Token | They |
Token ID | 2990 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
Blood | +0.027 |
arrest | +0.027 |
robbers | +0.026 |
blown | +0.025 |
growth | +0.025 |
Neg logit contributions | |
Deal | -0.033 |
difference | -0.032 |
ement | -0.031 |
decision | -0.030 |
isn | -0.030 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Blood | +0.012 |
arrest | +0.012 |
robbers | +0.012 |
blown | +0.012 |
growth | +0.012 |
Neg logit contributions | |
Deal | -0.018 |
difference | -0.017 |
ement | -0.016 |
decision | -0.016 |
isn | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Deal | +0.043 |
ement | +0.040 |
difference | +0.039 |
promise | +0.036 |
decision | +0.036 |
Neg logit contributions | |
arrest | -0.014 |
blown | -0.014 |
robbers | -0.013 |
Blood | -0.013 |
spilled | -0.012 |
Token | have |
Token ID | 423 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Deal | +0.044 |
ement | +0.039 |
difference | +0.037 |
decision | +0.036 |
exactly | +0.036 |
Neg logit contributions | |
arrest | -0.034 |
robbers | -0.032 |
growth | -0.032 |
blown | -0.031 |
Blood | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Blood | +0.002 |
arrest | +0.002 |
robbers | +0.002 |
blown | +0.002 |
growth | +0.002 |
Neg logit contributions | |
Deal | -0.004 |
difference | -0.004 |
ement | -0.004 |
isn | -0.003 |
why | -0.003 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.92e+00 |
Pos logit contributions | |
Deal | +0.046 |
ement | +0.042 |
uct | +0.038 |
difference | +0.038 |
isn | +0.037 |
Neg logit contributions | |
robbers | -0.041 |
arrest | -0.039 |
Blood | -0.039 |
spilled | -0.039 |
blown | -0.038 |
Token | cy |
Token ID | 948 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Deal | +0.191 |
difference | +0.173 |
why | +0.168 |
isn | +0.167 |
int | +0.165 |
Neg logit contributions | |
arrest | -0.101 |
Blood | -0.096 |
growth | -0.094 |
robbers | -0.093 |
spilled | -0.092 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
arrest | +0.027 |
robbers | +0.027 |
spilled | +0.026 |
blown | +0.025 |
Blood | +0.025 |
Neg logit contributions | |
Deal | -0.039 |
difference | -0.033 |
ement | -0.033 |
why | -0.033 |
exactly | -0.033 |
Token | party |
Token ID | 2151 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
Deal | +0.039 |
difference | +0.034 |
ement | +0.033 |
isn | +0.032 |
decision | +0.032 |
Neg logit contributions | |
arrest | -0.034 |
robbers | -0.034 |
Blood | -0.034 |
blown | -0.034 |
mice | -0.032 |
Token | with |
Token ID | 351 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
Deal | +0.009 |
difference | +0.008 |
ement | +0.008 |
isn | +0.008 |
decision | +0.008 |
Neg logit contributions | |
robbers | -0.007 |
arrest | -0.007 |
blown | -0.007 |
mice | -0.007 |
spilled | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
Deal | +0.001 |
difference | +0.001 |
decision | +0.001 |
ement | +0.001 |
promise | +0.001 |
Neg logit contributions | |
arrest | -0.000 |
Blood | -0.000 |
robbers | -0.000 |
blown | -0.000 |
growth | -0.000 |
Token | time |
Token ID | 640 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
Deal | +0.028 |
difference | +0.026 |
ement | +0.025 |
isn | +0.024 |
why | +0.024 |
Neg logit contributions | |
Blood | -0.026 |
arrest | -0.025 |
spilled | -0.024 |
robbers | -0.024 |
growth | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Deal | +0.006 |
difference | +0.005 |
decision | +0.005 |
ement | +0.004 |
promise | +0.004 |
Neg logit contributions | |
Blood | -0.007 |
arrest | -0.006 |
robbers | -0.006 |
spilled | -0.006 |
blown | -0.006 |
Token | there |
Token ID | 612 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
Deal | +0.030 |
difference | +0.026 |
ement | +0.025 |
decision | +0.025 |
isn | +0.025 |
Neg logit contributions | |
Blood | -0.025 |
arrest | -0.024 |
robbers | -0.024 |
blown | -0.024 |
growth | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
Deal | +0.001 |
difference | +0.001 |
izes | +0.001 |
int | +0.001 |
ieved | +0.001 |
Neg logit contributions | |
spilled | -0.001 |
Blood | -0.001 |
growth | -0.001 |
blown | -0.001 |
arrest | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
alks | +0.054 |
arrest | +0.053 |
plets | +0.052 |
Blood | +0.052 |
robbers | +0.052 |
Neg logit contributions | |
Deal | -0.051 |
exactly | -0.045 |
equally | -0.043 |
why | -0.043 |
isn | -0.042 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.73e+00 |
Pos logit contributions | |
Deal | +0.005 |
difference | +0.004 |
ement | +0.004 |
isn | +0.004 |
why | +0.004 |
Neg logit contributions | |
Blood | -0.003 |
arrest | -0.002 |
robbers | -0.002 |
blown | -0.002 |
growth | -0.002 |
Token | cy |
Token ID | 948 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
Deal | +0.167 |
difference | +0.149 |
ement | +0.143 |
isn | +0.142 |
why | +0.142 |
Neg logit contributions | |
Blood | -0.122 |
arrest | -0.121 |
robbers | -0.117 |
blown | -0.116 |
growth | -0.115 |
Token | little |
Token ID | 1310 |
Feature activation | -9.31e-01 |
Pos logit contributions | |
Blood | +0.010 |
arrest | +0.010 |
blown | +0.010 |
robbers | +0.010 |
growth | +0.009 |
Neg logit contributions | |
Deal | -0.015 |
difference | -0.013 |
ement | -0.013 |
isn | -0.013 |
decision | -0.012 |
Token | bunny |
Token ID | 44915 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Blood | +0.023 |
growth | +0.023 |
arrest | +0.023 |
blown | +0.022 |
alks | +0.022 |
Neg logit contributions | |
Deal | -0.024 |
difference | -0.022 |
promise | -0.021 |
isn | -0.021 |
decision | -0.020 |
Token | called |
Token ID | 1444 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Deal | +0.033 |
difference | +0.029 |
why | +0.028 |
ement | +0.028 |
exactly | +0.028 |
Neg logit contributions | |
Blood | -0.030 |
arrest | -0.030 |
blown | -0.029 |
robbers | -0.029 |
growth | -0.029 |
Token | Bob |
Token ID | 5811 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
Blood | +0.044 |
blown | +0.044 |
arrest | +0.042 |
spilled | +0.041 |
alks | +0.041 |
Neg logit contributions | |
Deal | -0.049 |
difference | -0.043 |
why | -0.043 |
isn | -0.043 |
exactly | -0.042 |
Token | time |
Token ID | 640 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
Deal | +0.035 |
difference | +0.032 |
ement | +0.031 |
isn | +0.030 |
why | +0.030 |
Neg logit contributions | |
Blood | -0.030 |
arrest | -0.028 |
spilled | -0.027 |
growth | -0.027 |
robbers | -0.026 |
Token | there |
Token ID | 612 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.012 |
decision | +0.011 |
promise | +0.011 |
ement | +0.011 |
Neg logit contributions | |
Blood | -0.016 |
arrest | -0.015 |
robbers | -0.015 |
spilled | -0.014 |
blown | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
Deal | +0.004 |
difference | +0.004 |
izes | +0.004 |
int | +0.004 |
decision | +0.003 |
Neg logit contributions | |
spilled | -0.003 |
Blood | -0.003 |
growth | -0.003 |
blown | -0.003 |
ges | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
alks | +0.053 |
arrest | +0.052 |
robbers | +0.051 |
plets | +0.050 |
Blood | +0.050 |
Neg logit contributions | |
Deal | -0.050 |
exactly | -0.042 |
why | -0.040 |
equally | -0.040 |
isn | -0.040 |
Token | f |
Token ID | 277 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
Deal | +0.010 |
difference | +0.009 |
ement | +0.009 |
isn | +0.009 |
why | +0.009 |
Neg logit contributions | |
Blood | -0.006 |
arrest | -0.006 |
robbers | -0.006 |
blown | -0.006 |
growth | -0.005 |
Token | auc |
Token ID | 14272 |
Feature activation | +5.71e+00 |
Pos logit contributions | |
Deal | +0.052 |
difference | +0.046 |
ement | +0.044 |
why | +0.043 |
isn | +0.043 |
Neg logit contributions | |
Blood | -0.042 |
arrest | -0.041 |
robbers | -0.040 |
blown | -0.040 |
growth | -0.040 |
Token | et |
Token ID | 316 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
difference | +0.116 |
Deal | +0.115 |
exactly | +0.111 |
ement | +0.109 |
isn | +0.106 |
Neg logit contributions | |
Blood | -0.152 |
arrest | -0.150 |
spilled | -0.149 |
robbers | -0.148 |
mice | -0.145 |
Token | , |
Token ID | 11 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
Deal | +0.023 |
difference | +0.020 |
isn | +0.019 |
ement | +0.019 |
decision | +0.019 |
Neg logit contributions | |
Blood | -0.022 |
arrest | -0.022 |
blown | -0.021 |
robbers | -0.020 |
growth | -0.020 |
Token | who |
Token ID | 508 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
Deal | +0.000 |
exactly | +0.000 |
why | +0.000 |
isn | +0.000 |
difference | +0.000 |
Neg logit contributions | |
arrest | -0.000 |
growth | -0.000 |
blown | -0.000 |
alks | -0.000 |
robbers | -0.000 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
Blood | +0.015 |
arrest | +0.014 |
spilled | +0.014 |
blown | +0.014 |
robbers | +0.014 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.010 |
ement | -0.010 |
exactly | -0.009 |
decision | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
arrest | +0.058 |
Blood | +0.057 |
robbers | +0.055 |
plets | +0.054 |
growth | +0.054 |
Neg logit contributions | |
Deal | -0.059 |
why | -0.056 |
difference | -0.056 |
exactly | -0.055 |
isn | -0.054 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Deal | +0.007 |
difference | +0.006 |
decision | +0.006 |
ement | +0.006 |
isn | +0.006 |
Neg logit contributions | |
Blood | -0.005 |
arrest | -0.005 |
blown | -0.005 |
growth | -0.005 |
robbers | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -9.26e-01 |
Pos logit contributions | |
Deal | +0.031 |
difference | +0.026 |
ement | +0.025 |
isn | +0.025 |
why | +0.025 |
Neg logit contributions | |
Blood | -0.035 |
arrest | -0.035 |
robbers | -0.034 |
blown | -0.034 |
growth | -0.034 |
Token | It |
Token ID | 632 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
arrest | +0.019 |
Blood | +0.018 |
blown | +0.018 |
robbers | +0.018 |
spilled | +0.017 |
Neg logit contributions | |
Deal | -0.028 |
difference | -0.025 |
isn | -0.024 |
ement | -0.024 |
why | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
arrest | +0.045 |
Blood | +0.044 |
robbers | +0.044 |
alks | +0.043 |
growth | +0.043 |
Neg logit contributions | |
Deal | -0.030 |
isn | -0.025 |
difference | -0.024 |
why | -0.023 |
promise | -0.022 |
Token | so |
Token ID | 523 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
Blood | +0.040 |
arrest | +0.040 |
robbers | +0.039 |
blown | +0.039 |
growth | +0.038 |
Neg logit contributions | |
Deal | -0.050 |
difference | -0.044 |
ement | -0.042 |
isn | -0.042 |
why | -0.042 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.69e+00 |
Pos logit contributions | |
arrest | +0.013 |
Blood | +0.013 |
robbers | +0.013 |
growth | +0.013 |
blown | +0.013 |
Neg logit contributions | |
Deal | -0.017 |
difference | -0.015 |
isn | -0.015 |
ement | -0.015 |
exactly | -0.014 |
Token | cy |
Token ID | 948 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
Deal | +0.168 |
difference | +0.155 |
decision | +0.147 |
why | +0.145 |
ement | +0.145 |
Neg logit contributions | |
Blood | -0.118 |
arrest | -0.114 |
blown | -0.110 |
growth | -0.109 |
robbers | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
arrest | +0.001 |
Blood | +0.001 |
blown | +0.001 |
robbers | +0.001 |
spilled | +0.001 |
Neg logit contributions | |
Deal | -0.001 |
difference | -0.001 |
ement | -0.001 |
decision | -0.001 |
why | -0.001 |
Token | fun |
Token ID | 1257 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
arrest | +0.018 |
Blood | +0.018 |
growth | +0.018 |
robbers | +0.018 |
blown | +0.017 |
Neg logit contributions | |
Deal | -0.020 |
difference | -0.017 |
isn | -0.017 |
why | -0.016 |
exactly | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
Blood | +0.009 |
arrest | +0.009 |
robbers | +0.009 |
blown | +0.009 |
mice | +0.009 |
Neg logit contributions | |
Deal | -0.010 |
difference | -0.009 |
ement | -0.009 |
decision | -0.008 |
promise | -0.008 |
Token | play |
Token ID | 711 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
Deal | +0.026 |
difference | +0.023 |
ement | +0.022 |
isn | +0.022 |
exactly | +0.022 |
Neg logit contributions | |
Blood | -0.023 |
arrest | -0.022 |
robbers | -0.022 |
blown | -0.022 |
growth | -0.021 |
Token | time |
Token ID | 640 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
Deal | +0.028 |
difference | +0.025 |
ement | +0.024 |
why | +0.024 |
isn | +0.024 |
Neg logit contributions | |
Blood | -0.026 |
arrest | -0.024 |
spilled | -0.024 |
growth | -0.023 |
robbers | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.012 |
decision | +0.011 |
ement | +0.011 |
promise | +0.011 |
Neg logit contributions | |
Blood | -0.018 |
arrest | -0.017 |
robbers | -0.016 |
spilled | -0.016 |
blown | -0.016 |
Token | there |
Token ID | 612 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
Deal | +0.034 |
difference | +0.030 |
ement | +0.029 |
decision | +0.028 |
isn | +0.028 |
Neg logit contributions | |
Blood | -0.029 |
arrest | -0.028 |
robbers | -0.028 |
blown | -0.027 |
growth | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
Deal | +0.005 |
difference | +0.005 |
izes | +0.005 |
int | +0.005 |
ieved | +0.005 |
Neg logit contributions | |
spilled | -0.004 |
Blood | -0.004 |
growth | -0.003 |
ges | -0.003 |
arrest | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
alks | +0.049 |
arrest | +0.048 |
robbers | +0.047 |
Blood | +0.047 |
plets | +0.047 |
Neg logit contributions | |
Deal | -0.049 |
exactly | -0.042 |
equally | -0.041 |
why | -0.040 |
difference | -0.040 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.68e+00 |
Pos logit contributions | |
Deal | +0.004 |
difference | +0.004 |
ement | +0.004 |
isn | +0.004 |
why | +0.004 |
Neg logit contributions | |
Blood | -0.002 |
arrest | -0.002 |
robbers | -0.002 |
blown | -0.002 |
growth | -0.002 |
Token | cy |
Token ID | 948 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
Deal | +0.163 |
difference | +0.145 |
ement | +0.139 |
isn | +0.138 |
why | +0.138 |
Neg logit contributions | |
Blood | -0.124 |
arrest | -0.123 |
robbers | -0.119 |
blown | -0.118 |
growth | -0.117 |
Token | rubber |
Token ID | 14239 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
Blood | +0.004 |
arrest | +0.004 |
blown | +0.004 |
robbers | +0.004 |
growth | +0.004 |
Neg logit contributions | |
Deal | -0.006 |
difference | -0.006 |
ement | -0.005 |
isn | -0.005 |
why | -0.005 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
Deal | +0.011 |
difference | +0.010 |
isn | +0.009 |
promise | +0.009 |
why | +0.009 |
Neg logit contributions | |
Blood | -0.008 |
arrest | -0.008 |
growth | -0.008 |
blown | -0.008 |
alks | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
Deal | +0.031 |
difference | +0.026 |
ement | +0.025 |
isn | +0.025 |
why | +0.025 |
Neg logit contributions | |
Blood | -0.036 |
arrest | -0.035 |
robbers | -0.035 |
blown | -0.034 |
growth | -0.034 |
Token | It |
Token ID | 632 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Blood | +0.014 |
arrest | +0.013 |
robbers | +0.013 |
blown | +0.013 |
growth | +0.013 |
Neg logit contributions | |
Deal | -0.022 |
difference | -0.020 |
ement | -0.019 |
isn | -0.019 |
decision | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
Deal | +0.028 |
difference | +0.023 |
ement | +0.023 |
decision | +0.023 |
promise | +0.023 |
Neg logit contributions | |
arrest | -0.021 |
mice | -0.020 |
blown | -0.020 |
growth | -0.020 |
robbers | -0.020 |
Token | deer |
Token ID | 20096 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
robbers | +0.003 |
mice | +0.003 |
arrest | +0.003 |
blown | +0.003 |
bushes | +0.003 |
Neg logit contributions | |
Deal | -0.007 |
ieved | -0.006 |
ement | -0.006 |
why | -0.006 |
exactly | -0.006 |
Token | again |
Token ID | 757 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
arrest | +0.016 |
blown | +0.016 |
robbers | +0.016 |
mice | +0.016 |
spilled | +0.015 |
Neg logit contributions | |
Deal | -0.018 |
ement | -0.015 |
difference | -0.014 |
isn | -0.014 |
uct | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
arrest | +0.001 |
Blood | +0.001 |
robbers | +0.001 |
growth | +0.001 |
blown | +0.001 |
Neg logit contributions | |
Deal | -0.001 |
difference | -0.001 |
ement | -0.001 |
isn | -0.001 |
decision | -0.001 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
Deal | +0.016 |
difference | +0.014 |
decision | +0.014 |
ement | +0.014 |
promise | +0.014 |
Neg logit contributions | |
Blood | -0.013 |
growth | -0.012 |
arrest | -0.012 |
robbers | -0.012 |
alks | -0.011 |
Token | â |
Token ID | 22940 |
Feature activation | +5.61e+00 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.012 |
ement | +0.012 |
decision | +0.012 |
why | +0.012 |
Neg logit contributions | |
Blood | -0.012 |
arrest | -0.011 |
robbers | -0.011 |
blown | -0.011 |
growth | -0.011 |
Token | € |
Token ID | 26391 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
Deal | +0.138 |
exactly | +0.133 |
difference | +0.131 |
decision | +0.127 |
why | +0.125 |
Neg logit contributions | |
Blood | -0.134 |
growth | -0.131 |
arrest | -0.129 |
blown | -0.126 |
robbers | -0.123 |
Token | � |
Token ID | 129 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
Deal | +0.012 |
difference | +0.011 |
decision | +0.011 |
ement | +0.010 |
int | +0.010 |
Neg logit contributions | |
Blood | -0.009 |
arrest | -0.009 |
growth | -0.008 |
robbers | -0.008 |
alks | -0.008 |
Token | � |
Token ID | 241 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Deal | +0.083 |
difference | +0.078 |
decision | +0.075 |
choice | +0.074 |
exactly | +0.071 |
Neg logit contributions | |
arrest | -0.080 |
Blood | -0.076 |
growth | -0.072 |
blown | -0.068 |
robbers | -0.068 |
Token | Can |
Token ID | 6090 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
Deal | +0.043 |
difference | +0.041 |
ement | +0.039 |
decision | +0.038 |
promise | +0.037 |
Neg logit contributions | |
Blood | -0.036 |
arrest | -0.033 |
growth | -0.033 |
blown | -0.033 |
robbers | -0.032 |
Token | I |
Token ID | 314 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
Blood | +0.003 |
arrest | +0.003 |
robbers | +0.003 |
blown | +0.003 |
growth | +0.002 |
Neg logit contributions | |
Deal | -0.003 |
difference | -0.003 |
ement | -0.003 |
decision | -0.003 |
exactly | -0.003 |
Token | there |
Token ID | 612 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
Deal | +0.035 |
difference | +0.031 |
ement | +0.030 |
isn | +0.029 |
why | +0.029 |
Neg logit contributions | |
Blood | -0.032 |
arrest | -0.031 |
robbers | -0.031 |
blown | -0.030 |
growth | -0.030 |
Token | was |
Token ID | 373 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Deal | +0.009 |
difference | +0.008 |
izes | +0.007 |
int | +0.007 |
decision | +0.007 |
Neg logit contributions | |
spilled | -0.007 |
Blood | -0.007 |
growth | -0.006 |
arrest | -0.006 |
blown | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
alks | +0.043 |
arrest | +0.042 |
plets | +0.042 |
robbers | +0.042 |
Blood | +0.042 |
Neg logit contributions | |
Deal | -0.044 |
exactly | -0.039 |
why | -0.038 |
difference | -0.038 |
equally | -0.037 |
Token | big |
Token ID | 1263 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
Deal | +0.005 |
difference | +0.005 |
ement | +0.004 |
isn | +0.004 |
why | +0.004 |
Neg logit contributions | |
Blood | -0.003 |
arrest | -0.003 |
blown | -0.003 |
robbers | -0.003 |
growth | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
Deal | +0.057 |
ement | +0.050 |
ieved | +0.049 |
exactly | +0.049 |
difference | +0.049 |
Neg logit contributions | |
robbers | -0.037 |
arrest | -0.036 |
mice | -0.036 |
Blood | -0.035 |
spilled | -0.034 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.60e+00 |
Pos logit contributions | |
arrest | +0.025 |
growth | +0.025 |
Blood | +0.024 |
blown | +0.024 |
alks | +0.024 |
Neg logit contributions | |
Deal | -0.031 |
difference | -0.027 |
isn | -0.026 |
promise | -0.026 |
ement | -0.026 |
Token | cy |
Token ID | 948 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
Deal | +0.159 |
difference | +0.142 |
ement | +0.136 |
isn | +0.135 |
why | +0.135 |
Neg logit contributions | |
Blood | -0.124 |
arrest | -0.123 |
robbers | -0.119 |
blown | -0.118 |
growth | -0.116 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
arrest | +0.005 |
Blood | +0.005 |
robbers | +0.004 |
blown | +0.004 |
mice | +0.004 |
Neg logit contributions | |
Deal | -0.007 |
difference | -0.006 |
ement | -0.006 |
why | -0.006 |
exactly | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
Deal | +0.029 |
difference | +0.024 |
ement | +0.023 |
isn | +0.023 |
why | +0.023 |
Neg logit contributions | |
Blood | -0.035 |
arrest | -0.034 |
robbers | -0.034 |
blown | -0.034 |
growth | -0.033 |
Token | It |
Token ID | 632 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
arrest | +0.014 |
Blood | +0.014 |
blown | +0.014 |
robbers | +0.013 |
growth | +0.013 |
Neg logit contributions | |
Deal | -0.021 |
difference | -0.019 |
isn | -0.019 |
ement | -0.018 |
why | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
arrest | +0.032 |
Blood | +0.032 |
robbers | +0.032 |
blown | +0.031 |
growth | +0.031 |
Neg logit contributions | |
Deal | -0.023 |
difference | -0.019 |
isn | -0.019 |
ement | -0.018 |
why | -0.018 |
Token | up |
Token ID | 510 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Blood | +0.052 |
arrest | +0.051 |
robbers | +0.050 |
growth | +0.050 |
blown | +0.050 |
Neg logit contributions | |
Deal | -0.029 |
difference | -0.024 |
ement | -0.022 |
isn | -0.022 |
why | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Blood | +0.024 |
arrest | +0.024 |
robbers | +0.024 |
growth | +0.023 |
blown | +0.023 |
Neg logit contributions | |
Deal | -0.026 |
difference | -0.023 |
ement | -0.022 |
exactly | -0.022 |
isn | -0.022 |
Token | down |
Token ID | 866 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
Deal | +0.033 |
difference | +0.030 |
ement | +0.030 |
decision | +0.028 |
promise | +0.028 |
Neg logit contributions | |
Blood | -0.025 |
blown | -0.024 |
spilled | -0.024 |
robbers | -0.024 |
arrest | -0.023 |
Token | like |
Token ID | 588 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
Blood | +0.042 |
arrest | +0.042 |
robbers | +0.041 |
blown | +0.041 |
growth | +0.040 |
Neg logit contributions | |
Deal | -0.040 |
difference | -0.035 |
ement | -0.033 |
isn | -0.033 |
why | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
Blood | +0.005 |
arrest | +0.005 |
alks | +0.005 |
growth | +0.005 |
blown | +0.005 |
Neg logit contributions | |
Deal | -0.005 |
why | -0.005 |
exactly | -0.004 |
difference | -0.004 |
isn | -0.004 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.60e+00 |
Pos logit contributions | |
Deal | +0.021 |
ement | +0.019 |
ieved | +0.019 |
uct | +0.019 |
why | +0.018 |
Neg logit contributions | |
robbers | -0.012 |
spilled | -0.012 |
Blood | -0.012 |
arrest | -0.012 |
blown | -0.012 |
Token | cy |
Token ID | 948 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
Deal | +0.169 |
difference | +0.153 |
why | +0.147 |
ement | +0.146 |
isn | +0.146 |
Neg logit contributions | |
Blood | -0.111 |
arrest | -0.110 |
robbers | -0.104 |
blown | -0.104 |
growth | -0.104 |
Token | ball |
Token ID | 2613 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
Blood | +0.039 |
arrest | +0.039 |
blown | +0.038 |
robbers | +0.038 |
spilled | +0.037 |
Neg logit contributions | |
Deal | -0.053 |
difference | -0.047 |
ement | -0.045 |
why | -0.045 |
exactly | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Deal | +0.012 |
difference | +0.011 |
isn | +0.010 |
ement | +0.010 |
why | +0.010 |
Neg logit contributions | |
Blood | -0.014 |
arrest | -0.014 |
robbers | -0.014 |
growth | -0.014 |
blown | -0.014 |
Token | |
Token ID | 220 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.013 |
decision | +0.012 |
promise | +0.012 |
ement | +0.012 |
Neg logit contributions | |
Blood | -0.011 |
arrest | -0.009 |
robbers | -0.009 |
spilled | -0.009 |
alks | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Blood | +0.025 |
arrest | +0.022 |
blown | +0.022 |
alks | +0.022 |
robbers | +0.022 |
Neg logit contributions | |
Deal | -0.030 |
difference | -0.026 |
decision | -0.025 |
ement | -0.025 |
promise | -0.024 |
Token | round |
Token ID | 2835 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
Blood | +0.021 |
arrest | +0.021 |
robbers | +0.020 |
blown | +0.020 |
growth | +0.020 |
Neg logit contributions | |
Deal | -0.028 |
difference | -0.025 |
ement | -0.024 |
isn | -0.024 |
why | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
arrest | +0.022 |
Blood | +0.021 |
robbers | +0.021 |
growth | +0.021 |
blown | +0.021 |
Neg logit contributions | |
Deal | -0.014 |
isn | -0.012 |
ement | -0.012 |
why | -0.012 |
difference | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
Blood | +0.027 |
arrest | +0.025 |
blown | +0.025 |
robbers | +0.025 |
spilled | +0.025 |
Neg logit contributions | |
Deal | -0.029 |
difference | -0.026 |
ement | -0.025 |
decision | -0.025 |
promise | -0.024 |
Token | ever |
Token ID | 1683 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
Blood | +0.017 |
arrest | +0.017 |
robbers | +0.016 |
blown | +0.016 |
growth | +0.016 |
Neg logit contributions | |
Deal | -0.016 |
difference | -0.014 |
isn | -0.014 |
ement | -0.014 |
why | -0.014 |
Token | so |
Token ID | 523 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
Blood | +0.010 |
blown | +0.010 |
robbers | +0.010 |
spilled | +0.010 |
arrest | +0.010 |
Neg logit contributions | |
Deal | -0.012 |
difference | -0.010 |
ement | -0.010 |
decision | -0.010 |
exactly | -0.010 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
Deal | +0.015 |
difference | +0.014 |
ement | +0.013 |
decision | +0.013 |
why | +0.013 |
Neg logit contributions | |
Blood | -0.014 |
arrest | -0.014 |
blown | -0.014 |
robbers | -0.014 |
mice | -0.013 |
Token | cy |
Token ID | 948 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
Deal | +0.169 |
difference | +0.156 |
decision | +0.148 |
why | +0.147 |
ement | +0.146 |
Neg logit contributions | |
Blood | -0.112 |
arrest | -0.107 |
growth | -0.104 |
blown | -0.104 |
spilled | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
Blood | +0.007 |
arrest | +0.007 |
robbers | +0.007 |
blown | +0.007 |
growth | +0.007 |
Neg logit contributions | |
Deal | -0.007 |
difference | -0.006 |
ement | -0.006 |
isn | -0.006 |
why | -0.006 |
Token | Every |
Token ID | 3887 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Deal | +0.001 |
difference | +0.001 |
decision | +0.001 |
exactly | +0.001 |
promise | +0.001 |
Neg logit contributions | |
Blood | -0.000 |
arrest | -0.000 |
growth | -0.000 |
robbers | -0.000 |
mice | -0.000 |
Token | day |
Token ID | 1110 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
Deal | +0.037 |
ement | +0.032 |
difference | +0.032 |
exactly | +0.031 |
ieved | +0.030 |
Neg logit contributions | |
Blood | -0.038 |
spilled | -0.036 |
growth | -0.036 |
blown | -0.036 |
arrest | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
Blood | +0.073 |
arrest | +0.072 |
robbers | +0.071 |
blown | +0.070 |
growth | +0.070 |
Neg logit contributions | |
Deal | -0.041 |
difference | -0.035 |
ement | -0.032 |
isn | -0.032 |
why | -0.031 |
Token | next |
Token ID | 1306 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
Deal | +0.017 |
difference | +0.015 |
ement | +0.014 |
exactly | +0.013 |
ieved | +0.013 |
Neg logit contributions | |
arrest | -0.014 |
blown | -0.014 |
Blood | -0.014 |
robbers | -0.014 |
mice | -0.014 |
Token | time |
Token ID | 640 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
Deal | +0.062 |
ement | +0.052 |
difference | +0.052 |
ieved | +0.051 |
exactly | +0.050 |
Neg logit contributions | |
Blood | -0.046 |
blown | -0.046 |
arrest | -0.046 |
mice | -0.044 |
robbers | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
arrest | +0.043 |
Blood | +0.042 |
robbers | +0.042 |
blown | +0.042 |
spilled | +0.040 |
Neg logit contributions | |
Deal | -0.046 |
difference | -0.040 |
ement | -0.039 |
decision | -0.037 |
exactly | -0.036 |
Token | And |
Token ID | 843 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
Deal | +0.021 |
decision | +0.019 |
difference | +0.019 |
ement | +0.018 |
exactly | +0.018 |
Neg logit contributions | |
Blood | -0.018 |
robbers | -0.017 |
arrest | -0.017 |
growth | -0.017 |
mice | -0.016 |
Token | don |
Token ID | 836 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
ement | +0.041 |
Deal | +0.041 |
difference | +0.038 |
ieved | +0.036 |
decision | +0.035 |
Neg logit contributions | |
arrest | -0.034 |
robbers | -0.033 |
mice | -0.033 |
Blood | -0.031 |
blown | -0.031 |
Token | â |
Token ID | 22940 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
Deal | +0.000 |
difference | +0.000 |
ement | +0.000 |
decision | +0.000 |
exactly | +0.000 |
Neg logit contributions | |
Blood | -0.000 |
arrest | -0.000 |
blown | -0.000 |
robbers | -0.000 |
growth | -0.000 |
Token | € |
Token ID | 26391 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
exactly | +0.102 |
amaz | +0.095 |
difference | +0.095 |
isn | +0.093 |
why | +0.091 |
Neg logit contributions | |
Blood | -0.151 |
arrest | -0.149 |
blown | -0.146 |
growth | -0.142 |
robbers | -0.141 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.003 |
decision | +0.003 |
ement | +0.003 |
int | +0.003 |
Neg logit contributions | |
Blood | -0.003 |
arrest | -0.003 |
alks | -0.002 |
robbers | -0.002 |
plets | -0.002 |
Token | t |
Token ID | 83 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
Deal | +0.034 |
difference | +0.032 |
decision | +0.030 |
why | +0.029 |
ement | +0.029 |
Neg logit contributions | |
Blood | -0.019 |
arrest | -0.018 |
spilled | -0.017 |
blown | -0.017 |
alks | -0.017 |
Token | be |
Token ID | 307 |
Feature activation | +7.30e-01 |
Pos logit contributions | |
Deal | +0.011 |
ement | +0.010 |
difference | +0.009 |
exactly | +0.009 |
decision | +0.009 |
Neg logit contributions | |
Blood | -0.009 |
arrest | -0.009 |
robbers | -0.009 |
blown | -0.009 |
spilled | -0.008 |
Token | so |
Token ID | 523 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Deal | +0.018 |
difference | +0.015 |
ement | +0.015 |
decision | +0.014 |
isn | +0.014 |
Neg logit contributions | |
Blood | -0.020 |
blown | -0.020 |
arrest | -0.019 |
robbers | -0.019 |
mice | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
Blood | +0.024 |
arrest | +0.024 |
blown | +0.023 |
robbers | +0.023 |
growth | +0.022 |
Neg logit contributions | |
Deal | -0.037 |
difference | -0.033 |
isn | -0.032 |
why | -0.032 |
exactly | -0.032 |
Token | shiny |
Token ID | 22441 |
Feature activation | +8.26e-01 |
Pos logit contributions | |
Deal | +0.020 |
uct | +0.018 |
ieved | +0.018 |
ement | +0.018 |
why | +0.017 |
Neg logit contributions | |
robbers | -0.012 |
spilled | -0.012 |
Blood | -0.012 |
arrest | -0.011 |
mice | -0.011 |
Token | rock |
Token ID | 3881 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
Deal | +0.025 |
uct | +0.021 |
difference | +0.021 |
ieved | +0.021 |
ement | +0.021 |
Neg logit contributions | |
robbers | -0.017 |
Blood | -0.017 |
spilled | -0.017 |
arrest | -0.017 |
alks | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
Deal | +0.017 |
difference | +0.015 |
ement | +0.014 |
isn | +0.014 |
why | +0.014 |
Neg logit contributions | |
Blood | -0.020 |
arrest | -0.020 |
robbers | -0.020 |
blown | -0.019 |
growth | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
Blood | +0.022 |
arrest | +0.021 |
alks | +0.021 |
blown | +0.020 |
growth | +0.020 |
Neg logit contributions | |
Deal | -0.027 |
why | -0.024 |
isn | -0.024 |
exactly | -0.024 |
difference | -0.024 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
robbers | +0.002 |
Blood | +0.002 |
spilled | +0.002 |
arrest | +0.002 |
blown | +0.002 |
Neg logit contributions | |
Deal | -0.003 |
ement | -0.003 |
ieved | -0.003 |
uct | -0.003 |
why | -0.003 |
Token | cy |
Token ID | 948 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
Deal | +0.166 |
difference | +0.148 |
why | +0.142 |
ement | +0.142 |
exactly | +0.140 |
Neg logit contributions | |
Blood | -0.115 |
arrest | -0.115 |
robbers | -0.110 |
growth | -0.109 |
blown | -0.109 |
Token | ball |
Token ID | 2613 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
Blood | +0.034 |
arrest | +0.034 |
robbers | +0.033 |
blown | +0.032 |
spilled | +0.032 |
Neg logit contributions | |
Deal | -0.050 |
difference | -0.043 |
ement | -0.042 |
why | -0.042 |
exactly | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.013 |
isn | +0.012 |
ement | +0.012 |
why | +0.012 |
Neg logit contributions | |
Blood | -0.017 |
arrest | -0.017 |
robbers | -0.017 |
growth | -0.017 |
blown | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
Deal | +0.016 |
difference | +0.014 |
ement | +0.014 |
promise | +0.014 |
decision | +0.014 |
Neg logit contributions | |
Blood | -0.012 |
robbers | -0.011 |
arrest | -0.011 |
alks | -0.010 |
spilled | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
Blood | +0.001 |
arrest | +0.001 |
growth | +0.001 |
robbers | +0.001 |
blown | +0.001 |
Neg logit contributions | |
Deal | -0.001 |
difference | -0.001 |
ement | -0.001 |
decision | -0.001 |
promise | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +7.10e-01 |
Pos logit contributions | |
arrest | +0.002 |
mice | +0.002 |
blown | +0.002 |
spilled | +0.002 |
alks | +0.002 |
Neg logit contributions | |
Deal | -0.004 |
difference | -0.003 |
ement | -0.003 |
ieved | -0.003 |
int | -0.003 |
Token | â |
Token ID | 22940 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
Deal | +0.021 |
ement | +0.018 |
difference | +0.017 |
ieved | +0.016 |
decision | +0.016 |
Neg logit contributions | |
arrest | -0.016 |
robbers | -0.016 |
blown | -0.015 |
spilled | -0.015 |
mice | -0.015 |
Token | € |
Token ID | 26391 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
exactly | +0.097 |
difference | +0.094 |
Deal | +0.090 |
amaz | +0.088 |
why | +0.087 |
Neg logit contributions | |
Blood | -0.145 |
arrest | -0.136 |
plets | -0.135 |
blown | -0.132 |
growth | -0.132 |
Token | said |
Token ID | 531 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
Deal | +0.005 |
difference | +0.005 |
ement | +0.004 |
decision | +0.004 |
why | +0.004 |
Neg logit contributions | |
arrest | -0.005 |
Blood | -0.005 |
robbers | -0.004 |
blown | -0.004 |
growth | -0.004 |
Token | Ali |
Token ID | 12104 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Deal | +0.026 |
difference | +0.023 |
ieved | +0.022 |
ement | +0.022 |
uct | +0.020 |
Neg logit contributions | |
Blood | -0.018 |
arrest | -0.017 |
spilled | -0.017 |
blown | -0.017 |
Po | -0.017 |
Token | â |
Token ID | 22940 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
Deal | +0.085 |
difference | +0.073 |
ement | +0.071 |
ieved | +0.066 |
decision | +0.065 |
Neg logit contributions | |
arrest | -0.069 |
Blood | -0.069 |
robbers | -0.069 |
mice | -0.067 |
blown | -0.067 |
Token | € |
Token ID | 26391 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
exactly | +0.111 |
difference | +0.106 |
Deal | +0.101 |
why | +0.099 |
amaz | +0.099 |
Neg logit contributions | |
Blood | -0.151 |
robbers | -0.144 |
arrest | -0.142 |
growth | -0.139 |
blown | -0.137 |
Token | ™ |
Token ID | 8151 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
Deal | +0.018 |
difference | +0.017 |
ement | +0.016 |
decision | +0.016 |
why | +0.016 |
Neg logit contributions | |
arrest | -0.014 |
Blood | -0.014 |
robbers | -0.013 |
growth | -0.012 |
blown | -0.012 |
Token | s |
Token ID | 82 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
Deal | +0.070 |
difference | +0.063 |
ement | +0.061 |
isn | +0.060 |
why | +0.060 |
Neg logit contributions | |
Blood | -0.040 |
arrest | -0.039 |
robbers | -0.037 |
blown | -0.037 |
growth | -0.036 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
Deal | +0.052 |
ement | +0.043 |
ieved | +0.042 |
difference | +0.041 |
uct | +0.040 |
Neg logit contributions | |
Blood | -0.055 |
arrest | -0.055 |
alks | -0.054 |
robbers | -0.053 |
blown | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Deal | +0.006 |
difference | +0.005 |
ement | +0.005 |
decision | +0.005 |
deal | +0.005 |
Neg logit contributions | |
robbers | -0.004 |
Blood | -0.004 |
arrest | -0.004 |
spilled | -0.004 |
mice | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
Deal | +0.042 |
difference | +0.037 |
why | +0.036 |
ement | +0.036 |
isn | +0.036 |
Neg logit contributions | |
Blood | -0.037 |
arrest | -0.036 |
robbers | -0.035 |
blown | -0.035 |
growth | -0.034 |
Token | He |
Token ID | 679 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
blown | +0.017 |
arrest | +0.017 |
spilled | +0.016 |
robbers | +0.016 |
Blood | +0.016 |
Neg logit contributions | |
Deal | -0.018 |
difference | -0.016 |
isn | -0.016 |
why | -0.016 |
ement | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
arrest | +0.018 |
alks | +0.018 |
robbers | +0.017 |
growth | +0.017 |
Blood | +0.017 |
Neg logit contributions | |
Deal | -0.008 |
isn | -0.006 |
why | -0.006 |
exactly | -0.005 |
difference | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
arrest | +0.016 |
growth | +0.016 |
Blood | +0.015 |
robbers | +0.015 |
alks | +0.015 |
Neg logit contributions | |
Deal | -0.009 |
exactly | -0.007 |
why | -0.007 |
difference | -0.007 |
isn | -0.006 |
Token | very |
Token ID | 845 |
Feature activation | +9.19e-02 |
Pos logit contributions | |
Blood | +0.009 |
arrest | +0.009 |
robbers | +0.009 |
blown | +0.009 |
mice | +0.008 |
Neg logit contributions | |
Deal | -0.009 |
difference | -0.008 |
ement | -0.008 |
isn | -0.008 |
why | -0.008 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.53e+00 |
Pos logit contributions | |
Deal | +0.002 |
difference | +0.002 |
isn | +0.002 |
ement | +0.002 |
promise | +0.002 |
Neg logit contributions | |
arrest | -0.003 |
Blood | -0.003 |
growth | -0.003 |
blown | -0.003 |
robbers | -0.002 |
Token | cy |
Token ID | 948 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Deal | +0.161 |
difference | +0.144 |
ement | +0.138 |
why | +0.138 |
isn | +0.138 |
Neg logit contributions | |
Blood | -0.117 |
arrest | -0.116 |
robbers | -0.113 |
blown | -0.111 |
growth | -0.110 |
Token | bunny |
Token ID | 44915 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
Blood | +0.024 |
arrest | +0.024 |
robbers | +0.023 |
blown | +0.023 |
growth | +0.022 |
Neg logit contributions | |
Deal | -0.049 |
difference | -0.045 |
ement | -0.043 |
why | -0.043 |
isn | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
Deal | +0.020 |
difference | +0.018 |
ement | +0.017 |
decision | +0.017 |
isn | +0.017 |
Neg logit contributions | |
arrest | -0.016 |
mice | -0.016 |
Blood | -0.016 |
robbers | -0.016 |
blown | -0.016 |
Token | all |
Token ID | 477 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
arrest | +0.015 |
Blood | +0.015 |
blown | +0.014 |
robbers | +0.014 |
growth | +0.014 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.009 |
isn | -0.009 |
why | -0.009 |
exactly | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
Blood | +0.017 |
arrest | +0.016 |
blown | +0.016 |
robbers | +0.016 |
spilled | +0.016 |
Neg logit contributions | |
Deal | -0.015 |
difference | -0.014 |
ement | -0.013 |
decision | -0.013 |
exactly | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
Deal | +0.010 |
difference | +0.009 |
decision | +0.008 |
promise | +0.008 |
ement | +0.008 |
Neg logit contributions | |
Blood | -0.012 |
robbers | -0.011 |
arrest | -0.011 |
spilled | -0.011 |
mice | -0.010 |
Token | there |
Token ID | 612 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
Deal | +0.033 |
difference | +0.030 |
ement | +0.028 |
decision | +0.028 |
isn | +0.028 |
Neg logit contributions | |
Blood | -0.027 |
arrest | -0.026 |
robbers | -0.025 |
blown | -0.025 |
growth | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
Deal | +0.004 |
difference | +0.003 |
izes | +0.003 |
int | +0.003 |
ieved | +0.003 |
Neg logit contributions | |
spilled | -0.003 |
Blood | -0.003 |
growth | -0.003 |
ges | -0.002 |
blown | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
alks | +0.050 |
arrest | +0.049 |
Blood | +0.048 |
robbers | +0.048 |
plets | +0.048 |
Neg logit contributions | |
Deal | -0.048 |
exactly | -0.042 |
why | -0.040 |
equally | -0.040 |
difference | -0.040 |
Token | small |
Token ID | 1402 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
Deal | +0.009 |
difference | +0.008 |
ement | +0.008 |
isn | +0.008 |
why | +0.008 |
Neg logit contributions | |
Blood | -0.005 |
arrest | -0.005 |
robbers | -0.005 |
blown | -0.005 |
growth | -0.005 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.52e+00 |
Pos logit contributions | |
Deal | +0.016 |
difference | +0.015 |
ement | +0.014 |
isn | +0.014 |
why | +0.014 |
Neg logit contributions | |
Blood | -0.010 |
arrest | -0.010 |
blown | -0.010 |
growth | -0.010 |
robbers | -0.010 |
Token | cy |
Token ID | 948 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
Deal | +0.160 |
difference | +0.143 |
ement | +0.137 |
isn | +0.136 |
why | +0.136 |
Neg logit contributions | |
Blood | -0.118 |
arrest | -0.117 |
robbers | -0.114 |
blown | -0.113 |
growth | -0.111 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
Blood | +0.008 |
arrest | +0.007 |
robbers | +0.007 |
blown | +0.007 |
growth | +0.007 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.010 |
ement | -0.010 |
isn | -0.010 |
why | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
Deal | +0.046 |
difference | +0.040 |
ement | +0.038 |
isn | +0.038 |
decision | +0.037 |
Neg logit contributions | |
Blood | -0.044 |
arrest | -0.043 |
robbers | -0.043 |
blown | -0.042 |
growth | -0.042 |
Token | It |
Token ID | 632 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Blood | +0.018 |
arrest | +0.018 |
blown | +0.018 |
robbers | +0.017 |
growth | +0.017 |
Neg logit contributions | |
Deal | -0.023 |
difference | -0.021 |
isn | -0.020 |
ement | -0.020 |
why | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
Blood | +0.036 |
arrest | +0.035 |
blown | +0.034 |
robbers | +0.034 |
growth | +0.034 |
Neg logit contributions | |
Deal | -0.026 |
difference | -0.022 |
ement | -0.021 |
decision | -0.021 |
exactly | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
Deal | +0.071 |
difference | +0.065 |
ement | +0.061 |
decision | +0.058 |
ieved | +0.057 |
Neg logit contributions | |
robbers | -0.048 |
spilled | -0.046 |
Blood | -0.046 |
blown | -0.045 |
arrest | -0.044 |
Token | knew |
Token ID | 2993 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
Deal | +0.019 |
difference | +0.019 |
ement | +0.017 |
decision | +0.016 |
int | +0.016 |
Neg logit contributions | |
spilled | -0.017 |
Blood | -0.016 |
blown | -0.016 |
arrest | -0.016 |
robbers | -0.015 |
Token | his |
Token ID | 465 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
Deal | +0.025 |
difference | +0.022 |
ement | +0.021 |
promise | +0.021 |
decision | +0.020 |
Neg logit contributions | |
arrest | -0.021 |
Blood | -0.021 |
robbers | -0.021 |
blown | -0.020 |
spilled | -0.020 |
Token | mom |
Token ID | 1995 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
Deal | +0.084 |
ieved | +0.076 |
uct | +0.073 |
ement | +0.073 |
izes | +0.069 |
Neg logit contributions | |
robbers | -0.050 |
arrest | -0.048 |
spilled | -0.048 |
mice | -0.047 |
blown | -0.046 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
spilled | +0.013 |
blown | +0.012 |
robbers | +0.012 |
mice | +0.012 |
Blood | +0.012 |
Neg logit contributions | |
Deal | -0.013 |
difference | -0.012 |
ement | -0.011 |
decision | -0.011 |
int | -0.011 |
Token | â |
Token ID | 22940 |
Feature activation | +5.52e+00 |
Pos logit contributions | |
Deal | +0.011 |
difference | +0.010 |
isn | +0.010 |
ement | +0.010 |
why | +0.010 |
Neg logit contributions | |
robbers | -0.008 |
Blood | -0.008 |
mice | -0.007 |
arrest | -0.007 |
growth | -0.007 |
Token | € |
Token ID | 26391 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
Deal | +0.098 |
difference | +0.096 |
exactly | +0.096 |
ement | +0.089 |
isn | +0.088 |
Neg logit contributions | |
Blood | -0.163 |
arrest | -0.163 |
robbers | -0.159 |
blown | -0.157 |
growth | -0.154 |
Token | ™ |
Token ID | 8151 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
Deal | +0.001 |
difference | +0.001 |
ement | +0.001 |
promise | +0.001 |
decision | +0.001 |
Neg logit contributions | |
arrest | -0.001 |
Blood | -0.001 |
robbers | -0.001 |
blown | -0.001 |
growth | -0.001 |
Token | t |
Token ID | 83 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
Deal | +0.032 |
difference | +0.029 |
ement | +0.029 |
isn | +0.028 |
why | +0.028 |
Neg logit contributions | |
Blood | -0.015 |
arrest | -0.015 |
robbers | -0.014 |
blown | -0.014 |
growth | -0.014 |
Token | let |
Token ID | 1309 |
Feature activation | -7.86e-01 |
Pos logit contributions | |
Blood | +0.008 |
arrest | +0.008 |
robbers | +0.008 |
blown | +0.008 |
spilled | +0.007 |
Neg logit contributions | |
Deal | -0.010 |
ement | -0.009 |
difference | -0.008 |
isn | -0.008 |
exactly | -0.008 |
Token | him |
Token ID | 683 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
arrest | +0.019 |
Blood | +0.018 |
robbers | +0.018 |
blown | +0.018 |
spilled | +0.017 |
Neg logit contributions | |
Deal | -0.022 |
ement | -0.019 |
difference | -0.018 |
ieved | -0.017 |
exactly | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.012 |
ement | +0.012 |
ieved | +0.011 |
decision | +0.011 |
Neg logit contributions | |
arrest | -0.016 |
Blood | -0.016 |
spilled | -0.016 |
blown | -0.016 |
robbers | -0.015 |
Token | â |
Token ID | 22940 |
Feature activation | +4.87e+00 |
Pos logit contributions | |
Deal | +0.021 |
difference | +0.019 |
ement | +0.019 |
decision | +0.017 |
ieved | +0.017 |
Neg logit contributions | |
arrest | -0.016 |
robbers | -0.015 |
spilled | -0.015 |
Blood | -0.015 |
blown | -0.015 |
Token | € |
Token ID | 26391 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
exactly | +0.094 |
difference | +0.091 |
Deal | +0.085 |
why | +0.083 |
amaz | +0.083 |
Neg logit contributions | |
Blood | -0.138 |
arrest | -0.133 |
plets | -0.131 |
blown | -0.128 |
alks | -0.127 |
Token | said |
Token ID | 531 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
Blood | +0.020 |
arrest | +0.019 |
alks | +0.018 |
robbers | +0.018 |
blown | +0.018 |
Neg logit contributions | |
Deal | -0.021 |
difference | -0.019 |
ement | -0.018 |
decision | -0.018 |
why | -0.017 |
Token | Sarah |
Token ID | 10490 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Deal | +0.020 |
difference | +0.018 |
ement | +0.017 |
decision | +0.017 |
ieved | +0.016 |
Neg logit contributions | |
Blood | -0.018 |
arrest | -0.017 |
spilled | -0.017 |
blown | -0.016 |
robbers | -0.016 |
Token | â |
Token ID | 22940 |
Feature activation | +5.47e+00 |
Pos logit contributions | |
Deal | +0.039 |
difference | +0.036 |
ement | +0.033 |
decision | +0.032 |
deal | +0.032 |
Neg logit contributions | |
Blood | -0.039 |
spilled | -0.038 |
arrest | -0.038 |
blown | -0.037 |
robbers | -0.037 |
Token | € |
Token ID | 26391 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
exactly | +0.110 |
difference | +0.105 |
why | +0.097 |
amaz | +0.097 |
Deal | +0.094 |
Neg logit contributions | |
Blood | -0.149 |
plets | -0.140 |
arrest | -0.140 |
robbers | -0.139 |
alks | -0.137 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
Blood | +0.006 |
arrest | +0.006 |
robbers | +0.005 |
alks | +0.005 |
plets | +0.005 |
Neg logit contributions | |
Deal | -0.007 |
difference | -0.007 |
ement | -0.006 |
decision | -0.006 |
exactly | -0.006 |
Token | s |
Token ID | 82 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
Deal | +0.052 |
difference | +0.049 |
decision | +0.046 |
why | +0.046 |
promise | +0.046 |
Neg logit contributions | |
Blood | -0.025 |
arrest | -0.023 |
spilled | -0.022 |
alks | -0.022 |
blown | -0.022 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Deal | +0.070 |
difference | +0.058 |
ement | +0.057 |
ieved | +0.056 |
uct | +0.053 |
Neg logit contributions | |
Blood | -0.072 |
arrest | -0.069 |
alks | -0.069 |
spilled | -0.067 |
robbers | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +7.74e-01 |
Pos logit contributions | |
spilled | +0.004 |
robbers | +0.004 |
Blood | +0.004 |
arrest | +0.004 |
mice | +0.004 |
Neg logit contributions | |
Deal | -0.005 |
difference | -0.004 |
ement | -0.004 |
decision | -0.004 |
int | -0.004 |
Token | Dad |
Token ID | 17415 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
arrest | +0.005 |
Blood | +0.005 |
robbers | +0.005 |
blown | +0.005 |
growth | +0.005 |
Neg logit contributions | |
Deal | -0.006 |
difference | -0.006 |
isn | -0.005 |
ement | -0.005 |
why | -0.005 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
Deal | +0.026 |
difference | +0.024 |
decision | +0.022 |
ement | +0.022 |
exactly | +0.021 |
Neg logit contributions | |
spilled | -0.018 |
robbers | -0.018 |
Blood | -0.018 |
blown | -0.018 |
arrest | -0.018 |
Token | it |
Token ID | 340 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
Deal | +0.007 |
difference | +0.006 |
ement | +0.006 |
isn | +0.006 |
decision | +0.005 |
Neg logit contributions | |
Blood | -0.006 |
arrest | -0.005 |
robbers | -0.005 |
blown | -0.005 |
growth | -0.005 |
Token | first |
Token ID | 717 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Deal | +0.023 |
difference | +0.020 |
ement | +0.019 |
isn | +0.019 |
decision | +0.018 |
Neg logit contributions | |
blown | -0.012 |
robbers | -0.012 |
arrest | -0.012 |
spilled | -0.011 |
mice | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
Deal | +0.040 |
difference | +0.035 |
ement | +0.032 |
ieved | +0.032 |
uct | +0.032 |
Neg logit contributions | |
spilled | -0.023 |
arrest | -0.022 |
blown | -0.022 |
Blood | -0.022 |
alks | -0.021 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.46e+00 |
Pos logit contributions | |
Deal | +0.023 |
difference | +0.021 |
ement | +0.020 |
decision | +0.020 |
isn | +0.019 |
Neg logit contributions | |
blown | -0.016 |
arrest | -0.015 |
robbers | -0.015 |
Blood | -0.015 |
spilled | -0.015 |
Token | ed |
Token ID | 276 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
Deal | +0.179 |
difference | +0.168 |
decision | +0.157 |
why | +0.156 |
exactly | +0.154 |
Neg logit contributions | |
Blood | -0.094 |
arrest | -0.091 |
growth | -0.088 |
spilled | -0.087 |
ges | -0.087 |
Token | it |
Token ID | 340 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
Deal | +0.026 |
difference | +0.023 |
decision | +0.021 |
promise | +0.021 |
ement | +0.021 |
Neg logit contributions | |
arrest | -0.017 |
mice | -0.016 |
ges | -0.016 |
Blood | -0.016 |
growth | -0.016 |
Token | around |
Token ID | 1088 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
Deal | +0.017 |
difference | +0.015 |
ement | +0.014 |
decision | +0.014 |
promise | +0.014 |
Neg logit contributions | |
blown | -0.010 |
spilled | -0.010 |
Blood | -0.010 |
arrest | -0.009 |
robbers | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
Deal | +0.051 |
difference | +0.045 |
ement | +0.043 |
decision | +0.042 |
isn | +0.042 |
Neg logit contributions | |
Blood | -0.041 |
arrest | -0.040 |
robbers | -0.039 |
blown | -0.039 |
growth | -0.038 |
Token | room |
Token ID | 2119 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
Blood | +0.011 |
arrest | +0.011 |
robbers | +0.011 |
blown | +0.011 |
growth | +0.011 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.010 |
ement | -0.009 |
isn | -0.009 |
why | -0.009 |
Token | then |
Token ID | 788 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
Deal | +0.035 |
difference | +0.032 |
ement | +0.030 |
decision | +0.029 |
isn | +0.029 |
Neg logit contributions | |
Blood | -0.043 |
arrest | -0.043 |
robbers | -0.043 |
mice | -0.042 |
blown | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Blood | +0.004 |
arrest | +0.004 |
robbers | +0.004 |
blown | +0.004 |
growth | +0.004 |
Neg logit contributions | |
Deal | -0.004 |
difference | -0.004 |
ement | -0.003 |
decision | -0.003 |
isn | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Deal | +0.038 |
difference | +0.034 |
why | +0.033 |
exactly | +0.033 |
isn | +0.033 |
Neg logit contributions | |
Blood | -0.029 |
arrest | -0.028 |
blown | -0.027 |
robbers | -0.027 |
alks | -0.027 |
Token | big |
Token ID | 1263 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
Deal | +0.031 |
difference | +0.028 |
ement | +0.026 |
isn | +0.026 |
why | +0.026 |
Neg logit contributions | |
Blood | -0.024 |
arrest | -0.024 |
blown | -0.023 |
robbers | -0.023 |
growth | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
Deal | +0.071 |
difference | +0.064 |
ement | +0.062 |
isn | +0.061 |
why | +0.061 |
Neg logit contributions | |
Blood | -0.048 |
arrest | -0.047 |
robbers | -0.046 |
blown | -0.046 |
growth | -0.045 |
Token | boun |
Token ID | 31283 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
Deal | +0.008 |
difference | +0.008 |
isn | +0.007 |
promise | +0.007 |
why | +0.007 |
Neg logit contributions | |
arrest | -0.008 |
Blood | -0.008 |
growth | -0.008 |
alks | -0.007 |
blown | -0.007 |
Token | cy |
Token ID | 948 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
Deal | +0.153 |
difference | +0.137 |
ement | +0.131 |
isn | +0.130 |
why | +0.129 |
Neg logit contributions | |
Blood | -0.121 |
arrest | -0.120 |
robbers | -0.117 |
blown | -0.116 |
growth | -0.114 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Blood | +0.006 |
arrest | +0.006 |
robbers | +0.006 |
blown | +0.006 |
growth | +0.006 |
Neg logit contributions | |
Deal | -0.009 |
difference | -0.008 |
ement | -0.008 |
isn | -0.008 |
decision | -0.008 |
Token | floated |
Token ID | 29742 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
Deal | +0.028 |
difference | +0.024 |
isn | +0.024 |
why | +0.024 |
exactly | +0.024 |
Neg logit contributions | |
Blood | -0.027 |
growth | -0.026 |
alks | -0.026 |
arrest | -0.026 |
robbers | -0.025 |
Token | by |
Token ID | 416 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.003 |
ement | +0.003 |
decision | +0.003 |
promise | +0.003 |
Neg logit contributions | |
Blood | -0.003 |
blown | -0.003 |
robbers | -0.003 |
arrest | -0.003 |
spilled | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Deal | +0.022 |
difference | +0.019 |
ement | +0.018 |
isn | +0.018 |
why | +0.018 |
Neg logit contributions | |
Blood | -0.023 |
arrest | -0.023 |
robbers | -0.022 |
blown | -0.022 |
growth | -0.022 |
Token | about |
Token ID | 546 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Blood | +0.015 |
blown | +0.014 |
arrest | +0.014 |
spilled | +0.014 |
robbers | +0.014 |
Neg logit contributions | |
Deal | -0.013 |
difference | -0.011 |
promise | -0.010 |
decision | -0.010 |
ement | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
Deal | +0.031 |
ement | +0.025 |
difference | +0.025 |
ieved | +0.024 |
promise | +0.024 |
Neg logit contributions | |
arrest | -0.039 |
robbers | -0.038 |
Blood | -0.038 |
mice | -0.037 |
blown | -0.036 |
Token | but |
Token ID | 475 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
Deal | +0.005 |
difference | +0.004 |
ement | +0.004 |
decision | +0.004 |
aman | +0.004 |
Neg logit contributions | |
robbers | -0.003 |
mice | -0.003 |
Blood | -0.003 |
spilled | -0.003 |
blown | -0.003 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
Deal | +0.104 |
difference | +0.099 |
ement | +0.096 |
decision | +0.089 |
promise | +0.088 |
Neg logit contributions | |
Blood | -0.061 |
robbers | -0.061 |
blown | -0.058 |
spilled | -0.057 |
arrest | -0.056 |
Token | didn |
Token ID | 1422 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
difference | +0.054 |
Deal | +0.052 |
ement | +0.048 |
int | +0.046 |
amaz | +0.046 |
Neg logit contributions | |
spilled | -0.054 |
Blood | -0.051 |
blown | -0.050 |
bitten | -0.049 |
robbers | -0.048 |
Token | â |
Token ID | 22940 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
Blood | +0.011 |
arrest | +0.011 |
robbers | +0.010 |
blown | +0.010 |
growth | +0.010 |
Neg logit contributions | |
Deal | -0.022 |
difference | -0.020 |
ement | -0.020 |
isn | -0.020 |
why | -0.020 |
Token | € |
Token ID | 26391 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Deal | +0.087 |
difference | +0.080 |
exactly | +0.075 |
ement | +0.073 |
why | +0.072 |
Neg logit contributions | |
Blood | -0.180 |
arrest | -0.176 |
robbers | -0.172 |
blown | -0.171 |
growth | -0.171 |
Token | ™ |
Token ID | 8151 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.003 |
ement | +0.003 |
decision | +0.003 |
promise | +0.003 |
Neg logit contributions | |
Blood | -0.002 |
arrest | -0.002 |
robbers | -0.002 |
blown | -0.002 |
alks | -0.002 |
Token | t |
Token ID | 83 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
Deal | +0.036 |
difference | +0.033 |
ement | +0.032 |
isn | +0.032 |
why | +0.032 |
Neg logit contributions | |
Blood | -0.016 |
arrest | -0.016 |
robbers | -0.015 |
blown | -0.015 |
growth | -0.015 |
Token | listen |
Token ID | 6004 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
Deal | +0.010 |
ement | +0.009 |
difference | +0.008 |
isn | +0.008 |
exactly | +0.008 |
Neg logit contributions | |
Blood | -0.010 |
robbers | -0.010 |
arrest | -0.010 |
blown | -0.010 |
spilled | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
Deal | +0.070 |
difference | +0.064 |
ement | +0.058 |
isn | +0.058 |
ieved | +0.057 |
Neg logit contributions | |
Blood | -0.032 |
spilled | -0.031 |
blown | -0.030 |
arrest | -0.029 |
robbers | -0.029 |
Token | door |
Token ID | 3420 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
Deal | +0.013 |
difference | +0.011 |
ieved | +0.011 |
why | +0.011 |
ement | +0.011 |
Neg logit contributions | |
arrest | -0.006 |
Blood | -0.006 |
spilled | -0.006 |
robbers | -0.006 |
blown | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
Blood | +0.007 |
blown | +0.007 |
arrest | +0.007 |
alks | +0.007 |
growth | +0.007 |
Neg logit contributions | |
Deal | -0.012 |
difference | -0.011 |
deal | -0.010 |
promise | -0.010 |
isn | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
Deal | +0.011 |
difference | +0.010 |
ement | +0.009 |
isn | +0.009 |
why | +0.009 |
Neg logit contributions | |
Blood | -0.010 |
arrest | -0.010 |
robbers | -0.010 |
blown | -0.010 |
growth | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
Deal | +0.008 |
difference | +0.007 |
ement | +0.007 |
isn | +0.007 |
decision | +0.007 |
Neg logit contributions | |
Blood | -0.006 |
arrest | -0.006 |
robbers | -0.006 |
blown | -0.006 |
growth | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Deal | +0.037 |
difference | +0.033 |
ement | +0.032 |
isn | +0.032 |
why | +0.031 |
Neg logit contributions | |
Blood | -0.030 |
arrest | -0.029 |
robbers | -0.028 |
blown | -0.028 |
growth | -0.028 |
Token | amaz |
Token ID | 40642 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
Deal | +0.028 |
ieved | +0.021 |
uct | +0.020 |
izes | +0.020 |
isn | +0.019 |
Neg logit contributions | |
growth | -0.028 |
arrest | -0.028 |
mice | -0.026 |
robbers | -0.026 |
alks | -0.026 |
Token | ement |
Token ID | 972 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
Deal | +0.066 |
difference | +0.042 |
promise | +0.039 |
isn | +0.034 |
why | +0.033 |
Neg logit contributions | |
Blood | -0.207 |
blown | -0.203 |
growth | -0.203 |
arrest | -0.201 |
alks | -0.198 |
Token | , |
Token ID | 11 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
Deal | +0.040 |
difference | +0.034 |
isn | +0.032 |
ement | +0.031 |
why | +0.031 |
Neg logit contributions | |
Blood | -0.037 |
blown | -0.037 |
arrest | -0.036 |
alks | -0.036 |
spilled | -0.035 |
Token | she |
Token ID | 673 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
Deal | +0.010 |
difference | +0.009 |
ement | +0.009 |
decision | +0.008 |
promise | +0.008 |
Neg logit contributions | |
Blood | -0.006 |
blown | -0.006 |
arrest | -0.006 |
robbers | -0.006 |
spilled | -0.006 |
Token | could |
Token ID | 714 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Deal | +0.012 |
difference | +0.011 |
decision | +0.010 |
ement | +0.010 |
promise | +0.010 |
Neg logit contributions | |
Blood | -0.007 |
blown | -0.007 |
arrest | -0.007 |
spilled | -0.007 |
growth | -0.007 |
Token | see |
Token ID | 766 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
arrest | +0.024 |
Blood | +0.023 |
blown | +0.022 |
growth | +0.022 |
robbers | +0.022 |
Neg logit contributions | |
Deal | -0.043 |
difference | -0.039 |
ement | -0.037 |
isn | -0.037 |
decision | -0.036 |
Token | park |
Token ID | 3952 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Deal | +0.020 |
ement | +0.019 |
isn | +0.019 |
difference | +0.019 |
why | +0.018 |
Neg logit contributions | |
bushes | -0.012 |
arrest | -0.012 |
robbers | -0.012 |
spilled | -0.011 |
blown | -0.011 |
Token | they |
Token ID | 484 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
robbers | +0.003 |
Blood | +0.003 |
arrest | +0.003 |
spilled | +0.003 |
blown | +0.003 |
Neg logit contributions | |
Deal | -0.003 |
difference | -0.003 |
ement | -0.003 |
ieved | -0.003 |
uct | -0.003 |
Token | found |
Token ID | 1043 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
arrest | +0.006 |
Blood | +0.006 |
robbers | +0.006 |
blown | +0.006 |
spilled | +0.006 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.010 |
ement | -0.010 |
decision | -0.010 |
isn | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
Deal | +0.012 |
difference | +0.011 |
decision | +0.010 |
promise | +0.010 |
ement | +0.010 |
Neg logit contributions | |
blown | -0.005 |
spilled | -0.005 |
arrest | -0.005 |
mice | -0.005 |
growth | -0.005 |
Token | biggest |
Token ID | 4094 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
robbers | +0.004 |
blown | +0.004 |
bushes | +0.004 |
arrest | +0.004 |
spilled | +0.004 |
Neg logit contributions | |
Deal | -0.005 |
ement | -0.004 |
why | -0.004 |
difference | -0.004 |
isn | -0.004 |
Token | ear |
Token ID | 1027 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
arrest | +0.019 |
robbers | +0.019 |
spilled | +0.019 |
blown | +0.018 |
bushes | +0.018 |
Neg logit contributions | |
Deal | -0.025 |
uct | -0.022 |
ieved | -0.021 |
int | -0.021 |
why | -0.021 |
Token | Tom |
Token ID | 4186 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
Blood | +0.139 |
arrest | +0.137 |
robbers | +0.135 |
blown | +0.132 |
growth | +0.132 |
Neg logit contributions | |
Deal | -0.158 |
difference | -0.139 |
ement | -0.135 |
isn | -0.134 |
exactly | -0.133 |
Token | had |
Token ID | 550 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
Blood | +0.049 |
arrest | +0.048 |
robbers | +0.047 |
blown | +0.047 |
growth | +0.046 |
Neg logit contributions | |
Deal | -0.078 |
difference | -0.070 |
ement | -0.068 |
isn | -0.067 |
why | -0.067 |
Token | ever |
Token ID | 1683 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
Blood | +0.084 |
arrest | +0.080 |
robbers | +0.080 |
growth | +0.078 |
blown | +0.078 |
Neg logit contributions | |
Deal | -0.111 |
difference | -0.099 |
ement | -0.096 |
isn | -0.096 |
exactly | -0.096 |
Token | seen |
Token ID | 1775 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
Blood | +0.072 |
arrest | +0.065 |
alks | +0.065 |
growth | +0.064 |
robbers | +0.064 |
Neg logit contributions | |
Deal | -0.100 |
difference | -0.092 |
why | -0.091 |
promise | -0.091 |
isn | -0.090 |
Token | ! |
Token ID | 0 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
arrest | +0.016 |
blown | +0.016 |
mice | +0.016 |
spilled | +0.016 |
robbers | +0.015 |
Neg logit contributions | |
Deal | -0.024 |
difference | -0.021 |
promise | -0.020 |
decision | -0.020 |
deal | -0.019 |
Token | always |
Token ID | 1464 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.003 |
ement | +0.002 |
why | +0.002 |
exactly | +0.002 |
Neg logit contributions | |
Blood | -0.005 |
arrest | -0.005 |
blown | -0.005 |
robbers | -0.005 |
growth | -0.005 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
blown | +0.006 |
spilled | +0.006 |
Blood | +0.006 |
robbers | +0.005 |
arrest | +0.005 |
Neg logit contributions | |
Deal | -0.008 |
difference | -0.007 |
ement | -0.007 |
decision | -0.006 |
int | -0.006 |
Token | like |
Token ID | 588 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
spilled | +0.004 |
blown | +0.004 |
Blood | +0.004 |
robbers | +0.004 |
mice | +0.004 |
Neg logit contributions | |
Deal | -0.005 |
difference | -0.004 |
ement | -0.004 |
decision | -0.004 |
promise | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -9.42e-01 |
Pos logit contributions | |
Blood | +0.021 |
arrest | +0.021 |
robbers | +0.020 |
growth | +0.020 |
alks | +0.020 |
Neg logit contributions | |
Deal | -0.027 |
difference | -0.024 |
why | -0.024 |
isn | -0.024 |
exactly | -0.024 |
Token | tornado |
Token ID | 33718 |
Feature activation | +9.49e-01 |
Pos logit contributions | |
Blood | +0.019 |
robbers | +0.019 |
arrest | +0.019 |
spilled | +0.018 |
mice | +0.018 |
Neg logit contributions | |
Deal | -0.029 |
ement | -0.026 |
difference | -0.025 |
ieved | -0.025 |
why | -0.024 |
Token | had |
Token ID | 550 |
Feature activation | -5.80e+00 |
Pos logit contributions | |
Deal | +0.025 |
difference | +0.021 |
ement | +0.021 |
decision | +0.020 |
exactly | +0.020 |
Neg logit contributions | |
arrest | -0.024 |
blown | -0.024 |
spilled | -0.024 |
Blood | -0.023 |
robbers | -0.023 |
Token | gone |
Token ID | 3750 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
robbers | +0.136 |
Blood | +0.134 |
alks | +0.131 |
mice | +0.124 |
growth | +0.123 |
Neg logit contributions | |
isn | -0.140 |
exactly | -0.134 |
Deal | -0.131 |
about | -0.128 |
why | -0.127 |
Token | through |
Token ID | 832 |
Feature activation | -8.87e-01 |
Pos logit contributions | |
Blood | +0.039 |
arrest | +0.039 |
blown | +0.039 |
robbers | +0.038 |
spilled | +0.038 |
Neg logit contributions | |
Deal | -0.044 |
difference | -0.038 |
ement | -0.036 |
promise | -0.035 |
decision | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
robbers | +0.024 |
Blood | +0.024 |
alks | +0.023 |
plets | +0.023 |
blown | +0.022 |
Neg logit contributions | |
difference | -0.017 |
isn | -0.017 |
why | -0.017 |
Deal | -0.017 |
ement | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
Blood | +0.008 |
arrest | +0.008 |
blown | +0.008 |
spilled | +0.008 |
robbers | +0.008 |
Neg logit contributions | |
Deal | -0.010 |
difference | -0.008 |
ement | -0.008 |
decision | -0.008 |
why | -0.008 |
Token | One |
Token ID | 1881 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
Deal | +0.002 |
difference | +0.002 |
ement | +0.002 |
isn | +0.002 |
why | +0.002 |
Neg logit contributions | |
Blood | -0.002 |
arrest | -0.002 |
blown | -0.002 |
robbers | -0.002 |
growth | -0.002 |
Token | I |
Token ID | 40 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
difference | +0.041 |
Deal | +0.041 |
ement | +0.039 |
decision | +0.039 |
exactly | +0.038 |
Neg logit contributions | |
Blood | -0.026 |
growth | -0.023 |
robbers | -0.023 |
arrest | -0.023 |
blown | -0.022 |
Token | like |
Token ID | 588 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
Deal | +0.030 |
ement | +0.029 |
uct | +0.027 |
difference | +0.026 |
ieved | +0.025 |
Neg logit contributions | |
spilled | -0.025 |
Blood | -0.023 |
blown | -0.023 |
arrest | -0.022 |
bitten | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Blood | +0.000 |
arrest | +0.000 |
robbers | +0.000 |
blown | +0.000 |
growth | +0.000 |
Neg logit contributions | |
Deal | -0.000 |
difference | -0.000 |
ement | -0.000 |
decision | -0.000 |
exactly | -0.000 |
Token | w |
Token ID | 266 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Blood | +0.032 |
blown | +0.032 |
arrest | +0.031 |
robbers | +0.031 |
mice | +0.030 |
Neg logit contributions | |
Deal | -0.041 |
difference | -0.036 |
ement | -0.036 |
exactly | -0.035 |
why | -0.035 |
Token | igg |
Token ID | 6950 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
robbers | +0.054 |
blown | +0.053 |
Blood | +0.053 |
arrest | +0.053 |
growth | +0.052 |
Neg logit contributions | |
Deal | -0.050 |
ement | -0.046 |
isn | -0.044 |
difference | -0.044 |
decision | -0.042 |
Token | ly |
Token ID | 306 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
Blood | +0.021 |
arrest | +0.021 |
blown | +0.020 |
robbers | +0.020 |
alks | +0.020 |
Neg logit contributions | |
Deal | -0.041 |
difference | -0.037 |
ement | -0.035 |
exactly | -0.035 |
why | -0.035 |
Token | dance |
Token ID | 9280 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
Blood | +0.132 |
arrest | +0.130 |
robbers | +0.127 |
blown | +0.126 |
growth | +0.124 |
Neg logit contributions | |
Deal | -0.154 |
difference | -0.137 |
ement | -0.131 |
isn | -0.129 |
why | -0.129 |
Token | the |
Token ID | 262 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
robbers | +0.023 |
Blood | +0.022 |
arrest | +0.022 |
mice | +0.022 |
blown | +0.022 |
Neg logit contributions | |
Deal | -0.022 |
difference | -0.020 |
ement | -0.019 |
exactly | -0.019 |
uct | -0.018 |
Token | most |
Token ID | 749 |
Feature activation | +9.39e-02 |
Pos logit contributions | |
Blood | +0.029 |
blown | +0.029 |
robbers | +0.029 |
arrest | +0.029 |
spilled | +0.028 |
Neg logit contributions | |
Deal | -0.034 |
ement | -0.030 |
difference | -0.030 |
exactly | -0.030 |
why | -0.029 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
Deal | +0.003 |
aman | +0.002 |
ement | +0.002 |
difference | +0.002 |
ieved | +0.002 |
Neg logit contributions | |
blown | -0.002 |
mice | -0.002 |
spilled | -0.002 |
arrest | -0.002 |
queens | -0.002 |
Token | So |
Token ID | 1406 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
Blood | +0.007 |
robbers | +0.006 |
arrest | +0.006 |
alks | +0.006 |
spilled | +0.006 |
Neg logit contributions | |
Deal | -0.008 |
difference | -0.007 |
ement | -0.007 |
decision | -0.007 |
exactly | -0.007 |
Token | It |
Token ID | 632 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
Blood | +0.027 |
growth | +0.025 |
arrest | +0.024 |
alks | +0.024 |
robbers | +0.024 |
Neg logit contributions | |
Deal | -0.033 |
difference | -0.029 |
decision | -0.029 |
promise | -0.028 |
exactly | -0.028 |
Token | was |
Token ID | 373 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
Blood | +0.041 |
blown | +0.039 |
spilled | +0.039 |
arrest | +0.039 |
robbers | +0.039 |
Neg logit contributions | |
Deal | -0.029 |
difference | -0.025 |
ement | -0.023 |
decision | -0.023 |
exactly | -0.022 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Blood | +0.054 |
blown | +0.052 |
arrest | +0.051 |
robbers | +0.050 |
spilled | +0.049 |
Neg logit contributions | |
Deal | -0.094 |
difference | -0.084 |
ement | -0.082 |
decision | -0.080 |
isn | -0.079 |
Token | pink |
Token ID | 11398 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
Blood | +0.026 |
arrest | +0.026 |
blown | +0.025 |
robbers | +0.025 |
growth | +0.025 |
Neg logit contributions | |
Deal | -0.035 |
difference | -0.031 |
ement | -0.029 |
decision | -0.029 |
why | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
Blood | +0.034 |
arrest | +0.033 |
robbers | +0.033 |
blown | +0.033 |
growth | +0.032 |
Neg logit contributions | |
Deal | -0.030 |
difference | -0.026 |
ement | -0.025 |
isn | -0.025 |
exactly | -0.025 |
Token | had |
Token ID | 550 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
Blood | +0.035 |
blown | +0.033 |
robbers | +0.033 |
spilled | +0.032 |
mice | +0.031 |
Neg logit contributions | |
Deal | -0.057 |
difference | -0.054 |
decision | -0.051 |
ement | -0.051 |
promise | -0.050 |
Token | lots |
Token ID | 6041 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
robbers | +0.149 |
Blood | +0.144 |
arrest | +0.143 |
alks | +0.142 |
plets | +0.136 |
Neg logit contributions | |
isn | -0.118 |
exactly | -0.113 |
Deal | -0.106 |
why | -0.104 |
hairst | -0.103 |
Token | of |
Token ID | 286 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
Blood | +0.067 |
arrest | +0.066 |
robbers | +0.064 |
blown | +0.063 |
growth | +0.063 |
Neg logit contributions | |
Deal | -0.083 |
difference | -0.076 |
ement | -0.074 |
isn | -0.073 |
why | -0.072 |
Token | spark |
Token ID | 9009 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Blood | +0.083 |
arrest | +0.082 |
robbers | +0.080 |
blown | +0.079 |
growth | +0.078 |
Neg logit contributions | |
Deal | -0.109 |
difference | -0.097 |
ement | -0.094 |
isn | -0.094 |
why | -0.092 |
Token | les |
Token ID | 829 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
Blood | +0.028 |
arrest | +0.027 |
robbers | +0.027 |
growth | +0.026 |
blown | +0.025 |
Neg logit contributions | |
Deal | -0.028 |
ement | -0.027 |
isn | -0.026 |
difference | -0.025 |
exactly | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
Blood | +0.062 |
arrest | +0.059 |
robbers | +0.059 |
growth | +0.058 |
blown | +0.057 |
Neg logit contributions | |
Deal | -0.051 |
difference | -0.046 |
ement | -0.045 |
exactly | -0.045 |
isn | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
Blood | +0.001 |
arrest | +0.001 |
robbers | +0.001 |
growth | +0.001 |
blown | +0.001 |
Neg logit contributions | |
Deal | -0.001 |
difference | -0.001 |
decision | -0.001 |
ement | -0.001 |
promise | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Blood | +0.017 |
growth | +0.016 |
arrest | +0.015 |
alks | +0.015 |
blown | +0.015 |
Neg logit contributions | |
Deal | -0.025 |
difference | -0.022 |
decision | -0.022 |
ement | -0.021 |
promise | -0.021 |
Token | Finally |
Token ID | 11158 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
Blood | +0.027 |
arrest | +0.026 |
robbers | +0.025 |
growth | +0.025 |
blown | +0.024 |
Neg logit contributions | |
Deal | -0.034 |
difference | -0.032 |
decision | -0.031 |
ement | -0.030 |
isn | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
arrest | +0.012 |
Blood | +0.012 |
robbers | +0.012 |
blown | +0.012 |
growth | +0.012 |
Neg logit contributions | |
Deal | -0.012 |
difference | -0.010 |
ement | -0.009 |
decision | -0.009 |
promise | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
Deal | +0.025 |
difference | +0.022 |
ement | +0.022 |
decision | +0.021 |
isn | +0.021 |
Neg logit contributions | |
arrest | -0.017 |
Blood | -0.017 |
blown | -0.016 |
growth | -0.016 |
robbers | -0.016 |
Token | ear |
Token ID | 1027 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
Blood | +0.014 |
arrest | +0.014 |
robbers | +0.014 |
blown | +0.014 |
growth | +0.013 |
Neg logit contributions | |
Deal | -0.023 |
difference | -0.021 |
ement | -0.020 |
exactly | -0.020 |
why | -0.020 |
Token | had |
Token ID | 550 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Blood | +0.132 |
arrest | +0.130 |
blown | +0.128 |
growth | +0.125 |
robbers | +0.124 |
Neg logit contributions | |
Deal | -0.153 |
difference | -0.135 |
ement | -0.127 |
decision | -0.125 |
isn | -0.124 |
Token | counted |
Token ID | 14789 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
Blood | +0.048 |
robbers | +0.047 |
arrest | +0.046 |
mice | +0.045 |
growth | +0.045 |
Neg logit contributions | |
Deal | -0.059 |
difference | -0.054 |
exactly | -0.053 |
isn | -0.053 |
why | -0.052 |
Token | all |
Token ID | 477 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Blood | +0.012 |
blown | +0.012 |
alks | +0.011 |
arrest | +0.011 |
robbers | +0.011 |
Neg logit contributions | |
Deal | -0.016 |
difference | -0.014 |
promise | -0.013 |
decision | -0.013 |
ement | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
blown | +0.019 |
arrest | +0.018 |
alks | +0.017 |
Blood | +0.017 |
spilled | +0.017 |
Neg logit contributions | |
Deal | -0.035 |
difference | -0.028 |
promise | -0.028 |
deal | -0.028 |
olog | -0.027 |
Token | way |
Token ID | 835 |
Feature activation | -8.89e-01 |
Pos logit contributions | |
robbers | +0.029 |
arrest | +0.028 |
blown | +0.028 |
mice | +0.028 |
Blood | +0.027 |
Neg logit contributions | |
Deal | -0.035 |
ement | -0.029 |
difference | -0.029 |
why | -0.029 |
isn | -0.028 |
Token | was |
Token ID | 373 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Deal | +0.010 |
izes | +0.009 |
difference | +0.009 |
int | +0.009 |
ieved | +0.008 |
Neg logit contributions | |
spilled | -0.006 |
Blood | -0.006 |
ges | -0.005 |
blown | -0.005 |
bitten | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
Blood | +0.030 |
arrest | +0.030 |
robbers | +0.029 |
alks | +0.029 |
blown | +0.029 |
Neg logit contributions | |
Deal | -0.045 |
difference | -0.040 |
exactly | -0.039 |
ement | -0.039 |
why | -0.039 |
Token | ranch |
Token ID | 28752 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
Deal | +0.008 |
difference | +0.007 |
ement | +0.007 |
isn | +0.007 |
why | +0.007 |
Neg logit contributions | |
Blood | -0.004 |
arrest | -0.004 |
robbers | -0.004 |
blown | -0.004 |
growth | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Deal | +0.041 |
difference | +0.037 |
ement | +0.035 |
decision | +0.035 |
isn | +0.035 |
Neg logit contributions | |
Blood | -0.037 |
arrest | -0.037 |
mice | -0.036 |
robbers | -0.036 |
blown | -0.035 |
Token | It |
Token ID | 632 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
arrest | +0.031 |
blown | +0.031 |
Blood | +0.030 |
spilled | +0.029 |
robbers | +0.029 |
Neg logit contributions | |
Deal | -0.033 |
isn | -0.029 |
why | -0.028 |
difference | -0.028 |
ement | -0.028 |
Token | had |
Token ID | 550 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
Blood | +0.057 |
arrest | +0.056 |
robbers | +0.055 |
blown | +0.055 |
growth | +0.054 |
Neg logit contributions | |
Deal | -0.033 |
difference | -0.028 |
ement | -0.026 |
isn | -0.026 |
why | -0.025 |
Token | been |
Token ID | 587 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
alks | +0.158 |
Blood | +0.155 |
plets | +0.151 |
arrest | +0.148 |
ges | +0.148 |
Neg logit contributions | |
Deal | -0.097 |
isn | -0.096 |
about | -0.093 |
exactly | -0.090 |
wonderful | -0.088 |
Token | used |
Token ID | 973 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
growth | +0.096 |
arrest | +0.094 |
alks | +0.088 |
Blood | +0.087 |
plets | +0.086 |
Neg logit contributions | |
exactly | -0.056 |
Deal | -0.056 |
why | -0.052 |
equally | -0.049 |
about | -0.048 |
Token | by |
Token ID | 416 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
Blood | +0.071 |
arrest | +0.071 |
growth | +0.069 |
robbers | +0.068 |
blown | +0.066 |
Neg logit contributions | |
Deal | -0.078 |
exactly | -0.070 |
difference | -0.069 |
why | -0.068 |
isn | -0.068 |
Token | many |
Token ID | 867 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Blood | +0.010 |
arrest | +0.010 |
robbers | +0.010 |
blown | +0.010 |
growth | +0.010 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.010 |
why | -0.010 |
exactly | -0.010 |
isn | -0.010 |
Token | animals |
Token ID | 4695 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
Blood | +0.020 |
arrest | +0.020 |
robbers | +0.019 |
blown | +0.019 |
growth | +0.019 |
Neg logit contributions | |
Deal | -0.032 |
difference | -0.029 |
ement | -0.028 |
isn | -0.028 |
why | -0.027 |
Token | big |
Token ID | 1263 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Deal | +0.017 |
difference | +0.015 |
ement | +0.014 |
why | +0.014 |
exactly | +0.014 |
Neg logit contributions | |
Blood | -0.010 |
robbers | -0.009 |
arrest | -0.009 |
blown | -0.009 |
spilled | -0.009 |
Token | do |
Token ID | 466 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
Deal | +0.052 |
difference | +0.043 |
ement | +0.043 |
exactly | +0.043 |
ieved | +0.042 |
Neg logit contributions | |
arrest | -0.037 |
robbers | -0.037 |
Blood | -0.036 |
mice | -0.035 |
blown | -0.035 |
Token | ggy |
Token ID | 19970 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
Blood | +0.126 |
arrest | +0.124 |
robbers | +0.121 |
blown | +0.120 |
growth | +0.119 |
Neg logit contributions | |
Deal | -0.080 |
difference | -0.068 |
ement | -0.062 |
why | -0.061 |
isn | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
arrest | +0.008 |
robbers | +0.008 |
spilled | +0.008 |
Blood | +0.007 |
blown | +0.007 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.010 |
ement | -0.009 |
decision | -0.009 |
isn | -0.009 |
Token | The |
Token ID | 383 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
Blood | +0.038 |
arrest | +0.037 |
robbers | +0.036 |
blown | +0.035 |
growth | +0.035 |
Neg logit contributions | |
Deal | -0.042 |
difference | -0.036 |
ement | -0.034 |
exactly | -0.034 |
decision | -0.034 |
Token | do |
Token ID | 466 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
Blood | +0.043 |
arrest | +0.043 |
blown | +0.042 |
robbers | +0.041 |
growth | +0.041 |
Neg logit contributions | |
Deal | -0.050 |
difference | -0.044 |
isn | -0.042 |
ement | -0.042 |
decision | -0.041 |
Token | ggy |
Token ID | 19970 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
Blood | +0.175 |
arrest | +0.171 |
robbers | +0.167 |
blown | +0.166 |
growth | +0.163 |
Neg logit contributions | |
Deal | -0.106 |
difference | -0.091 |
ement | -0.083 |
why | -0.082 |
isn | -0.081 |
Token | bark |
Token ID | 21405 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
Blood | +0.050 |
spilled | +0.048 |
robbers | +0.048 |
arrest | +0.048 |
blown | +0.047 |
Neg logit contributions | |
Deal | -0.050 |
difference | -0.045 |
ement | -0.042 |
decision | -0.041 |
promise | -0.040 |
Token | ed |
Token ID | 276 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
Blood | +0.013 |
arrest | +0.013 |
robbers | +0.012 |
blown | +0.012 |
growth | +0.012 |
Neg logit contributions | |
Deal | -0.025 |
difference | -0.022 |
ement | -0.022 |
isn | -0.022 |
why | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
Deal | +0.011 |
difference | +0.009 |
ieved | +0.009 |
uct | +0.008 |
ement | +0.008 |
Neg logit contributions | |
mice | -0.006 |
robbers | -0.005 |
Blood | -0.005 |
spilled | -0.005 |
blown | -0.005 |
Token | Tommy |
Token ID | 19919 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
Blood | +0.007 |
arrest | +0.007 |
robbers | +0.007 |
blown | +0.007 |
growth | +0.007 |
Neg logit contributions | |
Deal | -0.009 |
difference | -0.008 |
ement | -0.008 |
isn | -0.008 |
decision | -0.008 |
Token | interesting |
Token ID | 3499 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
Deal | +0.031 |
ieved | +0.024 |
uct | +0.024 |
difference | +0.023 |
rats | +0.022 |
Neg logit contributions | |
mice | -0.032 |
blown | -0.031 |
robbers | -0.030 |
arrest | -0.030 |
growth | -0.030 |
Token | things |
Token ID | 1243 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
Deal | +0.006 |
ieved | +0.005 |
uct | +0.005 |
difference | +0.005 |
ement | +0.005 |
Neg logit contributions | |
mice | -0.005 |
robbers | -0.005 |
arrest | -0.005 |
growth | -0.005 |
alks | -0.005 |
Token | about |
Token ID | 546 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
Deal | +0.059 |
difference | +0.051 |
ieved | +0.049 |
decision | +0.049 |
ement | +0.049 |
Neg logit contributions | |
blown | -0.040 |
spilled | -0.038 |
robbers | -0.038 |
arrest | -0.037 |
Blood | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
Deal | +0.010 |
difference | +0.009 |
ement | +0.009 |
decision | +0.008 |
promise | +0.008 |
Neg logit contributions | |
arrest | -0.008 |
Blood | -0.008 |
robbers | -0.008 |
blown | -0.008 |
growth | -0.008 |
Token | unknown |
Token ID | 6439 |
Feature activation | -7.63e-01 |
Pos logit contributions | |
Deal | +0.015 |
ieved | +0.013 |
ement | +0.012 |
uct | +0.012 |
exactly | +0.012 |
Neg logit contributions | |
arrest | -0.012 |
robbers | -0.012 |
mice | -0.011 |
blown | -0.011 |
spilled | -0.011 |
Token | ear |
Token ID | 1027 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
arrest | +0.017 |
blown | +0.016 |
Blood | +0.016 |
robbers | +0.016 |
growth | +0.016 |
Neg logit contributions | |
Deal | -0.022 |
ieved | -0.018 |
uct | -0.018 |
difference | -0.018 |
ement | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
Blood | +0.144 |
arrest | +0.141 |
robbers | +0.139 |
blown | +0.137 |
growth | +0.136 |
Neg logit contributions | |
Deal | -0.131 |
difference | -0.116 |
ement | -0.111 |
isn | -0.110 |
why | -0.109 |
Token | He |
Token ID | 679 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.012 |
decision | +0.011 |
promise | +0.011 |
ement | +0.011 |
Neg logit contributions | |
Blood | -0.012 |
alks | -0.012 |
growth | -0.012 |
arrest | -0.012 |
robbers | -0.011 |
Token | still |
Token ID | 991 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
Deal | +0.001 |
difference | +0.001 |
ement | +0.001 |
decision | +0.001 |
deal | +0.001 |
Neg logit contributions | |
blown | -0.001 |
arrest | -0.001 |
robbers | -0.001 |
Blood | -0.001 |
growth | -0.001 |
Token | visits |
Token ID | 11864 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
Deal | +0.015 |
difference | +0.013 |
ement | +0.013 |
decision | +0.013 |
exactly | +0.012 |
Neg logit contributions | |
Blood | -0.011 |
arrest | -0.011 |
blown | -0.011 |
robbers | -0.011 |
spilled | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
Deal | +0.018 |
difference | +0.015 |
ement | +0.015 |
promise | +0.015 |
decision | +0.014 |
Neg logit contributions | |
arrest | -0.012 |
Blood | -0.012 |
robbers | -0.012 |
blown | -0.012 |
growth | -0.012 |
Token | that |
Token ID | 326 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
Deal | +0.031 |
difference | +0.028 |
exactly | +0.027 |
ement | +0.027 |
why | +0.027 |
Neg logit contributions | |
Blood | -0.033 |
arrest | -0.032 |
robbers | -0.032 |
growth | -0.032 |
blown | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
Deal | +0.072 |
ieved | +0.062 |
exactly | +0.060 |
ement | +0.059 |
uct | +0.059 |
Neg logit contributions | |
mice | -0.055 |
spilled | -0.054 |
blown | -0.053 |
arrest | -0.053 |
robbers | -0.052 |
Token | Whenever |
Token ID | 21326 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
Blood | +0.005 |
arrest | +0.005 |
blown | +0.005 |
robbers | +0.005 |
growth | +0.005 |
Neg logit contributions | |
Deal | -0.007 |
difference | -0.007 |
isn | -0.006 |
ement | -0.006 |
why | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
Blood | +0.006 |
arrest | +0.006 |
robbers | +0.005 |
blown | +0.005 |
growth | +0.005 |
Neg logit contributions | |
Deal | -0.009 |
difference | -0.008 |
ement | -0.008 |
isn | -0.008 |
decision | -0.008 |
Token | chip |
Token ID | 11594 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
Blood | +0.038 |
arrest | +0.037 |
growth | +0.037 |
robbers | +0.036 |
blown | +0.036 |
Neg logit contributions | |
Deal | -0.052 |
difference | -0.047 |
decision | -0.045 |
promise | -0.045 |
ement | -0.044 |
Token | m |
Token ID | 76 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
Blood | +0.066 |
arrest | +0.065 |
robbers | +0.064 |
blown | +0.063 |
growth | +0.062 |
Neg logit contributions | |
Deal | -0.078 |
difference | -0.069 |
ement | -0.066 |
isn | -0.066 |
why | -0.065 |
Token | unk |
Token ID | 2954 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
robbers | +0.107 |
growth | +0.105 |
arrest | +0.104 |
blown | +0.096 |
Blood | +0.096 |
Neg logit contributions | |
Deal | -0.160 |
decision | -0.155 |
isn | -0.151 |
difference | -0.151 |
ement | -0.150 |
Token | needed |
Token ID | 2622 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Blood | +0.049 |
arrest | +0.049 |
blown | +0.048 |
robbers | +0.048 |
mice | +0.047 |
Neg logit contributions | |
Deal | -0.053 |
difference | -0.046 |
ement | -0.045 |
decision | -0.044 |
exactly | -0.044 |
Token | help |
Token ID | 1037 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
Blood | +0.022 |
arrest | +0.022 |
robbers | +0.021 |
blown | +0.021 |
growth | +0.021 |
Neg logit contributions | |
Deal | -0.030 |
difference | -0.027 |
ement | -0.026 |
why | -0.026 |
isn | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
Blood | +0.015 |
arrest | +0.015 |
mice | +0.015 |
robbers | +0.015 |
blown | +0.014 |
Neg logit contributions | |
Deal | -0.014 |
ement | -0.011 |
difference | -0.011 |
ieved | -0.011 |
uct | -0.011 |
Token | Sam |
Token ID | 3409 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.003 |
ement | +0.003 |
isn | +0.003 |
decision | +0.003 |
Neg logit contributions | |
Blood | -0.002 |
arrest | -0.002 |
robbers | -0.002 |
blown | -0.002 |
mice | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
alks | +0.058 |
arrest | +0.057 |
plets | +0.056 |
robbers | +0.056 |
Blood | +0.055 |
Neg logit contributions | |
Deal | -0.053 |
exactly | -0.046 |
equally | -0.043 |
why | -0.043 |
difference | -0.043 |
Token | big |
Token ID | 1263 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
Deal | +0.008 |
difference | +0.007 |
ement | +0.007 |
isn | +0.007 |
exactly | +0.007 |
Neg logit contributions | |
Blood | -0.004 |
arrest | -0.004 |
robbers | -0.004 |
blown | -0.004 |
mice | -0.004 |
Token | house |
Token ID | 2156 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
Deal | +0.058 |
ieved | +0.050 |
ement | +0.049 |
exactly | +0.049 |
uct | +0.049 |
Neg logit contributions | |
robbers | -0.039 |
arrest | -0.038 |
mice | -0.037 |
Blood | -0.037 |
spilled | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Deal | +0.006 |
difference | +0.006 |
ement | +0.005 |
isn | +0.005 |
decision | +0.005 |
Neg logit contributions | |
Blood | -0.006 |
arrest | -0.006 |
blown | -0.006 |
robbers | -0.006 |
spilled | -0.006 |
Token | It |
Token ID | 632 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
blown | +0.033 |
arrest | +0.033 |
spilled | +0.031 |
alks | +0.031 |
Blood | +0.031 |
Neg logit contributions | |
Deal | -0.036 |
isn | -0.032 |
difference | -0.031 |
why | -0.031 |
ement | -0.030 |
Token | had |
Token ID | 550 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
Blood | +0.050 |
spilled | +0.048 |
blown | +0.048 |
arrest | +0.046 |
robbers | +0.046 |
Neg logit contributions | |
Deal | -0.041 |
difference | -0.037 |
ement | -0.035 |
exactly | -0.034 |
decision | -0.034 |
Token | lots |
Token ID | 6041 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
alks | +0.141 |
plets | +0.140 |
robbers | +0.140 |
Blood | +0.138 |
arrest | +0.136 |
Neg logit contributions | |
Deal | -0.112 |
isn | -0.109 |
exactly | -0.107 |
why | -0.103 |
difference | -0.102 |
Token | of |
Token ID | 286 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
Blood | +0.063 |
arrest | +0.061 |
robbers | +0.060 |
growth | +0.060 |
blown | +0.059 |
Neg logit contributions | |
Deal | -0.065 |
difference | -0.060 |
ement | -0.058 |
isn | -0.058 |
why | -0.056 |
Token | surprises |
Token ID | 24072 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
Blood | +0.074 |
arrest | +0.072 |
robbers | +0.071 |
blown | +0.070 |
growth | +0.068 |
Neg logit contributions | |
Deal | -0.126 |
difference | -0.113 |
ement | -0.110 |
isn | -0.108 |
why | -0.108 |
Token | inside |
Token ID | 2641 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
Deal | +0.040 |
difference | +0.036 |
ement | +0.034 |
ieved | +0.033 |
decision | +0.033 |
Neg logit contributions | |
blown | -0.031 |
robbers | -0.030 |
spilled | -0.030 |
Blood | -0.030 |
arrest | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
Blood | +0.014 |
arrest | +0.013 |
mice | +0.013 |
blown | +0.013 |
robbers | +0.013 |
Neg logit contributions | |
Deal | -0.020 |
difference | -0.018 |
ement | -0.017 |
decision | -0.017 |
promise | -0.017 |
Token | " |
Token ID | 366 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Deal | +0.004 |
difference | +0.003 |
why | +0.003 |
ement | +0.003 |
isn | +0.003 |
Neg logit contributions | |
arrest | -0.005 |
Blood | -0.005 |
robbers | -0.005 |
blown | -0.005 |
growth | -0.004 |
Token | Let |
Token ID | 5756 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Deal | +0.044 |
difference | +0.041 |
ement | +0.039 |
decision | +0.038 |
isn | +0.038 |
Neg logit contributions | |
Blood | -0.034 |
arrest | -0.033 |
blown | -0.032 |
robbers | -0.032 |
growth | -0.032 |
Token | 's |
Token ID | 338 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
Deal | +0.004 |
ement | +0.003 |
difference | +0.003 |
exactly | +0.003 |
decision | +0.003 |
Neg logit contributions | |
arrest | -0.006 |
Blood | -0.006 |
blown | -0.005 |
spilled | -0.005 |
robbers | -0.005 |
Token | replace |
Token ID | 6330 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
Deal | +0.087 |
ement | +0.078 |
difference | +0.073 |
isn | +0.071 |
decision | +0.071 |
Neg logit contributions | |
arrest | -0.061 |
robbers | -0.058 |
Blood | -0.058 |
blown | -0.058 |
growth | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Deal | +0.038 |
difference | +0.032 |
ement | +0.032 |
decision | +0.031 |
promise | +0.030 |
Neg logit contributions | |
arrest | -0.036 |
Blood | -0.035 |
mice | -0.035 |
robbers | -0.035 |
blown | -0.034 |
Token | ear |
Token ID | 1027 |
Feature activation | -5.43e+00 |
Pos logit contributions | |
Blood | +0.030 |
arrest | +0.030 |
robbers | +0.029 |
blown | +0.029 |
mice | +0.028 |
Neg logit contributions | |
Deal | -0.041 |
difference | -0.036 |
ement | -0.036 |
why | -0.035 |
exactly | -0.035 |
Token | so |
Token ID | 523 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
Blood | +0.136 |
arrest | +0.134 |
blown | +0.130 |
robbers | +0.130 |
growth | +0.127 |
Neg logit contributions | |
Deal | -0.140 |
difference | -0.123 |
ement | -0.117 |
isn | -0.114 |
decision | -0.114 |
Token | your |
Token ID | 534 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Deal | +0.015 |
difference | +0.013 |
ement | +0.013 |
decision | +0.012 |
isn | +0.012 |
Neg logit contributions | |
Blood | -0.017 |
arrest | -0.016 |
blown | -0.016 |
robbers | -0.016 |
mice | -0.016 |
Token | toy |
Token ID | 13373 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
Deal | +0.009 |
ement | +0.008 |
difference | +0.008 |
why | +0.007 |
ieved | +0.007 |
Neg logit contributions | |
Blood | -0.006 |
arrest | -0.006 |
blown | -0.006 |
robbers | -0.006 |
mice | -0.006 |
Token | cat |
Token ID | 3797 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
arrest | +0.018 |
Blood | +0.017 |
robbers | +0.017 |
blown | +0.017 |
mice | +0.017 |
Neg logit contributions | |
Deal | -0.017 |
difference | -0.015 |
ement | -0.015 |
decision | -0.014 |
why | -0.014 |
Token | can |
Token ID | 460 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
arrest | +0.015 |
Blood | +0.015 |
blown | +0.015 |
robbers | +0.015 |
spilled | +0.015 |
Neg logit contributions | |
Deal | -0.013 |
difference | -0.011 |
ement | -0.011 |
decision | -0.010 |
int | -0.010 |
Token | some |
Token ID | 617 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
arrest | +0.021 |
Blood | +0.021 |
alks | +0.020 |
robbers | +0.020 |
blown | +0.020 |
Neg logit contributions | |
Deal | -0.020 |
why | -0.018 |
isn | -0.018 |
exactly | -0.017 |
ement | -0.017 |
Token | are |
Token ID | 389 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Blood | +0.053 |
arrest | +0.052 |
robbers | +0.050 |
growth | +0.050 |
blown | +0.049 |
Neg logit contributions | |
Deal | -0.048 |
difference | -0.043 |
isn | -0.042 |
why | -0.041 |
promise | -0.041 |
Token | blue |
Token ID | 4171 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
arrest | +0.050 |
Blood | +0.050 |
growth | +0.048 |
robbers | +0.048 |
blown | +0.047 |
Neg logit contributions | |
Deal | -0.061 |
difference | -0.055 |
why | -0.054 |
exactly | -0.054 |
isn | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Blood | +0.019 |
arrest | +0.018 |
robbers | +0.018 |
growth | +0.018 |
blown | +0.018 |
Neg logit contributions | |
Deal | -0.008 |
isn | -0.007 |
difference | -0.007 |
ement | -0.007 |
why | -0.006 |
Token | some |
Token ID | 617 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
arrest | +0.033 |
Blood | +0.032 |
robbers | +0.032 |
alks | +0.031 |
growth | +0.031 |
Neg logit contributions | |
Deal | -0.033 |
isn | -0.030 |
why | -0.029 |
exactly | -0.029 |
ement | -0.029 |
Token | have |
Token ID | 423 |
Feature activation | -5.42e+00 |
Pos logit contributions | |
Blood | +0.057 |
arrest | +0.056 |
robbers | +0.055 |
growth | +0.054 |
blown | +0.053 |
Neg logit contributions | |
Deal | -0.057 |
difference | -0.051 |
isn | -0.049 |
why | -0.048 |
promise | -0.048 |
Token | flowers |
Token ID | 12734 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
Blood | +0.136 |
arrest | +0.131 |
robbers | +0.130 |
alks | +0.125 |
growth | +0.121 |
Neg logit contributions | |
isn | -0.121 |
exactly | -0.114 |
why | -0.113 |
Deal | -0.111 |
hairst | -0.109 |
Token | , |
Token ID | 11 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
Blood | +0.019 |
arrest | +0.018 |
robbers | +0.018 |
growth | +0.018 |
mice | +0.017 |
Neg logit contributions | |
Deal | -0.010 |
isn | -0.009 |
why | -0.009 |
difference | -0.009 |
exactly | -0.009 |
Token | some |
Token ID | 617 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
Blood | +0.040 |
arrest | +0.039 |
alks | +0.038 |
robbers | +0.038 |
blown | +0.038 |
Neg logit contributions | |
Deal | -0.038 |
isn | -0.038 |
why | -0.037 |
exactly | -0.035 |
ement | -0.034 |
Token | have |
Token ID | 423 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
Blood | +0.045 |
arrest | +0.044 |
robbers | +0.043 |
growth | +0.043 |
blown | +0.041 |
Neg logit contributions | |
Deal | -0.056 |
difference | -0.051 |
isn | -0.051 |
why | -0.049 |
ement | -0.049 |
Token | dots |
Token ID | 22969 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
Blood | +0.130 |
robbers | +0.125 |
arrest | +0.122 |
alks | +0.117 |
plets | +0.113 |
Neg logit contributions | |
isn | -0.126 |
exactly | -0.117 |
why | -0.117 |
Deal | -0.116 |
hairst | -0.115 |
Token | back |
Token ID | 736 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Deal | +0.004 |
ement | +0.004 |
exactly | +0.004 |
difference | +0.004 |
why | +0.004 |
Neg logit contributions | |
arrest | -0.004 |
Blood | -0.004 |
robbers | -0.004 |
growth | -0.004 |
alks | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +9.15e-01 |
Pos logit contributions | |
Blood | +0.027 |
arrest | +0.026 |
blown | +0.026 |
robbers | +0.026 |
growth | +0.025 |
Neg logit contributions | |
Deal | -0.035 |
difference | -0.031 |
ement | -0.030 |
isn | -0.029 |
decision | -0.029 |
Token | admired |
Token ID | 29382 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
Deal | +0.023 |
difference | +0.020 |
ement | +0.020 |
decision | +0.019 |
isn | +0.018 |
Neg logit contributions | |
blown | -0.024 |
spilled | -0.024 |
Blood | -0.024 |
arrest | -0.023 |
robbers | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Blood | +0.011 |
arrest | +0.011 |
blown | +0.011 |
robbers | +0.011 |
growth | +0.011 |
Neg logit contributions | |
Deal | -0.012 |
difference | -0.011 |
ement | -0.010 |
decision | -0.010 |
promise | -0.010 |
Token | spark |
Token ID | 9009 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
robbers | +0.023 |
blown | +0.023 |
arrest | +0.023 |
mice | +0.022 |
spilled | +0.022 |
Neg logit contributions | |
Deal | -0.040 |
ement | -0.035 |
ieved | -0.035 |
uct | -0.035 |
why | -0.034 |
Token | ly |
Token ID | 306 |
Feature activation | -5.41e+00 |
Pos logit contributions | |
arrest | +0.024 |
Blood | +0.023 |
blown | +0.023 |
mice | +0.023 |
spilled | +0.023 |
Neg logit contributions | |
Deal | -0.033 |
difference | -0.028 |
exactly | -0.027 |
why | -0.027 |
ement | -0.027 |
Token | space |
Token ID | 2272 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
Blood | +0.126 |
arrest | +0.122 |
robbers | +0.120 |
blown | +0.119 |
growth | +0.117 |
Neg logit contributions | |
Deal | -0.146 |
difference | -0.134 |
isn | -0.128 |
ement | -0.126 |
decision | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
blown | +0.008 |
arrest | +0.008 |
mice | +0.008 |
alks | +0.008 |
spilled | +0.008 |
Neg logit contributions | |
Deal | -0.012 |
difference | -0.010 |
ieved | -0.010 |
uct | -0.009 |
ement | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
Deal | +0.001 |
difference | +0.001 |
isn | +0.001 |
ement | +0.001 |
why | +0.001 |
Neg logit contributions | |
arrest | -0.001 |
robbers | -0.001 |
blown | -0.001 |
Blood | -0.001 |
spilled | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Blood | +0.015 |
arrest | +0.014 |
growth | +0.014 |
blown | +0.014 |
alks | +0.014 |
Neg logit contributions | |
Deal | -0.019 |
difference | -0.017 |
decision | -0.016 |
ement | -0.016 |
isn | -0.015 |
Token | The |
Token ID | 464 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
Blood | +0.031 |
arrest | +0.030 |
blown | +0.030 |
robbers | +0.030 |
growth | +0.029 |
Neg logit contributions | |
Deal | -0.030 |
difference | -0.027 |
ement | -0.026 |
decision | -0.025 |
isn | -0.025 |
Token | some |
Token ID | 617 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
arrest | +0.033 |
Blood | +0.032 |
robbers | +0.032 |
alks | +0.031 |
growth | +0.031 |
Neg logit contributions | |
Deal | -0.033 |
isn | -0.030 |
why | -0.029 |
exactly | -0.029 |
ement | -0.029 |
Token | have |
Token ID | 423 |
Feature activation | -5.42e+00 |
Pos logit contributions | |
Blood | +0.057 |
arrest | +0.056 |
robbers | +0.055 |
growth | +0.054 |
blown | +0.053 |
Neg logit contributions | |
Deal | -0.057 |
difference | -0.051 |
isn | -0.049 |
why | -0.048 |
promise | -0.048 |
Token | flowers |
Token ID | 12734 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
Blood | +0.136 |
arrest | +0.131 |
robbers | +0.130 |
alks | +0.125 |
growth | +0.121 |
Neg logit contributions | |
isn | -0.121 |
exactly | -0.114 |
why | -0.113 |
Deal | -0.111 |
hairst | -0.109 |
Token | , |
Token ID | 11 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
Blood | +0.019 |
arrest | +0.018 |
robbers | +0.018 |
growth | +0.018 |
mice | +0.017 |
Neg logit contributions | |
Deal | -0.010 |
isn | -0.009 |
why | -0.009 |
difference | -0.009 |
exactly | -0.009 |
Token | some |
Token ID | 617 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
Blood | +0.040 |
arrest | +0.039 |
alks | +0.038 |
robbers | +0.038 |
blown | +0.038 |
Neg logit contributions | |
Deal | -0.038 |
isn | -0.038 |
why | -0.037 |
exactly | -0.035 |
ement | -0.034 |
Token | have |
Token ID | 423 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
Blood | +0.045 |
arrest | +0.044 |
robbers | +0.043 |
growth | +0.043 |
blown | +0.041 |
Neg logit contributions | |
Deal | -0.056 |
difference | -0.051 |
isn | -0.051 |
why | -0.049 |
ement | -0.049 |
Token | dots |
Token ID | 22969 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
Blood | +0.130 |
robbers | +0.125 |
arrest | +0.122 |
alks | +0.117 |
plets | +0.113 |
Neg logit contributions | |
isn | -0.126 |
exactly | -0.117 |
why | -0.117 |
Deal | -0.116 |
hairst | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
arrest | +0.010 |
robbers | +0.009 |
Blood | +0.009 |
growth | +0.009 |
blown | +0.009 |
Neg logit contributions | |
Deal | -0.005 |
why | -0.004 |
isn | -0.004 |
exactly | -0.004 |
ement | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | -9.33e-01 |
Pos logit contributions | |
Deal | +0.009 |
difference | +0.008 |
decision | +0.008 |
ement | +0.008 |
exactly | +0.008 |
Neg logit contributions | |
Blood | -0.006 |
arrest | -0.006 |
robbers | -0.006 |
blown | -0.006 |
growth | -0.006 |
Token | likes |
Token ID | 7832 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
Blood | +0.023 |
arrest | +0.023 |
blown | +0.022 |
robbers | +0.022 |
growth | +0.022 |
Neg logit contributions | |
Deal | -0.024 |
difference | -0.021 |
ement | -0.020 |
decision | -0.020 |
exactly | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
Blood | +0.015 |
arrest | +0.015 |
growth | +0.014 |
robbers | +0.014 |
plets | +0.014 |
Neg logit contributions | |
Deal | -0.016 |
why | -0.015 |
isn | -0.015 |
difference | -0.015 |
exactly | -0.014 |
Token | box |
Token ID | 3091 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
Deal | +0.030 |
ieved | +0.025 |
uct | +0.025 |
ement | +0.025 |
exactly | +0.024 |
Neg logit contributions | |
robbers | -0.020 |
arrest | -0.019 |
spilled | -0.019 |
mice | -0.018 |
Blood | -0.018 |
Token | full |
Token ID | 1336 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
Deal | +0.020 |
difference | +0.017 |
ement | +0.017 |
why | +0.017 |
exactly | +0.017 |
Neg logit contributions | |
Blood | -0.020 |
arrest | -0.019 |
robbers | -0.019 |
blown | -0.019 |
growth | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
Blood | +0.062 |
arrest | +0.061 |
robbers | +0.061 |
blown | +0.060 |
mice | +0.060 |
Neg logit contributions | |
Deal | -0.118 |
difference | -0.103 |
ement | -0.100 |
isn | -0.098 |
decision | -0.098 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
robbers | +0.070 |
mice | +0.070 |
arrest | +0.069 |
Blood | +0.068 |
blown | +0.067 |
Neg logit contributions | |
Deal | -0.119 |
difference | -0.103 |
ement | -0.102 |
ieved | -0.100 |
decision | -0.099 |
Token | or |
Token ID | 393 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
arrest | +0.052 |
Blood | +0.052 |
robbers | +0.051 |
blown | +0.050 |
mice | +0.050 |
Neg logit contributions | |
Deal | -0.072 |
difference | -0.063 |
ement | -0.061 |
why | -0.060 |
exactly | -0.060 |
Token | nam |
Token ID | 7402 |
Feature activation | -5.38e+00 |
Pos logit contributions | |
Blood | +0.008 |
robbers | +0.008 |
arrest | +0.008 |
blown | +0.008 |
mice | +0.007 |
Neg logit contributions | |
Deal | -0.012 |
difference | -0.011 |
ement | -0.011 |
isn | -0.010 |
decision | -0.010 |
Token | ents |
Token ID | 658 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
robbers | +0.119 |
growth | +0.113 |
Blood | +0.112 |
arrest | +0.112 |
spilled | +0.109 |
Neg logit contributions | |
Deal | -0.140 |
equally | -0.136 |
exactly | -0.135 |
ement | -0.134 |
difference | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
Blood | +0.049 |
arrest | +0.047 |
robbers | +0.047 |
growth | +0.046 |
alks | +0.045 |
Neg logit contributions | |
Deal | -0.041 |
difference | -0.036 |
exactly | -0.035 |
ement | -0.035 |
why | -0.035 |
Token | |
Token ID | 220 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
arrest | +0.019 |
Blood | +0.019 |
blown | +0.019 |
robbers | +0.019 |
alks | +0.018 |
Neg logit contributions | |
Deal | -0.024 |
difference | -0.021 |
isn | -0.020 |
ement | -0.020 |
why | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Blood | +0.053 |
arrest | +0.051 |
robbers | +0.050 |
blown | +0.050 |
growth | +0.049 |
Neg logit contributions | |
Deal | -0.052 |
difference | -0.045 |
ement | -0.043 |
isn | -0.043 |
decision | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
Blood | +0.026 |
arrest | +0.025 |
robbers | +0.025 |
growth | +0.024 |
blown | +0.024 |
Neg logit contributions | |
Deal | -0.031 |
difference | -0.027 |
decision | -0.026 |
ement | -0.026 |
isn | -0.026 |
Token | other |
Token ID | 584 |
Feature activation | -4.28e+00 |
Pos logit contributions | |
arrest | +0.057 |
Blood | +0.057 |
growth | +0.056 |
blown | +0.054 |
alks | +0.052 |
Neg logit contributions | |
Deal | -0.073 |
difference | -0.068 |
decision | -0.064 |
promise | -0.064 |
why | -0.061 |
Token | was |
Token ID | 373 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
Blood | +0.095 |
arrest | +0.094 |
blown | +0.090 |
alks | +0.089 |
growth | +0.087 |
Neg logit contributions | |
Deal | -0.125 |
difference | -0.113 |
isn | -0.112 |
why | -0.109 |
decision | -0.108 |
Token | sleepy |
Token ID | 42023 |
Feature activation | -9.27e-01 |
Pos logit contributions | |
Blood | +0.024 |
arrest | +0.024 |
robbers | +0.024 |
growth | +0.024 |
blown | +0.023 |
Neg logit contributions | |
Deal | -0.022 |
difference | -0.019 |
why | -0.018 |
ement | -0.018 |
exactly | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
Blood | +0.022 |
robbers | +0.022 |
arrest | +0.022 |
mice | +0.022 |
blown | +0.021 |
Neg logit contributions | |
Deal | -0.025 |
difference | -0.021 |
ement | -0.021 |
exactly | -0.020 |
ieved | -0.020 |
Token | The |
Token ID | 383 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
blown | +0.026 |
arrest | +0.026 |
Blood | +0.024 |
spilled | +0.024 |
growth | +0.024 |
Neg logit contributions | |
Deal | -0.030 |
difference | -0.026 |
isn | -0.026 |
why | -0.026 |
ement | -0.025 |
Token | alert |
Token ID | 7995 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
arrest | +0.033 |
Blood | +0.033 |
growth | +0.032 |
blown | +0.031 |
alks | +0.031 |
Neg logit contributions | |
Deal | -0.042 |
difference | -0.039 |
promise | -0.037 |
decision | -0.037 |
why | -0.036 |
Token | human |
Token ID | 1692 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Blood | +0.110 |
arrest | +0.105 |
growth | +0.103 |
blown | +0.103 |
bitten | +0.097 |
Neg logit contributions | |
decision | -0.149 |
difference | -0.146 |
why | -0.142 |
isn | -0.140 |
Deal | -0.140 |
Token | said |
Token ID | 531 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
Blood | +0.033 |
mice | +0.033 |
robbers | +0.033 |
spilled | +0.033 |
arrest | +0.032 |
Neg logit contributions | |
Deal | -0.040 |
difference | -0.036 |
ement | -0.035 |
decision | -0.034 |
exactly | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
Blood | +0.023 |
mice | +0.022 |
robbers | +0.021 |
arrest | +0.021 |
spilled | +0.021 |
Neg logit contributions | |
Deal | -0.024 |
difference | -0.021 |
ement | -0.020 |
ieved | -0.019 |
decision | -0.019 |
Token | " |
Token ID | 366 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
Blood | +0.008 |
robbers | +0.008 |
arrest | +0.008 |
mice | +0.008 |
blown | +0.008 |
Neg logit contributions | |
Deal | -0.009 |
difference | -0.009 |
decision | -0.008 |
ement | -0.008 |
promise | -0.008 |
Token | Let |
Token ID | 5756 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
Deal | +0.010 |
difference | +0.009 |
isn | +0.009 |
ement | +0.009 |
why | +0.009 |
Neg logit contributions | |
Blood | -0.008 |
arrest | -0.008 |
robbers | -0.007 |
blown | -0.007 |
alks | -0.007 |
Token | room |
Token ID | 2119 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
Blood | +0.001 |
arrest | +0.001 |
robbers | +0.001 |
blown | +0.001 |
growth | +0.001 |
Neg logit contributions | |
Deal | -0.001 |
difference | -0.001 |
ement | -0.001 |
isn | -0.001 |
why | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.013 |
ement | +0.012 |
isn | +0.012 |
why | +0.012 |
Neg logit contributions | |
Blood | -0.015 |
arrest | -0.015 |
robbers | -0.014 |
blown | -0.014 |
growth | -0.014 |
Token | He |
Token ID | 679 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
arrest | +0.036 |
Blood | +0.036 |
blown | +0.035 |
robbers | +0.034 |
growth | +0.034 |
Neg logit contributions | |
Deal | -0.047 |
difference | -0.042 |
isn | -0.040 |
why | -0.040 |
ement | -0.040 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
Blood | +0.009 |
robbers | +0.009 |
arrest | +0.009 |
blown | +0.009 |
spilled | +0.008 |
Neg logit contributions | |
Deal | -0.013 |
difference | -0.012 |
ement | -0.012 |
decision | -0.011 |
exactly | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
why | +0.009 |
Deal | +0.009 |
exactly | +0.008 |
difference | +0.008 |
isn | +0.008 |
Neg logit contributions | |
plets | -0.012 |
alks | -0.012 |
Blood | -0.012 |
arrest | -0.012 |
growth | -0.012 |
Token | lonely |
Token ID | 21757 |
Feature activation | -5.36e+00 |
Pos logit contributions | |
Blood | +0.025 |
arrest | +0.025 |
growth | +0.025 |
robbers | +0.024 |
blown | +0.024 |
Neg logit contributions | |
Deal | -0.023 |
difference | -0.020 |
decision | -0.019 |
isn | -0.019 |
ement | -0.019 |
Token | machine |
Token ID | 4572 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
growth | +0.169 |
arrest | +0.166 |
blown | +0.164 |
Blood | +0.162 |
robbers | +0.160 |
Neg logit contributions | |
Deal | -0.094 |
decision | -0.089 |
difference | -0.088 |
promise | -0.083 |
isn | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Deal | +0.037 |
difference | +0.031 |
ement | +0.030 |
decision | +0.030 |
ieved | +0.030 |
Neg logit contributions | |
spilled | -0.023 |
blown | -0.022 |
Blood | -0.022 |
arrest | -0.022 |
robbers | -0.022 |
Token | had |
Token ID | 550 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Deal | +0.004 |
difference | +0.004 |
ement | +0.003 |
isn | +0.003 |
why | +0.003 |
Neg logit contributions | |
Blood | -0.004 |
arrest | -0.004 |
robbers | -0.004 |
blown | -0.004 |
growth | -0.003 |
Token | an |
Token ID | 281 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
Blood | +0.031 |
arrest | +0.030 |
robbers | +0.029 |
growth | +0.029 |
blown | +0.029 |
Neg logit contributions | |
Deal | -0.045 |
difference | -0.041 |
isn | -0.039 |
exactly | -0.039 |
why | -0.039 |
Token | idea |
Token ID | 2126 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
Deal | +0.027 |
difference | +0.020 |
isn | +0.019 |
ement | +0.018 |
why | +0.017 |
Neg logit contributions | |
Blood | -0.062 |
arrest | -0.061 |
robbers | -0.061 |
blown | -0.061 |
mice | -0.060 |
Token | counted |
Token ID | 14789 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Blood | +0.057 |
arrest | +0.057 |
blown | +0.055 |
robbers | +0.055 |
growth | +0.054 |
Neg logit contributions | |
Deal | -0.078 |
difference | -0.070 |
ement | -0.067 |
isn | -0.066 |
why | -0.066 |
Token | that |
Token ID | 326 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
blown | +0.026 |
alks | +0.026 |
Blood | +0.025 |
arrest | +0.024 |
robbers | +0.024 |
Neg logit contributions | |
Deal | -0.036 |
difference | -0.031 |
promise | -0.029 |
decision | -0.029 |
ement | -0.028 |
Token | high |
Token ID | 1029 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
Deal | +0.018 |
ement | +0.015 |
difference | +0.015 |
ieved | +0.015 |
decision | +0.014 |
Neg logit contributions | |
blown | -0.014 |
spilled | -0.013 |
Blood | -0.013 |
mice | -0.013 |
arrest | -0.013 |
Token | ! |
Token ID | 0 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
Deal | +0.005 |
difference | +0.005 |
ement | +0.004 |
promise | +0.004 |
exactly | +0.004 |
Neg logit contributions | |
blown | -0.005 |
arrest | -0.005 |
Blood | -0.005 |
growth | -0.005 |
spilled | -0.005 |
Token | The |
Token ID | 383 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
Deal | +0.012 |
difference | +0.011 |
decision | +0.010 |
ement | +0.010 |
why | +0.010 |
Neg logit contributions | |
Blood | -0.011 |
arrest | -0.011 |
growth | -0.011 |
robbers | -0.011 |
blown | -0.011 |
Token | ear |
Token ID | 1027 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
arrest | +0.006 |
Blood | +0.006 |
blown | +0.006 |
robbers | +0.006 |
growth | +0.006 |
Neg logit contributions | |
Deal | -0.010 |
difference | -0.008 |
ement | -0.008 |
exactly | -0.008 |
isn | -0.008 |
Token | knew |
Token ID | 2993 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Blood | +0.139 |
arrest | +0.137 |
blown | +0.136 |
spilled | +0.131 |
growth | +0.131 |
Neg logit contributions | |
Deal | -0.134 |
difference | -0.118 |
ement | -0.109 |
decision | -0.108 |
why | -0.105 |
Token | it |
Token ID | 340 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
Deal | +0.030 |
difference | +0.025 |
ement | +0.023 |
decision | +0.023 |
Amen | +0.023 |
Neg logit contributions | |
blown | -0.025 |
arrest | -0.023 |
Blood | -0.022 |
spilled | -0.022 |
robbers | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
Deal | +0.007 |
difference | +0.007 |
ement | +0.006 |
decision | +0.006 |
deal | +0.006 |
Neg logit contributions | |
blown | -0.005 |
spilled | -0.005 |
Blood | -0.005 |
robbers | -0.005 |
mice | -0.005 |
Token | getting |
Token ID | 1972 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
Deal | +0.048 |
ement | +0.041 |
difference | +0.040 |
olog | +0.037 |
oc | +0.037 |
Neg logit contributions | |
blown | -0.031 |
Blood | -0.028 |
bitten | -0.027 |
spilled | -0.027 |
alks | -0.026 |
Token | stronger |
Token ID | 7387 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
blown | +0.019 |
Blood | +0.018 |
robbers | +0.017 |
alks | +0.017 |
spilled | +0.017 |
Neg logit contributions | |
Deal | -0.020 |
difference | -0.018 |
ement | -0.017 |
decision | -0.017 |
isn | -0.017 |
Token | finally |
Token ID | 3443 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
arrest | +0.015 |
blown | +0.015 |
robbers | +0.014 |
spilled | +0.014 |
Blood | +0.014 |
Neg logit contributions | |
Deal | -0.020 |
difference | -0.017 |
ement | -0.017 |
decision | -0.016 |
isn | -0.015 |
Token | understood |
Token ID | 7247 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
arrest | +0.029 |
blown | +0.028 |
Blood | +0.028 |
robbers | +0.027 |
growth | +0.027 |
Neg logit contributions | |
Deal | -0.043 |
difference | -0.038 |
ement | -0.037 |
isn | -0.036 |
decision | -0.036 |
Token | why |
Token ID | 1521 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Deal | +0.057 |
ement | +0.047 |
difference | +0.046 |
ieved | +0.046 |
isn | +0.044 |
Neg logit contributions | |
arrest | -0.038 |
blown | -0.037 |
robbers | -0.037 |
mice | -0.036 |
growth | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Deal | +0.033 |
difference | +0.029 |
ement | +0.028 |
why | +0.028 |
isn | +0.028 |
Neg logit contributions | |
Blood | -0.029 |
arrest | -0.029 |
robbers | -0.028 |
blown | -0.028 |
growth | -0.028 |
Token | chip |
Token ID | 11594 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
Blood | +0.028 |
arrest | +0.028 |
robbers | +0.027 |
growth | +0.027 |
blown | +0.027 |
Neg logit contributions | |
Deal | -0.035 |
difference | -0.032 |
decision | -0.030 |
ement | -0.030 |
isn | -0.030 |
Token | m |
Token ID | 76 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
Blood | +0.056 |
arrest | +0.055 |
robbers | +0.054 |
blown | +0.054 |
growth | +0.053 |
Neg logit contributions | |
Deal | -0.062 |
difference | -0.055 |
ement | -0.052 |
isn | -0.052 |
why | -0.051 |
Token | unk |
Token ID | 2954 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
robbers | +0.112 |
growth | +0.111 |
arrest | +0.109 |
blown | +0.099 |
bandits | +0.098 |
Neg logit contributions | |
decision | -0.149 |
Deal | -0.148 |
isn | -0.144 |
promise | -0.142 |
ement | -0.142 |
Token | had |
Token ID | 550 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Blood | +0.047 |
arrest | +0.047 |
blown | +0.046 |
robbers | +0.046 |
spilled | +0.045 |
Neg logit contributions | |
Deal | -0.042 |
difference | -0.037 |
ement | -0.035 |
decision | -0.034 |
exactly | -0.033 |
Token | been |
Token ID | 587 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
Blood | +0.026 |
arrest | +0.025 |
growth | +0.025 |
robbers | +0.025 |
alks | +0.024 |
Neg logit contributions | |
Deal | -0.034 |
difference | -0.031 |
exactly | -0.030 |
isn | -0.030 |
why | -0.030 |
Token | so |
Token ID | 523 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
Blood | +0.014 |
arrest | +0.014 |
blown | +0.014 |
robbers | +0.014 |
mice | +0.013 |
Neg logit contributions | |
Deal | -0.015 |
difference | -0.013 |
ement | -0.012 |
isn | -0.012 |
decision | -0.012 |
Token | boss |
Token ID | 6478 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
Deal | +0.035 |
difference | +0.031 |
ement | +0.029 |
why | +0.029 |
exactly | +0.029 |
Neg logit contributions | |
Blood | -0.036 |
arrest | -0.036 |
robbers | -0.035 |
growth | -0.035 |
blown | -0.034 |
Token | had |
Token ID | 550 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
Deal | +0.011 |
difference | +0.010 |
decision | +0.010 |
ement | +0.009 |
exactly | +0.009 |
Neg logit contributions | |
Blood | -0.010 |
arrest | -0.009 |
blown | -0.009 |
robbers | -0.009 |
spilled | -0.009 |
Token | pictures |
Token ID | 5986 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Blood | +0.100 |
plets | +0.095 |
robbers | +0.094 |
arrest | +0.093 |
alks | +0.093 |
Neg logit contributions | |
Deal | -0.095 |
why | -0.089 |
isn | -0.089 |
exactly | -0.089 |
difference | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
Blood | +0.029 |
arrest | +0.028 |
robbers | +0.028 |
blown | +0.028 |
growth | +0.027 |
Neg logit contributions | |
Deal | -0.031 |
difference | -0.027 |
ement | -0.026 |
isn | -0.026 |
why | -0.026 |
Token | some |
Token ID | 617 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
Blood | +0.019 |
arrest | +0.019 |
blown | +0.019 |
alks | +0.018 |
growth | +0.018 |
Neg logit contributions | |
Deal | -0.023 |
why | -0.020 |
difference | -0.020 |
isn | -0.020 |
exactly | -0.020 |
Token | books |
Token ID | 3835 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
Blood | +0.037 |
arrest | +0.037 |
blown | +0.036 |
robbers | +0.035 |
growth | +0.035 |
Neg logit contributions | |
Deal | -0.043 |
difference | -0.039 |
isn | -0.037 |
why | -0.037 |
ement | -0.037 |
Token | had |
Token ID | 550 |
Feature activation | -5.34e+00 |
Pos logit contributions | |
Deal | +0.009 |
difference | +0.008 |
ement | +0.007 |
isn | +0.007 |
why | +0.007 |
Neg logit contributions | |
Blood | -0.014 |
arrest | -0.013 |
robbers | -0.013 |
blown | -0.013 |
growth | -0.013 |
Token | stories |
Token ID | 3923 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
Blood | +0.134 |
robbers | +0.126 |
arrest | +0.124 |
blown | +0.122 |
alks | +0.122 |
Neg logit contributions | |
Deal | -0.126 |
isn | -0.121 |
exactly | -0.117 |
why | -0.116 |
difference | -0.113 |
Token | , |
Token ID | 11 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
Blood | +0.043 |
arrest | +0.042 |
robbers | +0.041 |
blown | +0.041 |
alks | +0.040 |
Neg logit contributions | |
Deal | -0.039 |
difference | -0.033 |
decision | -0.031 |
ement | -0.031 |
promise | -0.031 |
Token | some |
Token ID | 617 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Blood | +0.040 |
arrest | +0.039 |
blown | +0.038 |
robbers | +0.038 |
growth | +0.037 |
Neg logit contributions | |
Deal | -0.042 |
difference | -0.037 |
isn | -0.036 |
why | -0.036 |
ement | -0.035 |
Token | books |
Token ID | 3835 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
Blood | +0.035 |
arrest | +0.035 |
robbers | +0.034 |
blown | +0.034 |
growth | +0.033 |
Neg logit contributions | |
Deal | -0.041 |
difference | -0.037 |
ement | -0.035 |
isn | -0.035 |
why | -0.035 |
Token | had |
Token ID | 550 |
Feature activation | -5.34e+00 |
Pos logit contributions | |
Deal | +0.010 |
difference | +0.008 |
ement | +0.008 |
isn | +0.008 |
why | +0.008 |
Neg logit contributions | |
Blood | -0.014 |
arrest | -0.014 |
robbers | -0.014 |
blown | -0.013 |
growth | -0.013 |
Token | replied |
Token ID | 8712 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
difference | +0.016 |
roads | +0.014 |
Deal | +0.014 |
decision | +0.014 |
destinations | +0.014 |
Neg logit contributions | |
spilled | -0.013 |
Blood | -0.012 |
bitten | -0.011 |
mice | -0.011 |
blown | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
Blood | +0.013 |
arrest | +0.012 |
blown | +0.012 |
robbers | +0.012 |
spilled | +0.012 |
Neg logit contributions | |
Deal | -0.008 |
difference | -0.007 |
ement | -0.006 |
decision | -0.006 |
isn | -0.006 |
Token | " |
Token ID | 366 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Deal | +0.011 |
difference | +0.010 |
ement | +0.009 |
isn | +0.009 |
decision | +0.009 |
Neg logit contributions | |
Blood | -0.012 |
arrest | -0.012 |
robbers | -0.011 |
blown | -0.011 |
growth | -0.011 |
Token | Sure |
Token ID | 19457 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
difference | +0.030 |
ement | +0.030 |
Deal | +0.029 |
decision | +0.029 |
exactly | +0.028 |
Neg logit contributions | |
Blood | -0.048 |
arrest | -0.044 |
blown | -0.044 |
mice | -0.044 |
robbers | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
Deal | +0.050 |
difference | +0.043 |
int | +0.042 |
ement | +0.042 |
ieved | +0.040 |
Neg logit contributions | |
arrest | -0.054 |
Blood | -0.054 |
robbers | -0.052 |
mice | -0.051 |
spilled | -0.049 |
Token | but |
Token ID | 475 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
Deal | +0.008 |
ement | +0.007 |
difference | +0.007 |
decision | +0.007 |
int | +0.007 |
Neg logit contributions | |
Blood | -0.006 |
robbers | -0.006 |
arrest | -0.006 |
mice | -0.006 |
spilled | -0.005 |
Token | first |
Token ID | 717 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
Deal | +0.111 |
ement | +0.105 |
difference | +0.101 |
int | +0.095 |
ieved | +0.094 |
Neg logit contributions | |
Blood | -0.104 |
robbers | -0.098 |
blown | -0.098 |
arrest | -0.097 |
mice | -0.096 |
Token | you |
Token ID | 345 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Deal | +0.049 |
ement | +0.046 |
int | +0.045 |
difference | +0.043 |
ieved | +0.042 |
Neg logit contributions | |
mice | -0.045 |
robbers | -0.044 |
arrest | -0.043 |
Blood | -0.043 |
spilled | -0.043 |
Token | need |
Token ID | 761 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
Deal | +0.033 |
difference | +0.029 |
ement | +0.028 |
isn | +0.028 |
why | +0.028 |
Neg logit contributions | |
Blood | -0.023 |
arrest | -0.023 |
robbers | -0.022 |
blown | -0.022 |
growth | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
Deal | +0.013 |
why | +0.012 |
difference | +0.012 |
isn | +0.011 |
exactly | +0.011 |
Neg logit contributions | |
Blood | -0.013 |
arrest | -0.013 |
growth | -0.013 |
plets | -0.013 |
robbers | -0.013 |
Token | yield |
Token ID | 7800 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
Deal | +0.026 |
ement | +0.024 |
difference | +0.023 |
isn | +0.022 |
exactly | +0.022 |
Neg logit contributions | |
arrest | -0.020 |
Blood | -0.019 |
robbers | -0.019 |
blown | -0.018 |
mice | -0.018 |
Token | never |
Token ID | 1239 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.002 |
ement | +0.002 |
decision | +0.002 |
ieved | +0.002 |
Neg logit contributions | |
arrest | -0.002 |
Blood | -0.002 |
robbers | -0.002 |
blown | -0.002 |
spilled | -0.002 |
Token | take |
Token ID | 1011 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
Blood | +0.027 |
arrest | +0.027 |
blown | +0.027 |
robbers | +0.027 |
spilled | +0.026 |
Neg logit contributions | |
Deal | -0.039 |
difference | -0.034 |
ement | -0.034 |
exactly | -0.032 |
decision | -0.032 |
Token | something |
Token ID | 1223 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
Blood | +0.019 |
arrest | +0.019 |
robbers | +0.018 |
blown | +0.018 |
growth | +0.018 |
Neg logit contributions | |
Deal | -0.021 |
difference | -0.019 |
ement | -0.018 |
isn | -0.018 |
why | -0.018 |
Token | that |
Token ID | 326 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
Deal | +0.055 |
difference | +0.045 |
ement | +0.045 |
isn | +0.042 |
decision | +0.042 |
Neg logit contributions | |
blown | -0.051 |
arrest | -0.051 |
Blood | -0.051 |
robbers | -0.050 |
spilled | -0.050 |
Token | doesn |
Token ID | 1595 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Deal | +0.015 |
ement | +0.013 |
difference | +0.012 |
ieved | +0.012 |
exactly | +0.012 |
Neg logit contributions | |
spilled | -0.020 |
Blood | -0.019 |
blown | -0.019 |
arrest | -0.019 |
robbers | -0.018 |
Token | â |
Token ID | 22940 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
Blood | +0.022 |
robbers | +0.022 |
arrest | +0.021 |
growth | +0.021 |
blown | +0.021 |
Neg logit contributions | |
Deal | -0.037 |
difference | -0.034 |
isn | -0.034 |
why | -0.033 |
ement | -0.033 |
Token | € |
Token ID | 26391 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
Deal | +0.071 |
difference | +0.065 |
exactly | +0.063 |
ement | +0.060 |
why | +0.058 |
Neg logit contributions | |
Blood | -0.133 |
arrest | -0.132 |
robbers | -0.128 |
blown | -0.128 |
growth | -0.126 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
Blood | +0.038 |
arrest | +0.037 |
robbers | +0.036 |
blown | +0.036 |
growth | +0.035 |
Neg logit contributions | |
Deal | -0.044 |
difference | -0.039 |
ement | -0.037 |
why | -0.037 |
decision | -0.037 |
Token | t |
Token ID | 83 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
Blood | +0.002 |
arrest | +0.002 |
robbers | +0.002 |
blown | +0.002 |
growth | +0.002 |
Neg logit contributions | |
Deal | -0.004 |
difference | -0.004 |
ement | -0.004 |
isn | -0.004 |
why | -0.004 |
Token | belong |
Token ID | 5594 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
Blood | +0.012 |
arrest | +0.012 |
growth | +0.012 |
robbers | +0.012 |
alks | +0.012 |
Neg logit contributions | |
Deal | -0.009 |
difference | -0.008 |
why | -0.008 |
isn | -0.008 |
promise | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
Deal | +0.021 |
difference | +0.020 |
ement | +0.019 |
decision | +0.018 |
int | +0.018 |
Neg logit contributions | |
blown | -0.012 |
arrest | -0.012 |
Blood | -0.012 |
robbers | -0.011 |
spilled | -0.011 |
Token | said |
Token ID | 531 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Deal | +0.023 |
difference | +0.020 |
ement | +0.019 |
isn | +0.019 |
decision | +0.019 |
Neg logit contributions | |
Blood | -0.028 |
arrest | -0.028 |
robbers | -0.028 |
blown | -0.028 |
growth | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.002 |
decision | +0.002 |
ement | +0.002 |
ieved | +0.002 |
Neg logit contributions | |
Blood | -0.003 |
spilled | -0.003 |
arrest | -0.003 |
blown | -0.003 |
mice | -0.003 |
Token | " |
Token ID | 366 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
Deal | +0.016 |
difference | +0.014 |
ement | +0.013 |
isn | +0.013 |
decision | +0.013 |
Neg logit contributions | |
Blood | -0.017 |
arrest | -0.016 |
robbers | -0.016 |
blown | -0.016 |
growth | -0.016 |
Token | Okay |
Token ID | 16454 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
Deal | +0.036 |
difference | +0.033 |
ement | +0.031 |
decision | +0.030 |
why | +0.030 |
Neg logit contributions | |
Blood | -0.059 |
arrest | -0.057 |
robbers | -0.057 |
blown | -0.057 |
growth | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
Deal | +0.024 |
difference | +0.020 |
ement | +0.020 |
roads | +0.019 |
int | +0.019 |
Neg logit contributions | |
Blood | -0.021 |
robbers | -0.021 |
arrest | -0.021 |
mice | -0.021 |
spilled | -0.020 |
Token | but |
Token ID | 475 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
robbers | +0.008 |
arrest | +0.007 |
mice | +0.007 |
Blood | +0.007 |
blown | +0.007 |
Neg logit contributions | |
Deal | -0.011 |
ement | -0.011 |
difference | -0.010 |
roads | -0.010 |
decision | -0.010 |
Token | don |
Token ID | 836 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
ement | +0.109 |
Deal | +0.108 |
difference | +0.098 |
ieved | +0.097 |
uct | +0.093 |
Neg logit contributions | |
robbers | -0.084 |
Blood | -0.084 |
arrest | -0.081 |
blown | -0.080 |
mice | -0.079 |
Token | 't |
Token ID | 470 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
Blood | +0.008 |
robbers | +0.007 |
arrest | +0.007 |
growth | +0.007 |
blown | +0.007 |
Neg logit contributions | |
Deal | -0.014 |
difference | -0.013 |
isn | -0.013 |
ement | -0.013 |
why | -0.013 |
Token | forget |
Token ID | 6044 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
Deal | +0.004 |
difference | +0.004 |
ement | +0.003 |
isn | +0.003 |
why | +0.003 |
Neg logit contributions | |
Blood | -0.004 |
arrest | -0.004 |
robbers | -0.004 |
blown | -0.004 |
growth | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
Deal | +0.024 |
ement | +0.021 |
difference | +0.021 |
decision | +0.020 |
ieved | +0.020 |
Neg logit contributions | |
arrest | -0.011 |
Blood | -0.011 |
mice | -0.011 |
robbers | -0.011 |
spilled | -0.010 |
Token | put |
Token ID | 1234 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Deal | +0.034 |
ement | +0.032 |
difference | +0.030 |
isn | +0.029 |
exactly | +0.028 |
Neg logit contributions | |
arrest | -0.029 |
robbers | -0.028 |
Blood | -0.027 |
blown | -0.027 |
growth | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +7.10e-01 |
Pos logit contributions | |
arrest | +0.002 |
mice | +0.002 |
blown | +0.002 |
spilled | +0.002 |
alks | +0.002 |
Neg logit contributions | |
Deal | -0.004 |
difference | -0.003 |
ement | -0.003 |
ieved | -0.003 |
int | -0.003 |
Token | â |
Token ID | 22940 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
Deal | +0.021 |
ement | +0.018 |
difference | +0.017 |
ieved | +0.016 |
decision | +0.016 |
Neg logit contributions | |
arrest | -0.016 |
robbers | -0.016 |
blown | -0.015 |
spilled | -0.015 |
mice | -0.015 |
Token | € |
Token ID | 26391 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
exactly | +0.097 |
difference | +0.094 |
Deal | +0.090 |
amaz | +0.088 |
why | +0.087 |
Neg logit contributions | |
Blood | -0.145 |
arrest | -0.136 |
plets | -0.135 |
blown | -0.132 |
growth | -0.132 |
Token | said |
Token ID | 531 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
Deal | +0.005 |
difference | +0.005 |
ement | +0.004 |
decision | +0.004 |
why | +0.004 |
Neg logit contributions | |
arrest | -0.005 |
Blood | -0.005 |
robbers | -0.004 |
blown | -0.004 |
growth | -0.004 |
Token | Ali |
Token ID | 12104 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Deal | +0.026 |
difference | +0.023 |
ieved | +0.022 |
ement | +0.022 |
uct | +0.020 |
Neg logit contributions | |
Blood | -0.018 |
arrest | -0.017 |
spilled | -0.017 |
blown | -0.017 |
Po | -0.017 |
Token | â |
Token ID | 22940 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
Deal | +0.085 |
difference | +0.073 |
ement | +0.071 |
ieved | +0.066 |
decision | +0.065 |
Neg logit contributions | |
arrest | -0.069 |
Blood | -0.069 |
robbers | -0.069 |
mice | -0.067 |
blown | -0.067 |
Token | € |
Token ID | 26391 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
exactly | +0.111 |
difference | +0.106 |
Deal | +0.101 |
why | +0.099 |
amaz | +0.099 |
Neg logit contributions | |
Blood | -0.151 |
robbers | -0.144 |
arrest | -0.142 |
growth | -0.139 |
blown | -0.137 |
Token | ™ |
Token ID | 8151 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
Deal | +0.018 |
difference | +0.017 |
ement | +0.016 |
decision | +0.016 |
why | +0.016 |
Neg logit contributions | |
arrest | -0.014 |
Blood | -0.014 |
robbers | -0.013 |
growth | -0.012 |
blown | -0.012 |
Token | s |
Token ID | 82 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
Deal | +0.070 |
difference | +0.063 |
ement | +0.061 |
isn | +0.060 |
why | +0.060 |
Neg logit contributions | |
Blood | -0.040 |
arrest | -0.039 |
robbers | -0.037 |
blown | -0.037 |
growth | -0.036 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
Deal | +0.052 |
ement | +0.043 |
ieved | +0.042 |
difference | +0.041 |
uct | +0.040 |
Neg logit contributions | |
Blood | -0.055 |
arrest | -0.055 |
alks | -0.054 |
robbers | -0.053 |
blown | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Deal | +0.006 |
difference | +0.005 |
ement | +0.005 |
decision | +0.005 |
deal | +0.005 |
Neg logit contributions | |
robbers | -0.004 |
Blood | -0.004 |
arrest | -0.004 |
spilled | -0.004 |
mice | -0.004 |
Token | today |
Token ID | 1909 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
Deal | +0.061 |
difference | +0.058 |
ement | +0.056 |
decision | +0.054 |
ieved | +0.051 |
Neg logit contributions | |
Blood | -0.043 |
robbers | -0.042 |
spilled | -0.040 |
mice | -0.038 |
arrest | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
Deal | +0.012 |
ieved | +0.010 |
ement | +0.010 |
uct | +0.010 |
int | +0.010 |
Neg logit contributions | |
spilled | -0.010 |
Blood | -0.010 |
arrest | -0.009 |
mice | -0.009 |
robbers | -0.009 |
Token | they |
Token ID | 484 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
Deal | +0.013 |
difference | +0.012 |
ement | +0.011 |
decision | +0.011 |
promise | +0.011 |
Neg logit contributions | |
robbers | -0.007 |
Blood | -0.007 |
arrest | -0.007 |
mice | -0.007 |
blown | -0.007 |
Token | have |
Token ID | 423 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
arrest | +0.008 |
robbers | +0.008 |
Blood | +0.008 |
spilled | +0.008 |
blown | +0.008 |
Neg logit contributions | |
Deal | -0.010 |
ement | -0.009 |
difference | -0.009 |
exactly | -0.008 |
decision | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
Blood | +0.039 |
plets | +0.038 |
alks | +0.038 |
growth | +0.038 |
arrest | +0.038 |
Neg logit contributions | |
Deal | -0.041 |
why | -0.037 |
isn | -0.037 |
exactly | -0.036 |
difference | -0.036 |
Token | un |
Token ID | 555 |
Feature activation | +3.93e+00 |
Pos logit contributions | |
Deal | +0.014 |
ement | +0.013 |
difference | +0.012 |
isn | +0.012 |
exactly | +0.011 |
Neg logit contributions | |
arrest | -0.012 |
robbers | -0.012 |
Blood | -0.012 |
growth | -0.011 |
blown | -0.011 |
Token | pack |
Token ID | 8002 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
difference | +0.119 |
Deal | +0.117 |
isn | +0.113 |
ement | +0.111 |
decision | +0.109 |
Neg logit contributions | |
robbers | -0.076 |
Blood | -0.074 |
arrest | -0.074 |
blown | -0.068 |
mice | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
Deal | +0.051 |
difference | +0.046 |
ieved | +0.043 |
ement | +0.043 |
uct | +0.042 |
Neg logit contributions | |
spilled | -0.024 |
Blood | -0.023 |
mice | -0.023 |
robbers | -0.023 |
alks | -0.022 |
Token | Their |
Token ID | 5334 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
arrest | +0.013 |
Blood | +0.013 |
blown | +0.012 |
robbers | +0.012 |
growth | +0.012 |
Neg logit contributions | |
Deal | -0.016 |
difference | -0.014 |
isn | -0.014 |
ement | -0.014 |
why | -0.013 |
Token | family |
Token ID | 1641 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
Deal | +0.030 |
ieved | +0.028 |
ement | +0.027 |
difference | +0.027 |
uct | +0.026 |
Neg logit contributions | |
robbers | -0.017 |
Blood | -0.017 |
spilled | -0.016 |
mice | -0.016 |
arrest | -0.015 |
Token | moved |
Token ID | 3888 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
spilled | +0.015 |
robbers | +0.014 |
arrest | +0.014 |
mice | +0.013 |
blown | +0.013 |
Neg logit contributions | |
Deal | -0.013 |
ement | -0.012 |
difference | -0.011 |
ieved | -0.011 |
uct | -0.011 |
Token | bed |
Token ID | 3996 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
difference | +0.023 |
promise | +0.021 |
decision | +0.020 |
Deal | +0.019 |
wish | +0.019 |
Neg logit contributions | |
plets | -0.029 |
Blood | -0.028 |
blown | -0.028 |
robbers | -0.027 |
bitten | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
Deal | +0.008 |
ement | +0.007 |
exactly | +0.007 |
difference | +0.007 |
decision | +0.006 |
Neg logit contributions | |
robbers | -0.013 |
arrest | -0.013 |
Blood | -0.013 |
growth | -0.013 |
blown | -0.013 |
Token | The |
Token ID | 383 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
arrest | +0.037 |
blown | +0.037 |
robbers | +0.036 |
growth | +0.035 |
Blood | +0.035 |
Neg logit contributions | |
Deal | -0.048 |
isn | -0.042 |
difference | -0.041 |
why | -0.041 |
decision | -0.039 |
Token | gem |
Token ID | 16840 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
arrest | +0.013 |
Blood | +0.013 |
growth | +0.013 |
robbers | +0.012 |
blown | +0.012 |
Neg logit contributions | |
Deal | -0.014 |
difference | -0.013 |
isn | -0.012 |
ement | -0.012 |
decision | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Blood | +0.029 |
arrest | +0.029 |
robbers | +0.028 |
blown | +0.028 |
growth | +0.028 |
Neg logit contributions | |
Deal | -0.026 |
difference | -0.023 |
ement | -0.022 |
isn | -0.022 |
why | -0.021 |
Token | shiny |
Token ID | 22441 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
Blood | +0.027 |
blown | +0.026 |
arrest | +0.026 |
robbers | +0.026 |
spilled | +0.025 |
Neg logit contributions | |
Deal | -0.028 |
difference | -0.024 |
ement | -0.023 |
isn | -0.023 |
decision | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -8.62e-01 |
Pos logit contributions | |
Deal | +0.023 |
difference | +0.021 |
ement | +0.020 |
decision | +0.020 |
why | +0.019 |
Neg logit contributions | |
Blood | -0.019 |
arrest | -0.018 |
blown | -0.018 |
robbers | -0.018 |
mice | -0.018 |
Token | blue |
Token ID | 4171 |
Feature activation | +6.10e-02 |
Pos logit contributions | |
Blood | +0.017 |
blown | +0.016 |
robbers | +0.016 |
spilled | +0.015 |
mice | +0.015 |
Neg logit contributions | |
Deal | -0.027 |
difference | -0.025 |
ement | -0.024 |
decision | -0.024 |
ieved | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Deal | +0.001 |
isn | +0.001 |
ement | +0.001 |
difference | +0.001 |
why | +0.001 |
Neg logit contributions | |
arrest | -0.002 |
Blood | -0.002 |
robbers | -0.002 |
growth | -0.002 |
alks | -0.002 |
Token | Kitty |
Token ID | 21168 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
arrest | +0.031 |
blown | +0.031 |
growth | +0.030 |
robbers | +0.030 |
Blood | +0.030 |
Neg logit contributions | |
Deal | -0.038 |
isn | -0.033 |
difference | -0.032 |
why | -0.032 |
ement | -0.031 |
Token | showed |
Token ID | 3751 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
Blood | +0.007 |
arrest | +0.007 |
robbers | +0.007 |
blown | +0.007 |
spilled | +0.007 |
Neg logit contributions | |
Deal | -0.008 |
difference | -0.007 |
ement | -0.007 |
decision | -0.006 |
exactly | -0.006 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
Deal | +0.044 |
difference | +0.038 |
ement | +0.038 |
decision | +0.037 |
promise | +0.037 |
Neg logit contributions | |
Blood | -0.038 |
arrest | -0.037 |
growth | -0.036 |
robbers | -0.036 |
alks | -0.035 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
Deal | +0.021 |
difference | +0.019 |
ement | +0.018 |
isn | +0.018 |
why | +0.018 |
Neg logit contributions | |
Blood | -0.023 |
arrest | -0.022 |
robbers | -0.022 |
blown | -0.022 |
growth | -0.021 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
Deal | +0.051 |
difference | +0.046 |
ement | +0.044 |
isn | +0.044 |
why | +0.043 |
Neg logit contributions | |
Blood | -0.037 |
arrest | -0.036 |
robbers | -0.035 |
blown | -0.035 |
growth | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Deal | +0.004 |
why | +0.003 |
difference | +0.003 |
exactly | +0.003 |
isn | +0.003 |
Neg logit contributions | |
alks | -0.003 |
arrest | -0.003 |
blown | -0.003 |
robbers | -0.003 |
Blood | -0.003 |
Token | time |
Token ID | 640 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Deal | +0.032 |
difference | +0.029 |
ement | +0.029 |
why | +0.028 |
isn | +0.028 |
Neg logit contributions | |
Blood | -0.029 |
arrest | -0.027 |
spilled | -0.027 |
growth | -0.026 |
robbers | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.003 |
decision | +0.003 |
ement | +0.003 |
promise | +0.003 |
Neg logit contributions | |
Blood | -0.004 |
arrest | -0.004 |
robbers | -0.004 |
spilled | -0.004 |
blown | -0.004 |
Token | there |
Token ID | 612 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
Deal | +0.028 |
difference | +0.025 |
ement | +0.024 |
isn | +0.024 |
decision | +0.024 |
Neg logit contributions | |
Blood | -0.024 |
arrest | -0.023 |
robbers | -0.023 |
blown | -0.023 |
growth | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
Blood | +0.001 |
spilled | +0.001 |
arrest | +0.001 |
growth | +0.001 |
blown | +0.001 |
Neg logit contributions | |
Deal | -0.001 |
difference | -0.001 |
int | -0.001 |
ieved | -0.001 |
izes | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
alks | +0.042 |
plets | +0.041 |
arrest | +0.041 |
robbers | +0.040 |
blown | +0.040 |
Neg logit contributions | |
Deal | -0.043 |
exactly | -0.037 |
why | -0.036 |
equally | -0.035 |
difference | -0.035 |
Token | rich |
Token ID | 5527 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.003 |
ement | +0.003 |
isn | +0.003 |
why | +0.003 |
Neg logit contributions | |
Blood | -0.002 |
arrest | -0.002 |
blown | -0.002 |
robbers | -0.002 |
growth | -0.002 |
Token | man |
Token ID | 582 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
Deal | +0.016 |
difference | +0.015 |
decision | +0.014 |
promise | +0.013 |
ement | +0.013 |
Neg logit contributions | |
Blood | -0.018 |
growth | -0.017 |
blown | -0.017 |
arrest | -0.017 |
alks | -0.016 |
Token | M |
Token ID | 44 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
Deal | +0.018 |
difference | +0.016 |
ement | +0.015 |
isn | +0.015 |
why | +0.015 |
Neg logit contributions | |
Blood | -0.012 |
arrest | -0.011 |
robbers | -0.011 |
blown | -0.011 |
growth | -0.011 |
Token | ummy |
Token ID | 13513 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
Deal | +0.079 |
difference | +0.068 |
isn | +0.064 |
why | +0.063 |
decision | +0.062 |
Neg logit contributions | |
Blood | -0.069 |
alks | -0.068 |
arrest | -0.067 |
robbers | -0.065 |
spilled | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
Deal | +0.078 |
difference | +0.069 |
ement | +0.066 |
decision | +0.065 |
isn | +0.064 |
Neg logit contributions | |
Blood | -0.069 |
arrest | -0.068 |
robbers | -0.066 |
blown | -0.066 |
growth | -0.065 |
Token | Daddy |
Token ID | 26926 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Blood | +0.003 |
arrest | +0.003 |
blown | +0.003 |
robbers | +0.003 |
growth | +0.003 |
Neg logit contributions | |
Deal | -0.004 |
difference | -0.004 |
why | -0.004 |
isn | -0.003 |
ement | -0.003 |
Token | took |
Token ID | 1718 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
Deal | +0.050 |
difference | +0.044 |
ement | +0.043 |
decision | +0.042 |
isn | +0.042 |
Neg logit contributions | |
Blood | -0.035 |
arrest | -0.035 |
robbers | -0.034 |
blown | -0.034 |
growth | -0.033 |
Token | Peter |
Token ID | 5613 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Blood | +0.014 |
arrest | +0.014 |
blown | +0.014 |
robbers | +0.014 |
growth | +0.013 |
Neg logit contributions | |
Deal | -0.016 |
difference | -0.015 |
ement | -0.014 |
isn | -0.014 |
why | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
Deal | +0.034 |
difference | +0.030 |
ement | +0.029 |
why | +0.029 |
isn | +0.029 |
Neg logit contributions | |
Blood | -0.026 |
arrest | -0.025 |
robbers | -0.025 |
blown | -0.024 |
growth | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
arrest | +0.004 |
Blood | +0.004 |
robbers | +0.004 |
growth | +0.004 |
blown | +0.004 |
Neg logit contributions | |
Deal | -0.008 |
difference | -0.007 |
ement | -0.007 |
isn | -0.007 |
promise | -0.007 |
Token | special |
Token ID | 2041 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
Deal | +0.034 |
difference | +0.029 |
ement | +0.029 |
isn | +0.028 |
exactly | +0.028 |
Neg logit contributions | |
Blood | -0.022 |
arrest | -0.022 |
robbers | -0.021 |
blown | -0.021 |
growth | -0.021 |
Token | adventure |
Token ID | 8855 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
Deal | +0.100 |
ement | +0.084 |
exactly | +0.082 |
difference | +0.082 |
isn | +0.081 |
Neg logit contributions | |
robbers | -0.081 |
arrest | -0.081 |
Blood | -0.079 |
spilled | -0.078 |
growth | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
Deal | +0.053 |
difference | +0.048 |
ement | +0.045 |
isn | +0.045 |
promise | +0.044 |
Neg logit contributions | |
arrest | -0.037 |
robbers | -0.036 |
blown | -0.036 |
Blood | -0.035 |
alks | -0.035 |
Token | knew |
Token ID | 2993 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
spilled | +0.015 |
Blood | +0.015 |
arrest | +0.015 |
blown | +0.015 |
robbers | +0.014 |
Neg logit contributions | |
Deal | -0.019 |
difference | -0.018 |
ement | -0.017 |
decision | -0.017 |
why | -0.016 |
Token | her |
Token ID | 607 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Deal | +0.013 |
difference | +0.012 |
decision | +0.012 |
ement | +0.011 |
promise | +0.011 |
Neg logit contributions | |
arrest | -0.009 |
robbers | -0.009 |
Blood | -0.009 |
blown | -0.009 |
mice | -0.009 |
Token | dad |
Token ID | 9955 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
Deal | +0.055 |
difference | +0.048 |
exactly | +0.047 |
ieved | +0.047 |
ement | +0.047 |
Neg logit contributions | |
robbers | -0.027 |
arrest | -0.025 |
Blood | -0.024 |
spilled | -0.024 |
blown | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
spilled | +0.021 |
mice | +0.020 |
blown | +0.019 |
Blood | +0.019 |
arrest | +0.019 |
Neg logit contributions | |
Deal | -0.022 |
difference | -0.021 |
int | -0.020 |
ement | -0.019 |
decision | -0.019 |
Token | right |
Token ID | 826 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
Deal | +0.039 |
difference | +0.033 |
ement | +0.032 |
isn | +0.032 |
decision | +0.032 |
Neg logit contributions | |
blown | -0.048 |
Blood | -0.047 |
robbers | -0.047 |
arrest | -0.046 |
mice | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
Deal | +0.117 |
difference | +0.109 |
ieved | +0.102 |
isn | +0.102 |
olog | +0.101 |
Neg logit contributions | |
mice | -0.049 |
robbers | -0.048 |
spilled | -0.047 |
bushes | -0.046 |
blown | -0.044 |
Token | She |
Token ID | 1375 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
Deal | +0.023 |
difference | +0.020 |
ement | +0.020 |
decision | +0.020 |
promise | +0.019 |
Neg logit contributions | |
Blood | -0.013 |
arrest | -0.013 |
robbers | -0.013 |
growth | -0.012 |
blown | -0.012 |
Token | still |
Token ID | 991 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
spilled | +0.006 |
blown | +0.006 |
Blood | +0.006 |
arrest | +0.006 |
robbers | +0.006 |
Neg logit contributions | |
Deal | -0.007 |
difference | -0.007 |
ement | -0.006 |
decision | -0.006 |
isn | -0.006 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
Deal | +0.009 |
ement | +0.009 |
difference | +0.009 |
decision | +0.008 |
int | +0.008 |
Neg logit contributions | |
Blood | -0.006 |
spilled | -0.006 |
blown | -0.006 |
robbers | -0.006 |
arrest | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
Blood | +0.012 |
arrest | +0.012 |
robbers | +0.012 |
blown | +0.011 |
mice | +0.011 |
Neg logit contributions | |
Deal | -0.016 |
difference | -0.015 |
ement | -0.014 |
isn | -0.014 |
decision | -0.014 |
Token | run |
Token ID | 1057 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
Deal | +0.003 |
ement | +0.002 |
difference | +0.002 |
exactly | +0.002 |
isn | +0.002 |
Neg logit contributions | |
arrest | -0.002 |
robbers | -0.002 |
Blood | -0.002 |
blown | -0.002 |
growth | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
Deal | +0.024 |
difference | +0.021 |
isn | +0.020 |
decision | +0.019 |
promise | +0.019 |
Neg logit contributions | |
Blood | -0.018 |
blown | -0.017 |
arrest | -0.017 |
growth | -0.017 |
alks | -0.017 |
Token | sw |
Token ID | 1509 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
Deal | +0.013 |
difference | +0.011 |
ement | +0.011 |
decision | +0.011 |
isn | +0.011 |
Neg logit contributions | |
Blood | -0.009 |
arrest | -0.009 |
blown | -0.009 |
robbers | -0.009 |
growth | -0.009 |
Token | am |
Token ID | 321 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
Deal | +0.049 |
difference | +0.044 |
why | +0.040 |
ement | +0.039 |
decision | +0.039 |
Neg logit contributions | |
Blood | -0.048 |
alks | -0.047 |
arrest | -0.046 |
blown | -0.045 |
growth | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Deal | +0.012 |
difference | +0.010 |
ement | +0.010 |
uct | +0.009 |
aman | +0.009 |
Neg logit contributions | |
Blood | -0.007 |
blown | -0.006 |
spilled | -0.006 |
robbers | -0.006 |
alks | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Deal | +0.040 |
difference | +0.036 |
why | +0.035 |
exactly | +0.034 |
isn | +0.034 |
Neg logit contributions | |
Blood | -0.037 |
arrest | -0.036 |
blown | -0.036 |
robbers | -0.035 |
alks | -0.035 |
Token | pearl |
Token ID | 43836 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Deal | +0.036 |
ement | +0.031 |
difference | +0.030 |
isn | +0.030 |
exactly | +0.029 |
Neg logit contributions | |
arrest | -0.025 |
robbers | -0.025 |
Blood | -0.025 |
blown | -0.024 |
growth | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
Deal | +0.050 |
difference | +0.042 |
isn | +0.040 |
ieved | +0.040 |
ement | +0.040 |
Neg logit contributions | |
alks | -0.031 |
blown | -0.031 |
Blood | -0.031 |
arrest | -0.031 |
spilled | -0.030 |
Token | It |
Token ID | 632 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
Blood | +0.016 |
arrest | +0.016 |
blown | +0.016 |
robbers | +0.016 |
growth | +0.015 |
Neg logit contributions | |
Deal | -0.019 |
difference | -0.017 |
isn | -0.017 |
ement | -0.017 |
why | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
blown | +0.004 |
spilled | +0.004 |
Blood | +0.004 |
arrest | +0.003 |
growth | +0.003 |
Neg logit contributions | |
Deal | -0.005 |
difference | -0.004 |
ement | -0.004 |
decision | -0.004 |
isn | -0.004 |
Token | hard |
Token ID | 1327 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
blown | +0.009 |
Blood | +0.009 |
spilled | +0.009 |
robbers | +0.008 |
mice | +0.008 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.009 |
ement | -0.009 |
decision | -0.008 |
promise | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
Deal | +0.027 |
difference | +0.025 |
isn | +0.024 |
exactly | +0.023 |
ement | +0.023 |
Neg logit contributions | |
Blood | -0.029 |
arrest | -0.028 |
robbers | -0.028 |
growth | -0.028 |
blown | -0.027 |
Token | were |
Token ID | 547 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
Deal | +0.036 |
difference | +0.033 |
ement | +0.032 |
decision | +0.031 |
exactly | +0.031 |
Neg logit contributions | |
arrest | -0.023 |
Blood | -0.022 |
robbers | -0.022 |
blown | -0.022 |
spilled | -0.022 |
Token | brother |
Token ID | 3956 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
Deal | +0.023 |
difference | +0.020 |
ement | +0.019 |
isn | +0.019 |
why | +0.019 |
Neg logit contributions | |
Blood | -0.019 |
arrest | -0.018 |
robbers | -0.018 |
blown | -0.018 |
growth | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
robbers | +0.039 |
arrest | +0.038 |
mice | +0.038 |
Blood | +0.038 |
spilled | +0.036 |
Neg logit contributions | |
Deal | -0.053 |
difference | -0.047 |
decision | -0.045 |
ement | -0.044 |
isn | -0.044 |
Token | sister |
Token ID | 6621 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
growth | +0.053 |
plets | +0.052 |
Blood | +0.052 |
alks | +0.052 |
blown | +0.051 |
Neg logit contributions | |
Deal | -0.047 |
why | -0.041 |
promise | -0.038 |
exactly | -0.037 |
replied | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
robbers | +0.027 |
arrest | +0.026 |
spilled | +0.026 |
mice | +0.026 |
Blood | +0.025 |
Neg logit contributions | |
Deal | -0.042 |
difference | -0.037 |
ement | -0.036 |
decision | -0.036 |
ieved | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
arrest | +0.020 |
Blood | +0.020 |
blown | +0.020 |
robbers | +0.019 |
growth | +0.019 |
Neg logit contributions | |
Deal | -0.033 |
difference | -0.029 |
ement | -0.028 |
isn | -0.028 |
why | -0.028 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Blood | +0.028 |
arrest | +0.028 |
robbers | +0.027 |
blown | +0.027 |
growth | +0.027 |
Neg logit contributions | |
Deal | -0.024 |
difference | -0.021 |
ement | -0.020 |
isn | -0.020 |
why | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
Blood | +0.027 |
arrest | +0.026 |
growth | +0.026 |
blown | +0.025 |
plets | +0.025 |
Neg logit contributions | |
Deal | -0.032 |
difference | -0.029 |
why | -0.029 |
isn | -0.028 |
exactly | -0.028 |
Token | play |
Token ID | 711 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
Blood | +0.009 |
arrest | +0.009 |
blown | +0.008 |
robbers | +0.008 |
growth | +0.008 |
Neg logit contributions | |
Deal | -0.009 |
difference | -0.008 |
ement | -0.008 |
isn | -0.008 |
why | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
Blood | +0.002 |
mice | +0.002 |
robbers | +0.002 |
arrest | +0.001 |
spilled | +0.001 |
Neg logit contributions | |
Deal | -0.002 |
difference | -0.002 |
decision | -0.002 |
ement | -0.002 |
uct | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
Deal | +0.001 |
difference | +0.001 |
why | +0.001 |
isn | +0.001 |
decision | +0.001 |
Neg logit contributions | |
plets | -0.002 |
Blood | -0.002 |
blown | -0.002 |
arrest | -0.002 |
alks | -0.002 |
Token | way |
Token ID | 835 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
Deal | +0.037 |
ement | +0.031 |
difference | +0.030 |
exactly | +0.030 |
isn | +0.030 |
Neg logit contributions | |
arrest | -0.041 |
robbers | -0.039 |
Blood | -0.039 |
blown | -0.039 |
mice | -0.038 |
Token | home |
Token ID | 1363 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Deal | +0.008 |
difference | +0.007 |
ement | +0.007 |
isn | +0.007 |
why | +0.007 |
Neg logit contributions | |
Blood | -0.011 |
arrest | -0.011 |
robbers | -0.011 |
blown | -0.011 |
growth | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
Blood | +0.035 |
arrest | +0.034 |
spilled | +0.034 |
blown | +0.034 |
robbers | +0.033 |
Neg logit contributions | |
Deal | -0.027 |
difference | -0.023 |
ement | -0.021 |
isn | -0.020 |
decision | -0.020 |
Token | they |
Token ID | 484 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
Deal | +0.017 |
difference | +0.016 |
ement | +0.015 |
isn | +0.014 |
decision | +0.014 |
Neg logit contributions | |
Blood | -0.014 |
arrest | -0.014 |
robbers | -0.013 |
blown | -0.013 |
spilled | -0.013 |
Token | got |
Token ID | 1392 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
Blood | +0.005 |
arrest | +0.005 |
robbers | +0.004 |
blown | +0.004 |
growth | +0.004 |
Neg logit contributions | |
Deal | -0.006 |
difference | -0.005 |
ement | -0.005 |
isn | -0.005 |
why | -0.005 |
Token | into |
Token ID | 656 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
growth | +0.023 |
arrest | +0.022 |
Blood | +0.022 |
robbers | +0.022 |
alks | +0.021 |
Neg logit contributions | |
Deal | -0.021 |
exactly | -0.020 |
difference | -0.019 |
why | -0.019 |
promise | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
Deal | +0.011 |
difference | +0.011 |
decision | +0.011 |
promise | +0.010 |
exactly | +0.010 |
Neg logit contributions | |
Blood | -0.012 |
robbers | -0.012 |
plets | -0.012 |
blown | -0.012 |
arrest | -0.012 |
Token | car |
Token ID | 1097 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
Deal | +0.026 |
ement | +0.023 |
ieved | +0.022 |
uct | +0.021 |
isn | +0.021 |
Neg logit contributions | |
spilled | -0.015 |
robbers | -0.015 |
arrest | -0.014 |
Blood | -0.014 |
mice | -0.014 |
Token | accident |
Token ID | 5778 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
arrest | +0.003 |
blown | +0.003 |
robbers | +0.003 |
Blood | +0.003 |
spilled | +0.003 |
Neg logit contributions | |
Deal | -0.005 |
difference | -0.004 |
ement | -0.004 |
isn | -0.004 |
decision | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
Deal | +0.008 |
difference | +0.007 |
ement | +0.007 |
isn | +0.007 |
exactly | +0.007 |
Neg logit contributions | |
Blood | -0.010 |
arrest | -0.010 |
robbers | -0.010 |
growth | -0.010 |
alks | -0.010 |
Token | Sarah |
Token ID | 10490 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
arrest | +0.008 |
blown | +0.008 |
robbers | +0.008 |
growth | +0.008 |
alks | +0.008 |
Neg logit contributions | |
Deal | -0.006 |
isn | -0.006 |
difference | -0.006 |
ement | -0.006 |
why | -0.005 |
Token | Mark |
Token ID | 2940 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
arrest | +0.047 |
Blood | +0.047 |
robbers | +0.046 |
mice | +0.044 |
growth | +0.044 |
Neg logit contributions | |
Deal | -0.047 |
difference | -0.042 |
ement | -0.041 |
decision | -0.039 |
exactly | -0.039 |
Token | think |
Token ID | 892 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
arrest | +0.004 |
robbers | +0.003 |
blown | +0.003 |
spilled | +0.003 |
growth | +0.003 |
Neg logit contributions | |
Deal | -0.005 |
ement | -0.004 |
exactly | -0.004 |
decision | -0.004 |
difference | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
Blood | +0.015 |
arrest | +0.015 |
blown | +0.015 |
robbers | +0.015 |
mice | +0.015 |
Neg logit contributions | |
Deal | -0.017 |
difference | -0.014 |
ement | -0.014 |
decision | -0.014 |
exactly | -0.013 |
Token | is |
Token ID | 318 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
Deal | +0.039 |
ement | +0.036 |
ieved | +0.036 |
uct | +0.033 |
difference | +0.032 |
Neg logit contributions | |
mice | -0.047 |
Blood | -0.044 |
spilled | -0.044 |
robbers | -0.044 |
blown | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Deal | +0.052 |
ement | +0.047 |
difference | +0.047 |
decision | +0.046 |
ieved | +0.045 |
Neg logit contributions | |
mice | -0.025 |
Blood | -0.025 |
robbers | -0.024 |
blown | -0.024 |
spilled | -0.023 |
Token | good |
Token ID | 922 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
ement | +0.039 |
Deal | +0.039 |
uct | +0.038 |
ieved | +0.037 |
correctly | +0.035 |
Neg logit contributions | |
robbers | -0.035 |
blown | -0.033 |
mice | -0.033 |
Blood | -0.032 |
spilled | -0.032 |
Token | name |
Token ID | 1438 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
Deal | +0.010 |
uct | +0.009 |
ement | +0.009 |
ieved | +0.009 |
exactly | +0.009 |
Neg logit contributions | |
robbers | -0.009 |
blown | -0.009 |
spilled | -0.009 |
arrest | -0.009 |
Blood | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
mice | +0.047 |
Blood | +0.046 |
robbers | +0.046 |
arrest | +0.046 |
blown | +0.045 |
Neg logit contributions | |
Deal | -0.059 |
difference | -0.054 |
decision | -0.051 |
ement | -0.050 |
exactly | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | +2.70e-03 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.013 |
decision | +0.012 |
ement | +0.012 |
exactly | +0.012 |
Neg logit contributions | |
Blood | -0.007 |
robbers | -0.006 |
arrest | -0.006 |
alks | -0.006 |
mice | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
Deal | +0.000 |
difference | +0.000 |
decision | +0.000 |
ement | +0.000 |
exactly | +0.000 |
Neg logit contributions | |
Blood | -0.000 |
growth | -0.000 |
arrest | -0.000 |
alks | -0.000 |
blown | -0.000 |
Token | " |
Token ID | 1 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
Blood | +0.017 |
arrest | +0.016 |
robbers | +0.016 |
growth | +0.015 |
mice | +0.015 |
Neg logit contributions | |
Deal | -0.020 |
difference | -0.020 |
ement | -0.019 |
decision | -0.019 |
exactly | -0.018 |
Token | if |
Token ID | 611 |
Feature activation | +9.35e-01 |
Pos logit contributions | |
Deal | +0.012 |
difference | +0.011 |
ement | +0.010 |
decision | +0.010 |
why | +0.010 |
Neg logit contributions | |
Blood | -0.012 |
arrest | -0.011 |
blown | -0.011 |
robbers | -0.011 |
mice | -0.011 |
Token | we |
Token ID | 356 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
Deal | +0.022 |
difference | +0.019 |
isn | +0.019 |
why | +0.019 |
exactly | +0.018 |
Neg logit contributions | |
Blood | -0.025 |
arrest | -0.024 |
growth | -0.024 |
robbers | -0.024 |
blown | -0.024 |
Token | find |
Token ID | 1064 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
Blood | +0.008 |
arrest | +0.007 |
growth | +0.007 |
alks | +0.007 |
robbers | +0.007 |
Neg logit contributions | |
Deal | -0.006 |
promise | -0.005 |
difference | -0.005 |
why | -0.005 |
isn | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
Deal | +0.000 |
difference | +0.000 |
ement | +0.000 |
isn | +0.000 |
why | +0.000 |
Neg logit contributions | |
Blood | -0.001 |
arrest | -0.001 |
robbers | -0.001 |
blown | -0.001 |
growth | -0.001 |
Token | ." |
Token ID | 526 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
Blood | +0.008 |
arrest | +0.008 |
robbers | +0.008 |
blown | +0.008 |
growth | +0.008 |
Neg logit contributions | |
Deal | -0.007 |
difference | -0.006 |
ement | -0.006 |
isn | -0.006 |
why | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Deal | +0.013 |
difference | +0.011 |
isn | +0.011 |
ement | +0.010 |
why | +0.010 |
Neg logit contributions | |
blown | -0.016 |
arrest | -0.016 |
robbers | -0.016 |
Blood | -0.016 |
growth | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
Deal | +0.005 |
difference | +0.004 |
ement | +0.004 |
decision | +0.004 |
isn | +0.004 |
Neg logit contributions | |
Blood | -0.005 |
arrest | -0.004 |
robbers | -0.004 |
blown | -0.004 |
growth | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
Blood | +0.005 |
arrest | +0.005 |
robbers | +0.005 |
blown | +0.005 |
mice | +0.005 |
Neg logit contributions | |
Deal | -0.006 |
difference | -0.006 |
ement | -0.006 |
decision | -0.005 |
isn | -0.005 |
Token | boy |
Token ID | 2933 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
Deal | +0.006 |
difference | +0.005 |
ement | +0.005 |
isn | +0.005 |
why | +0.005 |
Neg logit contributions | |
Blood | -0.006 |
arrest | -0.006 |
blown | -0.006 |
robbers | -0.006 |
growth | -0.006 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
Blood | +0.017 |
spilled | +0.017 |
mice | +0.016 |
arrest | +0.016 |
blown | +0.016 |
Neg logit contributions | |
Deal | -0.022 |
difference | -0.020 |
ement | -0.019 |
decision | -0.018 |
isn | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
Blood | +0.003 |
robbers | +0.003 |
blown | +0.003 |
arrest | +0.003 |
mice | +0.003 |
Neg logit contributions | |
Deal | -0.005 |
difference | -0.004 |
isn | -0.004 |
ement | -0.004 |
decision | -0.004 |
Token | there |
Token ID | 612 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
Deal | +0.035 |
difference | +0.031 |
ement | +0.030 |
decision | +0.029 |
isn | +0.029 |
Neg logit contributions | |
Blood | -0.029 |
arrest | -0.028 |
robbers | -0.028 |
blown | -0.028 |
growth | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
Deal | +0.004 |
difference | +0.004 |
izes | +0.004 |
int | +0.004 |
decision | +0.004 |
Neg logit contributions | |
spilled | -0.004 |
Blood | -0.003 |
growth | -0.003 |
blown | -0.003 |
arrest | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
alks | +0.054 |
arrest | +0.054 |
robbers | +0.052 |
plets | +0.052 |
Blood | +0.051 |
Neg logit contributions | |
Deal | -0.044 |
exactly | -0.039 |
why | -0.037 |
equally | -0.037 |
� | -0.036 |
Token | pretty |
Token ID | 2495 |
Feature activation | +6.59e-03 |
Pos logit contributions | |
Deal | +0.008 |
difference | +0.007 |
ement | +0.007 |
isn | +0.007 |
exactly | +0.007 |
Neg logit contributions | |
Blood | -0.004 |
arrest | -0.004 |
robbers | -0.004 |
blown | -0.004 |
mice | -0.004 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
Deal | +0.000 |
difference | +0.000 |
ement | +0.000 |
isn | +0.000 |
why | +0.000 |
Neg logit contributions | |
Blood | -0.000 |
arrest | -0.000 |
robbers | -0.000 |
blown | -0.000 |
growth | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -8.94e-01 |
Pos logit contributions | |
Blood | +0.009 |
arrest | +0.009 |
robbers | +0.009 |
blown | +0.009 |
growth | +0.009 |
Neg logit contributions | |
Deal | -0.011 |
difference | -0.010 |
why | -0.009 |
ement | -0.009 |
isn | -0.009 |
Token | She |
Token ID | 1375 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
arrest | +0.021 |
blown | +0.020 |
robbers | +0.020 |
Blood | +0.019 |
alks | +0.019 |
Neg logit contributions | |
Deal | -0.025 |
difference | -0.023 |
isn | -0.022 |
why | -0.022 |
ement | -0.021 |
Token | had |
Token ID | 550 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
arrest | +0.033 |
robbers | +0.032 |
alks | +0.032 |
Blood | +0.032 |
growth | +0.032 |
Neg logit contributions | |
Deal | -0.023 |
isn | -0.018 |
difference | -0.018 |
why | -0.017 |
promise | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
alks | +0.086 |
robbers | +0.084 |
plets | +0.084 |
arrest | +0.082 |
Blood | +0.080 |
Neg logit contributions | |
Deal | -0.075 |
exactly | -0.067 |
isn | -0.065 |
why | -0.063 |
wonderful | -0.063 |
Token | massage |
Token ID | 26900 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Blood | +0.009 |
robbers | +0.009 |
arrest | +0.008 |
spilled | +0.008 |
blown | +0.008 |
Neg logit contributions | |
Deal | -0.013 |
difference | -0.012 |
ement | -0.011 |
why | -0.011 |
isn | -0.011 |
Token | every |
Token ID | 790 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
Deal | +0.033 |
difference | +0.029 |
ement | +0.028 |
isn | +0.027 |
why | +0.027 |
Neg logit contributions | |
Blood | -0.031 |
arrest | -0.031 |
robbers | -0.030 |
blown | -0.030 |
growth | -0.029 |
Token | much |
Token ID | 881 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
Deal | +0.011 |
difference | +0.010 |
ement | +0.010 |
decision | +0.009 |
isn | +0.009 |
Neg logit contributions | |
blown | -0.009 |
Blood | -0.009 |
arrest | -0.008 |
robbers | -0.008 |
mice | -0.008 |
Token | happier |
Token ID | 23030 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
blown | +0.030 |
mice | +0.030 |
robbers | +0.029 |
Blood | +0.028 |
spilled | +0.028 |
Neg logit contributions | |
Deal | -0.026 |
ement | -0.022 |
isn | -0.021 |
ieved | -0.020 |
uct | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
arrest | +0.016 |
robbers | +0.016 |
blown | +0.016 |
Blood | +0.015 |
mice | +0.015 |
Neg logit contributions | |
Deal | -0.019 |
ement | -0.016 |
ieved | -0.015 |
difference | -0.015 |
uct | -0.015 |
Token | all |
Token ID | 477 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
blown | +0.001 |
arrest | +0.001 |
Blood | +0.001 |
robbers | +0.001 |
mice | +0.001 |
Neg logit contributions | |
Deal | -0.001 |
ement | -0.001 |
difference | -0.001 |
decision | -0.001 |
ieved | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Blood | +0.008 |
arrest | +0.008 |
blown | +0.008 |
robbers | +0.008 |
mice | +0.008 |
Neg logit contributions | |
Deal | -0.012 |
difference | -0.011 |
ement | -0.010 |
decision | -0.010 |
exactly | -0.010 |
Token | little |
Token ID | 1310 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
Blood | +0.028 |
arrest | +0.027 |
robbers | +0.027 |
blown | +0.026 |
growth | +0.026 |
Neg logit contributions | |
Deal | -0.043 |
difference | -0.039 |
ement | -0.038 |
isn | -0.037 |
why | -0.037 |
Token | animals |
Token ID | 4695 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
Blood | +0.066 |
arrest | +0.063 |
growth | +0.063 |
alks | +0.061 |
blown | +0.060 |
Neg logit contributions | |
Deal | -0.092 |
difference | -0.088 |
isn | -0.083 |
decision | -0.081 |
ement | -0.081 |
Token | came |
Token ID | 1625 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
arrest | +0.009 |
Blood | +0.009 |
blown | +0.009 |
robbers | +0.009 |
spilled | +0.009 |
Neg logit contributions | |
Deal | -0.012 |
difference | -0.010 |
ement | -0.010 |
decision | -0.010 |
isn | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
Blood | +0.056 |
arrest | +0.056 |
blown | +0.055 |
robbers | +0.054 |
alks | +0.052 |
Neg logit contributions | |
Deal | -0.085 |
difference | -0.075 |
ement | -0.073 |
isn | -0.071 |
decision | -0.071 |
Token | play |
Token ID | 711 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
Deal | +0.055 |
ement | +0.048 |
difference | +0.047 |
isn | +0.046 |
decision | +0.045 |
Neg logit contributions | |
arrest | -0.043 |
Blood | -0.041 |
robbers | -0.041 |
growth | -0.040 |
blown | -0.040 |
Token | around |
Token ID | 1088 |
Feature activation | +8.17e-01 |
Pos logit contributions | |
Deal | +0.009 |
difference | +0.008 |
aman | +0.008 |
decision | +0.008 |
uct | +0.007 |
Neg logit contributions | |
mice | -0.008 |
arrest | -0.008 |
robbers | -0.008 |
Blood | -0.008 |
growth | -0.008 |
Token | trains |
Token ID | 13404 |
Feature activation | +7.55e-01 |
Pos logit contributions | |
alks | +0.063 |
Blood | +0.062 |
growth | +0.060 |
arrest | +0.060 |
plets | +0.059 |
Neg logit contributions | |
Deal | -0.047 |
why | -0.044 |
isn | -0.042 |
exactly | -0.039 |
difference | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
Deal | +0.014 |
difference | +0.012 |
ement | +0.011 |
isn | +0.011 |
why | +0.011 |
Neg logit contributions | |
Blood | -0.024 |
arrest | -0.024 |
robbers | -0.023 |
blown | -0.023 |
growth | -0.023 |
Token | planes |
Token ID | 13016 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
Blood | +0.073 |
alks | +0.070 |
growth | +0.066 |
blown | +0.066 |
arrest | +0.066 |
Neg logit contributions | |
Deal | -0.085 |
why | -0.079 |
difference | -0.077 |
isn | -0.077 |
deal | -0.073 |
Token | that |
Token ID | 326 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
Deal | +0.022 |
difference | +0.020 |
ement | +0.019 |
decision | +0.018 |
isn | +0.018 |
Neg logit contributions | |
Blood | -0.022 |
arrest | -0.022 |
robbers | -0.021 |
blown | -0.021 |
growth | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Blood | +0.015 |
arrest | +0.015 |
robbers | +0.015 |
blown | +0.014 |
growth | +0.014 |
Neg logit contributions | |
Deal | -0.018 |
difference | -0.016 |
isn | -0.015 |
why | -0.015 |
ement | -0.015 |
Token | could |
Token ID | 714 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
Blood | +0.030 |
arrest | +0.029 |
growth | +0.029 |
alks | +0.029 |
robbers | +0.028 |
Neg logit contributions | |
Deal | -0.027 |
isn | -0.023 |
why | -0.022 |
difference | -0.022 |
exactly | -0.021 |
Token | zoom |
Token ID | 19792 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Blood | +0.065 |
arrest | +0.063 |
robbers | +0.062 |
blown | +0.061 |
growth | +0.061 |
Neg logit contributions | |
Deal | -0.069 |
difference | -0.061 |
why | -0.058 |
isn | -0.057 |
promise | -0.057 |
Token | around |
Token ID | 1088 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Blood | +0.016 |
arrest | +0.015 |
robbers | +0.015 |
blown | +0.015 |
growth | +0.015 |
Neg logit contributions | |
Deal | -0.015 |
difference | -0.013 |
ement | -0.012 |
decision | -0.012 |
promise | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
arrest | +0.003 |
Blood | +0.003 |
robbers | +0.002 |
growth | +0.002 |
mice | +0.002 |
Neg logit contributions | |
Deal | -0.003 |
difference | -0.003 |
ement | -0.002 |
promise | -0.002 |
decision | -0.002 |
Token | floor |
Token ID | 4314 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
arrest | +0.017 |
robbers | +0.017 |
mice | +0.016 |
Blood | +0.016 |
spilled | +0.016 |
Neg logit contributions | |
Deal | -0.027 |
ement | -0.024 |
exactly | -0.024 |
ieved | -0.023 |
why | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
Deal | +0.002 |
difference | +0.002 |
ement | +0.001 |
isn | +0.001 |
decision | +0.001 |
Neg logit contributions | |
Blood | -0.001 |
arrest | -0.001 |
blown | -0.001 |
robbers | -0.001 |
spilled | -0.001 |
Token | as |
Token ID | 355 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
robbers | +0.013 |
arrest | +0.013 |
blown | +0.012 |
mice | +0.012 |
Blood | +0.012 |
Neg logit contributions | |
Deal | -0.014 |
difference | -0.012 |
ement | -0.012 |
isn | -0.011 |
decision | -0.011 |
Token | fast |
Token ID | 3049 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
growth | +0.035 |
arrest | +0.035 |
Blood | +0.035 |
alks | +0.033 |
robbers | +0.033 |
Neg logit contributions | |
Deal | -0.033 |
exactly | -0.028 |
equally | -0.026 |
why | -0.026 |
difference | -0.026 |
Token | as |
Token ID | 355 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
Blood | +0.031 |
robbers | +0.028 |
alks | +0.028 |
umbs | +0.028 |
growth | +0.028 |
Neg logit contributions | |
Deal | -0.020 |
why | -0.019 |
isn | -0.018 |
ement | -0.017 |
exactly | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
Blood | +0.042 |
arrest | +0.042 |
robbers | +0.040 |
growth | +0.040 |
blown | +0.040 |
Neg logit contributions | |
Deal | -0.048 |
difference | -0.043 |
exactly | -0.042 |
why | -0.042 |
isn | -0.042 |
Token | could |
Token ID | 714 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
Blood | +0.067 |
arrest | +0.067 |
robbers | +0.065 |
alks | +0.063 |
growth | +0.063 |
Neg logit contributions | |
Deal | -0.058 |
isn | -0.049 |
difference | -0.049 |
why | -0.049 |
exactly | -0.048 |
Token | away |
Token ID | 1497 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
Blood | +0.082 |
arrest | +0.081 |
robbers | +0.080 |
blown | +0.078 |
growth | +0.078 |
Neg logit contributions | |
Deal | -0.071 |
difference | -0.063 |
ement | -0.059 |
why | -0.059 |
isn | -0.059 |
Token | from |
Token ID | 422 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
Blood | +0.076 |
arrest | +0.073 |
robbers | +0.072 |
blown | +0.071 |
growth | +0.070 |
Neg logit contributions | |
Deal | -0.067 |
difference | -0.060 |
ement | -0.056 |
why | -0.056 |
isn | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
Deal | +0.051 |
difference | +0.047 |
why | +0.046 |
exactly | +0.045 |
ement | +0.044 |
Neg logit contributions | |
Blood | -0.054 |
arrest | -0.051 |
robbers | -0.050 |
blown | -0.049 |
alks | -0.049 |
Token | weird |
Token ID | 7650 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
Blood | +0.006 |
arrest | +0.006 |
robbers | +0.006 |
blown | +0.006 |
growth | +0.006 |
Neg logit contributions | |
Deal | -0.008 |
difference | -0.007 |
ement | -0.007 |
isn | -0.007 |
decision | -0.007 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Blood | +0.020 |
arrest | +0.020 |
robbers | +0.019 |
blown | +0.019 |
growth | +0.019 |
Neg logit contributions | |
Deal | -0.025 |
difference | -0.022 |
ement | -0.021 |
isn | -0.021 |
why | -0.021 |
Token | drinking |
Token ID | 7722 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Deal | +0.032 |
difference | +0.028 |
decision | +0.027 |
ement | +0.026 |
promise | +0.026 |
Neg logit contributions | |
arrest | -0.025 |
blown | -0.025 |
Blood | -0.025 |
spilled | -0.024 |
robbers | -0.024 |
Token | by |
Token ID | 416 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
arrest | +0.013 |
Blood | +0.013 |
blown | +0.013 |
robbers | +0.013 |
spilled | +0.013 |
Neg logit contributions | |
Deal | -0.019 |
difference | -0.017 |
isn | -0.017 |
ement | -0.016 |
ieved | -0.016 |
Token | giving |
Token ID | 3501 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
Deal | +0.025 |
why | +0.024 |
difference | +0.024 |
exactly | +0.023 |
decision | +0.022 |
Neg logit contributions | |
Blood | -0.024 |
arrest | -0.023 |
growth | -0.023 |
robbers | -0.023 |
plets | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
mice | +0.002 |
Blood | +0.002 |
arrest | +0.002 |
blown | +0.002 |
spilled | +0.002 |
Neg logit contributions | |
Deal | -0.004 |
difference | -0.003 |
ement | -0.003 |
decision | -0.003 |
isn | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
Deal | +0.014 |
ement | +0.012 |
difference | +0.011 |
ieved | +0.011 |
isn | +0.011 |
Neg logit contributions | |
mice | -0.007 |
arrest | -0.007 |
blown | -0.007 |
spilled | -0.007 |
growth | -0.007 |
Token | big |
Token ID | 1263 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Deal | +0.023 |
ement | +0.019 |
difference | +0.018 |
uct | +0.018 |
isn | +0.018 |
Neg logit contributions | |
Blood | -0.027 |
arrest | -0.026 |
blown | -0.026 |
spilled | -0.026 |
robbers | -0.026 |
Token | hug |
Token ID | 16225 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
Deal | +0.060 |
uct | +0.050 |
exactly | +0.050 |
ieved | +0.049 |
ement | +0.049 |
Neg logit contributions | |
arrest | -0.045 |
robbers | -0.044 |
Blood | -0.044 |
spilled | -0.044 |
mice | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | +9.09e-02 |
Pos logit contributions | |
spilled | +0.058 |
blown | +0.058 |
mice | +0.058 |
Blood | +0.058 |
robbers | +0.057 |
Neg logit contributions | |
Deal | -0.087 |
difference | -0.078 |
ement | -0.074 |
decision | -0.073 |
ieved | -0.072 |
Token | saying |
Token ID | 2282 |
Feature activation | +9.39e-02 |
Pos logit contributions | |
Deal | +0.003 |
difference | +0.002 |
ement | +0.002 |
isn | +0.002 |
decision | +0.002 |
Neg logit contributions | |
Blood | -0.002 |
arrest | -0.002 |
robbers | -0.002 |
blown | -0.002 |
growth | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Deal | +0.002 |
difference | +0.002 |
ieved | +0.002 |
decision | +0.002 |
promise | +0.002 |
Neg logit contributions | |
Blood | -0.002 |
mice | -0.002 |
spilled | -0.002 |
alks | -0.002 |
arrest | -0.002 |
Token | w |
Token ID | 86 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
Deal | +0.035 |
difference | +0.030 |
ement | +0.029 |
isn | +0.029 |
why | +0.029 |
Neg logit contributions | |
arrest | -0.033 |
Blood | -0.032 |
robbers | -0.032 |
growth | -0.031 |
blown | -0.031 |
Token | elcome |
Token ID | 9571 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
Blood | +0.004 |
arrest | +0.004 |
spilled | +0.004 |
alks | +0.004 |
blown | +0.004 |
Neg logit contributions | |
Deal | -0.003 |
difference | -0.003 |
ement | -0.003 |
decision | -0.002 |
promise | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
Blood | +0.034 |
robbers | +0.032 |
blown | +0.032 |
plets | +0.032 |
arrest | +0.032 |
Neg logit contributions | |
Deal | -0.027 |
difference | -0.026 |
why | -0.026 |
exactly | -0.024 |
ement | -0.024 |
Token | street |
Token ID | 4675 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
Blood | +0.008 |
blown | +0.007 |
arrest | +0.007 |
alks | +0.007 |
plets | +0.007 |
Neg logit contributions | |
Deal | -0.009 |
difference | -0.008 |
decision | -0.008 |
promise | -0.008 |
isn | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
arrest | +0.012 |
Blood | +0.012 |
blown | +0.011 |
spilled | +0.011 |
robbers | +0.011 |
Neg logit contributions | |
Deal | -0.018 |
difference | -0.016 |
ement | -0.015 |
ieved | -0.015 |
isn | -0.015 |
Token | The |
Token ID | 383 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
arrest | +0.033 |
blown | +0.031 |
robbers | +0.030 |
growth | +0.029 |
Blood | +0.029 |
Neg logit contributions | |
Deal | -0.027 |
isn | -0.026 |
difference | -0.025 |
� | -0.025 |
why | -0.025 |
Token | van |
Token ID | 5719 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
Blood | +0.020 |
arrest | +0.020 |
robbers | +0.020 |
blown | +0.020 |
growth | +0.019 |
Neg logit contributions | |
Deal | -0.016 |
difference | -0.014 |
ement | -0.013 |
isn | -0.013 |
why | -0.013 |
Token | has |
Token ID | 468 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
Blood | +0.022 |
arrest | +0.022 |
spilled | +0.021 |
blown | +0.021 |
robbers | +0.020 |
Neg logit contributions | |
Deal | -0.026 |
difference | -0.023 |
why | -0.021 |
ieved | -0.021 |
decision | -0.021 |
Token | many |
Token ID | 867 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
Blood | +0.091 |
plets | +0.089 |
alks | +0.088 |
ges | +0.085 |
growth | +0.085 |
Neg logit contributions | |
exactly | -0.080 |
Deal | -0.079 |
isn | -0.078 |
why | -0.075 |
difference | -0.073 |
Token | colors |
Token ID | 7577 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
Blood | +0.045 |
arrest | +0.044 |
robbers | +0.043 |
blown | +0.042 |
growth | +0.041 |
Neg logit contributions | |
Deal | -0.071 |
difference | -0.064 |
ement | -0.061 |
why | -0.061 |
isn | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
arrest | +0.021 |
Blood | +0.021 |
robbers | +0.021 |
growth | +0.020 |
mice | +0.020 |
Neg logit contributions | |
Deal | -0.017 |
ement | -0.015 |
difference | -0.015 |
isn | -0.015 |
exactly | -0.015 |
Token | pictures |
Token ID | 5986 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
Blood | +0.047 |
arrest | +0.047 |
robbers | +0.046 |
blown | +0.045 |
growth | +0.045 |
Neg logit contributions | |
Deal | -0.054 |
difference | -0.047 |
ement | -0.045 |
isn | -0.044 |
why | -0.044 |
Token | on |
Token ID | 319 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
Blood | +0.015 |
blown | +0.015 |
arrest | +0.015 |
robbers | +0.015 |
alks | +0.014 |
Neg logit contributions | |
Deal | -0.019 |
difference | -0.016 |
decision | -0.015 |
ement | -0.015 |
promise | -0.015 |