Token | thought |
Token ID | 1807 |
Feature activation | -6.99e-02 |
Pos logit contributions | |
hon | +0.021 |
notice | +0.018 |
boom | +0.017 |
explode | +0.017 |
announcements | +0.017 |
Neg logit contributions | |
favourite | -0.027 |
possession | -0.026 |
favorite | -0.025 |
wrist | -0.024 |
purse | -0.023 |
Token | for |
Token ID | 329 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
possession | +0.111 |
favourite | +0.108 |
favorite | +0.097 |
purse | +0.090 |
wrist | +0.085 |
Neg logit contributions | |
hon | -0.138 |
notice | -0.118 |
hon | -0.117 |
rep | -0.116 |
cal | -0.115 |
Token | a |
Token ID | 257 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
possession | +0.074 |
favourite | +0.071 |
favorite | +0.066 |
purse | +0.064 |
wrist | +0.062 |
Neg logit contributions | |
hon | -0.049 |
notice | -0.042 |
rep | -0.042 |
hon | -0.041 |
cal | -0.041 |
Token | moment |
Token ID | 2589 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
hon | +0.028 |
overcome | +0.023 |
cal | +0.022 |
announcements | +0.022 |
hon | +0.021 |
Neg logit contributions | |
favourite | -0.034 |
possession | -0.034 |
favorite | -0.032 |
purse | -0.030 |
brand | -0.029 |
Token | before |
Token ID | 878 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
possession | +0.048 |
favourite | +0.047 |
favorite | +0.043 |
purse | +0.042 |
wrist | +0.040 |
Neg logit contributions | |
hon | -0.038 |
notice | -0.032 |
hon | -0.031 |
cal | -0.031 |
rep | -0.031 |
Token | rep |
Token ID | 1128 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
hon | +0.027 |
RO | +0.023 |
cal | +0.022 |
rr | +0.022 |
Hon | +0.022 |
Neg logit contributions | |
favourite | -0.022 |
possession | -0.021 |
purse | -0.020 |
brand | -0.020 |
favorite | -0.019 |
Token | lying |
Token ID | 3157 |
Feature activation | -5.73e-03 |
Pos logit contributions | |
hon | +0.334 |
notice | +0.276 |
hon | +0.272 |
rep | +0.271 |
cal | +0.270 |
Neg logit contributions | |
possession | -0.532 |
favourite | -0.515 |
favorite | -0.479 |
purse | -0.468 |
wrist | -0.453 |
Token | , |
Token ID | 11 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
possession | +0.012 |
favourite | +0.012 |
favorite | +0.011 |
purse | +0.011 |
wrist | +0.010 |
Neg logit contributions | |
hon | -0.008 |
notice | -0.006 |
hon | -0.006 |
cal | -0.006 |
rep | -0.006 |
Token | " |
Token ID | 366 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
possession | +0.031 |
favourite | +0.030 |
favorite | +0.026 |
purse | +0.023 |
wrist | +0.022 |
Neg logit contributions | |
hon | -0.082 |
notice | -0.074 |
cal | -0.072 |
hon | -0.072 |
rep | -0.072 |
Token | I |
Token ID | 40 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
possession | +0.172 |
favourite | +0.164 |
favorite | +0.149 |
purse | +0.144 |
wrist | +0.136 |
Neg logit contributions | |
hon | -0.192 |
notice | -0.168 |
rep | -0.166 |
hon | -0.164 |
cal | -0.164 |
Token | think |
Token ID | 892 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
hon | +0.073 |
announcements | +0.063 |
louder | +0.063 |
boom | +0.061 |
thunder | +0.061 |
Neg logit contributions | |
favourite | -0.085 |
possession | -0.084 |
favorite | -0.079 |
favourites | -0.071 |
purse | -0.071 |
Token | show |
Token ID | 905 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
hon | +0.195 |
Jim | +0.160 |
announcements | +0.157 |
louder | +0.155 |
rep | +0.154 |
Neg logit contributions | |
favourite | -0.226 |
possession | -0.220 |
favorite | -0.212 |
purse | -0.192 |
favorites | -0.190 |
Token | understanding |
Token ID | 4547 |
Feature activation | -3.57e-03 |
Pos logit contributions | |
possession | +0.045 |
favourite | +0.044 |
favorite | +0.041 |
purse | +0.039 |
wrist | +0.038 |
Neg logit contributions | |
hon | -0.034 |
notice | -0.029 |
rep | -0.028 |
hon | -0.028 |
cal | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +8.05e-03 |
Pos logit contributions | |
possession | +0.007 |
favourite | +0.007 |
favorite | +0.006 |
purse | +0.006 |
wrist | +0.006 |
Neg logit contributions | |
hon | -0.006 |
notice | -0.005 |
rep | -0.005 |
hon | -0.005 |
cal | -0.005 |
Token | care |
Token ID | 1337 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
hon | +0.014 |
notice | +0.012 |
rep | +0.011 |
hon | +0.011 |
oblig | +0.011 |
Neg logit contributions | |
possession | -0.015 |
favourite | -0.014 |
favorite | -0.013 |
purse | -0.012 |
wrist | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
hon | +0.065 |
notice | +0.057 |
rep | +0.057 |
hon | +0.056 |
cal | +0.056 |
Neg logit contributions | |
possession | -0.060 |
favourite | -0.057 |
favorite | -0.052 |
purse | -0.051 |
wrist | -0.048 |
Token | gent |
Token ID | 25049 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
hon | +0.032 |
notice | +0.026 |
rep | +0.026 |
hon | +0.026 |
cal | +0.026 |
Neg logit contributions | |
possession | -0.041 |
favourite | -0.039 |
favorite | -0.037 |
purse | -0.035 |
wrist | -0.034 |
Token | leness |
Token ID | 48795 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
hon | +0.382 |
slit | +0.351 |
thunder | +0.342 |
rep | +0.341 |
cal | +0.338 |
Neg logit contributions | |
possession | -0.467 |
favourite | -0.399 |
possessions | -0.387 |
favorite | -0.365 |
attered | -0.354 |
Token | . |
Token ID | 13 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
possession | +0.044 |
favourite | +0.042 |
favorite | +0.039 |
purse | +0.037 |
wrist | +0.036 |
Neg logit contributions | |
hon | -0.035 |
notice | -0.029 |
rep | -0.029 |
hon | -0.028 |
oblig | -0.028 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.04e-03 |
Pos logit contributions | |
hon | +0.029 |
notice | +0.026 |
hon | +0.025 |
cal | +0.025 |
rep | +0.025 |
Neg logit contributions | |
possession | -0.025 |
favourite | -0.024 |
favorite | -0.022 |
purse | -0.021 |
wrist | -0.020 |
Token | Johnny |
Token ID | 44960 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
hon | +0.002 |
announcements | +0.002 |
louder | +0.002 |
notice | +0.002 |
hon | +0.002 |
Neg logit contributions | |
possession | -0.001 |
favourite | -0.001 |
favorite | -0.001 |
robe | -0.001 |
purse | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -5.15e-02 |
Pos logit contributions | |
hon | +0.056 |
louder | +0.050 |
announcements | +0.050 |
notice | +0.048 |
oblig | +0.048 |
Neg logit contributions | |
possession | -0.045 |
favourite | -0.043 |
favorite | -0.040 |
purse | -0.039 |
wrist | -0.037 |
Token | thought |
Token ID | 1807 |
Feature activation | -7.55e-02 |
Pos logit contributions | |
hon | +0.066 |
notice | +0.058 |
announcements | +0.057 |
boom | +0.056 |
explode | +0.055 |
Neg logit contributions | |
favourite | -0.071 |
possession | -0.071 |
favorite | -0.065 |
wrist | -0.061 |
purse | -0.060 |
Token | for |
Token ID | 329 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
possession | +0.099 |
favourite | +0.098 |
favorite | +0.086 |
purse | +0.075 |
brand | +0.072 |
Neg logit contributions | |
hon | -0.169 |
notice | -0.151 |
announcements | -0.147 |
hon | -0.147 |
rep | -0.146 |
Token | a |
Token ID | 257 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
possession | +0.086 |
favourite | +0.083 |
favorite | +0.077 |
purse | +0.074 |
wrist | +0.072 |
Neg logit contributions | |
hon | -0.064 |
notice | -0.053 |
rep | -0.052 |
hon | -0.052 |
cal | -0.052 |
Token | moment |
Token ID | 2589 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
hon | +0.022 |
announcements | +0.018 |
overcome | +0.018 |
cal | +0.018 |
hon | +0.017 |
Neg logit contributions | |
favourite | -0.025 |
possession | -0.025 |
favorite | -0.023 |
purse | -0.021 |
brand | -0.021 |
Token | before |
Token ID | 878 |
Feature activation | +6.33e-03 |
Pos logit contributions | |
possession | +0.059 |
favourite | +0.057 |
favorite | +0.053 |
purse | +0.050 |
wrist | +0.048 |
Neg logit contributions | |
hon | -0.058 |
notice | -0.049 |
cal | -0.047 |
hon | -0.047 |
announcements | -0.047 |
Token | rep |
Token ID | 1128 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
hon | +0.012 |
RO | +0.010 |
rr | +0.010 |
Hon | +0.010 |
cal | +0.010 |
Neg logit contributions | |
favourite | -0.009 |
possession | -0.009 |
brand | -0.008 |
favorite | -0.008 |
wallet | -0.008 |
Token | lying |
Token ID | 3157 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
hon | +0.331 |
notice | +0.274 |
rep | +0.270 |
hon | +0.269 |
cal | +0.267 |
Neg logit contributions | |
possession | -0.532 |
favourite | -0.514 |
favorite | -0.480 |
purse | -0.467 |
wrist | -0.452 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
possession | +0.022 |
favourite | +0.021 |
favorite | +0.020 |
purse | +0.019 |
wrist | +0.018 |
Neg logit contributions | |
hon | -0.017 |
notice | -0.014 |
RO | -0.014 |
hon | -0.014 |
announcements | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
possession | +0.037 |
favourite | +0.036 |
favorite | +0.032 |
purse | +0.029 |
wrist | +0.028 |
Neg logit contributions | |
hon | -0.061 |
notice | -0.055 |
hon | -0.054 |
rep | -0.053 |
cal | -0.053 |
Token | Yes |
Token ID | 5297 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
possession | +0.190 |
favourite | +0.181 |
favorite | +0.167 |
purse | +0.163 |
wrist | +0.154 |
Neg logit contributions | |
hon | -0.173 |
notice | -0.149 |
rep | -0.147 |
cal | -0.145 |
hon | -0.145 |
Token | I |
Token ID | 314 |
Feature activation | +1.97e-03 |
Pos logit contributions | |
hon | +0.170 |
announcements | +0.147 |
notice | +0.143 |
overcome | +0.142 |
RO | +0.142 |
Neg logit contributions | |
favourite | -0.170 |
possession | -0.165 |
favorite | -0.159 |
purse | -0.146 |
brand | -0.138 |
Token | cle |
Token ID | 1190 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
possession | +0.060 |
favourite | +0.059 |
favorite | +0.054 |
purse | +0.051 |
brand | +0.047 |
Neg logit contributions | |
hon | -0.064 |
rep | -0.054 |
notice | -0.054 |
oblig | -0.053 |
announcements | -0.053 |
Token | vere |
Token ID | 4119 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
hon | +0.161 |
notice | +0.145 |
thunder | +0.145 |
hon | +0.143 |
cal | +0.142 |
Neg logit contributions | |
possession | -0.166 |
favourite | -0.158 |
favorite | -0.139 |
purse | -0.135 |
wrist | -0.134 |
Token | r |
Token ID | 81 |
Feature activation | +9.06e-03 |
Pos logit contributions | |
hon | +0.282 |
cal | +0.246 |
rep | +0.240 |
announcements | +0.237 |
hon | +0.236 |
Neg logit contributions | |
possession | -0.502 |
favourite | -0.450 |
purse | -0.420 |
favorite | -0.419 |
wrist | -0.407 |
Token | and |
Token ID | 290 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
hon | +0.016 |
cal | +0.014 |
RO | +0.014 |
Run | +0.014 |
notice | +0.014 |
Neg logit contributions | |
possession | -0.016 |
favourite | -0.015 |
favorite | -0.014 |
purse | -0.013 |
wrist | -0.012 |
Token | cle |
Token ID | 1190 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
possession | +0.050 |
favourite | +0.049 |
favorite | +0.045 |
brand | +0.043 |
purse | +0.042 |
Neg logit contributions | |
hon | -0.042 |
rep | -0.031 |
rr | -0.031 |
Jim | -0.031 |
notice | -0.031 |
Token | vere |
Token ID | 4119 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
hon | +0.198 |
thunder | +0.182 |
notice | +0.180 |
rep | +0.175 |
cal | +0.174 |
Neg logit contributions | |
possession | -0.235 |
favourite | -0.223 |
favorite | -0.198 |
wrist | -0.194 |
purse | -0.190 |
Token | r |
Token ID | 81 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
hon | +0.318 |
cal | +0.285 |
slit | +0.276 |
rep | +0.274 |
announcements | +0.273 |
Neg logit contributions | |
possession | -0.522 |
favourite | -0.452 |
purse | -0.423 |
favorite | -0.419 |
wrist | -0.413 |
Token | . |
Token ID | 13 |
Feature activation | -7.72e-03 |
Pos logit contributions | |
hon | +0.048 |
Run | +0.043 |
cal | +0.042 |
RO | +0.042 |
rep | +0.041 |
Neg logit contributions | |
possession | -0.048 |
favourite | -0.046 |
favorite | -0.042 |
possessions | -0.037 |
purse | -0.037 |
Token | Eventually |
Token ID | 16178 |
Feature activation | +8.44e-02 |
Pos logit contributions | |
possession | +0.011 |
favourite | +0.011 |
favorite | +0.009 |
purse | +0.009 |
wrist | +0.008 |
Neg logit contributions | |
hon | -0.016 |
notice | -0.014 |
hon | -0.014 |
rep | -0.014 |
cal | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
hon | +0.181 |
rr | +0.157 |
announcements | +0.147 |
RO | +0.142 |
Hon | +0.142 |
Neg logit contributions | |
possession | -0.111 |
favourite | -0.108 |
purse | -0.097 |
microscope | -0.089 |
favorite | -0.087 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -5.50e-03 |
Pos logit contributions | |
hon | +0.121 |
notice | +0.106 |
hon | +0.104 |
cal | +0.103 |
rep | +0.103 |
Neg logit contributions | |
possession | -0.089 |
favourite | -0.087 |
favorite | -0.079 |
purse | -0.074 |
wrist | -0.069 |
Token | You |
Token ID | 921 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.023 |
favorite | +0.021 |
purse | +0.020 |
wrist | +0.019 |
Neg logit contributions | |
hon | -0.030 |
notice | -0.026 |
rep | -0.026 |
hon | -0.026 |
cal | -0.026 |
Token | never |
Token ID | 1239 |
Feature activation | +6.60e-02 |
Pos logit contributions | |
hon | +0.087 |
louder | +0.077 |
rep | +0.076 |
announcements | +0.076 |
oblig | +0.074 |
Neg logit contributions | |
possession | -0.073 |
favourite | -0.073 |
favorite | -0.068 |
purse | -0.064 |
wallet | -0.058 |
Token | know |
Token ID | 760 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
hon | +0.103 |
announcements | +0.087 |
Jim | +0.085 |
oblig | +0.083 |
louder | +0.082 |
Neg logit contributions | |
possession | -0.119 |
favourite | -0.113 |
favorite | -0.110 |
purse | -0.102 |
wallet | -0.100 |
Token | what |
Token ID | 644 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
possession | +0.063 |
favourite | +0.060 |
favorite | +0.054 |
purse | +0.051 |
wrist | +0.047 |
Neg logit contributions | |
hon | -0.087 |
notice | -0.078 |
oblig | -0.076 |
rep | -0.076 |
hon | -0.075 |
Token | dangers |
Token ID | 16169 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
possession | +0.179 |
favourite | +0.172 |
favorite | +0.160 |
purse | +0.151 |
wrist | +0.143 |
Neg logit contributions | |
hon | -0.164 |
notice | -0.141 |
rep | -0.139 |
hon | -0.138 |
cal | -0.137 |
Token | lur |
Token ID | 20605 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
hon | +0.169 |
rep | +0.146 |
oblig | +0.145 |
announcements | +0.144 |
cal | +0.144 |
Neg logit contributions | |
possession | -0.160 |
favourite | -0.150 |
favorite | -0.139 |
purse | -0.133 |
robe | -0.127 |
Token | k |
Token ID | 74 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
hon | +0.403 |
rep | +0.368 |
cal | +0.362 |
notice | +0.352 |
overcome | +0.352 |
Neg logit contributions | |
possession | -0.410 |
favourite | -0.387 |
favorite | -0.362 |
purse | -0.339 |
attered | -0.329 |
Token | around |
Token ID | 1088 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
hon | +0.038 |
rep | +0.034 |
slit | +0.034 |
cal | +0.033 |
oblig | +0.033 |
Neg logit contributions | |
possession | -0.046 |
favourite | -0.044 |
favorite | -0.041 |
purse | -0.039 |
wrist | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
possession | +0.079 |
favourite | +0.075 |
favorite | +0.069 |
purse | +0.067 |
wrist | +0.064 |
Neg logit contributions | |
hon | -0.064 |
rep | -0.053 |
cal | -0.053 |
hon | -0.053 |
oblig | -0.053 |
Token | corner |
Token ID | 5228 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
hon | +0.074 |
notice | +0.068 |
cal | +0.067 |
hon | +0.066 |
rep | +0.066 |
Neg logit contributions | |
possession | -0.062 |
favourite | -0.058 |
favorite | -0.053 |
purse | -0.051 |
wrist | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
possession | +0.232 |
favourite | +0.222 |
favorite | +0.208 |
purse | +0.198 |
wrist | +0.193 |
Neg logit contributions | |
hon | -0.184 |
rep | -0.153 |
notice | -0.153 |
hon | -0.152 |
oblig | -0.151 |
Token | and |
Token ID | 290 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
possession | +0.126 |
favourite | +0.120 |
favorite | +0.110 |
purse | +0.106 |
wrist | +0.100 |
Neg logit contributions | |
hon | -0.127 |
notice | -0.110 |
rep | -0.109 |
cal | -0.108 |
oblig | -0.107 |
Token | watched |
Token ID | 7342 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
possession | +0.081 |
favourite | +0.078 |
favorite | +0.074 |
purse | +0.072 |
wrist | +0.069 |
Neg logit contributions | |
hon | -0.063 |
rr | -0.054 |
oblig | -0.052 |
rep | -0.050 |
notice | -0.049 |
Token | as |
Token ID | 355 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
hon | +0.142 |
announcements | +0.121 |
rr | +0.121 |
notice | +0.119 |
oblig | +0.117 |
Neg logit contributions | |
favourite | -0.116 |
possession | -0.111 |
favorite | -0.106 |
wrist | -0.093 |
purse | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
hon | +0.063 |
rr | +0.058 |
oblig | +0.058 |
announcements | +0.056 |
notice | +0.054 |
Neg logit contributions | |
possession | -0.058 |
favourite | -0.054 |
favorite | -0.052 |
attic | -0.045 |
purse | -0.044 |
Token | snake |
Token ID | 17522 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
hon | +0.135 |
notice | +0.118 |
rep | +0.116 |
cal | +0.115 |
hon | +0.114 |
Neg logit contributions | |
possession | -0.118 |
favourite | -0.114 |
favorite | -0.104 |
purse | -0.099 |
wrist | -0.093 |
Token | slit |
Token ID | 40724 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
hon | +0.022 |
oblig | +0.019 |
rr | +0.019 |
rep | +0.019 |
Jim | +0.018 |
Neg logit contributions | |
favourite | -0.027 |
favorite | -0.026 |
possession | -0.025 |
purse | -0.023 |
wrist | -0.022 |
Token | hered |
Token ID | 6083 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
hon | +0.342 |
announcements | +0.305 |
cal | +0.300 |
rep | +0.299 |
thunder | +0.297 |
Neg logit contributions | |
favourite | -0.436 |
possession | -0.436 |
purse | -0.404 |
favorite | -0.395 |
attered | -0.383 |
Token | away |
Token ID | 1497 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
hon | +0.077 |
rep | +0.073 |
oblig | +0.069 |
cal | +0.066 |
announcements | +0.066 |
Neg logit contributions | |
possession | -0.062 |
favourite | -0.058 |
favorite | -0.053 |
purse | -0.048 |
item | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -9.75e-03 |
Pos logit contributions | |
possession | +0.030 |
favourite | +0.029 |
favorite | +0.027 |
attered | +0.023 |
purse | +0.023 |
Neg logit contributions | |
hon | -0.039 |
oblig | -0.034 |
Jim | -0.034 |
rr | -0.032 |
announcements | -0.032 |
Token | But |
Token ID | 887 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
possession | +0.015 |
favourite | +0.014 |
favorite | +0.013 |
purse | +0.012 |
wrist | +0.012 |
Neg logit contributions | |
hon | -0.019 |
notice | -0.017 |
rep | -0.017 |
hon | -0.017 |
cal | -0.017 |
Token | he |
Token ID | 339 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
hon | +0.021 |
notice | +0.018 |
rep | +0.018 |
cal | +0.017 |
oblig | +0.017 |
Neg logit contributions | |
possession | -0.026 |
favourite | -0.026 |
favorite | -0.024 |
purse | -0.023 |
wrist | -0.022 |
Token | running |
Token ID | 2491 |
Feature activation | +6.41e-02 |
Pos logit contributions | |
possession | +0.090 |
favourite | +0.089 |
favorite | +0.083 |
purse | +0.077 |
brand | +0.074 |
Neg logit contributions | |
hon | -0.075 |
notice | -0.062 |
oblig | -0.061 |
cal | -0.060 |
announcements | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | -5.35e-03 |
Pos logit contributions | |
hon | +0.116 |
rep | +0.104 |
oblig | +0.104 |
notice | +0.100 |
cal | +0.099 |
Neg logit contributions | |
possession | -0.109 |
favourite | -0.106 |
favorite | -0.098 |
purse | -0.090 |
attered | -0.085 |
Token | but |
Token ID | 475 |
Feature activation | -3.68e-03 |
Pos logit contributions | |
favourite | +0.009 |
possession | +0.009 |
favorite | +0.009 |
purse | +0.008 |
wrist | +0.008 |
Neg logit contributions | |
hon | -0.009 |
oblig | -0.008 |
notice | -0.008 |
rep | -0.008 |
cal | -0.008 |
Token | slower |
Token ID | 13611 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
possession | +0.007 |
favourite | +0.007 |
favorite | +0.006 |
purse | +0.006 |
attered | +0.006 |
Neg logit contributions | |
hon | -0.005 |
notice | -0.005 |
rep | -0.005 |
hon | -0.005 |
cal | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +5.92e-03 |
Pos logit contributions | |
hon | +0.209 |
oblig | +0.171 |
RO | +0.170 |
cal | +0.165 |
notice | +0.165 |
Neg logit contributions | |
possession | -0.254 |
favourite | -0.236 |
favorite | -0.221 |
wrist | -0.211 |
purse | -0.207 |
Token | gent |
Token ID | 25049 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
hon | +0.009 |
oblig | +0.008 |
rep | +0.007 |
notice | +0.007 |
announcements | +0.007 |
Neg logit contributions | |
possession | -0.012 |
favourite | -0.011 |
favorite | -0.011 |
purse | -0.010 |
wrist | -0.010 |
Token | ler |
Token ID | 1754 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
hon | +0.340 |
thunder | +0.320 |
cal | +0.319 |
slit | +0.313 |
overcome | +0.310 |
Neg logit contributions | |
possession | -0.452 |
favourite | -0.389 |
possessions | -0.372 |
favorite | -0.357 |
attered | -0.357 |
Token | ." |
Token ID | 526 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
hon | +0.031 |
rep | +0.025 |
Jim | +0.025 |
announcements | +0.025 |
notice | +0.025 |
Neg logit contributions | |
possession | -0.038 |
favourite | -0.036 |
favorite | -0.034 |
robe | -0.032 |
purse | -0.031 |
Token | The |
Token ID | 383 |
Feature activation | +8.89e-03 |
Pos logit contributions | |
hon | +0.065 |
announcements | +0.058 |
notice | +0.057 |
boom | +0.057 |
hon | +0.057 |
Neg logit contributions | |
possession | -0.047 |
favourite | -0.046 |
favorite | -0.041 |
purse | -0.040 |
wrist | -0.039 |
Token | car |
Token ID | 1097 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
hon | +0.016 |
notice | +0.014 |
rep | +0.013 |
hon | +0.013 |
cal | +0.013 |
Neg logit contributions | |
possession | -0.015 |
favourite | -0.015 |
favorite | -0.014 |
purse | -0.013 |
wrist | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
favourite | +0.037 |
possession | +0.035 |
favorite | +0.035 |
purse | +0.031 |
wrist | +0.030 |
Neg logit contributions | |
hon | -0.030 |
notice | -0.027 |
boom | -0.025 |
oblig | -0.025 |
rep | -0.025 |
Token | one |
Token ID | 530 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
possession | +0.010 |
favourite | +0.009 |
favorite | +0.009 |
purse | +0.008 |
wrist | +0.008 |
Neg logit contributions | |
hon | -0.008 |
notice | -0.007 |
rep | -0.007 |
cal | -0.007 |
hon | -0.007 |
Token | of |
Token ID | 286 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
possession | +0.052 |
favourite | +0.050 |
favorite | +0.046 |
purse | +0.044 |
wrist | +0.043 |
Neg logit contributions | |
hon | -0.045 |
notice | -0.038 |
rep | -0.038 |
cal | -0.037 |
hon | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
hon | +0.082 |
notice | +0.069 |
rep | +0.067 |
cal | +0.067 |
hon | +0.066 |
Neg logit contributions | |
possession | -0.118 |
favourite | -0.113 |
favorite | -0.105 |
purse | -0.101 |
wrist | -0.098 |
Token | biggest |
Token ID | 4094 |
Feature activation | -8.91e-03 |
Pos logit contributions | |
hon | +0.090 |
notice | +0.076 |
rep | +0.074 |
cal | +0.074 |
oblig | +0.074 |
Neg logit contributions | |
possession | -0.080 |
favourite | -0.080 |
favorite | -0.074 |
purse | -0.069 |
wrist | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
possession | +0.016 |
favourite | +0.015 |
favorite | +0.014 |
purse | +0.013 |
wrist | +0.013 |
Neg logit contributions | |
hon | -0.016 |
notice | -0.013 |
cal | -0.013 |
overcome | -0.013 |
rep | -0.013 |
Token | gent |
Token ID | 25049 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
hon | +0.035 |
notice | +0.031 |
rep | +0.030 |
overcome | +0.029 |
oblig | +0.029 |
Neg logit contributions | |
possession | -0.033 |
favourite | -0.032 |
favorite | -0.030 |
purse | -0.027 |
brand | -0.026 |
Token | lest |
Token ID | 32712 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
hon | +0.316 |
overcome | +0.300 |
cal | +0.300 |
slit | +0.297 |
thunder | +0.291 |
Neg logit contributions | |
possession | -0.490 |
favourite | -0.428 |
possessions | -0.403 |
favorite | -0.398 |
item | -0.386 |
Token | dinosaurs |
Token ID | 27199 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
possession | +0.027 |
favourite | +0.026 |
favorite | +0.024 |
purse | +0.023 |
wrist | +0.022 |
Neg logit contributions | |
hon | -0.024 |
notice | -0.021 |
rep | -0.020 |
cal | -0.020 |
overcome | -0.020 |
Token | ever |
Token ID | 1683 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
hon | +0.035 |
rep | +0.029 |
cal | +0.029 |
notice | +0.029 |
announcements | +0.028 |
Neg logit contributions | |
possession | -0.049 |
favourite | -0.048 |
favorite | -0.045 |
purse | -0.042 |
wrist | -0.041 |
Token | ," |
Token ID | 553 |
Feature activation | +7.13e-02 |
Pos logit contributions | |
hon | +0.203 |
Jim | +0.177 |
oblig | +0.173 |
Hon | +0.173 |
slit | +0.172 |
Neg logit contributions | |
possession | -0.210 |
favourite | -0.198 |
favorite | -0.197 |
favorites | -0.168 |
possessions | -0.166 |
Token | mom |
Token ID | 1995 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
hon | +0.115 |
notice | +0.101 |
cal | +0.097 |
overcome | +0.095 |
announcements | +0.095 |
Neg logit contributions | |
possession | -0.133 |
favourite | -0.129 |
favorite | -0.121 |
purse | -0.120 |
wrist | -0.111 |
Token | in |
Token ID | 287 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
hon | +0.040 |
cal | +0.035 |
notice | +0.034 |
rep | +0.034 |
hon | +0.034 |
Neg logit contributions | |
possession | -0.027 |
favourite | -0.027 |
favorite | -0.024 |
purse | -0.022 |
wrist | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +9.01e-03 |
Pos logit contributions | |
possession | +0.048 |
favourite | +0.047 |
favorite | +0.044 |
purse | +0.043 |
attered | +0.042 |
Neg logit contributions | |
hon | -0.018 |
notice | -0.015 |
rep | -0.015 |
hon | -0.014 |
cal | -0.014 |
Token | park |
Token ID | 3952 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
hon | +0.017 |
notice | +0.015 |
rep | +0.015 |
hon | +0.015 |
cal | +0.015 |
Neg logit contributions | |
possession | -0.015 |
favourite | -0.014 |
favorite | -0.013 |
purse | -0.012 |
wrist | -0.012 |
Token | on |
Token ID | 319 |
Feature activation | -6.35e-03 |
Pos logit contributions | |
possession | +0.039 |
favourite | +0.038 |
favorite | +0.035 |
purse | +0.034 |
wrist | +0.033 |
Neg logit contributions | |
hon | -0.028 |
cal | -0.023 |
notice | -0.023 |
hon | -0.022 |
rep | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
possession | +0.014 |
favourite | +0.014 |
favorite | +0.013 |
purse | +0.012 |
wrist | +0.012 |
Neg logit contributions | |
hon | -0.008 |
notice | -0.007 |
hon | -0.007 |
cal | -0.007 |
rep | -0.007 |
Token | bree |
Token ID | 20780 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
hon | +0.049 |
rep | +0.042 |
notice | +0.041 |
cal | +0.041 |
oblig | +0.041 |
Neg logit contributions | |
possession | -0.048 |
favourite | -0.047 |
favorite | -0.043 |
purse | -0.041 |
wrist | -0.038 |
Token | zy |
Token ID | 7357 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
hon | +0.383 |
notice | +0.327 |
rep | +0.323 |
cal | +0.323 |
hon | +0.318 |
Neg logit contributions | |
possession | -0.437 |
favourite | -0.420 |
favorite | -0.387 |
purse | -0.375 |
wrist | -0.362 |
Token | day |
Token ID | 1110 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
hon | +0.019 |
cal | +0.017 |
notice | +0.016 |
oblig | +0.016 |
announcements | +0.016 |
Neg logit contributions | |
possession | -0.018 |
favourite | -0.018 |
favorite | -0.016 |
purse | -0.016 |
wrist | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
hon | +0.021 |
cal | +0.018 |
hon | +0.017 |
oblig | +0.017 |
slit | +0.017 |
Neg logit contributions | |
favourite | -0.023 |
possession | -0.023 |
favorite | -0.021 |
purse | -0.020 |
wrist | -0.019 |
Token | Mary |
Token ID | 5335 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
possession | +0.123 |
favourite | +0.115 |
favorite | +0.107 |
purse | +0.106 |
wrist | +0.103 |
Neg logit contributions | |
hon | -0.073 |
notice | -0.064 |
rep | -0.060 |
hon | -0.059 |
cal | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
hon | +0.045 |
notice | +0.038 |
hon | +0.038 |
rep | +0.038 |
cal | +0.037 |
Neg logit contributions | |
possession | -0.051 |
favourite | -0.048 |
favorite | -0.045 |
purse | -0.043 |
wrist | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -4.49e-03 |
Pos logit contributions | |
hon | +0.034 |
RO | +0.030 |
notice | +0.030 |
Run | +0.029 |
slit | +0.029 |
Neg logit contributions | |
favourite | -0.030 |
possession | -0.029 |
favorite | -0.027 |
brand | -0.023 |
item | -0.022 |
Token | The |
Token ID | 383 |
Feature activation | -2.69e-03 |
Pos logit contributions | |
possession | +0.008 |
favourite | +0.007 |
favorite | +0.007 |
purse | +0.007 |
wrist | +0.006 |
Neg logit contributions | |
hon | -0.008 |
notice | -0.007 |
rep | -0.007 |
hon | -0.007 |
cal | -0.007 |
Token | next |
Token ID | 1306 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
possession | +0.005 |
favourite | +0.004 |
purse | +0.004 |
favorite | +0.004 |
wrist | +0.004 |
Neg logit contributions | |
hon | -0.004 |
hon | -0.004 |
rep | -0.004 |
notice | -0.004 |
cal | -0.004 |
Token | day |
Token ID | 1110 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
hon | +0.153 |
announcements | +0.139 |
notice | +0.136 |
overcome | +0.134 |
cal | +0.132 |
Neg logit contributions | |
favourite | -0.102 |
possession | -0.101 |
favorite | -0.094 |
wallet | -0.076 |
purse | -0.076 |
Token | they |
Token ID | 484 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
hon | +0.029 |
rr | +0.024 |
oblig | +0.024 |
announcements | +0.024 |
boom | +0.023 |
Neg logit contributions | |
favourite | -0.029 |
possession | -0.027 |
favorite | -0.026 |
purse | -0.024 |
wallet | -0.022 |
Token | rep |
Token ID | 1128 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
hon | +0.120 |
announcements | +0.098 |
rep | +0.097 |
louder | +0.097 |
oblig | +0.097 |
Neg logit contributions | |
favourite | -0.106 |
possession | -0.104 |
favorite | -0.099 |
purse | -0.091 |
favourites | -0.087 |
Token | ainted |
Token ID | 14215 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
hon | +0.394 |
notice | +0.340 |
rep | +0.335 |
hon | +0.333 |
cal | +0.332 |
Neg logit contributions | |
possession | -0.427 |
favourite | -0.410 |
favorite | -0.377 |
purse | -0.363 |
wrist | -0.348 |
Token | the |
Token ID | 262 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
hon | +0.028 |
oblig | +0.023 |
rr | +0.022 |
RO | +0.022 |
cal | +0.022 |
Neg logit contributions | |
possession | -0.028 |
favourite | -0.027 |
favorite | -0.025 |
purse | -0.024 |
wrist | -0.023 |
Token | tank |
Token ID | 6873 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
hon | +0.031 |
notice | +0.026 |
cal | +0.025 |
oblig | +0.025 |
RO | +0.024 |
Neg logit contributions | |
favourite | -0.030 |
possession | -0.029 |
favorite | -0.028 |
purse | -0.025 |
wallet | -0.025 |
Token | green |
Token ID | 4077 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
possession | +0.079 |
favourite | +0.078 |
favorite | +0.072 |
purse | +0.067 |
brand | +0.063 |
Neg logit contributions | |
hon | -0.087 |
announcements | -0.073 |
oblig | -0.072 |
RO | -0.072 |
rep | -0.071 |
Token | , |
Token ID | 11 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
possession | +0.090 |
favourite | +0.088 |
favorite | +0.081 |
purse | +0.076 |
wrist | +0.072 |
Neg logit contributions | |
hon | -0.085 |
notice | -0.074 |
cal | -0.072 |
oblig | -0.072 |
rep | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
hon | +0.030 |
oblig | +0.024 |
rep | +0.024 |
notice | +0.024 |
cal | +0.023 |
Neg logit contributions | |
possession | -0.042 |
favourite | -0.042 |
favorite | -0.039 |
purse | -0.037 |
wrist | -0.036 |
Token | shore |
Token ID | 15191 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
hon | +0.086 |
rep | +0.075 |
notice | +0.074 |
oblig | +0.073 |
cal | +0.073 |
Neg logit contributions | |
possession | -0.072 |
favourite | -0.070 |
favorite | -0.065 |
purse | -0.061 |
wrist | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
hon | +0.070 |
oblig | +0.061 |
RO | +0.060 |
notice | +0.060 |
cal | +0.059 |
Neg logit contributions | |
possession | -0.064 |
favourite | -0.063 |
favorite | -0.060 |
purse | -0.056 |
wrist | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
hon | +0.017 |
oblig | +0.014 |
notice | +0.014 |
rep | +0.014 |
cal | +0.014 |
Neg logit contributions | |
possession | -0.019 |
favourite | -0.019 |
favorite | -0.018 |
purse | -0.017 |
wrist | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +6.75e-02 |
Pos logit contributions | |
hon | +0.031 |
oblig | +0.025 |
rep | +0.025 |
notice | +0.024 |
cal | +0.024 |
Neg logit contributions | |
possession | -0.045 |
favourite | -0.044 |
favorite | -0.042 |
purse | -0.039 |
wrist | -0.038 |
Token | se |
Token ID | 384 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
hon | +0.120 |
oblig | +0.101 |
notice | +0.099 |
overcome | +0.098 |
rep | +0.097 |
Neg logit contributions | |
favourite | -0.118 |
possession | -0.117 |
favorite | -0.111 |
purse | -0.104 |
wallet | -0.099 |
Token | ag |
Token ID | 363 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
rep | +0.370 |
hon | +0.335 |
cal | +0.330 |
thunder | +0.329 |
notice | +0.327 |
Neg logit contributions | |
possession | -0.444 |
favourite | -0.404 |
favorite | -0.388 |
attered | -0.365 |
shirt | -0.355 |
Token | ull |
Token ID | 724 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
hon | +0.115 |
cal | +0.097 |
rep | +0.094 |
notice | +0.092 |
oblig | +0.086 |
Neg logit contributions | |
possession | -0.298 |
favourite | -0.285 |
favorite | -0.271 |
purse | -0.254 |
attered | -0.252 |
Token | s |
Token ID | 82 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
hon | +0.213 |
oblig | +0.202 |
notice | +0.198 |
cal | +0.198 |
rep | +0.197 |
Neg logit contributions | |
favourite | -0.124 |
possession | -0.118 |
favorite | -0.114 |
purse | -0.096 |
attered | -0.095 |
Token | flying |
Token ID | 7348 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
favourite | +0.060 |
favorite | +0.056 |
possession | +0.054 |
purse | +0.049 |
wrist | +0.047 |
Neg logit contributions | |
hon | -0.072 |
oblig | -0.066 |
cal | -0.062 |
rr | -0.061 |
announcements | -0.061 |
Token | in |
Token ID | 287 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
hon | +0.126 |
oblig | +0.116 |
rep | +0.112 |
notice | +0.110 |
announcements | +0.107 |
Neg logit contributions | |
possession | -0.087 |
favourite | -0.086 |
favorite | -0.082 |
attered | -0.068 |
purse | -0.065 |
Token | Gary |
Token ID | 10936 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
favourite | +0.009 |
possession | +0.009 |
favorite | +0.008 |
favorites | +0.007 |
favourites | +0.007 |
Neg logit contributions | |
hon | -0.009 |
rep | -0.007 |
slit | -0.007 |
notice | -0.007 |
RO | -0.007 |
Token | lessons |
Token ID | 11658 |
Feature activation | +7.41e-03 |
Pos logit contributions | |
hon | +0.029 |
notice | +0.025 |
rep | +0.025 |
hon | +0.025 |
cal | +0.024 |
Neg logit contributions | |
possession | -0.022 |
favourite | -0.021 |
favorite | -0.020 |
purse | -0.018 |
wrist | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | -1.80e-03 |
Pos logit contributions | |
hon | +0.012 |
notice | +0.010 |
rep | +0.010 |
cal | +0.009 |
oblig | +0.009 |
Neg logit contributions | |
possession | -0.015 |
favourite | -0.014 |
favorite | -0.013 |
purse | -0.012 |
wrist | -0.012 |
Token | kindness |
Token ID | 23887 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
possession | +0.004 |
favourite | +0.004 |
favorite | +0.003 |
purse | +0.003 |
wrist | +0.003 |
Neg logit contributions | |
hon | -0.003 |
rep | -0.002 |
cal | -0.002 |
hon | -0.002 |
notice | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -9.32e-03 |
Pos logit contributions | |
hon | +0.027 |
notice | +0.022 |
rep | +0.022 |
oblig | +0.022 |
hon | +0.022 |
Neg logit contributions | |
possession | -0.033 |
favourite | -0.031 |
favorite | -0.029 |
purse | -0.027 |
wrist | -0.027 |
Token | gent |
Token ID | 25049 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
possession | +0.017 |
favourite | +0.017 |
favorite | +0.016 |
purse | +0.014 |
wrist | +0.014 |
Neg logit contributions | |
hon | -0.016 |
notice | -0.013 |
rep | -0.013 |
oblig | -0.013 |
cal | -0.013 |
Token | leness |
Token ID | 48795 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
hon | +0.363 |
cal | +0.350 |
thunder | +0.349 |
rep | +0.342 |
slit | +0.341 |
Neg logit contributions | |
possession | -0.427 |
possessions | -0.360 |
favourite | -0.349 |
attered | -0.331 |
favorite | -0.321 |
Token | . |
Token ID | 13 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
possession | +0.056 |
favourite | +0.054 |
favorite | +0.051 |
purse | +0.047 |
wrist | +0.046 |
Neg logit contributions | |
hon | -0.044 |
notice | -0.038 |
rep | -0.036 |
oblig | -0.036 |
cal | -0.035 |
Token | Gary |
Token ID | 10936 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
possession | +0.019 |
favourite | +0.018 |
favorite | +0.017 |
purse | +0.016 |
wrist | +0.016 |
Neg logit contributions | |
hon | -0.017 |
notice | -0.015 |
rep | -0.014 |
hon | -0.014 |
cal | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
hon | +0.114 |
boom | +0.099 |
notice | +0.099 |
rep | +0.098 |
announcements | +0.097 |
Neg logit contributions | |
favourite | -0.093 |
possession | -0.091 |
favorite | -0.086 |
wrist | -0.073 |
purse | -0.072 |
Token | so |
Token ID | 523 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
favourite | +0.022 |
possession | +0.022 |
favorite | +0.020 |
brand | +0.018 |
purse | +0.018 |
Neg logit contributions | |
hon | -0.025 |
Run | -0.021 |
rr | -0.020 |
boom | -0.020 |
slit | -0.020 |
Token | new |
Token ID | 649 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
possession | +0.009 |
favourite | +0.009 |
favorite | +0.008 |
purse | +0.008 |
item | +0.007 |
Neg logit contributions | |
hon | -0.008 |
rep | -0.007 |
announcements | -0.007 |
cal | -0.007 |
boom | -0.006 |
Token | number |
Token ID | 1271 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
possession | +0.083 |
favourite | +0.082 |
favorite | +0.074 |
purse | +0.068 |
favorites | +0.066 |
Neg logit contributions | |
hon | -0.090 |
notice | -0.076 |
rep | -0.076 |
cal | -0.076 |
overcome | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | -4.76e-03 |
Pos logit contributions | |
favourite | +0.086 |
possession | +0.086 |
favorite | +0.080 |
purse | +0.075 |
wrist | +0.071 |
Neg logit contributions | |
hon | -0.073 |
notice | -0.062 |
cal | -0.060 |
rep | -0.060 |
RO | -0.060 |
Token | louder |
Token ID | 27089 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
favourite | +0.009 |
possession | +0.008 |
favorite | +0.008 |
purse | +0.008 |
wrist | +0.007 |
Neg logit contributions | |
hon | -0.008 |
rr | -0.006 |
oblig | -0.006 |
notice | -0.006 |
RO | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
hon | +0.446 |
RO | +0.394 |
cal | +0.388 |
oblig | +0.378 |
rep | +0.375 |
Neg logit contributions | |
possession | -0.311 |
favourite | -0.295 |
favorite | -0.270 |
wrist | -0.245 |
purse | -0.232 |
Token | louder |
Token ID | 27089 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
hon | +0.032 |
rr | +0.027 |
RO | +0.025 |
oblig | +0.024 |
iver | +0.023 |
Neg logit contributions | |
favourite | -0.037 |
favorite | -0.034 |
possession | -0.033 |
brand | -0.033 |
purse | -0.029 |
Token | Pe |
Token ID | 2631 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
hon | +0.454 |
RO | +0.398 |
cal | +0.393 |
oblig | +0.383 |
Jim | +0.380 |
Neg logit contributions | |
possession | -0.343 |
favourite | -0.330 |
favorite | -0.300 |
wrist | -0.266 |
purse | -0.248 |
Token | aches |
Token ID | 3694 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
hon | +0.184 |
notice | +0.184 |
thunder | +0.174 |
boom | +0.172 |
rep | +0.171 |
Neg logit contributions | |
possession | -0.194 |
favourite | -0.194 |
favorite | -0.179 |
belt | -0.176 |
purse | -0.175 |
Token | â |
Token ID | 22940 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
hon | +0.079 |
announcements | +0.069 |
Hon | +0.067 |
cal | +0.067 |
rep | +0.067 |
Neg logit contributions | |
favourite | -0.074 |
favorite | -0.068 |
possession | -0.066 |
attered | -0.065 |
robe | -0.063 |
Token | € |
Token ID | 26391 |
Feature activation | +8.04e-02 |
Pos logit contributions | |
hon | +0.210 |
notice | +0.199 |
louder | +0.193 |
announcements | +0.190 |
swarm | +0.187 |
Neg logit contributions | |
possession | -0.258 |
favourite | -0.240 |
wrist | -0.230 |
attered | -0.230 |
purse | -0.223 |
Token | ™ |
Token ID | 8151 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
hon | +0.171 |
louder | +0.168 |
cal | +0.162 |
rep | +0.157 |
notice | +0.157 |
Neg logit contributions | |
favourite | -0.100 |
possession | -0.099 |
favorite | -0.087 |
purse | -0.084 |
wrist | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -1.28e-03 |
Pos logit contributions | |
possession | +0.000 |
favourite | +0.000 |
favorite | +0.000 |
purse | +0.000 |
brand | +0.000 |
Neg logit contributions | |
hon | -0.000 |
oblig | -0.000 |
announcements | -0.000 |
RO | -0.000 |
hon | -0.000 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
possession | +0.002 |
favourite | +0.002 |
favorite | +0.002 |
purse | +0.002 |
wrist | +0.002 |
Neg logit contributions | |
hon | -0.002 |
notice | -0.002 |
hon | -0.002 |
cal | -0.002 |
rep | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
hon | +0.046 |
announcements | +0.038 |
rr | +0.037 |
hon | +0.037 |
RO | +0.037 |
Neg logit contributions | |
favourite | -0.037 |
possession | -0.036 |
favorite | -0.032 |
purse | -0.030 |
microscope | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
hon | +0.024 |
rr | +0.020 |
announcements | +0.020 |
qu | +0.019 |
hon | +0.019 |
Neg logit contributions | |
favourite | -0.021 |
possession | -0.020 |
favorite | -0.018 |
brand | -0.017 |
purse | -0.016 |
Token | worm |
Token ID | 23849 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
hon | +0.091 |
notice | +0.078 |
hon | +0.077 |
announcements | +0.077 |
rep | +0.077 |
Neg logit contributions | |
favourite | -0.069 |
possession | -0.069 |
favorite | -0.063 |
purse | -0.060 |
wallet | -0.055 |
Token | slit |
Token ID | 40724 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
hon | +0.016 |
oblig | +0.013 |
rep | +0.013 |
notice | +0.013 |
announcements | +0.013 |
Neg logit contributions | |
favourite | -0.020 |
possession | -0.019 |
favorite | -0.018 |
purse | -0.017 |
wrist | -0.016 |
Token | hered |
Token ID | 6083 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
hon | +0.357 |
announcements | +0.311 |
cal | +0.306 |
rep | +0.304 |
notice | +0.304 |
Neg logit contributions | |
favourite | -0.412 |
possession | -0.399 |
purse | -0.377 |
robe | -0.363 |
favorite | -0.361 |
Token | away |
Token ID | 1497 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
hon | +0.062 |
rep | +0.056 |
oblig | +0.054 |
notice | +0.053 |
cal | +0.053 |
Neg logit contributions | |
possession | -0.044 |
favourite | -0.042 |
favorite | -0.038 |
purse | -0.035 |
attered | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
possession | +0.041 |
favourite | +0.040 |
favorite | +0.036 |
purse | +0.033 |
brand | +0.032 |
Neg logit contributions | |
hon | -0.055 |
oblig | -0.047 |
announcements | -0.046 |
rr | -0.046 |
Jim | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
possession | +0.048 |
favourite | +0.046 |
favorite | +0.042 |
purse | +0.040 |
brand | +0.039 |
Neg logit contributions | |
hon | -0.040 |
announcements | -0.030 |
oblig | -0.030 |
hon | -0.030 |
notice | -0.030 |
Token | area |
Token ID | 1989 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
hon | +0.102 |
notice | +0.087 |
hon | +0.087 |
rep | +0.086 |
announcements | +0.086 |
Neg logit contributions | |
possession | -0.077 |
favourite | -0.076 |
favorite | -0.069 |
purse | -0.066 |
wrist | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
possession | +0.117 |
favourite | +0.115 |
favorite | +0.111 |
purse | +0.104 |
wrist | +0.098 |
Neg logit contributions | |
hon | -0.098 |
cal | -0.085 |
announcements | -0.085 |
rep | -0.084 |
oblig | -0.084 |
Token | it |
Token ID | 340 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
favourite | +0.028 |
possession | +0.028 |
favorite | +0.026 |
purse | +0.024 |
wrist | +0.023 |
Neg logit contributions | |
hon | -0.030 |
oblig | -0.026 |
overcome | -0.025 |
notice | -0.025 |
rep | -0.025 |
Token | twisted |
Token ID | 19074 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
favourite | +0.054 |
favorite | +0.052 |
possession | +0.050 |
attered | +0.042 |
purse | +0.041 |
Neg logit contributions | |
hon | -0.062 |
oblig | -0.056 |
rep | -0.055 |
notice | -0.054 |
announcements | -0.053 |
Token | around |
Token ID | 1088 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
hon | +0.018 |
oblig | +0.016 |
rep | +0.016 |
notice | +0.016 |
overcome | +0.015 |
Neg logit contributions | |
possession | -0.018 |
favourite | -0.017 |
favorite | -0.016 |
purse | -0.015 |
brand | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
possession | +0.116 |
favourite | +0.111 |
favorite | +0.103 |
purse | +0.097 |
wrist | +0.093 |
Neg logit contributions | |
hon | -0.097 |
rep | -0.078 |
oblig | -0.078 |
cal | -0.078 |
announcements | -0.076 |
Token | slit |
Token ID | 40724 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
possession | +0.039 |
favourite | +0.038 |
favorite | +0.036 |
purse | +0.033 |
brand | +0.032 |
Neg logit contributions | |
hon | -0.035 |
oblig | -0.028 |
rep | -0.028 |
rr | -0.027 |
notice | -0.027 |
Token | hered |
Token ID | 6083 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
hon | +0.362 |
announcements | +0.325 |
cal | +0.318 |
rep | +0.312 |
thunder | +0.312 |
Neg logit contributions | |
favourite | -0.407 |
possession | -0.404 |
purse | -0.369 |
robe | -0.365 |
favorite | -0.363 |
Token | away |
Token ID | 1497 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
hon | +0.049 |
rep | +0.045 |
oblig | +0.042 |
cal | +0.041 |
overcome | +0.040 |
Neg logit contributions | |
possession | -0.037 |
favourite | -0.034 |
favorite | -0.032 |
purse | -0.029 |
wrist | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -6.80e-03 |
Pos logit contributions | |
possession | +0.036 |
favourite | +0.034 |
favorite | +0.031 |
purse | +0.029 |
robe | +0.027 |
Neg logit contributions | |
hon | -0.042 |
oblig | -0.035 |
Jim | -0.034 |
announcements | -0.033 |
rr | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
possession | +0.011 |
favourite | +0.010 |
favorite | +0.009 |
purse | +0.009 |
wrist | +0.009 |
Neg logit contributions | |
hon | -0.013 |
notice | -0.012 |
cal | -0.011 |
hon | -0.011 |
announcements | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
hon | +0.069 |
notice | +0.061 |
rep | +0.060 |
cal | +0.060 |
hon | +0.059 |
Neg logit contributions | |
possession | -0.061 |
favourite | -0.059 |
favorite | -0.053 |
purse | -0.051 |
wrist | -0.048 |
Token | ignorant |
Token ID | 20109 |
Feature activation | -7.51e-03 |
Pos logit contributions | |
possession | +0.032 |
favourite | +0.031 |
favorite | +0.029 |
purse | +0.027 |
wrist | +0.026 |
Neg logit contributions | |
hon | -0.028 |
notice | -0.023 |
rep | -0.023 |
hon | -0.023 |
cal | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | +4.93e-03 |
Pos logit contributions | |
possession | +0.014 |
favourite | +0.014 |
favorite | +0.013 |
purse | +0.012 |
wrist | +0.012 |
Neg logit contributions | |
hon | -0.012 |
notice | -0.010 |
hon | -0.010 |
rep | -0.010 |
cal | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
hon | +0.007 |
notice | +0.006 |
rep | +0.006 |
hon | +0.005 |
cal | +0.005 |
Neg logit contributions | |
possession | -0.011 |
favourite | -0.010 |
favorite | -0.010 |
purse | -0.009 |
wrist | -0.009 |
Token | dangers |
Token ID | 16169 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
hon | +0.045 |
rep | +0.039 |
notice | +0.038 |
oblig | +0.038 |
cal | +0.038 |
Neg logit contributions | |
possession | -0.040 |
favourite | -0.039 |
favorite | -0.036 |
purse | -0.034 |
favorites | -0.031 |
Token | that |
Token ID | 326 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
hon | +0.178 |
rep | +0.159 |
oblig | +0.156 |
notice | +0.153 |
overcome | +0.153 |
Neg logit contributions | |
possession | -0.168 |
favourite | -0.153 |
favorite | -0.142 |
purse | -0.142 |
wrist | -0.126 |
Token | lur |
Token ID | 20605 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
possession | +0.040 |
favourite | +0.038 |
favorite | +0.036 |
purse | +0.035 |
robe | +0.032 |
Neg logit contributions | |
hon | -0.035 |
rep | -0.029 |
oblig | -0.029 |
cal | -0.029 |
louder | -0.029 |
Token | ked |
Token ID | 9091 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
hon | +0.404 |
rep | +0.372 |
cal | +0.364 |
notice | +0.353 |
oblig | +0.353 |
Neg logit contributions | |
possession | -0.394 |
favourite | -0.372 |
favorite | -0.349 |
purse | -0.329 |
attered | -0.319 |
Token | in |
Token ID | 287 |
Feature activation | -2.18e-03 |
Pos logit contributions | |
possession | +0.018 |
favourite | +0.018 |
favorite | +0.017 |
purse | +0.015 |
attered | +0.014 |
Neg logit contributions | |
hon | -0.021 |
oblig | -0.019 |
rep | -0.018 |
notice | -0.018 |
cal | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
possession | +0.005 |
favourite | +0.004 |
favorite | +0.004 |
purse | +0.004 |
wrist | +0.004 |
Neg logit contributions | |
hon | -0.003 |
rep | -0.002 |
oblig | -0.002 |
hon | -0.002 |
cal | -0.002 |
Token | shadows |
Token ID | 16187 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
hon | +0.050 |
rep | +0.044 |
oblig | +0.043 |
notice | +0.042 |
overcome | +0.042 |
Neg logit contributions | |
possession | -0.042 |
favourite | -0.040 |
favorite | -0.037 |
purse | -0.036 |
wallet | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
hon | +0.040 |
oblig | +0.035 |
rep | +0.035 |
notice | +0.034 |
cal | +0.033 |
Neg logit contributions | |
possession | -0.042 |
favourite | -0.041 |
favorite | -0.038 |
purse | -0.036 |
wrist | -0.034 |
Token | The |
Token ID | 383 |
Feature activation | +9.08e-03 |
Pos logit contributions | |
possession | +0.003 |
favourite | +0.003 |
favorite | +0.002 |
purse | +0.002 |
wrist | +0.002 |
Neg logit contributions | |
hon | -0.004 |
notice | -0.004 |
rep | -0.004 |
cal | -0.004 |
hon | -0.004 |
Token | snake |
Token ID | 17522 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
hon | +0.015 |
hon | +0.014 |
louder | +0.013 |
notice | +0.013 |
boom | +0.013 |
Neg logit contributions | |
possession | -0.017 |
favourite | -0.016 |
favorite | -0.014 |
purse | -0.014 |
attered | -0.014 |
Token | let |
Token ID | 1309 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
favourite | +0.023 |
possession | +0.023 |
favorite | +0.022 |
purse | +0.021 |
wrist | +0.020 |
Neg logit contributions | |
hon | -0.018 |
rep | -0.016 |
oblig | -0.015 |
notice | -0.015 |
announcements | -0.015 |
Token | go |
Token ID | 467 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
hon | +0.137 |
rep | +0.129 |
oblig | +0.119 |
announcements | +0.116 |
rr | +0.114 |
Neg logit contributions | |
favourite | -0.165 |
possession | -0.163 |
favorite | -0.160 |
purse | -0.157 |
robe | -0.141 |
Token | and |
Token ID | 290 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
hon | +0.159 |
rep | +0.142 |
announcements | +0.138 |
cal | +0.133 |
notice | +0.133 |
Neg logit contributions | |
possession | -0.185 |
favourite | -0.173 |
favorite | -0.162 |
purse | -0.161 |
wrist | -0.148 |
Token | slit |
Token ID | 40724 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
possession | +0.018 |
favourite | +0.017 |
purse | +0.017 |
favorite | +0.016 |
wallet | +0.015 |
Neg logit contributions | |
hon | -0.023 |
rr | -0.023 |
announcements | -0.020 |
rep | -0.019 |
oblig | -0.019 |
Token | hered |
Token ID | 6083 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
hon | +0.326 |
rep | +0.284 |
announcements | +0.275 |
cal | +0.275 |
notice | +0.273 |
Neg logit contributions | |
possession | -0.458 |
favourite | -0.447 |
favorite | -0.414 |
purse | -0.412 |
wrist | -0.387 |
Token | away |
Token ID | 1497 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
hon | +0.062 |
rep | +0.062 |
oblig | +0.058 |
cal | +0.054 |
announcements | +0.053 |
Neg logit contributions | |
possession | -0.054 |
favourite | -0.048 |
favorite | -0.047 |
purse | -0.045 |
wrist | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +8.33e-04 |
Pos logit contributions | |
possession | +0.027 |
favourite | +0.024 |
favorite | +0.024 |
purse | +0.021 |
where | +0.020 |
Neg logit contributions | |
hon | -0.038 |
oblig | -0.034 |
Jim | -0.032 |
announcements | -0.031 |
Andrew | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
hon | +0.002 |
notice | +0.001 |
rep | +0.001 |
cal | +0.001 |
hon | +0.001 |
Neg logit contributions | |
possession | -0.001 |
favourite | -0.001 |
favorite | -0.001 |
purse | -0.001 |
wrist | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
hon | +0.082 |
notice | +0.071 |
rep | +0.070 |
hon | +0.069 |
cal | +0.069 |
Neg logit contributions | |
possession | -0.092 |
favourite | -0.088 |
favorite | -0.081 |
purse | -0.078 |
wrist | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -3.14e-03 |
Pos logit contributions | |
possession | +0.032 |
favourite | +0.031 |
favorite | +0.029 |
purse | +0.027 |
wrist | +0.026 |
Neg logit contributions | |
hon | -0.035 |
notice | -0.030 |
rep | -0.030 |
cal | -0.030 |
hon | -0.029 |
Token | soon |
Token ID | 2582 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
possession | +0.006 |
favourite | +0.006 |
favorite | +0.006 |
purse | +0.005 |
wrist | +0.005 |
Neg logit contributions | |
hon | -0.005 |
notice | -0.004 |
rep | -0.004 |
cal | -0.004 |
hon | -0.004 |
Token | noticed |
Token ID | 6810 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
hon | +0.114 |
rr | +0.102 |
boom | +0.096 |
oblig | +0.094 |
rep | +0.093 |
Neg logit contributions | |
favourite | -0.099 |
possession | -0.095 |
favorite | -0.092 |
purse | -0.080 |
wallet | -0.078 |
Token | a |
Token ID | 257 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
possession | +0.069 |
favourite | +0.067 |
favorite | +0.062 |
purse | +0.057 |
brand | +0.054 |
Neg logit contributions | |
hon | -0.075 |
notice | -0.065 |
oblig | -0.063 |
rep | -0.062 |
cal | -0.062 |
Token | narrow |
Token ID | 7135 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
hon | +0.075 |
notice | +0.067 |
hon | +0.066 |
rep | +0.066 |
cal | +0.066 |
Neg logit contributions | |
possession | -0.082 |
favourite | -0.078 |
favorite | -0.072 |
purse | -0.069 |
wrist | -0.067 |
Token | slit |
Token ID | 40724 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
hon | +0.093 |
notice | +0.086 |
hon | +0.086 |
rep | +0.084 |
cal | +0.083 |
Neg logit contributions | |
possession | -0.072 |
favourite | -0.071 |
favorite | -0.064 |
purse | -0.058 |
attered | -0.057 |
Token | in |
Token ID | 287 |
Feature activation | +5.57e-04 |
Pos logit contributions | |
hon | +0.359 |
cal | +0.314 |
notice | +0.312 |
announcements | +0.305 |
rep | +0.304 |
Neg logit contributions | |
favourite | -0.393 |
possession | -0.388 |
purse | -0.368 |
robe | -0.357 |
favorite | -0.344 |
Token | the |
Token ID | 262 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
hon | +0.001 |
rep | +0.001 |
oblig | +0.001 |
notice | +0.001 |
overcome | +0.001 |
Neg logit contributions | |
possession | -0.001 |
favourite | -0.001 |
favorite | -0.001 |
purse | -0.001 |
wrist | -0.001 |
Token | books |
Token ID | 3835 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
thunder | +0.065 |
notice | +0.063 |
cal | +0.062 |
boom | +0.061 |
hon | +0.060 |
Neg logit contributions | |
possession | -0.053 |
favourite | -0.051 |
uggets | -0.046 |
wrist | -0.045 |
favorite | -0.044 |
Token | he |
Token ID | 258 |
Feature activation | -9.57e-03 |
Pos logit contributions | |
possession | +0.016 |
favourite | +0.015 |
purse | +0.014 |
favorite | +0.014 |
wrist | +0.013 |
Neg logit contributions | |
hon | -0.021 |
oblig | -0.018 |
cal | -0.018 |
rep | -0.018 |
announcements | -0.018 |
Token | lf |
Token ID | 1652 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
possession | +0.015 |
favourite | +0.014 |
favorite | +0.013 |
purse | +0.013 |
wrist | +0.012 |
Neg logit contributions | |
hon | -0.019 |
cal | -0.016 |
rep | -0.016 |
notice | -0.016 |
hon | -0.016 |
Token | shells |
Token ID | 19679 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
hon | +0.112 |
notice | +0.095 |
oblig | +0.095 |
rep | +0.094 |
overcome | +0.092 |
Neg logit contributions | |
possession | -0.102 |
favourite | -0.101 |
favorite | -0.094 |
purse | -0.089 |
wrist | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
hon | +0.074 |
oblig | +0.066 |
announcements | +0.065 |
cal | +0.064 |
RO | +0.064 |
Neg logit contributions | |
possession | -0.072 |
favourite | -0.071 |
favorite | -0.068 |
purse | -0.066 |
wrist | -0.063 |
Token | They |
Token ID | 1119 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
hon | +0.024 |
notice | +0.020 |
rep | +0.020 |
hon | +0.020 |
cal | +0.020 |
Neg logit contributions | |
possession | -0.025 |
favourite | -0.024 |
favorite | -0.022 |
purse | -0.021 |
wrist | -0.020 |
Token | hear |
Token ID | 3285 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
hon | +0.124 |
louder | +0.100 |
oblig | +0.098 |
rep | +0.097 |
rr | +0.097 |
Neg logit contributions | |
favourite | -0.146 |
favorite | -0.141 |
wrist | -0.132 |
purse | -0.132 |
possession | -0.130 |
Token | the |
Token ID | 262 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
hon | +0.199 |
oblig | +0.177 |
overcome | +0.175 |
notice | +0.170 |
rep | +0.167 |
Neg logit contributions | |
favourite | -0.193 |
favorite | -0.187 |
possession | -0.183 |
purse | -0.167 |
brand | -0.158 |
Token | se |
Token ID | 384 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
hon | +0.093 |
notice | +0.077 |
oblig | +0.076 |
overcome | +0.076 |
RO | +0.075 |
Neg logit contributions | |
favourite | -0.101 |
possession | -0.099 |
favorite | -0.097 |
purse | -0.090 |
wallet | -0.088 |
Token | ag |
Token ID | 363 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
rep | +0.380 |
hon | +0.337 |
thunder | +0.331 |
overcome | +0.326 |
notice | +0.324 |
Neg logit contributions | |
possession | -0.417 |
favourite | -0.380 |
favorite | -0.375 |
shirt | -0.346 |
attered | -0.335 |
Token | ull |
Token ID | 724 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
hon | +0.110 |
cal | +0.096 |
rep | +0.095 |
notice | +0.093 |
oblig | +0.086 |
Neg logit contributions | |
possession | -0.265 |
favourite | -0.251 |
favorite | -0.240 |
robe | -0.224 |
wrist | -0.223 |
Token | s |
Token ID | 82 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
oblig | +0.191 |
Jim | +0.189 |
hon | +0.186 |
rep | +0.186 |
notice | +0.179 |
Neg logit contributions | |
favorite | -0.098 |
favourite | -0.097 |
possession | -0.089 |
purse | -0.082 |
brand | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
favourite | +0.038 |
favorite | +0.038 |
possession | +0.036 |
wrist | +0.036 |
purse | +0.033 |
Neg logit contributions | |
hon | -0.079 |
oblig | -0.078 |
RO | -0.075 |
swarm | -0.074 |
Jim | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
hon | +0.023 |
qu | +0.020 |
rr | +0.019 |
Jim | +0.019 |
oblig | +0.018 |
Neg logit contributions | |
favourite | -0.029 |
favorite | -0.029 |
possession | -0.028 |
wrist | -0.027 |
purse | -0.027 |
Token | he |
Token ID | 339 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
hon | +0.001 |
rr | +0.001 |
oblig | +0.001 |
announcements | +0.001 |
notice | +0.001 |
Neg logit contributions | |
favourite | -0.001 |
possession | -0.001 |
favorite | -0.001 |
purse | -0.001 |
brand | -0.001 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
hon | +0.070 |
rep | +0.058 |
notice | +0.057 |
oblig | +0.057 |
announcements | +0.057 |
Neg logit contributions | |
favourite | -0.073 |
possession | -0.070 |
favorite | -0.068 |
wrist | -0.060 |
brand | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
hon | +0.031 |
rep | +0.026 |
notice | +0.026 |
hon | +0.025 |
announcements | +0.025 |
Neg logit contributions | |
possession | -0.039 |
favourite | -0.038 |
favorite | -0.035 |
purse | -0.034 |
wrist | -0.032 |
Token | healing |
Token ID | 11516 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
hon | +0.081 |
notice | +0.066 |
announcements | +0.065 |
hon | +0.065 |
oblig | +0.065 |
Neg logit contributions | |
favourite | -0.074 |
possession | -0.072 |
favorite | -0.067 |
purse | -0.061 |
wrist | -0.058 |
Token | power |
Token ID | 1176 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
hon | +0.073 |
notice | +0.063 |
rep | +0.062 |
hon | +0.061 |
cal | +0.061 |
Neg logit contributions | |
possession | -0.072 |
favourite | -0.068 |
favorite | -0.062 |
purse | -0.060 |
wrist | -0.057 |
Token | cour |
Token ID | 1093 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
hon | +0.032 |
notice | +0.027 |
announcements | +0.026 |
rep | +0.026 |
hon | +0.026 |
Neg logit contributions | |
possession | -0.035 |
favourite | -0.033 |
favorite | -0.030 |
purse | -0.029 |
brand | -0.028 |
Token | sing |
Token ID | 12215 |
Feature activation | +8.84e-02 |
Pos logit contributions | |
hon | +0.380 |
notice | +0.327 |
rep | +0.322 |
hon | +0.321 |
cal | +0.316 |
Neg logit contributions | |
possession | -0.422 |
favourite | -0.404 |
favorite | -0.372 |
purse | -0.362 |
wrist | -0.343 |
Token | through |
Token ID | 832 |
Feature activation | +5.31e-03 |
Pos logit contributions | |
hon | +0.164 |
rep | +0.148 |
oblig | +0.145 |
notice | +0.139 |
overcome | +0.137 |
Neg logit contributions | |
possession | -0.149 |
favourite | -0.138 |
favorite | -0.129 |
purse | -0.114 |
item | -0.110 |
Token | his |
Token ID | 465 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
hon | +0.008 |
oblig | +0.007 |
announcements | +0.006 |
rep | +0.006 |
overcome | +0.006 |
Neg logit contributions | |
possession | -0.011 |
favourite | -0.010 |
favorite | -0.009 |
purse | -0.009 |
wrist | -0.009 |
Token | body |
Token ID | 1767 |
Feature activation | +4.35e-03 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.023 |
favorite | +0.020 |
purse | +0.019 |
wrist | +0.018 |
Neg logit contributions | |
hon | -0.034 |
notice | -0.030 |
rep | -0.030 |
hon | -0.029 |
cal | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.23e-03 |
Pos logit contributions | |
hon | +0.007 |
notice | +0.006 |
RO | +0.006 |
oblig | +0.006 |
announcements | +0.006 |
Neg logit contributions | |
possession | -0.008 |
favourite | -0.008 |
favorite | -0.007 |
purse | -0.007 |
wrist | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
hon | +0.034 |
RO | +0.031 |
notice | +0.029 |
Jim | +0.029 |
announcements | +0.029 |
Neg logit contributions | |
favourite | -0.028 |
possession | -0.027 |
favorite | -0.027 |
purse | -0.024 |
wrist | -0.023 |
Token | He |
Token ID | 679 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
favourite | +0.059 |
possession | +0.058 |
favorite | +0.055 |
purse | +0.049 |
robe | +0.046 |
Neg logit contributions | |
hon | -0.080 |
announcements | -0.069 |
hon | -0.069 |
notice | -0.068 |
overcome | -0.068 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
possession | +0.052 |
favourite | +0.051 |
favorite | +0.047 |
purse | +0.046 |
wrist | +0.044 |
Neg logit contributions | |
hon | -0.034 |
notice | -0.029 |
rep | -0.028 |
cal | -0.027 |
oblig | -0.027 |
Token | Lily |
Token ID | 20037 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
possession | +0.250 |
favourite | +0.242 |
favorite | +0.226 |
purse | +0.217 |
wrist | +0.206 |
Neg logit contributions | |
hon | -0.178 |
notice | -0.150 |
rep | -0.149 |
hon | -0.146 |
cal | -0.144 |
Token | his |
Token ID | 465 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
possession | +0.129 |
favourite | +0.125 |
favorite | +0.115 |
purse | +0.111 |
wrist | +0.107 |
Neg logit contributions | |
hon | -0.101 |
notice | -0.086 |
hon | -0.085 |
cal | -0.084 |
rep | -0.084 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
possession | +0.225 |
favourite | +0.207 |
attered | +0.186 |
purse | +0.180 |
favorite | +0.179 |
Neg logit contributions | |
hon | -0.252 |
hon | -0.234 |
rep | -0.230 |
notice | -0.224 |
sne | -0.222 |
Token | gun |
Token ID | 2485 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
possession | +0.422 |
favourite | +0.408 |
favourites | +0.356 |
favorite | +0.356 |
favorites | +0.352 |
Neg logit contributions | |
hon | -0.342 |
hon | -0.316 |
louder | -0.294 |
boom | -0.287 |
rep | -0.287 |
Token | . |
Token ID | 13 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
possession | +0.172 |
favourite | +0.165 |
favorite | +0.152 |
purse | +0.149 |
wrist | +0.145 |
Neg logit contributions | |
hon | -0.095 |
hon | -0.080 |
notice | -0.077 |
rep | -0.077 |
louder | -0.077 |
Token | It |
Token ID | 632 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
possession | +0.076 |
favourite | +0.074 |
favorite | +0.069 |
purse | +0.065 |
wrist | +0.063 |
Neg logit contributions | |
hon | -0.074 |
notice | -0.063 |
rep | -0.063 |
hon | -0.063 |
cal | -0.062 |
Token | was |
Token ID | 373 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
possession | +0.177 |
favourite | +0.173 |
favorite | +0.160 |
purse | +0.151 |
wrist | +0.143 |
Neg logit contributions | |
hon | -0.177 |
notice | -0.155 |
rep | -0.151 |
oblig | -0.150 |
cal | -0.149 |
Token | red |
Token ID | 2266 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
possession | +0.049 |
favourite | +0.046 |
favorite | +0.042 |
purse | +0.042 |
wrist | +0.041 |
Neg logit contributions | |
hon | -0.037 |
hon | -0.032 |
louder | -0.032 |
rep | -0.031 |
notice | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -6.83e-03 |
Pos logit contributions | |
possession | +0.077 |
favourite | +0.076 |
favorite | +0.071 |
purse | +0.069 |
wrist | +0.065 |
Neg logit contributions | |
hon | -0.046 |
hon | -0.036 |
cal | -0.035 |
rep | -0.035 |
notice | -0.035 |
Token | Emma |
Token ID | 18966 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
possession | +0.014 |
favourite | +0.013 |
favorite | +0.012 |
purse | +0.012 |
wrist | +0.012 |
Neg logit contributions | |
hon | -0.010 |
notice | -0.009 |
rep | -0.008 |
hon | -0.008 |
cal | -0.008 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
hon | +0.024 |
notice | +0.020 |
rep | +0.020 |
cal | +0.020 |
hon | +0.020 |
Neg logit contributions | |
possession | -0.027 |
favourite | -0.026 |
favorite | -0.024 |
purse | -0.023 |
wrist | -0.022 |
Token | Lily |
Token ID | 20037 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
possession | +0.242 |
favourite | +0.226 |
wrist | +0.212 |
attered | +0.211 |
purse | +0.209 |
Neg logit contributions | |
hon | -0.120 |
louder | -0.114 |
cal | -0.111 |
notice | -0.110 |
thunder | -0.104 |
Token | her |
Token ID | 607 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
possession | +0.131 |
favourite | +0.123 |
attered | +0.113 |
favorite | +0.111 |
purse | +0.109 |
Neg logit contributions | |
hon | -0.086 |
overcome | -0.078 |
cal | -0.077 |
notice | -0.076 |
hon | -0.076 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
attered | +0.199 |
possession | +0.188 |
resses | +0.164 |
rolley | +0.162 |
umes | +0.162 |
Neg logit contributions | |
hon | -0.260 |
cal | -0.253 |
sne | -0.250 |
hon | -0.249 |
mad | -0.247 |
Token | and |
Token ID | 290 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
possession | +0.419 |
favourite | +0.397 |
wrist | +0.373 |
favourites | +0.365 |
shirt | +0.361 |
Neg logit contributions | |
hon | -0.284 |
hon | -0.281 |
louder | -0.249 |
sne | -0.249 |
boom | -0.247 |
Token | they |
Token ID | 484 |
Feature activation | +6.10e-02 |
Pos logit contributions | |
possession | +0.099 |
favourite | +0.095 |
favorite | +0.088 |
purse | +0.085 |
wrist | +0.083 |
Neg logit contributions | |
hon | -0.061 |
notice | -0.054 |
rep | -0.053 |
hon | -0.052 |
cal | -0.051 |
Token | played |
Token ID | 2826 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
hon | +0.103 |
notice | +0.086 |
rep | +0.086 |
cal | +0.086 |
hon | +0.084 |
Neg logit contributions | |
possession | -0.111 |
favourite | -0.109 |
favorite | -0.101 |
purse | -0.096 |
wrist | -0.092 |
Token | together |
Token ID | 1978 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
hon | +0.035 |
notice | +0.030 |
rep | +0.029 |
hon | +0.029 |
cal | +0.029 |
Neg logit contributions | |
possession | -0.040 |
favourite | -0.039 |
favorite | -0.036 |
purse | -0.035 |
wrist | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
hon | +0.054 |
notice | +0.046 |
rep | +0.046 |
hon | +0.045 |
cal | +0.045 |
Neg logit contributions | |
possession | -0.065 |
favourite | -0.062 |
favorite | -0.058 |
purse | -0.056 |
wrist | -0.053 |
Token | cooler |
Token ID | 19346 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
hon | +0.060 |
notice | +0.053 |
rep | +0.053 |
hon | +0.052 |
cal | +0.052 |
Neg logit contributions | |
possession | -0.042 |
favourite | -0.040 |
favorite | -0.036 |
purse | -0.034 |
wrist | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
hon | +0.037 |
notice | +0.031 |
rep | +0.030 |
cal | +0.030 |
hon | +0.030 |
Neg logit contributions | |
possession | -0.048 |
favourite | -0.046 |
favorite | -0.043 |
purse | -0.041 |
wrist | -0.040 |
Token | Mike |
Token ID | 4995 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
possession | +0.026 |
favourite | +0.026 |
favorite | +0.023 |
purse | +0.022 |
wrist | +0.022 |
Neg logit contributions | |
hon | -0.030 |
notice | -0.026 |
cal | -0.026 |
hon | -0.026 |
rep | -0.025 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
hon | +0.087 |
rep | +0.080 |
louder | +0.079 |
thunder | +0.078 |
boom | +0.078 |
Neg logit contributions | |
possession | -0.070 |
favourite | -0.070 |
favorite | -0.065 |
purse | -0.059 |
wrist | -0.057 |
Token | his |
Token ID | 465 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
possession | +0.235 |
favourite | +0.229 |
favorite | +0.214 |
purse | +0.205 |
wrist | +0.195 |
Neg logit contributions | |
hon | -0.159 |
notice | -0.135 |
rep | -0.132 |
cal | -0.128 |
hon | -0.127 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
possession | +0.196 |
favourite | +0.174 |
attered | +0.162 |
favorite | +0.154 |
purse | +0.152 |
Neg logit contributions | |
hon | -0.240 |
hon | -0.221 |
cal | -0.218 |
rep | -0.217 |
notice | -0.214 |
Token | car |
Token ID | 1097 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
possession | +0.400 |
favourite | +0.372 |
wrist | +0.346 |
attered | +0.343 |
favourites | +0.340 |
Neg logit contributions | |
hon | -0.300 |
hon | -0.298 |
sne | -0.268 |
slit | -0.264 |
boom | -0.253 |
Token | to |
Token ID | 284 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
possession | +0.216 |
favourite | +0.206 |
favorite | +0.186 |
purse | +0.181 |
wrist | +0.177 |
Neg logit contributions | |
hon | -0.189 |
hon | -0.168 |
cal | -0.161 |
notice | -0.160 |
rep | -0.159 |
Token | everyone |
Token ID | 2506 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
hon | +0.038 |
cal | +0.030 |
rep | +0.030 |
notice | +0.030 |
announcements | +0.030 |
Neg logit contributions | |
possession | -0.053 |
favourite | -0.051 |
favorite | -0.048 |
purse | -0.046 |
wrist | -0.045 |
Token | who |
Token ID | 508 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
hon | +0.053 |
notice | +0.044 |
rep | +0.044 |
cal | +0.043 |
announcements | +0.043 |
Neg logit contributions | |
possession | -0.060 |
favourite | -0.059 |
favorite | -0.054 |
purse | -0.051 |
wrist | -0.050 |
Token | walked |
Token ID | 6807 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
possession | +0.155 |
favourite | +0.149 |
favorite | +0.137 |
purse | +0.132 |
wrist | +0.127 |
Neg logit contributions | |
hon | -0.140 |
notice | -0.120 |
hon | -0.119 |
rep | -0.118 |
cal | -0.118 |
Token | Sally |
Token ID | 25737 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
hon | +0.034 |
notice | +0.028 |
rep | +0.028 |
hon | +0.027 |
cal | +0.027 |
Neg logit contributions | |
possession | -0.059 |
favourite | -0.057 |
favorite | -0.053 |
purse | -0.052 |
wrist | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | -5.10e-02 |
Pos logit contributions | |
possession | +0.074 |
favourite | +0.069 |
favorite | +0.063 |
purse | +0.062 |
attered | +0.061 |
Neg logit contributions | |
hon | -0.049 |
notice | -0.043 |
overcome | -0.043 |
rep | -0.042 |
hon | -0.042 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
possession | +0.125 |
favourite | +0.121 |
favorite | +0.114 |
purse | +0.111 |
wrist | +0.108 |
Neg logit contributions | |
hon | -0.054 |
notice | -0.042 |
rep | -0.041 |
hon | -0.041 |
cal | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
possession | +0.263 |
favourite | +0.255 |
favorite | +0.234 |
wrist | +0.233 |
attered | +0.226 |
Neg logit contributions | |
hon | -0.164 |
cal | -0.142 |
notice | -0.141 |
rep | -0.135 |
louder | -0.134 |
Token | his |
Token ID | 465 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
possession | +0.356 |
favourite | +0.343 |
attered | +0.337 |
wrist | +0.317 |
favourites | +0.314 |
Neg logit contributions | |
hon | -0.119 |
cal | -0.111 |
notice | -0.106 |
hon | -0.103 |
hear | -0.101 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
possession | +0.206 |
ase | +0.206 |
attered | +0.196 |
umes | +0.191 |
wrist | +0.187 |
Neg logit contributions | |
sne | -0.209 |
hon | -0.205 |
rep | -0.195 |
mad | -0.192 |
hon | -0.190 |
Token | pistol |
Token ID | 16790 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
possession | +0.393 |
favourite | +0.383 |
wrist | +0.378 |
artments | +0.366 |
favorites | +0.364 |
Neg logit contributions | |
hon | -0.268 |
hon | -0.268 |
sne | -0.259 |
louder | -0.246 |
slit | -0.235 |
Token | . |
Token ID | 13 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
possession | +0.332 |
favourite | +0.321 |
favorite | +0.292 |
purse | +0.292 |
favorites | +0.289 |
Neg logit contributions | |
hon | -0.116 |
hon | -0.098 |
louder | -0.093 |
rep | -0.093 |
cal | -0.089 |
Token | Sally |
Token ID | 25737 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
possession | +0.074 |
favourite | +0.069 |
favorite | +0.064 |
wrist | +0.061 |
purse | +0.061 |
Neg logit contributions | |
hon | -0.038 |
rep | -0.034 |
notice | -0.033 |
cal | -0.031 |
hon | -0.031 |
Token | 's |
Token ID | 338 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
possession | +0.038 |
favourite | +0.036 |
favorite | +0.033 |
purse | +0.032 |
wrist | +0.031 |
Neg logit contributions | |
hon | -0.030 |
notice | -0.025 |
hon | -0.025 |
rep | -0.025 |
cal | -0.025 |
Token | face |
Token ID | 1986 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
possession | +0.124 |
umes | +0.116 |
ase | +0.113 |
artments | +0.108 |
attered | +0.107 |
Neg logit contributions | |
sne | -0.115 |
hon | -0.105 |
mad | -0.105 |
louder | -0.102 |
loud | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
hon | +0.037 |
rep | +0.032 |
cal | +0.032 |
notice | +0.032 |
hon | +0.032 |
Neg logit contributions | |
possession | -0.032 |
favourite | -0.031 |
favorite | -0.029 |
purse | -0.027 |
wrist | -0.025 |
Token | L |
Token ID | 43 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
hon | +0.025 |
rep | +0.022 |
notice | +0.022 |
cal | +0.021 |
announcements | +0.021 |
Neg logit contributions | |
possession | -0.029 |
favourite | -0.028 |
favorite | -0.026 |
purse | -0.025 |
wrist | -0.025 |
Token | ily |
Token ID | 813 |
Feature activation | -7.66e-03 |
Pos logit contributions | |
hon | +0.108 |
notice | +0.089 |
rep | +0.088 |
cal | +0.087 |
hon | +0.086 |
Neg logit contributions | |
possession | -0.143 |
favourite | -0.140 |
favorite | -0.129 |
purse | -0.124 |
wrist | -0.120 |
Token | handed |
Token ID | 10158 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
possession | +0.017 |
favourite | +0.016 |
favorite | +0.015 |
purse | +0.015 |
wrist | +0.014 |
Neg logit contributions | |
hon | -0.010 |
notice | -0.008 |
rep | -0.008 |
hon | -0.008 |
cal | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
possession | +0.287 |
favourite | +0.279 |
wrist | +0.259 |
favorite | +0.256 |
attered | +0.249 |
Neg logit contributions | |
hon | -0.100 |
louder | -0.075 |
hon | -0.074 |
notice | -0.071 |
cal | -0.071 |
Token | purse |
Token ID | 27517 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
attered | +0.237 |
possession | +0.217 |
resses | +0.203 |
uggets | +0.201 |
favourite | +0.196 |
Neg logit contributions | |
the | -0.202 |
come | -0.199 |
mad | -0.197 |
hear | -0.196 |
notice | -0.194 |
Token | to |
Token ID | 284 |
Feature activation | -5.44e-03 |
Pos logit contributions | |
possession | +0.479 |
favourites | +0.471 |
favourite | +0.464 |
favorites | +0.464 |
attered | +0.426 |
Neg logit contributions | |
louder | -0.179 |
hon | -0.175 |
hon | -0.172 |
cal | -0.149 |
come | -0.148 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
possession | +0.010 |
favourite | +0.010 |
favorite | +0.009 |
purse | +0.009 |
wrist | +0.009 |
Neg logit contributions | |
hon | -0.009 |
announcements | -0.007 |
notice | -0.007 |
cal | -0.007 |
overcome | -0.007 |
Token | my |
Token ID | 1820 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.024 |
favorite | +0.022 |
purse | +0.021 |
wrist | +0.020 |
Neg logit contributions | |
hon | -0.016 |
notice | -0.013 |
rep | -0.013 |
cal | -0.013 |
hon | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
possession | +0.140 |
favourite | +0.135 |
favorite | +0.125 |
purse | +0.122 |
attered | +0.120 |
Neg logit contributions | |
hon | -0.068 |
rep | -0.055 |
notice | -0.055 |
cal | -0.054 |
hon | -0.054 |
Token | but |
Token ID | 475 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
possession | +0.091 |
favourite | +0.087 |
favorite | +0.080 |
purse | +0.078 |
wrist | +0.075 |
Neg logit contributions | |
hon | -0.071 |
notice | -0.061 |
rep | -0.060 |
hon | -0.060 |
cal | -0.059 |
Token | Everyday |
Token ID | 48154 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
possession | +0.126 |
favourite | +0.119 |
favorite | +0.109 |
brand | +0.106 |
possessions | +0.105 |
Neg logit contributions | |
hon | -0.097 |
notice | -0.089 |
cal | -0.085 |
rep | -0.085 |
oblig | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
favourite | +0.069 |
possession | +0.069 |
favorite | +0.063 |
purse | +0.062 |
wrist | +0.056 |
Neg logit contributions | |
hon | -0.076 |
hon | -0.064 |
notice | -0.063 |
overcome | -0.063 |
cal | -0.063 |
Token | she |
Token ID | 673 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
possession | +0.068 |
favourite | +0.066 |
favorite | +0.061 |
purse | +0.059 |
wrist | +0.056 |
Neg logit contributions | |
hon | -0.064 |
hon | -0.054 |
notice | -0.054 |
cal | -0.054 |
rep | -0.054 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
possession | +0.029 |
favourite | +0.028 |
favorite | +0.025 |
purse | +0.025 |
wrist | +0.024 |
Neg logit contributions | |
hon | -0.026 |
notice | -0.022 |
hon | -0.022 |
rep | -0.022 |
cal | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
possession | +0.303 |
attered | +0.297 |
belt | +0.286 |
robe | +0.285 |
favourite | +0.282 |
Neg logit contributions | |
hon | -0.107 |
cal | -0.092 |
hear | -0.091 |
louder | -0.091 |
overcome | -0.086 |
Token | purse |
Token ID | 27517 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
attered | +0.305 |
resses | +0.256 |
possession | +0.249 |
artments | +0.249 |
belt | +0.245 |
Neg logit contributions | |
the | -0.177 |
sne | -0.176 |
rep | -0.174 |
," | -0.169 |
cal | -0.164 |
Token | to |
Token ID | 284 |
Feature activation | -1.46e-04 |
Pos logit contributions | |
possession | +0.461 |
favorites | +0.433 |
favourite | +0.432 |
favourites | +0.428 |
attered | +0.425 |
Neg logit contributions | |
come | -0.168 |
overcome | -0.162 |
louder | -0.160 |
hon | -0.150 |
cal | -0.146 |
Token | her |
Token ID | 607 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
possession | +0.000 |
attered | +0.000 |
favourite | +0.000 |
ometer | +0.000 |
favorite | +0.000 |
Neg logit contributions | |
hear | -0.000 |
notice | -0.000 |
hon | -0.000 |
rep | -0.000 |
come | -0.000 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.94e-03 |
Pos logit contributions | |
attered | +0.223 |
possession | +0.222 |
shirt | +0.194 |
artments | +0.193 |
robe | +0.190 |
Neg logit contributions | |
rep | -0.126 |
sne | -0.121 |
overcome | -0.119 |
cal | -0.119 |
hear | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
hon | +0.003 |
notice | +0.002 |
rep | +0.002 |
hon | +0.002 |
cal | +0.002 |
Neg logit contributions | |
possession | -0.004 |
favourite | -0.004 |
favorite | -0.004 |
purse | -0.004 |
wrist | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
possession | +0.100 |
favourite | +0.088 |
purse | +0.083 |
possessions | +0.082 |
wallet | +0.081 |
Neg logit contributions | |
hon | -0.064 |
rep | -0.057 |
notice | -0.056 |
hear | -0.052 |
oblig | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-04 |
Pos logit contributions | |
favourite | +0.083 |
favorite | +0.080 |
possession | +0.077 |
brand | +0.064 |
purse | +0.064 |
Neg logit contributions | |
hon | -0.100 |
oblig | -0.092 |
Jim | -0.092 |
notice | -0.091 |
rep | -0.086 |
Token | L |
Token ID | 43 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
possession | +0.000 |
favourite | +0.000 |
favorite | +0.000 |
purse | +0.000 |
wrist | +0.000 |
Neg logit contributions | |
hon | -0.000 |
cal | -0.000 |
notice | -0.000 |
rep | -0.000 |
hon | -0.000 |
Token | ily |
Token ID | 813 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
hon | +0.088 |
notice | +0.079 |
hon | +0.079 |
rep | +0.079 |
announcements | +0.078 |
Neg logit contributions | |
possession | -0.078 |
favourite | -0.071 |
favorite | -0.065 |
purse | -0.065 |
wrist | -0.061 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
favourite | +0.040 |
favorite | +0.037 |
possession | +0.036 |
purse | +0.034 |
wrist | +0.033 |
Neg logit contributions | |
hon | -0.055 |
announcements | -0.048 |
boom | -0.048 |
explode | -0.047 |
oblig | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | -8.02e-02 |
Pos logit contributions | |
possession | +0.298 |
favourite | +0.285 |
favorite | +0.260 |
purse | +0.256 |
wrist | +0.254 |
Neg logit contributions | |
hon | -0.211 |
cal | -0.180 |
notice | -0.179 |
louder | -0.176 |
hon | -0.175 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
possession | +0.122 |
favourite | +0.117 |
favorite | +0.106 |
purse | +0.103 |
wrist | +0.096 |
Neg logit contributions | |
hon | -0.160 |
notice | -0.141 |
rep | -0.138 |
cal | -0.138 |
hon | -0.137 |
Token | to |
Token ID | 284 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
favourite | +0.397 |
possession | +0.397 |
favourites | +0.381 |
favorites | +0.373 |
artments | +0.360 |
Neg logit contributions | |
hon | -0.265 |
hon | -0.259 |
louder | -0.248 |
hear | -0.232 |
boom | -0.228 |
Token | her |
Token ID | 607 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
possession | +0.022 |
favourite | +0.021 |
favorite | +0.020 |
purse | +0.019 |
wrist | +0.018 |
Neg logit contributions | |
hon | -0.021 |
notice | -0.018 |
rep | -0.017 |
cal | -0.017 |
hon | -0.017 |
Token | friend |
Token ID | 1545 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
possession | +0.223 |
attered | +0.206 |
favourite | +0.204 |
artments | +0.183 |
shirt | +0.182 |
Neg logit contributions | |
hon | -0.183 |
rep | -0.177 |
hon | -0.175 |
notice | -0.171 |
overcome | -0.170 |
Token | , |
Token ID | 11 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
favourite | +0.146 |
possession | +0.143 |
favorite | +0.137 |
purse | +0.133 |
wallet | +0.124 |
Neg logit contributions | |
hon | -0.118 |
announcements | -0.096 |
notice | -0.095 |
RO | -0.093 |
hon | -0.092 |
Token | Tim |
Token ID | 5045 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
possession | +0.077 |
favourite | +0.075 |
favorite | +0.070 |
purse | +0.068 |
wrist | +0.064 |
Neg logit contributions | |
hon | -0.075 |
notice | -0.063 |
announcements | -0.063 |
oblig | -0.062 |
cal | -0.062 |
Token | Tim |
Token ID | 5045 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
possession | +0.040 |
favourite | +0.039 |
favorite | +0.037 |
purse | +0.035 |
wallet | +0.032 |
Neg logit contributions | |
hon | -0.031 |
cal | -0.026 |
announcements | -0.026 |
oblig | -0.025 |
notice | -0.025 |
Token | wanted |
Token ID | 2227 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
hon | +0.080 |
oblig | +0.073 |
announcements | +0.073 |
cal | +0.071 |
rep | +0.071 |
Neg logit contributions | |
possession | -0.066 |
favourite | -0.064 |
favorite | -0.062 |
purse | -0.058 |
wallet | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
possession | +0.145 |
favourite | +0.138 |
attered | +0.129 |
wrist | +0.127 |
purse | +0.125 |
Neg logit contributions | |
hon | -0.097 |
hon | -0.089 |
rep | -0.083 |
thunder | -0.083 |
overcome | -0.083 |
Token | show |
Token ID | 905 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
hon | +0.033 |
notice | +0.027 |
rep | +0.027 |
cal | +0.027 |
hon | +0.027 |
Neg logit contributions | |
possession | -0.041 |
favourite | -0.040 |
favorite | -0.037 |
purse | -0.036 |
wrist | -0.034 |
Token | his |
Token ID | 465 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
possession | +0.142 |
favourite | +0.138 |
favorite | +0.128 |
attered | +0.126 |
purse | +0.125 |
Neg logit contributions | |
hon | -0.068 |
hon | -0.056 |
notice | -0.056 |
louder | -0.054 |
cal | -0.054 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
possession | +0.195 |
attered | +0.176 |
ase | +0.174 |
artments | +0.169 |
ocket | +0.169 |
Neg logit contributions | |
hon | -0.189 |
hon | -0.182 |
rep | -0.171 |
sne | -0.169 |
thunder | -0.165 |
Token | to |
Token ID | 284 |
Feature activation | +5.44e-03 |
Pos logit contributions | |
possession | +0.370 |
favourite | +0.339 |
wrist | +0.319 |
favourites | +0.311 |
attered | +0.308 |
Neg logit contributions | |
hon | -0.337 |
hon | -0.327 |
louder | -0.296 |
boom | -0.292 |
sne | -0.282 |
Token | his |
Token ID | 465 |
Feature activation | -9.14e-02 |
Pos logit contributions | |
hon | +0.006 |
notice | +0.005 |
hon | +0.005 |
rep | +0.005 |
overcome | +0.005 |
Neg logit contributions | |
possession | -0.013 |
favourite | -0.013 |
favorite | -0.012 |
purse | -0.011 |
attered | -0.011 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.77e-03 |
Pos logit contributions | |
possession | +0.176 |
attered | +0.162 |
shirt | +0.153 |
artments | +0.152 |
favourite | +0.151 |
Neg logit contributions | |
hon | -0.119 |
hon | -0.117 |
rep | -0.112 |
sne | -0.109 |
cal | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
hon | +0.004 |
notice | +0.003 |
rep | +0.003 |
hon | +0.003 |
cal | +0.003 |
Neg logit contributions | |
possession | -0.006 |
favourite | -0.006 |
favorite | -0.006 |
purse | -0.005 |
wrist | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
possession | +0.070 |
favourite | +0.065 |
purse | +0.060 |
favorite | +0.059 |
wrist | +0.058 |
Neg logit contributions | |
hon | -0.046 |
rep | -0.038 |
hon | -0.038 |
notice | -0.037 |
overcome | -0.036 |
Token | just |
Token ID | 655 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
favourite | +0.122 |
favorite | +0.116 |
possession | +0.116 |
purse | +0.098 |
wallet | +0.097 |
Neg logit contributions | |
hon | -0.136 |
cal | -0.118 |
oblig | -0.116 |
Run | -0.116 |
notice | -0.115 |
Token | buried |
Token ID | 11694 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
hon | +0.037 |
notice | +0.032 |
hon | +0.032 |
rep | +0.032 |
cal | +0.031 |
Neg logit contributions | |
possession | -0.054 |
favourite | -0.053 |
favorite | -0.048 |
purse | -0.048 |
wrist | -0.047 |
Token | his |
Token ID | 465 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
possession | +0.125 |
favourite | +0.120 |
purse | +0.111 |
wrist | +0.110 |
favorite | +0.109 |
Neg logit contributions | |
hon | -0.072 |
hon | -0.062 |
louder | -0.061 |
notice | -0.061 |
cal | -0.061 |
Token | most |
Token ID | 749 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
possession | +0.120 |
favourite | +0.103 |
attered | +0.097 |
wrist | +0.094 |
purse | +0.093 |
Neg logit contributions | |
hon | -0.158 |
hon | -0.150 |
sne | -0.149 |
rep | -0.147 |
notice | -0.145 |
Token | prized |
Token ID | 37768 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
possession | +0.035 |
favourite | +0.033 |
favorite | +0.030 |
purse | +0.030 |
wrist | +0.029 |
Neg logit contributions | |
hon | -0.038 |
cal | -0.034 |
notice | -0.034 |
hon | -0.034 |
overcome | -0.033 |
Token | possession |
Token ID | 7797 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
favourite | +0.075 |
attered | +0.071 |
wrist | +0.065 |
brand | +0.063 |
umes | +0.062 |
Neg logit contributions | |
hon | -0.175 |
hon | -0.162 |
notice | -0.152 |
sne | -0.150 |
cal | -0.150 |
Token | a |
Token ID | 257 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
possession | +0.526 |
favourite | +0.525 |
attered | +0.495 |
purse | +0.493 |
favorite | +0.491 |
Neg logit contributions | |
hon | -0.117 |
hon | -0.098 |
rep | -0.096 |
overcome | -0.086 |
notice | -0.086 |
Token | few |
Token ID | 1178 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
hon | +0.032 |
hon | +0.029 |
notice | +0.029 |
rep | +0.029 |
cal | +0.028 |
Neg logit contributions | |
possession | -0.029 |
favourite | -0.027 |
favorite | -0.024 |
attered | -0.024 |
purse | -0.024 |
Token | inches |
Token ID | 8331 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
possession | +0.071 |
favourite | +0.070 |
favorite | +0.063 |
purse | +0.062 |
wrist | +0.062 |
Neg logit contributions | |
hon | -0.063 |
notice | -0.056 |
hon | -0.055 |
cal | -0.055 |
rep | -0.054 |
Token | underground |
Token ID | 11447 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
hon | +0.116 |
notice | +0.098 |
hon | +0.097 |
rep | +0.094 |
cal | +0.093 |
Neg logit contributions | |
possession | -0.199 |
favourite | -0.194 |
favorite | -0.179 |
purse | -0.176 |
wrist | -0.173 |
Token | . |
Token ID | 13 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
possession | +0.128 |
favourite | +0.123 |
favorite | +0.115 |
purse | +0.110 |
wrist | +0.106 |
Neg logit contributions | |
hon | -0.085 |
notice | -0.070 |
rep | -0.069 |
cal | -0.068 |
hon | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | +9.18e-03 |
Pos logit contributions | |
hon | +0.112 |
rep | +0.095 |
announcements | +0.095 |
notice | +0.094 |
cal | +0.093 |
Neg logit contributions | |
possession | -0.135 |
favourite | -0.124 |
purse | -0.118 |
favorite | -0.116 |
wrist | -0.111 |
Token | little |
Token ID | 1310 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
hon | +0.017 |
notice | +0.015 |
oblig | +0.014 |
rep | +0.014 |
cal | +0.014 |
Neg logit contributions | |
possession | -0.016 |
favourite | -0.015 |
favorite | -0.014 |
purse | -0.014 |
wallet | -0.013 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
hon | +0.045 |
oblig | +0.041 |
overcome | +0.041 |
notice | +0.040 |
announcements | +0.040 |
Neg logit contributions | |
possession | -0.038 |
favourite | -0.034 |
purse | -0.033 |
favorite | -0.033 |
wallet | -0.029 |
Token | find |
Token ID | 1064 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.022 |
favorite | +0.021 |
purse | +0.020 |
attered | +0.020 |
Neg logit contributions | |
hon | -0.016 |
notice | -0.014 |
overcome | -0.014 |
cal | -0.013 |
rep | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
possession | +0.285 |
favourite | +0.284 |
favorite | +0.266 |
attered | +0.264 |
wrist | +0.262 |
Neg logit contributions | |
hon | -0.133 |
cal | -0.125 |
notice | -0.118 |
louder | -0.116 |
overcome | -0.110 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
attered | +0.164 |
ase | +0.157 |
possession | +0.147 |
umes | +0.144 |
shirt | +0.140 |
Neg logit contributions | |
sne | -0.191 |
mad | -0.188 |
rep | -0.187 |
fall | -0.185 |
time | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
possession | +0.482 |
favourite | +0.482 |
favorites | +0.448 |
favourites | +0.448 |
attered | +0.438 |
Neg logit contributions | |
hon | -0.169 |
hon | -0.156 |
overcome | -0.144 |
louder | -0.142 |
rep | -0.137 |
Token | After |
Token ID | 2293 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
possession | +0.047 |
favourite | +0.044 |
favorite | +0.041 |
purse | +0.038 |
wrist | +0.036 |
Neg logit contributions | |
hon | -0.052 |
cal | -0.046 |
notice | -0.046 |
rep | -0.045 |
hon | -0.045 |
Token | searching |
Token ID | 10342 |
Feature activation | -4.19e-03 |
Pos logit contributions | |
possession | +0.031 |
favourite | +0.030 |
favorite | +0.028 |
purse | +0.027 |
wrist | +0.025 |
Neg logit contributions | |
hon | -0.029 |
notice | -0.025 |
cal | -0.024 |
hon | -0.024 |
rep | -0.024 |
Token | for |
Token ID | 329 |
Feature activation | -9.52e-03 |
Pos logit contributions | |
possession | +0.008 |
favourite | +0.008 |
wrist | +0.008 |
favorite | +0.007 |
attered | +0.007 |
Neg logit contributions | |
hon | -0.006 |
notice | -0.006 |
cal | -0.005 |
louder | -0.005 |
overcome | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
possession | +0.023 |
favourite | +0.023 |
attered | +0.022 |
favorite | +0.021 |
wrist | +0.021 |
Neg logit contributions | |
notice | -0.008 |
hon | -0.008 |
cal | -0.008 |
louder | -0.007 |
overcome | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
possession | +0.034 |
favourite | +0.032 |
favorite | +0.030 |
purse | +0.029 |
wrist | +0.028 |
Neg logit contributions | |
hon | -0.026 |
notice | -0.022 |
rep | -0.022 |
hon | -0.021 |
cal | -0.021 |
Token | Tom |
Token ID | 4186 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
possession | +0.088 |
favourite | +0.085 |
favorite | +0.080 |
purse | +0.078 |
wrist | +0.075 |
Neg logit contributions | |
hon | -0.046 |
notice | -0.037 |
rep | -0.036 |
hon | -0.036 |
cal | -0.035 |
Token | shows |
Token ID | 2523 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
hon | +0.134 |
notice | +0.116 |
rep | +0.115 |
cal | +0.112 |
hon | +0.112 |
Neg logit contributions | |
possession | -0.137 |
favourite | -0.133 |
favorite | -0.123 |
purse | -0.117 |
wrist | -0.112 |
Token | Mia |
Token ID | 32189 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
possession | +0.229 |
favourite | +0.222 |
favorite | +0.211 |
purse | +0.205 |
wrist | +0.193 |
Neg logit contributions | |
hon | -0.144 |
rep | -0.118 |
notice | -0.116 |
overcome | -0.114 |
cal | -0.113 |
Token | his |
Token ID | 465 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
possession | +0.047 |
favourite | +0.045 |
favorite | +0.042 |
purse | +0.041 |
wrist | +0.039 |
Neg logit contributions | |
hon | -0.030 |
notice | -0.025 |
hon | -0.024 |
rep | -0.024 |
cal | -0.024 |
Token | model |
Token ID | 2746 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
possession | +0.192 |
favourite | +0.168 |
attered | +0.150 |
favorite | +0.142 |
wrist | +0.141 |
Neg logit contributions | |
hon | -0.288 |
hon | -0.268 |
rep | -0.265 |
thunder | -0.257 |
notice | -0.257 |
Token | car |
Token ID | 1097 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
possession | +0.345 |
favourite | +0.329 |
favorite | +0.283 |
favourites | +0.282 |
attered | +0.277 |
Neg logit contributions | |
hon | -0.355 |
hon | -0.331 |
rep | -0.311 |
hear | -0.308 |
boom | -0.307 |
Token | . |
Token ID | 13 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
possession | +0.254 |
favourite | +0.243 |
favorite | +0.222 |
purse | +0.217 |
wrist | +0.211 |
Neg logit contributions | |
hon | -0.166 |
hon | -0.143 |
rep | -0.138 |
notice | -0.138 |
cal | -0.134 |
Token | He |
Token ID | 679 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
possession | +0.086 |
favourite | +0.082 |
favorite | +0.076 |
purse | +0.074 |
wrist | +0.071 |
Neg logit contributions | |
hon | -0.064 |
notice | -0.054 |
rep | -0.053 |
hon | -0.053 |
cal | -0.052 |
Token | says |
Token ID | 1139 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
possession | +0.055 |
favourite | +0.052 |
favorite | +0.048 |
purse | +0.047 |
wrist | +0.046 |
Neg logit contributions | |
hon | -0.036 |
notice | -0.030 |
hon | -0.030 |
rep | -0.030 |
cal | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | -5.52e-02 |
Pos logit contributions | |
favourite | +0.018 |
possession | +0.018 |
favorite | +0.018 |
purse | +0.017 |
wallet | +0.016 |
Neg logit contributions | |
hon | -0.017 |
notice | -0.015 |
overcome | -0.015 |
cal | -0.015 |
Run | -0.015 |
Token | ily |
Token ID | 813 |
Feature activation | -7.21e-03 |
Pos logit contributions | |
hon | +0.131 |
notice | +0.122 |
announcements | +0.122 |
cal | +0.122 |
hon | +0.120 |
Neg logit contributions | |
possession | -0.128 |
favourite | -0.114 |
favorite | -0.105 |
purse | -0.104 |
robe | -0.100 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
favourite | +0.013 |
possession | +0.013 |
favorite | +0.012 |
purse | +0.011 |
wrist | +0.011 |
Neg logit contributions | |
hon | -0.012 |
notice | -0.010 |
boom | -0.010 |
oblig | -0.010 |
rep | -0.010 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
possession | +0.316 |
favourite | +0.308 |
favorite | +0.291 |
purse | +0.282 |
wrist | +0.269 |
Neg logit contributions | |
hon | -0.136 |
notice | -0.109 |
rep | -0.107 |
cal | -0.103 |
hon | -0.102 |
Token | her |
Token ID | 607 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
possession | +0.047 |
favourite | +0.044 |
favorite | +0.040 |
purse | +0.040 |
attered | +0.039 |
Neg logit contributions | |
hon | -0.035 |
cal | -0.031 |
hon | -0.030 |
notice | -0.030 |
overcome | -0.030 |
Token | favorite |
Token ID | 4004 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
possession | +0.245 |
favourite | +0.226 |
attered | +0.218 |
favorite | +0.196 |
purse | +0.195 |
Neg logit contributions | |
hon | -0.287 |
hon | -0.269 |
rep | -0.264 |
cal | -0.261 |
notice | -0.259 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
attered | +0.305 |
possession | +0.264 |
umes | +0.264 |
belt | +0.257 |
artments | +0.255 |
Neg logit contributions | |
hon | -0.267 |
roar | -0.253 |
sne | -0.250 |
thunder | -0.244 |
laugh | -0.243 |
Token | , |
Token ID | 11 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
possession | +0.501 |
favourite | +0.468 |
attered | +0.452 |
favourites | +0.447 |
favorites | +0.444 |
Neg logit contributions | |
hon | -0.177 |
hon | -0.168 |
sne | -0.144 |
rep | -0.139 |
louder | -0.136 |
Token | a |
Token ID | 257 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
possession | +0.098 |
favourite | +0.093 |
favorite | +0.085 |
purse | +0.083 |
attered | +0.080 |
Neg logit contributions | |
hon | -0.085 |
hon | -0.074 |
rep | -0.073 |
notice | -0.073 |
cal | -0.072 |
Token | magnet |
Token ID | 19972 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
hon | +0.040 |
hon | +0.038 |
rep | +0.037 |
louder | +0.036 |
notice | +0.036 |
Neg logit contributions | |
possession | -0.036 |
favourite | -0.033 |
attered | -0.031 |
favorite | -0.029 |
favorites | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
possession | +0.148 |
favourite | +0.144 |
favorite | +0.131 |
favorites | +0.130 |
favourites | +0.129 |
Neg logit contributions | |
hon | -0.058 |
hon | -0.051 |
rep | -0.045 |
sne | -0.042 |
slit | -0.041 |
Token | " |
Token ID | 366 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
possession | +0.062 |
favourite | +0.058 |
favorite | +0.054 |
purse | +0.052 |
attered | +0.050 |
Neg logit contributions | |
hon | -0.044 |
rep | -0.037 |
notice | -0.037 |
hon | -0.036 |
louder | -0.036 |
Token | It |
Token ID | 632 |
Feature activation | -9.98e-02 |
Pos logit contributions | |
possession | +0.088 |
favourite | +0.084 |
favorite | +0.077 |
purse | +0.075 |
wrist | +0.072 |
Neg logit contributions | |
hon | -0.082 |
notice | -0.071 |
rep | -0.070 |
hon | -0.070 |
cal | -0.069 |
Token | was |
Token ID | 373 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
possession | +0.230 |
favourite | +0.214 |
wrist | +0.212 |
purse | +0.209 |
attered | +0.204 |
Neg logit contributions | |
hon | -0.114 |
hon | -0.105 |
roared | -0.103 |
slit | -0.102 |
ne | -0.100 |
Token | his |
Token ID | 465 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
possession | +0.060 |
wrist | +0.054 |
favourite | +0.054 |
attered | +0.053 |
purse | +0.051 |
Neg logit contributions | |
hon | -0.030 |
louder | -0.027 |
rep | -0.027 |
hon | -0.027 |
sne | -0.026 |
Token | most |
Token ID | 749 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
attered | +0.089 |
possession | +0.089 |
shirt | +0.078 |
rolley | +0.076 |
ase | +0.074 |
Neg logit contributions | |
hon | -0.105 |
rep | -0.105 |
sne | -0.100 |
hon | -0.099 |
roar | -0.098 |
Token | prized |
Token ID | 37768 |
Feature activation | -7.69e-02 |
Pos logit contributions | |
possession | +0.020 |
robe | +0.019 |
attered | +0.019 |
artments | +0.019 |
favourite | +0.018 |
Neg logit contributions | |
hon | -0.015 |
overcome | -0.014 |
cal | -0.014 |
bra | -0.014 |
hon | -0.013 |
Token | possession |
Token ID | 7797 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
favourite | +0.052 |
attered | +0.042 |
favourites | +0.042 |
wrist | +0.040 |
robe | +0.039 |
Neg logit contributions | |
hon | -0.178 |
hon | -0.162 |
," | -0.159 |
cal | -0.156 |
notice | -0.154 |
Token | , |
Token ID | 11 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
favourite | +0.530 |
possession | +0.517 |
attered | +0.509 |
favorites | +0.500 |
favorite | +0.491 |
Neg logit contributions | |
hon | -0.094 |
hon | -0.094 |
rep | -0.074 |
," | -0.071 |
sne | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | -8.85e-03 |
Pos logit contributions | |
possession | +0.059 |
favourite | +0.057 |
purse | +0.052 |
attered | +0.052 |
favorite | +0.052 |
Neg logit contributions | |
hon | -0.034 |
hon | -0.030 |
rep | -0.028 |
notice | -0.028 |
thunder | -0.027 |
Token | he |
Token ID | 339 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
possession | +0.021 |
favourite | +0.020 |
attered | +0.019 |
purse | +0.019 |
wrist | +0.018 |
Neg logit contributions | |
hon | -0.008 |
hon | -0.008 |
rep | -0.007 |
notice | -0.007 |
overcome | -0.007 |
Token | never |
Token ID | 1239 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
hon | +0.044 |
hon | +0.042 |
overcome | +0.040 |
sne | +0.039 |
roared | +0.038 |
Neg logit contributions | |
possession | -0.056 |
favourite | -0.051 |
purse | -0.050 |
wrist | -0.047 |
brand | -0.046 |
Token | let |
Token ID | 1309 |
Feature activation | +8.77e-02 |
Pos logit contributions | |
hon | +0.137 |
Jim | +0.117 |
announcements | +0.115 |
boom | +0.115 |
Run | +0.115 |
Neg logit contributions | |
possession | -0.113 |
favourite | -0.102 |
favorite | -0.100 |
purse | -0.093 |
wallet | -0.089 |
Token | so |
Token ID | 523 |
Feature activation | +3.29e-04 |
Pos logit contributions | |
possession | +0.061 |
favourite | +0.060 |
favorite | +0.055 |
purse | +0.052 |
wrist | +0.049 |
Neg logit contributions | |
hon | -0.063 |
notice | -0.054 |
rep | -0.054 |
cal | -0.053 |
hon | -0.052 |
Token | excited |
Token ID | 6568 |
Feature activation | +7.26e-03 |
Pos logit contributions | |
hon | +0.001 |
cal | +0.001 |
rep | +0.001 |
oblig | +0.001 |
notice | +0.001 |
Neg logit contributions | |
favourite | -0.001 |
possession | -0.001 |
favorite | -0.000 |
purse | -0.000 |
attered | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
hon | +0.016 |
cal | +0.014 |
oblig | +0.014 |
notice | +0.014 |
announcements | +0.014 |
Neg logit contributions | |
possession | -0.010 |
favourite | -0.010 |
favorite | -0.009 |
purse | -0.008 |
wallet | -0.007 |
Token | show |
Token ID | 905 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
hon | +0.053 |
cal | +0.044 |
rep | +0.042 |
louder | +0.041 |
slit | +0.040 |
Neg logit contributions | |
possession | -0.092 |
favourite | -0.091 |
favorite | -0.085 |
purse | -0.081 |
wrist | -0.081 |
Token | his |
Token ID | 465 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
possession | +0.118 |
favourite | +0.114 |
favorite | +0.107 |
purse | +0.103 |
wrist | +0.098 |
Neg logit contributions | |
hon | -0.081 |
notice | -0.067 |
rep | -0.067 |
cal | -0.065 |
overcome | -0.064 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
possession | +0.200 |
favourite | +0.176 |
attered | +0.165 |
wrist | +0.164 |
purse | +0.164 |
Neg logit contributions | |
hon | -0.192 |
hon | -0.183 |
thunder | -0.173 |
boom | -0.173 |
notice | -0.172 |
Token | trains |
Token ID | 13404 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
possession | +0.396 |
favourite | +0.371 |
wrist | +0.342 |
purse | +0.337 |
favorite | +0.328 |
Neg logit contributions | |
hon | -0.324 |
hon | -0.296 |
boom | -0.282 |
louder | -0.280 |
hear | -0.272 |
Token | to |
Token ID | 284 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
possession | +0.112 |
favourite | +0.108 |
favorite | +0.101 |
purse | +0.095 |
wrist | +0.090 |
Neg logit contributions | |
hon | -0.119 |
notice | -0.101 |
cal | -0.101 |
rep | -0.101 |
oblig | -0.099 |
Token | everyone |
Token ID | 2506 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
hon | +0.039 |
notice | +0.032 |
rep | +0.032 |
cal | +0.032 |
hon | +0.031 |
Neg logit contributions | |
possession | -0.058 |
favourite | -0.056 |
favorite | -0.052 |
purse | -0.050 |
wrist | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
hon | +0.047 |
notice | +0.040 |
cal | +0.039 |
rep | +0.039 |
announcements | +0.039 |
Neg logit contributions | |
possession | -0.061 |
favourite | -0.059 |
favorite | -0.055 |
purse | -0.052 |
wrist | -0.050 |
Token | But |
Token ID | 887 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
possession | +0.050 |
favourite | +0.043 |
purse | +0.043 |
ocket | +0.043 |
wrist | +0.041 |
Neg logit contributions | |
hon | -0.024 |
hon | -0.022 |
rep | -0.020 |
notice | -0.020 |
louder | -0.019 |
Token | nice |
Token ID | 3621 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
favourite | +0.042 |
possession | +0.041 |
favorite | +0.040 |
purse | +0.036 |
wrist | +0.035 |
Neg logit contributions | |
hon | -0.041 |
notice | -0.035 |
rep | -0.034 |
slit | -0.034 |
cal | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
possession | +0.041 |
favourite | +0.039 |
favorite | +0.037 |
purse | +0.036 |
wrist | +0.034 |
Neg logit contributions | |
hon | -0.029 |
notice | -0.024 |
rep | -0.024 |
cal | -0.023 |
oblig | -0.023 |
Token | He |
Token ID | 679 |
Feature activation | -9.34e-03 |
Pos logit contributions | |
possession | +0.037 |
favourite | +0.037 |
favorite | +0.034 |
purse | +0.032 |
wrist | +0.030 |
Neg logit contributions | |
hon | -0.039 |
notice | -0.034 |
overcome | -0.034 |
rep | -0.034 |
cal | -0.033 |
Token | offers |
Token ID | 4394 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
favourite | +0.020 |
possession | +0.020 |
favorite | +0.019 |
purse | +0.018 |
favorites | +0.017 |
Neg logit contributions | |
hon | -0.012 |
notice | -0.010 |
rep | -0.010 |
oblig | -0.010 |
overcome | -0.009 |
Token | his |
Token ID | 465 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
possession | +0.177 |
favourite | +0.171 |
favorite | +0.160 |
purse | +0.155 |
wrist | +0.150 |
Neg logit contributions | |
hon | -0.106 |
notice | -0.088 |
rep | -0.086 |
hon | -0.085 |
cal | -0.085 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
possession | +0.180 |
favourite | +0.162 |
attered | +0.150 |
purse | +0.141 |
wrist | +0.140 |
Neg logit contributions | |
hon | -0.202 |
hon | -0.190 |
rep | -0.177 |
boom | -0.177 |
notice | -0.176 |
Token | train |
Token ID | 4512 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
possession | +0.399 |
favourite | +0.368 |
wrist | +0.344 |
favourites | +0.336 |
attered | +0.326 |
Neg logit contributions | |
hon | -0.312 |
hon | -0.299 |
louder | -0.279 |
boom | -0.272 |
hear | -0.258 |
Token | to |
Token ID | 284 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
possession | +0.207 |
favourite | +0.199 |
favorite | +0.183 |
purse | +0.178 |
wrist | +0.171 |
Neg logit contributions | |
hon | -0.162 |
notice | -0.137 |
rep | -0.135 |
hon | -0.135 |
cal | -0.134 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
hon | +0.029 |
cal | +0.024 |
rep | +0.023 |
announcements | +0.023 |
overcome | +0.023 |
Neg logit contributions | |
possession | -0.032 |
favourite | -0.031 |
favorite | -0.030 |
purse | -0.028 |
wrist | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
possession | +0.061 |
favourite | +0.059 |
favorite | +0.056 |
purse | +0.054 |
wrist | +0.052 |
Neg logit contributions | |
hon | -0.037 |
notice | -0.031 |
rep | -0.030 |
hon | -0.030 |
cal | -0.030 |
Token | Sam |
Token ID | 3409 |
Feature activation | +5.49e-02 |
Pos logit contributions | |
possession | +0.058 |
favourite | +0.058 |
favorite | +0.056 |
purse | +0.053 |
wrist | +0.050 |
Neg logit contributions | |
hon | -0.037 |
announcements | -0.029 |
notice | -0.028 |
rr | -0.028 |
boom | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.023 |
favorite | +0.021 |
purse | +0.020 |
attered | +0.020 |
Neg logit contributions | |
hon | -0.017 |
rep | -0.014 |
notice | -0.014 |
cal | -0.013 |
hon | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
hon | +0.057 |
notice | +0.048 |
hon | +0.047 |
rep | +0.047 |
overcome | +0.046 |
Neg logit contributions | |
possession | -0.078 |
favourite | -0.075 |
favorite | -0.070 |
purse | -0.068 |
attered | -0.065 |
Token | She |
Token ID | 3347 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
hon | +0.051 |
notice | +0.044 |
rep | +0.043 |
hon | +0.043 |
cal | +0.043 |
Neg logit contributions | |
possession | -0.065 |
favourite | -0.063 |
favorite | -0.058 |
purse | -0.056 |
wrist | -0.054 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
possession | +0.077 |
favourite | +0.077 |
favorite | +0.072 |
purse | +0.068 |
wrist | +0.066 |
Neg logit contributions | |
hon | -0.062 |
notice | -0.053 |
rep | -0.052 |
announcements | -0.051 |
cal | -0.051 |
Token | her |
Token ID | 607 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
possession | +0.222 |
favourite | +0.213 |
favorite | +0.195 |
purse | +0.193 |
attered | +0.192 |
Neg logit contributions | |
hon | -0.141 |
louder | -0.117 |
hon | -0.117 |
notice | -0.116 |
cal | -0.116 |
Token | model |
Token ID | 2746 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
possession | +0.229 |
attered | +0.216 |
favourite | +0.207 |
shirt | +0.192 |
artments | +0.190 |
Neg logit contributions | |
hon | -0.233 |
hon | -0.230 |
rep | -0.218 |
boom | -0.215 |
thunder | -0.214 |
Token | to |
Token ID | 284 |
Feature activation | +3.29e-03 |
Pos logit contributions | |
possession | +0.363 |
favourite | +0.347 |
favourites | +0.308 |
attered | +0.305 |
favorite | +0.304 |
Neg logit contributions | |
hon | -0.336 |
hon | -0.305 |
louder | -0.290 |
boom | -0.288 |
hear | -0.287 |
Token | her |
Token ID | 607 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
hon | +0.005 |
notice | +0.005 |
rep | +0.004 |
hon | +0.004 |
cal | +0.004 |
Neg logit contributions | |
possession | -0.006 |
favourite | -0.006 |
favorite | -0.006 |
purse | -0.005 |
wrist | -0.005 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
possession | +0.179 |
favourite | +0.162 |
attered | +0.155 |
shirt | +0.146 |
purse | +0.144 |
Neg logit contributions | |
hon | -0.165 |
hon | -0.157 |
rep | -0.154 |
notice | -0.153 |
cal | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
possession | +0.065 |
favourite | +0.063 |
favorite | +0.059 |
purse | +0.057 |
wrist | +0.055 |
Neg logit contributions | |
hon | -0.058 |
notice | -0.049 |
rep | -0.048 |
hon | -0.048 |
cal | -0.048 |
Token | " |
Token ID | 366 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
possession | +0.041 |
favourite | +0.037 |
possessions | +0.033 |
favorite | +0.032 |
attered | +0.032 |
Neg logit contributions | |
hon | -0.037 |
rep | -0.035 |
notice | -0.034 |
hear | -0.033 |
thunder | -0.033 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
possession | +0.033 |
favourite | +0.031 |
favorite | +0.029 |
purse | +0.029 |
brand | +0.027 |
Neg logit contributions | |
hon | -0.031 |
oblig | -0.026 |
notice | -0.025 |
announcements | -0.025 |
cal | -0.025 |
Token | my |
Token ID | 1820 |
Feature activation | +5.89e-03 |
Pos logit contributions | |
hon | +0.117 |
announcements | +0.104 |
Run | +0.104 |
RO | +0.104 |
oblig | +0.102 |
Neg logit contributions | |
possession | -0.071 |
favorite | -0.062 |
favourite | -0.061 |
purse | -0.056 |
wallet | -0.052 |
Token | gave |
Token ID | 2921 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
hon | +0.010 |
announcements | +0.009 |
boom | +0.009 |
rr | +0.009 |
RO | +0.009 |
Neg logit contributions | |
possession | -0.010 |
favourite | -0.010 |
favorite | -0.009 |
purse | -0.009 |
wallet | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
possession | +0.131 |
favourite | +0.127 |
favorite | +0.117 |
purse | +0.114 |
wrist | +0.112 |
Neg logit contributions | |
hon | -0.083 |
notice | -0.071 |
cal | -0.069 |
rep | -0.068 |
hon | -0.068 |
Token | his |
Token ID | 465 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
possession | +0.231 |
favourite | +0.219 |
wrist | +0.196 |
favorite | +0.194 |
purse | +0.192 |
Neg logit contributions | |
hon | -0.198 |
notice | -0.176 |
cal | -0.174 |
hon | -0.172 |
louder | -0.170 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
possession | +0.176 |
favourite | +0.157 |
wrist | +0.143 |
attered | +0.141 |
purse | +0.140 |
Neg logit contributions | |
hon | -0.211 |
hon | -0.195 |
notice | -0.194 |
rep | -0.192 |
cal | -0.189 |
Token | makeup |
Token ID | 16029 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
possession | +0.354 |
favourite | +0.330 |
wrist | +0.302 |
purse | +0.287 |
favourites | +0.284 |
Neg logit contributions | |
hon | -0.337 |
hon | -0.306 |
louder | -0.301 |
boom | -0.299 |
notice | -0.290 |
Token | set |
Token ID | 900 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
possession | +0.238 |
favourite | +0.226 |
favorite | +0.204 |
purse | +0.201 |
wrist | +0.198 |
Neg logit contributions | |
hon | -0.141 |
notice | -0.120 |
rep | -0.120 |
hon | -0.119 |
cal | -0.118 |
Token | to |
Token ID | 284 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
hon | +0.108 |
announcements | +0.093 |
rep | +0.093 |
notice | +0.093 |
RO | +0.092 |
Neg logit contributions | |
possession | -0.104 |
favourite | -0.098 |
favorite | -0.091 |
purse | -0.088 |
wrist | -0.082 |
Token | play |
Token ID | 711 |
Feature activation | -8.31e-03 |
Pos logit contributions | |
hon | +0.084 |
Run | +0.070 |
announcements | +0.068 |
RO | +0.067 |
Jim | +0.066 |
Neg logit contributions | |
possession | -0.071 |
favourite | -0.070 |
favorite | -0.066 |
brand | -0.063 |
purse | -0.062 |
Token | with |
Token ID | 351 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
possession | +0.015 |
favourite | +0.014 |
favorite | +0.013 |
purse | +0.013 |
attered | +0.012 |
Neg logit contributions | |
hon | -0.014 |
notice | -0.012 |
louder | -0.012 |
hon | -0.012 |
rep | -0.012 |
Token | So |
Token ID | 1406 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
possession | +0.039 |
favourite | +0.035 |
wrist | +0.032 |
purse | +0.032 |
favorite | +0.032 |
Neg logit contributions | |
hon | -0.021 |
notice | -0.019 |
rep | -0.018 |
louder | -0.018 |
cal | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
hon | +0.094 |
announcements | +0.073 |
hon | +0.073 |
oblig | +0.073 |
cal | +0.072 |
Neg logit contributions | |
possession | -0.117 |
favourite | -0.117 |
favorite | -0.110 |
purse | -0.106 |
wrist | -0.097 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
possession | +0.022 |
favourite | +0.021 |
favorite | +0.020 |
purse | +0.019 |
wrist | +0.019 |
Neg logit contributions | |
hon | -0.015 |
notice | -0.012 |
rep | -0.012 |
hon | -0.012 |
cal | -0.012 |
Token | sold |
Token ID | 2702 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
hon | +0.034 |
notice | +0.029 |
rep | +0.029 |
cal | +0.028 |
hon | +0.028 |
Neg logit contributions | |
possession | -0.043 |
favourite | -0.041 |
favorite | -0.038 |
purse | -0.037 |
wrist | -0.036 |
Token | her |
Token ID | 607 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
possession | +0.155 |
favourite | +0.143 |
attered | +0.141 |
robe | +0.137 |
wrist | +0.134 |
Neg logit contributions | |
hon | -0.071 |
cal | -0.068 |
louder | -0.067 |
notice | -0.067 |
hon | -0.066 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
attered | +0.178 |
possession | +0.158 |
ase | +0.154 |
umes | +0.148 |
rolley | +0.147 |
Neg logit contributions | |
time | -0.179 |
hon | -0.175 |
sne | -0.174 |
rep | -0.171 |
mad | -0.170 |
Token | to |
Token ID | 284 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
possession | +0.389 |
favourite | +0.370 |
wrist | +0.352 |
artments | +0.348 |
attered | +0.344 |
Neg logit contributions | |
hon | -0.240 |
hon | -0.237 |
," | -0.201 |
sne | -0.188 |
rep | -0.185 |
Token | the |
Token ID | 262 |
Feature activation | -6.68e-03 |
Pos logit contributions | |
hon | +0.015 |
notice | +0.014 |
hon | +0.013 |
rep | +0.013 |
cal | +0.012 |
Neg logit contributions | |
possession | -0.025 |
favourite | -0.024 |
favorite | -0.022 |
attered | -0.022 |
purse | -0.021 |
Token | woman |
Token ID | 2415 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
possession | +0.012 |
favourite | +0.011 |
attered | +0.010 |
favorite | +0.009 |
robe | +0.009 |
Neg logit contributions | |
hon | -0.011 |
notice | -0.010 |
hon | -0.010 |
rep | -0.010 |
louder | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
possession | +0.102 |
favourite | +0.097 |
favorite | +0.090 |
purse | +0.087 |
wrist | +0.084 |
Neg logit contributions | |
hon | -0.070 |
notice | -0.060 |
rep | -0.059 |
hon | -0.058 |
cal | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
possession | +0.076 |
favourite | +0.070 |
favorite | +0.063 |
possessions | +0.062 |
purse | +0.062 |
Neg logit contributions | |
hon | -0.050 |
notice | -0.044 |
rep | -0.043 |
cal | -0.042 |
overcome | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
possession | +0.007 |
favourite | +0.007 |
favorite | +0.006 |
purse | +0.006 |
wrist | +0.006 |
Neg logit contributions | |
hon | -0.005 |
notice | -0.004 |
cal | -0.004 |
hon | -0.004 |
rep | -0.004 |
Token | gave |
Token ID | 2921 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
possession | +0.081 |
favourite | +0.079 |
favorite | +0.074 |
purse | +0.071 |
wrist | +0.067 |
Neg logit contributions | |
hon | -0.058 |
notice | -0.048 |
rep | -0.046 |
announcements | -0.045 |
hon | -0.045 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
possession | +0.122 |
favourite | +0.118 |
favorite | +0.110 |
purse | +0.107 |
wrist | +0.105 |
Neg logit contributions | |
hon | -0.066 |
notice | -0.054 |
rep | -0.053 |
hon | -0.053 |
cal | -0.053 |
Token | my |
Token ID | 1820 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
possession | +0.040 |
favourite | +0.039 |
favorite | +0.036 |
purse | +0.035 |
wrist | +0.034 |
Neg logit contributions | |
hon | -0.028 |
notice | -0.024 |
cal | -0.023 |
rep | -0.023 |
hon | -0.023 |
Token | his |
Token ID | 465 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
possession | +0.151 |
favourite | +0.144 |
wrist | +0.135 |
purse | +0.132 |
attered | +0.132 |
Neg logit contributions | |
hon | -0.075 |
cal | -0.072 |
louder | -0.064 |
notice | -0.064 |
overcome | -0.064 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
possession | +0.222 |
favourite | +0.203 |
attered | +0.198 |
wrist | +0.196 |
purse | +0.186 |
Neg logit contributions | |
hon | -0.159 |
sne | -0.152 |
notice | -0.151 |
cal | -0.151 |
louder | -0.149 |
Token | back |
Token ID | 736 |
Feature activation | -8.72e-02 |
Pos logit contributions | |
possession | +0.397 |
favourite | +0.360 |
wrist | +0.355 |
artments | +0.340 |
purse | +0.335 |
Neg logit contributions | |
louder | -0.259 |
hon | -0.259 |
sne | -0.247 |
hon | -0.246 |
boom | -0.241 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
possession | +0.196 |
favourite | +0.191 |
favorite | +0.176 |
purse | +0.173 |
wrist | +0.170 |
Neg logit contributions | |
hon | -0.100 |
notice | -0.086 |
hon | -0.083 |
rep | -0.083 |
cal | -0.081 |
Token | |
Token ID | 220 |
Feature activation | -3.83e-03 |
Pos logit contributions | |
possession | +0.068 |
favourite | +0.064 |
favorite | +0.060 |
purse | +0.059 |
wrist | +0.058 |
Neg logit contributions | |
hon | -0.031 |
rep | -0.027 |
hon | -0.026 |
notice | -0.024 |
cal | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
possession | +0.008 |
favourite | +0.007 |
favorite | +0.007 |
purse | +0.007 |
wrist | +0.006 |
Neg logit contributions | |
hon | -0.006 |
notice | -0.005 |
rep | -0.005 |
hon | -0.005 |
overcome | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
hon | +0.044 |
notice | +0.037 |
rep | +0.036 |
hon | +0.036 |
overcome | +0.036 |
Neg logit contributions | |
possession | -0.059 |
favourite | -0.056 |
favorite | -0.052 |
purse | -0.051 |
wrist | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
possession | +0.039 |
favourite | +0.038 |
favorite | +0.035 |
purse | +0.034 |
wrist | +0.032 |
Neg logit contributions | |
hon | -0.031 |
notice | -0.027 |
rep | -0.026 |
cal | -0.026 |
hon | -0.026 |
Token | show |
Token ID | 905 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
hon | +0.060 |
cal | +0.050 |
announcements | +0.048 |
rep | +0.048 |
louder | +0.047 |
Neg logit contributions | |
possession | -0.078 |
favourite | -0.076 |
favorite | -0.071 |
purse | -0.067 |
wrist | -0.067 |
Token | his |
Token ID | 465 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
possession | +0.102 |
favourite | +0.095 |
favorite | +0.089 |
purse | +0.087 |
wrist | +0.082 |
Neg logit contributions | |
hon | -0.087 |
notice | -0.072 |
rep | -0.070 |
cal | -0.070 |
slit | -0.069 |
Token | friend |
Token ID | 1545 |
Feature activation | -8.75e-02 |
Pos logit contributions | |
possession | +0.196 |
favourite | +0.181 |
purse | +0.163 |
attered | +0.162 |
favorite | +0.161 |
Neg logit contributions | |
hon | -0.199 |
hon | -0.179 |
rep | -0.176 |
cal | -0.174 |
notice | -0.174 |
Token | his |
Token ID | 465 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
possession | +0.184 |
favourite | +0.176 |
favorite | +0.166 |
purse | +0.160 |
wallet | +0.150 |
Neg logit contributions | |
hon | -0.130 |
notice | -0.107 |
rep | -0.105 |
announcements | -0.104 |
oblig | -0.102 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
possession | +0.197 |
favourite | +0.179 |
purse | +0.163 |
attered | +0.162 |
favorite | +0.159 |
Neg logit contributions | |
hon | -0.207 |
hon | -0.186 |
rep | -0.186 |
notice | -0.182 |
cal | -0.181 |
Token | car |
Token ID | 1097 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
possession | +0.378 |
favourite | +0.357 |
favorite | +0.321 |
purse | +0.319 |
wrist | +0.314 |
Neg logit contributions | |
hon | -0.326 |
hon | -0.291 |
rep | -0.280 |
louder | -0.275 |
overcome | -0.272 |
Token | . |
Token ID | 13 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
possession | +0.212 |
favourite | +0.205 |
favorite | +0.193 |
purse | +0.185 |
wrist | +0.178 |
Neg logit contributions | |
hon | -0.155 |
notice | -0.130 |
cal | -0.127 |
announcements | -0.125 |
rep | -0.125 |
Token | His |
Token ID | 2399 |
Feature activation | -7.94e-02 |
Pos logit contributions | |
possession | +0.060 |
favourite | +0.056 |
favorite | +0.051 |
purse | +0.051 |
wrist | +0.049 |
Neg logit contributions | |
hon | -0.047 |
rep | -0.041 |
hon | -0.041 |
notice | -0.039 |
cal | -0.038 |
Token | friend |
Token ID | 1545 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
possession | +0.149 |
favourite | +0.130 |
attered | +0.123 |
purse | +0.122 |
wrist | +0.119 |
Neg logit contributions | |
hon | -0.124 |
hon | -0.114 |
rep | -0.110 |
louder | -0.110 |
overcome | -0.109 |
Token | liked |
Token ID | 8288 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
favourite | +0.092 |
possession | +0.087 |
favorite | +0.087 |
purse | +0.077 |
wallet | +0.074 |
Neg logit contributions | |
hon | -0.096 |
announcements | -0.081 |
boom | -0.080 |
notice | -0.079 |
explode | -0.078 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
possession | +0.019 |
favourite | +0.017 |
favorite | +0.016 |
purse | +0.016 |
wrist | +0.015 |
Neg logit contributions | |
hon | -0.013 |
hon | -0.011 |
rep | -0.011 |
notice | -0.011 |
cal | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
hon | +0.048 |
oblig | +0.043 |
notice | +0.042 |
announcements | +0.041 |
rr | +0.041 |
Neg logit contributions | |
favourite | -0.044 |
possession | -0.042 |
favorite | -0.042 |
purse | -0.041 |
wallet | -0.038 |
Token | he |
Token ID | 339 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
hon | +0.022 |
rr | +0.018 |
oblig | +0.018 |
notice | +0.018 |
rep | +0.017 |
Neg logit contributions | |
possession | -0.026 |
favourite | -0.026 |
favorite | -0.025 |
purse | -0.024 |
wallet | -0.022 |
Token | started |
Token ID | 2067 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
hon | +0.071 |
rep | +0.059 |
oblig | +0.058 |
notice | +0.058 |
cal | +0.057 |
Neg logit contributions | |
possession | -0.101 |
favourite | -0.101 |
favorite | -0.095 |
purse | -0.090 |
wrist | -0.086 |
Token | to |
Token ID | 284 |
Feature activation | +6.97e-02 |
Pos logit contributions | |
possession | +0.041 |
favourite | +0.039 |
favorite | +0.036 |
purse | +0.034 |
wrist | +0.031 |
Neg logit contributions | |
hon | -0.057 |
notice | -0.050 |
oblig | -0.049 |
rep | -0.049 |
cal | -0.049 |
Token | sne |
Token ID | 10505 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
hon | +0.120 |
rr | +0.104 |
Jim | +0.103 |
ye | +0.099 |
oblig | +0.097 |
Neg logit contributions | |
favourite | -0.110 |
favorite | -0.108 |
purse | -0.104 |
microscope | -0.101 |
possession | -0.101 |
Token | e |
Token ID | 68 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
rep | +0.237 |
hon | +0.224 |
announcements | +0.212 |
hon | +0.203 |
louder | +0.203 |
Neg logit contributions | |
possession | -0.277 |
favourite | -0.239 |
purse | -0.231 |
favorite | -0.224 |
ase | -0.222 |
Token | ze |
Token ID | 2736 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
favourite | +0.063 |
possession | +0.061 |
favorite | +0.060 |
purse | +0.057 |
wrist | +0.054 |
Neg logit contributions | |
hon | -0.061 |
oblig | -0.057 |
rep | -0.056 |
notice | -0.056 |
cal | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
hon | +0.045 |
oblig | +0.043 |
rep | +0.041 |
notice | +0.039 |
overcome | +0.039 |
Neg logit contributions | |
possession | -0.032 |
favourite | -0.031 |
favorite | -0.030 |
robe | -0.029 |
purse | -0.028 |
Token | " |
Token ID | 366 |
Feature activation | -9.19e-02 |
Pos logit contributions | |
possession | +0.026 |
favourite | +0.025 |
favorite | +0.023 |
purse | +0.022 |
wrist | +0.021 |
Neg logit contributions | |
hon | -0.021 |
notice | -0.018 |
hon | -0.018 |
rep | -0.018 |
cal | -0.017 |
Token | A |
Token ID | 32 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
possession | +0.163 |
favourite | +0.158 |
favorite | +0.144 |
purse | +0.143 |
wrist | +0.137 |
Neg logit contributions | |
hon | -0.147 |
rep | -0.132 |
oblig | -0.128 |
cal | -0.127 |
overcome | -0.127 |
Token | made |
Token ID | 925 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
possession | +0.139 |
favourite | +0.136 |
favorite | +0.127 |
purse | +0.122 |
wrist | +0.117 |
Neg logit contributions | |
hon | -0.096 |
notice | -0.080 |
rep | -0.079 |
cal | -0.078 |
oblig | -0.077 |
Token | a |
Token ID | 257 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
hon | +0.072 |
rep | +0.067 |
oblig | +0.067 |
notice | +0.065 |
overcome | +0.063 |
Neg logit contributions | |
favourite | -0.060 |
favorite | -0.058 |
possession | -0.056 |
purse | -0.049 |
brand | -0.049 |
Token | loud |
Token ID | 7812 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
hon | +0.055 |
rep | +0.044 |
oblig | +0.043 |
notice | +0.040 |
overcome | +0.040 |
Neg logit contributions | |
favourite | -0.083 |
possession | -0.081 |
favorite | -0.080 |
purse | -0.076 |
brand | -0.073 |
Token | and |
Token ID | 290 |
Feature activation | -7.94e-04 |
Pos logit contributions | |
hon | +0.208 |
overcome | +0.201 |
Jim | +0.195 |
oblig | +0.190 |
rep | +0.188 |
Neg logit contributions | |
possession | -0.258 |
purse | -0.240 |
favourite | -0.235 |
wrist | -0.235 |
favorite | -0.233 |
Token | messy |
Token ID | 24097 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
favourite | +0.001 |
favorite | +0.001 |
possession | +0.001 |
brand | +0.001 |
purse | +0.001 |
Neg logit contributions | |
hon | -0.001 |
rr | -0.001 |
oblig | -0.001 |
overcome | -0.001 |
rep | -0.001 |
Token | noise |
Token ID | 7838 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
hon | +0.096 |
overcome | +0.083 |
cal | +0.082 |
rep | +0.081 |
oblig | +0.081 |
Neg logit contributions | |
possession | -0.080 |
purse | -0.071 |
favourite | -0.070 |
favorite | -0.067 |
wrist | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
hon | +0.235 |
oblig | +0.211 |
rep | +0.203 |
overcome | +0.201 |
rr | +0.194 |
Neg logit contributions | |
possession | -0.219 |
favourite | -0.213 |
favorite | -0.205 |
purse | -0.186 |
brand | -0.178 |
Token |
|
Token ID | 198 |
Feature activation | +5.58e-02 |
Pos logit contributions | |
hon | +0.021 |
notice | +0.019 |
hon | +0.019 |
cal | +0.019 |
oblig | +0.019 |
Neg logit contributions | |
possession | -0.009 |
favourite | -0.009 |
favorite | -0.008 |
purse | -0.007 |
wrist | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
hon | +0.104 |
rep | +0.092 |
notice | +0.091 |
cal | +0.091 |
hon | +0.091 |
Neg logit contributions | |
possession | -0.086 |
favourite | -0.083 |
favorite | -0.075 |
purse | -0.071 |
wrist | -0.068 |
Token | " |
Token ID | 1 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
hon | +0.077 |
rep | +0.071 |
oblig | +0.069 |
cal | +0.068 |
hon | +0.067 |
Neg logit contributions | |
possession | -0.085 |
favourite | -0.079 |
purse | -0.075 |
favorite | -0.071 |
wrist | -0.070 |
Token | Hey |
Token ID | 10814 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
favourite | +0.168 |
possession | +0.166 |
favorite | +0.156 |
purse | +0.155 |
wrist | +0.145 |
Neg logit contributions | |
hon | -0.172 |
overcome | -0.162 |
oblig | -0.160 |
cal | -0.154 |
rep | -0.153 |
Token | . |
Token ID | 13 |
Feature activation | +7.40e-03 |
Pos logit contributions | |
possession | +0.044 |
favourite | +0.044 |
favorite | +0.042 |
purse | +0.040 |
wrist | +0.037 |
Neg logit contributions | |
hon | -0.046 |
oblig | -0.042 |
overcome | -0.041 |
Jim | -0.041 |
rep | -0.040 |
Token | The |
Token ID | 383 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
hon | +0.014 |
notice | +0.013 |
cal | +0.012 |
oblig | +0.012 |
overcome | +0.012 |
Neg logit contributions | |
possession | -0.011 |
favourite | -0.011 |
favorite | -0.010 |
purse | -0.010 |
wrist | -0.009 |
Token | rocket |
Token ID | 10701 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
hon | +0.028 |
notice | +0.024 |
rep | +0.024 |
oblig | +0.023 |
cal | +0.023 |
Neg logit contributions | |
possession | -0.022 |
favourite | -0.021 |
favorite | -0.019 |
purse | -0.018 |
wrist | -0.017 |
Token | makes |
Token ID | 1838 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
hon | +0.020 |
rep | +0.018 |
oblig | +0.018 |
Jim | +0.017 |
notice | +0.017 |
Neg logit contributions | |
favourite | -0.022 |
possession | -0.022 |
favorite | -0.021 |
purse | -0.019 |
wrist | -0.019 |
Token | another |
Token ID | 1194 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
hon | +0.089 |
rep | +0.086 |
oblig | +0.086 |
notice | +0.080 |
Run | +0.078 |
Neg logit contributions | |
favourite | -0.081 |
favorite | -0.079 |
possession | -0.074 |
brand | -0.064 |
purse | -0.061 |
Token | loud |
Token ID | 7812 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
hon | +0.045 |
Jim | +0.037 |
oblig | +0.037 |
overcome | +0.036 |
rep | +0.036 |
Neg logit contributions | |
possession | -0.052 |
favourite | -0.051 |
favorite | -0.048 |
brand | -0.045 |
wrist | -0.045 |
Token | sound |
Token ID | 2128 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Jud | +0.224 |
Jim | +0.222 |
overcome | +0.211 |
hon | +0.210 |
oblig | +0.201 |
Neg logit contributions | |
possession | -0.264 |
favorite | -0.243 |
favourite | -0.239 |
wrist | -0.236 |
brand | -0.232 |
Token | and |
Token ID | 290 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
hon | +0.220 |
oblig | +0.210 |
rep | +0.206 |
Jud | +0.203 |
Jim | +0.200 |
Neg logit contributions | |
possession | -0.146 |
favourite | -0.143 |
favorite | -0.141 |
brand | -0.122 |
where | -0.122 |
Token | lifts |
Token ID | 27103 |
Feature activation | +8.14e-02 |
Pos logit contributions | |
hon | +0.015 |
oblig | +0.014 |
Jim | +0.013 |
rr | +0.013 |
Jud | +0.013 |
Neg logit contributions | |
favourite | -0.019 |
favorite | -0.019 |
possession | -0.018 |
brand | -0.017 |
purse | -0.016 |
Token | off |
Token ID | 572 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
hon | +0.153 |
oblig | +0.144 |
rep | +0.143 |
overcome | +0.132 |
achieve | +0.131 |
Neg logit contributions | |
possession | -0.130 |
favourite | -0.124 |
favorite | -0.124 |
purse | -0.106 |
item | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | +6.18e-03 |
Pos logit contributions | |
hon | +0.055 |
oblig | +0.050 |
overcome | +0.050 |
rep | +0.048 |
rr | +0.047 |
Neg logit contributions | |
possession | -0.051 |
favourite | -0.048 |
favorite | -0.047 |
purse | -0.045 |
item | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
hon | +0.011 |
rr | +0.009 |
oblig | +0.009 |
announcements | +0.008 |
cal | +0.008 |
Neg logit contributions | |
favourite | -0.010 |
possession | -0.010 |
favorite | -0.009 |
purse | -0.008 |
wallet | -0.008 |
Token | mill |
Token ID | 3939 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
hon | +0.084 |
notice | +0.073 |
rep | +0.072 |
hon | +0.072 |
cal | +0.071 |
Neg logit contributions | |
possession | -0.075 |
favourite | -0.072 |
favorite | -0.066 |
purse | -0.063 |
wrist | -0.060 |
Token | er |
Token ID | 263 |
Feature activation | +4.14e-03 |
Pos logit contributions | |
hon | +0.058 |
notice | +0.052 |
rr | +0.051 |
oblig | +0.050 |
boom | +0.048 |
Neg logit contributions | |
possession | -0.046 |
favourite | -0.046 |
purse | -0.042 |
favorite | -0.041 |
wallet | -0.041 |
Token | heard |
Token ID | 2982 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
hon | +0.007 |
notice | +0.006 |
cal | +0.006 |
oblig | +0.006 |
announcements | +0.006 |
Neg logit contributions | |
possession | -0.007 |
favourite | -0.007 |
favorite | -0.006 |
purse | -0.006 |
attered | -0.006 |
Token | loud |
Token ID | 7812 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
hon | +0.124 |
oblig | +0.108 |
notice | +0.106 |
rep | +0.103 |
overcome | +0.100 |
Neg logit contributions | |
favourite | -0.107 |
possession | -0.105 |
favorite | -0.098 |
brand | -0.092 |
purse | -0.089 |
Token | noises |
Token ID | 26782 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
hon | +0.277 |
oblig | +0.256 |
overcome | +0.251 |
notice | +0.246 |
Jim | +0.245 |
Neg logit contributions | |
possession | -0.209 |
favourite | -0.202 |
favorite | -0.186 |
brand | -0.175 |
purse | -0.173 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
hon | +0.263 |
oblig | +0.239 |
Hon | +0.224 |
overcome | +0.221 |
rr | +0.221 |
Neg logit contributions | |
possession | -0.164 |
favourite | -0.163 |
favorite | -0.144 |
brand | -0.143 |
purse | -0.140 |
Token | . |
Token ID | 13 |
Feature activation | +3.00e-03 |
Pos logit contributions | |
hon | +0.043 |
oblig | +0.037 |
cal | +0.035 |
RO | +0.034 |
Hon | +0.034 |
Neg logit contributions | |
possession | -0.037 |
favourite | -0.034 |
favorite | -0.031 |
purse | -0.030 |
attic | -0.029 |
Token | He |
Token ID | 679 |
Feature activation | +6.32e-03 |
Pos logit contributions | |
hon | +0.006 |
notice | +0.006 |
rep | +0.005 |
hon | +0.005 |
cal | +0.005 |
Neg logit contributions | |
possession | -0.004 |
favourite | -0.004 |
favorite | -0.004 |
purse | -0.003 |
wrist | -0.003 |
Token | knew |
Token ID | 2993 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
hon | +0.012 |
rep | +0.010 |
oblig | +0.010 |
notice | +0.010 |
boom | +0.010 |
Neg logit contributions | |
favourite | -0.010 |
possession | -0.010 |
favorite | -0.009 |
purse | -0.008 |
brand | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
possession | +0.155 |
favourite | +0.148 |
favorite | +0.136 |
purse | +0.129 |
wallet | +0.120 |
Neg logit contributions | |
hon | -0.155 |
notice | -0.130 |
oblig | -0.126 |
announcements | -0.124 |
boom | -0.123 |
Token | who |
Token ID | 508 |
Feature activation | -8.59e-02 |
Pos logit contributions | |
hon | +0.045 |
overcome | +0.041 |
cal | +0.040 |
notice | +0.040 |
hon | +0.039 |
Neg logit contributions | |
possession | -0.062 |
favourite | -0.060 |
attered | -0.056 |
purse | -0.055 |
favorite | -0.054 |
Token | can |
Token ID | 460 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
possession | +0.175 |
favourite | +0.169 |
favorite | +0.155 |
purse | +0.155 |
wrist | +0.151 |
Neg logit contributions | |
hon | -0.122 |
notice | -0.105 |
overcome | -0.102 |
rep | -0.102 |
hon | -0.101 |
Token | help |
Token ID | 1037 |
Feature activation | +9.19e-02 |
Pos logit contributions | |
hon | +0.025 |
announcements | +0.021 |
cal | +0.021 |
thunder | +0.021 |
louder | +0.021 |
Neg logit contributions | |
favourite | -0.033 |
possession | -0.033 |
favorite | -0.031 |
purse | -0.029 |
wrist | -0.029 |
Token | you |
Token ID | 345 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
hon | +0.145 |
rep | +0.122 |
cal | +0.120 |
hon | +0.120 |
announcements | +0.118 |
Neg logit contributions | |
possession | -0.183 |
favourite | -0.174 |
favorite | -0.164 |
purse | -0.154 |
wrist | -0.148 |
Token | when |
Token ID | 618 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
hon | +0.164 |
announcements | +0.135 |
hon | +0.134 |
cal | +0.134 |
rep | +0.134 |
Neg logit contributions | |
possession | -0.184 |
favourite | -0.179 |
favorite | -0.167 |
purse | -0.156 |
wrist | -0.153 |
Token | you |
Token ID | 345 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
hon | +0.055 |
cal | +0.045 |
notice | +0.044 |
hon | +0.044 |
oblig | +0.043 |
Neg logit contributions | |
possession | -0.081 |
favourite | -0.079 |
favorite | -0.074 |
purse | -0.070 |
wrist | -0.067 |
Token | need |
Token ID | 761 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
hon | +0.210 |
announcements | +0.184 |
louder | +0.177 |
thunder | +0.172 |
hon | +0.172 |
Neg logit contributions | |
favourite | -0.244 |
possession | -0.238 |
favorite | -0.228 |
favorites | -0.210 |
purse | -0.210 |
Token | it |
Token ID | 340 |
Feature activation | -6.48e-02 |
Pos logit contributions | |
possession | +0.063 |
favourite | +0.061 |
favorite | +0.056 |
purse | +0.055 |
attered | +0.053 |
Neg logit contributions | |
hon | -0.047 |
notice | -0.041 |
rep | -0.041 |
hon | -0.040 |
cal | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
possession | +0.126 |
favourite | +0.125 |
favorite | +0.119 |
purse | +0.106 |
brand | +0.104 |
Neg logit contributions | |
hon | -0.106 |
announcements | -0.089 |
hon | -0.088 |
notice | -0.086 |
RO | -0.085 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.024 |
favorite | +0.022 |
purse | +0.022 |
attered | +0.021 |
Neg logit contributions | |
hon | -0.013 |
cal | -0.011 |
rep | -0.011 |
overcome | -0.010 |
notice | -0.010 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
possession | +0.030 |
favourite | +0.029 |
favorite | +0.026 |
purse | +0.025 |
wrist | +0.024 |
Neg logit contributions | |
hon | -0.036 |
notice | -0.031 |
announcements | -0.031 |
louder | -0.031 |
hon | -0.031 |
Token | sad |
Token ID | 6507 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
possession | +0.059 |
favourite | +0.056 |
favorite | +0.051 |
purse | +0.051 |
wrist | +0.049 |
Neg logit contributions | |
hon | -0.038 |
hon | -0.033 |
cal | -0.033 |
notice | -0.033 |
louder | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
possession | +0.073 |
favourite | +0.070 |
favorite | +0.066 |
purse | +0.062 |
wrist | +0.058 |
Neg logit contributions | |
hon | -0.069 |
rep | -0.057 |
hon | -0.057 |
oblig | -0.057 |
cal | -0.057 |
Token | went |
Token ID | 1816 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
possession | +0.065 |
favourite | +0.064 |
favorite | +0.063 |
purse | +0.059 |
wallet | +0.055 |
Neg logit contributions | |
hon | -0.055 |
rr | -0.045 |
oblig | -0.045 |
notice | -0.043 |
overcome | -0.042 |
Token | away |
Token ID | 1497 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
hon | +0.109 |
notice | +0.103 |
hon | +0.102 |
louder | +0.099 |
cal | +0.099 |
Neg logit contributions | |
possession | -0.114 |
favourite | -0.110 |
favorite | -0.098 |
wrist | -0.095 |
purse | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
possession | +0.095 |
favourite | +0.089 |
favorite | +0.084 |
purse | +0.080 |
wrist | +0.075 |
Neg logit contributions | |
hon | -0.085 |
announcements | -0.067 |
oblig | -0.067 |
notice | -0.066 |
Jim | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
possession | +0.069 |
favourite | +0.064 |
purse | +0.060 |
favorite | +0.059 |
wrist | +0.057 |
Neg logit contributions | |
hon | -0.050 |
notice | -0.043 |
rep | -0.042 |
overcome | -0.042 |
cal | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
hon | +0.043 |
notice | +0.037 |
rep | +0.037 |
hon | +0.036 |
cal | +0.036 |
Neg logit contributions | |
possession | -0.050 |
favourite | -0.048 |
favorite | -0.044 |
purse | -0.043 |
wrist | -0.041 |
Token | Later |
Token ID | 18602 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
hon | +0.034 |
notice | +0.028 |
rep | +0.028 |
cal | +0.028 |
hon | +0.028 |
Neg logit contributions | |
possession | -0.046 |
favourite | -0.044 |
favorite | -0.041 |
purse | -0.039 |
wrist | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
possession | +0.053 |
favourite | +0.051 |
favorite | +0.047 |
purse | +0.047 |
wallet | +0.042 |
Neg logit contributions | |
hon | -0.078 |
announcements | -0.068 |
overcome | -0.066 |
notice | -0.065 |
hon | -0.065 |
Token | Fred |
Token ID | 8559 |
Feature activation | +6.01e-02 |
Pos logit contributions | |
favourite | +0.038 |
possession | +0.036 |
favorite | +0.035 |
purse | +0.031 |
favourites | +0.030 |
Neg logit contributions | |
hon | -0.031 |
cal | -0.025 |
announcements | -0.025 |
hon | -0.025 |
notice | -0.025 |
Token | heard |
Token ID | 2982 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
hon | +0.122 |
boom | +0.110 |
louder | +0.109 |
oblig | +0.109 |
announcements | +0.108 |
Neg logit contributions | |
favourite | -0.077 |
possession | -0.074 |
favorite | -0.074 |
purse | -0.067 |
wallet | -0.063 |
Token | very |
Token ID | 845 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
favourite | +0.002 |
favorite | +0.002 |
brand | +0.002 |
possession | +0.001 |
attered | +0.001 |
Neg logit contributions | |
hon | -0.002 |
oblig | -0.002 |
Jim | -0.002 |
inconvenience | -0.002 |
Josh | -0.002 |
Token | dark |
Token ID | 3223 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
favourite | +0.058 |
favorite | +0.055 |
possession | +0.054 |
brand | +0.049 |
attered | +0.043 |
Neg logit contributions | |
hon | -0.073 |
oblig | -0.062 |
rr | -0.060 |
Jim | -0.059 |
rep | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e-03 |
Pos logit contributions | |
hon | +0.053 |
oblig | +0.047 |
Jim | +0.044 |
overcome | +0.044 |
notice | +0.044 |
Neg logit contributions | |
possession | -0.042 |
favourite | -0.041 |
favorite | -0.038 |
brand | -0.034 |
purse | -0.033 |
Token | The |
Token ID | 383 |
Feature activation | +2.20e-03 |
Pos logit contributions | |
hon | +0.003 |
notice | +0.002 |
hon | +0.002 |
cal | +0.002 |
rep | +0.002 |
Neg logit contributions | |
favourite | -0.001 |
possession | -0.001 |
favorite | -0.001 |
purse | -0.001 |
wrist | -0.001 |
Token | little |
Token ID | 1310 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
hon | +0.004 |
notice | +0.004 |
oblig | +0.004 |
rep | +0.003 |
overcome | +0.003 |
Neg logit contributions | |
favourite | -0.004 |
possession | -0.004 |
favorite | -0.003 |
purse | -0.003 |
wrist | -0.003 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
hon | +0.066 |
oblig | +0.058 |
notice | +0.058 |
overcome | +0.057 |
rep | +0.056 |
Neg logit contributions | |
possession | -0.058 |
favourite | -0.056 |
favorite | -0.053 |
purse | -0.051 |
wrist | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
hon | +0.034 |
rep | +0.030 |
oblig | +0.029 |
announcements | +0.029 |
notice | +0.028 |
Neg logit contributions | |
favourite | -0.031 |
possession | -0.030 |
favorite | -0.028 |
purse | -0.024 |
attered | -0.023 |
Token | so |
Token ID | 523 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
favourite | +0.028 |
possession | +0.028 |
favorite | +0.026 |
purse | +0.024 |
brand | +0.023 |
Neg logit contributions | |
hon | -0.032 |
rep | -0.026 |
oblig | -0.026 |
notice | -0.026 |
Run | -0.025 |
Token | high |
Token ID | 1029 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
hon | +0.022 |
oblig | +0.018 |
rep | +0.018 |
hon | +0.018 |
announcements | +0.018 |
Neg logit contributions | |
favourite | -0.020 |
possession | -0.020 |
favorite | -0.019 |
brand | -0.017 |
purse | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
hon | +0.160 |
oblig | +0.146 |
rr | +0.141 |
announcements | +0.140 |
notice | +0.140 |
Neg logit contributions | |
favourite | -0.092 |
possession | -0.092 |
favorite | -0.084 |
brand | -0.071 |
purse | -0.069 |
Token | it |
Token ID | 340 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
possession | +0.019 |
favourite | +0.019 |
favorite | +0.017 |
purse | +0.016 |
brand | +0.015 |
Neg logit contributions | |
hon | -0.021 |
oblig | -0.018 |
rr | -0.017 |
notice | -0.017 |
rep | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
possession | +0.050 |
favourite | +0.048 |
purse | +0.044 |
favorite | +0.043 |
wrist | +0.043 |
Neg logit contributions | |
hon | -0.052 |
notice | -0.044 |
RO | -0.044 |
announcements | -0.044 |
oblig | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -1.75e-03 |
Pos logit contributions | |
possession | +0.040 |
favourite | +0.038 |
favorite | +0.035 |
purse | +0.034 |
wrist | +0.032 |
Neg logit contributions | |
hon | -0.042 |
notice | -0.036 |
oblig | -0.035 |
rep | -0.035 |
announcements | -0.034 |
Token | he |
Token ID | 339 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
possession | +0.004 |
favourite | +0.003 |
favorite | +0.003 |
purse | +0.003 |
wrist | +0.003 |
Neg logit contributions | |
hon | -0.003 |
notice | -0.002 |
oblig | -0.002 |
rep | -0.002 |
announcements | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -7.86e-03 |
Pos logit contributions | |
hon | +0.068 |
notice | +0.059 |
rep | +0.057 |
cal | +0.056 |
announcements | +0.055 |
Neg logit contributions | |
possession | -0.078 |
favourite | -0.075 |
favorite | -0.070 |
purse | -0.067 |
wrist | -0.064 |
Token | trying |
Token ID | 2111 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
possession | +0.013 |
favourite | +0.012 |
favorite | +0.011 |
purse | +0.011 |
brand | +0.010 |
Neg logit contributions | |
hon | -0.015 |
notice | -0.013 |
rep | -0.012 |
announcements | -0.012 |
Run | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.023 |
purse | +0.020 |
favorite | +0.020 |
wrist | +0.020 |
Neg logit contributions | |
hon | -0.022 |
notice | -0.020 |
louder | -0.020 |
overcome | -0.020 |
cal | -0.020 |
Token | collect |
Token ID | 2824 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
hon | +0.059 |
announcements | +0.051 |
louder | +0.047 |
rep | +0.047 |
boom | +0.046 |
Neg logit contributions | |
possession | -0.056 |
favourite | -0.053 |
wrist | -0.050 |
purse | -0.049 |
favorite | -0.049 |
Token | wood |
Token ID | 4898 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
possession | +0.098 |
favourite | +0.094 |
favorite | +0.086 |
purse | +0.083 |
wrist | +0.077 |
Neg logit contributions | |
hon | -0.093 |
notice | -0.081 |
rep | -0.080 |
hon | -0.079 |
oblig | -0.077 |
Token | for |
Token ID | 329 |
Feature activation | -7.19e-03 |
Pos logit contributions | |
possession | +0.050 |
favourite | +0.048 |
favorite | +0.044 |
purse | +0.042 |
wrist | +0.040 |
Neg logit contributions | |
hon | -0.043 |
notice | -0.037 |
rep | -0.036 |
hon | -0.035 |
cal | -0.035 |
Token | his |
Token ID | 465 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
possession | +0.015 |
favourite | +0.014 |
favorite | +0.013 |
purse | +0.013 |
brand | +0.012 |
Neg logit contributions | |
hon | -0.011 |
notice | -0.009 |
rep | -0.009 |
cal | -0.009 |
hon | -0.008 |
Token | job |
Token ID | 1693 |
Feature activation | +9.48e-04 |
Pos logit contributions | |
possession | +0.095 |
favourite | +0.091 |
favorite | +0.083 |
purse | +0.079 |
wrist | +0.075 |
Neg logit contributions | |
hon | -0.113 |
notice | -0.099 |
rep | -0.098 |
hon | -0.097 |
cal | -0.097 |
Token | 's |
Token ID | 338 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
possession | +0.021 |
favourite | +0.020 |
favorite | +0.019 |
purse | +0.018 |
wrist | +0.017 |
Neg logit contributions | |
hon | -0.015 |
notice | -0.013 |
rep | -0.012 |
hon | -0.012 |
cal | -0.012 |
Token | hiding |
Token ID | 11816 |
Feature activation | +1.15e-03 |
Pos logit contributions | |
possession | +0.059 |
favourite | +0.056 |
favorite | +0.051 |
purse | +0.050 |
wrist | +0.048 |
Neg logit contributions | |
hon | -0.058 |
notice | -0.051 |
rep | -0.050 |
hon | -0.050 |
cal | -0.050 |
Token | spot |
Token ID | 4136 |
Feature activation | -6.95e-02 |
Pos logit contributions | |
hon | +0.002 |
notice | +0.002 |
rep | +0.002 |
hon | +0.002 |
cal | +0.002 |
Neg logit contributions | |
possession | -0.002 |
favourite | -0.002 |
favorite | -0.002 |
purse | -0.002 |
wrist | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e-03 |
Pos logit contributions | |
possession | +0.147 |
favourite | +0.142 |
favorite | +0.132 |
purse | +0.128 |
wrist | +0.123 |
Neg logit contributions | |
hon | -0.097 |
notice | -0.081 |
rep | -0.079 |
hon | -0.079 |
cal | -0.078 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
possession | +0.004 |
favourite | +0.004 |
favorite | +0.004 |
purse | +0.004 |
wrist | +0.003 |
Neg logit contributions | |
hon | -0.005 |
notice | -0.004 |
rep | -0.004 |
hon | -0.004 |
cal | -0.004 |
Token | John |
Token ID | 7554 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
possession | +0.020 |
favourite | +0.019 |
favorite | +0.017 |
purse | +0.016 |
robe | +0.016 |
Neg logit contributions | |
hon | -0.027 |
announcements | -0.024 |
louder | -0.024 |
notice | -0.024 |
hon | -0.024 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
hon | +0.069 |
louder | +0.062 |
announcements | +0.062 |
rep | +0.061 |
notice | +0.060 |
Neg logit contributions | |
possession | -0.059 |
favourite | -0.057 |
favorite | -0.053 |
purse | -0.050 |
favorites | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
hon | +0.025 |
notice | +0.023 |
oblig | +0.023 |
overcome | +0.022 |
slit | +0.022 |
Neg logit contributions | |
favourite | -0.020 |
possession | -0.020 |
favorite | -0.019 |
brand | -0.017 |
purse | -0.017 |
Token | noise |
Token ID | 7838 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
hon | +0.017 |
notice | +0.013 |
overcome | +0.013 |
oblig | +0.013 |
rep | +0.012 |
Neg logit contributions | |
favourite | -0.026 |
possession | -0.026 |
favorite | -0.024 |
purse | -0.024 |
brand | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
hon | +0.190 |
overcome | +0.167 |
oblig | +0.165 |
slit | +0.164 |
notice | +0.162 |
Neg logit contributions | |
possession | -0.148 |
favourite | -0.148 |
favorite | -0.136 |
brand | -0.126 |
purse | -0.125 |
Token | He |
Token ID | 679 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
possession | +0.095 |
favourite | +0.090 |
favorite | +0.082 |
purse | +0.078 |
wrist | +0.075 |
Neg logit contributions | |
hon | -0.119 |
notice | -0.104 |
cal | -0.103 |
overcome | -0.101 |
rep | -0.101 |
Token | the |
Token ID | 262 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
possession | +0.059 |
favourite | +0.059 |
favorite | +0.055 |
brand | +0.053 |
purse | +0.050 |
Neg logit contributions | |
hon | -0.060 |
rr | -0.051 |
oblig | -0.051 |
notice | -0.050 |
overcome | -0.047 |
Token | sky |
Token ID | 6766 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
hon | +0.083 |
oblig | +0.071 |
qu | +0.067 |
rr | +0.066 |
Jim | +0.065 |
Neg logit contributions | |
favourite | -0.084 |
favorite | -0.082 |
possession | -0.079 |
attic | -0.075 |
purse | -0.073 |
Token | 's |
Token ID | 338 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
hon | +0.110 |
oblig | +0.100 |
notice | +0.096 |
rr | +0.096 |
rep | +0.094 |
Neg logit contributions | |
possession | -0.080 |
favourite | -0.078 |
favorite | -0.073 |
brand | -0.064 |
purse | -0.064 |
Token | voice |
Token ID | 3809 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
favourite | +0.030 |
possession | +0.030 |
favorite | +0.029 |
purse | +0.026 |
brand | +0.025 |
Neg logit contributions | |
hon | -0.025 |
oblig | -0.021 |
notice | -0.020 |
Jim | -0.020 |
cal | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
hon | +0.094 |
Jim | +0.087 |
oblig | +0.086 |
rep | +0.083 |
Jud | +0.081 |
Neg logit contributions | |
favourite | -0.078 |
possession | -0.077 |
favorite | -0.075 |
purse | -0.066 |
attered | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
possession | +0.021 |
favourite | +0.020 |
favorite | +0.019 |
purse | +0.017 |
wrist | +0.017 |
Neg logit contributions | |
hon | -0.026 |
notice | -0.023 |
rep | -0.023 |
cal | -0.023 |
hon | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
hon | +0.061 |
rep | +0.053 |
cal | +0.053 |
notice | +0.053 |
louder | +0.052 |
Neg logit contributions | |
possession | -0.056 |
favourite | -0.055 |
favorite | -0.050 |
purse | -0.046 |
wrist | -0.045 |
Token | The |
Token ID | 464 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
hon | +0.044 |
rep | +0.038 |
notice | +0.037 |
cal | +0.037 |
oblig | +0.037 |
Neg logit contributions | |
possession | -0.052 |
favourite | -0.051 |
favorite | -0.047 |
purse | -0.045 |
wrist | -0.043 |
Token | wise |
Token ID | 10787 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
favourite | +0.023 |
favorite | +0.022 |
possession | +0.022 |
purse | +0.021 |
wallet | +0.020 |
Neg logit contributions | |
hon | -0.025 |
oblig | -0.020 |
qu | -0.020 |
notice | -0.020 |
rr | -0.020 |
Token | man |
Token ID | 582 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
hon | +0.107 |
RO | +0.095 |
notice | +0.095 |
overcome | +0.093 |
qu | +0.093 |
Neg logit contributions | |
possession | -0.070 |
favourite | -0.060 |
favorite | -0.059 |
purse | -0.058 |
wallet | -0.056 |
Token | went |
Token ID | 1816 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
hon | +0.036 |
oblig | +0.033 |
notice | +0.032 |
rep | +0.032 |
announcements | +0.032 |
Neg logit contributions | |
favourite | -0.026 |
favorite | -0.026 |
possession | -0.025 |
purse | -0.023 |
wallet | -0.022 |
Token | shadow |
Token ID | 9082 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
hon | +0.041 |
notice | +0.036 |
hon | +0.036 |
louder | +0.036 |
thunder | +0.036 |
Neg logit contributions | |
possession | -0.047 |
favourite | -0.044 |
favorite | -0.040 |
purse | -0.040 |
wrist | -0.038 |
Token | moved |
Token ID | 3888 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
possession | +0.082 |
favourite | +0.079 |
favorite | +0.073 |
purse | +0.070 |
wrist | +0.067 |
Neg logit contributions | |
hon | -0.069 |
notice | -0.059 |
rep | -0.059 |
cal | -0.058 |
hon | -0.058 |
Token | too |
Token ID | 1165 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
hon | +0.059 |
oblig | +0.055 |
rep | +0.052 |
notice | +0.050 |
announcements | +0.050 |
Neg logit contributions | |
possession | -0.054 |
favourite | -0.052 |
favorite | -0.050 |
purse | -0.044 |
attered | -0.043 |
Token | ! |
Token ID | 0 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
hon | +0.093 |
oblig | +0.090 |
announcements | +0.086 |
rep | +0.083 |
RO | +0.080 |
Neg logit contributions | |
possession | -0.083 |
favourite | -0.074 |
favorite | -0.073 |
purse | -0.067 |
wallet | -0.061 |
Token | Max |
Token ID | 5436 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
possession | +0.044 |
favourite | +0.042 |
favorite | +0.039 |
purse | +0.038 |
wrist | +0.037 |
Neg logit contributions | |
hon | -0.030 |
notice | -0.025 |
rep | -0.025 |
hon | -0.025 |
cal | -0.025 |
Token | thought |
Token ID | 1807 |
Feature activation | -7.40e-02 |
Pos logit contributions | |
hon | +0.068 |
rep | +0.059 |
oblig | +0.058 |
notice | +0.057 |
cal | +0.057 |
Neg logit contributions | |
possession | -0.078 |
favourite | -0.077 |
favorite | -0.072 |
purse | -0.068 |
wrist | -0.065 |
Token | it |
Token ID | 340 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
possession | +0.141 |
favourite | +0.136 |
favorite | +0.127 |
purse | +0.120 |
wrist | +0.114 |
Neg logit contributions | |
hon | -0.120 |
notice | -0.102 |
rep | -0.102 |
cal | -0.100 |
hon | -0.099 |
Token | was |
Token ID | 373 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
possession | +0.179 |
favourite | +0.171 |
favorite | +0.157 |
purse | +0.152 |
wrist | +0.145 |
Neg logit contributions | |
hon | -0.171 |
notice | -0.148 |
rep | -0.146 |
hon | -0.146 |
cal | -0.145 |
Token | funny |
Token ID | 8258 |
Feature activation | -1.62e-03 |
Pos logit contributions | |
favourite | +0.106 |
possession | +0.105 |
favorite | +0.104 |
brand | +0.089 |
purse | +0.086 |
Neg logit contributions | |
hon | -0.107 |
oblig | -0.092 |
notice | -0.089 |
rep | -0.087 |
announcements | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
possession | +0.003 |
favourite | +0.003 |
favorite | +0.003 |
purse | +0.003 |
wrist | +0.003 |
Neg logit contributions | |
hon | -0.003 |
notice | -0.002 |
rep | -0.002 |
oblig | -0.002 |
cal | -0.002 |
Token | He |
Token ID | 679 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
possession | +0.060 |
favourite | +0.056 |
favorite | +0.051 |
purse | +0.051 |
wrist | +0.049 |
Neg logit contributions | |
hon | -0.049 |
notice | -0.042 |
hon | -0.042 |
rep | -0.041 |
cal | -0.041 |
Token | frustrated |
Token ID | 14718 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.023 |
favorite | +0.022 |
purse | +0.021 |
wrist | +0.020 |
Neg logit contributions | |
hon | -0.016 |
notice | -0.013 |
rep | -0.013 |
hon | -0.013 |
cal | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
possession | +0.038 |
favourite | +0.037 |
favorite | +0.034 |
purse | +0.031 |
brand | +0.030 |
Neg logit contributions | |
hon | -0.039 |
oblig | -0.033 |
RO | -0.033 |
announcements | -0.032 |
notice | -0.032 |
Token | when |
Token ID | 618 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
possession | +0.051 |
favourite | +0.050 |
favorite | +0.046 |
purse | +0.043 |
wrist | +0.041 |
Neg logit contributions | |
hon | -0.054 |
oblig | -0.045 |
rep | -0.044 |
cal | -0.044 |
notice | -0.044 |
Token | he |
Token ID | 339 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
hon | +0.059 |
cal | +0.046 |
rr | +0.046 |
notice | +0.045 |
announcements | +0.045 |
Neg logit contributions | |
possession | -0.071 |
favourite | -0.071 |
favorite | -0.064 |
purse | -0.058 |
brand | -0.057 |
Token | spotted |
Token ID | 13489 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
hon | +0.095 |
boom | +0.077 |
hon | +0.077 |
announcements | +0.076 |
rep | +0.076 |
Neg logit contributions | |
possession | -0.106 |
favourite | -0.105 |
favorite | -0.097 |
purse | -0.093 |
wallet | -0.088 |
Token | something |
Token ID | 1223 |
Feature activation | -9.85e-03 |
Pos logit contributions | |
possession | +0.099 |
favourite | +0.094 |
favorite | +0.088 |
purse | +0.086 |
wrist | +0.085 |
Neg logit contributions | |
hon | -0.069 |
rep | -0.060 |
notice | -0.059 |
hon | -0.059 |
cal | -0.059 |
Token | in |
Token ID | 287 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
possession | +0.019 |
favourite | +0.017 |
wrist | +0.016 |
favorite | +0.016 |
purse | +0.016 |
Neg logit contributions | |
hon | -0.015 |
rep | -0.014 |
louder | -0.014 |
hon | -0.014 |
notice | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
hon | +0.014 |
notice | +0.011 |
rep | +0.011 |
hon | +0.011 |
cal | +0.011 |
Neg logit contributions | |
possession | -0.027 |
favourite | -0.026 |
favorite | -0.024 |
purse | -0.024 |
wrist | -0.023 |
Token | bushes |
Token ID | 37413 |
Feature activation | -2.99e-03 |
Pos logit contributions | |
hon | +0.057 |
hon | +0.057 |
cal | +0.056 |
notice | +0.056 |
thunder | +0.055 |
Neg logit contributions | |
possession | -0.062 |
favourite | -0.060 |
favorite | -0.054 |
favourites | -0.051 |
favorites | -0.051 |
Token | ! |
Token ID | 0 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
possession | +0.006 |
favourite | +0.006 |
favorite | +0.006 |
purse | +0.006 |
wrist | +0.005 |
Neg logit contributions | |
hon | -0.004 |
notice | -0.003 |
rep | -0.003 |
hon | -0.003 |
cal | -0.003 |
Token | It |
Token ID | 632 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
possession | +0.030 |
favourite | +0.026 |
wrist | +0.026 |
favorite | +0.026 |
purse | +0.026 |
Neg logit contributions | |
hon | -0.018 |
rep | -0.017 |
thunder | -0.015 |
Run | -0.015 |
hon | -0.015 |
Token | wet |
Token ID | 9583 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
hon | +0.058 |
notice | +0.050 |
hon | +0.049 |
rep | +0.049 |
cal | +0.049 |
Neg logit contributions | |
possession | -0.066 |
favourite | -0.063 |
favorite | -0.057 |
purse | -0.056 |
wrist | -0.054 |
Token | floor |
Token ID | 4314 |
Feature activation | -1.34e-03 |
Pos logit contributions | |
hon | +0.091 |
notice | +0.077 |
hon | +0.076 |
rep | +0.076 |
cal | +0.075 |
Neg logit contributions | |
possession | -0.122 |
favourite | -0.118 |
favorite | -0.109 |
purse | -0.105 |
wrist | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | +6.98e-03 |
Pos logit contributions | |
possession | +0.002 |
favourite | +0.002 |
favorite | +0.002 |
purse | +0.002 |
wrist | +0.002 |
Neg logit contributions | |
hon | -0.002 |
oblig | -0.002 |
RO | -0.002 |
rep | -0.002 |
announcements | -0.002 |
Token | He |
Token ID | 679 |
Feature activation | +9.64e-03 |
Pos logit contributions | |
hon | +0.011 |
notice | +0.009 |
rep | +0.009 |
hon | +0.009 |
cal | +0.009 |
Neg logit contributions | |
possession | -0.014 |
favourite | -0.013 |
purse | -0.012 |
favorite | -0.012 |
wrist | -0.012 |
Token | fell |
Token ID | 3214 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
hon | +0.014 |
rep | +0.012 |
oblig | +0.011 |
notice | +0.011 |
cal | +0.011 |
Neg logit contributions | |
possession | -0.020 |
favourite | -0.019 |
favorite | -0.019 |
purse | -0.017 |
attered | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
hon | +0.096 |
oblig | +0.084 |
rep | +0.083 |
notice | +0.083 |
cal | +0.082 |
Neg logit contributions | |
possession | -0.085 |
favourite | -0.081 |
favorite | -0.076 |
purse | -0.072 |
attered | -0.068 |
Token | hurt |
Token ID | 5938 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
possession | +0.042 |
favourite | +0.041 |
favorite | +0.040 |
attered | +0.037 |
purse | +0.037 |
Neg logit contributions | |
hon | -0.029 |
oblig | -0.026 |
rep | -0.025 |
rr | -0.024 |
announcements | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
hon | +0.106 |
oblig | +0.093 |
rep | +0.091 |
notice | +0.087 |
cal | +0.087 |
Neg logit contributions | |
possession | -0.113 |
favourite | -0.108 |
favorite | -0.103 |
purse | -0.097 |
wrist | -0.091 |
Token | knee |
Token ID | 10329 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
possession | +0.064 |
favourite | +0.060 |
favorite | +0.054 |
purse | +0.052 |
wrist | +0.049 |
Neg logit contributions | |
hon | -0.084 |
notice | -0.074 |
hon | -0.074 |
rep | -0.074 |
cal | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +6.77e-03 |
Pos logit contributions | |
possession | +0.024 |
favourite | +0.024 |
favorite | +0.023 |
purse | +0.022 |
attered | +0.021 |
Neg logit contributions | |
hon | -0.026 |
oblig | -0.025 |
Jim | -0.023 |
rep | -0.023 |
RO | -0.022 |
Token | He |
Token ID | 679 |
Feature activation | +8.15e-03 |
Pos logit contributions | |
hon | +0.010 |
notice | +0.009 |
rep | +0.009 |
hon | +0.009 |
cal | +0.009 |
Neg logit contributions | |
possession | -0.013 |
favourite | -0.013 |
favorite | -0.012 |
purse | -0.012 |
wrist | -0.011 |
Token | One |
Token ID | 3198 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
possession | +0.031 |
favourite | +0.031 |
favorite | +0.027 |
purse | +0.026 |
robe | +0.026 |
Neg logit contributions | |
hon | -0.040 |
announcements | -0.036 |
notice | -0.035 |
hon | -0.035 |
louder | -0.035 |
Token | day |
Token ID | 1110 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
possession | +0.092 |
favourite | +0.091 |
favorite | +0.083 |
purse | +0.073 |
wrist | +0.066 |
Neg logit contributions | |
hon | -0.127 |
announcements | -0.112 |
notice | -0.110 |
rep | -0.109 |
cal | -0.109 |
Token | , |
Token ID | 11 |
Feature activation | -5.84e-02 |
Pos logit contributions | |
favourite | +0.044 |
possession | +0.041 |
favorite | +0.040 |
purse | +0.038 |
wallet | +0.034 |
Neg logit contributions | |
hon | -0.038 |
cal | -0.034 |
announcements | -0.033 |
hon | -0.032 |
explode | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
possession | +0.129 |
favourite | +0.122 |
favorite | +0.114 |
purse | +0.112 |
wrist | +0.111 |
Neg logit contributions | |
hon | -0.076 |
notice | -0.063 |
rep | -0.062 |
hon | -0.061 |
louder | -0.058 |
Token | brave |
Token ID | 14802 |
Feature activation | +7.94e-02 |
Pos logit contributions | |
hon | +0.032 |
notice | +0.027 |
overcome | +0.026 |
rep | +0.026 |
announcements | +0.026 |
Neg logit contributions | |
possession | -0.033 |
favourite | -0.032 |
favorite | -0.030 |
purse | -0.029 |
wallet | -0.027 |
Token | pirate |
Token ID | 25868 |
Feature activation | +2.65e-03 |
Pos logit contributions | |
hon | +0.152 |
overcome | +0.146 |
notice | +0.136 |
collapse | +0.134 |
announcements | +0.134 |
Neg logit contributions | |
possession | -0.125 |
favourite | -0.117 |
favorite | -0.113 |
purse | -0.110 |
wallet | -0.100 |
Token | went |
Token ID | 1816 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
hon | +0.005 |
cal | +0.005 |
louder | +0.005 |
announcements | +0.005 |
notice | +0.005 |
Neg logit contributions | |
possession | -0.004 |
favourite | -0.004 |
favorite | -0.004 |
purse | -0.003 |
wallet | -0.003 |
Token | on |
Token ID | 319 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
hon | +0.075 |
notice | +0.068 |
cal | +0.067 |
announcements | +0.067 |
rep | +0.067 |
Neg logit contributions | |
possession | -0.043 |
favourite | -0.042 |
favorite | -0.039 |
purse | -0.033 |
brand | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +7.28e-03 |
Pos logit contributions | |
possession | +0.074 |
favourite | +0.072 |
favorite | +0.067 |
purse | +0.062 |
brand | +0.059 |
Neg logit contributions | |
hon | -0.068 |
notice | -0.057 |
cal | -0.057 |
oblig | -0.056 |
louder | -0.055 |
Token | journey |
Token ID | 7002 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
hon | +0.013 |
rep | +0.011 |
cal | +0.011 |
notice | +0.011 |
oblig | +0.011 |
Neg logit contributions | |
possession | -0.013 |
favourite | -0.013 |
favorite | -0.012 |
purse | -0.011 |
brand | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
hon | +0.085 |
cal | +0.076 |
announcements | +0.074 |
rep | +0.072 |
slit | +0.072 |
Neg logit contributions | |
possession | -0.080 |
favourite | -0.073 |
favorite | -0.068 |
purse | -0.060 |
wrist | -0.058 |
Token | they |
Token ID | 484 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
possession | +0.030 |
favourite | +0.029 |
purse | +0.027 |
favorite | +0.027 |
wrist | +0.026 |
Neg logit contributions | |
hon | -0.014 |
hon | -0.012 |
rep | -0.012 |
notice | -0.012 |
louder | -0.011 |
Token | both |
Token ID | 1111 |
Feature activation | -6.67e-03 |
Pos logit contributions | |
hon | +0.094 |
notice | +0.082 |
hon | +0.079 |
rep | +0.078 |
overcome | +0.078 |
Neg logit contributions | |
possession | -0.139 |
favourite | -0.132 |
favorite | -0.121 |
purse | -0.121 |
wrist | -0.117 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
possession | +0.009 |
favourite | +0.009 |
favorite | +0.008 |
purse | +0.007 |
attered | +0.006 |
Neg logit contributions | |
hon | -0.015 |
rep | -0.013 |
cal | -0.013 |
louder | -0.013 |
announcements | -0.013 |
Token | animals |
Token ID | 4695 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
possession | +0.090 |
favourite | +0.086 |
favorite | +0.080 |
purse | +0.073 |
wrist | +0.069 |
Neg logit contributions | |
hon | -0.108 |
notice | -0.094 |
cal | -0.093 |
oblig | -0.093 |
rep | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
hon | +0.040 |
notice | +0.033 |
rep | +0.033 |
oblig | +0.032 |
cal | +0.032 |
Neg logit contributions | |
possession | -0.055 |
favourite | -0.053 |
favorite | -0.051 |
purse | -0.048 |
wrist | -0.045 |
Token | One |
Token ID | 1881 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
possession | +0.035 |
favourite | +0.031 |
purse | +0.030 |
wrist | +0.029 |
favorite | +0.029 |
Neg logit contributions | |
hon | -0.022 |
rep | -0.018 |
notice | -0.018 |
hon | -0.018 |
thunder | -0.017 |
Token | day |
Token ID | 1110 |
Feature activation | +6.38e-03 |
Pos logit contributions | |
possession | +0.060 |
favourite | +0.057 |
favorite | +0.052 |
purse | +0.051 |
wrist | +0.049 |
Neg logit contributions | |
hon | -0.058 |
notice | -0.050 |
hon | -0.050 |
rep | -0.050 |
cal | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
hon | +0.007 |
notice | +0.006 |
rep | +0.006 |
cal | +0.006 |
hon | +0.006 |
Neg logit contributions | |
possession | -0.015 |
favourite | -0.015 |
favorite | -0.014 |
purse | -0.013 |
wrist | -0.013 |
Token | there |
Token ID | 612 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
possession | +0.035 |
favourite | +0.034 |
favorite | +0.032 |
purse | +0.031 |
wrist | +0.030 |
Neg logit contributions | |
hon | -0.017 |
notice | -0.014 |
rep | -0.014 |
cal | -0.014 |
hon | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
possession | +0.050 |
favourite | +0.046 |
purse | +0.043 |
wrist | +0.042 |
favorite | +0.042 |
Neg logit contributions | |
hon | -0.036 |
notice | -0.031 |
overcome | -0.031 |
hon | -0.030 |
hear | -0.030 |
Token | an |
Token ID | 281 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
possession | +0.020 |
favourite | +0.020 |
favorite | +0.018 |
purse | +0.017 |
wrist | +0.017 |
Neg logit contributions | |
hon | -0.015 |
notice | -0.013 |
rep | -0.013 |
cal | -0.013 |
hon | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
favourite | +0.038 |
possession | +0.036 |
favorite | +0.036 |
purse | +0.034 |
wallet | +0.032 |
Neg logit contributions | |
hon | -0.040 |
overcome | -0.033 |
announcements | -0.033 |
cal | -0.033 |
notice | -0.033 |
Token | pron |
Token ID | 31186 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
hon | +0.110 |
notice | +0.097 |
rep | +0.096 |
cal | +0.096 |
hon | +0.095 |
Neg logit contributions | |
possession | -0.080 |
favourite | -0.077 |
favorite | -0.070 |
purse | -0.067 |
wrist | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
possession | +0.067 |
favourite | +0.065 |
favorite | +0.061 |
purse | +0.059 |
wrist | +0.057 |
Neg logit contributions | |
hon | -0.049 |
notice | -0.041 |
cal | -0.041 |
rep | -0.040 |
overcome | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
possession | +0.027 |
favourite | +0.026 |
favorite | +0.024 |
purse | +0.023 |
wrist | +0.022 |
Neg logit contributions | |
hon | -0.035 |
notice | -0.031 |
rep | -0.030 |
hon | -0.030 |
cal | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
hon | +0.069 |
notice | +0.060 |
rep | +0.060 |
cal | +0.060 |
hon | +0.059 |
Neg logit contributions | |
possession | -0.067 |
favourite | -0.064 |
favorite | -0.059 |
purse | -0.057 |
wrist | -0.054 |
Token | " |
Token ID | 1 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
hon | +0.059 |
cal | +0.054 |
slit | +0.053 |
rep | +0.053 |
hon | +0.052 |
Neg logit contributions | |
possession | -0.050 |
favourite | -0.048 |
purse | -0.045 |
wrist | -0.044 |
favorite | -0.042 |
Token | I |
Token ID | 40 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
possession | +0.201 |
favourite | +0.197 |
favorite | +0.182 |
purse | +0.172 |
wrist | +0.166 |
Neg logit contributions | |
hon | -0.210 |
notice | -0.187 |
cal | -0.187 |
rep | -0.183 |
overcome | -0.182 |
Token | do |
Token ID | 466 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
hon | +0.056 |
louder | +0.050 |
thunder | +0.049 |
announcements | +0.049 |
cal | +0.049 |
Neg logit contributions | |
favourite | -0.067 |
possession | -0.065 |
favorite | -0.064 |
purse | -0.057 |
favorites | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -2.60e-03 |
Pos logit contributions | |
hon | +0.058 |
notice | +0.050 |
cal | +0.050 |
rep | +0.048 |
louder | +0.047 |
Neg logit contributions | |
possession | -0.066 |
favourite | -0.065 |
favorite | -0.060 |
purse | -0.056 |
wrist | -0.055 |
Token | I |
Token ID | 314 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
favourite | +0.005 |
possession | +0.004 |
favorite | +0.004 |
purse | +0.004 |
wrist | +0.004 |
Neg logit contributions | |
hon | -0.005 |
notice | -0.004 |
overcome | -0.004 |
oblig | -0.004 |
announcements | -0.004 |
Token | do |
Token ID | 466 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
hon | +0.071 |
announcements | +0.063 |
thunder | +0.062 |
louder | +0.061 |
cal | +0.061 |
Neg logit contributions | |
favourite | -0.077 |
possession | -0.073 |
favorite | -0.073 |
favorites | -0.065 |
favourites | -0.065 |
Token | was |
Token ID | 373 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
possession | +0.004 |
favourite | +0.004 |
favorite | +0.004 |
purse | +0.004 |
wrist | +0.003 |
Neg logit contributions | |
hon | -0.006 |
hon | -0.005 |
notice | -0.005 |
rep | -0.005 |
cal | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
possession | +0.099 |
favourite | +0.092 |
purse | +0.083 |
favorite | +0.083 |
wrist | +0.081 |
Neg logit contributions | |
hon | -0.089 |
hon | -0.078 |
notice | -0.076 |
rep | -0.076 |
cal | -0.075 |
Token | very |
Token ID | 845 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
hon | +0.020 |
hon | +0.019 |
notice | +0.018 |
sne | +0.017 |
rep | +0.017 |
Neg logit contributions | |
possession | -0.028 |
favourite | -0.026 |
attered | -0.024 |
purse | -0.024 |
wrist | -0.024 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
possession | +0.074 |
favourite | +0.070 |
favorite | +0.063 |
purse | +0.063 |
wrist | +0.061 |
Neg logit contributions | |
hon | -0.065 |
hon | -0.056 |
notice | -0.056 |
rep | -0.055 |
cal | -0.054 |
Token | seal |
Token ID | 13810 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
possession | +0.076 |
favourite | +0.073 |
favorite | +0.067 |
purse | +0.065 |
wrist | +0.062 |
Neg logit contributions | |
hon | -0.067 |
notice | -0.058 |
hon | -0.057 |
rep | -0.057 |
cal | -0.056 |
Token | who |
Token ID | 508 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
possession | +0.156 |
favourite | +0.151 |
purse | +0.141 |
favorite | +0.138 |
wrist | +0.137 |
Neg logit contributions | |
hon | -0.096 |
hon | -0.082 |
rep | -0.076 |
cal | -0.075 |
notice | -0.074 |
Token | loved |
Token ID | 6151 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
possession | +0.186 |
favourite | +0.175 |
purse | +0.164 |
wrist | +0.156 |
favorite | +0.153 |
Neg logit contributions | |
hon | -0.229 |
hon | -0.212 |
notice | -0.194 |
cal | -0.191 |
overcome | -0.191 |
Token | to |
Token ID | 284 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
possession | +0.140 |
favourite | +0.134 |
favorite | +0.121 |
purse | +0.121 |
wrist | +0.117 |
Neg logit contributions | |
hon | -0.144 |
hon | -0.127 |
rep | -0.124 |
notice | -0.124 |
cal | -0.121 |
Token | hop |
Token ID | 1725 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
hon | +0.081 |
cal | +0.071 |
louder | +0.070 |
announcements | +0.068 |
rep | +0.067 |
Neg logit contributions | |
favourite | -0.094 |
possession | -0.093 |
favorite | -0.088 |
purse | -0.081 |
wrist | -0.081 |
Token | around |
Token ID | 1088 |
Feature activation | -7.19e-02 |
Pos logit contributions | |
hon | +0.033 |
notice | +0.029 |
hon | +0.029 |
rep | +0.028 |
cal | +0.028 |
Neg logit contributions | |
possession | -0.024 |
favourite | -0.023 |
favorite | -0.020 |
purse | -0.020 |
wrist | -0.019 |
Token | on |
Token ID | 319 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
possession | +0.136 |
favourite | +0.133 |
favorite | +0.123 |
purse | +0.112 |
wrist | +0.109 |
Neg logit contributions | |
hon | -0.118 |
cal | -0.100 |
notice | -0.100 |
announcements | -0.099 |
rep | -0.097 |
Token | hug |
Token ID | 16225 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
hon | +0.053 |
notice | +0.044 |
rep | +0.042 |
hon | +0.042 |
overcome | +0.042 |
Neg logit contributions | |
possession | -0.083 |
favourite | -0.080 |
favorite | -0.074 |
purse | -0.073 |
wrist | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
possession | +0.040 |
favourite | +0.039 |
favorite | +0.036 |
purse | +0.035 |
wrist | +0.034 |
Neg logit contributions | |
hon | -0.024 |
notice | -0.020 |
rep | -0.020 |
hon | -0.019 |
cal | -0.019 |
Token | She |
Token ID | 1375 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
possession | +0.055 |
favourite | +0.051 |
favorite | +0.048 |
purse | +0.046 |
attered | +0.045 |
Neg logit contributions | |
hon | -0.042 |
rep | -0.037 |
hon | -0.035 |
notice | -0.034 |
cal | -0.034 |
Token | said |
Token ID | 531 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
possession | +0.101 |
favourite | +0.099 |
favorite | +0.093 |
purse | +0.088 |
wrist | +0.085 |
Neg logit contributions | |
hon | -0.074 |
notice | -0.062 |
rep | -0.061 |
hon | -0.060 |
cal | -0.060 |
Token | : |
Token ID | 25 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
favourite | +0.045 |
possession | +0.045 |
favorite | +0.043 |
purse | +0.041 |
brand | +0.038 |
Neg logit contributions | |
hon | -0.036 |
notice | -0.032 |
RO | -0.030 |
overcome | -0.030 |
announcements | -0.030 |
Token | " |
Token ID | 366 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
possession | +0.184 |
favourite | +0.169 |
favorite | +0.149 |
purse | +0.141 |
wrist | +0.129 |
Neg logit contributions | |
hon | -0.341 |
rep | -0.307 |
notice | -0.306 |
hon | -0.303 |
cal | -0.301 |
Token | It |
Token ID | 1026 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
possession | +0.205 |
favourite | +0.198 |
favorite | +0.176 |
purse | +0.168 |
wrist | +0.165 |
Neg logit contributions | |
hon | -0.269 |
cal | -0.240 |
notice | -0.240 |
hon | -0.239 |
overcome | -0.238 |
Token | 's |
Token ID | 338 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
favourite | +0.083 |
possession | +0.080 |
favorite | +0.078 |
robe | +0.066 |
purse | +0.065 |
Neg logit contributions | |
hon | -0.104 |
notice | -0.094 |
oblig | -0.089 |
announcements | -0.089 |
rep | -0.088 |
Token | all |
Token ID | 477 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
favourite | +0.135 |
possession | +0.131 |
favorite | +0.130 |
brand | +0.118 |
purse | +0.115 |
Neg logit contributions | |
hon | -0.088 |
oblig | -0.067 |
notice | -0.067 |
announcements | -0.066 |
slit | -0.064 |
Token | right |
Token ID | 826 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
possession | +0.064 |
favourite | +0.062 |
favorite | +0.060 |
brand | +0.053 |
purse | +0.050 |
Neg logit contributions | |
hon | -0.070 |
notice | -0.060 |
oblig | -0.059 |
slit | -0.059 |
announcements | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
hon | +0.120 |
notice | +0.100 |
hon | +0.098 |
rep | +0.097 |
cal | +0.097 |
Neg logit contributions | |
possession | -0.154 |
favourite | -0.147 |
favorite | -0.138 |
purse | -0.130 |
wrist | -0.124 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
possession | +0.103 |
favourite | +0.088 |
attered | +0.083 |
wrist | +0.080 |
favorite | +0.079 |
Neg logit contributions | |
hon | -0.111 |
rep | -0.104 |
notice | -0.101 |
overcome | -0.100 |
cal | -0.099 |
Token | my |
Token ID | 1820 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
favourite | +0.019 |
possession | +0.018 |
favorite | +0.017 |
purse | +0.017 |
wallet | +0.016 |
Neg logit contributions | |
hon | -0.017 |
notice | -0.014 |
announcements | -0.014 |
boom | -0.014 |
cal | -0.014 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
favourite | +0.022 |
possession | +0.022 |
favorite | +0.020 |
purse | +0.020 |
wallet | +0.018 |
Neg logit contributions | |
hon | -0.020 |
boom | -0.017 |
notice | -0.017 |
announcements | -0.017 |
louder | -0.016 |
Token | her |
Token ID | 607 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
possession | +0.221 |
favourite | +0.212 |
favorite | +0.195 |
purse | +0.188 |
wrist | +0.178 |
Neg logit contributions | |
hon | -0.214 |
notice | -0.186 |
rep | -0.183 |
hon | -0.182 |
cal | -0.181 |
Token | a |
Token ID | 257 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
possession | +0.319 |
favourite | +0.302 |
favorite | +0.279 |
attered | +0.279 |
wrist | +0.273 |
Neg logit contributions | |
hon | -0.214 |
rep | -0.188 |
notice | -0.186 |
cal | -0.186 |
hon | -0.181 |
Token | pair |
Token ID | 5166 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
possession | +0.077 |
favourite | +0.071 |
attered | +0.068 |
favorite | +0.065 |
wrist | +0.064 |
Neg logit contributions | |
hon | -0.070 |
rep | -0.065 |
hon | -0.065 |
notice | -0.064 |
cal | -0.063 |
Token | of |
Token ID | 286 |
Feature activation | -5.74e-03 |
Pos logit contributions | |
favourite | +0.350 |
possession | +0.347 |
favorite | +0.333 |
favorites | +0.330 |
favourites | +0.319 |
Neg logit contributions | |
hon | -0.117 |
notice | -0.095 |
hear | -0.090 |
hon | -0.086 |
rep | -0.085 |
Token | sunglasses |
Token ID | 36251 |
Feature activation | -7.57e-02 |
Pos logit contributions | |
possession | +0.009 |
favourite | +0.008 |
favorite | +0.007 |
purse | +0.007 |
wrist | +0.007 |
Neg logit contributions | |
hon | -0.011 |
notice | -0.010 |
rep | -0.010 |
hon | -0.010 |
cal | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
possession | +0.156 |
favourite | +0.148 |
favorite | +0.137 |
purse | +0.131 |
wrist | +0.128 |
Neg logit contributions | |
hon | -0.108 |
rep | -0.094 |
notice | -0.092 |
hon | -0.091 |
overcome | -0.088 |
Token | They |
Token ID | 1119 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
possession | +0.062 |
favourite | +0.057 |
favorite | +0.052 |
purse | +0.049 |
wrist | +0.048 |
Neg logit contributions | |
hon | -0.063 |
notice | -0.056 |
rep | -0.056 |
hon | -0.054 |
louder | -0.053 |
Token | were |
Token ID | 547 |
Feature activation | -1.19e-03 |
Pos logit contributions | |
hon | +0.060 |
notice | +0.053 |
rep | +0.052 |
hon | +0.051 |
overcome | +0.048 |
Neg logit contributions | |
possession | -0.107 |
favourite | -0.098 |
favorite | -0.092 |
purse | -0.091 |
wrist | -0.089 |
Token | L |
Token ID | 406 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
possession | +0.099 |
favourite | +0.095 |
favorite | +0.088 |
purse | +0.086 |
wrist | +0.082 |
Neg logit contributions | |
hon | -0.092 |
notice | -0.081 |
hon | -0.077 |
announcements | -0.077 |
cal | -0.077 |
Token | ulu |
Token ID | 15712 |
Feature activation | -5.33e-03 |
Pos logit contributions | |
hon | +0.071 |
notice | +0.059 |
rep | +0.059 |
cal | +0.058 |
hon | +0.058 |
Neg logit contributions | |
possession | -0.080 |
favourite | -0.078 |
favorite | -0.072 |
purse | -0.068 |
wrist | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
possession | +0.009 |
favourite | +0.008 |
purse | +0.008 |
favorite | +0.008 |
wrist | +0.007 |
Neg logit contributions | |
hon | -0.010 |
announcements | -0.009 |
RO | -0.009 |
notice | -0.008 |
Hon | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
possession | +0.073 |
favourite | +0.067 |
favorite | +0.060 |
purse | +0.059 |
wrist | +0.056 |
Neg logit contributions | |
hon | -0.089 |
rep | -0.080 |
hon | -0.079 |
notice | -0.077 |
cal | -0.077 |
Token |
|
Token ID | 198 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
hon | +0.033 |
notice | +0.029 |
cal | +0.028 |
rep | +0.028 |
hon | +0.028 |
Neg logit contributions | |
possession | -0.035 |
favourite | -0.034 |
favorite | -0.031 |
purse | -0.029 |
wrist | -0.028 |
Token | " |
Token ID | 1 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
hon | +0.025 |
notice | +0.021 |
rep | +0.020 |
cal | +0.020 |
hon | +0.020 |
Neg logit contributions | |
possession | -0.034 |
favourite | -0.032 |
favorite | -0.030 |
purse | -0.029 |
wrist | -0.028 |
Token | My |
Token ID | 3666 |
Feature activation | -9.35e-02 |
Pos logit contributions | |
possession | +0.206 |
favourite | +0.199 |
favorite | +0.180 |
purse | +0.173 |
wrist | +0.165 |
Neg logit contributions | |
hon | -0.249 |
notice | -0.220 |
cal | -0.216 |
hon | -0.215 |
rep | -0.214 |
Token | tail |
Token ID | 7894 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
possession | +0.153 |
favourite | +0.128 |
attered | +0.121 |
purse | +0.119 |
artments | +0.114 |
Neg logit contributions | |
hon | -0.161 |
sne | -0.151 |
notice | -0.147 |
hear | -0.147 |
louder | -0.147 |
Token | is |
Token ID | 318 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
possession | +0.114 |
favourite | +0.110 |
favorite | +0.103 |
purse | +0.098 |
wrist | +0.094 |
Neg logit contributions | |
hon | -0.106 |
notice | -0.090 |
rep | -0.089 |
oblig | -0.088 |
announcements | -0.088 |
Token | stuck |
Token ID | 7819 |
Feature activation | -5.45e-03 |
Pos logit contributions | |
possession | +0.065 |
favourite | +0.063 |
favorite | +0.058 |
purse | +0.055 |
wrist | +0.052 |
Neg logit contributions | |
hon | -0.062 |
notice | -0.053 |
rep | -0.052 |
hon | -0.051 |
cal | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
possession | +0.010 |
favourite | +0.009 |
favorite | +0.009 |
purse | +0.008 |
attered | +0.008 |
Neg logit contributions | |
hon | -0.009 |
announcements | -0.008 |
oblig | -0.008 |
cal | -0.008 |
notice | -0.008 |