Token | US |
Token ID | 1294 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
horizont | +0.237 |
MK | +0.209 |
Dash | +0.183 |
HIP | +0.182 |
MK | +0.182 |
Neg logit contributions | |
McCartney | -0.164 |
aba | -0.161 |
Reloaded | -0.159 |
igible | -0.157 |
complement | -0.155 |
Token | economy |
Token ID | 3773 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
horizont | +0.164 |
MK | +0.146 |
Dash | +0.129 |
HIP | +0.128 |
MK | +0.128 |
Neg logit contributions | |
McCartney | -0.098 |
aba | -0.096 |
Reloaded | -0.094 |
igible | -0.093 |
complement | -0.092 |
Token | , |
Token ID | 11 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
horizont | +0.060 |
MK | +0.052 |
Dash | +0.044 |
MK | +0.043 |
HIP | +0.043 |
Neg logit contributions | |
McCartney | -0.060 |
aba | -0.059 |
Reloaded | -0.058 |
igible | -0.058 |
irs | -0.057 |
Token | effects |
Token ID | 3048 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
horizont | +0.279 |
MK | +0.243 |
Dash | +0.210 |
HIP | +0.208 |
MK | +0.208 |
Neg logit contributions | |
McCartney | -0.239 |
aba | -0.234 |
Reloaded | -0.232 |
igible | -0.229 |
stadt | -0.226 |
Token | of |
Token ID | 286 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
horizont | +0.199 |
MK | +0.169 |
Dash | +0.142 |
HIP | +0.140 |
MK | +0.139 |
Neg logit contributions | |
McCartney | -0.231 |
aba | -0.227 |
Reloaded | -0.224 |
igible | -0.222 |
complement | -0.221 |
Token | sequest |
Token ID | 46314 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
horizont | +0.023 |
MK | +0.020 |
Dash | +0.017 |
HIP | +0.017 |
MK | +0.017 |
Neg logit contributions | |
McCartney | -0.017 |
aba | -0.017 |
Reloaded | -0.017 |
igible | -0.016 |
complement | -0.016 |
Token | ration |
Token ID | 1358 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
McCartney | +0.237 |
aba | +0.235 |
irs | +0.182 |
Reloaded | +0.177 |
igible | +0.165 |
Neg logit contributions | |
horizont | -3.772 |
MK | -3.462 |
Dash | -3.219 |
HIP | -3.203 |
MK | -3.178 |
Token | and |
Token ID | 290 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
McCartney | +0.327 |
aba | +0.322 |
Reloaded | +0.316 |
igible | +0.313 |
complement | +0.313 |
Neg logit contributions | |
horizont | -0.342 |
MK | -0.296 |
Dash | -0.253 |
HIP | -0.249 |
MK | -0.248 |
Token | the |
Token ID | 262 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
horizont | +0.310 |
MK | +0.273 |
Dash | +0.239 |
HIP | +0.237 |
MK | +0.237 |
Neg logit contributions | |
McCartney | -0.219 |
aba | -0.214 |
Reloaded | -0.211 |
igible | -0.208 |
irs | -0.206 |
Token | debt |
Token ID | 5057 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
horizont | +0.244 |
MK | +0.214 |
Dash | +0.187 |
HIP | +0.185 |
MK | +0.185 |
Neg logit contributions | |
McCartney | -0.181 |
aba | -0.177 |
Reloaded | -0.175 |
igible | -0.173 |
complement | -0.171 |
Token | ceiling |
Token ID | 13387 |
Feature activation | -9.22e-01 |
Pos logit contributions | |
horizont | +0.475 |
MK | +0.398 |
Dash | +0.327 |
HIP | +0.321 |
MK | +0.319 |
Neg logit contributions | |
McCartney | -0.637 |
aba | -0.628 |
Reloaded | -0.620 |
igible | -0.615 |
complement | -0.613 |
Token | , |
Token ID | 11 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
horizont | +0.024 |
MK | +0.021 |
Dash | +0.018 |
HIP | +0.018 |
MK | +0.018 |
Neg logit contributions | |
McCartney | -0.023 |
aba | -0.023 |
Reloaded | -0.022 |
igible | -0.022 |
irs | -0.022 |
Token | contained |
Token ID | 7763 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
McCartney | +0.258 |
aba | +0.253 |
Reloaded | +0.248 |
igible | +0.246 |
complement | +0.246 |
Neg logit contributions | |
horizont | -0.309 |
MK | -0.270 |
Dash | -0.234 |
HIP | -0.231 |
MK | -0.230 |
Token | an |
Token ID | 281 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
McCartney | +0.253 |
aba | +0.248 |
Reloaded | +0.243 |
igible | +0.240 |
complement | +0.239 |
Neg logit contributions | |
horizont | -0.360 |
MK | -0.318 |
Dash | -0.279 |
HIP | -0.275 |
MK | -0.274 |
Token | iron |
Token ID | 6953 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
McCartney | +0.023 |
aba | +0.022 |
Reloaded | +0.022 |
igible | +0.022 |
complement | +0.022 |
Neg logit contributions | |
horizont | -0.032 |
MK | -0.028 |
Dash | -0.025 |
HIP | -0.024 |
MK | -0.024 |
Token | rod |
Token ID | 15299 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
horizont | +0.545 |
MK | +0.482 |
Dash | +0.422 |
HIP | +0.419 |
MK | +0.418 |
Neg logit contributions | |
McCartney | -0.379 |
aba | -0.370 |
Reloaded | -0.365 |
igible | -0.359 |
irs | -0.357 |
Token | enc |
Token ID | 2207 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
McCartney | +0.487 |
aba | +0.475 |
Reloaded | +0.472 |
igible | +0.463 |
irs | +0.463 |
Neg logit contributions | |
horizont | -0.580 |
MK | -0.506 |
Dash | -0.437 |
HIP | -0.433 |
MK | -0.433 |
Token | ased |
Token ID | 839 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
McCartney | +0.894 |
aba | +0.880 |
irs | +0.835 |
igible | +0.825 |
Reloaded | +0.819 |
Neg logit contributions | |
horizont | -3.034 |
MK | -2.752 |
Dash | -2.501 |
HIP | -2.476 |
MK | -2.468 |
Token | in |
Token ID | 287 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
McCartney | +0.645 |
aba | +0.634 |
complement | +0.622 |
Reloaded | +0.622 |
igible | +0.621 |
Neg logit contributions | |
horizont | -0.642 |
MK | -0.558 |
Dash | -0.477 |
HIP | -0.464 |
MK | -0.461 |
Token | copper |
Token ID | 15317 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
McCartney | +0.166 |
aba | +0.162 |
Reloaded | +0.161 |
igible | +0.158 |
irs | +0.157 |
Neg logit contributions | |
horizont | -0.229 |
MK | -0.201 |
Dash | -0.176 |
HIP | -0.175 |
MK | -0.174 |
Token | and |
Token ID | 290 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
horizont | +0.091 |
MK | +0.079 |
Dash | +0.068 |
HIP | +0.067 |
MK | +0.067 |
Neg logit contributions | |
McCartney | -0.082 |
aba | -0.080 |
Reloaded | -0.079 |
igible | -0.078 |
irs | -0.078 |
Token | dated |
Token ID | 14567 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
McCartney | +0.179 |
aba | +0.175 |
Reloaded | +0.171 |
igible | +0.169 |
complement | +0.169 |
Neg logit contributions | |
horizont | -0.265 |
MK | -0.234 |
Dash | -0.206 |
HIP | -0.203 |
MK | -0.202 |
Token | prof |
Token ID | 1534 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
McCartney | +0.053 |
aba | +0.052 |
Reloaded | +0.051 |
igible | +0.051 |
irs | +0.050 |
Neg logit contributions | |
horizont | -0.074 |
MK | -0.065 |
Dash | -0.057 |
HIP | -0.057 |
MK | -0.057 |
Token | use |
Token ID | 1904 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
horizont | +0.033 |
MK | +0.029 |
Dash | +0.025 |
HIP | +0.025 |
MK | +0.024 |
Neg logit contributions | |
McCartney | -0.029 |
aba | -0.029 |
Reloaded | -0.028 |
igible | -0.028 |
complement | -0.028 |
Token | apologies |
Token ID | 32920 |
Feature activation | +9.46e-01 |
Pos logit contributions | |
horizont | +0.170 |
MK | +0.152 |
Dash | +0.135 |
HIP | +0.134 |
MK | +0.134 |
Neg logit contributions | |
McCartney | -0.090 |
aba | -0.088 |
Reloaded | -0.087 |
igible | -0.085 |
irs | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
McCartney | +0.630 |
aba | +0.619 |
Reloaded | +0.612 |
igible | +0.605 |
irs | +0.602 |
Neg logit contributions | |
horizont | -0.700 |
MK | -0.609 |
Dash | -0.524 |
HIP | -0.517 |
MK | -0.516 |
Token | the |
Token ID | 262 |
Feature activation | +7.74e-02 |
Pos logit contributions | |
horizont | +0.127 |
MK | +0.111 |
Dash | +0.097 |
HIP | +0.097 |
MK | +0.096 |
Neg logit contributions | |
McCartney | -0.093 |
aba | -0.091 |
Reloaded | -0.090 |
igible | -0.089 |
irs | -0.088 |
Token | Motion |
Token ID | 20843 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
McCartney | +0.045 |
aba | +0.044 |
Reloaded | +0.043 |
igible | +0.043 |
stadt | +0.042 |
Neg logit contributions | |
horizont | -0.064 |
MK | -0.056 |
Dash | -0.049 |
HIP | -0.049 |
MK | -0.049 |
Token | Picture |
Token ID | 17741 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
McCartney | +1.505 |
aba | +1.474 |
Reloaded | +1.436 |
igible | +1.426 |
irs | +1.412 |
Neg logit contributions | |
horizont | -2.266 |
MK | -1.999 |
Dash | -1.747 |
ivably | -1.738 |
HIP | -1.732 |
Token | Academy |
Token ID | 8581 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
McCartney | +0.087 |
aba | +0.085 |
Reloaded | +0.084 |
igible | +0.083 |
complement | +0.082 |
Neg logit contributions | |
horizont | -0.157 |
MK | -0.139 |
Dash | -0.124 |
HIP | -0.123 |
MK | -0.122 |
Token | and |
Token ID | 290 |
Feature activation | -6.62e-02 |
Pos logit contributions | |
horizont | +0.221 |
MK | +0.192 |
Dash | +0.165 |
HIP | +0.163 |
MK | +0.163 |
Neg logit contributions | |
McCartney | -0.199 |
aba | -0.196 |
Reloaded | -0.193 |
igible | -0.191 |
irs | -0.190 |
Token | to |
Token ID | 284 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
horizont | +0.054 |
MK | +0.047 |
Dash | +0.041 |
HIP | +0.041 |
MK | +0.041 |
Neg logit contributions | |
McCartney | -0.040 |
aba | -0.039 |
Reloaded | -0.038 |
igible | -0.038 |
complement | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
horizont | +0.059 |
MK | +0.052 |
Dash | +0.045 |
HIP | +0.045 |
MK | +0.045 |
Neg logit contributions | |
McCartney | -0.042 |
aba | -0.042 |
Reloaded | -0.041 |
igible | -0.040 |
irs | -0.040 |
Token | influence |
Token ID | 4588 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
McCartney | +0.182 |
aba | +0.179 |
Reloaded | +0.176 |
igible | +0.174 |
irs | +0.173 |
Neg logit contributions | |
horizont | -0.226 |
MK | -0.197 |
Dash | -0.171 |
HIP | -0.169 |
MK | -0.169 |
Token | and |
Token ID | 290 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
horizont | +0.282 |
MK | +0.243 |
Dash | +0.208 |
HIP | +0.205 |
MK | +0.204 |
Neg logit contributions | |
McCartney | -0.274 |
aba | -0.270 |
Reloaded | -0.266 |
igible | -0.263 |
irs | -0.262 |
Token | friendship |
Token ID | 14738 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
horizont | +0.110 |
MK | +0.096 |
Dash | +0.084 |
HIP | +0.083 |
MK | +0.082 |
Neg logit contributions | |
McCartney | -0.085 |
aba | -0.084 |
Reloaded | -0.082 |
igible | -0.081 |
complement | -0.081 |
Token | with |
Token ID | 351 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
horizont | +0.283 |
MK | +0.244 |
Dash | +0.208 |
HIP | +0.205 |
MK | +0.205 |
Neg logit contributions | |
McCartney | -0.278 |
aba | -0.273 |
Reloaded | -0.270 |
igible | -0.267 |
irs | -0.265 |
Token | Eleanor |
Token ID | 45566 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
McCartney | +0.013 |
aba | +0.013 |
Reloaded | +0.012 |
igible | +0.012 |
irs | +0.012 |
Neg logit contributions | |
horizont | -0.018 |
MK | -0.016 |
Dash | -0.014 |
HIP | -0.014 |
MK | -0.014 |
Token | Roosevelt |
Token ID | 19297 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
McCartney | +0.715 |
aba | +0.695 |
complement | +0.677 |
Reloaded | +0.675 |
igible | +0.669 |
Neg logit contributions | |
horizont | -0.937 |
MK | -0.819 |
Dash | -0.715 |
HIP | -0.704 |
MK | -0.699 |
Token | to |
Token ID | 284 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
McCartney | +1.736 |
aba | +1.693 |
complement | +1.657 |
Reloaded | +1.647 |
igible | +1.641 |
Neg logit contributions | |
horizont | -2.041 |
MK | -1.772 |
Dash | -1.535 |
HIP | -1.511 |
ivably | -1.496 |
Token | secure |
Token ID | 5713 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
horizont | +0.050 |
MK | +0.044 |
Dash | +0.038 |
HIP | +0.038 |
MK | +0.037 |
Neg logit contributions | |
McCartney | -0.040 |
aba | -0.040 |
Reloaded | -0.039 |
igible | -0.039 |
irs | -0.038 |
Token | luxury |
Token ID | 13064 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
McCartney | +0.060 |
aba | +0.059 |
Reloaded | +0.059 |
igible | +0.058 |
irs | +0.057 |
Neg logit contributions | |
horizont | -0.089 |
MK | -0.079 |
Dash | -0.069 |
HIP | -0.069 |
MK | -0.069 |
Token | travel |
Token ID | 3067 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
horizont | +0.325 |
MK | +0.287 |
Dash | +0.251 |
HIP | +0.248 |
MK | +0.247 |
Neg logit contributions | |
McCartney | -0.227 |
aba | -0.222 |
Reloaded | -0.218 |
igible | -0.215 |
complement | -0.215 |
Token | buses |
Token ID | 16893 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
McCartney | +0.169 |
aba | +0.166 |
Reloaded | +0.163 |
igible | +0.161 |
complement | +0.161 |
Neg logit contributions | |
horizont | -0.199 |
MK | -0.173 |
Dash | -0.150 |
HIP | -0.148 |
MK | -0.147 |
Token | commanded |
Token ID | 22419 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
horizont | +0.189 |
MK | +0.166 |
Dash | +0.144 |
MK | +0.143 |
HIP | +0.143 |
Neg logit contributions | |
McCartney | -0.142 |
aba | -0.139 |
Reloaded | -0.138 |
igible | -0.136 |
irs | -0.135 |
Token | the |
Token ID | 262 |
Feature activation | +2.41e-03 |
Pos logit contributions | |
McCartney | +0.415 |
aba | +0.408 |
Reloaded | +0.402 |
igible | +0.397 |
irs | +0.394 |
Neg logit contributions | |
horizont | -0.587 |
MK | -0.517 |
Dash | -0.451 |
MK | -0.450 |
HIP | -0.446 |
Token | troops |
Token ID | 6553 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
McCartney | +0.001 |
aba | +0.001 |
Reloaded | +0.001 |
igible | +0.001 |
irs | +0.001 |
Neg logit contributions | |
horizont | -0.002 |
MK | -0.002 |
Dash | -0.002 |
MK | -0.002 |
HIP | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
McCartney | +0.030 |
aba | +0.030 |
Reloaded | +0.030 |
igible | +0.029 |
irs | +0.029 |
Neg logit contributions | |
horizont | -0.031 |
MK | -0.027 |
Dash | -0.023 |
MK | -0.023 |
SOU | -0.023 |
Token | to |
Token ID | 1462 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
horizont | +0.017 |
MK | +0.015 |
Dash | +0.013 |
HIP | +0.013 |
MK | +0.013 |
Neg logit contributions | |
McCartney | -0.013 |
aba | -0.013 |
Reloaded | -0.013 |
igible | -0.012 |
complement | -0.012 |
Token | enc |
Token ID | 2207 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
McCartney | +0.237 |
aba | +0.233 |
Reloaded | +0.229 |
igible | +0.227 |
complement | +0.226 |
Neg logit contributions | |
horizont | -0.267 |
MK | -0.232 |
Dash | -0.200 |
HIP | -0.198 |
MK | -0.197 |
Token | irc |
Token ID | 1980 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
McCartney | +1.563 |
aba | +1.555 |
irs | +1.514 |
igible | +1.501 |
complement | +1.490 |
Neg logit contributions | |
horizont | -2.212 |
MK | -1.929 |
Dash | -1.674 |
HIP | -1.670 |
MK | -1.654 |
Token | le |
Token ID | 293 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
McCartney | +0.218 |
aba | +0.215 |
complement | +0.209 |
Reloaded | +0.209 |
irs | +0.208 |
Neg logit contributions | |
horizont | -0.223 |
MK | -0.189 |
Dash | -0.162 |
HIP | -0.159 |
MK | -0.158 |
Token | and |
Token ID | 290 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
horizont | +0.287 |
MK | +0.250 |
Dash | +0.217 |
HIP | +0.214 |
MK | +0.214 |
Neg logit contributions | |
McCartney | -0.235 |
aba | -0.230 |
Reloaded | -0.227 |
igible | -0.224 |
irs | -0.223 |
Token | destroy |
Token ID | 4117 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
horizont | +0.043 |
MK | +0.037 |
Dash | +0.032 |
HIP | +0.032 |
MK | +0.032 |
Neg logit contributions | |
McCartney | -0.038 |
aba | -0.037 |
Reloaded | -0.036 |
igible | -0.036 |
irs | -0.036 |
Token | [ |
Token ID | 685 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
McCartney | +0.911 |
aba | +0.893 |
Reloaded | +0.875 |
complement | +0.873 |
igible | +0.870 |
Neg logit contributions | |
horizont | -1.089 |
MK | -0.954 |
Dash | -0.830 |
HIP | -0.812 |
MK | -0.806 |
Token | that |
Token ID | 326 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
McCartney | +0.459 |
aba | +0.446 |
Reloaded | +0.434 |
complement | +0.434 |
igible | +0.433 |
Neg logit contributions | |
horizont | -0.625 |
MK | -0.550 |
Dash | -0.480 |
HIP | -0.470 |
MK | -0.469 |
Token | would |
Token ID | 561 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
McCartney | +0.099 |
aba | +0.097 |
Reloaded | +0.095 |
igible | +0.094 |
complement | +0.094 |
Neg logit contributions | |
horizont | -0.139 |
MK | -0.123 |
Dash | -0.107 |
HIP | -0.106 |
MK | -0.105 |
Token | naturally |
Token ID | 8752 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
McCartney | +0.129 |
aba | +0.126 |
complement | +0.124 |
igible | +0.122 |
irs | +0.122 |
Neg logit contributions | |
horizont | -0.157 |
MK | -0.137 |
Dash | -0.119 |
HIP | -0.117 |
MK | -0.116 |
Token | include |
Token ID | 2291 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
McCartney | +0.101 |
aba | +0.099 |
complement | +0.099 |
igible | +0.096 |
irs | +0.096 |
Neg logit contributions | |
horizont | -0.124 |
MK | -0.109 |
Dash | -0.095 |
HIP | -0.093 |
MK | -0.092 |
Token | B |
Token ID | 347 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
McCartney | +0.060 |
aba | +0.059 |
Reloaded | +0.058 |
igible | +0.057 |
complement | +0.057 |
Neg logit contributions | |
horizont | -0.090 |
MK | -0.080 |
Dash | -0.070 |
HIP | -0.069 |
MK | -0.069 |
Token | abs |
Token ID | 8937 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
McCartney | +0.210 |
aba | +0.204 |
Reloaded | +0.203 |
igible | +0.200 |
complement | +0.199 |
Neg logit contributions | |
horizont | -0.313 |
MK | -0.279 |
Dash | -0.245 |
HIP | -0.241 |
MK | -0.241 |
Token | , |
Token ID | 11 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
McCartney | +1.696 |
aba | +1.652 |
complement | +1.610 |
Reloaded | +1.599 |
irs | +1.585 |
Neg logit contributions | |
horizont | -1.998 |
MK | -1.731 |
Dash | -1.489 |
HIP | -1.480 |
ivably | -1.475 |
Token | right |
Token ID | 826 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
McCartney | +0.295 |
aba | +0.289 |
Reloaded | +0.283 |
igible | +0.280 |
complement | +0.278 |
Neg logit contributions | |
horizont | -0.431 |
MK | -0.380 |
Dash | -0.334 |
HIP | -0.330 |
MK | -0.329 |
Token | ? |
Token ID | 30 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
McCartney | +0.358 |
aba | +0.350 |
Reloaded | +0.343 |
igible | +0.340 |
complement | +0.338 |
Neg logit contributions | |
horizont | -0.532 |
MK | -0.471 |
Dash | -0.414 |
HIP | -0.409 |
MK | -0.408 |
Token | Not |
Token ID | 1892 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
McCartney | +0.270 |
aba | +0.264 |
Reloaded | +0.260 |
igible | +0.258 |
complement | +0.257 |
Neg logit contributions | |
horizont | -0.237 |
MK | -0.201 |
Dash | -0.169 |
HIP | -0.165 |
MK | -0.164 |
Token | so |
Token ID | 523 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
McCartney | +0.400 |
aba | +0.391 |
Reloaded | +0.383 |
complement | +0.380 |
igible | +0.380 |
Neg logit contributions | |
horizont | -0.460 |
MK | -0.399 |
Dash | -0.346 |
HIP | -0.338 |
MK | -0.337 |
Token | times |
Token ID | 1661 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
McCartney | +0.315 |
aba | +0.307 |
Reloaded | +0.299 |
complement | +0.299 |
igible | +0.297 |
Neg logit contributions | |
horizont | -0.439 |
MK | -0.387 |
Dash | -0.340 |
HIP | -0.334 |
ivably | -0.332 |
Token | for |
Token ID | 329 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
McCartney | +0.385 |
aba | +0.378 |
Reloaded | +0.373 |
igible | +0.369 |
irs | +0.367 |
Neg logit contributions | |
horizont | -0.449 |
MK | -0.392 |
Dash | -0.338 |
HIP | -0.334 |
MK | -0.334 |
Token | Monk |
Token ID | 28854 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
McCartney | +0.171 |
aba | +0.168 |
Reloaded | +0.165 |
igible | +0.163 |
irs | +0.162 |
Neg logit contributions | |
horizont | -0.233 |
MK | -0.206 |
Dash | -0.180 |
HIP | -0.178 |
MK | -0.178 |
Token | . |
Token ID | 13 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
McCartney | +0.896 |
aba | +0.878 |
Reloaded | +0.862 |
igible | +0.852 |
stadt | +0.851 |
Neg logit contributions | |
horizont | -1.002 |
MK | -0.863 |
Dash | -0.741 |
HIP | -0.732 |
MK | -0.726 |
Token | ( |
Token ID | 357 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
McCartney | +0.180 |
aba | +0.176 |
Reloaded | +0.173 |
igible | +0.171 |
complement | +0.170 |
Neg logit contributions | |
horizont | -0.190 |
MK | -0.162 |
Dash | -0.139 |
HIP | -0.136 |
ivably | -0.136 |
Token | Fun |
Token ID | 24629 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
McCartney | +0.026 |
aba | +0.026 |
Reloaded | +0.026 |
igible | +0.025 |
complement | +0.025 |
Neg logit contributions | |
horizont | -0.033 |
MK | -0.029 |
Dash | -0.025 |
HIP | -0.025 |
MK | -0.024 |
Token | fact |
Token ID | 1109 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
McCartney | +1.946 |
aba | +1.911 |
complement | +1.870 |
igible | +1.865 |
irs | +1.864 |
Neg logit contributions | |
horizont | -1.700 |
MK | -1.454 |
Dash | -1.213 |
HIP | -1.193 |
MK | -1.191 |
Token | : |
Token ID | 25 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
McCartney | +0.041 |
aba | +0.041 |
Reloaded | +0.040 |
igible | +0.040 |
complement | +0.040 |
Neg logit contributions | |
horizont | -0.041 |
MK | -0.035 |
Dash | -0.030 |
HIP | -0.030 |
MK | -0.030 |
Token | Gu |
Token ID | 1962 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
McCartney | +0.029 |
aba | +0.028 |
Reloaded | +0.028 |
igible | +0.027 |
complement | +0.027 |
Neg logit contributions | |
horizont | -0.037 |
MK | -0.033 |
Dash | -0.029 |
HIP | -0.028 |
MK | -0.028 |
Token | illa |
Token ID | 5049 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
horizont | +0.049 |
MK | +0.043 |
Dash | +0.037 |
HIP | +0.036 |
MK | +0.036 |
Neg logit contributions | |
McCartney | -0.044 |
aba | -0.043 |
Reloaded | -0.042 |
igible | -0.042 |
complement | -0.042 |
Token | ume |
Token ID | 2454 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
horizont | +0.701 |
MK | +0.586 |
Dash | +0.478 |
HIP | +0.469 |
MK | +0.467 |
Neg logit contributions | |
McCartney | -0.984 |
aba | -0.969 |
Reloaded | -0.958 |
igible | -0.949 |
complement | -0.946 |
Token | Genesis |
Token ID | 18993 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
horizont | +0.206 |
MK | +0.179 |
Dash | +0.154 |
HIP | +0.152 |
MK | +0.151 |
Neg logit contributions | |
McCartney | -0.181 |
aba | -0.178 |
Reloaded | -0.173 |
igible | -0.173 |
irs | -0.172 |
Token | , |
Token ID | 11 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
horizont | +0.247 |
MK | +0.215 |
Dash | +0.185 |
MK | +0.184 |
HIP | +0.183 |
Neg logit contributions | |
McCartney | -0.230 |
aba | -0.224 |
Reloaded | -0.223 |
igible | -0.220 |
complement | -0.219 |
Token | to |
Token ID | 284 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
McCartney | +0.116 |
aba | +0.114 |
Reloaded | +0.112 |
igible | +0.111 |
irs | +0.110 |
Neg logit contributions | |
horizont | -0.176 |
MK | -0.156 |
Dash | -0.137 |
HIP | -0.136 |
MK | -0.136 |
Token | face |
Token ID | 1986 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
McCartney | +0.203 |
aba | +0.200 |
Reloaded | +0.196 |
igible | +0.194 |
complement | +0.194 |
Neg logit contributions | |
horizont | -0.231 |
MK | -0.200 |
Dash | -0.173 |
HIP | -0.170 |
MK | -0.170 |
Token | Dark |
Token ID | 3801 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
McCartney | +0.699 |
aba | +0.683 |
Reloaded | +0.670 |
igible | +0.664 |
complement | +0.663 |
Neg logit contributions | |
horizont | -0.958 |
MK | -0.843 |
Dash | -0.738 |
HIP | -0.727 |
SOU | -0.721 |
Token | se |
Token ID | 325 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
McCartney | +0.048 |
aba | +0.048 |
Reloaded | +0.047 |
igible | +0.047 |
irs | +0.047 |
Neg logit contributions | |
horizont | -0.034 |
MK | -0.028 |
Dash | -0.023 |
HIP | -0.022 |
MK | -0.022 |
Token | id |
Token ID | 312 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
McCartney | +1.435 |
aba | +1.409 |
igible | +1.366 |
Reloaded | +1.365 |
irs | +1.360 |
Neg logit contributions | |
horizont | -2.213 |
MK | -1.950 |
Dash | -1.716 |
HIP | -1.699 |
MK | -1.688 |
Token | 's |
Token ID | 338 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
McCartney | +0.859 |
aba | +0.840 |
Reloaded | +0.829 |
igible | +0.822 |
complement | +0.816 |
Neg logit contributions | |
horizont | -0.984 |
MK | -0.857 |
Dash | -0.739 |
HIP | -0.730 |
MK | -0.729 |
Token | army |
Token ID | 5428 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
McCartney | +0.153 |
aba | +0.149 |
Reloaded | +0.147 |
igible | +0.145 |
irs | +0.144 |
Neg logit contributions | |
horizont | -0.232 |
MK | -0.205 |
Dash | -0.180 |
MK | -0.179 |
HIP | -0.179 |
Token | , |
Token ID | 11 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
horizont | +0.548 |
MK | +0.471 |
Dash | +0.398 |
MK | +0.395 |
HIP | +0.391 |
Neg logit contributions | |
McCartney | -0.592 |
aba | -0.579 |
Reloaded | -0.578 |
igible | -0.570 |
irs | -0.567 |
Token | and |
Token ID | 290 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
McCartney | +0.135 |
aba | +0.133 |
Reloaded | +0.131 |
igible | +0.129 |
irs | +0.129 |
Neg logit contributions | |
horizont | -0.163 |
MK | -0.142 |
Dash | -0.123 |
HIP | -0.122 |
MK | -0.121 |
Token | and |
Token ID | 290 |
Feature activation | -7.44e-02 |
Pos logit contributions | |
McCartney | +0.193 |
aba | +0.189 |
Reloaded | +0.187 |
igible | +0.185 |
irs | +0.184 |
Neg logit contributions | |
horizont | -0.196 |
MK | -0.169 |
Dash | -0.144 |
MK | -0.142 |
HIP | -0.142 |
Token | that |
Token ID | 326 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
horizont | +0.059 |
MK | +0.051 |
Dash | +0.045 |
HIP | +0.044 |
MK | +0.044 |
Neg logit contributions | |
McCartney | -0.046 |
aba | -0.045 |
Reloaded | -0.044 |
igible | -0.044 |
complement | -0.044 |
Token | he |
Token ID | 339 |
Feature activation | +6.11e-02 |
Pos logit contributions | |
horizont | +0.229 |
MK | +0.199 |
Dash | +0.171 |
HIP | +0.169 |
MK | +0.168 |
Neg logit contributions | |
McCartney | -0.212 |
aba | -0.208 |
Reloaded | -0.205 |
igible | -0.203 |
complement | -0.202 |
Token | will |
Token ID | 481 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
McCartney | +0.040 |
aba | +0.040 |
Reloaded | +0.039 |
igible | +0.039 |
complement | +0.039 |
Neg logit contributions | |
horizont | -0.046 |
MK | -0.040 |
Dash | -0.034 |
HIP | -0.034 |
MK | -0.034 |
Token | be |
Token ID | 307 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
horizont | +0.122 |
MK | +0.106 |
Dash | +0.092 |
HIP | +0.091 |
MK | +0.090 |
Neg logit contributions | |
McCartney | -0.108 |
aba | -0.106 |
Reloaded | -0.104 |
igible | -0.103 |
complement | -0.103 |
Token | impe |
Token ID | 18516 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
horizont | +0.171 |
MK | +0.152 |
Dash | +0.134 |
HIP | +0.133 |
MK | +0.133 |
Neg logit contributions | |
McCartney | -0.108 |
aba | -0.106 |
Reloaded | -0.104 |
igible | -0.103 |
irs | -0.102 |
Token | ached |
Token ID | 2317 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
McCartney | +0.762 |
aba | +0.760 |
igible | +0.707 |
irs | +0.701 |
Reloaded | +0.701 |
Neg logit contributions | |
horizont | -2.864 |
MK | -2.580 |
Dash | -2.366 |
HIP | -2.359 |
SOU | -2.327 |
Token | , |
Token ID | 11 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
McCartney | +0.277 |
aba | +0.272 |
Reloaded | +0.269 |
igible | +0.266 |
irs | +0.264 |
Neg logit contributions | |
horizont | -0.291 |
MK | -0.252 |
Dash | -0.215 |
HIP | -0.212 |
MK | -0.212 |
Token | � |
Token ID | 447 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
horizont | +0.216 |
MK | +0.185 |
Dash | +0.156 |
MK | +0.153 |
HIP | +0.153 |
Neg logit contributions | |
McCartney | -0.251 |
aba | -0.247 |
Reloaded | -0.245 |
irs | -0.242 |
igible | -0.242 |
Token | � |
Token ID | 251 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
horizont | +0.324 |
MK | +0.297 |
Dash | +0.271 |
MK | +0.269 |
HIP | +0.269 |
Neg logit contributions | |
McCartney | -0.093 |
aba | -0.089 |
Reloaded | -0.088 |
igible | -0.085 |
complement | -0.084 |
Token | she |
Token ID | 673 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
McCartney | +0.086 |
aba | +0.085 |
Reloaded | +0.083 |
igible | +0.082 |
irs | +0.082 |
Neg logit contributions | |
horizont | -0.113 |
MK | -0.099 |
Dash | -0.086 |
HIP | -0.085 |
MK | -0.085 |
Token | from |
Token ID | 422 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
McCartney | +0.871 |
aba | +0.857 |
Reloaded | +0.841 |
igible | +0.835 |
complement | +0.833 |
Neg logit contributions | |
horizont | -0.907 |
MK | -0.782 |
Dash | -0.670 |
HIP | -0.657 |
MK | -0.654 |
Token | the |
Token ID | 262 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
McCartney | +0.264 |
aba | +0.259 |
Reloaded | +0.255 |
igible | +0.252 |
complement | +0.251 |
Neg logit contributions | |
horizont | -0.338 |
MK | -0.296 |
Dash | -0.258 |
HIP | -0.255 |
MK | -0.254 |
Token | forces |
Token ID | 3386 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
McCartney | +0.324 |
aba | +0.318 |
Reloaded | +0.312 |
irs | +0.309 |
complement | +0.309 |
Neg logit contributions | |
horizont | -0.389 |
MK | -0.339 |
Dash | -0.294 |
HIP | -0.289 |
ivably | -0.287 |
Token | of |
Token ID | 286 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
horizont | +0.236 |
MK | +0.201 |
Dash | +0.168 |
HIP | +0.165 |
MK | +0.164 |
Neg logit contributions | |
McCartney | -0.283 |
aba | -0.278 |
Reloaded | -0.275 |
igible | -0.272 |
irs | -0.271 |
Token | Dark |
Token ID | 3801 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
McCartney | +0.223 |
aba | +0.219 |
Reloaded | +0.215 |
complement | +0.213 |
igible | +0.213 |
Neg logit contributions | |
horizont | -0.238 |
MK | -0.206 |
Dash | -0.177 |
HIP | -0.174 |
MK | -0.172 |
Token | se |
Token ID | 325 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
McCartney | +0.019 |
aba | +0.019 |
Reloaded | +0.018 |
igible | +0.018 |
stadt | +0.018 |
Neg logit contributions | |
horizont | -0.014 |
MK | -0.011 |
Dash | -0.009 |
HIP | -0.009 |
ivably | -0.009 |
Token | id |
Token ID | 312 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
McCartney | +1.357 |
aba | +1.327 |
irs | +1.293 |
igible | +1.276 |
stadt | +1.261 |
Neg logit contributions | |
horizont | -2.278 |
MK | -2.009 |
Dash | -1.778 |
HIP | -1.755 |
ledge | -1.734 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
McCartney | +0.962 |
aba | +0.946 |
Reloaded | +0.930 |
igible | +0.921 |
complement | +0.917 |
Neg logit contributions | |
horizont | -1.070 |
MK | -0.928 |
Dash | -0.799 |
HIP | -0.788 |
MK | -0.784 |
Token |
|
Token ID | 198 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
McCartney | +0.148 |
aba | +0.146 |
Reloaded | +0.143 |
igible | +0.142 |
complement | +0.141 |
Neg logit contributions | |
horizont | -0.134 |
MK | -0.114 |
Dash | -0.096 |
HIP | -0.094 |
MK | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
McCartney | +0.245 |
aba | +0.241 |
Reloaded | +0.238 |
igible | +0.236 |
complement | +0.235 |
Neg logit contributions | |
horizont | -0.177 |
MK | -0.147 |
Dash | -0.120 |
HIP | -0.118 |
MK | -0.118 |
Token | The |
Token ID | 464 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
McCartney | +0.087 |
aba | +0.085 |
Reloaded | +0.083 |
igible | +0.083 |
complement | +0.083 |
Neg logit contributions | |
horizont | -0.103 |
MK | -0.092 |
Dash | -0.079 |
MK | -0.077 |
HIP | -0.077 |
Token | which |
Token ID | 543 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
horizont | +0.046 |
MK | +0.040 |
Dash | +0.035 |
HIP | +0.034 |
MK | +0.034 |
Neg logit contributions | |
McCartney | -0.038 |
aba | -0.038 |
Reloaded | -0.037 |
igible | -0.037 |
irs | -0.037 |
Token | were |
Token ID | 547 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
horizont | +0.234 |
MK | +0.205 |
Dash | +0.178 |
HIP | +0.176 |
MK | +0.175 |
Neg logit contributions | |
McCartney | -0.187 |
aba | -0.184 |
Reloaded | -0.181 |
igible | -0.179 |
irs | -0.178 |
Token | considered |
Token ID | 3177 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
horizont | +0.143 |
MK | +0.125 |
Dash | +0.109 |
HIP | +0.107 |
MK | +0.107 |
Neg logit contributions | |
McCartney | -0.111 |
aba | -0.109 |
Reloaded | -0.107 |
igible | -0.106 |
complement | -0.105 |
Token | likely |
Token ID | 1884 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
horizont | +0.031 |
MK | +0.027 |
Dash | +0.023 |
HIP | +0.023 |
MK | +0.023 |
Neg logit contributions | |
McCartney | -0.024 |
aba | -0.024 |
Reloaded | -0.024 |
igible | -0.023 |
irs | -0.023 |
Token | Apollo |
Token ID | 17508 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
horizont | +0.040 |
MK | +0.035 |
Dash | +0.030 |
HIP | +0.030 |
MK | +0.030 |
Neg logit contributions | |
McCartney | -0.034 |
aba | -0.034 |
Reloaded | -0.033 |
igible | -0.033 |
irs | -0.033 |
Token | landing |
Token ID | 9581 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
horizont | +0.974 |
MK | +0.851 |
MK | +0.742 |
HIP | +0.740 |
Dash | +0.735 |
Neg logit contributions | |
McCartney | -0.769 |
aba | -0.753 |
Reloaded | -0.752 |
irs | -0.738 |
igible | -0.735 |
Token | sites |
Token ID | 5043 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
McCartney | +1.238 |
aba | +1.212 |
Reloaded | +1.175 |
igible | +1.170 |
complement | +1.169 |
Neg logit contributions | |
horizont | -2.327 |
MK | -2.086 |
Dash | -1.861 |
HIP | -1.829 |
MK | -1.824 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
McCartney | +0.055 |
aba | +0.054 |
Reloaded | +0.054 |
igible | +0.053 |
irs | +0.053 |
Neg logit contributions | |
horizont | -0.058 |
MK | -0.050 |
Dash | -0.042 |
HIP | -0.042 |
MK | -0.042 |
Token | On |
Token ID | 1550 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
horizont | +0.133 |
MK | +0.117 |
Dash | +0.102 |
HIP | +0.101 |
MK | +0.101 |
Neg logit contributions | |
McCartney | -0.104 |
aba | -0.102 |
Reloaded | -0.100 |
igible | -0.099 |
irs | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
McCartney | +0.173 |
aba | +0.170 |
Reloaded | +0.168 |
irs | +0.165 |
igible | +0.164 |
Neg logit contributions | |
horizont | -0.223 |
MK | -0.196 |
Dash | -0.170 |
MK | -0.170 |
HIP | -0.169 |
Token | first |
Token ID | 717 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
McCartney | +0.052 |
aba | +0.052 |
Reloaded | +0.051 |
igible | +0.050 |
irs | +0.050 |
Neg logit contributions | |
horizont | -0.062 |
MK | -0.054 |
Dash | -0.047 |
HIP | -0.046 |
MK | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
horizont | +0.336 |
MK | +0.290 |
Dash | +0.249 |
HIP | +0.245 |
MK | +0.244 |
Neg logit contributions | |
McCartney | -0.318 |
aba | -0.312 |
Reloaded | -0.307 |
igible | -0.304 |
complement | -0.303 |
Token | it |
Token ID | 340 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
McCartney | +0.083 |
aba | +0.081 |
Reloaded | +0.080 |
igible | +0.079 |
irs | +0.078 |
Neg logit contributions | |
horizont | -0.096 |
MK | -0.083 |
Dash | -0.072 |
HIP | -0.071 |
MK | -0.071 |
Token | is |
Token ID | 318 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
McCartney | +0.147 |
aba | +0.145 |
Reloaded | +0.144 |
igible | +0.142 |
irs | +0.140 |
Neg logit contributions | |
horizont | -0.158 |
MK | -0.137 |
Dash | -0.117 |
MK | -0.116 |
HIP | -0.116 |
Token | a |
Token ID | 257 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
McCartney | +0.272 |
aba | +0.266 |
Reloaded | +0.262 |
igible | +0.259 |
complement | +0.259 |
Neg logit contributions | |
horizont | -0.316 |
MK | -0.275 |
Dash | -0.238 |
HIP | -0.235 |
MK | -0.234 |
Token | bl |
Token ID | 698 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
McCartney | +0.187 |
aba | +0.183 |
Reloaded | +0.180 |
complement | +0.179 |
igible | +0.178 |
Neg logit contributions | |
horizont | -0.224 |
MK | -0.195 |
Dash | -0.170 |
HIP | -0.166 |
MK | -0.166 |
Token | under |
Token ID | 4625 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
horizont | +0.182 |
MK | +0.153 |
Dash | +0.127 |
HIP | +0.124 |
MK | +0.124 |
Neg logit contributions | |
McCartney | -0.231 |
aba | -0.227 |
Reloaded | -0.223 |
igible | -0.222 |
irs | -0.221 |
Token | . |
Token ID | 13 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
McCartney | +1.655 |
aba | +1.633 |
Reloaded | +1.596 |
complement | +1.589 |
igible | +1.583 |
Neg logit contributions | |
horizont | -1.916 |
MK | -1.662 |
Dash | -1.437 |
HIP | -1.420 |
ivably | -1.408 |
Token | If |
Token ID | 1002 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
McCartney | +0.301 |
aba | +0.296 |
Reloaded | +0.291 |
igible | +0.289 |
complement | +0.288 |
Neg logit contributions | |
horizont | -0.354 |
MK | -0.310 |
Dash | -0.268 |
HIP | -0.265 |
MK | -0.264 |
Token | policymakers |
Token ID | 29484 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
horizont | +0.051 |
MK | +0.045 |
Dash | +0.039 |
HIP | +0.039 |
MK | +0.039 |
Neg logit contributions | |
McCartney | -0.038 |
aba | -0.038 |
Reloaded | -0.037 |
igible | -0.037 |
irs | -0.037 |
Token | listened |
Token ID | 16399 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
McCartney | +0.316 |
aba | +0.310 |
Reloaded | +0.303 |
complement | +0.302 |
igible | +0.301 |
Neg logit contributions | |
horizont | -0.372 |
MK | -0.324 |
Dash | -0.281 |
HIP | -0.276 |
MK | -0.275 |
Token | to |
Token ID | 284 |
Feature activation | +7.86e-02 |
Pos logit contributions | |
McCartney | +0.605 |
aba | +0.594 |
Reloaded | +0.584 |
igible | +0.578 |
complement | +0.576 |
Neg logit contributions | |
horizont | -0.698 |
MK | -0.608 |
Dash | -0.525 |
HIP | -0.517 |
MK | -0.515 |
Token | during |
Token ID | 1141 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
McCartney | +0.365 |
aba | +0.360 |
Reloaded | +0.356 |
igible | +0.353 |
irs | +0.350 |
Neg logit contributions | |
horizont | -0.389 |
MK | -0.337 |
Dash | -0.288 |
MK | -0.285 |
HIP | -0.285 |
Token | Van |
Token ID | 6656 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
McCartney | +0.437 |
aba | +0.429 |
Reloaded | +0.422 |
igible | +0.418 |
irs | +0.415 |
Neg logit contributions | |
horizont | -0.628 |
MK | -0.556 |
Dash | -0.487 |
HIP | -0.483 |
MK | -0.483 |
Token | E |
Token ID | 412 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
horizont | +0.193 |
MK | +0.166 |
Dash | +0.141 |
HIP | +0.140 |
MK | +0.138 |
Neg logit contributions | |
McCartney | -0.187 |
aba | -0.183 |
Reloaded | -0.178 |
irs | -0.178 |
igible | -0.178 |
Token | tten |
Token ID | 32407 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
McCartney | +0.318 |
aba | +0.311 |
Reloaded | +0.307 |
igible | +0.302 |
complement | +0.301 |
Neg logit contributions | |
horizont | -0.443 |
MK | -0.392 |
Dash | -0.341 |
MK | -0.338 |
HIP | -0.337 |
Token | 's |
Token ID | 338 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
McCartney | +0.178 |
aba | +0.174 |
Reloaded | +0.171 |
igible | +0.169 |
complement | +0.168 |
Neg logit contributions | |
horizont | -0.251 |
MK | -0.221 |
Dash | -0.194 |
HIP | -0.192 |
MK | -0.191 |
Token | enc |
Token ID | 2207 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
McCartney | +0.048 |
aba | +0.047 |
Reloaded | +0.046 |
igible | +0.046 |
irs | +0.046 |
Neg logit contributions | |
horizont | -0.064 |
MK | -0.057 |
Dash | -0.049 |
HIP | -0.049 |
MK | -0.049 |
Token | ore |
Token ID | 382 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
McCartney | +1.672 |
aba | +1.648 |
Reloaded | +1.605 |
igible | +1.600 |
irs | +1.599 |
Neg logit contributions | |
horizont | -1.876 |
MK | -1.624 |
Dash | -1.393 |
HIP | -1.378 |
MK | -1.369 |
Token | in |
Token ID | 287 |
Feature activation | +6.99e-02 |
Pos logit contributions | |
McCartney | +0.098 |
aba | +0.097 |
Reloaded | +0.095 |
igible | +0.094 |
irs | +0.094 |
Neg logit contributions | |
horizont | -0.110 |
MK | -0.096 |
Dash | -0.083 |
HIP | -0.082 |
MK | -0.081 |
Token | Minneapolis |
Token ID | 20455 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
McCartney | +0.042 |
aba | +0.041 |
Reloaded | +0.040 |
igible | +0.040 |
irs | +0.040 |
Neg logit contributions | |
horizont | -0.057 |
MK | -0.050 |
Dash | -0.044 |
HIP | -0.043 |
MK | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
McCartney | +0.495 |
aba | +0.488 |
Reloaded | +0.484 |
igible | +0.480 |
irs | +0.478 |
Neg logit contributions | |
horizont | -0.494 |
MK | -0.426 |
Dash | -0.363 |
MK | -0.360 |
HIP | -0.357 |
Token |
|
Token ID | 198 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
horizont | +0.121 |
MK | +0.103 |
Dash | +0.084 |
MK | +0.081 |
HIP | +0.081 |
Neg logit contributions | |
McCartney | -0.177 |
aba | -0.174 |
Reloaded | -0.172 |
igible | -0.171 |
complement | -0.171 |
Token | has |
Token ID | 468 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
McCartney | +0.192 |
aba | +0.188 |
Reloaded | +0.185 |
igible | +0.183 |
irs | +0.182 |
Neg logit contributions | |
horizont | -0.243 |
MK | -0.213 |
Dash | -0.185 |
HIP | -0.183 |
MK | -0.182 |
Token | constructed |
Token ID | 12006 |
Feature activation | +9.75e-01 |
Pos logit contributions | |
McCartney | +0.330 |
aba | +0.323 |
Reloaded | +0.316 |
complement | +0.314 |
igible | +0.314 |
Neg logit contributions | |
horizont | -0.408 |
MK | -0.358 |
Dash | -0.310 |
HIP | -0.306 |
MK | -0.305 |
Token | in |
Token ID | 287 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
McCartney | +0.667 |
aba | +0.656 |
Reloaded | +0.644 |
igible | +0.638 |
complement | +0.637 |
Neg logit contributions | |
horizont | -0.706 |
MK | -0.611 |
Dash | -0.524 |
HIP | -0.516 |
MK | -0.514 |
Token | order |
Token ID | 1502 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
McCartney | +0.319 |
aba | +0.312 |
Reloaded | +0.307 |
igible | +0.304 |
complement | +0.302 |
Neg logit contributions | |
horizont | -0.439 |
MK | -0.386 |
Dash | -0.338 |
HIP | -0.334 |
MK | -0.333 |
Token | to |
Token ID | 284 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
McCartney | +0.230 |
aba | +0.226 |
Reloaded | +0.222 |
igible | +0.220 |
complement | +0.219 |
Neg logit contributions | |
horizont | -0.262 |
MK | -0.228 |
Dash | -0.196 |
HIP | -0.194 |
MK | -0.193 |
Token | enc |
Token ID | 2207 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
McCartney | +0.205 |
aba | +0.202 |
Reloaded | +0.198 |
complement | +0.197 |
igible | +0.196 |
Neg logit contributions | |
horizont | -0.224 |
MK | -0.194 |
Dash | -0.167 |
HIP | -0.164 |
MK | -0.163 |
Token | ase |
Token ID | 589 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
aba | +1.491 |
McCartney | +1.460 |
irs | +1.434 |
igible | +1.418 |
complement | +1.402 |
Neg logit contributions | |
horizont | -2.099 |
MK | -1.811 |
HIP | -1.574 |
Dash | -1.564 |
MK | -1.547 |
Token | the |
Token ID | 262 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
McCartney | +0.350 |
aba | +0.342 |
Reloaded | +0.335 |
igible | +0.332 |
complement | +0.330 |
Neg logit contributions | |
horizont | -0.489 |
MK | -0.430 |
Dash | -0.378 |
HIP | -0.372 |
MK | -0.369 |
Token | super |
Token ID | 2208 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
McCartney | +0.280 |
aba | +0.274 |
Reloaded | +0.269 |
igible | +0.267 |
irs | +0.267 |
Neg logit contributions | |
horizont | -0.340 |
MK | -0.297 |
Dash | -0.258 |
HIP | -0.253 |
MK | -0.252 |
Token | secret |
Token ID | 3200 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
horizont | +0.353 |
MK | +0.305 |
Dash | +0.260 |
HIP | +0.257 |
MK | +0.255 |
Neg logit contributions | |
McCartney | -0.340 |
aba | -0.334 |
Reloaded | -0.328 |
igible | -0.326 |
irs | -0.325 |
Token | Cooper |
Token ID | 10382 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
McCartney | +0.112 |
aba | +0.110 |
Reloaded | +0.108 |
igible | +0.107 |
complement | +0.106 |
Neg logit contributions | |
horizont | -0.138 |
MK | -0.120 |
Dash | -0.104 |
HIP | -0.103 |
MK | -0.102 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
McCartney | +0.072 |
aba | +0.071 |
Reloaded | +0.069 |
igible | +0.069 |
complement | +0.068 |
Neg logit contributions | |
horizont | -0.085 |
MK | -0.074 |
Dash | -0.064 |
HIP | -0.063 |
MK | -0.063 |
Token | development |
Token ID | 2478 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
McCartney | +0.105 |
aba | +0.103 |
Reloaded | +0.101 |
igible | +0.100 |
irs | +0.100 |
Neg logit contributions | |
horizont | -0.120 |
MK | -0.104 |
Dash | -0.090 |
HIP | -0.089 |
MK | -0.089 |
Token | of |
Token ID | 286 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
horizont | +0.811 |
MK | +0.689 |
Dash | +0.572 |
HIP | +0.561 |
MK | +0.559 |
Neg logit contributions | |
McCartney | -1.009 |
aba | -0.989 |
Reloaded | -0.982 |
igible | -0.974 |
irs | -0.968 |
Token | the |
Token ID | 262 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
McCartney | +0.147 |
aba | +0.144 |
Reloaded | +0.142 |
igible | +0.142 |
irs | +0.140 |
Neg logit contributions | |
horizont | -0.193 |
MK | -0.169 |
Dash | -0.146 |
HIP | -0.146 |
MK | -0.145 |
Token | ov |
Token ID | 19643 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
McCartney | +0.104 |
aba | +0.102 |
Reloaded | +0.101 |
igible | +0.100 |
irs | +0.099 |
Neg logit contributions | |
horizont | -0.124 |
MK | -0.109 |
Dash | -0.094 |
HIP | -0.093 |
MK | -0.093 |
Token | aries |
Token ID | 3166 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
horizont | +0.519 |
MK | +0.459 |
Dash | +0.405 |
SOU | +0.401 |
HIP | +0.400 |
Neg logit contributions | |
McCartney | -0.387 |
Reloaded | -0.378 |
aba | -0.370 |
igible | -0.364 |
herty | -0.363 |
Token | — |
Token ID | 851 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
McCartney | +1.711 |
aba | +1.686 |
Reloaded | +1.646 |
complement | +1.639 |
igible | +1.637 |
Neg logit contributions | |
horizont | -1.787 |
MK | -1.541 |
Dash | -1.320 |
HIP | -1.298 |
MK | -1.293 |
Token | and |
Token ID | 290 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
McCartney | +0.112 |
aba | +0.110 |
Reloaded | +0.108 |
igible | +0.107 |
irs | +0.107 |
Neg logit contributions | |
horizont | -0.123 |
MK | -0.107 |
Dash | -0.092 |
HIP | -0.091 |
MK | -0.090 |
Token | one |
Token ID | 530 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
McCartney | +0.192 |
aba | +0.188 |
Reloaded | +0.185 |
igible | +0.184 |
irs | +0.182 |
Neg logit contributions | |
horizont | -0.261 |
MK | -0.229 |
Dash | -0.200 |
HIP | -0.198 |
MK | -0.198 |
Token | process |
Token ID | 1429 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
McCartney | +0.169 |
aba | +0.165 |
Reloaded | +0.163 |
igible | +0.161 |
complement | +0.160 |
Neg logit contributions | |
horizont | -0.203 |
MK | -0.178 |
Dash | -0.154 |
HIP | -0.152 |
MK | -0.151 |
Token | would |
Token ID | 561 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
McCartney | +0.360 |
aba | +0.353 |
Reloaded | +0.348 |
igible | +0.344 |
irs | +0.343 |
Neg logit contributions | |
horizont | -0.404 |
MK | -0.351 |
Dash | -0.302 |
HIP | -0.298 |
MK | -0.297 |
Token | tell |
Token ID | 1560 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
horizont | +0.229 |
MK | +0.200 |
Dash | +0.174 |
HIP | +0.172 |
MK | +0.171 |
Neg logit contributions | |
McCartney | -0.186 |
aba | -0.182 |
Reloaded | -0.179 |
igible | -0.177 |
irs | -0.176 |
Token | es |
Token ID | 274 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
McCartney | +0.269 |
aba | +0.263 |
Reloaded | +0.260 |
igible | +0.257 |
complement | +0.256 |
Neg logit contributions | |
horizont | -0.312 |
MK | -0.272 |
Dash | -0.235 |
HIP | -0.232 |
MK | -0.231 |
Token | actions |
Token ID | 4028 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
McCartney | +0.124 |
aba | +0.121 |
Reloaded | +0.119 |
igible | +0.118 |
complement | +0.118 |
Neg logit contributions | |
horizont | -0.151 |
MK | -0.133 |
Dash | -0.115 |
HIP | -0.113 |
MK | -0.113 |
Token | s |
Token ID | 264 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
McCartney | +0.731 |
aba | +0.718 |
Reloaded | +0.704 |
igible | +0.699 |
complement | +0.699 |
Neg logit contributions | |
horizont | -0.868 |
MK | -0.762 |
Dash | -0.660 |
HIP | -0.649 |
MK | -0.646 |
Token | ont |
Token ID | 756 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
horizont | +0.492 |
MK | +0.427 |
Dash | +0.369 |
HIP | +0.363 |
MK | +0.362 |
Neg logit contributions | |
McCartney | -0.435 |
aba | -0.425 |
Reloaded | -0.421 |
igible | -0.415 |
complement | -0.414 |
Token | enc |
Token ID | 2207 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
horizont | +0.623 |
MK | +0.544 |
Dash | +0.473 |
HIP | +0.468 |
MK | +0.467 |
Neg logit contributions | |
McCartney | -0.488 |
aba | -0.478 |
Reloaded | -0.470 |
igible | -0.466 |
irs | -0.462 |
Token | ore |
Token ID | 382 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
aba | +1.572 |
McCartney | +1.560 |
irs | +1.529 |
igible | +1.501 |
herty | +1.491 |
Neg logit contributions | |
horizont | -1.951 |
MK | -1.695 |
Dash | -1.463 |
HIP | -1.453 |
MK | -1.443 |
Token | obsc |
Token ID | 10666 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
McCartney | +0.227 |
aba | +0.223 |
Reloaded | +0.219 |
igible | +0.217 |
complement | +0.216 |
Neg logit contributions | |
horizont | -0.275 |
MK | -0.240 |
Dash | -0.208 |
HIP | -0.206 |
MK | -0.205 |
Token | ures |
Token ID | 942 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
McCartney | +0.500 |
aba | +0.491 |
Reloaded | +0.480 |
igible | +0.477 |
irs | +0.475 |
Neg logit contributions | |
horizont | -0.637 |
MK | -0.558 |
Dash | -0.486 |
HIP | -0.480 |
MK | -0.478 |
Token | : |
Token ID | 1058 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
horizont | +0.030 |
MK | +0.026 |
Dash | +0.023 |
HIP | +0.022 |
MK | +0.022 |
Neg logit contributions | |
McCartney | -0.026 |
aba | -0.025 |
Reloaded | -0.025 |
igible | -0.025 |
complement | -0.024 |
Token | les |
Token ID | 10287 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
horizont | +0.560 |
MK | +0.491 |
Dash | +0.427 |
HIP | +0.422 |
MK | +0.421 |
Neg logit contributions | |
McCartney | -0.437 |
aba | -0.428 |
Reloaded | -0.420 |
igible | -0.418 |
complement | -0.414 |
Token | safe |
Token ID | 3338 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
McCartney | +0.145 |
aba | +0.143 |
Reloaded | +0.141 |
igible | +0.139 |
irs | +0.138 |
Neg logit contributions | |
horizont | -0.156 |
MK | -0.135 |
Dash | -0.116 |
HIP | -0.114 |
MK | -0.114 |
Token | ! |
Token ID | 0 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
horizont | +0.268 |
MK | +0.233 |
Dash | +0.200 |
HIP | +0.198 |
MK | +0.198 |
Neg logit contributions | |
McCartney | -0.246 |
aba | -0.241 |
Reloaded | -0.240 |
igible | -0.236 |
irs | -0.235 |
Token | God |
Token ID | 1793 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
McCartney | +0.029 |
aba | +0.029 |
Reloaded | +0.028 |
igible | +0.028 |
complement | +0.028 |
Neg logit contributions | |
horizont | -0.034 |
MK | -0.030 |
Dash | -0.026 |
HIP | -0.025 |
MK | -0.025 |
Token | bless |
Token ID | 12012 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
McCartney | +0.185 |
aba | +0.178 |
Reloaded | +0.175 |
igible | +0.172 |
complement | +0.170 |
Neg logit contributions | |
horizont | -0.465 |
MK | -0.421 |
Dash | -0.380 |
HIP | -0.377 |
MK | -0.376 |
Token | y |
Token ID | 331 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
McCartney | +0.397 |
aba | +0.389 |
Reloaded | +0.382 |
igible | +0.379 |
complement | +0.376 |
Neg logit contributions | |
horizont | -0.518 |
MK | -0.455 |
Dash | -0.396 |
HIP | -0.392 |
MK | -0.391 |
Token | ' |
Token ID | 6 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
McCartney | +0.277 |
aba | +0.274 |
Reloaded | +0.267 |
irs | +0.264 |
igible | +0.264 |
Neg logit contributions | |
horizont | -0.348 |
MK | -0.302 |
Dash | -0.263 |
HIP | -0.259 |
ivably | -0.258 |
Token | all |
Token ID | 439 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
McCartney | +1.910 |
aba | +1.898 |
Reloaded | +1.858 |
irs | +1.853 |
igible | +1.840 |
Neg logit contributions | |
horizont | -1.577 |
MK | -1.312 |
Dash | -1.090 |
HIP | -1.073 |
ivably | -1.065 |
Token | ! |
Token ID | 0 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
McCartney | +0.635 |
aba | +0.623 |
Reloaded | +0.612 |
igible | +0.606 |
complement | +0.603 |
Neg logit contributions | |
horizont | -0.775 |
MK | -0.677 |
Dash | -0.588 |
HIP | -0.580 |
MK | -0.578 |
Token | - |
Token ID | 532 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
McCartney | +0.079 |
aba | +0.078 |
Reloaded | +0.077 |
igible | +0.076 |
complement | +0.076 |
Neg logit contributions | |
horizont | -0.084 |
MK | -0.072 |
Dash | -0.062 |
HIP | -0.061 |
MK | -0.061 |
Token | Derek |
Token ID | 20893 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
McCartney | +0.460 |
aba | +0.451 |
Reloaded | +0.443 |
igible | +0.438 |
complement | +0.435 |
Neg logit contributions | |
horizont | -0.628 |
MK | -0.553 |
Dash | -0.483 |
HIP | -0.478 |
MK | -0.476 |
Token | Carr |
Token ID | 20765 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
McCartney | +0.966 |
aba | +0.941 |
Reloaded | +0.917 |
igible | +0.908 |
complement | +0.904 |
Neg logit contributions | |
horizont | -1.267 |
MK | -1.104 |
Dash | -0.965 |
HIP | -0.952 |
ivably | -0.949 |
Token | seek |
Token ID | 5380 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
McCartney | +0.189 |
aba | +0.185 |
Reloaded | +0.182 |
igible | +0.180 |
complement | +0.180 |
Neg logit contributions | |
horizont | -0.217 |
MK | -0.189 |
Dash | -0.163 |
HIP | -0.161 |
MK | -0.161 |
Token | partnerships |
Token ID | 22867 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
McCartney | +0.404 |
aba | +0.396 |
Reloaded | +0.389 |
igible | +0.386 |
complement | +0.385 |
Neg logit contributions | |
horizont | -0.483 |
MK | -0.422 |
Dash | -0.366 |
HIP | -0.360 |
MK | -0.359 |
Token | — |
Token ID | 960 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
McCartney | +0.206 |
aba | +0.202 |
Reloaded | +0.201 |
igible | +0.197 |
irs | +0.196 |
Neg logit contributions | |
horizont | -0.198 |
MK | -0.170 |
Dash | -0.143 |
HIP | -0.142 |
MK | -0.142 |
Token | even |
Token ID | 10197 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
McCartney | +0.394 |
aba | +0.387 |
Reloaded | +0.381 |
igible | +0.377 |
complement | +0.376 |
Neg logit contributions | |
horizont | -0.443 |
MK | -0.385 |
Dash | -0.331 |
HIP | -0.327 |
MK | -0.326 |
Token | with |
Token ID | 351 |
Feature activation | +7.10e-01 |
Pos logit contributions | |
McCartney | +0.820 |
aba | +0.804 |
Reloaded | +0.787 |
igible | +0.782 |
complement | +0.781 |
Neg logit contributions | |
horizont | -1.057 |
MK | -0.930 |
Dash | -0.811 |
HIP | -0.797 |
MK | -0.794 |
Token | direct |
Token ID | 1277 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
McCartney | +0.421 |
aba | +0.413 |
Reloaded | +0.407 |
igible | +0.401 |
irs | +0.399 |
Neg logit contributions | |
horizont | -0.580 |
MK | -0.510 |
Dash | -0.445 |
HIP | -0.443 |
MK | -0.441 |
Token | competitors |
Token ID | 13861 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
McCartney | +1.381 |
aba | +1.349 |
complement | +1.322 |
igible | +1.320 |
Reloaded | +1.310 |
Neg logit contributions | |
horizont | -2.050 |
MK | -1.821 |
Dash | -1.602 |
HIP | -1.569 |
ivably | -1.568 |
Token | — |
Token ID | 960 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
McCartney | +0.715 |
aba | +0.701 |
Reloaded | +0.695 |
igible | +0.686 |
irs | +0.683 |
Neg logit contributions | |
horizont | -0.743 |
MK | -0.644 |
Dash | -0.549 |
HIP | -0.543 |
MK | -0.543 |
Token | in |
Token ID | 259 |
Feature activation | +7.61e-01 |
Pos logit contributions | |
McCartney | +0.476 |
aba | +0.463 |
Reloaded | +0.458 |
igible | +0.451 |
complement | +0.448 |
Neg logit contributions | |
horizont | -0.712 |
MK | -0.634 |
Dash | -0.558 |
MK | -0.549 |
HIP | -0.549 |
Token | areas |
Token ID | 3006 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
McCartney | +0.443 |
aba | +0.434 |
Reloaded | +0.428 |
igible | +0.422 |
irs | +0.420 |
Neg logit contributions | |
horizont | -0.629 |
MK | -0.554 |
Dash | -0.485 |
HIP | -0.483 |
MK | -0.481 |
Token | of |
Token ID | 286 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
horizont | +0.002 |
MK | +0.001 |
Dash | +0.001 |
HIP | +0.001 |
MK | +0.001 |
Neg logit contributions | |
McCartney | -0.001 |
aba | -0.001 |
Reloaded | -0.001 |
igible | -0.001 |
irs | -0.001 |
Token | June |
Token ID | 2795 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
McCartney | +0.325 |
aba | +0.319 |
Reloaded | +0.313 |
igible | +0.310 |
complement | +0.309 |
Neg logit contributions | |
horizont | -0.380 |
MK | -0.331 |
Dash | -0.286 |
HIP | -0.283 |
MK | -0.281 |
Token | " |
Token ID | 1 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
horizont | +0.122 |
MK | +0.108 |
Dash | +0.094 |
HIP | +0.093 |
MK | +0.093 |
Neg logit contributions | |
McCartney | -0.088 |
aba | -0.086 |
Reloaded | -0.085 |
igible | -0.084 |
irs | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
McCartney | +0.570 |
aba | +0.560 |
Reloaded | +0.552 |
igible | +0.547 |
complement | +0.544 |
Neg logit contributions | |
horizont | -0.567 |
MK | -0.488 |
Dash | -0.416 |
HIP | -0.410 |
MK | -0.408 |
Token | " |
Token ID | 366 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
horizont | +0.168 |
MK | +0.147 |
Dash | +0.128 |
MK | +0.127 |
HIP | +0.127 |
Neg logit contributions | |
McCartney | -0.124 |
aba | -0.122 |
Reloaded | -0.119 |
igible | -0.118 |
irs | -0.117 |
Token | to |
Token ID | 1462 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
horizont | +0.353 |
MK | +0.315 |
Dash | +0.279 |
MK | +0.274 |
HIP | +0.274 |
Neg logit contributions | |
McCartney | -0.217 |
aba | -0.211 |
Reloaded | -0.207 |
igible | -0.206 |
complement | -0.204 |
Token | enc |
Token ID | 2207 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
McCartney | +0.148 |
aba | +0.145 |
Reloaded | +0.143 |
igible | +0.142 |
complement | +0.141 |
Neg logit contributions | |
horizont | -0.167 |
MK | -0.145 |
Dash | -0.125 |
HIP | -0.123 |
MK | -0.123 |
Token | irc |
Token ID | 1980 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
aba | +1.399 |
McCartney | +1.395 |
irs | +1.356 |
igible | +1.345 |
complement | +1.333 |
Neg logit contributions | |
horizont | -2.038 |
MK | -1.763 |
Dash | -1.540 |
HIP | -1.535 |
ledge | -1.514 |
Token | le |
Token ID | 293 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
McCartney | +0.022 |
aba | +0.022 |
Reloaded | +0.022 |
igible | +0.021 |
complement | +0.021 |
Neg logit contributions | |
horizont | -0.022 |
MK | -0.019 |
Dash | -0.016 |
HIP | -0.016 |
MK | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +1.08e-03 |
Pos logit contributions | |
horizont | +0.248 |
MK | +0.216 |
Dash | +0.187 |
HIP | +0.185 |
MK | +0.184 |
Neg logit contributions | |
McCartney | -0.212 |
aba | -0.208 |
Reloaded | -0.204 |
igible | -0.202 |
complement | -0.201 |
Token | destroy |
Token ID | 4117 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
McCartney | +0.001 |
aba | +0.001 |
Reloaded | +0.001 |
igible | +0.001 |
complement | +0.001 |
Neg logit contributions | |
horizont | -0.001 |
MK | -0.001 |
Dash | -0.001 |
HIP | -0.001 |
MK | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
McCartney | +0.865 |
aba | +0.846 |
Reloaded | +0.831 |
complement | +0.826 |
igible | +0.824 |
Neg logit contributions | |
horizont | -1.084 |
MK | -0.951 |
Dash | -0.831 |
HIP | -0.814 |
MK | -0.808 |
Token | li |
Token ID | 4528 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
McCartney | +0.039 |
aba | +0.038 |
Reloaded | +0.037 |
irs | +0.037 |
igible | +0.037 |
Neg logit contributions | |
horizont | -0.061 |
MK | -0.054 |
Dash | -0.048 |
HIP | -0.047 |
MK | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
horizont | +0.195 |
MK | +0.170 |
Dash | +0.146 |
MK | +0.143 |
HIP | +0.143 |
Neg logit contributions | |
McCartney | -0.186 |
aba | -0.183 |
Reloaded | -0.181 |
igible | -0.179 |
complement | -0.178 |
Token | Dwarf |
Token ID | 33196 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
McCartney | +0.438 |
aba | +0.433 |
Reloaded | +0.429 |
igible | +0.421 |
stadt | +0.419 |
Neg logit contributions | |
horizont | -0.599 |
MK | -0.525 |
Dash | -0.462 |
MK | -0.458 |
HIP | -0.457 |
Token | , |
Token ID | 11 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
McCartney | +0.094 |
aba | +0.092 |
Reloaded | +0.091 |
igible | +0.090 |
complement | +0.090 |
Neg logit contributions | |
horizont | -0.102 |
MK | -0.088 |
Dash | -0.076 |
HIP | -0.075 |
MK | -0.075 |
Token | Leg |
Token ID | 3564 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
McCartney | +0.339 |
aba | +0.333 |
Reloaded | +0.328 |
igible | +0.324 |
complement | +0.323 |
Neg logit contributions | |
horizont | -0.418 |
MK | -0.366 |
Dash | -0.317 |
HIP | -0.313 |
MK | -0.312 |
Token | olas |
Token ID | 12456 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
McCartney | +0.415 |
aba | +0.404 |
irs | +0.373 |
igible | +0.369 |
herty | +0.369 |
Neg logit contributions | |
horizont | -1.284 |
MK | -1.164 |
Dash | -1.054 |
HIP | -1.046 |
MK | -1.036 |
Token | the |
Token ID | 262 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
McCartney | +1.560 |
aba | +1.525 |
Reloaded | +1.493 |
complement | +1.480 |
irs | +1.476 |
Neg logit contributions | |
horizont | -1.845 |
MK | -1.599 |
Dash | -1.386 |
HIP | -1.371 |
ivably | -1.357 |
Token | Elf |
Token ID | 19067 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
McCartney | +0.541 |
aba | +0.531 |
Reloaded | +0.522 |
igible | +0.516 |
complement | +0.513 |
Neg logit contributions | |
horizont | -0.742 |
MK | -0.653 |
Dash | -0.571 |
HIP | -0.565 |
MK | -0.563 |
Token | , |
Token ID | 11 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
McCartney | +0.197 |
aba | +0.193 |
Reloaded | +0.190 |
igible | +0.188 |
complement | +0.187 |
Neg logit contributions | |
horizont | -0.215 |
MK | -0.186 |
Dash | -0.160 |
HIP | -0.158 |
MK | -0.157 |
Token | and |
Token ID | 290 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
McCartney | +0.395 |
aba | +0.387 |
Reloaded | +0.380 |
igible | +0.376 |
complement | +0.375 |
Neg logit contributions | |
horizont | -0.490 |
MK | -0.428 |
Dash | -0.372 |
HIP | -0.367 |
MK | -0.365 |
Token | the |
Token ID | 262 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
McCartney | +0.247 |
aba | +0.242 |
Reloaded | +0.237 |
igible | +0.234 |
complement | +0.234 |
Neg logit contributions | |
horizont | -0.325 |
MK | -0.285 |
Dash | -0.248 |
HIP | -0.246 |
MK | -0.244 |
Token | images |
Token ID | 17566 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
McCartney | +0.226 |
Reloaded | +0.218 |
aba | +0.218 |
complement | +0.215 |
igible | +0.214 |
Neg logit contributions | |
horizont | -0.250 |
MK | -0.221 |
Dash | -0.191 |
MK | -0.186 |
HIP | -0.186 |
Token | ( |
Token ID | 7 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
horizont | +0.463 |
MK | +0.413 |
Dash | +0.351 |
MK | +0.347 |
HIP | +0.344 |
Neg logit contributions | |
McCartney | -0.466 |
aba | -0.454 |
Reloaded | -0.452 |
igible | -0.446 |
complement | -0.445 |
Token | args |
Token ID | 22046 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
McCartney | +0.053 |
aba | +0.051 |
Reloaded | +0.051 |
complement | +0.050 |
igible | +0.049 |
Neg logit contributions | |
horizont | -0.079 |
MK | -0.071 |
Dash | -0.062 |
HIP | -0.062 |
MK | -0.061 |
Token | [" |
Token ID | 14692 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
McCartney | +0.823 |
aba | +0.810 |
Reloaded | +0.798 |
igible | +0.791 |
irs | +0.789 |
Neg logit contributions | |
horizont | -0.735 |
MK | -0.625 |
Dash | -0.526 |
HIP | -0.518 |
MK | -0.515 |
Token | images |
Token ID | 17566 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
horizont | +0.399 |
MK | +0.354 |
Dash | +0.306 |
HIP | +0.298 |
MK | +0.298 |
Neg logit contributions | |
McCartney | -0.365 |
Reloaded | -0.352 |
aba | -0.352 |
complement | -0.347 |
igible | -0.345 |
Token | "] |
Token ID | 8973 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
horizont | +0.411 |
MK | +0.349 |
Dash | +0.281 |
MK | +0.273 |
HIP | +0.271 |
Neg logit contributions | |
McCartney | -0.663 |
aba | -0.647 |
Reloaded | -0.646 |
igible | -0.642 |
complement | -0.635 |
Token | ): |
Token ID | 2599 |
Feature activation | -9.19e-01 |
Pos logit contributions | |
horizont | +2.094 |
MK | +2.053 |
MK | +1.714 |
Dash | +1.658 |
SOU | +1.638 |
Neg logit contributions | |
McCartney | -2.729 |
Reloaded | -2.659 |
igible | -2.649 |
complement | -2.623 |
aba | -2.620 |
Token | # |
Token ID | 1303 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
horizont | +0.677 |
MK | +0.591 |
MK | +0.507 |
HIP | +0.506 |
Dash | +0.504 |
Neg logit contributions | |
McCartney | -0.616 |
Reloaded | -0.600 |
aba | -0.600 |
igible | -0.591 |
stadt | -0.584 |
Token | load |
Token ID | 3440 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
horizont | +1.827 |
MK | +1.657 |
HIP | +1.489 |
MK | +1.482 |
SOU | +1.465 |
Neg logit contributions | |
McCartney | -1.061 |
Reloaded | -1.022 |
aba | -1.015 |
herty | -0.997 |
igible | -0.996 |
Token | the |
Token ID | 262 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
McCartney | +0.364 |
aba | +0.357 |
Reloaded | +0.351 |
igible | +0.347 |
irs | +0.345 |
Neg logit contributions | |
horizont | -0.516 |
MK | -0.460 |
Dash | -0.403 |
HIP | -0.399 |
MK | -0.399 |
Token | image |
Token ID | 2939 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
horizont | +0.207 |
MK | +0.172 |
HIP | +0.139 |
Dash | +0.139 |
MK | +0.139 |
Neg logit contributions | |
McCartney | -0.307 |
aba | -0.303 |
Reloaded | -0.301 |
igible | -0.297 |
irs | -0.296 |
Token | images |
Token ID | 17566 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
McCartney | +0.469 |
aba | +0.463 |
Reloaded | +0.452 |
irs | +0.451 |
igible | +0.451 |
Neg logit contributions | |
horizont | -0.505 |
MK | -0.436 |
Dash | -0.374 |
HIP | -0.371 |
MK | -0.368 |
Token | ( |
Token ID | 7 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
horizont | +0.058 |
MK | +0.051 |
Dash | +0.044 |
HIP | +0.044 |
MK | +0.044 |
Neg logit contributions | |
McCartney | -0.046 |
aba | -0.045 |
Reloaded | -0.044 |
igible | -0.044 |
complement | -0.043 |
Token | args |
Token ID | 22046 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
McCartney | +0.235 |
aba | +0.231 |
Reloaded | +0.228 |
igible | +0.225 |
complement | +0.224 |
Neg logit contributions | |
horizont | -0.249 |
MK | -0.215 |
Dash | -0.185 |
HIP | -0.182 |
MK | -0.181 |
Token | [" |
Token ID | 14692 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
McCartney | +1.023 |
aba | +1.005 |
Reloaded | +0.986 |
igible | +0.979 |
irs | +0.976 |
Neg logit contributions | |
horizont | -1.191 |
MK | -1.036 |
Dash | -0.895 |
HIP | -0.883 |
MK | -0.879 |
Token | images |
Token ID | 17566 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
horizont | +0.217 |
MK | +0.188 |
Dash | +0.161 |
HIP | +0.158 |
MK | +0.158 |
Neg logit contributions | |
McCartney | -0.207 |
aba | -0.203 |
Reloaded | -0.200 |
igible | -0.198 |
complement | -0.197 |
Token | "] |
Token ID | 8973 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
horizont | +0.118 |
MK | +0.099 |
Dash | +0.081 |
HIP | +0.079 |
MK | +0.079 |
Neg logit contributions | |
McCartney | -0.163 |
aba | -0.161 |
Reloaded | -0.159 |
igible | -0.158 |
complement | -0.157 |
Token | ): |
Token ID | 2599 |
Feature activation | -7.71e-01 |
Pos logit contributions | |
horizont | +2.447 |
MK | +2.143 |
MK | +1.845 |
HIP | +1.791 |
Dash | +1.788 |
Neg logit contributions | |
McCartney | -2.554 |
Reloaded | -2.509 |
aba | -2.488 |
igible | -2.441 |
herty | -2.420 |
Token | # |
Token ID | 1303 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
horizont | +0.598 |
MK | +0.519 |
Dash | +0.449 |
HIP | +0.446 |
MK | +0.446 |
Neg logit contributions | |
McCartney | -0.491 |
aba | -0.480 |
Reloaded | -0.476 |
igible | -0.469 |
irs | -0.465 |
Token | load |
Token ID | 3440 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
horizont | +1.448 |
MK | +1.257 |
Dash | +1.082 |
MK | +1.078 |
HIP | +1.073 |
Neg logit contributions | |
McCartney | -1.248 |
aba | -1.220 |
Reloaded | -1.213 |
igible | -1.199 |
herty | -1.182 |
Token | the |
Token ID | 262 |
Feature activation | -8.22e-02 |
Pos logit contributions | |
McCartney | +0.475 |
aba | +0.466 |
Reloaded | +0.458 |
igible | +0.454 |
complement | +0.451 |
Neg logit contributions | |
horizont | -0.571 |
MK | -0.499 |
Dash | -0.432 |
HIP | -0.427 |
MK | -0.425 |
Token | image |
Token ID | 2939 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
horizont | +0.059 |
MK | +0.050 |
Dash | +0.043 |
HIP | +0.043 |
MK | +0.043 |
Neg logit contributions | |
McCartney | -0.057 |
aba | -0.056 |
Reloaded | -0.056 |
igible | -0.055 |
stadt | -0.055 |
Token | by |
Token ID | 416 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
horizont | +0.128 |
MK | +0.111 |
Dash | +0.096 |
HIP | +0.094 |
MK | +0.094 |
Neg logit contributions | |
McCartney | -0.116 |
aba | -0.114 |
Reloaded | -0.113 |
igible | -0.111 |
irs | -0.111 |
Token | a |
Token ID | 257 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
horizont | +0.155 |
MK | +0.137 |
Dash | +0.120 |
HIP | +0.119 |
MK | +0.119 |
Neg logit contributions | |
McCartney | -0.109 |
aba | -0.107 |
Reloaded | -0.106 |
igible | -0.105 |
irs | -0.104 |
Token | professor |
Token ID | 6240 |
Feature activation | -6.33e-01 |
Pos logit contributions | |
horizont | +0.193 |
MK | +0.168 |
Dash | +0.144 |
HIP | +0.143 |
MK | +0.142 |
Neg logit contributions | |
McCartney | -0.173 |
aba | -0.170 |
Reloaded | -0.168 |
igible | -0.166 |
irs | -0.165 |
Token | at |
Token ID | 379 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
horizont | +0.442 |
MK | +0.381 |
Dash | +0.325 |
HIP | +0.322 |
MK | +0.321 |
Neg logit contributions | |
McCartney | -0.446 |
aba | -0.439 |
Reloaded | -0.436 |
igible | -0.430 |
irs | -0.428 |
Token | St |
Token ID | 520 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
horizont | +0.586 |
MK | +0.521 |
HIP | +0.455 |
Dash | +0.454 |
MK | +0.454 |
Neg logit contributions | |
McCartney | -0.406 |
aba | -0.399 |
Reloaded | -0.398 |
igible | -0.391 |
irs | -0.387 |
Token | ony |
Token ID | 1647 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
horizont | +0.441 |
MK | +0.385 |
Dash | +0.311 |
HIP | +0.308 |
MK | +0.308 |
Neg logit contributions | |
McCartney | -0.622 |
Reloaded | -0.617 |
aba | -0.607 |
igible | -0.604 |
complement | -0.602 |
Token | Brook |
Token ID | 7590 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
horizont | +2.413 |
MK | +2.161 |
Dash | +1.920 |
HIP | +1.915 |
MK | +1.889 |
Neg logit contributions | |
McCartney | -1.551 |
aba | -1.523 |
Reloaded | -1.522 |
igible | -1.505 |
irs | -1.486 |
Token | � |
Token ID | 447 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
horizont | +1.704 |
MK | +1.515 |
MK | +1.315 |
SOU | +1.315 |
HIP | +1.309 |
Neg logit contributions | |
McCartney | -1.300 |
aba | -1.272 |
Reloaded | -1.268 |
igible | -1.253 |
irs | -1.248 |
Token | � |
Token ID | 247 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
horizont | +0.175 |
MK | +0.157 |
Dash | +0.138 |
HIP | +0.135 |
MK | +0.135 |
Neg logit contributions | |
McCartney | -0.145 |
aba | -0.142 |
Reloaded | -0.140 |
igible | -0.140 |
complement | -0.138 |
Token | s |
Token ID | 82 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
horizont | +0.409 |
MK | +0.349 |
Dash | +0.292 |
HIP | +0.287 |
MK | +0.286 |
Neg logit contributions | |
McCartney | -0.484 |
aba | -0.475 |
Reloaded | -0.472 |
igible | -0.466 |
complement | -0.464 |
Token | School |
Token ID | 3961 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
horizont | +0.361 |
MK | +0.320 |
Dash | +0.280 |
MK | +0.279 |
HIP | +0.278 |
Neg logit contributions | |
McCartney | -0.253 |
aba | -0.249 |
Reloaded | -0.247 |
igible | -0.242 |
irs | -0.239 |
Token | } |
Token ID | 1782 |
Feature activation | +9.48e-03 |
Pos logit contributions | |
horizont | +0.028 |
MK | +0.024 |
Dash | +0.020 |
HIP | +0.020 |
MK | +0.020 |
Neg logit contributions | |
McCartney | -0.032 |
aba | -0.031 |
Reloaded | -0.031 |
igible | -0.031 |
irs | -0.030 |
Token | End |
Token ID | 5268 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
McCartney | +0.007 |
aba | +0.007 |
Reloaded | +0.006 |
igible | +0.006 |
complement | +0.006 |
Neg logit contributions | |
horizont | -0.007 |
MK | -0.006 |
Dash | -0.005 |
HIP | -0.005 |
MK | -0.005 |
Token | { |
Token ID | 1391 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
horizont | +0.118 |
MK | +0.101 |
Dash | +0.086 |
HIP | +0.085 |
MK | +0.085 |
Neg logit contributions | |
McCartney | -0.111 |
aba | -0.109 |
Reloaded | -0.107 |
igible | -0.106 |
irs | -0.106 |
Token | Try |
Token ID | 9993 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
McCartney | +0.364 |
aba | +0.358 |
Reloaded | +0.350 |
complement | +0.348 |
igible | +0.348 |
Neg logit contributions | |
horizont | -0.432 |
MK | -0.377 |
Dash | -0.326 |
HIP | -0.321 |
MK | -0.319 |
Token | { |
Token ID | 1391 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
horizont | +0.671 |
MK | +0.559 |
Dash | +0.458 |
HIP | +0.448 |
MK | +0.446 |
Neg logit contributions | |
McCartney | -0.937 |
aba | -0.923 |
Reloaded | -0.913 |
igible | -0.905 |
irs | -0.901 |
Token | # |
Token ID | 1303 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
McCartney | +0.322 |
aba | +0.316 |
Reloaded | +0.310 |
igible | +0.308 |
complement | +0.307 |
Neg logit contributions | |
horizont | -0.394 |
MK | -0.344 |
Dash | -0.299 |
HIP | -0.294 |
MK | -0.293 |
Token | Close |
Token ID | 13872 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
horizont | +2.019 |
MK | +1.744 |
Dash | +1.508 |
HIP | +1.496 |
MK | +1.494 |
Neg logit contributions | |
McCartney | -1.874 |
aba | -1.850 |
Reloaded | -1.825 |
igible | -1.806 |
irs | -1.784 |
Token | Graphics |
Token ID | 19840 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
horizont | +1.348 |
MK | +1.172 |
Dash | +1.026 |
MK | +1.019 |
HIP | +1.018 |
Neg logit contributions | |
McCartney | -1.047 |
aba | -1.019 |
Reloaded | -1.008 |
stadt | -0.986 |
herty | -0.986 |
Token | $ |
Token ID | 720 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
McCartney | +0.157 |
aba | +0.154 |
Reloaded | +0.151 |
igible | +0.150 |
complement | +0.149 |
Neg logit contributions | |
horizont | -0.197 |
MK | -0.172 |
Dash | -0.150 |
HIP | -0.148 |
MK | -0.148 |
Token | image |
Token ID | 9060 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
horizont | +0.384 |
MK | +0.337 |
Dash | +0.292 |
HIP | +0.286 |
MK | +0.285 |
Neg logit contributions | |
McCartney | -0.326 |
aba | -0.318 |
Reloaded | -0.313 |
igible | -0.311 |
complement | -0.310 |
Token | . |
Token ID | 13 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
horizont | +0.342 |
MK | +0.288 |
Dash | +0.238 |
HIP | +0.234 |
MK | +0.233 |
Neg logit contributions | |
McCartney | -0.443 |
aba | -0.436 |
Reloaded | -0.430 |
igible | -0.427 |
complement | -0.425 |
Token | well |
Token ID | 880 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
horizont | +0.253 |
MK | +0.222 |
Dash | +0.193 |
HIP | +0.191 |
MK | +0.190 |
Neg logit contributions | |
McCartney | -0.201 |
aba | -0.198 |
Reloaded | -0.195 |
igible | -0.193 |
irs | -0.191 |
Token | as |
Token ID | 355 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
McCartney | +0.059 |
aba | +0.058 |
Reloaded | +0.057 |
igible | +0.056 |
irs | +0.056 |
Neg logit contributions | |
horizont | -0.064 |
MK | -0.056 |
Dash | -0.048 |
HIP | -0.048 |
MK | -0.048 |
Token | any |
Token ID | 597 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
horizont | +0.245 |
MK | +0.215 |
Dash | +0.188 |
HIP | +0.187 |
MK | +0.186 |
Neg logit contributions | |
McCartney | -0.177 |
aba | -0.174 |
Reloaded | -0.171 |
igible | -0.169 |
irs | -0.168 |
Token | sense |
Token ID | 2565 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
horizont | +0.580 |
MK | +0.506 |
Dash | +0.438 |
HIP | +0.433 |
MK | +0.432 |
Neg logit contributions | |
McCartney | -0.482 |
aba | -0.473 |
Reloaded | -0.466 |
igible | -0.461 |
irs | -0.458 |
Token | of |
Token ID | 286 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
horizont | +0.517 |
MK | +0.445 |
Dash | +0.377 |
HIP | +0.372 |
MK | +0.371 |
Neg logit contributions | |
McCartney | -0.536 |
aba | -0.527 |
Reloaded | -0.522 |
igible | -0.514 |
irs | -0.512 |
Token | inev |
Token ID | 9026 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
horizont | +0.231 |
MK | +0.203 |
Dash | +0.177 |
HIP | +0.175 |
MK | +0.175 |
Neg logit contributions | |
McCartney | -0.169 |
aba | -0.166 |
Reloaded | -0.163 |
igible | -0.161 |
irs | -0.160 |
Token | itability |
Token ID | 34147 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
horizont | +2.273 |
MK | +2.000 |
Dash | +1.763 |
ivably | +1.742 |
MK | +1.731 |
Neg logit contributions | |
McCartney | -1.585 |
Reloaded | -1.557 |
aba | -1.507 |
complement | -1.485 |
igible | -1.481 |
Token | about |
Token ID | 546 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
McCartney | +0.164 |
aba | +0.161 |
Reloaded | +0.159 |
igible | +0.157 |
irs | +0.156 |
Neg logit contributions | |
horizont | -0.172 |
MK | -0.148 |
Dash | -0.127 |
HIP | -0.125 |
MK | -0.125 |
Token | what |
Token ID | 644 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
McCartney | +0.114 |
aba | +0.111 |
Reloaded | +0.109 |
igible | +0.108 |
complement | +0.107 |
Neg logit contributions | |
horizont | -0.171 |
MK | -0.152 |
Dash | -0.134 |
HIP | -0.132 |
MK | -0.131 |
Token | happens |
Token ID | 4325 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
horizont | +0.204 |
MK | +0.179 |
Dash | +0.156 |
HIP | +0.154 |
MK | +0.153 |
Neg logit contributions | |
McCartney | -0.158 |
aba | -0.154 |
Reloaded | -0.152 |
igible | -0.150 |
complement | -0.150 |
Token | in |
Token ID | 287 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
horizont | +0.478 |
MK | +0.416 |
Dash | +0.359 |
HIP | +0.357 |
MK | +0.356 |
Neg logit contributions | |
McCartney | -0.399 |
aba | -0.392 |
Reloaded | -0.389 |
igible | -0.383 |
irs | -0.380 |
Token | "] |
Token ID | 8973 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
horizont | +0.407 |
MK | +0.336 |
Dash | +0.274 |
HIP | +0.269 |
MK | +0.267 |
Neg logit contributions | |
McCartney | -0.564 |
aba | -0.556 |
Reloaded | -0.548 |
igible | -0.543 |
complement | -0.542 |
Token | [ |
Token ID | 685 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
horizont | +1.408 |
MK | +1.308 |
MK | +1.090 |
SOU | +1.077 |
HIP | +1.067 |
Neg logit contributions | |
igible | -1.954 |
McCartney | -1.952 |
Reloaded | -1.944 |
aba | -1.932 |
keyes | -1.930 |
Token | Round |
Token ID | 22685 |
Feature activation | -9.69e-01 |
Pos logit contributions | |
McCartney | +0.513 |
aba | +0.507 |
Reloaded | +0.498 |
igible | +0.496 |
irs | +0.494 |
Neg logit contributions | |
horizont | -0.316 |
MK | -0.254 |
Dash | -0.202 |
HIP | -0.198 |
MK | -0.196 |
Token | " |
Token ID | 366 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
horizont | +0.864 |
MK | +0.770 |
Dash | +0.683 |
HIP | +0.675 |
MK | +0.673 |
Neg logit contributions | |
McCartney | -0.501 |
aba | -0.489 |
Reloaded | -0.479 |
igible | -0.473 |
complement | -0.470 |
Token | 11 |
Token ID | 1157 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
horizont | +0.333 |
MK | +0.293 |
Dash | +0.252 |
MK | +0.248 |
HIP | +0.248 |
Neg logit contributions | |
McCartney | -0.301 |
aba | -0.294 |
Reloaded | -0.289 |
igible | -0.287 |
complement | -0.286 |
Token | "] |
Token ID | 8973 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
horizont | +0.237 |
MK | +0.202 |
Dash | +0.170 |
HIP | +0.168 |
MK | +0.167 |
Neg logit contributions | |
McCartney | -0.264 |
aba | -0.260 |
Reloaded | -0.256 |
igible | -0.254 |
complement | -0.253 |
Token | [ |
Token ID | 685 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
horizont | +1.536 |
MK | +1.427 |
MK | +1.197 |
Dash | +1.170 |
HIP | +1.160 |
Neg logit contributions | |
McCartney | -2.150 |
Reloaded | -2.148 |
igible | -2.127 |
aba | -2.116 |
irs | -2.091 |
Token | White |
Token ID | 12256 |
Feature activation | -6.78e-02 |
Pos logit contributions | |
McCartney | +0.367 |
aba | +0.361 |
Reloaded | +0.355 |
igible | +0.352 |
complement | +0.350 |
Neg logit contributions | |
horizont | -0.394 |
MK | -0.341 |
Dash | -0.293 |
HIP | -0.289 |
MK | -0.288 |
Token | " |
Token ID | 366 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
horizont | +0.061 |
MK | +0.054 |
Dash | +0.048 |
HIP | +0.048 |
MK | +0.047 |
Neg logit contributions | |
McCartney | -0.035 |
aba | -0.034 |
Reloaded | -0.033 |
igible | -0.033 |
complement | -0.033 |
Token | Ab |
Token ID | 4826 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
horizont | +0.390 |
MK | +0.345 |
Dash | +0.299 |
MK | +0.292 |
HIP | +0.292 |
Neg logit contributions | |
McCartney | -0.324 |
aba | -0.317 |
Reloaded | -0.311 |
igible | -0.310 |
complement | -0.309 |
Token | as |
Token ID | 292 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
horizont | +0.800 |
MK | +0.692 |
Dash | +0.592 |
HIP | +0.582 |
MK | +0.581 |
Neg logit contributions | |
McCartney | -0.769 |
aba | -0.754 |
Reloaded | -0.745 |
igible | -0.738 |
complement | -0.734 |
Token | Chrom |
Token ID | 18255 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
McCartney | +0.044 |
aba | +0.043 |
Reloaded | +0.042 |
igible | +0.042 |
complement | +0.042 |
Neg logit contributions | |
horizont | -0.050 |
MK | -0.043 |
Dash | -0.037 |
HIP | -0.037 |
MK | -0.036 |
Token | ecast |
Token ID | 43299 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
horizont | +0.301 |
MK | +0.270 |
Dash | +0.236 |
MK | +0.235 |
HIP | +0.233 |
Neg logit contributions | |
McCartney | -0.206 |
aba | -0.199 |
Reloaded | -0.198 |
complement | -0.196 |
igible | -0.195 |
Token | or |
Token ID | 393 |
Feature activation | -9.17e-02 |
Pos logit contributions | |
horizont | +0.241 |
MK | +0.210 |
Dash | +0.180 |
MK | +0.179 |
HIP | +0.177 |
Neg logit contributions | |
McCartney | -0.233 |
aba | -0.228 |
Reloaded | -0.225 |
igible | -0.222 |
irs | -0.221 |
Token | Fire |
Token ID | 3764 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
horizont | +0.069 |
MK | +0.059 |
Dash | +0.051 |
HIP | +0.050 |
MK | +0.050 |
Neg logit contributions | |
McCartney | -0.061 |
aba | -0.060 |
Reloaded | -0.059 |
igible | -0.058 |
complement | -0.058 |
Token | St |
Token ID | 1273 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
horizont | +0.064 |
MK | +0.055 |
Dash | +0.048 |
HIP | +0.048 |
MK | +0.047 |
Neg logit contributions | |
McCartney | -0.051 |
aba | -0.051 |
Reloaded | -0.050 |
igible | -0.049 |
complement | -0.049 |
Token | ick |
Token ID | 624 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
horizont | +0.404 |
MK | +0.357 |
Dash | +0.307 |
MK | +0.305 |
HIP | +0.304 |
Neg logit contributions | |
McCartney | -0.349 |
aba | -0.339 |
Reloaded | -0.338 |
igible | -0.334 |
complement | -0.333 |
Token | . |
Token ID | 13 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
horizont | +1.925 |
MK | +1.694 |
Dash | +1.451 |
MK | +1.441 |
HIP | +1.423 |
Neg logit contributions | |
McCartney | -1.894 |
aba | -1.847 |
Reloaded | -1.814 |
herty | -1.788 |
igible | -1.787 |
Token |
|
Token ID | 198 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
McCartney | +0.158 |
aba | +0.155 |
Reloaded | +0.153 |
igible | +0.152 |
complement | +0.151 |
Neg logit contributions | |
horizont | -0.151 |
MK | -0.129 |
Dash | -0.110 |
HIP | -0.108 |
MK | -0.108 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
horizont | +0.015 |
MK | +0.012 |
Dash | +0.010 |
HIP | +0.010 |
MK | +0.010 |
Neg logit contributions | |
McCartney | -0.021 |
aba | -0.020 |
Reloaded | -0.020 |
igible | -0.020 |
complement | -0.020 |
Token | Check |
Token ID | 9787 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
horizont | +0.080 |
MK | +0.070 |
Dash | +0.061 |
HIP | +0.060 |
MK | +0.060 |
Neg logit contributions | |
McCartney | -0.064 |
aba | -0.063 |
Reloaded | -0.062 |
igible | -0.061 |
complement | -0.061 |
Token | out |
Token ID | 448 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
McCartney | +0.022 |
aba | +0.022 |
Reloaded | +0.021 |
igible | +0.021 |
irs | +0.021 |
Neg logit contributions | |
horizont | -0.026 |
MK | -0.023 |
Dash | -0.020 |
HIP | -0.020 |
MK | -0.019 |
Token | largest |
Token ID | 4387 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
horizont | +0.046 |
MK | +0.040 |
Dash | +0.035 |
HIP | +0.035 |
MK | +0.035 |
Neg logit contributions | |
McCartney | -0.034 |
aba | -0.033 |
Reloaded | -0.033 |
igible | -0.032 |
irs | -0.032 |
Token | party |
Token ID | 2151 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
McCartney | +0.519 |
aba | +0.507 |
Reloaded | +0.497 |
igible | +0.492 |
complement | +0.492 |
Neg logit contributions | |
horizont | -0.686 |
MK | -0.606 |
Dash | -0.529 |
HIP | -0.520 |
MK | -0.518 |
Token | abandoned |
Token ID | 9958 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
McCartney | +0.239 |
aba | +0.235 |
Reloaded | +0.230 |
igible | +0.228 |
complement | +0.227 |
Neg logit contributions | |
horizont | -0.288 |
MK | -0.252 |
Dash | -0.218 |
HIP | -0.215 |
MK | -0.213 |
Token | President |
Token ID | 1992 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
horizont | +0.271 |
MK | +0.238 |
Dash | +0.208 |
HIP | +0.207 |
MK | +0.207 |
Neg logit contributions | |
McCartney | -0.196 |
aba | -0.193 |
Reloaded | -0.190 |
igible | -0.188 |
irs | -0.187 |
Token | Dil |
Token ID | 29328 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
horizont | +0.252 |
MK | +0.218 |
Dash | +0.191 |
HIP | +0.189 |
MK | +0.189 |
Neg logit contributions | |
McCartney | -0.194 |
Reloaded | -0.190 |
aba | -0.189 |
igible | -0.187 |
herty | -0.185 |
Token | ma |
Token ID | 2611 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
horizont | +0.422 |
MK | +0.368 |
Dash | +0.318 |
HIP | +0.313 |
MK | +0.313 |
Neg logit contributions | |
McCartney | -0.368 |
Reloaded | -0.359 |
aba | -0.359 |
igible | -0.353 |
complement | -0.352 |
Token | Rousse |
Token ID | 42849 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
horizont | +2.029 |
MK | +1.794 |
SOU | +1.548 |
HIP | +1.541 |
MK | +1.539 |
Neg logit contributions | |
McCartney | -1.752 |
keyes | -1.726 |
igible | -1.722 |
aba | -1.718 |
Reloaded | -1.690 |
Token | ff |
Token ID | 487 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
McCartney | +0.277 |
Reloaded | +0.270 |
aba | +0.267 |
igible | +0.266 |
CRC | +0.260 |
Neg logit contributions | |
horizont | -0.394 |
MK | -0.352 |
Dash | -0.307 |
MK | -0.305 |
HIP | -0.303 |
Token | � |
Token ID | 447 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
horizont | +1.000 |
MK | +0.877 |
Dash | +0.758 |
MK | +0.755 |
HIP | +0.752 |
Neg logit contributions | |
McCartney | -0.808 |
aba | -0.790 |
Reloaded | -0.790 |
igible | -0.777 |
irs | -0.770 |
Token | � |
Token ID | 247 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
McCartney | +0.033 |
aba | +0.032 |
Reloaded | +0.031 |
igible | +0.031 |
complement | +0.031 |
Neg logit contributions | |
horizont | -0.045 |
MK | -0.040 |
Dash | -0.035 |
HIP | -0.034 |
MK | -0.034 |
Token | s |
Token ID | 82 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
horizont | +0.238 |
MK | +0.202 |
Dash | +0.169 |
HIP | +0.166 |
MK | +0.166 |
Neg logit contributions | |
McCartney | -0.281 |
aba | -0.276 |
Reloaded | -0.274 |
igible | -0.271 |
complement | -0.270 |
Token | bit |
Token ID | 2545 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
horizont | +0.195 |
MK | +0.175 |
Dash | +0.150 |
MK | +0.148 |
HIP | +0.148 |
Neg logit contributions | |
McCartney | -0.173 |
Reloaded | -0.169 |
aba | -0.169 |
igible | -0.166 |
irs | -0.164 |
Token | 256 |
Token ID | 17759 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
horizont | +1.072 |
MK | +0.955 |
HIP | +0.848 |
Dash | +0.846 |
MK | +0.843 |
Neg logit contributions | |
McCartney | -0.680 |
aba | -0.669 |
Reloaded | -0.663 |
igible | -0.652 |
irs | -0.650 |
Token | - |
Token ID | 12 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
horizont | +0.186 |
MK | +0.164 |
MK | +0.139 |
HIP | +0.138 |
Dash | +0.138 |
Neg logit contributions | |
McCartney | -0.194 |
aba | -0.192 |
Reloaded | -0.191 |
irs | -0.188 |
igible | -0.188 |
Token | bit |
Token ID | 2545 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
horizont | +0.208 |
MK | +0.187 |
Dash | +0.161 |
MK | +0.159 |
HIP | +0.159 |
Neg logit contributions | |
McCartney | -0.187 |
Reloaded | -0.183 |
aba | -0.182 |
igible | -0.179 |
irs | -0.177 |
Token | R |
Token ID | 371 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
horizont | +1.105 |
MK | +0.985 |
HIP | +0.874 |
Dash | +0.871 |
MK | +0.869 |
Neg logit contributions | |
McCartney | -0.722 |
aba | -0.711 |
Reloaded | -0.705 |
igible | -0.692 |
irs | -0.690 |
Token | OP |
Token ID | 3185 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
horizont | +0.744 |
MK | +0.649 |
Dash | +0.540 |
HIP | +0.532 |
SOU | +0.524 |
Neg logit contributions | |
McCartney | -0.955 |
stadt | -0.941 |
aba | -0.938 |
igible | -0.937 |
Reloaded | -0.934 |
Token | s |
Token ID | 82 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
horizont | +1.471 |
MK | +1.231 |
MK | +1.005 |
HIP | +1.001 |
Dash | +1.000 |
Neg logit contributions | |
aba | -2.292 |
McCartney | -2.291 |
Reloaded | -2.264 |
igible | -2.248 |
irs | -2.247 |
Token | 32 |
Token ID | 3933 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
horizont | +0.583 |
MK | +0.522 |
Dash | +0.466 |
HIP | +0.463 |
MK | +0.461 |
Neg logit contributions | |
McCartney | -0.305 |
aba | -0.298 |
Reloaded | -0.292 |
igible | -0.288 |
irs | -0.286 |
Token | 32 |
Token ID | 3933 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
McCartney | +0.248 |
aba | +0.243 |
Reloaded | +0.238 |
igible | +0.236 |
complement | +0.235 |
Neg logit contributions | |
horizont | -0.359 |
MK | -0.317 |
Dash | -0.279 |
HIP | -0.275 |
MK | -0.274 |
Token | 56 |
Token ID | 7265 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
McCartney | +0.185 |
aba | +0.180 |
Reloaded | +0.176 |
igible | +0.174 |
irs | +0.173 |
Neg logit contributions | |
horizont | -0.343 |
MK | -0.307 |
Dash | -0.273 |
HIP | -0.271 |
MK | -0.270 |
Token | 48 |
Token ID | 4764 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
McCartney | +0.310 |
aba | +0.302 |
Reloaded | +0.296 |
igible | +0.292 |
complement | +0.290 |
Neg logit contributions | |
horizont | -0.607 |
MK | -0.543 |
Dash | -0.485 |
HIP | -0.480 |
MK | -0.479 |
Token | referred |
Token ID | 6412 |
Feature activation | -9.75e-02 |
Pos logit contributions | |
McCartney | +0.042 |
aba | +0.041 |
Reloaded | +0.041 |
igible | +0.040 |
irs | +0.040 |
Neg logit contributions | |
horizont | -0.055 |
MK | -0.048 |
Dash | -0.042 |
HIP | -0.042 |
MK | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
horizont | +0.073 |
MK | +0.064 |
Dash | +0.055 |
HIP | +0.055 |
MK | +0.055 |
Neg logit contributions | |
McCartney | -0.064 |
aba | -0.063 |
Reloaded | -0.062 |
igible | -0.061 |
irs | -0.061 |
Token | as |
Token ID | 355 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
horizont | +0.259 |
MK | +0.225 |
Dash | +0.194 |
HIP | +0.192 |
MK | +0.191 |
Neg logit contributions | |
McCartney | -0.234 |
aba | -0.231 |
Reloaded | -0.228 |
igible | -0.226 |
irs | -0.224 |
Token | " |
Token ID | 366 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
horizont | +0.515 |
MK | +0.457 |
Dash | +0.402 |
HIP | +0.399 |
MK | +0.398 |
Neg logit contributions | |
McCartney | -0.340 |
aba | -0.334 |
Reloaded | -0.329 |
igible | -0.324 |
irs | -0.322 |
Token | John |
Token ID | 7554 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
horizont | +0.543 |
MK | +0.483 |
Dash | +0.422 |
HIP | +0.411 |
MK | +0.411 |
Neg logit contributions | |
McCartney | -0.436 |
aba | -0.425 |
Reloaded | -0.418 |
igible | -0.416 |
complement | -0.415 |
Token | Doe |
Token ID | 31780 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
McCartney | +0.146 |
aba | +0.144 |
Reloaded | +0.142 |
igible | +0.140 |
complement | +0.139 |
Neg logit contributions | |
horizont | -0.162 |
MK | -0.141 |
Dash | -0.122 |
HIP | -0.120 |
MK | -0.120 |
Token | " |
Token ID | 1 |
Feature activation | -5.97e-03 |
Pos logit contributions | |
horizont | +1.883 |
MK | +1.644 |
Dash | +1.417 |
MK | +1.407 |
HIP | +1.397 |
Neg logit contributions | |
McCartney | -1.853 |
Reloaded | -1.842 |
aba | -1.826 |
igible | -1.803 |
irs | -1.783 |
Token | contends |
Token ID | 33799 |
Feature activation | -7.16e-03 |
Pos logit contributions | |
horizont | +0.005 |
MK | +0.004 |
Dash | +0.004 |
HIP | +0.003 |
MK | +0.003 |
Neg logit contributions | |
McCartney | -0.004 |
aba | -0.004 |
Reloaded | -0.004 |
igible | -0.004 |
irs | -0.004 |
Token | he |
Token ID | 339 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
horizont | +0.006 |
MK | +0.005 |
Dash | +0.004 |
HIP | +0.004 |
MK | +0.004 |
Neg logit contributions | |
McCartney | -0.004 |
aba | -0.004 |
Reloaded | -0.004 |
igible | -0.004 |
complement | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
horizont | +0.103 |
MK | +0.090 |
Dash | +0.078 |
HIP | +0.077 |
MK | +0.077 |
Neg logit contributions | |
McCartney | -0.082 |
aba | -0.081 |
Reloaded | -0.080 |
igible | -0.079 |
irs | -0.078 |
Token | denied |
Token ID | 6699 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
horizont | +0.040 |
MK | +0.035 |
Dash | +0.031 |
HIP | +0.030 |
MK | +0.030 |
Neg logit contributions | |
McCartney | -0.030 |
aba | -0.029 |
Reloaded | -0.029 |
complement | -0.029 |
igible | -0.029 |
Token | both |
Token ID | 1111 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
McCartney | +0.157 |
aba | +0.154 |
Reloaded | +0.152 |
igible | +0.150 |
irs | +0.149 |
Neg logit contributions | |
horizont | -0.230 |
MK | -0.203 |
Dash | -0.178 |
HIP | -0.177 |
MK | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
McCartney | +0.364 |
aba | +0.356 |
Reloaded | +0.350 |
igible | +0.347 |
complement | +0.345 |
Neg logit contributions | |
horizont | -0.454 |
MK | -0.397 |
Dash | -0.345 |
HIP | -0.341 |
MK | -0.339 |
Token |
|
Token ID | 198 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
McCartney | +0.262 |
aba | +0.258 |
Reloaded | +0.255 |
igible | +0.253 |
irs | +0.252 |
Neg logit contributions | |
horizont | -0.223 |
MK | -0.192 |
Dash | -0.160 |
HIP | -0.159 |
MK | -0.158 |
Token |
|
Token ID | 198 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
McCartney | +0.432 |
aba | +0.425 |
Reloaded | +0.419 |
igible | +0.416 |
complement | +0.415 |
Neg logit contributions | |
horizont | -0.315 |
MK | -0.261 |
Dash | -0.214 |
HIP | -0.210 |
MK | -0.209 |
Token | Al |
Token ID | 2348 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
McCartney | +0.247 |
aba | +0.240 |
Reloaded | +0.236 |
igible | +0.234 |
complement | +0.233 |
Neg logit contributions | |
horizont | -0.317 |
MK | -0.282 |
Dash | -0.246 |
HIP | -0.241 |
MK | -0.240 |
Token | ma |
Token ID | 2611 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
horizont | +0.211 |
MK | +0.180 |
Dash | +0.152 |
HIP | +0.150 |
MK | +0.149 |
Neg logit contributions | |
McCartney | -0.232 |
aba | -0.228 |
Reloaded | -0.226 |
igible | -0.223 |
complement | -0.222 |
Token | from |
Token ID | 422 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
horizont | +2.070 |
MK | +1.858 |
Dash | +1.614 |
HIP | +1.605 |
MK | +1.598 |
Neg logit contributions | |
McCartney | -1.597 |
aba | -1.559 |
igible | -1.532 |
Reloaded | -1.529 |
keyes | -1.527 |
Token | Ex |
Token ID | 1475 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
McCartney | +0.287 |
aba | +0.281 |
Reloaded | +0.277 |
igible | +0.274 |
irs | +0.272 |
Neg logit contributions | |
horizont | -0.381 |
MK | -0.335 |
Dash | -0.292 |
HIP | -0.290 |
MK | -0.289 |
Token | Lib |
Token ID | 7980 |
Feature activation | -7.74e-02 |
Pos logit contributions | |
McCartney | +0.457 |
aba | +0.450 |
Reloaded | +0.441 |
igible | +0.438 |
irs | +0.436 |
Neg logit contributions | |
horizont | -0.513 |
MK | -0.445 |
Dash | -0.383 |
HIP | -0.378 |
MK | -0.377 |
Token | ris |
Token ID | 2442 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
horizont | +0.069 |
MK | +0.061 |
Dash | +0.054 |
HIP | +0.054 |
MK | +0.054 |
Neg logit contributions | |
McCartney | -0.040 |
aba | -0.039 |
Reloaded | -0.039 |
igible | -0.038 |
complement | -0.038 |
Token | has |
Token ID | 468 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
McCartney | +0.502 |
aba | +0.492 |
Reloaded | +0.484 |
igible | +0.479 |
complement | +0.477 |
Neg logit contributions | |
horizont | -0.602 |
MK | -0.525 |
Dash | -0.455 |
HIP | -0.449 |
MK | -0.447 |
Token | the |
Token ID | 262 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
McCartney | +0.135 |
aba | +0.133 |
Reloaded | +0.131 |
igible | +0.129 |
irs | +0.128 |
Neg logit contributions | |
horizont | -0.185 |
MK | -0.162 |
Dash | -0.142 |
HIP | -0.141 |
MK | -0.141 |
Token | so |
Token ID | 523 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
horizont | +0.053 |
MK | +0.047 |
Dash | +0.041 |
HIP | +0.041 |
MK | +0.040 |
Neg logit contributions | |
McCartney | -0.039 |
aba | -0.038 |
Reloaded | -0.038 |
igible | -0.037 |
complement | -0.037 |
Token | - |
Token ID | 12 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
McCartney | +0.284 |
aba | +0.277 |
Reloaded | +0.276 |
complement | +0.271 |
stadt | +0.271 |
Neg logit contributions | |
horizont | -0.277 |
MK | -0.240 |
Dash | -0.204 |
HIP | -0.200 |
MK | -0.200 |
Token | called |
Token ID | 7174 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
horizont | +0.160 |
MK | +0.138 |
Dash | +0.118 |
HIP | +0.116 |
MK | +0.116 |
Neg logit contributions | |
McCartney | -0.155 |
aba | -0.153 |
Reloaded | -0.150 |
igible | -0.149 |
irs | -0.148 |
Token | John |
Token ID | 1757 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
horizont | +0.779 |
MK | +0.687 |
Dash | +0.604 |
HIP | +0.598 |
MK | +0.596 |
Neg logit contributions | |
McCartney | -0.542 |
aba | -0.531 |
Reloaded | -0.523 |
igible | -0.516 |
complement | -0.512 |
Token | Doe |
Token ID | 31780 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
McCartney | +0.176 |
aba | +0.172 |
Reloaded | +0.168 |
igible | +0.167 |
irs | +0.166 |
Neg logit contributions | |
horizont | -0.213 |
MK | -0.185 |
Dash | -0.161 |
HIP | -0.159 |
MK | -0.158 |
Token | investigation |
Token ID | 3645 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
horizont | +1.937 |
MK | +1.685 |
Dash | +1.453 |
HIP | +1.442 |
MK | +1.442 |
Neg logit contributions | |
McCartney | -1.770 |
aba | -1.735 |
Reloaded | -1.734 |
igible | -1.700 |
irs | -1.688 |
Token | into |
Token ID | 656 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
horizont | +0.130 |
MK | +0.112 |
Dash | +0.096 |
HIP | +0.094 |
MK | +0.094 |
Neg logit contributions | |
McCartney | -0.130 |
aba | -0.128 |
Reloaded | -0.126 |
igible | -0.125 |
irs | -0.124 |
Token | alleged |
Token ID | 4260 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
McCartney | +0.270 |
aba | +0.265 |
Reloaded | +0.261 |
igible | +0.258 |
complement | +0.256 |
Neg logit contributions | |
horizont | -0.358 |
MK | -0.314 |
Dash | -0.274 |
HIP | -0.272 |
MK | -0.271 |
Token | campaign |
Token ID | 1923 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
McCartney | +0.459 |
aba | +0.449 |
complement | +0.436 |
igible | +0.435 |
Reloaded | +0.435 |
Neg logit contributions | |
horizont | -0.591 |
MK | -0.516 |
Dash | -0.450 |
ivably | -0.439 |
HIP | -0.438 |
Token | - |
Token ID | 12 |
Feature activation | +2.93e-03 |
Pos logit contributions | |
McCartney | +0.129 |
aba | +0.127 |
Reloaded | +0.123 |
complement | +0.123 |
igible | +0.122 |
Neg logit contributions | |
horizont | -0.185 |
MK | -0.164 |
Dash | -0.143 |
HIP | -0.141 |
MK | -0.140 |
Token | sweats |
Token ID | 39387 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
horizont | +0.527 |
MK | +0.464 |
Dash | +0.408 |
HIP | +0.405 |
MK | +0.404 |
Neg logit contributions | |
McCartney | -0.359 |
aba | -0.353 |
Reloaded | -0.347 |
igible | -0.343 |
complement | -0.340 |
Token | h |
Token ID | 71 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
horizont | +0.047 |
MK | +0.040 |
Dash | +0.034 |
HIP | +0.033 |
MK | +0.033 |
Neg logit contributions | |
McCartney | -0.048 |
aba | -0.048 |
Reloaded | -0.047 |
igible | -0.046 |
complement | -0.046 |
Token | irts |
Token ID | 9682 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
horizont | +0.137 |
MK | +0.125 |
Dash | +0.113 |
MK | +0.111 |
HIP | +0.111 |
Neg logit contributions | |
McCartney | -0.052 |
Reloaded | -0.050 |
aba | -0.050 |
igible | -0.048 |
stadt | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -6.78e-03 |
Pos logit contributions | |
horizont | +0.152 |
MK | +0.131 |
Dash | +0.112 |
HIP | +0.110 |
MK | +0.110 |
Neg logit contributions | |
McCartney | -0.147 |
aba | -0.144 |
Reloaded | -0.143 |
igible | -0.141 |
irs | -0.140 |
Token | M |
Token ID | 337 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
horizont | +0.005 |
MK | +0.005 |
Dash | +0.004 |
HIP | +0.004 |
MK | +0.004 |
Neg logit contributions | |
McCartney | -0.004 |
aba | -0.004 |
Reloaded | -0.004 |
igible | -0.004 |
complement | -0.004 |
Token | ardi |
Token ID | 22490 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
McCartney | +0.130 |
aba | +0.127 |
Reloaded | +0.126 |
igible | +0.124 |
complement | +0.124 |
Neg logit contributions | |
horizont | -0.137 |
MK | -0.119 |
Dash | -0.102 |
HIP | -0.100 |
MK | -0.100 |
Token | Gr |
Token ID | 1902 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
horizont | +1.535 |
MK | +1.293 |
Dash | +1.075 |
MK | +1.043 |
HIP | +1.032 |
Neg logit contributions | |
McCartney | -2.162 |
aba | -2.129 |
Reloaded | -2.125 |
igible | -2.094 |
stadt | -2.092 |
Token | as |
Token ID | 292 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
McCartney | +0.739 |
aba | +0.726 |
Reloaded | +0.719 |
igible | +0.710 |
complement | +0.710 |
Neg logit contributions | |
horizont | -0.636 |
MK | -0.541 |
Dash | -0.454 |
HIP | -0.446 |
MK | -0.445 |
Token | beads |
Token ID | 36116 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
McCartney | +0.151 |
aba | +0.148 |
Reloaded | +0.145 |
igible | +0.143 |
complement | +0.143 |
Neg logit contributions | |
horizont | -0.212 |
MK | -0.187 |
Dash | -0.164 |
HIP | -0.162 |
MK | -0.161 |
Token | and |
Token ID | 290 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
McCartney | +0.205 |
aba | +0.201 |
Reloaded | +0.198 |
igible | +0.196 |
irs | +0.195 |
Neg logit contributions | |
horizont | -0.215 |
MK | -0.186 |
Dash | -0.159 |
HIP | -0.157 |
MK | -0.156 |
Token | CD |
Token ID | 6458 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
McCartney | +0.024 |
aba | +0.024 |
Reloaded | +0.023 |
igible | +0.023 |
complement | +0.023 |
Neg logit contributions | |
horizont | -0.031 |
MK | -0.027 |
Dash | -0.024 |
HIP | -0.023 |
MK | -0.023 |
Token | as |
Token ID | 355 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
McCartney | +0.041 |
aba | +0.040 |
Reloaded | +0.040 |
igible | +0.039 |
stadt | +0.039 |
Neg logit contributions | |
horizont | -0.040 |
MK | -0.034 |
Dash | -0.029 |
HIP | -0.029 |
MK | -0.029 |
Token | follows |
Token ID | 5679 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
horizont | +0.229 |
MK | +0.201 |
Dash | +0.176 |
HIP | +0.174 |
MK | +0.173 |
Neg logit contributions | |
McCartney | -0.185 |
aba | -0.180 |
Reloaded | -0.179 |
igible | -0.177 |
stadt | -0.175 |
Token | . |
Token ID | 13 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
horizont | +1.092 |
MK | +0.923 |
Dash | +0.792 |
HIP | +0.778 |
MK | +0.773 |
Neg logit contributions | |
McCartney | -1.193 |
Reloaded | -1.173 |
aba | -1.165 |
igible | -1.144 |
stadt | -1.144 |
Token |
|
Token ID | 198 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
horizont | +0.197 |
MK | +0.166 |
Dash | +0.137 |
HIP | +0.135 |
MK | +0.134 |
Neg logit contributions | |
McCartney | -0.258 |
aba | -0.253 |
Reloaded | -0.251 |
igible | -0.249 |
complement | -0.247 |
Token |
|
Token ID | 198 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
horizont | +0.168 |
MK | +0.142 |
Dash | +0.115 |
HIP | +0.113 |
MK | +0.113 |
Neg logit contributions | |
McCartney | -0.244 |
aba | -0.240 |
Reloaded | -0.238 |
igible | -0.236 |
complement | -0.235 |
Token | # |
Token ID | 2 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
horizont | +0.188 |
MK | +0.160 |
Dash | +0.133 |
HIP | +0.130 |
MK | +0.130 |
Neg logit contributions | |
McCartney | -0.230 |
aba | -0.225 |
Reloaded | -0.223 |
igible | -0.221 |
complement | -0.220 |
Token | !/ |
Token ID | 48443 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
horizont | +1.700 |
MK | +1.457 |
Dash | +1.219 |
MK | +1.196 |
HIP | +1.195 |
Neg logit contributions | |
McCartney | -2.015 |
Reloaded | -1.952 |
aba | -1.950 |
igible | -1.918 |
stadt | -1.909 |
Token | bin |
Token ID | 8800 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
horizont | +0.320 |
MK | +0.282 |
Dash | +0.248 |
HIP | +0.245 |
MK | +0.244 |
Neg logit contributions | |
McCartney | -0.218 |
aba | -0.214 |
Reloaded | -0.209 |
igible | -0.207 |
irs | -0.206 |
Token | / |
Token ID | 14 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
horizont | +0.782 |
MK | +0.657 |
Dash | +0.526 |
HIP | +0.512 |
MK | +0.511 |
Neg logit contributions | |
McCartney | -1.296 |
aba | -1.265 |
Reloaded | -1.261 |
igible | -1.253 |
stadt | -1.246 |
Token | sh |
Token ID | 1477 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
horizont | +0.182 |
MK | +0.158 |
HIP | +0.138 |
Dash | +0.137 |
MK | +0.136 |
Neg logit contributions | |
McCartney | -0.147 |
aba | -0.146 |
irs | -0.141 |
Reloaded | -0.141 |
igible | -0.141 |
Token | / |
Token ID | 1220 |
Feature activation | -9.63e-01 |
Pos logit contributions | |
horizont | +0.711 |
MK | +0.614 |
Dash | +0.525 |
HIP | +0.517 |
MK | +0.516 |
Neg logit contributions | |
McCartney | -0.691 |
aba | -0.678 |
Reloaded | -0.670 |
igible | -0.662 |
complement | -0.658 |
Token | / |
Token ID | 14 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
McCartney | +0.019 |
aba | +0.019 |
Reloaded | +0.019 |
igible | +0.019 |
complement | +0.019 |
Neg logit contributions | |
horizont | -0.013 |
MK | -0.011 |
Dash | -0.009 |
HIP | -0.009 |
MK | -0.009 |
Token | any |
Token ID | 1092 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
horizont | +0.129 |
MK | +0.108 |
Dash | +0.090 |
HIP | +0.088 |
MK | +0.088 |
Neg logit contributions | |
McCartney | -0.162 |
aba | -0.160 |
Reloaded | -0.157 |
igible | -0.156 |
complement | -0.156 |
Token | Mode |
Token ID | 10363 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
horizont | +0.371 |
MK | +0.329 |
Dash | +0.290 |
HIP | +0.287 |
MK | +0.286 |
Neg logit contributions | |
McCartney | -0.238 |
aba | -0.233 |
Reloaded | -0.228 |
igible | -0.225 |
complement | -0.224 |
Token | : |
Token ID | 25 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
McCartney | +0.056 |
aba | +0.055 |
Reloaded | +0.054 |
igible | +0.054 |
irs | +0.053 |
Neg logit contributions | |
horizont | -0.053 |
MK | -0.047 |
Dash | -0.040 |
MK | -0.039 |
HIP | -0.039 |
Token | Man |
Token ID | 5124 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
McCartney | +0.024 |
aba | +0.023 |
Reloaded | +0.023 |
igible | +0.023 |
complement | +0.023 |
Neg logit contributions | |
horizont | -0.026 |
MK | -0.023 |
Dash | -0.019 |
HIP | -0.019 |
MK | -0.019 |
Token | aged |
Token ID | 1886 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
horizont | +0.797 |
MK | +0.724 |
Dash | +0.628 |
MK | +0.624 |
ivably | +0.623 |
Neg logit contributions | |
McCartney | -0.626 |
Reloaded | -0.610 |
aba | -0.601 |
complement | -0.597 |
CRC | -0.588 |
Token | Access |
Token ID | 8798 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
horizont | +1.539 |
MK | +1.355 |
HIP | +1.147 |
MK | +1.142 |
Dash | +1.117 |
Neg logit contributions | |
McCartney | -2.097 |
Reloaded | -2.063 |
aba | -2.059 |
igible | -2.031 |
irs | -2.020 |
Token | Point |
Token ID | 6252 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
horizont | +1.356 |
MK | +1.223 |
HIP | +1.070 |
Dash | +1.067 |
MK | +1.065 |
Neg logit contributions | |
McCartney | -1.056 |
Reloaded | -1.043 |
aba | -1.039 |
igible | -1.015 |
irs | -1.012 |
Token | : |
Token ID | 25 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
horizont | +0.560 |
MK | +0.502 |
Dash | +0.424 |
MK | +0.422 |
HIP | +0.420 |
Neg logit contributions | |
McCartney | -0.642 |
aba | -0.630 |
Reloaded | -0.627 |
igible | -0.619 |
complement | -0.618 |
Token | Not |
Token ID | 1892 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
McCartney | +0.056 |
aba | +0.055 |
Reloaded | +0.054 |
igible | +0.054 |
complement | +0.054 |
Neg logit contributions | |
horizont | -0.066 |
MK | -0.058 |
Dash | -0.050 |
HIP | -0.049 |
MK | -0.049 |
Token | - |
Token ID | 12 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
horizont | +0.078 |
MK | +0.068 |
Dash | +0.058 |
HIP | +0.058 |
MK | +0.057 |
Neg logit contributions | |
McCartney | -0.080 |
aba | -0.079 |
Reloaded | -0.078 |
igible | -0.077 |
complement | -0.077 |
Token | Second |
Token ID | 5498 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
horizont | +0.072 |
MK | +0.062 |
Dash | +0.054 |
MK | +0.053 |
HIP | +0.053 |
Neg logit contributions | |
McCartney | -0.067 |
aba | -0.066 |
Reloaded | -0.066 |
igible | -0.065 |
irs | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
horizont | +0.174 |
MK | +0.151 |
Dash | +0.130 |
HIP | +0.128 |
MK | +0.128 |
Neg logit contributions | |
McCartney | -0.162 |
aba | -0.159 |
Reloaded | -0.157 |
igible | -0.155 |
irs | -0.154 |
Token | and |
Token ID | 290 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
horizont | +0.082 |
MK | +0.071 |
Dash | +0.060 |
HIP | +0.059 |
MK | +0.059 |
Neg logit contributions | |
McCartney | -0.085 |
aba | -0.084 |
Reloaded | -0.083 |
igible | -0.082 |
complement | -0.082 |
Token | Fourth |
Token ID | 15692 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
horizont | +0.270 |
MK | +0.229 |
Dash | +0.192 |
HIP | +0.189 |
MK | +0.188 |
Neg logit contributions | |
McCartney | -0.319 |
aba | -0.314 |
Reloaded | -0.310 |
igible | -0.307 |
irs | -0.306 |
Token | Div |
Token ID | 4777 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
McCartney | +0.088 |
aba | +0.086 |
Reloaded | +0.084 |
igible | +0.083 |
irs | +0.082 |
Neg logit contributions | |
horizont | -0.155 |
MK | -0.138 |
Dash | -0.123 |
HIP | -0.122 |
MK | -0.121 |
Token | isions |
Token ID | 3279 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
horizont | +0.173 |
MK | +0.150 |
Dash | +0.129 |
HIP | +0.126 |
MK | +0.126 |
Neg logit contributions | |
McCartney | -0.165 |
aba | -0.162 |
Reloaded | -0.160 |
igible | -0.158 |
complement | -0.157 |
Token | all |
Token ID | 477 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
horizont | +1.829 |
MK | +1.579 |
Dash | +1.361 |
SOU | +1.351 |
MK | +1.349 |
Neg logit contributions | |
McCartney | -1.820 |
Reloaded | -1.802 |
aba | -1.777 |
igible | -1.759 |
irs | -1.747 |
Token | at |
Token ID | 379 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
horizont | +0.044 |
MK | +0.038 |
Dash | +0.033 |
HIP | +0.033 |
MK | +0.032 |
Neg logit contributions | |
McCartney | -0.040 |
aba | -0.040 |
Reloaded | -0.039 |
igible | -0.039 |
irs | -0.038 |
Token | once |
Token ID | 1752 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
horizont | +0.477 |
MK | +0.416 |
Dash | +0.358 |
HIP | +0.356 |
MK | +0.355 |
Neg logit contributions | |
McCartney | -0.416 |
aba | -0.408 |
Reloaded | -0.403 |
igible | -0.399 |
irs | -0.396 |
Token | on |
Token ID | 319 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
horizont | +0.264 |
MK | +0.228 |
Dash | +0.196 |
HIP | +0.193 |
MK | +0.193 |
Neg logit contributions | |
McCartney | -0.248 |
aba | -0.243 |
Reloaded | -0.241 |
igible | -0.238 |
irs | -0.237 |
Token | S |
Token ID | 311 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
horizont | +0.335 |
MK | +0.297 |
Dash | +0.261 |
HIP | +0.259 |
MK | +0.259 |
Neg logit contributions | |
McCartney | -0.226 |
aba | -0.221 |
Reloaded | -0.218 |
igible | -0.215 |
irs | -0.214 |
Token | pushed |
Token ID | 7121 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
horizont | +0.344 |
MK | +0.302 |
Dash | +0.264 |
MK | +0.263 |
HIP | +0.262 |
Neg logit contributions | |
McCartney | -0.245 |
aba | -0.242 |
Reloaded | -0.238 |
igible | -0.236 |
irs | -0.234 |
Token | them |
Token ID | 606 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
horizont | +0.786 |
MK | +0.687 |
Dash | +0.596 |
MK | +0.589 |
HIP | +0.588 |
Neg logit contributions | |
McCartney | -0.636 |
aba | -0.626 |
Reloaded | -0.617 |
igible | -0.611 |
irs | -0.606 |
Token | well |
Token ID | 880 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
horizont | +0.646 |
MK | +0.557 |
Dash | +0.474 |
MK | +0.471 |
HIP | +0.469 |
Neg logit contributions | |
McCartney | -0.617 |
aba | -0.608 |
Reloaded | -0.602 |
igible | -0.597 |
irs | -0.591 |
Token | beyond |
Token ID | 3675 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
McCartney | +0.073 |
aba | +0.072 |
Reloaded | +0.071 |
igible | +0.070 |
complement | +0.070 |
Neg logit contributions | |
horizont | -0.074 |
MK | -0.064 |
Dash | -0.055 |
HIP | -0.054 |
MK | -0.054 |
Token | 90 |
Token ID | 4101 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
horizont | +0.161 |
MK | +0.142 |
Dash | +0.125 |
MK | +0.125 |
HIP | +0.124 |
Neg logit contributions | |
McCartney | -0.101 |
aba | -0.100 |
Reloaded | -0.099 |
igible | -0.097 |
irs | -0.096 |
Token | dB |
Token ID | 36077 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
McCartney | +0.467 |
aba | +0.459 |
Reloaded | +0.453 |
igible | +0.450 |
irs | +0.449 |
Neg logit contributions | |
horizont | -0.313 |
MK | -0.257 |
Dash | -0.209 |
HIP | -0.205 |
MK | -0.202 |
Token | did |
Token ID | 750 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
horizont | +1.885 |
MK | +1.652 |
Dash | +1.410 |
MK | +1.398 |
HIP | +1.386 |
Neg logit contributions | |
McCartney | -1.709 |
aba | -1.680 |
Reloaded | -1.674 |
igible | -1.638 |
irs | -1.634 |
Token | their |
Token ID | 511 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
McCartney | +0.082 |
aba | +0.081 |
Reloaded | +0.080 |
igible | +0.079 |
irs | +0.079 |
Neg logit contributions | |
horizont | -0.088 |
MK | -0.077 |
Dash | -0.066 |
MK | -0.065 |
HIP | -0.065 |
Token | limitations |
Token ID | 11247 |
Feature activation | -8.24e-01 |
Pos logit contributions | |
horizont | +0.140 |
MK | +0.123 |
Dash | +0.107 |
MK | +0.106 |
HIP | +0.106 |
Neg logit contributions | |
McCartney | -0.104 |
aba | -0.102 |
Reloaded | -0.100 |
igible | -0.099 |
irs | -0.098 |
Token | become |
Token ID | 1716 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
horizont | +0.602 |
MK | +0.521 |
Dash | +0.446 |
HIP | +0.441 |
MK | +0.440 |
Neg logit contributions | |
McCartney | -0.558 |
aba | -0.548 |
Reloaded | -0.541 |
igible | -0.535 |
irs | -0.532 |
Token | audible |
Token ID | 35282 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
horizont | +0.206 |
MK | +0.180 |
Dash | +0.157 |
HIP | +0.155 |
MK | +0.155 |
Neg logit contributions | |
McCartney | -0.163 |
aba | -0.160 |
Reloaded | -0.157 |
igible | -0.156 |
irs | -0.155 |
Token | that |
Token ID | 326 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
horizont | +0.583 |
MK | +0.514 |
MK | +0.452 |
Dash | +0.450 |
SOU | +0.446 |
Neg logit contributions | |
McCartney | -0.403 |
aba | -0.399 |
Reloaded | -0.391 |
igible | -0.387 |
irs | -0.387 |
Token | moment |
Token ID | 2589 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
horizont | +0.503 |
MK | +0.438 |
Dash | +0.377 |
MK | +0.374 |
HIP | +0.374 |
Neg logit contributions | |
McCartney | -0.429 |
aba | -0.422 |
Reloaded | -0.416 |
igible | -0.411 |
irs | -0.408 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
horizont | +1.038 |
MK | +0.903 |
Dash | +0.770 |
MK | +0.769 |
HIP | +0.760 |
Neg logit contributions | |
McCartney | -0.978 |
aba | -0.966 |
Reloaded | -0.958 |
irs | -0.945 |
igible | -0.945 |
Token | the |
Token ID | 262 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
horizont | +0.108 |
MK | +0.095 |
Dash | +0.083 |
MK | +0.082 |
HIP | +0.082 |
Neg logit contributions | |
McCartney | -0.077 |
aba | -0.076 |
Reloaded | -0.075 |
igible | -0.074 |
irs | -0.073 |
Token | general |
Token ID | 2276 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
horizont | +0.251 |
MK | +0.220 |
Dash | +0.192 |
HIP | +0.189 |
MK | +0.189 |
Neg logit contributions | |
McCartney | -0.192 |
aba | -0.188 |
Reloaded | -0.185 |
igible | -0.183 |
complement | -0.182 |
Token | consensus |
Token ID | 11529 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
McCartney | +0.099 |
aba | +0.097 |
Reloaded | +0.095 |
igible | +0.094 |
complement | +0.094 |
Neg logit contributions | |
horizont | -0.123 |
MK | -0.108 |
Dash | -0.094 |
HIP | -0.092 |
MK | -0.092 |
Token | had |
Token ID | 550 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
horizont | +1.858 |
MK | +1.616 |
Dash | +1.384 |
MK | +1.382 |
HIP | +1.381 |
Neg logit contributions | |
McCartney | -1.708 |
aba | -1.690 |
Reloaded | -1.689 |
irs | -1.653 |
igible | -1.643 |
Token | been |
Token ID | 587 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
horizont | +0.179 |
MK | +0.156 |
Dash | +0.135 |
HIP | +0.134 |
MK | +0.133 |
Neg logit contributions | |
McCartney | -0.148 |
aba | -0.145 |
Reloaded | -0.143 |
igible | -0.141 |
complement | -0.141 |
Token | that |
Token ID | 326 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
horizont | +0.337 |
MK | +0.297 |
Dash | +0.261 |
HIP | +0.258 |
MK | +0.257 |
Neg logit contributions | |
McCartney | -0.233 |
aba | -0.229 |
Reloaded | -0.225 |
igible | -0.222 |
irs | -0.221 |
Token | regulation |
Token ID | 9001 |
Feature activation | -9.41e-03 |
Pos logit contributions | |
horizont | +0.189 |
MK | +0.167 |
Dash | +0.146 |
MK | +0.145 |
HIP | +0.145 |
Neg logit contributions | |
McCartney | -0.131 |
aba | -0.128 |
Reloaded | -0.127 |
igible | -0.125 |
irs | -0.124 |
Token | was |
Token ID | 373 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
horizont | +0.007 |
MK | +0.006 |
Dash | +0.005 |
HIP | +0.005 |
MK | +0.005 |
Neg logit contributions | |
McCartney | -0.006 |
aba | -0.006 |
Reloaded | -0.006 |
igible | -0.006 |
irs | -0.006 |
Token | gender |
Token ID | 8388 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
horizont | +0.375 |
MK | +0.334 |
Dash | +0.295 |
HIP | +0.289 |
MK | +0.289 |
Neg logit contributions | |
McCartney | -0.258 |
aba | -0.250 |
Reloaded | -0.249 |
igible | -0.244 |
complement | -0.243 |
Token | comparison |
Token ID | 7208 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
horizont | +0.974 |
MK | +0.847 |
Dash | +0.734 |
HIP | +0.723 |
MK | +0.720 |
Neg logit contributions | |
McCartney | -0.873 |
aba | -0.855 |
Reloaded | -0.852 |
igible | -0.837 |
irs | -0.833 |
Token | , |
Token ID | 11 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
McCartney | +0.217 |
aba | +0.213 |
Reloaded | +0.210 |
igible | +0.208 |
complement | +0.207 |
Neg logit contributions | |
horizont | -0.234 |
MK | -0.203 |
Dash | -0.174 |
HIP | -0.172 |
MK | -0.171 |
Token | declaring |
Token ID | 18684 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
horizont | +0.270 |
MK | +0.238 |
Dash | +0.207 |
MK | +0.205 |
HIP | +0.205 |
Neg logit contributions | |
McCartney | -0.218 |
aba | -0.215 |
Reloaded | -0.213 |
igible | -0.209 |
irs | -0.208 |
Token | Do |
Token ID | 2141 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
horizont | +0.277 |
MK | +0.243 |
Dash | +0.212 |
HIP | +0.210 |
MK | +0.209 |
Neg logit contributions | |
McCartney | -0.213 |
aba | -0.208 |
Reloaded | -0.205 |
igible | -0.203 |
irs | -0.202 |
Token | lez |
Token ID | 36858 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
horizont | +0.451 |
MK | +0.393 |
Dash | +0.322 |
ivably | +0.316 |
HIP | +0.316 |
Neg logit contributions | |
McCartney | -0.609 |
Reloaded | -0.595 |
aba | -0.593 |
igible | -0.580 |
complement | -0.580 |
Token | al |
Token ID | 282 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
horizont | +1.662 |
MK | +1.490 |
Dash | +1.288 |
ivably | +1.273 |
MK | +1.241 |
Neg logit contributions | |
McCartney | -1.858 |
Reloaded | -1.819 |
aba | -1.762 |
igible | -1.760 |
complement | -1.750 |
Token | � |
Token ID | 564 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
horizont | +0.212 |
MK | +0.185 |
Dash | +0.160 |
HIP | +0.158 |
MK | +0.157 |
Neg logit contributions | |
McCartney | -0.186 |
aba | -0.183 |
Reloaded | -0.181 |
igible | -0.179 |
irs | -0.178 |
Token | � |
Token ID | 250 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
horizont | +0.242 |
MK | +0.221 |
Dash | +0.196 |
HIP | +0.193 |
MK | +0.193 |
Neg logit contributions | |
McCartney | -0.161 |
aba | -0.156 |
Reloaded | -0.156 |
igible | -0.153 |
complement | -0.152 |
Token | the |
Token ID | 1169 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
horizont | +0.497 |
MK | +0.435 |
Dash | +0.375 |
HIP | +0.368 |
MK | +0.366 |
Neg logit contributions | |
McCartney | -0.462 |
aba | -0.451 |
Reloaded | -0.448 |
igible | -0.441 |
complement | -0.440 |
Token | very |
Token ID | 845 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
horizont | +0.723 |
MK | +0.633 |
Dash | +0.551 |
HIP | +0.545 |
MK | +0.542 |
Neg logit contributions | |
McCartney | -0.590 |
aba | -0.579 |
Reloaded | -0.571 |
igible | -0.562 |
complement | -0.560 |
Token | made |
Token ID | 925 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
McCartney | +0.443 |
aba | +0.434 |
Reloaded | +0.424 |
igible | +0.423 |
complement | +0.423 |
Neg logit contributions | |
horizont | -0.520 |
MK | -0.454 |
Dash | -0.393 |
HIP | -0.386 |
MK | -0.384 |
Token | the |
Token ID | 262 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
McCartney | +0.078 |
aba | +0.076 |
Reloaded | +0.075 |
igible | +0.074 |
complement | +0.074 |
Neg logit contributions | |
horizont | -0.104 |
MK | -0.092 |
Dash | -0.080 |
HIP | -0.079 |
MK | -0.079 |
Token | proceeding |
Token ID | 18788 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
horizont | +0.012 |
MK | +0.010 |
Dash | +0.009 |
HIP | +0.009 |
MK | +0.009 |
Neg logit contributions | |
McCartney | -0.009 |
aba | -0.009 |
Reloaded | -0.009 |
igible | -0.009 |
irs | -0.008 |
Token | against |
Token ID | 1028 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
McCartney | +0.203 |
aba | +0.199 |
Reloaded | +0.195 |
complement | +0.194 |
igible | +0.193 |
Neg logit contributions | |
horizont | -0.260 |
MK | -0.229 |
Dash | -0.199 |
HIP | -0.196 |
MK | -0.195 |
Token | John |
Token ID | 1757 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
McCartney | +0.034 |
aba | +0.033 |
Reloaded | +0.033 |
igible | +0.032 |
irs | +0.032 |
Neg logit contributions | |
horizont | -0.047 |
MK | -0.041 |
Dash | -0.036 |
HIP | -0.036 |
MK | -0.036 |
Token | Doe |
Token ID | 31780 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
McCartney | +0.384 |
aba | +0.376 |
Reloaded | +0.369 |
igible | +0.366 |
complement | +0.365 |
Neg logit contributions | |
horizont | -0.441 |
MK | -0.383 |
Dash | -0.330 |
HIP | -0.326 |
MK | -0.324 |
Token | fundamentally |
Token ID | 17640 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
horizont | +1.991 |
MK | +1.751 |
Dash | +1.512 |
HIP | +1.507 |
MK | +1.501 |
Neg logit contributions | |
McCartney | -1.602 |
aba | -1.573 |
Reloaded | -1.553 |
igible | -1.529 |
irs | -1.518 |
Token | unfair |
Token ID | 11675 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
horizont | +0.319 |
MK | +0.279 |
Dash | +0.243 |
MK | +0.239 |
HIP | +0.239 |
Neg logit contributions | |
McCartney | -0.246 |
aba | -0.242 |
Reloaded | -0.237 |
complement | -0.235 |
igible | -0.235 |
Token | ," |
Token ID | 553 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
horizont | +0.249 |
MK | +0.214 |
Dash | +0.182 |
HIP | +0.180 |
MK | +0.179 |
Neg logit contributions | |
McCartney | -0.255 |
aba | -0.250 |
Reloaded | -0.247 |
igible | -0.245 |
irs | -0.243 |
Token | the |
Token ID | 262 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
McCartney | +0.167 |
aba | +0.163 |
Reloaded | +0.160 |
igible | +0.159 |
complement | +0.158 |
Neg logit contributions | |
horizont | -0.240 |
MK | -0.212 |
Dash | -0.186 |
HIP | -0.184 |
MK | -0.183 |
Token | ruling |
Token ID | 6922 |
Feature activation | -4.96e-02 |
Pos logit contributions | |
horizont | +0.308 |
MK | +0.273 |
Dash | +0.241 |
HIP | +0.240 |
MK | +0.239 |
Neg logit contributions | |
McCartney | -0.194 |
aba | -0.191 |
Reloaded | -0.188 |
igible | -0.186 |
irs | -0.183 |
Token | � |
Token ID | 12466 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
McCartney | +0.525 |
aba | +0.516 |
Reloaded | +0.509 |
igible | +0.505 |
complement | +0.503 |
Neg logit contributions | |
horizont | -0.454 |
MK | -0.386 |
Dash | -0.324 |
HIP | -0.319 |
MK | -0.318 |
Token | � |
Token ID | 109 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
McCartney | +0.615 |
aba | +0.599 |
Reloaded | +0.584 |
igible | +0.576 |
irs | +0.572 |
Neg logit contributions | |
horizont | -1.313 |
MK | -1.179 |
Dash | -1.056 |
HIP | -1.046 |
MK | -1.043 |
Token | ы |
Token ID | 45035 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
horizont | +0.120 |
MK | +0.100 |
Dash | +0.082 |
HIP | +0.080 |
MK | +0.080 |
Neg logit contributions | |
McCartney | -0.173 |
aba | -0.170 |
Reloaded | -0.168 |
igible | -0.166 |
complement | -0.166 |
Token | л |
Token ID | 30143 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
McCartney | +0.550 |
aba | +0.540 |
Reloaded | +0.532 |
igible | +0.527 |
complement | +0.524 |
Neg logit contributions | |
horizont | -0.578 |
MK | -0.499 |
Dash | -0.427 |
HIP | -0.421 |
MK | -0.420 |
Token | а |
Token ID | 16142 |
Feature activation | +7.55e-01 |
Pos logit contributions | |
McCartney | +0.124 |
aba | +0.122 |
Reloaded | +0.120 |
igible | +0.119 |
complement | +0.119 |
Neg logit contributions | |
horizont | -0.100 |
MK | -0.084 |
Dash | -0.070 |
HIP | -0.069 |
MK | -0.068 |
Token | � |
Token ID | 12466 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
McCartney | +0.510 |
aba | +0.501 |
Reloaded | +0.493 |
igible | +0.488 |
complement | +0.486 |
Neg logit contributions | |
horizont | -0.554 |
MK | -0.480 |
Dash | -0.412 |
HIP | -0.407 |
MK | -0.405 |
Token | � |
Token ID | 123 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
McCartney | +0.605 |
aba | +0.589 |
Reloaded | +0.574 |
igible | +0.566 |
irs | +0.563 |
Neg logit contributions | |
horizont | -1.291 |
MK | -1.159 |
Dash | -1.038 |
HIP | -1.029 |
MK | -1.025 |
Token | е |
Token ID | 16843 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
horizont | +0.112 |
MK | +0.092 |
Dash | +0.074 |
HIP | +0.073 |
MK | +0.073 |
Neg logit contributions | |
McCartney | -0.171 |
aba | -0.168 |
Reloaded | -0.167 |
igible | -0.165 |
complement | -0.165 |
Token | р |
Token ID | 21169 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
McCartney | +0.371 |
aba | +0.365 |
Reloaded | +0.360 |
igible | +0.358 |
complement | +0.356 |
Neg logit contributions | |
horizont | -0.303 |
MK | -0.255 |
Dash | -0.213 |
HIP | -0.209 |
MK | -0.208 |
Token | е |
Token ID | 16843 |
Feature activation | +4.41e-01 |
Pos logit contributions | |
horizont | +0.146 |
MK | +0.124 |
Dash | +0.104 |
HIP | +0.103 |
MK | +0.102 |
Neg logit contributions | |
McCartney | -0.170 |
aba | -0.167 |
Reloaded | -0.165 |
igible | -0.163 |
complement | -0.163 |
Token | д |
Token ID | 43666 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
McCartney | +0.300 |
aba | +0.294 |
Reloaded | +0.290 |
igible | +0.287 |
complement | +0.286 |
Neg logit contributions | |
horizont | -0.322 |
MK | -0.278 |
Dash | -0.239 |
HIP | -0.235 |
MK | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
horizont | +0.233 |
MK | +0.202 |
Dash | +0.174 |
HIP | +0.172 |
MK | +0.171 |
Neg logit contributions | |
McCartney | -0.212 |
aba | -0.208 |
Reloaded | -0.205 |
igible | -0.203 |
complement | -0.201 |
Token | She |
Token ID | 1375 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
horizont | +0.092 |
MK | +0.080 |
Dash | +0.069 |
HIP | +0.068 |
MK | +0.068 |
Neg logit contributions | |
McCartney | -0.084 |
aba | -0.082 |
Reloaded | -0.081 |
igible | -0.080 |
irs | -0.080 |
Token | left |
Token ID | 1364 |
Feature activation | -8.78e-01 |
Pos logit contributions | |
horizont | +0.099 |
MK | +0.086 |
Dash | +0.075 |
HIP | +0.074 |
MK | +0.073 |
Neg logit contributions | |
McCartney | -0.087 |
aba | -0.085 |
Reloaded | -0.085 |
igible | -0.083 |
irs | -0.082 |
Token | with |
Token ID | 351 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
horizont | +0.664 |
MK | +0.578 |
Dash | +0.499 |
HIP | +0.495 |
MK | +0.493 |
Neg logit contributions | |
McCartney | -0.571 |
aba | -0.559 |
Reloaded | -0.555 |
igible | -0.548 |
irs | -0.545 |
Token | Y |
Token ID | 575 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
horizont | +0.277 |
MK | +0.243 |
Dash | +0.211 |
HIP | +0.209 |
MK | +0.208 |
Neg logit contributions | |
McCartney | -0.221 |
aba | -0.216 |
Reloaded | -0.214 |
igible | -0.211 |
irs | -0.210 |
Token | oru |
Token ID | 27786 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
McCartney | +0.154 |
aba | +0.150 |
Reloaded | +0.148 |
igible | +0.146 |
complement | +0.145 |
Neg logit contributions | |
horizont | -0.225 |
MK | -0.199 |
Dash | -0.175 |
HIP | -0.172 |
MK | -0.172 |
Token | ichi |
Token ID | 16590 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
McCartney | +0.808 |
aba | +0.799 |
Reloaded | +0.770 |
igible | +0.770 |
complement | +0.765 |
Neg logit contributions | |
horizont | -1.297 |
MK | -1.146 |
Dash | -1.011 |
HIP | -1.002 |
MK | -0.996 |
Token | - |
Token ID | 12 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
McCartney | +0.110 |
aba | +0.108 |
Reloaded | +0.107 |
igible | +0.106 |
irs | +0.106 |
Neg logit contributions | |
horizont | -0.121 |
MK | -0.105 |
Dash | -0.091 |
HIP | -0.089 |
MK | -0.089 |
Token | san |
Token ID | 12807 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
horizont | +0.048 |
MK | +0.041 |
Dash | +0.035 |
HIP | +0.035 |
MK | +0.035 |
Neg logit contributions | |
McCartney | -0.045 |
aba | -0.044 |
Reloaded | -0.044 |
igible | -0.043 |
complement | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
McCartney | +0.032 |
aba | +0.031 |
Reloaded | +0.031 |
igible | +0.031 |
irs | +0.031 |
Neg logit contributions | |
horizont | -0.034 |
MK | -0.030 |
Dash | -0.026 |
HIP | -0.025 |
MK | -0.025 |
Token | � |
Token ID | 447 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
horizont | +0.181 |
MK | +0.158 |
Dash | +0.136 |
HIP | +0.135 |
MK | +0.134 |
Neg logit contributions | |
McCartney | -0.160 |
aba | -0.157 |
Reloaded | -0.155 |
igible | -0.154 |
complement | -0.153 |
Token | set |
Token ID | 900 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
horizont | +0.117 |
MK | +0.102 |
Dash | +0.088 |
HIP | +0.087 |
MK | +0.086 |
Neg logit contributions | |
McCartney | -0.100 |
aba | -0.098 |
Reloaded | -0.096 |
igible | -0.095 |
complement | -0.095 |
Token | him |
Token ID | 683 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
horizont | +1.019 |
MK | +0.879 |
Dash | +0.752 |
MK | +0.750 |
HIP | +0.749 |
Neg logit contributions | |
McCartney | -0.997 |
aba | -0.986 |
Reloaded | -0.976 |
irs | -0.965 |
igible | -0.964 |
Token | up |
Token ID | 510 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
McCartney | +0.254 |
aba | +0.249 |
Reloaded | +0.245 |
igible | +0.244 |
complement | +0.243 |
Neg logit contributions | |
horizont | -0.233 |
MK | -0.199 |
Dash | -0.168 |
HIP | -0.165 |
MK | -0.164 |
Token | for |
Token ID | 329 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
McCartney | +0.352 |
aba | +0.345 |
Reloaded | +0.341 |
igible | +0.337 |
irs | +0.335 |
Neg logit contributions | |
horizont | -0.398 |
MK | -0.346 |
Dash | -0.298 |
HIP | -0.294 |
MK | -0.294 |
Token | the |
Token ID | 262 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
McCartney | +0.113 |
aba | +0.111 |
Reloaded | +0.109 |
igible | +0.108 |
irs | +0.107 |
Neg logit contributions | |
horizont | -0.168 |
MK | -0.149 |
Dash | -0.131 |
HIP | -0.130 |
MK | -0.129 |
Token | win |
Token ID | 1592 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
horizont | +0.096 |
MK | +0.084 |
Dash | +0.074 |
HIP | +0.073 |
MK | +0.072 |
Neg logit contributions | |
McCartney | -0.073 |
aba | -0.071 |
Reloaded | -0.070 |
igible | -0.069 |
complement | -0.069 |
Token | after |
Token ID | 706 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
McCartney | +1.001 |
aba | +0.983 |
Reloaded | +0.963 |
igible | +0.959 |
complement | +0.955 |
Neg logit contributions | |
horizont | -1.105 |
MK | -0.957 |
Dash | -0.823 |
HIP | -0.812 |
MK | -0.808 |
Token | yesterday |
Token ID | 7415 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
McCartney | +0.223 |
aba | +0.219 |
Reloaded | +0.215 |
igible | +0.213 |
irs | +0.212 |
Neg logit contributions | |
horizont | -0.312 |
MK | -0.275 |
Dash | -0.241 |
HIP | -0.239 |
MK | -0.238 |
Token | � |
Token ID | 447 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
McCartney | +0.186 |
aba | +0.182 |
Reloaded | +0.180 |
igible | +0.178 |
irs | +0.176 |
Neg logit contributions | |
horizont | -0.267 |
MK | -0.236 |
Dash | -0.207 |
MK | -0.205 |
HIP | -0.204 |
Token | � |
Token ID | 247 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
McCartney | +0.181 |
aba | +0.178 |
Reloaded | +0.175 |
igible | +0.173 |
complement | +0.172 |
Neg logit contributions | |
horizont | -0.253 |
MK | -0.224 |
Dash | -0.197 |
HIP | -0.194 |
MK | -0.193 |
Token | s |
Token ID | 82 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
horizont | +0.211 |
MK | +0.179 |
Dash | +0.150 |
HIP | +0.148 |
MK | +0.147 |
Neg logit contributions | |
McCartney | -0.247 |
aba | -0.242 |
Reloaded | -0.240 |
igible | -0.237 |
complement | -0.236 |
Token | AV |
Token ID | 10116 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
horizont | +0.447 |
MK | +0.395 |
Dash | +0.343 |
HIP | +0.338 |
MK | +0.337 |
Neg logit contributions | |
McCartney | -0.361 |
aba | -0.352 |
Reloaded | -0.348 |
igible | -0.345 |
stadt | -0.343 |
Token | ING |
Token ID | 2751 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
horizont | +0.119 |
MK | +0.105 |
Dash | +0.092 |
HIP | +0.091 |
MK | +0.091 |
Neg logit contributions | |
McCartney | -0.078 |
aba | -0.076 |
Reloaded | -0.075 |
igible | -0.074 |
complement | -0.073 |
Token | library |
Token ID | 5888 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
horizont | +0.843 |
MK | +0.732 |
Dash | +0.633 |
HIP | +0.629 |
SOU | +0.627 |
Neg logit contributions | |
McCartney | -0.712 |
aba | -0.701 |
Reloaded | -0.693 |
igible | -0.682 |
stadt | -0.677 |
Token | is |
Token ID | 318 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
horizont | +0.191 |
MK | +0.161 |
Dash | +0.134 |
HIP | +0.132 |
MK | +0.131 |
Neg logit contributions | |
McCartney | -0.238 |
aba | -0.235 |
Reloaded | -0.232 |
igible | -0.230 |
irs | -0.229 |
Token | not |
Token ID | 407 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
horizont | +0.132 |
MK | +0.116 |
Dash | +0.100 |
HIP | +0.099 |
MK | +0.099 |
Neg logit contributions | |
McCartney | -0.108 |
aba | -0.106 |
Reloaded | -0.104 |
igible | -0.103 |
irs | -0.102 |
Token | null |
Token ID | 9242 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
horizont | +0.260 |
MK | +0.230 |
Dash | +0.201 |
HIP | +0.200 |
MK | +0.199 |
Neg logit contributions | |
McCartney | -0.184 |
aba | -0.181 |
Reloaded | -0.177 |
igible | -0.175 |
irs | -0.174 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
McCartney | +0.987 |
aba | +0.973 |
igible | +0.957 |
complement | +0.952 |
irs | +0.948 |
Neg logit contributions | |
horizont | -1.059 |
MK | -0.915 |
Dash | -0.789 |
MK | -0.767 |
HIP | -0.765 |
Token | not |
Token ID | 407 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
McCartney | +0.142 |
aba | +0.139 |
Reloaded | +0.137 |
igible | +0.135 |
irs | +0.134 |
Neg logit contributions | |
horizont | -0.174 |
MK | -0.152 |
Dash | -0.132 |
HIP | -0.131 |
MK | -0.130 |
Token | library |
Token ID | 5888 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
horizont | +0.462 |
MK | +0.408 |
Dash | +0.357 |
HIP | +0.355 |
MK | +0.354 |
Neg logit contributions | |
McCartney | -0.334 |
aba | -0.327 |
Reloaded | -0.321 |
igible | -0.317 |
irs | -0.315 |
Token | contains |
Token ID | 4909 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
horizont | +0.051 |
MK | +0.044 |
Dash | +0.037 |
HIP | +0.037 |
MK | +0.037 |
Neg logit contributions | |
McCartney | -0.049 |
aba | -0.048 |
Reloaded | -0.047 |
igible | -0.047 |
irs | -0.047 |
Token | ' |
Token ID | 705 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
horizont | +0.090 |
MK | +0.079 |
Dash | +0.068 |
HIP | +0.068 |
MK | +0.068 |
Neg logit contributions | |
McCartney | -0.070 |
aba | -0.069 |
Reloaded | -0.067 |
igible | -0.067 |
stadt | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
horizont | +0.271 |
MK | +0.247 |
Dash | +0.208 |
MK | +0.207 |
HIP | +0.206 |
Neg logit contributions | |
McCartney | -0.294 |
Reloaded | -0.288 |
aba | -0.288 |
complement | -0.283 |
irs | -0.283 |
Token | View |
Token ID | 3582 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
horizont | +0.254 |
MK | +0.229 |
Dash | +0.204 |
HIP | +0.203 |
MK | +0.202 |
Neg logit contributions | |
McCartney | -0.133 |
aba | -0.131 |
Reloaded | -0.128 |
igible | -0.126 |
complement | -0.126 |
Token | all |
Token ID | 477 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
McCartney | +1.306 |
aba | +1.283 |
complement | +1.257 |
Reloaded | +1.247 |
igible | +1.242 |
Neg logit contributions | |
horizont | -1.394 |
MK | -1.157 |
Dash | -0.995 |
HIP | -0.964 |
dimension | -0.951 |
Token | New |
Token ID | 968 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
horizont | +0.034 |
MK | +0.030 |
HIP | +0.026 |
Dash | +0.026 |
MK | +0.026 |
Neg logit contributions | |
McCartney | -0.033 |
aba | -0.033 |
Reloaded | -0.032 |
igible | -0.032 |
irs | -0.032 |
Token | York |
Token ID | 1971 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
horizont | +0.541 |
MK | +0.476 |
HIP | +0.410 |
Dash | +0.410 |
MK | +0.407 |
Neg logit contributions | |
McCartney | -0.508 |
aba | -0.495 |
Reloaded | -0.494 |
igible | -0.482 |
complement | -0.482 |
Token | Times |
Token ID | 3782 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
McCartney | +0.187 |
aba | +0.184 |
Reloaded | +0.181 |
igible | +0.180 |
irs | +0.178 |
Neg logit contributions | |
horizont | -0.179 |
MK | -0.156 |
Dash | -0.133 |
HIP | -0.131 |
MK | -0.131 |
Token | newsletters |
Token ID | 16983 |
Feature activation | +7.67e-01 |
Pos logit contributions | |
McCartney | +0.941 |
aba | +0.934 |
complement | +0.914 |
irs | +0.909 |
Reloaded | +0.907 |
Neg logit contributions | |
horizont | -0.957 |
MK | -0.807 |
Dash | -0.689 |
ivably | -0.667 |
dimension | -0.667 |
Token | . |
Token ID | 13 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
McCartney | +0.533 |
aba | +0.522 |
Reloaded | +0.520 |
igible | +0.511 |
irs | +0.510 |
Neg logit contributions | |
horizont | -0.534 |
MK | -0.474 |
Dash | -0.404 |
MK | -0.399 |
HIP | -0.397 |
Token |
|
Token ID | 198 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
horizont | +0.185 |
MK | +0.154 |
Dash | +0.126 |
HIP | +0.124 |
MK | +0.123 |
Neg logit contributions | |
McCartney | -0.256 |
aba | -0.252 |
Reloaded | -0.248 |
igible | -0.247 |
complement | -0.246 |
Token |
|
Token ID | 198 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
McCartney | +0.189 |
aba | +0.186 |
Reloaded | +0.183 |
igible | +0.182 |
complement | +0.181 |
Neg logit contributions | |
horizont | -0.138 |
MK | -0.115 |
Dash | -0.094 |
HIP | -0.092 |
MK | -0.092 |
Token | The |
Token ID | 464 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
McCartney | +0.061 |
aba | +0.059 |
Reloaded | +0.058 |
igible | +0.058 |
complement | +0.058 |
Neg logit contributions | |
horizont | -0.073 |
MK | -0.065 |
Dash | -0.056 |
MK | -0.055 |
HIP | -0.055 |
Token | floor |
Token ID | 4314 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
McCartney | +0.001 |
aba | +0.001 |
Reloaded | +0.001 |
igible | +0.001 |
irs | +0.001 |
Neg logit contributions | |
horizont | -0.001 |
MK | -0.001 |
Dash | -0.001 |
HIP | -0.001 |
MK | -0.001 |
Token | this |
Token ID | 428 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
horizont | +0.143 |
MK | +0.125 |
Dash | +0.108 |
HIP | +0.107 |
MK | +0.107 |
Neg logit contributions | |
McCartney | -0.119 |
aba | -0.117 |
Reloaded | -0.116 |
igible | -0.114 |
irs | -0.114 |
Token | week |
Token ID | 1285 |
Feature activation | -7.28e-02 |
Pos logit contributions | |
McCartney | +0.114 |
aba | +0.112 |
Reloaded | +0.110 |
igible | +0.109 |
irs | +0.108 |
Neg logit contributions | |
horizont | -0.154 |
MK | -0.136 |
Dash | -0.119 |
HIP | -0.117 |
MK | -0.117 |
Token | in |
Token ID | 287 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
horizont | +0.052 |
MK | +0.045 |
Dash | +0.039 |
MK | +0.038 |
HIP | +0.038 |
Neg logit contributions | |
McCartney | -0.050 |
aba | -0.049 |
Reloaded | -0.049 |
igible | -0.048 |
irs | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
horizont | +0.282 |
MK | +0.246 |
Dash | +0.213 |
HIP | +0.213 |
MK | +0.213 |
Neg logit contributions | |
McCartney | -0.228 |
aba | -0.224 |
Reloaded | -0.222 |
irs | -0.218 |
igible | -0.218 |
Token | standard |
Token ID | 3210 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
horizont | +0.102 |
MK | +0.089 |
Dash | +0.078 |
HIP | +0.077 |
MK | +0.077 |
Neg logit contributions | |
McCartney | -0.084 |
aba | -0.083 |
Reloaded | -0.082 |
igible | -0.081 |
irs | -0.081 |
Token | league |
Token ID | 4652 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
horizont | +0.137 |
MK | +0.119 |
HIP | +0.104 |
Dash | +0.103 |
MK | +0.103 |
Neg logit contributions | |
McCartney | -0.119 |
aba | -0.118 |
Reloaded | -0.117 |
irs | -0.115 |
igible | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
McCartney | +0.069 |
aba | +0.067 |
Reloaded | +0.067 |
irs | +0.066 |
igible | +0.066 |
Neg logit contributions | |
horizont | -0.074 |
MK | -0.064 |
Dash | -0.055 |
HIP | -0.054 |
MK | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
horizont | +0.129 |
MK | +0.110 |
Dash | +0.093 |
HIP | +0.091 |
MK | +0.091 |
Neg logit contributions | |
McCartney | -0.144 |
aba | -0.141 |
Reloaded | -0.139 |
igible | -0.138 |
complement | -0.138 |
Token |
|
Token ID | 198 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
McCartney | +0.137 |
aba | +0.135 |
Reloaded | +0.134 |
igible | +0.132 |
complement | +0.132 |
Neg logit contributions | |
horizont | -0.095 |
MK | -0.080 |
Dash | -0.065 |
HIP | -0.064 |
MK | -0.064 |
Token | The |
Token ID | 464 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
horizont | +0.029 |
MK | +0.025 |
Dash | +0.022 |
HIP | +0.022 |
MK | +0.022 |
Neg logit contributions | |
McCartney | -0.024 |
aba | -0.023 |
Reloaded | -0.023 |
igible | -0.023 |
complement | -0.022 |
Token | [ |
Token ID | 685 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
McCartney | +1.030 |
aba | +1.018 |
complement | +0.999 |
Reloaded | +0.993 |
igible | +0.993 |
Neg logit contributions | |
horizont | -1.030 |
MK | -0.886 |
Dash | -0.759 |
HIP | -0.739 |
ivably | -0.736 |
Token | address |
Token ID | 21975 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
McCartney | +0.215 |
aba | +0.212 |
Reloaded | +0.207 |
igible | +0.206 |
complement | +0.205 |
Neg logit contributions | |
horizont | -0.258 |
MK | -0.225 |
Dash | -0.195 |
HIP | -0.192 |
MK | -0.192 |
Token | of |
Token ID | 286 |
Feature activation | -8.44e-02 |
Pos logit contributions | |
McCartney | +0.193 |
aba | +0.190 |
Reloaded | +0.188 |
igible | +0.186 |
irs | +0.186 |
Neg logit contributions | |
horizont | -0.143 |
MK | -0.119 |
Dash | -0.098 |
HIP | -0.096 |
MK | -0.096 |
Token | system |
Token ID | 1080 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
horizont | +0.068 |
MK | +0.059 |
Dash | +0.052 |
HIP | +0.051 |
MK | +0.051 |
Neg logit contributions | |
McCartney | -0.051 |
aba | -0.050 |
Reloaded | -0.050 |
igible | -0.049 |
irs | -0.048 |
Token | ] |
Token ID | 60 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
McCartney | +0.444 |
aba | +0.436 |
Reloaded | +0.430 |
igible | +0.426 |
complement | +0.424 |
Neg logit contributions | |
horizont | -0.458 |
MK | -0.395 |
Dash | -0.338 |
HIP | -0.333 |
MK | -0.331 |
Token | | |
Token ID | 930 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
horizont | +0.280 |
MK | +0.240 |
Dash | +0.203 |
HIP | +0.200 |
MK | +0.200 |
Neg logit contributions | |
McCartney | -0.296 |
aba | -0.291 |
Reloaded | -0.287 |
igible | -0.284 |
irs | -0.283 |
Token | < |
Token ID | 1279 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
McCartney | +0.762 |
aba | +0.751 |
Reloaded | +0.734 |
complement | +0.733 |
igible | +0.731 |
Neg logit contributions | |
horizont | -0.889 |
MK | -0.775 |
Dash | -0.671 |
HIP | -0.659 |
MK | -0.656 |
Token | -- |
Token ID | 438 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
McCartney | +0.297 |
aba | +0.293 |
Reloaded | +0.288 |
igible | +0.286 |
complement | +0.286 |
Neg logit contributions | |
horizont | -0.255 |
MK | -0.216 |
Dash | -0.181 |
HIP | -0.178 |
MK | -0.177 |
Token | Stack |
Token ID | 23881 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
horizont | +0.320 |
MK | +0.280 |
Dash | +0.244 |
HIP | +0.242 |
MK | +0.241 |
Neg logit contributions | |
McCartney | -0.247 |
aba | -0.242 |
Reloaded | -0.238 |
igible | -0.235 |
irs | -0.233 |
Token | pointer |
Token ID | 17562 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
horizont | +0.396 |
MK | +0.342 |
Dash | +0.293 |
HIP | +0.290 |
MK | +0.289 |
Neg logit contributions | |
McCartney | -0.381 |
aba | -0.372 |
Reloaded | -0.368 |
igible | -0.363 |
irs | -0.362 |
Token | + |
Token ID | 1343 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
McCartney | +0.193 |
aba | +0.190 |
Reloaded | +0.187 |
igible | +0.185 |
irs | +0.184 |
Neg logit contributions | |
horizont | -0.218 |
MK | -0.190 |
Dash | -0.163 |
HIP | -0.161 |
MK | -0.161 |
Token | your |
Token ID | 534 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
horizont | +0.678 |
MK | +0.596 |
Dash | +0.518 |
HIP | +0.513 |
MK | +0.513 |
Neg logit contributions | |
McCartney | -0.533 |
aba | -0.521 |
Reloaded | -0.516 |
igible | -0.511 |
irs | -0.506 |
Token | character |
Token ID | 2095 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
McCartney | +0.098 |
aba | +0.096 |
Reloaded | +0.095 |
igible | +0.094 |
irs | +0.094 |
Neg logit contributions | |
horizont | -0.095 |
MK | -0.082 |
Dash | -0.070 |
MK | -0.069 |
HIP | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
horizont | +1.023 |
MK | +0.888 |
Dash | +0.759 |
MK | +0.751 |
HIP | +0.749 |
Neg logit contributions | |
McCartney | -0.964 |
aba | -0.944 |
Reloaded | -0.938 |
igible | -0.927 |
irs | -0.916 |
Token | but |
Token ID | 475 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
McCartney | +0.097 |
aba | +0.095 |
Reloaded | +0.094 |
igible | +0.093 |
irs | +0.092 |
Neg logit contributions | |
horizont | -0.111 |
MK | -0.097 |
Dash | -0.083 |
HIP | -0.082 |
MK | -0.082 |
Token | I |
Token ID | 314 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
McCartney | +0.065 |
aba | +0.064 |
Reloaded | +0.063 |
igible | +0.062 |
irs | +0.062 |
Neg logit contributions | |
horizont | -0.079 |
MK | -0.069 |
Dash | -0.060 |
HIP | -0.059 |
MK | -0.059 |
Token | wish |
Token ID | 4601 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
horizont | +0.114 |
MK | +0.099 |
Dash | +0.086 |
HIP | +0.085 |
MK | +0.085 |
Neg logit contributions | |
McCartney | -0.093 |
aba | -0.091 |
Reloaded | -0.091 |
igible | -0.089 |
irs | -0.089 |
Token | it |
Token ID | 340 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
horizont | +0.087 |
MK | +0.077 |
Dash | +0.066 |
HIP | +0.066 |
MK | +0.065 |
Neg logit contributions | |
McCartney | -0.072 |
aba | -0.070 |
Reloaded | -0.069 |
igible | -0.068 |
irs | -0.068 |
Token | had |
Token ID | 550 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
horizont | +0.079 |
MK | +0.069 |
Dash | +0.060 |
MK | +0.060 |
HIP | +0.060 |
Neg logit contributions | |
McCartney | -0.064 |
aba | -0.063 |
Reloaded | -0.062 |
igible | -0.062 |
irs | -0.061 |
Token | been |
Token ID | 587 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
horizont | +0.145 |
MK | +0.125 |
Dash | +0.109 |
HIP | +0.108 |
MK | +0.108 |
Neg logit contributions | |
McCartney | -0.122 |
aba | -0.120 |
Reloaded | -0.119 |
igible | -0.118 |
irs | -0.117 |
Token | addressed |
Token ID | 9469 |
Feature activation | -6.99e-01 |
Pos logit contributions | |
horizont | +0.243 |
MK | +0.213 |
Dash | +0.186 |
HIP | +0.184 |
MK | +0.184 |
Neg logit contributions | |
McCartney | -0.187 |
aba | -0.184 |
Reloaded | -0.181 |
igible | -0.179 |
irs | -0.178 |
Token | between |
Token ID | 1022 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
horizont | +0.528 |
MK | +0.458 |
Dash | +0.395 |
MK | +0.392 |
HIP | +0.391 |
Neg logit contributions | |
McCartney | -0.457 |
aba | -0.449 |
Reloaded | -0.444 |
igible | -0.440 |
irs | -0.435 |
Token | where |
Token ID | 810 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
McCartney | +0.740 |
aba | +0.724 |
Reloaded | +0.719 |
igible | +0.708 |
irs | +0.705 |
Neg logit contributions | |
horizont | -0.788 |
MK | -0.682 |
Dash | -0.584 |
HIP | -0.578 |
MK | -0.577 |
Token | the |
Token ID | 262 |
Feature activation | -7.89e-02 |
Pos logit contributions | |
horizont | +0.062 |
MK | +0.055 |
Dash | +0.047 |
HIP | +0.047 |
MK | +0.047 |
Neg logit contributions | |
McCartney | -0.048 |
aba | -0.047 |
Reloaded | -0.046 |
igible | -0.046 |
complement | -0.046 |
Token | attack |
Token ID | 1368 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
horizont | +0.059 |
MK | +0.051 |
Dash | +0.044 |
HIP | +0.044 |
MK | +0.044 |
Neg logit contributions | |
McCartney | -0.052 |
aba | -0.051 |
Reloaded | -0.050 |
igible | -0.050 |
complement | -0.049 |
Token | originated |
Token ID | 20973 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
horizont | +0.150 |
MK | +0.131 |
Dash | +0.114 |
HIP | +0.113 |
MK | +0.113 |
Neg logit contributions | |
McCartney | -0.116 |
aba | -0.114 |
Reloaded | -0.112 |
igible | -0.111 |
complement | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -6.02e-02 |
Pos logit contributions | |
McCartney | +0.214 |
aba | +0.210 |
Reloaded | +0.209 |
igible | +0.205 |
irs | +0.205 |
Neg logit contributions | |
horizont | -0.217 |
MK | -0.187 |
Dash | -0.160 |
HIP | -0.157 |
MK | -0.157 |
Token | � |
Token ID | 447 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
horizont | +0.042 |
MK | +0.037 |
Dash | +0.031 |
HIP | +0.031 |
MK | +0.031 |
Neg logit contributions | |
McCartney | -0.042 |
aba | -0.042 |
Reloaded | -0.041 |
igible | -0.041 |
irs | -0.041 |
Token | � |
Token ID | 251 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
McCartney | +0.057 |
aba | +0.056 |
Reloaded | +0.055 |
igible | +0.054 |
complement | +0.054 |
Neg logit contributions | |
horizont | -0.097 |
MK | -0.087 |
Dash | -0.077 |
HIP | -0.076 |
MK | -0.076 |
Token | And |
Token ID | 843 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
McCartney | +0.163 |
aba | +0.160 |
Reloaded | +0.158 |
igible | +0.156 |
complement | +0.156 |
Neg logit contributions | |
horizont | -0.169 |
MK | -0.145 |
Dash | -0.124 |
HIP | -0.123 |
MK | -0.122 |
Token | while |
Token ID | 981 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
horizont | +0.218 |
MK | +0.191 |
Dash | +0.165 |
HIP | +0.164 |
MK | +0.164 |
Neg logit contributions | |
McCartney | -0.175 |
aba | -0.172 |
Reloaded | -0.170 |
igible | -0.168 |
irs | -0.167 |
Token | people |
Token ID | 661 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
horizont | +0.410 |
MK | +0.359 |
Dash | +0.311 |
HIP | +0.311 |
MK | +0.311 |
Neg logit contributions | |
McCartney | -0.318 |
aba | -0.314 |
Reloaded | -0.310 |
igible | -0.306 |
irs | -0.305 |
Token | � |
Token ID | 564 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
horizont | +0.599 |
MK | +0.520 |
Dash | +0.447 |
HIP | +0.443 |
MK | +0.441 |
Neg logit contributions | |
McCartney | -0.537 |
aba | -0.528 |
Reloaded | -0.522 |
igible | -0.515 |
irs | -0.512 |
Token | be |
Token ID | 307 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
McCartney | +0.220 |
aba | +0.216 |
complement | +0.212 |
igible | +0.210 |
Reloaded | +0.210 |
Neg logit contributions | |
horizont | -0.284 |
MK | -0.249 |
Dash | -0.218 |
HIP | -0.214 |
MK | -0.212 |
Token | weird |
Token ID | 7650 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
McCartney | +0.286 |
aba | +0.279 |
complement | +0.274 |
Reloaded | +0.271 |
igible | +0.271 |
Neg logit contributions | |
horizont | -0.339 |
MK | -0.296 |
Dash | -0.258 |
HIP | -0.252 |
ivably | -0.250 |
Token | ," |
Token ID | 553 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
horizont | +0.010 |
MK | +0.009 |
Dash | +0.008 |
HIP | +0.007 |
MK | +0.007 |
Neg logit contributions | |
McCartney | -0.011 |
aba | -0.010 |
Reloaded | -0.010 |
igible | -0.010 |
irs | -0.010 |
Token | said |
Token ID | 531 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
McCartney | +0.567 |
aba | +0.557 |
Reloaded | +0.548 |
igible | +0.540 |
irs | +0.537 |
Neg logit contributions | |
horizont | -0.810 |
MK | -0.715 |
Dash | -0.627 |
HIP | -0.620 |
MK | -0.619 |
Token | L |
Token ID | 406 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
horizont | +0.034 |
MK | +0.030 |
MK | +0.026 |
Dash | +0.026 |
HIP | +0.026 |
Neg logit contributions | |
McCartney | -0.025 |
aba | -0.024 |
Reloaded | -0.024 |
igible | -0.023 |
irs | -0.023 |
Token | add |
Token ID | 2860 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
McCartney | +0.113 |
aba | +0.111 |
Reloaded | +0.110 |
igible | +0.108 |
complement | +0.108 |
Neg logit contributions | |
horizont | -0.125 |
MK | -0.109 |
Dash | -0.093 |
HIP | -0.092 |
MK | -0.092 |
Token | , |
Token ID | 11 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
McCartney | +0.143 |
aba | +0.140 |
Reloaded | +0.140 |
igible | +0.137 |
irs | +0.136 |
Neg logit contributions | |
horizont | -0.149 |
MK | -0.129 |
Dash | -0.110 |
MK | -0.109 |
HIP | -0.109 |
Token | who |
Token ID | 508 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
horizont | +0.188 |
MK | +0.169 |
MK | +0.149 |
Dash | +0.149 |
HIP | +0.149 |
Neg logit contributions | |
McCartney | -0.107 |
aba | -0.105 |
Reloaded | -0.104 |
igible | -0.102 |
irs | -0.101 |
Token | has |
Token ID | 468 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
horizont | +0.086 |
MK | +0.075 |
Dash | +0.066 |
HIP | +0.065 |
MK | +0.065 |
Neg logit contributions | |
McCartney | -0.066 |
aba | -0.065 |
Reloaded | -0.065 |
igible | -0.063 |
irs | -0.063 |
Token | two |
Token ID | 734 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
horizont | +0.387 |
MK | +0.340 |
Dash | +0.297 |
HIP | +0.294 |
MK | +0.294 |
Neg logit contributions | |
McCartney | -0.288 |
aba | -0.284 |
Reloaded | -0.280 |
igible | -0.276 |
irs | -0.274 |
Token | goals |
Token ID | 4661 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
McCartney | +0.102 |
aba | +0.100 |
Reloaded | +0.098 |
igible | +0.097 |
irs | +0.096 |
Neg logit contributions | |
horizont | -0.143 |
MK | -0.126 |
Dash | -0.110 |
HIP | -0.109 |
MK | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
horizont | +0.036 |
MK | +0.031 |
Dash | +0.027 |
HIP | +0.027 |
MK | +0.027 |
Neg logit contributions | |
McCartney | -0.028 |
aba | -0.028 |
Reloaded | -0.027 |
igible | -0.027 |
complement | -0.027 |
Token | day |
Token ID | 1110 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
McCartney | +0.124 |
aba | +0.121 |
Reloaded | +0.119 |
igible | +0.118 |
complement | +0.118 |
Neg logit contributions | |
horizont | -0.131 |
MK | -0.114 |
Dash | -0.097 |
HIP | -0.095 |
MK | -0.095 |
Token | after |
Token ID | 706 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
horizont | +0.201 |
MK | +0.175 |
Dash | +0.151 |
HIP | +0.149 |
MK | +0.148 |
Neg logit contributions | |
McCartney | -0.180 |
aba | -0.177 |
Reloaded | -0.175 |
igible | -0.173 |
irs | -0.172 |
Token | G |
Token ID | 402 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
McCartney | +0.236 |
aba | +0.232 |
Reloaded | +0.228 |
igible | +0.226 |
irs | +0.225 |
Neg logit contributions | |
horizont | -0.296 |
MK | -0.259 |
Dash | -0.225 |
HIP | -0.222 |
MK | -0.222 |
Token | ant |
Token ID | 415 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
McCartney | +0.015 |
aba | +0.014 |
Reloaded | +0.014 |
igible | +0.014 |
complement | +0.014 |
Neg logit contributions | |
horizont | -0.020 |
MK | -0.017 |
Dash | -0.015 |
MK | -0.015 |
HIP | -0.015 |
Token | Heart |
Token ID | 8894 |
Feature activation | -7.14e-01 |
Pos logit contributions | |
horizont | +0.546 |
MK | +0.474 |
Dash | +0.408 |
HIP | +0.403 |
MK | +0.401 |
Neg logit contributions | |
McCartney | -0.485 |
aba | -0.477 |
Reloaded | -0.469 |
igible | -0.464 |
irs | -0.462 |
Token | . |
Token ID | 13 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
horizont | +0.555 |
MK | +0.486 |
Dash | +0.426 |
HIP | +0.420 |
MK | +0.418 |
Neg logit contributions | |
McCartney | -0.444 |
aba | -0.436 |
Reloaded | -0.433 |
igible | -0.428 |
complement | -0.425 |
Token | He |
Token ID | 679 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
McCartney | +0.095 |
aba | +0.093 |
Reloaded | +0.091 |
igible | +0.091 |
complement | +0.090 |
Neg logit contributions | |
horizont | -0.110 |
MK | -0.095 |
Dash | -0.082 |
HIP | -0.081 |
MK | -0.081 |
Token | was |
Token ID | 373 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
horizont | +0.270 |
MK | +0.237 |
Dash | +0.206 |
HIP | +0.204 |
MK | +0.203 |
Neg logit contributions | |
McCartney | -0.218 |
aba | -0.214 |
Reloaded | -0.212 |
igible | -0.209 |
irs | -0.207 |
Token | visiting |
Token ID | 10013 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
McCartney | +0.056 |
aba | +0.055 |
Reloaded | +0.054 |
igible | +0.054 |
complement | +0.053 |
Neg logit contributions | |
horizont | -0.077 |
MK | -0.068 |
Dash | -0.059 |
HIP | -0.059 |
MK | -0.058 |
Token | from |
Token ID | 422 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
McCartney | +0.375 |
aba | +0.367 |
Reloaded | +0.360 |
igible | +0.356 |
complement | +0.355 |
Neg logit contributions | |
horizont | -0.499 |
MK | -0.438 |
Dash | -0.383 |
HIP | -0.377 |
MK | -0.375 |
Token | a |
Token ID | 257 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
McCartney | +0.027 |
aba | +0.026 |
Reloaded | +0.026 |
igible | +0.026 |
irs | +0.026 |
Neg logit contributions | |
horizont | -0.036 |
MK | -0.032 |
Dash | -0.028 |
HIP | -0.028 |
MK | -0.028 |
Token | Pokémon |
Token ID | 9887 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
McCartney | +0.091 |
aba | +0.089 |
Reloaded | +0.087 |
complement | +0.086 |
igible | +0.086 |
Neg logit contributions | |
horizont | -0.118 |
MK | -0.103 |
Dash | -0.090 |
HIP | -0.089 |
MK | -0.088 |
Token | ... |
Token ID | 986 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
horizont | +1.223 |
MK | +1.071 |
Dash | +0.930 |
HIP | +0.919 |
MK | +0.918 |
Neg logit contributions | |
McCartney | -0.974 |
aba | -0.955 |
Reloaded | -0.941 |
igible | -0.930 |
irs | -0.925 |
Token | Wait |
Token ID | 16314 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
horizont | +0.354 |
MK | +0.301 |
Dash | +0.251 |
HIP | +0.247 |
MK | +0.246 |
Neg logit contributions | |
McCartney | -0.431 |
aba | -0.424 |
Reloaded | -0.419 |
igible | -0.415 |
irs | -0.414 |
Token | ... |
Token ID | 986 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
horizont | +0.037 |
MK | +0.032 |
Dash | +0.027 |
HIP | +0.027 |
MK | +0.027 |
Neg logit contributions | |
McCartney | -0.034 |
aba | -0.034 |
Reloaded | -0.033 |
igible | -0.033 |
complement | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
horizont | +0.469 |
MK | +0.410 |
Dash | +0.354 |
HIP | +0.350 |
MK | +0.349 |
Neg logit contributions | |
McCartney | -0.400 |
aba | -0.393 |
Reloaded | -0.387 |
igible | -0.383 |
complement | -0.381 |
Token |
|
Token ID | 198 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
horizont | +0.252 |
MK | +0.214 |
Dash | +0.174 |
HIP | +0.170 |
MK | +0.170 |
Neg logit contributions | |
McCartney | -0.374 |
aba | -0.368 |
Reloaded | -0.365 |
igible | -0.362 |
complement | -0.361 |
Token | Did |
Token ID | 11633 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
horizont | +0.314 |
MK | +0.277 |
Dash | +0.242 |
HIP | +0.236 |
MK | +0.235 |
Neg logit contributions | |
McCartney | -0.255 |
aba | -0.248 |
Reloaded | -0.244 |
igible | -0.243 |
complement | -0.243 |
Token | that |
Token ID | 326 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
horizont | +0.353 |
MK | +0.311 |
Dash | +0.272 |
HIP | +0.269 |
MK | +0.269 |
Neg logit contributions | |
McCartney | -0.254 |
aba | -0.249 |
Reloaded | -0.244 |
igible | -0.242 |
complement | -0.240 |
Token | Pokémon |
Token ID | 9887 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
horizont | +0.089 |
MK | +0.078 |
Dash | +0.068 |
HIP | +0.067 |
MK | +0.066 |
Neg logit contributions | |
McCartney | -0.074 |
aba | -0.072 |
Reloaded | -0.070 |
complement | -0.070 |
igible | -0.070 |
Token | just |
Token ID | 655 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
horizont | +1.273 |
MK | +1.106 |
Dash | +0.957 |
HIP | +0.945 |
MK | +0.943 |
Neg logit contributions | |
McCartney | -1.104 |
aba | -1.086 |
Reloaded | -1.073 |
igible | -1.060 |
irs | -1.053 |
Token | dialog |
Token ID | 17310 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
McCartney | +0.150 |
aba | +0.146 |
Reloaded | +0.143 |
igible | +0.142 |
complement | +0.141 |
Neg logit contributions | |
horizont | -0.235 |
MK | -0.209 |
Dash | -0.184 |
HIP | -0.182 |
MK | -0.181 |
Token | or |
Token ID | 393 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
horizont | +0.092 |
MK | +0.080 |
Dash | +0.068 |
HIP | +0.067 |
MK | +0.067 |
Neg logit contributions | |
McCartney | -0.090 |
aba | -0.089 |
Reloaded | -0.088 |
igible | -0.087 |
complement | -0.086 |
Token | on |
Token ID | 319 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
McCartney | +0.013 |
aba | +0.013 |
Reloaded | +0.012 |
complement | +0.012 |
igible | +0.012 |
Neg logit contributions | |
horizont | -0.015 |
MK | -0.013 |
Dash | -0.012 |
HIP | -0.011 |
MK | -0.011 |
Token | om |
Token ID | 296 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
McCartney | +0.132 |
aba | +0.130 |
Reloaded | +0.127 |
igible | +0.126 |
irs | +0.126 |
Neg logit contributions | |
horizont | -0.138 |
MK | -0.119 |
Dash | -0.102 |
HIP | -0.101 |
MK | -0.100 |
Token | at |
Token ID | 265 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
McCartney | +0.236 |
aba | +0.232 |
igible | +0.224 |
irs | +0.223 |
Reloaded | +0.223 |
Neg logit contributions | |
horizont | -0.416 |
MK | -0.369 |
Dash | -0.328 |
HIP | -0.326 |
MK | -0.323 |
Token | op |
Token ID | 404 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
McCartney | +0.404 |
aba | +0.400 |
igible | +0.390 |
Reloaded | +0.389 |
complement | +0.388 |
Neg logit contributions | |
horizont | -0.370 |
MK | -0.315 |
Dash | -0.266 |
HIP | -0.261 |
MK | -0.259 |
Token | o |
Token ID | 78 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
horizont | +0.460 |
MK | +0.392 |
Dash | +0.325 |
MK | +0.321 |
HIP | +0.320 |
Neg logit contributions | |
McCartney | -0.556 |
aba | -0.544 |
Reloaded | -0.542 |
igible | -0.532 |
irs | -0.531 |
Token | etic |
Token ID | 5139 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
McCartney | +0.298 |
aba | +0.290 |
Reloaded | +0.281 |
igible | +0.279 |
irs | +0.277 |
Neg logit contributions | |
horizont | -0.649 |
MK | -0.582 |
Dash | -0.523 |
HIP | -0.519 |
MK | -0.516 |
Token | words |
Token ID | 2456 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
McCartney | +0.210 |
aba | +0.206 |
Reloaded | +0.202 |
igible | +0.200 |
complement | +0.199 |
Neg logit contributions | |
horizont | -0.272 |
MK | -0.239 |
Dash | -0.208 |
HIP | -0.206 |
MK | -0.204 |
Token | . |
Token ID | 13 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
McCartney | +0.029 |
aba | +0.029 |
Reloaded | +0.029 |
igible | +0.028 |
irs | +0.028 |
Neg logit contributions | |
horizont | -0.030 |
MK | -0.026 |
Dash | -0.022 |
HIP | -0.022 |
MK | -0.022 |
Token | People |
Token ID | 4380 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
McCartney | +0.420 |
aba | +0.411 |
Reloaded | +0.403 |
igible | +0.398 |
complement | +0.396 |
Neg logit contributions | |
horizont | -0.464 |
MK | -0.400 |
Dash | -0.343 |
HIP | -0.337 |
ivably | -0.337 |
Token | to |
Token ID | 284 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
McCartney | +0.176 |
aba | +0.172 |
Reloaded | +0.171 |
igible | +0.168 |
irs | +0.167 |
Neg logit contributions | |
horizont | -0.211 |
MK | -0.184 |
Dash | -0.160 |
HIP | -0.158 |
MK | -0.157 |
Token | realise |
Token ID | 18996 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
McCartney | +0.007 |
aba | +0.007 |
Reloaded | +0.007 |
igible | +0.007 |
irs | +0.007 |
Neg logit contributions | |
horizont | -0.008 |
MK | -0.007 |
Dash | -0.006 |
HIP | -0.006 |
MK | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
McCartney | +0.051 |
aba | +0.050 |
Reloaded | +0.049 |
igible | +0.048 |
complement | +0.048 |
Neg logit contributions | |
horizont | -0.072 |
MK | -0.063 |
Dash | -0.055 |
HIP | -0.055 |
MK | -0.054 |
Token | a |
Token ID | 257 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
McCartney | +0.190 |
aba | +0.186 |
Reloaded | +0.183 |
igible | +0.181 |
complement | +0.180 |
Neg logit contributions | |
horizont | -0.228 |
MK | -0.199 |
Dash | -0.173 |
HIP | -0.170 |
MK | -0.170 |
Token | life |
Token ID | 1204 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
McCartney | +0.125 |
aba | +0.122 |
Reloaded | +0.120 |
igible | +0.119 |
irs | +0.119 |
Neg logit contributions | |
horizont | -0.151 |
MK | -0.132 |
Dash | -0.115 |
HIP | -0.113 |
MK | -0.113 |
Token | time |
Token ID | 640 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
horizont | +0.369 |
MK | +0.319 |
Dash | +0.273 |
HIP | +0.270 |
MK | +0.269 |
Neg logit contributions | |
McCartney | -0.346 |
aba | -0.340 |
Reloaded | -0.336 |
igible | -0.332 |
irs | -0.331 |
Token | of |
Token ID | 286 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
horizont | +0.314 |
MK | +0.273 |
Dash | +0.236 |
HIP | +0.233 |
MK | +0.232 |
Neg logit contributions | |
McCartney | -0.275 |
aba | -0.269 |
Reloaded | -0.265 |
igible | -0.263 |
complement | -0.262 |
Token | learning |
Token ID | 4673 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
McCartney | +0.026 |
aba | +0.026 |
Reloaded | +0.025 |
igible | +0.025 |
irs | +0.025 |
Neg logit contributions | |
horizont | -0.034 |
MK | -0.030 |
Dash | -0.026 |
HIP | -0.026 |
MK | -0.026 |
Token | will |
Token ID | 481 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
McCartney | +0.143 |
aba | +0.141 |
Reloaded | +0.139 |
igible | +0.137 |
complement | +0.136 |
Neg logit contributions | |
horizont | -0.169 |
MK | -0.148 |
Dash | -0.128 |
HIP | -0.126 |
MK | -0.125 |
Token | not |
Token ID | 407 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
horizont | +0.267 |
MK | +0.234 |
Dash | +0.204 |
HIP | +0.201 |
MK | +0.200 |
Neg logit contributions | |
McCartney | -0.211 |
aba | -0.206 |
Reloaded | -0.203 |
igible | -0.201 |
complement | -0.200 |
Token | even |
Token ID | 772 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
horizont | +0.129 |
MK | +0.113 |
Dash | +0.098 |
HIP | +0.097 |
MK | +0.096 |
Neg logit contributions | |
McCartney | -0.103 |
aba | -0.101 |
Reloaded | -0.099 |
complement | -0.098 |
igible | -0.098 |
Token | I |
Token ID | 314 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
horizont | +0.117 |
MK | +0.101 |
Dash | +0.087 |
HIP | +0.086 |
MK | +0.085 |
Neg logit contributions | |
McCartney | -0.113 |
aba | -0.111 |
Reloaded | -0.109 |
igible | -0.108 |
irs | -0.108 |
Token | needed |
Token ID | 2622 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
horizont | +0.052 |
MK | +0.046 |
Dash | +0.040 |
HIP | +0.040 |
MK | +0.040 |
Neg logit contributions | |
McCartney | -0.036 |
aba | -0.036 |
Reloaded | -0.035 |
igible | -0.035 |
irs | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
McCartney | +0.158 |
aba | +0.155 |
Reloaded | +0.153 |
igible | +0.151 |
irs | +0.150 |
Neg logit contributions | |
horizont | -0.179 |
MK | -0.156 |
Dash | -0.135 |
HIP | -0.133 |
MK | -0.133 |
Token | ask |
Token ID | 1265 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
horizont | +0.216 |
MK | +0.187 |
Dash | +0.160 |
HIP | +0.157 |
MK | +0.156 |
Neg logit contributions | |
McCartney | -0.208 |
aba | -0.205 |
Reloaded | -0.201 |
igible | -0.200 |
complement | -0.199 |
Token | a |
Token ID | 257 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
horizont | +0.178 |
MK | +0.156 |
Dash | +0.134 |
MK | +0.134 |
HIP | +0.133 |
Neg logit contributions | |
McCartney | -0.153 |
aba | -0.151 |
Reloaded | -0.149 |
igible | -0.148 |
irs | -0.147 |
Token | dear |
Token ID | 13674 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
horizont | +0.236 |
MK | +0.210 |
Dash | +0.182 |
MK | +0.182 |
HIP | +0.182 |
Neg logit contributions | |
McCartney | -0.170 |
aba | -0.167 |
Reloaded | -0.164 |
igible | -0.163 |
irs | -0.162 |
Token | friend |
Token ID | 1545 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
horizont | +0.362 |
MK | +0.317 |
MK | +0.272 |
Dash | +0.271 |
HIP | +0.270 |
Neg logit contributions | |
McCartney | -0.317 |
aba | -0.313 |
Reloaded | -0.310 |
igible | -0.305 |
irs | -0.305 |
Token | for |
Token ID | 329 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
horizont | +0.568 |
MK | +0.497 |
Dash | +0.426 |
MK | +0.424 |
HIP | +0.423 |
Neg logit contributions | |
McCartney | -0.509 |
aba | -0.501 |
Reloaded | -0.498 |
igible | -0.491 |
irs | -0.489 |
Token | a |
Token ID | 257 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
horizont | +0.236 |
MK | +0.208 |
Dash | +0.183 |
MK | +0.182 |
HIP | +0.182 |
Neg logit contributions | |
McCartney | -0.160 |
aba | -0.157 |
Reloaded | -0.155 |
igible | -0.153 |
irs | -0.152 |
Token | blessing |
Token ID | 20027 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
horizont | +0.151 |
MK | +0.133 |
Dash | +0.115 |
HIP | +0.115 |
MK | +0.115 |
Neg logit contributions | |
McCartney | -0.115 |
aba | -0.113 |
Reloaded | -0.111 |
igible | -0.110 |
irs | -0.109 |
Token | for |
Token ID | 329 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
McCartney | +0.132 |
aba | +0.130 |
Reloaded | +0.128 |
igible | +0.127 |
irs | +0.127 |
Neg logit contributions | |
horizont | -0.141 |
MK | -0.122 |
Dash | -0.104 |
MK | -0.103 |
HIP | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | -5.75e-03 |
Pos logit contributions | |
horizont | +0.697 |
MK | +0.605 |
Dash | +0.520 |
MK | +0.515 |
HIP | +0.512 |
Neg logit contributions | |
McCartney | -0.625 |
aba | -0.613 |
Reloaded | -0.606 |
igible | -0.599 |
irs | -0.596 |
Token | Main |
Token ID | 8774 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
horizont | +0.005 |
MK | +0.004 |
Dash | +0.003 |
HIP | +0.003 |
MK | +0.003 |
Neg logit contributions | |
McCartney | -0.004 |
aba | -0.004 |
Reloaded | -0.003 |
igible | -0.003 |
complement | -0.003 |
Token | stream |
Token ID | 5532 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
McCartney | +1.014 |
aba | +0.999 |
Reloaded | +0.979 |
igible | +0.975 |
stadt | +0.974 |
Neg logit contributions | |
horizont | -0.899 |
MK | -0.767 |
Dash | -0.639 |
HIP | -0.638 |
ivably | -0.631 |
Token | media |
Token ID | 2056 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
horizont | +0.103 |
MK | +0.091 |
Dash | +0.080 |
HIP | +0.079 |
MK | +0.079 |
Neg logit contributions | |
McCartney | -0.073 |
aba | -0.071 |
Reloaded | -0.070 |
igible | -0.069 |
irs | -0.069 |
Token | was |
Token ID | 373 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
horizont | +0.418 |
MK | +0.366 |
Dash | +0.317 |
HIP | +0.314 |
MK | +0.313 |
Neg logit contributions | |
McCartney | -0.333 |
aba | -0.327 |
Reloaded | -0.322 |
igible | -0.319 |
irs | -0.316 |
Token | dup |
Token ID | 32597 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
horizont | +0.017 |
MK | +0.015 |
Dash | +0.013 |
HIP | +0.013 |
MK | +0.013 |
Neg logit contributions | |
McCartney | -0.013 |
aba | -0.012 |
Reloaded | -0.012 |
igible | -0.012 |
complement | -0.012 |
Token | ed |
Token ID | 276 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
horizont | +1.581 |
MK | +1.399 |
MK | +1.202 |
Dash | +1.200 |
ivably | +1.191 |
Neg logit contributions | |
McCartney | -1.359 |
Reloaded | -1.327 |
aba | -1.311 |
complement | -1.279 |
igible | -1.277 |
Token | into |
Token ID | 656 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
horizont | +0.164 |
MK | +0.139 |
Dash | +0.116 |
HIP | +0.115 |
MK | +0.114 |
Neg logit contributions | |
McCartney | -0.192 |
aba | -0.189 |
Reloaded | -0.186 |
igible | -0.185 |
irs | -0.184 |
Token | fueling |
Token ID | 43881 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
horizont | +0.180 |
MK | +0.161 |
Dash | +0.143 |
HIP | +0.142 |
MK | +0.141 |
Neg logit contributions | |
McCartney | -0.097 |
aba | -0.095 |
Reloaded | -0.093 |
igible | -0.092 |
irs | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
horizont | +0.042 |
MK | +0.038 |
Dash | +0.033 |
HIP | +0.033 |
MK | +0.033 |
Neg logit contributions | |
McCartney | -0.026 |
aba | -0.025 |
Reloaded | -0.025 |
igible | -0.025 |
irs | -0.025 |
Token | hate |
Token ID | 5465 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
horizont | +0.164 |
MK | +0.143 |
Dash | +0.123 |
HIP | +0.122 |
MK | +0.121 |
Neg logit contributions | |
McCartney | -0.141 |
aba | -0.138 |
Reloaded | -0.136 |
igible | -0.134 |
complement | -0.134 |
Token | n |
Token ID | 77 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
McCartney | +0.032 |
aba | +0.031 |
Reloaded | +0.031 |
igible | +0.031 |
complement | +0.031 |
Neg logit contributions | |
horizont | -0.036 |
MK | -0.032 |
Dash | -0.027 |
HIP | -0.027 |
MK | -0.027 |
Token | owski |
Token ID | 12079 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
horizont | +0.309 |
MK | +0.277 |
Dash | +0.247 |
HIP | +0.245 |
MK | +0.244 |
Neg logit contributions | |
McCartney | -0.154 |
aba | -0.150 |
Reloaded | -0.147 |
igible | -0.145 |
complement | -0.144 |
Token | , |
Token ID | 11 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
horizont | +0.173 |
MK | +0.150 |
Dash | +0.128 |
HIP | +0.126 |
MK | +0.126 |
Neg logit contributions | |
McCartney | -0.169 |
aba | -0.166 |
Reloaded | -0.164 |
igible | -0.163 |
complement | -0.162 |
Token | expert |
Token ID | 5887 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
horizont | +0.051 |
MK | +0.045 |
Dash | +0.039 |
HIP | +0.039 |
MK | +0.038 |
Neg logit contributions | |
McCartney | -0.037 |
aba | -0.036 |
Reloaded | -0.036 |
igible | -0.035 |
irs | -0.035 |
Token | senior |
Token ID | 4664 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
horizont | +0.281 |
MK | +0.238 |
Dash | +0.200 |
HIP | +0.196 |
MK | +0.195 |
Neg logit contributions | |
McCartney | -0.332 |
aba | -0.327 |
Reloaded | -0.323 |
igible | -0.320 |
irs | -0.318 |
Token | meteor |
Token ID | 19999 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
horizont | +0.411 |
MK | +0.357 |
Dash | +0.307 |
HIP | +0.304 |
MK | +0.302 |
Neg logit contributions | |
McCartney | -0.369 |
aba | -0.362 |
Reloaded | -0.357 |
igible | -0.353 |
irs | -0.352 |
Token | ologist |
Token ID | 7451 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
horizont | +2.287 |
MK | +2.072 |
Dash | +1.876 |
MK | +1.860 |
HIP | +1.853 |
Neg logit contributions | |
McCartney | -0.829 |
Reloaded | -0.795 |
aba | -0.783 |
igible | -0.752 |
irs | -0.746 |
Token | for |
Token ID | 329 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
McCartney | +0.016 |
aba | +0.016 |
Reloaded | +0.016 |
igible | +0.016 |
irs | +0.016 |
Neg logit contributions | |
horizont | -0.019 |
MK | -0.017 |
Dash | -0.014 |
HIP | -0.014 |
MK | -0.014 |
Token | Acc |
Token ID | 6366 |
Feature activation | +8.58e-01 |
Pos logit contributions | |
horizont | +0.403 |
MK | +0.355 |
Dash | +0.311 |
HIP | +0.308 |
MK | +0.306 |
Neg logit contributions | |
McCartney | -0.289 |
aba | -0.283 |
Reloaded | -0.278 |
igible | -0.275 |
irs | -0.273 |
Token | u |
Token ID | 84 |
Feature activation | -7.87e-02 |
Pos logit contributions | |
McCartney | +0.610 |
aba | +0.609 |
igible | +0.588 |
irs | +0.586 |
herty | +0.585 |
Neg logit contributions | |
horizont | -0.600 |
MK | -0.512 |
Dash | -0.434 |
HIP | -0.430 |
MK | -0.427 |
Token | Weather |
Token ID | 41865 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
horizont | +0.052 |
MK | +0.044 |
Dash | +0.037 |
HIP | +0.037 |
MK | +0.037 |
Neg logit contributions | |
McCartney | -0.059 |
aba | -0.058 |
Reloaded | -0.057 |
igible | -0.056 |
irs | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
horizont | +0.123 |
MK | +0.109 |
Dash | +0.095 |
MK | +0.094 |
HIP | +0.094 |
Neg logit contributions | |
McCartney | -0.092 |
aba | -0.091 |
Reloaded | -0.090 |
igible | -0.089 |
irs | -0.089 |
Token | soundtrack |
Token ID | 21751 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
horizont | +0.137 |
MK | +0.120 |
Dash | +0.105 |
HIP | +0.104 |
MK | +0.104 |
Neg logit contributions | |
McCartney | -0.103 |
aba | -0.101 |
Reloaded | -0.100 |
igible | -0.099 |
complement | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
horizont | +0.102 |
MK | +0.089 |
Dash | +0.076 |
MK | +0.076 |
HIP | +0.075 |
Neg logit contributions | |
McCartney | -0.091 |
aba | -0.089 |
Reloaded | -0.088 |
igible | -0.088 |
irs | -0.087 |
Token | Maybe |
Token ID | 6674 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
horizont | +0.316 |
MK | +0.277 |
Dash | +0.238 |
HIP | +0.235 |
MK | +0.235 |
Neg logit contributions | |
McCartney | -0.286 |
aba | -0.282 |
Reloaded | -0.278 |
igible | -0.276 |
complement | -0.274 |
Token | my |
Token ID | 616 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
McCartney | +0.532 |
aba | +0.520 |
Reloaded | +0.510 |
complement | +0.505 |
igible | +0.505 |
Neg logit contributions | |
horizont | -0.648 |
MK | -0.566 |
Dash | -0.492 |
HIP | -0.483 |
MK | -0.480 |
Token | character |
Token ID | 2095 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
horizont | +0.417 |
MK | +0.365 |
Dash | +0.317 |
HIP | +0.314 |
MK | +0.313 |
Neg logit contributions | |
McCartney | -0.329 |
aba | -0.323 |
Reloaded | -0.318 |
igible | -0.315 |
complement | -0.312 |
Token | suffered |
Token ID | 6989 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
horizont | +1.708 |
MK | +1.497 |
Dash | +1.304 |
MK | +1.304 |
HIP | +1.301 |
Neg logit contributions | |
McCartney | -1.342 |
aba | -1.339 |
Reloaded | -1.327 |
igible | -1.309 |
irs | -1.292 |
Token | a |
Token ID | 257 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
horizont | +0.281 |
MK | +0.244 |
Dash | +0.209 |
MK | +0.207 |
HIP | +0.207 |
Neg logit contributions | |
McCartney | -0.254 |
aba | -0.250 |
Reloaded | -0.247 |
igible | -0.244 |
irs | -0.243 |
Token | little |
Token ID | 1310 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
horizont | +0.174 |
MK | +0.154 |
Dash | +0.136 |
HIP | +0.134 |
MK | +0.134 |
Neg logit contributions | |
McCartney | -0.111 |
aba | -0.108 |
Reloaded | -0.106 |
igible | -0.105 |
complement | -0.104 |
Token | bit |
Token ID | 1643 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
horizont | +0.348 |
MK | +0.305 |
Dash | +0.267 |
HIP | +0.264 |
MK | +0.264 |
Neg logit contributions | |
McCartney | -0.258 |
aba | -0.253 |
Reloaded | -0.249 |
igible | -0.246 |
irs | -0.245 |
Token | as |
Token ID | 355 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
horizont | +0.457 |
MK | +0.394 |
Dash | +0.337 |
MK | +0.334 |
HIP | +0.333 |
Neg logit contributions | |
McCartney | -0.432 |
aba | -0.426 |
Reloaded | -0.421 |
igible | -0.417 |
irs | -0.414 |
Token | "] |
Token ID | 8973 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
horizont | +0.407 |
MK | +0.336 |
Dash | +0.274 |
HIP | +0.269 |
MK | +0.267 |
Neg logit contributions | |
McCartney | -0.564 |
aba | -0.556 |
Reloaded | -0.548 |
igible | -0.543 |
complement | -0.542 |
Token | [ |
Token ID | 685 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
horizont | +1.408 |
MK | +1.308 |
MK | +1.090 |
SOU | +1.077 |
HIP | +1.067 |
Neg logit contributions | |
igible | -1.954 |
McCartney | -1.952 |
Reloaded | -1.944 |
aba | -1.932 |
keyes | -1.930 |
Token | Round |
Token ID | 22685 |
Feature activation | -9.69e-01 |
Pos logit contributions | |
McCartney | +0.513 |
aba | +0.507 |
Reloaded | +0.498 |
igible | +0.496 |
irs | +0.494 |
Neg logit contributions | |
horizont | -0.316 |
MK | -0.254 |
Dash | -0.202 |
HIP | -0.198 |
MK | -0.196 |
Token | " |
Token ID | 366 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
horizont | +0.864 |
MK | +0.770 |
Dash | +0.683 |
HIP | +0.675 |
MK | +0.673 |
Neg logit contributions | |
McCartney | -0.501 |
aba | -0.489 |
Reloaded | -0.479 |
igible | -0.473 |
complement | -0.470 |
Token | 11 |
Token ID | 1157 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
horizont | +0.333 |
MK | +0.293 |
Dash | +0.252 |
MK | +0.248 |
HIP | +0.248 |
Neg logit contributions | |
McCartney | -0.301 |
aba | -0.294 |
Reloaded | -0.289 |
igible | -0.287 |
complement | -0.286 |
Token | "] |
Token ID | 8973 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
horizont | +0.237 |
MK | +0.202 |
Dash | +0.170 |
HIP | +0.168 |
MK | +0.167 |
Neg logit contributions | |
McCartney | -0.264 |
aba | -0.260 |
Reloaded | -0.256 |
igible | -0.254 |
complement | -0.253 |
Token | [ |
Token ID | 685 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
horizont | +1.536 |
MK | +1.427 |
MK | +1.197 |
Dash | +1.170 |
HIP | +1.160 |
Neg logit contributions | |
McCartney | -2.150 |
Reloaded | -2.148 |
igible | -2.127 |
aba | -2.116 |
irs | -2.091 |
Token | White |
Token ID | 12256 |
Feature activation | -6.78e-02 |
Pos logit contributions | |
McCartney | +0.367 |
aba | +0.361 |
Reloaded | +0.355 |
igible | +0.352 |
complement | +0.350 |
Neg logit contributions | |
horizont | -0.394 |
MK | -0.341 |
Dash | -0.293 |
HIP | -0.289 |
MK | -0.288 |
Token | " |
Token ID | 366 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
horizont | +0.061 |
MK | +0.054 |
Dash | +0.048 |
HIP | +0.048 |
MK | +0.047 |
Neg logit contributions | |
McCartney | -0.035 |
aba | -0.034 |
Reloaded | -0.033 |
igible | -0.033 |
complement | -0.033 |
Token | Ab |
Token ID | 4826 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
horizont | +0.390 |
MK | +0.345 |
Dash | +0.299 |
MK | +0.292 |
HIP | +0.292 |
Neg logit contributions | |
McCartney | -0.324 |
aba | -0.317 |
Reloaded | -0.311 |
igible | -0.310 |
complement | -0.309 |
Token | as |
Token ID | 292 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
horizont | +0.800 |
MK | +0.692 |
Dash | +0.592 |
HIP | +0.582 |
MK | +0.581 |
Neg logit contributions | |
McCartney | -0.769 |
aba | -0.754 |
Reloaded | -0.745 |
igible | -0.738 |
complement | -0.734 |
Token | the |
Token ID | 262 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
McCartney | +0.135 |
aba | +0.133 |
Reloaded | +0.131 |
igible | +0.129 |
irs | +0.128 |
Neg logit contributions | |
horizont | -0.185 |
MK | -0.162 |
Dash | -0.142 |
HIP | -0.141 |
MK | -0.141 |
Token | so |
Token ID | 523 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
horizont | +0.053 |
MK | +0.047 |
Dash | +0.041 |
HIP | +0.041 |
MK | +0.040 |
Neg logit contributions | |
McCartney | -0.039 |
aba | -0.038 |
Reloaded | -0.038 |
igible | -0.037 |
complement | -0.037 |
Token | - |
Token ID | 12 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
McCartney | +0.284 |
aba | +0.277 |
Reloaded | +0.276 |
complement | +0.271 |
stadt | +0.271 |
Neg logit contributions | |
horizont | -0.277 |
MK | -0.240 |
Dash | -0.204 |
HIP | -0.200 |
MK | -0.200 |
Token | called |
Token ID | 7174 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
horizont | +0.160 |
MK | +0.138 |
Dash | +0.118 |
HIP | +0.116 |
MK | +0.116 |
Neg logit contributions | |
McCartney | -0.155 |
aba | -0.153 |
Reloaded | -0.150 |
igible | -0.149 |
irs | -0.148 |
Token | John |
Token ID | 1757 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
horizont | +0.779 |
MK | +0.687 |
Dash | +0.604 |
HIP | +0.598 |
MK | +0.596 |
Neg logit contributions | |
McCartney | -0.542 |
aba | -0.531 |
Reloaded | -0.523 |
igible | -0.516 |
complement | -0.512 |
Token | Doe |
Token ID | 31780 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
McCartney | +0.176 |
aba | +0.172 |
Reloaded | +0.168 |
igible | +0.167 |
irs | +0.166 |
Neg logit contributions | |
horizont | -0.213 |
MK | -0.185 |
Dash | -0.161 |
HIP | -0.159 |
MK | -0.158 |
Token | investigation |
Token ID | 3645 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
horizont | +1.937 |
MK | +1.685 |
Dash | +1.453 |
HIP | +1.442 |
MK | +1.442 |
Neg logit contributions | |
McCartney | -1.770 |
aba | -1.735 |
Reloaded | -1.734 |
igible | -1.700 |
irs | -1.688 |
Token | into |
Token ID | 656 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
horizont | +0.130 |
MK | +0.112 |
Dash | +0.096 |
HIP | +0.094 |
MK | +0.094 |
Neg logit contributions | |
McCartney | -0.130 |
aba | -0.128 |
Reloaded | -0.126 |
igible | -0.125 |
irs | -0.124 |
Token | alleged |
Token ID | 4260 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
McCartney | +0.270 |
aba | +0.265 |
Reloaded | +0.261 |
igible | +0.258 |
complement | +0.256 |
Neg logit contributions | |
horizont | -0.358 |
MK | -0.314 |
Dash | -0.274 |
HIP | -0.272 |
MK | -0.271 |
Token | campaign |
Token ID | 1923 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
McCartney | +0.459 |
aba | +0.449 |
complement | +0.436 |
igible | +0.435 |
Reloaded | +0.435 |
Neg logit contributions | |
horizont | -0.591 |
MK | -0.516 |
Dash | -0.450 |
ivably | -0.439 |
HIP | -0.438 |
Token | - |
Token ID | 12 |
Feature activation | +2.93e-03 |
Pos logit contributions | |
McCartney | +0.129 |
aba | +0.127 |
Reloaded | +0.123 |
complement | +0.123 |
igible | +0.122 |
Neg logit contributions | |
horizont | -0.185 |
MK | -0.164 |
Dash | -0.143 |
HIP | -0.141 |
MK | -0.140 |