature activation-0.006
Top resid features:
disc
Tokendisc
Feature activation+0.006
Top resid features:
ussed
Tokenussed
Feature activation+0.099
Top resid features:
move
Token move
Feature activation+0.885
Top resid features:
from
Token from
Feature activation+0.777
Top resid features:
Oakland
Token Oakland
Feature activation+0.015
Top resid features:
to
Token to
Feature activation+0.740
Top resid features:
Las
Token Las
Feature activation+0.006
Top resid features:
Vegas
Token Vegas
Feature activation+0.056
Top resid features:
converts
Token converts
Feature activation+0.074
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
The
Token The
Feature activation+0.031
Top resid features:
process
Token process
Feature activation+0.041
Top resid features:
of
Token of
Feature activation+0.062
Top resid features:
transitioning
Token transitioning
Feature activation+0.733
Top resid features:
into
Token into
Feature activation+0.516
Top resid features:
Islam
Token Islam
Feature activation+0.070
Top resid features:
from
Token from
Feature activation+0.609
Top resid features:
a
Token a
Feature activation+0.077
Top resid features:
previous
Token previous
Feature activation+0.020
Top resid features:
these
Token these
Feature activation+0.002
Top resid features:
very
Token very
Feature activation-0.000
Top resid features:
pages
Token pages
Feature activation+0.016
Top resid features:
,
Token,
Feature activation+0.063
Top resid features:
the
Token the
Feature activation+0.039
Top resid features:
demise
Token demise
Feature activation+1.547
Top resid features:
of
Token of
Feature activation+0.857
Top resid features:
Pakistan
Token Pakistan
Feature activation+0.054
Top resid features:
is
Token is
Feature activation+0.046
Top resid features:
"
Token "
Feature activation+0.000
Top resid features:
ine
Tokenine
Feature activation+0.000
Top resid features:
that
Token that
Feature activation+0.034
Top resid features:
backing
Token backing
Feature activation+0.258
Top resid features:
the
Token the
Feature activation+0.107
Top resid features:
UK
Token UK
Feature activation-0.003
Top resid features:
's
Token's
Feature activation+0.051
Top resid features:
exit
Token exit
Feature activation+1.499
Top resid features:
from
Token from
Feature activation+0.521
Top resid features:
the
Token the
Feature activation+0.139
Top resid features:
European
Token European
Feature activation+0.001
Top resid features:
Union
Token Union
Feature activation+0.064
Top resid features:
(
Token (
Feature activation+0.000
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.285
Top resid features:
.
Token.
Feature activation-0.025
Top resid features:
His
Token His
Feature activation+0.043
Top resid features:
move
Token move
Feature activation+0.275
Top resid features:
toward
Token toward
Feature activation+1.567
Top resid features:
libertarian
Token libertarian
Feature activation+0.307
Top resid features:
ism
Tokenism
Feature activation+0.187
Top resid features:
coincided
Token coincided
Feature activation+0.000
Top resid features:
with
Token with
Feature activation+0.000
Top resid features:
a
Token a
Feature activation+0.000
Top resid features:
told
Token told
Feature activation-0.012
Top resid features:
RT
Token RT
Feature activation-0.005
Top resid features:
that
Token that
Feature activation+0.018
Top resid features:
her
Token her
Feature activation+0.037
Top resid features:
daughter
Token daughter
Feature activation+0.007
Top resid features:
becoming
Token becoming
Feature activation+1.508
Top resid features:
an
Token an
Feature activation+0.212
Top resid features:
Internet
Token Internet
Feature activation+0.019
Top resid features:
sensation
Token sensation
Feature activation+0.382
Top resid features:
is
Token is
Feature activation+0.059
Top resid features:
âĢ
Token âĢ
Feature activation+0.000
Top resid features:
s
Tokens
Feature activation-0.007
Top resid features:
roman
Tokenroman
Feature activation-0.031
Top resid features:
or
Token or
Feature activation+0.043
Top resid features:
coming
Token coming
Feature activation+0.074
Top resid features:
-
Token-
Feature activation+0.344
Top resid features:
of
Tokenof
Feature activation+0.714
Top resid features:
-
Token-
Feature activation+0.361
Top resid features:
age
Tokenage
Feature activation+0.532
Top resid features:
story
Token story
Feature activation+0.000
Top resid features:
,
Token,
Feature activation+0.000
Top resid features:
focused
Token focused
Feature activation+0.000
Top resid features:
Du
TokenDu
Feature activation+0.002
Top resid features:
jar
Tokenjar
Feature activation+0.002
Top resid features:
ric
Tokenric
Feature activation+0.012
Top resid features:
said
Token said
Feature activation+0.027
Top resid features:
the
Token the
Feature activation+0.024
Top resid features:
removal
Token removal
Feature activation+0.913
Top resid features:
of
Token of
Feature activation+0.840
Top resid features:
the
Token the
Feature activation+0.196
Top resid features:
coalition
Token coalition
Feature activation+0.166
Top resid features:
was
Token was
Feature activation+0.079
Top resid features:
pending
Token pending
Feature activation+0.000
Top resid features:
that
Token that
Feature activation+0.019
Top resid features:
backing
Token backing
Feature activation+0.320
Top resid features:
the
Token the
Feature activation+0.111
Top resid features:
UK
Token UK
Feature activation+0.036
Top resid features:
's
Token's
Feature activation+0.036
Top resid features:
exit
Token exit
Feature activation+0.873
Top resid features:
from
Token from
Feature activation+0.232
Top resid features:
the
Token the
Feature activation+0.070
Top resid features:
European
Token European
Feature activation+0.010
Top resid features:
Union
Token Union
Feature activation+0.150
Top resid features:
(
Token (
Feature activation+0.008
Top resid features:
converts
Token converts
Feature activation+0.055
Top resid features:
.
Token.
Feature activation+0.011
Top resid features:
The
Token The
Feature activation+0.014
Top resid features:
process
Token process
Feature activation-0.000
Top resid features:
of
Token of
Feature activation+0.033
Top resid features:
transitioning
Token transitioning
Feature activation+0.597
Top resid features:
into
Token into
Feature activation+0.266
Top resid features:
Islam
Token Islam
Feature activation+0.131
Top resid features:
from
Token from
Feature activation+0.361
Top resid features:
a
Token a
Feature activation+0.082
Top resid features:
previous
Token previous
Feature activation+0.012
Top resid features:
hurdle
Token hurdle
Feature activation-0.000
Top resid features:
.
Token.
Feature activation+0.004
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation-0.068
Top resid features:
The
TokenThe
Feature activation+0.006
Top resid features:
birth
Token birth
Feature activation+0.479
Top resid features:
of
Token of
Feature activation+1.159
Top resid features:
the
Token the
Feature activation+0.271
Top resid features:
Travel
Token Travel
Feature activation+0.018
Top resid features:
ing
Tokening
Feature activation+0.014
Top resid features:
Wil
Token Wil
Feature activation+0.003
Top resid features:
bur
Tokenbur
Feature activation+0.037
Top resid features:
that
Token that
Feature activation+0.018
Top resid features:
backing
Token backing
Feature activation+0.174
Top resid features:
the
Token the
Feature activation+0.045
Top resid features:
UK
Token UK
Feature activation+0.011
Top resid features:
's
Token's
Feature activation+0.020
Top resid features:
exit
Token exit
Feature activation+1.122
Top resid features:
from
Token from
Feature activation+0.424
Top resid features:
the
Token the
Feature activation+0.119
Top resid features:
European
Token European
Feature activation+0.002
Top resid features:
Union
Token Union
Feature activation+0.112
Top resid features:
(
Token (
Feature activation+0.025
Top resid features:
they
Token they
Feature activation-0.003
Top resid features:
could
Token could
Feature activation-0.004
Top resid features:
that
Token that
Feature activation-0.041
Top resid features:
its
Token its
Feature activation+0.028
Top resid features:
temporary
Token temporary
Feature activation+0.251
Top resid features:
closure
Token closure
Feature activation+1.590
Top resid features:
was
Token was
Feature activation+0.105
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
most
Token most
Feature activation+0.000
Top resid features:
dire
Token dire
Feature activation+0.000
Top resid features:
effect
Token effect
Feature activation+0.000
Top resid features:
diseases
Token diseases
Feature activation+0.011
Top resid features:
where
Token where
Feature activation+0.018
Top resid features:
the
Token the
Feature activation+0.022
Top resid features:
partial
Token partial
Feature activation+0.121
Top resid features:
removal
Token removal
Feature activation+0.409
Top resid features:
of
Token of
Feature activation+0.431
Top resid features:
the
Token the
Feature activation+0.046
Top resid features:
liver
Token liver
Feature activation+0.070
Top resid features:
due
Token due
Feature activation+0.261
Top resid features:
to
Token to
Feature activation+0.213
Top resid features:
fib
Token fib
Feature activation+0.006
Top resid features:
assistants
Token assistants
Feature activation-0.003
Top resid features:
.
Token.
Feature activation+0.013
Top resid features:
Where
Token Where
Feature activation-0.009
Top resid features:
it
Token it
Feature activation-0.006
Top resid features:
goes
Token goes
Feature activation+0.174
Top resid features:
from
Token from
Feature activation+1.602
Top resid features:
here
Token here
Feature activation+0.164
Top resid features:
will
Token will
Feature activation-0.079
Top resid features:
help
Token help
Feature activation+0.000
Top resid features:
define
Token define
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.004
Top resid features:
two
Token two
Feature activation-0.007
Top resid features:
countries
Token countries
Feature activation-0.015
Top resid features:
must
Token must
Feature activation-0.034
Top resid features:
make
Token make
Feature activation-0.103
Top resid features:
connecting
Token connecting
Feature activation+2.038
Top resid features:
flights
Token flights
Feature activation+0.121
Top resid features:
in
Token in
Feature activation+0.000
Top resid features:
Hong
Token Hong
Feature activation+0.000
Top resid features:
Kong
Token Kong
Feature activation+0.000
Top resid features:
or
Token or
Feature activation+0.000
Top resid features:
That
Token That
Feature activation+0.048
Top resid features:
somehow
Token somehow
Feature activation+0.048
Top resid features:
,
Token,
Feature activation+0.006
Top resid features:
the
Token the
Feature activation+0.060
Top resid features:
loss
Token loss
Feature activation+0.329
Top resid features:
of
Token of
Feature activation+0.490
Top resid features:
life
Token life
Feature activation+0.337
Top resid features:
from
Token from
Feature activation+0.179
Top resid features:
an
Token an
Feature activation+0.072
Top resid features:
ambush
Token ambush
Feature activation+0.143
Top resid features:
in
Token in
Feature activation+0.048
Top resid features:
Ļ
TokenĻ
Feature activation+0.034
Top resid features:
s
Tokens
Feature activation+0.023
Top resid features:
a
Token a
Feature activation+0.010
Top resid features:
coming
Token coming
Feature activation+0.133
Top resid features:
of
Token of
Feature activation+0.380
Top resid features:
age
Token age
Feature activation+1.287
Top resid features:
story
Token story
Feature activation+0.000
Top resid features:
that
Token that
Feature activation+0.000
Top resid features:
âĢ
TokenâĢ
Feature activation+0.000
Top resid features:
Ļ
TokenĻ
Feature activation+0.000
Top resid features:
s
Tokens
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.003
Top resid features:
Ċ
TokenĊ
Feature activation-0.006
Top resid features:
The
TokenThe
Feature activation+0.034
Top resid features:
re
Token re
Feature activation+0.233
Top resid features:
class
Tokenclass
Feature activation+0.432
Top resid features:
ification
Tokenification
Feature activation+0.622
Top resid features:
gave
Token gave
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
FCC
Token FCC
Feature activation+0.000
Top resid features:
the
Token the
Feature activation+0.000
Top resid features:
power
Token power
Feature activation+0.000
Top resid features:

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.02

Head 2: 0.05

Head 3: 0.04

Head 4: 0.06

Head 5: 0.13

Head 6: 0.03

Head 7: 0.09

Head 8: 0.05

Head 9: 0.38

Head 10: 0.09

Head 11: 0.03

Positive logits

postponed3.77

delayed3.47

irreversible3.44

precip3.41

prophecy3.38

prophes3.30

ceremony3.28

foreseen3.26

cataly3.24

postp3.21

timed3.19

preceded3.19

sequence3.16

gradual3.16

versible3.11

event3.09

triggered3.04

awaited3.04

suspense3.04

occurred3.03

Negative logits

databases-3.03

Diversity-2.87

BuyableInstoreAndOnline-2.76

tagging-2.75

diversity-2.74

brands-2.72

Yelp-2.67

inspecting-2.60

typing-2.60

License-2.58

subreddits-2.56

iversity-2.56

formats-2.52

Ort-2.45

interfaces-2.45

filters-2.43

workplaces-2.43

interns-2.40

Collections-2.39

licenses-2.39

INTERVAL 2.205 - 2.450
CONTAINS 0.000%

there
Token there
Feature activation+0.000
is
Token is
Feature activation+0.000
no
Token no
Feature activation+0.000
God
Token God
Feature activation+0.000
)
Token)
Feature activation+1.896
is
Token is
Feature activation+2.450
daunting
Token daunting
Feature activation+1.311
:
Token:
Feature activation+0.082
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.104
What
TokenWhat
Feature activation+0.000

INTERVAL 1.960 - 2.205
CONTAINS 0.000%

Las
Token Las
Feature activation+0.000
Vegas
Token Vegas
Feature activation+0.736
now
Token now
Feature activation+0.583
seems
Token seems
Feature activation+0.935
all
Token all
Feature activation+0.211
but
Token but
Feature activation+1.990
certain
Token certain
Feature activation+0.687
to
Token to
Feature activation+0.620
happen
Token happen
Feature activation+0.482
.
Token.
Feature activation+0.000
After
Token After
Feature activation+0.000

INTERVAL 1.715 - 1.960
CONTAINS 0.000%

,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
demise
Token demise
Feature activation+0.000
of
Token of
Feature activation+0.000
Pakistan
Token Pakistan
Feature activation+0.000
is
Token is
Feature activation+1.841
"
Token "
Feature activation+0.581
ine
Tokenine
Feature activation+0.000
v
Tokenv
Feature activation+0.000
itable
Tokenitable
Feature activation+0.417
"
Token"
Feature activation+0.000
believe
Token believe
Feature activation+0.000
there
Token there
Feature activation+0.000
is
Token is
Feature activation+0.000
no
Token no
Feature activation+0.000
God
Token God
Feature activation+0.000
)
Token)
Feature activation+1.896
is
Token is
Feature activation+2.450
daunting
Token daunting
Feature activation+1.311
:
Token:
Feature activation+0.082
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.104
's
Token's
Feature activation+0.000
exit
Token exit
Feature activation+0.781
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
European
Token European
Feature activation+0.000
Union
Token Union
Feature activation+1.807
(
Token (
Feature activation+0.000
EU
TokenEU
Feature activation+0.000
)
Token)
Feature activation+1.276
was
Token was
Feature activation+1.157
a
Token a
Feature activation+1.320

INTERVAL 1.470 - 1.715
CONTAINS 0.000%

.
Token.
Feature activation+0.000
His
Token His
Feature activation+0.000
move
Token move
Feature activation+0.000
toward
Token toward
Feature activation+0.000
libertarian
Token libertarian
Feature activation+0.000
ism
Tokenism
Feature activation+1.621
coincided
Token coincided
Feature activation+0.047
with
Token with
Feature activation+0.432
a
Token a
Feature activation+0.203
larger
Token larger
Feature activation+0.033
role
Token role
Feature activation+0.000

INTERVAL 1.225 - 1.470
CONTAINS 0.001%

or
Token or
Feature activation+0.000
coming
Token coming
Feature activation+0.000
-
Token-
Feature activation+0.000
of
Tokenof
Feature activation+0.000
-
Token-
Feature activation+0.000
age
Tokenage
Feature activation+1.405
story
Token story
Feature activation+0.000
,
Token,
Feature activation+0.000
focused
Token focused
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
Travel
Token Travel
Feature activation+0.000
ing
Tokening
Feature activation+0.000
Wil
Token Wil
Feature activation+0.000
bur
Tokenbur
Feature activation+0.000
ys
Tokenys
Feature activation+0.189
was
Token was
Feature activation+1.280
a
Token a
Feature activation+1.080
happy
Token happy
Feature activation+0.398
accident
Token accident
Feature activation+0.000
.
Token.
Feature activation+0.000
Warner
Token Warner
Feature activation+0.000
the
Token the
Feature activation+0.000
European
Token European
Feature activation+0.000
Union
Token Union
Feature activation+1.807
(
Token (
Feature activation+0.000
EU
TokenEU
Feature activation+0.000
)
Token)
Feature activation+1.276
was
Token was
Feature activation+1.157
a
Token a
Feature activation+1.320
strategic
Token strategic
Feature activation+0.430
error
Token error
Feature activation+0.000
.
Token.
Feature activation+0.000
Union
Token Union
Feature activation+1.807
(
Token (
Feature activation+0.000
EU
TokenEU
Feature activation+0.000
)
Token)
Feature activation+1.276
was
Token was
Feature activation+1.157
a
Token a
Feature activation+1.320
strategic
Token strategic
Feature activation+0.430
error
Token error
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
is
Token is
Feature activation+0.000
no
Token no
Feature activation+0.000
God
Token God
Feature activation+0.000
)
Token)
Feature activation+1.896
is
Token is
Feature activation+2.450
daunting
Token daunting
Feature activation+1.311
:
Token:
Feature activation+0.082
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.104
What
TokenWhat
Feature activation+0.000
will
Token will
Feature activation+0.000

INTERVAL 0.980 - 1.225
CONTAINS 0.003%

fact
Token fact
Feature activation+0.000
that
Token that
Feature activation+0.000
asc
Token asc
Feature activation+0.000
ension
Tokenension
Feature activation+0.292
had
Token had
Feature activation+0.731
taken
Token taken
Feature activation+1.137
place
Token place
Feature activation+0.468
(
Token (
Feature activation+0.000
M
TokenM
Feature activation+0.000
oses
Tokenoses
Feature activation+0.000
).
Token).
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
re
Token re
Feature activation+0.000
class
Tokenclass
Feature activation+0.000
ification
Tokenification
Feature activation+1.168
gave
Token gave
Feature activation+0.000
the
Token the
Feature activation+0.000
FCC
Token FCC
Feature activation+0.000
the
Token the
Feature activation+0.000
power
Token power
Feature activation+0.000
European
Token European
Feature activation+0.000
Union
Token Union
Feature activation+1.807
(
Token (
Feature activation+0.000
EU
TokenEU
Feature activation+0.000
)
Token)
Feature activation+1.276
was
Token was
Feature activation+1.157
a
Token a
Feature activation+1.320
strategic
Token strategic
Feature activation+0.430
error
Token error
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Windows
Token Windows
Feature activation+0.000
NT
Token NT
Feature activation+0.000
4
Token 4
Feature activation+0.000
to
Token to
Feature activation+0.000
Ubuntu
Token Ubuntu
Feature activation+0.000
Linux
Token Linux
Feature activation+1.039
Terminal
Token Terminal
Feature activation+0.000
Server
Token Server
Feature activation+0.289
Project
Token Project
Feature activation+0.000
(
Token (
Feature activation+0.000
LT
TokenLT
Feature activation+0.000
(
Token (
Feature activation+0.000
LT
TokenLT
Feature activation+0.000
SP
TokenSP
Feature activation+0.000
)
Token)
Feature activation+0.724
might
Token might
Feature activation+0.870
not
Token not
Feature activation+1.081
get
Token get
Feature activation+0.719
an
Token an
Feature activation+0.451
A
Token A
Feature activation+0.000
+
Token+
Feature activation+0.000
mark
Token mark
Feature activation+0.000

INTERVAL 0.735 - 0.980
CONTAINS 0.004%

the
Token the
Feature activation+0.000
purchase
Token purchase
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
palace
Token palace
Feature activation+0.000
was
Token was
Feature activation+0.957
a
Token a
Feature activation+0.597
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
private
Tokenprivate
Feature activation+0.049
deal
Token deal
Feature activation+0.131
by
Token by
Feature activation+0.000
Scotland
Token Scotland
Feature activation+0.000
's
Token's
Feature activation+0.000
forthcoming
Token forthcoming
Feature activation+0.000
independence
Token independence
Feature activation+0.018
referendum
Token referendum
Feature activation+0.798
-
Token -
Feature activation+0.000
and
Token and
Feature activation+0.000
many
Token many
Feature activation+0.000
also
Token also
Feature activation+0.000
waved
Token waved
Feature activation+0.000
that
Token that
Feature activation+0.000
backing
Token backing
Feature activation+0.000
the
Token the
Feature activation+0.000
UK
Token UK
Feature activation+0.000
's
Token's
Feature activation+0.000
exit
Token exit
Feature activation+0.781
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
European
Token European
Feature activation+0.000
Union
Token Union
Feature activation+1.807
(
Token (
Feature activation+0.000
country
Token country
Feature activation+0.000
of
Token of
Feature activation+0.000
Niger
Token Niger
Feature activation+0.000
is
Token is
Feature activation+1.173
not
Token not
Feature activation+0.898
just
Token just
Feature activation+0.757
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Ben
TokenBen
Feature activation+0.000
ghazi
Tokenghazi
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
#
Token #
Feature activation+0.000
Net
TokenNet
Feature activation+0.000
Ne
TokenNe
Feature activation+0.000
utral
Tokenutral
Feature activation+0.000
ity
Tokenity
Feature activation+0.000
is
Token is
Feature activation+0.788
an
Token an
Feature activation+0.650
attack
Token attack
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
LGBTQ
Token LGBTQ
Feature activation+0.000

INTERVAL 0.490 - 0.735
CONTAINS 0.015%

me
Token me
Feature activation+0.000
,
Token,
Feature activation+0.000
electing
Token electing
Feature activation+0.000
the
Token the
Feature activation+0.000
SNP
Token SNP
Feature activation+0.000
is
Token is
Feature activation+0.595
the
Token the
Feature activation+0.320
new
Token new
Feature activation+0.110
politics
Token politics
Feature activation+0.000
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
coming
Token coming
Feature activation+0.000
demographic
Token demographic
Feature activation+0.000
shift
Token shift
Feature activation+0.177
has
Token has
Feature activation+0.590
been
Token been
Feature activation+0.518
the
Token the
Feature activation+0.142
subject
Token subject
Feature activation+0.000
of
Token of
Feature activation+0.000
much
Token much
Feature activation+0.000
.
Token.
Feature activation+0.000
That
Token That
Feature activation+0.000
shouldn
Token shouldn
Feature activation+0.196
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.580
be
Token be
Feature activation+0.484
the
Token the
Feature activation+0.280
case
Token case
Feature activation+0.126
.
Token.
Feature activation+0.000
That
Token That
Feature activation+0.000
ians
Tokenians
Feature activation+0.000
,
Token,
Feature activation+0.000
winning
Token winning
Feature activation+0.000
the
Token the
Feature activation+0.000
war
Token war
Feature activation+0.000
was
Token was
Feature activation+0.576
a
Token a
Feature activation+0.320
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
double
Tokendouble
Feature activation+0.000
victory
Token victory
Feature activation+0.000
regional
Token regional
Feature activation+0.000
government
Token government
Feature activation+0.000
has
Token has
Feature activation+0.000
called
Token called
Feature activation+0.000
a
Token a
Feature activation+0.000
referendum
Token referendum
Feature activation+0.531
for
Token for
Feature activation+0.058
9
Token 9
Feature activation+0.000
November
Token November
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000

INTERVAL 0.245 - 0.490
CONTAINS 0.038%

alien
Token alien
Feature activation+0.000
âĢĵ
TokenâĢĵ
Feature activation+0.000
whose
Tokenwhose
Feature activation+0.000
identity
Token identity
Feature activation+0.000
is
Token is
Feature activation+0.257
kept
Token kept
Feature activation+0.401
anonymous
Token anonymous
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
indictment
Token indictment
Feature activation+0.000
âĢĵ
TokenâĢĵ
Feature activation+0.000
(
Token (
Feature activation+0.000
EU
TokenEU
Feature activation+0.000
)
Token)
Feature activation+1.276
was
Token was
Feature activation+1.157
a
Token a
Feature activation+1.320
strategic
Token strategic
Feature activation+0.430
error
Token error
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
He
TokenHe
Feature activation+0.000
make
Token make
Feature activation+0.000
connecting
Token connecting
Feature activation+0.000
flights
Token flights
Feature activation+1.176
in
Token in
Feature activation+0.258
Hong
Token Hong
Feature activation+0.000
Kong
Token Kong
Feature activation+0.371
or
Token or
Feature activation+0.000
Seoul
Token Seoul
Feature activation+0.385
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
urban
Token urban
Feature activation+0.000
freeway
Token freeway
Feature activation+0.000
removal
Token removal
Feature activation+0.000
has
Token has
Feature activation+0.142
become
Token become
Feature activation+0.307
a
Token a
Feature activation+0.247
trend
Token trend
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
the
Token the
Feature activation+0.000
Rays
Token Rays
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
Dodgers
Token Dodgers
Feature activation+0.000
in
Token in
Feature activation+0.261
the
Token the
Feature activation+0.336
fall
Token fall
Feature activation+0.250
of
Token of
Feature activation+0.000
2014
Token 2014
Feature activation+0.140
.
Token.
Feature activation+0.000

INTERVAL 0.000 - 0.245
CONTAINS 99.938%

among
Token among
Feature activation+0.000
the
Token the
Feature activation+0.000
Ing
Token Ing
Feature activation+0.000
vae
Tokenvae
Feature activation+0.000
onic
Tokenonic
Feature activation+0.000
languages
Token languages
Feature activation+0.000
,
Token,
Feature activation+0.000
together
Token together
Feature activation+0.000
with
Token with
Feature activation+0.000
Low
Token Low
Feature activation+0.000
German
Token German
Feature activation+0.000
rally
Token rally
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
Ohio
Token Ohio
Feature activation+0.000
State
Token State
Feature activation+0.000
University
Token University
Feature activation+0.000
.
Token.
Feature activation+0.000
Hide
Token Hide
Feature activation+0.000
Caption
Token Caption
Feature activation+0.000
4
Token 4
Feature activation+0.000
of
Token of
Feature activation+0.000
are
Token are
Feature activation+0.000
visitors
Token visitors
Feature activation+0.000
from
Token from
Feature activation+0.000
outside
Token outside
Feature activation+0.000
the
Token the
Feature activation+0.000
Las
Token Las
Feature activation+0.000
Vegas
Token Vegas
Feature activation+0.000
area
Token area
Feature activation+0.000
and
Token and
Feature activation+0.000
here
Token here
Feature activation+0.000
for
Token for
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Add
TokenAdd
Feature activation+0.000
in
Token in
Feature activation+0.000
your
Token your
Feature activation+0.000
melted
Token melted
Feature activation+0.000
butter
Token butter
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Then
TokenThen
Feature activation+0.000
home
Token home
Feature activation+0.000
address
Token address
Feature activation+0.000
and
Token and
Feature activation+0.000
phone
Token phone
Feature activation+0.000
number
Token number
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Gro
TokenGro
Feature activation+0.000
oming
Tokenoming
Feature activation+0.000
Your
Token Your
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

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

Top feature 9 in H8.9: (feature 10894)

TOP ACTIVATIONS
MAX = 2.589

X
TokenX
Feature activation+0.000
ML
TokenML
Feature activation+0.220
X
Token X
Feature activation+0.000
t
Tokent
Feature activation+0.000
ract
Tokenract
Feature activation+0.000
,
Token,
Feature activation+2.589
but
Token but
Feature activation+1.341
they
Token they
Feature activation+0.590
all
Token all
Feature activation+0.293
have
Token have
Feature activation+0.391
space
Token space
Feature activation+0.000
Dragon
Token Dragon
Feature activation+0.000
's
Token's
Feature activation+0.000
Lair
Token Lair
Feature activation+0.000
and
Token and
Feature activation+0.729
many
Token many
Feature activation+1.812
more
Token more
Feature activation+2.215
.
Token.
Feature activation+0.326
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
A
TokenA
Feature activation+0.000
visitor
Token visitor
Feature activation+0.000
tours
Token tours
Feature activation+0.000
ark
Tokenark
Feature activation+0.000
ali
Tokenali
Feature activation+0.375
to
Token to
Feature activation+0.959
name
Token name
Feature activation+1.027
a
Token a
Feature activation+0.474
few
Token few
Feature activation+2.085
.
Token.
Feature activation+1.462
He
Token He
Feature activation+0.409
worked
Token worked
Feature activation+0.688
with
Token with
Feature activation+0.867
each
Token each
Feature activation+1.746
and
Token and
Feature activation+0.164
The
Token The
Feature activation+0.000
Guth
Token Guth
Feature activation+0.000
rie
Tokenrie
Feature activation+0.000
Theater
Token Theater
Feature activation+0.459
,
Token,
Feature activation+1.978
among
Token among
Feature activation+1.256
others
Token others
Feature activation+1.399
.
Token.
Feature activation+0.137
He
Token He
Feature activation+0.000
was
Token was
Feature activation+0.017
to
Token to
Feature activation+0.000
Ar
Token Ar
Feature activation+0.000
sh
Tokensh
Feature activation+0.000
in
Tokenin
Feature activation+0.000
ov
Tokenov
Feature activation+0.009
,
Token,
Feature activation+1.952
and
Token and
Feature activation+1.061
especially
Token especially
Feature activation+0.597
amongst
Token amongst
Feature activation+0.313
the
Token the
Feature activation+0.011
rank
Token rank
Feature activation+0.000
ug
Tokenug
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.493
Caroline
Token Caroline
Feature activation+0.000
Shaw
Token Shaw
Feature activation+0.000
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.441
Ċ
TokenĊ
Feature activation+0.672
The
TokenThe
Feature activation+0.515
album
Token album
Feature activation+0.000
was
Token was
Feature activation+0.000
Powell
Token Powell
Feature activation+0.000
and
Token and
Feature activation+0.571
Patrick
Token Patrick
Feature activation+0.000
McN
Token McN
Feature activation+0.000
air
Tokenair
Feature activation+0.386
,
Token,
Feature activation+1.921
but
Token but
Feature activation+1.213
these
Token these
Feature activation+1.406
players
Token players
Feature activation+0.186
cannot
Token cannot
Feature activation+0.000
do
Token do
Feature activation+0.000
Kenn
Token Kenn
Feature activation+0.000
es
Tokenes
Feature activation+0.000
aw
Tokenaw
Feature activation+0.000
and
Token and
Feature activation+0.000
Atlanta
Token Atlanta
Feature activation+0.109
,
Token,
Feature activation+1.902
where
Token where
Feature activation+0.280
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
wounded
Token wounded
Feature activation+0.000
twice
Token twice
Feature activation+0.000
.
Token.
Feature activation+0.672
com
Tokencom
Feature activation+0.312
,
Token,
Feature activation+0.557
and
Token and
Feature activation+0.625
Starbucks
Token Starbucks
Feature activation+0.371
.
Token.
Feature activation+1.841
Ċ
TokenĊ
Feature activation+0.409
Ċ
TokenĊ
Feature activation+1.115
Today
TokenToday
Feature activation+1.075
,
Token,
Feature activation+1.072
the
Token the
Feature activation+1.209
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Smithsonian
Token Smithsonian
Feature activation+0.000
(
Token (
Feature activation+0.451
2013
Token2013
Feature activation+0.471
),
Token),
Feature activation+1.827
but
Token but
Feature activation+1.625
the
Token the
Feature activation+0.563
discovery
Token discovery
Feature activation+0.000
of
Token of
Feature activation+0.094
his
Token his
Feature activation+0.000
,
Token,
Feature activation+0.101
Dragon
Token Dragon
Feature activation+0.000
's
Token's
Feature activation+0.000
Lair
Token Lair
Feature activation+0.000
and
Token and
Feature activation+0.729
many
Token many
Feature activation+1.812
more
Token more
Feature activation+2.215
.
Token.
Feature activation+0.326
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
A
TokenA
Feature activation+0.000
visitor
Token visitor
Feature activation+0.000
Secret
Token Secret
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Learn
TokenLearn
Feature activation+0.000
more
Token more
Feature activation+0.381
about
Token about
Feature activation+1.806
Christina
Token Christina
Feature activation+0.101
and
Token and
Feature activation+0.782
follow
Token follow
Feature activation+0.000
her
Token her
Feature activation+0.000
@
Token @
Feature activation+0.000
employers
Token employers
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
assistance
Token assistance
Feature activation+0.000
allows
Token allows
Feature activation+0.000
these
Token these
Feature activation+1.777
companies
Token companies
Feature activation+0.000
to
Token to
Feature activation+0.000
pay
Token pay
Feature activation+0.000
their
Token their
Feature activation+0.218
workers
Token workers
Feature activation+0.000
few
Token few
Feature activation+2.085
.
Token.
Feature activation+1.462
He
Token He
Feature activation+0.409
worked
Token worked
Feature activation+0.688
with
Token with
Feature activation+0.867
each
Token each
Feature activation+1.746
and
Token and
Feature activation+0.000
every
Token every
Feature activation+0.914
well
Token well
Feature activation+0.117
known
Token known
Feature activation+1.019
music
Token music
Feature activation+0.000
sy
Tokensy
Feature activation+0.000
were
Token were
Feature activation+1.296
the
Token the
Feature activation+0.799
panel
Token panel
Feature activation+0.000
ists
Tokenists
Feature activation+0.000
.
Token.
Feature activation+1.720
Ċ
TokenĊ
Feature activation+0.295
Ċ
TokenĊ
Feature activation+0.779
Interestingly
TokenInterestingly
Feature activation+0.589
enough
Token enough
Feature activation+0.684
,
Token,
Feature activation+1.387
tab
Token tab
Feature activation+0.000
un
Tokenun
Feature activation+0.000
,
Token,
Feature activation+0.000
V
Token V
Feature activation+0.000
X
TokenX
Feature activation+0.000
.
Token.
Feature activation+1.719
Ċ
TokenĊ
Feature activation+0.103
Ċ
TokenĊ
Feature activation+1.323
3
Token3
Feature activation+0.097
.
Token.
Feature activation+0.288
During
Token During
Feature activation+0.478
Evangel
Token Evangel
Feature activation+0.000
ical
Tokenical
Feature activation+0.000
,
Token,
Feature activation+0.280
say
Token say
Feature activation+1.137
officials
Token officials
Feature activation+0.000
of
Token of
Feature activation+1.710
the
Token the
Feature activation+0.763
Massachusetts
Token Massachusetts
Feature activation+0.000
Bay
Token Bay
Feature activation+0.000
Transportation
Token Transportation
Feature activation+0.000
Authority
Token Authority
Feature activation+0.000
ako
Tokenako
Feature activation+0.214
and
Token and
Feature activation+0.894
th
Token th
Feature activation+0.000
res
Tokenres
Feature activation+0.000
her
Tokenher
Feature activation+0.000
are
Token are
Feature activation+1.695
among
Token among
Feature activation+1.264
the
Token the
Feature activation+0.581
species
Token species
Feature activation+0.290
that
Token that
Feature activation+0.000
can
Token can
Feature activation+0.000
Rush
Token Rush
Feature activation+0.000
Limbaugh
Token Limbaugh
Feature activation+0.000
,
Token,
Feature activation+0.014
Sean
Token Sean
Feature activation+0.000
Hannity
Token Hannity
Feature activation+0.433
.
Token.
Feature activation+1.679
Ċ
TokenĊ
Feature activation+0.567
Ċ
TokenĊ
Feature activation+1.148
In
TokenIn
Feature activation+0.403
a
Token a
Feature activation+0.099
recently
Token recently
Feature activation+0.000
and
Token and
Feature activation+0.421
Tal
Token Tal
Feature activation+0.000
im
Tokenim
Feature activation+0.000
h
Tokenh
Feature activation+0.000
ane
Tokenane
Feature activation+0.013
,
Token,
Feature activation+1.643
all
Token all
Feature activation+0.123
popular
Token popular
Feature activation+0.317
hang
Token hang
Feature activation+0.000
-
Token-
Feature activation+0.000
out
Tokenout
Feature activation+0.000

Top DFA by src position
MAX = 1.378

oup
Tokenoup
Feature activation+0.016
Top resid features:
,
Token,
Feature activation+0.467
Top resid features:
H
Token H
Feature activation+0.004
Top resid features:
XT
TokenXT
Feature activation+0.016
Top resid features:
,
Token,
Feature activation+0.730
Top resid features:
and
Token and
Feature activation+1.057
Top resid features:
Ha
Token Ha
Feature activation+0.002
Top resid features:
X
TokenX
Feature activation+0.003
Top resid features:
ML
TokenML
Feature activation+0.197
Top resid features:
X
Token X
Feature activation+0.007
Top resid features:
t
Tokent
Feature activation-0.001
Top resid features:
Man
TokenMan
Feature activation+0.019
Top resid features:
,
Token,
Feature activation+0.420
Top resid features:
Street
Token Street
Feature activation+0.019
Top resid features:
Fighter
Token Fighter
Feature activation+0.027
Top resid features:
2
Token 2
Feature activation+0.063
Top resid features:
,
Token,
Feature activation+0.473
Top resid features:
Gal
Token Gal
Feature activation+0.014
Top resid features:
aga
Tokenaga
Feature activation+0.041
Top resid features:
,
Token,
Feature activation+0.407
Top resid features:
Cent
Token Cent
Feature activation+0.005
Top resid features:
ip
Tokenip
Feature activation+0.003
Top resid features:
Sang
Token Sang
Feature activation+0.001
Top resid features:
am
Tokenam
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.144
Top resid features:
Sri
Token Sri
Feature activation+0.004
Top resid features:
420
Token 420
Feature activation+0.002
Top resid features:
,
Token,
Feature activation+0.312
Top resid features:
Jag
Token Jag
Feature activation+0.004
Top resid features:
te
Tokente
Feature activation+0.002
Top resid features:
R
Token R
Feature activation+0.001
Top resid features:
aho
Tokenaho
Feature activation+0.008
Top resid features:
,
Token,
Feature activation+0.244
Top resid features:
,
Token,
Feature activation+0.278
Top resid features:
The
Token The
Feature activation+0.051
Top resid features:
Round
Token Round
Feature activation+0.001
Top resid features:
about
Tokenabout
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.335
Top resid features:
and
Token and
Feature activation+0.818
Top resid features:
The
Token The
Feature activation+0.072
Top resid features:
Guth
Token Guth
Feature activation+0.011
Top resid features:
rie
Tokenrie
Feature activation+0.023
Top resid features:
Theater
Token Theater
Feature activation+0.308
Top resid features:
,
Token,
Feature activation+0.060
Top resid features:
ure
Tokenure
Feature activation+0.003
Top resid features:
,
Token,
Feature activation+0.141
Top resid features:
from
Token from
Feature activation+0.262
Top resid features:
Maxim
Token Maxim
Feature activation+0.002
Top resid features:
off
Tokenoff
Feature activation+0.046
Top resid features:
to
Token to
Feature activation+0.310
Top resid features:
Ar
Token Ar
Feature activation+0.020
Top resid features:
sh
Tokensh
Feature activation+0.004
Top resid features:
in
Tokenin
Feature activation+0.004
Top resid features:
ov
Tokenov
Feature activation+0.085
Top resid features:
,
Token,
Feature activation+0.235
Top resid features:
<|endoftext|>
Token<|endoftext|>
Feature activation+0.234
Top resid features:
,
Token,
Feature activation+0.012
Top resid features:
including
Token including
Feature activation+0.039
Top resid features:
Chris
Token Chris
Feature activation-0.000
Top resid features:
Brown
Token Brown
Feature activation+0.005
Top resid features:
,
Token,
Feature activation+0.122
Top resid features:
James
Token James
Feature activation+0.001
Top resid features:
Wilson
Token Wilson
Feature activation+0.020
Top resid features:
,
Token,
Feature activation+0.451
Top resid features:
Nick
Token Nick
Feature activation+0.002
Top resid features:
Powell
Token Powell
Feature activation+0.045
Top resid features:
and
Token and
Feature activation+1.143
Top resid features:
Patrick
Token Patrick
Feature activation+0.008
Top resid features:
McN
Token McN
Feature activation+0.002
Top resid features:
air
Tokenair
Feature activation+0.108
Top resid features:
,
Token,
Feature activation+0.106
Top resid features:
but
Token but
Feature activation+0.000
Top resid features:
boro
Tokenboro
Feature activation+0.158
Top resid features:
,
Token,
Feature activation+0.294
Top resid features:
Kenn
Token Kenn
Feature activation+0.004
Top resid features:
es
Tokenes
Feature activation-0.009
Top resid features:
aw
Tokenaw
Feature activation+0.039
Top resid features:
and
Token and
Feature activation+1.378
Top resid features:
Atlanta
Token Atlanta
Feature activation+0.385
Top resid features:
,
Token,
Feature activation+0.109
Top resid features:
where
Token where
Feature activation+0.000
Top resid features:
he
Token he
Feature activation+0.000
Top resid features:
was
Token was
Feature activation+0.000
Top resid features:
Depot
Token Depot
Feature activation+0.098
Top resid features:
,
Token,
Feature activation+0.150
Top resid features:
Intel
Token Intel
Feature activation+0.061
Top resid features:
,
Token,
Feature activation+0.138
Top resid features:
and
Token and
Feature activation+0.128
Top resid features:
Dell
Token Dell
Feature activation+0.241
Top resid features:
Computer
Token Computer
Feature activation+0.084
Top resid features:
;
Token;
Feature activation+0.168
Top resid features:
the
Token the
Feature activation+0.054
Top resid features:
1990
Token 1990
Feature activation+0.013
Top resid features:
s
Tokens
Feature activation+0.056
Top resid features:
gen
Tokengen
Feature activation+0.001
Top resid features:
heim
Tokenheim
Feature activation+0.025
Top resid features:
(
Token (
Feature activation+0.027
Top resid features:
2000
Token2000
Feature activation+0.008
Top resid features:
)
Token)
Feature activation+0.327
Top resid features:
and
Token and
Feature activation+1.002
Top resid features:
the
Token the
Feature activation+0.493
Top resid features:
Smithsonian
Token Smithsonian
Feature activation+0.200
Top resid features:
(
Token (
Feature activation+0.022
Top resid features:
2013
Token2013
Feature activation+0.006
Top resid features:
),
Token),
Feature activation+0.175
Top resid features:
Man
TokenMan
Feature activation+0.027
Top resid features:
,
Token,
Feature activation+0.323
Top resid features:
Street
Token Street
Feature activation+0.038
Top resid features:
Fighter
Token Fighter
Feature activation+0.050
Top resid features:
2
Token 2
Feature activation+0.061
Top resid features:
,
Token,
Feature activation+0.368
Top resid features:
Gal
Token Gal
Feature activation+0.018
Top resid features:
aga
Tokenaga
Feature activation+0.056
Top resid features:
,
Token,
Feature activation+0.333
Top resid features:
Cent
Token Cent
Feature activation+0.009
Top resid features:
ip
Tokenip
Feature activation+0.004
Top resid features:
âĢ
TokenâĢ
Feature activation+0.002
Top resid features:
Ļ
TokenĻ
Feature activation+0.003
Top resid features:
s
Tokens
Feature activation+0.055
Top resid features:
Secret
Token Secret
Feature activation+0.019
Top resid features:
Ċ
TokenĊ
Feature activation+0.076
Top resid features:
Ċ
TokenĊ
Feature activation+0.375
Top resid features:
Learn
TokenLearn
Feature activation+0.023
Top resid features:
more
Token more
Feature activation-0.020
Top resid features:
about
Token about
Feature activation+0.019
Top resid features:
Christina
Token Christina
Feature activation+0.000
Top resid features:
and
Token and
Feature activation+0.000
Top resid features:
to
Token to
Feature activation-0.024
Top resid features:
Walmart
Token Walmart
Feature activation-0.018
Top resid features:
,
Token,
Feature activation+0.012
Top resid features:
McDonald
Token McDonald
Feature activation+0.243
Top resid features:
s
Tokens
Feature activation+0.252
Top resid features:
,
Token,
Feature activation+0.909
Top resid features:
and
Token and
Feature activation+0.672
Top resid features:
other
Token other
Feature activation+0.120
Top resid features:
low
Token low
Feature activation-0.003
Top resid features:
-
Token-
Feature activation+0.001
Top resid features:
wage
Tokenwage
Feature activation+0.002
Top resid features:
Guide
Token Guide
Feature activation+0.003
Top resid features:
,
Token,
Feature activation+0.305
Top resid features:
Kath
Token Kath
Feature activation+0.002
Top resid features:
put
Tokenput
Feature activation+0.007
Top resid features:
li
Tokenli
Feature activation+0.020
Top resid features:
,
Token,
Feature activation+0.310
Top resid features:
Band
Token Band
Feature activation+0.009
Top resid features:
ini
Tokenini
Feature activation+0.029
Top resid features:
,
Token,
Feature activation+0.300
Top resid features:
An
Token An
Feature activation+0.004
Top resid features:
ark
Tokenark
Feature activation+0.004
Top resid features:
product
Token product
Feature activation+0.000
Top resid features:
manager
Token manager
Feature activation+0.001
Top resid features:
of
Token of
Feature activation+0.012
Top resid features:
Zero
Token Zero
Feature activation+0.001
Top resid features:
Block
TokenBlock
Feature activation+0.001
Top resid features:
,
Token,
Feature activation+0.256
Top resid features:
Ole
Token Ole
Feature activation+0.003
Top resid features:
h
Tokenh
Feature activation+0.001
Top resid features:
Z
Token Z
Feature activation+0.002
Top resid features:
ad
Tokenad
Feature activation-0.000
Top resid features:
ore
Tokenore
Feature activation+0.018
Top resid features:
and
Token and
Feature activation+0.028
Top resid features:
ph
Token ph
Feature activation+0.001
Top resid features:
os
Tokenos
Feature activation+0.001
Top resid features:
g
Tokeng
Feature activation-0.001
Top resid features:
ene
Tokenene
Feature activation+0.090
Top resid features:
;
Token;
Feature activation+0.296
Top resid features:
blister
Token blister
Feature activation-0.000
Top resid features:
agents
Token agents
Feature activation+0.111
Top resid features:
âĢĶ
TokenâĢĶ
Feature activation+0.173
Top resid features:
must
Tokenmust
Feature activation+0.003
Top resid features:
ard
Tokenard
Feature activation+0.010
Top resid features:
of
Token of
Feature activation+0.000
Top resid features:
Christ
Token Christ
Feature activation+0.005
Top resid features:
,
Token,
Feature activation+0.021
Top resid features:
Scientist
Token Scientist
Feature activation+0.004
Top resid features:
;
Token;
Feature activation+0.194
Top resid features:
and
Token and
Feature activation+0.366
Top resid features:
the
Token the
Feature activation+0.177
Top resid features:
Boston
Token Boston
Feature activation+0.017
Top resid features:
Chinese
Token Chinese
Feature activation-0.001
Top resid features:
Evangel
Token Evangel
Feature activation+0.013
Top resid features:
ical
Tokenical
Feature activation+0.022
Top resid features:
ocean
Token ocean
Feature activation+0.000
Top resid features:
ic
Tokenic
Feature activation+0.006
Top resid features:
whit
Token whit
Feature activation+0.009
Top resid features:
et
Tokenet
Feature activation+0.000
Top resid features:
ip
Tokenip
Feature activation+0.138
Top resid features:
,
Token,
Feature activation+0.379
Top resid features:
short
Token short
Feature activation+0.031
Top resid features:
fin
Tokenfin
Feature activation+0.109
Top resid features:
m
Token m
Feature activation+0.003
Top resid features:
ako
Tokenako
Feature activation+0.062
Top resid features:
and
Token and
Feature activation+0.306
Top resid features:
Fox
Token Fox
Feature activation+0.015
Top resid features:
News
Token News
Feature activation-0.037
Top resid features:
,
Token,
Feature activation+0.148
Top resid features:
Rush
Token Rush
Feature activation+0.069
Top resid features:
Limbaugh
Token Limbaugh
Feature activation+0.229
Top resid features:
,
Token,
Feature activation+0.832
Top resid features:
Sean
Token Sean
Feature activation+0.054
Top resid features:
Hannity
Token Hannity
Feature activation+0.375
Top resid features:
.
Token.
Feature activation+0.502
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
Ċ
TokenĊ
Feature activation+0.000
Top resid features:
T
Token T
Feature activation+0.009
Top resid features:
arl
Tokenarl
Feature activation+0.003
Top resid features:
aba
Tokenaba
Feature activation+0.037
Top resid features:
ÅŁ
TokenÅŁ
Feature activation+0.030
Top resid features:
ı
Tokenı
Feature activation+0.025
Top resid features:
and
Token and
Feature activation+0.924
Top resid features:
Tal
Token Tal
Feature activation+0.036
Top resid features:
im
Tokenim
Feature activation+0.037
Top resid features:
h
Tokenh
Feature activation+0.027
Top resid features:
ane
Tokenane
Feature activation+0.093
Top resid features:
,
Token,
Feature activation+0.237
Top resid features:

Decoder Weights Distribution

Head 0: 0.04

Head 1: 0.04

Head 2: 0.05

Head 3: 0.10

Head 4: 0.03

Head 5: 0.05

Head 6: 0.03

Head 7: 0.04

Head 8: 0.10

Head 9: 0.38

Head 10: 0.11

Head 11: 0.03

Positive logits

These3.77

these3.55

these3.20

names3.17

These3.16

combinations2.96

categories2.96

THESE2.96

Lists2.91

identities2.89

list2.76

those2.71

lists2.70

labels2.69

listings2.68

none2.66

ones2.64

rosters2.63

favorites2.60

distinctions2.55

Negative logits

awa-2.25

baum-2.15

onse-2.12

hran-2.07

oin-2.06

arah-2.05

veland-2.01

olean-2.00

lvl-1.97

-1.96

bra-1.94

-1.93

vec-1.91

KA-1.89

oria-1.89

Lv-1.88

haw-1.88

Assistant-1.88

biologist-1.87

rate-1.87

INTERVAL 2.330 - 2.589
CONTAINS 0.000%

X
TokenX
Feature activation+0.000
ML
TokenML
Feature activation+0.220
X
Token X
Feature activation+0.000
t
Tokent
Feature activation+0.000
ract
Tokenract
Feature activation+0.000
,
Token,
Feature activation+2.589
but
Token but
Feature activation+1.341
they
Token they
Feature activation+0.590
all
Token all
Feature activation+0.293
have
Token have
Feature activation+0.391
space
Token space
Feature activation+0.000

INTERVAL 2.071 - 2.330
CONTAINS 0.000%

Dragon
Token Dragon
Feature activation+0.000
's
Token's
Feature activation+0.000
Lair
Token Lair
Feature activation+0.000
and
Token and
Feature activation+0.729
many
Token many
Feature activation+1.812
more
Token more
Feature activation+2.215
.
Token.
Feature activation+0.326
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
A
TokenA
Feature activation+0.000
visitor
Token visitor
Feature activation+0.000
tours
Token tours
Feature activation+0.000
ark
Tokenark
Feature activation+0.000
ali
Tokenali
Feature activation+0.375
to
Token to
Feature activation+0.959
name
Token name
Feature activation+1.027
a
Token a
Feature activation+0.474
few
Token few
Feature activation+2.085
.
Token.
Feature activation+1.462
He
Token He
Feature activation+0.409
worked
Token worked
Feature activation+0.688
with
Token with
Feature activation+0.867
each
Token each
Feature activation+1.746

INTERVAL 1.812 - 2.071
CONTAINS 0.001%

.
Token.
Feature activation+0.672
com
Tokencom
Feature activation+0.312
,
Token,
Feature activation+0.557
and
Token and
Feature activation+0.625
Starbucks
Token Starbucks
Feature activation+0.371
.
Token.
Feature activation+1.841
Ċ
TokenĊ
Feature activation+0.409
Ċ
TokenĊ
Feature activation+1.115
Today
TokenToday
Feature activation+1.075
,
Token,
Feature activation+1.072
the
Token the
Feature activation+1.209
ug
Tokenug
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.493
Caroline
Token Caroline
Feature activation+0.000
Shaw
Token Shaw
Feature activation+0.000
.
Token.
Feature activation+1.949
Ċ
TokenĊ
Feature activation+0.441
Ċ
TokenĊ
Feature activation+0.672
The
TokenThe
Feature activation+0.515
album
Token album
Feature activation+0.000
was
Token was
Feature activation+0.000
Powell
Token Powell
Feature activation+0.000
and
Token and
Feature activation+0.571
Patrick
Token Patrick
Feature activation+0.000
McN
Token McN
Feature activation+0.000
air
Tokenair
Feature activation+0.386
,
Token,
Feature activation+1.921
but
Token but
Feature activation+1.213
these
Token these
Feature activation+1.406
players
Token players
Feature activation+0.186
cannot
Token cannot
Feature activation+0.000
do
Token do
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Smithsonian
Token Smithsonian
Feature activation+0.000
(
Token (
Feature activation+0.451
2013
Token2013
Feature activation+0.471
),
Token),
Feature activation+1.827
but
Token but
Feature activation+1.625
the
Token the
Feature activation+0.563
discovery
Token discovery
Feature activation+0.000
of
Token of
Feature activation+0.094
his
Token his
Feature activation+0.000
and
Token and
Feature activation+0.164
The
Token The
Feature activation+0.000
Guth
Token Guth
Feature activation+0.000
rie
Tokenrie
Feature activation+0.000
Theater
Token Theater
Feature activation+0.459
,
Token,
Feature activation+1.978
among
Token among
Feature activation+1.256
others
Token others
Feature activation+1.399
.
Token.
Feature activation+0.137
He
Token He
Feature activation+0.000
was
Token was
Feature activation+0.017

INTERVAL 1.554 - 1.812
CONTAINS 0.002%

sy
Tokensy
Feature activation+0.000
were
Token were
Feature activation+1.296
the
Token the
Feature activation+0.799
panel
Token panel
Feature activation+0.000
ists
Tokenists
Feature activation+0.000
.
Token.
Feature activation+1.720
Ċ
TokenĊ
Feature activation+0.295
Ċ
TokenĊ
Feature activation+0.779
Interestingly
TokenInterestingly
Feature activation+0.589
enough
Token enough
Feature activation+0.684
,
Token,
Feature activation+1.387
few
Token few
Feature activation+2.085
.
Token.
Feature activation+1.462
He
Token He
Feature activation+0.409
worked
Token worked
Feature activation+0.688
with
Token with
Feature activation+0.867
each
Token each
Feature activation+1.746
and
Token and
Feature activation+0.000
every
Token every
Feature activation+0.914
well
Token well
Feature activation+0.117
known
Token known
Feature activation+1.019
music
Token music
Feature activation+0.000
,
Token,
Feature activation+0.101
Dragon
Token Dragon
Feature activation+0.000
's
Token's
Feature activation+0.000
Lair
Token Lair
Feature activation+0.000
and
Token and
Feature activation+0.729
many
Token many
Feature activation+1.812
more
Token more
Feature activation+2.215
.
Token.
Feature activation+0.326
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
A
TokenA
Feature activation+0.000
visitor
Token visitor
Feature activation+0.000
drug
Token drug
Feature activation+0.000
used
Token used
Feature activation+0.000
to
Token to
Feature activation+0.000
induce
Token induce
Feature activation+0.000
abortions
Token abortions
Feature activation+0.000
.
Token.
Feature activation+1.640
Several
Token Several
Feature activation+1.226
funds
Token funds
Feature activation+0.198
in
Token in
Feature activation+0.857
the
Token the
Feature activation+0.426
Hobby
Token Hobby
Feature activation+0.000
the
Token the
Feature activation+0.000
Smithsonian
Token Smithsonian
Feature activation+0.000
(
Token (
Feature activation+0.451
2013
Token2013
Feature activation+0.471
),
Token),
Feature activation+1.827
but
Token but
Feature activation+1.625
the
Token the
Feature activation+0.563
discovery
Token discovery
Feature activation+0.000
of
Token of
Feature activation+0.094
his
Token his
Feature activation+0.000
computer
Token computer
Feature activation+0.000

INTERVAL 1.295 - 1.554
CONTAINS 0.003%

fentanyl
Token fentanyl
Feature activation+0.000
-
Token-
Feature activation+0.000
hero
Tokenhero
Feature activation+0.000
in
Tokenin
Feature activation+0.000
mixture
Token mixture
Feature activation+0.000
.
Token.
Feature activation+1.333
Ċ
TokenĊ
Feature activation+0.038
Ċ
TokenĊ
Feature activation+0.996
De
TokenDe
Feature activation+0.000
utsch
Tokenutsch
Feature activation+0.000
,
Token,
Feature activation+0.167
Retirement
Token Retirement
Feature activation+0.000
System
Token System
Feature activation+0.100
,
Token,
Feature activation+0.195
among
Token among
Feature activation+1.029
others
Token others
Feature activation+1.320
.
Token.
Feature activation+1.328
Ċ
TokenĊ
Feature activation+0.211
Ċ
TokenĊ
Feature activation+1.140
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Climate
TokenClimate
Feature activation+0.000
Human
Token Human
Feature activation+0.000
Rights
Token Rights
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
Labor
Token Labor
Feature activation+0.000
.
Token.
Feature activation+1.506
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.385
Continue
TokenContinue
Feature activation+0.000
reading
Token reading
Feature activation+0.000
.
Token.
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Retirement
Token Retirement
Feature activation+0.000
System
Token System
Feature activation+0.100
,
Token,
Feature activation+0.195
among
Token among
Feature activation+1.029
others
Token others
Feature activation+1.320
.
Token.
Feature activation+1.328
Ċ
TokenĊ
Feature activation+0.211
Ċ
TokenĊ
Feature activation+1.140
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
ot
Tokenot
Feature activation+0.000
kin
Tokenkin
Feature activation+0.000
to
Token to
Feature activation+0.153
Fa
Token Fa
Feature activation+0.000
ure
Tokenure
Feature activation+0.000
,
Token,
Feature activation+1.498
from
Token from
Feature activation+0.054
Maxim
Token Maxim
Feature activation+0.000
off
Tokenoff
Feature activation+0.060
to
Token to
Feature activation+0.000
Ar
Token Ar
Feature activation+0.000

INTERVAL 1.036 - 1.295
CONTAINS 0.007%

later
Token later
Feature activation+0.785
Taliban
Token Taliban
Feature activation+0.148
(
Token (
Feature activation+0.099
light
Tokenlight
Feature activation+0.000
green
Token green
Feature activation+0.000
).
Token).
Feature activation+1.279
Ċ
TokenĊ
Feature activation+0.637
Ċ
TokenĊ
Feature activation+0.507
After
TokenAfter
Feature activation+0.397
the
Token the
Feature activation+0.194
fall
Token fall
Feature activation+0.000
I
Token I
Feature activation+0.089
've
Token've
Feature activation+0.360
explained
Token explained
Feature activation+0.385
elsewhere
Token elsewhere
Feature activation+0.126
,
Token,
Feature activation+0.717
these
Token these
Feature activation+1.087
all
Token all
Feature activation+0.803
came
Token came
Feature activation+0.242
about
Token about
Feature activation+0.051
in
Token in
Feature activation+0.000
an
Token an
Feature activation+0.000
imon
Tokenimon
Feature activation+0.000
bar
Token bar
Feature activation+0.000
G
Token G
Feature activation+0.000
ior
Tokenior
Feature activation+0.000
a
Tokena
Feature activation+0.000
;
Token;
Feature activation+1.199
died
Token died
Feature activation+0.000
70
Token 70
Feature activation+0.000
CE
Token CE
Feature activation+0.000
)
Token)
Feature activation+0.000
was
Token was
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.245
Just
Token Just
Feature activation+0.000
Shoot
Token Shoot
Feature activation+0.000
Me
Token Me
Feature activation+0.000
.
Token.
Feature activation+1.052
She
Token She
Feature activation+0.590
also
Token also
Feature activation+0.325
had
Token had
Feature activation+0.087
small
Token small
Feature activation+0.173
roles
Token roles
Feature activation+0.000
po
Tokenpo
Feature activation+0.000
County
Token County
Feature activation+0.000
Sheriff
Token Sheriff
Feature activation+0.000
's
Token's
Feature activation+0.000
Office
Token Office
Feature activation+0.208
.
Token.
Feature activation+1.086
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
R
TokenR
Feature activation+0.000
af
Tokenaf
Feature activation+0.000
ah
Tokenah
Feature activation+0.000
Hospital
Token Hospital
Feature activation+0.000

INTERVAL 0.777 - 1.036
CONTAINS 0.016%

UPS
Token UPS
Feature activation+0.000
and
Token and
Feature activation+0.000
Lockheed
Token Lockheed
Feature activation+0.000
Martin
Token Martin
Feature activation+0.117
,
Token,
Feature activation+0.920
among
Token among
Feature activation+0.894
others
Token others
Feature activation+1.268
.
Token.
Feature activation+0.890
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.576
âĢ
TokenâĢ
Feature activation+0.000
H
Token H
Feature activation+0.000
ao
Tokenao
Feature activation+0.000
and
Token and
Feature activation+0.000
MM
Token MM
Feature activation+0.000
Y
TokenY
Feature activation+0.000
,
Token,
Feature activation+0.967
European
Token European
Feature activation+0.000
players
Token players
Feature activation+0.000
like
Token like
Feature activation+0.000
F
Token F
Feature activation+0.000
ATA
TokenATA
Feature activation+0.000
ine
Tokenine
Feature activation+0.000
âĢ¦
TokenâĢ¦
Feature activation+0.673
you
Token you
Feature activation+0.735
name
Token name
Feature activation+0.043
it
Token it
Feature activation+0.249
,
Token,
Feature activation+0.989
I
Token I
Feature activation+0.508
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+0.780
probably
Token probably
Feature activation+0.672
with
Token with
Feature activation+0.026
$
Token $
Feature activation+0.331
6
Token6
Feature activation+0.000
33
Token33
Feature activation+0.000
million
Token million
Feature activation+0.072
.
Token.
Feature activation+0.898
Ċ
TokenĊ
Feature activation+0.212
Ċ
TokenĊ
Feature activation+0.257
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Des
TokenDes
Feature activation+0.000
a
Token a
Feature activation+0.474
few
Token few
Feature activation+2.085
.
Token.
Feature activation+1.462
He
Token He
Feature activation+0.409
worked
Token worked
Feature activation+0.688
with
Token with
Feature activation+0.867
each
Token each
Feature activation+1.746
and
Token and
Feature activation+0.000
every
Token every
Feature activation+0.914
well
Token well
Feature activation+0.117
known
Token known
Feature activation+1.019

INTERVAL 0.518 - 0.777
CONTAINS 0.040%

ard
Tokenard
Feature activation+0.000
rop
Tokenrop
Feature activation+0.000
Attacks
Token Attacks
Feature activation+0.000
,
Token,
Feature activation+0.858
among
Token among
Feature activation+0.520
many
Token many
Feature activation+0.765
others
Token others
Feature activation+0.725
.
Token.
Feature activation+0.341
These
Token These
Feature activation+0.583
attacks
Token attacks
Feature activation+0.000
look
Token look
Feature activation+0.000
10
Token10
Feature activation+0.000
.
Token.
Feature activation+0.000
10
Token10
Feature activation+0.086
,
Token,
Feature activation+0.537
and
Token and
Feature activation+0.423
some
Token some
Feature activation+0.594
paid
Token paid
Feature activation+0.000
$
Token $
Feature activation+0.000
10
Token10
Feature activation+0.000
.
Token.
Feature activation+0.006
10
Token10
Feature activation+0.010
by
Token by
Feature activation+0.184
Sch
Token Sch
Feature activation+0.000
iller
Tokeniller
Feature activation+0.260
and
Token and
Feature activation+0.000
Go
Token Go
Feature activation+0.000
ethe
Tokenethe
Feature activation+0.654
;
Token;
Feature activation+0.165
heroic
Token heroic
Feature activation+0.000
tales
Token tales
Feature activation+0.000
by
Token by
Feature activation+0.000
C
Token C
Feature activation+0.000
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.829
name
Token name
Feature activation+0.007
it
Token it
Feature activation+1.521
âĢĶ
Token âĢĶ
Feature activation+1.108
from
Token from
Feature activation+0.644
soccer
Token soccer
Feature activation+0.000
moms
Token moms
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
the
Token the
Feature activation+0.006
%
Token%
Feature activation+0.000
in
Token in
Feature activation+0.000
Ukraine
Token Ukraine
Feature activation+0.000
.
Token.
Feature activation+0.312
Even
Token Even
Feature activation+0.000
with
Token with
Feature activation+0.587
its
Token its
Feature activation+0.046
proclaimed
Token proclaimed
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ĺ
Tokenĺ
Feature activation+0.000
independence
Tokenindependence
Feature activation+0.000

INTERVAL 0.259 - 0.518
CONTAINS 0.114%

rye
Token rye
Feature activation+0.000
,
Token,
Feature activation+0.005
as
Token as
Feature activation+0.017
well
Token well
Feature activation+0.000
as
Token as
Feature activation+0.000
other
Token other
Feature activation+0.266
foods
Token foods
Feature activation+0.000
including
Token including
Feature activation+0.000
apples
Token apples
Feature activation+0.000
,
Token,
Feature activation+0.000
cabbage
Token cabbage
Feature activation+0.000
and
Token and
Feature activation+0.000
Hiro
Token Hiro
Feature activation+0.000
ki
Tokenki
Feature activation+0.000
Kuro
Token Kuro
Feature activation+0.000
da
Tokenda
Feature activation+0.000
.
Token.
Feature activation+0.263
Don
Token Don
Feature activation+0.000
't
Token't
Feature activation+0.000
get
Token get
Feature activation+0.000
me
Token me
Feature activation+0.000
wrong
Token wrong
Feature activation+0.000
D
Token D
Feature activation+0.000
umber
Tokenumber
Feature activation+0.000
:
Token:
Feature activation+0.000
Sound
Token Sound
Feature activation+0.000
track
Tokentrack
Feature activation+0.000
".
Token".
Feature activation+0.312
At
Token At
Feature activation+0.000
the
Token the
Feature activation+0.000
forefront
Token forefront
Feature activation+0.000
is
Token is
Feature activation+0.000
Nick
Token Nick
Feature activation+0.000
are
Tokenare
Feature activation+0.000
and
Token and
Feature activation+0.000
David
Token David
Feature activation+0.000
Hyde
Token Hyde
Feature activation+0.000
Pierce
Token Pierce
Feature activation+0.000
,
Token,
Feature activation+0.440
among
Token among
Feature activation+0.560
others
Token others
Feature activation+0.500
.
Token.
Feature activation+0.168
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
it
Token it
Feature activation+0.280
!
Token!
Feature activation+0.550
I
Token I
Feature activation+0.098
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+0.394
pretty
Token pretty
Feature activation+0.078
much
Token much
Feature activation+0.141
given
Token given
Feature activation+0.295
up
Token up
Feature activation+0.065
on
Token on
Feature activation+0.509

INTERVAL 0.000 - 0.259
CONTAINS 99.817%

central
Token central
Feature activation+0.000
ised
Tokenised
Feature activation+0.000
utility
Token utility
Feature activation+0.000
stream
Token stream
Feature activation+0.000
lines
Tokenlines
Feature activation+0.000
the
Token the
Feature activation+0.000
process
Token process
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
August
Token August
Feature activation+0.000
31
Token 31
Feature activation+0.000
or
Token or
Feature activation+0.000
any
Token any
Feature activation+0.000
time
Token time
Feature activation+0.000
since
Token since
Feature activation+0.000
,
Token,
Feature activation+0.000
is
Token is
Feature activation+0.000
asked
Token asked
Feature activation+0.000
to
Token to
Feature activation+0.000
contact
Token contact
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
harmful
Token harmful
Feature activation+0.000
effects
Token effects
Feature activation+0.000
of
Token of
Feature activation+0.000
societal
Token societal
Feature activation+0.000
beauty
Token beauty
Feature activation+0.000
standards
Token standards
Feature activation+0.000
on
Token on
Feature activation+0.000
women
Token women
Feature activation+0.000
have
Token have
Feature activation+0.000
ships
Token ships
Feature activation+0.000
.
Token.
Feature activation+0.000
Raid
Token Raid
Feature activation+0.000
s
Tokens
Feature activation+0.000
on
Token on
Feature activation+0.000
fort
Token fort
Feature activation+0.000
resses
Tokenresses
Feature activation+0.000
.
Token.
Feature activation+0.000
Add
Token Add
Feature activation+0.000
in
Token in
Feature activation+0.000
aspects
Token aspects
Feature activation+0.000
dramatic
Token dramatic
Feature activation+0.000
a
Token a
Feature activation+0.000
change
Token change
Feature activation+0.000
for
Token for
Feature activation+0.000
them
Token them
Feature activation+0.000
,"
Token,"
Feature activation+0.000
said
Token said
Feature activation+0.000
Pal
Token Pal
Feature activation+0.000
iw
Tokeniw
Feature activation+0.000
al
Tokenal
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