Token | brother |
Token ID | 3956 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
amaz | +0.013 |
gasp | +0.012 |
angles | +0.011 |
ill | +0.010 |
horror | +0.010 |
Neg logit contributions | |
Nut | -0.031 |
Rust | -0.028 |
Nut | -0.027 |
br | -0.027 |
con | -0.027 |
Token | would |
Token ID | 561 |
Feature activation | -3.65e-03 |
Pos logit contributions | |
Nut | +0.073 |
Rust | +0.064 |
br | +0.062 |
rob | +0.062 |
str | +0.062 |
Neg logit contributions | |
amaz | -0.040 |
gasp | -0.036 |
angles | -0.034 |
horror | -0.033 |
ill | -0.032 |
Token | wear |
Token ID | 5806 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
angles | +0.005 |
horror | +0.004 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.008 |
Rust | -0.007 |
br | -0.007 |
str | -0.007 |
rob | -0.007 |
Token | their |
Token ID | 511 |
Feature activation | -8.98e-03 |
Pos logit contributions | |
Nut | +0.076 |
str | +0.069 |
age | +0.067 |
cone | +0.064 |
Rust | +0.064 |
Neg logit contributions | |
amaz | -0.039 |
bowls | -0.033 |
gasp | -0.033 |
faces | -0.031 |
cups | -0.030 |
Token | bath |
Token ID | 7837 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
amaz | +0.015 |
gasp | +0.015 |
ill | +0.014 |
angles | +0.014 |
horror | +0.013 |
Neg logit contributions | |
Nut | -0.017 |
Rust | -0.015 |
br | -0.015 |
rob | -0.014 |
Nut | -0.014 |
Token | ro |
Token ID | 305 |
Feature activation | +8.65e-02 |
Pos logit contributions | |
Nut | +0.035 |
Rust | +0.029 |
str | +0.029 |
Nut | +0.028 |
Silver | +0.028 |
Neg logit contributions | |
amaz | -0.029 |
gasp | -0.027 |
angles | -0.026 |
bowls | -0.026 |
ill | -0.025 |
Token | bes |
Token ID | 12636 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
Nut | +0.169 |
Nut | +0.154 |
cone | +0.146 |
Sy | +0.146 |
Rust | +0.142 |
Neg logit contributions | |
faces | -0.124 |
cloth | -0.123 |
gasp | -0.120 |
amaz | -0.113 |
piles | -0.113 |
Token | together |
Token ID | 1978 |
Feature activation | -6.98e-04 |
Pos logit contributions | |
Nut | +0.041 |
str | +0.034 |
Rust | +0.034 |
rob | +0.033 |
Sy | +0.033 |
Neg logit contributions | |
amaz | -0.025 |
gasp | -0.024 |
ill | -0.024 |
bowls | -0.023 |
angles | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +8.38e-03 |
Pos logit contributions | |
amaz | +0.001 |
gasp | +0.001 |
angles | +0.001 |
horror | +0.001 |
ill | +0.001 |
Neg logit contributions | |
Nut | -0.002 |
Rust | -0.001 |
rob | -0.001 |
str | -0.001 |
Nut | -0.001 |
Token | play |
Token ID | 711 |
Feature activation | +9.26e-03 |
Pos logit contributions | |
Nut | +0.017 |
Rust | +0.015 |
br | +0.015 |
Nut | +0.015 |
str | +0.014 |
Neg logit contributions | |
amaz | -0.014 |
gasp | -0.013 |
angles | -0.012 |
horror | -0.012 |
ill | -0.011 |
Token | games |
Token ID | 1830 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
Nut | +0.020 |
str | +0.017 |
Rust | +0.017 |
Nut | +0.016 |
con | +0.016 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.014 |
angles | -0.013 |
horror | -0.013 |
bowls | -0.012 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.42e-03 |
Pos logit contributions | |
Nut | +0.018 |
Rust | +0.016 |
rob | +0.016 |
br | +0.016 |
str | +0.015 |
Neg logit contributions | |
amaz | -0.009 |
gasp | -0.008 |
horror | -0.007 |
angles | -0.007 |
ill | -0.007 |
Token | 's |
Token ID | 338 |
Feature activation | -2.21e-03 |
Pos logit contributions | |
Nut | +0.015 |
Rust | +0.013 |
br | +0.013 |
rob | +0.013 |
str | +0.013 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.008 |
horror | -0.007 |
angles | -0.007 |
ill | -0.007 |
Token | mother |
Token ID | 2802 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
amaz | +0.003 |
gasp | +0.002 |
angles | +0.002 |
horror | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.006 |
Rust | -0.005 |
str | -0.005 |
br | -0.005 |
Nut | -0.005 |
Token | wrapped |
Token ID | 12908 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
amaz | +0.016 |
gasp | +0.015 |
horror | +0.014 |
angles | +0.013 |
ill | +0.013 |
Neg logit contributions | |
Nut | -0.026 |
Rust | -0.022 |
rob | -0.022 |
br | -0.022 |
str | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
Nut | +0.031 |
str | +0.031 |
br | +0.030 |
age | +0.028 |
rob | +0.028 |
Neg logit contributions | |
amaz | -0.014 |
angles | -0.013 |
gasp | -0.013 |
eyes | -0.013 |
Mom | -0.013 |
Token | band |
Token ID | 4097 |
Feature activation | +8.14e-02 |
Pos logit contributions | |
Nut | +0.031 |
rob | +0.030 |
con | +0.028 |
Sy | +0.028 |
Rust | +0.027 |
Neg logit contributions | |
amaz | -0.018 |
gasp | -0.018 |
wish | -0.017 |
spit | -0.016 |
bowls | -0.016 |
Token | age |
Token ID | 496 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
Nut | +0.130 |
rob | +0.130 |
str | +0.120 |
hawk | +0.118 |
Barkley | +0.117 |
Neg logit contributions | |
of | -0.138 |
gasp | -0.136 |
eye | -0.134 |
hands | -0.130 |
face | -0.129 |
Token | around |
Token ID | 1088 |
Feature activation | +9.76e-03 |
Pos logit contributions | |
Nut | +0.101 |
rob | +0.090 |
str | +0.089 |
Rust | +0.085 |
con | +0.084 |
Neg logit contributions | |
gasp | -0.035 |
amaz | -0.034 |
horror | -0.034 |
open | -0.032 |
ill | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | +7.73e-03 |
Pos logit contributions | |
Nut | +0.025 |
str | +0.023 |
br | +0.023 |
age | +0.023 |
rob | +0.023 |
Neg logit contributions | |
amaz | -0.011 |
gasp | -0.009 |
horror | -0.009 |
ill | -0.009 |
angles | -0.009 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
Nut | +0.019 |
rob | +0.018 |
br | +0.018 |
Rust | +0.017 |
str | +0.017 |
Neg logit contributions | |
amaz | -0.009 |
gasp | -0.009 |
eyes | -0.009 |
faces | -0.008 |
face | -0.008 |
Token | 's |
Token ID | 338 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
Nut | +0.047 |
rob | +0.043 |
br | +0.042 |
str | +0.041 |
Rust | +0.041 |
Neg logit contributions | |
amaz | -0.020 |
gasp | -0.018 |
horror | -0.017 |
bowls | -0.016 |
angles | -0.015 |
Token | so |
Token ID | 523 |
Feature activation | +9.67e-05 |
Pos logit contributions | |
Nut | +0.048 |
Rust | +0.041 |
str | +0.040 |
Silver | +0.040 |
rob | +0.039 |
Neg logit contributions | |
amaz | -0.030 |
gasp | -0.029 |
ill | -0.028 |
angles | -0.027 |
bowls | -0.025 |
Token | she |
Token ID | 673 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
Nut | +0.000 |
Rust | +0.000 |
rob | +0.000 |
str | +0.000 |
br | +0.000 |
Neg logit contributions | |
amaz | -0.000 |
gasp | -0.000 |
angles | -0.000 |
horror | -0.000 |
ill | -0.000 |
Token | could |
Token ID | 714 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
amaz | +0.015 |
gasp | +0.014 |
angles | +0.013 |
horror | +0.012 |
ill | +0.012 |
Neg logit contributions | |
Nut | -0.023 |
Rust | -0.019 |
str | -0.019 |
rob | -0.019 |
br | -0.019 |
Token | run |
Token ID | 1057 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
amaz | +0.028 |
angles | +0.025 |
horror | +0.025 |
gasp | +0.024 |
bowls | +0.023 |
Neg logit contributions | |
Nut | -0.035 |
br | -0.031 |
age | -0.031 |
Rust | -0.030 |
K | -0.030 |
Token | fr |
Token ID | 1216 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
Nut | +0.061 |
str | +0.059 |
wl | +0.055 |
Silver | +0.054 |
Mic | +0.053 |
Neg logit contributions | |
gasp | -0.020 |
decide | -0.019 |
angles | -0.018 |
ill | -0.017 |
open | -0.017 |
Token | ree |
Token ID | 631 |
Feature activation | +8.11e-02 |
Pos logit contributions | |
amaz | +0.035 |
gasp | +0.034 |
ill | +0.033 |
angles | +0.033 |
reflections | +0.030 |
Neg logit contributions | |
Nut | -0.037 |
Rust | -0.032 |
Nut | -0.030 |
str | -0.030 |
Silver | -0.030 |
Token | ly |
Token ID | 306 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Nut | +0.140 |
str | +0.124 |
Nut | +0.118 |
Mic | +0.116 |
Sy | +0.116 |
Neg logit contributions | |
gasp | -0.136 |
faces | -0.129 |
ill | -0.120 |
gasped | -0.120 |
amaz | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Nut | +0.054 |
str | +0.046 |
Rust | +0.046 |
Nut | +0.045 |
Silver | +0.044 |
Neg logit contributions | |
amaz | -0.033 |
gasp | -0.030 |
angles | -0.030 |
horror | -0.027 |
bowls | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
Nut | +0.026 |
rob | +0.023 |
br | +0.023 |
Rust | +0.022 |
str | +0.022 |
Neg logit contributions | |
amaz | -0.016 |
gasp | -0.014 |
ill | -0.014 |
Mom | -0.013 |
angles | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +4.46e-03 |
Pos logit contributions | |
Nut | +0.033 |
br | +0.029 |
con | +0.028 |
Sy | +0.028 |
rob | +0.028 |
Neg logit contributions | |
amaz | -0.020 |
gasp | -0.020 |
ill | -0.020 |
angles | -0.018 |
gasped | -0.018 |
Token | The |
Token ID | 464 |
Feature activation | +1.60e-03 |
Pos logit contributions | |
con | +0.008 |
Nut | +0.008 |
rob | +0.008 |
Nut | +0.007 |
str | +0.007 |
Neg logit contributions | |
amaz | -0.007 |
ill | -0.007 |
urious | -0.007 |
decide | -0.007 |
wonder | -0.007 |
Token | so |
Token ID | 523 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
Nut | +0.056 |
rob | +0.049 |
Rust | +0.047 |
str | +0.047 |
br | +0.046 |
Neg logit contributions | |
amaz | -0.051 |
gasp | -0.047 |
horror | -0.044 |
angles | -0.044 |
ill | -0.042 |
Token | he |
Token ID | 339 |
Feature activation | +4.48e-03 |
Pos logit contributions | |
Nut | +0.019 |
Rust | +0.016 |
br | +0.016 |
str | +0.015 |
rob | +0.015 |
Neg logit contributions | |
amaz | -0.021 |
gasp | -0.020 |
angles | -0.019 |
horror | -0.019 |
ill | -0.018 |
Token | promised |
Token ID | 8072 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
Nut | +0.011 |
rob | +0.009 |
Rust | +0.009 |
br | +0.009 |
str | +0.009 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.005 |
angles | -0.005 |
horror | -0.005 |
ill | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +7.61e-03 |
Pos logit contributions | |
rob | +0.059 |
Nut | +0.059 |
br | +0.055 |
Rust | +0.053 |
str | +0.053 |
Neg logit contributions | |
amaz | -0.035 |
Dad | -0.034 |
ill | -0.033 |
decide | -0.033 |
gasp | -0.033 |
Token | be |
Token ID | 307 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
Nut | +0.018 |
Rust | +0.015 |
rob | +0.015 |
br | +0.015 |
str | +0.015 |
Neg logit contributions | |
amaz | -0.011 |
gasp | -0.010 |
angles | -0.009 |
horror | -0.009 |
ill | -0.009 |
Token | Beat |
Token ID | 12568 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
Nut | +0.049 |
rob | +0.042 |
Rust | +0.041 |
con | +0.041 |
str | +0.041 |
Neg logit contributions | |
amaz | -0.036 |
gasp | -0.032 |
angles | -0.031 |
horror | -0.031 |
ill | -0.029 |
Token | rice |
Token ID | 20970 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
Nut | +0.146 |
rob | +0.136 |
br | +0.127 |
Rust | +0.122 |
Nut | +0.119 |
Neg logit contributions | |
gasp | -0.137 |
amaz | -0.129 |
decide | -0.119 |
ill | -0.118 |
gasped | -0.115 |
Token | 's |
Token ID | 338 |
Feature activation | +9.38e-03 |
Pos logit contributions | |
Nut | +0.052 |
rob | +0.044 |
Rust | +0.044 |
br | +0.043 |
str | +0.043 |
Neg logit contributions | |
amaz | -0.042 |
gasp | -0.039 |
angles | -0.037 |
horror | -0.036 |
ill | -0.036 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
Nut | +0.020 |
Rust | +0.017 |
Nut | +0.017 |
str | +0.017 |
age | +0.017 |
Neg logit contributions | |
amaz | -0.014 |
gasp | -0.013 |
angles | -0.012 |
ill | -0.012 |
horror | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
Nut | +0.032 |
rob | +0.027 |
Rust | +0.027 |
str | +0.027 |
br | +0.027 |
Neg logit contributions | |
amaz | -0.019 |
gasp | -0.017 |
angles | -0.016 |
horror | -0.016 |
ill | -0.016 |
Token | From |
Token ID | 3574 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
Nut | +0.030 |
str | +0.028 |
rob | +0.027 |
br | +0.026 |
Rust | +0.025 |
Neg logit contributions | |
amaz | -0.019 |
horror | -0.016 |
gasp | -0.016 |
angles | -0.015 |
ill | -0.015 |
Token | your |
Token ID | 534 |
Feature activation | +6.91e-03 |
Pos logit contributions | |
Nut | +0.021 |
Rust | +0.018 |
rob | +0.018 |
br | +0.018 |
str | +0.018 |
Neg logit contributions | |
amaz | -0.016 |
gasp | -0.015 |
angles | -0.014 |
horror | -0.014 |
ill | -0.013 |
Token | parents |
Token ID | 3397 |
Feature activation | -5.04e-03 |
Pos logit contributions | |
Nut | +0.018 |
Rust | +0.015 |
str | +0.015 |
br | +0.015 |
Nut | +0.015 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.007 |
angles | -0.007 |
horror | -0.007 |
ill | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
angles | +0.008 |
horror | +0.008 |
ill | +0.008 |
Neg logit contributions | |
Nut | -0.010 |
Rust | -0.008 |
str | -0.008 |
br | -0.008 |
rob | -0.008 |
Token | not |
Token ID | 407 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
Nut | +0.056 |
Rust | +0.046 |
Nut | +0.046 |
age | +0.045 |
str | +0.045 |
Neg logit contributions | |
amaz | -0.049 |
gasp | -0.046 |
angles | -0.044 |
horror | -0.043 |
ill | -0.043 |
Token | take |
Token ID | 1011 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
Nut | +0.040 |
Rust | +0.035 |
str | +0.034 |
br | +0.034 |
Nut | +0.034 |
Neg logit contributions | |
amaz | -0.026 |
gasp | -0.023 |
angles | -0.022 |
horror | -0.022 |
ill | -0.021 |
Token | unnecessary |
Token ID | 13114 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
Nut | +0.055 |
rob | +0.050 |
str | +0.049 |
K | +0.047 |
Rust | +0.047 |
Neg logit contributions | |
amaz | -0.039 |
gasp | -0.036 |
angles | -0.035 |
horror | -0.035 |
faces | -0.033 |
Token | risks |
Token ID | 7476 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
Nut | +0.157 |
rob | +0.139 |
str | +0.139 |
K | +0.135 |
cone | +0.129 |
Neg logit contributions | |
amaz | -0.134 |
gasp | -0.131 |
faces | -0.122 |
wish | -0.118 |
angles | -0.116 |
Token | . |
Token ID | 13 |
Feature activation | +2.97e-03 |
Pos logit contributions | |
Nut | +0.027 |
Rust | +0.023 |
rob | +0.023 |
str | +0.023 |
br | +0.023 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.015 |
angles | -0.014 |
horror | -0.014 |
ill | -0.014 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Nut | +0.006 |
rob | +0.005 |
br | +0.005 |
Rust | +0.005 |
str | +0.005 |
Neg logit contributions | |
amaz | -0.005 |
gasp | -0.005 |
horror | -0.005 |
angles | -0.005 |
ill | -0.005 |
Token | One |
Token ID | 3198 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Nut | +0.028 |
Rust | +0.023 |
str | +0.023 |
br | +0.022 |
rob | +0.022 |
Neg logit contributions | |
amaz | -0.032 |
gasp | -0.030 |
angles | -0.028 |
horror | -0.028 |
ill | -0.028 |
Token | day |
Token ID | 1110 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
amaz | +0.034 |
gasp | +0.032 |
angles | +0.032 |
horror | +0.032 |
ill | +0.030 |
Neg logit contributions | |
Nut | -0.017 |
str | -0.014 |
br | -0.014 |
Rust | -0.013 |
Nut | -0.013 |
Token | more |
Token ID | 517 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
Nut | +0.081 |
rob | +0.079 |
con | +0.076 |
cone | +0.074 |
str | +0.073 |
Neg logit contributions | |
amaz | -0.043 |
horror | -0.036 |
gasp | -0.035 |
wonders | -0.035 |
wonder | -0.035 |
Token | cautious |
Token ID | 21205 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
Nut | +0.045 |
rob | +0.039 |
Rust | +0.039 |
br | +0.038 |
str | +0.038 |
Neg logit contributions | |
amaz | -0.034 |
gasp | -0.031 |
angles | -0.030 |
horror | -0.029 |
ill | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
Nut | +0.050 |
Rust | +0.043 |
str | +0.042 |
rob | +0.042 |
K | +0.041 |
Neg logit contributions | |
amaz | -0.036 |
gasp | -0.034 |
angles | -0.033 |
horror | -0.032 |
ill | -0.031 |
Token | not |
Token ID | 407 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
Nut | +0.057 |
Rust | +0.047 |
rob | +0.046 |
br | +0.046 |
str | +0.045 |
Neg logit contributions | |
amaz | -0.065 |
gasp | -0.061 |
angles | -0.059 |
horror | -0.058 |
ill | -0.055 |
Token | take |
Token ID | 1011 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
Nut | +0.036 |
Rust | +0.030 |
rob | +0.030 |
br | +0.029 |
str | +0.029 |
Neg logit contributions | |
amaz | -0.038 |
gasp | -0.036 |
angles | -0.034 |
horror | -0.034 |
ill | -0.033 |
Token | unnecessary |
Token ID | 13114 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
Nut | +0.054 |
rob | +0.051 |
str | +0.050 |
K | +0.047 |
Rust | +0.046 |
Neg logit contributions | |
amaz | -0.040 |
ways | -0.038 |
eyes | -0.037 |
angles | -0.037 |
horror | -0.037 |
Token | risks |
Token ID | 7476 |
Feature activation | +9.64e-03 |
Pos logit contributions | |
Nut | +0.143 |
rob | +0.130 |
str | +0.127 |
K | +0.124 |
Rust | +0.118 |
Neg logit contributions | |
amaz | -0.145 |
gasp | -0.137 |
faces | -0.129 |
angles | -0.129 |
ways | -0.127 |
Token | . |
Token ID | 13 |
Feature activation | +2.68e-03 |
Pos logit contributions | |
Nut | +0.023 |
Rust | +0.019 |
rob | +0.019 |
str | +0.019 |
K | +0.019 |
Neg logit contributions | |
amaz | -0.013 |
gasp | -0.012 |
angles | -0.012 |
horror | -0.011 |
ill | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
Nut | +0.006 |
rob | +0.006 |
Rust | +0.006 |
br | +0.006 |
Silver | +0.005 |
Neg logit contributions | |
amaz | -0.004 |
horror | -0.003 |
ill | -0.003 |
gasp | -0.003 |
angles | -0.003 |
Token | memory |
Token ID | 4088 |
Feature activation | -2.63e-03 |
Pos logit contributions | |
Nut | +0.034 |
rob | +0.030 |
br | +0.029 |
Rust | +0.029 |
con | +0.029 |
Neg logit contributions | |
amaz | -0.018 |
gasp | -0.015 |
horror | -0.015 |
angles | -0.014 |
wonder | -0.013 |
Token | of |
Token ID | 286 |
Feature activation | +6.09e-03 |
Pos logit contributions | |
amaz | +0.004 |
gasp | +0.003 |
angles | +0.003 |
horror | +0.003 |
ill | +0.003 |
Neg logit contributions | |
Nut | -0.006 |
Rust | -0.005 |
br | -0.005 |
str | -0.005 |
rob | -0.005 |
Token | be |
Token ID | 307 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
Nut | +0.038 |
Rust | +0.032 |
Nut | +0.032 |
str | +0.031 |
age | +0.031 |
Neg logit contributions | |
amaz | -0.026 |
gasp | -0.023 |
angles | -0.022 |
horror | -0.022 |
ill | -0.021 |
Token | prepared |
Token ID | 5597 |
Feature activation | +2.46e-03 |
Pos logit contributions | |
Nut | +0.056 |
rob | +0.049 |
Rust | +0.048 |
con | +0.047 |
str | +0.047 |
Neg logit contributions | |
amaz | -0.045 |
gasp | -0.041 |
angles | -0.039 |
horror | -0.038 |
ill | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
Nut | +0.005 |
rob | +0.004 |
Rust | +0.004 |
str | +0.004 |
K | +0.004 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.004 |
angles | -0.004 |
horror | -0.004 |
ill | -0.004 |
Token | not |
Token ID | 407 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
Nut | +0.059 |
Rust | +0.049 |
Nut | +0.049 |
age | +0.049 |
str | +0.048 |
Neg logit contributions | |
amaz | -0.054 |
gasp | -0.050 |
ill | -0.049 |
angles | -0.049 |
horror | -0.049 |
Token | take |
Token ID | 1011 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
Nut | +0.032 |
Rust | +0.027 |
Nut | +0.027 |
str | +0.027 |
br | +0.027 |
Neg logit contributions | |
amaz | -0.025 |
gasp | -0.022 |
angles | -0.022 |
horror | -0.021 |
ill | -0.021 |
Token | unnecessary |
Token ID | 13114 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
Nut | +0.039 |
rob | +0.035 |
str | +0.033 |
K | +0.033 |
Rust | +0.032 |
Neg logit contributions | |
amaz | -0.035 |
gasp | -0.032 |
angles | -0.031 |
horror | -0.031 |
wonders | -0.029 |
Token | risks |
Token ID | 7476 |
Feature activation | +5.69e-03 |
Pos logit contributions | |
Nut | +0.136 |
rob | +0.118 |
K | +0.115 |
str | +0.115 |
Rust | +0.109 |
Neg logit contributions | |
amaz | -0.156 |
gasp | -0.147 |
angles | -0.138 |
horror | -0.134 |
faces | -0.133 |
Token | . |
Token ID | 13 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
Nut | +0.013 |
Rust | +0.011 |
rob | +0.011 |
str | +0.011 |
br | +0.011 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.007 |
angles | -0.007 |
horror | -0.007 |
ill | -0.007 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
amaz | +0.008 |
gasp | +0.007 |
horror | +0.007 |
angles | +0.007 |
ill | +0.007 |
Neg logit contributions | |
Nut | -0.009 |
rob | -0.008 |
br | -0.008 |
con | -0.007 |
Rust | -0.007 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Nut | +0.019 |
Rust | +0.015 |
str | +0.015 |
br | +0.015 |
rob | +0.015 |
Neg logit contributions | |
amaz | -0.020 |
gasp | -0.019 |
angles | -0.018 |
ill | -0.017 |
horror | -0.017 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.012 |
horror | +0.012 |
angles | +0.011 |
spit | +0.011 |
Neg logit contributions | |
Nut | -0.025 |
str | -0.023 |
Nut | -0.023 |
Rust | -0.022 |
br | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
Nut | +0.010 |
Rust | +0.009 |
rob | +0.009 |
br | +0.009 |
str | +0.009 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.005 |
angles | -0.005 |
horror | -0.005 |
ill | -0.004 |
Token | He |
Token ID | 679 |
Feature activation | -1.65e-03 |
Pos logit contributions | |
Nut | +0.012 |
rob | +0.010 |
Rust | +0.010 |
br | +0.010 |
str | +0.010 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.007 |
horror | -0.007 |
angles | -0.007 |
ill | -0.007 |
Token | gave |
Token ID | 2921 |
Feature activation | -3.89e-03 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
horror | +0.002 |
angles | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.004 |
Rust | -0.003 |
rob | -0.003 |
br | -0.003 |
str | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.005 |
angles | +0.005 |
horror | +0.005 |
ill | +0.005 |
Neg logit contributions | |
Nut | -0.009 |
Rust | -0.007 |
br | -0.007 |
str | -0.007 |
rob | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
Nut | +0.042 |
rob | +0.037 |
br | +0.036 |
str | +0.036 |
Rust | +0.036 |
Neg logit contributions | |
amaz | -0.028 |
gasp | -0.026 |
horror | -0.024 |
bowls | -0.024 |
ill | -0.023 |
Token | band |
Token ID | 4097 |
Feature activation | +7.71e-02 |
Pos logit contributions | |
Nut | +0.083 |
rob | +0.077 |
Rust | +0.072 |
Sy | +0.072 |
K | +0.070 |
Neg logit contributions | |
amaz | -0.046 |
gasp | -0.046 |
horror | -0.038 |
wish | -0.038 |
bowls | -0.036 |
Token | - |
Token ID | 12 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
Nut | +0.142 |
rob | +0.135 |
Rust | +0.126 |
str | +0.121 |
Silver | +0.121 |
Neg logit contributions | |
gasp | -0.139 |
amaz | -0.129 |
eye | -0.121 |
face | -0.119 |
hands | -0.117 |
Token | aid |
Token ID | 1698 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
Nut | +0.075 |
rob | +0.064 |
Rust | +0.064 |
Nut | +0.063 |
Barkley | +0.063 |
Neg logit contributions | |
amaz | -0.092 |
gasp | -0.092 |
ill | -0.089 |
spit | -0.085 |
decide | -0.083 |
Token | for |
Token ID | 329 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
Nut | +0.094 |
rob | +0.081 |
Rust | +0.080 |
str | +0.079 |
Sy | +0.078 |
Neg logit contributions | |
gasp | -0.042 |
amaz | -0.038 |
ill | -0.036 |
hands | -0.035 |
eye | -0.034 |
Token | her |
Token ID | 607 |
Feature activation | +2.68e-03 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.006 |
horror | +0.006 |
angles | +0.006 |
bowls | +0.005 |
Neg logit contributions | |
Nut | -0.009 |
rob | -0.008 |
str | -0.008 |
Rust | -0.008 |
br | -0.008 |
Token | knee |
Token ID | 10329 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
Nut | +0.006 |
Rust | +0.005 |
br | +0.005 |
str | +0.005 |
rob | +0.005 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.004 |
angles | -0.004 |
horror | -0.004 |
ill | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +6.24e-03 |
Pos logit contributions | |
amaz | +0.008 |
gasp | +0.008 |
horror | +0.007 |
angles | +0.007 |
bowls | +0.007 |
Neg logit contributions | |
Nut | -0.009 |
br | -0.007 |
con | -0.007 |
Rust | -0.007 |
Nut | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | -7.91e-04 |
Pos logit contributions | |
Nut | +0.013 |
Rust | +0.011 |
Nut | +0.010 |
br | +0.010 |
str | +0.010 |
Neg logit contributions | |
amaz | -0.011 |
gasp | -0.010 |
angles | -0.009 |
horror | -0.009 |
ill | -0.009 |
Token | played |
Token ID | 2826 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
amaz | +0.001 |
gasp | +0.001 |
angles | +0.001 |
horror | +0.001 |
ill | +0.001 |
Neg logit contributions | |
Nut | -0.002 |
Rust | -0.001 |
Nut | -0.001 |
K | -0.001 |
str | -0.001 |
Token | hop |
Token ID | 1725 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
Nut | +0.039 |
str | +0.034 |
Rust | +0.033 |
br | +0.032 |
Sy | +0.031 |
Neg logit contributions | |
amaz | -0.027 |
gasp | -0.025 |
horror | -0.022 |
angles | -0.022 |
ill | -0.022 |
Token | sc |
Token ID | 1416 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
Nut | +0.079 |
str | +0.070 |
Rust | +0.066 |
Cle | +0.066 |
Sy | +0.065 |
Neg logit contributions | |
gasp | -0.033 |
gasped | -0.032 |
ill | -0.030 |
teams | -0.030 |
groups | -0.030 |
Token | ot |
Token ID | 313 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
Nut | +0.081 |
rob | +0.075 |
Rust | +0.074 |
str | +0.072 |
con | +0.071 |
Neg logit contributions | |
amaz | -0.034 |
gasp | -0.033 |
ill | -0.029 |
angles | -0.028 |
horror | -0.027 |
Token | ch |
Token ID | 354 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
Nut | +0.119 |
str | +0.099 |
Sy | +0.097 |
hawk | +0.091 |
Rust | +0.090 |
Neg logit contributions | |
gasp | -0.146 |
faces | -0.139 |
of | -0.138 |
cups | -0.133 |
are | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
Nut | +0.098 |
Rust | +0.085 |
Charl | +0.081 |
Sy | +0.081 |
Nut | +0.080 |
Neg logit contributions | |
gasp | -0.072 |
ill | -0.066 |
faces | -0.066 |
hands | -0.063 |
of | -0.062 |
Token | picked |
Token ID | 6497 |
Feature activation | +7.50e-03 |
Pos logit contributions | |
Nut | +0.041 |
Rust | +0.036 |
str | +0.036 |
age | +0.035 |
br | +0.035 |
Neg logit contributions | |
gasp | -0.019 |
amaz | -0.018 |
gasped | -0.016 |
angles | -0.016 |
ill | -0.016 |
Token | more |
Token ID | 517 |
Feature activation | +8.85e-03 |
Pos logit contributions | |
Nut | +0.016 |
Rust | +0.014 |
str | +0.013 |
age | +0.013 |
Silver | +0.013 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.011 |
angles | -0.011 |
horror | -0.010 |
ill | -0.010 |
Token | da |
Token ID | 12379 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Nut | +0.018 |
Rust | +0.015 |
rob | +0.014 |
Nut | +0.014 |
str | +0.014 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.014 |
angles | -0.014 |
horror | -0.013 |
ill | -0.013 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
Nut | +0.030 |
Rust | +0.026 |
str | +0.025 |
br | +0.025 |
Nut | +0.025 |
Neg logit contributions | |
amaz | -0.019 |
gasp | -0.017 |
angles | -0.017 |
horror | -0.016 |
ill | -0.015 |
Token | him |
Token ID | 683 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.017 |
angles | +0.017 |
horror | +0.016 |
ill | +0.016 |
Neg logit contributions | |
Nut | -0.027 |
Rust | -0.023 |
str | -0.023 |
br | -0.023 |
rob | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | +5.62e-03 |
Pos logit contributions | |
Nut | +0.035 |
Rust | +0.029 |
br | +0.028 |
str | +0.028 |
Nut | +0.028 |
Neg logit contributions | |
amaz | -0.036 |
gasp | -0.033 |
angles | -0.032 |
horror | -0.031 |
ill | -0.030 |
Token | other |
Token ID | 584 |
Feature activation | +8.47e-03 |
Pos logit contributions | |
Nut | +0.012 |
Rust | +0.010 |
br | +0.010 |
rob | +0.010 |
str | +0.010 |
Neg logit contributions | |
amaz | -0.009 |
gasp | -0.008 |
angles | -0.008 |
horror | -0.008 |
ill | -0.007 |
Token | bath |
Token ID | 7837 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
Nut | +0.018 |
Rust | +0.016 |
age | +0.016 |
con | +0.016 |
Nut | +0.016 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.011 |
ill | -0.011 |
horror | -0.010 |
angles | -0.010 |
Token | ro |
Token ID | 305 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
Nut | +0.020 |
Rust | +0.016 |
br | +0.016 |
rob | +0.016 |
str | +0.016 |
Neg logit contributions | |
amaz | -0.018 |
gasp | -0.017 |
angles | -0.016 |
horror | -0.016 |
ill | -0.015 |
Token | bes |
Token ID | 12636 |
Feature activation | +7.76e-03 |
Pos logit contributions | |
Nut | +0.137 |
Nut | +0.119 |
Rust | +0.117 |
cone | +0.117 |
K | +0.115 |
Neg logit contributions | |
gasp | -0.128 |
cloth | -0.124 |
faces | -0.122 |
amaz | -0.122 |
gasped | -0.120 |
Token | . |
Token ID | 13 |
Feature activation | +2.79e-03 |
Pos logit contributions | |
Nut | +0.016 |
Rust | +0.013 |
br | +0.013 |
rob | +0.013 |
str | +0.013 |
Neg logit contributions | |
amaz | -0.013 |
gasp | -0.012 |
angles | -0.012 |
horror | -0.011 |
ill | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | -2.10e-03 |
Pos logit contributions | |
Nut | +0.006 |
rob | +0.006 |
str | +0.006 |
br | +0.006 |
Rust | +0.006 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.003 |
horror | -0.003 |
ill | -0.003 |
angles | -0.003 |
Token | had |
Token ID | 550 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
amaz | +0.003 |
gasp | +0.003 |
angles | +0.003 |
horror | +0.003 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.005 |
Rust | -0.004 |
str | -0.004 |
br | -0.004 |
Nut | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +3.22e-03 |
Pos logit contributions | |
Nut | +0.048 |
Rust | +0.040 |
rob | +0.039 |
str | +0.039 |
br | +0.039 |
Neg logit contributions | |
amaz | -0.053 |
gasp | -0.050 |
angles | -0.048 |
horror | -0.047 |
ill | -0.046 |
Token | mom |
Token ID | 1995 |
Feature activation | -9.25e-03 |
Pos logit contributions | |
amaz | +0.024 |
angles | +0.023 |
ill | +0.023 |
gasp | +0.022 |
adds | +0.019 |
Neg logit contributions | |
Nut | -0.032 |
Rust | -0.028 |
Nut | -0.027 |
str | -0.027 |
age | -0.027 |
Token | my |
Token ID | 1820 |
Feature activation | +4.83e-03 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.012 |
horror | +0.012 |
ill | +0.011 |
Neg logit contributions | |
Nut | -0.021 |
Rust | -0.018 |
br | -0.018 |
rob | -0.018 |
str | -0.017 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.60e-03 |
Pos logit contributions | |
Nut | +0.012 |
rob | +0.010 |
br | +0.010 |
Rust | +0.010 |
con | +0.010 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.006 |
horror | -0.005 |
angles | -0.005 |
ill | -0.005 |
Token | her |
Token ID | 607 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
angles | +0.002 |
horror | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.004 |
Rust | -0.003 |
str | -0.003 |
br | -0.003 |
rob | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
Nut | +0.076 |
rob | +0.070 |
br | +0.067 |
str | +0.065 |
Rust | +0.064 |
Neg logit contributions | |
amaz | -0.057 |
gasp | -0.052 |
bowls | -0.050 |
horror | -0.049 |
eyes | -0.048 |
Token | band |
Token ID | 4097 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
Nut | +0.078 |
rob | +0.074 |
Rust | +0.068 |
Sy | +0.067 |
con | +0.067 |
Neg logit contributions | |
amaz | -0.038 |
gasp | -0.037 |
wish | -0.032 |
horror | -0.031 |
wonder | -0.030 |
Token | - |
Token ID | 12 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
Nut | +0.115 |
rob | +0.108 |
hawk | +0.098 |
str | +0.098 |
Rust | +0.098 |
Neg logit contributions | |
gasp | -0.152 |
eye | -0.142 |
amaz | -0.140 |
of | -0.139 |
face | -0.138 |
Token | aid |
Token ID | 1698 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
Nut | +0.081 |
Barkley | +0.075 |
rob | +0.073 |
Rust | +0.072 |
Nut | +0.071 |
Neg logit contributions | |
ill | -0.076 |
gasp | -0.075 |
amaz | -0.074 |
spit | -0.073 |
ucks | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
Nut | +0.084 |
str | +0.072 |
rob | +0.072 |
Rust | +0.070 |
Sy | +0.070 |
Neg logit contributions | |
gasp | -0.036 |
amaz | -0.033 |
hands | -0.033 |
eye | -0.033 |
ill | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Nut | +0.026 |
Rust | +0.022 |
Nut | +0.022 |
br | +0.022 |
str | +0.021 |
Neg logit contributions | |
amaz | -0.018 |
gasp | -0.017 |
angles | -0.016 |
horror | -0.016 |
ill | -0.015 |
Token | kiss |
Token ID | 9245 |
Feature activation | +4.67e-04 |
Pos logit contributions | |
Nut | +0.034 |
rob | +0.030 |
Rust | +0.029 |
Nut | +0.028 |
Sy | +0.028 |
Neg logit contributions | |
amaz | -0.026 |
gasp | -0.025 |
horror | -0.022 |
bowls | -0.021 |
angles | -0.021 |
Token | L |
Token ID | 43 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
ill | +0.002 |
angles | +0.002 |
horror | +0.002 |
Neg logit contributions | |
Nut | -0.002 |
rob | -0.002 |
con | -0.002 |
Rust | -0.002 |
Nut | -0.002 |
Token | ily |
Token ID | 813 |
Feature activation | +6.82e-03 |
Pos logit contributions | |
amaz | +0.015 |
gasp | +0.014 |
angles | +0.013 |
horror | +0.013 |
bowls | +0.012 |
Neg logit contributions | |
Nut | -0.023 |
br | -0.020 |
rob | -0.019 |
Rust | -0.019 |
str | -0.019 |
Token | gave |
Token ID | 2921 |
Feature activation | +3.94e-03 |
Pos logit contributions | |
Nut | +0.017 |
Rust | +0.015 |
rob | +0.014 |
br | +0.014 |
str | +0.014 |
Neg logit contributions | |
amaz | -0.009 |
gasp | -0.008 |
horror | -0.007 |
angles | -0.007 |
ill | -0.007 |
Token | him |
Token ID | 683 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
Nut | +0.010 |
br | +0.009 |
str | +0.008 |
age | +0.008 |
Rust | +0.008 |
Neg logit contributions | |
amaz | -0.005 |
gasp | -0.005 |
angles | -0.004 |
horror | -0.004 |
ill | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
Nut | +0.075 |
rob | +0.065 |
br | +0.064 |
str | +0.063 |
Rust | +0.062 |
Neg logit contributions | |
amaz | -0.065 |
gasp | -0.060 |
horror | -0.057 |
bowls | -0.056 |
ill | -0.055 |
Token | band |
Token ID | 4097 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
Nut | +0.091 |
rob | +0.085 |
Sy | +0.079 |
Rust | +0.078 |
con | +0.077 |
Neg logit contributions | |
gasp | -0.049 |
amaz | -0.048 |
wish | -0.040 |
wonder | -0.039 |
horror | -0.037 |
Token | - |
Token ID | 12 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
Nut | +0.139 |
rob | +0.127 |
Rust | +0.124 |
Silver | +0.119 |
Barkley | +0.116 |
Neg logit contributions | |
gasp | -0.136 |
amaz | -0.126 |
eye | -0.119 |
face | -0.115 |
of | -0.113 |
Token | aid |
Token ID | 1698 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
Nut | +0.071 |
Rust | +0.062 |
rob | +0.061 |
Barkley | +0.061 |
Nut | +0.060 |
Neg logit contributions | |
amaz | -0.089 |
gasp | -0.088 |
ill | -0.085 |
spit | -0.082 |
wonder | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
Nut | +0.095 |
Rust | +0.080 |
str | +0.080 |
rob | +0.079 |
Sy | +0.079 |
Neg logit contributions | |
gasp | -0.044 |
amaz | -0.041 |
ill | -0.037 |
horror | -0.037 |
eye | -0.037 |
Token | they |
Token ID | 484 |
Feature activation | -9.27e-03 |
Pos logit contributions | |
Nut | +0.047 |
Rust | +0.041 |
br | +0.040 |
rob | +0.040 |
str | +0.040 |
Neg logit contributions | |
amaz | -0.028 |
gasp | -0.025 |
angles | -0.023 |
horror | -0.023 |
ill | -0.022 |
Token | continued |
Token ID | 3767 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
amaz | +0.012 |
gasp | +0.011 |
angles | +0.011 |
horror | +0.010 |
ill | +0.010 |
Neg logit contributions | |
Nut | -0.022 |
Rust | -0.019 |
br | -0.019 |
str | -0.019 |
rob | -0.019 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
amaz | +0.023 |
angles | +0.021 |
gasp | +0.021 |
ill | +0.020 |
horror | +0.019 |
Neg logit contributions | |
Nut | -0.039 |
str | -0.033 |
Rust | -0.033 |
Nut | -0.033 |
K | -0.032 |
Token | my |
Token ID | 1820 |
Feature activation | -7.33e-03 |
Pos logit contributions | |
amaz | +0.030 |
gasp | +0.027 |
angles | +0.026 |
horror | +0.025 |
ill | +0.024 |
Neg logit contributions | |
Nut | -0.043 |
Rust | -0.037 |
str | -0.036 |
br | -0.036 |
Nut | -0.036 |
Token | gave |
Token ID | 2921 |
Feature activation | -5.92e-03 |
Pos logit contributions | |
amaz | +0.008 |
gasp | +0.007 |
gasped | +0.006 |
horror | +0.006 |
ill | +0.006 |
Neg logit contributions | |
Nut | -0.020 |
rob | -0.017 |
Rust | -0.017 |
br | -0.017 |
con | -0.017 |
Token | her |
Token ID | 607 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
angles | +0.008 |
horror | +0.008 |
ill | +0.007 |
Neg logit contributions | |
Nut | -0.013 |
Rust | -0.011 |
str | -0.011 |
br | -0.011 |
rob | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
Nut | +0.052 |
rob | +0.046 |
br | +0.044 |
Rust | +0.044 |
str | +0.042 |
Neg logit contributions | |
amaz | -0.047 |
gasp | -0.043 |
horror | -0.041 |
bowls | -0.040 |
angles | -0.039 |
Token | band |
Token ID | 4097 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
Nut | +0.068 |
rob | +0.064 |
Rust | +0.060 |
Sy | +0.059 |
con | +0.058 |
Neg logit contributions | |
amaz | -0.035 |
gasp | -0.035 |
wish | -0.028 |
wonder | -0.027 |
bowls | -0.026 |
Token | - |
Token ID | 12 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
Nut | +0.116 |
rob | +0.110 |
Rust | +0.099 |
Cle | +0.094 |
str | +0.094 |
Neg logit contributions | |
gasp | -0.157 |
amaz | -0.146 |
eye | -0.143 |
of | -0.139 |
face | -0.137 |
Token | aid |
Token ID | 1698 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
Nut | +0.074 |
rob | +0.065 |
Rust | +0.065 |
Barkley | +0.065 |
Nut | +0.062 |
Neg logit contributions | |
gasp | -0.084 |
amaz | -0.083 |
ill | -0.082 |
wonder | -0.077 |
spit | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
Nut | +0.090 |
rob | +0.077 |
str | +0.076 |
Rust | +0.076 |
Sy | +0.075 |
Neg logit contributions | |
gasp | -0.043 |
amaz | -0.040 |
hands | -0.038 |
ill | -0.036 |
horror | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
Nut | +0.027 |
Rust | +0.023 |
rob | +0.022 |
str | +0.022 |
br | +0.022 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.016 |
horror | -0.014 |
angles | -0.014 |
bowls | -0.014 |
Token | kiss |
Token ID | 9245 |
Feature activation | +2.60e-03 |
Pos logit contributions | |
Nut | +0.024 |
rob | +0.021 |
Rust | +0.021 |
Sy | +0.020 |
Nut | +0.020 |
Neg logit contributions | |
amaz | -0.016 |
gasp | -0.015 |
horror | -0.013 |
bowls | -0.013 |
wish | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
Nut | +0.033 |
Rust | +0.029 |
rob | +0.028 |
br | +0.028 |
str | +0.028 |
Neg logit contributions | |
amaz | -0.022 |
gasp | -0.021 |
angles | -0.019 |
horror | -0.019 |
ill | -0.018 |
Token | She |
Token ID | 1375 |
Feature activation | +7.37e-03 |
Pos logit contributions | |
Nut | +0.021 |
Rust | +0.018 |
rob | +0.018 |
br | +0.018 |
str | +0.018 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.016 |
angles | -0.015 |
horror | -0.015 |
ill | -0.014 |
Token | gave |
Token ID | 2921 |
Feature activation | +2.59e-04 |
Pos logit contributions | |
Nut | +0.017 |
Rust | +0.015 |
str | +0.014 |
br | +0.014 |
rob | +0.014 |
Neg logit contributions | |
amaz | -0.011 |
gasp | -0.010 |
angles | -0.009 |
horror | -0.009 |
ill | -0.008 |
Token | them |
Token ID | 606 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
Nut | +0.001 |
Rust | +0.000 |
br | +0.000 |
str | +0.000 |
rob | +0.000 |
Neg logit contributions | |
amaz | -0.000 |
gasp | -0.000 |
angles | -0.000 |
horror | -0.000 |
ill | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
Nut | +0.088 |
Rust | +0.073 |
rob | +0.073 |
str | +0.072 |
br | +0.072 |
Neg logit contributions | |
amaz | -0.079 |
gasp | -0.074 |
angles | -0.069 |
horror | -0.069 |
bowls | -0.068 |
Token | band |
Token ID | 4097 |
Feature activation | +7.50e-02 |
Pos logit contributions | |
Nut | +0.103 |
rob | +0.093 |
Rust | +0.090 |
Sy | +0.089 |
K | +0.088 |
Neg logit contributions | |
amaz | -0.041 |
gasp | -0.040 |
bowls | -0.032 |
horror | -0.031 |
wish | -0.031 |
Token | - |
Token ID | 12 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
Nut | +0.116 |
rob | +0.095 |
Rust | +0.094 |
str | +0.094 |
Silver | +0.094 |
Neg logit contributions | |
gasp | -0.159 |
amaz | -0.149 |
eye | -0.142 |
faces | -0.139 |
face | -0.139 |
Token | aid |
Token ID | 1698 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
Nut | +0.083 |
Rust | +0.069 |
Nut | +0.069 |
str | +0.068 |
rob | +0.068 |
Neg logit contributions | |
gasp | -0.098 |
amaz | -0.097 |
ill | -0.094 |
ucks | -0.090 |
faces | -0.088 |
Token | for |
Token ID | 329 |
Feature activation | -1.13e-03 |
Pos logit contributions | |
Nut | +0.091 |
Rust | +0.077 |
str | +0.077 |
Sy | +0.076 |
Silver | +0.076 |
Neg logit contributions | |
gasp | -0.036 |
amaz | -0.033 |
ill | -0.032 |
bowls | -0.032 |
hands | -0.031 |
Token | Jen |
Token ID | 13364 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
angles | +0.002 |
horror | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.002 |
Rust | -0.002 |
rob | -0.002 |
br | -0.002 |
str | -0.002 |
Token | 's |
Token ID | 338 |
Feature activation | +9.79e-03 |
Pos logit contributions | |
Nut | +0.063 |
rob | +0.054 |
Rust | +0.053 |
br | +0.053 |
Sy | +0.052 |
Neg logit contributions | |
amaz | -0.039 |
gasp | -0.039 |
ill | -0.036 |
horror | -0.035 |
bowls | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
Nut | +0.012 |
br | +0.010 |
Rust | +0.010 |
rob | +0.010 |
str | +0.010 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.011 |
angles | -0.010 |
horror | -0.010 |
ill | -0.010 |
Token | They |
Token ID | 1119 |
Feature activation | +5.78e-04 |
Pos logit contributions | |
amaz | +0.025 |
gasp | +0.024 |
angles | +0.023 |
horror | +0.022 |
bowls | +0.021 |
Neg logit contributions | |
Nut | -0.030 |
Nut | -0.026 |
Rust | -0.025 |
Sy | -0.025 |
str | -0.024 |
Token | played |
Token ID | 2826 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Nut | +0.001 |
Rust | +0.001 |
str | +0.001 |
Nut | +0.001 |
K | +0.001 |
Neg logit contributions | |
amaz | -0.001 |
gasp | -0.001 |
angles | -0.001 |
horror | -0.001 |
ill | -0.001 |
Token | hop |
Token ID | 1725 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
Nut | +0.037 |
str | +0.031 |
Rust | +0.031 |
Nut | +0.030 |
rob | +0.030 |
Neg logit contributions | |
amaz | -0.029 |
gasp | -0.027 |
angles | -0.025 |
horror | -0.025 |
ill | -0.024 |
Token | sc |
Token ID | 1416 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
Nut | +0.088 |
str | +0.074 |
Rust | +0.072 |
Nut | +0.071 |
Silver | +0.071 |
Neg logit contributions | |
gasp | -0.044 |
amaz | -0.040 |
ill | -0.039 |
gasped | -0.038 |
teams | -0.037 |
Token | ot |
Token ID | 313 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
Nut | +0.094 |
rob | +0.086 |
Rust | +0.084 |
str | +0.083 |
Silver | +0.082 |
Neg logit contributions | |
amaz | -0.041 |
gasp | -0.039 |
ill | -0.035 |
angles | -0.034 |
horror | -0.033 |
Token | ch |
Token ID | 354 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
Nut | +0.119 |
str | +0.098 |
Sy | +0.096 |
hawk | +0.088 |
Nut | +0.087 |
Neg logit contributions | |
gasp | -0.149 |
faces | -0.137 |
of | -0.136 |
teams | -0.136 |
hands | -0.134 |
Token | , |
Token ID | 11 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
Nut | +0.116 |
Rust | +0.100 |
Sy | +0.099 |
Nut | +0.097 |
Silver | +0.097 |
Neg logit contributions | |
gasp | -0.064 |
ill | -0.055 |
faces | -0.053 |
hands | -0.052 |
teams | -0.052 |
Token | jumped |
Token ID | 11687 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
Nut | +0.071 |
Rust | +0.060 |
str | +0.060 |
rob | +0.059 |
br | +0.058 |
Neg logit contributions | |
amaz | -0.049 |
gasp | -0.047 |
horror | -0.041 |
angles | -0.041 |
bowls | -0.040 |
Token | rope |
Token ID | 17182 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
Nut | +0.036 |
str | +0.031 |
rob | +0.031 |
Rust | +0.031 |
Silver | +0.029 |
Neg logit contributions | |
gasp | -0.022 |
amaz | -0.022 |
ill | -0.019 |
angles | -0.019 |
bowls | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
Nut | +0.098 |
Rust | +0.085 |
str | +0.085 |
rob | +0.082 |
Silver | +0.082 |
Neg logit contributions | |
gasp | -0.060 |
amaz | -0.060 |
ill | -0.057 |
angles | -0.056 |
bowls | -0.052 |
Token | walk |
Token ID | 2513 |
Feature activation | +8.81e-03 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
angles | +0.007 |
horror | +0.007 |
ill | +0.007 |
Neg logit contributions | |
Nut | -0.015 |
Rust | -0.013 |
br | -0.013 |
rob | -0.013 |
str | -0.013 |
Token | around |
Token ID | 1088 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
Nut | +0.020 |
str | +0.017 |
Rust | +0.017 |
rob | +0.016 |
br | +0.016 |
Neg logit contributions | |
amaz | -0.013 |
gasp | -0.012 |
angles | -0.012 |
ill | -0.011 |
wonder | -0.010 |
Token | on |
Token ID | 319 |
Feature activation | +7.23e-03 |
Pos logit contributions | |
Nut | +0.015 |
str | +0.013 |
Rust | +0.013 |
br | +0.013 |
age | +0.012 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.011 |
angles | -0.010 |
horror | -0.010 |
ill | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -7.28e-04 |
Pos logit contributions | |
Nut | +0.015 |
str | +0.013 |
Rust | +0.013 |
rob | +0.013 |
br | +0.013 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.011 |
angles | -0.011 |
horror | -0.010 |
bowls | -0.010 |
Token | tight |
Token ID | 5381 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
amaz | +0.001 |
gasp | +0.001 |
angles | +0.001 |
bowls | +0.001 |
horror | +0.001 |
Neg logit contributions | |
Nut | -0.002 |
rob | -0.001 |
Rust | -0.001 |
Sy | -0.001 |
Nut | -0.001 |
Token | ro |
Token ID | 305 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
Nut | +0.039 |
rob | +0.034 |
Rust | +0.034 |
str | +0.033 |
K | +0.032 |
Neg logit contributions | |
gasp | -0.023 |
amaz | -0.022 |
angles | -0.021 |
ill | -0.019 |
eyes | -0.019 |
Token | pe |
Token ID | 431 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
Nut | +0.121 |
Rust | +0.103 |
Nut | +0.102 |
Sy | +0.097 |
Barkley | +0.091 |
Neg logit contributions | |
gasp | -0.143 |
angles | -0.139 |
ill | -0.139 |
cloth | -0.137 |
amaz | -0.135 |
Token | , |
Token ID | 11 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
Nut | +0.063 |
Rust | +0.055 |
Nut | +0.054 |
Sy | +0.052 |
rob | +0.051 |
Neg logit contributions | |
gasp | -0.037 |
amaz | -0.036 |
angles | -0.036 |
ill | -0.032 |
decide | -0.030 |
Token | he |
Token ID | 339 |
Feature activation | -5.90e-03 |
Pos logit contributions | |
Nut | +0.067 |
Rust | +0.057 |
rob | +0.056 |
str | +0.056 |
br | +0.054 |
Neg logit contributions | |
amaz | -0.049 |
gasp | -0.046 |
angles | -0.042 |
horror | -0.041 |
wonder | -0.039 |
Token | could |
Token ID | 714 |
Feature activation | -4.18e-03 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.009 |
angles | +0.008 |
horror | +0.008 |
ill | +0.008 |
Neg logit contributions | |
Nut | -0.013 |
Rust | -0.011 |
rob | -0.011 |
Nut | -0.010 |
str | -0.010 |
Token | even |
Token ID | 772 |
Feature activation | -3.72e-03 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
angles | +0.004 |
horror | +0.004 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.010 |
Rust | -0.009 |
rob | -0.009 |
str | -0.009 |
br | -0.009 |
Token | be |
Token ID | 307 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
Nut | +0.008 |
Rust | +0.007 |
rob | +0.007 |
br | +0.007 |
str | +0.006 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.004 |
angles | -0.004 |
horror | -0.004 |
ill | -0.003 |
Token | careful |
Token ID | 8161 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
Nut | +0.047 |
rob | +0.041 |
Rust | +0.040 |
con | +0.039 |
str | +0.039 |
Neg logit contributions | |
amaz | -0.033 |
gasp | -0.030 |
angles | -0.029 |
horror | -0.028 |
ill | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
Nut | +0.034 |
Rust | +0.029 |
br | +0.028 |
str | +0.028 |
rob | +0.028 |
Neg logit contributions | |
amaz | -0.029 |
gasp | -0.027 |
angles | -0.025 |
horror | -0.025 |
ill | -0.024 |
Token | not |
Token ID | 407 |
Feature activation | +8.95e-03 |
Pos logit contributions | |
Nut | +0.050 |
Rust | +0.041 |
str | +0.041 |
Nut | +0.040 |
br | +0.040 |
Neg logit contributions | |
amaz | -0.052 |
gasp | -0.048 |
angles | -0.046 |
horror | -0.046 |
ill | -0.045 |
Token | take |
Token ID | 1011 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
Nut | +0.020 |
Rust | +0.017 |
br | +0.016 |
str | +0.016 |
Nut | +0.016 |
Neg logit contributions | |
amaz | -0.014 |
gasp | -0.013 |
angles | -0.012 |
horror | -0.012 |
ill | -0.011 |
Token | unnecessary |
Token ID | 13114 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
Nut | +0.055 |
rob | +0.052 |
str | +0.050 |
Rust | +0.047 |
age | +0.046 |
Neg logit contributions | |
amaz | -0.038 |
angles | -0.034 |
gasp | -0.034 |
horror | -0.034 |
eyes | -0.034 |
Token | risks |
Token ID | 7476 |
Feature activation | +2.67e-03 |
Pos logit contributions | |
Nut | +0.136 |
rob | +0.120 |
str | +0.117 |
K | +0.113 |
Rust | +0.111 |
Neg logit contributions | |
amaz | -0.142 |
gasp | -0.134 |
angles | -0.126 |
horror | -0.123 |
faces | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +6.99e-03 |
Pos logit contributions | |
Nut | +0.006 |
Rust | +0.005 |
rob | +0.005 |
str | +0.005 |
br | +0.005 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.003 |
angles | -0.003 |
horror | -0.003 |
ill | -0.003 |
Token | The |
Token ID | 383 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
Nut | +0.013 |
rob | +0.012 |
br | +0.012 |
Rust | +0.012 |
str | +0.011 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.011 |
horror | -0.011 |
angles | -0.011 |
ill | -0.011 |
Token | moral |
Token ID | 6573 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
Nut | +0.027 |
rob | +0.024 |
Rust | +0.024 |
br | +0.023 |
con | +0.023 |
Neg logit contributions | |
amaz | -0.013 |
gasp | -0.011 |
horror | -0.011 |
angles | -0.010 |
spit | -0.009 |
Token | of |
Token ID | 286 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
Nut | +0.082 |
Rust | +0.075 |
rob | +0.074 |
str | +0.073 |
con | +0.072 |
Neg logit contributions | |
amaz | -0.010 |
gasp | -0.008 |
horror | -0.006 |
angles | -0.006 |
ill | -0.005 |
Token | children |
Token ID | 1751 |
Feature activation | +3.60e-04 |
Pos logit contributions | |
amaz | +0.016 |
angles | +0.014 |
gasp | +0.014 |
horror | +0.014 |
ill | +0.013 |
Neg logit contributions | |
Nut | -0.021 |
Rust | -0.018 |
Nut | -0.018 |
str | -0.017 |
br | -0.017 |
Token | put |
Token ID | 1234 |
Feature activation | +4.50e-03 |
Pos logit contributions | |
Nut | +0.001 |
str | +0.001 |
Rust | +0.001 |
age | +0.001 |
br | +0.001 |
Neg logit contributions | |
amaz | -0.001 |
gasp | -0.001 |
horror | -0.001 |
angles | -0.001 |
bowls | -0.000 |
Token | on |
Token ID | 319 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
Nut | +0.009 |
Rust | +0.007 |
str | +0.007 |
br | +0.007 |
rob | +0.007 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.007 |
angles | -0.007 |
horror | -0.007 |
ill | -0.007 |
Token | their |
Token ID | 511 |
Feature activation | -3.72e-03 |
Pos logit contributions | |
Nut | +0.044 |
Rust | +0.037 |
rob | +0.037 |
str | +0.037 |
br | +0.036 |
Neg logit contributions | |
amaz | -0.031 |
gasp | -0.029 |
angles | -0.027 |
horror | -0.026 |
bowls | -0.026 |
Token | bath |
Token ID | 7837 |
Feature activation | +9.74e-03 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.005 |
angles | +0.005 |
ill | +0.005 |
horror | +0.005 |
Neg logit contributions | |
Nut | -0.008 |
Rust | -0.007 |
br | -0.007 |
str | -0.007 |
con | -0.007 |
Token | ro |
Token ID | 305 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
Nut | +0.022 |
Rust | +0.018 |
str | +0.018 |
Nut | +0.018 |
br | +0.018 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.014 |
angles | -0.013 |
horror | -0.013 |
bowls | -0.012 |
Token | bes |
Token ID | 12636 |
Feature activation | -6.65e-04 |
Pos logit contributions | |
Nut | +0.150 |
Nut | +0.128 |
Sy | +0.125 |
Rust | +0.124 |
Silver | +0.121 |
Neg logit contributions | |
gasp | -0.115 |
cloth | -0.113 |
amaz | -0.108 |
angles | -0.107 |
hands | -0.106 |
Token | and |
Token ID | 290 |
Feature activation | +3.68e-03 |
Pos logit contributions | |
amaz | +0.001 |
gasp | +0.001 |
angles | +0.001 |
ill | +0.001 |
bowls | +0.001 |
Neg logit contributions | |
Nut | -0.001 |
Rust | -0.001 |
str | -0.001 |
Nut | -0.001 |
br | -0.001 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
Nut | +0.007 |
Rust | +0.006 |
Nut | +0.006 |
br | +0.006 |
rob | +0.006 |
Neg logit contributions | |
amaz | -0.006 |
angles | -0.006 |
gasp | -0.006 |
horror | -0.005 |
ill | -0.005 |
Token | around |
Token ID | 1088 |
Feature activation | +8.83e-03 |
Pos logit contributions | |
Nut | +0.051 |
str | +0.044 |
rob | +0.042 |
Rust | +0.041 |
Silver | +0.041 |
Neg logit contributions | |
gasp | -0.033 |
amaz | -0.033 |
angles | -0.031 |
horror | -0.030 |
ill | -0.029 |
Token | in |
Token ID | 287 |
Feature activation | -4.23e-03 |
Pos logit contributions | |
Nut | +0.019 |
Rust | +0.016 |
str | +0.015 |
br | +0.015 |
rob | +0.015 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.013 |
angles | -0.013 |
horror | -0.013 |
ill | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +6.91e-03 |
Pos logit contributions | |
Nut | +0.000 |
Rust | +0.000 |
Nut | +0.000 |
str | +0.000 |
age | +0.000 |
Neg logit contributions | |
amaz | -0.000 |
gasp | -0.000 |
angles | -0.000 |
horror | -0.000 |
ill | -0.000 |
Token | Her |
Token ID | 2332 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
Nut | +0.014 |
Rust | +0.012 |
Nut | +0.012 |
str | +0.012 |
br | +0.012 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.011 |
angles | -0.010 |
horror | -0.010 |
ill | -0.010 |
Token | mom |
Token ID | 1995 |
Feature activation | -8.58e-03 |
Pos logit contributions | |
amaz | +0.022 |
angles | +0.021 |
gasp | +0.021 |
ill | +0.021 |
horror | +0.018 |
Neg logit contributions | |
Nut | -0.030 |
Nut | -0.026 |
Rust | -0.025 |
age | -0.024 |
Bun | -0.024 |
Token | put |
Token ID | 1234 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
amaz | +0.011 |
gasp | +0.010 |
horror | +0.010 |
angles | +0.009 |
ill | +0.009 |
Neg logit contributions | |
Nut | -0.020 |
rob | -0.018 |
Rust | -0.018 |
br | -0.018 |
str | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +3.43e-03 |
Pos logit contributions | |
Nut | +0.027 |
str | +0.025 |
br | +0.024 |
Rust | +0.023 |
con | +0.023 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.016 |
bowls | -0.015 |
horror | -0.015 |
angles | -0.014 |
Token | band |
Token ID | 4097 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
Nut | +0.007 |
rob | +0.006 |
Rust | +0.006 |
con | +0.006 |
Sy | +0.006 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.006 |
wish | -0.005 |
horror | -0.005 |
bowls | -0.005 |
Token | - |
Token ID | 12 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
Nut | +0.117 |
str | +0.105 |
rob | +0.104 |
con | +0.098 |
Sy | +0.097 |
Neg logit contributions | |
gasp | -0.137 |
amaz | -0.131 |
of | -0.127 |
eye | -0.125 |
hands | -0.124 |
Token | aid |
Token ID | 1698 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
Nut | +0.046 |
rob | +0.039 |
str | +0.039 |
Rust | +0.038 |
con | +0.037 |
Neg logit contributions | |
amaz | -0.069 |
gasp | -0.068 |
ill | -0.064 |
bowls | -0.063 |
spit | -0.062 |
Token | on |
Token ID | 319 |
Feature activation | +2.39e-03 |
Pos logit contributions | |
Nut | +0.076 |
str | +0.065 |
Sy | +0.063 |
rob | +0.062 |
con | +0.060 |
Neg logit contributions | |
gasp | -0.040 |
amaz | -0.039 |
hands | -0.038 |
horror | -0.036 |
eye | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
Nut | +0.006 |
str | +0.006 |
rob | +0.006 |
con | +0.005 |
br | +0.005 |
Neg logit contributions | |
amaz | -0.003 |
gasp | -0.002 |
horror | -0.002 |
eyes | -0.002 |
Mom | -0.002 |
Token | knee |
Token ID | 10329 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
amaz | +0.022 |
gasp | +0.020 |
angles | +0.019 |
horror | +0.018 |
bowls | +0.017 |
Neg logit contributions | |
Nut | -0.041 |
Rust | -0.036 |
br | -0.035 |
rob | -0.035 |
str | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -3.48e-03 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
angles | +0.008 |
horror | +0.007 |
ill | +0.007 |
Neg logit contributions | |
Nut | -0.013 |
Rust | -0.011 |
br | -0.011 |
rob | -0.011 |
str | -0.011 |
Token | They |
Token ID | 1119 |
Feature activation | +7.06e-03 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
angles | +0.004 |
horror | +0.004 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.008 |
Rust | -0.007 |
Nut | -0.007 |
br | -0.007 |
str | -0.007 |
Token | gave |
Token ID | 2921 |
Feature activation | -2.34e-03 |
Pos logit contributions | |
Nut | +0.014 |
Rust | +0.013 |
K | +0.012 |
Nut | +0.012 |
str | +0.012 |
Neg logit contributions | |
amaz | -0.012 |
angles | -0.011 |
gasp | -0.010 |
horror | -0.010 |
ill | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
amaz | +0.004 |
gasp | +0.004 |
angles | +0.003 |
horror | +0.003 |
bowls | +0.003 |
Neg logit contributions | |
Nut | -0.005 |
Rust | -0.004 |
br | -0.004 |
str | -0.004 |
rob | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
Nut | +0.048 |
rob | +0.041 |
Rust | +0.040 |
str | +0.040 |
br | +0.038 |
Neg logit contributions | |
amaz | -0.037 |
gasp | -0.033 |
bowls | -0.032 |
horror | -0.031 |
eyes | -0.030 |
Token | band |
Token ID | 4097 |
Feature activation | +7.32e-02 |
Pos logit contributions | |
Nut | +0.099 |
rob | +0.089 |
Sy | +0.084 |
Rust | +0.084 |
K | +0.083 |
Neg logit contributions | |
amaz | -0.045 |
gasp | -0.045 |
wish | -0.037 |
bowls | -0.037 |
wonder | -0.035 |
Token | age |
Token ID | 496 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
Nut | +0.118 |
Rust | +0.096 |
Sy | +0.094 |
rob | +0.094 |
Cle | +0.093 |
Neg logit contributions | |
gasp | -0.155 |
amaz | -0.145 |
eye | -0.136 |
hands | -0.136 |
faces | -0.134 |
Token | and |
Token ID | 290 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
Nut | +0.069 |
Rust | +0.059 |
rob | +0.058 |
str | +0.058 |
Silver | +0.057 |
Neg logit contributions | |
amaz | -0.036 |
gasp | -0.035 |
ill | -0.032 |
bowls | -0.030 |
horror | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
Nut | +0.061 |
Rust | +0.052 |
br | +0.051 |
Nut | +0.051 |
rob | +0.050 |
Neg logit contributions | |
amaz | -0.054 |
angles | -0.050 |
gasp | -0.049 |
horror | -0.047 |
ill | -0.046 |
Token | spoon |
Token ID | 24556 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
Nut | +0.051 |
rob | +0.044 |
Rust | +0.043 |
Sy | +0.042 |
Nut | +0.042 |
Neg logit contributions | |
amaz | -0.035 |
gasp | -0.033 |
horror | -0.029 |
bowls | -0.029 |
angles | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
Nut | +0.037 |
age | +0.031 |
br | +0.031 |
Rust | +0.031 |
str | +0.030 |
Neg logit contributions | |
amaz | -0.032 |
gasp | -0.029 |
angles | -0.029 |
horror | -0.028 |
bowls | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -7.06e-03 |
Pos logit contributions | |
amaz | +0.078 |
gasp | +0.074 |
angles | +0.071 |
horror | +0.069 |
bowls | +0.068 |
Neg logit contributions | |
Nut | -0.065 |
Nut | -0.053 |
Rust | -0.053 |
Sy | -0.052 |
br | -0.051 |
Token | pointed |
Token ID | 6235 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
amaz | +0.011 |
gasp | +0.010 |
angles | +0.009 |
ill | +0.009 |
horror | +0.009 |
Neg logit contributions | |
Nut | -0.015 |
Rust | -0.013 |
Nut | -0.013 |
str | -0.013 |
Sy | -0.013 |
Token | at |
Token ID | 379 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
amaz | +0.044 |
gasp | +0.041 |
angles | +0.040 |
bowls | +0.040 |
horror | +0.040 |
Neg logit contributions | |
Nut | -0.036 |
Nut | -0.030 |
Rust | -0.029 |
Sy | -0.029 |
br | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | -4.08e-03 |
Pos logit contributions | |
amaz | +0.048 |
gasp | +0.043 |
angles | +0.041 |
horror | +0.041 |
bowls | +0.038 |
Neg logit contributions | |
Nut | -0.064 |
br | -0.054 |
Rust | -0.054 |
rob | -0.054 |
str | -0.054 |
Token | in |
Token ID | 287 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
amaz | +0.007 |
gasp | +0.007 |
angles | +0.006 |
horror | +0.006 |
bowls | +0.006 |
Neg logit contributions | |
Nut | -0.008 |
Rust | -0.006 |
br | -0.006 |
Nut | -0.006 |
con | -0.006 |
Token | amaz |
Token ID | 40642 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
gasp | +0.018 |
amaz | +0.018 |
angles | +0.018 |
ill | +0.017 |
urious | +0.017 |
Neg logit contributions | |
Nut | -0.028 |
Rust | -0.025 |
Nut | -0.024 |
K | -0.024 |
br | -0.024 |
Token | ement |
Token ID | 972 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
gasp | +0.103 |
amaz | +0.099 |
horror | +0.090 |
bowls | +0.090 |
decide | +0.088 |
Neg logit contributions | |
Nut | -0.119 |
br | -0.114 |
rob | -0.109 |
pe | -0.105 |
co | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
amaz | +0.034 |
gasp | +0.030 |
horror | +0.030 |
angles | +0.030 |
bowls | +0.029 |
Neg logit contributions | |
Nut | -0.037 |
br | -0.032 |
con | -0.032 |
str | -0.031 |
age | -0.031 |
Token | The |
Token ID | 383 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
amaz | +0.026 |
gasp | +0.024 |
horror | +0.023 |
angles | +0.022 |
bowls | +0.021 |
Neg logit contributions | |
Nut | -0.037 |
rob | -0.032 |
str | -0.031 |
Rust | -0.031 |
br | -0.031 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
amaz | +0.019 |
ill | +0.019 |
gasp | +0.018 |
angles | +0.018 |
ucks | +0.018 |
Neg logit contributions | |
Nut | -0.018 |
str | -0.015 |
Rust | -0.015 |
Silver | -0.015 |
hawk | -0.015 |
Token | stopped |
Token ID | 5025 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
amaz | +0.016 |
gasp | +0.014 |
angles | +0.013 |
horror | +0.013 |
ill | +0.013 |
Neg logit contributions | |
Nut | -0.024 |
Rust | -0.021 |
rob | -0.021 |
br | -0.020 |
str | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -4.33e-03 |
Pos logit contributions | |
amaz | +0.024 |
gasp | +0.022 |
angles | +0.021 |
horror | +0.020 |
ill | +0.020 |
Neg logit contributions | |
Nut | -0.036 |
Rust | -0.031 |
br | -0.030 |
rob | -0.030 |
str | -0.030 |
Token | The |
Token ID | 383 |
Feature activation | -7.10e-03 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.005 |
gasped | +0.005 |
horror | +0.005 |
Dad | +0.005 |
Neg logit contributions | |
Nut | -0.010 |
rob | -0.010 |
str | -0.009 |
Rust | -0.009 |
br | -0.009 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -7.71e-03 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.013 |
horror | +0.012 |
ill | +0.012 |
Neg logit contributions | |
Nut | -0.013 |
Rust | -0.010 |
rob | -0.010 |
br | -0.010 |
str | -0.010 |
Token | watched |
Token ID | 7342 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
amaz | +0.011 |
gasp | +0.010 |
angles | +0.009 |
horror | +0.009 |
ill | +0.009 |
Neg logit contributions | |
Nut | -0.018 |
Rust | -0.016 |
rob | -0.016 |
br | -0.015 |
str | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
amaz | +0.070 |
gasp | +0.066 |
angles | +0.062 |
horror | +0.062 |
ill | +0.060 |
Neg logit contributions | |
Nut | -0.073 |
Rust | -0.061 |
rob | -0.061 |
str | -0.061 |
br | -0.060 |
Token | amaz |
Token ID | 40642 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
ucks | +0.010 |
ateful | +0.009 |
ill | +0.009 |
bedrooms | +0.009 |
angles | +0.008 |
Neg logit contributions | |
Nut | -0.055 |
Nut | -0.051 |
Rust | -0.050 |
Silver | -0.049 |
Sy | -0.049 |
Token | ement |
Token ID | 972 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
gasp | +0.090 |
amaz | +0.089 |
horror | +0.080 |
bowls | +0.079 |
angles | +0.078 |
Neg logit contributions | |
Nut | -0.132 |
br | -0.123 |
rob | -0.119 |
Rust | -0.114 |
Nut | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-03 |
Pos logit contributions | |
amaz | +0.046 |
gasp | +0.042 |
angles | +0.041 |
horror | +0.041 |
bowls | +0.039 |
Neg logit contributions | |
Nut | -0.034 |
con | -0.028 |
Rust | -0.027 |
Nut | -0.027 |
br | -0.027 |
Token | After |
Token ID | 2293 |
Feature activation | +8.30e-03 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
horror | +0.002 |
ill | +0.002 |
gasped | +0.001 |
Neg logit contributions | |
Nut | -0.003 |
rob | -0.003 |
str | -0.003 |
Rust | -0.003 |
br | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | -7.44e-04 |
Pos logit contributions | |
Nut | +0.016 |
rob | +0.014 |
Rust | +0.014 |
str | +0.014 |
age | +0.013 |
Neg logit contributions | |
amaz | -0.016 |
gasp | -0.014 |
horror | -0.013 |
bowls | -0.013 |
eyes | -0.012 |
Token | stopped |
Token ID | 5025 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
amaz | +0.001 |
angles | +0.001 |
gasp | +0.001 |
horror | +0.001 |
bowls | +0.001 |
Neg logit contributions | |
Nut | -0.001 |
br | -0.001 |
str | -0.001 |
Nut | -0.001 |
Rust | -0.001 |
Token | upstairs |
Token ID | 26148 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
amaz | +0.037 |
angles | +0.034 |
gasp | +0.033 |
ill | +0.032 |
horror | +0.031 |
Neg logit contributions | |
Nut | -0.044 |
str | -0.039 |
Nut | -0.038 |
Sy | -0.038 |
Rust | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -5.66e-03 |
Pos logit contributions | |
amaz | +0.065 |
gasp | +0.060 |
angles | +0.060 |
spit | +0.060 |
horror | +0.059 |
Neg logit contributions | |
br | -0.029 |
Nut | -0.027 |
co | -0.026 |
str | -0.026 |
con | -0.026 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
amaz | +0.009 |
angles | +0.009 |
gasp | +0.008 |
horror | +0.008 |
ill | +0.008 |
Neg logit contributions | |
Nut | -0.011 |
Nut | -0.010 |
Rust | -0.010 |
K | -0.010 |
str | -0.010 |
Token | around |
Token ID | 1088 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
amaz | +0.040 |
gasp | +0.036 |
angles | +0.036 |
ill | +0.036 |
horror | +0.035 |
Neg logit contributions | |
Nut | -0.037 |
br | -0.033 |
Rust | -0.033 |
Nut | -0.033 |
Sy | -0.032 |
Token | in |
Token ID | 287 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
amaz | +0.046 |
gasp | +0.043 |
ill | +0.042 |
angles | +0.041 |
urious | +0.041 |
Neg logit contributions | |
Nut | -0.040 |
Rust | -0.035 |
Nut | -0.035 |
Sy | -0.034 |
br | -0.034 |
Token | amaz |
Token ID | 40642 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
urious | +0.064 |
azed | +0.064 |
spit | +0.063 |
ill | +0.063 |
ays | +0.063 |
Neg logit contributions | |
Nut | -0.024 |
K | -0.022 |
Rust | -0.021 |
Nut | -0.020 |
Charl | -0.019 |
Token | ement |
Token ID | 972 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
gasp | +0.099 |
amaz | +0.094 |
horror | +0.087 |
angles | +0.084 |
forget | +0.084 |
Neg logit contributions | |
Nut | -0.117 |
br | -0.115 |
rob | -0.112 |
co | -0.105 |
Rust | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
amaz | +0.044 |
gasp | +0.040 |
horror | +0.040 |
angles | +0.039 |
bowls | +0.037 |
Neg logit contributions | |
Nut | -0.039 |
br | -0.034 |
con | -0.034 |
str | -0.033 |
Rust | -0.033 |
Token | Everything |
Token ID | 11391 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
amaz | +0.031 |
gasp | +0.029 |
angles | +0.029 |
horror | +0.027 |
ill | +0.026 |
Neg logit contributions | |
Nut | -0.033 |
Nut | -0.028 |
Rust | -0.028 |
Sy | -0.028 |
br | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.017 |
angles | +0.016 |
horror | +0.016 |
ill | +0.016 |
Neg logit contributions | |
Nut | -0.021 |
br | -0.019 |
Rust | -0.018 |
str | -0.018 |
rob | -0.018 |
Token | so |
Token ID | 523 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
amaz | +0.022 |
gasp | +0.021 |
angles | +0.020 |
horror | +0.020 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.019 |
Rust | -0.015 |
Nut | -0.015 |
br | -0.015 |
str | -0.015 |
Token | But |
Token ID | 1537 |
Feature activation | +4.83e-04 |
Pos logit contributions | |
amaz | +0.026 |
ill | +0.024 |
angles | +0.024 |
gasp | +0.024 |
decide | +0.024 |
Neg logit contributions | |
Nut | -0.021 |
con | -0.018 |
Nut | -0.018 |
rob | -0.017 |
Rust | -0.017 |
Token | then |
Token ID | 788 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
Nut | +0.001 |
Rust | +0.001 |
br | +0.001 |
rob | +0.001 |
age | +0.001 |
Neg logit contributions | |
amaz | -0.001 |
gasp | -0.001 |
angles | -0.001 |
horror | -0.001 |
ill | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -8.31e-03 |
Pos logit contributions | |
amaz | +0.020 |
gasp | +0.019 |
angles | +0.018 |
horror | +0.018 |
ill | +0.017 |
Neg logit contributions | |
Nut | -0.021 |
Rust | -0.018 |
rob | -0.017 |
str | -0.017 |
br | -0.017 |
Token | sun |
Token ID | 4252 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
amaz | +0.015 |
gasp | +0.013 |
angles | +0.013 |
horror | +0.012 |
ill | +0.012 |
Neg logit contributions | |
Nut | -0.017 |
br | -0.014 |
str | -0.014 |
Nut | -0.014 |
Rust | -0.014 |
Token | comes |
Token ID | 2058 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
amaz | +0.029 |
angles | +0.026 |
gasp | +0.025 |
horror | +0.024 |
ill | +0.024 |
Neg logit contributions | |
Nut | -0.034 |
br | -0.032 |
age | -0.031 |
str | -0.030 |
ch | -0.029 |
Token | out |
Token ID | 503 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
amaz | +0.069 |
gasp | +0.065 |
angles | +0.061 |
horror | +0.060 |
ill | +0.059 |
Neg logit contributions | |
Nut | -0.086 |
Rust | -0.073 |
rob | -0.073 |
str | -0.071 |
con | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
amaz | +0.136 |
angles | +0.129 |
gasp | +0.124 |
horror | +0.117 |
reflections | +0.116 |
Neg logit contributions | |
Nut | -0.078 |
br | -0.077 |
Nut | -0.068 |
ch | -0.068 |
K | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
amaz | +0.032 |
gasp | +0.029 |
angles | +0.028 |
horror | +0.028 |
ill | +0.027 |
Neg logit contributions | |
Nut | -0.035 |
Rust | -0.029 |
br | -0.028 |
str | -0.028 |
rob | -0.028 |
Token | mud |
Token ID | 17492 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
amaz | +0.021 |
gasp | +0.019 |
angles | +0.018 |
horror | +0.018 |
ill | +0.018 |
Neg logit contributions | |
Nut | -0.022 |
con | -0.019 |
Nut | -0.019 |
str | -0.019 |
br | -0.018 |
Token | starts |
Token ID | 4940 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
amaz | +0.018 |
gasp | +0.016 |
angles | +0.015 |
horror | +0.015 |
awe | +0.014 |
Neg logit contributions | |
Nut | -0.021 |
age | -0.019 |
br | -0.019 |
str | -0.018 |
con | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
amaz | +0.078 |
gasp | +0.072 |
angles | +0.070 |
horror | +0.069 |
wish | +0.065 |
Neg logit contributions | |
Nut | -0.047 |
str | -0.037 |
br | -0.036 |
K | -0.036 |
Nut | -0.036 |
Token | furry |
Token ID | 40857 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
Nut | +0.047 |
Rust | +0.040 |
br | +0.039 |
str | +0.039 |
Nut | +0.039 |
Neg logit contributions | |
amaz | -0.035 |
gasp | -0.032 |
angles | -0.031 |
horror | -0.030 |
ill | -0.029 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -2.34e-03 |
Pos logit contributions | |
amaz | +0.023 |
gasp | +0.021 |
angles | +0.021 |
horror | +0.020 |
ill | +0.020 |
Neg logit contributions | |
Nut | -0.027 |
Rust | -0.023 |
br | -0.023 |
str | -0.023 |
Nut | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +9.17e-03 |
Pos logit contributions | |
amaz | +0.004 |
gasp | +0.004 |
angles | +0.004 |
horror | +0.004 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.004 |
Rust | -0.004 |
br | -0.004 |
Nut | -0.004 |
str | -0.003 |
Token | stared |
Token ID | 18484 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
Nut | +0.020 |
Rust | +0.017 |
Nut | +0.017 |
br | +0.016 |
Sy | +0.016 |
Neg logit contributions | |
amaz | -0.014 |
angles | -0.013 |
gasp | -0.013 |
horror | -0.013 |
ill | -0.013 |
Token | in |
Token ID | 287 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
amaz | +0.077 |
gasp | +0.075 |
angles | +0.074 |
bowls | +0.073 |
horror | +0.073 |
Neg logit contributions | |
Nut | -0.065 |
Nut | -0.058 |
Sy | -0.058 |
Dove | -0.055 |
br | -0.054 |
Token | amaz |
Token ID | 40642 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
ill | +0.031 |
auc | +0.031 |
ateful | +0.030 |
urious | +0.030 |
ays | +0.029 |
Neg logit contributions | |
Nut | -0.051 |
Nut | -0.047 |
Dove | -0.046 |
Sy | -0.045 |
Rust | -0.045 |
Token | ement |
Token ID | 972 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
gasp | +0.103 |
amaz | +0.097 |
horror | +0.090 |
forget | +0.090 |
decide | +0.089 |
Neg logit contributions | |
Nut | -0.112 |
br | -0.110 |
rob | -0.104 |
co | -0.098 |
pe | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
amaz | +0.051 |
horror | +0.047 |
gasp | +0.045 |
angles | +0.044 |
bowls | +0.044 |
Neg logit contributions | |
Nut | -0.043 |
con | -0.037 |
br | -0.037 |
pe | -0.036 |
age | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
amaz | +0.025 |
gasp | +0.024 |
angles | +0.023 |
horror | +0.023 |
ill | +0.022 |
Neg logit contributions | |
Nut | -0.021 |
Rust | -0.017 |
Nut | -0.017 |
br | -0.016 |
str | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
amaz | +0.031 |
gasp | +0.029 |
angles | +0.028 |
horror | +0.028 |
bowls | +0.026 |
Neg logit contributions | |
Nut | -0.021 |
Rust | -0.017 |
br | -0.017 |
str | -0.016 |
Nut | -0.016 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
amaz | +0.059 |
gasp | +0.056 |
horror | +0.055 |
angles | +0.054 |
bowls | +0.051 |
Neg logit contributions | |
Nut | -0.026 |
Rust | -0.020 |
br | -0.020 |
K | -0.018 |
Silver | -0.018 |
Token | green |
Token ID | 4077 |
Feature activation | -8.26e-03 |
Pos logit contributions | |
amaz | +0.030 |
gasp | +0.028 |
angles | +0.026 |
horror | +0.026 |
ill | +0.025 |
Neg logit contributions | |
Nut | -0.037 |
Rust | -0.032 |
rob | -0.031 |
str | -0.031 |
br | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -1.52e-03 |
Pos logit contributions | |
amaz | +0.016 |
gasp | +0.014 |
angles | +0.014 |
horror | +0.014 |
ill | +0.013 |
Neg logit contributions | |
Nut | -0.015 |
Rust | -0.013 |
br | -0.012 |
rob | -0.012 |
str | -0.012 |
Token | yellow |
Token ID | 7872 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
angles | +0.002 |
horror | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.003 |
Rust | -0.003 |
br | -0.003 |
str | -0.003 |
rob | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.35e-03 |
Pos logit contributions | |
amaz | +0.032 |
gasp | +0.028 |
horror | +0.028 |
angles | +0.028 |
urious | +0.025 |
Neg logit contributions | |
Nut | -0.032 |
br | -0.029 |
con | -0.027 |
Nut | -0.027 |
Sy | -0.027 |
Token | look |
Token ID | 804 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
Nut | +0.003 |
Rust | +0.002 |
br | +0.002 |
rob | +0.002 |
str | +0.002 |
Neg logit contributions | |
amaz | -0.002 |
gasp | -0.002 |
angles | -0.002 |
horror | -0.002 |
ill | -0.002 |
Token | y |
Token ID | 331 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
amaz | +0.071 |
gasp | +0.065 |
ill | +0.063 |
horror | +0.063 |
angles | +0.063 |
Neg logit contributions | |
Nut | -0.062 |
br | -0.058 |
Sy | -0.052 |
Rust | -0.050 |
Nut | -0.050 |
Token | ummy |
Token ID | 13513 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
amaz | +0.105 |
gasp | +0.090 |
horror | +0.089 |
gasped | +0.082 |
angles | +0.082 |
Neg logit contributions | |
Nut | -0.131 |
br | -0.114 |
Nut | -0.112 |
con | -0.109 |
Sy | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
amaz | +0.065 |
gasp | +0.059 |
horror | +0.057 |
angles | +0.055 |
ill | +0.054 |
Neg logit contributions | |
Nut | -0.058 |
br | -0.054 |
con | -0.051 |
cone | -0.050 |
rob | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +1.07e-03 |
Pos logit contributions | |
amaz | +0.044 |
gasp | +0.041 |
angles | +0.040 |
horror | +0.039 |
ill | +0.037 |
Neg logit contributions | |
Nut | -0.033 |
Rust | -0.028 |
K | -0.027 |
br | -0.027 |
str | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Nut | +0.002 |
Rust | +0.001 |
rob | +0.001 |
br | +0.001 |
str | +0.001 |
Neg logit contributions | |
amaz | -0.002 |
gasp | -0.002 |
angles | -0.002 |
horror | -0.002 |
ill | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
amaz | +0.022 |
gasp | +0.021 |
angles | +0.020 |
horror | +0.020 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.016 |
Rust | -0.013 |
br | -0.012 |
str | -0.012 |
rob | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
amaz | +0.012 |
gasp | +0.011 |
gasped | +0.010 |
horror | +0.010 |
angles | +0.009 |
Neg logit contributions | |
Nut | -0.026 |
rob | -0.024 |
Rust | -0.023 |
br | -0.023 |
str | -0.022 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
amaz | +0.024 |
gasp | +0.022 |
angles | +0.021 |
horror | +0.021 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.043 |
Rust | -0.037 |
str | -0.037 |
br | -0.036 |
Nut | -0.036 |
Token | at |
Token ID | 379 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
amaz | +0.026 |
gasp | +0.024 |
angles | +0.023 |
horror | +0.022 |
ill | +0.022 |
Neg logit contributions | |
Nut | -0.032 |
Rust | -0.027 |
Nut | -0.027 |
str | -0.027 |
br | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
amaz | +0.023 |
angles | +0.021 |
gasp | +0.021 |
ill | +0.020 |
horror | +0.019 |
Neg logit contributions | |
Nut | -0.031 |
Rust | -0.028 |
br | -0.027 |
age | -0.027 |
str | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
amaz | +0.028 |
gasp | +0.026 |
angles | +0.026 |
horror | +0.025 |
bowls | +0.024 |
Neg logit contributions | |
Nut | -0.025 |
Rust | -0.020 |
br | -0.020 |
Nut | -0.020 |
str | -0.020 |
Token | amaz |
Token ID | 40642 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
gasp | +0.018 |
bowls | +0.017 |
ucks | +0.017 |
ays | +0.017 |
adds | +0.017 |
Neg logit contributions | |
Nut | -0.029 |
K | -0.025 |
Nut | -0.025 |
Rust | -0.025 |
Sy | -0.025 |
Token | ement |
Token ID | 972 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
gasp | +0.090 |
amaz | +0.089 |
horror | +0.082 |
bowls | +0.080 |
angles | +0.078 |
Neg logit contributions | |
Nut | -0.128 |
br | -0.117 |
rob | -0.115 |
str | -0.109 |
Nut | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-03 |
Pos logit contributions | |
amaz | +0.038 |
horror | +0.034 |
gasp | +0.034 |
angles | +0.033 |
bowls | +0.032 |
Neg logit contributions | |
Nut | -0.040 |
str | -0.033 |
Nut | -0.033 |
con | -0.033 |
br | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +2.94e-03 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
angles | +0.002 |
gasped | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.004 |
rob | -0.004 |
Rust | -0.004 |
br | -0.004 |
str | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -5.07e-03 |
Pos logit contributions | |
Nut | +0.006 |
rob | +0.005 |
br | +0.005 |
con | +0.005 |
Rust | +0.005 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.004 |
ill | -0.004 |
gasped | -0.004 |
angles | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
gasped | +0.008 |
wondered | +0.008 |
urious | +0.008 |
Neg logit contributions | |
rob | -0.009 |
Nut | -0.009 |
con | -0.008 |
Nut | -0.007 |
br | -0.007 |
Token | But |
Token ID | 1537 |
Feature activation | -1.48e-03 |
Pos logit contributions | |
amaz | +0.029 |
gasp | +0.027 |
angles | +0.027 |
horror | +0.026 |
ill | +0.026 |
Neg logit contributions | |
Nut | -0.022 |
Rust | -0.018 |
rob | -0.018 |
con | -0.017 |
br | -0.017 |
Token | then |
Token ID | 788 |
Feature activation | -6.95e-03 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
horror | +0.002 |
angles | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.003 |
Rust | -0.003 |
rob | -0.003 |
br | -0.003 |
str | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +6.77e-04 |
Pos logit contributions | |
amaz | +0.012 |
gasp | +0.011 |
horror | +0.011 |
angles | +0.011 |
ill | +0.010 |
Neg logit contributions | |
Nut | -0.014 |
Rust | -0.011 |
rob | -0.011 |
age | -0.011 |
str | -0.011 |
Token | sun |
Token ID | 4252 |
Feature activation | -8.05e-03 |
Pos logit contributions | |
Nut | +0.001 |
Rust | +0.001 |
br | +0.001 |
str | +0.001 |
rob | +0.001 |
Neg logit contributions | |
amaz | -0.001 |
gasp | -0.001 |
angles | -0.001 |
horror | -0.001 |
ill | -0.001 |
Token | came |
Token ID | 1625 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
amaz | +0.014 |
angles | +0.013 |
gasp | +0.013 |
horror | +0.012 |
ill | +0.011 |
Neg logit contributions | |
Nut | -0.015 |
br | -0.014 |
age | -0.014 |
str | -0.014 |
Rust | -0.013 |
Token | out |
Token ID | 503 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
amaz | +0.090 |
gasp | +0.084 |
angles | +0.080 |
horror | +0.079 |
ill | +0.076 |
Neg logit contributions | |
Nut | -0.099 |
Rust | -0.083 |
br | -0.081 |
str | -0.081 |
rob | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
amaz | +0.145 |
angles | +0.134 |
gasp | +0.128 |
reflections | +0.125 |
horror | +0.120 |
Neg logit contributions | |
Nut | -0.071 |
br | -0.070 |
ch | -0.062 |
K | -0.060 |
Nut | -0.059 |
Token | it |
Token ID | 340 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
amaz | +0.025 |
gasp | +0.024 |
angles | +0.022 |
horror | +0.022 |
bowls | +0.021 |
Neg logit contributions | |
Nut | -0.033 |
Rust | -0.028 |
rob | -0.027 |
str | -0.027 |
br | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
amaz | +0.036 |
gasp | +0.032 |
angles | +0.032 |
horror | +0.030 |
ill | +0.029 |
Neg logit contributions | |
Nut | -0.044 |
br | -0.039 |
Rust | -0.038 |
str | -0.038 |
rob | -0.037 |
Token | very |
Token ID | 845 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
amaz | +0.040 |
gasp | +0.037 |
horror | +0.035 |
angles | +0.035 |
bowls | +0.034 |
Neg logit contributions | |
Nut | -0.047 |
Rust | -0.040 |
rob | -0.040 |
con | -0.039 |
str | -0.039 |
Token | bright |
Token ID | 6016 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
amaz | +0.068 |
angles | +0.062 |
gasp | +0.060 |
ill | +0.060 |
horror | +0.059 |
Neg logit contributions | |
Nut | -0.054 |
str | -0.048 |
br | -0.046 |
Nut | -0.044 |
Rust | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
gasp | +0.059 |
amaz | +0.059 |
angles | +0.055 |
horror | +0.055 |
ill | +0.054 |
Neg logit contributions | |
Nut | -0.033 |
Nut | -0.025 |
Rust | -0.025 |
str | -0.025 |
pe | -0.024 |
Token | pot |
Token ID | 1787 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
amaz | +0.036 |
gasp | +0.033 |
angles | +0.032 |
ill | +0.031 |
horror | +0.031 |
Neg logit contributions | |
Nut | -0.053 |
br | -0.046 |
str | -0.046 |
Rust | -0.046 |
age | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -7.92e-03 |
Pos logit contributions | |
amaz | +0.064 |
gasp | +0.057 |
horror | +0.055 |
angles | +0.055 |
gasped | +0.049 |
Neg logit contributions | |
Nut | -0.072 |
age | -0.064 |
br | -0.064 |
str | -0.062 |
con | -0.061 |
Token | smell |
Token ID | 8508 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
amaz | +0.011 |
ill | +0.011 |
angles | +0.011 |
horror | +0.010 |
urious | +0.010 |
Neg logit contributions | |
Nut | -0.016 |
Nut | -0.014 |
str | -0.013 |
Rust | -0.013 |
con | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
amaz | +0.078 |
gasp | +0.074 |
ill | +0.072 |
angles | +0.071 |
horror | +0.071 |
Neg logit contributions | |
Nut | -0.060 |
br | -0.056 |
str | -0.050 |
Nut | -0.050 |
con | -0.048 |
Token | y |
Token ID | 331 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
amaz | +0.027 |
ill | +0.025 |
gasp | +0.025 |
horror | +0.024 |
angles | +0.024 |
Neg logit contributions | |
Nut | -0.028 |
br | -0.025 |
Rust | -0.024 |
str | -0.024 |
con | -0.023 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
amaz | +0.093 |
gasp | +0.084 |
gasped | +0.081 |
angles | +0.079 |
horror | +0.077 |
Neg logit contributions | |
Nut | -0.124 |
br | -0.114 |
ww | -0.109 |
str | -0.109 |
con | -0.108 |
Token | sauce |
Token ID | 10746 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
amaz | +0.053 |
ill | +0.048 |
horror | +0.047 |
angles | +0.047 |
gasp | +0.047 |
Neg logit contributions | |
Nut | -0.044 |
br | -0.043 |
con | -0.040 |
str | -0.039 |
rob | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
amaz | +0.041 |
horror | +0.036 |
angles | +0.035 |
gasp | +0.035 |
urious | +0.033 |
Neg logit contributions | |
br | -0.036 |
Nut | -0.035 |
age | -0.033 |
con | -0.032 |
pe | -0.032 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
amaz | +0.042 |
gasp | +0.041 |
angles | +0.039 |
horror | +0.036 |
ill | +0.035 |
Neg logit contributions | |
Nut | -0.040 |
K | -0.034 |
Nut | -0.034 |
Sy | -0.033 |
Rust | -0.033 |
Token | said |
Token ID | 531 |
Feature activation | +2.67e-03 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.017 |
angles | +0.017 |
horror | +0.016 |
ill | +0.015 |
Neg logit contributions | |
Nut | -0.027 |
Rust | -0.024 |
str | -0.023 |
br | -0.023 |
rob | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
amaz | +0.038 |
gasp | +0.035 |
angles | +0.035 |
horror | +0.035 |
urious | +0.032 |
Neg logit contributions | |
Nut | -0.028 |
br | -0.025 |
cone | -0.024 |
age | -0.024 |
con | -0.023 |
Token | It |
Token ID | 632 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
amaz | +0.020 |
gasp | +0.018 |
angles | +0.016 |
horror | +0.016 |
ill | +0.016 |
Neg logit contributions | |
Nut | -0.032 |
Rust | -0.028 |
rob | -0.027 |
br | -0.027 |
str | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
amaz | +0.023 |
gasp | +0.021 |
angles | +0.020 |
horror | +0.020 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.025 |
br | -0.021 |
str | -0.021 |
Nut | -0.021 |
Rust | -0.020 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
amaz | +0.022 |
gasp | +0.021 |
angles | +0.020 |
horror | +0.020 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.021 |
Rust | -0.017 |
rob | -0.017 |
br | -0.017 |
str | -0.017 |
Token | with |
Token ID | 351 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
amaz | +0.020 |
gasp | +0.019 |
ill | +0.018 |
angles | +0.018 |
bowls | +0.016 |
Neg logit contributions | |
Nut | -0.041 |
Rust | -0.036 |
br | -0.036 |
con | -0.036 |
rob | -0.035 |
Token | y |
Token ID | 331 |
Feature activation | -6.25e-02 |
Pos logit contributions | |
amaz | +0.020 |
gasp | +0.020 |
horror | +0.019 |
ucks | +0.019 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.018 |
Nut | -0.015 |
br | -0.014 |
Sy | -0.014 |
Rust | -0.013 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
amaz | +0.092 |
gasp | +0.083 |
horror | +0.081 |
angles | +0.077 |
gasped | +0.076 |
Neg logit contributions | |
Nut | -0.135 |
br | -0.119 |
Nut | -0.116 |
con | -0.111 |
cone | -0.109 |
Token | vegetables |
Token ID | 13701 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
amaz | +0.048 |
horror | +0.046 |
angles | +0.044 |
gasp | +0.044 |
ill | +0.042 |
Neg logit contributions | |
Nut | -0.042 |
br | -0.040 |
con | -0.037 |
Nut | -0.036 |
rob | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -1.78e-04 |
Pos logit contributions | |
amaz | +0.020 |
gasp | +0.018 |
angles | +0.017 |
horror | +0.017 |
ill | +0.016 |
Neg logit contributions | |
Nut | -0.021 |
br | -0.017 |
Nut | -0.017 |
Rust | -0.017 |
con | -0.017 |
Token | looked |
Token ID | 3114 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
amaz | +0.000 |
gasp | +0.000 |
horror | +0.000 |
ill | +0.000 |
angles | +0.000 |
Neg logit contributions | |
Nut | -0.000 |
Nut | -0.000 |
br | -0.000 |
con | -0.000 |
Sy | -0.000 |
Token | so |
Token ID | 523 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
amaz | +0.068 |
ill | +0.067 |
gasp | +0.064 |
angles | +0.063 |
ucks | +0.061 |
Neg logit contributions | |
Nut | -0.048 |
br | -0.042 |
Sy | -0.041 |
Nut | -0.040 |
str | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
amaz | +0.018 |
gasp | +0.016 |
angles | +0.015 |
horror | +0.015 |
bowls | +0.015 |
Neg logit contributions | |
Nut | -0.022 |
Rust | -0.018 |
br | -0.018 |
con | -0.018 |
str | -0.018 |
Token | ocean |
Token ID | 9151 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
amaz | +0.064 |
ill | +0.063 |
horror | +0.062 |
gasp | +0.062 |
angles | +0.059 |
Neg logit contributions | |
Nut | -0.050 |
br | -0.048 |
rob | -0.048 |
str | -0.047 |
ch | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
amaz | +0.059 |
horror | +0.055 |
gasp | +0.053 |
angles | +0.051 |
ill | +0.050 |
Neg logit contributions | |
Nut | -0.025 |
str | -0.024 |
br | -0.023 |
age | -0.021 |
rob | -0.020 |
Token | there |
Token ID | 612 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
amaz | +0.101 |
horror | +0.098 |
angles | +0.096 |
gasp | +0.096 |
ill | +0.090 |
Neg logit contributions | |
Nut | -0.061 |
Nut | -0.053 |
Sy | -0.051 |
K | -0.049 |
str | -0.049 |
Token | was |
Token ID | 373 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
amaz | +0.048 |
horror | +0.046 |
angles | +0.045 |
gasp | +0.045 |
bowls | +0.042 |
Neg logit contributions | |
Nut | -0.030 |
str | -0.026 |
Nut | -0.026 |
K | -0.025 |
br | -0.025 |
Token | an |
Token ID | 281 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
amaz | +0.024 |
gasp | +0.023 |
angles | +0.022 |
horror | +0.022 |
ill | +0.021 |
Neg logit contributions | |
Nut | -0.018 |
Rust | -0.015 |
br | -0.014 |
rob | -0.014 |
str | -0.014 |
Token | oct |
Token ID | 19318 |
Feature activation | -7.97e-03 |
Pos logit contributions | |
amaz | +0.105 |
gasp | +0.099 |
ill | +0.096 |
angles | +0.094 |
horror | +0.092 |
Neg logit contributions | |
Nut | -0.119 |
str | -0.104 |
Nut | -0.103 |
Sy | -0.100 |
rob | -0.097 |
Token | opus |
Token ID | 25790 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
amaz | +0.016 |
gasp | +0.016 |
angles | +0.015 |
ill | +0.014 |
horror | +0.014 |
Neg logit contributions | |
Nut | -0.013 |
Rust | -0.011 |
Nut | -0.010 |
str | -0.010 |
K | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
amaz | +0.017 |
gasp | +0.016 |
angles | +0.015 |
horror | +0.015 |
ill | +0.014 |
Neg logit contributions | |
Nut | -0.025 |
Rust | -0.022 |
br | -0.021 |
rob | -0.021 |
str | -0.021 |
Token | The |
Token ID | 383 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
amaz | +0.030 |
gasp | +0.027 |
angles | +0.027 |
horror | +0.026 |
ill | +0.025 |
Neg logit contributions | |
Nut | -0.030 |
Rust | -0.025 |
Nut | -0.025 |
br | -0.025 |
str | -0.025 |
Token | oct |
Token ID | 19318 |
Feature activation | -9.74e-03 |
Pos logit contributions | |
amaz | +0.040 |
gasp | +0.036 |
ill | +0.036 |
horror | +0.036 |
angles | +0.035 |
Neg logit contributions | |
Nut | -0.043 |
str | -0.037 |
Nut | -0.036 |
rob | -0.036 |
br | -0.035 |
Token | that |
Token ID | 326 |
Feature activation | -2.34e-03 |
Pos logit contributions | |
amaz | +0.050 |
gasp | +0.045 |
angles | +0.045 |
horror | +0.044 |
ill | +0.042 |
Neg logit contributions | |
Nut | -0.037 |
br | -0.030 |
Rust | -0.030 |
Nut | -0.029 |
str | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | -5.97e-03 |
Pos logit contributions | |
amaz | +0.003 |
gasp | +0.003 |
angles | +0.003 |
horror | +0.003 |
ill | +0.003 |
Neg logit contributions | |
Nut | -0.005 |
Rust | -0.005 |
rob | -0.004 |
br | -0.004 |
str | -0.004 |
Token | green |
Token ID | 4077 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
amaz | +0.010 |
gasp | +0.009 |
angles | +0.009 |
horror | +0.008 |
ill | +0.008 |
Neg logit contributions | |
Nut | -0.012 |
Rust | -0.011 |
rob | -0.011 |
str | -0.010 |
br | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +3.63e-03 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.018 |
angles | +0.017 |
horror | +0.017 |
ill | +0.016 |
Neg logit contributions | |
Nut | -0.021 |
Rust | -0.018 |
br | -0.017 |
rob | -0.017 |
str | -0.017 |
Token | looked |
Token ID | 3114 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
Nut | +0.008 |
Rust | +0.006 |
br | +0.006 |
str | +0.006 |
rob | +0.006 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.006 |
angles | -0.005 |
horror | -0.005 |
ill | -0.005 |
Token | y |
Token ID | 331 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
amaz | +0.084 |
gasp | +0.080 |
ill | +0.080 |
angles | +0.077 |
horror | +0.076 |
Neg logit contributions | |
Nut | -0.051 |
Nut | -0.041 |
Sy | -0.040 |
br | -0.039 |
Rust | -0.037 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
amaz | +0.103 |
gasp | +0.089 |
horror | +0.086 |
angles | +0.082 |
gasped | +0.080 |
Neg logit contributions | |
Nut | -0.134 |
br | -0.115 |
Nut | -0.115 |
Sy | -0.109 |
cone | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | -6.87e-03 |
Pos logit contributions | |
amaz | +0.051 |
gasp | +0.047 |
horror | +0.044 |
angles | +0.043 |
ill | +0.043 |
Neg logit contributions | |
Nut | -0.053 |
br | -0.047 |
con | -0.046 |
cone | -0.046 |
Nut | -0.045 |
Token | It |
Token ID | 632 |
Feature activation | -9.58e-03 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
horror | +0.007 |
bowls | +0.007 |
wonder | +0.007 |
Neg logit contributions | |
Nut | -0.015 |
rob | -0.014 |
br | -0.014 |
str | -0.013 |
Rust | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.012 |
horror | +0.012 |
ill | +0.011 |
Neg logit contributions | |
Nut | -0.022 |
Rust | -0.019 |
rob | -0.018 |
br | -0.018 |
str | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
amaz | +0.022 |
gasp | +0.020 |
horror | +0.019 |
angles | +0.019 |
bowls | +0.018 |
Neg logit contributions | |
Nut | -0.021 |
rob | -0.018 |
Rust | -0.018 |
str | -0.017 |
br | -0.017 |
Token | then |
Token ID | 788 |
Feature activation | -8.94e-03 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
horror | +0.005 |
angles | +0.005 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.006 |
Rust | -0.005 |
rob | -0.005 |
br | -0.005 |
str | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.018 |
angles | +0.017 |
horror | +0.017 |
ill | +0.017 |
Neg logit contributions | |
Nut | -0.014 |
br | -0.012 |
Nut | -0.012 |
Rust | -0.012 |
str | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +5.72e-04 |
Pos logit contributions | |
amaz | +0.045 |
angles | +0.043 |
gasp | +0.043 |
ill | +0.040 |
horror | +0.040 |
Neg logit contributions | |
Nut | -0.033 |
Nut | -0.027 |
Rust | -0.026 |
br | -0.026 |
Sy | -0.026 |
Token | sun |
Token ID | 4252 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Nut | +0.001 |
str | +0.001 |
Rust | +0.001 |
br | +0.001 |
con | +0.001 |
Neg logit contributions | |
amaz | -0.001 |
gasp | -0.001 |
angles | -0.001 |
horror | -0.001 |
ill | -0.001 |
Token | comes |
Token ID | 2058 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
amaz | +0.033 |
angles | +0.029 |
gasp | +0.028 |
ill | +0.027 |
horror | +0.026 |
Neg logit contributions | |
br | -0.039 |
Nut | -0.038 |
age | -0.036 |
str | -0.035 |
ch | -0.034 |
Token | out |
Token ID | 503 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
amaz | +0.079 |
gasp | +0.073 |
angles | +0.071 |
horror | +0.069 |
ill | +0.066 |
Neg logit contributions | |
Nut | -0.080 |
Rust | -0.067 |
br | -0.066 |
Nut | -0.065 |
str | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
amaz | +0.141 |
angles | +0.128 |
gasp | +0.126 |
reflections | +0.119 |
bedrooms | +0.114 |
Neg logit contributions | |
br | -0.076 |
Nut | -0.070 |
ch | -0.068 |
Bun | -0.061 |
Nut | -0.060 |
Token | It |
Token ID | 632 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
amaz | +0.067 |
angles | +0.064 |
gasp | +0.062 |
horror | +0.057 |
bowls | +0.056 |
Neg logit contributions | |
Nut | -0.070 |
br | -0.059 |
Nut | -0.058 |
Rust | -0.057 |
Sy | -0.056 |
Token | is |
Token ID | 318 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
amaz | +0.039 |
angles | +0.035 |
ill | +0.034 |
gasp | +0.034 |
horror | +0.031 |
Neg logit contributions | |
br | -0.047 |
Nut | -0.047 |
str | -0.042 |
rob | -0.042 |
con | -0.041 |
Token | warm |
Token ID | 5814 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
amaz | +0.037 |
gasp | +0.035 |
angles | +0.033 |
horror | +0.033 |
ill | +0.032 |
Neg logit contributions | |
Nut | -0.042 |
Rust | -0.035 |
rob | -0.034 |
br | -0.034 |
str | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -8.05e-03 |
Pos logit contributions | |
amaz | +0.030 |
gasp | +0.028 |
angles | +0.027 |
horror | +0.026 |
ill | +0.025 |
Neg logit contributions | |
Nut | -0.023 |
br | -0.018 |
Rust | -0.018 |
rob | -0.018 |
Nut | -0.018 |
Token | hello |
Token ID | 23748 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
Nut | +0.005 |
rob | +0.005 |
Rust | +0.004 |
con | +0.004 |
cone | +0.004 |
Neg logit contributions | |
amaz | -0.003 |
gasp | -0.003 |
horror | -0.003 |
angles | -0.003 |
ill | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
amaz | +0.028 |
gasp | +0.026 |
angles | +0.025 |
horror | +0.024 |
bowls | +0.023 |
Neg logit contributions | |
Nut | -0.033 |
Rust | -0.028 |
Nut | -0.027 |
br | -0.027 |
str | -0.027 |
Token | The |
Token ID | 383 |
Feature activation | +7.91e-03 |
Pos logit contributions | |
amaz | +0.020 |
gasp | +0.018 |
angles | +0.017 |
horror | +0.017 |
ill | +0.016 |
Neg logit contributions | |
Nut | -0.029 |
Rust | -0.025 |
br | -0.024 |
str | -0.024 |
rob | -0.024 |
Token | dog |
Token ID | 3290 |
Feature activation | +5.43e-03 |
Pos logit contributions | |
Nut | +0.014 |
Rust | +0.012 |
Nut | +0.011 |
br | +0.011 |
rob | +0.011 |
Neg logit contributions | |
amaz | -0.016 |
gasp | -0.015 |
angles | -0.014 |
horror | -0.014 |
ill | -0.014 |
Token | l |
Token ID | 300 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
Nut | +0.012 |
Rust | +0.010 |
rob | +0.010 |
br | +0.010 |
str | +0.010 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.007 |
horror | -0.007 |
angles | -0.007 |
gasped | -0.007 |
Token | icked |
Token ID | 9484 |
Feature activation | -6.19e-02 |
Pos logit contributions | |
amaz | +0.038 |
gasp | +0.034 |
angles | +0.032 |
horror | +0.030 |
bowls | +0.029 |
Neg logit contributions | |
Nut | -0.067 |
Rust | -0.058 |
br | -0.058 |
rob | -0.057 |
Nut | -0.056 |
Token | his |
Token ID | 465 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
amaz | +0.141 |
angles | +0.126 |
gasp | +0.125 |
urious | +0.119 |
horror | +0.115 |
Neg logit contributions | |
Nut | -0.093 |
Nut | -0.079 |
Dove | -0.072 |
Sy | -0.072 |
Bun | -0.070 |
Token | face |
Token ID | 1986 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.017 |
angles | +0.013 |
horror | +0.013 |
ill | +0.012 |
Neg logit contributions | |
Nut | -0.065 |
Rust | -0.058 |
rob | -0.057 |
str | -0.057 |
br | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
amaz | +0.031 |
gasp | +0.029 |
angles | +0.028 |
horror | +0.027 |
ill | +0.026 |
Neg logit contributions | |
Nut | -0.037 |
Rust | -0.031 |
br | -0.030 |
str | -0.030 |
rob | -0.030 |
Token | w |
Token ID | 266 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
Nut | +0.020 |
Rust | +0.017 |
str | +0.017 |
br | +0.017 |
rob | +0.017 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.016 |
angles | -0.015 |
horror | -0.015 |
ill | -0.014 |
Token | agged |
Token ID | 14655 |
Feature activation | -2.40e-03 |
Pos logit contributions | |
amaz | +0.044 |
gasp | +0.041 |
angles | +0.039 |
horror | +0.039 |
ill | +0.038 |
Neg logit contributions | |
Nut | -0.052 |
Rust | -0.044 |
str | -0.043 |
br | -0.043 |
Nut | -0.043 |
Token | around |
Token ID | 1088 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
amaz | +0.038 |
horror | +0.035 |
angles | +0.033 |
bowls | +0.033 |
gasp | +0.033 |
Neg logit contributions | |
Nut | -0.041 |
str | -0.039 |
br | -0.037 |
Rust | -0.036 |
Silver | -0.036 |
Token | started |
Token ID | 2067 |
Feature activation | +9.59e-03 |
Pos logit contributions | |
amaz | +0.041 |
gasp | +0.038 |
angles | +0.037 |
horror | +0.037 |
urious | +0.035 |
Neg logit contributions | |
Nut | -0.039 |
br | -0.034 |
Nut | -0.034 |
Rust | -0.033 |
str | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
Nut | +0.016 |
Rust | +0.013 |
br | +0.013 |
str | +0.013 |
Nut | +0.013 |
Neg logit contributions | |
amaz | -0.019 |
gasp | -0.018 |
angles | -0.017 |
horror | -0.017 |
ill | -0.017 |
Token | gasp |
Token ID | 47440 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
amaz | +0.023 |
angles | +0.022 |
horror | +0.021 |
gasp | +0.021 |
bowls | +0.019 |
Neg logit contributions | |
Nut | -0.041 |
Rust | -0.035 |
Nut | -0.035 |
br | -0.035 |
str | -0.034 |
Token | in |
Token ID | 287 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
amaz | +0.106 |
angles | +0.105 |
gasp | +0.103 |
horror | +0.099 |
bowls | +0.095 |
Neg logit contributions | |
Nut | -0.079 |
br | -0.079 |
Bun | -0.070 |
Rust | -0.066 |
Nut | -0.066 |
Token | amaz |
Token ID | 40642 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
ill | +0.022 |
urious | +0.021 |
auc | +0.020 |
ucks | +0.020 |
gasp | +0.019 |
Neg logit contributions | |
Nut | -0.054 |
Nut | -0.050 |
Sy | -0.049 |
Rust | -0.048 |
Dove | -0.048 |
Token | ement |
Token ID | 972 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
gasp | +0.094 |
amaz | +0.090 |
horror | +0.083 |
bowls | +0.081 |
angles | +0.079 |
Neg logit contributions | |
Nut | -0.119 |
br | -0.115 |
rob | -0.109 |
Rust | -0.103 |
str | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | -8.63e-03 |
Pos logit contributions | |
amaz | +0.044 |
horror | +0.040 |
gasp | +0.039 |
angles | +0.039 |
bowls | +0.038 |
Neg logit contributions | |
Nut | -0.044 |
br | -0.039 |
con | -0.039 |
age | -0.038 |
str | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
amaz | +0.015 |
gasp | +0.013 |
angles | +0.013 |
horror | +0.013 |
ill | +0.012 |
Neg logit contributions | |
Nut | -0.018 |
Rust | -0.015 |
Nut | -0.015 |
str | -0.014 |
br | -0.014 |
Token | magical |
Token ID | 10883 |
Feature activation | -3.66e-03 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.018 |
ill | +0.018 |
angles | +0.018 |
horror | +0.017 |
Neg logit contributions | |
Nut | -0.019 |
br | -0.017 |
Silver | -0.017 |
Rust | -0.017 |
str | -0.016 |
Token | cut |
Token ID | 2005 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.006 |
horror | +0.005 |
angles | +0.005 |
ill | +0.005 |
Neg logit contributions | |
Nut | -0.007 |
br | -0.006 |
Rust | -0.006 |
str | -0.006 |
con | -0.006 |
Token | would |
Token ID | 561 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
amaz | +0.045 |
horror | +0.039 |
gasp | +0.039 |
angles | +0.037 |
ill | +0.036 |
Neg logit contributions | |
Nut | -0.050 |
br | -0.043 |
str | -0.043 |
Rust | -0.043 |
rob | -0.042 |
Token | stop |
Token ID | 2245 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
amaz | +0.022 |
gasp | +0.020 |
angles | +0.019 |
horror | +0.018 |
ill | +0.017 |
Neg logit contributions | |
Nut | -0.036 |
Rust | -0.031 |
rob | -0.030 |
str | -0.030 |
br | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
amaz | +0.058 |
gasp | +0.054 |
horror | +0.051 |
angles | +0.050 |
bowls | +0.049 |
Neg logit contributions | |
Nut | -0.055 |
Nut | -0.045 |
Rust | -0.045 |
rob | -0.044 |
Sy | -0.044 |
Token | stare |
Token ID | 24170 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
amaz | +0.023 |
gasp | +0.020 |
angles | +0.020 |
horror | +0.020 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.032 |
Nut | -0.027 |
Rust | -0.027 |
con | -0.027 |
br | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
amaz | +0.068 |
horror | +0.063 |
gasp | +0.062 |
angles | +0.062 |
ill | +0.057 |
Neg logit contributions | |
Nut | -0.077 |
br | -0.073 |
Nut | -0.068 |
Rust | -0.067 |
Sy | -0.067 |
Token | amaz |
Token ID | 40642 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
gasp | +0.049 |
ill | +0.047 |
urious | +0.047 |
spit | +0.046 |
amaz | +0.045 |
Neg logit contributions | |
Nut | -0.034 |
Nut | -0.029 |
Rust | -0.027 |
Sy | -0.027 |
br | -0.025 |
Token | ement |
Token ID | 972 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
gasp | +0.091 |
amaz | +0.089 |
horror | +0.080 |
angles | +0.077 |
bowls | +0.074 |
Neg logit contributions | |
Nut | -0.127 |
br | -0.122 |
rob | -0.119 |
Rust | -0.111 |
con | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | +5.93e-03 |
Pos logit contributions | |
amaz | +0.054 |
horror | +0.049 |
gasp | +0.047 |
angles | +0.046 |
bowls | +0.044 |
Neg logit contributions | |
Nut | -0.051 |
br | -0.047 |
con | -0.045 |
age | -0.044 |
zag | -0.043 |
Token | |
Token ID | 220 |
Feature activation | +3.51e-03 |
Pos logit contributions | |
Nut | +0.013 |
br | +0.012 |
Rust | +0.011 |
Charl | +0.011 |
Silver | +0.011 |
Neg logit contributions | |
amaz | -0.010 |
gasp | -0.008 |
wonder | -0.008 |
angles | -0.008 |
ill | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
Nut | +0.006 |
Rust | +0.005 |
rob | +0.005 |
br | +0.005 |
str | +0.005 |
Neg logit contributions | |
amaz | -0.007 |
gasp | -0.006 |
angles | -0.006 |
horror | -0.006 |
ill | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
Nut | +0.041 |
br | +0.036 |
con | +0.036 |
Rust | +0.036 |
cone | +0.036 |
Neg logit contributions | |
ill | -0.021 |
amaz | -0.021 |
gasp | -0.020 |
angles | -0.020 |
wonders | -0.019 |
Token | down |
Token ID | 866 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
amaz | +0.013 |
gasp | +0.013 |
angles | +0.012 |
horror | +0.011 |
ill | +0.011 |
Neg logit contributions | |
Nut | -0.028 |
rob | -0.025 |
Rust | -0.025 |
str | -0.025 |
con | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
amaz | +0.041 |
gasp | +0.038 |
angles | +0.036 |
horror | +0.035 |
ill | +0.033 |
Neg logit contributions | |
Nut | -0.055 |
Rust | -0.047 |
br | -0.047 |
Nut | -0.046 |
rob | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +3.90e-03 |
Pos logit contributions | |
Nut | +0.028 |
Rust | +0.024 |
br | +0.023 |
str | +0.023 |
rob | +0.023 |
Neg logit contributions | |
amaz | -0.022 |
gasp | -0.021 |
angles | -0.019 |
horror | -0.019 |
bowls | -0.018 |
Token | stars |
Token ID | 5788 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
Nut | +0.008 |
Rust | +0.007 |
br | +0.007 |
rob | +0.007 |
str | +0.007 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.006 |
angles | -0.005 |
horror | -0.005 |
ill | -0.005 |
Token | came |
Token ID | 1625 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
amaz | +0.043 |
angles | +0.039 |
horror | +0.038 |
gasp | +0.037 |
urious | +0.035 |
Neg logit contributions | |
Nut | -0.039 |
br | -0.039 |
age | -0.036 |
str | -0.034 |
Rust | -0.034 |
Token | out |
Token ID | 503 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
amaz | +0.073 |
gasp | +0.067 |
angles | +0.065 |
horror | +0.064 |
bowls | +0.062 |
Neg logit contributions | |
Nut | -0.068 |
br | -0.056 |
Rust | -0.056 |
Nut | -0.056 |
K | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
amaz | +0.126 |
gasp | +0.120 |
spit | +0.120 |
angles | +0.115 |
horror | +0.111 |
Neg logit contributions | |
br | -0.077 |
Nut | -0.071 |
ch | -0.068 |
Bun | -0.064 |
cho | -0.062 |
Token | They |
Token ID | 1119 |
Feature activation | -8.98e-03 |
Pos logit contributions | |
amaz | +0.018 |
gasp | +0.017 |
horror | +0.016 |
angles | +0.016 |
ill | +0.015 |
Neg logit contributions | |
Nut | -0.021 |
Rust | -0.018 |
rob | -0.017 |
str | -0.017 |
br | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | -8.11e-03 |
Pos logit contributions | |
amaz | +0.014 |
angles | +0.013 |
gasp | +0.013 |
horror | +0.012 |
urious | +0.012 |
Neg logit contributions | |
Nut | -0.019 |
Rust | -0.016 |
str | -0.016 |
K | -0.016 |
Nut | -0.016 |
Token | goodbye |
Token ID | 24829 |
Feature activation | +6.05e-03 |
Pos logit contributions | |
amaz | +0.015 |
gasp | +0.014 |
angles | +0.014 |
horror | +0.013 |
ill | +0.013 |
Neg logit contributions | |
Nut | -0.015 |
Rust | -0.013 |
rob | -0.012 |
str | -0.012 |
br | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Nut | +0.015 |
rob | +0.014 |
Rust | +0.013 |
Mic | +0.012 |
str | +0.012 |
Neg logit contributions | |
gasp | -0.008 |
amaz | -0.008 |
ill | -0.007 |
eyes | -0.007 |
faces | -0.007 |
Token | m |
Token ID | 285 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
Nut | +0.012 |
rob | +0.011 |
age | +0.011 |
Rust | +0.011 |
con | +0.010 |
Neg logit contributions | |
amaz | -0.005 |
gasp | -0.004 |
horror | -0.004 |
wish | -0.004 |
wonder | -0.004 |
Token | ama |
Token ID | 1689 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
amaz | +0.030 |
gasp | +0.028 |
angles | +0.027 |
horror | +0.027 |
ill | +0.026 |
Neg logit contributions | |
Nut | -0.020 |
Rust | -0.016 |
br | -0.015 |
rob | -0.015 |
Nut | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +3.09e-03 |
Pos logit contributions | |
amaz | +0.030 |
gasp | +0.027 |
angles | +0.026 |
horror | +0.026 |
ill | +0.025 |
Neg logit contributions | |
Nut | -0.036 |
Rust | -0.030 |
str | -0.030 |
br | -0.030 |
Nut | -0.029 |
Token | dad |
Token ID | 9955 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
Nut | +0.006 |
Rust | +0.005 |
br | +0.005 |
rob | +0.005 |
str | +0.005 |
Neg logit contributions | |
amaz | -0.005 |
gasp | -0.005 |
angles | -0.005 |
horror | -0.005 |
ill | -0.004 |
Token | lay |
Token ID | 3830 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
amaz | +0.023 |
angles | +0.021 |
horror | +0.020 |
gasp | +0.020 |
reflections | +0.018 |
Neg logit contributions | |
Nut | -0.024 |
Nut | -0.021 |
K | -0.020 |
str | -0.020 |
Sy | -0.020 |
Token | an |
Token ID | 281 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
amaz | +0.038 |
gasp | +0.035 |
angles | +0.033 |
horror | +0.033 |
bowls | +0.032 |
Neg logit contributions | |
Nut | -0.046 |
Rust | -0.039 |
br | -0.038 |
str | -0.038 |
rob | -0.038 |
Token | infant |
Token ID | 11212 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
amaz | +0.106 |
gasp | +0.097 |
angles | +0.096 |
ill | +0.092 |
horror | +0.088 |
Neg logit contributions | |
Nut | -0.119 |
Rust | -0.100 |
str | -0.100 |
age | -0.100 |
Nut | -0.099 |
Token | in |
Token ID | 287 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
amaz | +0.044 |
angles | +0.040 |
gasp | +0.040 |
horror | +0.039 |
ill | +0.038 |
Neg logit contributions | |
Nut | -0.034 |
Rust | -0.030 |
age | -0.029 |
br | -0.028 |
K | -0.028 |
Token | their |
Token ID | 511 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
amaz | +0.026 |
gasp | +0.025 |
angles | +0.024 |
horror | +0.023 |
ill | +0.023 |
Neg logit contributions | |
Nut | -0.022 |
Rust | -0.018 |
Nut | -0.018 |
br | -0.017 |
rob | -0.017 |
Token | bed |
Token ID | 3996 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
amaz | +0.045 |
ill | +0.043 |
gasp | +0.042 |
angles | +0.042 |
horror | +0.038 |
Neg logit contributions | |
Nut | -0.063 |
Rust | -0.056 |
age | -0.054 |
str | -0.054 |
Nut | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -9.91e-03 |
Pos logit contributions | |
amaz | +0.034 |
gasp | +0.031 |
angles | +0.030 |
horror | +0.029 |
ill | +0.027 |
Neg logit contributions | |
Nut | -0.048 |
Rust | -0.042 |
br | -0.041 |
age | -0.041 |
str | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | -7.07e-03 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.006 |
angles | +0.006 |
horror | +0.006 |
ill | +0.005 |
Neg logit contributions | |
Nut | -0.009 |
Rust | -0.008 |
str | -0.008 |
Nut | -0.008 |
br | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +7.55e-03 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
horror | +0.013 |
angles | +0.012 |
bowls | +0.012 |
Neg logit contributions | |
Nut | -0.012 |
str | -0.010 |
Nut | -0.010 |
con | -0.010 |
br | -0.009 |
Token | but |
Token ID | 475 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
Nut | +0.013 |
Rust | +0.011 |
br | +0.011 |
rob | +0.011 |
str | +0.011 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.014 |
angles | -0.013 |
horror | -0.013 |
ill | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -5.48e-03 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.006 |
angles | +0.006 |
ucks | +0.006 |
ill | +0.006 |
Neg logit contributions | |
Nut | -0.003 |
Sy | -0.003 |
Nut | -0.003 |
Dove | -0.003 |
K | -0.003 |
Token | looked |
Token ID | 3114 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
angles | +0.008 |
horror | +0.007 |
ill | +0.007 |
Neg logit contributions | |
Nut | -0.012 |
Rust | -0.010 |
str | -0.010 |
br | -0.010 |
Nut | -0.010 |
Token | y |
Token ID | 331 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
amaz | +0.062 |
gasp | +0.060 |
ill | +0.058 |
angles | +0.058 |
horror | +0.056 |
Neg logit contributions | |
Nut | -0.049 |
Nut | -0.040 |
Sy | -0.040 |
br | -0.039 |
Rust | -0.038 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
amaz | +0.101 |
gasp | +0.089 |
horror | +0.086 |
angles | +0.082 |
awe | +0.081 |
Neg logit contributions | |
Nut | -0.124 |
br | -0.107 |
Nut | -0.105 |
cone | -0.104 |
con | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
amaz | +0.045 |
gasp | +0.042 |
horror | +0.040 |
angles | +0.039 |
ill | +0.038 |
Neg logit contributions | |
Nut | -0.045 |
con | -0.040 |
cone | -0.040 |
br | -0.040 |
rob | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
amaz | +0.041 |
gasp | +0.038 |
angles | +0.037 |
horror | +0.036 |
bowls | +0.034 |
Neg logit contributions | |
Nut | -0.029 |
Nut | -0.024 |
str | -0.023 |
K | -0.023 |
Sy | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
amaz | +0.003 |
gasp | +0.003 |
angles | +0.003 |
horror | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.002 |
Rust | -0.002 |
rob | -0.002 |
br | -0.002 |
str | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
amaz | +0.028 |
gasp | +0.026 |
angles | +0.025 |
horror | +0.025 |
ill | +0.024 |
Neg logit contributions | |
Nut | -0.017 |
Rust | -0.014 |
str | -0.013 |
br | -0.013 |
K | -0.013 |
Token | melt |
Token ID | 16867 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
amaz | +0.059 |
gasp | +0.053 |
angles | +0.051 |
horror | +0.049 |
ill | +0.047 |
Neg logit contributions | |
Nut | -0.090 |
Rust | -0.077 |
br | -0.076 |
str | -0.076 |
Nut | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | -7.74e-03 |
Pos logit contributions | |
amaz | +0.066 |
gasp | +0.061 |
horror | +0.058 |
angles | +0.057 |
ill | +0.052 |
Neg logit contributions | |
Nut | -0.071 |
br | -0.060 |
Nut | -0.060 |
rob | -0.060 |
Rust | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | -8.06e-03 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.013 |
ill | +0.012 |
horror | +0.012 |
Neg logit contributions | |
Nut | -0.015 |
Nut | -0.012 |
Rust | -0.012 |
br | -0.012 |
str | -0.012 |
Token | sun |
Token ID | 4252 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
amaz | +0.015 |
gasp | +0.014 |
ill | +0.014 |
angles | +0.014 |
horror | +0.013 |
Neg logit contributions | |
Nut | -0.014 |
str | -0.012 |
Rust | -0.011 |
Nut | -0.011 |
age | -0.011 |
Token | came |
Token ID | 1625 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
amaz | +0.024 |
angles | +0.021 |
gasp | +0.020 |
horror | +0.020 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.024 |
age | -0.022 |
br | -0.022 |
str | -0.021 |
Nut | -0.020 |
Token | out |
Token ID | 503 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
amaz | +0.067 |
gasp | +0.062 |
angles | +0.060 |
horror | +0.058 |
ill | +0.056 |
Neg logit contributions | |
Nut | -0.067 |
Rust | -0.055 |
br | -0.054 |
Nut | -0.054 |
str | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -4.81e-03 |
Pos logit contributions | |
amaz | +0.128 |
angles | +0.120 |
gasp | +0.120 |
spit | +0.114 |
ateful | +0.113 |
Neg logit contributions | |
br | -0.074 |
Nut | -0.072 |
Bun | -0.068 |
ch | -0.067 |
Nut | -0.067 |
Token | People |
Token ID | 4380 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
amaz | +0.006 |
eyes | +0.005 |
horror | +0.005 |
gasp | +0.005 |
Mom | +0.005 |
Neg logit contributions | |
Nut | -0.011 |
rob | -0.010 |
str | -0.010 |
Silver | -0.010 |
pe | -0.009 |
Token | were |
Token ID | 547 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
amaz | +0.034 |
angles | +0.030 |
gasp | +0.029 |
horror | +0.029 |
bowls | +0.027 |
Neg logit contributions | |
Nut | -0.043 |
str | -0.038 |
Rust | -0.038 |
br | -0.037 |
age | -0.036 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
amaz | +0.034 |
angles | +0.031 |
gasp | +0.031 |
ill | +0.030 |
urious | +0.029 |
Neg logit contributions | |
Nut | -0.032 |
str | -0.027 |
Rust | -0.027 |
Nut | -0.027 |
age | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
amaz | +0.024 |
gasp | +0.023 |
angles | +0.023 |
horror | +0.022 |
bowls | +0.021 |
Neg logit contributions | |
Nut | -0.016 |
Nut | -0.014 |
br | -0.013 |
con | -0.013 |
Rust | -0.013 |
Token | The |
Token ID | 383 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
amaz | +0.032 |
gasp | +0.030 |
angles | +0.029 |
horror | +0.028 |
ill | +0.027 |
Neg logit contributions | |
Nut | -0.043 |
Rust | -0.036 |
Nut | -0.035 |
br | -0.035 |
str | -0.035 |
Token | road |
Token ID | 2975 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
Nut | +0.028 |
Rust | +0.024 |
br | +0.023 |
str | +0.023 |
rob | +0.023 |
Neg logit contributions | |
amaz | -0.022 |
gasp | -0.021 |
angles | -0.020 |
horror | -0.019 |
ill | -0.019 |
Token | is |
Token ID | 318 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
Nut | +0.030 |
Rust | +0.026 |
rob | +0.025 |
br | +0.025 |
str | +0.025 |
Neg logit contributions | |
amaz | -0.016 |
gasp | -0.015 |
angles | -0.014 |
horror | -0.013 |
ill | -0.013 |
Token | for |
Token ID | 329 |
Feature activation | -3.04e-03 |
Pos logit contributions | |
Nut | +0.032 |
rob | +0.028 |
Rust | +0.027 |
con | +0.026 |
Silver | +0.026 |
Neg logit contributions | |
amaz | -0.030 |
gasp | -0.028 |
horror | -0.027 |
angles | -0.026 |
bowls | -0.025 |
Token | cars |
Token ID | 5006 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
angles | +0.004 |
horror | +0.004 |
bowls | +0.004 |
Neg logit contributions | |
Nut | -0.006 |
Rust | -0.005 |
rob | -0.005 |
str | -0.005 |
br | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +5.69e-02 |
Pos logit contributions | |
Nut | +0.065 |
Rust | +0.057 |
rob | +0.055 |
str | +0.054 |
Silver | +0.054 |
Neg logit contributions | |
amaz | -0.039 |
gasp | -0.037 |
angles | -0.034 |
horror | -0.034 |
ill | -0.033 |
Token | trucks |
Token ID | 13960 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Nut | +0.123 |
str | +0.105 |
Rust | +0.105 |
rob | +0.101 |
age | +0.101 |
Neg logit contributions | |
amaz | -0.091 |
gasp | -0.083 |
angles | -0.078 |
horror | -0.077 |
bowls | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Nut | +0.052 |
Rust | +0.045 |
rob | +0.044 |
str | +0.044 |
br | +0.044 |
Neg logit contributions | |
amaz | -0.029 |
gasp | -0.027 |
angles | -0.025 |
horror | -0.024 |
ill | -0.023 |
Token | You |
Token ID | 921 |
Feature activation | +5.51e-03 |
Pos logit contributions | |
amaz | +0.035 |
gasp | +0.032 |
angles | +0.031 |
horror | +0.029 |
ill | +0.028 |
Neg logit contributions | |
Nut | -0.047 |
Rust | -0.040 |
br | -0.039 |
Nut | -0.039 |
rob | -0.039 |
Token | could |
Token ID | 714 |
Feature activation | -1.70e-03 |
Pos logit contributions | |
Nut | +0.014 |
Rust | +0.012 |
rob | +0.012 |
br | +0.012 |
str | +0.012 |
Neg logit contributions | |
amaz | -0.007 |
gasp | -0.006 |
angles | -0.005 |
horror | -0.005 |
ill | -0.005 |
Token | get |
Token ID | 651 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
horror | +0.002 |
angles | +0.002 |
decide | +0.002 |
Neg logit contributions | |
Nut | -0.004 |
rob | -0.004 |
br | -0.004 |
Rust | -0.004 |
str | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
Nut | +0.021 |
con | +0.021 |
rob | +0.021 |
Sy | +0.019 |
cone | +0.019 |
Neg logit contributions | |
amaz | -0.019 |
ill | -0.017 |
gasp | -0.016 |
decide | -0.016 |
horror | -0.016 |
Token | ch |
Token ID | 442 |
Feature activation | -5.48e-03 |
Pos logit contributions | |
Nut | +0.019 |
Rust | +0.016 |
rob | +0.016 |
br | +0.016 |
str | +0.016 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.013 |
angles | -0.013 |
horror | -0.013 |
bowls | -0.012 |
Token | ubby |
Token ID | 38393 |
Feature activation | +6.82e-03 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.005 |
angles | +0.005 |
horror | +0.005 |
ill | +0.005 |
Neg logit contributions | |
Nut | -0.014 |
Rust | -0.013 |
rob | -0.013 |
str | -0.012 |
Nut | -0.012 |
Token | little |
Token ID | 1310 |
Feature activation | +9.98e-03 |
Pos logit contributions | |
Nut | +0.010 |
Rust | +0.008 |
br | +0.008 |
str | +0.008 |
Nut | +0.008 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.014 |
angles | -0.014 |
horror | -0.013 |
ill | -0.013 |
Token | bear |
Token ID | 6842 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
Nut | +0.017 |
Rust | +0.014 |
Nut | +0.014 |
br | +0.014 |
rob | +0.014 |
Neg logit contributions | |
amaz | -0.020 |
gasp | -0.019 |
angles | -0.018 |
horror | -0.018 |
ill | -0.017 |
Token | learned |
Token ID | 4499 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
Nut | +0.047 |
Rust | +0.040 |
rob | +0.040 |
br | +0.040 |
str | +0.039 |
Neg logit contributions | |
amaz | -0.029 |
gasp | -0.027 |
angles | -0.025 |
horror | -0.025 |
ill | -0.024 |
Token | an |
Token ID | 281 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
Nut | +0.109 |
rob | +0.106 |
str | +0.102 |
br | +0.101 |
cone | +0.098 |
Neg logit contributions | |
amaz | -0.070 |
eyes | -0.063 |
ways | -0.060 |
wonders | -0.059 |
horror | -0.058 |
Token | important |
Token ID | 1593 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
Nut | +0.021 |
rob | +0.018 |
Rust | +0.018 |
br | +0.017 |
str | +0.017 |
Neg logit contributions | |
amaz | -0.020 |
angles | -0.019 |
gasp | -0.018 |
horror | -0.018 |
wish | -0.017 |
Token | lesson |
Token ID | 11483 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
rob | +0.159 |
Nut | +0.157 |
Mic | +0.141 |
Rust | +0.139 |
Wh | +0.136 |
Neg logit contributions | |
wish | -0.080 |
wonder | -0.069 |
request | -0.065 |
promise | -0.063 |
amaz | -0.059 |
Token | - |
Token ID | 532 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
Nut | +0.060 |
Rust | +0.052 |
rob | +0.050 |
str | +0.049 |
Silver | +0.049 |
Neg logit contributions | |
amaz | -0.046 |
angles | -0.043 |
gasp | -0.041 |
horror | -0.041 |
ill | -0.040 |
Token | never |
Token ID | 1239 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
Nut | +0.078 |
rob | +0.068 |
Rust | +0.066 |
str | +0.064 |
br | +0.064 |
Neg logit contributions | |
amaz | -0.078 |
gasp | -0.070 |
angles | -0.069 |
horror | -0.068 |
wonder | -0.063 |
Token | store |
Token ID | 3650 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
Nut | +0.003 |
Rust | +0.003 |
br | +0.002 |
str | +0.002 |
rob | +0.002 |
Neg logit contributions | |
amaz | -0.003 |
gasp | -0.002 |
angles | -0.002 |
horror | -0.002 |
ill | -0.002 |
Token | keeper |
Token ID | 13884 |
Feature activation | -6.42e-03 |
Pos logit contributions | |
Nut | +0.021 |
Rust | +0.018 |
br | +0.017 |
rob | +0.017 |
str | +0.017 |
Neg logit contributions | |
amaz | -0.019 |
gasp | -0.017 |
angles | -0.016 |
horror | -0.016 |
ill | -0.016 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.009 |
angles | +0.009 |
horror | +0.008 |
ill | +0.008 |
Neg logit contributions | |
Nut | -0.014 |
str | -0.012 |
Rust | -0.012 |
Nut | -0.012 |
br | -0.012 |
Token | â |
Token ID | 22940 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
Nut | +0.025 |
rob | +0.020 |
str | +0.020 |
Rust | +0.019 |
con | +0.019 |
Neg logit contributions | |
gasp | -0.028 |
amaz | -0.028 |
horror | -0.026 |
ill | -0.025 |
decide | -0.025 |
Token | € |
Token ID | 26391 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
Nut | +0.020 |
Rust | +0.016 |
br | +0.016 |
Nut | +0.016 |
str | +0.015 |
Neg logit contributions | |
amaz | -0.022 |
gasp | -0.021 |
ill | -0.020 |
angles | -0.020 |
horror | -0.020 |
Token | â„¢ |
Token ID | 8151 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
Nut | +0.075 |
Rust | +0.062 |
rob | +0.061 |
br | +0.061 |
Silver | +0.059 |
Neg logit contributions | |
amaz | -0.090 |
gasp | -0.085 |
ill | -0.080 |
horror | -0.076 |
gasped | -0.074 |
Token | t |
Token ID | 83 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
Nut | +0.078 |
Rust | +0.063 |
rob | +0.062 |
br | +0.060 |
con | +0.059 |
Neg logit contributions | |
ill | -0.117 |
gasp | -0.114 |
amaz | -0.113 |
angles | -0.107 |
horror | -0.106 |
Token | want |
Token ID | 765 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
Nut | +0.039 |
Rust | +0.036 |
rob | +0.036 |
con | +0.034 |
str | +0.034 |
Neg logit contributions | |
gasp | -0.014 |
amaz | -0.012 |
ill | -0.011 |
angles | -0.011 |
horror | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -8.74e-03 |
Pos logit contributions | |
Nut | +0.034 |
Rust | +0.028 |
rob | +0.028 |
str | +0.027 |
K | +0.027 |
Neg logit contributions | |
amaz | -0.036 |
gasp | -0.033 |
angles | -0.033 |
horror | -0.032 |
ill | -0.031 |
Token | do |
Token ID | 466 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
amaz | +0.013 |
gasp | +0.011 |
angles | +0.011 |
horror | +0.010 |
ill | +0.010 |
Neg logit contributions | |
Nut | -0.020 |
Rust | -0.017 |
Nut | -0.017 |
str | -0.017 |
Sy | -0.017 |
Token | it |
Token ID | 340 |
Feature activation | +2.60e-03 |
Pos logit contributions | |
Nut | +0.064 |
Rust | +0.054 |
rob | +0.053 |
str | +0.052 |
K | +0.052 |
Neg logit contributions | |
amaz | -0.057 |
gasp | -0.051 |
angles | -0.049 |
horror | -0.049 |
wish | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -5.73e-03 |
Pos logit contributions | |
amaz | +0.007 |
gasp | +0.006 |
angles | +0.006 |
horror | +0.006 |
ill | +0.005 |
Neg logit contributions | |
Nut | -0.009 |
Rust | -0.008 |
rob | -0.008 |
str | -0.008 |
br | -0.008 |
Token | do |
Token ID | 466 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
amaz | +0.008 |
gasp | +0.007 |
angles | +0.007 |
horror | +0.007 |
bowls | +0.006 |
Neg logit contributions | |
Nut | -0.014 |
Rust | -0.012 |
str | -0.012 |
Nut | -0.012 |
br | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +7.48e-04 |
Pos logit contributions | |
Nut | +0.027 |
Rust | +0.023 |
rob | +0.023 |
str | +0.022 |
K | +0.022 |
Neg logit contributions | |
amaz | -0.018 |
gasp | -0.017 |
angles | -0.016 |
horror | -0.016 |
ill | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | +8.46e-03 |
Pos logit contributions | |
Nut | +0.002 |
rob | +0.002 |
br | +0.002 |
Rust | +0.002 |
str | +0.001 |
Neg logit contributions | |
amaz | -0.001 |
horror | -0.001 |
gasp | -0.001 |
angles | -0.001 |
ill | -0.001 |
Token | also |
Token ID | 635 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
Nut | +0.018 |
Rust | +0.016 |
rob | +0.015 |
br | +0.015 |
str | +0.015 |
Neg logit contributions | |
amaz | -0.013 |
gasp | -0.012 |
angles | -0.012 |
horror | -0.012 |
ill | -0.011 |
Token | learned |
Token ID | 4499 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
Nut | +0.036 |
Rust | +0.031 |
str | +0.030 |
Nut | +0.030 |
age | +0.030 |
Neg logit contributions | |
amaz | -0.029 |
gasp | -0.027 |
angles | -0.026 |
ill | -0.025 |
horror | -0.025 |
Token | that |
Token ID | 326 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
rob | +0.111 |
Nut | +0.107 |
br | +0.100 |
str | +0.097 |
Rust | +0.097 |
Neg logit contributions | |
amaz | -0.073 |
eyes | -0.069 |
ways | -0.068 |
Dad | -0.067 |
wonders | -0.067 |
Token | sometimes |
Token ID | 3360 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Nut | +0.053 |
Rust | +0.044 |
rob | +0.043 |
br | +0.043 |
str | +0.043 |
Neg logit contributions | |
amaz | -0.050 |
gasp | -0.046 |
angles | -0.045 |
horror | -0.044 |
ill | -0.042 |
Token | people |
Token ID | 661 |
Feature activation | +1.34e-03 |
Pos logit contributions | |
Nut | +0.027 |
Rust | +0.023 |
br | +0.022 |
rob | +0.022 |
str | +0.022 |
Neg logit contributions | |
amaz | -0.026 |
gasp | -0.024 |
angles | -0.023 |
horror | -0.023 |
ill | -0.022 |
Token | can |
Token ID | 460 |
Feature activation | -6.47e-03 |
Pos logit contributions | |
Nut | +0.003 |
Rust | +0.003 |
rob | +0.003 |
br | +0.003 |
str | +0.003 |
Neg logit contributions | |
amaz | -0.002 |
gasp | -0.002 |
angles | -0.001 |
horror | -0.001 |
ill | -0.001 |
Token | be |
Token ID | 307 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
angles | +0.008 |
horror | +0.008 |
ill | +0.007 |
Neg logit contributions | |
Nut | -0.015 |
Rust | -0.013 |
rob | -0.013 |
br | -0.013 |
str | -0.013 |
Token | They |
Token ID | 1119 |
Feature activation | -5.26e-03 |
Pos logit contributions | |
amaz | +0.011 |
gasp | +0.010 |
angles | +0.009 |
horror | +0.009 |
ill | +0.009 |
Neg logit contributions | |
Nut | -0.016 |
Rust | -0.014 |
rob | -0.014 |
br | -0.014 |
str | -0.014 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.77e-03 |
Pos logit contributions | |
amaz | +0.008 |
angles | +0.007 |
gasp | +0.007 |
horror | +0.007 |
urious | +0.006 |
Neg logit contributions | |
Nut | -0.012 |
Rust | -0.010 |
str | -0.010 |
Nut | -0.010 |
K | -0.010 |
Token | people |
Token ID | 661 |
Feature activation | -4.32e-03 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.018 |
angles | +0.017 |
horror | +0.017 |
ill | +0.017 |
Neg logit contributions | |
Nut | -0.017 |
Rust | -0.014 |
Nut | -0.014 |
str | -0.014 |
br | -0.014 |
Token | wearing |
Token ID | 5762 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
amaz | +0.008 |
angles | +0.007 |
gasp | +0.007 |
horror | +0.007 |
ill | +0.007 |
Neg logit contributions | |
Nut | -0.007 |
str | -0.006 |
Rust | -0.006 |
br | -0.006 |
Nut | -0.006 |
Token | hats |
Token ID | 23910 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
Nut | +0.073 |
str | +0.062 |
Rust | +0.061 |
rob | +0.060 |
age | +0.060 |
Neg logit contributions | |
amaz | -0.055 |
gasp | -0.050 |
bowls | -0.046 |
angles | -0.046 |
horror | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
Nut | +0.039 |
Rust | +0.033 |
str | +0.032 |
rob | +0.032 |
br | +0.032 |
Neg logit contributions | |
amaz | -0.032 |
gasp | -0.030 |
angles | -0.028 |
ill | -0.028 |
horror | -0.027 |
Token | scar |
Token ID | 10153 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
Nut | +0.100 |
Rust | +0.084 |
str | +0.083 |
br | +0.082 |
rob | +0.082 |
Neg logit contributions | |
amaz | -0.087 |
gasp | -0.081 |
angles | -0.076 |
horror | -0.075 |
ill | -0.072 |
Token | ves |
Token ID | 1158 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
Nut | +0.044 |
Sy | +0.035 |
Nut | +0.034 |
Rust | +0.033 |
Silver | +0.032 |
Neg logit contributions | |
gasp | -0.066 |
amaz | -0.064 |
angles | -0.063 |
faces | -0.062 |
bowls | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
Nut | +0.051 |
Rust | +0.044 |
rob | +0.043 |
str | +0.043 |
Sy | +0.042 |
Neg logit contributions | |
amaz | -0.036 |
gasp | -0.034 |
angles | -0.031 |
ill | -0.030 |
horror | -0.030 |
Token | gloves |
Token ID | 18051 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
Nut | +0.095 |
Rust | +0.080 |
br | +0.079 |
rob | +0.078 |
Nut | +0.078 |
Neg logit contributions | |
amaz | -0.084 |
gasp | -0.078 |
angles | -0.075 |
horror | -0.073 |
ill | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -6.76e-03 |
Pos logit contributions | |
Nut | +0.084 |
Rust | +0.074 |
str | +0.072 |
Silver | +0.072 |
br | +0.071 |
Neg logit contributions | |
amaz | -0.043 |
gasp | -0.042 |
ill | -0.038 |
angles | -0.038 |
horror | -0.036 |
Token | stream |
Token ID | 4269 |
Feature activation | +4.46e-03 |
Pos logit contributions | |
amaz | +0.031 |
gasp | +0.029 |
horror | +0.028 |
angles | +0.027 |
ill | +0.027 |
Neg logit contributions | |
Nut | -0.027 |
str | -0.022 |
br | -0.022 |
Rust | -0.022 |
Nut | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
Nut | +0.011 |
Rust | +0.009 |
br | +0.009 |
rob | +0.009 |
str | +0.009 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.005 |
angles | -0.005 |
ill | -0.005 |
horror | -0.005 |
Token | but |
Token ID | 475 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
Nut | +0.029 |
rob | +0.024 |
Rust | +0.024 |
br | +0.024 |
str | +0.024 |
Neg logit contributions | |
amaz | -0.028 |
gasp | -0.027 |
angles | -0.025 |
horror | -0.025 |
ill | -0.024 |
Token | it |
Token ID | 340 |
Feature activation | +1.63e-03 |
Pos logit contributions | |
Nut | +0.021 |
Rust | +0.018 |
rob | +0.018 |
br | +0.018 |
con | +0.017 |
Neg logit contributions | |
amaz | -0.020 |
gasp | -0.018 |
angles | -0.017 |
horror | -0.017 |
ill | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | +1.62e-03 |
Pos logit contributions | |
Nut | +0.004 |
Rust | +0.003 |
br | +0.003 |
str | +0.003 |
rob | +0.003 |
Neg logit contributions | |
amaz | -0.002 |
gasp | -0.002 |
angles | -0.002 |
horror | -0.002 |
ill | -0.002 |
Token | too |
Token ID | 1165 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
Nut | +0.004 |
rob | +0.003 |
con | +0.003 |
Rust | +0.003 |
br | +0.003 |
Neg logit contributions | |
amaz | -0.002 |
gasp | -0.002 |
horror | -0.002 |
angles | -0.002 |
bowls | -0.002 |
Token | high |
Token ID | 1029 |
Feature activation | +8.83e-03 |
Pos logit contributions | |
Nut | +0.026 |
rob | +0.023 |
Rust | +0.022 |
con | +0.022 |
age | +0.022 |
Neg logit contributions | |
amaz | -0.019 |
gasp | -0.019 |
bowls | -0.017 |
ill | -0.017 |
angles | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
Nut | +0.021 |
str | +0.018 |
Rust | +0.018 |
Silver | +0.018 |
K | +0.018 |
Neg logit contributions | |
amaz | -0.011 |
gasp | -0.011 |
angles | -0.010 |
ill | -0.010 |
horror | -0.010 |
Token | cross |
Token ID | 3272 |
Feature activation | +2.67e-03 |
Pos logit contributions | |
amaz | +0.026 |
gasp | +0.024 |
angles | +0.023 |
horror | +0.022 |
ill | +0.021 |
Neg logit contributions | |
Nut | -0.045 |
Rust | -0.039 |
br | -0.038 |
str | -0.038 |
rob | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +4.93e-03 |
Pos logit contributions | |
Nut | +0.007 |
Rust | +0.006 |
br | +0.006 |
str | +0.006 |
rob | +0.006 |
Neg logit contributions | |
amaz | -0.003 |
gasp | -0.003 |
angles | -0.003 |
ill | -0.003 |
horror | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
Nut | +0.010 |
br | +0.009 |
rob | +0.009 |
Rust | +0.009 |
str | +0.009 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.007 |
horror | -0.007 |
ill | -0.007 |
angles | -0.007 |
Token | duck |
Token ID | 22045 |
Feature activation | +2.86e-03 |
Pos logit contributions | |
Nut | +0.011 |
Rust | +0.009 |
rob | +0.009 |
str | +0.009 |
K | +0.009 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.008 |
horror | -0.007 |
angles | -0.007 |
bowls | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +2.51e-03 |
Pos logit contributions | |
Nut | +0.006 |
Rust | +0.005 |
br | +0.005 |
Nut | +0.005 |
str | +0.005 |
Neg logit contributions | |
amaz | -0.005 |
gasp | -0.004 |
angles | -0.004 |
horror | -0.004 |
ill | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | -4.70e-04 |
Pos logit contributions | |
Nut | +0.006 |
Rust | +0.005 |
rob | +0.005 |
str | +0.005 |
br | +0.005 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.003 |
horror | -0.003 |
angles | -0.003 |
bowls | -0.003 |
Token | turtle |
Token ID | 28699 |
Feature activation | +4.13e-03 |
Pos logit contributions | |
amaz | +0.001 |
gasp | +0.001 |
angles | +0.001 |
horror | +0.001 |
ill | +0.001 |
Neg logit contributions | |
Nut | -0.001 |
Rust | -0.001 |
br | -0.001 |
str | -0.001 |
rob | -0.001 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
Nut | +0.010 |
Rust | +0.009 |
rob | +0.009 |
br | +0.009 |
str | +0.009 |
Neg logit contributions | |
amaz | -0.005 |
gasp | -0.005 |
angles | -0.004 |
horror | -0.004 |
ill | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
amaz | +0.025 |
gasp | +0.023 |
horror | +0.022 |
angles | +0.022 |
ill | +0.021 |
Neg logit contributions | |
Nut | -0.031 |
str | -0.027 |
rob | -0.027 |
Rust | -0.026 |
con | -0.026 |
Token | said |
Token ID | 531 |
Feature activation | +2.96e-03 |
Pos logit contributions | |
Nut | +0.048 |
age | +0.042 |
str | +0.042 |
Rust | +0.042 |
br | +0.041 |
Neg logit contributions | |
amaz | -0.020 |
gasp | -0.018 |
angles | -0.016 |
horror | -0.015 |
gasped | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
Nut | +0.007 |
rob | +0.006 |
Rust | +0.006 |
age | +0.006 |
K | +0.005 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.004 |
horror | -0.004 |
angles | -0.004 |
eyes | -0.004 |
Token | " |
Token ID | 366 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
Nut | +0.069 |
rob | +0.067 |
age | +0.061 |
Rust | +0.061 |
str | +0.059 |
Neg logit contributions | |
amaz | -0.029 |
gasp | -0.028 |
Dad | -0.027 |
eyes | -0.027 |
Mom | -0.026 |
Token | That |
Token ID | 2504 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
Nut | +0.047 |
str | +0.042 |
Nut | +0.042 |
con | +0.042 |
Sy | +0.041 |
Neg logit contributions | |
amaz | -0.029 |
angles | -0.027 |
ill | -0.027 |
gasp | -0.027 |
wonders | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
Nut | +0.052 |
Rust | +0.045 |
br | +0.045 |
str | +0.044 |
rob | +0.044 |
Neg logit contributions | |
amaz | -0.032 |
gasp | -0.029 |
angles | -0.028 |
horror | -0.027 |
ill | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +5.32e-04 |
Pos logit contributions | |
Nut | +0.007 |
rob | +0.006 |
br | +0.006 |
con | +0.006 |
Rust | +0.006 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.006 |
ill | -0.006 |
angles | -0.005 |
horror | -0.005 |
Token | The |
Token ID | 464 |
Feature activation | +3.15e-03 |
Pos logit contributions | |
Nut | +0.001 |
rob | +0.001 |
con | +0.001 |
br | +0.001 |
str | +0.001 |
Neg logit contributions | |
amaz | -0.001 |
gasp | -0.001 |
ill | -0.001 |
angles | -0.001 |
horror | -0.001 |
Token | monkey |
Token ID | 21657 |
Feature activation | +7.49e-03 |
Pos logit contributions | |
Nut | +0.007 |
Rust | +0.006 |
rob | +0.006 |
br | +0.006 |
Nut | +0.006 |
Neg logit contributions | |
amaz | -0.005 |
gasp | -0.004 |
angles | -0.004 |
horror | -0.004 |
bowls | -0.004 |
Token | tried |
Token ID | 3088 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
Nut | +0.018 |
Rust | +0.015 |
rob | +0.015 |
br | +0.015 |
str | +0.015 |
Neg logit contributions | |
amaz | -0.010 |
gasp | -0.009 |
angles | -0.009 |
horror | -0.008 |
ill | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Nut | +0.066 |
rob | +0.064 |
Rust | +0.058 |
str | +0.058 |
br | +0.058 |
Neg logit contributions | |
amaz | -0.052 |
gasp | -0.049 |
angles | -0.046 |
horror | -0.045 |
ill | -0.044 |
Token | swing |
Token ID | 9628 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
amaz | +0.017 |
gasp | +0.015 |
angles | +0.015 |
horror | +0.014 |
ill | +0.014 |
Neg logit contributions | |
Nut | -0.025 |
Rust | -0.022 |
Nut | -0.021 |
br | -0.021 |
str | -0.021 |
Token | on |
Token ID | 319 |
Feature activation | +7.04e-03 |
Pos logit contributions | |
Nut | +0.080 |
Rust | +0.071 |
str | +0.069 |
rob | +0.067 |
Silver | +0.067 |
Neg logit contributions | |
amaz | -0.050 |
gasp | -0.049 |
ill | -0.048 |
angles | -0.045 |
horror | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
Nut | +0.017 |
str | +0.015 |
rob | +0.015 |
age | +0.015 |
Rust | +0.015 |
Neg logit contributions | |
amaz | -0.009 |
gasp | -0.008 |
angles | -0.008 |
horror | -0.008 |
ill | -0.008 |
Token | rope |
Token ID | 17182 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
Nut | +0.023 |
Rust | +0.020 |
rob | +0.019 |
br | +0.019 |
Nut | +0.019 |
Neg logit contributions | |
amaz | -0.016 |
gasp | -0.016 |
angles | -0.015 |
horror | -0.014 |
bowls | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
Nut | +0.050 |
Rust | +0.043 |
rob | +0.042 |
str | +0.041 |
Silver | +0.041 |
Neg logit contributions | |
amaz | -0.036 |
gasp | -0.035 |
ill | -0.034 |
angles | -0.033 |
horror | -0.031 |
Token | but |
Token ID | 475 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
Nut | +0.060 |
rob | +0.055 |
br | +0.052 |
Rust | +0.052 |
age | +0.052 |
Neg logit contributions | |
amaz | -0.039 |
gasp | -0.037 |
horror | -0.034 |
angles | -0.034 |
ill | -0.032 |
Token | he |
Token ID | 339 |
Feature activation | +3.94e-03 |
Pos logit contributions | |
Nut | +0.009 |
Rust | +0.007 |
rob | +0.007 |
str | +0.007 |
age | +0.007 |
Neg logit contributions | |
amaz | -0.007 |
gasp | -0.007 |
horror | -0.006 |
angles | -0.006 |
bowls | -0.006 |
Token | could |
Token ID | 714 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
Nut | +0.008 |
Rust | +0.007 |
age | +0.007 |
Nut | +0.007 |
str | +0.006 |
Neg logit contributions | |
amaz | -0.007 |
gasp | -0.006 |
angles | -0.006 |
horror | -0.006 |
ill | -0.006 |
Token | hug |
Token ID | 16225 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
amaz | +0.015 |
gasp | +0.013 |
angles | +0.013 |
horror | +0.013 |
ill | +0.012 |
Neg logit contributions | |
Nut | -0.026 |
Rust | -0.023 |
br | -0.022 |
Nut | -0.022 |
age | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | -5.48e-05 |
Pos logit contributions | |
amaz | +0.016 |
gasp | +0.015 |
angles | +0.014 |
horror | +0.014 |
ill | +0.014 |
Neg logit contributions | |
Nut | -0.029 |
str | -0.026 |
Rust | -0.025 |
br | -0.025 |
age | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
amaz | +0.000 |
gasp | +0.000 |
angles | +0.000 |
horror | +0.000 |
ill | +0.000 |
Neg logit contributions | |
Nut | -0.000 |
Rust | -0.000 |
br | -0.000 |
str | -0.000 |
rob | -0.000 |
Token | play |
Token ID | 711 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
Nut | +0.029 |
br | +0.025 |
Rust | +0.025 |
Nut | +0.025 |
con | +0.025 |
Neg logit contributions | |
amaz | -0.019 |
angles | -0.018 |
gasp | -0.017 |
ill | -0.017 |
horror | -0.016 |
Token | with |
Token ID | 351 |
Feature activation | -9.54e-03 |
Pos logit contributions | |
Nut | +0.025 |
Rust | +0.021 |
str | +0.021 |
br | +0.021 |
rob | +0.021 |
Neg logit contributions | |
amaz | -0.020 |
gasp | -0.018 |
angles | -0.017 |
horror | -0.017 |
ill | -0.016 |
Token | it |
Token ID | 340 |
Feature activation | -6.56e-03 |
Pos logit contributions | |
amaz | +0.018 |
gasp | +0.017 |
angles | +0.016 |
horror | +0.016 |
ill | +0.015 |
Neg logit contributions | |
Nut | -0.018 |
Rust | -0.015 |
Nut | -0.015 |
br | -0.014 |
con | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -4.52e-03 |
Pos logit contributions | |
amaz | +0.011 |
gasp | +0.010 |
angles | +0.010 |
horror | +0.009 |
ill | +0.009 |
Neg logit contributions | |
Nut | -0.013 |
br | -0.011 |
Rust | -0.011 |
Nut | -0.011 |
con | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | +1.17e-03 |
Pos logit contributions | |
amaz | +0.008 |
gasp | +0.007 |
angles | +0.007 |
horror | +0.007 |
ill | +0.006 |
Neg logit contributions | |
Nut | -0.009 |
Rust | -0.008 |
str | -0.008 |
br | -0.008 |
rob | -0.008 |
Token | said |
Token ID | 531 |
Feature activation | -3.10e-04 |
Pos logit contributions | |
Nut | +0.002 |
Rust | +0.002 |
br | +0.002 |
rob | +0.002 |
con | +0.002 |
Neg logit contributions | |
amaz | -0.002 |
angles | -0.002 |
gasp | -0.002 |
horror | -0.002 |
ill | -0.002 |
Token | She |
Token ID | 1375 |
Feature activation | -5.23e-04 |
Pos logit contributions | |
Nut | +0.006 |
Rust | +0.005 |
br | +0.005 |
str | +0.005 |
rob | +0.005 |
Neg logit contributions | |
amaz | -0.004 |
gasp | -0.003 |
angles | -0.003 |
horror | -0.003 |
ill | -0.003 |
Token | told |
Token ID | 1297 |
Feature activation | -7.22e-03 |
Pos logit contributions | |
amaz | +0.001 |
angles | +0.001 |
gasp | +0.001 |
horror | +0.001 |
ill | +0.001 |
Neg logit contributions | |
Nut | -0.001 |
Rust | -0.001 |
Nut | -0.001 |
str | -0.001 |
br | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | -5.24e-03 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
horror | +0.008 |
angles | +0.008 |
faces | +0.007 |
Neg logit contributions | |
Nut | -0.018 |
str | -0.016 |
rob | -0.016 |
Rust | -0.016 |
br | -0.016 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
amaz | +0.008 |
angles | +0.008 |
gasp | +0.008 |
ill | +0.007 |
horror | +0.007 |
Neg logit contributions | |
Nut | -0.011 |
Nut | -0.010 |
Rust | -0.009 |
br | -0.009 |
str | -0.009 |
Token | my |
Token ID | 1820 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
Nut | +0.030 |
Rust | +0.026 |
rob | +0.025 |
br | +0.025 |
str | +0.025 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.016 |
angles | -0.015 |
horror | -0.014 |
ill | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
Nut | +0.052 |
rob | +0.044 |
Rust | +0.043 |
br | +0.042 |
str | +0.042 |
Neg logit contributions | |
amaz | -0.056 |
gasp | -0.052 |
angles | -0.049 |
horror | -0.049 |
ill | -0.048 |
Token | " |
Token ID | 366 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
Nut | +0.066 |
rob | +0.063 |
br | +0.060 |
con | +0.059 |
age | +0.058 |
Neg logit contributions | |
amaz | -0.039 |
Dad | -0.035 |
horror | -0.034 |
gasp | -0.033 |
Mom | -0.033 |
Token | I |
Token ID | 40 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
Nut | +0.024 |
Rust | +0.021 |
str | +0.021 |
br | +0.020 |
rob | +0.020 |
Neg logit contributions | |
amaz | -0.019 |
gasp | -0.017 |
angles | -0.017 |
ill | -0.016 |
horror | -0.016 |
Token | learned |
Token ID | 4499 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
Nut | +0.025 |
str | +0.022 |
Nut | +0.021 |
age | +0.021 |
Rust | +0.021 |
Neg logit contributions | |
amaz | -0.019 |
angles | -0.019 |
gasp | -0.017 |
bowls | -0.017 |
horror | -0.017 |
Token | that |
Token ID | 326 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
Nut | +0.067 |
str | +0.059 |
rob | +0.059 |
age | +0.058 |
Rust | +0.056 |
Neg logit contributions | |
amaz | -0.058 |
eyes | -0.052 |
ways | -0.049 |
Mom | -0.049 |
wonders | -0.049 |
Token | exercise |
Token ID | 5517 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Nut | +0.044 |
Rust | +0.037 |
rob | +0.036 |
str | +0.036 |
br | +0.036 |
Neg logit contributions | |
amaz | -0.038 |
gasp | -0.035 |
angles | -0.033 |
horror | -0.033 |
ill | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -7.79e-03 |
Pos logit contributions | |
Nut | +0.032 |
Rust | +0.026 |
rob | +0.026 |
str | +0.026 |
br | +0.026 |
Neg logit contributions | |
amaz | -0.039 |
gasp | -0.036 |
angles | -0.035 |
horror | -0.034 |
bowls | -0.033 |
Token | very |
Token ID | 845 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
amaz | +0.012 |
gasp | +0.011 |
wish | +0.010 |
wonder | +0.010 |
horror | +0.010 |
Neg logit contributions | |
Nut | -0.017 |
rob | -0.016 |
Rust | -0.015 |
con | -0.015 |
br | -0.015 |
Token | special |
Token ID | 2041 |
Feature activation | +1.25e-03 |
Pos logit contributions | |
amaz | +0.035 |
gasp | +0.031 |
ill | +0.031 |
angles | +0.031 |
horror | +0.030 |
Neg logit contributions | |
Nut | -0.032 |
str | -0.028 |
br | -0.027 |
Rust | -0.026 |
age | -0.026 |
Token | gem |
Token ID | 16840 |
Feature activation | +2.68e-03 |
Pos logit contributions | |
Nut | +0.003 |
Rust | +0.002 |
rob | +0.002 |
K | +0.002 |
str | +0.002 |
Neg logit contributions | |
amaz | -0.002 |
gasp | -0.002 |
wish | -0.002 |
horror | -0.002 |
angles | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -6.81e-03 |
Pos logit contributions | |
Nut | +0.006 |
Rust | +0.006 |
str | +0.006 |
rob | +0.005 |
br | +0.005 |
Neg logit contributions | |
amaz | -0.003 |
gasp | -0.003 |
angles | -0.003 |
horror | -0.003 |
ill | -0.003 |
Token | Every |
Token ID | 3887 |
Feature activation | +9.20e-03 |
Pos logit contributions | |
amaz | +0.011 |
gasp | +0.010 |
angles | +0.010 |
horror | +0.009 |
ill | +0.009 |
Neg logit contributions | |
Nut | -0.014 |
Rust | -0.012 |
str | -0.012 |
br | -0.012 |
rob | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | -3.35e-03 |
Pos logit contributions | |
Nut | +0.021 |
rob | +0.018 |
Rust | +0.018 |
Sy | +0.018 |
con | +0.018 |
Neg logit contributions | |
amaz | -0.014 |
gasp | -0.013 |
wish | -0.012 |
wonder | -0.012 |
horror | -0.012 |
Token | he |
Token ID | 339 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.005 |
horror | +0.005 |
angles | +0.005 |
bowls | +0.005 |
Neg logit contributions | |
Nut | -0.007 |
Rust | -0.006 |
rob | -0.006 |
age | -0.006 |
str | -0.006 |
Token | brought |
Token ID | 3181 |
Feature activation | -2.89e-03 |
Pos logit contributions | |
amaz | +0.031 |
gasp | +0.029 |
angles | +0.028 |
horror | +0.027 |
bowls | +0.027 |
Neg logit contributions | |
Nut | -0.035 |
Rust | -0.029 |
str | -0.028 |
rob | -0.028 |
br | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
amaz | +0.006 |
gasp | +0.005 |
angles | +0.005 |
horror | +0.005 |
ill | +0.005 |
Neg logit contributions | |
Nut | -0.005 |
Rust | -0.004 |
str | -0.004 |
rob | -0.004 |
br | -0.004 |
Token | out |
Token ID | 503 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
Nut | +0.039 |
rob | +0.035 |
Rust | +0.034 |
str | +0.034 |
K | +0.033 |
Neg logit contributions | |
amaz | -0.032 |
gasp | -0.030 |
angles | -0.028 |
bowls | -0.028 |
ill | -0.027 |
Token | " |
Token ID | 1 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
Nut | +0.005 |
Rust | +0.004 |
br | +0.003 |
str | +0.003 |
rob | +0.003 |
Neg logit contributions | |
amaz | -0.007 |
gasp | -0.007 |
angles | -0.007 |
horror | -0.007 |
ill | -0.007 |
Token | What |
Token ID | 2061 |
Feature activation | -8.54e-04 |
Pos logit contributions | |
Nut | +0.021 |
str | +0.018 |
age | +0.017 |
con | +0.017 |
br | +0.017 |
Neg logit contributions | |
amaz | -0.018 |
gasp | -0.017 |
angles | -0.016 |
horror | -0.015 |
ill | -0.015 |
Token | 's |
Token ID | 338 |
Feature activation | +3.59e-03 |
Pos logit contributions | |
amaz | +0.001 |
angles | +0.001 |
horror | +0.001 |
ishment | +0.001 |
gasp | +0.001 |
Neg logit contributions | |
Nut | -0.002 |
br | -0.002 |
str | -0.002 |
Nut | -0.001 |
rob | -0.001 |
Token | wrong |
Token ID | 2642 |
Feature activation | +5.20e-03 |
Pos logit contributions | |
Nut | +0.008 |
age | +0.006 |
con | +0.006 |
Rust | +0.006 |
K | +0.006 |
Neg logit contributions | |
amaz | -0.005 |
wish | -0.005 |
Dad | -0.005 |
Mom | -0.005 |
says | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -8.56e-03 |
Pos logit contributions | |
Nut | +0.009 |
Rust | +0.008 |
br | +0.008 |
Nut | +0.008 |
str | +0.007 |
Neg logit contributions | |
amaz | -0.010 |
gasp | -0.009 |
angles | -0.009 |
horror | -0.009 |
ill | -0.009 |
Token | Tim |
Token ID | 5045 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.012 |
bowls | +0.012 |
horror | +0.012 |
Neg logit contributions | |
Nut | -0.018 |
Nut | -0.016 |
Sy | -0.015 |
K | -0.014 |
Bun | -0.014 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.48e-03 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.012 |
horror | +0.012 |
ill | +0.012 |
Neg logit contributions | |
Nut | -0.018 |
Rust | -0.015 |
br | -0.015 |
str | -0.015 |
rob | -0.015 |
Token | she |
Token ID | 673 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
Nut | +0.002 |
Nut | +0.002 |
Sy | +0.002 |
K | +0.002 |
Rust | +0.002 |
Neg logit contributions | |
amaz | -0.003 |
gasp | -0.003 |
angles | -0.003 |
ill | -0.003 |
bowls | -0.003 |
Token | asks |
Token ID | 7893 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
amaz | +0.030 |
ill | +0.026 |
gasp | +0.024 |
angles | +0.022 |
bowls | +0.022 |
Neg logit contributions | |
Nut | -0.067 |
Rust | -0.063 |
Silver | -0.061 |
Nut | -0.061 |
br | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
amaz | +0.048 |
gasp | +0.041 |
angles | +0.040 |
horror | +0.038 |
wish | +0.037 |
Neg logit contributions | |
Nut | -0.048 |
Nut | -0.041 |
Sy | -0.040 |
Rust | -0.039 |
con | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +8.21e-03 |
Pos logit contributions | |
Nut | +0.034 |
rob | +0.031 |
str | +0.029 |
Rust | +0.029 |
br | +0.029 |
Neg logit contributions | |
amaz | -0.025 |
gasp | -0.024 |
horror | -0.023 |
ill | -0.022 |
gasped | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
Nut | +0.053 |
Rust | +0.045 |
rob | +0.044 |
str | +0.044 |
Silver | +0.043 |
Neg logit contributions | |
amaz | -0.041 |
gasp | -0.039 |
ill | -0.037 |
angles | -0.036 |
horror | -0.035 |
Token | got |
Token ID | 1392 |
Feature activation | -2.57e-03 |
Pos logit contributions | |
Nut | +0.026 |
Nut | +0.022 |
Rust | +0.022 |
con | +0.022 |
br | +0.022 |
Neg logit contributions | |
amaz | -0.019 |
gasp | -0.018 |
angles | -0.018 |
horror | -0.017 |
ill | -0.017 |
Token | ready |
Token ID | 3492 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
angles | +0.004 |
horror | +0.004 |
bowls | +0.004 |
Neg logit contributions | |
Nut | -0.005 |
Rust | -0.004 |
rob | -0.004 |
br | -0.004 |
str | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
amaz | +0.010 |
gasp | +0.009 |
angles | +0.009 |
horror | +0.009 |
ill | +0.008 |
Neg logit contributions | |
Nut | -0.009 |
Rust | -0.007 |
rob | -0.007 |
str | -0.007 |
br | -0.007 |
Token | leave |
Token ID | 2666 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
amaz | +0.026 |
angles | +0.024 |
gasp | +0.023 |
horror | +0.023 |
ill | +0.023 |
Neg logit contributions | |
Nut | -0.036 |
Nut | -0.032 |
age | -0.030 |
Rust | -0.030 |
br | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +4.08e-03 |
Pos logit contributions | |
amaz | +0.017 |
gasp | +0.016 |
angles | +0.015 |
horror | +0.014 |
ill | +0.014 |
Neg logit contributions | |
Nut | -0.027 |
Rust | -0.024 |
str | -0.023 |
br | -0.023 |
rob | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
Nut | +0.008 |
br | +0.007 |
rob | +0.007 |
Rust | +0.007 |
str | +0.007 |
Neg logit contributions | |
amaz | -0.007 |
gasp | -0.006 |
horror | -0.006 |
angles | -0.006 |
ill | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +3.95e-03 |
Pos logit contributions | |
Nut | +0.025 |
br | +0.021 |
Rust | +0.021 |
rob | +0.021 |
Sy | +0.021 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.017 |
ill | -0.016 |
angles | -0.015 |
gasped | -0.015 |
Token | He |
Token ID | 1544 |
Feature activation | -2.06e-04 |
Pos logit contributions | |
Nut | +0.006 |
rob | +0.006 |
con | +0.006 |
Sy | +0.006 |
str | +0.006 |
Neg logit contributions | |
amaz | -0.007 |
gasp | -0.007 |
ill | -0.007 |
urious | -0.006 |
decide | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
amaz | +0.000 |
gasp | +0.000 |
angles | +0.000 |
horror | +0.000 |
ill | +0.000 |
Neg logit contributions | |
Nut | -0.000 |
Rust | -0.000 |
br | -0.000 |
str | -0.000 |
rob | -0.000 |
Token | goodbye |
Token ID | 24829 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Nut | +0.026 |
rob | +0.023 |
Rust | +0.022 |
K | +0.021 |
str | +0.021 |
Neg logit contributions | |
amaz | -0.020 |
gasp | -0.019 |
horror | -0.018 |
angles | -0.018 |
eyes | -0.017 |
Token | zoom |
Token ID | 19792 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
Nut | +0.007 |
Rust | +0.006 |
str | +0.005 |
rob | +0.005 |
br | +0.005 |
Neg logit contributions | |
amaz | -0.006 |
gasp | -0.006 |
angles | -0.006 |
horror | -0.005 |
ill | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
Nut | +0.027 |
Rust | +0.023 |
str | +0.023 |
K | +0.021 |
Nut | +0.021 |
Neg logit contributions | |
amaz | -0.019 |
gasp | -0.018 |
angles | -0.017 |
ill | -0.016 |
horror | -0.015 |
Token | blast |
Token ID | 11975 |
Feature activation | +7.72e-03 |
Pos logit contributions | |
Nut | +0.026 |
Rust | +0.023 |
br | +0.022 |
rob | +0.022 |
str | +0.022 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.015 |
angles | -0.014 |
horror | -0.014 |
ill | -0.014 |
Token | noises |
Token ID | 26782 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
Nut | +0.017 |
Rust | +0.015 |
str | +0.014 |
Nut | +0.014 |
rob | +0.014 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.011 |
angles | -0.011 |
horror | -0.010 |
ill | -0.010 |
Token | with |
Token ID | 351 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
amaz | +0.035 |
gasp | +0.032 |
angles | +0.030 |
horror | +0.029 |
ill | +0.028 |
Neg logit contributions | |
Nut | -0.061 |
Rust | -0.053 |
str | -0.052 |
rob | -0.052 |
br | -0.052 |
Token | his |
Token ID | 465 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
amaz | +0.037 |
gasp | +0.035 |
urious | +0.034 |
horror | +0.033 |
ill | +0.033 |
Neg logit contributions | |
Nut | -0.028 |
Nut | -0.023 |
br | -0.023 |
Rust | -0.022 |
con | -0.021 |
Token | mouth |
Token ID | 5422 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
amaz | +0.044 |
gasp | +0.040 |
ill | +0.040 |
angles | +0.039 |
horror | +0.039 |
Neg logit contributions | |
Nut | -0.049 |
br | -0.042 |
Rust | -0.042 |
con | -0.041 |
Nut | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -7.34e-03 |
Pos logit contributions | |
amaz | +0.029 |
horror | +0.025 |
gasp | +0.025 |
angles | +0.025 |
bowls | +0.023 |
Neg logit contributions | |
Nut | -0.032 |
br | -0.030 |
age | -0.029 |
con | -0.027 |
Rust | -0.027 |
Token | He |
Token ID | 679 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.012 |
horror | +0.012 |
ill | +0.011 |
Neg logit contributions | |
Nut | -0.014 |
Rust | -0.011 |
br | -0.011 |
rob | -0.011 |
str | -0.011 |
Token | shoots |
Token ID | 20611 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
amaz | +0.025 |
gasp | +0.022 |
angles | +0.021 |
horror | +0.021 |
ill | +0.020 |
Neg logit contributions | |
Nut | -0.039 |
br | -0.035 |
Rust | -0.034 |
rob | -0.034 |
con | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -5.00e-03 |
Pos logit contributions | |
amaz | +0.020 |
gasp | +0.019 |
angles | +0.018 |
horror | +0.018 |
ill | +0.018 |
Neg logit contributions | |
Nut | -0.020 |
br | -0.017 |
Rust | -0.017 |
Nut | -0.016 |
con | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +5.87e-03 |
Pos logit contributions | |
Nut | +0.028 |
Nut | +0.024 |
Rust | +0.024 |
Sy | +0.024 |
str | +0.023 |
Neg logit contributions | |
amaz | -0.025 |
gasp | -0.024 |
horror | -0.022 |
angles | -0.022 |
ill | -0.022 |
Token | animal |
Token ID | 5044 |
Feature activation | -5.17e-03 |
Pos logit contributions | |
Nut | +0.012 |
Rust | +0.010 |
rob | +0.010 |
br | +0.010 |
str | +0.010 |
Neg logit contributions | |
amaz | -0.010 |
gasp | -0.009 |
angles | -0.009 |
horror | -0.009 |
ill | -0.008 |
Token | kept |
Token ID | 4030 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
amaz | +0.008 |
gasp | +0.008 |
angles | +0.007 |
horror | +0.007 |
ill | +0.007 |
Neg logit contributions | |
Nut | -0.010 |
str | -0.009 |
age | -0.009 |
Rust | -0.009 |
br | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +8.58e-03 |
Pos logit contributions | |
Nut | +0.044 |
Rust | +0.037 |
rob | +0.037 |
br | +0.036 |
str | +0.036 |
Neg logit contributions | |
amaz | -0.040 |
gasp | -0.037 |
angles | -0.035 |
horror | -0.034 |
ill | -0.033 |
Token | jewel |
Token ID | 16927 |
Feature activation | +8.57e-03 |
Pos logit contributions | |
Nut | +0.017 |
Rust | +0.014 |
br | +0.014 |
rob | +0.014 |
str | +0.014 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.014 |
angles | -0.013 |
horror | -0.013 |
ill | -0.013 |
Token | safe |
Token ID | 3338 |
Feature activation | -5.90e-04 |
Pos logit contributions | |
Nut | +0.016 |
Rust | +0.014 |
br | +0.013 |
str | +0.013 |
Nut | +0.013 |
Neg logit contributions | |
amaz | -0.016 |
gasp | -0.014 |
angles | -0.014 |
horror | -0.014 |
ill | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-03 |
Pos logit contributions | |
amaz | +0.001 |
gasp | +0.001 |
angles | +0.001 |
horror | +0.001 |
ill | +0.001 |
Neg logit contributions | |
Nut | -0.001 |
Rust | -0.001 |
Nut | -0.001 |
br | -0.001 |
age | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +9.33e-03 |
Pos logit contributions | |
amaz | +0.002 |
gasp | +0.002 |
angles | +0.002 |
horror | +0.002 |
ill | +0.002 |
Neg logit contributions | |
Nut | -0.003 |
rob | -0.002 |
br | -0.002 |
Rust | -0.002 |
str | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +4.62e-04 |
Pos logit contributions | |
Nut | +0.018 |
rob | +0.016 |
br | +0.016 |
cone | +0.016 |
con | +0.016 |
Neg logit contributions | |
amaz | -0.015 |
gasp | -0.014 |
ill | -0.013 |
angles | -0.013 |
gasped | -0.013 |
Token | The |
Token ID | 464 |
Feature activation | +5.44e-03 |
Pos logit contributions | |
Nut | +0.001 |
rob | +0.001 |
con | +0.001 |
cone | +0.001 |
Nut | +0.001 |
Neg logit contributions | |
amaz | -0.001 |
ill | -0.001 |
angles | -0.001 |
gasp | -0.001 |
gasped | -0.001 |
Token | animal |
Token ID | 5044 |
Feature activation | +4.07e-04 |
Pos logit contributions | |
Nut | +0.011 |
Rust | +0.009 |
rob | +0.009 |
br | +0.009 |
str | +0.009 |
Neg logit contributions | |
amaz | -0.009 |
gasp | -0.008 |
angles | -0.008 |
horror | -0.008 |
bowls | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +8.35e-06 |
Pos logit contributions | |
amaz | +0.023 |
gasp | +0.021 |
horror | +0.020 |
angles | +0.020 |
ill | +0.018 |
Neg logit contributions | |
Nut | -0.029 |
br | -0.025 |
Rust | -0.024 |
con | -0.024 |
Nut | -0.024 |
Token | His |
Token ID | 2399 |
Feature activation | -4.01e-03 |
Token | collar |
Token ID | 19908 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
amaz | +0.007 |
gasp | +0.006 |
eyes | +0.006 |
horror | +0.006 |
wish | +0.006 |
Neg logit contributions | |
Nut | -0.008 |
rob | -0.007 |
br | -0.007 |
Rust | -0.007 |
str | -0.006 |
Token | helped |
Token ID | 4193 |
Feature activation | -9.01e-03 |
Pos logit contributions | |
Nut | +0.038 |
Rust | +0.034 |
str | +0.033 |
rob | +0.033 |
Silver | +0.032 |
Neg logit contributions | |
amaz | -0.017 |
gasp | -0.016 |
angles | -0.014 |
gasped | -0.014 |
horror | -0.013 |
Token | his |
Token ID | 465 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
amaz | +0.010 |
gasp | +0.009 |
angles | +0.008 |
horror | +0.008 |
decide | +0.008 |
Neg logit contributions | |
Nut | -0.023 |
str | -0.020 |
rob | -0.020 |
Rust | -0.020 |
br | -0.020 |
Token | owner |
Token ID | 4870 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
amaz | +0.028 |
gasp | +0.026 |
horror | +0.025 |
angles | +0.024 |
wish | +0.023 |
Neg logit contributions | |
Nut | -0.033 |
rob | -0.029 |
Rust | -0.029 |
br | -0.028 |
str | -0.028 |
Token | find |
Token ID | 1064 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
amaz | +0.053 |
horror | +0.046 |
gasp | +0.046 |
angles | +0.045 |
ill | +0.040 |
Neg logit contributions | |
Nut | -0.075 |
age | -0.065 |
Nut | -0.063 |
br | -0.062 |
str | -0.062 |
Token | him |
Token ID | 683 |
Feature activation | +6.96e-03 |
Pos logit contributions | |
amaz | +0.019 |
gasp | +0.017 |
angles | +0.017 |
horror | +0.017 |
bowls | +0.016 |
Neg logit contributions | |
Nut | -0.023 |
str | -0.020 |
Rust | -0.019 |
rob | -0.019 |
br | -0.019 |
Token | if |
Token ID | 611 |
Feature activation | -2.20e-03 |
Pos logit contributions | |
Nut | +0.015 |
rob | +0.013 |
Rust | +0.013 |
str | +0.013 |
br | +0.013 |
Neg logit contributions | |
amaz | -0.011 |
gasp | -0.010 |
angles | -0.009 |
bowls | -0.009 |
horror | -0.009 |
Token | he |
Token ID | 339 |
Feature activation | -7.94e-03 |
Pos logit contributions | |
amaz | +0.004 |
gasp | +0.004 |
angles | +0.004 |
horror | +0.004 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.004 |
Rust | -0.003 |
br | -0.003 |
Nut | -0.003 |
str | -0.003 |
Token | ever |
Token ID | 1683 |
Feature activation | -7.13e-03 |
Pos logit contributions | |
amaz | +0.012 |
gasp | +0.011 |
angles | +0.011 |
horror | +0.010 |
ill | +0.010 |
Neg logit contributions | |
Nut | -0.018 |
Rust | -0.015 |
Nut | -0.015 |
br | -0.015 |
str | -0.015 |
Token | nuts |
Token ID | 14380 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
amaz | +0.010 |
gasp | +0.009 |
angles | +0.009 |
bowls | +0.008 |
wonders | +0.008 |
Neg logit contributions | |
Nut | -0.012 |
Rust | -0.011 |
str | -0.010 |
age | -0.010 |
rob | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +8.54e-03 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
angles | +0.005 |
ill | +0.005 |
bowls | +0.005 |
Neg logit contributions | |
Nut | -0.009 |
Rust | -0.008 |
rob | -0.008 |
str | -0.008 |
K | -0.008 |
Token | store |
Token ID | 3650 |
Feature activation | +6.14e-03 |
Pos logit contributions | |
Nut | +0.020 |
Rust | +0.018 |
str | +0.017 |
br | +0.017 |
rob | +0.017 |
Neg logit contributions | |
amaz | -0.012 |
gasp | -0.010 |
angles | -0.010 |
horror | -0.009 |
ill | -0.009 |
Token | them |
Token ID | 606 |
Feature activation | -2.93e-03 |
Pos logit contributions | |
Nut | +0.011 |
str | +0.010 |
Rust | +0.010 |
rob | +0.010 |
br | +0.010 |
Neg logit contributions | |
amaz | -0.011 |
angles | -0.010 |
gasp | -0.010 |
bowls | -0.009 |
eyes | -0.009 |
Token | in |
Token ID | 287 |
Feature activation | -6.97e-03 |
Pos logit contributions | |
amaz | +0.004 |
angles | +0.004 |
gasp | +0.004 |
bowls | +0.004 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.006 |
Rust | -0.005 |
rob | -0.005 |
str | -0.005 |
br | -0.005 |
Token | his |
Token ID | 465 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
amaz | +0.013 |
gasp | +0.012 |
angles | +0.011 |
horror | +0.011 |
ill | +0.011 |
Neg logit contributions | |
Nut | -0.013 |
Rust | -0.011 |
str | -0.011 |
br | -0.011 |
rob | -0.011 |
Token | tree |
Token ID | 5509 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
amaz | +0.042 |
gasp | +0.035 |
horror | +0.033 |
bowls | +0.032 |
angles | +0.032 |
Neg logit contributions | |
Nut | -0.081 |
rob | -0.073 |
Rust | -0.072 |
br | -0.072 |
str | -0.070 |
Token | house |
Token ID | 4803 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
amaz | +0.018 |
gasp | +0.016 |
angles | +0.015 |
horror | +0.015 |
ill | +0.014 |
Neg logit contributions | |
Nut | -0.028 |
Rust | -0.024 |
str | -0.023 |
br | -0.023 |
Nut | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +5.18e-03 |
Pos logit contributions | |
amaz | +0.022 |
gasp | +0.019 |
angles | +0.019 |
bowls | +0.018 |
ill | +0.018 |
Neg logit contributions | |
Nut | -0.047 |
Rust | -0.042 |
rob | -0.041 |
str | -0.040 |
br | -0.040 |
Token | One |
Token ID | 1881 |
Feature activation | -7.54e-03 |
Pos logit contributions | |
iversity | +0.012 |
br | +0.012 |
str | +0.012 |
Nut | +0.012 |
rob | +0.012 |
Neg logit contributions | |
amaz | -0.005 |
Mom | -0.004 |
asks | -0.004 |
wonder | -0.004 |
eyes | -0.004 |
Token | day |
Token ID | 1110 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
amaz | +0.012 |
gasp | +0.012 |
ill | +0.011 |
wonder | +0.011 |
horror | +0.011 |
Neg logit contributions | |
Nut | -0.015 |
rob | -0.013 |
Rust | -0.013 |
con | -0.013 |
br | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -7.44e-03 |
Pos logit contributions | |
Nut | +0.009 |
rob | +0.008 |
Rust | +0.008 |
str | +0.008 |
br | +0.008 |
Neg logit contributions | |
amaz | -0.008 |
gasp | -0.008 |
ill | -0.007 |
angles | -0.007 |
horror | -0.007 |
Token | It |
Token ID | 1026 |
Feature activation | -2.90e-03 |
Pos logit contributions | |
amaz | +0.017 |
gasp | +0.016 |
angles | +0.015 |
horror | +0.015 |
ill | +0.015 |
Neg logit contributions | |
Nut | -0.009 |
rob | -0.008 |
str | -0.008 |
con | -0.007 |
Sy | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.004 |
angles | +0.004 |
horror | +0.004 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.006 |
Rust | -0.005 |
br | -0.005 |
Nut | -0.005 |
str | -0.005 |
Token | very |
Token ID | 845 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
amaz | +0.025 |
gasp | +0.023 |
horror | +0.022 |
angles | +0.022 |
ill | +0.021 |
Neg logit contributions | |
Nut | -0.023 |
Rust | -0.020 |
rob | -0.019 |
br | -0.019 |
str | -0.019 |
Token | quiet |
Token ID | 5897 |
Feature activation | -2.91e-03 |
Pos logit contributions | |
amaz | +0.038 |
angles | +0.036 |
gasp | +0.034 |
ill | +0.034 |
horror | +0.032 |
Neg logit contributions | |
Nut | -0.033 |
Nut | -0.028 |
str | -0.027 |
K | -0.027 |
Bug | -0.026 |
Token | inside |
Token ID | 2641 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
angles | +0.005 |
horror | +0.005 |
ill | +0.004 |
Neg logit contributions | |
Nut | -0.006 |
Rust | -0.005 |
str | -0.005 |
br | -0.005 |
rob | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -9.76e-03 |
Pos logit contributions | |
amaz | +0.088 |
gasp | +0.084 |
angles | +0.079 |
horror | +0.079 |
ill | +0.077 |
Neg logit contributions | |
Nut | -0.059 |
Nut | -0.050 |
Rust | -0.046 |
Bun | -0.046 |
Sy | -0.045 |
Token | house |
Token ID | 2156 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.013 |
ill | +0.013 |
adds | +0.011 |
Neg logit contributions | |
Nut | -0.023 |
Rust | -0.020 |
Nut | -0.020 |
K | -0.020 |
Bug | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -2.60e-03 |
Pos logit contributions | |
amaz | +0.053 |
gasp | +0.049 |
angles | +0.048 |
horror | +0.047 |
spit | +0.044 |
Neg logit contributions | |
Nut | -0.044 |
Nut | -0.037 |
age | -0.037 |
con | -0.036 |
str | -0.036 |
Token | Benny |
Token ID | 44275 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
amaz | +0.005 |
gasp | +0.005 |
angles | +0.005 |
ill | +0.005 |
horror | +0.005 |
Neg logit contributions | |
Nut | -0.004 |
Nut | -0.004 |
Rust | -0.003 |
Bun | -0.003 |
Sy | -0.003 |
Token | felt |
Token ID | 2936 |
Feature activation | -8.72e-03 |
Pos logit contributions | |
Nut | +0.027 |
Rust | +0.023 |
Nut | +0.023 |
str | +0.023 |
age | +0.022 |
Neg logit contributions | |
amaz | -0.018 |
gasp | -0.016 |
angles | -0.015 |
horror | -0.015 |
ill | -0.014 |
Token | aroo |
Token ID | 38049 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Nut | +0.021 |
Rust | +0.016 |
rob | +0.015 |
Nut | +0.015 |
br | +0.015 |
Neg logit contributions | |
amaz | -0.040 |
gasp | -0.040 |
angles | -0.038 |
ill | -0.038 |
horror | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
amaz | +0.024 |
gasp | +0.022 |
angles | +0.021 |
horror | +0.021 |
ill | +0.020 |
Neg logit contributions | |
Nut | -0.030 |
Rust | -0.025 |
Nut | -0.025 |
str | -0.025 |
age | -0.024 |
Token | He |
Token ID | 679 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
amaz | +0.039 |
gasp | +0.037 |
angles | +0.035 |
horror | +0.034 |
bowls | +0.032 |
Neg logit contributions | |
Nut | -0.044 |
Nut | -0.038 |
Rust | -0.036 |
Sy | -0.036 |
str | -0.036 |
Token | stopped |
Token ID | 5025 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
amaz | +0.014 |
gasp | +0.013 |
angles | +0.012 |
gasped | +0.011 |
horror | +0.011 |
Neg logit contributions | |
Nut | -0.033 |
rob | -0.029 |
Rust | -0.029 |
br | -0.029 |
age | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +7.52e-03 |
Pos logit contributions | |
amaz | +0.025 |
gasp | +0.024 |
angles | +0.022 |
horror | +0.021 |
ill | +0.021 |
Neg logit contributions | |
Nut | -0.032 |
rob | -0.029 |
br | -0.028 |
Rust | -0.028 |
age | -0.028 |
Token | stared |
Token ID | 18484 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
Nut | +0.018 |
Rust | +0.015 |
str | +0.015 |
Nut | +0.015 |
br | +0.015 |
Neg logit contributions | |
amaz | -0.010 |
gasp | -0.009 |
angles | -0.009 |
horror | -0.009 |
ill | -0.008 |
Token | at |
Token ID | 379 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
amaz | +0.057 |
gasp | +0.053 |
angles | +0.051 |
horror | +0.051 |
bowls | +0.050 |
Neg logit contributions | |
Nut | -0.072 |
Nut | -0.064 |
Sy | -0.062 |
br | -0.061 |
Rust | -0.060 |
Token | it |
Token ID | 340 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
amaz | +0.033 |
gasp | +0.030 |
angles | +0.030 |
horror | +0.028 |
bowls | +0.027 |
Neg logit contributions | |
Nut | -0.039 |
br | -0.034 |
age | -0.033 |
str | -0.033 |
Rust | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
amaz | +0.042 |
gasp | +0.039 |
horror | +0.038 |
angles | +0.037 |
bowls | +0.037 |
Neg logit contributions | |
Nut | -0.044 |
br | -0.038 |
con | -0.037 |
age | -0.037 |
Rust | -0.037 |
Token | He |
Token ID | 679 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
amaz | +0.031 |
gasp | +0.029 |
angles | +0.028 |
horror | +0.027 |
ill | +0.026 |
Neg logit contributions | |
Nut | -0.035 |
Rust | -0.029 |
Nut | -0.029 |
str | -0.028 |
br | -0.028 |
Token | was |
Token ID | 373 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
amaz | +0.024 |
gasp | +0.022 |
angles | +0.021 |
horror | +0.020 |
ill | +0.019 |
Neg logit contributions | |
Nut | -0.043 |
Rust | -0.037 |
str | -0.036 |
br | -0.036 |
Nut | -0.036 |
Token | its |
Token ID | 663 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Nut | +0.015 |
Rust | +0.013 |
str | +0.012 |
rob | +0.012 |
con | +0.012 |
Neg logit contributions | |
amaz | -0.011 |
gasp | -0.010 |
angles | -0.009 |
horror | -0.009 |
ill | -0.009 |
Token | tail |
Token ID | 7894 |
Feature activation | -4.90e-03 |
Pos logit contributions | |
amaz | +0.020 |
gasp | +0.018 |
angles | +0.018 |
horror | +0.017 |
ill | +0.017 |
Neg logit contributions | |
Nut | -0.018 |
Rust | -0.015 |
rob | -0.015 |
br | -0.014 |
str | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
amaz | +0.009 |
gasp | +0.008 |
angles | +0.008 |
horror | +0.007 |
bowls | +0.007 |
Neg logit contributions | |
Nut | -0.009 |
age | -0.008 |
Rust | -0.008 |
br | -0.008 |
pe | -0.007 |
Token | It |
Token ID | 632 |
Feature activation | -8.31e-03 |
Pos logit contributions | |
amaz | +0.035 |
gasp | +0.033 |
angles | +0.033 |
horror | +0.030 |
ill | +0.030 |
Neg logit contributions | |
Nut | -0.037 |
Rust | -0.032 |
br | -0.032 |
Nut | -0.031 |
Silver | -0.030 |
Token | l |
Token ID | 300 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
amaz | +0.012 |
gasp | +0.011 |
ill | +0.011 |
angles | +0.011 |
horror | +0.010 |
Neg logit contributions | |
Nut | -0.017 |
Rust | -0.015 |
br | -0.015 |
Silver | -0.015 |
rob | -0.015 |
Token | icks |
Token ID | 3378 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
amaz | +0.054 |
gasp | +0.047 |
angles | +0.047 |
bowls | +0.043 |
horror | +0.042 |
Neg logit contributions | |
Nut | -0.082 |
br | -0.073 |
Rust | -0.072 |
Silver | -0.070 |
age | -0.070 |
Token | Anna |
Token ID | 11735 |
Feature activation | +3.88e-03 |
Pos logit contributions | |
amaz | +0.079 |
angles | +0.070 |
gasp | +0.069 |
urious | +0.064 |
horror | +0.063 |
Neg logit contributions | |
Nut | -0.076 |
br | -0.067 |
Rust | -0.067 |
Silver | -0.064 |
K | -0.062 |
Token | 's |
Token ID | 338 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
Nut | +0.008 |
Rust | +0.007 |
age | +0.007 |
br | +0.007 |
Silver | +0.006 |
Neg logit contributions | |
amaz | -0.006 |
angles | -0.006 |
gasp | -0.006 |
horror | -0.005 |
ill | -0.005 |
Token | hand |
Token ID | 1021 |
Feature activation | -8.15e-03 |
Pos logit contributions | |
amaz | +0.012 |
angles | +0.011 |
gasp | +0.010 |
ill | +0.010 |
horror | +0.010 |
Neg logit contributions | |
Nut | -0.029 |
age | -0.026 |
Rust | -0.025 |
br | -0.025 |
Nut | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
amaz | +0.013 |
gasp | +0.012 |
angles | +0.012 |
horror | +0.011 |
bowls | +0.011 |
Neg logit contributions | |
Nut | -0.016 |
age | -0.014 |
Rust | -0.014 |
br | -0.014 |
con | -0.014 |
Token | Anna |
Token ID | 11735 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
amaz | +0.028 |
gasp | +0.026 |
angles | +0.025 |
horror | +0.024 |
ill | +0.023 |
Neg logit contributions | |
Nut | -0.031 |
Rust | -0.026 |
br | -0.026 |
Nut | -0.025 |
rob | -0.025 |