Token | two |
Token ID | 734 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
bur | +0.005 |
fourteen | +0.005 |
blew | +0.005 |
eling | +0.005 |
ps | +0.005 |
Neg logit contributions | |
and | -0.006 |
urry | -0.005 |
Follow | -0.005 |
apping | -0.005 |
Nice | -0.005 |
Token | squirrel |
Token ID | 33039 |
Feature activation | -5.90e-04 |
Pos logit contributions | |
eling | +0.057 |
blew | +0.055 |
bur | +0.055 |
fourteen | +0.054 |
ck | +0.053 |
Neg logit contributions | |
and | -0.054 |
Follow | -0.046 |
urry | -0.045 |
Nice | -0.041 |
Today | -0.040 |
Token | s |
Token ID | 82 |
Feature activation | -9.64e-03 |
Pos logit contributions | |
and | +0.001 |
urry | +0.001 |
Follow | +0.001 |
of | +0.001 |
tum | +0.001 |
Neg logit contributions | |
bur | -0.001 |
fourteen | -0.001 |
ps | -0.001 |
eling | -0.001 |
IT | -0.001 |
Token | kept |
Token ID | 4030 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
and | +0.017 |
Follow | +0.015 |
urry | +0.015 |
Nice | +0.014 |
Today | +0.013 |
Neg logit contributions | |
bur | -0.013 |
fourteen | -0.012 |
ps | -0.012 |
blew | -0.012 |
eling | -0.012 |
Token | qu |
Token ID | 627 |
Feature activation | +8.26e-03 |
Pos logit contributions | |
fourteen | +0.020 |
ami | +0.019 |
IT | +0.018 |
ible | +0.018 |
bur | +0.018 |
Neg logit contributions | |
and | -0.014 |
urry | -0.012 |
Go | -0.012 |
Stay | -0.012 |
of | -0.012 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.94e-02 |
Pos logit contributions | |
bur | +0.011 |
eling | +0.011 |
fourteen | +0.011 |
ps | +0.010 |
blew | +0.010 |
Neg logit contributions | |
and | -0.015 |
Follow | -0.013 |
urry | -0.013 |
Nice | -0.012 |
tum | -0.012 |
Token | ling |
Token ID | 1359 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
bur | +0.113 |
fourteen | +0.110 |
ps | +0.100 |
eling | +0.099 |
blew | +0.099 |
Neg logit contributions | |
and | -0.201 |
urry | -0.172 |
Follow | -0.171 |
apping | -0.158 |
Nice | -0.154 |
Token | until |
Token ID | 1566 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
bur | +0.017 |
fourteen | +0.017 |
eling | +0.016 |
ps | +0.016 |
blew | +0.016 |
Neg logit contributions | |
and | -0.023 |
urry | -0.020 |
Follow | -0.020 |
Nice | -0.018 |
tum | -0.018 |
Token | finally |
Token ID | 3443 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
ps | +0.025 |
cking | +0.023 |
eling | +0.023 |
bur | +0.023 |
ible | +0.023 |
Neg logit contributions | |
and | -0.024 |
urry | -0.023 |
tum | -0.023 |
of | -0.022 |
birthday | -0.021 |
Token | Sammy |
Token ID | 44067 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
bur | +0.034 |
eling | +0.034 |
ps | +0.034 |
cking | +0.033 |
fourteen | +0.032 |
Neg logit contributions | |
and | -0.032 |
urry | -0.030 |
Follow | -0.027 |
tum | -0.027 |
of | -0.027 |
Token | said |
Token ID | 531 |
Feature activation | -2.24e-03 |
Pos logit contributions | |
ps | +0.059 |
cking | +0.059 |
bur | +0.058 |
ries | +0.058 |
fourteen | +0.058 |
Neg logit contributions | |
and | -0.062 |
urry | -0.054 |
Follow | -0.051 |
of | -0.049 |
tum | -0.048 |
Token | mum |
Token ID | 25682 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
fourteen | +0.015 |
ps | +0.015 |
bur | +0.014 |
ck | +0.014 |
blew | +0.014 |
Neg logit contributions | |
and | -0.019 |
Follow | -0.016 |
urry | -0.016 |
Nice | -0.015 |
apping | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +5.64e-04 |
Pos logit contributions | |
bur | +0.016 |
fourteen | +0.016 |
eling | +0.015 |
ps | +0.015 |
blew | +0.015 |
Neg logit contributions | |
and | -0.017 |
urry | -0.014 |
Follow | -0.014 |
Nice | -0.013 |
tum | -0.013 |
Token | understood |
Token ID | 7247 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
eling | +0.001 |
ps | +0.001 |
ible | +0.001 |
Neg logit contributions | |
and | -0.001 |
urry | -0.001 |
Follow | -0.001 |
tum | -0.001 |
Stay | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | -7.70e-03 |
Pos logit contributions | |
bur | +0.014 |
fourteen | +0.013 |
ps | +0.013 |
IT | +0.013 |
rob | +0.013 |
Neg logit contributions | |
and | -0.018 |
of | -0.015 |
urry | -0.015 |
story | -0.015 |
tum | -0.014 |
Token | qu |
Token ID | 627 |
Feature activation | +5.06e-03 |
Pos logit contributions | |
and | +0.016 |
Follow | +0.014 |
urry | +0.014 |
Nice | +0.013 |
apping | +0.013 |
Neg logit contributions | |
bur | -0.008 |
blew | -0.007 |
fourteen | -0.007 |
eling | -0.007 |
ps | -0.007 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
bur | +0.006 |
eling | +0.006 |
fourteen | +0.006 |
ps | +0.006 |
blew | +0.006 |
Neg logit contributions | |
and | -0.010 |
urry | -0.008 |
Follow | -0.008 |
Nice | -0.008 |
tum | -0.008 |
Token | lling |
Token ID | 2680 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
bur | +0.102 |
fourteen | +0.099 |
abi | +0.087 |
ps | +0.087 |
IT | +0.086 |
Neg logit contributions | |
and | -0.208 |
Follow | -0.178 |
urry | -0.178 |
of | -0.165 |
apping | -0.165 |
Token | was |
Token ID | 373 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
bur | +0.032 |
fourteen | +0.031 |
eling | +0.030 |
ps | +0.029 |
ck | +0.029 |
Neg logit contributions | |
and | -0.031 |
urry | -0.026 |
Follow | -0.026 |
tum | -0.023 |
Nice | -0.023 |
Token | not |
Token ID | 407 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
rob | +0.029 |
bur | +0.028 |
ami | +0.026 |
fourteen | +0.025 |
ries | +0.025 |
Neg logit contributions | |
and | -0.023 |
of | -0.022 |
welcome | -0.022 |
urry | -0.021 |
Stay | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -1.76e-03 |
Pos logit contributions | |
rob | +0.050 |
bur | +0.050 |
fourteen | +0.047 |
ami | +0.047 |
eling | +0.046 |
Neg logit contributions | |
and | -0.047 |
of | -0.042 |
urry | -0.040 |
welcome | -0.039 |
birthday | -0.038 |
Token | good |
Token ID | 922 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
and | +0.003 |
urry | +0.002 |
Follow | +0.002 |
tum | +0.002 |
Nice | +0.002 |
Neg logit contributions | |
bur | -0.003 |
eling | -0.003 |
fourteen | -0.003 |
ps | -0.003 |
ck | -0.003 |
Token | them |
Token ID | 606 |
Feature activation | -7.26e-03 |
Pos logit contributions | |
and | +0.061 |
urry | +0.052 |
Follow | +0.050 |
birthday | +0.050 |
tum | +0.050 |
Neg logit contributions | |
bur | -0.035 |
fourteen | -0.034 |
blew | -0.032 |
ps | -0.031 |
IT | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -3.36e-03 |
Pos logit contributions | |
and | +0.013 |
Follow | +0.011 |
urry | +0.011 |
Nice | +0.011 |
tum | +0.010 |
Neg logit contributions | |
bur | -0.009 |
fourteen | -0.009 |
ps | -0.009 |
eling | -0.009 |
blew | -0.008 |
Token | never |
Token ID | 1239 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
and | +0.006 |
Follow | +0.005 |
urry | +0.005 |
tum | +0.005 |
Stay | +0.005 |
Neg logit contributions | |
bur | -0.004 |
fourteen | -0.004 |
eling | -0.004 |
ps | -0.004 |
ible | -0.004 |
Token | threw |
Token ID | 9617 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
and | +0.022 |
urry | +0.019 |
Follow | +0.019 |
tum | +0.017 |
Nice | +0.017 |
Neg logit contributions | |
bur | -0.017 |
fourteen | -0.017 |
eling | -0.016 |
ps | -0.016 |
ible | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
bur | +0.031 |
eling | +0.031 |
fourteen | +0.030 |
blew | +0.030 |
IT | +0.028 |
Neg logit contributions | |
and | -0.061 |
urry | -0.051 |
Follow | -0.051 |
tum | -0.051 |
of | -0.049 |
Token | tant |
Token ID | 24246 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
and | +0.020 |
Follow | +0.019 |
urry | +0.018 |
School | +0.017 |
Today | +0.017 |
Neg logit contributions | |
bur | -0.012 |
blew | -0.012 |
fourteen | -0.011 |
ck | -0.011 |
comb | -0.011 |
Token | rum |
Token ID | 6582 |
Feature activation | -3.24e-03 |
Pos logit contributions | |
bur | +0.150 |
fourteen | +0.148 |
blew | +0.148 |
eling | +0.140 |
IT | +0.139 |
Neg logit contributions | |
and | -0.157 |
urry | -0.126 |
Follow | -0.116 |
aring | -0.114 |
apping | -0.114 |
Token | again |
Token ID | 757 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
and | +0.006 |
urry | +0.005 |
Follow | +0.005 |
Nice | +0.005 |
tum | +0.005 |
Neg logit contributions | |
bur | -0.004 |
fourteen | -0.004 |
eling | -0.004 |
ps | -0.004 |
blew | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +5.87e-03 |
Pos logit contributions | |
and | +0.036 |
urry | +0.031 |
Follow | +0.029 |
tum | +0.028 |
of | +0.028 |
Neg logit contributions | |
bur | -0.032 |
fourteen | -0.031 |
ps | -0.031 |
eling | -0.030 |
lled | -0.030 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.33e-03 |
Pos logit contributions | |
bur | +0.010 |
fourteen | +0.010 |
eling | +0.010 |
ps | +0.010 |
blew | +0.010 |
Neg logit contributions | |
and | -0.008 |
urry | -0.007 |
Follow | -0.007 |
Nice | -0.006 |
tum | -0.006 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
and | +0.002 |
urry | +0.002 |
Follow | +0.002 |
aring | +0.002 |
tum | +0.002 |
Neg logit contributions | |
bur | -0.002 |
fourteen | -0.002 |
ps | -0.002 |
blew | -0.002 |
IT | -0.002 |
Token | love |
Token ID | 1842 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
and | +0.018 |
urry | +0.015 |
Follow | +0.015 |
Nice | +0.014 |
apping | +0.013 |
Neg logit contributions | |
bur | -0.015 |
fourteen | -0.015 |
eling | -0.015 |
ps | -0.015 |
blew | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
and | +0.018 |
urry | +0.015 |
Follow | +0.015 |
tum | +0.013 |
Nice | +0.013 |
Neg logit contributions | |
bur | -0.017 |
fourteen | -0.016 |
eling | -0.016 |
ps | -0.016 |
blew | -0.016 |
Token | have |
Token ID | 423 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
bur | +0.029 |
fourteen | +0.028 |
eling | +0.028 |
ps | +0.028 |
blew | +0.027 |
Neg logit contributions | |
and | -0.032 |
urry | -0.027 |
Follow | -0.027 |
Nice | -0.024 |
tum | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -8.33e-03 |
Pos logit contributions | |
bur | +0.021 |
fourteen | +0.020 |
eling | +0.020 |
ps | +0.020 |
blew | +0.019 |
Neg logit contributions | |
and | -0.028 |
Follow | -0.024 |
urry | -0.024 |
tum | -0.023 |
story | -0.022 |
Token | new |
Token ID | 649 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
and | +0.015 |
Follow | +0.013 |
urry | +0.013 |
apping | +0.012 |
Go | +0.011 |
Neg logit contributions | |
bur | -0.011 |
blew | -0.010 |
fourteen | -0.010 |
abi | -0.010 |
ible | -0.010 |
Token | hairst |
Token ID | 42658 |
Feature activation | +9.66e-02 |
Pos logit contributions | |
bur | +0.015 |
fourteen | +0.014 |
blew | +0.014 |
ible | +0.014 |
ps | +0.014 |
Neg logit contributions | |
and | -0.019 |
urry | -0.016 |
Follow | -0.016 |
rav | -0.015 |
aring | -0.014 |
Token | yle |
Token ID | 2349 |
Feature activation | -6.20e-03 |
Pos logit contributions | |
fourteen | +0.136 |
IT | +0.133 |
bur | +0.132 |
blew | +0.130 |
ps | +0.129 |
Neg logit contributions | |
and | -0.166 |
apping | -0.138 |
urry | -0.134 |
Follow | -0.131 |
of | -0.130 |
Token | ". |
Token ID | 1911 |
Feature activation | -7.29e-04 |
Pos logit contributions | |
and | +0.010 |
urry | +0.008 |
Follow | +0.008 |
Nice | +0.007 |
tum | +0.007 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.009 |
eling | -0.009 |
ps | -0.009 |
blew | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.24e-03 |
Pos logit contributions | |
and | +0.001 |
Follow | +0.001 |
urry | +0.001 |
Today | +0.001 |
tum | +0.001 |
Neg logit contributions | |
eling | -0.001 |
ps | -0.001 |
bur | -0.001 |
cking | -0.001 |
ck | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.60e-04 |
Pos logit contributions | |
and | +0.002 |
Follow | +0.002 |
urry | +0.002 |
tum | +0.002 |
Nice | +0.002 |
Neg logit contributions | |
bur | -0.002 |
fourteen | -0.002 |
ck | -0.002 |
eling | -0.002 |
blew | -0.002 |
Token | The |
Token ID | 464 |
Feature activation | -5.18e-03 |
Pos logit contributions | |
bur | +0.000 |
ps | +0.000 |
fourteen | +0.000 |
eling | +0.000 |
ible | +0.000 |
Neg logit contributions | |
and | -0.000 |
Follow | -0.000 |
urry | -0.000 |
apping | -0.000 |
Nice | -0.000 |
Token | In |
Token ID | 554 |
Feature activation | +3.93e-03 |
Pos logit contributions | |
bur | +0.022 |
eling | +0.022 |
ps | +0.022 |
fourteen | +0.022 |
ck | +0.021 |
Neg logit contributions | |
and | -0.020 |
urry | -0.017 |
Follow | -0.017 |
tum | -0.016 |
Nice | -0.015 |
Token | fact |
Token ID | 1109 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
eling | +0.005 |
IT | +0.005 |
ck | +0.005 |
bur | +0.005 |
cking | +0.005 |
Neg logit contributions | |
of | -0.006 |
tum | -0.006 |
and | -0.006 |
story | -0.006 |
urry | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -3.33e-03 |
Pos logit contributions | |
bur | +0.047 |
fourteen | +0.044 |
ps | +0.044 |
IT | +0.044 |
eling | +0.044 |
Neg logit contributions | |
and | -0.016 |
urry | -0.014 |
Follow | -0.012 |
of | -0.011 |
tum | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +6.48e-03 |
Pos logit contributions | |
and | +0.005 |
of | +0.005 |
urry | +0.005 |
tum | +0.005 |
Follow | +0.005 |
Neg logit contributions | |
ps | -0.005 |
fourteen | -0.005 |
bur | -0.005 |
eling | -0.005 |
IT | -0.004 |
Token | special |
Token ID | 2041 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
bur | +0.008 |
eling | +0.008 |
blew | +0.008 |
ps | +0.008 |
fourteen | +0.008 |
Neg logit contributions | |
and | -0.012 |
urry | -0.011 |
Follow | -0.011 |
Nice | -0.010 |
apping | -0.010 |
Token | bal |
Token ID | 3652 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
blew | +0.025 |
bur | +0.025 |
eling | +0.024 |
ps | +0.023 |
ible | +0.022 |
Neg logit contributions | |
and | -0.038 |
Follow | -0.033 |
urry | -0.033 |
izza | -0.030 |
Nice | -0.030 |
Token | m |
Token ID | 76 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
bur | +0.168 |
fourteen | +0.161 |
eling | +0.159 |
blew | +0.157 |
ps | +0.156 |
Neg logit contributions | |
and | -0.125 |
Follow | -0.106 |
urry | -0.104 |
tum | -0.098 |
Nice | -0.092 |
Token | made |
Token ID | 925 |
Feature activation | -3.28e-03 |
Pos logit contributions | |
and | +0.025 |
urry | +0.021 |
tum | +0.020 |
Follow | +0.019 |
apping | +0.019 |
Neg logit contributions | |
fourteen | -0.025 |
bur | -0.024 |
eling | -0.023 |
IT | -0.023 |
ck | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | +6.43e-03 |
Pos logit contributions | |
and | +0.006 |
of | +0.006 |
urry | +0.005 |
story | +0.005 |
ay | +0.005 |
Neg logit contributions | |
fourteen | -0.004 |
bur | -0.004 |
rob | -0.004 |
eling | -0.004 |
ps | -0.003 |
Token | ankle |
Token ID | 19847 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
bur | +0.010 |
fourteen | +0.010 |
eling | +0.010 |
ps | +0.010 |
blew | +0.010 |
Neg logit contributions | |
and | -0.010 |
urry | -0.008 |
Follow | -0.008 |
Nice | -0.007 |
tum | -0.007 |
Token | look |
Token ID | 804 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
and | +0.022 |
urry | +0.019 |
Follow | +0.019 |
Nice | +0.017 |
tum | +0.017 |
Neg logit contributions | |
bur | -0.020 |
fourteen | -0.020 |
eling | -0.019 |
ps | -0.019 |
blew | -0.019 |
Token | wanted |
Token ID | 2227 |
Feature activation | -5.77e-03 |
Pos logit contributions | |
fourteen | +0.014 |
bur | +0.014 |
ps | +0.013 |
eling | +0.013 |
cking | +0.013 |
Neg logit contributions | |
and | -0.015 |
urry | -0.013 |
Follow | -0.013 |
tum | -0.012 |
School | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
and | +0.008 |
urry | +0.007 |
Follow | +0.007 |
Nice | +0.006 |
tum | +0.006 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.010 |
eling | -0.010 |
ps | -0.010 |
blew | -0.009 |
Token | have |
Token ID | 423 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
bur | +0.034 |
fourteen | +0.034 |
eling | +0.033 |
ps | +0.033 |
ible | +0.032 |
Neg logit contributions | |
and | -0.029 |
urry | -0.025 |
Follow | -0.024 |
tum | -0.022 |
Stay | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -7.02e-03 |
Pos logit contributions | |
bur | +0.020 |
fourteen | +0.019 |
eling | +0.018 |
ps | +0.018 |
ami | +0.018 |
Neg logit contributions | |
and | -0.027 |
tum | -0.023 |
story | -0.022 |
birthday | -0.022 |
urry | -0.022 |
Token | cool |
Token ID | 3608 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
and | +0.012 |
Follow | +0.010 |
urry | +0.010 |
Nice | +0.009 |
apping | +0.009 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.010 |
eling | -0.010 |
blew | -0.010 |
ps | -0.009 |
Token | hairst |
Token ID | 42658 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
bur | +0.015 |
blew | +0.015 |
ible | +0.014 |
fourteen | +0.014 |
IT | +0.014 |
Neg logit contributions | |
and | -0.020 |
Follow | -0.017 |
urry | -0.017 |
vere | -0.016 |
rav | -0.015 |
Token | yle |
Token ID | 2349 |
Feature activation | -8.70e-03 |
Pos logit contributions | |
fourteen | +0.137 |
ps | +0.132 |
IT | +0.132 |
blew | +0.130 |
bur | +0.129 |
Neg logit contributions | |
and | -0.154 |
apping | -0.128 |
of | -0.125 |
urry | -0.121 |
Follow | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
and | +0.015 |
urry | +0.013 |
Follow | +0.012 |
Nice | +0.011 |
tum | +0.011 |
Neg logit contributions | |
bur | -0.013 |
fourteen | -0.013 |
eling | -0.012 |
ps | -0.012 |
blew | -0.012 |
Token | but |
Token ID | 475 |
Feature activation | -4.04e-03 |
Pos logit contributions | |
and | +0.019 |
urry | +0.017 |
Follow | +0.017 |
rav | +0.015 |
apping | +0.015 |
Neg logit contributions | |
bur | -0.016 |
fourteen | -0.015 |
eling | -0.015 |
blew | -0.015 |
ps | -0.015 |
Token | he |
Token ID | 339 |
Feature activation | +8.36e-03 |
Pos logit contributions | |
and | +0.009 |
urry | +0.008 |
Follow | +0.008 |
apping | +0.007 |
Nice | +0.007 |
Neg logit contributions | |
bur | -0.004 |
fourteen | -0.003 |
eling | -0.003 |
blew | -0.003 |
ps | -0.003 |
Token | didn |
Token ID | 1422 |
Feature activation | -7.55e-03 |
Pos logit contributions | |
fourteen | +0.013 |
ps | +0.013 |
bur | +0.013 |
eling | +0.012 |
cking | +0.012 |
Neg logit contributions | |
and | -0.013 |
urry | -0.011 |
Follow | -0.011 |
tum | -0.010 |
Stay | -0.010 |
Token | together |
Token ID | 1978 |
Feature activation | +6.83e-03 |
Pos logit contributions | |
bur | +0.017 |
fourteen | +0.017 |
eling | +0.016 |
ps | +0.016 |
blew | +0.016 |
Neg logit contributions | |
and | -0.027 |
urry | -0.024 |
Follow | -0.023 |
tum | -0.022 |
Nice | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
bur | +0.010 |
fourteen | +0.009 |
eling | +0.009 |
ps | +0.009 |
blew | +0.009 |
Neg logit contributions | |
and | -0.012 |
urry | -0.010 |
Follow | -0.010 |
tum | -0.009 |
of | -0.009 |
Token | They |
Token ID | 1119 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
bur | +0.016 |
fourteen | +0.016 |
eling | +0.015 |
ps | +0.015 |
blew | +0.015 |
Neg logit contributions | |
and | -0.016 |
urry | -0.014 |
Follow | -0.013 |
tum | -0.012 |
Nice | -0.012 |
Token | never |
Token ID | 1239 |
Feature activation | -9.08e-03 |
Pos logit contributions | |
bur | +0.029 |
fourteen | +0.029 |
ps | +0.028 |
eling | +0.028 |
ries | +0.027 |
Neg logit contributions | |
and | -0.033 |
urry | -0.029 |
Follow | -0.029 |
tum | -0.026 |
of | -0.026 |
Token | qu |
Token ID | 627 |
Feature activation | +4.21e-03 |
Pos logit contributions | |
and | +0.018 |
urry | +0.016 |
Follow | +0.015 |
Nice | +0.014 |
tum | +0.014 |
Neg logit contributions | |
bur | -0.011 |
fourteen | -0.010 |
eling | -0.010 |
ps | -0.010 |
blew | -0.010 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
bur | +0.005 |
fourteen | +0.005 |
eling | +0.005 |
ps | +0.005 |
blew | +0.005 |
Neg logit contributions | |
and | -0.008 |
urry | -0.007 |
Follow | -0.007 |
tum | -0.006 |
Nice | -0.006 |
Token | led |
Token ID | 992 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
bur | +0.075 |
fourteen | +0.070 |
abi | +0.062 |
eling | +0.057 |
ps | +0.057 |
Neg logit contributions | |
and | -0.228 |
Follow | -0.197 |
urry | -0.196 |
apping | -0.186 |
of | -0.185 |
Token | again |
Token ID | 757 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
bur | +0.034 |
fourteen | +0.033 |
eling | +0.033 |
ps | +0.032 |
blew | +0.032 |
Neg logit contributions | |
and | -0.039 |
urry | -0.033 |
Follow | -0.032 |
Nice | -0.030 |
tum | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
and | +0.031 |
urry | +0.026 |
Follow | +0.024 |
of | +0.023 |
story | +0.023 |
Neg logit contributions | |
bur | -0.039 |
fourteen | -0.038 |
ps | -0.038 |
eling | -0.037 |
ck | -0.036 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.99e-03 |
Pos logit contributions | |
bur | +0.018 |
fourteen | +0.018 |
eling | +0.017 |
ps | +0.017 |
ible | +0.017 |
Neg logit contributions | |
and | -0.017 |
urry | -0.014 |
Follow | -0.014 |
tum | -0.013 |
Nice | -0.013 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
bur | +0.008 |
fourteen | +0.007 |
ps | +0.007 |
blew | +0.007 |
eling | +0.007 |
Neg logit contributions | |
and | -0.011 |
urry | -0.009 |
Follow | -0.009 |
tum | -0.009 |
Nice | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | -7.14e-03 |
Pos logit contributions | |
bur | +0.017 |
fourteen | +0.017 |
ps | +0.016 |
blew | +0.016 |
eling | +0.016 |
Neg logit contributions | |
and | -0.027 |
Follow | -0.023 |
urry | -0.023 |
Nice | -0.021 |
tum | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
and | +0.008 |
urry | +0.006 |
Follow | +0.006 |
tum | +0.005 |
Nice | +0.005 |
Neg logit contributions | |
bur | -0.014 |
fourteen | -0.014 |
eling | -0.014 |
ps | -0.014 |
ck | -0.014 |
Token | they |
Token ID | 484 |
Feature activation | +2.89e-03 |
Pos logit contributions | |
and | +0.017 |
urry | +0.014 |
of | +0.014 |
Follow | +0.014 |
Nice | +0.013 |
Neg logit contributions | |
bur | -0.016 |
ps | -0.015 |
eling | -0.015 |
cking | -0.014 |
ck | -0.014 |
Token | never |
Token ID | 1239 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
fourteen | +0.005 |
ries | +0.005 |
ps | +0.005 |
cking | +0.005 |
eling | +0.005 |
Neg logit contributions | |
and | -0.003 |
urry | -0.003 |
of | -0.003 |
story | -0.003 |
Follow | -0.003 |
Token | qu |
Token ID | 627 |
Feature activation | +1.15e-03 |
Pos logit contributions | |
and | +0.023 |
urry | +0.021 |
Follow | +0.020 |
tum | +0.019 |
Nice | +0.019 |
Neg logit contributions | |
bur | -0.015 |
fourteen | -0.014 |
ps | -0.014 |
eling | -0.014 |
ck | -0.013 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.44e-02 |
Pos logit contributions | |
bur | +0.002 |
fourteen | +0.002 |
eling | +0.001 |
ps | +0.001 |
blew | +0.001 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
Follow | -0.002 |
apping | -0.002 |
tum | -0.002 |
Token | led |
Token ID | 992 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
bur | +0.094 |
fourteen | +0.089 |
abi | +0.081 |
ps | +0.074 |
IT | +0.072 |
Neg logit contributions | |
and | -0.213 |
Follow | -0.176 |
urry | -0.175 |
of | -0.171 |
apping | -0.170 |
Token | again |
Token ID | 757 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
bur | +0.031 |
eling | +0.030 |
fourteen | +0.030 |
ps | +0.030 |
blew | +0.029 |
Neg logit contributions | |
and | -0.033 |
urry | -0.027 |
Follow | -0.027 |
of | -0.025 |
Nice | -0.025 |
Token | over |
Token ID | 625 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
and | +0.037 |
urry | +0.030 |
of | +0.028 |
Follow | +0.028 |
tum | +0.027 |
Neg logit contributions | |
bur | -0.039 |
fourteen | -0.038 |
ps | -0.037 |
eling | -0.036 |
IT | -0.035 |
Token | clothes |
Token ID | 8242 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
bur | +0.015 |
fourteen | +0.014 |
ck | +0.014 |
eling | +0.014 |
ps | +0.014 |
Neg logit contributions | |
and | -0.020 |
urry | -0.017 |
of | -0.016 |
tum | -0.016 |
birthday | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
and | +0.022 |
urry | +0.019 |
Follow | +0.017 |
tum | +0.017 |
apping | +0.017 |
Neg logit contributions | |
bur | -0.021 |
eling | -0.020 |
fourteen | -0.020 |
ps | -0.020 |
ck | -0.020 |
Token | ribbon |
Token ID | 29092 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
fourteen | +0.022 |
bur | +0.022 |
blew | +0.021 |
ps | +0.021 |
eling | +0.020 |
Neg logit contributions | |
and | -0.033 |
urry | -0.029 |
Follow | -0.028 |
aring | -0.025 |
ooters | -0.025 |
Token | for |
Token ID | 329 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
and | +0.018 |
urry | +0.015 |
Follow | +0.014 |
tum | +0.013 |
of | +0.013 |
Neg logit contributions | |
eling | -0.018 |
bur | -0.018 |
fourteen | -0.018 |
ps | -0.017 |
ible | -0.017 |
Token | being |
Token ID | 852 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
and | +0.051 |
urry | +0.044 |
Follow | +0.043 |
of | +0.042 |
story | +0.042 |
Neg logit contributions | |
bur | -0.024 |
fourteen | -0.022 |
ible | -0.021 |
eling | -0.021 |
blew | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +8.88e-04 |
Pos logit contributions | |
and | +0.060 |
Follow | +0.050 |
of | +0.048 |
urry | +0.048 |
welcome | +0.046 |
Neg logit contributions | |
bur | -0.049 |
eling | -0.048 |
fourteen | -0.048 |
IT | -0.045 |
ible | -0.044 |
Token | best |
Token ID | 1266 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
eling | +0.001 |
ps | +0.001 |
ck | +0.001 |
Neg logit contributions | |
and | -0.001 |
Follow | -0.001 |
urry | -0.001 |
tum | -0.001 |
Nice | -0.001 |
Token | bal |
Token ID | 3652 |
Feature activation | +9.43e-02 |
Pos logit contributions | |
bur | +0.034 |
fourteen | +0.033 |
eling | +0.033 |
ps | +0.033 |
ck | +0.032 |
Neg logit contributions | |
and | -0.036 |
urry | -0.031 |
Follow | -0.030 |
tum | -0.027 |
Nice | -0.027 |
Token | ancer |
Token ID | 8250 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
bur | +0.182 |
fourteen | +0.176 |
eling | +0.174 |
ps | +0.171 |
blew | +0.171 |
Neg logit contributions | |
and | -0.108 |
Follow | -0.090 |
urry | -0.088 |
tum | -0.083 |
Nice | -0.077 |
Token | in |
Token ID | 287 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
and | +0.091 |
urry | +0.079 |
Follow | +0.078 |
Nice | +0.071 |
tum | +0.070 |
Neg logit contributions | |
bur | -0.073 |
fourteen | -0.071 |
eling | -0.070 |
ps | -0.069 |
blew | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | +1.44e-03 |
Pos logit contributions | |
and | +0.062 |
of | +0.056 |
urry | +0.053 |
story | +0.052 |
welcome | +0.052 |
Neg logit contributions | |
bur | -0.040 |
eling | -0.038 |
fourteen | -0.037 |
cking | -0.037 |
ck | -0.037 |
Token | competition |
Token ID | 5449 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
blew | +0.003 |
eling | +0.002 |
bur | +0.002 |
ible | +0.002 |
fourteen | +0.002 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
Follow | -0.002 |
Nice | -0.001 |
aring | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.54e-03 |
Pos logit contributions | |
bur | +0.022 |
fourteen | +0.021 |
eling | +0.021 |
ps | +0.021 |
ck | +0.020 |
Neg logit contributions | |
and | -0.031 |
urry | -0.027 |
Follow | -0.027 |
Nice | -0.024 |
tum | -0.024 |
Token | heal |
Token ID | 12035 |
Feature activation | +8.48e-04 |
Pos logit contributions | |
and | +0.013 |
Follow | +0.011 |
urry | +0.011 |
Nice | +0.010 |
tum | +0.010 |
Neg logit contributions | |
bur | -0.017 |
eling | -0.016 |
fourteen | -0.016 |
ps | -0.016 |
ck | -0.016 |
Token | you |
Token ID | 345 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
fourteen | +0.001 |
bur | +0.001 |
lled | +0.001 |
blew | +0.001 |
eling | +0.001 |
Neg logit contributions | |
and | -0.001 |
Follow | -0.001 |
urry | -0.001 |
Go | -0.001 |
Stay | -0.001 |
Token | with |
Token ID | 351 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
and | +0.037 |
Follow | +0.033 |
urry | +0.032 |
Nice | +0.031 |
rav | +0.030 |
Neg logit contributions | |
ps | -0.025 |
bur | -0.025 |
eling | -0.024 |
fourteen | -0.023 |
ck | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
fourteen | +0.012 |
bur | +0.012 |
lled | +0.012 |
ami | +0.012 |
ps | +0.012 |
Neg logit contributions | |
and | -0.015 |
urry | -0.013 |
Follow | -0.013 |
tum | -0.013 |
Stay | -0.012 |
Token | special |
Token ID | 2041 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
and | +0.014 |
Follow | +0.012 |
urry | +0.012 |
Nice | +0.011 |
Go | +0.011 |
Neg logit contributions | |
bur | -0.006 |
blew | -0.006 |
eling | -0.006 |
fourteen | -0.006 |
ps | -0.006 |
Token | bal |
Token ID | 3652 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
blew | +0.020 |
bur | +0.020 |
ps | +0.019 |
eling | +0.019 |
fourteen | +0.018 |
Neg logit contributions | |
and | -0.036 |
Follow | -0.032 |
urry | -0.031 |
Nice | -0.029 |
engu | -0.029 |
Token | m |
Token ID | 76 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
bur | +0.160 |
eling | +0.145 |
ries | +0.142 |
fourteen | +0.142 |
blew | +0.141 |
Neg logit contributions | |
and | -0.134 |
Follow | -0.122 |
urry | -0.114 |
tum | -0.113 |
Nice | -0.108 |
Token | ". |
Token ID | 1911 |
Feature activation | +5.54e-03 |
Pos logit contributions | |
and | +0.024 |
urry | +0.021 |
Follow | +0.020 |
tum | +0.019 |
apping | +0.019 |
Neg logit contributions | |
fourteen | -0.024 |
bur | -0.024 |
eling | -0.022 |
IT | -0.022 |
ps | -0.022 |
Token | She |
Token ID | 1375 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
eling | +0.010 |
bur | +0.010 |
ps | +0.010 |
fourteen | +0.010 |
ck | +0.010 |
Neg logit contributions | |
and | -0.008 |
urry | -0.006 |
Follow | -0.006 |
Today | -0.006 |
Nice | -0.006 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.81e-03 |
Pos logit contributions | |
fourteen | +0.018 |
bur | +0.018 |
eling | +0.018 |
ps | +0.017 |
ible | +0.017 |
Neg logit contributions | |
and | -0.029 |
Follow | -0.026 |
urry | -0.026 |
tum | -0.024 |
of | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
and | +0.004 |
urry | +0.003 |
Follow | +0.003 |
Nice | +0.003 |
of | +0.003 |
Neg logit contributions | |
eling | -0.005 |
bur | -0.005 |
fourteen | -0.005 |
ible | -0.005 |
cking | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | -6.98e-03 |
Pos logit contributions | |
bur | +0.014 |
fourteen | +0.014 |
blew | +0.014 |
ps | +0.013 |
cking | +0.012 |
Neg logit contributions | |
and | -0.030 |
Follow | -0.027 |
engu | -0.025 |
urry | -0.025 |
Nice | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
and | +0.009 |
Follow | +0.007 |
urry | +0.007 |
Nice | +0.006 |
Stay | +0.006 |
Neg logit contributions | |
bur | -0.013 |
fourteen | -0.013 |
blew | -0.012 |
eling | -0.012 |
ps | -0.012 |
Token | they |
Token ID | 484 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
and | +0.011 |
urry | +0.010 |
of | +0.009 |
tum | +0.009 |
birthday | +0.009 |
Neg logit contributions | |
ps | -0.012 |
eling | -0.011 |
cking | -0.011 |
ck | -0.011 |
bur | -0.011 |
Token | never |
Token ID | 1239 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
ps | +0.003 |
fourteen | +0.003 |
ries | +0.003 |
eling | +0.003 |
cking | +0.003 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
of | -0.002 |
Follow | -0.002 |
rav | -0.002 |
Token | qu |
Token ID | 627 |
Feature activation | +3.84e-03 |
Pos logit contributions | |
and | +0.023 |
Follow | +0.020 |
urry | +0.020 |
Nice | +0.018 |
vere | +0.018 |
Neg logit contributions | |
bur | -0.012 |
blew | -0.011 |
fourteen | -0.011 |
eling | -0.011 |
ps | -0.011 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
bur | +0.005 |
fourteen | +0.005 |
eling | +0.005 |
ps | +0.005 |
blew | +0.004 |
Neg logit contributions | |
and | -0.007 |
urry | -0.006 |
Follow | -0.006 |
tum | -0.006 |
Nice | -0.006 |
Token | led |
Token ID | 992 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
bur | +0.095 |
fourteen | +0.092 |
abi | +0.083 |
ps | +0.076 |
ible | +0.075 |
Neg logit contributions | |
and | -0.210 |
Follow | -0.174 |
urry | -0.173 |
apping | -0.167 |
of | -0.167 |
Token | again |
Token ID | 757 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
bur | +0.036 |
eling | +0.035 |
fourteen | +0.035 |
ps | +0.034 |
IT | +0.034 |
Neg logit contributions | |
and | -0.037 |
urry | -0.030 |
Follow | -0.030 |
of | -0.028 |
apping | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -3.96e-03 |
Pos logit contributions | |
and | +0.037 |
urry | +0.030 |
of | +0.027 |
Follow | +0.027 |
tum | +0.027 |
Neg logit contributions | |
bur | -0.040 |
fourteen | -0.039 |
ps | -0.038 |
eling | -0.038 |
IT | -0.036 |
Token | played |
Token ID | 2826 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
and | +0.007 |
urry | +0.006 |
tum | +0.006 |
Follow | +0.006 |
Stay | +0.006 |
Neg logit contributions | |
ps | -0.005 |
bur | -0.005 |
fourteen | -0.005 |
eling | -0.005 |
cking | -0.005 |
Token | together |
Token ID | 1978 |
Feature activation | +3.07e-03 |
Pos logit contributions | |
bur | +0.016 |
fourteen | +0.015 |
eling | +0.015 |
ps | +0.014 |
blew | +0.014 |
Neg logit contributions | |
and | -0.018 |
Follow | -0.015 |
urry | -0.015 |
tum | -0.014 |
apping | -0.014 |
Token | but |
Token ID | 475 |
Feature activation | +4.77e-05 |
Pos logit contributions | |
and | +0.018 |
urry | +0.015 |
Follow | +0.014 |
Nice | +0.014 |
tum | +0.013 |
Neg logit contributions | |
bur | -0.017 |
eling | -0.017 |
ps | -0.017 |
fourteen | -0.017 |
ck | -0.016 |
Token | today |
Token ID | 1909 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
ps | +0.000 |
bur | +0.000 |
eling | +0.000 |
fourteen | +0.000 |
cking | +0.000 |
Neg logit contributions | |
and | -0.000 |
urry | -0.000 |
Follow | -0.000 |
Today | -0.000 |
Nice | -0.000 |
Token | they |
Token ID | 484 |
Feature activation | +8.34e-03 |
Pos logit contributions | |
and | +0.093 |
urry | +0.083 |
of | +0.079 |
Follow | +0.079 |
tum | +0.078 |
Neg logit contributions | |
bur | -0.081 |
ps | -0.080 |
fourteen | -0.078 |
eling | -0.078 |
cking | -0.075 |
Token | were |
Token ID | 547 |
Feature activation | +6.38e-03 |
Pos logit contributions | |
bur | +0.014 |
fourteen | +0.014 |
ps | +0.014 |
cking | +0.013 |
eling | +0.013 |
Neg logit contributions | |
and | -0.011 |
urry | -0.010 |
Follow | -0.009 |
tum | -0.009 |
of | -0.009 |
Token | qu |
Token ID | 627 |
Feature activation | +2.19e-03 |
Pos logit contributions | |
bur | +0.009 |
fourteen | +0.009 |
eling | +0.008 |
ible | +0.008 |
ps | +0.008 |
Neg logit contributions | |
and | -0.011 |
urry | -0.009 |
Follow | -0.009 |
of | -0.009 |
apping | -0.009 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.40e-02 |
Pos logit contributions | |
bur | +0.003 |
fourteen | +0.003 |
ps | +0.003 |
eling | +0.003 |
blew | +0.003 |
Neg logit contributions | |
and | -0.004 |
urry | -0.003 |
Follow | -0.003 |
apping | -0.003 |
aring | -0.003 |
Token | ling |
Token ID | 1359 |
Feature activation | +5.21e-03 |
Pos logit contributions | |
bur | +0.108 |
fourteen | +0.105 |
abi | +0.094 |
IT | +0.090 |
ible | +0.089 |
Neg logit contributions | |
and | -0.194 |
urry | -0.160 |
Follow | -0.159 |
apping | -0.155 |
of | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | +4.83e-03 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.006 |
ps | +0.006 |
blew | +0.006 |
Neg logit contributions | |
and | -0.010 |
urry | -0.008 |
Follow | -0.008 |
tum | -0.008 |
Nice | -0.008 |
Token | Jenna |
Token ID | 45377 |
Feature activation | +8.28e-03 |
Pos logit contributions | |
bur | +0.008 |
ps | +0.008 |
eling | +0.008 |
fourteen | +0.008 |
ck | +0.008 |
Neg logit contributions | |
and | -0.007 |
urry | -0.006 |
Follow | -0.006 |
Today | -0.005 |
Nice | -0.005 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.15e-03 |
Pos logit contributions | |
ps | +0.012 |
bur | +0.012 |
fourteen | +0.011 |
cking | +0.011 |
eling | +0.011 |
Neg logit contributions | |
and | -0.013 |
urry | -0.012 |
Follow | -0.011 |
tum | -0.011 |
of | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
and | +0.010 |
urry | +0.008 |
Follow | +0.007 |
tum | +0.007 |
of | +0.007 |
Neg logit contributions | |
eling | -0.015 |
bur | -0.015 |
ps | -0.015 |
fourteen | -0.015 |
cking | -0.015 |
Token | temper |
Token ID | 4124 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
bur | +0.022 |
fourteen | +0.021 |
eling | +0.021 |
ps | +0.021 |
blew | +0.020 |
Neg logit contributions | |
and | -0.022 |
urry | -0.019 |
Follow | -0.019 |
Nice | -0.017 |
tum | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -9.85e-05 |
Pos logit contributions | |
bur | +0.029 |
fourteen | +0.028 |
eling | +0.028 |
ps | +0.028 |
blew | +0.027 |
Neg logit contributions | |
and | -0.035 |
urry | -0.031 |
Follow | -0.031 |
Nice | -0.028 |
tum | -0.028 |
Token | often |
Token ID | 1690 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
and | +0.000 |
urry | +0.000 |
Follow | +0.000 |
tum | +0.000 |
Nice | +0.000 |
Neg logit contributions | |
bur | -0.000 |
fourteen | -0.000 |
eling | -0.000 |
ps | -0.000 |
blew | -0.000 |
Token | threw |
Token ID | 9617 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
and | +0.029 |
urry | +0.025 |
Follow | +0.025 |
tum | +0.023 |
Stay | +0.023 |
Neg logit contributions | |
bur | -0.020 |
fourteen | -0.019 |
ps | -0.018 |
eling | -0.018 |
cking | -0.018 |
Token | temper |
Token ID | 4124 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
bur | +0.039 |
fourteen | +0.038 |
eling | +0.037 |
ps | +0.036 |
blew | +0.036 |
Neg logit contributions | |
and | -0.065 |
urry | -0.055 |
Follow | -0.055 |
tum | -0.052 |
Nice | -0.051 |
Token | tant |
Token ID | 24246 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
bur | +0.032 |
fourteen | +0.031 |
eling | +0.030 |
ps | +0.029 |
blew | +0.029 |
Neg logit contributions | |
and | -0.035 |
urry | -0.029 |
Follow | -0.029 |
tum | -0.026 |
Nice | -0.026 |
Token | rums |
Token ID | 45241 |
Feature activation | +5.10e-03 |
Pos logit contributions | |
bur | +0.146 |
blew | +0.146 |
fourteen | +0.145 |
eling | +0.134 |
IT | +0.133 |
Neg logit contributions | |
and | -0.152 |
urry | -0.120 |
Follow | -0.113 |
aring | -0.108 |
apping | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | +3.55e-03 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.006 |
ps | +0.006 |
blew | +0.005 |
Neg logit contributions | |
and | -0.010 |
urry | -0.009 |
Follow | -0.009 |
Nice | -0.008 |
tum | -0.008 |
Token | One |
Token ID | 1881 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
fourteen | +0.005 |
bur | +0.005 |
eling | +0.005 |
IT | +0.005 |
ps | +0.005 |
Neg logit contributions | |
and | -0.005 |
urry | -0.005 |
Follow | -0.005 |
rav | -0.004 |
apping | -0.004 |
Token | day |
Token ID | 1110 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
and | +0.037 |
Follow | +0.032 |
urry | +0.030 |
vere | +0.029 |
rav | +0.027 |
Neg logit contributions | |
blew | -0.045 |
bur | -0.044 |
eling | -0.042 |
fourteen | -0.042 |
ible | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -5.22e-03 |
Pos logit contributions | |
bur | +0.026 |
fourteen | +0.026 |
eling | +0.025 |
ps | +0.025 |
blew | +0.025 |
Neg logit contributions | |
and | -0.020 |
urry | -0.017 |
Follow | -0.017 |
tum | -0.015 |
Nice | -0.015 |
Token | alone |
Token ID | 3436 |
Feature activation | -6.94e-03 |
Pos logit contributions | |
and | +0.065 |
urry | +0.057 |
Follow | +0.056 |
Nice | +0.051 |
tum | +0.051 |
Neg logit contributions | |
bur | -0.052 |
fourteen | -0.051 |
eling | -0.050 |
ps | -0.049 |
blew | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -7.13e-03 |
Pos logit contributions | |
and | +0.011 |
urry | +0.009 |
Follow | +0.009 |
tum | +0.008 |
Nice | +0.008 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.010 |
eling | -0.010 |
ps | -0.010 |
blew | -0.010 |
Token | without |
Token ID | 1231 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
and | +0.012 |
urry | +0.010 |
Follow | +0.010 |
tum | +0.009 |
Nice | +0.009 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.010 |
eling | -0.010 |
ps | -0.010 |
ible | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -8.32e-03 |
Pos logit contributions | |
bur | +0.038 |
fourteen | +0.036 |
IT | +0.034 |
ami | +0.034 |
blew | +0.034 |
Neg logit contributions | |
and | -0.065 |
urry | -0.057 |
tum | -0.055 |
Stay | -0.054 |
Follow | -0.054 |
Token | cool |
Token ID | 3608 |
Feature activation | +9.08e-03 |
Pos logit contributions | |
and | +0.014 |
Follow | +0.012 |
urry | +0.012 |
Go | +0.011 |
School | +0.011 |
Neg logit contributions | |
blew | -0.011 |
bur | -0.011 |
fourteen | -0.011 |
ps | -0.010 |
eling | -0.010 |
Token | hairst |
Token ID | 42658 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
bur | +0.014 |
blew | +0.013 |
fourteen | +0.013 |
ps | +0.013 |
ck | +0.012 |
Neg logit contributions | |
and | -0.015 |
Follow | -0.013 |
urry | -0.012 |
vere | -0.011 |
Go | -0.011 |
Token | yle |
Token ID | 2349 |
Feature activation | -1.10e-03 |
Pos logit contributions | |
IT | +0.141 |
fourteen | +0.139 |
ible | +0.135 |
bur | +0.133 |
blew | +0.132 |
Neg logit contributions | |
and | -0.149 |
apping | -0.129 |
of | -0.121 |
urry | -0.117 |
vere | -0.113 |
Token | and |
Token ID | 290 |
Feature activation | -9.50e-03 |
Pos logit contributions | |
and | +0.002 |
urry | +0.002 |
Follow | +0.002 |
Nice | +0.001 |
tum | +0.001 |
Neg logit contributions | |
bur | -0.002 |
fourteen | -0.002 |
eling | -0.001 |
ps | -0.001 |
blew | -0.001 |
Token | without |
Token ID | 1231 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
and | +0.017 |
Follow | +0.015 |
tum | +0.015 |
urry | +0.015 |
of | +0.014 |
Neg logit contributions | |
bur | -0.012 |
fourteen | -0.011 |
eling | -0.011 |
ami | -0.010 |
ible | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -9.76e-03 |
Pos logit contributions | |
bur | +0.043 |
fourteen | +0.041 |
ami | +0.041 |
rob | +0.040 |
IT | +0.040 |
Neg logit contributions | |
and | -0.068 |
urry | -0.059 |
welcome | -0.058 |
tum | -0.058 |
Stay | -0.056 |
Token | solution |
Token ID | 4610 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
and | +0.015 |
urry | +0.013 |
Follow | +0.012 |
tum | +0.011 |
Nice | +0.011 |
Neg logit contributions | |
bur | -0.015 |
fourteen | -0.015 |
eling | -0.015 |
ps | -0.015 |
ck | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +3.68e-03 |
Pos logit contributions | |
fourteen | +0.032 |
ible | +0.031 |
rob | +0.030 |
cking | +0.029 |
lled | +0.029 |
Neg logit contributions | |
and | -0.034 |
urry | -0.031 |
welcome | -0.029 |
Follow | -0.029 |
of | -0.028 |
Token | store |
Token ID | 3650 |
Feature activation | +9.27e-03 |
Pos logit contributions | |
bur | +0.005 |
eling | +0.005 |
blew | +0.005 |
fourteen | +0.004 |
ps | +0.004 |
Neg logit contributions | |
and | -0.007 |
urry | -0.006 |
Follow | -0.006 |
Nice | -0.005 |
rav | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -4.00e-03 |
Pos logit contributions | |
bur | +0.012 |
fourteen | +0.011 |
blew | +0.011 |
ps | +0.011 |
eling | +0.010 |
Neg logit contributions | |
and | -0.017 |
Follow | -0.015 |
urry | -0.015 |
Nice | -0.014 |
rav | -0.014 |
Token | got |
Token ID | 1392 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
and | +0.007 |
urry | +0.006 |
Follow | +0.006 |
tum | +0.006 |
of | +0.006 |
Neg logit contributions | |
fourteen | -0.005 |
ps | -0.005 |
eling | -0.005 |
ible | -0.005 |
bur | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +4.47e-03 |
Pos logit contributions | |
bur | +0.027 |
fourteen | +0.026 |
eling | +0.025 |
ps | +0.025 |
blew | +0.024 |
Neg logit contributions | |
and | -0.061 |
urry | -0.055 |
Follow | -0.055 |
Nice | -0.051 |
tum | -0.051 |
Token | bal |
Token ID | 3652 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.005 |
ps | +0.005 |
blew | +0.005 |
Neg logit contributions | |
and | -0.008 |
urry | -0.007 |
Follow | -0.007 |
tum | -0.006 |
Nice | -0.006 |
Token | m |
Token ID | 76 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
bur | +0.129 |
eling | +0.117 |
fourteen | +0.115 |
blew | +0.115 |
ries | +0.112 |
Neg logit contributions | |
and | -0.160 |
Follow | -0.147 |
urry | -0.143 |
tum | -0.140 |
Nice | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | +6.48e-03 |
Pos logit contributions | |
and | +0.025 |
urry | +0.022 |
Follow | +0.020 |
tum | +0.020 |
apping | +0.019 |
Neg logit contributions | |
bur | -0.026 |
fourteen | -0.026 |
eling | -0.025 |
ps | -0.025 |
IT | -0.025 |
Token | She |
Token ID | 1375 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
bur | +0.010 |
eling | +0.010 |
fourteen | +0.010 |
ps | +0.010 |
ck | +0.010 |
Neg logit contributions | |
and | -0.010 |
urry | -0.008 |
Follow | -0.008 |
tum | -0.008 |
Nice | -0.007 |
Token | carefully |
Token ID | 7773 |
Feature activation | +7.72e-05 |
Pos logit contributions | |
fourteen | +0.016 |
bur | +0.015 |
ps | +0.015 |
eling | +0.015 |
IT | +0.014 |
Neg logit contributions | |
and | -0.026 |
urry | -0.023 |
Follow | -0.023 |
tum | -0.022 |
Stay | -0.021 |
Token | rubbed |
Token ID | 31862 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
fourteen | +0.000 |
bur | +0.000 |
eling | +0.000 |
ible | +0.000 |
cks | +0.000 |
Neg logit contributions | |
and | -0.000 |
urry | -0.000 |
tum | -0.000 |
Follow | -0.000 |
of | -0.000 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
and | +0.005 |
of | +0.005 |
urry | +0.005 |
Follow | +0.004 |
tum | +0.004 |
Neg logit contributions | |
bur | -0.006 |
ps | -0.006 |
eling | -0.006 |
cking | -0.005 |
ck | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -2.30e-03 |
Pos logit contributions | |
eling | +0.027 |
cking | +0.026 |
bur | +0.026 |
fourteen | +0.026 |
ck | +0.026 |
Neg logit contributions | |
and | -0.021 |
Follow | -0.018 |
urry | -0.017 |
of | -0.017 |
story | -0.016 |
Token | Jack |
Token ID | 3619 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
and | +0.002 |
urry | +0.002 |
Stay | +0.002 |
School | +0.002 |
Today | +0.002 |
Neg logit contributions | |
ps | -0.004 |
cking | -0.004 |
bur | -0.004 |
fourteen | -0.004 |
ck | -0.004 |
Token | never |
Token ID | 1239 |
Feature activation | -9.57e-03 |
Pos logit contributions | |
eling | +0.019 |
ries | +0.019 |
cking | +0.019 |
bur | +0.019 |
fourteen | +0.018 |
Neg logit contributions | |
and | -0.013 |
urry | -0.011 |
of | -0.011 |
Follow | -0.010 |
story | -0.010 |
Token | qu |
Token ID | 627 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
and | +0.019 |
urry | +0.017 |
Follow | +0.017 |
tum | +0.015 |
Nice | +0.015 |
Neg logit contributions | |
bur | -0.011 |
fourteen | -0.010 |
eling | -0.010 |
ps | -0.010 |
ck | -0.010 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
bur | +0.008 |
fourteen | +0.007 |
eling | +0.007 |
ps | +0.007 |
blew | +0.007 |
Neg logit contributions | |
and | -0.010 |
urry | -0.008 |
Follow | -0.008 |
apping | -0.008 |
tum | -0.008 |
Token | led |
Token ID | 992 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
bur | +0.097 |
fourteen | +0.092 |
abi | +0.084 |
afar | +0.076 |
ps | +0.076 |
Neg logit contributions | |
and | -0.208 |
urry | -0.170 |
Follow | -0.170 |
of | -0.168 |
apping | -0.167 |
Token | again |
Token ID | 757 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
bur | +0.038 |
eling | +0.037 |
fourteen | +0.036 |
ps | +0.036 |
IT | +0.035 |
Neg logit contributions | |
and | -0.035 |
urry | -0.027 |
of | -0.027 |
Follow | -0.027 |
apping | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +9.59e-03 |
Pos logit contributions | |
and | +0.034 |
urry | +0.027 |
of | +0.025 |
apping | +0.025 |
story | +0.024 |
Neg logit contributions | |
bur | -0.040 |
fourteen | -0.038 |
ps | -0.037 |
eling | -0.037 |
rob | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
eling | +0.019 |
bur | +0.019 |
ps | +0.018 |
ck | +0.018 |
cking | +0.018 |
Neg logit contributions | |
and | -0.013 |
urry | -0.010 |
Follow | -0.009 |
Nice | -0.009 |
Today | -0.008 |
Token | learned |
Token ID | 4499 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
ries | +0.035 |
ps | +0.033 |
cking | +0.033 |
fourteen | +0.033 |
bur | +0.033 |
Neg logit contributions | |
and | -0.019 |
urry | -0.018 |
Follow | -0.016 |
of | -0.016 |
Stay | -0.016 |
Token | she |
Token ID | 673 |
Feature activation | +6.79e-03 |
Pos logit contributions | |
and | +0.063 |
Follow | +0.057 |
urry | +0.056 |
Today | +0.052 |
suits | +0.052 |
Neg logit contributions | |
fourteen | -0.028 |
ps | -0.027 |
eling | -0.026 |
bur | -0.026 |
blew | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
bur | +0.011 |
fourteen | +0.010 |
eling | +0.010 |
ps | +0.010 |
cking | +0.010 |
Neg logit contributions | |
and | -0.011 |
urry | -0.009 |
Follow | -0.009 |
tum | -0.008 |
Stay | -0.008 |
Token | sorry |
Token ID | 7926 |
Feature activation | -1.74e-03 |
Pos logit contributions | |
rob | +0.035 |
bur | +0.034 |
ible | +0.034 |
ami | +0.033 |
IT | +0.033 |
Neg logit contributions | |
welcome | -0.018 |
and | -0.017 |
of | -0.017 |
urry | -0.016 |
Stay | -0.015 |
Token | for |
Token ID | 329 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
and | +0.003 |
urry | +0.003 |
Follow | +0.003 |
tum | +0.002 |
Nice | +0.002 |
Neg logit contributions | |
bur | -0.002 |
fourteen | -0.002 |
eling | -0.002 |
ps | -0.002 |
blew | -0.002 |
Token | qu |
Token ID | 627 |
Feature activation | +1.44e-03 |
Pos logit contributions | |
and | +0.038 |
urry | +0.032 |
birthday | +0.030 |
story | +0.030 |
vere | +0.030 |
Neg logit contributions | |
bur | -0.034 |
fourteen | -0.033 |
ible | -0.032 |
rob | -0.031 |
blew | -0.030 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.35e-02 |
Pos logit contributions | |
bur | +0.002 |
fourteen | +0.002 |
eling | +0.002 |
ps | +0.002 |
blew | +0.002 |
Neg logit contributions | |
and | -0.003 |
urry | -0.002 |
Follow | -0.002 |
tum | -0.002 |
Nice | -0.002 |
Token | ling |
Token ID | 1359 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
bur | +0.121 |
fourteen | +0.115 |
abi | +0.104 |
IT | +0.102 |
ible | +0.101 |
Neg logit contributions | |
and | -0.184 |
urry | -0.151 |
Follow | -0.149 |
apping | -0.142 |
of | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | +1.42e-03 |
Pos logit contributions | |
bur | +0.005 |
fourteen | +0.004 |
eling | +0.004 |
ps | +0.004 |
blew | +0.004 |
Neg logit contributions | |
and | -0.007 |
urry | -0.006 |
Follow | -0.006 |
Nice | -0.005 |
tum | -0.005 |
Token | From |
Token ID | 3574 |
Feature activation | +1.18e-03 |
Pos logit contributions | |
bur | +0.003 |
eling | +0.003 |
ck | +0.003 |
ps | +0.003 |
cking | +0.002 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
Follow | -0.002 |
Nice | -0.001 |
Today | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | -8.99e-03 |
Pos logit contributions | |
ps | +0.002 |
ible | +0.002 |
fourteen | +0.002 |
bur | +0.002 |
eling | +0.002 |
Neg logit contributions | |
urry | -0.002 |
of | -0.001 |
and | -0.001 |
Follow | -0.001 |
welcome | -0.001 |
Token | day |
Token ID | 1110 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
and | +0.013 |
Follow | +0.011 |
urry | +0.010 |
ooters | +0.010 |
vere | +0.009 |
Neg logit contributions | |
blew | -0.015 |
bur | -0.015 |
fourteen | -0.015 |
ible | -0.014 |
eling | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +4.19e-03 |
Pos logit contributions | |
bur | +0.004 |
fourteen | +0.004 |
eling | +0.003 |
ps | +0.003 |
blew | +0.003 |
Neg logit contributions | |
and | -0.009 |
urry | -0.008 |
Follow | -0.008 |
story | -0.007 |
Stay | -0.007 |
Token | story |
Token ID | 1621 |
Feature activation | -5.81e-03 |
Pos logit contributions | |
blew | +0.008 |
ps | +0.008 |
fourteen | +0.008 |
bur | +0.008 |
ck | +0.008 |
Neg logit contributions | |
and | -0.005 |
suits | -0.004 |
urry | -0.004 |
Follow | -0.004 |
apping | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
and | +0.011 |
adows | +0.010 |
urry | +0.009 |
tum | +0.009 |
Follow | +0.009 |
Neg logit contributions | |
fourteen | -0.006 |
ps | -0.006 |
blew | -0.006 |
mill | -0.006 |
blows | -0.006 |
Token | : |
Token ID | 25 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
fourteen | +0.009 |
bur | +0.009 |
blew | +0.009 |
ps | +0.009 |
eling | +0.009 |
Neg logit contributions | |
and | -0.030 |
Follow | -0.026 |
suits | -0.026 |
urry | -0.025 |
tum | -0.025 |
Token | Qu |
Token ID | 2264 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
and | +0.025 |
suits | +0.022 |
tum | +0.021 |
urry | +0.021 |
Follow | +0.020 |
Neg logit contributions | |
fourteen | -0.023 |
blew | -0.022 |
ck | -0.021 |
bur | -0.021 |
Why | -0.021 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.34e-02 |
Pos logit contributions | |
and | +0.033 |
Follow | +0.029 |
urry | +0.029 |
tum | +0.028 |
Nice | +0.027 |
Neg logit contributions | |
bur | -0.028 |
eling | -0.028 |
ps | -0.027 |
fourteen | -0.027 |
ible | -0.027 |
Token | lling |
Token ID | 2680 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
eling | +0.081 |
lled | +0.079 |
ps | +0.079 |
blew | +0.076 |
bur | +0.075 |
Neg logit contributions | |
and | -0.204 |
urry | -0.190 |
Follow | -0.187 |
tum | -0.183 |
Nice | -0.179 |
Token | over |
Token ID | 625 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ps | +0.025 |
fourteen | +0.024 |
blew | +0.024 |
eling | +0.024 |
ible | +0.023 |
Neg logit contributions | |
and | -0.023 |
urry | -0.019 |
Follow | -0.019 |
tum | -0.017 |
Nice | -0.017 |
Token | small |
Token ID | 1402 |
Feature activation | +2.81e-03 |
Pos logit contributions | |
fourteen | +0.015 |
ps | +0.015 |
eling | +0.015 |
ible | +0.015 |
bur | +0.014 |
Neg logit contributions | |
and | -0.019 |
Follow | -0.017 |
urry | -0.017 |
tum | -0.016 |
Nice | -0.016 |
Token | things |
Token ID | 1243 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
ible | +0.003 |
eling | +0.003 |
ps | +0.003 |
blew | +0.003 |
fourteen | +0.003 |
Neg logit contributions | |
and | -0.005 |
urry | -0.005 |
Follow | -0.005 |
Nice | -0.004 |
rav | -0.004 |
Token | is |
Token ID | 318 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
bur | +0.052 |
fourteen | +0.051 |
eling | +0.050 |
ps | +0.050 |
ck | +0.049 |
Neg logit contributions | |
and | -0.037 |
urry | -0.031 |
Follow | -0.030 |
Nice | -0.027 |
tum | -0.026 |
Token | good |
Token ID | 922 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
bur | +0.014 |
IT | +0.014 |
fourteen | +0.014 |
rob | +0.013 |
ami | +0.013 |
Neg logit contributions | |
and | -0.009 |
of | -0.008 |
urry | -0.008 |
tum | -0.008 |
welcome | -0.007 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
bur | +0.053 |
fourteen | +0.052 |
ps | +0.051 |
eling | +0.051 |
blew | +0.050 |
Neg logit contributions | |
and | -0.048 |
Follow | -0.040 |
urry | -0.040 |
Nice | -0.036 |
tum | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -3.06e-03 |
Pos logit contributions | |
bur | +0.034 |
fourteen | +0.033 |
eling | +0.033 |
ps | +0.033 |
blew | +0.032 |
Neg logit contributions | |
and | -0.047 |
urry | -0.040 |
Follow | -0.040 |
Nice | -0.037 |
tum | -0.036 |
Token | never |
Token ID | 1239 |
Feature activation | -8.36e-03 |
Pos logit contributions | |
and | +0.006 |
urry | +0.005 |
Follow | +0.005 |
Nice | +0.004 |
tum | +0.004 |
Neg logit contributions | |
bur | -0.004 |
fourteen | -0.004 |
eling | -0.004 |
ps | -0.004 |
blew | -0.004 |
Token | qu |
Token ID | 627 |
Feature activation | +5.35e-03 |
Pos logit contributions | |
and | +0.017 |
urry | +0.015 |
Follow | +0.015 |
Nice | +0.014 |
tum | +0.014 |
Neg logit contributions | |
bur | -0.009 |
fourteen | -0.009 |
eling | -0.009 |
ps | -0.009 |
blew | -0.008 |
Token | arre |
Token ID | 9624 |
Feature activation | +9.32e-02 |
Pos logit contributions | |
bur | +0.007 |
fourteen | +0.007 |
eling | +0.007 |
ps | +0.007 |
blew | +0.007 |
Neg logit contributions | |
and | -0.009 |
urry | -0.008 |
Follow | -0.008 |
apping | -0.007 |
aring | -0.007 |
Token | led |
Token ID | 992 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
bur | +0.105 |
fourteen | +0.101 |
abi | +0.094 |
ami | +0.084 |
afar | +0.084 |
Neg logit contributions | |
and | -0.202 |
urry | -0.161 |
Follow | -0.161 |
of | -0.159 |
apping | -0.158 |
Token | again |
Token ID | 757 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
bur | +0.034 |
fourteen | +0.033 |
eling | +0.033 |
ps | +0.032 |
blew | +0.032 |
Neg logit contributions | |
and | -0.037 |
urry | -0.031 |
Follow | -0.031 |
Nice | -0.028 |
tum | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +5.17e-03 |
Pos logit contributions | |
and | +0.040 |
urry | +0.032 |
of | +0.031 |
Follow | +0.030 |
birthday | +0.029 |
Neg logit contributions | |
bur | -0.040 |
fourteen | -0.038 |
ps | -0.037 |
eling | -0.037 |
IT | -0.035 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +7.34e-04 |
Pos logit contributions | |
bur | +0.009 |
eling | +0.009 |
ps | +0.009 |
ck | +0.009 |
blew | +0.009 |
Neg logit contributions | |
and | -0.007 |
urry | -0.006 |
Follow | -0.006 |
Nice | -0.005 |
Today | -0.005 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
ps | +0.001 |
blew | +0.001 |
eling | +0.001 |
Neg logit contributions | |
and | -0.001 |
urry | -0.001 |
Follow | -0.001 |
tum | -0.001 |
Nice | -0.001 |
Token | kept |
Token ID | 4030 |
Feature activation | +4.69e-03 |
Pos logit contributions | |
fourteen | +0.014 |
bur | +0.014 |
ps | +0.014 |
cking | +0.013 |
eling | +0.013 |
Neg logit contributions | |
and | -0.020 |
urry | -0.018 |
Follow | -0.018 |
tum | -0.018 |
Stay | -0.017 |
Token | crying |
Token ID | 13774 |
Feature activation | -4.69e-03 |
Pos logit contributions | |
fourteen | +0.008 |
lled | +0.007 |
ami | +0.007 |
ible | +0.007 |
rob | +0.007 |
Neg logit contributions | |
and | -0.006 |
story | -0.005 |
tum | -0.005 |
Stay | -0.005 |
urry | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -4.73e-03 |
Pos logit contributions | |
and | +0.007 |
urry | +0.006 |
Follow | +0.006 |
tum | +0.006 |
apping | +0.006 |
Neg logit contributions | |
bur | -0.007 |
fourteen | -0.007 |
eling | -0.007 |
ps | -0.007 |
blew | -0.007 |
Token | throwing |
Token ID | 9644 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
and | +0.009 |
urry | +0.008 |
Follow | +0.008 |
tum | +0.008 |
Stay | +0.007 |
Neg logit contributions | |
bur | -0.005 |
fourteen | -0.005 |
ps | -0.005 |
ible | -0.005 |
eling | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -6.14e-03 |
Pos logit contributions | |
bur | +0.026 |
ami | +0.025 |
eling | +0.025 |
fourteen | +0.025 |
cking | +0.024 |
Neg logit contributions | |
and | -0.043 |
tum | -0.039 |
of | -0.038 |
Stay | -0.035 |
welcome | -0.034 |
Token | tant |
Token ID | 24246 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
and | +0.011 |
Follow | +0.011 |
urry | +0.010 |
School | +0.010 |
Nice | +0.010 |
Neg logit contributions | |
blew | -0.007 |
bur | -0.007 |
fourteen | -0.007 |
ps | -0.006 |
ck | -0.006 |
Token | rum |
Token ID | 6582 |
Feature activation | -7.02e-03 |
Pos logit contributions | |
bur | +0.131 |
fourteen | +0.129 |
eling | +0.126 |
blew | +0.123 |
ps | +0.123 |
Neg logit contributions | |
and | -0.159 |
urry | -0.136 |
Follow | -0.133 |
Nice | -0.121 |
tum | -0.120 |
Token | . |
Token ID | 13 |
Feature activation | +5.60e-03 |
Pos logit contributions | |
and | +0.013 |
urry | +0.012 |
Follow | +0.011 |
tum | +0.011 |
Nice | +0.011 |
Neg logit contributions | |
bur | -0.009 |
fourteen | -0.008 |
eling | -0.008 |
ps | -0.008 |
blew | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +2.59e-03 |
Pos logit contributions | |
fourteen | +0.007 |
bur | +0.007 |
IT | +0.007 |
blew | +0.007 |
ps | +0.007 |
Neg logit contributions | |
and | -0.009 |
Follow | -0.008 |
urry | -0.008 |
rav | -0.007 |
vere | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.99e-03 |
Pos logit contributions | |
bur | +0.003 |
fourteen | +0.003 |
ck | +0.003 |
blew | +0.003 |
lled | +0.003 |
Neg logit contributions | |
and | -0.004 |
Follow | -0.004 |
urry | -0.004 |
tum | -0.004 |
Nice | -0.004 |
Token | Finally |
Token ID | 11158 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
fourteen | +0.003 |
bur | +0.003 |
eling | +0.003 |
blew | +0.003 |
ck | +0.003 |
Neg logit contributions | |
and | -0.006 |
Follow | -0.005 |
urry | -0.005 |
Nice | -0.005 |
tum | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
and | +0.002 |
urry | +0.002 |
Follow | +0.002 |
vere | +0.001 |
rav | +0.001 |
Neg logit contributions | |
fourteen | -0.002 |
bur | -0.002 |
ps | -0.002 |
blew | -0.002 |
eling | -0.002 |
Token | said |
Token ID | 531 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
bur | +0.018 |
fourteen | +0.018 |
eling | +0.017 |
ps | +0.017 |
IT | +0.017 |
Neg logit contributions | |
and | -0.023 |
urry | -0.020 |
Follow | -0.020 |
tum | -0.019 |
suits | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
and | +0.031 |
Follow | +0.029 |
urry | +0.028 |
case | +0.025 |
ney | +0.025 |
Neg logit contributions | |
her | -0.021 |
fourteen | -0.020 |
blew | -0.019 |
ps | -0.019 |
how | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
and | +0.025 |
urry | +0.022 |
Follow | +0.021 |
tum | +0.019 |
vere | +0.019 |
Neg logit contributions | |
fourteen | -0.023 |
blew | -0.023 |
bur | -0.022 |
ps | -0.022 |
abi | -0.020 |
Token | L |
Token ID | 43 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
and | +0.024 |
Follow | +0.020 |
urry | +0.019 |
School | +0.019 |
rav | +0.019 |
Neg logit contributions | |
eling | -0.015 |
bur | -0.014 |
ps | -0.014 |
cking | -0.013 |
ck | -0.013 |
Token | adies |
Token ID | 50192 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
bur | +0.018 |
eling | +0.018 |
fourteen | +0.017 |
blew | +0.017 |
ps | +0.017 |
Neg logit contributions | |
and | -0.014 |
urry | -0.012 |
Follow | -0.011 |
apping | -0.010 |
Go | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -7.55e-03 |
Pos logit contributions | |
and | +0.185 |
urry | +0.162 |
Follow | +0.162 |
Nice | +0.146 |
tum | +0.145 |
Neg logit contributions | |
bur | -0.145 |
fourteen | -0.142 |
ps | -0.138 |
eling | -0.135 |
blew | -0.134 |
Token | gentlemen |
Token ID | 28527 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
and | +0.010 |
urry | +0.008 |
Follow | +0.008 |
tum | +0.007 |
Nice | +0.007 |
Neg logit contributions | |
fourteen | -0.013 |
bur | -0.013 |
blew | -0.013 |
ck | -0.013 |
eling | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
and | +0.101 |
Follow | +0.082 |
urry | +0.082 |
tum | +0.075 |
Nice | +0.072 |
Neg logit contributions | |
fourteen | -0.112 |
blew | -0.106 |
bur | -0.105 |
ps | -0.105 |
eling | -0.103 |
Token | welcome |
Token ID | 7062 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
and | +0.024 |
urry | +0.021 |
Follow | +0.020 |
tum | +0.020 |
Nice | +0.019 |
Neg logit contributions | |
fourteen | -0.019 |
blew | -0.019 |
ps | -0.018 |
bur | -0.018 |
eling | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
and | +0.066 |
urry | +0.057 |
Follow | +0.056 |
tum | +0.050 |
Nice | +0.050 |
Neg logit contributions | |
bur | -0.064 |
fourteen | -0.063 |
ps | -0.062 |
eling | -0.061 |
blew | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | -4.94e-03 |
Pos logit contributions | |
bur | +0.019 |
fourteen | +0.018 |
ps | +0.018 |
eling | +0.018 |
blew | +0.017 |
Neg logit contributions | |
and | -0.033 |
urry | -0.030 |
Follow | -0.029 |
Nice | -0.027 |
tum | -0.027 |
Token | He |
Token ID | 679 |
Feature activation | +5.45e-04 |
Pos logit contributions | |
and | +0.008 |
urry | +0.007 |
Follow | +0.006 |
rav | +0.006 |
apping | +0.006 |
Neg logit contributions | |
fourteen | -0.007 |
ps | -0.007 |
bur | -0.007 |
blew | -0.007 |
eling | -0.007 |
Token | says |
Token ID | 1139 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
eling | +0.001 |
ps | +0.001 |
lled | +0.001 |
Neg logit contributions | |
and | -0.001 |
Follow | -0.001 |
urry | -0.001 |
tum | -0.001 |
Stay | -0.001 |
Token | " |
Token ID | 366 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
and | +0.044 |
urry | +0.039 |
Follow | +0.039 |
uggle | +0.036 |
anny | +0.036 |
Neg logit contributions | |
fourteen | -0.029 |
blew | -0.026 |
her | -0.026 |
ps | -0.026 |
how | -0.024 |
Token | L |
Token ID | 43 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
and | +0.026 |
Follow | +0.022 |
urry | +0.021 |
Nice | +0.020 |
rav | +0.019 |
Neg logit contributions | |
eling | -0.021 |
ps | -0.021 |
bur | -0.020 |
fourteen | -0.020 |
cking | -0.019 |
Token | adies |
Token ID | 50192 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
bur | +0.023 |
eling | +0.022 |
blew | +0.022 |
fourteen | +0.022 |
cking | +0.021 |
Neg logit contributions | |
and | -0.018 |
urry | -0.015 |
Follow | -0.014 |
Go | -0.012 |
of | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -6.30e-03 |
Pos logit contributions | |
and | +0.181 |
urry | +0.159 |
Follow | +0.159 |
Nice | +0.143 |
tum | +0.142 |
Neg logit contributions | |
bur | -0.138 |
fourteen | -0.135 |
ps | -0.132 |
eling | -0.128 |
blew | -0.126 |
Token | gentlemen |
Token ID | 28527 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
and | +0.008 |
urry | +0.007 |
Follow | +0.006 |
tum | +0.006 |
Nice | +0.006 |
Neg logit contributions | |
fourteen | -0.011 |
bur | -0.011 |
blew | -0.011 |
eling | -0.011 |
ck | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
and | +0.099 |
urry | +0.081 |
Follow | +0.080 |
tum | +0.075 |
Nice | +0.070 |
Neg logit contributions | |
fourteen | -0.114 |
bur | -0.108 |
ps | -0.108 |
blew | -0.107 |
eling | -0.105 |
Token | boys |
Token ID | 6510 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
and | +0.019 |
urry | +0.017 |
Follow | +0.016 |
tum | +0.016 |
Nice | +0.015 |
Neg logit contributions | |
fourteen | -0.014 |
blew | -0.013 |
ps | -0.013 |
bur | -0.013 |
eling | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -5.80e-03 |
Pos logit contributions | |
and | +0.019 |
urry | +0.016 |
Follow | +0.016 |
Go | +0.014 |
story | +0.014 |
Neg logit contributions | |
bur | -0.023 |
eling | -0.022 |
fourteen | -0.022 |
ps | -0.021 |
lled | -0.021 |
Token | they |
Token ID | 484 |
Feature activation | +4.60e-03 |
Pos logit contributions | |
bur | +0.000 |
eling | +0.000 |
fourteen | +0.000 |
ck | +0.000 |
ps | +0.000 |
Neg logit contributions | |
and | -0.000 |
Follow | -0.000 |
urry | -0.000 |
tum | -0.000 |
Nice | -0.000 |
Token | had |
Token ID | 550 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
fourteen | +0.009 |
bur | +0.009 |
IT | +0.009 |
ami | +0.009 |
abi | +0.009 |
Neg logit contributions | |
and | -0.005 |
urry | -0.004 |
Follow | -0.004 |
of | -0.004 |
Stay | -0.003 |
Token | been |
Token ID | 587 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
and | +0.026 |
urry | +0.021 |
of | +0.020 |
Follow | +0.020 |
birthday | +0.019 |
Neg logit contributions | |
bur | -0.026 |
eling | -0.025 |
fourteen | -0.024 |
IT | -0.024 |
ami | -0.024 |
Token | there |
Token ID | 612 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
and | +0.032 |
urry | +0.028 |
Follow | +0.026 |
of | +0.025 |
apping | +0.024 |
Neg logit contributions | |
fourteen | -0.032 |
bur | -0.031 |
IT | -0.030 |
eling | -0.030 |
ami | -0.029 |
Token | for |
Token ID | 329 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
and | +0.033 |
Follow | +0.029 |
urry | +0.029 |
tum | +0.026 |
Nice | +0.026 |
Neg logit contributions | |
bur | -0.022 |
fourteen | -0.022 |
ps | -0.021 |
eling | -0.021 |
blew | -0.021 |
Token | centuries |
Token ID | 10675 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
and | +0.048 |
urry | +0.041 |
Follow | +0.040 |
birthday | +0.038 |
tum | +0.038 |
Neg logit contributions | |
bur | -0.026 |
fourteen | -0.024 |
eling | -0.024 |
blew | -0.023 |
ps | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +9.63e-03 |
Pos logit contributions | |
and | +0.193 |
Follow | +0.188 |
tum | +0.178 |
urry | +0.176 |
rav | +0.173 |
Neg logit contributions | |
ps | -0.074 |
fourteen | -0.071 |
et | -0.069 |
bur | -0.068 |
ible | -0.066 |
Token | The |
Token ID | 383 |
Feature activation | +1.57e-04 |
Pos logit contributions | |
fourteen | +0.012 |
bur | +0.012 |
ps | +0.012 |
eling | +0.011 |
blew | +0.011 |
Neg logit contributions | |
and | -0.017 |
urry | -0.015 |
Follow | -0.015 |
Nice | -0.014 |
rav | -0.013 |
Token | dragon |
Token ID | 10441 |
Feature activation | +2.06e-03 |
Pos logit contributions | |
ps | +0.000 |
fourteen | +0.000 |
bur | +0.000 |
eling | +0.000 |
blew | +0.000 |
Neg logit contributions | |
and | -0.000 |
urry | -0.000 |
Follow | -0.000 |
Nice | -0.000 |
apping | -0.000 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
bur | +0.003 |
fourteen | +0.003 |
eling | +0.003 |
ps | +0.003 |
IT | +0.003 |
Neg logit contributions | |
and | -0.003 |
Follow | -0.003 |
urry | -0.003 |
tum | -0.003 |
of | -0.003 |
Token | at |
Token ID | 379 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
bur | +0.032 |
fourteen | +0.032 |
eling | +0.031 |
ps | +0.031 |
blew | +0.030 |
Neg logit contributions | |
and | -0.037 |
urry | -0.032 |
Follow | -0.031 |
tum | -0.028 |
Nice | -0.028 |
Token | hurt |
Token ID | 5938 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
bur | +0.021 |
fourteen | +0.021 |
eling | +0.020 |
ps | +0.020 |
blew | +0.019 |
Neg logit contributions | |
and | -0.027 |
urry | -0.024 |
Follow | -0.023 |
tum | -0.021 |
Nice | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
bur | +0.042 |
fourteen | +0.041 |
eling | +0.040 |
ps | +0.040 |
blew | +0.039 |
Neg logit contributions | |
and | -0.060 |
urry | -0.052 |
Follow | -0.052 |
tum | -0.048 |
Nice | -0.048 |
Token | He |
Token ID | 679 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
fourteen | +0.017 |
bur | +0.017 |
eling | +0.016 |
blew | +0.016 |
ps | +0.016 |
Neg logit contributions | |
and | -0.017 |
urry | -0.014 |
Follow | -0.014 |
tum | -0.013 |
vere | -0.012 |
Token | thanked |
Token ID | 26280 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
bur | +0.017 |
fourteen | +0.016 |
IT | +0.015 |
ries | +0.015 |
cking | +0.015 |
Neg logit contributions | |
and | -0.019 |
Follow | -0.017 |
urry | -0.017 |
tum | -0.016 |
Stay | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
fourteen | +0.018 |
eling | +0.018 |
bur | +0.018 |
blew | +0.017 |
ible | +0.017 |
Neg logit contributions | |
and | -0.038 |
urry | -0.033 |
Follow | -0.033 |
story | -0.032 |
Nice | -0.032 |
Token | poss |
Token ID | 1184 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
bur | +0.014 |
blew | +0.014 |
fourteen | +0.014 |
ps | +0.014 |
eling | +0.014 |
Neg logit contributions | |
and | -0.021 |
urry | -0.018 |
Follow | -0.018 |
Nice | -0.016 |
apping | -0.016 |
Token | um |
Token ID | 388 |
Feature activation | +4.20e-03 |
Pos logit contributions | |
and | +0.127 |
Follow | +0.112 |
urry | +0.102 |
Nice | +0.101 |
School | +0.097 |
Neg logit contributions | |
bur | -0.155 |
ps | -0.147 |
fourteen | -0.147 |
ck | -0.145 |
ami | -0.141 |
Token | and |
Token ID | 290 |
Feature activation | +6.44e-03 |
Pos logit contributions | |
bur | +0.008 |
fourteen | +0.007 |
eling | +0.007 |
ps | +0.007 |
ck | +0.007 |
Neg logit contributions | |
and | -0.005 |
urry | -0.004 |
Follow | -0.004 |
Nice | -0.004 |
tum | -0.004 |
Token | hopped |
Token ID | 45230 |
Feature activation | -2.06e-03 |
Pos logit contributions | |
fourteen | +0.009 |
ible | +0.008 |
bur | +0.008 |
ps | +0.008 |
eling | +0.008 |
Neg logit contributions | |
and | -0.011 |
Follow | -0.010 |
urry | -0.010 |
tum | -0.009 |
Stay | -0.009 |
Token | away |
Token ID | 1497 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
and | +0.004 |
urry | +0.004 |
Follow | +0.004 |
Nice | +0.003 |
School | +0.003 |
Neg logit contributions | |
bur | -0.002 |
ps | -0.002 |
fourteen | -0.002 |
blew | -0.002 |
ck | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -4.29e-03 |
Pos logit contributions | |
bur | +0.040 |
eling | +0.039 |
fourteen | +0.039 |
ible | +0.038 |
IT | +0.037 |
Neg logit contributions | |
and | -0.037 |
urry | -0.031 |
Follow | -0.030 |
tum | -0.029 |
of | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
and | +0.023 |
Follow | +0.020 |
urry | +0.020 |
tum | +0.018 |
of | +0.018 |
Neg logit contributions | |
bur | -0.019 |
fourteen | -0.019 |
ps | -0.018 |
eling | -0.018 |
cking | -0.018 |
Token | ' |
Token ID | 705 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
and | +0.020 |
Follow | +0.017 |
urry | +0.017 |
Today | +0.015 |
Stay | +0.015 |
Neg logit contributions | |
fourteen | -0.020 |
ps | -0.020 |
blew | -0.020 |
bur | -0.019 |
eling | -0.019 |
Token | R |
Token ID | 49 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
ps | +0.017 |
fourteen | +0.016 |
eling | +0.015 |
cking | +0.015 |
ck | +0.014 |
Neg logit contributions | |
and | -0.022 |
Follow | -0.018 |
urry | -0.017 |
rav | -0.016 |
Go | -0.016 |
Token | ever |
Token ID | 964 |
Feature activation | +3.88e-03 |
Pos logit contributions | |
and | +0.061 |
Follow | +0.052 |
urry | +0.050 |
Nice | +0.049 |
tum | +0.048 |
Neg logit contributions | |
bur | -0.064 |
ible | -0.063 |
fourteen | -0.062 |
eling | -0.062 |
ps | -0.061 |
Token | se |
Token ID | 325 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
ps | +0.006 |
eling | +0.006 |
blew | +0.005 |
Neg logit contributions | |
and | -0.006 |
urry | -0.005 |
Follow | -0.005 |
tum | -0.005 |
Nice | -0.005 |
Token | Mass |
Token ID | 5674 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
bur | +0.063 |
fourteen | +0.061 |
eling | +0.060 |
ps | +0.060 |
blew | +0.059 |
Neg logit contributions | |
and | -0.047 |
urry | -0.039 |
Follow | -0.038 |
tum | -0.033 |
Nice | -0.033 |
Token | age |
Token ID | 496 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
and | +0.138 |
Nice | +0.121 |
tum | +0.118 |
urry | +0.114 |
Follow | +0.114 |
Neg logit contributions | |
bur | -0.125 |
ps | -0.123 |
ible | -0.121 |
fourteen | -0.120 |
IT | -0.118 |
Token | '. |
Token ID | 4458 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
bur | +0.025 |
fourteen | +0.024 |
ps | +0.024 |
eling | +0.024 |
blew | +0.023 |
Neg logit contributions | |
and | -0.026 |
urry | -0.022 |
Follow | -0.022 |
Nice | -0.020 |
tum | -0.020 |
Token | She |
Token ID | 1375 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
and | +0.020 |
Follow | +0.017 |
urry | +0.017 |
Nice | +0.015 |
tum | +0.015 |
Neg logit contributions | |
bur | -0.023 |
eling | -0.023 |
fourteen | -0.022 |
ps | -0.022 |
ck | -0.022 |
Token | pressed |
Token ID | 12070 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
bur | +0.019 |
fourteen | +0.018 |
eling | +0.018 |
ps | +0.017 |
cking | +0.017 |
Neg logit contributions | |
and | -0.024 |
Follow | -0.021 |
urry | -0.021 |
of | -0.020 |
tum | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +3.06e-03 |
Pos logit contributions | |
bur | +0.041 |
eling | +0.038 |
IT | +0.035 |
cking | +0.035 |
fourteen | +0.035 |
Neg logit contributions | |
and | -0.060 |
of | -0.052 |
urry | -0.051 |
Follow | -0.051 |
tum | -0.048 |
Token | with |
Token ID | 351 |
Feature activation | +8.30e-03 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
eling | +0.001 |
ps | +0.001 |
blew | +0.001 |
Neg logit contributions | |
and | -0.003 |
urry | -0.003 |
Follow | -0.003 |
Nice | -0.002 |
tum | -0.002 |
Token | lots |
Token ID | 6041 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
bur | +0.009 |
fourteen | +0.009 |
eling | +0.009 |
ps | +0.008 |
IT | +0.008 |
Neg logit contributions | |
and | -0.017 |
Follow | -0.015 |
urry | -0.015 |
tum | -0.014 |
Stay | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +2.03e-03 |
Pos logit contributions | |
and | +0.024 |
Follow | +0.022 |
urry | +0.021 |
rav | +0.020 |
School | +0.019 |
Neg logit contributions | |
ries | -0.044 |
ps | -0.044 |
fourteen | -0.044 |
her | -0.043 |
colored | -0.042 |
Token | food |
Token ID | 2057 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
ps | +0.002 |
eling | +0.002 |
colored | +0.002 |
blew | +0.002 |
her | +0.002 |
Neg logit contributions | |
and | -0.004 |
urry | -0.004 |
Follow | -0.004 |
aring | -0.003 |
apping | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +4.44e-03 |
Pos logit contributions | |
and | +0.044 |
urry | +0.039 |
Follow | +0.038 |
Nice | +0.035 |
tum | +0.035 |
Neg logit contributions | |
bur | -0.028 |
fourteen | -0.027 |
eling | -0.026 |
ps | -0.026 |
blew | -0.026 |
Token | Them |
Token ID | 26460 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
fourteen | +0.005 |
bur | +0.005 |
IT | +0.005 |
blew | +0.005 |
ible | +0.005 |
Neg logit contributions | |
and | -0.008 |
urry | -0.007 |
rav | -0.007 |
Follow | -0.007 |
vere | -0.007 |
Token | ouse |
Token ID | 1076 |
Feature activation | +2.53e-03 |
Pos logit contributions | |
and | +0.153 |
School | +0.132 |
vere | +0.131 |
Nice | +0.131 |
Follow | +0.130 |
Neg logit contributions | |
blew | -0.104 |
bur | -0.098 |
ck | -0.095 |
ps | -0.094 |
lled | -0.092 |
Token | 's |
Token ID | 338 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
bur | +0.004 |
fourteen | +0.004 |
eling | +0.003 |
ps | +0.003 |
blew | +0.003 |
Neg logit contributions | |
and | -0.004 |
urry | -0.004 |
Follow | -0.004 |
tum | -0.003 |
Nice | -0.003 |
Token | mouth |
Token ID | 5422 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
and | +0.040 |
urry | +0.034 |
Follow | +0.033 |
rav | +0.031 |
ooters | +0.030 |
Neg logit contributions | |
bur | -0.031 |
ps | -0.030 |
fourteen | -0.030 |
blew | -0.029 |
eling | -0.029 |
Token | watered |
Token ID | 48024 |
Feature activation | -8.26e-03 |
Pos logit contributions | |
and | +0.031 |
Follow | +0.027 |
urry | +0.027 |
rav | +0.024 |
School | +0.024 |
Neg logit contributions | |
bur | -0.025 |
fourteen | -0.025 |
blew | -0.025 |
ps | -0.025 |
eling | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -9.57e-03 |
Pos logit contributions | |
and | +0.015 |
urry | +0.014 |
Follow | +0.013 |
tum | +0.012 |
apping | +0.012 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.010 |
ps | -0.010 |
eling | -0.010 |
ible | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | -3.48e-03 |
Pos logit contributions | |
and | +0.055 |
urry | +0.049 |
Follow | +0.049 |
tum | +0.045 |
Nice | +0.045 |
Neg logit contributions | |
bur | -0.030 |
fourteen | -0.029 |
eling | -0.029 |
ps | -0.028 |
blew | -0.027 |
Token | few |
Token ID | 1178 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
and | +0.006 |
Follow | +0.005 |
urry | +0.005 |
Nice | +0.005 |
apping | +0.005 |
Neg logit contributions | |
eling | -0.004 |
blew | -0.004 |
bur | -0.004 |
sc | -0.004 |
cking | -0.004 |
Token | days |
Token ID | 1528 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
eling | +0.035 |
fourteen | +0.032 |
ps | +0.032 |
blew | +0.032 |
ck | +0.032 |
Neg logit contributions | |
and | -0.043 |
urry | -0.039 |
Follow | -0.038 |
Nice | -0.037 |
rav | -0.035 |
Token | when |
Token ID | 618 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
and | +0.106 |
Follow | +0.092 |
urry | +0.091 |
Nice | +0.084 |
tum | +0.083 |
Neg logit contributions | |
bur | -0.087 |
fourteen | -0.086 |
eling | -0.084 |
ps | -0.083 |
blew | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +4.12e-03 |
Pos logit contributions | |
bur | +0.029 |
fourteen | +0.027 |
ps | +0.027 |
eling | +0.027 |
ck | +0.026 |
Neg logit contributions | |
and | -0.043 |
urry | -0.038 |
Follow | -0.038 |
Nice | -0.036 |
tum | -0.035 |
Token | renovations |
Token ID | 44481 |
Feature activation | -8.85e-02 |
Pos logit contributions | |
bur | +0.006 |
blew | +0.006 |
ps | +0.005 |
eling | +0.005 |
fourteen | +0.005 |
Neg logit contributions | |
and | -0.007 |
urry | -0.006 |
Follow | -0.006 |
apping | -0.005 |
Nice | -0.005 |
Token | were |
Token ID | 547 |
Feature activation | +8.38e-03 |
Pos logit contributions | |
and | +0.134 |
urry | +0.113 |
Follow | +0.112 |
tum | +0.102 |
Nice | +0.101 |
Neg logit contributions | |
bur | -0.140 |
fourteen | -0.136 |
ps | -0.134 |
eling | -0.133 |
blew | -0.132 |
Token | done |
Token ID | 1760 |
Feature activation | -2.02e-03 |
Pos logit contributions | |
bur | +0.011 |
fourteen | +0.010 |
ps | +0.010 |
eling | +0.010 |
blew | +0.010 |
Neg logit contributions | |
and | -0.015 |
Follow | -0.014 |
urry | -0.014 |
Nice | -0.013 |
tum | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
and | +0.004 |
urry | +0.003 |
Follow | +0.003 |
tum | +0.003 |
Nice | +0.003 |
Neg logit contributions | |
bur | -0.003 |
fourteen | -0.002 |
eling | -0.002 |
ps | -0.002 |
blew | -0.002 |
Token | Everyone |
Token ID | 11075 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
bur | +0.021 |
eling | +0.020 |
fourteen | +0.020 |
ps | +0.020 |
ck | +0.019 |
Neg logit contributions | |
and | -0.015 |
Follow | -0.013 |
urry | -0.013 |
Nice | -0.011 |
tum | -0.011 |
Token | could |
Token ID | 714 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
bur | +0.062 |
fourteen | +0.060 |
ps | +0.056 |
eling | +0.056 |
IT | +0.055 |
Neg logit contributions | |
and | -0.063 |
Follow | -0.054 |
urry | -0.053 |
of | -0.050 |
apping | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -7.69e-03 |
Pos logit contributions | |
and | +0.068 |
urry | +0.060 |
Follow | +0.059 |
story | +0.057 |
Today | +0.056 |
Neg logit contributions | |
bur | -0.034 |
fourteen | -0.033 |
eling | -0.032 |
ps | -0.031 |
blew | -0.031 |
Token | few |
Token ID | 1178 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
and | +0.013 |
Follow | +0.012 |
urry | +0.011 |
Nice | +0.011 |
rav | +0.011 |
Neg logit contributions | |
blew | -0.010 |
bur | -0.010 |
eling | -0.010 |
sc | -0.009 |
afar | -0.009 |
Token | days |
Token ID | 1528 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
eling | +0.033 |
fourteen | +0.032 |
ps | +0.032 |
bur | +0.032 |
ck | +0.032 |
Neg logit contributions | |
and | -0.036 |
urry | -0.032 |
Follow | -0.031 |
Nice | -0.029 |
rav | -0.028 |
Token | when |
Token ID | 618 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
and | +0.118 |
Follow | +0.102 |
urry | +0.102 |
Nice | +0.094 |
tum | +0.093 |
Neg logit contributions | |
bur | -0.090 |
fourteen | -0.088 |
eling | -0.086 |
ps | -0.085 |
blew | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | +2.59e-03 |
Pos logit contributions | |
bur | +0.029 |
ps | +0.027 |
fourteen | +0.027 |
eling | +0.027 |
ck | +0.026 |
Neg logit contributions | |
and | -0.041 |
urry | -0.036 |
Follow | -0.035 |
Nice | -0.033 |
tum | -0.033 |
Token | renovations |
Token ID | 44481 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
bur | +0.004 |
fourteen | +0.004 |
blew | +0.003 |
ps | +0.003 |
eling | +0.003 |
Neg logit contributions | |
and | -0.005 |
urry | -0.004 |
Follow | -0.004 |
Nice | -0.003 |
apping | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | +8.50e-03 |
Pos logit contributions | |
and | +0.136 |
urry | +0.116 |
Follow | +0.115 |
tum | +0.105 |
Nice | +0.103 |
Neg logit contributions | |
bur | -0.136 |
fourteen | -0.133 |
ps | -0.131 |
eling | -0.129 |
blew | -0.128 |
Token | done |
Token ID | 1760 |
Feature activation | -1.13e-03 |
Pos logit contributions | |
bur | +0.012 |
fourteen | +0.011 |
eling | +0.011 |
ps | +0.011 |
blew | +0.011 |
Neg logit contributions | |
and | -0.015 |
urry | -0.013 |
Follow | -0.013 |
tum | -0.012 |
Nice | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +3.31e-03 |
Pos logit contributions | |
and | +0.002 |
urry | +0.002 |
Follow | +0.002 |
tum | +0.002 |
Nice | +0.002 |
Neg logit contributions | |
bur | -0.001 |
fourteen | -0.001 |
eling | -0.001 |
ps | -0.001 |
blew | -0.001 |
Token | Everyone |
Token ID | 11075 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
bur | +0.006 |
eling | +0.005 |
ps | +0.005 |
fourteen | +0.005 |
ck | +0.005 |
Neg logit contributions | |
and | -0.005 |
urry | -0.004 |
Follow | -0.004 |
Nice | -0.004 |
tum | -0.004 |
Token | could |
Token ID | 714 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
bur | +0.065 |
fourteen | +0.063 |
ps | +0.059 |
IT | +0.059 |
eling | +0.059 |
Neg logit contributions | |
and | -0.066 |
Follow | -0.056 |
urry | -0.055 |
of | -0.052 |
apping | -0.050 |
Token | olly |
Token ID | 5098 |
Feature activation | -6.37e-03 |
Pos logit contributions | |
bur | +0.023 |
fourteen | +0.023 |
eling | +0.022 |
ps | +0.022 |
blew | +0.022 |
Neg logit contributions | |
and | -0.018 |
urry | -0.015 |
Follow | -0.015 |
Nice | -0.013 |
tum | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +4.08e-03 |
Pos logit contributions | |
and | +0.011 |
urry | +0.009 |
Follow | +0.009 |
Nice | +0.008 |
tum | +0.008 |
Neg logit contributions | |
bur | -0.009 |
fourteen | -0.009 |
eling | -0.009 |
ps | -0.009 |
blew | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.006 |
ps | +0.006 |
ible | +0.005 |
Neg logit contributions | |
and | -0.007 |
urry | -0.006 |
Follow | -0.006 |
tum | -0.005 |
apping | -0.005 |
Token | three |
Token ID | 1115 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
and | +0.018 |
urry | +0.016 |
Follow | +0.016 |
tum | +0.014 |
Nice | +0.014 |
Neg logit contributions | |
bur | -0.021 |
fourteen | -0.020 |
ps | -0.020 |
eling | -0.020 |
lled | -0.020 |
Token | - |
Token ID | 12 |
Feature activation | -5.10e-03 |
Pos logit contributions | |
and | +0.035 |
Follow | +0.030 |
urry | +0.030 |
Nice | +0.027 |
rav | +0.026 |
Neg logit contributions | |
eling | -0.045 |
bur | -0.044 |
ps | -0.043 |
fourteen | -0.043 |
ck | -0.043 |
Token | year |
Token ID | 1941 |
Feature activation | -8.74e-02 |
Pos logit contributions | |
and | +0.005 |
urry | +0.004 |
Follow | +0.004 |
tum | +0.003 |
Nice | +0.003 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.010 |
eling | -0.010 |
ps | -0.010 |
blew | -0.010 |
Token | old |
Token ID | 1468 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
and | +0.156 |
Follow | +0.144 |
Nice | +0.138 |
urry | +0.138 |
School | +0.133 |
Neg logit contributions | |
ps | -0.102 |
ck | -0.102 |
bur | -0.101 |
fourteen | -0.099 |
eling | -0.097 |
Token | girl |
Token ID | 2576 |
Feature activation | -9.91e-03 |
Pos logit contributions | |
and | +0.024 |
urry | +0.020 |
Follow | +0.020 |
Nice | +0.018 |
tum | +0.018 |
Neg logit contributions | |
bur | -0.023 |
fourteen | -0.023 |
eling | -0.022 |
ps | -0.022 |
blew | -0.022 |
Token | who |
Token ID | 508 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
and | +0.019 |
urry | +0.017 |
Follow | +0.017 |
Nice | +0.016 |
rav | +0.015 |
Neg logit contributions | |
bur | -0.011 |
fourteen | -0.011 |
blew | -0.011 |
eling | -0.011 |
ps | -0.010 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.95e-04 |
Pos logit contributions | |
and | +0.015 |
Follow | +0.013 |
urry | +0.013 |
School | +0.011 |
Stay | +0.011 |
Neg logit contributions | |
bur | -0.022 |
fourteen | -0.022 |
ps | -0.021 |
cking | -0.020 |
eling | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
and | +0.001 |
urry | +0.001 |
Follow | +0.001 |
tum | +0.001 |
Nice | +0.001 |
Neg logit contributions | |
bur | -0.001 |
fourteen | -0.001 |
eling | -0.001 |
ps | -0.001 |
ck | -0.001 |
Token | oy |
Token ID | 726 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
and | +0.016 |
Follow | +0.014 |
urry | +0.013 |
Nice | +0.012 |
Today | +0.012 |
Neg logit contributions | |
fourteen | -0.017 |
bur | -0.016 |
eling | -0.016 |
ck | -0.016 |
ps | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
and | +0.022 |
Follow | +0.019 |
urry | +0.018 |
tum | +0.017 |
Nice | +0.016 |
Neg logit contributions | |
bur | -0.019 |
fourteen | -0.018 |
ps | -0.018 |
ible | -0.017 |
ami | -0.017 |
Token | mate |
Token ID | 16133 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
and | +0.019 |
urry | +0.016 |
Follow | +0.016 |
apping | +0.015 |
vere | +0.014 |
Neg logit contributions | |
fourteen | -0.017 |
blew | -0.016 |
ps | -0.016 |
bur | -0.016 |
ck | -0.015 |
Token | y |
Token ID | 88 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
and | +0.044 |
urry | +0.038 |
Follow | +0.037 |
Nice | +0.034 |
tum | +0.034 |
Neg logit contributions | |
bur | -0.038 |
fourteen | -0.037 |
eling | -0.037 |
ps | -0.036 |
blew | -0.035 |
Token | ! |
Token ID | 0 |
Feature activation | -9.61e-03 |
Pos logit contributions | |
and | +0.026 |
urry | +0.023 |
Follow | +0.023 |
tum | +0.021 |
Nice | +0.021 |
Neg logit contributions | |
fourteen | -0.015 |
bur | -0.015 |
ps | -0.015 |
blew | -0.014 |
ible | -0.014 |
Token | Land |
Token ID | 6379 |
Feature activation | -8.73e-02 |
Pos logit contributions | |
and | +0.013 |
urry | +0.009 |
tum | +0.009 |
vere | +0.009 |
story | +0.009 |
Neg logit contributions | |
fourteen | -0.017 |
blew | -0.016 |
bur | -0.016 |
ps | -0.015 |
IT | -0.015 |
Token | ho |
Token ID | 8169 |
Feature activation | -5.47e-02 |
Pos logit contributions | |
and | +0.144 |
tum | +0.126 |
urry | +0.124 |
Follow | +0.123 |
vere | +0.110 |
Neg logit contributions | |
ps | -0.121 |
fourteen | -0.119 |
blew | -0.113 |
lled | -0.109 |
bur | -0.108 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.17e-03 |
Pos logit contributions | |
and | +0.089 |
Follow | +0.078 |
tum | +0.075 |
urry | +0.071 |
Nice | +0.070 |
Neg logit contributions | |
fourteen | -0.084 |
bur | -0.083 |
ps | -0.082 |
ck | -0.076 |
blew | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | +1.40e-03 |
Pos logit contributions | |
and | +0.002 |
urry | +0.001 |
Follow | +0.001 |
tum | +0.001 |
Nice | +0.001 |
Neg logit contributions | |
bur | -0.002 |
fourteen | -0.002 |
eling | -0.002 |
ps | -0.002 |
blew | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
fourteen | +0.002 |
bur | +0.002 |
ck | +0.002 |
blew | +0.002 |
ps | +0.002 |
Neg logit contributions | |
and | -0.002 |
Follow | -0.002 |
urry | -0.002 |
Nice | -0.002 |
School | -0.002 |
Token | But |
Token ID | 1537 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
bur | +0.002 |
fourteen | +0.002 |
eling | +0.002 |
ps | +0.002 |
ible | +0.002 |
Neg logit contributions | |
and | -0.004 |
urry | -0.003 |
Follow | -0.003 |
tum | -0.003 |
Nice | -0.003 |
Token | ?" |
Token ID | 1701 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
and | +0.000 |
urry | +0.000 |
Follow | +0.000 |
tum | +0.000 |
Nice | +0.000 |
Neg logit contributions | |
bur | -0.000 |
fourteen | -0.000 |
eling | -0.000 |
ps | -0.000 |
blew | -0.000 |
Token | asked |
Token ID | 1965 |
Feature activation | +4.51e-03 |
Pos logit contributions | |
ps | +0.069 |
fourteen | +0.069 |
bur | +0.069 |
lled | +0.065 |
ible | +0.064 |
Neg logit contributions | |
and | -0.078 |
Follow | -0.074 |
tum | -0.073 |
urry | -0.072 |
of | -0.071 |
Token | a |
Token ID | 257 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
fourteen | +0.007 |
eling | +0.007 |
cking | +0.007 |
bur | +0.006 |
ps | +0.006 |
Neg logit contributions | |
and | -0.007 |
of | -0.006 |
urry | -0.006 |
Stay | -0.006 |
tum | -0.006 |
Token | three |
Token ID | 1115 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
and | +0.013 |
urry | +0.011 |
Follow | +0.011 |
Go | +0.011 |
Nice | +0.010 |
Neg logit contributions | |
blew | -0.009 |
bur | -0.008 |
comb | -0.008 |
fourteen | -0.008 |
pressed | -0.008 |
Token | - |
Token ID | 12 |
Feature activation | +3.32e-03 |
Pos logit contributions | |
and | +0.027 |
Follow | +0.023 |
urry | +0.023 |
Nice | +0.020 |
Stay | +0.019 |
Neg logit contributions | |
bur | -0.036 |
fourteen | -0.036 |
eling | -0.036 |
ps | -0.035 |
ck | -0.035 |
Token | year |
Token ID | 1941 |
Feature activation | -8.69e-02 |
Pos logit contributions | |
bur | +0.007 |
eling | +0.007 |
fourteen | +0.007 |
ps | +0.007 |
blew | +0.007 |
Neg logit contributions | |
and | -0.003 |
urry | -0.002 |
Follow | -0.002 |
of | -0.002 |
Nice | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
and | +0.159 |
Follow | +0.145 |
Nice | +0.137 |
urry | +0.137 |
School | +0.132 |
Neg logit contributions | |
ck | -0.103 |
ps | -0.100 |
bur | -0.099 |
fourteen | -0.094 |
eling | -0.094 |
Token | old |
Token ID | 727 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.006 |
ps | +0.006 |
ck | +0.006 |
Neg logit contributions | |
and | -0.004 |
urry | -0.003 |
Follow | -0.003 |
tum | -0.003 |
Nice | -0.003 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
bur | +0.059 |
fourteen | +0.058 |
eling | +0.056 |
ps | +0.056 |
blew | +0.055 |
Neg logit contributions | |
and | -0.061 |
urry | -0.052 |
Follow | -0.052 |
tum | -0.046 |
Nice | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +2.03e-03 |
Pos logit contributions | |
and | +0.025 |
urry | +0.022 |
Follow | +0.022 |
Nice | +0.021 |
tum | +0.021 |
Neg logit contributions | |
bur | -0.014 |
eling | -0.013 |
fourteen | -0.013 |
ps | -0.013 |
ck | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.17e-03 |
Pos logit contributions | |
bur | +0.003 |
fourteen | +0.003 |
blew | +0.003 |
ps | +0.003 |
eling | +0.003 |
Neg logit contributions | |
and | -0.003 |
urry | -0.003 |
Follow | -0.003 |
tum | -0.002 |
vere | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
and | +0.045 |
Follow | +0.037 |
vere | +0.035 |
urry | +0.035 |
Go | +0.033 |
Neg logit contributions | |
blew | -0.050 |
bur | -0.050 |
eling | -0.049 |
ck | -0.048 |
fourteen | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -7.32e-03 |
Pos logit contributions | |
bur | +0.025 |
fourteen | +0.024 |
eling | +0.024 |
ps | +0.024 |
blew | +0.024 |
Neg logit contributions | |
and | -0.016 |
urry | -0.013 |
Follow | -0.013 |
tum | -0.011 |
Nice | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -7.40e-03 |
Pos logit contributions | |
and | +0.012 |
urry | +0.011 |
Follow | +0.011 |
Nice | +0.010 |
of | +0.010 |
Neg logit contributions | |
ps | -0.010 |
eling | -0.010 |
cking | -0.010 |
bur | -0.010 |
fourteen | -0.010 |
Token | three |
Token ID | 1115 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
and | +0.015 |
Follow | +0.012 |
urry | +0.012 |
Go | +0.012 |
Today | +0.011 |
Neg logit contributions | |
blew | -0.009 |
bur | -0.008 |
pin | -0.008 |
sc | -0.008 |
abi | -0.008 |
Token | - |
Token ID | 12 |
Feature activation | -4.95e-04 |
Pos logit contributions | |
and | +0.027 |
Follow | +0.022 |
urry | +0.021 |
Nice | +0.019 |
rav | +0.018 |
Neg logit contributions | |
eling | -0.040 |
ck | -0.039 |
blew | -0.039 |
bur | -0.039 |
fourteen | -0.039 |
Token | year |
Token ID | 1941 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
and | +0.001 |
urry | +0.000 |
Follow | +0.000 |
tum | +0.000 |
Nice | +0.000 |
Neg logit contributions | |
bur | -0.001 |
fourteen | -0.001 |
eling | -0.001 |
ps | -0.001 |
blew | -0.001 |
Token | - |
Token ID | 12 |
Feature activation | -1.02e-03 |
Pos logit contributions | |
and | +0.155 |
Follow | +0.138 |
Nice | +0.131 |
urry | +0.130 |
School | +0.127 |
Neg logit contributions | |
ck | -0.107 |
ps | -0.106 |
bur | -0.105 |
fourteen | -0.105 |
eling | -0.100 |
Token | old |
Token ID | 727 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
and | +0.001 |
Follow | +0.001 |
urry | +0.001 |
tum | +0.001 |
Nice | +0.001 |
Neg logit contributions | |
bur | -0.002 |
fourteen | -0.002 |
ck | -0.002 |
eling | -0.002 |
ps | -0.002 |
Token | boy |
Token ID | 2933 |
Feature activation | -5.40e-03 |
Pos logit contributions | |
bur | +0.062 |
fourteen | +0.061 |
eling | +0.060 |
ps | +0.059 |
blew | +0.058 |
Neg logit contributions | |
and | -0.061 |
urry | -0.052 |
Follow | -0.052 |
Nice | -0.046 |
tum | -0.046 |
Token | decided |
Token ID | 3066 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
and | +0.009 |
urry | +0.008 |
Follow | +0.008 |
tum | +0.007 |
Nice | +0.007 |
Neg logit contributions | |
bur | -0.007 |
fourteen | -0.007 |
eling | -0.007 |
ps | -0.007 |
ck | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
and | +0.022 |
urry | +0.018 |
Follow | +0.018 |
Nice | +0.016 |
tum | +0.016 |
Neg logit contributions | |
bur | -0.031 |
fourteen | -0.030 |
eling | -0.030 |
ps | -0.030 |
ible | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | +6.87e-03 |
Pos logit contributions | |
and | +0.019 |
urry | +0.016 |
Follow | +0.016 |
of | +0.015 |
tum | +0.015 |
Neg logit contributions | |
fourteen | -0.024 |
ps | -0.024 |
cking | -0.023 |
bur | -0.023 |
eling | -0.023 |
Token | an |
Token ID | 281 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
bur | +0.010 |
fourteen | +0.010 |
eling | +0.010 |
ps | +0.010 |
ible | +0.009 |
Neg logit contributions | |
and | -0.011 |
urry | -0.010 |
Follow | -0.010 |
tum | -0.009 |
apping | -0.009 |
Token | eager |
Token ID | 11069 |
Feature activation | -6.22e-03 |
Pos logit contributions | |
and | +0.046 |
Follow | +0.040 |
urry | +0.039 |
Nice | +0.036 |
rav | +0.036 |
Neg logit contributions | |
bur | -0.039 |
eling | -0.037 |
fourteen | -0.037 |
ps | -0.036 |
ck | -0.036 |
Token | 3 |
Token ID | 513 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
and | +0.011 |
Follow | +0.010 |
vere | +0.009 |
urry | +0.009 |
adows | +0.008 |
Neg logit contributions | |
bur | -0.008 |
blew | -0.008 |
her | -0.007 |
ami | -0.007 |
comb | -0.007 |
Token | - |
Token ID | 12 |
Feature activation | +3.87e-03 |
Pos logit contributions | |
and | +0.049 |
Follow | +0.041 |
urry | +0.040 |
Nice | +0.035 |
tum | +0.034 |
Neg logit contributions | |
bur | -0.082 |
eling | -0.080 |
fourteen | -0.080 |
ps | -0.079 |
blew | -0.078 |
Token | year |
Token ID | 1941 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
bur | +0.008 |
fourteen | +0.007 |
eling | +0.007 |
ck | +0.007 |
ps | +0.007 |
Neg logit contributions | |
and | -0.004 |
Follow | -0.003 |
urry | -0.003 |
tum | -0.003 |
Nice | -0.003 |
Token | - |
Token ID | 12 |
Feature activation | +3.91e-03 |
Pos logit contributions | |
and | +0.158 |
Follow | +0.142 |
urry | +0.136 |
Nice | +0.134 |
Today | +0.128 |
Neg logit contributions | |
bur | -0.100 |
ps | -0.099 |
ck | -0.099 |
eling | -0.097 |
fourteen | -0.096 |
Token | old |
Token ID | 727 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
bur | +0.007 |
eling | +0.007 |
fourteen | +0.007 |
ck | +0.007 |
ps | +0.007 |
Neg logit contributions | |
and | -0.004 |
Follow | -0.004 |
urry | -0.004 |
Nice | -0.003 |
tum | -0.003 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
bur | +0.053 |
fourteen | +0.053 |
eling | +0.051 |
ps | +0.051 |
blew | +0.049 |
Neg logit contributions | |
and | -0.067 |
urry | -0.057 |
Follow | -0.057 |
Nice | -0.052 |
tum | -0.052 |
Token | who |
Token ID | 508 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
and | +0.018 |
urry | +0.016 |
Follow | +0.016 |
apping | +0.014 |
Nice | +0.014 |
Neg logit contributions | |
bur | -0.013 |
fourteen | -0.013 |
eling | -0.013 |
ps | -0.013 |
lled | -0.012 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.32e-04 |
Pos logit contributions | |
and | +0.018 |
Follow | +0.016 |
urry | +0.016 |
School | +0.014 |
Stay | +0.014 |
Neg logit contributions | |
fourteen | -0.029 |
bur | -0.028 |
cking | -0.028 |
ps | -0.028 |
ck | -0.027 |
Token | sign |
Token ID | 1051 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
blew | +0.016 |
bur | +0.016 |
fourteen | +0.015 |
ible | +0.015 |
ries | +0.015 |
Neg logit contributions | |
and | -0.027 |
Follow | -0.021 |
urry | -0.021 |
vere | -0.019 |
anny | -0.019 |
Token | that |
Token ID | 326 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
bur | +0.017 |
fourteen | +0.016 |
eling | +0.016 |
ps | +0.016 |
ible | +0.015 |
Neg logit contributions | |
and | -0.033 |
urry | -0.030 |
Follow | -0.030 |
tum | -0.028 |
Nice | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
and | +0.026 |
Follow | +0.024 |
urry | +0.023 |
tum | +0.023 |
story | +0.022 |
Neg logit contributions | |
fourteen | -0.020 |
bur | -0.019 |
eling | -0.019 |
ps | -0.018 |
ible | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
and | +0.019 |
Follow | +0.016 |
urry | +0.016 |
ooters | +0.014 |
suits | +0.014 |
Neg logit contributions | |
fourteen | -0.016 |
blew | -0.015 |
ps | -0.015 |
her | -0.014 |
bur | -0.014 |
Token | F |
Token ID | 37 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
and | +0.019 |
urry | +0.016 |
Follow | +0.016 |
Nice | +0.015 |
tum | +0.015 |
Neg logit contributions | |
bur | -0.014 |
fourteen | -0.014 |
eling | -0.014 |
ps | -0.014 |
blew | -0.013 |
Token | estival |
Token ID | 6743 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
and | +0.107 |
Follow | +0.088 |
vere | +0.087 |
tum | +0.086 |
Nice | +0.085 |
Neg logit contributions | |
bur | -0.105 |
fourteen | -0.103 |
ck | -0.102 |
IT | -0.100 |
ps | -0.099 |
Token | Today |
Token ID | 6288 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
vere | +0.112 |
villains | +0.103 |
and | +0.097 |
Follow | +0.096 |
tum | +0.092 |
Neg logit contributions | |
fourteen | -0.134 |
IT | -0.132 |
bur | -0.126 |
et | -0.116 |
least | -0.115 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.94e-03 |
Pos logit contributions | |
and | +0.075 |
Follow | +0.066 |
urry | +0.062 |
ooters | +0.060 |
rav | +0.058 |
Neg logit contributions | |
fourteen | -0.044 |
blew | -0.042 |
ps | -0.039 |
IT | -0.038 |
bur | -0.038 |
Token | She |
Token ID | 1375 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
bur | +0.005 |
fourteen | +0.005 |
eling | +0.005 |
ps | +0.005 |
ck | +0.005 |
Neg logit contributions | |
and | -0.004 |
urry | -0.004 |
Follow | -0.003 |
tum | -0.003 |
Nice | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | +6.76e-03 |
Pos logit contributions | |
bur | +0.015 |
fourteen | +0.015 |
eling | +0.015 |
ps | +0.015 |
cking | +0.014 |
Neg logit contributions | |
and | -0.021 |
urry | -0.019 |
Follow | -0.019 |
tum | -0.018 |
Stay | -0.017 |
Token | very |
Token ID | 845 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
bur | +0.010 |
ps | +0.010 |
fourteen | +0.010 |
eling | +0.010 |
blew | +0.010 |
Neg logit contributions | |
and | -0.011 |
urry | -0.009 |
Follow | -0.009 |
Nice | -0.008 |
rav | -0.008 |
Token | my |
Token ID | 1820 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
fourteen | +0.010 |
cking | +0.010 |
ck | +0.010 |
eling | +0.010 |
ps | +0.009 |
Neg logit contributions | |
and | -0.009 |
urry | -0.008 |
Follow | -0.007 |
Today | -0.007 |
tum | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
bur | +0.024 |
fourteen | +0.024 |
ps | +0.023 |
eling | +0.023 |
ck | +0.022 |
Neg logit contributions | |
and | -0.023 |
urry | -0.020 |
Follow | -0.020 |
tum | -0.018 |
Stay | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -8.36e-03 |
Pos logit contributions | |
bur | +0.008 |
fourteen | +0.008 |
eling | +0.008 |
ible | +0.008 |
ps | +0.008 |
Neg logit contributions | |
and | -0.009 |
urry | -0.008 |
Follow | -0.007 |
tum | -0.007 |
apping | -0.007 |
Token | three |
Token ID | 1115 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
and | +0.012 |
urry | +0.010 |
Follow | +0.010 |
tum | +0.009 |
Nice | +0.009 |
Neg logit contributions | |
bur | -0.014 |
fourteen | -0.014 |
ps | -0.013 |
eling | -0.013 |
lled | -0.013 |
Token | - |
Token ID | 12 |
Feature activation | +3.40e-03 |
Pos logit contributions | |
and | +0.030 |
urry | +0.025 |
Follow | +0.025 |
Nice | +0.021 |
tum | +0.021 |
Neg logit contributions | |
bur | -0.041 |
fourteen | -0.040 |
eling | -0.040 |
ps | -0.039 |
blew | -0.039 |
Token | year |
Token ID | 1941 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
bur | +0.007 |
fourteen | +0.007 |
eling | +0.007 |
ps | +0.007 |
ck | +0.007 |
Neg logit contributions | |
and | -0.004 |
urry | -0.003 |
Follow | -0.003 |
tum | -0.003 |
Nice | -0.003 |
Token | - |
Token ID | 12 |
Feature activation | +3.46e-03 |
Pos logit contributions | |
and | +0.164 |
Follow | +0.149 |
Nice | +0.145 |
urry | +0.141 |
Today | +0.137 |
Neg logit contributions | |
ck | -0.090 |
ps | -0.090 |
eling | -0.088 |
bur | -0.087 |
et | -0.086 |
Token | old |
Token ID | 727 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
bur | +0.006 |
ck | +0.006 |
eling | +0.006 |
fourteen | +0.006 |
cking | +0.006 |
Neg logit contributions | |
and | -0.004 |
Follow | -0.004 |
urry | -0.004 |
Nice | -0.003 |
tum | -0.003 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
bur | +0.055 |
fourteen | +0.054 |
eling | +0.053 |
ps | +0.053 |
blew | +0.051 |
Neg logit contributions | |
and | -0.063 |
urry | -0.054 |
Follow | -0.054 |
tum | -0.049 |
Nice | -0.049 |
Token | who |
Token ID | 508 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
and | +0.022 |
urry | +0.019 |
Follow | +0.019 |
tum | +0.018 |
Nice | +0.018 |
Neg logit contributions | |
bur | -0.015 |
fourteen | -0.014 |
eling | -0.014 |
ps | -0.014 |
lled | -0.013 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.09e-05 |
Pos logit contributions | |
and | +0.022 |
urry | +0.019 |
Follow | +0.018 |
School | +0.017 |
Stay | +0.017 |
Neg logit contributions | |
fourteen | -0.033 |
bur | -0.032 |
ps | -0.031 |
cking | -0.031 |
ries | -0.031 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.44e-03 |
Pos logit contributions | |
and | +0.017 |
Follow | +0.015 |
urry | +0.014 |
Nice | +0.013 |
tum | +0.013 |
Neg logit contributions | |
bur | -0.016 |
fourteen | -0.016 |
blew | -0.016 |
eling | -0.015 |
ps | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
fourteen | +0.002 |
eling | +0.002 |
ps | +0.002 |
bur | +0.002 |
cking | +0.002 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
Follow | -0.002 |
of | -0.002 |
story | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | +7.34e-03 |
Pos logit contributions | |
and | +0.004 |
urry | +0.004 |
Follow | +0.004 |
birthday | +0.004 |
tum | +0.004 |
Neg logit contributions | |
fourteen | -0.002 |
ps | -0.002 |
eling | -0.002 |
bur | -0.002 |
ible | -0.002 |
Token | 3 |
Token ID | 513 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
bur | +0.012 |
fourteen | +0.012 |
eling | +0.012 |
ps | +0.011 |
blew | +0.011 |
Neg logit contributions | |
and | -0.011 |
urry | -0.009 |
Follow | -0.009 |
tum | -0.008 |
Nice | -0.008 |
Token | - |
Token ID | 12 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
and | +0.050 |
Follow | +0.041 |
urry | +0.040 |
Nice | +0.036 |
vere | +0.033 |
Neg logit contributions | |
bur | -0.088 |
eling | -0.087 |
ps | -0.086 |
fourteen | -0.085 |
ck | -0.085 |
Token | year |
Token ID | 1941 |
Feature activation | -8.62e-02 |
Pos logit contributions | |
and | +0.002 |
Follow | +0.002 |
urry | +0.002 |
Nice | +0.002 |
tum | +0.002 |
Neg logit contributions | |
bur | -0.004 |
fourteen | -0.004 |
ps | -0.004 |
eling | -0.004 |
ck | -0.004 |
Token | - |
Token ID | 12 |
Feature activation | -2.54e-03 |
Pos logit contributions | |
and | +0.156 |
Follow | +0.143 |
Nice | +0.137 |
urry | +0.135 |
Today | +0.130 |
Neg logit contributions | |
ps | -0.104 |
ck | -0.102 |
bur | -0.102 |
fourteen | -0.098 |
eling | -0.096 |
Token | old |
Token ID | 727 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
and | +0.003 |
Follow | +0.002 |
urry | +0.002 |
Nice | +0.002 |
tum | +0.002 |
Neg logit contributions | |
bur | -0.005 |
fourteen | -0.005 |
eling | -0.005 |
ps | -0.005 |
ck | -0.005 |
Token | daughter |
Token ID | 4957 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
bur | +0.059 |
fourteen | +0.058 |
eling | +0.057 |
ps | +0.056 |
blew | +0.055 |
Neg logit contributions | |
and | -0.057 |
urry | -0.049 |
Follow | -0.048 |
Nice | -0.043 |
tum | -0.043 |
Token | decided |
Token ID | 3066 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
and | +0.029 |
urry | +0.025 |
Follow | +0.025 |
Nice | +0.023 |
vere | +0.022 |
Neg logit contributions | |
bur | -0.022 |
blew | -0.022 |
fourteen | -0.022 |
ible | -0.021 |
ps | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
and | +0.023 |
urry | +0.019 |
Follow | +0.019 |
tum | +0.017 |
Nice | +0.017 |
Neg logit contributions | |
bur | -0.033 |
fourteen | -0.032 |
eling | -0.032 |
ps | -0.031 |
ible | -0.031 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
and | +0.030 |
urry | +0.025 |
Follow | +0.025 |
tum | +0.021 |
Nice | +0.021 |
Neg logit contributions | |
bur | -0.043 |
fourteen | -0.042 |
eling | -0.041 |
ps | -0.041 |
blew | -0.040 |
Token | shouted |
Token ID | 17293 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
bur | +0.044 |
fourteen | +0.044 |
ps | +0.044 |
eling | +0.043 |
lled | +0.042 |
Neg logit contributions | |
and | -0.042 |
urry | -0.038 |
Follow | -0.038 |
tum | -0.038 |
of | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | +2.25e-03 |
Pos logit contributions | |
and | +0.026 |
urry | +0.024 |
tum | +0.024 |
Follow | +0.023 |
Stay | +0.021 |
Neg logit contributions | |
eling | -0.031 |
bur | -0.031 |
fourteen | -0.030 |
cking | -0.029 |
lled | -0.029 |
Token | three |
Token ID | 1115 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
bur | +0.003 |
fourteen | +0.003 |
eling | +0.003 |
ps | +0.003 |
ck | +0.003 |
Neg logit contributions | |
and | -0.004 |
urry | -0.003 |
Follow | -0.003 |
tum | -0.003 |
Nice | -0.003 |
Token | - |
Token ID | 12 |
Feature activation | -1.66e-03 |
Pos logit contributions | |
and | +0.034 |
urry | +0.029 |
Follow | +0.029 |
Nice | +0.026 |
tum | +0.025 |
Neg logit contributions | |
bur | -0.036 |
fourteen | -0.036 |
eling | -0.035 |
ps | -0.035 |
blew | -0.034 |
Token | year |
Token ID | 1941 |
Feature activation | -8.62e-02 |
Pos logit contributions | |
and | +0.002 |
urry | +0.001 |
Follow | +0.001 |
of | +0.001 |
Go | +0.001 |
Neg logit contributions | |
bur | -0.004 |
eling | -0.004 |
fourteen | -0.004 |
ps | -0.004 |
ible | -0.003 |
Token | - |
Token ID | 12 |
Feature activation | -1.64e-03 |
Pos logit contributions | |
and | +0.143 |
Follow | +0.126 |
urry | +0.123 |
Nice | +0.116 |
Stay | +0.112 |
Neg logit contributions | |
bur | -0.119 |
fourteen | -0.118 |
ps | -0.117 |
eling | -0.115 |
ck | -0.114 |
Token | old |
Token ID | 727 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
and | +0.002 |
urry | +0.002 |
Follow | +0.002 |
Nice | +0.001 |
tum | +0.001 |
Neg logit contributions | |
bur | -0.003 |
fourteen | -0.003 |
eling | -0.003 |
ps | -0.003 |
ck | -0.003 |
Token | son |
Token ID | 3367 |
Feature activation | -6.95e-03 |
Pos logit contributions | |
bur | +0.056 |
fourteen | +0.054 |
eling | +0.053 |
ps | +0.053 |
blew | +0.052 |
Neg logit contributions | |
and | -0.062 |
urry | -0.054 |
Follow | -0.053 |
Nice | -0.048 |
tum | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
and | +0.013 |
urry | +0.011 |
Follow | +0.011 |
Nice | +0.010 |
tum | +0.010 |
Neg logit contributions | |
bur | -0.009 |
fourteen | -0.008 |
eling | -0.008 |
ps | -0.008 |
blew | -0.008 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
and | +0.014 |
Follow | +0.012 |
urry | +0.012 |
Stay | +0.012 |
tum | +0.012 |
Neg logit contributions | |
bur | -0.019 |
lled | -0.018 |
ible | -0.018 |
fourteen | -0.018 |
rob | -0.018 |
Token | cube |
Token ID | 23441 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
and | +0.000 |
urry | +0.000 |
Follow | +0.000 |
Nice | +0.000 |
tum | +0.000 |
Neg logit contributions | |
bur | -0.000 |
fourteen | -0.000 |
eling | -0.000 |
ps | -0.000 |
blew | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | -5.62e-03 |
Pos logit contributions | |
fourteen | +0.073 |
eling | +0.073 |
bur | +0.072 |
ps | +0.070 |
IT | +0.070 |
Neg logit contributions | |
and | -0.094 |
Follow | -0.082 |
urry | -0.082 |
tum | -0.081 |
story | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | +2.41e-04 |
Pos logit contributions | |
and | +0.011 |
Follow | +0.010 |
of | +0.009 |
tum | +0.009 |
urry | +0.009 |
Neg logit contributions | |
bur | -0.007 |
fourteen | -0.007 |
ps | -0.006 |
eling | -0.006 |
ible | -0.006 |
Token | 3 |
Token ID | 513 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
bur | +0.000 |
fourteen | +0.000 |
eling | +0.000 |
ps | +0.000 |
blew | +0.000 |
Neg logit contributions | |
and | -0.000 |
urry | -0.000 |
Follow | -0.000 |
Nice | -0.000 |
tum | -0.000 |
Token | - |
Token ID | 12 |
Feature activation | +4.10e-03 |
Pos logit contributions | |
and | +0.044 |
urry | +0.034 |
Follow | +0.033 |
Nice | +0.028 |
tum | +0.027 |
Neg logit contributions | |
bur | -0.092 |
fourteen | -0.091 |
eling | -0.090 |
ps | -0.089 |
blew | -0.088 |
Token | year |
Token ID | 1941 |
Feature activation | -8.61e-02 |
Pos logit contributions | |
bur | +0.008 |
fourteen | +0.008 |
eling | +0.008 |
ps | +0.008 |
ck | +0.008 |
Neg logit contributions | |
and | -0.005 |
urry | -0.004 |
Follow | -0.004 |
Nice | -0.003 |
tum | -0.003 |
Token | - |
Token ID | 12 |
Feature activation | +4.21e-03 |
Pos logit contributions | |
and | +0.161 |
Follow | +0.145 |
Nice | +0.140 |
urry | +0.136 |
Today | +0.133 |
Neg logit contributions | |
ck | -0.106 |
ps | -0.105 |
fourteen | -0.096 |
eling | -0.094 |
bur | -0.094 |
Token | old |
Token ID | 727 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
bur | +0.007 |
ck | +0.007 |
fourteen | +0.007 |
eling | +0.007 |
ps | +0.007 |
Neg logit contributions | |
and | -0.005 |
Follow | -0.004 |
urry | -0.004 |
Nice | -0.004 |
Today | -0.004 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
bur | +0.049 |
fourteen | +0.047 |
eling | +0.046 |
ps | +0.046 |
blew | +0.045 |
Neg logit contributions | |
and | -0.070 |
urry | -0.062 |
Follow | -0.061 |
Nice | -0.056 |
tum | -0.056 |
Token | were |
Token ID | 547 |
Feature activation | +3.21e-03 |
Pos logit contributions | |
and | +0.027 |
urry | +0.024 |
Follow | +0.024 |
tum | +0.022 |
Stay | +0.022 |
Neg logit contributions | |
bur | -0.020 |
fourteen | -0.020 |
eling | -0.020 |
ps | -0.019 |
IT | -0.019 |
Token | very |
Token ID | 845 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
fourteen | +0.005 |
bur | +0.005 |
eling | +0.005 |
ible | +0.005 |
IT | +0.005 |
Neg logit contributions | |
and | -0.005 |
urry | -0.004 |
Follow | -0.004 |
tum | -0.004 |
of | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +4.21e-04 |
Pos logit contributions | |
bur | +0.024 |
fourteen | +0.023 |
eling | +0.022 |
ps | +0.022 |
blew | +0.022 |
Neg logit contributions | |
and | -0.046 |
urry | -0.041 |
Follow | -0.040 |
Nice | -0.038 |
tum | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -3.61e-03 |
Pos logit contributions | |
eling | +0.001 |
bur | +0.001 |
cking | +0.001 |
ck | +0.001 |
lled | +0.001 |
Neg logit contributions | |
and | -0.001 |
Follow | -0.000 |
urry | -0.000 |
Today | -0.000 |
Nice | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -2.57e-03 |
Pos logit contributions | |
and | +0.005 |
urry | +0.004 |
Follow | +0.004 |
Nice | +0.004 |
tum | +0.004 |
Neg logit contributions | |
bur | -0.006 |
fourteen | -0.006 |
eling | -0.006 |
ps | -0.006 |
blew | -0.006 |
Token | " |
Token ID | 1 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
and | +0.005 |
urry | +0.004 |
Follow | +0.004 |
Stay | +0.004 |
Go | +0.004 |
Neg logit contributions | |
bur | -0.004 |
ps | -0.003 |
fourteen | -0.003 |
eling | -0.003 |
ible | -0.003 |
Token | Ro |
Token ID | 15450 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
eling | +0.049 |
ries | +0.048 |
ps | +0.048 |
abi | +0.047 |
ible | +0.047 |
Neg logit contributions | |
and | -0.047 |
Follow | -0.040 |
Stay | -0.039 |
Go | -0.039 |
Watch | -0.038 |
Token | aaa |
Token ID | 46071 |
Feature activation | -8.61e-02 |
Pos logit contributions | |
and | +0.038 |
urry | +0.032 |
Follow | +0.032 |
tum | +0.028 |
Nice | +0.028 |
Neg logit contributions | |
bur | -0.044 |
fourteen | -0.043 |
eling | -0.042 |
ps | -0.042 |
blew | -0.041 |
Token | ar |
Token ID | 283 |
Feature activation | +3.92e-03 |
Pos logit contributions | |
and | +0.145 |
Follow | +0.127 |
vere | +0.123 |
halls | +0.117 |
urry | +0.116 |
Neg logit contributions | |
bur | -0.110 |
fourteen | -0.109 |
ible | -0.103 |
et | -0.103 |
blew | -0.101 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.006 |
ps | +0.006 |
blew | +0.006 |
Neg logit contributions | |
and | -0.006 |
urry | -0.005 |
Follow | -0.005 |
Nice | -0.005 |
tum | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | +6.78e-03 |
Pos logit contributions | |
bur | +0.045 |
eling | +0.042 |
cking | +0.041 |
lled | +0.041 |
ps | +0.041 |
Neg logit contributions | |
and | -0.042 |
Follow | -0.038 |
tum | -0.037 |
of | -0.036 |
urry | -0.036 |
Token | shouted |
Token ID | 17293 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
bur | +0.009 |
fourteen | +0.009 |
eling | +0.008 |
ps | +0.008 |
IT | +0.008 |
Neg logit contributions | |
and | -0.012 |
Follow | -0.011 |
urry | -0.011 |
tum | -0.010 |
Stay | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
and | +0.033 |
urry | +0.029 |
Follow | +0.029 |
Nice | +0.026 |
tum | +0.026 |
Neg logit contributions | |
bur | -0.021 |
fourteen | -0.021 |
ps | -0.020 |
eling | -0.020 |
blew | -0.020 |
Token | said |
Token ID | 531 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
and | +0.012 |
urry | +0.011 |
Follow | +0.011 |
tum | +0.010 |
Stay | +0.010 |
Neg logit contributions | |
fourteen | -0.008 |
bur | -0.008 |
ps | -0.007 |
ible | -0.007 |
lled | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
and | +0.024 |
Follow | +0.023 |
urry | +0.020 |
suits | +0.020 |
ooters | +0.019 |
Neg logit contributions | |
her | -0.017 |
fourteen | -0.016 |
blew | -0.015 |
how | -0.014 |
just | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | +8.41e-04 |
Pos logit contributions | |
cking | +0.019 |
fourteen | +0.019 |
lled | +0.018 |
blew | +0.018 |
ps | +0.018 |
Neg logit contributions | |
and | -0.036 |
School | -0.035 |
Stay | -0.035 |
urry | -0.034 |
Go | -0.033 |
Token | 3 |
Token ID | 513 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
eling | +0.001 |
ps | +0.001 |
blew | +0.001 |
Neg logit contributions | |
and | -0.001 |
urry | -0.001 |
Follow | -0.001 |
Nice | -0.001 |
tum | -0.001 |
Token | - |
Token ID | 12 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
and | +0.058 |
Follow | +0.051 |
urry | +0.048 |
Nice | +0.043 |
rav | +0.043 |
Neg logit contributions | |
eling | -0.074 |
bur | -0.073 |
ps | -0.073 |
ck | -0.072 |
fourteen | -0.071 |
Token | year |
Token ID | 1941 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
bur | +0.007 |
fourteen | +0.007 |
ps | +0.007 |
eling | +0.007 |
blew | +0.007 |
Neg logit contributions | |
and | -0.003 |
urry | -0.002 |
Follow | -0.002 |
Stay | -0.002 |
aring | -0.002 |
Token | - |
Token ID | 12 |
Feature activation | +3.23e-03 |
Pos logit contributions | |
and | +0.156 |
Follow | +0.141 |
urry | +0.133 |
Nice | +0.131 |
rav | +0.127 |
Neg logit contributions | |
ps | -0.104 |
ck | -0.102 |
bur | -0.102 |
eling | -0.100 |
fourteen | -0.098 |
Token | old |
Token ID | 727 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.006 |
ps | +0.006 |
ck | +0.006 |
Neg logit contributions | |
and | -0.004 |
Follow | -0.003 |
urry | -0.003 |
Nice | -0.003 |
tum | -0.003 |
Token | son |
Token ID | 3367 |
Feature activation | -8.78e-03 |
Pos logit contributions | |
bur | +0.061 |
fourteen | +0.060 |
eling | +0.059 |
ps | +0.059 |
blew | +0.057 |
Neg logit contributions | |
and | -0.056 |
urry | -0.047 |
Follow | -0.047 |
tum | -0.042 |
Nice | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -9.97e-03 |
Pos logit contributions | |
and | +0.012 |
Follow | +0.010 |
urry | +0.010 |
tum | +0.008 |
vere | +0.008 |
Neg logit contributions | |
bur | -0.015 |
fourteen | -0.015 |
ps | -0.014 |
eling | -0.014 |
ible | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
and | +0.015 |
urry | +0.013 |
Follow | +0.012 |
Stay | +0.012 |
Nice | +0.011 |
Neg logit contributions | |
ps | -0.016 |
bur | -0.016 |
eling | -0.016 |
fourteen | -0.016 |
lled | -0.016 |
Token | your |
Token ID | 534 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
IT | +0.024 |
ries | +0.023 |
fourteen | +0.023 |
ami | +0.023 |
ible | +0.023 |
Neg logit contributions | |
and | -0.021 |
welcome | -0.020 |
urry | -0.019 |
of | -0.018 |
Follow | -0.018 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
and | +0.028 |
urry | +0.024 |
Follow | +0.023 |
tum | +0.021 |
Nice | +0.021 |
Neg logit contributions | |
bur | -0.031 |
fourteen | -0.031 |
eling | -0.031 |
ps | -0.030 |
blew | -0.029 |
Token | ." |
Token ID | 526 |
Feature activation | +8.86e-03 |
Pos logit contributions | |
bur | +0.023 |
fourteen | +0.022 |
ps | +0.022 |
eling | +0.021 |
blew | +0.021 |
Neg logit contributions | |
and | -0.038 |
urry | -0.033 |
Follow | -0.033 |
Nice | -0.030 |
tum | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +3.55e-05 |
Pos logit contributions | |
bur | +0.014 |
fourteen | +0.014 |
eling | +0.013 |
ps | +0.013 |
blew | +0.013 |
Neg logit contributions | |
and | -0.013 |
urry | -0.011 |
Follow | -0.011 |
Nice | -0.010 |
tum | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +2.57e-04 |
Pos logit contributions | |
bur | +0.000 |
fourteen | +0.000 |
eling | +0.000 |
ps | +0.000 |
blew | +0.000 |
Neg logit contributions | |
and | -0.000 |
Follow | -0.000 |
urry | -0.000 |
Nice | -0.000 |
tum | -0.000 |
Token | S |
Token ID | 50 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
ps | +0.000 |
bur | +0.000 |
ible | +0.000 |
fourteen | +0.000 |
IT | +0.000 |
Neg logit contributions | |
and | -0.000 |
urry | -0.000 |
Follow | -0.000 |
Stay | -0.000 |
Go | -0.000 |
Token | ara |
Token ID | 3301 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
bur | +0.089 |
lled | +0.088 |
ps | +0.088 |
fourteen | +0.087 |
cking | +0.086 |
Neg logit contributions | |
and | -0.111 |
urry | -0.100 |
Follow | -0.095 |
story | -0.092 |
tum | -0.092 |
Token | is |
Token ID | 318 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
and | +0.026 |
Follow | +0.022 |
urry | +0.022 |
tum | +0.020 |
Stay | +0.020 |
Neg logit contributions | |
bur | -0.025 |
fourteen | -0.025 |
eling | -0.024 |
ps | -0.023 |
blew | -0.023 |
Token | not |
Token ID | 407 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
bur | +0.029 |
fourteen | +0.029 |
ible | +0.029 |
eling | +0.028 |
ami | +0.028 |
Neg logit contributions | |
and | -0.021 |
urry | -0.018 |
Follow | -0.018 |
tum | -0.016 |
Stay | -0.016 |
Token | sure |
Token ID | 1654 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
fourteen | +0.057 |
IT | +0.055 |
neath | +0.055 |
eling | +0.054 |
ible | +0.053 |
Neg logit contributions | |
and | -0.033 |
welcome | -0.027 |
urry | -0.026 |
of | -0.023 |
story | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-03 |
Pos logit contributions | |
bur | +0.027 |
fourteen | +0.026 |
ps | +0.025 |
eling | +0.025 |
blew | +0.025 |
Neg logit contributions | |
and | -0.041 |
urry | -0.036 |
Follow | -0.035 |
tum | -0.033 |
Nice | -0.033 |
Token | very |
Token ID | 845 |
Feature activation | +5.31e-03 |
Pos logit contributions | |
bur | +0.008 |
fourteen | +0.008 |
ible | +0.008 |
eling | +0.008 |
IT | +0.008 |
Neg logit contributions | |
and | -0.006 |
urry | -0.005 |
Follow | -0.005 |
apping | -0.004 |
tum | -0.004 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
bur | +0.009 |
fourteen | +0.009 |
eling | +0.009 |
ps | +0.008 |
blew | +0.008 |
Neg logit contributions | |
and | -0.008 |
urry | -0.006 |
Follow | -0.006 |
tum | -0.006 |
Nice | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -3.99e-03 |
Pos logit contributions | |
bur | +0.019 |
eling | +0.018 |
fourteen | +0.018 |
ps | +0.017 |
IT | +0.017 |
Neg logit contributions | |
and | -0.021 |
urry | -0.017 |
of | -0.017 |
birthday | -0.017 |
apping | -0.016 |
Token | They |
Token ID | 1119 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
and | +0.005 |
urry | +0.004 |
Follow | +0.004 |
Nice | +0.004 |
tum | +0.004 |
Neg logit contributions | |
bur | -0.007 |
eling | -0.007 |
fourteen | -0.007 |
ps | -0.007 |
ck | -0.007 |
Token | took |
Token ID | 1718 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
bur | +0.018 |
eling | +0.017 |
fourteen | +0.017 |
ries | +0.017 |
ps | +0.016 |
Neg logit contributions | |
and | -0.018 |
urry | -0.016 |
Follow | -0.016 |
tum | -0.015 |
of | -0.015 |
Token | turns |
Token ID | 4962 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
eling | +0.043 |
bur | +0.043 |
IT | +0.039 |
ami | +0.039 |
rob | +0.038 |
Neg logit contributions | |
and | -0.056 |
tum | -0.052 |
of | -0.051 |
urry | -0.049 |
Follow | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
bur | +0.097 |
eling | +0.095 |
fourteen | +0.094 |
ps | +0.091 |
ible | +0.091 |
Neg logit contributions | |
and | -0.085 |
urry | -0.070 |
Follow | -0.067 |
tum | -0.063 |
apping | -0.061 |
Token | sit |
Token ID | 1650 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
elle | +0.032 |
rob | +0.032 |
ible | +0.032 |
awed | +0.031 |
eling | +0.031 |
Neg logit contributions | |
birthday | -0.016 |
of | -0.016 |
welcome | -0.016 |
tum | -0.015 |
please | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
and | +0.019 |
urry | +0.016 |
Follow | +0.016 |
apping | +0.015 |
tum | +0.014 |
Neg logit contributions | |
fourteen | -0.021 |
eling | -0.021 |
bur | -0.020 |
ps | -0.020 |
lled | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -2.93e-03 |
Pos logit contributions | |
and | +0.061 |
of | +0.056 |
birthday | +0.056 |
story | +0.054 |
tum | +0.052 |
Neg logit contributions | |
eling | -0.053 |
bur | -0.050 |
ible | -0.048 |
fourteen | -0.047 |
cking | -0.047 |
Token | pumpkin |
Token ID | 30089 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
and | +0.005 |
urry | +0.004 |
Follow | +0.004 |
tum | +0.004 |
birthday | +0.004 |
Neg logit contributions | |
bur | -0.004 |
eling | -0.004 |
fourteen | -0.004 |
ps | -0.004 |
lled | -0.004 |
Token | Her |
Token ID | 2332 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
and | +0.003 |
urry | +0.002 |
Follow | +0.002 |
Nice | +0.002 |
tum | +0.002 |
Neg logit contributions | |
bur | -0.003 |
fourteen | -0.003 |
ps | -0.003 |
eling | -0.003 |
blew | -0.002 |
Token | dad |
Token ID | 9955 |
Feature activation | +9.69e-04 |
Pos logit contributions | |
bur | +0.018 |
ps | +0.017 |
fourteen | +0.017 |
blew | +0.017 |
eling | +0.017 |
Neg logit contributions | |
and | -0.022 |
Follow | -0.019 |
urry | -0.019 |
Nice | -0.018 |
Stay | -0.017 |
Token | asked |
Token ID | 1965 |
Feature activation | +5.76e-03 |
Pos logit contributions | |
fourteen | +0.001 |
bur | +0.001 |
ps | +0.001 |
eling | +0.001 |
IT | +0.001 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
Follow | -0.002 |
tum | -0.001 |
Stay | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -9.05e-03 |
Pos logit contributions | |
bur | +0.002 |
fourteen | +0.002 |
ps | +0.002 |
eling | +0.002 |
her | +0.002 |
Neg logit contributions | |
and | -0.015 |
Follow | -0.014 |
urry | -0.014 |
tum | -0.013 |
Nice | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
and | +0.012 |
urry | +0.010 |
Follow | +0.010 |
tum | +0.008 |
Nice | +0.008 |
Neg logit contributions | |
bur | -0.016 |
fourteen | -0.016 |
eling | -0.016 |
ps | -0.016 |
blew | -0.015 |
Token | Why |
Token ID | 5195 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
and | +0.022 |
Follow | +0.018 |
Stay | +0.017 |
urry | +0.017 |
Come | +0.017 |
Neg logit contributions | |
ps | -0.011 |
ible | -0.010 |
fourteen | -0.010 |
pointer | -0.010 |
abi | -0.009 |
Token | are |
Token ID | 389 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
blew | +0.091 |
bur | +0.090 |
ps | +0.088 |
eling | +0.087 |
fourteen | +0.087 |
Neg logit contributions | |
and | -0.089 |
Follow | -0.074 |
urry | -0.071 |
Nice | -0.066 |
apping | -0.063 |
Token | you |
Token ID | 345 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.006 |
ps | +0.006 |
blew | +0.006 |
Neg logit contributions | |
and | -0.004 |
urry | -0.003 |
Follow | -0.003 |
tum | -0.003 |
Nice | -0.003 |
Token | crying |
Token ID | 13774 |
Feature activation | -3.95e-03 |
Pos logit contributions | |
and | +0.034 |
urry | +0.030 |
Follow | +0.029 |
tum | +0.026 |
Nice | +0.026 |
Neg logit contributions | |
bur | -0.034 |
fourteen | -0.034 |
eling | -0.033 |
ps | -0.032 |
ible | -0.032 |
Token | ?". |
Token ID | 43634 |
Feature activation | +7.15e-03 |
Pos logit contributions | |
and | +0.007 |
Follow | +0.006 |
urry | +0.006 |
rav | +0.005 |
engu | +0.005 |
Neg logit contributions | |
bur | -0.006 |
fourteen | -0.005 |
blew | -0.005 |
ps | -0.005 |
eling | -0.005 |
Token | Mary |
Token ID | 5335 |
Feature activation | -3.47e-03 |
Pos logit contributions | |
bur | +0.012 |
ps | +0.011 |
eling | +0.011 |
cking | +0.011 |
fourteen | +0.011 |
Neg logit contributions | |
and | -0.011 |
Follow | -0.009 |
urry | -0.009 |
Today | -0.009 |
Stay | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.36e-03 |
Pos logit contributions | |
and | +0.048 |
Follow | +0.044 |
urry | +0.040 |
halls | +0.040 |
uggle | +0.040 |
Neg logit contributions | |
fourteen | -0.022 |
ps | -0.021 |
bur | -0.021 |
just | -0.020 |
her | -0.020 |
Token | I |
Token ID | 314 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
and | +0.002 |
halls | +0.001 |
suits | +0.001 |
vere | +0.001 |
story | +0.001 |
Neg logit contributions | |
fourteen | -0.002 |
bur | -0.002 |
IT | -0.002 |
blew | -0.002 |
ps | -0.002 |
Token | don |
Token ID | 836 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
and | +0.044 |
urry | +0.039 |
Follow | +0.037 |
tum | +0.034 |
Stay | +0.033 |
Neg logit contributions | |
bur | -0.049 |
fourteen | -0.049 |
eling | -0.047 |
ps | -0.047 |
ries | -0.046 |
Token | 't |
Token ID | 470 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
and | +0.047 |
Follow | +0.041 |
urry | +0.040 |
tum | +0.038 |
Nice | +0.037 |
Neg logit contributions | |
bur | -0.027 |
fourteen | -0.027 |
ps | -0.027 |
blew | -0.027 |
eling | -0.026 |
Token | like |
Token ID | 588 |
Feature activation | -2.84e-03 |
Pos logit contributions | |
bur | +0.025 |
fourteen | +0.024 |
eling | +0.024 |
ps | +0.024 |
blew | +0.023 |
Neg logit contributions | |
and | -0.021 |
Follow | -0.018 |
urry | -0.018 |
Nice | -0.016 |
tum | -0.016 |
Token | brushes |
Token ID | 36377 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
and | +0.005 |
urry | +0.004 |
Follow | +0.004 |
tum | +0.004 |
Go | +0.004 |
Neg logit contributions | |
bur | -0.004 |
fourteen | -0.004 |
eling | -0.003 |
lled | -0.003 |
ps | -0.003 |
Token | ." |
Token ID | 526 |
Feature activation | +7.85e-03 |
Pos logit contributions | |
bur | +0.086 |
eling | +0.084 |
fourteen | +0.083 |
ps | +0.081 |
blew | +0.081 |
Neg logit contributions | |
and | -0.088 |
urry | -0.074 |
Follow | -0.072 |
tum | -0.068 |
Nice | -0.067 |
Token | |
Token ID | 220 |
Feature activation | -6.07e-03 |
Pos logit contributions | |
eling | +0.016 |
lled | +0.016 |
ps | +0.015 |
fourteen | +0.015 |
bur | +0.015 |
Neg logit contributions | |
and | -0.009 |
urry | -0.007 |
Follow | -0.007 |
tum | -0.007 |
Nice | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +6.36e-04 |
Pos logit contributions | |
and | +0.009 |
urry | +0.007 |
Follow | +0.007 |
tum | +0.007 |
Nice | +0.006 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.010 |
eling | -0.010 |
ps | -0.010 |
blew | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
eling | +0.001 |
ps | +0.001 |
blew | +0.001 |
Neg logit contributions | |
and | -0.001 |
urry | -0.001 |
Follow | -0.001 |
tum | -0.001 |
Nice | -0.001 |
Token | L |
Token ID | 43 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
ps | +0.002 |
fourteen | +0.002 |
bur | +0.002 |
eling | +0.002 |
ible | +0.002 |
Neg logit contributions | |
and | -0.004 |
urry | -0.003 |
Follow | -0.003 |
Stay | -0.003 |
tum | -0.003 |
Token | There |
Token ID | 1318 |
Feature activation | +3.89e-03 |
Pos logit contributions | |
her | +0.013 |
IT | +0.011 |
neath | +0.010 |
Why | +0.010 |
fourteen | +0.010 |
Neg logit contributions | |
and | -0.027 |
vere | -0.023 |
ye | -0.022 |
rav | -0.022 |
urry | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +9.71e-03 |
Pos logit contributions | |
blew | +0.004 |
her | +0.004 |
bur | +0.004 |
lled | +0.004 |
ck | +0.004 |
Neg logit contributions | |
and | -0.008 |
Follow | -0.007 |
urry | -0.007 |
aring | -0.007 |
ooters | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -4.33e-03 |
Pos logit contributions | |
bur | +0.010 |
eling | +0.009 |
fourteen | +0.009 |
ps | +0.009 |
ck | +0.009 |
Neg logit contributions | |
and | -0.020 |
Follow | -0.018 |
urry | -0.018 |
tum | -0.017 |
Nice | -0.017 |
Token | big |
Token ID | 1263 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
and | +0.009 |
urry | +0.007 |
Follow | +0.007 |
Go | +0.007 |
aring | +0.007 |
Neg logit contributions | |
blew | -0.005 |
bur | -0.005 |
comb | -0.005 |
pin | -0.005 |
sc | -0.005 |
Token | shiny |
Token ID | 22441 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
blew | +0.021 |
bur | +0.021 |
fourteen | +0.021 |
ible | +0.020 |
et | +0.019 |
Neg logit contributions | |
and | -0.033 |
Follow | -0.026 |
urry | -0.026 |
rav | -0.024 |
anny | -0.024 |
Token | crane |
Token ID | 41175 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
bur | +0.028 |
fourteen | +0.027 |
blew | +0.027 |
ps | +0.026 |
eling | +0.026 |
Neg logit contributions | |
and | -0.041 |
urry | -0.036 |
Follow | -0.036 |
Nice | -0.032 |
Stay | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e-03 |
Pos logit contributions | |
bur | +0.072 |
fourteen | +0.072 |
ps | +0.070 |
eling | +0.069 |
ible | +0.066 |
Neg logit contributions | |
and | -0.102 |
urry | -0.090 |
Follow | -0.088 |
Nice | -0.081 |
tum | -0.081 |
Token | with |
Token ID | 351 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
and | +0.002 |
Follow | +0.002 |
urry | +0.002 |
Nice | +0.002 |
Stay | +0.002 |
Neg logit contributions | |
bur | -0.002 |
eling | -0.002 |
fourteen | -0.002 |
ps | -0.002 |
ible | -0.002 |
Token | two |
Token ID | 734 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
bur | +0.015 |
eling | +0.014 |
fourteen | +0.014 |
IT | +0.013 |
ps | +0.013 |
Neg logit contributions | |
and | -0.019 |
Follow | -0.016 |
of | -0.015 |
urry | -0.015 |
tum | -0.015 |
Token | people |
Token ID | 661 |
Feature activation | +7.23e-04 |
Pos logit contributions | |
fourteen | +0.032 |
ps | +0.031 |
bur | +0.031 |
eling | +0.031 |
ible | +0.031 |
Neg logit contributions | |
and | -0.046 |
Follow | -0.041 |
urry | -0.041 |
School | -0.037 |
uggle | -0.036 |
Token | operating |
Token ID | 5361 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
eling | +0.001 |
ps | +0.001 |
IT | +0.001 |
Neg logit contributions | |
and | -0.001 |
urry | -0.001 |
Follow | -0.001 |
apping | -0.001 |
Nice | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
and | +0.003 |
Follow | +0.003 |
urry | +0.003 |
story | +0.003 |
tum | +0.003 |
Neg logit contributions | |
fourteen | -0.004 |
eling | -0.003 |
cking | -0.003 |
ps | -0.003 |
bur | -0.003 |
Token | playing |
Token ID | 2712 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
bur | +0.013 |
eling | +0.013 |
fourteen | +0.013 |
ps | +0.013 |
blew | +0.013 |
Neg logit contributions | |
and | -0.018 |
Follow | -0.016 |
urry | -0.016 |
Nice | -0.015 |
rav | -0.014 |
Token | near |
Token ID | 1474 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
and | +0.026 |
Follow | +0.021 |
urry | +0.021 |
tum | +0.019 |
Stay | +0.018 |
Neg logit contributions | |
bur | -0.028 |
fourteen | -0.027 |
eling | -0.026 |
ps | -0.026 |
ible | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e-03 |
Pos logit contributions | |
and | +0.060 |
of | +0.052 |
tum | +0.049 |
Stay | +0.049 |
story | +0.048 |
Neg logit contributions | |
fourteen | -0.050 |
bur | -0.046 |
ps | -0.046 |
eling | -0.045 |
rob | -0.045 |
Token | door |
Token ID | 3420 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
bur | +0.002 |
fourteen | +0.002 |
blew | +0.002 |
ps | +0.002 |
eling | +0.002 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
Follow | -0.002 |
Nice | -0.002 |
rav | -0.002 |
Token | when |
Token ID | 618 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
and | +0.077 |
urry | +0.069 |
Follow | +0.069 |
rav | +0.064 |
Nice | +0.063 |
Neg logit contributions | |
fourteen | -0.059 |
bur | -0.059 |
eling | -0.057 |
ps | -0.057 |
blew | -0.056 |
Token | his |
Token ID | 465 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ps | +0.025 |
fourteen | +0.025 |
bur | +0.024 |
ck | +0.023 |
eling | +0.023 |
Neg logit contributions | |
and | -0.034 |
urry | -0.031 |
tum | -0.030 |
Follow | -0.030 |
Stay | -0.029 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.57e-03 |
Pos logit contributions | |
bur | +0.013 |
blew | +0.013 |
ps | +0.013 |
sc | +0.012 |
comb | +0.012 |
Neg logit contributions | |
and | -0.023 |
Follow | -0.021 |
urry | -0.020 |
Nice | -0.019 |
aring | -0.019 |
Token | my |
Token ID | 1820 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
and | +0.008 |
urry | +0.007 |
Follow | +0.007 |
tum | +0.006 |
Nice | +0.006 |
Neg logit contributions | |
bur | -0.006 |
fourteen | -0.006 |
eling | -0.006 |
ps | -0.006 |
blew | -0.006 |
Token | told |
Token ID | 1297 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
fourteen | +0.013 |
bur | +0.013 |
eling | +0.012 |
ps | +0.012 |
ck | +0.011 |
Neg logit contributions | |
and | -0.019 |
urry | -0.016 |
Follow | -0.016 |
tum | -0.015 |
Nice | -0.015 |
Token | him |
Token ID | 683 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
and | +0.034 |
urry | +0.030 |
tum | +0.030 |
Stay | +0.029 |
story | +0.029 |
Neg logit contributions | |
fourteen | -0.025 |
bur | -0.024 |
cking | -0.023 |
blew | -0.023 |
ps | -0.023 |
Token | waving |
Token ID | 25849 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
ible | +0.006 |
fourteen | +0.006 |
ami | +0.006 |
IT | +0.006 |
wy | +0.006 |
Neg logit contributions | |
and | -0.004 |
welcome | -0.004 |
story | -0.004 |
birthday | -0.004 |
apping | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | +1.28e-03 |
Pos logit contributions | |
and | +0.043 |
urry | +0.040 |
Follow | +0.040 |
aring | +0.037 |
engu | +0.037 |
Neg logit contributions | |
fourteen | -0.008 |
ps | -0.007 |
bur | -0.007 |
blew | -0.007 |
eling | -0.006 |
Token | hand |
Token ID | 1021 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
ps | +0.002 |
bur | +0.002 |
blew | +0.002 |
fourteen | +0.002 |
eling | +0.002 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
Follow | -0.002 |
Nice | -0.001 |
Stay | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -5.91e-03 |
Pos logit contributions | |
and | +0.037 |
Follow | +0.034 |
urry | +0.033 |
rav | +0.031 |
Nice | +0.031 |
Neg logit contributions | |
bur | -0.027 |
fourteen | -0.027 |
ps | -0.027 |
blew | -0.026 |
ible | -0.025 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.39e-03 |
Pos logit contributions | |
and | +0.011 |
urry | +0.010 |
Follow | +0.010 |
Nice | +0.009 |
tum | +0.009 |
Neg logit contributions | |
bur | -0.007 |
fourteen | -0.007 |
eling | -0.007 |
ps | -0.007 |
blew | -0.007 |
Token | some |
Token ID | 617 |
Feature activation | +5.07e-03 |
Pos logit contributions | |
and | +0.016 |
Follow | +0.013 |
urry | +0.013 |
tum | +0.013 |
of | +0.013 |
Neg logit contributions | |
bur | -0.008 |
fourteen | -0.007 |
eling | -0.007 |
ps | -0.007 |
ck | -0.007 |
Token | dolphins |
Token ID | 36798 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
blew | +0.006 |
eling | +0.006 |
ps | +0.006 |
fourteen | +0.005 |
lled | +0.005 |
Neg logit contributions | |
and | -0.010 |
urry | -0.009 |
Follow | -0.009 |
Today | -0.008 |
Nice | -0.008 |
Token | swimming |
Token ID | 14899 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
and | +0.076 |
urry | +0.066 |
Follow | +0.066 |
tum | +0.062 |
rav | +0.061 |
Neg logit contributions | |
bur | -0.044 |
ps | -0.044 |
fourteen | -0.043 |
blew | -0.043 |
eling | -0.043 |
Token | by |
Token ID | 416 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
and | +0.064 |
urry | +0.060 |
Follow | +0.058 |
tum | +0.056 |
rav | +0.055 |
Neg logit contributions | |
ps | -0.036 |
bur | -0.036 |
blew | -0.035 |
lled | -0.035 |
fourteen | -0.033 |
Token | in |
Token ID | 287 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
and | +0.016 |
urry | +0.013 |
of | +0.012 |
birthday | +0.012 |
Follow | +0.012 |
Neg logit contributions | |
bur | -0.015 |
fourteen | -0.014 |
eling | -0.013 |
blew | -0.013 |
ible | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e-03 |
Pos logit contributions | |
and | +0.056 |
of | +0.050 |
story | +0.049 |
welcome | +0.048 |
birthday | +0.048 |
Neg logit contributions | |
bur | -0.038 |
fourteen | -0.037 |
eling | -0.037 |
ible | -0.036 |
blew | -0.036 |
Token | sad |
Token ID | 6507 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
bur | +0.030 |
eling | +0.029 |
blew | +0.029 |
fourteen | +0.029 |
ck | +0.029 |
Neg logit contributions | |
and | -0.030 |
Follow | -0.025 |
urry | -0.024 |
Nice | -0.023 |
Today | -0.022 |
Token | but |
Token ID | 475 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
bur | +0.032 |
eling | +0.030 |
fourteen | +0.030 |
ps | +0.030 |
ck | +0.029 |
Neg logit contributions | |
and | -0.030 |
urry | -0.025 |
Follow | -0.023 |
tum | -0.023 |
of | -0.022 |
Token | he |
Token ID | 339 |
Feature activation | +5.32e-03 |
Pos logit contributions | |
and | +0.003 |
urry | +0.003 |
Follow | +0.003 |
Nice | +0.003 |
tum | +0.003 |
Neg logit contributions | |
bur | -0.002 |
eling | -0.002 |
fourteen | -0.002 |
blew | -0.002 |
ps | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
fourteen | +0.008 |
bur | +0.008 |
ps | +0.007 |
eling | +0.007 |
ries | +0.007 |
Neg logit contributions | |
and | -0.009 |
Follow | -0.008 |
urry | -0.008 |
tum | -0.007 |
Stay | -0.007 |
Token | also |
Token ID | 635 |
Feature activation | +6.48e-03 |
Pos logit contributions | |
fourteen | +0.021 |
bur | +0.021 |
ible | +0.020 |
eling | +0.020 |
IT | +0.020 |
Neg logit contributions | |
and | -0.018 |
urry | -0.016 |
Follow | -0.015 |
tum | -0.014 |
apping | -0.014 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
bur | +0.010 |
fourteen | +0.010 |
eling | +0.010 |
ps | +0.010 |
blew | +0.010 |
Neg logit contributions | |
and | -0.010 |
urry | -0.008 |
Follow | -0.008 |
Nice | -0.007 |
tum | -0.007 |
Token | because |
Token ID | 780 |
Feature activation | +8.64e-03 |
Pos logit contributions | |
bur | +0.034 |
fourteen | +0.033 |
eling | +0.033 |
ps | +0.033 |
blew | +0.032 |
Neg logit contributions | |
and | -0.026 |
urry | -0.021 |
Follow | -0.020 |
tum | -0.018 |
Nice | -0.018 |
Token | he |
Token ID | 339 |
Feature activation | +5.37e-03 |
Pos logit contributions | |
bur | +0.009 |
fourteen | +0.008 |
eling | +0.008 |
blew | +0.008 |
ps | +0.008 |
Neg logit contributions | |
and | -0.018 |
urry | -0.016 |
Follow | -0.016 |
tum | -0.015 |
Nice | -0.015 |
Token | knew |
Token ID | 2993 |
Feature activation | -7.33e-03 |
Pos logit contributions | |
fourteen | +0.009 |
bur | +0.009 |
ps | +0.009 |
ries | +0.008 |
eling | +0.008 |
Neg logit contributions | |
and | -0.008 |
Follow | -0.007 |
urry | -0.007 |
tum | -0.006 |
Stay | -0.006 |
Token | his |
Token ID | 465 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
and | +0.013 |
urry | +0.012 |
Follow | +0.011 |
of | +0.011 |
tum | +0.011 |
Neg logit contributions | |
fourteen | -0.009 |
bur | -0.009 |
ps | -0.009 |
eling | -0.009 |
IT | -0.009 |
Token | dad |
Token ID | 9955 |
Feature activation | -2.08e-04 |
Pos logit contributions | |
bur | +0.013 |
fourteen | +0.012 |
ps | +0.012 |
blew | +0.012 |
eling | +0.012 |
Neg logit contributions | |
and | -0.021 |
Follow | -0.018 |
urry | -0.017 |
Nice | -0.017 |
Go | -0.016 |
Token | I |
Token ID | 314 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
fourteen | +0.004 |
bur | +0.004 |
blew | +0.004 |
ps | +0.004 |
IT | +0.004 |
Neg logit contributions | |
and | -0.004 |
suits | -0.003 |
halls | -0.002 |
urry | -0.002 |
story | -0.002 |
Token | 'll |
Token ID | 1183 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
and | +0.036 |
urry | +0.030 |
Follow | +0.029 |
tum | +0.026 |
Nice | +0.026 |
Neg logit contributions | |
bur | -0.049 |
fourteen | -0.048 |
eling | -0.048 |
ps | -0.047 |
blew | -0.047 |
Token | help |
Token ID | 1037 |
Feature activation | -9.82e-03 |
Pos logit contributions | |
urry | +0.024 |
and | +0.023 |
Follow | +0.022 |
Stay | +0.020 |
tum | +0.019 |
Neg logit contributions | |
fourteen | -0.041 |
ries | -0.040 |
bur | -0.040 |
rob | -0.040 |
lled | -0.040 |
Token | you |
Token ID | 345 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
and | +0.014 |
urry | +0.012 |
Follow | +0.012 |
Nice | +0.011 |
tum | +0.011 |
Neg logit contributions | |
bur | -0.016 |
eling | -0.016 |
fourteen | -0.016 |
ps | -0.016 |
ck | -0.015 |
Token | ." |
Token ID | 526 |
Feature activation | +9.11e-03 |
Pos logit contributions | |
and | +0.035 |
urry | +0.030 |
Follow | +0.030 |
tum | +0.027 |
Nice | +0.027 |
Neg logit contributions | |
bur | -0.033 |
fourteen | -0.032 |
eling | -0.032 |
ps | -0.031 |
blew | -0.031 |
Token | Roc |
Token ID | 13545 |
Feature activation | +7.46e-03 |
Pos logit contributions | |
bur | +0.016 |
fourteen | +0.016 |
eling | +0.016 |
ps | +0.015 |
blew | +0.015 |
Neg logit contributions | |
and | -0.012 |
urry | -0.010 |
Follow | -0.010 |
tum | -0.009 |
Nice | -0.009 |
Token | co |
Token ID | 1073 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
fourteen | +0.008 |
bur | +0.008 |
ps | +0.008 |
blew | +0.008 |
lled | +0.008 |
Neg logit contributions | |
and | -0.015 |
urry | -0.013 |
Follow | -0.013 |
aring | -0.012 |
Nice | -0.012 |
Token | pulled |
Token ID | 5954 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
bur | +0.034 |
fourteen | +0.033 |
ps | +0.033 |
eling | +0.032 |
IT | +0.031 |
Neg logit contributions | |
and | -0.046 |
urry | -0.041 |
Follow | -0.039 |
of | -0.036 |
Stay | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +6.44e-04 |
Pos logit contributions | |
bur | +0.042 |
fourteen | +0.042 |
eling | +0.041 |
blew | +0.039 |
ps | +0.039 |
Neg logit contributions | |
and | -0.067 |
urry | -0.059 |
Follow | -0.058 |
Nice | -0.053 |
Stay | -0.053 |
Token | carrot |
Token ID | 40562 |
Feature activation | -3.33e-03 |
Pos logit contributions | |
bur | +0.001 |
fourteen | +0.001 |
ps | +0.001 |
eling | +0.001 |
blew | +0.001 |
Neg logit contributions | |
and | -0.001 |
urry | -0.001 |
Follow | -0.001 |
apping | -0.001 |
School | -0.001 |
Token | out |
Token ID | 503 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
and | +0.006 |
urry | +0.005 |
Follow | +0.005 |
Nice | +0.004 |
tum | +0.004 |
Neg logit contributions | |
bur | -0.005 |
fourteen | -0.005 |
ps | -0.004 |
eling | -0.004 |
blew | -0.004 |
Token | no |
Token ID | 645 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
and | +0.003 |
Follow | +0.003 |
urry | +0.003 |
Nice | +0.002 |
tum | +0.002 |
Neg logit contributions | |
bur | -0.001 |
eling | -0.001 |
ps | -0.001 |
fourteen | -0.001 |
ck | -0.001 |
Token | worry |
Token ID | 5490 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
and | +0.034 |
urry | +0.031 |
tum | +0.030 |
of | +0.030 |
Follow | +0.030 |
Neg logit contributions | |
bur | -0.017 |
ps | -0.016 |
fourteen | -0.016 |
eling | -0.016 |
IT | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
and | +0.050 |
urry | +0.044 |
Follow | +0.044 |
Nice | +0.040 |
tum | +0.040 |
Neg logit contributions | |
bur | -0.035 |
fourteen | -0.034 |
eling | -0.033 |
ps | -0.033 |
blew | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | +6.24e-03 |
Pos logit contributions | |
and | +0.050 |
of | +0.045 |
urry | +0.044 |
birthday | +0.044 |
story | +0.044 |
Neg logit contributions | |
bur | -0.037 |
fourteen | -0.035 |
rob | -0.035 |
eling | -0.034 |
ami | -0.033 |
Token | world |
Token ID | 995 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
bur | +0.009 |
fourteen | +0.009 |
eling | +0.009 |
ps | +0.009 |
blew | +0.009 |
Neg logit contributions | |
and | -0.010 |
urry | -0.009 |
Follow | -0.009 |
Nice | -0.008 |
apping | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +7.76e-03 |
Pos logit contributions | |
and | +0.080 |
urry | +0.070 |
Follow | +0.068 |
Nice | +0.063 |
of | +0.063 |
Neg logit contributions | |
bur | -0.058 |
fourteen | -0.056 |
eling | -0.055 |
ps | -0.054 |
ck | -0.052 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.87e-03 |
Pos logit contributions | |
fourteen | +0.012 |
bur | +0.012 |
ible | +0.011 |
eling | +0.011 |
IT | +0.011 |
Neg logit contributions | |
and | -0.012 |
urry | -0.010 |
Follow | -0.010 |
vere | -0.009 |
tum | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
bur | +0.004 |
fourteen | +0.004 |
ps | +0.004 |
blew | +0.004 |
eling | +0.004 |
Neg logit contributions | |
and | -0.007 |
urry | -0.007 |
Follow | -0.007 |
tum | -0.006 |
Nice | -0.006 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
bur | +0.065 |
fourteen | +0.062 |
blew | +0.061 |
eling | +0.061 |
ck | +0.060 |
Neg logit contributions | |
and | -0.051 |
Follow | -0.046 |
urry | -0.038 |
Go | -0.037 |
Stay | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | -5.27e-03 |
Pos logit contributions | |
and | +0.074 |
urry | +0.063 |
Follow | +0.062 |
Stay | +0.060 |
of | +0.059 |
Neg logit contributions | |
bur | -0.044 |
fourteen | -0.043 |
eling | -0.041 |
ible | -0.040 |
ps | -0.040 |
Token | time |
Token ID | 640 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
and | +0.008 |
Follow | +0.007 |
urry | +0.007 |
Nice | +0.006 |
tum | +0.006 |
Neg logit contributions | |
bur | -0.008 |
fourteen | -0.008 |
eling | -0.008 |
ps | -0.007 |
blew | -0.007 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
and | +0.007 |
Follow | +0.006 |
urry | +0.006 |
tum | +0.006 |
Stay | +0.006 |
Neg logit contributions | |
bur | -0.005 |
fourteen | -0.005 |
ps | -0.005 |
eling | -0.004 |
ible | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | +6.26e-03 |
Pos logit contributions | |
bur | +0.046 |
eling | +0.044 |
blew | +0.044 |
fourteen | +0.044 |
IT | +0.042 |
Neg logit contributions | |
and | -0.107 |
Follow | -0.096 |
urry | -0.094 |
tum | -0.092 |
of | -0.091 |
Token | arm |
Token ID | 3211 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
bur | +0.010 |
fourteen | +0.010 |
eling | +0.009 |
ps | +0.009 |
blew | +0.009 |
Neg logit contributions | |
and | -0.010 |
urry | -0.008 |
Follow | -0.008 |
tum | -0.008 |
Nice | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -3.38e-03 |
Pos logit contributions | |
and | +0.034 |
Follow | +0.032 |
urry | +0.031 |
rav | +0.030 |
Nice | +0.029 |
Neg logit contributions | |
bur | -0.021 |
fourteen | -0.020 |
ps | -0.020 |
blew | -0.019 |
eling | -0.019 |
Token | Jen |
Token ID | 13364 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
and | +0.005 |
Follow | +0.004 |
urry | +0.004 |
Nice | +0.004 |
tum | +0.004 |
Neg logit contributions | |
bur | -0.006 |
eling | -0.006 |
ps | -0.006 |
fourteen | -0.006 |
ck | -0.005 |
Token | started |
Token ID | 2067 |
Feature activation | -2.91e-03 |
Pos logit contributions | |
fourteen | +0.034 |
bur | +0.032 |
cking | +0.031 |
ps | +0.031 |
IT | +0.031 |
Neg logit contributions | |
and | -0.035 |
urry | -0.031 |
Follow | -0.030 |
of | -0.028 |
tum | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
and | +0.003 |
urry | +0.003 |
Follow | +0.002 |
tum | +0.002 |
Stay | +0.002 |
Neg logit contributions | |
fourteen | -0.006 |
bur | -0.006 |
eling | -0.006 |
ible | -0.006 |
ps | -0.005 |
Token | cry |
Token ID | 3960 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
fourteen | +0.032 |
bur | +0.030 |
ible | +0.029 |
ami | +0.029 |
ps | +0.029 |
Neg logit contributions | |
and | -0.021 |
urry | -0.019 |
Follow | -0.018 |
Stay | -0.017 |
tum | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -8.02e-03 |
Pos logit contributions | |
and | +0.053 |
urry | +0.046 |
Follow | +0.045 |
Nice | +0.041 |
tum | +0.041 |
Neg logit contributions | |
bur | -0.046 |
fourteen | -0.045 |
eling | -0.044 |
ps | -0.044 |
blew | -0.043 |
Token | said |
Token ID | 531 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
and | +0.014 |
Follow | +0.013 |
urry | +0.013 |
tum | +0.012 |
Stay | +0.012 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.010 |
ps | -0.010 |
ible | -0.010 |
eling | -0.010 |
Token | " |
Token ID | 366 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
and | +0.027 |
Follow | +0.023 |
urry | +0.023 |
Today | +0.020 |
Nice | +0.019 |
Neg logit contributions | |
fourteen | -0.028 |
ps | -0.027 |
blew | -0.027 |
eling | -0.026 |
bur | -0.026 |
Token | each |
Token ID | 1123 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
lled | +0.026 |
bur | +0.026 |
fourteen | +0.025 |
ps | +0.025 |
eling | +0.025 |
Neg logit contributions | |
and | -0.032 |
urry | -0.028 |
tum | -0.027 |
Stay | -0.026 |
Follow | -0.026 |
Token | other |
Token ID | 584 |
Feature activation | -3.88e-03 |
Pos logit contributions | |
and | +0.024 |
urry | +0.020 |
Follow | +0.020 |
tum | +0.018 |
Stay | +0.018 |
Neg logit contributions | |
bur | -0.021 |
fourteen | -0.021 |
eling | -0.021 |
ps | -0.020 |
blew | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -7.88e-03 |
Pos logit contributions | |
and | +0.007 |
urry | +0.006 |
Follow | +0.006 |
tum | +0.006 |
Nice | +0.006 |
Neg logit contributions | |
bur | -0.005 |
fourteen | -0.005 |
eling | -0.005 |
ps | -0.005 |
blew | -0.004 |
Token | hug |
Token ID | 16225 |
Feature activation | -6.91e-03 |
Pos logit contributions | |
and | +0.013 |
urry | +0.011 |
Follow | +0.011 |
Nice | +0.011 |
Today | +0.010 |
Neg logit contributions | |
bur | -0.011 |
fourteen | -0.011 |
ps | -0.011 |
blew | -0.011 |
eling | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -5.74e-03 |
Pos logit contributions | |
and | +0.014 |
urry | +0.012 |
Follow | +0.012 |
Nice | +0.011 |
tum | +0.011 |
Neg logit contributions | |
bur | -0.008 |
fourteen | -0.007 |
ps | -0.007 |
eling | -0.007 |
blew | -0.007 |
Token | Fighting |
Token ID | 19098 |
Feature activation | -3.52e-03 |
Pos logit contributions | |
and | +0.008 |
urry | +0.007 |
rav | +0.006 |
Follow | +0.006 |
suits | +0.006 |
Neg logit contributions | |
fourteen | -0.009 |
bur | -0.009 |
ps | -0.009 |
ible | -0.009 |
abi | -0.008 |
Token | is |
Token ID | 318 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
and | +0.005 |
urry | +0.005 |
Follow | +0.005 |
tum | +0.004 |
Nice | +0.004 |
Neg logit contributions | |
ps | -0.005 |
blew | -0.005 |
bur | -0.005 |
fourteen | -0.005 |
mill | -0.005 |
Token | not |
Token ID | 407 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
bur | +0.018 |
ami | +0.018 |
rob | +0.018 |
ible | +0.017 |
IT | +0.017 |
Neg logit contributions | |
and | -0.014 |
urry | -0.013 |
Stay | -0.013 |
Follow | -0.013 |
of | -0.013 |
Token | good |
Token ID | 922 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
bur | +0.043 |
eling | +0.042 |
fourteen | +0.042 |
ami | +0.041 |
IT | +0.040 |
Neg logit contributions | |
and | -0.031 |
urry | -0.025 |
Follow | -0.024 |
Stay | -0.023 |
of | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -4.38e-03 |
Pos logit contributions | |
bur | +0.038 |
fourteen | +0.037 |
eling | +0.037 |
ps | +0.036 |
ck | +0.035 |
Neg logit contributions | |
and | -0.044 |
urry | -0.038 |
Follow | -0.038 |
tum | -0.035 |
Nice | -0.034 |
Token | It |
Token ID | 632 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
and | +0.005 |
urry | +0.004 |
Follow | +0.004 |
tum | +0.004 |
Nice | +0.004 |
Neg logit contributions | |
bur | -0.008 |
fourteen | -0.008 |
eling | -0.008 |
ps | -0.008 |
blew | -0.008 |
Token | Let |
Token ID | 5756 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
and | +0.011 |
Follow | +0.009 |
Go | +0.008 |
Stay | +0.008 |
urry | +0.008 |
Neg logit contributions | |
ps | -0.009 |
ries | -0.009 |
eling | -0.009 |
fourteen | -0.009 |
ible | -0.008 |
Token | me |
Token ID | 502 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
fourteen | +0.021 |
bur | +0.021 |
ps | +0.021 |
eling | +0.021 |
cking | +0.020 |
Neg logit contributions | |
and | -0.013 |
urry | -0.011 |
Follow | -0.010 |
tum | -0.009 |
Stay | -0.009 |
Token | go |
Token ID | 467 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
and | +0.062 |
Follow | +0.051 |
urry | +0.051 |
Nice | +0.049 |
tum | +0.047 |
Neg logit contributions | |
bur | -0.059 |
fourteen | -0.056 |
ps | -0.056 |
eling | -0.055 |
blew | -0.055 |
Token | ! |
Token ID | 0 |
Feature activation | -3.96e-03 |
Pos logit contributions | |
and | +0.018 |
urry | +0.015 |
Follow | +0.015 |
Nice | +0.014 |
tum | +0.014 |
Neg logit contributions | |
bur | -0.017 |
fourteen | -0.017 |
eling | -0.017 |
ps | -0.017 |
blew | -0.016 |
Token | Please |
Token ID | 4222 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
and | +0.005 |
halls | +0.004 |
urry | +0.004 |
suits | +0.004 |
tum | +0.004 |
Neg logit contributions | |
fourteen | -0.007 |
bur | -0.007 |
IT | -0.007 |
blew | -0.006 |
ps | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -6.83e-03 |
Pos logit contributions | |
and | +0.079 |
Follow | +0.063 |
urry | +0.063 |
Nice | +0.060 |
tum | +0.060 |
Neg logit contributions | |
bur | -0.072 |
ps | -0.068 |
fourteen | -0.067 |
blew | -0.066 |
eling | -0.066 |
Token | I |
Token ID | 314 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
and | +0.011 |
halls | +0.009 |
apping | +0.009 |
urry | +0.009 |
Follow | +0.009 |
Neg logit contributions | |
bur | -0.010 |
blew | -0.009 |
fourteen | -0.009 |
ps | -0.009 |
ck | -0.009 |
Token | want |
Token ID | 765 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
and | +0.040 |
Follow | +0.032 |
urry | +0.032 |
Nice | +0.029 |
tum | +0.029 |
Neg logit contributions | |
bur | -0.045 |
fourteen | -0.044 |
eling | -0.043 |
blew | -0.043 |
ps | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
and | +0.029 |
urry | +0.026 |
Follow | +0.026 |
apping | +0.024 |
Today | +0.023 |
Neg logit contributions | |
bur | -0.027 |
fourteen | -0.026 |
blew | -0.025 |
ps | -0.025 |
ck | -0.024 |
Token | go |
Token ID | 467 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
fourteen | +0.036 |
bur | +0.035 |
eling | +0.035 |
ible | +0.035 |
lled | +0.034 |
Neg logit contributions | |
and | -0.025 |
urry | -0.022 |
Follow | -0.021 |
Stay | -0.020 |
tum | -0.020 |
Token | home |
Token ID | 1363 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
and | +0.021 |
urry | +0.018 |
Follow | +0.018 |
tum | +0.016 |
Nice | +0.016 |
Neg logit contributions | |
bur | -0.019 |
fourteen | -0.018 |
ps | -0.018 |
eling | -0.018 |
blew | -0.017 |
Token | day |
Token ID | 1110 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
and | +0.014 |
Follow | +0.013 |
urry | +0.012 |
ooters | +0.011 |
vere | +0.011 |
Neg logit contributions | |
bur | -0.017 |
blew | -0.017 |
fourteen | -0.017 |
ible | -0.016 |
eling | -0.016 |
Token | on |
Token ID | 319 |
Feature activation | -4.98e-03 |
Pos logit contributions | |
bur | +0.017 |
fourteen | +0.017 |
blew | +0.017 |
ps | +0.017 |
eling | +0.016 |
Neg logit contributions | |
and | -0.031 |
Follow | -0.027 |
urry | -0.026 |
Nice | -0.025 |
engu | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -5.02e-03 |
Pos logit contributions | |
and | +0.006 |
urry | +0.005 |
Follow | +0.005 |
Nice | +0.004 |
tum | +0.004 |
Neg logit contributions | |
bur | -0.010 |
fourteen | -0.009 |
eling | -0.009 |
ps | -0.009 |
blew | -0.009 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
and | +0.007 |
of | +0.006 |
tum | +0.006 |
urry | +0.006 |
birthday | +0.006 |
Neg logit contributions | |
bur | -0.009 |
ps | -0.009 |
eling | -0.009 |
cking | -0.008 |
ible | -0.008 |
Token | learned |
Token ID | 4499 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
fourteen | +0.022 |
eling | +0.022 |
bur | +0.022 |
cking | +0.021 |
ck | +0.021 |
Neg logit contributions | |
and | -0.023 |
urry | -0.019 |
Follow | -0.019 |
tum | -0.018 |
of | -0.018 |
Token | that |
Token ID | 326 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
and | +0.030 |
of | +0.024 |
urry | +0.022 |
story | +0.022 |
birthday | +0.021 |
Neg logit contributions | |
bur | -0.042 |
eling | -0.042 |
ible | -0.041 |
fourteen | -0.040 |
ps | -0.039 |
Token | it |
Token ID | 340 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
and | +0.026 |
Follow | +0.025 |
suits | +0.025 |
urry | +0.025 |
ari | +0.024 |
Neg logit contributions | |
blew | -0.011 |
her | -0.010 |
Why | -0.009 |
bur | -0.009 |
mill | -0.009 |
Token | 's |
Token ID | 338 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
and | +0.011 |
urry | +0.009 |
Follow | +0.009 |
tum | +0.008 |
Nice | +0.008 |
Neg logit contributions | |
bur | -0.015 |
fourteen | -0.015 |
eling | -0.014 |
ps | -0.014 |
blew | -0.014 |
Token | okay |
Token ID | 8788 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
and | +0.020 |
urry | +0.017 |
Follow | +0.015 |
Stay | +0.014 |
story | +0.014 |
Neg logit contributions | |
fourteen | -0.044 |
ries | -0.042 |
eling | -0.042 |
rob | -0.042 |
ps | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
and | +0.031 |
Follow | +0.029 |
urry | +0.028 |
vere | +0.027 |
engu | +0.026 |
Neg logit contributions | |
blew | -0.027 |
bur | -0.027 |
ps | -0.025 |
fourteen | -0.025 |
ible | -0.025 |
Token | make |
Token ID | 787 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
bur | +0.032 |
fourteen | +0.032 |
ps | +0.031 |
eling | +0.031 |
lled | +0.030 |
Neg logit contributions | |
and | -0.031 |
urry | -0.028 |
Follow | -0.026 |
Stay | -0.024 |
tum | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +5.91e-04 |
Pos logit contributions | |
bur | +0.021 |
ible | +0.020 |
fourteen | +0.020 |
eling | +0.020 |
rob | +0.019 |
Neg logit contributions | |
and | -0.025 |
of | -0.023 |
urry | -0.022 |
tum | -0.022 |
Today | -0.022 |
Token | cold |
Token ID | 4692 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
bur | +0.001 |
ps | +0.001 |
ck | +0.001 |
blew | +0.001 |
cking | +0.001 |
Neg logit contributions | |
and | -0.001 |
urry | -0.001 |
apping | -0.001 |
Follow | -0.001 |
rav | -0.001 |
Token | waves |
Token ID | 9813 |
Feature activation | -8.11e-03 |
Pos logit contributions | |
and | +0.050 |
Follow | +0.047 |
urry | +0.045 |
vere | +0.043 |
rav | +0.042 |
Neg logit contributions | |
bur | -0.029 |
fourteen | -0.026 |
blew | -0.026 |
ps | -0.026 |
ible | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +1.80e-03 |
Pos logit contributions | |
and | +0.016 |
Follow | +0.014 |
urry | +0.014 |
Nice | +0.013 |
tum | +0.013 |
Neg logit contributions | |
bur | -0.009 |
fourteen | -0.009 |
ps | -0.008 |
eling | -0.008 |
blew | -0.008 |
Token | After |
Token ID | 2293 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
IT | +0.002 |
fourteen | +0.002 |
her | +0.002 |
ps | +0.002 |
pressed | +0.002 |
Neg logit contributions | |
and | -0.003 |
urry | -0.003 |
vere | -0.003 |
ye | -0.003 |
Follow | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
bur | +0.012 |
fourteen | +0.012 |
eling | +0.011 |
ps | +0.011 |
blew | +0.011 |
Neg logit contributions | |
and | -0.030 |
urry | -0.027 |
Follow | -0.027 |
Nice | -0.025 |
tum | -0.025 |
Token | while |
Token ID | 981 |
Feature activation | -2.43e-03 |
Pos logit contributions | |
and | +0.011 |
Follow | +0.010 |
urry | +0.009 |
Go | +0.009 |
Stay | +0.008 |
Neg logit contributions | |
bur | -0.011 |
fourteen | -0.011 |
blew | -0.010 |
eling | -0.010 |
ck | -0.010 |
Token | of |
Token ID | 286 |
Feature activation | +2.01e-03 |
Pos logit contributions | |
and | +0.004 |
Follow | +0.003 |
urry | +0.003 |
aring | +0.003 |
ooters | +0.003 |
Neg logit contributions | |
fourteen | -0.004 |
bur | -0.004 |
blew | -0.004 |
ps | -0.004 |
eling | -0.004 |
Token | ro |
Token ID | 686 |
Feature activation | -2.55e-03 |
Pos logit contributions | |
bur | +0.002 |
fourteen | +0.002 |
ible | +0.002 |
IT | +0.002 |
eling | +0.002 |
Neg logit contributions | |
and | -0.004 |
urry | -0.003 |
Follow | -0.003 |
of | -0.003 |
story | -0.003 |
Token | wing |
Token ID | 5469 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
and | +0.003 |
Nice | +0.003 |
Follow | +0.003 |
birthday | +0.003 |
urry | +0.003 |
Neg logit contributions | |
bur | -0.005 |
eling | -0.005 |
ck | -0.004 |
ps | -0.004 |
lled | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -8.34e-03 |
Pos logit contributions | |
and | +0.007 |
Follow | +0.006 |
urry | +0.006 |
aring | +0.005 |
Nice | +0.005 |
Neg logit contributions | |
bur | -0.007 |
fourteen | -0.007 |
ps | -0.007 |
blew | -0.007 |
eling | -0.007 |
Token | her |
Token ID | 607 |
Feature activation | +1.50e-03 |
Pos logit contributions | |
and | +0.054 |
Follow | +0.046 |
urry | +0.045 |
of | +0.045 |
tum | +0.045 |
Neg logit contributions | |
bur | -0.031 |
eling | -0.029 |
fourteen | -0.029 |
blew | -0.028 |
IT | -0.028 |
Token | folder |
Token ID | 9483 |
Feature activation | +5.95e-03 |
Pos logit contributions | |
bur | +0.002 |
fourteen | +0.002 |
eling | +0.002 |
ps | +0.002 |
blew | +0.002 |
Neg logit contributions | |
and | -0.002 |
urry | -0.002 |
Follow | -0.002 |
tum | -0.002 |
Nice | -0.002 |
Token | ! |
Token ID | 0 |
Feature activation | -6.39e-03 |
Pos logit contributions | |
bur | +0.007 |
fourteen | +0.007 |
ps | +0.007 |
blew | +0.006 |
eling | +0.006 |
Neg logit contributions | |
and | -0.011 |
urry | -0.010 |
Follow | -0.010 |
Nice | -0.009 |
Stay | -0.009 |
Token | It |
Token ID | 632 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
and | +0.010 |
urry | +0.009 |
Follow | +0.008 |
tum | +0.008 |
rav | +0.008 |
Neg logit contributions | |
bur | -0.009 |
fourteen | -0.009 |
blew | -0.008 |
ps | -0.008 |
IT | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +5.39e-03 |
Pos logit contributions | |
and | +0.018 |
Follow | +0.016 |
urry | +0.015 |
of | +0.014 |
case | +0.014 |
Neg logit contributions | |
bur | -0.023 |
fourteen | -0.022 |
eling | -0.021 |
ps | -0.021 |
cking | -0.021 |
Token | tucked |
Token ID | 29779 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
bur | +0.007 |
eling | +0.007 |
fourteen | +0.007 |
ible | +0.007 |
ps | +0.007 |
Neg logit contributions | |
and | -0.009 |
urry | -0.008 |
Follow | -0.008 |
tum | -0.008 |
Nice | -0.007 |
Token | away |
Token ID | 1497 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
and | +0.128 |
urry | +0.114 |
Follow | +0.113 |
tum | +0.104 |
Nice | +0.104 |
Neg logit contributions | |
bur | -0.066 |
fourteen | -0.064 |
eling | -0.062 |
ps | -0.062 |
blew | -0.060 |
Token | in |
Token ID | 287 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
bur | +0.021 |
fourteen | +0.021 |
ps | +0.020 |
eling | +0.020 |
blew | +0.020 |
Neg logit contributions | |
and | -0.025 |
urry | -0.022 |
Follow | -0.022 |
Nice | -0.020 |
tum | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -1.38e-03 |
Pos logit contributions | |
and | +0.076 |
urry | +0.067 |
Follow | +0.066 |
tum | +0.063 |
Nice | +0.062 |
Neg logit contributions | |
bur | -0.035 |
fourteen | -0.034 |
eling | -0.033 |
ps | -0.032 |
blew | -0.031 |
Token | back |
Token ID | 736 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
and | +0.002 |
urry | +0.002 |
rav | +0.002 |
aring | +0.002 |
apping | +0.002 |
Neg logit contributions | |
ps | -0.002 |
blew | -0.002 |
fourteen | -0.002 |
her | -0.002 |
afar | -0.002 |
Token | of |
Token ID | 286 |
Feature activation | -1.32e-03 |
Pos logit contributions | |
and | +0.039 |
Follow | +0.038 |
urry | +0.038 |
rav | +0.037 |
vere | +0.035 |
Neg logit contributions | |
her | -0.046 |
ps | -0.044 |
ries | -0.042 |
bur | -0.042 |
colored | -0.042 |
Token | bust |
Token ID | 13076 |
Feature activation | +8.14e-03 |
Pos logit contributions | |
and | +0.004 |
urry | +0.003 |
Follow | +0.003 |
apping | +0.003 |
Today | +0.003 |
Neg logit contributions | |
blew | -0.002 |
eling | -0.002 |
bur | -0.002 |
fourteen | -0.002 |
lled | -0.002 |
Token | le |
Token ID | 293 |
Feature activation | +2.74e-03 |
Pos logit contributions | |
bur | +0.011 |
fourteen | +0.011 |
abi | +0.010 |
ps | +0.009 |
IT | +0.009 |
Neg logit contributions | |
and | -0.015 |
of | -0.012 |
Follow | -0.012 |
urry | -0.012 |
halls | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +6.18e-03 |
Pos logit contributions | |
bur | +0.004 |
fourteen | +0.003 |
eling | +0.003 |
ps | +0.003 |
blew | +0.003 |
Neg logit contributions | |
and | -0.005 |
urry | -0.004 |
Follow | -0.004 |
tum | -0.004 |
Nice | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
bur | +0.010 |
fourteen | +0.010 |
eling | +0.010 |
ps | +0.010 |
ck | +0.010 |
Neg logit contributions | |
and | -0.009 |
urry | -0.008 |
Follow | -0.007 |
Nice | -0.007 |
tum | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.73e-03 |
Pos logit contributions | |
bur | +0.002 |
ck | +0.002 |
blew | +0.002 |
fourteen | +0.002 |
lled | +0.002 |
Neg logit contributions | |
and | -0.003 |
Follow | -0.003 |
tum | -0.003 |
urry | -0.003 |
birthday | -0.003 |
Token | G |
Token ID | 38 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
ps | +0.003 |
eling | +0.003 |
bur | +0.003 |
fourteen | +0.003 |
ries | +0.003 |
Neg logit contributions | |
and | -0.005 |
urry | -0.005 |
Follow | -0.004 |
Nice | -0.004 |
Today | -0.004 |
Token | illy |
Token ID | 6548 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
and | +0.100 |
urry | +0.087 |
Follow | +0.080 |
rav | +0.074 |
Today | +0.074 |
Neg logit contributions | |
fourteen | -0.083 |
ps | -0.080 |
blew | -0.080 |
eling | -0.079 |
bur | -0.079 |
Token | and |
Token ID | 290 |
Feature activation | -9.58e-04 |
Pos logit contributions | |
and | +0.054 |
urry | +0.048 |
Follow | +0.047 |
tum | +0.044 |
Nice | +0.044 |
Neg logit contributions | |
bur | -0.030 |
fourteen | -0.029 |
eling | -0.028 |
ps | -0.028 |
blew | -0.027 |
Token | Jack |
Token ID | 3619 |
Feature activation | +7.62e-03 |
Pos logit contributions | |
and | +0.001 |
urry | +0.001 |
Today | +0.001 |
Nice | +0.001 |
Follow | +0.001 |
Neg logit contributions | |
cking | -0.002 |
eling | -0.002 |
fourteen | -0.002 |
ps | -0.002 |
ck | -0.002 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
fourteen | +0.010 |
ps | +0.010 |
bur | +0.010 |
eling | +0.010 |
abi | +0.010 |
Neg logit contributions | |
and | -0.013 |
urry | -0.012 |
Follow | -0.011 |
Stay | -0.010 |
of | -0.010 |
Token | up |
Token ID | 510 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
bur | +0.030 |
eling | +0.029 |
fourteen | +0.029 |
ps | +0.029 |
blew | +0.028 |
Neg logit contributions | |
and | -0.043 |
urry | -0.037 |
Follow | -0.037 |
Nice | -0.034 |
tum | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -7.55e-03 |
Pos logit contributions | |
blew | +0.040 |
fourteen | +0.037 |
eling | +0.036 |
et | +0.036 |
ps | +0.036 |
Neg logit contributions | |
and | -0.068 |
Follow | -0.062 |
urry | -0.062 |
Nice | -0.061 |
rav | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | -9.22e-03 |
Pos logit contributions | |
and | +0.017 |
aring | +0.016 |
urry | +0.015 |
Follow | +0.015 |
Nice | +0.014 |
Neg logit contributions | |
blew | -0.007 |
her | -0.006 |
ries | -0.006 |
fourteen | -0.006 |
eling | -0.006 |
Token | little |
Token ID | 1310 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
and | +0.018 |
Follow | +0.016 |
urry | +0.016 |
Go | +0.015 |
Nice | +0.014 |
Neg logit contributions | |
blew | -0.011 |
bur | -0.010 |
sc | -0.010 |
fourteen | -0.009 |
abi | -0.009 |
Token | girl |
Token ID | 2576 |
Feature activation | -9.57e-03 |
Pos logit contributions | |
blew | +0.013 |
bur | +0.013 |
ible | +0.012 |
fourteen | +0.012 |
mill | +0.012 |
Neg logit contributions | |
and | -0.022 |
urry | -0.019 |
Follow | -0.018 |
vere | -0.017 |
Nice | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +6.76e-03 |
Pos logit contributions | |
and | +0.018 |
urry | +0.016 |
Follow | +0.016 |
vere | +0.015 |
Nice | +0.015 |
Neg logit contributions | |
blew | -0.011 |
bur | -0.011 |
ck | -0.010 |
eling | -0.010 |
ible | -0.010 |
Token | sitting |
Token ID | 5586 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
bur | +0.011 |
fourteen | +0.011 |
eling | +0.011 |
ps | +0.011 |
ible | +0.011 |
Neg logit contributions | |
and | -0.010 |
urry | -0.008 |
Follow | -0.008 |
tum | -0.007 |
apping | -0.007 |
Token | on |
Token ID | 319 |
Feature activation | -7.13e-03 |
Pos logit contributions | |
and | +0.103 |
urry | +0.089 |
Follow | +0.088 |
tum | +0.080 |
Nice | +0.080 |
Neg logit contributions | |
bur | -0.090 |
fourteen | -0.088 |
eling | -0.086 |
ps | -0.086 |
blew | -0.084 |
Token | a |
Token ID | 257 |
Feature activation | -9.15e-03 |
Pos logit contributions | |
and | +0.010 |
birthday | +0.010 |
story | +0.009 |
School | +0.009 |
halls | +0.009 |
Neg logit contributions | |
cking | -0.011 |
eling | -0.011 |
ck | -0.011 |
bur | -0.011 |
fourteen | -0.010 |
Token | seat |
Token ID | 5852 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
and | +0.014 |
Follow | +0.012 |
urry | +0.012 |
Nice | +0.011 |
rav | +0.011 |
Neg logit contributions | |
bur | -0.014 |
fourteen | -0.014 |
blew | -0.014 |
eling | -0.013 |
ck | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +2.17e-03 |
Pos logit contributions | |
bur | +0.026 |
fourteen | +0.026 |
eling | +0.026 |
ps | +0.025 |
ck | +0.025 |
Neg logit contributions | |
and | -0.031 |
urry | -0.027 |
Follow | -0.026 |
tum | -0.025 |
apping | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
her | +0.002 |
bur | +0.002 |
IT | +0.002 |
fourteen | +0.002 |
blew | +0.002 |
Neg logit contributions | |
and | -0.004 |
urry | -0.004 |
vere | -0.004 |
rav | -0.004 |
Follow | -0.004 |
Token | how |
Token ID | 703 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
and | +0.039 |
of | +0.034 |
story | +0.032 |
Stay | +0.031 |
School | +0.030 |
Neg logit contributions | |
bur | -0.033 |
cking | -0.031 |
eling | -0.031 |
ck | -0.030 |
fourteen | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +4.76e-03 |
Pos logit contributions | |
bur | +0.029 |
fourteen | +0.029 |
eling | +0.028 |
ps | +0.028 |
blew | +0.028 |
Neg logit contributions | |
and | -0.023 |
urry | -0.019 |
Follow | -0.019 |
Nice | -0.017 |
tum | -0.017 |
Token | After |
Token ID | 2293 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
bur | +0.009 |
eling | +0.009 |
ck | +0.009 |
fourteen | +0.009 |
cking | +0.009 |
Neg logit contributions | |
and | -0.006 |
Follow | -0.005 |
urry | -0.005 |
Nice | -0.004 |
Today | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
bur | +0.026 |
ible | +0.024 |
lled | +0.023 |
ps | +0.023 |
IT | +0.023 |
Neg logit contributions | |
and | -0.031 |
urry | -0.027 |
story | -0.027 |
of | -0.027 |
birthday | -0.026 |
Token | few |
Token ID | 1178 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
and | +0.011 |
urry | +0.009 |
Follow | +0.009 |
Nice | +0.008 |
tum | +0.008 |
Neg logit contributions | |
bur | -0.014 |
fourteen | -0.014 |
eling | -0.014 |
ps | -0.013 |
blew | -0.013 |
Token | days |
Token ID | 1528 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
bur | +0.036 |
fourteen | +0.034 |
ps | +0.034 |
eling | +0.034 |
blew | +0.033 |
Neg logit contributions | |
and | -0.034 |
urry | -0.029 |
Follow | -0.029 |
tum | -0.026 |
Nice | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | -7.27e-03 |
Pos logit contributions | |
and | +0.074 |
urry | +0.063 |
Follow | +0.057 |
of | +0.055 |
tum | +0.051 |
Neg logit contributions | |
bur | -0.135 |
ps | -0.133 |
fourteen | -0.130 |
eling | -0.129 |
cking | -0.128 |
Token | Stamp |
Token ID | 40694 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
and | +0.011 |
of | +0.010 |
School | +0.009 |
tum | +0.009 |
urry | +0.009 |
Neg logit contributions | |
ps | -0.011 |
cking | -0.011 |
bur | -0.011 |
eling | -0.011 |
ck | -0.011 |
Token | started |
Token ID | 2067 |
Feature activation | +8.78e-04 |
Pos logit contributions | |
and | +0.025 |
Follow | +0.022 |
urry | +0.022 |
of | +0.021 |
apping | +0.020 |
Neg logit contributions | |
bur | -0.022 |
fourteen | -0.021 |
eling | -0.021 |
cking | -0.020 |
IT | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
bur | +0.002 |
fourteen | +0.002 |
eling | +0.002 |
ck | +0.002 |
ps | +0.002 |
Neg logit contributions | |
and | -0.001 |
tum | -0.001 |
story | -0.001 |
of | -0.001 |
urry | -0.001 |
Token | like |
Token ID | 588 |
Feature activation | -1.20e-03 |
Pos logit contributions | |
fourteen | +0.035 |
bur | +0.035 |
ries | +0.033 |
ps | +0.033 |
rob | +0.033 |
Neg logit contributions | |
and | -0.024 |
story | -0.023 |
urry | -0.021 |
birthday | -0.021 |
Follow | -0.021 |
Token | bird |
Token ID | 6512 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
bur | +0.011 |
fourteen | +0.011 |
eling | +0.010 |
ps | +0.010 |
lled | +0.010 |
Neg logit contributions | |
and | -0.011 |
urry | -0.010 |
Follow | -0.010 |
tum | -0.009 |
story | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
fourteen | +0.014 |
bur | +0.014 |
eling | +0.013 |
ps | +0.013 |
blew | +0.013 |
Neg logit contributions | |
and | -0.014 |
urry | -0.012 |
Follow | -0.012 |
tum | -0.011 |
Nice | -0.010 |
Token | After |
Token ID | 2293 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
bur | +0.019 |
fourteen | +0.018 |
eling | +0.018 |
ps | +0.018 |
blew | +0.018 |
Neg logit contributions | |
and | -0.018 |
urry | -0.016 |
Follow | -0.016 |
Nice | -0.014 |
tum | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -4.12e-03 |
Pos logit contributions | |
bur | +0.024 |
ible | +0.022 |
eling | +0.022 |
IT | +0.022 |
fourteen | +0.022 |
Neg logit contributions | |
and | -0.036 |
urry | -0.032 |
of | -0.031 |
Follow | -0.031 |
tum | -0.031 |
Token | few |
Token ID | 1178 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
and | +0.006 |
Follow | +0.006 |
urry | +0.005 |
Go | +0.005 |
Stay | +0.005 |
Neg logit contributions | |
fourteen | -0.006 |
blew | -0.006 |
bur | -0.006 |
ck | -0.006 |
eling | -0.006 |
Token | weeks |
Token ID | 2745 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
eling | +0.037 |
fourteen | +0.036 |
ps | +0.036 |
ck | +0.035 |
bur | +0.035 |
Neg logit contributions | |
and | -0.037 |
urry | -0.031 |
Follow | -0.030 |
Nice | -0.028 |
apping | -0.027 |
Token | of |
Token ID | 286 |
Feature activation | +4.73e-03 |
Pos logit contributions | |
and | +0.075 |
urry | +0.061 |
Follow | +0.060 |
tum | +0.051 |
Nice | +0.051 |
Neg logit contributions | |
bur | -0.123 |
fourteen | -0.121 |
eling | -0.119 |
ps | -0.118 |
blew | -0.117 |
Token | rest |
Token ID | 1334 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
bur | +0.006 |
fourteen | +0.006 |
eling | +0.006 |
ps | +0.006 |
blew | +0.006 |
Neg logit contributions | |
and | -0.008 |
urry | -0.007 |
Follow | -0.007 |
Nice | -0.007 |
tum | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -3.20e-03 |
Pos logit contributions | |
and | +0.017 |
Follow | +0.014 |
urry | +0.014 |
Nice | +0.012 |
School | +0.012 |
Neg logit contributions | |
bur | -0.023 |
fourteen | -0.023 |
eling | -0.023 |
ps | -0.022 |
blew | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +6.05e-03 |
Pos logit contributions | |
and | +0.006 |
urry | +0.005 |
Follow | +0.005 |
tum | +0.005 |
of | +0.005 |
Neg logit contributions | |
bur | -0.004 |
ps | -0.004 |
eling | -0.004 |
fourteen | -0.004 |
ck | -0.004 |
Token | bird |
Token ID | 6512 |
Feature activation | +8.99e-03 |
Pos logit contributions | |
bur | +0.009 |
fourteen | +0.008 |
eling | +0.008 |
ps | +0.008 |
blew | +0.008 |
Neg logit contributions | |
and | -0.010 |
urry | -0.009 |
Follow | -0.009 |
tum | -0.008 |
Nice | -0.008 |