Token | t |
Token ID | 83 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
scar | +0.455 |
illon | +0.436 |
Wil | +0.431 |
ubis | +0.429 |
Thor | +0.409 |
Neg logit contributions | |
IPA | -0.531 |
Rican | -0.492 |
prem | -0.480 |
Orig | -0.461 |
Mortgage | -0.458 |
Token | weet |
Token ID | 7277 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
scar | +0.556 |
illon | +0.526 |
Wil | +0.524 |
ubis | +0.520 |
Thor | +0.496 |
Neg logit contributions | |
IPA | -0.769 |
Rican | -0.716 |
prem | -0.701 |
Orig | -0.676 |
Mortgage | -0.671 |
Token | / |
Token ID | 14 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
IPA | +0.858 |
Rican | +0.787 |
prem | +0.766 |
Orig | +0.732 |
Mortgage | +0.723 |
Neg logit contributions | |
scar | -0.984 |
illon | -0.950 |
Wil | -0.943 |
ubis | -0.934 |
Thor | -0.901 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
IPA | +0.492 |
Rican | +0.457 |
prem | +0.446 |
Orig | +0.428 |
Mortgage | +0.427 |
Neg logit contributions | |
scar | -0.395 |
illon | -0.377 |
Wil | -0.373 |
ubis | -0.372 |
Thor | -0.354 |
Token | Story |
Token ID | 11605 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
IPA | +0.361 |
Rican | +0.333 |
prem | +0.325 |
Orig | +0.312 |
Mortgage | +0.311 |
Neg logit contributions | |
scar | -0.316 |
illon | -0.305 |
ubis | -0.300 |
Wil | -0.297 |
Thor | -0.282 |
Token | highlights |
Token ID | 11330 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
IPA | +0.589 |
Rican | +0.512 |
prem | +0.489 |
Orig | +0.455 |
Mortgage | +0.450 |
Neg logit contributions | |
scar | -1.249 |
illon | -1.213 |
ubis | -1.207 |
Wil | -1.206 |
Thor | -1.166 |
Token | Secretary |
Token ID | 4986 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
IPA | +2.275 |
Rican | +2.060 |
prem | +1.964 |
Orig | +1.914 |
Mortgage | +1.908 |
Neg logit contributions | |
scar | -2.823 |
illon | -2.723 |
ubis | -2.687 |
Wil | -2.685 |
Thor | -2.573 |
Token | of |
Token ID | 286 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
IPA | +0.129 |
Rican | +0.119 |
prem | +0.115 |
Orig | +0.110 |
Mortgage | +0.110 |
Neg logit contributions | |
scar | -0.128 |
illon | -0.123 |
Wil | -0.121 |
ubis | -0.121 |
Thor | -0.116 |
Token | Defense |
Token ID | 5947 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
scar | +0.065 |
illon | +0.063 |
Wil | +0.063 |
ubis | +0.062 |
Thor | +0.060 |
Neg logit contributions | |
IPA | -0.039 |
Rican | -0.035 |
prem | -0.034 |
Orig | -0.032 |
Mortgage | -0.032 |
Token | Chuck |
Token ID | 14663 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
IPA | +0.054 |
Rican | +0.050 |
prem | +0.048 |
Orig | +0.046 |
Mortgage | +0.046 |
Neg logit contributions | |
scar | -0.058 |
illon | -0.055 |
Wil | -0.055 |
ubis | -0.055 |
Thor | -0.052 |
Token | Hag |
Token ID | 21375 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
IPA | +0.417 |
Rican | +0.392 |
prem | +0.382 |
Orig | +0.372 |
Mortgage | +0.369 |
Neg logit contributions | |
scar | -0.212 |
illon | -0.199 |
Wil | -0.197 |
ubis | -0.195 |
Thor | -0.182 |
Token | / |
Token ID | 14 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
scar | +0.848 |
illon | +0.809 |
Wil | +0.801 |
ubis | +0.797 |
Thor | +0.762 |
Neg logit contributions | |
IPA | -0.949 |
Rican | -0.878 |
prem | -0.853 |
Orig | -0.819 |
Mortgage | -0.816 |
Token | Getty |
Token ID | 6633 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
scar | +0.237 |
illon | +0.220 |
ubis | +0.215 |
Wil | +0.214 |
Thor | +0.198 |
Neg logit contributions | |
IPA | -0.566 |
Rican | -0.535 |
prem | -0.523 |
Orig | -0.508 |
Mortgage | -0.507 |
Token | Images |
Token ID | 5382 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
scar | +0.435 |
illon | +0.417 |
Wil | +0.415 |
ubis | +0.412 |
Thor | +0.397 |
Neg logit contributions | |
IPA | -0.368 |
Rican | -0.338 |
prem | -0.326 |
Orig | -0.309 |
Mortgage | -0.309 |
Token | ) |
Token ID | 8 |
Feature activation | -8.65e-02 |
Pos logit contributions | |
scar | +0.153 |
illon | +0.147 |
Wil | +0.145 |
ubis | +0.145 |
Thor | +0.139 |
Neg logit contributions | |
IPA | -0.143 |
Rican | -0.131 |
prem | -0.127 |
Orig | -0.121 |
Mortgage | -0.121 |
Token | Story |
Token ID | 8362 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
scar | +0.075 |
illon | +0.072 |
Wil | +0.071 |
ubis | +0.071 |
Thor | +0.068 |
Neg logit contributions | |
IPA | -0.061 |
Rican | -0.055 |
prem | -0.053 |
Orig | -0.051 |
Mortgage | -0.051 |
Token | Highlights |
Token ID | 30776 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
IPA | +0.092 |
Rican | +0.082 |
prem | +0.079 |
Orig | +0.076 |
Mortgage | +0.074 |
Neg logit contributions | |
scar | -0.131 |
illon | -0.126 |
ubis | -0.126 |
Wil | -0.126 |
Thor | -0.121 |
Token | The |
Token ID | 383 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
IPA | +2.048 |
Rican | +1.843 |
prem | +1.751 |
Mortgage | +1.717 |
Orig | +1.705 |
Neg logit contributions | |
scar | -2.811 |
illon | -2.721 |
ubis | -2.695 |
Wil | -2.668 |
Thor | -2.574 |
Token | United |
Token ID | 1578 |
Feature activation | +9.15e-01 |
Pos logit contributions | |
scar | +0.027 |
illon | +0.026 |
Wil | +0.026 |
ubis | +0.026 |
Thor | +0.025 |
Neg logit contributions | |
IPA | -0.018 |
Rican | -0.016 |
prem | -0.015 |
Orig | -0.014 |
Mortgage | -0.014 |
Token | States |
Token ID | 1829 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
IPA | +0.616 |
Rican | +0.561 |
prem | +0.540 |
Orig | +0.512 |
Mortgage | +0.510 |
Neg logit contributions | |
scar | -0.815 |
illon | -0.784 |
Wil | -0.778 |
ubis | -0.774 |
Thor | -0.747 |
Token | will |
Token ID | 481 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
IPA | +0.296 |
Rican | +0.272 |
prem | +0.262 |
Orig | +0.250 |
Mortgage | +0.249 |
Neg logit contributions | |
scar | -0.339 |
illon | -0.325 |
Wil | -0.322 |
ubis | -0.321 |
Thor | -0.308 |
Token | run |
Token ID | 1057 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
IPA | +0.253 |
Rican | +0.230 |
prem | +0.227 |
Orig | +0.215 |
Mortgage | +0.214 |
Neg logit contributions | |
scar | -0.263 |
illon | -0.253 |
Wil | -0.253 |
ubis | -0.250 |
Thor | -0.241 |
Token | a |
Token ID | 257 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
IPA | +0.504 |
Rican | +0.462 |
prem | +0.450 |
Orig | +0.428 |
Mortgage | +0.425 |
Neg logit contributions | |
scar | -0.516 |
illon | -0.498 |
Wil | -0.492 |
ubis | -0.489 |
Thor | -0.468 |
Token | form |
Token ID | 1296 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
IPA | +0.114 |
Rican | +0.104 |
prem | +0.101 |
Orig | +0.096 |
Mortgage | +0.096 |
Neg logit contributions | |
scar | -0.135 |
illon | -0.130 |
Wil | -0.129 |
ubis | -0.128 |
Thor | -0.123 |
Token | of |
Token ID | 286 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
IPA | +0.046 |
Rican | +0.043 |
prem | +0.041 |
Orig | +0.040 |
Mortgage | +0.039 |
Neg logit contributions | |
scar | -0.046 |
illon | -0.044 |
Wil | -0.044 |
ubis | -0.043 |
Thor | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
IPA | +0.161 |
Rican | +0.146 |
prem | +0.143 |
Orig | +0.135 |
Mortgage | +0.134 |
Neg logit contributions | |
scar | -0.185 |
illon | -0.178 |
Wil | -0.177 |
ubis | -0.176 |
Thor | -0.169 |
Token | occupational |
Token ID | 34266 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
IPA | +0.194 |
Rican | +0.176 |
prem | +0.172 |
Orig | +0.162 |
Mortgage | +0.161 |
Neg logit contributions | |
scar | -0.231 |
illon | -0.223 |
Wil | -0.221 |
ubis | -0.220 |
Thor | -0.211 |
Token | deriv |
Token ID | 16124 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
IPA | +0.447 |
Rican | +0.406 |
prem | +0.395 |
Orig | +0.372 |
Mortgage | +0.371 |
Neg logit contributions | |
scar | -0.531 |
illon | -0.510 |
Wil | -0.507 |
ubis | -0.504 |
Thor | -0.485 |
Token | ational |
Token ID | 864 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
IPA | +2.501 |
Rican | +2.244 |
prem | +2.228 |
Orig | +2.122 |
Mortgage | +2.094 |
Neg logit contributions | |
scar | -2.309 |
Wil | -2.264 |
illon | -2.208 |
ubis | -2.206 |
Thor | -2.145 |
Token | suffix |
Token ID | 35488 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
IPA | +0.804 |
Rican | +0.718 |
prem | +0.699 |
Orig | +0.657 |
Mortgage | +0.652 |
Neg logit contributions | |
scar | -1.087 |
illon | -1.047 |
Wil | -1.041 |
ubis | -1.037 |
Thor | -0.999 |
Token | - |
Token ID | 532 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
IPA | +0.967 |
Rican | +0.875 |
prem | +0.862 |
Orig | +0.816 |
Mortgage | +0.806 |
Neg logit contributions | |
scar | -0.987 |
illon | -0.955 |
Wil | -0.947 |
ubis | -0.938 |
Thor | -0.900 |
Token | cin |
Token ID | 17879 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
scar | +0.298 |
illon | +0.284 |
Wil | +0.282 |
ubis | +0.279 |
Thor | +0.268 |
Neg logit contributions | |
IPA | -0.367 |
Rican | -0.340 |
prem | -0.331 |
Orig | -0.318 |
Mortgage | -0.316 |
Token | , |
Token ID | 11 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
scar | +0.028 |
illon | +0.027 |
Wil | +0.027 |
ubis | +0.026 |
Thor | +0.025 |
Neg logit contributions | |
IPA | -0.030 |
Rican | -0.027 |
prem | -0.026 |
Orig | -0.025 |
Mortgage | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
scar | +0.040 |
illon | +0.038 |
ubis | +0.038 |
Wil | +0.038 |
Thor | +0.036 |
Neg logit contributions | |
IPA | -0.043 |
Rican | -0.040 |
prem | -0.039 |
Orig | -0.037 |
Mortgage | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
scar | +0.099 |
illon | +0.094 |
Wil | +0.093 |
ubis | +0.093 |
Thor | +0.089 |
Neg logit contributions | |
IPA | -0.110 |
Rican | -0.101 |
prem | -0.098 |
Orig | -0.094 |
Mortgage | -0.094 |
Token | Ko |
Token ID | 48735 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
IPA | +0.038 |
Rican | +0.035 |
prem | +0.035 |
Orig | +0.033 |
Mortgage | +0.033 |
Neg logit contributions | |
scar | -0.031 |
illon | -0.030 |
ubis | -0.029 |
Wil | -0.029 |
Thor | -0.028 |
Token | Do |
Token ID | 5211 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
IPA | +0.598 |
Rican | +0.559 |
prem | +0.542 |
Orig | +0.522 |
Mortgage | +0.519 |
Neg logit contributions | |
scar | -0.442 |
illon | -0.417 |
Wil | -0.413 |
ubis | -0.412 |
Thor | -0.388 |
Token | Profile |
Token ID | 13118 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
IPA | +0.583 |
Rican | +0.540 |
prem | +0.524 |
Orig | +0.506 |
Mortgage | +0.502 |
Neg logit contributions | |
scar | -0.489 |
illon | -0.463 |
Wil | -0.459 |
ubis | -0.459 |
Thor | -0.433 |
Token | Joined |
Token ID | 16709 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
IPA | +0.531 |
Rican | +0.483 |
prem | +0.465 |
Orig | +0.442 |
Mortgage | +0.440 |
Neg logit contributions | |
scar | -0.695 |
illon | -0.668 |
Wil | -0.663 |
ubis | -0.660 |
Thor | -0.635 |
Token | December |
Token ID | 3426 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
IPA | +1.829 |
Rican | +1.636 |
prem | +1.566 |
Mortgage | +1.496 |
Orig | +1.492 |
Neg logit contributions | |
scar | -2.796 |
illon | -2.719 |
ubis | -2.679 |
Wil | -2.677 |
Thor | -2.574 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
IPA | +0.484 |
Rican | +0.421 |
prem | +0.396 |
Orig | +0.365 |
Mortgage | +0.363 |
Neg logit contributions | |
scar | -1.155 |
illon | -1.118 |
Wil | -1.112 |
ubis | -1.108 |
Thor | -1.076 |
Token | United |
Token ID | 1578 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
IPA | +0.341 |
Rican | +0.304 |
prem | +0.290 |
Orig | +0.273 |
Mortgage | +0.271 |
Neg logit contributions | |
scar | -0.609 |
illon | -0.589 |
Wil | -0.585 |
ubis | -0.582 |
Thor | -0.563 |
Token | States |
Token ID | 1829 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
IPA | +0.824 |
Rican | +0.742 |
prem | +0.711 |
Orig | +0.683 |
Mortgage | +0.672 |
Neg logit contributions | |
scar | -1.188 |
illon | -1.146 |
Wil | -1.136 |
ubis | -1.136 |
Thor | -1.087 |
Token | 6 |
Token ID | 718 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
IPA | +0.878 |
Rican | +0.789 |
prem | +0.760 |
Orig | +0.727 |
Mortgage | +0.724 |
Neg logit contributions | |
scar | -1.214 |
illon | -1.172 |
Wil | -1.159 |
ubis | -1.156 |
Thor | -1.112 |
Token | year |
Token ID | 614 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
IPA | +0.511 |
Rican | +0.473 |
prem | +0.461 |
Orig | +0.441 |
Mortgage | +0.439 |
Neg logit contributions | |
scar | -0.442 |
illon | -0.421 |
ubis | -0.417 |
Wil | -0.417 |
Thor | -0.396 |
Token | . |
Token ID | 13 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
IPA | +0.247 |
Rican | +0.228 |
prem | +0.223 |
Orig | +0.212 |
Mortgage | +0.211 |
Neg logit contributions | |
scar | -0.241 |
illon | -0.231 |
Wil | -0.229 |
ubis | -0.229 |
Thor | -0.219 |
Token | But |
Token ID | 887 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
IPA | +0.136 |
Rican | +0.124 |
prem | +0.119 |
Orig | +0.115 |
Mortgage | +0.114 |
Neg logit contributions | |
scar | -0.153 |
illon | -0.147 |
Wil | -0.146 |
ubis | -0.146 |
Thor | -0.140 |
Token | , |
Token ID | 11 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
IPA | +0.192 |
Rican | +0.176 |
prem | +0.172 |
Orig | +0.164 |
Mortgage | +0.163 |
Neg logit contributions | |
scar | -0.193 |
illon | -0.185 |
ubis | -0.184 |
Wil | -0.183 |
Thor | -0.175 |
Token | in |
Token ID | 287 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
IPA | +0.201 |
Rican | +0.185 |
prem | +0.180 |
Orig | +0.173 |
Mortgage | +0.171 |
Neg logit contributions | |
scar | -0.203 |
illon | -0.195 |
Wil | -0.194 |
ubis | -0.193 |
Thor | -0.185 |
Token | spite |
Token ID | 15275 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
IPA | +0.300 |
Rican | +0.273 |
prem | +0.267 |
Orig | +0.255 |
Mortgage | +0.252 |
Neg logit contributions | |
scar | -0.335 |
illon | -0.320 |
Wil | -0.318 |
ubis | -0.317 |
Thor | -0.305 |
Token | of |
Token ID | 286 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
IPA | +2.146 |
Rican | +1.940 |
prem | +1.916 |
Orig | +1.828 |
Mortgage | +1.797 |
Neg logit contributions | |
scar | -2.303 |
illon | -2.211 |
ubis | -2.202 |
Wil | -2.193 |
Thor | -2.109 |
Token | the |
Token ID | 262 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
scar | +0.096 |
illon | +0.092 |
Wil | +0.091 |
ubis | +0.091 |
Thor | +0.087 |
Neg logit contributions | |
IPA | -0.091 |
Rican | -0.083 |
prem | -0.081 |
Orig | -0.077 |
Mortgage | -0.076 |
Token | shared |
Token ID | 4888 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
IPA | +0.035 |
Rican | +0.032 |
prem | +0.031 |
Orig | +0.030 |
Mortgage | +0.029 |
Neg logit contributions | |
scar | -0.040 |
illon | -0.038 |
Wil | -0.038 |
ubis | -0.038 |
Thor | -0.036 |
Token | anx |
Token ID | 7296 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
scar | +0.033 |
illon | +0.031 |
Wil | +0.031 |
ubis | +0.031 |
Thor | +0.030 |
Neg logit contributions | |
IPA | -0.032 |
Rican | -0.029 |
prem | -0.028 |
Orig | -0.027 |
Mortgage | -0.027 |
Token | ieties |
Token ID | 9545 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
IPA | +0.503 |
Rican | +0.473 |
prem | +0.463 |
Orig | +0.448 |
Mortgage | +0.447 |
Neg logit contributions | |
scar | -0.253 |
illon | -0.235 |
Wil | -0.233 |
ubis | -0.231 |
Thor | -0.217 |
Token | on |
Token ID | 319 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
IPA | +0.073 |
Rican | +0.067 |
prem | +0.065 |
Orig | +0.062 |
Mortgage | +0.062 |
Neg logit contributions | |
scar | -0.073 |
illon | -0.069 |
Wil | -0.069 |
ubis | -0.068 |
Thor | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
IPA | +0.074 |
Rican | +0.067 |
prem | +0.065 |
Orig | +0.062 |
Mortgage | +0.062 |
Neg logit contributions | |
scar | -0.084 |
illon | -0.081 |
Wil | -0.080 |
ubis | -0.080 |
Thor | -0.077 |
Token | ground |
Token ID | 2323 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
IPA | +0.163 |
Rican | +0.147 |
prem | +0.142 |
Orig | +0.134 |
Mortgage | +0.134 |
Neg logit contributions | |
scar | -0.225 |
illon | -0.216 |
Wil | -0.215 |
ubis | -0.214 |
Thor | -0.206 |
Token | . |
Token ID | 13 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
IPA | +0.024 |
Rican | +0.022 |
prem | +0.022 |
Orig | +0.021 |
Mortgage | +0.021 |
Neg logit contributions | |
scar | -0.024 |
illon | -0.023 |
Wil | -0.023 |
ubis | -0.023 |
Thor | -0.022 |
Token | Targ |
Token ID | 31089 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
IPA | +0.018 |
Rican | +0.017 |
prem | +0.016 |
Orig | +0.015 |
Mortgage | +0.015 |
Neg logit contributions | |
scar | -0.023 |
illon | -0.022 |
Wil | -0.022 |
ubis | -0.022 |
Thor | -0.021 |
Token | ets |
Token ID | 1039 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
scar | +0.018 |
illon | +0.017 |
Wil | +0.016 |
ubis | +0.016 |
Thor | +0.016 |
Neg logit contributions | |
IPA | -0.024 |
Rican | -0.023 |
prem | -0.022 |
Orig | -0.021 |
Mortgage | -0.021 |
Token | will |
Token ID | 481 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
IPA | +2.078 |
Rican | +1.882 |
prem | +1.842 |
Orig | +1.742 |
Mortgage | +1.723 |
Neg logit contributions | |
scar | -2.379 |
illon | -2.280 |
ubis | -2.277 |
Wil | -2.270 |
Thor | -2.174 |
Token | be |
Token ID | 307 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
IPA | +0.327 |
Rican | +0.295 |
prem | +0.290 |
Orig | +0.273 |
Mortgage | +0.271 |
Neg logit contributions | |
scar | -0.395 |
illon | -0.380 |
Wil | -0.380 |
ubis | -0.377 |
Thor | -0.364 |
Token | chosen |
Token ID | 7147 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
IPA | +0.025 |
Rican | +0.023 |
prem | +0.022 |
Orig | +0.021 |
Mortgage | +0.021 |
Neg logit contributions | |
scar | -0.029 |
illon | -0.027 |
Wil | -0.027 |
ubis | -0.027 |
Thor | -0.026 |
Token | based |
Token ID | 1912 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
IPA | +0.046 |
Rican | +0.041 |
prem | +0.040 |
Orig | +0.037 |
Mortgage | +0.037 |
Neg logit contributions | |
scar | -0.070 |
illon | -0.067 |
Wil | -0.067 |
ubis | -0.067 |
Thor | -0.064 |
Token | on |
Token ID | 319 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
IPA | +0.221 |
Rican | +0.203 |
prem | +0.196 |
Orig | +0.187 |
Mortgage | +0.187 |
Neg logit contributions | |
scar | -0.238 |
illon | -0.228 |
Wil | -0.227 |
ubis | -0.225 |
Thor | -0.217 |
Token | vote |
Token ID | 3015 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
scar | +0.130 |
illon | +0.124 |
Wil | +0.124 |
ubis | +0.123 |
Thor | +0.118 |
Neg logit contributions | |
IPA | -0.115 |
Rican | -0.104 |
prem | -0.102 |
Orig | -0.096 |
Mortgage | -0.096 |
Token | accordingly |
Token ID | 16062 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
IPA | +0.536 |
Rican | +0.491 |
prem | +0.476 |
Orig | +0.452 |
Mortgage | +0.450 |
Neg logit contributions | |
scar | -0.585 |
illon | -0.564 |
Wil | -0.559 |
ubis | -0.558 |
Thor | -0.533 |
Token | . |
Token ID | 13 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
IPA | +1.000 |
Rican | +0.918 |
prem | +0.899 |
Orig | +0.857 |
Mortgage | +0.849 |
Neg logit contributions | |
scar | -0.989 |
illon | -0.951 |
Wil | -0.941 |
ubis | -0.939 |
Thor | -0.899 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
IPA | +0.286 |
Rican | +0.263 |
prem | +0.254 |
Orig | +0.247 |
Mortgage | +0.245 |
Neg logit contributions | |
scar | -0.252 |
illon | -0.243 |
ubis | -0.238 |
Wil | -0.238 |
Thor | -0.227 |
Token | Story |
Token ID | 11605 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
scar | +0.040 |
illon | +0.038 |
ubis | +0.038 |
Wil | +0.037 |
Thor | +0.035 |
Neg logit contributions | |
IPA | -0.045 |
Rican | -0.042 |
prem | -0.041 |
Orig | -0.039 |
Mortgage | -0.039 |
Token | highlights |
Token ID | 11330 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
IPA | +0.585 |
Rican | +0.509 |
prem | +0.486 |
Orig | +0.453 |
Mortgage | +0.448 |
Neg logit contributions | |
scar | -1.243 |
illon | -1.208 |
ubis | -1.201 |
Wil | -1.201 |
Thor | -1.161 |
Token | Police |
Token ID | 4287 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
IPA | +1.986 |
Rican | +1.799 |
prem | +1.716 |
Orig | +1.667 |
Mortgage | +1.663 |
Neg logit contributions | |
scar | -2.479 |
illon | -2.389 |
ubis | -2.358 |
Wil | -2.358 |
Thor | -2.260 |
Token | chief |
Token ID | 4039 |
Feature activation | +5.21e-02 |
Pos logit contributions | |
scar | +0.501 |
illon | +0.479 |
Wil | +0.475 |
ubis | +0.472 |
Thor | +0.453 |
Neg logit contributions | |
IPA | -0.486 |
Rican | -0.448 |
prem | -0.433 |
Orig | -0.414 |
Mortgage | -0.413 |
Token | vows |
Token ID | 36605 |
Feature activation | +4.24e-03 |
Pos logit contributions | |
IPA | +0.040 |
Rican | +0.037 |
prem | +0.036 |
Orig | +0.034 |
Mortgage | +0.034 |
Neg logit contributions | |
scar | -0.041 |
illon | -0.040 |
Wil | -0.039 |
ubis | -0.039 |
Thor | -0.038 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
IPA | +0.003 |
Rican | +0.003 |
prem | +0.003 |
Orig | +0.003 |
Mortgage | +0.003 |
Neg logit contributions | |
scar | -0.003 |
illon | -0.003 |
Wil | -0.003 |
ubis | -0.003 |
Thor | -0.003 |
Token | suspect |
Token ID | 4099 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
IPA | +0.118 |
Rican | +0.108 |
prem | +0.104 |
Orig | +0.099 |
Mortgage | +0.099 |
Neg logit contributions | |
scar | -0.141 |
illon | -0.135 |
Wil | -0.134 |
ubis | -0.133 |
Thor | -0.128 |
Token | 55 |
Token ID | 2816 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
IPA | +0.610 |
Rican | +0.561 |
prem | +0.545 |
Orig | +0.529 |
Mortgage | +0.526 |
Neg logit contributions | |
scar | -0.538 |
illon | -0.522 |
ubis | -0.511 |
Wil | -0.508 |
Thor | -0.480 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
IPA | +0.051 |
Rican | +0.046 |
prem | +0.044 |
Orig | +0.042 |
Mortgage | +0.042 |
Neg logit contributions | |
scar | -0.079 |
illon | -0.076 |
Wil | -0.076 |
ubis | -0.075 |
Thor | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
IPA | +1.471 |
Rican | +1.332 |
prem | +1.304 |
Orig | +1.254 |
Mortgage | +1.247 |
Neg logit contributions | |
scar | -1.467 |
illon | -1.413 |
Wil | -1.396 |
ubis | -1.395 |
Thor | -1.327 |
Token |
|
Token ID | 198 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
IPA | +0.548 |
Rican | +0.502 |
prem | +0.490 |
Orig | +0.472 |
Mortgage | +0.468 |
Neg logit contributions | |
scar | -0.511 |
illon | -0.491 |
ubis | -0.485 |
Wil | -0.485 |
Thor | -0.461 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
IPA | +0.348 |
Rican | +0.309 |
prem | +0.297 |
Orig | +0.280 |
Mortgage | +0.278 |
Neg logit contributions | |
scar | -0.608 |
illon | -0.591 |
Wil | -0.583 |
ubis | -0.582 |
Thor | -0.562 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
IPA | +0.859 |
Rican | +0.766 |
prem | +0.722 |
Mortgage | +0.691 |
Orig | +0.690 |
Neg logit contributions | |
scar | -1.459 |
illon | -1.396 |
Wil | -1.392 |
ubis | -1.391 |
Thor | -1.338 |
Token | this |
Token ID | 428 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
IPA | +1.790 |
Rican | +1.596 |
prem | +1.561 |
Orig | +1.466 |
Mortgage | +1.452 |
Neg logit contributions | |
scar | -2.618 |
illon | -2.520 |
Wil | -2.509 |
ubis | -2.500 |
Thor | -2.421 |
Token | comment |
Token ID | 2912 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
IPA | +0.367 |
Rican | +0.335 |
prem | +0.321 |
Orig | +0.306 |
Mortgage | +0.305 |
Neg logit contributions | |
scar | -0.478 |
illon | -0.458 |
Wil | -0.455 |
ubis | -0.453 |
Thor | -0.438 |
Token | on |
Token ID | 319 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
scar | +0.128 |
illon | +0.122 |
Wil | +0.121 |
ubis | +0.121 |
Thor | +0.116 |
Neg logit contributions | |
IPA | -0.119 |
Rican | -0.109 |
prem | -0.106 |
Orig | -0.101 |
Mortgage | -0.101 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
scar | +0.544 |
illon | +0.522 |
Wil | +0.520 |
ubis | +0.516 |
Thor | +0.501 |
Neg logit contributions | |
IPA | -0.406 |
Rican | -0.373 |
prem | -0.358 |
Orig | -0.337 |
Mortgage | -0.336 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
IPA | +0.318 |
Rican | +0.287 |
prem | +0.275 |
Orig | +0.261 |
Mortgage | +0.259 |
Neg logit contributions | |
scar | -0.479 |
illon | -0.462 |
Wil | -0.458 |
ubis | -0.457 |
Thor | -0.441 |
Token | 14 |
Token ID | 1415 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
IPA | +0.628 |
Rican | +0.579 |
prem | +0.561 |
Orig | +0.541 |
Mortgage | +0.540 |
Neg logit contributions | |
scar | -0.577 |
illon | -0.557 |
Wil | -0.546 |
ubis | -0.545 |
Thor | -0.518 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
scar | +0.365 |
illon | +0.354 |
Wil | +0.350 |
ubis | +0.349 |
Thor | +0.337 |
Neg logit contributions | |
IPA | -0.245 |
Rican | -0.221 |
prem | -0.211 |
Orig | -0.201 |
Mortgage | -0.200 |
Token |
|
Token ID | 198 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
IPA | +1.269 |
Rican | +1.156 |
prem | +1.123 |
Orig | +1.080 |
Mortgage | +1.079 |
Neg logit contributions | |
scar | -1.243 |
illon | -1.196 |
Wil | -1.184 |
ubis | -1.176 |
Thor | -1.124 |
Token |
|
Token ID | 198 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
IPA | +0.561 |
Rican | +0.514 |
prem | +0.501 |
Orig | +0.483 |
Mortgage | +0.480 |
Neg logit contributions | |
scar | -0.517 |
illon | -0.497 |
Wil | -0.490 |
ubis | -0.490 |
Thor | -0.467 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
IPA | +0.370 |
Rican | +0.334 |
prem | +0.320 |
Orig | +0.300 |
Mortgage | +0.299 |
Neg logit contributions | |
scar | -0.582 |
illon | -0.559 |
Wil | -0.557 |
ubis | -0.554 |
Thor | -0.536 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
IPA | +0.760 |
Rican | +0.677 |
prem | +0.633 |
Mortgage | +0.607 |
Orig | +0.604 |
Neg logit contributions | |
scar | -1.331 |
illon | -1.274 |
ubis | -1.269 |
Wil | -1.268 |
Thor | -1.219 |
Token | this |
Token ID | 428 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
IPA | +1.795 |
Rican | +1.615 |
prem | +1.560 |
Orig | +1.482 |
Mortgage | +1.474 |
Neg logit contributions | |
scar | -2.511 |
illon | -2.412 |
Wil | -2.396 |
ubis | -2.384 |
Thor | -2.312 |
Token | comment |
Token ID | 2912 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
IPA | +0.388 |
Rican | +0.355 |
prem | +0.341 |
Orig | +0.323 |
Mortgage | +0.322 |
Neg logit contributions | |
scar | -0.502 |
illon | -0.484 |
Wil | -0.478 |
ubis | -0.477 |
Thor | -0.461 |
Token | on |
Token ID | 319 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
scar | +0.252 |
illon | +0.241 |
Wil | +0.237 |
ubis | +0.237 |
Thor | +0.229 |
Neg logit contributions | |
IPA | -0.273 |
Rican | -0.259 |
prem | -0.244 |
Orig | -0.237 |
Mortgage | -0.233 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
scar | +0.514 |
illon | +0.496 |
Wil | +0.494 |
ubis | +0.487 |
Thor | +0.477 |
Neg logit contributions | |
IPA | -0.402 |
Rican | -0.371 |
prem | -0.356 |
Orig | -0.340 |
Mortgage | -0.338 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
IPA | +0.318 |
Rican | +0.287 |
prem | +0.275 |
Orig | +0.260 |
Mortgage | +0.259 |
Neg logit contributions | |
scar | -0.472 |
illon | -0.455 |
Wil | -0.450 |
ubis | -0.450 |
Thor | -0.434 |
Token |
|
Token ID | 198 |
Feature activation | -5.95e-02 |
Pos logit contributions | |
scar | +0.099 |
illon | +0.095 |
Wil | +0.094 |
ubis | +0.094 |
Thor | +0.089 |
Neg logit contributions | |
IPA | -0.110 |
Rican | -0.102 |
prem | -0.099 |
Orig | -0.095 |
Mortgage | -0.095 |
Token | Dark |
Token ID | 17367 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
scar | +0.041 |
illon | +0.040 |
Wil | +0.039 |
ubis | +0.039 |
Thor | +0.037 |
Neg logit contributions | |
IPA | -0.051 |
Rican | -0.047 |
prem | -0.046 |
Orig | -0.044 |
Mortgage | -0.044 |
Token | horse |
Token ID | 30527 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
scar | +0.588 |
illon | +0.560 |
Wil | +0.554 |
ubis | +0.551 |
Thor | +0.525 |
Neg logit contributions | |
IPA | -0.713 |
Rican | -0.662 |
prem | -0.643 |
Orig | -0.619 |
Mortgage | -0.616 |
Token | Profile |
Token ID | 13118 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
scar | +0.239 |
illon | +0.229 |
Wil | +0.227 |
ubis | +0.226 |
Thor | +0.216 |
Neg logit contributions | |
IPA | -0.225 |
Rican | -0.207 |
prem | -0.200 |
Orig | -0.192 |
Mortgage | -0.191 |
Token | Blog |
Token ID | 14001 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
IPA | +0.407 |
Rican | +0.371 |
prem | +0.358 |
Orig | +0.340 |
Mortgage | +0.339 |
Neg logit contributions | |
scar | -0.512 |
illon | -0.492 |
Wil | -0.488 |
ubis | -0.486 |
Thor | -0.468 |
Token | Joined |
Token ID | 16709 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
IPA | +0.317 |
Rican | +0.291 |
prem | +0.281 |
Orig | +0.268 |
Mortgage | +0.268 |
Neg logit contributions | |
scar | -0.353 |
illon | -0.338 |
Wil | -0.336 |
ubis | -0.334 |
Thor | -0.321 |
Token | December |
Token ID | 3426 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
IPA | +1.667 |
Rican | +1.492 |
prem | +1.430 |
Mortgage | +1.366 |
Orig | +1.365 |
Neg logit contributions | |
scar | -2.518 |
illon | -2.448 |
ubis | -2.411 |
Wil | -2.410 |
Thor | -2.318 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
IPA | +0.763 |
Rican | +0.661 |
prem | +0.620 |
Orig | +0.577 |
Mortgage | +0.573 |
Neg logit contributions | |
scar | -1.827 |
illon | -1.772 |
Wil | -1.760 |
ubis | -1.754 |
Thor | -1.703 |
Token | United |
Token ID | 1578 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
IPA | +0.624 |
Rican | +0.553 |
prem | +0.526 |
Orig | +0.497 |
Mortgage | +0.494 |
Neg logit contributions | |
scar | -1.166 |
illon | -1.130 |
Wil | -1.119 |
ubis | -1.117 |
Thor | -1.078 |
Token | States |
Token ID | 1829 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
IPA | +0.806 |
Rican | +0.731 |
prem | +0.703 |
Orig | +0.668 |
Mortgage | +0.663 |
Neg logit contributions | |
scar | -1.109 |
illon | -1.068 |
Wil | -1.059 |
ubis | -1.054 |
Thor | -1.017 |
Token | 228 |
Token ID | 29041 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
IPA | +0.852 |
Rican | +0.767 |
prem | +0.739 |
Orig | +0.708 |
Mortgage | +0.705 |
Neg logit contributions | |
scar | -1.176 |
illon | -1.136 |
Wil | -1.124 |
ubis | -1.120 |
Thor | -1.078 |
Token | 19 |
Token ID | 1129 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
IPA | +0.470 |
Rican | +0.432 |
prem | +0.419 |
Orig | +0.406 |
Mortgage | +0.404 |
Neg logit contributions | |
scar | -0.439 |
illon | -0.426 |
ubis | -0.418 |
Wil | -0.415 |
Thor | -0.394 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
IPA | +0.147 |
Rican | +0.132 |
prem | +0.127 |
Orig | +0.121 |
Mortgage | +0.121 |
Neg logit contributions | |
scar | -0.224 |
illon | -0.216 |
Wil | -0.215 |
ubis | -0.214 |
Thor | -0.207 |
Token |
|
Token ID | 198 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
IPA | +1.580 |
Rican | +1.432 |
prem | +1.403 |
Orig | +1.349 |
Mortgage | +1.341 |
Neg logit contributions | |
scar | -1.576 |
illon | -1.520 |
ubis | -1.502 |
Wil | -1.501 |
Thor | -1.427 |
Token |
|
Token ID | 198 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
IPA | +0.471 |
Rican | +0.431 |
prem | +0.421 |
Orig | +0.405 |
Mortgage | +0.402 |
Neg logit contributions | |
scar | -0.439 |
illon | -0.422 |
Wil | -0.417 |
ubis | -0.417 |
Thor | -0.397 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
IPA | +0.312 |
Rican | +0.276 |
prem | +0.266 |
Orig | +0.252 |
Mortgage | +0.248 |
Neg logit contributions | |
scar | -0.557 |
illon | -0.543 |
Wil | -0.536 |
ubis | -0.535 |
Thor | -0.515 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
IPA | +0.963 |
Rican | +0.856 |
prem | +0.815 |
Mortgage | +0.777 |
Orig | +0.776 |
Neg logit contributions | |
scar | -1.605 |
illon | -1.542 |
Wil | -1.537 |
ubis | -1.532 |
Thor | -1.481 |
Token | this |
Token ID | 428 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
IPA | +1.716 |
Rican | +1.532 |
prem | +1.496 |
Orig | +1.408 |
Mortgage | +1.394 |
Neg logit contributions | |
scar | -2.470 |
illon | -2.379 |
Wil | -2.362 |
ubis | -2.358 |
Thor | -2.277 |
Token | comment |
Token ID | 2912 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
IPA | +0.478 |
Rican | +0.437 |
prem | +0.418 |
Orig | +0.399 |
Mortgage | +0.398 |
Neg logit contributions | |
scar | -0.622 |
illon | -0.597 |
Wil | -0.592 |
ubis | -0.589 |
Thor | -0.570 |
Token | on |
Token ID | 319 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
IPA | +0.077 |
Rican | +0.070 |
prem | +0.068 |
Orig | +0.065 |
Mortgage | +0.065 |
Neg logit contributions | |
scar | -0.083 |
illon | -0.080 |
Wil | -0.079 |
ubis | -0.079 |
Thor | -0.076 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
scar | +0.390 |
illon | +0.375 |
Wil | +0.373 |
ubis | +0.371 |
Thor | +0.359 |
Neg logit contributions | |
IPA | -0.284 |
Rican | -0.260 |
prem | -0.250 |
Orig | -0.235 |
Mortgage | -0.234 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
IPA | +0.377 |
Rican | +0.339 |
prem | +0.326 |
Orig | +0.309 |
Mortgage | +0.307 |
Neg logit contributions | |
scar | -0.564 |
illon | -0.544 |
Wil | -0.540 |
ubis | -0.538 |
Thor | -0.520 |
Token | 27 |
Token ID | 1983 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
IPA | +0.522 |
Rican | +0.480 |
prem | +0.466 |
Orig | +0.452 |
Mortgage | +0.450 |
Neg logit contributions | |
scar | -0.482 |
illon | -0.468 |
ubis | -0.458 |
Wil | -0.455 |
Thor | -0.433 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
IPA | +0.118 |
Rican | +0.106 |
prem | +0.102 |
Orig | +0.098 |
Mortgage | +0.097 |
Neg logit contributions | |
scar | -0.178 |
illon | -0.172 |
Wil | -0.171 |
ubis | -0.170 |
Thor | -0.165 |
Token |
|
Token ID | 198 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
IPA | +1.520 |
Rican | +1.378 |
prem | +1.350 |
Orig | +1.299 |
Mortgage | +1.291 |
Neg logit contributions | |
scar | -1.504 |
illon | -1.449 |
Wil | -1.433 |
ubis | -1.432 |
Thor | -1.363 |
Token |
|
Token ID | 198 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
IPA | +0.443 |
Rican | +0.406 |
prem | +0.397 |
Orig | +0.382 |
Mortgage | +0.379 |
Neg logit contributions | |
scar | -0.412 |
illon | -0.396 |
Wil | -0.391 |
ubis | -0.390 |
Thor | -0.372 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
IPA | +0.287 |
Rican | +0.255 |
prem | +0.245 |
Orig | +0.231 |
Mortgage | +0.229 |
Neg logit contributions | |
scar | -0.504 |
illon | -0.490 |
Wil | -0.485 |
ubis | -0.484 |
Thor | -0.467 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
IPA | +0.905 |
Rican | +0.805 |
prem | +0.768 |
Mortgage | +0.733 |
Orig | +0.731 |
Neg logit contributions | |
scar | -1.491 |
illon | -1.432 |
Wil | -1.427 |
ubis | -1.423 |
Thor | -1.374 |
Token | this |
Token ID | 428 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
IPA | +1.696 |
Rican | +1.516 |
prem | +1.477 |
Orig | +1.394 |
Mortgage | +1.380 |
Neg logit contributions | |
scar | -2.428 |
illon | -2.339 |
Wil | -2.321 |
ubis | -2.316 |
Thor | -2.238 |
Token | comment |
Token ID | 2912 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
IPA | +0.356 |
Rican | +0.326 |
prem | +0.311 |
Orig | +0.297 |
Mortgage | +0.297 |
Neg logit contributions | |
scar | -0.458 |
illon | -0.440 |
Wil | -0.436 |
ubis | -0.434 |
Thor | -0.420 |
Token | on |
Token ID | 319 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
IPA | +0.024 |
Rican | +0.022 |
prem | +0.021 |
Orig | +0.020 |
Mortgage | +0.020 |
Neg logit contributions | |
scar | -0.026 |
illon | -0.025 |
Wil | -0.025 |
ubis | -0.024 |
Thor | -0.023 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
scar | +0.466 |
illon | +0.448 |
Wil | +0.445 |
ubis | +0.442 |
Thor | +0.428 |
Neg logit contributions | |
IPA | -0.338 |
Rican | -0.310 |
prem | -0.297 |
Orig | -0.280 |
Mortgage | -0.279 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
IPA | +0.257 |
Rican | +0.231 |
prem | +0.222 |
Orig | +0.211 |
Mortgage | +0.210 |
Neg logit contributions | |
scar | -0.382 |
illon | -0.368 |
Wil | -0.366 |
ubis | -0.364 |
Thor | -0.352 |
Token | K |
Token ID | 42 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
IPA | +1.113 |
Rican | +1.035 |
prem | +1.002 |
Orig | +0.968 |
Mortgage | +0.957 |
Neg logit contributions | |
scar | -0.935 |
illon | -0.895 |
ubis | -0.885 |
Wil | -0.881 |
Thor | -0.833 |
Token | ier |
Token ID | 959 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
IPA | +0.187 |
Rican | +0.174 |
prem | +0.168 |
Orig | +0.163 |
Mortgage | +0.162 |
Neg logit contributions | |
scar | -0.158 |
illon | -0.150 |
Wil | -0.149 |
ubis | -0.148 |
Thor | -0.142 |
Token | an |
Token ID | 272 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
IPA | +0.774 |
Rican | +0.719 |
prem | +0.698 |
Orig | +0.671 |
Mortgage | +0.667 |
Neg logit contributions | |
scar | -0.664 |
illon | -0.631 |
Wil | -0.628 |
ubis | -0.622 |
Thor | -0.595 |
Token | Profile |
Token ID | 13118 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
IPA | +0.824 |
Rican | +0.743 |
prem | +0.711 |
Orig | +0.671 |
Mortgage | +0.667 |
Neg logit contributions | |
scar | -1.298 |
illon | -1.247 |
Wil | -1.240 |
ubis | -1.237 |
Thor | -1.191 |
Token | Blog |
Token ID | 14001 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
IPA | +0.246 |
Rican | +0.223 |
prem | +0.215 |
Orig | +0.204 |
Mortgage | +0.204 |
Neg logit contributions | |
scar | -0.326 |
illon | -0.314 |
Wil | -0.311 |
ubis | -0.310 |
Thor | -0.298 |
Token | Joined |
Token ID | 16709 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
IPA | +0.207 |
Rican | +0.190 |
prem | +0.183 |
Orig | +0.175 |
Mortgage | +0.174 |
Neg logit contributions | |
scar | -0.229 |
illon | -0.219 |
Wil | -0.217 |
ubis | -0.216 |
Thor | -0.208 |
Token | August |
Token ID | 2932 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
IPA | +1.634 |
Rican | +1.464 |
prem | +1.403 |
Orig | +1.339 |
Mortgage | +1.337 |
Neg logit contributions | |
scar | -2.473 |
illon | -2.405 |
ubis | -2.369 |
Wil | -2.365 |
Thor | -2.276 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +7.14e-01 |
Pos logit contributions | |
IPA | +0.698 |
Rican | +0.603 |
prem | +0.568 |
Orig | +0.525 |
Mortgage | +0.521 |
Neg logit contributions | |
scar | -1.704 |
illon | -1.656 |
Wil | -1.644 |
ubis | -1.637 |
Thor | -1.589 |
Token | United |
Token ID | 1578 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
IPA | +0.391 |
Rican | +0.347 |
prem | +0.331 |
Orig | +0.311 |
Mortgage | +0.309 |
Neg logit contributions | |
scar | -0.726 |
illon | -0.703 |
Wil | -0.697 |
ubis | -0.695 |
Thor | -0.672 |
Token | States |
Token ID | 1829 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
IPA | +0.904 |
Rican | +0.820 |
prem | +0.789 |
Orig | +0.749 |
Mortgage | +0.744 |
Neg logit contributions | |
scar | -1.249 |
illon | -1.203 |
Wil | -1.194 |
ubis | -1.188 |
Thor | -1.146 |
Token | 6 |
Token ID | 718 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
IPA | +0.769 |
Rican | +0.693 |
prem | +0.668 |
Orig | +0.638 |
Mortgage | +0.636 |
Neg logit contributions | |
scar | -1.059 |
illon | -1.022 |
Wil | -1.012 |
ubis | -1.009 |
Thor | -0.971 |
Token | */ |
Token ID | 9466 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
IPA | +0.492 |
Rican | +0.453 |
prem | +0.443 |
Orig | +0.424 |
Mortgage | +0.419 |
Neg logit contributions | |
scar | -0.473 |
illon | -0.454 |
Wil | -0.450 |
ubis | -0.447 |
Thor | -0.432 |
Token | will |
Token ID | 481 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
scar | +0.393 |
illon | +0.377 |
Wil | +0.374 |
ubis | +0.372 |
Thor | +0.358 |
Neg logit contributions | |
IPA | -0.350 |
Rican | -0.321 |
prem | -0.311 |
Orig | -0.296 |
Mortgage | -0.295 |
Token | give |
Token ID | 1577 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
IPA | +0.306 |
Rican | +0.276 |
prem | +0.273 |
Orig | +0.259 |
Mortgage | +0.255 |
Neg logit contributions | |
scar | -0.338 |
Wil | -0.323 |
illon | -0.322 |
ubis | -0.320 |
Thor | -0.309 |
Token | an |
Token ID | 281 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
IPA | +0.160 |
Rican | +0.146 |
prem | +0.142 |
Orig | +0.136 |
Mortgage | +0.135 |
Neg logit contributions | |
scar | -0.181 |
illon | -0.174 |
Wil | -0.172 |
ubis | -0.172 |
Thor | -0.165 |
Token | error |
Token ID | 4049 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
IPA | +0.472 |
Rican | +0.415 |
prem | +0.402 |
Orig | +0.378 |
Mortgage | +0.372 |
Neg logit contributions | |
scar | -0.817 |
illon | -0.790 |
Wil | -0.786 |
ubis | -0.778 |
Thor | -0.758 |
Token | message |
Token ID | 3275 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
IPA | +1.014 |
Rican | +0.932 |
prem | +0.911 |
Orig | +0.867 |
Mortgage | +0.860 |
Neg logit contributions | |
scar | -1.043 |
Wil | -0.996 |
illon | -0.995 |
ubis | -0.990 |
Thor | -0.952 |
Token | for |
Token ID | 329 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
IPA | +2.074 |
Rican | +1.903 |
prem | +1.874 |
Orig | +1.785 |
Mortgage | +1.773 |
Neg logit contributions | |
scar | -2.013 |
Wil | -1.927 |
illon | -1.923 |
ubis | -1.911 |
Thor | -1.835 |
Token | the |
Token ID | 262 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
IPA | +0.784 |
Rican | +0.716 |
prem | +0.696 |
Orig | +0.663 |
Mortgage | +0.659 |
Neg logit contributions | |
scar | -0.898 |
illon | -0.860 |
Wil | -0.855 |
ubis | -0.850 |
Thor | -0.818 |
Token | trailing |
Token ID | 25462 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
IPA | +0.348 |
Rican | +0.317 |
prem | +0.307 |
Orig | +0.292 |
Mortgage | +0.290 |
Neg logit contributions | |
scar | -0.425 |
illon | -0.409 |
Wil | -0.406 |
ubis | -0.404 |
Thor | -0.389 |
Token | " |
Token ID | 366 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
IPA | +0.792 |
Rican | +0.716 |
prem | +0.698 |
Orig | +0.660 |
Mortgage | +0.657 |
Neg logit contributions | |
scar | -0.984 |
illon | -0.945 |
ubis | -0.936 |
Wil | -0.936 |
Thor | -0.900 |
Token | */ |
Token ID | 16208 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
IPA | +0.132 |
Rican | +0.121 |
prem | +0.118 |
Orig | +0.113 |
Mortgage | +0.112 |
Neg logit contributions | |
scar | -0.126 |
illon | -0.121 |
Wil | -0.120 |
ubis | -0.119 |
Thor | -0.114 |
Token | � |
Token ID | 447 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
IPA | +0.135 |
Rican | +0.124 |
prem | +0.121 |
Orig | +0.115 |
Mortgage | +0.115 |
Neg logit contributions | |
scar | -0.138 |
illon | -0.132 |
Wil | -0.131 |
ubis | -0.131 |
Thor | -0.126 |
Token | � |
Token ID | 247 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
IPA | +0.620 |
Rican | +0.582 |
prem | +0.568 |
Orig | +0.547 |
Mortgage | +0.545 |
Neg logit contributions | |
scar | -0.384 |
illon | -0.362 |
Wil | -0.357 |
ubis | -0.354 |
Thor | -0.336 |
Token | s |
Token ID | 82 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
IPA | +0.421 |
Rican | +0.383 |
prem | +0.371 |
Orig | +0.353 |
Mortgage | +0.350 |
Neg logit contributions | |
scar | -0.505 |
illon | -0.487 |
Wil | -0.486 |
ubis | -0.482 |
Thor | -0.466 |
Token | neo |
Token ID | 19102 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
IPA | +0.171 |
Rican | +0.155 |
prem | +0.150 |
Orig | +0.142 |
Mortgage | +0.141 |
Neg logit contributions | |
scar | -0.222 |
illon | -0.213 |
Wil | -0.212 |
ubis | -0.211 |
Thor | -0.203 |
Token | - |
Token ID | 12 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
IPA | +0.449 |
Rican | +0.408 |
prem | +0.394 |
Mortgage | +0.373 |
Orig | +0.372 |
Neg logit contributions | |
scar | -0.556 |
illon | -0.535 |
ubis | -0.529 |
Wil | -0.529 |
Thor | -0.508 |
Token | Ott |
Token ID | 49092 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
IPA | +0.781 |
Rican | +0.716 |
prem | +0.698 |
Orig | +0.662 |
Mortgage | +0.660 |
Neg logit contributions | |
scar | -0.792 |
illon | -0.766 |
ubis | -0.757 |
Wil | -0.749 |
Thor | -0.719 |
Token | oman |
Token ID | 5185 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
IPA | +1.675 |
Rican | +1.511 |
prem | +1.475 |
Orig | +1.378 |
Mortgage | +1.362 |
Neg logit contributions | |
scar | -2.385 |
illon | -2.285 |
Wil | -2.279 |
ubis | -2.276 |
Thor | -2.198 |
Token | an |
Token ID | 281 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
IPA | +0.251 |
Rican | +0.228 |
prem | +0.221 |
Orig | +0.209 |
Mortgage | +0.209 |
Neg logit contributions | |
scar | -0.312 |
illon | -0.300 |
Wil | -0.297 |
ubis | -0.296 |
Thor | -0.285 |
Token | vil |
Token ID | 2991 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
IPA | +0.403 |
Rican | +0.374 |
prem | +0.366 |
Orig | +0.351 |
Mortgage | +0.350 |
Neg logit contributions | |
scar | -0.296 |
illon | -0.284 |
Wil | -0.280 |
ubis | -0.275 |
Thor | -0.266 |
Token | fall |
Token ID | 2121 |
Feature activation | +2.51e-03 |
Pos logit contributions | |
scar | +0.550 |
illon | +0.526 |
Wil | +0.524 |
ubis | +0.520 |
Thor | +0.500 |
Neg logit contributions | |
IPA | -0.543 |
Rican | -0.499 |
prem | -0.486 |
Orig | -0.463 |
Mortgage | -0.461 |
Token | hardest |
Token ID | 17612 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
IPA | +0.002 |
Rican | +0.002 |
prem | +0.002 |
Orig | +0.002 |
Mortgage | +0.002 |
Neg logit contributions | |
scar | -0.002 |
illon | -0.002 |
Wil | -0.002 |
ubis | -0.002 |
Thor | -0.002 |
Token | early |
Token ID | 1903 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
IPA | +0.425 |
Rican | +0.391 |
prem | +0.381 |
Orig | +0.362 |
Mortgage | +0.359 |
Neg logit contributions | |
scar | -0.472 |
illon | -0.453 |
Wil | -0.450 |
ubis | -0.447 |
Thor | -0.430 |
Token | adolescent |
Token ID | 28680 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
IPA | +0.858 |
Rican | +0.789 |
prem | +0.781 |
Orig | +0.730 |
Mortgage | +0.724 |
Neg logit contributions | |
scar | -0.921 |
illon | -0.888 |
ubis | -0.875 |
Wil | -0.874 |
Thor | -0.842 |
Token | boys |
Token ID | 6510 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
IPA | +0.665 |
Rican | +0.613 |
prem | +0.604 |
Orig | +0.569 |
Mortgage | +0.564 |
Neg logit contributions | |
scar | -0.678 |
illon | -0.654 |
ubis | -0.648 |
Wil | -0.645 |
Thor | -0.618 |
Token | : |
Token ID | 25 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
IPA | +0.412 |
Rican | +0.376 |
prem | +0.371 |
Orig | +0.349 |
Mortgage | +0.347 |
Neg logit contributions | |
scar | -0.439 |
illon | -0.425 |
Wil | -0.422 |
ubis | -0.421 |
Thor | -0.404 |
Token | Aff |
Token ID | 6708 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
IPA | +0.332 |
Rican | +0.301 |
prem | +0.290 |
Orig | +0.274 |
Mortgage | +0.273 |
Neg logit contributions | |
scar | -0.471 |
illon | -0.454 |
Wil | -0.451 |
ubis | -0.449 |
Thor | -0.433 |
Token | ili |
Token ID | 2403 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
scar | +0.011 |
illon | +0.011 |
Wil | +0.011 |
ubis | +0.011 |
Thor | +0.010 |
Neg logit contributions | |
IPA | -0.015 |
Rican | -0.014 |
prem | -0.013 |
Orig | -0.013 |
Mortgage | -0.013 |
Token | ative |
Token ID | 876 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
IPA | +2.229 |
Rican | +2.055 |
prem | +2.017 |
Orig | +1.950 |
Mortgage | +1.926 |
Neg logit contributions | |
scar | -1.742 |
illon | -1.666 |
Wil | -1.654 |
ubis | -1.635 |
Thor | -1.572 |
Token | and |
Token ID | 290 |
Feature activation | +6.56e-02 |
Pos logit contributions | |
IPA | +0.120 |
Rican | +0.111 |
prem | +0.108 |
Orig | +0.102 |
Mortgage | +0.102 |
Neg logit contributions | |
scar | -0.132 |
illon | -0.126 |
Wil | -0.125 |
ubis | -0.125 |
Thor | -0.120 |
Token | aggressive |
Token ID | 8361 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
IPA | +0.048 |
Rican | +0.044 |
prem | +0.042 |
Orig | +0.040 |
Mortgage | +0.040 |
Neg logit contributions | |
scar | -0.055 |
illon | -0.053 |
Wil | -0.052 |
ubis | -0.052 |
Thor | -0.050 |
Token | dimensions |
Token ID | 15225 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
scar | +0.284 |
illon | +0.273 |
Wil | +0.270 |
ubis | +0.269 |
Thor | +0.259 |
Neg logit contributions | |
IPA | -0.260 |
Rican | -0.239 |
prem | -0.231 |
Orig | -0.221 |
Mortgage | -0.219 |
Token | and |
Token ID | 290 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
IPA | +0.809 |
Rican | +0.752 |
prem | +0.732 |
Orig | +0.702 |
Mortgage | +0.700 |
Neg logit contributions | |
scar | -0.667 |
illon | -0.635 |
Wil | -0.629 |
ubis | -0.626 |
Thor | -0.598 |
Token | of |
Token ID | 286 |
Feature activation | +5.49e-02 |
Pos logit contributions | |
scar | +0.693 |
illon | +0.660 |
Wil | +0.654 |
ubis | +0.650 |
Thor | +0.620 |
Neg logit contributions | |
IPA | -0.847 |
Rican | -0.787 |
prem | -0.765 |
Orig | -0.735 |
Mortgage | -0.733 |
Token | Dal |
Token ID | 12348 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
IPA | +0.043 |
Rican | +0.040 |
prem | +0.038 |
Orig | +0.037 |
Mortgage | +0.037 |
Neg logit contributions | |
scar | -0.043 |
illon | -0.041 |
Wil | -0.041 |
ubis | -0.041 |
Thor | -0.039 |
Token | K |
Token ID | 42 |
Feature activation | -8.63e-02 |
Pos logit contributions | |
scar | +0.260 |
illon | +0.247 |
Wil | +0.244 |
ubis | +0.243 |
Thor | +0.232 |
Neg logit contributions | |
IPA | -0.329 |
Rican | -0.307 |
prem | -0.298 |
Orig | -0.287 |
Mortgage | -0.286 |
Token | Profile |
Token ID | 13118 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
scar | +0.060 |
illon | +0.057 |
Wil | +0.057 |
ubis | +0.056 |
Thor | +0.054 |
Neg logit contributions | |
IPA | -0.075 |
Rican | -0.069 |
prem | -0.068 |
Orig | -0.065 |
Mortgage | -0.065 |
Token | Blog |
Token ID | 14001 |
Feature activation | -7.18e-02 |
Pos logit contributions | |
IPA | +0.277 |
Rican | +0.252 |
prem | +0.243 |
Orig | +0.231 |
Mortgage | +0.230 |
Neg logit contributions | |
scar | -0.362 |
illon | -0.348 |
Wil | -0.346 |
ubis | -0.344 |
Thor | -0.331 |
Token | Joined |
Token ID | 16709 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
scar | +0.058 |
illon | +0.056 |
Wil | +0.055 |
ubis | +0.055 |
Thor | +0.053 |
Neg logit contributions | |
IPA | -0.054 |
Rican | -0.050 |
prem | -0.048 |
Orig | -0.046 |
Mortgage | -0.045 |
Token | June |
Token ID | 2795 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
IPA | +1.595 |
Rican | +1.427 |
prem | +1.370 |
Mortgage | +1.307 |
Orig | +1.307 |
Neg logit contributions | |
scar | -2.406 |
illon | -2.338 |
ubis | -2.303 |
Wil | -2.302 |
Thor | -2.216 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
IPA | +0.397 |
Rican | +0.344 |
prem | +0.324 |
Orig | +0.300 |
Mortgage | +0.298 |
Neg logit contributions | |
scar | -0.949 |
illon | -0.921 |
Wil | -0.915 |
ubis | -0.912 |
Thor | -0.885 |
Token | Canada |
Token ID | 3340 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
IPA | +0.091 |
Rican | +0.081 |
prem | +0.077 |
Orig | +0.072 |
Mortgage | +0.072 |
Neg logit contributions | |
scar | -0.169 |
illon | -0.164 |
Wil | -0.163 |
ubis | -0.162 |
Thor | -0.157 |
Token | 16 |
Token ID | 1467 |
Feature activation | +1.96e-03 |
Pos logit contributions | |
IPA | +1.098 |
Rican | +0.988 |
prem | +0.951 |
Orig | +0.914 |
Mortgage | +0.912 |
Neg logit contributions | |
scar | -1.521 |
illon | -1.469 |
Wil | -1.451 |
ubis | -1.450 |
Thor | -1.394 |
Token | 10 |
Token ID | 940 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
IPA | +0.002 |
Rican | +0.002 |
prem | +0.001 |
Orig | +0.001 |
Mortgage | +0.001 |
Neg logit contributions | |
scar | -0.001 |
illon | -0.001 |
Wil | -0.001 |
ubis | -0.001 |
Thor | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
scar | +0.293 |
illon | +0.280 |
Wil | +0.277 |
ubis | +0.276 |
Thor | +0.263 |
Neg logit contributions | |
IPA | -0.329 |
Rican | -0.304 |
prem | -0.295 |
Orig | -0.284 |
Mortgage | -0.283 |
Token | re |
Token ID | 260 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
scar | +0.206 |
illon | +0.197 |
ubis | +0.194 |
Wil | +0.194 |
Thor | +0.184 |
Neg logit contributions | |
IPA | -0.252 |
Rican | -0.233 |
prem | -0.228 |
Orig | -0.219 |
Mortgage | -0.217 |
Token | i |
Token ID | 72 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
scar | +0.366 |
illon | +0.347 |
Wil | +0.345 |
ubis | +0.341 |
Thor | +0.326 |
Neg logit contributions | |
IPA | -0.462 |
Rican | -0.430 |
prem | -0.419 |
Orig | -0.402 |
Mortgage | -0.400 |
Token | Profile |
Token ID | 13118 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
scar | +0.369 |
illon | +0.351 |
Wil | +0.348 |
ubis | +0.346 |
Thor | +0.330 |
Neg logit contributions | |
IPA | -0.444 |
Rican | -0.414 |
prem | -0.402 |
Orig | -0.386 |
Mortgage | -0.384 |
Token | Blog |
Token ID | 14001 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
IPA | +0.290 |
Rican | +0.264 |
prem | +0.255 |
Orig | +0.243 |
Mortgage | +0.242 |
Neg logit contributions | |
scar | -0.357 |
illon | -0.343 |
Wil | -0.341 |
ubis | -0.339 |
Thor | -0.326 |
Token | Joined |
Token ID | 16709 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
IPA | +0.365 |
Rican | +0.336 |
prem | +0.325 |
Orig | +0.310 |
Mortgage | +0.309 |
Neg logit contributions | |
scar | -0.399 |
illon | -0.381 |
Wil | -0.378 |
ubis | -0.377 |
Thor | -0.362 |
Token | October |
Token ID | 3267 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
IPA | +1.601 |
Rican | +1.434 |
prem | +1.377 |
Mortgage | +1.315 |
Orig | +1.312 |
Neg logit contributions | |
scar | -2.400 |
illon | -2.331 |
ubis | -2.296 |
Wil | -2.295 |
Thor | -2.209 |
Token | 2002 |
Token ID | 6244 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
IPA | +0.969 |
Rican | +0.837 |
prem | +0.782 |
Orig | +0.725 |
Mortgage | +0.725 |
Neg logit contributions | |
scar | -2.408 |
illon | -2.340 |
Wil | -2.324 |
ubis | -2.317 |
Thor | -2.248 |
Token | United |
Token ID | 1578 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
scar | +0.021 |
illon | +0.020 |
Wil | +0.020 |
ubis | +0.020 |
Thor | +0.019 |
Neg logit contributions | |
IPA | -0.012 |
Rican | -0.011 |
prem | -0.010 |
Orig | -0.009 |
Mortgage | -0.009 |
Token | States |
Token ID | 1829 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
IPA | +0.808 |
Rican | +0.734 |
prem | +0.706 |
Orig | +0.671 |
Mortgage | +0.666 |
Neg logit contributions | |
scar | -1.109 |
illon | -1.068 |
Wil | -1.060 |
ubis | -1.054 |
Thor | -1.016 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
IPA | +0.736 |
Rican | +0.664 |
prem | +0.640 |
Orig | +0.611 |
Mortgage | +0.610 |
Neg logit contributions | |
scar | -1.008 |
illon | -0.972 |
Wil | -0.962 |
ubis | -0.959 |
Thor | -0.923 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
scar | +0.173 |
illon | +0.165 |
Wil | +0.163 |
ubis | +0.163 |
Thor | +0.155 |
Neg logit contributions | |
IPA | -0.194 |
Rican | -0.179 |
prem | -0.174 |
Orig | -0.167 |
Mortgage | -0.167 |
Token | Big |
Token ID | 12804 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
scar | +0.085 |
illon | +0.081 |
ubis | +0.080 |
Wil | +0.080 |
Thor | +0.075 |
Neg logit contributions | |
IPA | -0.103 |
Rican | -0.095 |
prem | -0.093 |
Orig | -0.089 |
Mortgage | -0.089 |
Token | Fan |
Token ID | 22480 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
scar | +0.015 |
illon | +0.014 |
Wil | +0.014 |
ubis | +0.014 |
Thor | +0.013 |
Neg logit contributions | |
IPA | -0.014 |
Rican | -0.013 |
prem | -0.013 |
Orig | -0.012 |
Mortgage | -0.012 |
Token | Profile |
Token ID | 13118 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
scar | +0.362 |
illon | +0.347 |
Wil | +0.344 |
ubis | +0.343 |
Thor | +0.329 |
Neg logit contributions | |
IPA | -0.306 |
Rican | -0.280 |
prem | -0.271 |
Orig | -0.258 |
Mortgage | -0.256 |
Token | Blog |
Token ID | 14001 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
IPA | +0.211 |
Rican | +0.192 |
prem | +0.185 |
Orig | +0.176 |
Mortgage | +0.175 |
Neg logit contributions | |
scar | -0.271 |
illon | -0.261 |
Wil | -0.259 |
ubis | -0.257 |
Thor | -0.248 |
Token | Joined |
Token ID | 16709 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
IPA | +0.137 |
Rican | +0.126 |
prem | +0.122 |
Orig | +0.116 |
Mortgage | +0.116 |
Neg logit contributions | |
scar | -0.152 |
illon | -0.146 |
Wil | -0.145 |
ubis | -0.144 |
Thor | -0.138 |
Token | December |
Token ID | 3426 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
IPA | +1.594 |
Rican | +1.427 |
prem | +1.367 |
Orig | +1.305 |
Mortgage | +1.305 |
Neg logit contributions | |
scar | -2.408 |
illon | -2.342 |
ubis | -2.307 |
Wil | -2.302 |
Thor | -2.214 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
IPA | +0.569 |
Rican | +0.494 |
prem | +0.464 |
Orig | +0.430 |
Mortgage | +0.427 |
Neg logit contributions | |
scar | -1.357 |
illon | -1.316 |
Wil | -1.307 |
ubis | -1.303 |
Thor | -1.265 |
Token | TL |
Token ID | 24811 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
IPA | +0.395 |
Rican | +0.349 |
prem | +0.333 |
Orig | +0.313 |
Mortgage | +0.311 |
Neg logit contributions | |
scar | -0.749 |
illon | -0.725 |
Wil | -0.719 |
ubis | -0.717 |
Thor | -0.693 |
Token | AD |
Token ID | 2885 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
IPA | +0.029 |
Rican | +0.026 |
prem | +0.025 |
Orig | +0.024 |
Mortgage | +0.024 |
Neg logit contributions | |
scar | -0.030 |
illon | -0.028 |
Wil | -0.028 |
ubis | -0.028 |
Thor | -0.027 |
Token | T |
Token ID | 51 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
IPA | +0.125 |
Rican | +0.115 |
prem | +0.112 |
Orig | +0.107 |
Mortgage | +0.106 |
Neg logit contributions | |
scar | -0.122 |
illon | -0.116 |
Wil | -0.116 |
ubis | -0.115 |
Thor | -0.110 |
Token | virtue |
Token ID | 14675 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
scar | +0.137 |
illon | +0.132 |
Wil | +0.130 |
ubis | +0.130 |
Thor | +0.125 |
Neg logit contributions | |
IPA | -0.116 |
Rican | -0.106 |
prem | -0.104 |
Orig | -0.098 |
Mortgage | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
scar | +0.337 |
illon | +0.322 |
Wil | +0.320 |
ubis | +0.318 |
Thor | +0.305 |
Neg logit contributions | |
IPA | -0.347 |
Rican | -0.320 |
prem | -0.312 |
Orig | -0.298 |
Mortgage | -0.296 |
Token |
|
Token ID | 198 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
scar | +0.274 |
illon | +0.263 |
Wil | +0.261 |
ubis | +0.260 |
Thor | +0.250 |
Neg logit contributions | |
IPA | -0.237 |
Rican | -0.217 |
prem | -0.210 |
Orig | -0.200 |
Mortgage | -0.199 |
Token |
|
Token ID | 198 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
scar | +0.035 |
illon | +0.033 |
Wil | +0.033 |
ubis | +0.033 |
Thor | +0.031 |
Neg logit contributions | |
IPA | -0.039 |
Rican | -0.036 |
prem | -0.035 |
Orig | -0.034 |
Mortgage | -0.034 |
Token | In |
Token ID | 818 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
IPA | +0.083 |
Rican | +0.077 |
prem | +0.075 |
Orig | +0.072 |
Mortgage | +0.071 |
Neg logit contributions | |
scar | -0.075 |
illon | -0.072 |
ubis | -0.071 |
Wil | -0.071 |
Thor | -0.067 |
Token | spite |
Token ID | 15275 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
scar | +0.193 |
illon | +0.185 |
Wil | +0.184 |
ubis | +0.183 |
Thor | +0.176 |
Neg logit contributions | |
IPA | -0.163 |
Rican | -0.149 |
prem | -0.144 |
Orig | -0.137 |
Mortgage | -0.136 |
Token | of |
Token ID | 286 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
IPA | +1.921 |
Rican | +1.737 |
prem | +1.725 |
Orig | +1.641 |
Mortgage | +1.619 |
Neg logit contributions | |
scar | -2.059 |
illon | -1.975 |
ubis | -1.955 |
Wil | -1.950 |
Thor | -1.888 |
Token | the |
Token ID | 262 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
scar | +0.141 |
illon | +0.135 |
Wil | +0.134 |
ubis | +0.134 |
Thor | +0.128 |
Neg logit contributions | |
IPA | -0.133 |
Rican | -0.122 |
prem | -0.119 |
Orig | -0.114 |
Mortgage | -0.113 |
Token | bre |
Token ID | 1449 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
scar | +0.270 |
illon | +0.259 |
Wil | +0.257 |
ubis | +0.256 |
Thor | +0.246 |
Neg logit contributions | |
IPA | -0.245 |
Rican | -0.224 |
prem | -0.218 |
Orig | -0.208 |
Mortgage | -0.206 |
Token | vity |
Token ID | 21319 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
scar | +0.921 |
illon | +0.878 |
Wil | +0.872 |
ubis | +0.867 |
Thor | +0.830 |
Neg logit contributions | |
IPA | -1.067 |
Rican | -0.988 |
prem | -0.961 |
Orig | -0.921 |
Mortgage | -0.920 |
Token | of |
Token ID | 286 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
scar | +0.987 |
illon | +0.944 |
Wil | +0.935 |
ubis | +0.927 |
Thor | +0.888 |
Neg logit contributions | |
IPA | -1.049 |
Rican | -0.971 |
prem | -0.938 |
Orig | -0.901 |
Mortgage | -0.900 |
Token | education |
Token ID | 3707 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
scar | +0.107 |
illon | +0.103 |
Wil | +0.102 |
ubis | +0.102 |
Thor | +0.098 |
Neg logit contributions | |
IPA | -0.091 |
Rican | -0.083 |
prem | -0.081 |
Mortgage | -0.076 |
Orig | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
scar | +0.258 |
illon | +0.249 |
Wil | +0.246 |
ubis | +0.245 |
Thor | +0.235 |
Neg logit contributions | |
IPA | -0.274 |
Rican | -0.253 |
prem | -0.246 |
Orig | -0.235 |
Mortgage | -0.235 |
Token | If |
Token ID | 1002 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
scar | +0.195 |
illon | +0.187 |
Wil | +0.184 |
ubis | +0.184 |
Thor | +0.176 |
Neg logit contributions | |
IPA | -0.178 |
Rican | -0.162 |
prem | -0.157 |
Orig | -0.151 |
Mortgage | -0.150 |
Token | you |
Token ID | 345 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
scar | +0.155 |
illon | +0.148 |
Wil | +0.147 |
ubis | +0.146 |
Thor | +0.141 |
Neg logit contributions | |
IPA | -0.146 |
Rican | -0.134 |
prem | -0.130 |
Orig | -0.124 |
Mortgage | -0.124 |
Token | would |
Token ID | 561 |
Feature activation | -9.33e-02 |
Pos logit contributions | |
scar | +0.404 |
illon | +0.387 |
Wil | +0.385 |
ubis | +0.383 |
Thor | +0.369 |
Neg logit contributions | |
IPA | -0.366 |
Rican | -0.334 |
prem | -0.325 |
Orig | -0.309 |
Mortgage | -0.307 |
Token | like |
Token ID | 588 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
scar | +0.072 |
Wil | +0.068 |
illon | +0.068 |
ubis | +0.068 |
Thor | +0.065 |
Neg logit contributions | |
IPA | -0.075 |
Rican | -0.068 |
prem | -0.068 |
Orig | -0.064 |
Mortgage | -0.064 |
Token | to |
Token ID | 284 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
scar | +2.139 |
illon | +2.054 |
Wil | +2.054 |
ubis | +2.024 |
Thor | +1.944 |
Neg logit contributions | |
IPA | -2.203 |
Rican | -2.035 |
prem | -1.965 |
Orig | -1.895 |
Mortgage | -1.883 |
Token | alter |
Token ID | 8343 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
scar | +1.192 |
illon | +1.138 |
Wil | +1.129 |
ubis | +1.122 |
Thor | +1.076 |
Neg logit contributions | |
IPA | -1.227 |
Rican | -1.133 |
prem | -1.099 |
Orig | -1.051 |
Mortgage | -1.048 |
Token | or |
Token ID | 393 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
scar | +0.458 |
illon | +0.438 |
Wil | +0.435 |
ubis | +0.432 |
Thor | +0.415 |
Neg logit contributions | |
IPA | -0.449 |
Rican | -0.414 |
prem | -0.401 |
Orig | -0.383 |
Mortgage | -0.382 |
Token | adjust |
Token ID | 4532 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
scar | +0.225 |
illon | +0.215 |
Wil | +0.213 |
ubis | +0.212 |
Thor | +0.203 |
Neg logit contributions | |
IPA | -0.235 |
Rican | -0.217 |
prem | -0.210 |
Orig | -0.201 |
Mortgage | -0.201 |
Token | the |
Token ID | 262 |
Feature activation | -8.77e-02 |
Pos logit contributions | |
IPA | +0.201 |
Rican | +0.185 |
prem | +0.179 |
Orig | +0.171 |
Mortgage | +0.171 |
Neg logit contributions | |
scar | -0.206 |
illon | -0.197 |
Wil | -0.195 |
ubis | -0.195 |
Thor | -0.187 |
Token | 41 |
Token ID | 6073 |
Feature activation | +5.21e-02 |
Pos logit contributions | |
IPA | +0.468 |
Rican | +0.428 |
prem | +0.411 |
Orig | +0.390 |
Mortgage | +0.387 |
Neg logit contributions | |
scar | -0.616 |
illon | -0.591 |
ubis | -0.588 |
Wil | -0.587 |
Thor | -0.564 |
Token | , |
Token ID | 11 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
IPA | +0.043 |
Rican | +0.040 |
prem | +0.039 |
Orig | +0.037 |
Mortgage | +0.037 |
Neg logit contributions | |
scar | -0.039 |
illon | -0.037 |
ubis | -0.036 |
Wil | -0.036 |
Thor | -0.035 |
Token | he |
Token ID | 339 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
IPA | +0.240 |
Rican | +0.217 |
prem | +0.207 |
Orig | +0.195 |
Mortgage | +0.194 |
Neg logit contributions | |
scar | -0.381 |
illon | -0.367 |
Wil | -0.365 |
ubis | -0.364 |
Thor | -0.352 |
Token | has |
Token ID | 468 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
scar | +0.227 |
illon | +0.217 |
Wil | +0.215 |
ubis | +0.214 |
Thor | +0.206 |
Neg logit contributions | |
IPA | -0.211 |
Rican | -0.194 |
prem | -0.187 |
Orig | -0.179 |
Mortgage | -0.178 |
Token | re |
Token ID | 302 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
scar | +0.104 |
illon | +0.100 |
Wil | +0.099 |
ubis | +0.098 |
Thor | +0.095 |
Neg logit contributions | |
IPA | -0.093 |
Rican | -0.086 |
prem | -0.083 |
Orig | -0.079 |
Mortgage | -0.079 |
Token | pped |
Token ID | 1496 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
scar | +0.163 |
illon | +0.157 |
Wil | +0.155 |
ubis | +0.155 |
Thor | +0.149 |
Neg logit contributions | |
IPA | -0.168 |
Rican | -0.155 |
prem | -0.150 |
Orig | -0.143 |
Mortgage | -0.143 |
Token | the |
Token ID | 262 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
scar | +2.340 |
illon | +2.250 |
Wil | +2.235 |
ubis | +2.204 |
Thor | +2.131 |
Neg logit contributions | |
IPA | -2.004 |
Rican | -1.839 |
prem | -1.770 |
Orig | -1.684 |
Mortgage | -1.682 |
Token | 305 |
Token ID | 32747 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
IPA | +0.157 |
Rican | +0.143 |
prem | +0.138 |
Orig | +0.131 |
Mortgage | +0.130 |
Neg logit contributions | |
scar | -0.214 |
illon | -0.206 |
Wil | -0.204 |
ubis | -0.203 |
Thor | -0.196 |
Token | since |
Token ID | 1201 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
IPA | +0.345 |
Rican | +0.317 |
prem | +0.307 |
Orig | +0.293 |
Mortgage | +0.293 |
Neg logit contributions | |
scar | -0.369 |
illon | -0.353 |
Wil | -0.350 |
ubis | -0.349 |
Thor | -0.335 |
Token | 1973 |
Token ID | 15674 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
scar | +0.448 |
illon | +0.432 |
Wil | +0.429 |
ubis | +0.428 |
Thor | +0.413 |
Neg logit contributions | |
IPA | -0.304 |
Rican | -0.274 |
prem | -0.264 |
Orig | -0.249 |
Mortgage | -0.248 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
IPA | +0.040 |
Rican | +0.037 |
prem | +0.036 |
Orig | +0.034 |
Mortgage | +0.034 |
Neg logit contributions | |
scar | -0.040 |
illon | -0.038 |
Wil | -0.038 |
ubis | -0.038 |
Thor | -0.036 |
Token | est |
Token ID | 1556 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
IPA | +0.121 |
Rican | +0.109 |
prem | +0.105 |
Orig | +0.099 |
Mortgage | +0.098 |
Neg logit contributions | |
scar | -0.180 |
illon | -0.173 |
Wil | -0.172 |
ubis | -0.171 |
Thor | -0.166 |
Token | el |
Token ID | 417 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
scar | +0.336 |
illon | +0.319 |
Wil | +0.316 |
ubis | +0.314 |
Thor | +0.300 |
Neg logit contributions | |
IPA | -0.412 |
Rican | -0.383 |
prem | -0.372 |
Orig | -0.358 |
Mortgage | -0.357 |
Token | ada |
Token ID | 4763 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
scar | +0.377 |
illon | +0.359 |
Wil | +0.355 |
ubis | +0.353 |
Thor | +0.337 |
Neg logit contributions | |
IPA | -0.455 |
Rican | -0.423 |
prem | -0.411 |
Orig | -0.395 |
Mortgage | -0.394 |
Token | flag |
Token ID | 6056 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
scar | +0.758 |
illon | +0.728 |
Wil | +0.724 |
ubis | +0.720 |
Thor | +0.693 |
Neg logit contributions | |
IPA | -0.671 |
Rican | -0.616 |
prem | -0.595 |
Orig | -0.567 |
Mortgage | -0.563 |
Token | cl |
Token ID | 537 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
scar | +0.178 |
illon | +0.171 |
Wil | +0.171 |
ubis | +0.170 |
Thor | +0.164 |
Neg logit contributions | |
IPA | -0.177 |
Rican | -0.163 |
prem | -0.158 |
Orig | -0.151 |
Mortgage | -0.149 |
Token | ung |
Token ID | 2150 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
scar | +0.290 |
Wil | +0.276 |
illon | +0.275 |
ubis | +0.273 |
Thor | +0.264 |
Neg logit contributions | |
IPA | -0.333 |
Rican | -0.305 |
prem | -0.298 |
Orig | -0.288 |
Mortgage | -0.286 |
Token | from |
Token ID | 422 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
scar | +2.158 |
illon | +2.051 |
Wil | +2.038 |
ubis | +2.021 |
Thor | +1.940 |
Neg logit contributions | |
IPA | -2.167 |
Rican | -2.009 |
prem | -1.933 |
Mortgage | -1.862 |
Orig | -1.859 |
Token | the |
Token ID | 262 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
IPA | +0.030 |
Rican | +0.028 |
prem | +0.027 |
Orig | +0.026 |
Mortgage | +0.026 |
Neg logit contributions | |
scar | -0.033 |
illon | -0.032 |
Wil | -0.032 |
ubis | -0.031 |
Thor | -0.030 |
Token | tract |
Token ID | 14998 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
IPA | +0.097 |
Rican | +0.088 |
prem | +0.085 |
Orig | +0.080 |
Mortgage | +0.080 |
Neg logit contributions | |
scar | -0.136 |
illon | -0.131 |
Wil | -0.130 |
ubis | -0.130 |
Thor | -0.125 |
Token | ors |
Token ID | 669 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
IPA | +0.305 |
Rican | +0.283 |
prem | +0.277 |
Mortgage | +0.267 |
Orig | +0.266 |
Neg logit contributions | |
scar | -0.227 |
illon | -0.215 |
Wil | -0.215 |
ubis | -0.215 |
Thor | -0.203 |
Token | in |
Token ID | 287 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
IPA | +0.012 |
Rican | +0.011 |
prem | +0.011 |
Orig | +0.010 |
Mortgage | +0.010 |
Neg logit contributions | |
scar | -0.012 |
illon | -0.012 |
Wil | -0.012 |
ubis | -0.012 |
Thor | -0.011 |
Token | uck |
Token ID | 1347 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
IPA | +0.031 |
Rican | +0.029 |
prem | +0.028 |
Orig | +0.028 |
Mortgage | +0.027 |
Neg logit contributions | |
scar | -0.022 |
illon | -0.020 |
ubis | -0.020 |
Wil | -0.020 |
Thor | -0.019 |
Token | ley |
Token ID | 1636 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
scar | +0.077 |
illon | +0.072 |
Wil | +0.072 |
ubis | +0.072 |
Thor | +0.067 |
Neg logit contributions | |
IPA | -0.135 |
Rican | -0.126 |
prem | -0.123 |
Orig | -0.119 |
Mortgage | -0.119 |
Token | NBA |
Token ID | 32470 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
IPA | +0.300 |
Rican | +0.279 |
prem | +0.273 |
Orig | +0.264 |
Mortgage | +0.263 |
Neg logit contributions | |
scar | -0.215 |
ubis | -0.202 |
illon | -0.202 |
Wil | -0.198 |
Thor | -0.189 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.03e-02 |
Pos logit contributions | |
scar | +0.556 |
illon | +0.526 |
ubis | +0.526 |
Wil | +0.520 |
Thor | +0.496 |
Neg logit contributions | |
IPA | -0.707 |
Rican | -0.653 |
prem | -0.640 |
Orig | -0.618 |
Mortgage | -0.615 |
Token | Ted |
Token ID | 38972 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
IPA | +0.075 |
Rican | +0.069 |
prem | +0.068 |
Orig | +0.065 |
Mortgage | +0.065 |
Neg logit contributions | |
scar | -0.066 |
illon | -0.063 |
ubis | -0.062 |
Wil | -0.062 |
Thor | -0.059 |
Token | Nug |
Token ID | 45777 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
scar | +0.515 |
illon | +0.494 |
Wil | +0.489 |
ubis | +0.488 |
Thor | +0.467 |
Neg logit contributions | |
IPA | -0.476 |
Rican | -0.438 |
prem | -0.423 |
Orig | -0.406 |
Mortgage | -0.403 |
Token | ent |
Token ID | 298 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
scar | +1.870 |
illon | +1.775 |
Wil | +1.758 |
ubis | +1.747 |
Thor | +1.665 |
Neg logit contributions | |
IPA | -2.407 |
Rican | -2.242 |
prem | -2.180 |
Orig | -2.097 |
Mortgage | -2.091 |
Token | promised |
Token ID | 8072 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
scar | +0.235 |
illon | +0.225 |
Wil | +0.223 |
ubis | +0.222 |
Thor | +0.213 |
Neg logit contributions | |
IPA | -0.219 |
Rican | -0.202 |
prem | -0.195 |
Orig | -0.186 |
Mortgage | -0.186 |
Token | Thursday |
Token ID | 3635 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
IPA | +0.367 |
Rican | +0.337 |
prem | +0.326 |
Orig | +0.310 |
Mortgage | +0.309 |
Neg logit contributions | |
scar | -0.411 |
illon | -0.395 |
Wil | -0.391 |
ubis | -0.389 |
Thor | -0.374 |
Token | that |
Token ID | 326 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
scar | +0.290 |
illon | +0.276 |
Wil | +0.275 |
ubis | +0.273 |
Thor | +0.262 |
Neg logit contributions | |
IPA | -0.276 |
Rican | -0.254 |
prem | -0.246 |
Orig | -0.235 |
Mortgage | -0.235 |
Token | he |
Token ID | 339 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
scar | +0.018 |
illon | +0.017 |
Wil | +0.017 |
ubis | +0.017 |
Thor | +0.017 |
Neg logit contributions | |
IPA | -0.013 |
Rican | -0.012 |
prem | -0.011 |
Orig | -0.010 |
Mortgage | -0.010 |
Token | Russia |
Token ID | 3284 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
scar | +0.299 |
illon | +0.287 |
Wil | +0.285 |
ubis | +0.283 |
Thor | +0.273 |
Neg logit contributions | |
IPA | -0.242 |
Rican | -0.221 |
prem | -0.214 |
Orig | -0.204 |
Mortgage | -0.202 |
Token | with |
Token ID | 351 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
scar | +0.076 |
illon | +0.072 |
Wil | +0.072 |
ubis | +0.071 |
Thor | +0.069 |
Neg logit contributions | |
IPA | -0.071 |
Rican | -0.065 |
prem | -0.063 |
Orig | -0.060 |
Mortgage | -0.060 |
Token | Love |
Token ID | 5896 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
scar | +0.460 |
illon | +0.440 |
Wil | +0.438 |
ubis | +0.436 |
Thor | +0.419 |
Neg logit contributions | |
IPA | -0.399 |
Rican | -0.366 |
prem | -0.352 |
Mortgage | -0.336 |
Orig | -0.335 |
Token | . |
Token ID | 13 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
IPA | +0.815 |
Rican | +0.749 |
prem | +0.729 |
Orig | +0.699 |
Mortgage | +0.696 |
Neg logit contributions | |
scar | -0.815 |
illon | -0.782 |
Wil | -0.779 |
ubis | -0.776 |
Thor | -0.741 |
Token | L |
Token ID | 406 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
IPA | +0.039 |
Rican | +0.036 |
prem | +0.035 |
Orig | +0.034 |
Mortgage | +0.034 |
Neg logit contributions | |
scar | -0.039 |
illon | -0.037 |
Wil | -0.037 |
ubis | -0.037 |
Thor | -0.035 |
Token | aid |
Token ID | 1698 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
IPA | +0.220 |
Rican | +0.205 |
prem | +0.201 |
Orig | +0.193 |
Mortgage | +0.192 |
Neg logit contributions | |
scar | -0.160 |
illon | -0.151 |
Wil | -0.150 |
ubis | -0.148 |
Thor | -0.142 |
Token | on |
Token ID | 319 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
scar | +2.036 |
illon | +1.917 |
Wil | +1.903 |
ubis | +1.900 |
Bow | +1.802 |
Neg logit contributions | |
IPA | -2.199 |
Rican | -2.056 |
prem | -1.968 |
Mortgage | -1.897 |
Orig | -1.894 |
Token | cloth |
Token ID | 16270 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
IPA | +0.212 |
Rican | +0.194 |
prem | +0.188 |
Orig | +0.179 |
Mortgage | +0.178 |
Neg logit contributions | |
scar | -0.251 |
illon | -0.240 |
Wil | -0.239 |
ubis | -0.237 |
Thor | -0.228 |
Token | and |
Token ID | 290 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
IPA | +0.030 |
Rican | +0.028 |
prem | +0.027 |
Orig | +0.026 |
Mortgage | +0.026 |
Neg logit contributions | |
scar | -0.032 |
illon | -0.030 |
Wil | -0.030 |
ubis | -0.030 |
Thor | -0.029 |
Token | packed |
Token ID | 11856 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
IPA | +0.361 |
Rican | +0.331 |
prem | +0.321 |
Orig | +0.306 |
Mortgage | +0.305 |
Neg logit contributions | |
scar | -0.391 |
illon | -0.375 |
Wil | -0.371 |
ubis | -0.370 |
Thor | -0.355 |
Token | in |
Token ID | 287 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
scar | +0.385 |
illon | +0.368 |
Wil | +0.364 |
ubis | +0.362 |
Thor | +0.347 |
Neg logit contributions | |
IPA | -0.413 |
Rican | -0.382 |
prem | -0.371 |
Orig | -0.355 |
Mortgage | -0.354 |
Token | a |
Token ID | 64 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
IPA | +0.578 |
Rican | +0.539 |
prem | +0.520 |
Orig | +0.505 |
Mortgage | +0.503 |
Neg logit contributions | |
scar | -0.444 |
Wil | -0.421 |
illon | -0.418 |
ubis | -0.413 |
Thor | -0.397 |
Token | S |
Token ID | 311 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
IPA | +0.235 |
Rican | +0.214 |
prem | +0.206 |
Orig | +0.196 |
Mortgage | +0.196 |
Neg logit contributions | |
scar | -0.297 |
illon | -0.284 |
Wil | -0.283 |
ubis | -0.281 |
Thor | -0.271 |
Token | utra |
Token ID | 35076 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
scar | +0.103 |
illon | +0.098 |
Wil | +0.097 |
ubis | +0.096 |
Thor | +0.092 |
Neg logit contributions | |
IPA | -0.123 |
Rican | -0.114 |
prem | -0.111 |
Orig | -0.106 |
Mortgage | -0.106 |
Token | , |
Token ID | 11 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
scar | +0.330 |
illon | +0.317 |
Wil | +0.314 |
ubis | +0.313 |
Thor | +0.301 |
Neg logit contributions | |
IPA | -0.264 |
Rican | -0.241 |
prem | -0.233 |
Orig | -0.221 |
Mortgage | -0.220 |
Token | L |
Token ID | 406 |
Feature activation | +2.42e-03 |
Pos logit contributions | |
scar | +0.104 |
illon | +0.100 |
Wil | +0.099 |
ubis | +0.099 |
Thor | +0.095 |
Neg logit contributions | |
IPA | -0.086 |
Rican | -0.078 |
prem | -0.075 |
Orig | -0.072 |
Mortgage | -0.071 |
Token | aid |
Token ID | 1698 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
IPA | +0.002 |
Rican | +0.002 |
prem | +0.002 |
Orig | +0.002 |
Mortgage | +0.002 |
Neg logit contributions | |
scar | -0.002 |
illon | -0.002 |
Wil | -0.002 |
ubis | -0.002 |
Thor | -0.001 |
Token | back |
Token ID | 1891 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
scar | +2.349 |
illon | +2.237 |
Wil | +2.223 |
ubis | +2.215 |
Bow | +2.132 |
Neg logit contributions | |
IPA | -1.868 |
Rican | -1.734 |
prem | -1.679 |
Orig | -1.573 |
Mortgage | -1.561 |
Token | Manor |
Token ID | 33555 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
scar | +0.892 |
illon | +0.855 |
Wil | +0.851 |
ubis | +0.845 |
Thor | +0.819 |
Neg logit contributions | |
IPA | -0.659 |
Rican | -0.604 |
prem | -0.587 |
Mortgage | -0.547 |
Orig | -0.546 |
Token | , |
Token ID | 11 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
IPA | +0.128 |
Rican | +0.118 |
prem | +0.114 |
Orig | +0.109 |
Mortgage | +0.109 |
Neg logit contributions | |
scar | -0.137 |
illon | -0.131 |
Wil | -0.130 |
ubis | -0.130 |
Thor | -0.124 |
Token | Y |
Token ID | 575 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
scar | +0.034 |
illon | +0.033 |
Wil | +0.032 |
ubis | +0.032 |
Thor | +0.031 |
Neg logit contributions | |
IPA | -0.028 |
Rican | -0.025 |
prem | -0.024 |
Orig | -0.023 |
Mortgage | -0.023 |
Token | atra |
Token ID | 26066 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
IPA | +0.630 |
Rican | +0.584 |
prem | +0.567 |
Orig | +0.551 |
Mortgage | +0.549 |
Neg logit contributions | |
scar | -0.462 |
illon | -0.436 |
Wil | -0.433 |
ubis | -0.428 |
Thor | -0.412 |
Token | Experimental |
Token ID | 32286 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
IPA | +0.357 |
Rican | +0.326 |
prem | +0.314 |
Orig | +0.299 |
Mortgage | +0.297 |
Neg logit contributions | |
scar | -0.456 |
illon | -0.438 |
Wil | -0.434 |
ubis | -0.432 |
Thor | -0.417 |
Token | subjects |
Token ID | 7481 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
scar | +0.431 |
illon | +0.414 |
Wil | +0.411 |
ubis | +0.409 |
Thor | +0.395 |
Neg logit contributions | |
IPA | -0.335 |
Rican | -0.305 |
prem | -0.295 |
Orig | -0.280 |
Mortgage | -0.278 |
Token | Three |
Token ID | 7683 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
IPA | +0.147 |
Rican | +0.135 |
prem | +0.130 |
Orig | +0.124 |
Mortgage | +0.124 |
Neg logit contributions | |
scar | -0.173 |
illon | -0.166 |
Wil | -0.164 |
ubis | -0.164 |
Thor | -0.157 |
Token | female |
Token ID | 4048 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
IPA | +0.269 |
Rican | +0.247 |
prem | +0.239 |
Orig | +0.229 |
Mortgage | +0.228 |
Neg logit contributions | |
scar | -0.284 |
illon | -0.272 |
Wil | -0.269 |
ubis | -0.268 |
Thor | -0.257 |
Token | r |
Token ID | 374 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
IPA | +0.497 |
Rican | +0.457 |
prem | +0.445 |
Orig | +0.423 |
Mortgage | +0.419 |
Neg logit contributions | |
scar | -0.544 |
illon | -0.521 |
Wil | -0.515 |
ubis | -0.513 |
Thor | -0.495 |
Token | hes |
Token ID | 956 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
scar | +0.105 |
illon | +0.099 |
Wil | +0.098 |
ubis | +0.097 |
Thor | +0.092 |
Neg logit contributions | |
IPA | -0.176 |
Rican | -0.166 |
prem | -0.162 |
Orig | -0.156 |
Mortgage | -0.156 |
Token | us |
Token ID | 385 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
scar | +1.129 |
illon | +1.039 |
Wil | +1.023 |
ubis | +1.012 |
Thor | +0.934 |
Neg logit contributions | |
IPA | -2.941 |
Rican | -2.784 |
prem | -2.724 |
Orig | -2.646 |
Mortgage | -2.640 |
Token | monkeys |
Token ID | 26697 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
scar | +0.426 |
illon | +0.407 |
Wil | +0.404 |
ubis | +0.401 |
Thor | +0.385 |
Neg logit contributions | |
IPA | -0.432 |
Rican | -0.398 |
prem | -0.385 |
Orig | -0.369 |
Mortgage | -0.368 |
Token | ( |
Token ID | 357 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
scar | +0.209 |
illon | +0.200 |
Wil | +0.198 |
ubis | +0.197 |
Thor | +0.190 |
Neg logit contributions | |
IPA | -0.187 |
Rican | -0.172 |
prem | -0.167 |
Orig | -0.159 |
Mortgage | -0.158 |
Token | Mac |
Token ID | 14155 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
scar | +0.144 |
illon | +0.138 |
ubis | +0.135 |
Wil | +0.134 |
Thor | +0.128 |
Neg logit contributions | |
IPA | -0.175 |
Rican | -0.163 |
prem | -0.159 |
Orig | -0.152 |
Mortgage | -0.151 |
Token | aca |
Token ID | 22260 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
scar | +0.048 |
illon | +0.046 |
Wil | +0.046 |
ubis | +0.045 |
Thor | +0.043 |
Neg logit contributions | |
IPA | -0.062 |
Rican | -0.057 |
prem | -0.056 |
Orig | -0.054 |
Mortgage | -0.053 |
Token | can |
Token ID | 460 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
scar | +0.215 |
illon | +0.207 |
Wil | +0.205 |
ubis | +0.205 |
Thor | +0.196 |
Neg logit contributions | |
IPA | -0.223 |
Rican | -0.205 |
prem | -0.201 |
Orig | -0.191 |
Mortgage | -0.190 |
Token | remember |
Token ID | 3505 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
IPA | +0.320 |
Rican | +0.294 |
prem | +0.289 |
Orig | +0.276 |
Mortgage | +0.273 |
Neg logit contributions | |
scar | -0.281 |
illon | -0.268 |
Wil | -0.267 |
ubis | -0.265 |
Thor | -0.255 |
Token | taking |
Token ID | 2263 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
scar | +0.115 |
illon | +0.111 |
Wil | +0.110 |
ubis | +0.109 |
Thor | +0.105 |
Neg logit contributions | |
IPA | -0.116 |
Rican | -0.107 |
prem | -0.104 |
Orig | -0.099 |
Mortgage | -0.098 |
Token | in |
Token ID | 287 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
IPA | +0.193 |
Rican | +0.177 |
prem | +0.172 |
Orig | +0.165 |
Mortgage | +0.162 |
Neg logit contributions | |
scar | -0.217 |
illon | -0.210 |
Wil | -0.206 |
ubis | -0.206 |
Thor | -0.198 |
Token | r |
Token ID | 374 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
scar | +0.732 |
illon | +0.701 |
Wil | +0.696 |
ubis | +0.692 |
Thor | +0.665 |
Neg logit contributions | |
IPA | -0.661 |
Rican | -0.607 |
prem | -0.586 |
Orig | -0.559 |
Mortgage | -0.558 |
Token | agged |
Token ID | 14655 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
IPA | +0.389 |
Rican | +0.360 |
prem | +0.350 |
Orig | +0.336 |
Mortgage | +0.335 |
Neg logit contributions | |
scar | -0.339 |
illon | -0.322 |
Wil | -0.322 |
ubis | -0.318 |
Thor | -0.307 |
Token | breaths |
Token ID | 45576 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
scar | +2.155 |
illon | +2.051 |
Wil | +2.031 |
ubis | +2.019 |
Thor | +1.940 |
Neg logit contributions | |
IPA | -1.877 |
Rican | -1.717 |
prem | -1.652 |
Mortgage | -1.588 |
Orig | -1.575 |
Token | and |
Token ID | 290 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
IPA | +0.154 |
Rican | +0.141 |
prem | +0.138 |
Orig | +0.131 |
Mortgage | +0.130 |
Neg logit contributions | |
scar | -0.157 |
illon | -0.152 |
Wil | -0.150 |
ubis | -0.150 |
Thor | -0.144 |
Token | feeling |
Token ID | 4203 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
IPA | +0.135 |
Rican | +0.124 |
prem | +0.121 |
Orig | +0.116 |
Mortgage | +0.115 |
Neg logit contributions | |
scar | -0.133 |
illon | -0.127 |
Wil | -0.126 |
ubis | -0.126 |
Thor | -0.120 |
Token | my |
Token ID | 616 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
IPA | +0.551 |
Rican | +0.506 |
prem | +0.494 |
Orig | +0.471 |
Mortgage | +0.466 |
Neg logit contributions | |
scar | -0.550 |
illon | -0.532 |
Wil | -0.525 |
ubis | -0.524 |
Thor | -0.500 |
Token | own |
Token ID | 898 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
IPA | +0.175 |
Rican | +0.160 |
prem | +0.155 |
Orig | +0.147 |
Mortgage | +0.146 |
Neg logit contributions | |
scar | -0.214 |
illon | -0.207 |
Wil | -0.205 |
ubis | -0.204 |
Thor | -0.196 |
Token | IFF |
Token ID | 29267 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
scar | +0.285 |
illon | +0.269 |
Wil | +0.264 |
ubis | +0.262 |
Thor | +0.247 |
Neg logit contributions | |
IPA | -0.481 |
Rican | -0.453 |
prem | -0.439 |
Mortgage | -0.425 |
Orig | -0.424 |
Token | original |
Token ID | 2656 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
scar | +0.240 |
illon | +0.237 |
Wil | +0.236 |
ubis | +0.225 |
Thor | +0.224 |
Neg logit contributions | |
IPA | -0.135 |
Rican | -0.128 |
prem | -0.113 |
Mortgage | -0.112 |
audits | -0.108 |
Token | image |
Token ID | 2939 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
IPA | +0.311 |
Rican | +0.283 |
prem | +0.273 |
Orig | +0.259 |
Mortgage | +0.257 |
Neg logit contributions | |
scar | -0.405 |
illon | -0.390 |
Wil | -0.387 |
ubis | -0.385 |
Thor | -0.371 |
Token | Download |
Token ID | 10472 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
scar | +0.380 |
illon | +0.364 |
Wil | +0.361 |
ubis | +0.358 |
Thor | +0.344 |
Neg logit contributions | |
IPA | -0.372 |
Rican | -0.344 |
prem | -0.333 |
Orig | -0.318 |
Mortgage | -0.317 |
Token | : |
Token ID | 25 |
Feature activation | -6.31e-02 |
Pos logit contributions | |
scar | +0.600 |
illon | +0.576 |
Wil | +0.571 |
ubis | +0.567 |
Thor | +0.545 |
Neg logit contributions | |
IPA | -0.545 |
Rican | -0.504 |
prem | -0.484 |
Mortgage | -0.463 |
Orig | -0.462 |
Token | Figure |
Token ID | 11291 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
scar | +0.060 |
illon | +0.057 |
Wil | +0.057 |
ubis | +0.057 |
Thor | +0.055 |
Neg logit contributions | |
IPA | -0.039 |
Rican | -0.035 |
prem | -0.034 |
Orig | -0.032 |
Mortgage | -0.032 |
Token | 1 |
Token ID | 352 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
scar | +2.108 |
illon | +2.025 |
Wil | +2.009 |
ubis | +2.001 |
Thor | +1.912 |
Neg logit contributions | |
IPA | -1.838 |
Rican | -1.727 |
prem | -1.645 |
Mortgage | -1.581 |
Orig | -1.572 |
Token | . |
Token ID | 13 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
scar | +0.415 |
illon | +0.394 |
Wil | +0.391 |
ubis | +0.388 |
Thor | +0.372 |
Neg logit contributions | |
IPA | -0.480 |
Rican | -0.450 |
prem | -0.434 |
Orig | -0.414 |
Mortgage | -0.414 |
Token | Maze |
Token ID | 33412 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
scar | +0.616 |
illon | +0.592 |
Wil | +0.589 |
ubis | +0.586 |
Thor | +0.565 |
Neg logit contributions | |
IPA | -0.477 |
Rican | -0.439 |
prem | -0.419 |
Orig | -0.395 |
Mortgage | -0.394 |
Token | and |
Token ID | 290 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
IPA | +0.181 |
Rican | +0.166 |
prem | +0.160 |
Orig | +0.152 |
Mortgage | +0.151 |
Neg logit contributions | |
scar | -0.229 |
illon | -0.220 |
Wil | -0.219 |
ubis | -0.218 |
Thor | -0.210 |
Token | task |
Token ID | 4876 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
IPA | +0.071 |
Rican | +0.065 |
prem | +0.063 |
Orig | +0.060 |
Mortgage | +0.060 |
Neg logit contributions | |
scar | -0.082 |
illon | -0.079 |
Wil | -0.078 |
ubis | -0.078 |
Thor | -0.075 |
Token | $ |
Token ID | 720 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
IPA | +0.032 |
Rican | +0.029 |
prem | +0.028 |
Orig | +0.027 |
Mortgage | +0.027 |
Neg logit contributions | |
scar | -0.043 |
illon | -0.041 |
Wil | -0.041 |
ubis | -0.040 |
Thor | -0.039 |
Token | 57 |
Token ID | 3553 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
scar | +0.272 |
illon | +0.260 |
Wil | +0.257 |
ubis | +0.256 |
Thor | +0.245 |
Neg logit contributions | |
IPA | -0.308 |
Rican | -0.285 |
prem | -0.277 |
Orig | -0.266 |
Mortgage | -0.265 |
Token | million |
Token ID | 1510 |
Feature activation | +6.64e-02 |
Pos logit contributions | |
scar | +0.214 |
illon | +0.204 |
Wil | +0.202 |
ubis | +0.201 |
Thor | +0.192 |
Neg logit contributions | |
IPA | -0.267 |
Rican | -0.249 |
prem | -0.242 |
Mortgage | -0.232 |
Orig | -0.232 |
Token | spent |
Token ID | 3377 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
IPA | +0.051 |
Rican | +0.047 |
prem | +0.046 |
Orig | +0.043 |
Mortgage | +0.043 |
Neg logit contributions | |
scar | -0.053 |
illon | -0.051 |
Wil | -0.050 |
ubis | -0.050 |
Thor | -0.048 |
Token | so |
Token ID | 523 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
scar | +0.817 |
illon | +0.782 |
Wil | +0.776 |
ubis | +0.771 |
Thor | +0.741 |
Neg logit contributions | |
IPA | -0.778 |
Rican | -0.717 |
prem | -0.694 |
Orig | -0.663 |
Mortgage | -0.660 |
Token | far |
Token ID | 1290 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
scar | +0.636 |
illon | +0.600 |
Wil | +0.594 |
ubis | +0.592 |
Thor | +0.557 |
Neg logit contributions | |
IPA | -1.078 |
Rican | -1.012 |
prem | -0.992 |
Orig | -0.952 |
Mortgage | -0.949 |
Token | , |
Token ID | 11 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
scar | +1.958 |
illon | +1.871 |
Wil | +1.850 |
ubis | +1.839 |
Thor | +1.760 |
Neg logit contributions | |
IPA | -2.008 |
Rican | -1.853 |
prem | -1.782 |
Orig | -1.719 |
Mortgage | -1.713 |
Token | there |
Token ID | 612 |
Feature activation | +6.96e-02 |
Pos logit contributions | |
scar | +0.205 |
illon | +0.196 |
Wil | +0.195 |
ubis | +0.195 |
Thor | +0.187 |
Neg logit contributions | |
IPA | -0.171 |
Rican | -0.156 |
prem | -0.151 |
Orig | -0.144 |
Mortgage | -0.143 |
Token | is |
Token ID | 318 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
IPA | +0.048 |
Rican | +0.044 |
prem | +0.043 |
Orig | +0.040 |
Mortgage | +0.040 |
Neg logit contributions | |
scar | -0.061 |
illon | -0.058 |
Wil | -0.058 |
ubis | -0.057 |
Thor | -0.055 |
Token | no |
Token ID | 645 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
scar | +0.209 |
illon | +0.200 |
Wil | +0.198 |
ubis | +0.198 |
Thor | +0.189 |
Neg logit contributions | |
IPA | -0.198 |
Rican | -0.182 |
prem | -0.177 |
Orig | -0.169 |
Mortgage | -0.168 |
Token | sign |
Token ID | 1051 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
scar | +0.649 |
illon | +0.622 |
Wil | +0.617 |
ubis | +0.614 |
Thor | +0.590 |
Neg logit contributions | |
IPA | -0.602 |
Rican | -0.553 |
prem | -0.536 |
Orig | -0.511 |
Mortgage | -0.509 |
Token | s |
Token ID | 82 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
IPA | +0.191 |
Rican | +0.175 |
prem | +0.169 |
Orig | +0.161 |
Mortgage | +0.161 |
Neg logit contributions | |
scar | -0.220 |
illon | -0.211 |
Wil | -0.209 |
ubis | -0.208 |
Thor | -0.200 |
Token | journalism |
Token ID | 14033 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
IPA | +0.193 |
Rican | +0.179 |
prem | +0.174 |
Orig | +0.167 |
Mortgage | +0.167 |
Neg logit contributions | |
scar | -0.156 |
illon | -0.148 |
Wil | -0.147 |
ubis | -0.147 |
Thor | -0.140 |
Token | . |
Token ID | 13 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
scar | +0.580 |
illon | +0.554 |
Wil | +0.549 |
ubis | +0.546 |
Thor | +0.523 |
Neg logit contributions | |
IPA | -0.628 |
Rican | -0.581 |
prem | -0.563 |
Orig | -0.540 |
Mortgage | -0.539 |
Token | You |
Token ID | 921 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
scar | +0.178 |
illon | +0.171 |
ubis | +0.169 |
Wil | +0.168 |
Thor | +0.161 |
Neg logit contributions | |
IPA | -0.165 |
Rican | -0.151 |
prem | -0.145 |
Mortgage | -0.141 |
Orig | -0.140 |
Token | can |
Token ID | 460 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
scar | +0.114 |
illon | +0.109 |
Wil | +0.108 |
ubis | +0.108 |
Thor | +0.104 |
Neg logit contributions | |
IPA | -0.089 |
Rican | -0.081 |
prem | -0.078 |
Orig | -0.074 |
Mortgage | -0.074 |
Token | read |
Token ID | 1100 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
IPA | +0.029 |
Rican | +0.027 |
prem | +0.026 |
Mortgage | +0.025 |
Orig | +0.025 |
Neg logit contributions | |
scar | -0.030 |
illon | -0.029 |
ubis | -0.029 |
Wil | -0.029 |
Thor | -0.027 |
Token | our |
Token ID | 674 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
scar | +2.139 |
illon | +2.037 |
Wil | +2.036 |
ubis | +2.000 |
Thor | +1.936 |
Neg logit contributions | |
IPA | -1.811 |
Rican | -1.670 |
prem | -1.595 |
Orig | -1.537 |
Mortgage | -1.534 |
Token | Privacy |
Token ID | 16777 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
scar | +1.245 |
illon | +1.200 |
Wil | +1.193 |
ubis | +1.185 |
Thor | +1.148 |
Neg logit contributions | |
IPA | -0.746 |
Rican | -0.670 |
prem | -0.638 |
Orig | -0.601 |
Mortgage | -0.596 |
Token | Policy |
Token ID | 7820 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
scar | +1.929 |
illon | +1.866 |
Wil | +1.854 |
ubis | +1.841 |
Thor | +1.793 |
Neg logit contributions | |
IPA | -0.809 |
Rican | -0.714 |
prem | -0.675 |
Orig | -0.615 |
audits | -0.607 |
Token | here |
Token ID | 994 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
scar | +0.863 |
illon | +0.815 |
Wil | +0.806 |
ubis | +0.803 |
Thor | +0.757 |
Neg logit contributions | |
IPA | -1.384 |
Rican | -1.293 |
prem | -1.266 |
Orig | -1.218 |
Mortgage | -1.215 |
Token | . |
Token ID | 13 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
scar | +0.383 |
illon | +0.367 |
Wil | +0.363 |
ubis | +0.362 |
Thor | +0.346 |
Neg logit contributions | |
IPA | -0.407 |
Rican | -0.375 |
prem | -0.364 |
Orig | -0.349 |
Mortgage | -0.348 |
Token | working |
Token ID | 1762 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
scar | +0.108 |
illon | +0.103 |
Wil | +0.103 |
ubis | +0.102 |
Thor | +0.098 |
Neg logit contributions | |
IPA | -0.103 |
Rican | -0.095 |
prem | -0.092 |
Orig | -0.088 |
Mortgage | -0.087 |
Token | downtown |
Token ID | 9436 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
IPA | +0.176 |
Rican | +0.161 |
prem | +0.157 |
Orig | +0.149 |
Mortgage | +0.148 |
Neg logit contributions | |
scar | -0.191 |
illon | -0.184 |
Wil | -0.182 |
ubis | -0.182 |
Thor | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
IPA | +0.669 |
Rican | +0.613 |
prem | +0.598 |
Orig | +0.569 |
Mortgage | +0.568 |
Neg logit contributions | |
scar | -0.695 |
illon | -0.666 |
ubis | -0.660 |
Wil | -0.659 |
Thor | -0.630 |
Token | ( |
Token ID | 357 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
scar | +0.038 |
illon | +0.036 |
Wil | +0.036 |
ubis | +0.036 |
Thor | +0.034 |
Neg logit contributions | |
IPA | -0.034 |
Rican | -0.031 |
prem | -0.030 |
Mortgage | -0.028 |
Orig | -0.028 |
Token | CBC |
Token ID | 29208 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
scar | +0.138 |
illon | +0.132 |
ubis | +0.130 |
Wil | +0.129 |
Thor | +0.123 |
Neg logit contributions | |
IPA | -0.162 |
Rican | -0.150 |
prem | -0.146 |
Orig | -0.140 |
Mortgage | -0.140 |
Token | News |
Token ID | 3000 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
IPA | +0.152 |
Rican | +0.139 |
prem | +0.135 |
Orig | +0.129 |
Mortgage | +0.129 |
Neg logit contributions | |
scar | -0.169 |
illon | -0.161 |
ubis | -0.160 |
Wil | -0.160 |
Thor | -0.153 |
Token | ) |
Token ID | 8 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
scar | +1.968 |
illon | +1.886 |
Wil | +1.869 |
ubis | +1.848 |
Thor | +1.778 |
Neg logit contributions | |
IPA | -1.934 |
Rican | -1.797 |
prem | -1.731 |
Orig | -1.644 |
Mortgage | -1.633 |
Token | Staff |
Token ID | 9983 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
IPA | +0.309 |
Rican | +0.285 |
prem | +0.276 |
Orig | +0.264 |
Mortgage | +0.264 |
Neg logit contributions | |
scar | -0.307 |
illon | -0.293 |
Wil | -0.290 |
ubis | -0.289 |
Thor | -0.277 |
Token | in |
Token ID | 287 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
scar | +1.130 |
illon | +1.081 |
Wil | +1.073 |
ubis | +1.068 |
Thor | +1.026 |
Neg logit contributions | |
IPA | -1.057 |
Rican | -0.973 |
prem | -0.941 |
Orig | -0.898 |
Mortgage | -0.896 |
Token | northern |
Token ID | 7840 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
IPA | +0.153 |
Rican | +0.139 |
prem | +0.135 |
Orig | +0.128 |
Mortgage | +0.127 |
Neg logit contributions | |
scar | -0.194 |
illon | -0.186 |
Wil | -0.184 |
ubis | -0.184 |
Thor | -0.177 |
Token | regions |
Token ID | 7652 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
IPA | +0.185 |
Rican | +0.158 |
prem | +0.149 |
Orig | +0.134 |
Mortgage | +0.133 |
Neg logit contributions | |
scar | -0.524 |
illon | -0.508 |
Wil | -0.505 |
ubis | -0.503 |
Thor | -0.489 |
Token | age |
Token ID | 2479 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
scar | +0.425 |
illon | +0.404 |
Wil | +0.400 |
ubis | +0.398 |
Thor | +0.380 |
Neg logit contributions | |
IPA | -0.517 |
Rican | -0.481 |
prem | -0.467 |
Orig | -0.449 |
Mortgage | -0.447 |
Token | groups |
Token ID | 2628 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
IPA | +0.145 |
Rican | +0.131 |
prem | +0.126 |
Orig | +0.119 |
Mortgage | +0.119 |
Neg logit contributions | |
scar | -0.207 |
illon | -0.199 |
Wil | -0.198 |
ubis | -0.197 |
Thor | -0.191 |
Token | . |
Token ID | 13 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
IPA | +0.353 |
Rican | +0.325 |
prem | +0.316 |
Orig | +0.301 |
Mortgage | +0.300 |
Neg logit contributions | |
scar | -0.358 |
illon | -0.344 |
Wil | -0.340 |
ubis | -0.339 |
Thor | -0.325 |
Token | You |
Token ID | 921 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
IPA | +0.040 |
Rican | +0.037 |
prem | +0.036 |
Orig | +0.034 |
Mortgage | +0.034 |
Neg logit contributions | |
scar | -0.041 |
illon | -0.039 |
Wil | -0.039 |
ubis | -0.039 |
Thor | -0.037 |
Token | can |
Token ID | 460 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
IPA | +0.086 |
Rican | +0.079 |
prem | +0.077 |
Orig | +0.072 |
Mortgage | +0.072 |
Neg logit contributions | |
scar | -0.101 |
illon | -0.097 |
ubis | -0.096 |
Wil | -0.096 |
Thor | -0.092 |
Token | read |
Token ID | 1100 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
IPA | +0.251 |
Rican | +0.230 |
prem | +0.227 |
Mortgage | +0.213 |
Orig | +0.213 |
Neg logit contributions | |
scar | -0.247 |
illon | -0.236 |
Wil | -0.235 |
ubis | -0.234 |
Thor | -0.224 |
Token | more |
Token ID | 517 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
scar | +1.933 |
illon | +1.854 |
Wil | +1.842 |
ubis | +1.815 |
Thor | +1.759 |
Neg logit contributions | |
IPA | -1.935 |
Rican | -1.784 |
prem | -1.717 |
Mortgage | -1.654 |
Orig | -1.651 |
Token | about |
Token ID | 546 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
scar | +0.615 |
illon | +0.591 |
Wil | +0.587 |
ubis | +0.584 |
Thor | +0.564 |
Neg logit contributions | |
IPA | -0.463 |
Rican | -0.421 |
prem | -0.406 |
Orig | -0.385 |
Mortgage | -0.383 |
Token | the |
Token ID | 262 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
scar | +1.207 |
illon | +1.157 |
Wil | +1.150 |
ubis | +1.141 |
Thor | +1.099 |
Neg logit contributions | |
IPA | -1.131 |
Rican | -1.042 |
prem | -1.005 |
Orig | -0.962 |
Mortgage | -0.960 |
Token | New |
Token ID | 968 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
scar | +0.728 |
illon | +0.702 |
Wil | +0.699 |
ubis | +0.694 |
Thor | +0.673 |
Neg logit contributions | |
IPA | -0.471 |
Rican | -0.426 |
prem | -0.407 |
Orig | -0.384 |
Mortgage | -0.383 |
Token | York |
Token ID | 1971 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
scar | +0.060 |
illon | +0.057 |
Wil | +0.057 |
ubis | +0.057 |
Thor | +0.054 |
Neg logit contributions | |
IPA | -0.049 |
Rican | -0.045 |
prem | -0.043 |
Orig | -0.041 |
Mortgage | -0.041 |
Token | m |
Token ID | 76 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
scar | +0.040 |
illon | +0.039 |
Wil | +0.038 |
ubis | +0.038 |
Thor | +0.037 |
Neg logit contributions | |
IPA | -0.046 |
Rican | -0.043 |
prem | -0.042 |
Orig | -0.039 |
Mortgage | -0.039 |
Token | eth |
Token ID | 2788 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
scar | +0.055 |
Wil | +0.052 |
illon | +0.051 |
ubis | +0.051 |
Thor | +0.048 |
Neg logit contributions | |
IPA | -0.101 |
Rican | -0.094 |
prem | -0.092 |
Orig | -0.089 |
Mortgage | -0.088 |
Token | ox |
Token ID | 1140 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
scar | +0.698 |
illon | +0.665 |
Wil | +0.660 |
ubis | +0.655 |
Thor | +0.627 |
Neg logit contributions | |
IPA | -0.804 |
Rican | -0.746 |
prem | -0.724 |
Orig | -0.695 |
Mortgage | -0.692 |
Token | ym |
Token ID | 4948 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
scar | +0.423 |
illon | +0.402 |
Wil | +0.397 |
ubis | +0.394 |
Thor | +0.375 |
Neg logit contributions | |
IPA | -0.578 |
Rican | -0.539 |
prem | -0.525 |
Orig | -0.505 |
Mortgage | -0.504 |
Token | ethyl |
Token ID | 21610 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
scar | +0.764 |
illon | +0.714 |
Wil | +0.709 |
ubis | +0.701 |
Thor | +0.661 |
Neg logit contributions | |
IPA | -1.472 |
Rican | -1.382 |
prem | -1.351 |
Orig | -1.306 |
Mortgage | -1.303 |
Token | phen |
Token ID | 31024 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
scar | +1.223 |
illon | +1.171 |
Wil | +1.162 |
ubis | +1.156 |
Thor | +1.111 |
Neg logit contributions | |
IPA | -1.117 |
Rican | -1.026 |
prem | -0.992 |
Orig | -0.946 |
Mortgage | -0.943 |
Token | yl |
Token ID | 2645 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
scar | +1.129 |
illon | +1.049 |
Wil | +1.026 |
ubis | +1.015 |
Thor | +0.941 |
Neg logit contributions | |
IPA | -2.741 |
Rican | -2.600 |
prem | -2.544 |
Orig | -2.468 |
Mortgage | -2.466 |
Token | )- |
Token ID | 13219 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
scar | +1.350 |
illon | +1.293 |
Wil | +1.282 |
ubis | +1.275 |
Thor | +1.226 |
Neg logit contributions | |
IPA | -1.233 |
Rican | -1.137 |
prem | -1.097 |
Orig | -1.049 |
Mortgage | -1.046 |
Token | 2 |
Token ID | 17 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
scar | +0.696 |
illon | +0.664 |
ubis | +0.657 |
Wil | +0.656 |
Thor | +0.623 |
Neg logit contributions | |
IPA | -0.840 |
Rican | -0.778 |
prem | -0.757 |
Orig | -0.724 |
Mortgage | -0.722 |
Token | - |
Token ID | 12 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
scar | +0.494 |
illon | +0.473 |
Wil | +0.469 |
ubis | +0.467 |
Thor | +0.448 |
Neg logit contributions | |
IPA | -0.463 |
Rican | -0.425 |
prem | -0.412 |
Orig | -0.393 |
Mortgage | -0.392 |
Token | ethyl |
Token ID | 21610 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
scar | +0.098 |
illon | +0.094 |
Wil | +0.093 |
ubis | +0.093 |
Thor | +0.088 |
Neg logit contributions | |
IPA | -0.107 |
Rican | -0.099 |
prem | -0.096 |
Orig | -0.092 |
Mortgage | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | +6.95e-03 |
Pos logit contributions | |
IPA | +0.081 |
Rican | +0.075 |
prem | +0.073 |
Orig | +0.069 |
Mortgage | +0.069 |
Neg logit contributions | |
scar | -0.080 |
illon | -0.077 |
Wil | -0.076 |
ubis | -0.076 |
Thor | -0.073 |
Token | Because |
Token ID | 4362 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
IPA | +0.005 |
Rican | +0.005 |
prem | +0.004 |
Orig | +0.004 |
Mortgage | +0.004 |
Neg logit contributions | |
scar | -0.006 |
illon | -0.006 |
Wil | -0.006 |
ubis | -0.006 |
Thor | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
IPA | +0.492 |
Rican | +0.452 |
prem | +0.439 |
Orig | +0.420 |
Mortgage | +0.415 |
Neg logit contributions | |
scar | -0.540 |
illon | -0.519 |
Wil | -0.513 |
ubis | -0.511 |
Thor | -0.489 |
Token | the |
Token ID | 262 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
scar | +0.095 |
illon | +0.091 |
Wil | +0.090 |
ubis | +0.090 |
Thor | +0.086 |
Neg logit contributions | |
IPA | -0.092 |
Rican | -0.084 |
prem | -0.082 |
Orig | -0.078 |
Mortgage | -0.078 |
Token | T |
Token ID | 309 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
scar | +0.093 |
illon | +0.090 |
Wil | +0.089 |
ubis | +0.089 |
Thor | +0.085 |
Neg logit contributions | |
IPA | -0.081 |
Rican | -0.074 |
prem | -0.072 |
Orig | -0.068 |
Mortgage | -0.068 |
Token | & |
Token ID | 5 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
scar | +0.091 |
illon | +0.086 |
Wil | +0.086 |
ubis | +0.085 |
Thor | +0.081 |
Neg logit contributions | |
IPA | -0.104 |
Rican | -0.096 |
prem | -0.094 |
Orig | -0.090 |
Mortgage | -0.089 |
Token | A |
Token ID | 32 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
scar | +1.920 |
illon | +1.830 |
Wil | +1.820 |
ubis | +1.802 |
Thor | +1.737 |
Neg logit contributions | |
IPA | -1.943 |
Rican | -1.798 |
prem | -1.746 |
Orig | -1.667 |
Mortgage | -1.663 |
Token | , |
Token ID | 11 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
scar | +0.300 |
illon | +0.288 |
Wil | +0.286 |
ubis | +0.284 |
Thor | +0.274 |
Neg logit contributions | |
IPA | -0.245 |
Rican | -0.224 |
prem | -0.216 |
Orig | -0.205 |
Mortgage | -0.205 |
Token | this |
Token ID | 428 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
IPA | +0.119 |
Rican | +0.109 |
prem | +0.106 |
Orig | +0.101 |
Mortgage | +0.100 |
Neg logit contributions | |
scar | -0.136 |
illon | -0.131 |
Wil | -0.130 |
ubis | -0.129 |
Thor | -0.124 |
Token | title |
Token ID | 3670 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
IPA | +0.287 |
Rican | +0.262 |
prem | +0.253 |
Orig | +0.241 |
Mortgage | +0.238 |
Neg logit contributions | |
scar | -0.356 |
illon | -0.343 |
Wil | -0.340 |
ubis | -0.338 |
Thor | -0.325 |
Token | is |
Token ID | 318 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
IPA | +0.223 |
Rican | +0.205 |
prem | +0.198 |
Orig | +0.189 |
Mortgage | +0.188 |
Neg logit contributions | |
scar | -0.249 |
illon | -0.239 |
Wil | -0.237 |
ubis | -0.236 |
Thor | -0.227 |
Token | is |
Token ID | 318 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
scar | +0.570 |
illon | +0.542 |
Wil | +0.539 |
ubis | +0.535 |
Thor | +0.516 |
Neg logit contributions | |
IPA | -0.497 |
Rican | -0.459 |
prem | -0.438 |
Mortgage | -0.421 |
Orig | -0.420 |
Token | totally |
Token ID | 6635 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
scar | +0.261 |
illon | +0.249 |
Wil | +0.247 |
ubis | +0.245 |
Thor | +0.235 |
Neg logit contributions | |
IPA | -0.266 |
Rican | -0.246 |
prem | -0.238 |
Orig | -0.228 |
Mortgage | -0.228 |
Token | rigged |
Token ID | 31043 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
scar | +0.026 |
illon | +0.025 |
Wil | +0.025 |
ubis | +0.025 |
Thor | +0.023 |
Neg logit contributions | |
IPA | -0.036 |
Rican | -0.034 |
prem | -0.033 |
Orig | -0.032 |
Mortgage | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
scar | +0.866 |
illon | +0.827 |
Wil | +0.821 |
ubis | +0.816 |
Thor | +0.782 |
Neg logit contributions | |
IPA | -0.918 |
Rican | -0.849 |
prem | -0.824 |
Orig | -0.788 |
Mortgage | -0.786 |
Token | Cro |
Token ID | 9325 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
scar | +0.049 |
illon | +0.047 |
Wil | +0.046 |
ubis | +0.046 |
Thor | +0.044 |
Neg logit contributions | |
IPA | -0.049 |
Rican | -0.045 |
prem | -0.044 |
Orig | -0.042 |
Mortgage | -0.042 |
Token | oked |
Token ID | 6545 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
scar | +0.335 |
illon | +0.321 |
Wil | +0.318 |
ubis | +0.316 |
Thor | +0.303 |
Neg logit contributions | |
IPA | -0.304 |
Rican | -0.279 |
prem | -0.270 |
Orig | -0.257 |
Mortgage | -0.257 |
Token | Hillary |
Token ID | 5073 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
scar | +1.937 |
illon | +1.844 |
Wil | +1.835 |
ubis | +1.815 |
Thor | +1.742 |
Neg logit contributions | |
IPA | -1.913 |
Rican | -1.766 |
prem | -1.712 |
Orig | -1.629 |
Mortgage | -1.623 |
Token | Clinton |
Token ID | 2605 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
scar | +0.100 |
illon | +0.096 |
Wil | +0.095 |
ubis | +0.094 |
Thor | +0.091 |
Neg logit contributions | |
IPA | -0.091 |
Rican | -0.084 |
prem | -0.081 |
Orig | -0.077 |
Mortgage | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
IPA | +0.506 |
Rican | +0.462 |
prem | +0.450 |
Orig | +0.426 |
Mortgage | +0.425 |
Neg logit contributions | |
scar | -0.602 |
illon | -0.579 |
Wil | -0.575 |
ubis | -0.573 |
Thor | -0.552 |
Token | Deborah |
Token ID | 36976 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
scar | +0.060 |
illon | +0.058 |
Wil | +0.057 |
ubis | +0.057 |
Thor | +0.055 |
Neg logit contributions | |
IPA | -0.055 |
Rican | -0.051 |
prem | -0.049 |
Mortgage | -0.047 |
Orig | -0.047 |
Token | Wasserman |
Token ID | 38997 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
scar | +0.065 |
illon | +0.062 |
Wil | +0.062 |
ubis | +0.061 |
Thor | +0.059 |
Neg logit contributions | |
IPA | -0.062 |
Rican | -0.057 |
prem | -0.056 |
Orig | -0.053 |
Mortgage | -0.053 |
Token | What |
Token ID | 1867 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
IPA | +0.276 |
Rican | +0.251 |
prem | +0.242 |
Orig | +0.234 |
Mortgage | +0.232 |
Neg logit contributions | |
scar | -0.312 |
illon | -0.300 |
ubis | -0.297 |
Wil | -0.295 |
Thor | -0.282 |
Token | we |
Token ID | 356 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
IPA | +0.040 |
Rican | +0.036 |
prem | +0.035 |
Orig | +0.034 |
Mortgage | +0.033 |
Neg logit contributions | |
scar | -0.046 |
illon | -0.045 |
Wil | -0.044 |
ubis | -0.044 |
Thor | -0.042 |
Token | 're |
Token ID | 821 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
IPA | +0.383 |
Rican | +0.349 |
prem | +0.343 |
Orig | +0.323 |
Mortgage | +0.322 |
Neg logit contributions | |
scar | -0.418 |
Wil | -0.402 |
illon | -0.401 |
ubis | -0.399 |
Thor | -0.384 |
Token | learning |
Token ID | 4673 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
scar | +0.089 |
illon | +0.085 |
Wil | +0.085 |
ubis | +0.084 |
Thor | +0.081 |
Neg logit contributions | |
IPA | -0.098 |
Rican | -0.090 |
prem | -0.088 |
Orig | -0.084 |
Mortgage | -0.084 |
Token | so |
Token ID | 523 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
scar | +0.934 |
illon | +0.892 |
Wil | +0.885 |
ubis | +0.879 |
Thor | +0.846 |
Neg logit contributions | |
IPA | -0.880 |
Rican | -0.811 |
prem | -0.782 |
Orig | -0.748 |
Mortgage | -0.747 |
Token | far |
Token ID | 1290 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
scar | +0.631 |
illon | +0.596 |
Wil | +0.590 |
ubis | +0.588 |
Thor | +0.550 |
Neg logit contributions | |
IPA | -1.073 |
Rican | -1.006 |
prem | -0.987 |
Orig | -0.949 |
Mortgage | -0.943 |
Token | isn |
Token ID | 2125 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
scar | +2.048 |
illon | +1.953 |
Wil | +1.935 |
ubis | +1.920 |
Thor | +1.862 |
Neg logit contributions | |
IPA | -1.763 |
Rican | -1.624 |
prem | -1.548 |
Mortgage | -1.487 |
Orig | -1.484 |
Token | 't |
Token ID | 470 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
scar | +0.127 |
illon | +0.121 |
Wil | +0.119 |
ubis | +0.119 |
Thor | +0.113 |
Neg logit contributions | |
IPA | -0.178 |
Rican | -0.165 |
prem | -0.161 |
Orig | -0.156 |
Mortgage | -0.155 |
Token | terribly |
Token ID | 22121 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
scar | +0.054 |
illon | +0.052 |
Wil | +0.052 |
ubis | +0.051 |
Thor | +0.049 |
Neg logit contributions | |
IPA | -0.055 |
Rican | -0.050 |
prem | -0.049 |
Orig | -0.047 |
Mortgage | -0.047 |
Token | illuminating |
Token ID | 46717 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
scar | +0.766 |
illon | +0.733 |
Wil | +0.729 |
ubis | +0.724 |
Thor | +0.699 |
Neg logit contributions | |
IPA | -0.619 |
Rican | -0.566 |
prem | -0.544 |
Orig | -0.519 |
Mortgage | -0.518 |
Token | , |
Token ID | 11 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
scar | +0.734 |
illon | +0.701 |
Wil | +0.695 |
ubis | +0.692 |
Thor | +0.663 |
Neg logit contributions | |
IPA | -0.760 |
Rican | -0.702 |
prem | -0.680 |
Orig | -0.651 |
Mortgage | -0.649 |
Token | Row |
Token ID | 25166 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
IPA | +0.056 |
Rican | +0.051 |
prem | +0.050 |
Orig | +0.048 |
Mortgage | +0.048 |
Neg logit contributions | |
scar | -0.050 |
illon | -0.049 |
ubis | -0.048 |
Wil | -0.047 |
Thor | -0.045 |
Token | lf |
Token ID | 1652 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
scar | +0.301 |
illon | +0.290 |
ubis | +0.287 |
Wil | +0.287 |
Thor | +0.273 |
Neg logit contributions | |
IPA | -0.348 |
Rican | -0.321 |
prem | -0.313 |
Orig | -0.299 |
Mortgage | -0.298 |
Token | achieved |
Token ID | 8793 |
Feature activation | -8.79e-02 |
Pos logit contributions | |
IPA | +0.460 |
Rican | +0.425 |
prem | +0.412 |
Orig | +0.398 |
Mortgage | +0.393 |
Neg logit contributions | |
scar | -0.448 |
ubis | -0.425 |
illon | -0.425 |
Wil | -0.423 |
Thor | -0.402 |
Token | movie |
Token ID | 3807 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
scar | +0.071 |
illon | +0.068 |
Wil | +0.067 |
ubis | +0.067 |
Thor | +0.064 |
Neg logit contributions | |
IPA | -0.067 |
Rican | -0.061 |
prem | -0.059 |
Orig | -0.056 |
Mortgage | -0.056 |
Token | st |
Token ID | 336 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
IPA | +0.069 |
Rican | +0.064 |
prem | +0.062 |
Orig | +0.059 |
Mortgage | +0.059 |
Neg logit contributions | |
scar | -0.072 |
illon | -0.069 |
Wil | -0.068 |
ubis | -0.068 |
Thor | -0.065 |
Token | ard |
Token ID | 446 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
scar | +0.223 |
illon | +0.206 |
Wil | +0.203 |
ubis | +0.201 |
Thor | +0.186 |
Neg logit contributions | |
IPA | -0.557 |
Rican | -0.526 |
prem | -0.515 |
Orig | -0.500 |
Mortgage | -0.499 |
Token | om |
Token ID | 296 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
scar | +1.557 |
illon | +1.480 |
Wil | +1.452 |
ubis | +1.443 |
Thor | +1.371 |
Neg logit contributions | |
IPA | -2.222 |
Rican | -2.095 |
prem | -2.008 |
Mortgage | -1.947 |
Orig | -1.941 |
Token | with |
Token ID | 351 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
scar | +0.803 |
illon | +0.764 |
Wil | +0.759 |
ubis | +0.753 |
Thor | +0.723 |
Neg logit contributions | |
IPA | -0.819 |
Rican | -0.758 |
prem | -0.731 |
Orig | -0.701 |
Mortgage | -0.701 |
Token | his |
Token ID | 465 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
IPA | +0.009 |
Rican | +0.009 |
prem | +0.008 |
Orig | +0.008 |
Mortgage | +0.008 |
Neg logit contributions | |
scar | -0.011 |
illon | -0.011 |
Wil | -0.011 |
ubis | -0.011 |
Thor | -0.010 |
Token | appearance |
Token ID | 5585 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
scar | +0.555 |
illon | +0.535 |
Wil | +0.531 |
ubis | +0.528 |
Thor | +0.510 |
Neg logit contributions | |
IPA | -0.404 |
Rican | -0.367 |
prem | -0.354 |
Orig | -0.335 |
Mortgage | -0.333 |
Token | in |
Token ID | 287 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
scar | +0.613 |
illon | +0.588 |
Wil | +0.583 |
ubis | +0.581 |
Thor | +0.556 |
Neg logit contributions | |
IPA | -0.586 |
Rican | -0.538 |
prem | -0.523 |
Orig | -0.499 |
Mortgage | -0.497 |
Token | , |
Token ID | 11 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
IPA | +0.225 |
Rican | +0.207 |
prem | +0.202 |
Orig | +0.193 |
Mortgage | +0.191 |
Neg logit contributions | |
scar | -0.226 |
illon | -0.216 |
Wil | -0.214 |
ubis | -0.214 |
Thor | -0.204 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
IPA | +0.030 |
Rican | +0.027 |
prem | +0.027 |
Orig | +0.025 |
Mortgage | +0.025 |
Neg logit contributions | |
scar | -0.034 |
illon | -0.033 |
ubis | -0.033 |
Wil | -0.032 |
Thor | -0.031 |
Token | Bulls |
Token ID | 18075 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
scar | +0.219 |
illon | +0.211 |
Wil | +0.210 |
ubis | +0.209 |
Thor | +0.202 |
Neg logit contributions | |
IPA | -0.148 |
Rican | -0.134 |
prem | -0.129 |
Orig | -0.122 |
Mortgage | -0.121 |
Token | would |
Token ID | 561 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
IPA | +0.065 |
Rican | +0.060 |
prem | +0.059 |
Orig | +0.056 |
Mortgage | +0.056 |
Neg logit contributions | |
scar | -0.064 |
illon | -0.061 |
Wil | -0.061 |
ubis | -0.061 |
Thor | -0.058 |
Token | have |
Token ID | 423 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
IPA | +0.029 |
Rican | +0.027 |
prem | +0.027 |
Orig | +0.025 |
Mortgage | +0.025 |
Neg logit contributions | |
scar | -0.028 |
Wil | -0.027 |
illon | -0.027 |
ubis | -0.026 |
Thor | -0.025 |
Token | preferred |
Token ID | 9871 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
scar | +0.294 |
illon | +0.281 |
Wil | +0.279 |
ubis | +0.277 |
Thor | +0.266 |
Neg logit contributions | |
IPA | -0.285 |
Rican | -0.262 |
prem | -0.254 |
Orig | -0.243 |
Mortgage | -0.242 |
Token | as |
Token ID | 355 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
scar | +1.951 |
illon | +1.874 |
Wil | +1.862 |
ubis | +1.834 |
Thor | +1.772 |
Neg logit contributions | |
IPA | -1.806 |
Rican | -1.656 |
prem | -1.596 |
Mortgage | -1.537 |
Orig | -1.523 |
Token | much |
Token ID | 881 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
scar | +0.384 |
illon | +0.366 |
Wil | +0.364 |
ubis | +0.362 |
Thor | +0.345 |
Neg logit contributions | |
IPA | -0.419 |
Rican | -0.388 |
prem | -0.377 |
Orig | -0.362 |
Mortgage | -0.358 |
Token | time |
Token ID | 640 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
scar | +0.134 |
illon | +0.128 |
Wil | +0.127 |
ubis | +0.127 |
Thor | +0.122 |
Neg logit contributions | |
IPA | -0.130 |
Rican | -0.120 |
prem | -0.117 |
Orig | -0.111 |
Mortgage | -0.111 |
Token | as |
Token ID | 355 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
IPA | +0.269 |
Rican | +0.248 |
prem | +0.240 |
Orig | +0.230 |
Mortgage | +0.229 |
Neg logit contributions | |
scar | -0.277 |
illon | -0.265 |
Wil | -0.263 |
ubis | -0.261 |
Thor | -0.251 |
Token | possible |
Token ID | 1744 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
IPA | +0.369 |
Rican | +0.340 |
prem | +0.330 |
Orig | +0.317 |
Mortgage | +0.314 |
Neg logit contributions | |
scar | -0.366 |
illon | -0.349 |
Wil | -0.346 |
ubis | -0.346 |
Thor | -0.330 |
Token | regional |
Token ID | 7915 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
IPA | +0.182 |
Rican | +0.168 |
prem | +0.162 |
Orig | +0.155 |
Mortgage | +0.154 |
Neg logit contributions | |
scar | -0.202 |
illon | -0.194 |
Wil | -0.192 |
ubis | -0.191 |
Thor | -0.184 |
Token | interests |
Token ID | 5353 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
IPA | +0.049 |
Rican | +0.045 |
prem | +0.044 |
Orig | +0.041 |
Mortgage | +0.041 |
Neg logit contributions | |
scar | -0.062 |
illon | -0.059 |
Wil | -0.059 |
ubis | -0.059 |
Thor | -0.056 |
Token | should |
Token ID | 815 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
IPA | +0.086 |
Rican | +0.078 |
prem | +0.076 |
Orig | +0.072 |
Mortgage | +0.072 |
Neg logit contributions | |
scar | -0.098 |
illon | -0.094 |
Wil | -0.093 |
ubis | -0.093 |
Thor | -0.089 |
Token | be |
Token ID | 307 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
IPA | +0.297 |
prem | +0.269 |
Rican | +0.269 |
Orig | +0.252 |
Mortgage | +0.250 |
Neg logit contributions | |
scar | -0.293 |
Wil | -0.281 |
illon | -0.280 |
ubis | -0.278 |
Thor | -0.267 |
Token | adequately |
Token ID | 22668 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
IPA | +0.203 |
Rican | +0.187 |
prem | +0.182 |
Orig | +0.173 |
Mortgage | +0.172 |
Neg logit contributions | |
scar | -0.210 |
illon | -0.202 |
Wil | -0.200 |
ubis | -0.199 |
Thor | -0.191 |
Token | represented |
Token ID | 7997 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
scar | +0.504 |
illon | +0.477 |
Wil | +0.473 |
ubis | +0.469 |
Thor | +0.448 |
Neg logit contributions | |
IPA | -0.661 |
Rican | -0.617 |
prem | -0.598 |
Orig | -0.579 |
Mortgage | -0.576 |
Token | , |
Token ID | 11 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
scar | +1.840 |
illon | +1.750 |
Wil | +1.737 |
ubis | +1.722 |
Thor | +1.652 |
Neg logit contributions | |
IPA | -1.897 |
Rican | -1.761 |
prem | -1.687 |
Orig | -1.631 |
Mortgage | -1.629 |
Token | gender |
Token ID | 5279 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
scar | +0.324 |
illon | +0.310 |
Wil | +0.307 |
ubis | +0.306 |
Thor | +0.294 |
Neg logit contributions | |
IPA | -0.319 |
Rican | -0.294 |
prem | -0.285 |
Orig | -0.272 |
Mortgage | -0.271 |
Token | should |
Token ID | 815 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
IPA | +0.212 |
Rican | +0.195 |
prem | +0.189 |
Orig | +0.180 |
Mortgage | +0.179 |
Neg logit contributions | |
scar | -0.232 |
illon | -0.223 |
Wil | -0.221 |
ubis | -0.220 |
Thor | -0.211 |
Token | be |
Token ID | 307 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
IPA | +0.155 |
Rican | +0.141 |
prem | +0.138 |
Orig | +0.130 |
Mortgage | +0.130 |
Neg logit contributions | |
scar | -0.170 |
illon | -0.163 |
Wil | -0.163 |
ubis | -0.161 |
Thor | -0.155 |
Token | adequately |
Token ID | 22668 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
IPA | +0.214 |
Rican | +0.196 |
prem | +0.190 |
Orig | +0.181 |
Mortgage | +0.180 |
Neg logit contributions | |
scar | -0.233 |
illon | -0.223 |
Wil | -0.221 |
ubis | -0.220 |
Thor | -0.212 |
Token | wind |
Token ID | 2344 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
scar | +0.483 |
illon | +0.464 |
Wil | +0.460 |
ubis | +0.459 |
Thor | +0.442 |
Neg logit contributions | |
IPA | -0.386 |
Rican | -0.352 |
prem | -0.340 |
Orig | -0.323 |
Mortgage | -0.322 |
Token | speeds |
Token ID | 12055 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
IPA | +0.438 |
Rican | +0.401 |
prem | +0.389 |
Orig | +0.372 |
Mortgage | +0.368 |
Neg logit contributions | |
scar | -0.470 |
illon | -0.453 |
Wil | -0.448 |
ubis | -0.448 |
Thor | -0.430 |
Token | generated |
Token ID | 7560 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
scar | +0.087 |
illon | +0.083 |
Wil | +0.082 |
ubis | +0.082 |
Thor | +0.079 |
Neg logit contributions | |
IPA | -0.084 |
Rican | -0.078 |
prem | -0.075 |
Orig | -0.072 |
Mortgage | -0.072 |
Token | by |
Token ID | 416 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
scar | +0.651 |
illon | +0.624 |
Wil | +0.619 |
ubis | +0.616 |
Thor | +0.593 |
Neg logit contributions | |
IPA | -0.559 |
Rican | -0.512 |
prem | -0.494 |
Orig | -0.471 |
Mortgage | -0.469 |
Token | Hurricane |
Token ID | 16004 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
scar | +0.429 |
illon | +0.414 |
Wil | +0.412 |
ubis | +0.407 |
Thor | +0.395 |
Neg logit contributions | |
IPA | -0.319 |
Rican | -0.290 |
prem | -0.278 |
Orig | -0.264 |
Mortgage | -0.263 |
Token | Matthew |
Token ID | 9308 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
IPA | +0.454 |
Rican | +0.415 |
prem | +0.400 |
Orig | +0.381 |
Mortgage | +0.380 |
Neg logit contributions | |
scar | -0.550 |
illon | -0.528 |
Wil | -0.523 |
ubis | -0.521 |
Thor | -0.502 |
Token | . |
Token ID | 13 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
IPA | +1.411 |
Rican | +1.297 |
prem | +1.259 |
Orig | +1.208 |
Mortgage | +1.202 |
Neg logit contributions | |
scar | -1.506 |
illon | -1.438 |
ubis | -1.433 |
Wil | -1.420 |
Thor | -1.371 |
Token | Credit |
Token ID | 10504 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
scar | +0.215 |
illon | +0.206 |
Wil | +0.204 |
ubis | +0.203 |
Thor | +0.195 |
Neg logit contributions | |
IPA | -0.203 |
Rican | -0.187 |
prem | -0.181 |
Orig | -0.172 |
Mortgage | -0.172 |
Token | : |
Token ID | 25 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
IPA | +0.050 |
Rican | +0.046 |
prem | +0.044 |
Orig | +0.042 |
Mortgage | +0.042 |
Neg logit contributions | |
scar | -0.057 |
illon | -0.055 |
Wil | -0.054 |
ubis | -0.054 |
Thor | -0.052 |
Token | NO |
Token ID | 15285 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
scar | +0.286 |
illon | +0.274 |
Wil | +0.273 |
ubis | +0.271 |
Thor | +0.262 |
Neg logit contributions | |
IPA | -0.229 |
Rican | -0.210 |
prem | -0.203 |
Orig | -0.192 |
Mortgage | -0.191 |
Token | AA |
Token ID | 3838 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
scar | +0.124 |
illon | +0.119 |
Wil | +0.118 |
ubis | +0.117 |
Thor | +0.113 |
Neg logit contributions | |
IPA | -0.116 |
Rican | -0.107 |
prem | -0.103 |
Orig | -0.099 |
Mortgage | -0.099 |
Token | ya |
Token ID | 3972 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
scar | +0.285 |
illon | +0.272 |
Wil | +0.270 |
ubis | +0.269 |
Thor | +0.258 |
Neg logit contributions | |
IPA | -0.292 |
Rican | -0.270 |
prem | -0.262 |
Orig | -0.250 |
Mortgage | -0.249 |
Token | Fisher |
Token ID | 14388 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
scar | +1.375 |
illon | +1.317 |
Wil | +1.306 |
ubis | +1.299 |
Thor | +1.249 |
Neg logit contributions | |
IPA | -1.262 |
Rican | -1.161 |
prem | -1.122 |
Orig | -1.071 |
Mortgage | -1.068 |
Token | , |
Token ID | 11 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
scar | +0.048 |
illon | +0.046 |
Wil | +0.046 |
ubis | +0.045 |
Thor | +0.043 |
Neg logit contributions | |
IPA | -0.046 |
Rican | -0.042 |
prem | -0.041 |
Orig | -0.039 |
Mortgage | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
IPA | +0.090 |
Rican | +0.082 |
prem | +0.080 |
Orig | +0.076 |
Mortgage | +0.076 |
Neg logit contributions | |
scar | -0.098 |
illon | -0.094 |
Wil | -0.093 |
ubis | -0.093 |
Thor | -0.089 |
Token | doctoral |
Token ID | 40995 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
scar | +0.239 |
illon | +0.230 |
Wil | +0.229 |
ubis | +0.227 |
Thor | +0.220 |
Neg logit contributions | |
IPA | -0.166 |
Rican | -0.150 |
prem | -0.145 |
Orig | -0.137 |
Mortgage | -0.136 |
Token | candidate |
Token ID | 4540 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
IPA | +0.408 |
Rican | +0.369 |
prem | +0.354 |
Orig | +0.335 |
Mortgage | +0.332 |
Neg logit contributions | |
scar | -0.598 |
illon | -0.577 |
Wil | -0.572 |
ubis | -0.571 |
Thor | -0.550 |
Token | in |
Token ID | 287 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
IPA | +1.491 |
Rican | +1.367 |
prem | +1.329 |
Orig | +1.257 |
Mortgage | +1.246 |
Neg logit contributions | |
scar | -1.616 |
illon | -1.563 |
ubis | -1.551 |
Wil | -1.547 |
Thor | -1.483 |
Token | lingu |
Token ID | 20280 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
IPA | +0.122 |
Rican | +0.110 |
prem | +0.106 |
Orig | +0.100 |
Mortgage | +0.100 |
Neg logit contributions | |
scar | -0.181 |
illon | -0.174 |
Wil | -0.173 |
ubis | -0.172 |
Thor | -0.166 |
Token | istics |
Token ID | 3969 |
Feature activation | -9.04e-02 |
Pos logit contributions | |
IPA | +0.043 |
Rican | +0.040 |
prem | +0.039 |
Orig | +0.037 |
Mortgage | +0.037 |
Neg logit contributions | |
scar | -0.043 |
illon | -0.041 |
Wil | -0.041 |
ubis | -0.041 |
Thor | -0.039 |
Token | at |
Token ID | 379 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
scar | +0.074 |
illon | +0.071 |
Wil | +0.070 |
ubis | +0.070 |
Thor | +0.067 |
Neg logit contributions | |
IPA | -0.068 |
Rican | -0.062 |
prem | -0.060 |
Orig | -0.057 |
Mortgage | -0.057 |
Token | Penn |
Token ID | 6595 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
IPA | +0.231 |
Rican | +0.210 |
prem | +0.203 |
Orig | +0.193 |
Mortgage | +0.192 |
Neg logit contributions | |
scar | -0.285 |
illon | -0.274 |
Wil | -0.271 |
ubis | -0.270 |
Thor | -0.260 |
Token | blocks |
Token ID | 7021 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
scar | +0.379 |
illon | +0.364 |
Wil | +0.361 |
ubis | +0.360 |
Thor | +0.347 |
Neg logit contributions | |
IPA | -0.293 |
Rican | -0.267 |
prem | -0.257 |
Orig | -0.244 |
Mortgage | -0.244 |
Token | and |
Token ID | 290 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
scar | +0.796 |
illon | +0.761 |
Wil | +0.756 |
ubis | +0.751 |
Thor | +0.722 |
Neg logit contributions | |
IPA | -0.759 |
Rican | -0.697 |
prem | -0.678 |
Orig | -0.647 |
Mortgage | -0.643 |
Token | balls |
Token ID | 11333 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
IPA | +0.196 |
Rican | +0.179 |
prem | +0.173 |
Orig | +0.165 |
Mortgage | +0.163 |
Neg logit contributions | |
scar | -0.238 |
illon | -0.228 |
Wil | -0.227 |
ubis | -0.225 |
Thor | -0.217 |
Token | can |
Token ID | 460 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
scar | +0.559 |
illon | +0.536 |
Wil | +0.533 |
ubis | +0.530 |
Thor | +0.510 |
Neg logit contributions | |
IPA | -0.475 |
Rican | -0.435 |
prem | -0.421 |
Orig | -0.401 |
Mortgage | -0.398 |
Token | be |
Token ID | 307 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
scar | +0.101 |
illon | +0.097 |
Wil | +0.096 |
ubis | +0.096 |
Thor | +0.092 |
Neg logit contributions | |
IPA | -0.095 |
Rican | -0.087 |
prem | -0.085 |
Orig | -0.081 |
Mortgage | -0.080 |
Token | played |
Token ID | 2826 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
IPA | +0.104 |
Rican | +0.095 |
prem | +0.093 |
Orig | +0.088 |
Mortgage | +0.087 |
Neg logit contributions | |
scar | -0.126 |
illon | -0.121 |
Wil | -0.120 |
ubis | -0.120 |
Thor | -0.116 |
Token | with |
Token ID | 351 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
IPA | +1.363 |
Rican | +1.248 |
prem | +1.234 |
Orig | +1.168 |
Mortgage | +1.148 |
Neg logit contributions | |
scar | -1.409 |
illon | -1.363 |
Wil | -1.357 |
ubis | -1.344 |
Thor | -1.301 |
Token | in |
Token ID | 287 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
IPA | +0.151 |
Rican | +0.139 |
prem | +0.135 |
Orig | +0.129 |
Mortgage | +0.127 |
Neg logit contributions | |
scar | -0.166 |
illon | -0.159 |
Wil | -0.158 |
ubis | -0.157 |
Thor | -0.152 |
Token | many |
Token ID | 867 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
IPA | +0.134 |
Rican | +0.123 |
prem | +0.119 |
Orig | +0.114 |
Mortgage | +0.113 |
Neg logit contributions | |
scar | -0.154 |
illon | -0.147 |
Wil | -0.146 |
ubis | -0.146 |
Thor | -0.140 |
Token | different |
Token ID | 1180 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
scar | +0.340 |
illon | +0.327 |
Wil | +0.324 |
ubis | +0.323 |
Thor | +0.311 |
Neg logit contributions | |
IPA | -0.265 |
Rican | -0.241 |
prem | -0.232 |
Orig | -0.221 |
Mortgage | -0.220 |
Token | ways |
Token ID | 2842 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
IPA | +0.683 |
Rican | +0.627 |
prem | +0.610 |
Orig | +0.580 |
Mortgage | +0.575 |
Neg logit contributions | |
scar | -0.758 |
illon | -0.728 |
Wil | -0.721 |
ubis | -0.717 |
Thor | -0.690 |
Token | a |
Token ID | 257 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
IPA | +0.240 |
Rican | +0.221 |
prem | +0.217 |
Orig | +0.206 |
Mortgage | +0.205 |
Neg logit contributions | |
scar | -0.239 |
illon | -0.231 |
Wil | -0.228 |
ubis | -0.227 |
Thor | -0.219 |
Token | World |
Token ID | 2159 |
Feature activation | -3.86e-03 |
Pos logit contributions | |
IPA | +0.202 |
Rican | +0.184 |
prem | +0.180 |
Orig | +0.170 |
Mortgage | +0.169 |
Neg logit contributions | |
scar | -0.241 |
illon | -0.232 |
Wil | -0.230 |
ubis | -0.230 |
Thor | -0.221 |
Token | War |
Token ID | 1810 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
scar | +0.003 |
illon | +0.003 |
Wil | +0.003 |
ubis | +0.003 |
Thor | +0.003 |
Neg logit contributions | |
IPA | -0.003 |
Rican | -0.003 |
prem | -0.003 |
Orig | -0.003 |
Mortgage | -0.003 |
Token | Two |
Token ID | 4930 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
IPA | +0.373 |
Rican | +0.330 |
prem | +0.318 |
Orig | +0.300 |
Mortgage | +0.297 |
Neg logit contributions | |
scar | -0.609 |
Wil | -0.586 |
illon | -0.586 |
ubis | -0.585 |
Thor | -0.565 |
Token | ( |
Token ID | 357 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
IPA | +0.216 |
Rican | +0.196 |
prem | +0.190 |
Orig | +0.181 |
Mortgage | +0.179 |
Neg logit contributions | |
scar | -0.269 |
illon | -0.259 |
Wil | -0.257 |
ubis | -0.256 |
Thor | -0.247 |
Token | circ |
Token ID | 21170 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
scar | +0.026 |
illon | +0.025 |
ubis | +0.024 |
Wil | +0.024 |
Thor | +0.023 |
Neg logit contributions | |
IPA | -0.034 |
Rican | -0.031 |
prem | -0.031 |
Orig | -0.030 |
Mortgage | -0.029 |
Token | a |
Token ID | 64 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
IPA | +1.580 |
Rican | +1.464 |
prem | +1.434 |
Orig | +1.369 |
Mortgage | +1.360 |
Neg logit contributions | |
scar | -1.342 |
Wil | -1.268 |
ubis | -1.267 |
illon | -1.264 |
Thor | -1.209 |
Token | 1941 |
Token ID | 23234 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
scar | +0.096 |
illon | +0.093 |
Wil | +0.092 |
ubis | +0.092 |
Thor | +0.089 |
Neg logit contributions | |
IPA | -0.065 |
Rican | -0.058 |
prem | -0.056 |
Orig | -0.053 |
Mortgage | -0.053 |
Token | ) |
Token ID | 8 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
scar | +0.161 |
illon | +0.155 |
Wil | +0.153 |
ubis | +0.153 |
Thor | +0.147 |
Neg logit contributions | |
IPA | -0.146 |
Rican | -0.134 |
prem | -0.130 |
Orig | -0.124 |
Mortgage | -0.123 |
Token | improvement |
Token ID | 9025 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
IPA | +0.086 |
Rican | +0.079 |
prem | +0.076 |
Orig | +0.073 |
Mortgage | +0.072 |
Neg logit contributions | |
scar | -0.107 |
illon | -0.103 |
Wil | -0.103 |
ubis | -0.102 |
Thor | -0.098 |
Token | on |
Token ID | 319 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
scar | +0.158 |
illon | +0.152 |
ubis | +0.151 |
Wil | +0.151 |
Thor | +0.145 |
Neg logit contributions | |
IPA | -0.154 |
Rican | -0.141 |
prem | -0.139 |
Orig | -0.132 |
Mortgage | -0.131 |
Token | over |
Token ID | 625 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
scar | +0.704 |
illon | +0.670 |
Wil | +0.666 |
ubis | +0.661 |
Thor | +0.635 |
Neg logit contributions | |
IPA | -0.685 |
Rican | -0.632 |
prem | -0.607 |
Orig | -0.583 |
Mortgage | -0.583 |
Token | that |
Token ID | 326 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
IPA | +0.046 |
Rican | +0.042 |
prem | +0.041 |
Orig | +0.039 |
Mortgage | +0.039 |
Neg logit contributions | |
scar | -0.046 |
illon | -0.044 |
Wil | -0.044 |
ubis | -0.044 |
Thor | -0.042 |
Token | debt |
Token ID | 5057 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
IPA | +0.096 |
Rican | +0.088 |
prem | +0.085 |
Orig | +0.081 |
Mortgage | +0.081 |
Neg logit contributions | |
scar | -0.109 |
illon | -0.105 |
Wil | -0.104 |
ubis | -0.104 |
Thor | -0.100 |
Token | / |
Token ID | 14 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
IPA | +0.355 |
Rican | +0.324 |
prem | +0.313 |
Orig | +0.298 |
Mortgage | +0.296 |
Neg logit contributions | |
scar | -0.435 |
illon | -0.418 |
Wil | -0.415 |
ubis | -0.414 |
Thor | -0.398 |
Token | G |
Token ID | 38 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
scar | +0.515 |
illon | +0.491 |
Wil | +0.487 |
ubis | +0.484 |
Thor | +0.464 |
Neg logit contributions | |
IPA | -0.538 |
Rican | -0.497 |
prem | -0.482 |
Orig | -0.461 |
Mortgage | -0.460 |
Token | DP |
Token ID | 6322 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
IPA | +0.107 |
Rican | +0.099 |
prem | +0.097 |
Mortgage | +0.093 |
Orig | +0.093 |
Neg logit contributions | |
scar | -0.083 |
illon | -0.079 |
Wil | -0.079 |
ubis | -0.078 |
Thor | -0.075 |
Token | number |
Token ID | 1271 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
IPA | +1.293 |
Rican | +1.175 |
prem | +1.141 |
Orig | +1.080 |
Mortgage | +1.078 |
Neg logit contributions | |
scar | -1.601 |
illon | -1.534 |
Wil | -1.525 |
ubis | -1.522 |
Thor | -1.463 |
Token | but |
Token ID | 475 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
IPA | +0.173 |
Rican | +0.160 |
prem | +0.155 |
Orig | +0.148 |
Mortgage | +0.148 |
Neg logit contributions | |
scar | -0.167 |
illon | -0.160 |
Wil | -0.159 |
ubis | -0.158 |
Thor | -0.151 |
Token | that |
Token ID | 326 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
IPA | +0.180 |
Rican | +0.165 |
prem | +0.161 |
Orig | +0.154 |
Mortgage | +0.153 |
Neg logit contributions | |
scar | -0.179 |
illon | -0.170 |
Wil | -0.169 |
ubis | -0.169 |
Thor | -0.162 |
Token | � |
Token ID | 447 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
scar | +0.217 |
illon | +0.208 |
Wil | +0.206 |
ubis | +0.205 |
Thor | +0.197 |
Neg logit contributions | |
IPA | -0.195 |
Rican | -0.179 |
prem | -0.173 |
Orig | -0.165 |
Mortgage | -0.164 |
Token | � |
Token ID | 247 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
IPA | +0.551 |
Rican | +0.517 |
prem | +0.505 |
Orig | +0.486 |
Mortgage | +0.486 |
Neg logit contributions | |
scar | -0.327 |
illon | -0.307 |
Wil | -0.302 |
ubis | -0.300 |
Thor | -0.282 |
Token | un |
Token ID | 403 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
scar | +0.188 |
illon | +0.175 |
Wil | +0.173 |
ubis | +0.172 |
Thor | +0.160 |
Neg logit contributions | |
IPA | -0.374 |
Rican | -0.353 |
prem | -0.344 |
Orig | -0.334 |
Mortgage | -0.333 |
Token | Kitchen |
Token ID | 22541 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
scar | +0.016 |
illon | +0.016 |
Wil | +0.015 |
ubis | +0.015 |
Thor | +0.015 |
Neg logit contributions | |
IPA | -0.013 |
Rican | -0.012 |
prem | -0.012 |
Orig | -0.011 |
Mortgage | -0.011 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
scar | +0.510 |
illon | +0.485 |
Wil | +0.480 |
ubis | +0.477 |
Thor | +0.455 |
Neg logit contributions | |
IPA | -0.647 |
Rican | -0.603 |
prem | -0.585 |
Orig | -0.563 |
Mortgage | -0.562 |
Token | Recently |
Token ID | 24661 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
scar | +0.201 |
illon | +0.192 |
ubis | +0.189 |
Wil | +0.189 |
Thor | +0.179 |
Neg logit contributions | |
IPA | -0.230 |
Rican | -0.212 |
prem | -0.207 |
Orig | -0.198 |
Mortgage | -0.198 |
Token | , |
Token ID | 11 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
scar | +0.361 |
illon | +0.345 |
Wil | +0.343 |
ubis | +0.341 |
Thor | +0.327 |
Neg logit contributions | |
IPA | -0.344 |
Rican | -0.316 |
prem | -0.306 |
Orig | -0.293 |
Mortgage | -0.292 |
Token | there |
Token ID | 612 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
IPA | +0.500 |
Rican | +0.457 |
prem | +0.442 |
Orig | +0.423 |
Mortgage | +0.420 |
Neg logit contributions | |
scar | -0.579 |
illon | -0.555 |
Wil | -0.550 |
ubis | -0.549 |
Thor | -0.527 |
Token | � |
Token ID | 447 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
IPA | +0.213 |
Rican | +0.196 |
prem | +0.190 |
Orig | +0.182 |
Mortgage | +0.181 |
Neg logit contributions | |
scar | -0.215 |
illon | -0.205 |
Wil | -0.203 |
ubis | -0.202 |
Thor | -0.194 |
Token | � |
Token ID | 247 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
scar | +0.197 |
illon | +0.186 |
Wil | +0.183 |
ubis | +0.182 |
Thor | +0.173 |
Neg logit contributions | |
IPA | -0.314 |
Rican | -0.294 |
prem | -0.287 |
Orig | -0.277 |
Mortgage | -0.276 |
Token | s |
Token ID | 82 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
IPA | +0.092 |
Rican | +0.085 |
prem | +0.083 |
Orig | +0.080 |
Mortgage | +0.079 |
Neg logit contributions | |
scar | -0.085 |
illon | -0.081 |
Wil | -0.080 |
ubis | -0.080 |
Thor | -0.076 |
Token | been |
Token ID | 587 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
scar | +0.398 |
illon | +0.382 |
Wil | +0.379 |
ubis | +0.377 |
Thor | +0.363 |
Neg logit contributions | |
IPA | -0.344 |
Rican | -0.316 |
prem | -0.305 |
Orig | -0.290 |
Mortgage | -0.290 |
Token | some |
Token ID | 617 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
IPA | +0.185 |
Rican | +0.169 |
prem | +0.164 |
Orig | +0.156 |
Mortgage | +0.155 |
Neg logit contributions | |
scar | -0.227 |
illon | -0.218 |
Wil | -0.217 |
ubis | -0.216 |
Thor | -0.208 |
Token | almost |
Token ID | 2048 |
Feature activation | +8.25e-01 |
Pos logit contributions | |
IPA | +0.206 |
Rican | +0.190 |
prem | +0.186 |
Orig | +0.177 |
Mortgage | +0.176 |
Neg logit contributions | |
scar | -0.180 |
illon | -0.173 |
Wil | -0.171 |
ubis | -0.171 |
Thor | -0.163 |
Token | always |
Token ID | 1464 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
IPA | +0.637 |
Rican | +0.587 |
prem | +0.572 |
Orig | +0.544 |
Mortgage | +0.540 |
Neg logit contributions | |
scar | -0.650 |
illon | -0.629 |
Wil | -0.620 |
ubis | -0.617 |
Thor | -0.593 |
Token | wide |
Token ID | 3094 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
IPA | +0.961 |
Rican | +0.884 |
prem | +0.867 |
Orig | +0.824 |
Mortgage | +0.811 |
Neg logit contributions | |
scar | -0.930 |
illon | -0.900 |
Wil | -0.891 |
ubis | -0.886 |
Thor | -0.849 |
Token | open |
Token ID | 1280 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
IPA | +0.411 |
Rican | +0.372 |
prem | +0.358 |
Orig | +0.338 |
Mortgage | +0.334 |
Neg logit contributions | |
scar | -0.594 |
illon | -0.572 |
Wil | -0.571 |
ubis | -0.567 |
Thor | -0.549 |
Token | when |
Token ID | 618 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
IPA | +0.017 |
Rican | +0.015 |
prem | +0.015 |
Orig | +0.014 |
Mortgage | +0.014 |
Neg logit contributions | |
scar | -0.017 |
illon | -0.016 |
Wil | -0.016 |
ubis | -0.016 |
Thor | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +7.01e-01 |
Pos logit contributions | |
scar | +0.161 |
illon | +0.154 |
Wil | +0.153 |
ubis | +0.152 |
Thor | +0.146 |
Neg logit contributions | |
IPA | -0.142 |
Rican | -0.130 |
prem | -0.126 |
Orig | -0.120 |
Mortgage | -0.119 |
Token | quarterback |
Token ID | 7880 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
IPA | +0.602 |
Rican | +0.559 |
prem | +0.544 |
Orig | +0.523 |
Mortgage | +0.520 |
Neg logit contributions | |
scar | -0.496 |
illon | -0.472 |
Wil | -0.467 |
ubis | -0.464 |
Thor | -0.443 |
Token | released |
Token ID | 2716 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
IPA | +0.542 |
Rican | +0.498 |
prem | +0.486 |
Orig | +0.464 |
Mortgage | +0.459 |
Neg logit contributions | |
scar | -0.546 |
illon | -0.524 |
Wil | -0.520 |
ubis | -0.517 |
Thor | -0.497 |
Token | the |
Token ID | 262 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
scar | +0.024 |
illon | +0.023 |
Wil | +0.023 |
ubis | +0.023 |
Thor | +0.022 |
Neg logit contributions | |
IPA | -0.022 |
Rican | -0.020 |
prem | -0.020 |
Orig | -0.019 |
Mortgage | -0.019 |
Token | football |
Token ID | 4346 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
scar | +0.025 |
illon | +0.024 |
Wil | +0.023 |
ubis | +0.023 |
Thor | +0.022 |
Neg logit contributions | |
IPA | -0.052 |
Rican | -0.049 |
prem | -0.048 |
Orig | -0.047 |
Mortgage | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
scar | +0.132 |
illon | +0.126 |
Wil | +0.126 |
ubis | +0.125 |
Thor | +0.120 |
Neg logit contributions | |
IPA | -0.136 |
Rican | -0.125 |
prem | -0.122 |
Orig | -0.117 |
Mortgage | -0.116 |
Token | ial |
Token ID | 498 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
IPA | +0.265 |
Rican | +0.239 |
prem | +0.227 |
Orig | +0.215 |
Mortgage | +0.214 |
Neg logit contributions | |
scar | -0.431 |
illon | -0.416 |
Wil | -0.412 |
ubis | -0.411 |
Thor | -0.397 |
Token | et |
Token ID | 316 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
scar | +0.213 |
illon | +0.202 |
Wil | +0.200 |
ubis | +0.199 |
Thor | +0.190 |
Neg logit contributions | |
IPA | -0.257 |
Rican | -0.239 |
prem | -0.232 |
Orig | -0.223 |
Mortgage | -0.222 |
Token | � |
Token ID | 447 |
Feature activation | +2.28e-03 |
Pos logit contributions | |
IPA | +0.196 |
Rican | +0.180 |
prem | +0.174 |
Orig | +0.166 |
Mortgage | +0.166 |
Neg logit contributions | |
scar | -0.215 |
illon | -0.206 |
Wil | -0.205 |
ubis | -0.204 |
Thor | -0.196 |
Token | � |
Token ID | 247 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
IPA | +0.002 |
Rican | +0.002 |
prem | +0.002 |
Orig | +0.002 |
Mortgage | +0.002 |
Neg logit contributions | |
scar | -0.001 |
illon | -0.001 |
Wil | -0.001 |
ubis | -0.001 |
Thor | -0.001 |
Token | s |
Token ID | 82 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
IPA | +0.025 |
Rican | +0.023 |
prem | +0.022 |
Orig | +0.021 |
Mortgage | +0.021 |
Neg logit contributions | |
scar | -0.029 |
illon | -0.027 |
Wil | -0.027 |
ubis | -0.027 |
Thor | -0.026 |
Token | little |
Token ID | 1310 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
scar | +0.056 |
illon | +0.054 |
Wil | +0.053 |
ubis | +0.053 |
Thor | +0.051 |
Neg logit contributions | |
IPA | -0.044 |
Rican | -0.041 |
prem | -0.039 |
Orig | -0.037 |
Mortgage | -0.037 |
Token | w |
Token ID | 266 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
IPA | +0.190 |
Rican | +0.173 |
prem | +0.167 |
Orig | +0.159 |
Mortgage | +0.159 |
Neg logit contributions | |
scar | -0.234 |
illon | -0.225 |
Wil | -0.223 |
ubis | -0.222 |
Thor | -0.214 |
Token | unders |
Token ID | 41116 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
scar | +0.420 |
illon | +0.395 |
Wil | +0.391 |
ubis | +0.388 |
Thor | +0.372 |
Neg logit contributions | |
IPA | -0.657 |
Rican | -0.614 |
prem | -0.598 |
Orig | -0.579 |
Mortgage | -0.576 |
Token | pe |
Token ID | 431 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
IPA | +1.091 |
Rican | +0.996 |
prem | +0.977 |
Orig | +0.932 |
Mortgage | +0.926 |
Neg logit contributions | |
scar | -1.078 |
Wil | -1.028 |
illon | -1.025 |
ubis | -1.021 |
Thor | -0.978 |
Token | aker |
Token ID | 3110 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
IPA | +0.180 |
Rican | +0.166 |
prem | +0.161 |
Orig | +0.154 |
Mortgage | +0.153 |
Neg logit contributions | |
scar | -0.186 |
illon | -0.178 |
Wil | -0.177 |
ubis | -0.176 |
Thor | -0.169 |
Token | be |
Token ID | 307 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
IPA | +0.805 |
Rican | +0.737 |
prem | +0.717 |
Orig | +0.684 |
Mortgage | +0.681 |
Neg logit contributions | |
scar | -0.860 |
illon | -0.822 |
Wil | -0.818 |
ubis | -0.813 |
Thor | -0.780 |
Token | grid |
Token ID | 10706 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
IPA | +0.235 |
Rican | +0.214 |
prem | +0.208 |
Orig | +0.197 |
Mortgage | +0.196 |
Neg logit contributions | |
scar | -0.285 |
illon | -0.274 |
Wil | -0.271 |
ubis | -0.270 |
Thor | -0.260 |
Token | resulting |
Token ID | 7186 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
IPA | +0.507 |
Rican | +0.465 |
prem | +0.453 |
Orig | +0.431 |
Mortgage | +0.426 |
Neg logit contributions | |
scar | -0.559 |
illon | -0.536 |
ubis | -0.532 |
Wil | -0.530 |
Thor | -0.508 |
Token | from |
Token ID | 422 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
IPA | +0.084 |
Rican | +0.077 |
prem | +0.075 |
Orig | +0.072 |
Mortgage | +0.071 |
Neg logit contributions | |
scar | -0.087 |
illon | -0.084 |
Wil | -0.083 |
ubis | -0.083 |
Thor | -0.079 |
Token | this |
Token ID | 428 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
scar | +0.248 |
illon | +0.238 |
Wil | +0.236 |
ubis | +0.235 |
Thor | +0.226 |
Neg logit contributions | |
IPA | -0.210 |
Rican | -0.192 |
prem | -0.186 |
Orig | -0.177 |
Mortgage | -0.176 |
Token | operation |
Token ID | 4905 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
IPA | +0.136 |
Rican | +0.124 |
prem | +0.120 |
Orig | +0.114 |
Mortgage | +0.113 |
Neg logit contributions | |
scar | -0.177 |
illon | -0.171 |
Wil | -0.169 |
ubis | -0.169 |
Thor | -0.163 |
Token | must |
Token ID | 1276 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
scar | +0.256 |
illon | +0.246 |
Wil | +0.244 |
ubis | +0.244 |
Thor | +0.234 |
Neg logit contributions | |
IPA | -0.228 |
Rican | -0.208 |
prem | -0.203 |
Orig | -0.194 |
Mortgage | -0.192 |
Token | preserve |
Token ID | 12201 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
IPA | +0.356 |
Rican | +0.323 |
prem | +0.318 |
Orig | +0.302 |
Mortgage | +0.299 |
Neg logit contributions | |
scar | -0.391 |
illon | -0.373 |
Wil | -0.372 |
ubis | -0.370 |
Thor | -0.356 |
Token | the |
Token ID | 262 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
scar | +0.625 |
illon | +0.598 |
Wil | +0.593 |
ubis | +0.590 |
Thor | +0.566 |
Neg logit contributions | |
IPA | -0.599 |
Rican | -0.552 |
prem | -0.534 |
Orig | -0.510 |
Mortgage | -0.508 |
Token | // |
Token ID | 3373 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
IPA | +0.053 |
Rican | +0.048 |
prem | +0.046 |
Orig | +0.044 |
Mortgage | +0.044 |
Neg logit contributions | |
scar | -0.069 |
illon | -0.067 |
Wil | -0.066 |
ubis | -0.066 |
Thor | -0.063 |
Token | same |
Token ID | 976 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
IPA | +0.042 |
Rican | +0.038 |
prem | +0.037 |
Orig | +0.035 |
Mortgage | +0.035 |
Neg logit contributions | |
scar | -0.053 |
illon | -0.051 |
Wil | -0.050 |
ubis | -0.050 |
Thor | -0.048 |
Token | total |
Token ID | 2472 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
IPA | +0.012 |
Rican | +0.011 |
prem | +0.011 |
Orig | +0.010 |
Mortgage | +0.010 |
Neg logit contributions | |
scar | -0.015 |
illon | -0.015 |
Wil | -0.015 |
ubis | -0.015 |
Thor | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
IPA | +0.381 |
Rican | +0.352 |
prem | +0.341 |
Orig | +0.326 |
Mortgage | +0.325 |
Neg logit contributions | |
scar | -0.369 |
illon | -0.352 |
Wil | -0.349 |
ubis | -0.347 |
Thor | -0.333 |
Token | intention |
Token ID | 6778 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
scar | +0.077 |
illon | +0.074 |
Wil | +0.073 |
ubis | +0.073 |
Thor | +0.070 |
Neg logit contributions | |
IPA | -0.069 |
Rican | -0.063 |
prem | -0.061 |
Orig | -0.058 |
Mortgage | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
scar | +0.183 |
illon | +0.174 |
Wil | +0.173 |
ubis | +0.172 |
Thor | +0.165 |
Neg logit contributions | |
IPA | -0.201 |
Rican | -0.185 |
prem | -0.181 |
Orig | -0.173 |
Mortgage | -0.172 |
Token | caus |
Token ID | 26846 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
scar | +0.118 |
illon | +0.113 |
Wil | +0.112 |
ubis | +0.111 |
Thor | +0.107 |
Neg logit contributions | |
IPA | -0.097 |
Rican | -0.089 |
prem | -0.086 |
Orig | -0.081 |
Mortgage | -0.081 |
Token | ality |
Token ID | 1483 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
scar | +0.097 |
illon | +0.092 |
Wil | +0.092 |
ubis | +0.091 |
Thor | +0.087 |
Neg logit contributions | |
IPA | -0.110 |
Rican | -0.102 |
prem | -0.099 |
Orig | -0.095 |
Mortgage | -0.095 |
Token | in |
Token ID | 287 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
IPA | +0.139 |
Rican | +0.128 |
prem | +0.125 |
Orig | +0.119 |
Mortgage | +0.118 |
Neg logit contributions | |
scar | -0.135 |
illon | -0.129 |
Wil | -0.128 |
ubis | -0.127 |
Thor | -0.122 |
Token | jazz |
Token ID | 21274 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
IPA | +0.162 |
Rican | +0.148 |
prem | +0.143 |
Orig | +0.136 |
Mortgage | +0.135 |
Neg logit contributions | |
scar | -0.204 |
illon | -0.196 |
Wil | -0.195 |
ubis | -0.194 |
Thor | -0.187 |
Token | improvis |
Token ID | 47833 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
scar | +0.299 |
illon | +0.286 |
Wil | +0.284 |
ubis | +0.282 |
Thor | +0.272 |
Neg logit contributions | |
IPA | -0.270 |
Rican | -0.248 |
prem | -0.240 |
Orig | -0.229 |
Mortgage | -0.228 |
Token | ation |
Token ID | 341 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
scar | +0.317 |
illon | +0.301 |
Wil | +0.299 |
ubis | +0.296 |
Thor | +0.283 |
Neg logit contributions | |
IPA | -0.394 |
Rican | -0.365 |
prem | -0.356 |
Orig | -0.342 |
Mortgage | -0.341 |
Token | ; |
Token ID | 26 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
scar | +0.345 |
illon | +0.330 |
Wil | +0.327 |
ubis | +0.325 |
Thor | +0.312 |
Neg logit contributions | |
IPA | -0.360 |
Rican | -0.331 |
prem | -0.323 |
Orig | -0.308 |
Mortgage | -0.307 |
Token | these |
Token ID | 777 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
scar | +0.047 |
illon | +0.045 |
Wil | +0.044 |
ubis | +0.044 |
Thor | +0.042 |
Neg logit contributions | |
IPA | -0.046 |
Rican | -0.042 |
prem | -0.041 |
Orig | -0.039 |
Mortgage | -0.039 |
Token | tool |
Token ID | 2891 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
IPA | +0.025 |
Rican | +0.023 |
prem | +0.022 |
Orig | +0.021 |
Mortgage | +0.021 |
Neg logit contributions | |
scar | -0.031 |
illon | -0.030 |
Wil | -0.030 |
ubis | -0.029 |
Thor | -0.028 |
Token | helpful |
Token ID | 7613 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
IPA | +0.386 |
Rican | +0.353 |
prem | +0.342 |
Orig | +0.324 |
Mortgage | +0.322 |
Neg logit contributions | |
scar | -0.456 |
illon | -0.439 |
Wil | -0.435 |
ubis | -0.433 |
Thor | -0.417 |
Token | and |
Token ID | 290 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
IPA | +0.254 |
Rican | +0.234 |
prem | +0.227 |
Orig | +0.217 |
Mortgage | +0.215 |
Neg logit contributions | |
scar | -0.255 |
illon | -0.245 |
Wil | -0.242 |
ubis | -0.241 |
Thor | -0.231 |
Token | you |
Token ID | 345 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
scar | +0.033 |
illon | +0.031 |
Wil | +0.031 |
ubis | +0.031 |
Thor | +0.030 |
Neg logit contributions | |
IPA | -0.032 |
Rican | -0.029 |
prem | -0.029 |
Orig | -0.027 |
Mortgage | -0.027 |
Token | can |
Token ID | 460 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
scar | +0.028 |
illon | +0.027 |
Wil | +0.027 |
ubis | +0.027 |
Thor | +0.026 |
Neg logit contributions | |
IPA | -0.027 |
Rican | -0.025 |
prem | -0.024 |
Orig | -0.023 |
Mortgage | -0.023 |
Token | find |
Token ID | 1064 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
IPA | +0.041 |
Rican | +0.037 |
prem | +0.037 |
Orig | +0.035 |
Mortgage | +0.034 |
Neg logit contributions | |
scar | -0.046 |
illon | -0.044 |
Wil | -0.044 |
ubis | -0.044 |
Thor | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
scar | +0.265 |
illon | +0.253 |
Wil | +0.251 |
ubis | +0.250 |
Thor | +0.240 |
Neg logit contributions | |
IPA | -0.255 |
Rican | -0.235 |
prem | -0.228 |
Orig | -0.217 |
Mortgage | -0.216 |
Token | way |
Token ID | 835 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
IPA | +0.120 |
Rican | +0.109 |
prem | +0.106 |
Orig | +0.101 |
Mortgage | +0.100 |
Neg logit contributions | |
scar | -0.140 |
illon | -0.135 |
Wil | -0.134 |
ubis | -0.133 |
Thor | -0.128 |
Token | to |
Token ID | 284 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
IPA | +0.415 |
Rican | +0.381 |
prem | +0.370 |
Orig | +0.353 |
Mortgage | +0.351 |
Neg logit contributions | |
scar | -0.440 |
illon | -0.421 |
Wil | -0.418 |
ubis | -0.416 |
Thor | -0.399 |
Token | use |
Token ID | 779 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
IPA | +0.060 |
Rican | +0.055 |
prem | +0.053 |
Orig | +0.051 |
Mortgage | +0.050 |
Neg logit contributions | |
scar | -0.064 |
illon | -0.061 |
Wil | -0.061 |
ubis | -0.061 |
Thor | -0.058 |
Token | it |
Token ID | 340 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
IPA | +0.012 |
Rican | +0.011 |
prem | +0.011 |
Orig | +0.011 |
Mortgage | +0.010 |
Neg logit contributions | |
scar | -0.011 |
illon | -0.010 |
Wil | -0.010 |
ubis | -0.010 |
Thor | -0.009 |
Token | hes |
Token ID | 956 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
scar | +0.164 |
illon | +0.154 |
Wil | +0.152 |
ubis | +0.151 |
Thor | +0.142 |
Neg logit contributions | |
IPA | -0.281 |
Rican | -0.264 |
prem | -0.257 |
Orig | -0.248 |
Mortgage | -0.248 |
Token | out |
Token ID | 503 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
scar | +1.425 |
illon | +1.351 |
Wil | +1.339 |
ubis | +1.326 |
Thor | +1.267 |
Neg logit contributions | |
IPA | -1.897 |
Rican | -1.775 |
prem | -1.720 |
Mortgage | -1.657 |
Orig | -1.656 |
Token | this |
Token ID | 428 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
IPA | +0.088 |
Rican | +0.080 |
prem | +0.078 |
Orig | +0.074 |
Mortgage | +0.074 |
Neg logit contributions | |
scar | -0.100 |
illon | -0.096 |
Wil | -0.095 |
ubis | -0.094 |
Thor | -0.091 |
Token | commitment |
Token ID | 7901 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
scar | +0.150 |
illon | +0.144 |
Wil | +0.143 |
ubis | +0.141 |
Thor | +0.137 |
Neg logit contributions | |
IPA | -0.131 |
Rican | -0.120 |
prem | -0.115 |
Orig | -0.110 |
Mortgage | -0.110 |
Token | to |
Token ID | 284 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
scar | +0.186 |
illon | +0.178 |
Wil | +0.176 |
ubis | +0.175 |
Thor | +0.168 |
Neg logit contributions | |
IPA | -0.180 |
Rican | -0.166 |
prem | -0.161 |
Orig | -0.154 |
Mortgage | -0.153 |
Token | the |
Token ID | 262 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
scar | +0.042 |
illon | +0.041 |
Wil | +0.040 |
ubis | +0.040 |
Thor | +0.039 |
Neg logit contributions | |
IPA | -0.039 |
Rican | -0.035 |
prem | -0.034 |
Mortgage | -0.033 |
Orig | -0.033 |
Token | univers |
Token ID | 5820 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
scar | +0.441 |
illon | +0.423 |
Wil | +0.421 |
ubis | +0.416 |
Thor | +0.402 |
Neg logit contributions | |
IPA | -0.374 |
Rican | -0.343 |
prem | -0.329 |
Mortgage | -0.315 |
Orig | -0.314 |
Token | ality |
Token ID | 1483 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
IPA | +0.333 |
Rican | +0.308 |
prem | +0.300 |
Orig | +0.287 |
Mortgage | +0.286 |
Neg logit contributions | |
scar | -0.299 |
illon | -0.285 |
Wil | -0.284 |
ubis | -0.281 |
Thor | -0.270 |
Token | and |
Token ID | 290 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
scar | +0.707 |
illon | +0.675 |
Wil | +0.669 |
ubis | +0.659 |
Thor | +0.634 |
Neg logit contributions | |
IPA | -0.786 |
Rican | -0.732 |
prem | -0.702 |
Orig | -0.680 |
Mortgage | -0.680 |
Token | object |
Token ID | 2134 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
scar | +0.103 |
illon | +0.099 |
Wil | +0.098 |
ubis | +0.098 |
Thor | +0.094 |
Neg logit contributions | |
IPA | -0.094 |
Rican | -0.086 |
prem | -0.084 |
Orig | -0.080 |
Mortgage | -0.079 |
Token | ivity |
Token ID | 3458 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
IPA | +0.609 |
Rican | +0.563 |
prem | +0.547 |
Orig | +0.523 |
Mortgage | +0.520 |
Neg logit contributions | |
scar | -0.578 |
illon | -0.550 |
Wil | -0.548 |
ubis | -0.545 |
Thor | -0.523 |
Token | , |
Token ID | 11 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
scar | +0.859 |
illon | +0.822 |
Wil | +0.815 |
ubis | +0.811 |
Thor | +0.778 |
Neg logit contributions | |
IPA | -0.876 |
Rican | -0.808 |
prem | -0.785 |
Orig | -0.750 |
Mortgage | -0.747 |
Token | and |
Token ID | 290 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
scar | +0.277 |
illon | +0.266 |
Wil | +0.263 |
ubis | +0.262 |
Thor | +0.250 |
Neg logit contributions | |
IPA | -0.303 |
Rican | -0.281 |
prem | -0.273 |
Orig | -0.261 |
Mortgage | -0.260 |
Token | his |
Token ID | 465 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
scar | +0.455 |
illon | +0.436 |
Wil | +0.432 |
ubis | +0.431 |
Thor | +0.413 |
Neg logit contributions | |
IPA | -0.476 |
Rican | -0.440 |
prem | -0.428 |
Orig | -0.409 |
Mortgage | -0.407 |
Token | forces |
Token ID | 3386 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
scar | +0.312 |
illon | +0.301 |
Wil | +0.299 |
ubis | +0.298 |
Thor | +0.288 |
Neg logit contributions | |
IPA | -0.190 |
Rican | -0.171 |
prem | -0.163 |
Orig | -0.154 |
Mortgage | -0.153 |
Token | step |
Token ID | 2239 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
IPA | +0.292 |
Rican | +0.267 |
prem | +0.261 |
Orig | +0.247 |
Mortgage | +0.245 |
Neg logit contributions | |
scar | -0.329 |
illon | -0.316 |
Wil | -0.314 |
ubis | -0.314 |
Thor | -0.302 |
Token | up |
Token ID | 510 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
scar | +0.396 |
illon | +0.379 |
Wil | +0.376 |
ubis | +0.374 |
Thor | +0.359 |
Neg logit contributions | |
IPA | -0.384 |
Rican | -0.354 |
prem | -0.343 |
Orig | -0.328 |
Mortgage | -0.326 |
Token | their |
Token ID | 511 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
scar | +0.135 |
illon | +0.129 |
ubis | +0.128 |
Wil | +0.128 |
Thor | +0.123 |
Neg logit contributions | |
IPA | -0.134 |
Rican | -0.123 |
prem | -0.120 |
Orig | -0.114 |
Mortgage | -0.114 |
Token | assaults |
Token ID | 21832 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
IPA | +0.094 |
Rican | +0.087 |
prem | +0.084 |
Orig | +0.080 |
Mortgage | +0.080 |
Neg logit contributions | |
scar | -0.096 |
illon | -0.092 |
Wil | -0.091 |
ubis | -0.091 |
Thor | -0.087 |
Token | on |
Token ID | 319 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
scar | +0.014 |
illon | +0.014 |
Wil | +0.014 |
ubis | +0.014 |
Thor | +0.013 |
Neg logit contributions | |
IPA | -0.014 |
Rican | -0.013 |
prem | -0.013 |
Orig | -0.012 |
Mortgage | -0.012 |
Token | cities |
Token ID | 4736 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
IPA | +0.173 |
Rican | +0.159 |
prem | +0.154 |
Orig | +0.146 |
Mortgage | +0.146 |
Neg logit contributions | |
scar | -0.186 |
illon | -0.178 |
Wil | -0.176 |
ubis | -0.176 |
Thor | -0.169 |
Token | like |
Token ID | 588 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
scar | +0.012 |
illon | +0.012 |
Wil | +0.012 |
ubis | +0.012 |
Thor | +0.011 |
Neg logit contributions | |
IPA | -0.012 |
Rican | -0.011 |
prem | -0.011 |
Orig | -0.010 |
Mortgage | -0.010 |
Token | he |
Token ID | 258 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
scar | +0.246 |
illon | +0.233 |
Wil | +0.231 |
ubis | +0.230 |
Thor | +0.219 |
Neg logit contributions | |
IPA | -0.329 |
Rican | -0.307 |
prem | -0.299 |
Orig | -0.287 |
Mortgage | -0.287 |
Token | als |
Token ID | 874 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
IPA | +0.032 |
Rican | +0.029 |
prem | +0.028 |
Orig | +0.027 |
Mortgage | +0.027 |
Neg logit contributions | |
scar | -0.027 |
illon | -0.026 |
Wil | -0.025 |
ubis | -0.025 |
Thor | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
IPA | +0.020 |
Rican | +0.018 |
prem | +0.018 |
Orig | +0.017 |
Mortgage | +0.017 |
Neg logit contributions | |
scar | -0.021 |
illon | -0.020 |
Wil | -0.020 |
ubis | -0.020 |
Thor | -0.019 |
Token | so |
Token ID | 523 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
scar | +0.036 |
illon | +0.035 |
Wil | +0.034 |
ubis | +0.034 |
Thor | +0.033 |
Neg logit contributions | |
IPA | -0.037 |
Rican | -0.034 |
prem | -0.033 |
Orig | -0.032 |
Mortgage | -0.032 |
Token | he |
Token ID | 339 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
IPA | +0.134 |
Rican | +0.123 |
prem | +0.120 |
Orig | +0.114 |
Mortgage | +0.113 |
Neg logit contributions | |
scar | -0.143 |
illon | -0.137 |
Wil | -0.136 |
ubis | -0.136 |
Thor | -0.130 |
Token | � |
Token ID | 447 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
scar | +0.015 |
illon | +0.014 |
Wil | +0.014 |
ubis | +0.014 |
Thor | +0.013 |
Neg logit contributions | |
IPA | -0.013 |
Rican | -0.011 |
prem | -0.011 |
Orig | -0.011 |
Mortgage | -0.011 |
Token | � |
Token ID | 247 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
IPA | +0.056 |
Rican | +0.052 |
prem | +0.051 |
Orig | +0.049 |
Mortgage | +0.049 |
Neg logit contributions | |
scar | -0.032 |
illon | -0.030 |
Wil | -0.030 |
ubis | -0.029 |
Thor | -0.028 |
Token | s |
Token ID | 82 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
IPA | +0.042 |
Rican | +0.038 |
prem | +0.037 |
Orig | +0.035 |
Mortgage | +0.035 |
Neg logit contributions | |
scar | -0.048 |
illon | -0.046 |
Wil | -0.046 |
ubis | -0.046 |
Thor | -0.044 |
Token | literally |
Token ID | 7360 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
IPA | +0.313 |
Rican | +0.289 |
prem | +0.280 |
Orig | +0.267 |
Mortgage | +0.266 |
Neg logit contributions | |
scar | -0.319 |
illon | -0.306 |
Wil | -0.304 |
ubis | -0.302 |
Thor | -0.290 |
Token | the |
Token ID | 262 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
IPA | +0.145 |
Rican | +0.134 |
prem | +0.130 |
Orig | +0.124 |
Mortgage | +0.123 |
Neg logit contributions | |
scar | -0.146 |
illon | -0.140 |
Wil | -0.139 |
ubis | -0.138 |
Thor | -0.132 |
Token | perfect |
Token ID | 2818 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
IPA | +0.316 |
Rican | +0.290 |
prem | +0.281 |
Orig | +0.269 |
Mortgage | +0.267 |
Neg logit contributions | |
scar | -0.338 |
illon | -0.325 |
Wil | -0.322 |
ubis | -0.320 |
Thor | -0.307 |
Token | at |
Token ID | 379 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
scar | +0.063 |
illon | +0.061 |
Wil | +0.060 |
ubis | +0.060 |
Thor | +0.058 |
Neg logit contributions | |
IPA | -0.066 |
Rican | -0.061 |
prem | -0.060 |
Orig | -0.057 |
Mortgage | -0.057 |
Token | traditional |
Token ID | 4569 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
IPA | +0.118 |
Rican | +0.106 |
prem | +0.103 |
Orig | +0.097 |
Mortgage | +0.096 |
Neg logit contributions | |
scar | -0.166 |
illon | -0.160 |
Wil | -0.159 |
ubis | -0.158 |
Thor | -0.153 |
Token | public |
Token ID | 1171 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
scar | +0.506 |
illon | +0.489 |
Wil | +0.485 |
ubis | +0.482 |
Thor | +0.467 |
Neg logit contributions | |
IPA | -0.365 |
Rican | -0.331 |
prem | -0.320 |
Orig | -0.300 |
Mortgage | -0.298 |
Token | schools |
Token ID | 4266 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
IPA | +0.358 |
Rican | +0.322 |
prem | +0.309 |
Orig | +0.286 |
Mortgage | +0.285 |
Neg logit contributions | |
scar | -0.604 |
illon | -0.585 |
Wil | -0.581 |
ubis | -0.579 |
Thor | -0.562 |
Token | because |
Token ID | 780 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
scar | +0.336 |
illon | +0.321 |
Wil | +0.318 |
ubis | +0.317 |
Thor | +0.304 |
Neg logit contributions | |
IPA | -0.354 |
Rican | -0.327 |
prem | -0.318 |
Orig | -0.303 |
Mortgage | -0.303 |
Token | unions |
Token ID | 11936 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
IPA | +0.188 |
Rican | +0.172 |
prem | +0.168 |
Orig | +0.160 |
Mortgage | +0.158 |
Neg logit contributions | |
scar | -0.197 |
illon | -0.190 |
Wil | -0.187 |
ubis | -0.187 |
Thor | -0.181 |
Token | prioritize |
Token ID | 32980 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
scar | +0.055 |
illon | +0.053 |
Wil | +0.053 |
ubis | +0.053 |
Thor | +0.050 |
Neg logit contributions | |
IPA | -0.050 |
Rican | -0.045 |
prem | -0.044 |
Orig | -0.042 |
Mortgage | -0.042 |
Token | pension |
Token ID | 13553 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
scar | +0.049 |
illon | +0.047 |
Wil | +0.047 |
ubis | +0.047 |
Thor | +0.045 |
Neg logit contributions | |
IPA | -0.046 |
Rican | -0.042 |
prem | -0.041 |
Orig | -0.039 |
Mortgage | -0.039 |
Token | benefits |
Token ID | 4034 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
IPA | +0.282 |
Rican | +0.258 |
prem | +0.252 |
Orig | +0.237 |
Mortgage | +0.235 |
Neg logit contributions | |
scar | -0.345 |
illon | -0.333 |
Wil | -0.333 |
ubis | -0.331 |
Thor | -0.319 |
Token | in |
Token ID | 287 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
IPA | +0.580 |
Rican | +0.532 |
prem | +0.527 |
Orig | +0.493 |
Mortgage | +0.492 |
Neg logit contributions | |
scar | -0.599 |
illon | -0.575 |
Wil | -0.573 |
ubis | -0.571 |
Thor | -0.548 |
Token | negotiations |
Token ID | 9825 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
IPA | +0.028 |
Rican | +0.026 |
prem | +0.025 |
Orig | +0.024 |
Mortgage | +0.024 |
Neg logit contributions | |
scar | -0.030 |
illon | -0.028 |
Wil | -0.028 |
ubis | -0.028 |
Thor | -0.027 |
Token | however |
Token ID | 2158 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
scar | +0.173 |
illon | +0.166 |
ubis | +0.165 |
Wil | +0.164 |
Thor | +0.157 |
Neg logit contributions | |
IPA | -0.155 |
Rican | -0.142 |
prem | -0.138 |
Orig | -0.132 |
Mortgage | -0.130 |
Token | , |
Token ID | 11 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
scar | +0.401 |
illon | +0.383 |
Wil | +0.380 |
ubis | +0.377 |
Thor | +0.362 |
Neg logit contributions | |
IPA | -0.419 |
Rican | -0.387 |
prem | -0.375 |
Orig | -0.359 |
Mortgage | -0.358 |
Token | there |
Token ID | 612 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
scar | +0.356 |
illon | +0.341 |
Wil | +0.338 |
ubis | +0.337 |
Thor | +0.324 |
Neg logit contributions | |
IPA | -0.316 |
Rican | -0.290 |
prem | -0.280 |
Orig | -0.267 |
Mortgage | -0.265 |
Token | 's |
Token ID | 338 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
IPA | +0.201 |
Rican | +0.184 |
prem | +0.177 |
Orig | +0.169 |
Mortgage | +0.168 |
Neg logit contributions | |
scar | -0.246 |
illon | -0.236 |
Wil | -0.234 |
ubis | -0.233 |
Thor | -0.224 |
Token | no |
Token ID | 645 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
scar | +0.050 |
illon | +0.048 |
Wil | +0.047 |
ubis | +0.047 |
Thor | +0.045 |
Neg logit contributions | |
IPA | -0.054 |
Rican | -0.050 |
prem | -0.049 |
Orig | -0.047 |
Mortgage | -0.046 |
Token | way |
Token ID | 835 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
scar | +0.622 |
illon | +0.596 |
Wil | +0.592 |
ubis | +0.589 |
Thor | +0.567 |
Neg logit contributions | |
IPA | -0.535 |
Rican | -0.490 |
prem | -0.473 |
Orig | -0.451 |
Mortgage | -0.449 |
Token | to |
Token ID | 284 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
scar | +1.063 |
illon | +1.015 |
Wil | +1.013 |
ubis | +0.998 |
Thor | +0.969 |
Neg logit contributions | |
IPA | -0.972 |
Rican | -0.892 |
prem | -0.862 |
Mortgage | -0.828 |
Orig | -0.819 |
Token | know |
Token ID | 760 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
scar | +0.913 |
illon | +0.871 |
Wil | +0.863 |
ubis | +0.858 |
Thor | +0.822 |
Neg logit contributions | |
IPA | -0.991 |
Rican | -0.918 |
prem | -0.890 |
Orig | -0.853 |
Mortgage | -0.850 |
Token | exactly |
Token ID | 3446 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
scar | +0.311 |
illon | +0.297 |
Wil | +0.294 |
ubis | +0.293 |
Thor | +0.280 |
Neg logit contributions | |
IPA | -0.335 |
Rican | -0.310 |
prem | -0.301 |
Orig | -0.288 |
Mortgage | -0.287 |
Token | how |
Token ID | 703 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
IPA | +0.822 |
Rican | +0.750 |
prem | +0.732 |
Orig | +0.700 |
Mortgage | +0.688 |
Neg logit contributions | |
scar | -0.873 |
illon | -0.837 |
Wil | -0.831 |
ubis | -0.831 |
Thor | -0.793 |
Token | long |
Token ID | 890 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
IPA | +0.393 |
Rican | +0.363 |
prem | +0.353 |
Orig | +0.339 |
Mortgage | +0.334 |
Neg logit contributions | |
scar | -0.371 |
illon | -0.357 |
ubis | -0.353 |
Wil | -0.352 |
Thor | -0.334 |
Token | 8 |
Token ID | 23 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
IPA | +0.245 |
Rican | +0.226 |
prem | +0.220 |
Orig | +0.212 |
Mortgage | +0.211 |
Neg logit contributions | |
scar | -0.214 |
illon | -0.206 |
ubis | -0.203 |
Wil | -0.202 |
Thor | -0.192 |
Token | 35 |
Token ID | 2327 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
IPA | +0.158 |
Rican | +0.145 |
prem | +0.141 |
Orig | +0.134 |
Mortgage | +0.134 |
Neg logit contributions | |
scar | -0.172 |
illon | -0.165 |
Wil | -0.164 |
ubis | -0.163 |
Thor | -0.157 |
Token | ) |
Token ID | 8 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
IPA | +0.226 |
Rican | +0.208 |
prem | +0.201 |
Orig | +0.192 |
Mortgage | +0.191 |
Neg logit contributions | |
scar | -0.248 |
illon | -0.237 |
Wil | -0.235 |
ubis | -0.234 |
Thor | -0.225 |
Token |
|
Token ID | 198 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
IPA | +0.221 |
Rican | +0.205 |
prem | +0.199 |
Orig | +0.191 |
Mortgage | +0.191 |
Neg logit contributions | |
scar | -0.196 |
illon | -0.186 |
Wil | -0.185 |
ubis | -0.184 |
Thor | -0.176 |
Token |
|
Token ID | 198 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
scar | +0.141 |
illon | +0.134 |
Wil | +0.133 |
ubis | +0.132 |
Thor | +0.126 |
Neg logit contributions | |
IPA | -0.158 |
Rican | -0.146 |
prem | -0.142 |
Orig | -0.136 |
Mortgage | -0.135 |
Token | Enter |
Token ID | 17469 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
scar | +0.049 |
illon | +0.047 |
ubis | +0.046 |
Wil | +0.046 |
Thor | +0.044 |
Neg logit contributions | |
IPA | -0.058 |
Rican | -0.053 |
prem | -0.052 |
Orig | -0.050 |
Mortgage | -0.050 |
Token | 4 |
Token ID | 604 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
scar | +1.314 |
illon | +1.259 |
Wil | +1.249 |
ubis | +1.243 |
Thor | +1.196 |
Neg logit contributions | |
IPA | -1.151 |
Rican | -1.056 |
prem | -1.019 |
Orig | -0.972 |
Mortgage | -0.969 |
Token | different |
Token ID | 1180 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
IPA | +0.136 |
Rican | +0.123 |
prem | +0.120 |
Orig | +0.114 |
Mortgage | +0.113 |
Neg logit contributions | |
scar | -0.166 |
illon | -0.160 |
Wil | -0.159 |
ubis | -0.157 |
Thor | -0.152 |
Token | portals |
Token ID | 42604 |
Feature activation | +6.56e-02 |
Pos logit contributions | |
IPA | +0.635 |
Rican | +0.577 |
prem | +0.563 |
Orig | +0.532 |
Mortgage | +0.529 |
Neg logit contributions | |
scar | -0.765 |
illon | -0.733 |
Wil | -0.731 |
ubis | -0.724 |
Thor | -0.698 |
Token | without |
Token ID | 1231 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
IPA | +0.051 |
Rican | +0.047 |
prem | +0.046 |
Orig | +0.044 |
Mortgage | +0.043 |
Neg logit contributions | |
scar | -0.052 |
illon | -0.049 |
Wil | -0.049 |
ubis | -0.049 |
Thor | -0.047 |
Token | touching |
Token ID | 15241 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
IPA | +0.106 |
Rican | +0.098 |
prem | +0.096 |
Orig | +0.091 |
Mortgage | +0.091 |
Neg logit contributions | |
scar | -0.105 |
illon | -0.101 |
Wil | -0.100 |
ubis | -0.100 |
Thor | -0.095 |
Token | 't |
Token ID | 470 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
scar | +0.188 |
illon | +0.179 |
Wil | +0.176 |
ubis | +0.176 |
Thor | +0.167 |
Neg logit contributions | |
IPA | -0.265 |
Rican | -0.247 |
prem | -0.240 |
Orig | -0.232 |
Mortgage | -0.231 |
Token | included |
Token ID | 3017 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
IPA | +0.195 |
Rican | +0.180 |
prem | +0.175 |
Orig | +0.168 |
Mortgage | +0.166 |
Neg logit contributions | |
scar | -0.188 |
illon | -0.182 |
Wil | -0.179 |
ubis | -0.178 |
Thor | -0.170 |
Token | in |
Token ID | 287 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
scar | +0.102 |
illon | +0.098 |
Wil | +0.098 |
ubis | +0.097 |
Thor | +0.093 |
Neg logit contributions | |
IPA | -0.101 |
Rican | -0.093 |
prem | -0.090 |
Orig | -0.086 |
Mortgage | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
scar | +0.135 |
illon | +0.130 |
Wil | +0.129 |
ubis | +0.128 |
Thor | +0.123 |
Neg logit contributions | |
IPA | -0.126 |
Rican | -0.116 |
prem | -0.112 |
Orig | -0.107 |
Mortgage | -0.106 |
Token | mont |
Token ID | 40689 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
scar | +0.208 |
illon | +0.200 |
Wil | +0.199 |
ubis | +0.198 |
Thor | +0.191 |
Neg logit contributions | |
IPA | -0.153 |
Rican | -0.139 |
prem | -0.134 |
Orig | -0.127 |
Mortgage | -0.127 |
Token | age |
Token ID | 496 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
scar | +0.097 |
illon | +0.091 |
Wil | +0.091 |
ubis | +0.090 |
Thor | +0.086 |
Neg logit contributions | |
IPA | -0.138 |
Rican | -0.129 |
prem | -0.126 |
Orig | -0.121 |
Mortgage | -0.121 |
Token | depicting |
Token ID | 27561 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
scar | +1.247 |
illon | +1.189 |
Wil | +1.178 |
ubis | +1.172 |
Thor | +1.125 |
Neg logit contributions | |
IPA | -1.258 |
Rican | -1.162 |
prem | -1.122 |
Orig | -1.075 |
Mortgage | -1.075 |
Token | the |
Token ID | 262 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
scar | +0.018 |
illon | +0.017 |
Wil | +0.017 |
ubis | +0.017 |
Thor | +0.016 |
Neg logit contributions | |
IPA | -0.016 |
Rican | -0.014 |
prem | -0.014 |
Orig | -0.013 |
Mortgage | -0.013 |
Token | principal |
Token ID | 10033 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
scar | +0.056 |
illon | +0.054 |
Wil | +0.053 |
ubis | +0.053 |
Thor | +0.051 |
Neg logit contributions | |
IPA | -0.044 |
Rican | -0.040 |
prem | -0.039 |
Orig | -0.037 |
Mortgage | -0.037 |
Token | M |
Token ID | 337 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
IPA | +0.573 |
Rican | +0.528 |
prem | +0.518 |
Orig | +0.493 |
Mortgage | +0.487 |
Neg logit contributions | |
scar | -0.602 |
illon | -0.578 |
Wil | -0.574 |
ubis | -0.567 |
Thor | -0.548 |
Token | upp |
Token ID | 7211 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
IPA | +0.216 |
Rican | +0.197 |
prem | +0.192 |
Orig | +0.183 |
Mortgage | +0.182 |
Neg logit contributions | |
scar | -0.258 |
illon | -0.247 |
Wil | -0.247 |
ubis | -0.244 |
Thor | -0.237 |
Token | lots |
Token ID | 6041 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
IPA | +0.385 |
Rican | +0.353 |
prem | +0.342 |
Orig | +0.325 |
Mortgage | +0.323 |
Neg logit contributions | |
scar | -0.452 |
illon | -0.435 |
Wil | -0.431 |
ubis | -0.429 |
Thor | -0.412 |
Token | of |
Token ID | 286 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
scar | +0.957 |
illon | +0.913 |
Wil | +0.906 |
ubis | +0.900 |
Thor | +0.863 |
Neg logit contributions | |
IPA | -1.011 |
Rican | -0.936 |
prem | -0.906 |
Orig | -0.869 |
Mortgage | -0.867 |
Token | fun |
Token ID | 1257 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
IPA | +0.146 |
Rican | +0.133 |
prem | +0.128 |
Orig | +0.122 |
Mortgage | +0.121 |
Neg logit contributions | |
scar | -0.186 |
illon | -0.179 |
Wil | -0.178 |
ubis | -0.177 |
Thor | -0.170 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
scar | +0.269 |
illon | +0.257 |
Wil | +0.255 |
ubis | +0.254 |
Thor | +0.243 |
Neg logit contributions | |
IPA | -0.275 |
Rican | -0.253 |
prem | -0.247 |
Orig | -0.235 |
Mortgage | -0.234 |
Token | You |
Token ID | 921 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
IPA | +0.133 |
Rican | +0.122 |
prem | +0.118 |
Orig | +0.114 |
Mortgage | +0.113 |
Neg logit contributions | |
scar | -0.131 |
illon | -0.127 |
Wil | -0.125 |
ubis | -0.124 |
Thor | -0.119 |
Token | mention |
Token ID | 3068 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
IPA | +0.108 |
Rican | +0.099 |
prem | +0.096 |
Orig | +0.092 |
Mortgage | +0.091 |
Neg logit contributions | |
scar | -0.114 |
illon | -0.109 |
Wil | -0.109 |
ubis | -0.108 |
Thor | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
scar | +1.015 |
illon | +0.971 |
Wil | +0.963 |
ubis | +0.957 |
Thor | +0.918 |
Neg logit contributions | |
IPA | -1.027 |
Rican | -0.947 |
prem | -0.920 |
Orig | -0.880 |
Mortgage | -0.875 |
Token | she |
Token ID | 673 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
scar | +0.190 |
illon | +0.182 |
Wil | +0.181 |
ubis | +0.179 |
Thor | +0.172 |
Neg logit contributions | |
IPA | -0.172 |
Rican | -0.157 |
prem | -0.153 |
Orig | -0.145 |
Mortgage | -0.144 |
Token | eping |
Token ID | 7213 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
scar | +0.563 |
illon | +0.534 |
Wil | +0.530 |
ubis | +0.526 |
Thor | +0.502 |
Neg logit contributions | |
IPA | -0.730 |
Rican | -0.680 |
prem | -0.661 |
Orig | -0.636 |
Mortgage | -0.634 |
Token | mechanic |
Token ID | 21600 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
IPA | +0.250 |
Rican | +0.228 |
prem | +0.223 |
Orig | +0.211 |
Mortgage | +0.210 |
Neg logit contributions | |
scar | -0.268 |
illon | -0.259 |
Wil | -0.258 |
ubis | -0.256 |
Thor | -0.244 |
Token | changed |
Token ID | 3421 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
IPA | +0.308 |
Rican | +0.282 |
prem | +0.278 |
Orig | +0.263 |
Mortgage | +0.261 |
Neg logit contributions | |
scar | -0.308 |
illon | -0.298 |
Wil | -0.295 |
ubis | -0.292 |
Thor | -0.280 |
Token | of |
Token ID | 286 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
IPA | +1.199 |
Rican | +1.094 |
prem | +1.071 |
Orig | +1.019 |
Mortgage | +1.011 |
Neg logit contributions | |
scar | -1.240 |
illon | -1.189 |
ubis | -1.179 |
Wil | -1.175 |
Thor | -1.130 |
Token | the |
Token ID | 262 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
IPA | +0.343 |
Rican | +0.312 |
prem | +0.303 |
Orig | +0.288 |
Mortgage | +0.286 |
Neg logit contributions | |
scar | -0.411 |
illon | -0.395 |
Wil | -0.392 |
ubis | -0.391 |
Thor | -0.375 |
Token | state |
Token ID | 1181 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
scar | +0.074 |
illon | +0.071 |
Wil | +0.071 |
ubis | +0.070 |
Thor | +0.068 |
Neg logit contributions | |
IPA | -0.059 |
Rican | -0.054 |
prem | -0.052 |
Orig | -0.050 |
Mortgage | -0.049 |
Token | under |
Token ID | 739 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
scar | +0.218 |
illon | +0.208 |
Wil | +0.206 |
ubis | +0.205 |
Thor | +0.197 |
Neg logit contributions | |
IPA | -0.221 |
Rican | -0.204 |
prem | -0.197 |
Orig | -0.189 |
Mortgage | -0.188 |
Token | the |
Token ID | 262 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
IPA | +0.593 |
Rican | +0.539 |
prem | +0.527 |
Orig | +0.500 |
Mortgage | +0.497 |
Neg logit contributions | |
scar | -0.665 |
illon | -0.639 |
Wil | -0.633 |
ubis | -0.631 |
Thor | -0.609 |
Token | influence |
Token ID | 4588 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
scar | +0.170 |
illon | +0.163 |
Wil | +0.162 |
ubis | +0.161 |
Thor | +0.155 |
Neg logit contributions | |
IPA | -0.131 |
Rican | -0.119 |
prem | -0.115 |
Orig | -0.109 |
Mortgage | -0.109 |
Token | of |
Token ID | 286 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
scar | +1.189 |
illon | +1.136 |
Wil | +1.127 |
ubis | +1.115 |
Thor | +1.064 |
Neg logit contributions | |
IPA | -1.306 |
Rican | -1.217 |
prem | -1.171 |
Orig | -1.127 |
Mortgage | -1.125 |
Token | high |
Token ID | 1029 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
IPA | +0.271 |
Rican | +0.245 |
prem | +0.239 |
Orig | +0.226 |
Mortgage | +0.224 |
Neg logit contributions | |
scar | -0.348 |
illon | -0.334 |
Wil | -0.331 |
ubis | -0.331 |
Thor | -0.319 |
Token | profile |
Token ID | 7034 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
scar | +0.372 |
illon | +0.357 |
Wil | +0.355 |
ubis | +0.353 |
Thor | +0.340 |
Neg logit contributions | |
IPA | -0.309 |
Rican | -0.283 |
prem | -0.273 |
Orig | -0.260 |
Mortgage | -0.259 |
Token | politicians |
Token ID | 7602 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
scar | +0.152 |
illon | +0.146 |
Wil | +0.145 |
ubis | +0.144 |
Thor | +0.139 |
Neg logit contributions | |
IPA | -0.111 |
Rican | -0.101 |
prem | -0.097 |
Orig | -0.092 |
Mortgage | -0.092 |
Token | � |
Token ID | 447 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
IPA | +0.338 |
Rican | +0.311 |
prem | +0.303 |
Orig | +0.290 |
Mortgage | +0.287 |
Neg logit contributions | |
scar | -0.345 |
illon | -0.330 |
Wil | -0.327 |
ubis | -0.326 |
Thor | -0.313 |