Token | be |
Token ID | 307 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
Roose | +0.178 |
muzzle | +0.159 |
ibaba | +0.158 |
rils | +0.149 |
bang | +0.148 |
Neg logit contributions | |
�� | -0.204 |
aur | -0.192 |
� | -0.177 |
aires | -0.170 |
�� | -0.165 |
Token | the |
Token ID | 262 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
Roose | +1.245 |
ibaba | +1.111 |
muzzle | +1.101 |
ICLE | +1.033 |
smart | +1.033 |
Neg logit contributions | |
�� | -1.463 |
aur | -1.382 |
� | -1.276 |
aires | -1.226 |
�� | -1.186 |
Token | case |
Token ID | 1339 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
�� | +0.511 |
aur | +0.485 |
� | +0.451 |
aires | +0.430 |
�� | +0.417 |
Neg logit contributions | |
Roose | -0.403 |
ibaba | -0.361 |
muzzle | -0.353 |
ICLE | -0.336 |
rils | -0.334 |
Token | with |
Token ID | 351 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
Roose | +1.595 |
ibaba | +1.430 |
muzzle | +1.424 |
smart | +1.346 |
ICLE | +1.339 |
Neg logit contributions | |
�� | -1.642 |
aur | -1.527 |
� | -1.414 |
aires | -1.343 |
�� | -1.305 |
Token | University |
Token ID | 2059 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
Roose | +0.910 |
ibaba | +0.811 |
muzzle | +0.806 |
ICLE | +0.757 |
smart | +0.756 |
Neg logit contributions | |
�� | -1.050 |
aur | -0.992 |
� | -0.916 |
aires | -0.876 |
�� | -0.849 |
Token | of |
Token ID | 286 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
�� | +0.395 |
aur | +0.374 |
� | +0.346 |
aires | +0.331 |
�� | +0.322 |
Neg logit contributions | |
Roose | -0.316 |
ibaba | -0.280 |
muzzle | -0.277 |
smart | -0.260 |
azines | -0.258 |
Token | Ottawa |
Token ID | 14074 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
aur | +12.103 |
�� | +11.739 |
aires | +10.588 |
� | +10.454 |
ARY | +9.689 |
Neg logit contributions | |
ibaba | -13.168 |
Roose | -12.210 |
rils | -11.912 |
ICLE | -11.854 |
smart | -11.772 |
Token | law |
Token ID | 1099 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
�� | +5.532 |
aur | +5.393 |
� | +4.896 |
aires | +4.726 |
equival | +4.535 |
Neg logit contributions | |
Roose | -4.482 |
ibaba | -4.204 |
ICLE | -3.884 |
muzzle | -3.841 |
rils | -3.831 |
Token | professor |
Token ID | 6240 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
�� | +6.286 |
aur | +6.154 |
� | +5.703 |
aires | +5.378 |
ional | +5.180 |
Neg logit contributions | |
Roose | -5.153 |
ibaba | -4.872 |
muzzle | -4.446 |
ICLE | -4.353 |
azines | -4.230 |
Token | Elizabeth |
Token ID | 10674 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
Roose | +0.902 |
ibaba | +0.796 |
muzzle | +0.792 |
ICLE | +0.742 |
smart | +0.742 |
Neg logit contributions | |
�� | -1.093 |
aur | -1.026 |
� | -0.953 |
aires | -0.911 |
�� | -0.887 |
Token | She |
Token ID | 1375 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
�� | +2.127 |
aur | +2.052 |
� | +1.886 |
aires | +1.823 |
�� | +1.743 |
Neg logit contributions | |
Roose | -1.754 |
ibaba | -1.597 |
muzzle | -1.571 |
ICLE | -1.490 |
rils | -1.469 |
Token | L |
Token ID | 406 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
�� | +2.329 |
aur | +2.236 |
� | +2.059 |
aires | +1.986 |
�� | +1.905 |
Neg logit contributions | |
Roose | -1.880 |
ibaba | -1.707 |
muzzle | -1.687 |
ICLE | -1.588 |
smart | -1.573 |
Token | inder |
Token ID | 5540 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
�� | +1.584 |
aur | +1.502 |
� | +1.388 |
aires | +1.329 |
�� | +1.288 |
Neg logit contributions | |
Roose | -1.309 |
ibaba | -1.169 |
muzzle | -1.155 |
ICLE | -1.089 |
smart | -1.084 |
Token | of |
Token ID | 286 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
Roose | +1.210 |
ibaba | +1.087 |
muzzle | +1.075 |
smart | +1.017 |
ICLE | +1.012 |
Neg logit contributions | |
�� | -1.321 |
aur | -1.237 |
� | -1.144 |
aires | -1.093 |
�� | -1.061 |
Token | the |
Token ID | 262 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
�� | +3.145 |
aur | +2.994 |
� | +2.740 |
aires | +2.592 |
�� | +2.522 |
Neg logit contributions | |
Roose | -2.959 |
ibaba | -2.739 |
muzzle | -2.661 |
ICLE | -2.559 |
azines | -2.532 |
Token | University |
Token ID | 2059 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
�� | +2.249 |
aur | +2.153 |
� | +2.001 |
aires | +1.905 |
�� | +1.857 |
Neg logit contributions | |
Roose | -1.597 |
ibaba | -1.462 |
muzzle | -1.397 |
ICLE | -1.330 |
rils | -1.321 |
Token | of |
Token ID | 286 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Roose | +0.979 |
ibaba | +0.920 |
muzzle | +0.892 |
ICLE | +0.853 |
rils | +0.850 |
Neg logit contributions | |
�� | -0.982 |
aur | -0.960 |
� | -0.864 |
aires | -0.841 |
�� | -0.788 |
Token | California |
Token ID | 3442 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
aur | +11.308 |
�� | +10.583 |
aires | +9.778 |
� | +9.671 |
ARY | +8.928 |
Neg logit contributions | |
ibaba | -13.721 |
Roose | -12.270 |
rils | -12.257 |
mosqu | -12.221 |
smart | -12.208 |
Token | , |
Token ID | 11 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
�� | +4.212 |
aur | +4.019 |
� | +3.649 |
aires | +3.480 |
�� | +3.304 |
Neg logit contributions | |
Roose | -4.695 |
ibaba | -4.360 |
muzzle | -4.266 |
ICLE | -4.105 |
rils | -4.050 |
Token | Berkeley |
Token ID | 14727 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
�� | +7.328 |
aur | +7.183 |
� | +6.435 |
aires | +6.252 |
�� | +5.942 |
Neg logit contributions | |
Roose | -6.572 |
ibaba | -6.135 |
muzzle | -6.040 |
ICLE | -5.672 |
rils | -5.649 |
Token | , |
Token ID | 11 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
Roose | +0.874 |
ibaba | +0.783 |
muzzle | +0.778 |
smart | +0.734 |
ICLE | +0.731 |
Neg logit contributions | |
�� | -0.925 |
aur | -0.866 |
� | -0.799 |
aires | -0.763 |
�� | -0.739 |
Token | who |
Token ID | 508 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
�� | +1.697 |
aur | +1.631 |
� | +1.494 |
aires | +1.418 |
�� | +1.362 |
Neg logit contributions | |
Roose | -1.612 |
ibaba | -1.473 |
muzzle | -1.437 |
ICLE | -1.383 |
azines | -1.365 |
Token | - |
Token ID | 12 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
Roose | +0.856 |
ibaba | +0.777 |
muzzle | +0.751 |
ICLE | +0.716 |
rils | +0.714 |
Neg logit contributions | |
�� | -0.922 |
aur | -0.896 |
� | -0.821 |
aires | -0.780 |
�� | -0.743 |
Token | year |
Token ID | 1941 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
Roose | +1.432 |
ibaba | +1.282 |
muzzle | +1.249 |
ICLE | +1.164 |
azines | +1.163 |
Neg logit contributions | |
�� | -1.766 |
aur | -1.736 |
� | -1.569 |
aires | -1.509 |
�� | -1.448 |
Token | - |
Token ID | 12 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
�� | +0.281 |
aur | +0.264 |
� | +0.246 |
aires | +0.238 |
�� | +0.234 |
Neg logit contributions | |
Roose | -0.197 |
muzzle | -0.171 |
ibaba | -0.170 |
smart | -0.160 |
ICLE | -0.159 |
Token | old |
Token ID | 727 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
aur | +0.400 |
�� | +0.386 |
� | +0.349 |
aires | +0.341 |
ional | +0.322 |
Neg logit contributions | |
Roose | -0.390 |
ibaba | -0.359 |
muzzle | -0.350 |
mosqu | -0.336 |
rils | -0.325 |
Token | University |
Token ID | 2059 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Roose | +2.751 |
muzzle | +2.427 |
ibaba | +2.395 |
smart | +2.243 |
Volume | +2.237 |
Neg logit contributions | |
�� | -3.486 |
aur | -3.250 |
� | -3.035 |
aires | -2.906 |
�� | -2.821 |
Token | of |
Token ID | 286 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
�� | +0.227 |
aur | +0.214 |
� | +0.198 |
aires | +0.191 |
�� | +0.184 |
Neg logit contributions | |
Roose | -0.171 |
ibaba | -0.149 |
muzzle | -0.148 |
smart | -0.139 |
Volume | -0.139 |
Token | Virginia |
Token ID | 6025 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
aur | +10.857 |
�� | +10.119 |
aires | +9.270 |
� | +9.263 |
ARY | +8.498 |
Neg logit contributions | |
ibaba | -13.877 |
rils | -12.673 |
ICLE | -12.484 |
smart | -12.430 |
Roose | -12.403 |
Token | undergrad |
Token ID | 43485 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Roose | +0.227 |
ibaba | +0.197 |
muzzle | +0.195 |
ICLE | +0.181 |
smart | +0.180 |
Neg logit contributions | |
�� | -0.366 |
aur | -0.347 |
� | -0.325 |
aires | -0.313 |
�� | -0.305 |
Token | who |
Token ID | 508 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
Roose | +1.004 |
ibaba | +0.902 |
muzzle | +0.899 |
ICLE | +0.852 |
smart | +0.852 |
Neg logit contributions | |
�� | -1.206 |
aur | -1.113 |
� | -1.027 |
aires | -0.993 |
�� | -0.965 |
Token | served |
Token ID | 4983 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
�� | +2.318 |
aur | +2.226 |
� | +2.057 |
aires | +1.955 |
�� | +1.862 |
Neg logit contributions | |
Roose | -2.355 |
ibaba | -2.182 |
muzzle | -2.106 |
ICLE | -2.045 |
rils | -2.032 |
Token | 18 |
Token ID | 1248 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
�� | +1.134 |
aur | +1.073 |
� | +0.983 |
aires | +0.935 |
�� | +0.904 |
Neg logit contributions | |
Roose | -1.135 |
ibaba | -1.028 |
muzzle | -1.011 |
ICLE | -0.968 |
rils | -0.960 |
Token | , |
Token ID | 11 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
�� | +3.365 |
aur | +3.215 |
� | +2.940 |
aires | +2.764 |
�� | +2.672 |
Neg logit contributions | |
Roose | -3.661 |
ibaba | -3.331 |
muzzle | -3.287 |
ICLE | -3.138 |
azines | -3.124 |
Token | professor |
Token ID | 6240 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
�� | +0.906 |
aur | +0.861 |
� | +0.796 |
aires | +0.757 |
�� | +0.731 |
Neg logit contributions | |
Roose | -0.813 |
ibaba | -0.734 |
muzzle | -0.718 |
smart | -0.684 |
azines | -0.683 |
Token | at |
Token ID | 379 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
Roose | +2.659 |
ibaba | +2.352 |
muzzle | +2.351 |
smart | +2.212 |
bang | +2.186 |
Neg logit contributions | |
�� | -3.536 |
aur | -3.307 |
� | -3.092 |
aires | -2.963 |
�� | -2.913 |
Token | the |
Token ID | 262 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
�� | +2.628 |
aur | +2.487 |
� | +2.276 |
aires | +2.154 |
�� | +2.087 |
Neg logit contributions | |
Roose | -2.558 |
ibaba | -2.359 |
muzzle | -2.307 |
ICLE | -2.202 |
smart | -2.192 |
Token | University |
Token ID | 2059 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
�� | +2.236 |
aur | +2.143 |
� | +1.997 |
aires | +1.901 |
�� | +1.842 |
Neg logit contributions | |
Roose | -1.528 |
ibaba | -1.414 |
muzzle | -1.346 |
ICLE | -1.279 |
smart | -1.274 |
Token | of |
Token ID | 286 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
Roose | +2.563 |
ibaba | +2.331 |
muzzle | +2.294 |
ICLE | +2.174 |
rils | +2.162 |
Neg logit contributions | |
�� | -2.780 |
aur | -2.665 |
� | -2.442 |
aires | -2.340 |
�� | -2.239 |
Token | Calgary |
Token ID | 17842 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
aur | +9.926 |
�� | +8.968 |
aires | +8.289 |
� | +8.158 |
ARY | +7.325 |
Neg logit contributions | |
ibaba | -14.399 |
Roose | -13.244 |
rils | -13.229 |
smart | -13.137 |
ICLE | -12.952 |
Token | , |
Token ID | 11 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
�� | +2.541 |
aur | +2.426 |
� | +2.211 |
aires | +2.107 |
�� | +2.034 |
Neg logit contributions | |
Roose | -2.622 |
ibaba | -2.374 |
muzzle | -2.343 |
ICLE | -2.239 |
rils | -2.214 |
Token | as |
Token ID | 355 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
�� | +2.135 |
aur | +2.042 |
� | +1.854 |
aires | +1.766 |
�� | +1.687 |
Neg logit contributions | |
Roose | -2.195 |
ibaba | -1.991 |
muzzle | -1.953 |
ICLE | -1.861 |
azines | -1.861 |
Token | both |
Token ID | 1111 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
�� | +1.779 |
aur | +1.701 |
� | +1.567 |
aires | +1.489 |
�� | +1.446 |
Neg logit contributions | |
Roose | -1.538 |
ibaba | -1.389 |
muzzle | -1.361 |
ICLE | -1.285 |
smart | -1.283 |
Token | cases |
Token ID | 2663 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
�� | +0.090 |
aur | +0.086 |
� | +0.079 |
aires | +0.075 |
�� | +0.072 |
Neg logit contributions | |
Roose | -0.082 |
ibaba | -0.074 |
muzzle | -0.072 |
ICLE | -0.069 |
smart | -0.069 |
Token | therapy |
Token ID | 9102 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Roose | +1.265 |
ibaba | +1.137 |
muzzle | +1.128 |
ICLE | +1.056 |
smart | +1.054 |
Neg logit contributions | |
�� | -1.405 |
aur | -1.318 |
� | -1.214 |
aires | -1.157 |
�� | -1.131 |
Token | experiment |
Token ID | 6306 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
�� | +1.351 |
aur | +1.303 |
� | +1.197 |
aires | +1.144 |
�� | +1.090 |
Neg logit contributions | |
Roose | -1.266 |
ibaba | -1.160 |
muzzle | -1.120 |
azines | -1.085 |
rils | -1.081 |
Token | at |
Token ID | 379 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
�� | +2.279 |
aur | +2.183 |
� | +1.977 |
aires | +1.881 |
�� | +1.782 |
Neg logit contributions | |
Roose | -2.568 |
ibaba | -2.364 |
muzzle | -2.298 |
azines | -2.210 |
rils | -2.203 |
Token | the |
Token ID | 262 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
Roose | +1.002 |
ibaba | +0.885 |
muzzle | +0.880 |
smart | +0.824 |
ICLE | +0.823 |
Neg logit contributions | |
�� | -1.185 |
aur | -1.115 |
� | -1.034 |
aires | -0.991 |
�� | -0.962 |
Token | University |
Token ID | 2059 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
Roose | +2.138 |
muzzle | +1.853 |
ibaba | +1.816 |
Volume | +1.691 |
ICLE | +1.689 |
Neg logit contributions | |
�� | -3.078 |
aur | -2.906 |
� | -2.702 |
aires | -2.617 |
�� | -2.553 |
Token | of |
Token ID | 286 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Roose | +1.868 |
ibaba | +1.728 |
muzzle | +1.680 |
ICLE | +1.604 |
rils | +1.597 |
Neg logit contributions | |
�� | -1.953 |
aur | -1.892 |
� | -1.715 |
aires | -1.658 |
�� | -1.569 |
Token | Pennsylvania |
Token ID | 9589 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
aur | +11.082 |
�� | +10.463 |
aires | +9.639 |
� | +9.495 |
ARY | +8.742 |
Neg logit contributions | |
ibaba | -12.966 |
Roose | -11.801 |
ICLE | -11.730 |
rils | -11.684 |
smart | -11.609 |
Token | . |
Token ID | 13 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
�� | +1.690 |
aur | +1.596 |
� | +1.458 |
aires | +1.387 |
�� | +1.334 |
Neg logit contributions | |
Roose | -1.789 |
ibaba | -1.628 |
muzzle | -1.604 |
ICLE | -1.528 |
smart | -1.522 |
Token | The |
Token ID | 383 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
�� | +1.576 |
aur | +1.526 |
� | +1.407 |
aires | +1.345 |
�� | +1.283 |
Neg logit contributions | |
Roose | -1.343 |
ibaba | -1.227 |
muzzle | -1.206 |
smart | -1.133 |
ICLE | -1.131 |
Token | event |
Token ID | 1785 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
�� | +0.965 |
aur | +0.925 |
� | +0.851 |
aires | +0.812 |
�� | +0.777 |
Neg logit contributions | |
Roose | -0.861 |
ibaba | -0.791 |
muzzle | -0.763 |
smart | -0.730 |
ICLE | -0.730 |
Token | triggers |
Token ID | 20022 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
�� | +1.678 |
aur | +1.602 |
� | +1.470 |
aires | +1.411 |
�� | +1.343 |
Neg logit contributions | |
Roose | -1.509 |
ibaba | -1.369 |
muzzle | -1.327 |
ICLE | -1.271 |
smart | -1.267 |
Token | os |
Token ID | 418 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
�� | +3.425 |
aur | +3.371 |
� | +3.113 |
aires | +3.028 |
�� | +2.893 |
Neg logit contributions | |
Roose | -1.910 |
ibaba | -1.647 |
muzzle | -1.625 |
ICLE | -1.528 |
smart | -1.469 |
Token | ky |
Token ID | 2584 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
�� | +1.843 |
aur | +1.754 |
� | +1.609 |
aires | +1.534 |
�� | +1.487 |
Neg logit contributions | |
Roose | -1.643 |
ibaba | -1.477 |
muzzle | -1.458 |
ICLE | -1.380 |
smart | -1.374 |
Token | of |
Token ID | 286 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
�� | +0.415 |
aur | +0.386 |
� | +0.356 |
aires | +0.346 |
�� | +0.332 |
Neg logit contributions | |
Roose | -0.384 |
ibaba | -0.341 |
muzzle | -0.339 |
smart | -0.323 |
rils | -0.318 |
Token | the |
Token ID | 262 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
�� | +2.231 |
aur | +2.120 |
� | +1.939 |
aires | +1.831 |
�� | +1.777 |
Neg logit contributions | |
Roose | -2.122 |
ibaba | -1.956 |
muzzle | -1.899 |
ICLE | -1.824 |
smart | -1.812 |
Token | University |
Token ID | 2059 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
�� | +2.093 |
aur | +2.003 |
� | +1.846 |
aires | +1.745 |
�� | +1.691 |
Neg logit contributions | |
Roose | -1.757 |
ibaba | -1.637 |
muzzle | -1.565 |
ICLE | -1.501 |
rils | -1.497 |
Token | of |
Token ID | 286 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
Roose | +1.719 |
ibaba | +1.598 |
muzzle | +1.550 |
ICLE | +1.476 |
rils | +1.475 |
Neg logit contributions | |
�� | -1.776 |
aur | -1.720 |
� | -1.561 |
aires | -1.507 |
�� | -1.428 |
Token | Colorado |
Token ID | 7492 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
aur | +10.359 |
�� | +9.337 |
aires | +8.804 |
� | +8.531 |
ARY | +8.045 |
Neg logit contributions | |
ibaba | -13.867 |
smart | -12.226 |
rils | -12.210 |
Roose | -12.209 |
mosqu | -12.170 |
Token | Boulder |
Token ID | 27437 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
�� | +2.973 |
aur | +2.818 |
� | +2.571 |
aires | +2.444 |
�� | +2.354 |
Neg logit contributions | |
Roose | -3.060 |
ibaba | -2.804 |
muzzle | -2.752 |
ICLE | -2.624 |
smart | -2.608 |
Token | , |
Token ID | 11 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
�� | +2.012 |
aur | +1.895 |
� | +1.738 |
aires | +1.654 |
�� | +1.599 |
Neg logit contributions | |
Roose | -2.021 |
ibaba | -1.827 |
muzzle | -1.806 |
ICLE | -1.713 |
smart | -1.708 |
Token | says |
Token ID | 1139 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
�� | +1.589 |
aur | +1.526 |
� | +1.396 |
aires | +1.323 |
�� | +1.279 |
Neg logit contributions | |
Roose | -1.526 |
ibaba | -1.400 |
muzzle | -1.366 |
ICLE | -1.312 |
smart | -1.295 |
Token | that |
Token ID | 326 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
Roose | +2.351 |
muzzle | +2.099 |
ibaba | +2.079 |
Volume | +1.956 |
bang | +1.955 |
Neg logit contributions | |
�� | -2.755 |
aur | -2.578 |
� | -2.392 |
aires | -2.292 |
�� | -2.235 |
Token | glac |
Token ID | 25169 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
�� | +0.892 |
aur | +0.851 |
� | +0.774 |
aires | +0.733 |
�� | +0.711 |
Neg logit contributions | |
Roose | -0.856 |
ibaba | -0.788 |
muzzle | -0.758 |
rils | -0.731 |
azines | -0.727 |
Token | iologist |
Token ID | 31599 |
Feature activation | +9.74e-02 |
Pos logit contributions | |
�� | +2.044 |
aur | +2.032 |
� | +1.852 |
aires | +1.805 |
ional | +1.731 |
Neg logit contributions | |
Roose | -1.398 |
ibaba | -1.204 |
muzzle | -1.193 |
rils | -1.107 |
ICLE | -1.091 |
Token | at |
Token ID | 379 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
�� | +0.797 |
aur | +0.743 |
� | +0.680 |
aires | +0.647 |
�� | +0.624 |
Neg logit contributions | |
Roose | -0.884 |
ibaba | -0.800 |
muzzle | -0.796 |
ICLE | -0.756 |
smart | -0.755 |
Token | the |
Token ID | 262 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
�� | +2.350 |
aur | +2.239 |
� | +2.055 |
aires | +1.943 |
�� | +1.886 |
Neg logit contributions | |
Roose | -2.186 |
ibaba | -1.997 |
muzzle | -1.964 |
ICLE | -1.865 |
smart | -1.858 |
Token | University |
Token ID | 2059 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
�� | +2.096 |
aur | +2.019 |
� | +1.874 |
aires | +1.778 |
�� | +1.726 |
Neg logit contributions | |
Roose | -1.500 |
ibaba | -1.383 |
muzzle | -1.328 |
smart | -1.254 |
ICLE | -1.253 |
Token | of |
Token ID | 286 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
Roose | +2.111 |
ibaba | +1.944 |
muzzle | +1.901 |
ICLE | +1.806 |
rils | +1.797 |
Neg logit contributions | |
�� | -2.256 |
aur | -2.175 |
� | -1.985 |
aires | -1.902 |
�� | -1.816 |
Token | Washington |
Token ID | 2669 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
aur | +11.334 |
�� | +10.163 |
� | +9.625 |
aires | +9.535 |
ARY | +8.751 |
Neg logit contributions | |
ibaba | -12.790 |
mosqu | -11.510 |
smart | -11.447 |
rils | -11.277 |
Roose | -11.219 |
Token | who |
Token ID | 508 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
�� | +3.781 |
aur | +3.634 |
� | +3.281 |
aires | +3.121 |
�� | +3.009 |
Neg logit contributions | |
Roose | -3.994 |
ibaba | -3.655 |
muzzle | -3.603 |
ICLE | -3.435 |
rils | -3.409 |
Token | has |
Token ID | 468 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
�� | +1.665 |
aur | +1.590 |
� | +1.463 |
aires | +1.382 |
�� | +1.333 |
Neg logit contributions | |
Roose | -1.629 |
ibaba | -1.485 |
muzzle | -1.459 |
ICLE | -1.393 |
azines | -1.380 |
Token | been |
Token ID | 587 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
Roose | +1.460 |
ibaba | +1.261 |
muzzle | +1.257 |
smart | +1.163 |
ICLE | +1.160 |
Neg logit contributions | |
�� | -2.329 |
aur | -2.204 |
� | -2.055 |
aires | -1.986 |
�� | -1.928 |
Token | studying |
Token ID | 11065 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
�� | +0.740 |
aur | +0.712 |
� | +0.664 |
aires | +0.632 |
�� | +0.619 |
Neg logit contributions | |
Roose | -0.513 |
ibaba | -0.459 |
muzzle | -0.446 |
ICLE | -0.423 |
azines | -0.419 |
Token | with |
Token ID | 351 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
�� | +0.478 |
aur | +0.459 |
� | +0.417 |
aires | +0.396 |
�� | +0.381 |
Neg logit contributions | |
Roose | -0.492 |
ibaba | -0.456 |
muzzle | -0.437 |
ICLE | -0.422 |
rils | -0.420 |
Token | bad |
Token ID | 2089 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
Roose | +5.299 |
muzzle | +4.772 |
ibaba | +4.526 |
bang | +4.429 |
disappro | +4.424 |
Neg logit contributions | |
�� | -6.373 |
aur | -5.982 |
� | -5.520 |
aires | -5.313 |
�� | -5.164 |
Token | manners |
Token ID | 33558 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
Roose | +3.575 |
muzzle | +3.190 |
ibaba | +3.129 |
bang | +2.978 |
azines | +2.972 |
Neg logit contributions | |
�� | -4.413 |
aur | -4.134 |
� | -3.843 |
aires | -3.686 |
�� | -3.596 |
Token | . |
Token ID | 13 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
Roose | +0.276 |
ibaba | +0.250 |
muzzle | +0.247 |
ICLE | +0.235 |
smart | +0.234 |
Neg logit contributions | |
�� | -0.271 |
aur | -0.256 |
� | -0.234 |
aires | -0.223 |
�� | -0.215 |
Token | - |
Token ID | 532 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
Roose | +0.336 |
ibaba | +0.306 |
muzzle | +0.298 |
ICLE | +0.282 |
rils | +0.282 |
Neg logit contributions | |
�� | -0.375 |
aur | -0.360 |
� | -0.329 |
aires | -0.316 |
�� | -0.304 |
Token | January |
Token ID | 3269 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
aur | +2.986 |
�� | +2.850 |
aires | +2.442 |
� | +2.390 |
ional | +2.236 |
Neg logit contributions | |
Roose | -4.210 |
ibaba | -3.884 |
rils | -3.728 |
muzzle | -3.661 |
azines | -3.614 |
Token | 4 |
Token ID | 604 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
aur | +3.993 |
aires | +3.410 |
�� | +2.312 |
ARY | +2.076 |
� | +1.671 |
Neg logit contributions | |
mosqu | -20.923 |
ibaba | -19.806 |
Roose | -19.242 |
ModLoader | -19.093 |
enthusi | -18.986 |
Token | , |
Token ID | 11 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
Roose | +3.464 |
ibaba | +3.147 |
muzzle | +3.100 |
ICLE | +2.945 |
smart | +2.930 |
Neg logit contributions | |
�� | -3.291 |
aur | -3.112 |
� | -2.841 |
aires | -2.706 |
�� | -2.602 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
�� | +2.039 |
aur | +2.038 |
aires | +1.800 |
� | +1.776 |
ional | +1.693 |
Neg logit contributions | |
Roose | -2.686 |
ibaba | -2.608 |
mosqu | -2.510 |
muzzle | -2.411 |
rils | -2.384 |
Token | User |
Token ID | 12982 |
Feature activation | -8.57e-01 |
Pos logit contributions | |
Roose | +1.002 |
ibaba | +0.876 |
muzzle | +0.863 |
ICLE | +0.805 |
smart | +0.801 |
Neg logit contributions | |
�� | -1.595 |
aur | -1.517 |
� | -1.418 |
aires | -1.363 |
�� | -1.328 |
Token | reviews |
Token ID | 8088 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
Roose | +4.330 |
muzzle | +3.655 |
ibaba | +3.575 |
azines | +3.269 |
bang | +3.260 |
Neg logit contributions | |
�� | -10.413 |
aur | -9.737 |
� | -9.290 |
aires | -8.879 |
�� | -8.852 |
Token | yoga |
Token ID | 20351 |
Feature activation | +8.24e-03 |
Pos logit contributions | |
�� | +1.692 |
aur | +1.612 |
� | +1.493 |
aires | +1.417 |
�� | +1.377 |
Neg logit contributions | |
Roose | -1.470 |
ibaba | -1.339 |
muzzle | -1.308 |
ICLE | -1.245 |
smart | -1.239 |
Token | classes |
Token ID | 6097 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
�� | +0.073 |
aur | +0.069 |
� | +0.063 |
aires | +0.060 |
�� | +0.058 |
Neg logit contributions | |
Roose | -0.069 |
ibaba | -0.062 |
muzzle | -0.061 |
ICLE | -0.058 |
smart | -0.058 |
Token | at |
Token ID | 379 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
�� | +1.987 |
aur | +1.887 |
� | +1.734 |
aires | +1.622 |
�� | +1.575 |
Neg logit contributions | |
Roose | -2.191 |
ibaba | -2.034 |
muzzle | -1.964 |
smart | -1.882 |
ICLE | -1.882 |
Token | the |
Token ID | 262 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
�� | +1.652 |
aur | +1.567 |
� | +1.447 |
aires | +1.377 |
�� | +1.337 |
Neg logit contributions | |
Roose | -1.432 |
ibaba | -1.302 |
muzzle | -1.275 |
ICLE | -1.211 |
rils | -1.202 |
Token | University |
Token ID | 2059 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
Roose | +1.261 |
muzzle | +1.097 |
ibaba | +1.094 |
smart | +1.019 |
ICLE | +1.017 |
Neg logit contributions | |
�� | -1.695 |
aur | -1.604 |
� | -1.488 |
aires | -1.436 |
�� | -1.394 |
Token | of |
Token ID | 286 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Roose | +0.796 |
ibaba | +0.733 |
muzzle | +0.716 |
ICLE | +0.683 |
rils | +0.679 |
Neg logit contributions | |
�� | -0.839 |
aur | -0.808 |
� | -0.738 |
aires | -0.710 |
�� | -0.678 |
Token | Ottawa |
Token ID | 14074 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
aur | +10.997 |
�� | +10.617 |
� | +9.638 |
aires | +9.628 |
ARY | +8.703 |
Neg logit contributions | |
ibaba | -12.171 |
Roose | -11.394 |
rils | -11.083 |
muzzle | -10.925 |
ICLE | -10.925 |
Token | in |
Token ID | 287 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
�� | +4.544 |
aur | +4.486 |
� | +4.111 |
aires | +3.927 |
ional | +3.732 |
Neg logit contributions | |
Roose | -4.314 |
ibaba | -4.076 |
muzzle | -3.801 |
ICLE | -3.742 |
smart | -3.696 |
Token | Canada |
Token ID | 3340 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
aur | +5.388 |
�� | +5.378 |
� | +4.720 |
aires | +4.536 |
equival | +4.263 |
Neg logit contributions | |
Roose | -6.813 |
ibaba | -6.485 |
muzzle | -5.976 |
ICLE | -5.966 |
rils | -5.880 |
Token | were |
Token ID | 547 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
�� | +0.201 |
aur | +0.191 |
� | +0.177 |
aires | +0.169 |
�� | +0.164 |
Neg logit contributions | |
Roose | -0.170 |
ibaba | -0.152 |
muzzle | -0.149 |
ICLE | -0.141 |
smart | -0.141 |
Token | abruptly |
Token ID | 25891 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
�� | +1.513 |
aur | +1.448 |
� | +1.348 |
aires | +1.278 |
�� | +1.242 |
Neg logit contributions | |
Roose | -1.144 |
ibaba | -1.037 |
muzzle | -0.988 |
ICLE | -0.956 |
smart | -0.946 |
Token | , |
Token ID | 11 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
Roose | +0.486 |
ibaba | +0.439 |
muzzle | +0.433 |
ICLE | +0.411 |
smart | +0.411 |
Neg logit contributions | |
�� | -0.513 |
aur | -0.477 |
� | -0.439 |
aires | -0.420 |
�� | -0.408 |
Token | according |
Token ID | 1864 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Roose | +2.366 |
ibaba | +2.109 |
muzzle | +2.106 |
smart | +1.971 |
ICLE | +1.963 |
Neg logit contributions | |
�� | -2.798 |
aur | -2.619 |
� | -2.412 |
aires | -2.327 |
�� | -2.257 |
Token | to |
Token ID | 284 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
aur | +1.232 |
�� | +1.231 |
� | +1.129 |
aires | +1.058 |
ional | +1.001 |
Neg logit contributions | |
Roose | -1.465 |
ibaba | -1.362 |
muzzle | -1.309 |
azines | -1.261 |
smart | -1.254 |
Token | the |
Token ID | 262 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
�� | +2.857 |
aur | +2.713 |
� | +2.514 |
aires | +2.364 |
�� | +2.310 |
Neg logit contributions | |
Roose | -2.759 |
ibaba | -2.499 |
muzzle | -2.459 |
azines | -2.340 |
ICLE | -2.339 |
Token | University |
Token ID | 2059 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
�� | +3.954 |
aur | +3.799 |
� | +3.505 |
aires | +3.340 |
�� | +3.199 |
Neg logit contributions | |
Roose | -3.668 |
ibaba | -3.385 |
muzzle | -3.267 |
rils | -3.166 |
azines | -3.153 |
Token | of |
Token ID | 286 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Roose | +3.202 |
ibaba | +2.805 |
muzzle | +2.742 |
Volume | +2.598 |
smart | +2.589 |
Neg logit contributions | |
�� | -4.258 |
aur | -3.992 |
� | -3.706 |
aires | -3.594 |
�� | -3.452 |
Token | Hawai |
Token ID | 10847 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
aur | +9.360 |
�� | +8.808 |
� | +7.940 |
aires | +7.755 |
�� | +7.016 |
Neg logit contributions | |
ibaba | -13.672 |
Roose | -13.306 |
rils | -12.670 |
muzzle | -12.648 |
smart | -12.573 |
Token | � |
Token ID | 447 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
aur | +6.264 |
�� | +6.165 |
� | +5.871 |
aires | +5.376 |
�� | +5.298 |
Neg logit contributions | |
Roose | -5.608 |
muzzle | -5.095 |
ibaba | -4.961 |
rils | -4.805 |
ICLE | -4.801 |
Token | � |
Token ID | 247 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
�� | +0.414 |
aur | +0.379 |
� | +0.330 |
aires | +0.308 |
�� | +0.289 |
Neg logit contributions | |
Roose | -0.794 |
ibaba | -0.736 |
muzzle | -0.729 |
ICLE | -0.702 |
smart | -0.700 |
Token | i |
Token ID | 72 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
�� | +3.312 |
aur | +3.227 |
� | +2.948 |
aires | +2.824 |
�� | +2.696 |
Neg logit contributions | |
Roose | -2.907 |
ibaba | -2.586 |
muzzle | -2.573 |
ICLE | -2.411 |
rils | -2.384 |
Token | , |
Token ID | 11 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
�� | +4.381 |
aur | +4.241 |
� | +3.873 |
aires | +3.670 |
�� | +3.557 |
Neg logit contributions | |
Roose | -4.358 |
ibaba | -3.949 |
muzzle | -3.909 |
ICLE | -3.733 |
rils | -3.717 |
Token | m |
Token ID | 76 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
Roose | +1.881 |
ibaba | +1.741 |
muzzle | +1.706 |
ICLE | +1.637 |
azines | +1.624 |
Neg logit contributions | |
�� | -1.222 |
aur | -1.192 |
� | -1.035 |
aires | -0.992 |
ional | -0.908 |
Token | . |
Token ID | 13 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
Roose | +3.003 |
muzzle | +2.705 |
ibaba | +2.647 |
smart | +2.532 |
bang | +2.521 |
Neg logit contributions | |
�� | -4.564 |
aur | -4.104 |
� | -3.856 |
�� | -3.700 |
aires | -3.612 |
Token | at |
Token ID | 379 |
Feature activation | +7.43e-03 |
Pos logit contributions | |
Roose | +2.845 |
muzzle | +2.521 |
ibaba | +2.489 |
smart | +2.356 |
Volume | +2.341 |
Neg logit contributions | |
�� | -3.238 |
aur | -3.034 |
� | -2.793 |
aires | -2.660 |
�� | -2.629 |
Token | the |
Token ID | 262 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
�� | +0.068 |
aur | +0.064 |
� | +0.059 |
aires | +0.056 |
�� | +0.055 |
Neg logit contributions | |
Roose | -0.060 |
ibaba | -0.054 |
muzzle | -0.053 |
ICLE | -0.050 |
smart | -0.050 |
Token | University |
Token ID | 2059 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
Roose | +2.860 |
muzzle | +2.485 |
ibaba | +2.397 |
bang | +2.309 |
Volume | +2.304 |
Neg logit contributions | |
�� | -3.407 |
aur | -3.228 |
� | -2.954 |
aires | -2.861 |
�� | -2.787 |
Token | of |
Token ID | 286 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Roose | +1.067 |
ibaba | +0.896 |
muzzle | +0.896 |
Volume | +0.850 |
smart | +0.847 |
Neg logit contributions | |
�� | -1.467 |
aur | -1.386 |
� | -1.283 |
�� | -1.220 |
aires | -1.216 |
Token | Cincinnati |
Token ID | 16137 |
Feature activation | +6.66e-02 |
Pos logit contributions | |
aur | +10.818 |
�� | +10.409 |
� | +9.391 |
aires | +9.340 |
�� | +8.454 |
Neg logit contributions | |
ibaba | -12.396 |
muzzle | -11.428 |
Roose | -11.348 |
smart | -11.318 |
ICLE | -11.215 |
Token | Medical |
Token ID | 8366 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
�� | +0.595 |
aur | +0.563 |
� | +0.519 |
aires | +0.495 |
�� | +0.478 |
Neg logit contributions | |
Roose | -0.549 |
ibaba | -0.496 |
muzzle | -0.489 |
ICLE | -0.463 |
smart | -0.461 |
Token | Center |
Token ID | 3337 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
�� | +1.639 |
aur | +1.568 |
� | +1.465 |
aires | +1.383 |
�� | +1.336 |
Neg logit contributions | |
Roose | -1.351 |
ibaba | -1.232 |
muzzle | -1.197 |
smart | -1.130 |
ICLE | -1.128 |
Token | , |
Token ID | 11 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
�� | +2.831 |
aur | +2.701 |
� | +2.474 |
aires | +2.358 |
�� | +2.238 |
Neg logit contributions | |
Roose | -3.085 |
ibaba | -2.880 |
muzzle | -2.796 |
azines | -2.671 |
ICLE | -2.665 |
Token | according |
Token ID | 1864 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
�� | +2.092 |
aur | +1.988 |
� | +1.845 |
aires | +1.736 |
�� | +1.677 |
Neg logit contributions | |
Roose | -2.053 |
ibaba | -1.926 |
muzzle | -1.856 |
smart | -1.781 |
azines | -1.778 |
Token | then |
Token ID | 788 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
�� | +2.368 |
aur | +2.266 |
� | +2.108 |
aires | +1.989 |
�� | +1.915 |
Neg logit contributions | |
Roose | -2.287 |
ibaba | -2.098 |
muzzle | -2.025 |
ICLE | -1.967 |
rils | -1.959 |
Token | went |
Token ID | 1816 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
�� | +2.599 |
aur | +2.512 |
� | +2.337 |
aires | +2.241 |
ional | +2.139 |
Neg logit contributions | |
Roose | -2.377 |
ibaba | -2.209 |
muzzle | -2.097 |
ICLE | -2.081 |
azines | -2.051 |
Token | to |
Token ID | 284 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
�� | +1.481 |
aur | +1.410 |
� | +1.305 |
aires | +1.234 |
�� | +1.198 |
Neg logit contributions | |
Roose | -1.464 |
ibaba | -1.323 |
muzzle | -1.295 |
ICLE | -1.245 |
rils | -1.238 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
�� | +2.516 |
aur | +2.391 |
� | +2.209 |
aires | +2.101 |
�� | +2.040 |
Neg logit contributions | |
Roose | -2.255 |
ibaba | -2.028 |
muzzle | -1.989 |
ICLE | -1.923 |
rils | -1.909 |
Token | University |
Token ID | 2059 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
�� | +1.020 |
aur | +0.965 |
� | +0.890 |
aires | +0.845 |
�� | +0.821 |
Neg logit contributions | |
Roose | -0.953 |
ibaba | -0.866 |
muzzle | -0.844 |
ICLE | -0.809 |
smart | -0.808 |
Token | of |
Token ID | 286 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Roose | +1.067 |
ibaba | +0.985 |
muzzle | +0.955 |
ICLE | +0.919 |
rils | +0.915 |
Neg logit contributions | |
�� | -1.078 |
aur | -1.045 |
� | -0.950 |
aires | -0.913 |
�� | -0.871 |
Token | Kentucky |
Token ID | 11867 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
aur | +10.702 |
�� | +10.293 |
aires | +9.429 |
� | +9.286 |
ARY | +8.664 |
Neg logit contributions | |
ibaba | -12.119 |
Roose | -11.447 |
rils | -11.101 |
smart | -10.939 |
muzzle | -10.750 |
Token | on |
Token ID | 319 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
�� | +0.748 |
aur | +0.705 |
� | +0.648 |
aires | +0.617 |
�� | +0.597 |
Neg logit contributions | |
Roose | -0.728 |
ibaba | -0.657 |
muzzle | -0.649 |
ICLE | -0.615 |
smart | -0.613 |
Token | a |
Token ID | 257 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
�� | +3.493 |
aur | +3.340 |
� | +3.069 |
aires | +2.931 |
�� | +2.814 |
Neg logit contributions | |
Roose | -3.244 |
ibaba | -2.903 |
muzzle | -2.860 |
rils | -2.748 |
azines | -2.741 |
Token | soccer |
Token ID | 11783 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
�� | +4.318 |
aur | +4.124 |
� | +3.767 |
aires | +3.561 |
�� | +3.491 |
Neg logit contributions | |
Roose | -4.261 |
ibaba | -3.880 |
muzzle | -3.747 |
rils | -3.729 |
azines | -3.689 |
Token | scholarship |
Token ID | 18652 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
�� | +3.278 |
aur | +3.092 |
� | +2.839 |
aires | +2.704 |
�� | +2.618 |
Neg logit contributions | |
Roose | -3.332 |
ibaba | -3.012 |
muzzle | -2.953 |
rils | -2.842 |
azines | -2.829 |
Token | eggs |
Token ID | 9653 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
�� | +0.428 |
aur | +0.399 |
� | +0.368 |
aires | +0.350 |
�� | +0.340 |
Neg logit contributions | |
Roose | -0.429 |
ibaba | -0.387 |
muzzle | -0.384 |
ICLE | -0.364 |
smart | -0.362 |
Token | ) |
Token ID | 8 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
�� | +4.311 |
aur | +4.245 |
� | +3.719 |
aires | +3.484 |
�� | +3.315 |
Neg logit contributions | |
Roose | -5.658 |
ibaba | -5.146 |
muzzle | -5.075 |
azines | -4.876 |
ICLE | -4.854 |
Token | and |
Token ID | 290 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
Roose | +0.385 |
ibaba | +0.346 |
muzzle | +0.342 |
ICLE | +0.324 |
smart | +0.323 |
Neg logit contributions | |
�� | -0.420 |
aur | -0.395 |
� | -0.364 |
aires | -0.348 |
�� | -0.337 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
�� | +0.216 |
aur | +0.206 |
� | +0.189 |
aires | +0.180 |
�� | +0.175 |
Neg logit contributions | |
Roose | -0.196 |
ibaba | -0.179 |
muzzle | -0.175 |
ICLE | -0.166 |
smart | -0.166 |
Token | University |
Token ID | 2059 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
Roose | +0.885 |
ibaba | +0.776 |
muzzle | +0.772 |
ICLE | +0.722 |
smart | +0.717 |
Neg logit contributions | |
�� | -1.201 |
aur | -1.135 |
� | -1.057 |
aires | -1.016 |
�� | -0.986 |
Token | of |
Token ID | 286 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Roose | +1.471 |
ibaba | +1.334 |
muzzle | +1.313 |
ICLE | +1.246 |
smart | +1.238 |
Neg logit contributions | |
�� | -1.633 |
aur | -1.556 |
� | -1.427 |
aires | -1.369 |
�� | -1.321 |
Token | Calgary |
Token ID | 17842 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
aur | +10.336 |
�� | +9.694 |
aires | +8.855 |
� | +8.758 |
ARY | +8.029 |
Neg logit contributions | |
ibaba | -12.660 |
Roose | -11.754 |
rils | -11.514 |
smart | -11.444 |
muzzle | -11.388 |
Token | ( |
Token ID | 357 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
�� | +3.963 |
aur | +3.896 |
� | +3.465 |
aires | +3.301 |
�� | +3.192 |
Neg logit contributions | |
Roose | -4.558 |
ibaba | -4.213 |
muzzle | -4.097 |
ICLE | -3.939 |
azines | -3.918 |
Token | which |
Token ID | 4758 |
Feature activation | +6.21e-02 |
Pos logit contributions | |
Roose | +0.318 |
ibaba | +0.283 |
muzzle | +0.280 |
ICLE | +0.264 |
smart | +0.263 |
Neg logit contributions | |
�� | -0.404 |
aur | -0.382 |
� | -0.355 |
aires | -0.340 |
�� | -0.330 |
Token | has |
Token ID | 468 |
Feature activation | +2.81e-03 |
Pos logit contributions | |
�� | +0.616 |
aur | +0.586 |
� | +0.545 |
aires | +0.522 |
�� | +0.506 |
Neg logit contributions | |
Roose | -0.450 |
ibaba | -0.404 |
muzzle | -0.394 |
ICLE | -0.372 |
smart | -0.369 |
Token | the |
Token ID | 262 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
�� | +0.025 |
aur | +0.024 |
� | +0.022 |
aires | +0.021 |
�� | +0.021 |
Neg logit contributions | |
Roose | -0.023 |
ibaba | -0.021 |
muzzle | -0.020 |
ICLE | -0.019 |
smart | -0.019 |
Token | clinical |
Token ID | 8668 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
Roose | +0.369 |
ibaba | +0.332 |
muzzle | +0.324 |
ICLE | +0.307 |
rils | +0.307 |
Neg logit contributions | |
�� | -0.447 |
aur | -0.426 |
� | -0.393 |
aires | -0.376 |
�� | -0.364 |
Token | researcher |
Token ID | 13453 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Roose | +2.593 |
muzzle | +2.309 |
ibaba | +2.274 |
smart | +2.119 |
ICLE | +2.111 |
Neg logit contributions | |
�� | -3.948 |
aur | -3.674 |
� | -3.471 |
aires | -3.354 |
�� | -3.259 |
Token | at |
Token ID | 379 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
�� | +1.366 |
aur | +1.294 |
� | +1.175 |
aires | +1.114 |
�� | +1.074 |
Neg logit contributions | |
Roose | -1.527 |
ibaba | -1.393 |
muzzle | -1.369 |
ICLE | -1.308 |
rils | -1.301 |
Token | the |
Token ID | 262 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
�� | +1.630 |
aur | +1.547 |
� | +1.423 |
aires | +1.356 |
�� | +1.315 |
Neg logit contributions | |
Roose | -1.458 |
ibaba | -1.320 |
muzzle | -1.299 |
ICLE | -1.233 |
smart | -1.225 |
Token | University |
Token ID | 2059 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
�� | +0.224 |
aur | +0.213 |
� | +0.197 |
aires | +0.188 |
�� | +0.183 |
Neg logit contributions | |
Roose | -0.176 |
ibaba | -0.158 |
muzzle | -0.155 |
ICLE | -0.146 |
smart | -0.145 |
Token | of |
Token ID | 286 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
Roose | +2.545 |
ibaba | +2.302 |
muzzle | +2.266 |
ICLE | +2.147 |
rils | +2.132 |
Neg logit contributions | |
�� | -2.857 |
aur | -2.722 |
� | -2.500 |
aires | -2.399 |
�� | -2.309 |
Token | Toronto |
Token ID | 6586 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
aur | +10.495 |
�� | +9.799 |
aires | +9.165 |
� | +8.994 |
ARY | +8.172 |
Neg logit contributions | |
ibaba | -12.302 |
Roose | -11.401 |
rils | -11.202 |
smart | -11.105 |
ICLE | -11.098 |
Token | , |
Token ID | 11 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
�� | +2.730 |
aur | +2.587 |
� | +2.363 |
aires | +2.250 |
�� | +2.166 |
Neg logit contributions | |
Roose | -2.832 |
ibaba | -2.566 |
muzzle | -2.529 |
ICLE | -2.416 |
smart | -2.396 |
Token | to |
Token ID | 284 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
�� | +0.731 |
aur | +0.679 |
� | +0.629 |
aires | +0.603 |
�� | +0.595 |
Neg logit contributions | |
Roose | -0.583 |
ibaba | -0.523 |
muzzle | -0.522 |
smart | -0.486 |
azines | -0.486 |
Token | withdraw |
Token ID | 8399 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
�� | +2.164 |
aur | +2.138 |
� | +1.933 |
aires | +1.867 |
ional | +1.726 |
Neg logit contributions | |
Roose | -2.333 |
ibaba | -2.118 |
ICLE | -2.043 |
muzzle | -1.992 |
rils | -1.983 |
Token | a |
Token ID | 257 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
Roose | +1.657 |
ibaba | +1.467 |
muzzle | +1.466 |
ICLE | +1.364 |
smart | +1.362 |
Neg logit contributions | |
�� | -2.158 |
aur | -2.017 |
� | -1.884 |
aires | -1.813 |
�� | -1.769 |
Token | of |
Token ID | 286 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
Roose | +3.580 |
muzzle | +3.161 |
ibaba | +3.160 |
Volume | +3.020 |
smart | +3.013 |
Neg logit contributions | |
�� | -4.639 |
aur | -4.273 |
� | -4.058 |
aires | -3.874 |
�� | -3.748 |
Token | Economics |
Token ID | 18963 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Roose | +1.872 |
ibaba | +1.659 |
muzzle | +1.648 |
ICLE | +1.545 |
Volume | +1.541 |
Neg logit contributions | |
�� | -2.239 |
aur | -2.111 |
� | -1.959 |
aires | -1.878 |
�� | -1.821 |
Token | at |
Token ID | 379 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
�� | +1.068 |
aur | +1.010 |
� | +0.918 |
aires | +0.870 |
�� | +0.840 |
Neg logit contributions | |
Roose | -1.203 |
ibaba | -1.097 |
muzzle | -1.080 |
ICLE | -1.029 |
smart | -1.024 |
Token | the |
Token ID | 262 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
�� | +1.464 |
aur | +1.387 |
� | +1.279 |
aires | +1.219 |
�� | +1.185 |
Neg logit contributions | |
Roose | -1.279 |
ibaba | -1.158 |
muzzle | -1.142 |
ICLE | -1.081 |
smart | -1.073 |
Token | University |
Token ID | 2059 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
�� | +0.381 |
aur | +0.363 |
� | +0.340 |
aires | +0.326 |
�� | +0.318 |
Neg logit contributions | |
Roose | -0.242 |
ibaba | -0.214 |
muzzle | -0.210 |
ICLE | -0.196 |
smart | -0.195 |
Token | of |
Token ID | 286 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
Roose | +2.762 |
ibaba | +2.529 |
muzzle | +2.476 |
ICLE | +2.344 |
rils | +2.341 |
Neg logit contributions | |
�� | -2.977 |
aur | -2.866 |
� | -2.616 |
aires | -2.515 |
�� | -2.406 |
Token | South |
Token ID | 2520 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
aur | +10.212 |
�� | +9.590 |
aires | +8.935 |
� | +8.773 |
ARY | +8.155 |
Neg logit contributions | |
ibaba | -12.457 |
rils | -11.333 |
Roose | -11.308 |
mosqu | -11.287 |
muzzle | -11.107 |
Token | Australia |
Token ID | 4505 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
�� | +2.594 |
aur | +2.491 |
� | +2.283 |
aires | +2.141 |
�� | +2.059 |
Neg logit contributions | |
Roose | -2.784 |
ibaba | -2.574 |
muzzle | -2.531 |
ICLE | -2.426 |
rils | -2.395 |
Token | , |
Token ID | 11 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
�� | +1.017 |
aur | +0.960 |
� | +0.880 |
aires | +0.838 |
�� | +0.811 |
Neg logit contributions | |
Roose | -1.013 |
ibaba | -0.915 |
muzzle | -0.906 |
ICLE | -0.859 |
smart | -0.855 |
Token | an |
Token ID | 281 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
�� | +2.191 |
aur | +2.136 |
� | +1.925 |
aires | +1.836 |
�� | +1.765 |
Neg logit contributions | |
Roose | -2.183 |
ibaba | -1.992 |
muzzle | -1.939 |
ICLE | -1.859 |
rils | -1.851 |
Token | Emer |
Token ID | 10320 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
Roose | +1.614 |
ibaba | +1.449 |
muzzle | +1.435 |
ICLE | +1.356 |
smart | +1.354 |
Neg logit contributions | |
�� | -1.713 |
aur | -1.601 |
� | -1.475 |
aires | -1.411 |
�� | -1.368 |
Token | ell |
Token ID | 695 |
Feature activation | -7.53e-02 |
Pos logit contributions | |
aur | +5.507 |
�� | +5.395 |
� | +4.819 |
aires | +4.794 |
ional | +4.460 |
Neg logit contributions | |
Roose | -4.538 |
muzzle | -4.044 |
ibaba | -4.014 |
ICLE | -3.862 |
bang | -3.729 |
Token | ert |
Token ID | 861 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
Roose | +0.604 |
ibaba | +0.542 |
muzzle | +0.533 |
smart | +0.505 |
ICLE | +0.503 |
Neg logit contributions | |
�� | -0.697 |
aur | -0.649 |
� | -0.605 |
aires | -0.575 |
�� | -0.563 |
Token | of |
Token ID | 286 |
Feature activation | +3.91e-03 |
Pos logit contributions | |
�� | +2.555 |
aur | +2.456 |
� | +2.222 |
aires | +2.125 |
�� | +2.034 |
Neg logit contributions | |
Roose | -2.561 |
ibaba | -2.332 |
muzzle | -2.298 |
ICLE | -2.185 |
azines | -2.171 |
Token | the |
Token ID | 262 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
�� | +0.035 |
aur | +0.033 |
� | +0.031 |
aires | +0.029 |
�� | +0.028 |
Neg logit contributions | |
Roose | -0.032 |
ibaba | -0.029 |
muzzle | -0.028 |
ICLE | -0.027 |
smart | -0.027 |
Token | University |
Token ID | 2059 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
�� | +0.229 |
aur | +0.218 |
� | +0.202 |
aires | +0.193 |
�� | +0.187 |
Neg logit contributions | |
Roose | -0.181 |
ibaba | -0.164 |
muzzle | -0.160 |
ICLE | -0.151 |
smart | -0.150 |
Token | of |
Token ID | 286 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
Roose | +2.157 |
ibaba | +1.978 |
muzzle | +1.937 |
ICLE | +1.843 |
rils | +1.831 |
Neg logit contributions | |
�� | -2.324 |
aur | -2.237 |
� | -2.040 |
aires | -1.966 |
�� | -1.873 |
Token | Gu |
Token ID | 1962 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
aur | +9.944 |
�� | +9.140 |
aires | +8.533 |
� | +8.429 |
ARY | +7.773 |
Neg logit contributions | |
ibaba | -12.711 |
ICLE | -11.517 |
Roose | -11.509 |
smart | -11.490 |
rils | -11.464 |
Token | el |
Token ID | 417 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
�� | +0.360 |
aur | +0.335 |
� | +0.307 |
aires | +0.289 |
�� | +0.280 |
Neg logit contributions | |
Roose | -0.411 |
ibaba | -0.375 |
muzzle | -0.370 |
smart | -0.352 |
ICLE | -0.352 |
Token | ph |
Token ID | 746 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
Roose | +0.969 |
ibaba | +0.818 |
muzzle | +0.795 |
smart | +0.700 |
azines | +0.699 |
Neg logit contributions | |
�� | -3.072 |
aur | -2.877 |
� | -2.763 |
aires | -2.648 |
�� | -2.643 |
Token | in |
Token ID | 287 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
�� | +3.054 |
aur | +2.929 |
� | +2.648 |
aires | +2.543 |
�� | +2.422 |
Neg logit contributions | |
Roose | -3.261 |
ibaba | -2.962 |
muzzle | -2.915 |
ICLE | -2.780 |
smart | -2.755 |
Token | Ontario |
Token ID | 10553 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
�� | +2.703 |
aur | +2.635 |
� | +2.353 |
aires | +2.267 |
�� | +2.123 |
Neg logit contributions | |
Roose | -2.989 |
ibaba | -2.752 |
muzzle | -2.672 |
ICLE | -2.576 |
smart | -2.554 |
Token | battalion |
Token ID | 36328 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
Roose | +1.333 |
muzzle | +1.197 |
ibaba | +1.185 |
smart | +1.114 |
bang | +1.109 |
Neg logit contributions | |
�� | -1.576 |
aur | -1.473 |
� | -1.364 |
aires | -1.300 |
�� | -1.267 |
Token | in |
Token ID | 287 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
�� | +1.342 |
aur | +1.294 |
� | +1.179 |
aires | +1.149 |
�� | +1.092 |
Neg logit contributions | |
Roose | -1.210 |
ibaba | -1.111 |
muzzle | -1.057 |
ICLE | -1.033 |
rils | -1.026 |
Token | the |
Token ID | 262 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
�� | +2.900 |
aur | +2.847 |
� | +2.562 |
aires | +2.505 |
�� | +2.368 |
Neg logit contributions | |
Roose | -2.508 |
ibaba | -2.300 |
muzzle | -2.186 |
ICLE | -2.147 |
rils | -2.143 |
Token | Democratic |
Token ID | 4390 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
�� | +0.994 |
aur | +0.961 |
� | +0.879 |
aires | +0.847 |
�� | +0.805 |
Neg logit contributions | |
Roose | -0.857 |
ibaba | -0.789 |
muzzle | -0.749 |
ICLE | -0.731 |
rils | -0.728 |
Token | Republic |
Token ID | 2066 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
�� | +0.499 |
aur | +0.476 |
� | +0.440 |
aires | +0.424 |
�� | +0.409 |
Neg logit contributions | |
Roose | -0.377 |
ibaba | -0.337 |
muzzle | -0.329 |
ICLE | -0.311 |
rils | -0.309 |
Token | of |
Token ID | 286 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
aur | +5.732 |
�� | +5.293 |
aires | +4.916 |
� | +4.894 |
ional | +4.617 |
Neg logit contributions | |
Roose | -7.880 |
ibaba | -7.677 |
mosqu | -7.618 |
ICLE | -7.347 |
rils | -7.196 |
Token | Congo |
Token ID | 25935 |
Feature activation | +9.84e-02 |
Pos logit contributions | |
aur | +7.762 |
�� | +6.882 |
� | +6.375 |
aires | +5.802 |
�� | +5.575 |
Neg logit contributions | |
Roose | -14.657 |
mosqu | -14.249 |
ibaba | -14.182 |
regor | -13.928 |
rils | -13.908 |
Token | , |
Token ID | 11 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
�� | +0.873 |
aur | +0.837 |
� | +0.769 |
aires | +0.738 |
�� | +0.705 |
Neg logit contributions | |
Roose | -0.802 |
ibaba | -0.733 |
muzzle | -0.706 |
ICLE | -0.684 |
rils | -0.676 |
Token | some |
Token ID | 617 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
�� | +2.374 |
aur | +2.308 |
� | +2.092 |
aires | +2.020 |
�� | +1.939 |
Neg logit contributions | |
Roose | -2.190 |
ibaba | -2.041 |
muzzle | -1.934 |
rils | -1.899 |
ICLE | -1.883 |
Token | of |
Token ID | 286 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
�� | +3.266 |
aur | +3.263 |
� | +2.873 |
aires | +2.802 |
ional | +2.628 |
Neg logit contributions | |
Roose | -4.626 |
ibaba | -4.414 |
muzzle | -4.131 |
ICLE | -4.131 |
rils | -4.105 |
Token | whose |
Token ID | 3025 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
�� | +2.400 |
aur | +2.301 |
� | +2.107 |
aires | +2.009 |
�� | +1.941 |
Neg logit contributions | |
Roose | -2.348 |
ibaba | -2.156 |
muzzle | -2.086 |
ICLE | -1.996 |
rils | -1.995 |
Token | percent |
Token ID | 1411 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
�� | +3.380 |
aur | +3.222 |
� | +3.023 |
aires | +2.826 |
�� | +2.752 |
Neg logit contributions | |
Roose | -3.072 |
ibaba | -2.790 |
muzzle | -2.713 |
smart | -2.599 |
azines | -2.574 |
Token | according |
Token ID | 1864 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
Roose | +3.189 |
muzzle | +2.870 |
ibaba | +2.865 |
bang | +2.673 |
ICLE | +2.668 |
Neg logit contributions | |
�� | -3.834 |
aur | -3.557 |
� | -3.300 |
aires | -3.168 |
�� | -3.110 |
Token | to |
Token ID | 284 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
aur | +1.391 |
�� | +1.385 |
� | +1.270 |
aires | +1.182 |
ional | +1.122 |
Neg logit contributions | |
Roose | -1.702 |
ibaba | -1.569 |
muzzle | -1.515 |
azines | -1.467 |
ICLE | -1.460 |
Token | the |
Token ID | 262 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
�� | +3.514 |
aur | +3.359 |
� | +3.117 |
aires | +2.920 |
�� | +2.848 |
Neg logit contributions | |
Roose | -3.416 |
ibaba | -3.103 |
muzzle | -3.047 |
azines | -2.928 |
rils | -2.925 |
Token | University |
Token ID | 2059 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
�� | +3.556 |
aur | +3.426 |
� | +3.145 |
aires | +2.983 |
�� | +2.866 |
Neg logit contributions | |
Roose | -3.421 |
ibaba | -3.161 |
muzzle | -3.054 |
rils | -2.974 |
azines | -2.959 |
Token | of |
Token ID | 286 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
Roose | +2.874 |
ibaba | +2.600 |
muzzle | +2.562 |
Volume | +2.429 |
smart | +2.421 |
Neg logit contributions | |
�� | -3.374 |
aur | -3.104 |
� | -2.852 |
aires | -2.771 |
�� | -2.694 |
Token | Hawaii |
Token ID | 13708 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
aur | +9.098 |
�� | +8.968 |
� | +7.934 |
aires | +7.470 |
�� | +7.007 |
Neg logit contributions | |
Roose | -13.487 |
ibaba | -12.904 |
muzzle | -12.505 |
rils | -12.335 |
smart | -12.201 |
Token | study |
Token ID | 2050 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
�� | +5.036 |
aur | +4.914 |
� | +4.521 |
aires | +4.269 |
�� | +4.101 |
Neg logit contributions | |
Roose | -5.081 |
ibaba | -4.547 |
muzzle | -4.514 |
rils | -4.351 |
ICLE | -4.351 |
Token | cited |
Token ID | 9181 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
�� | +0.982 |
aur | +0.923 |
� | +0.846 |
aires | +0.804 |
�� | +0.777 |
Neg logit contributions | |
Roose | -1.019 |
ibaba | -0.923 |
muzzle | -0.912 |
ICLE | -0.866 |
smart | -0.864 |
Token | previously |
Token ID | 4271 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
�� | +0.227 |
aur | +0.213 |
� | +0.195 |
aires | +0.186 |
�� | +0.180 |
Neg logit contributions | |
Roose | -0.230 |
ibaba | -0.209 |
muzzle | -0.207 |
ICLE | -0.196 |
smart | -0.196 |
Token | – |
Token ID | 784 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
�� | +1.470 |
aur | +1.386 |
� | +1.272 |
aires | +1.209 |
�� | +1.169 |
Neg logit contributions | |
Roose | -1.486 |
ibaba | -1.343 |
muzzle | -1.327 |
ICLE | -1.260 |
smart | -1.255 |
Token | a |
Token ID | 257 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
Roose | +2.054 |
muzzle | +1.817 |
ibaba | +1.816 |
smart | +1.691 |
bang | +1.690 |
Neg logit contributions | |
�� | -2.546 |
aur | -2.394 |
� | -2.219 |
aires | -2.116 |
�� | -2.065 |
Token | stop |
Token ID | 2245 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
Roose | +0.125 |
ibaba | +0.113 |
muzzle | +0.110 |
ICLE | +0.105 |
rils | +0.105 |
Neg logit contributions | |
�� | -0.141 |
aur | -0.133 |
� | -0.123 |
aires | -0.118 |
�� | -0.113 |
Token | at |
Token ID | 379 |
Feature activation | -7.92e-02 |
Pos logit contributions | |
Roose | +1.742 |
ibaba | +1.574 |
muzzle | +1.564 |
ICLE | +1.476 |
smart | +1.474 |
Neg logit contributions | |
�� | -1.834 |
aur | -1.711 |
� | -1.578 |
aires | -1.507 |
�� | -1.461 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Roose | +0.647 |
ibaba | +0.576 |
muzzle | +0.573 |
ICLE | +0.539 |
smart | +0.537 |
Neg logit contributions | |
�� | -0.720 |
aur | -0.679 |
� | -0.626 |
aires | -0.598 |
�� | -0.580 |
Token | University |
Token ID | 2059 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
Roose | +1.106 |
ibaba | +0.986 |
muzzle | +0.984 |
ICLE | +0.927 |
smart | +0.923 |
Neg logit contributions | |
�� | -1.166 |
aur | -1.099 |
� | -1.009 |
aires | -0.963 |
�� | -0.935 |
Token | of |
Token ID | 286 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Roose | +2.251 |
ibaba | +1.903 |
muzzle | +1.903 |
bang | +1.783 |
smart | +1.774 |
Neg logit contributions | |
�� | -3.274 |
aur | -3.086 |
� | -2.873 |
�� | -2.743 |
aires | -2.732 |
Token | Iowa |
Token ID | 9406 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
aur | +9.243 |
�� | +8.760 |
aires | +7.966 |
� | +7.945 |
ARY | +6.776 |
Neg logit contributions | |
ibaba | -13.271 |
Roose | -12.660 |
rils | -12.128 |
muzzle | -12.097 |
smart | -12.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
�� | +3.325 |
aur | +3.198 |
� | +2.930 |
aires | +2.792 |
�� | +2.650 |
Neg logit contributions | |
Roose | -3.512 |
ibaba | -3.223 |
muzzle | -3.137 |
ICLE | -3.007 |
rils | -2.997 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
Roose | +0.963 |
ibaba | +0.849 |
muzzle | +0.840 |
ICLE | +0.798 |
smart | +0.792 |
Neg logit contributions | |
�� | -1.149 |
aur | -1.079 |
� | -0.993 |
aires | -0.953 |
�� | -0.928 |
Token | If |
Token ID | 1532 |
Feature activation | -1.39e-03 |
Pos logit contributions | |
Roose | +1.598 |
ibaba | +1.437 |
muzzle | +1.416 |
smart | +1.367 |
ICLE | +1.358 |
Neg logit contributions | |
�� | -1.828 |
aur | -1.703 |
� | -1.583 |
aires | -1.514 |
�� | -1.472 |
Token | it |
Token ID | 340 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
Roose | +0.011 |
ibaba | +0.010 |
muzzle | +0.010 |
ICLE | +0.010 |
smart | +0.010 |
Neg logit contributions | |
�� | -0.013 |
aur | -0.012 |
� | -0.011 |
aires | -0.010 |
�� | -0.010 |
Token | Nation |
Token ID | 8741 |
Feature activation | +7.60e-02 |
Pos logit contributions | |
�� | +0.819 |
aur | +0.784 |
� | +0.722 |
aires | +0.692 |
�� | +0.669 |
Neg logit contributions | |
Roose | -0.675 |
ibaba | -0.607 |
muzzle | -0.597 |
ICLE | -0.564 |
rils | -0.560 |
Token | done |
Token ID | 1760 |
Feature activation | +9.94e-02 |
Pos logit contributions | |
�� | +0.650 |
aur | +0.614 |
� | +0.562 |
aires | +0.537 |
�� | +0.517 |
Neg logit contributions | |
Roose | -0.656 |
ibaba | -0.596 |
muzzle | -0.586 |
ICLE | -0.556 |
smart | -0.554 |
Token | by |
Token ID | 416 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
�� | +0.903 |
aur | +0.850 |
� | +0.786 |
aires | +0.749 |
�� | +0.727 |
Neg logit contributions | |
Roose | -0.812 |
ibaba | -0.729 |
muzzle | -0.721 |
ICLE | -0.680 |
smart | -0.680 |
Token | the |
Token ID | 262 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
�� | +1.060 |
aur | +1.007 |
� | +0.927 |
aires | +0.886 |
�� | +0.856 |
Neg logit contributions | |
Roose | -0.946 |
ibaba | -0.856 |
muzzle | -0.839 |
ICLE | -0.798 |
smart | -0.793 |
Token | University |
Token ID | 2059 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
�� | +1.920 |
aur | +1.843 |
� | +1.688 |
aires | +1.608 |
�� | +1.540 |
Neg logit contributions | |
Roose | -1.766 |
ibaba | -1.637 |
muzzle | -1.573 |
ICLE | -1.514 |
smart | -1.506 |
Token | of |
Token ID | 286 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Roose | +1.771 |
ibaba | +1.562 |
muzzle | +1.542 |
smart | +1.452 |
Volume | +1.441 |
Neg logit contributions | |
�� | -2.224 |
aur | -2.088 |
� | -1.944 |
aires | -1.854 |
�� | -1.817 |
Token | Victoria |
Token ID | 12313 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
aur | +9.106 |
�� | +8.417 |
aires | +7.474 |
� | +7.473 |
ARY | +6.735 |
Neg logit contributions | |
ibaba | -13.448 |
Roose | -12.534 |
rils | -12.283 |
smart | -12.085 |
ICLE | -12.052 |
Token | � |
Token ID | 447 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
�� | +4.013 |
aur | +3.951 |
� | +3.517 |
aires | +3.424 |
ional | +3.182 |
Neg logit contributions | |
Roose | -4.321 |
ibaba | -4.051 |
muzzle | -3.929 |
ICLE | -3.785 |
rils | -3.741 |
Token | � |
Token ID | 247 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
�� | +1.370 |
aur | +1.258 |
� | +1.125 |
aires | +1.026 |
�� | +0.986 |
Neg logit contributions | |
Roose | -2.418 |
ibaba | -2.238 |
muzzle | -2.219 |
ICLE | -2.128 |
smart | -2.126 |
Token | s |
Token ID | 82 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
�� | +4.276 |
aur | +4.187 |
� | +3.818 |
aires | +3.654 |
ional | +3.443 |
Neg logit contributions | |
Roose | -4.288 |
ibaba | -3.856 |
muzzle | -3.812 |
ICLE | -3.607 |
azines | -3.602 |
Token | Environmental |
Token ID | 13272 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
�� | +3.304 |
aur | +3.193 |
� | +2.919 |
aires | +2.783 |
�� | +2.677 |
Neg logit contributions | |
Roose | -2.681 |
ibaba | -2.471 |
muzzle | -2.387 |
ICLE | -2.282 |
smart | -2.258 |
Token | Pasadena |
Token ID | 41010 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
�� | +2.698 |
aur | +2.573 |
� | +2.369 |
aires | +2.273 |
�� | +2.200 |
Neg logit contributions | |
Roose | -2.246 |
ibaba | -2.049 |
muzzle | -1.999 |
ICLE | -1.896 |
rils | -1.890 |
Token | , |
Token ID | 11 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
Roose | +2.870 |
muzzle | +2.528 |
Volume | +2.476 |
ibaba | +2.469 |
smart | +2.458 |
Neg logit contributions | |
�� | -3.917 |
aur | -3.575 |
� | -3.363 |
aires | -3.181 |
�� | -3.124 |
Token | California |
Token ID | 3442 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
�� | +0.092 |
aur | +0.084 |
� | +0.076 |
aires | +0.071 |
�� | +0.069 |
Neg logit contributions | |
Roose | -0.123 |
muzzle | -0.114 |
ibaba | -0.113 |
ICLE | -0.109 |
Volume | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
Roose | +3.646 |
ibaba | +3.268 |
muzzle | +3.203 |
azines | +3.184 |
Volume | +3.136 |
Neg logit contributions | |
�� | -5.139 |
aur | -4.634 |
� | -4.366 |
aires | -4.079 |
�� | -4.041 |
Token | Hide |
Token ID | 10415 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
Roose | +2.897 |
ibaba | +2.523 |
muzzle | +2.518 |
Volume | +2.456 |
azines | +2.414 |
Neg logit contributions | |
�� | -3.410 |
aur | -3.137 |
� | -2.900 |
aires | -2.752 |
�� | -2.725 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
Roose | +4.581 |
muzzle | +3.968 |
ibaba | +3.967 |
ICLE | +3.707 |
Volume | +3.699 |
Neg logit contributions | |
�� | -7.157 |
aur | -6.680 |
� | -6.254 |
aires | -5.998 |
�� | -5.890 |
Token | 12 |
Token ID | 1105 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
Roose | +8.205 |
Volume | +6.872 |
Alm | +6.346 |
muzzle | +6.104 |
Salem | +6.100 |
Neg logit contributions | |
�� | -21.501 |
� | -18.448 |
aur | -18.387 |
�� | -18.117 |
�� | -17.756 |
Token | of |
Token ID | 286 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
aur | +1.230 |
�� | +1.139 |
aires | +1.041 |
� | +1.040 |
ional | +0.972 |
Neg logit contributions | |
Roose | -1.502 |
ibaba | -1.439 |
mosqu | -1.418 |
muzzle | -1.347 |
rils | -1.316 |
Token | 15 |
Token ID | 1315 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
aur | +2.288 |
�� | +2.169 |
� | +1.984 |
aires | +1.981 |
� | +1.761 |
Neg logit contributions | |
Roose | -3.429 |
ibaba | -3.333 |
mosqu | -3.230 |
rils | -3.176 |
muzzle | -3.066 |
Token | Photos |
Token ID | 9434 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
aur | +1.816 |
�� | +1.570 |
aires | +1.403 |
� | +1.360 |
ional | +1.114 |
Neg logit contributions | |
Roose | -5.105 |
ibaba | -4.958 |
mosqu | -4.951 |
muzzle | -4.806 |
rils | -4.636 |
Token | : |
Token ID | 25 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Roose | +5.502 |
muzzle | +4.781 |
ibaba | +4.619 |
smart | +4.570 |
Volume | +4.557 |
Neg logit contributions | |
�� | -8.007 |
aur | -7.174 |
� | -6.863 |
�� | -6.544 |
aires | -6.427 |
Token | 91 |
Token ID | 6420 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
�� | +1.043 |
aur | +1.039 |
� | +0.906 |
aires | +0.888 |
ional | +0.802 |
Neg logit contributions | |
Roose | -1.441 |
ibaba | -1.366 |
muzzle | -1.316 |
rils | -1.263 |
azines | -1.254 |
Token | x |
Token ID | 2124 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
aur | +5.123 |
�� | +4.808 |
aires | +4.457 |
� | +4.435 |
equival | +4.039 |
Neg logit contributions | |
Roose | -5.535 |
ibaba | -5.314 |
mosqu | -5.058 |
ICLE | -4.929 |
azines | -4.906 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
Roose | +3.783 |
ibaba | +3.422 |
muzzle | +3.389 |
ICLE | +3.220 |
smart | +3.212 |
Neg logit contributions | |
�� | -3.636 |
aur | -3.405 |
� | -3.125 |
aires | -2.965 |
�� | -2.873 |
Token | x |
Token ID | 87 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
Roose | +1.157 |
ibaba | +1.057 |
muzzle | +1.036 |
ICLE | +0.984 |
rils | +0.976 |
Neg logit contributions | |
�� | -0.964 |
aur | -0.954 |
� | -0.832 |
aires | -0.818 |
ional | -0.748 |
Token | . |
Token ID | 13 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
Roose | +6.863 |
muzzle | +6.070 |
ibaba | +5.925 |
smart | +5.911 |
Volume | +5.785 |
Neg logit contributions | |
�� | -9.078 |
aur | -7.797 |
� | -7.617 |
�� | -7.262 |
aires | -6.964 |
Token | x |
Token ID | 87 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
Roose | +3.710 |
ibaba | +3.352 |
muzzle | +3.312 |
ICLE | +3.141 |
smart | +3.123 |
Neg logit contributions | |
�� | -3.709 |
aur | -3.506 |
� | -3.208 |
aires | -3.062 |
�� | -2.950 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Roose | +9.676 |
smart | +8.738 |
muzzle | +8.404 |
ICLE | +8.072 |
Volume | +8.059 |
Neg logit contributions | |
�� | -17.715 |
� | -14.454 |
�� | -14.299 |
aur | -14.076 |
�� | -13.956 |
Token | 132 |
Token ID | 19924 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Roose | +5.485 |
smart | +4.689 |
muzzle | +4.553 |
ibaba | +4.374 |
rils | +4.146 |
Neg logit contributions | |
�� | -15.744 |
aur | -13.978 |
� | -13.892 |
�� | -13.501 |
aires | -13.390 |
Token | 10 |
Token ID | 838 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
�� | +1.247 |
aur | +1.219 |
� | +1.087 |
aires | +1.050 |
�� | +0.985 |
Neg logit contributions | |
Roose | -1.533 |
ibaba | -1.448 |
muzzle | -1.392 |
ICLE | -1.340 |
azines | -1.337 |
Token | 86 |
Token ID | 4521 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
�� | +0.667 |
aur | +0.647 |
� | +0.574 |
aires | +0.555 |
�� | +0.517 |
Neg logit contributions | |
Roose | -0.850 |
ibaba | -0.794 |
muzzle | -0.775 |
rils | -0.737 |
ICLE | -0.734 |
Token | 213 |
Token ID | 26427 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Roose | +1.006 |
ibaba | +0.913 |
muzzle | +0.902 |
ICLE | +0.856 |
smart | +0.853 |
Neg logit contributions | |
�� | -0.952 |
aur | -0.898 |
� | -0.821 |
aires | -0.781 |
�� | -0.753 |
Token | Haitian |
Token ID | 48723 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
�� | +1.703 |
aur | +1.656 |
� | +1.518 |
aires | +1.455 |
�� | +1.388 |
Neg logit contributions | |
Roose | -1.490 |
ibaba | -1.382 |
muzzle | -1.350 |
ICLE | -1.267 |
rils | -1.267 |
Token | government |
Token ID | 1230 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
�� | +1.203 |
aur | +1.165 |
� | +1.081 |
aires | +1.035 |
�� | +0.995 |
Neg logit contributions | |
Roose | -0.870 |
ibaba | -0.784 |
muzzle | -0.763 |
smart | -0.722 |
rils | -0.719 |
Token | officials |
Token ID | 2828 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
�� | +1.427 |
aur | +1.386 |
� | +1.282 |
aires | +1.238 |
�� | +1.181 |
Neg logit contributions | |
Roose | -1.057 |
ibaba | -0.958 |
muzzle | -0.930 |
ICLE | -0.874 |
smart | -0.873 |
Token | went |
Token ID | 1816 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
�� | +0.264 |
aur | +0.253 |
� | +0.233 |
aires | +0.222 |
�� | +0.215 |
Neg logit contributions | |
Roose | -0.230 |
ibaba | -0.208 |
muzzle | -0.204 |
ICLE | -0.193 |
smart | -0.192 |
Token | door |
Token ID | 3420 |
Feature activation | +9.81e-03 |
Pos logit contributions | |
�� | +2.529 |
aur | +2.450 |
� | +2.249 |
aires | +2.122 |
�� | +2.017 |
Neg logit contributions | |
Roose | -2.808 |
ibaba | -2.588 |
muzzle | -2.519 |
ICLE | -2.448 |
rils | -2.430 |
Token | to |
Token ID | 284 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
�� | +0.095 |
aur | +0.089 |
� | +0.083 |
aires | +0.079 |
�� | +0.077 |
Neg logit contributions | |
Roose | -0.075 |
ibaba | -0.066 |
muzzle | -0.066 |
smart | -0.062 |
ICLE | -0.062 |
Token | door |
Token ID | 3420 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
Roose | +4.727 |
muzzle | +4.158 |
bang | +3.940 |
ibaba | +3.089 |
shel | +2.960 |
Neg logit contributions | |
�� | -21.840 |
aur | -20.000 |
�� | -19.154 |
� | -19.144 |
aires | -18.331 |
Token | warning |
Token ID | 6509 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
Roose | +1.835 |
muzzle | +1.639 |
ibaba | +1.628 |
smart | +1.537 |
bang | +1.533 |
Neg logit contributions | |
�� | -2.028 |
aur | -1.879 |
� | -1.742 |
aires | -1.652 |
�� | -1.617 |
Token | residents |
Token ID | 5085 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Roose | +1.129 |
ibaba | +0.974 |
muzzle | +0.971 |
ICLE | +0.892 |
smart | +0.892 |
Neg logit contributions | |
�� | -1.859 |
aur | -1.761 |
� | -1.650 |
aires | -1.580 |
�� | -1.551 |
Token | . |
Token ID | 13 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
Roose | +0.900 |
ibaba | +0.800 |
muzzle | +0.797 |
ICLE | +0.750 |
smart | +0.749 |
Neg logit contributions | |
�� | -1.055 |
aur | -0.989 |
� | -0.916 |
aires | -0.875 |
�� | -0.853 |
Token | A |
Token ID | 317 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
�� | +0.564 |
aur | +0.543 |
� | +0.498 |
aires | +0.478 |
�� | +0.458 |
Neg logit contributions | |
Roose | -0.496 |
ibaba | -0.455 |
muzzle | -0.446 |
ICLE | -0.419 |
rils | -0.418 |
Token | This |
Token ID | 770 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
�� | +1.298 |
aur | +1.235 |
� | +1.147 |
aires | +1.082 |
�� | +1.042 |
Neg logit contributions | |
Roose | -1.372 |
ibaba | -1.228 |
muzzle | -1.214 |
ICLE | -1.173 |
smart | -1.158 |
Token | Story |
Token ID | 8362 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
Roose | +5.492 |
ibaba | +4.609 |
muzzle | +4.504 |
Volume | +4.455 |
ICLE | +4.321 |
Neg logit contributions | |
�� | -10.592 |
aur | -9.497 |
� | -9.328 |
aires | -9.082 |
�� | -8.927 |
Token | Send |
Token ID | 16290 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
Roose | +7.381 |
muzzle | +6.492 |
ibaba | +6.333 |
Volume | +6.317 |
azines | +6.278 |
Neg logit contributions | |
�� | -7.614 |
aur | -6.561 |
� | -6.178 |
aires | -5.792 |
�� | -5.735 |
Token | email |
Token ID | 3053 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
Roose | +1.665 |
ibaba | +1.466 |
muzzle | +1.455 |
ICLE | +1.335 |
smart | +1.327 |
Neg logit contributions | |
�� | -2.730 |
aur | -2.591 |
� | -2.409 |
aires | -2.337 |
�� | -2.272 |
Token | to |
Token ID | 284 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
Roose | +5.366 |
muzzle | +4.836 |
ibaba | +4.750 |
Volume | +4.598 |
smart | +4.480 |
Neg logit contributions | |
�� | -8.652 |
aur | -7.991 |
� | -7.455 |
�� | -7.153 |
aires | -7.076 |
Token | this |
Token ID | 428 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
Roose | +6.245 |
muzzle | +5.568 |
ibaba | +5.456 |
smart | +5.251 |
Volume | +5.236 |
Neg logit contributions | |
�� | -7.256 |
aur | -6.733 |
� | -6.197 |
aires | -5.945 |
�� | -5.873 |
Token | address |
Token ID | 2209 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Roose | +8.207 |
muzzle | +7.217 |
ibaba | +6.794 |
bang | +6.636 |
smart | +6.366 |
Neg logit contributions | |
�� | -18.016 |
aur | -16.271 |
� | -15.526 |
aires | -15.306 |
�� | -15.103 |
Token | Enter |
Token ID | 6062 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
Roose | +2.983 |
muzzle | +2.012 |
ibaba | +1.952 |
Volume | +1.833 |
bang | +1.736 |
Neg logit contributions | |
�� | -14.431 |
aur | -13.557 |
� | -12.858 |
�� | -12.496 |
aires | -12.464 |
Token | Your |
Token ID | 3406 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
Roose | +3.689 |
ibaba | +3.353 |
muzzle | +3.308 |
ICLE | +3.164 |
smart | +3.151 |
Neg logit contributions | |
�� | -3.184 |
aur | -2.991 |
� | -2.726 |
aires | -2.566 |
�� | -2.480 |
Token | Name |
Token ID | 6530 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
Roose | +3.705 |
ibaba | +3.192 |
muzzle | +3.181 |
Volume | +2.998 |
smart | +2.966 |
Neg logit contributions | |
�� | -5.819 |
aur | -5.397 |
� | -5.144 |
aires | -4.971 |
�� | -4.862 |
Token | Add |
Token ID | 3060 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
Roose | +7.021 |
muzzle | +6.373 |
ibaba | +6.352 |
Volume | +6.100 |
azines | +6.061 |
Neg logit contributions | |
�� | -5.936 |
aur | -5.286 |
� | -4.833 |
�� | -4.590 |
aires | -4.563 |
Token | signs |
Token ID | 5895 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Roose | +1.041 |
muzzle | +0.919 |
ibaba | +0.917 |
smart | +0.856 |
ICLE | +0.853 |
Neg logit contributions | |
�� | -1.312 |
aur | -1.235 |
� | -1.146 |
aires | -1.093 |
�� | -1.070 |
Token | about |
Token ID | 546 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
�� | +1.633 |
aur | +1.556 |
� | +1.427 |
aires | +1.348 |
�� | +1.302 |
Neg logit contributions | |
Roose | -1.621 |
ibaba | -1.480 |
muzzle | -1.446 |
ICLE | -1.384 |
azines | -1.373 |
Token | Hillary |
Token ID | 5073 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
Roose | +1.331 |
muzzle | +1.185 |
ibaba | +1.183 |
smart | +1.115 |
ICLE | +1.110 |
Neg logit contributions | |
�� | -1.396 |
aur | -1.312 |
� | -1.201 |
aires | -1.150 |
�� | -1.116 |
Token | the |
Token ID | 262 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
�� | +1.473 |
aur | +1.373 |
� | +1.240 |
aires | +1.166 |
�� | +1.119 |
Neg logit contributions | |
Roose | -2.006 |
ibaba | -1.838 |
muzzle | -1.820 |
ICLE | -1.741 |
smart | -1.735 |
Token | way |
Token ID | 835 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
Roose | +2.519 |
muzzle | +2.224 |
ibaba | +2.197 |
smart | +2.102 |
ICLE | +2.074 |
Neg logit contributions | |
�� | -2.755 |
aur | -2.590 |
� | -2.369 |
aires | -2.261 |
�� | -2.206 |
Token | Republicans |
Token ID | 4734 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Roose | +2.291 |
muzzle | +2.013 |
ibaba | +2.008 |
smart | +1.900 |
ICLE | +1.891 |
Neg logit contributions | |
�� | -2.487 |
aur | -2.336 |
� | -2.132 |
aires | -2.052 |
�� | -1.997 |
Token | ignored |
Token ID | 9514 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
Roose | +8.717 |
muzzle | +8.372 |
disappro | +7.404 |
bang | +7.367 |
shel | +7.252 |
Neg logit contributions | |
�� | -16.191 |
aur | -14.436 |
�� | -13.597 |
� | -13.406 |
aires | -13.065 |
Token | the |
Token ID | 262 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
Roose | +6.254 |
muzzle | +5.335 |
ibaba | +5.029 |
Volume | +5.005 |
disappro | +5.002 |
Neg logit contributions | |
�� | -7.650 |
aur | -7.144 |
� | -6.584 |
aires | -6.369 |
�� | -6.233 |
Token | ones |
Token ID | 3392 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
Roose | +4.521 |
muzzle | +3.860 |
ibaba | +3.704 |
bang | +3.605 |
disappro | +3.535 |
Neg logit contributions | |
�� | -5.632 |
aur | -5.286 |
� | -4.934 |
aires | -4.674 |
�� | -4.602 |
Token | raised |
Token ID | 4376 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
Roose | +5.477 |
muzzle | +4.734 |
ibaba | +4.616 |
smart | +4.486 |
disappro | +4.477 |
Neg logit contributions | |
�� | -6.006 |
aur | -5.593 |
� | -5.141 |
aires | -5.009 |
�� | -4.864 |
Token | about |
Token ID | 546 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
Roose | +5.068 |
muzzle | +4.418 |
ibaba | +4.247 |
bang | +4.175 |
disappro | +4.160 |
Neg logit contributions | |
�� | -7.015 |
aur | -6.440 |
� | -6.103 |
aires | -5.836 |
�� | -5.764 |
Token | 213 |
Token ID | 26427 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Roose | +1.006 |
ibaba | +0.913 |
muzzle | +0.902 |
ICLE | +0.856 |
smart | +0.853 |
Neg logit contributions | |
�� | -0.952 |
aur | -0.898 |
� | -0.821 |
aires | -0.781 |
�� | -0.753 |
Token | x |
Token ID | 2124 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
�� | +1.194 |
aur | +1.186 |
� | +1.072 |
aires | +1.054 |
�� | +0.982 |
Neg logit contributions | |
Roose | -1.001 |
ibaba | -0.930 |
muzzle | -0.863 |
ICLE | -0.858 |
azines | -0.847 |
Token | . |
Token ID | 13 |
Feature activation | -8.78e-02 |
Pos logit contributions | |
Roose | +3.110 |
ibaba | +2.835 |
muzzle | +2.790 |
ICLE | +2.651 |
rils | +2.640 |
Neg logit contributions | |
�� | -2.814 |
aur | -2.681 |
� | -2.432 |
aires | -2.324 |
�� | -2.218 |
Token | x |
Token ID | 87 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Roose | +0.825 |
ibaba | +0.754 |
muzzle | +0.738 |
ICLE | +0.702 |
rils | +0.696 |
Neg logit contributions | |
�� | -0.676 |
aur | -0.673 |
� | -0.583 |
aires | -0.575 |
ional | -0.526 |
Token | . |
Token ID | 13 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Roose | +7.298 |
muzzle | +6.445 |
smart | +6.343 |
ibaba | +6.246 |
Volume | +6.184 |
Neg logit contributions | |
�� | -10.346 |
aur | -8.760 |
� | -8.658 |
�� | -8.299 |
�� | -7.869 |
Token | x |
Token ID | 87 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
Roose | +4.867 |
ibaba | +4.381 |
muzzle | +4.348 |
smart | +4.162 |
ICLE | +4.115 |
Neg logit contributions | |
�� | -5.334 |
aur | -4.931 |
� | -4.619 |
aires | -4.345 |
�� | -4.286 |
Token | . |
Token ID | 13 |
Feature activation | -9.70e-01 |
Pos logit contributions | |
Roose | +9.324 |
smart | +8.700 |
muzzle | +8.288 |
iera | +8.123 |
ICLE | +8.055 |
Neg logit contributions | |
�� | -15.999 |
� | -13.074 |
�� | -12.854 |
aur | -12.747 |
�� | -12.452 |
Token | 133 |
Token ID | 16945 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Roose | +6.510 |
muzzle | +5.836 |
smart | +5.816 |
ibaba | +5.741 |
ICLE | +5.485 |
Neg logit contributions | |
�� | -9.784 |
aur | -8.703 |
� | -8.444 |
aires | -8.057 |
�� | -8.037 |
Token | 11 |
Token ID | 1367 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
�� | +0.836 |
aur | +0.817 |
� | +0.731 |
aires | +0.706 |
�� | +0.665 |
Neg logit contributions | |
Roose | -0.993 |
ibaba | -0.936 |
muzzle | -0.901 |
ICLE | -0.862 |
azines | -0.862 |
Token | 485 |
Token ID | 32642 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
Roose | +1.372 |
ibaba | +1.252 |
muzzle | +1.235 |
ICLE | +1.170 |
rils | +1.166 |
Neg logit contributions | |
�� | -1.248 |
aur | -1.180 |
� | -1.074 |
aires | -1.024 |
�� | -0.981 |
Token | 37 |
Token ID | 2718 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
aur | +2.238 |
�� | +2.109 |
aires | +1.906 |
� | +1.893 |
ional | +1.706 |
Neg logit contributions | |
Roose | -3.043 |
ibaba | -3.005 |
muzzle | -2.817 |
mosqu | -2.796 |
rils | -2.735 |
Token | 10 |
Token ID | 940 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
aur | +2.793 |
�� | +2.657 |
� | +2.375 |
aires | +2.369 |
ional | +2.130 |
Neg logit contributions | |
Roose | -3.736 |
ibaba | -3.699 |
muzzle | -3.438 |
mosqu | -3.383 |
rils | -3.323 |
Token | x |
Token ID | 2124 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
aur | +4.225 |
�� | +4.044 |
� | +3.676 |
aires | +3.640 |
equival | +3.418 |
Neg logit contributions | |
Roose | -4.200 |
ibaba | -3.981 |
ICLE | -3.732 |
rils | -3.680 |
azines | -3.671 |
Token | . |
Token ID | 13 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Roose | +2.684 |
ibaba | +2.456 |
muzzle | +2.410 |
ICLE | +2.289 |
rils | +2.285 |
Neg logit contributions | |
�� | -2.342 |
aur | -2.256 |
� | -2.031 |
aires | -1.948 |
�� | -1.842 |
Token | x |
Token ID | 87 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
Roose | +2.175 |
ibaba | +1.977 |
muzzle | +1.945 |
ICLE | +1.847 |
rils | +1.828 |
Neg logit contributions | |
�� | -1.959 |
aur | -1.902 |
� | -1.692 |
aires | -1.642 |
�� | -1.535 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Roose | +6.776 |
muzzle | +5.992 |
ibaba | +5.867 |
smart | +5.833 |
ICLE | +5.717 |
Neg logit contributions | |
�� | -8.888 |
aur | -7.681 |
� | -7.501 |
�� | -7.135 |
aires | -6.847 |
Token | x |
Token ID | 87 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Roose | +7.540 |
ibaba | +6.782 |
muzzle | +6.712 |
smart | +6.656 |
ICLE | +6.430 |
Neg logit contributions | |
�� | -10.040 |
aur | -8.884 |
� | -8.786 |
�� | -8.344 |
aires | -7.966 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
Roose | +9.477 |
smart | +8.210 |
ICLE | +8.198 |
muzzle | +8.190 |
ibaba | +7.832 |
Neg logit contributions | |
�� | -15.570 |
� | -13.057 |
aur | -13.012 |
�� | -12.854 |
�� | -11.967 |
Token | 125 |
Token ID | 11623 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Roose | +5.417 |
smart | +4.485 |
ibaba | +4.476 |
muzzle | +4.460 |
ICLE | +4.416 |
Neg logit contributions | |
�� | -12.569 |
aur | -11.563 |
� | -11.153 |
�� | -10.839 |
aires | -10.685 |
Token | 110 |
Token ID | 9796 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
Roose | +0.985 |
ibaba | +0.904 |
muzzle | +0.887 |
ICLE | +0.844 |
smart | +0.842 |
Neg logit contributions | |
�� | -0.891 |
aur | -0.845 |
� | -0.770 |
aires | -0.733 |
�� | -0.704 |
Token | 5 |
Token ID | 20 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
Roose | +0.942 |
ibaba | +0.869 |
muzzle | +0.856 |
ICLE | +0.813 |
rils | +0.812 |
Neg logit contributions | |
�� | -0.723 |
aur | -0.684 |
� | -0.614 |
aires | -0.583 |
�� | -0.554 |
Token | 172 |
Token ID | 23628 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
Roose | +0.548 |
ibaba | +0.494 |
muzzle | +0.489 |
ICLE | +0.465 |
smart | +0.463 |
Neg logit contributions | |
�� | -0.543 |
aur | -0.509 |
� | -0.468 |
aires | -0.445 |
�� | -0.431 |
Token | 41 |
Token ID | 3901 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
aur | +1.734 |
�� | +1.691 |
� | +1.500 |
aires | +1.483 |
ional | +1.336 |
Neg logit contributions | |
Roose | -2.239 |
ibaba | -2.171 |
muzzle | -2.044 |
mosqu | -1.981 |
rils | -1.971 |
Token | x |
Token ID | 2124 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
aur | +3.699 |
�� | +3.585 |
� | +3.256 |
aires | +3.231 |
ional | +3.017 |
Neg logit contributions | |
Roose | -3.635 |
ibaba | -3.467 |
ICLE | -3.196 |
mosqu | -3.193 |
azines | -3.165 |
Token | . |
Token ID | 13 |
Feature activation | +9.06e-03 |
Pos logit contributions | |
Roose | +3.710 |
ibaba | +3.361 |
muzzle | +3.325 |
ICLE | +3.159 |
smart | +3.149 |
Neg logit contributions | |
�� | -3.530 |
aur | -3.316 |
� | -3.037 |
aires | -2.885 |
�� | -2.789 |
Token | x |
Token ID | 87 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
aur | +0.068 |
�� | +0.067 |
� | +0.058 |
aires | +0.058 |
ional | +0.052 |
Neg logit contributions | |
Roose | -0.087 |
ibaba | -0.080 |
muzzle | -0.078 |
mosqu | -0.076 |
ICLE | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
Roose | +6.279 |
muzzle | +5.580 |
ibaba | +5.486 |
smart | +5.411 |
ICLE | +5.314 |
Neg logit contributions | |
�� | -7.753 |
aur | -6.780 |
� | -6.539 |
�� | -6.196 |
aires | -6.022 |
Token | x |
Token ID | 87 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Roose | +1.127 |
ibaba | +1.028 |
muzzle | +1.007 |
ICLE | +0.958 |
rils | +0.950 |
Neg logit contributions | |
�� | -0.943 |
aur | -0.934 |
� | -0.815 |
aires | -0.801 |
ional | -0.734 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Roose | +8.810 |
smart | +8.012 |
muzzle | +7.772 |
ibaba | +7.474 |
Volume | +7.454 |
Neg logit contributions | |
�� | -16.219 |
� | -13.509 |
�� | -13.273 |
aur | -13.066 |
�� | -12.934 |
Token | 136 |
Token ID | 20809 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Roose | +6.857 |
muzzle | +6.055 |
smart | +5.962 |
ibaba | +5.920 |
hea | +5.526 |
Neg logit contributions | |
�� | -12.299 |
aur | -11.055 |
� | -10.915 |
�� | -10.489 |
aires | -10.300 |
Token | 10 |
Token ID | 838 |
Feature activation | +9.40e-02 |
Pos logit contributions | |
�� | +0.812 |
aur | +0.793 |
� | +0.711 |
aires | +0.682 |
�� | +0.645 |
Neg logit contributions | |
Roose | -0.980 |
ibaba | -0.923 |
muzzle | -0.889 |
ICLE | -0.852 |
rils | -0.852 |
Token | 479 |
Token ID | 31714 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
�� | +0.683 |
aur | +0.663 |
� | +0.586 |
aires | +0.565 |
�� | +0.526 |
Neg logit contributions | |
Roose | -0.911 |
ibaba | -0.850 |
muzzle | -0.831 |
rils | -0.790 |
azines | -0.787 |
Token | 19 |
Token ID | 1129 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
aur | +2.067 |
�� | +1.992 |
� | +1.770 |
aires | +1.757 |
ional | +1.584 |
Neg logit contributions | |
Roose | -2.783 |
ibaba | -2.712 |
muzzle | -2.552 |
mosqu | -2.509 |
rils | -2.458 |
Token | 19 |
Token ID | 1129 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
aur | +2.067 |
�� | +1.992 |
� | +1.770 |
aires | +1.757 |
ional | +1.584 |
Neg logit contributions | |
Roose | -2.783 |
ibaba | -2.712 |
muzzle | -2.552 |
mosqu | -2.509 |
rils | -2.458 |
Token | x |
Token ID | 2124 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
aur | +4.469 |
�� | +4.211 |
� | +3.868 |
aires | +3.823 |
ional | +3.552 |
Neg logit contributions | |
Roose | -4.621 |
ibaba | -4.487 |
mosqu | -4.157 |
azines | -4.109 |
ICLE | -4.092 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
Roose | +3.568 |
ibaba | +3.238 |
muzzle | +3.198 |
ICLE | +3.039 |
smart | +3.026 |
Neg logit contributions | |
�� | -3.341 |
aur | -3.152 |
� | -2.879 |
aires | -2.739 |
�� | -2.637 |
Token | x |
Token ID | 87 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Roose | +1.443 |
ibaba | +1.316 |
muzzle | +1.290 |
ICLE | +1.227 |
rils | +1.215 |
Neg logit contributions | |
�� | -1.235 |
aur | -1.214 |
� | -1.066 |
aires | -1.043 |
�� | -0.959 |
Token | . |
Token ID | 13 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
Roose | +7.535 |
muzzle | +6.672 |
smart | +6.606 |
ibaba | +6.453 |
Volume | +6.394 |
Neg logit contributions | |
�� | -11.026 |
aur | -9.287 |
� | -9.233 |
�� | -8.886 |
�� | -8.446 |
Token | x |
Token ID | 87 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Roose | +5.743 |
ibaba | +5.193 |
muzzle | +5.160 |
smart | +4.994 |
ICLE | +4.848 |
Neg logit contributions | |
�� | -6.728 |
aur | -6.155 |
� | -5.861 |
�� | -5.478 |
aires | -5.461 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Roose | +8.982 |
smart | +8.384 |
muzzle | +8.139 |
ibaba | +7.917 |
iera | +7.816 |
Neg logit contributions | |
�� | -15.422 |
� | -12.959 |
aur | -12.871 |
�� | -12.697 |
�� | -12.215 |
Token | 137 |
Token ID | 19708 |
Feature activation | -4.96e-02 |
Pos logit contributions | |
Roose | +3.741 |
muzzle | +3.042 |
smart | +2.973 |
ibaba | +2.966 |
bang | +2.581 |
Neg logit contributions | |
�� | -14.336 |
aur | -13.347 |
� | -13.016 |
�� | -12.611 |
aires | -12.601 |
Token | 12 |
Token ID | 1105 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
Roose | +0.456 |
ibaba | +0.423 |
muzzle | +0.412 |
ICLE | +0.394 |
smart | +0.392 |
Neg logit contributions | |
�� | -0.389 |
aur | -0.372 |
� | -0.337 |
aires | -0.321 |
�� | -0.306 |
Token | 10 |
Token ID | 940 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
Roose | +0.298 |
ibaba | +0.274 |
muzzle | +0.268 |
rils | +0.254 |
ICLE | +0.253 |
Neg logit contributions | |
�� | -0.280 |
aur | -0.268 |
� | -0.243 |
aires | -0.234 |
�� | -0.221 |
Token | 584 |
Token ID | 46352 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
�� | +0.212 |
aur | +0.198 |
� | +0.182 |
aires | +0.172 |
�� | +0.167 |
Neg logit contributions | |
Roose | -0.221 |
ibaba | -0.200 |
muzzle | -0.198 |
ICLE | -0.188 |
smart | -0.188 |
Token | 487 |
Token ID | 35133 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
Roose | +1.850 |
ibaba | +1.656 |
muzzle | +1.647 |
smart | +1.571 |
ICLE | +1.570 |
Neg logit contributions | |
�� | -1.873 |
aur | -1.726 |
� | -1.589 |
aires | -1.501 |
�� | -1.474 |
Token | x |
Token ID | 2124 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
aur | +2.021 |
�� | +2.007 |
� | +1.823 |
aires | +1.783 |
�� | +1.663 |
Neg logit contributions | |
Roose | -1.786 |
ibaba | -1.681 |
ICLE | -1.550 |
muzzle | -1.541 |
azines | -1.534 |
Token | . |
Token ID | 13 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
Roose | +2.523 |
ibaba | +2.314 |
muzzle | +2.266 |
ICLE | +2.153 |
rils | +2.152 |
Neg logit contributions | |
�� | -2.177 |
aur | -2.104 |
� | -1.891 |
aires | -1.816 |
�� | -1.712 |
Token | x |
Token ID | 87 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
Roose | +0.867 |
ibaba | +0.791 |
muzzle | +0.775 |
ICLE | +0.736 |
rils | +0.730 |
Neg logit contributions | |
�� | -0.715 |
aur | -0.710 |
� | -0.617 |
aires | -0.608 |
ional | -0.555 |
Token | . |
Token ID | 13 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
Roose | +6.387 |
muzzle | +5.672 |
ibaba | +5.573 |
smart | +5.509 |
ICLE | +5.398 |
Neg logit contributions | |
�� | -7.947 |
aur | -6.944 |
� | -6.714 |
�� | -6.368 |
aires | -6.156 |
Token | x |
Token ID | 87 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Roose | +2.816 |
ibaba | +2.551 |
muzzle | +2.514 |
ICLE | +2.388 |
rils | +2.365 |
Neg logit contributions | |
�� | -2.646 |
aur | -2.542 |
� | -2.286 |
aires | -2.208 |
�� | -2.084 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Roose | +8.845 |
smart | +7.986 |
muzzle | +7.768 |
ibaba | +7.465 |
Volume | +7.456 |
Neg logit contributions | |
�� | -15.922 |
� | -13.343 |
�� | -13.145 |
aur | -12.957 |
�� | -12.695 |
Token | 135 |
Token ID | 17059 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Roose | +5.608 |
smart | +4.649 |
muzzle | +4.518 |
ibaba | +4.304 |
ICLE | +4.109 |
Neg logit contributions | |
�� | -15.996 |
aur | -14.488 |
� | -14.278 |
�� | -14.051 |
aires | -13.508 |
Token | 106 |
Token ID | 15696 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
�� | +1.082 |
aur | +1.049 |
� | +0.943 |
aires | +0.906 |
�� | +0.858 |
Neg logit contributions | |
Roose | -1.271 |
ibaba | -1.193 |
muzzle | -1.150 |
ICLE | -1.104 |
rils | -1.103 |
Token | 19 |
Token ID | 1129 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
�� | +1.144 |
aur | +1.138 |
� | +0.998 |
aires | +0.975 |
�� | +0.885 |
Neg logit contributions | |
Roose | -1.448 |
ibaba | -1.393 |
muzzle | -1.338 |
rils | -1.270 |
azines | -1.265 |
Token | 41 |
Token ID | 3901 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
aur | +1.734 |
�� | +1.691 |
� | +1.500 |
aires | +1.483 |
ional | +1.336 |
Neg logit contributions | |
Roose | -2.239 |
ibaba | -2.171 |
muzzle | -2.044 |
mosqu | -1.981 |
rils | -1.971 |
Token | of |
Token ID | 286 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
Roose | +3.945 |
muzzle | +3.429 |
smart | +3.425 |
ibaba | +3.419 |
bang | +3.241 |
Neg logit contributions | |
�� | -5.273 |
aur | -4.783 |
� | -4.470 |
aires | -4.307 |
�� | -4.240 |
Token | sight |
Token ID | 6504 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
�� | +3.557 |
aur | +3.416 |
� | +3.175 |
aires | +3.008 |
ional | +2.880 |
Neg logit contributions | |
Roose | -3.338 |
ibaba | -3.028 |
muzzle | -2.910 |
ICLE | -2.857 |
rils | -2.848 |
Token | does |
Token ID | 857 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
�� | +1.918 |
aur | +1.830 |
� | +1.677 |
aires | +1.599 |
�� | +1.538 |
Neg logit contributions | |
Roose | -1.762 |
ibaba | -1.588 |
muzzle | -1.547 |
ICLE | -1.475 |
rils | -1.464 |
Token | not |
Token ID | 407 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
Roose | +2.636 |
muzzle | +2.391 |
ibaba | +2.345 |
bang | +2.180 |
Volume | +2.140 |
Neg logit contributions | |
�� | -4.453 |
aur | -4.149 |
� | -3.936 |
aires | -3.809 |
�� | -3.723 |
Token | mean |
Token ID | 1612 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
Roose | +4.883 |
muzzle | +4.634 |
ibaba | +4.414 |
bang | +4.269 |
Volume | +4.212 |
Neg logit contributions | |
�� | -5.174 |
aur | -4.733 |
� | -4.517 |
aires | -4.186 |
�� | -4.184 |
Token | out |
Token ID | 503 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Roose | +4.037 |
muzzle | +3.751 |
bang | +3.516 |
ibaba | +3.498 |
smart | +3.395 |
Neg logit contributions | |
�� | -4.793 |
aur | -4.480 |
� | -4.170 |
aires | -4.006 |
�� | -3.954 |
Token | of |
Token ID | 286 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
smart | +7.255 |
Roose | +6.625 |
muzzle | +5.894 |
bang | +5.834 |
ICLE | +5.320 |
Neg logit contributions | |
�� | -16.377 |
aur | -14.292 |
� | -14.077 |
�� | -13.685 |
�� | -13.343 |
Token | the |
Token ID | 262 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
Roose | +4.901 |
muzzle | +4.437 |
bang | +4.111 |
smart | +3.944 |
ibaba | +3.921 |
Neg logit contributions | |
�� | -8.562 |
aur | -7.947 |
� | -7.478 |
�� | -7.223 |
aires | -7.071 |
Token | ocean |
Token ID | 9151 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
�� | +2.015 |
aur | +1.918 |
� | +1.775 |
aires | +1.718 |
�� | +1.640 |
Neg logit contributions | |
Roose | -1.612 |
ibaba | -1.462 |
muzzle | -1.403 |
rils | -1.339 |
smart | -1.338 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-03 |
Pos logit contributions | |
Roose | +2.079 |
muzzle | +1.911 |
ibaba | +1.897 |
smart | +1.799 |
ICLE | +1.785 |
Neg logit contributions | |
�� | -2.351 |
aur | -2.154 |
� | -2.021 |
aires | -1.921 |
�� | -1.870 |
Token | Life |
Token ID | 5155 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
Roose | +0.011 |
ibaba | +0.010 |
muzzle | +0.010 |
rils | +0.009 |
ICLE | +0.009 |
Neg logit contributions | |
�� | -0.012 |
aur | -0.012 |
� | -0.011 |
aires | -0.010 |
�� | -0.010 |
Token | Le |
Token ID | 1004 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Roose | +0.571 |
ibaba | +0.487 |
muzzle | +0.485 |
azines | +0.453 |
Volume | +0.451 |
Neg logit contributions | |
�� | -0.846 |
aur | -0.789 |
� | -0.736 |
aires | -0.710 |
�� | -0.695 |
Token | a |
Token ID | 64 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
Roose | +0.998 |
ibaba | +0.888 |
muzzle | +0.874 |
ICLE | +0.828 |
rils | +0.808 |
Neg logit contributions | |
aur | -1.317 |
�� | -1.313 |
� | -1.191 |
aires | -1.155 |
ional | -1.084 |
Token | Suzuki |
Token ID | 35807 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
aur | +1.012 |
�� | +1.003 |
� | +0.910 |
aires | +0.891 |
�� | +0.797 |
Neg logit contributions | |
Roose | -1.117 |
ibaba | -1.040 |
muzzle | -1.034 |
rils | -0.998 |
ICLE | -0.973 |
Token | , |
Token ID | 11 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
Roose | +0.981 |
ibaba | +0.922 |
muzzle | +0.885 |
rils | +0.856 |
ICLE | +0.849 |
Neg logit contributions | |
�� | -0.776 |
aur | -0.769 |
� | -0.684 |
aires | -0.655 |
ional | -0.610 |
Token | The |
Token ID | 383 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
Roose | +0.436 |
ibaba | +0.410 |
muzzle | +0.389 |
rils | +0.376 |
azines | +0.368 |
Neg logit contributions | |
�� | -0.541 |
aur | -0.534 |
� | -0.491 |
aires | -0.471 |
�� | -0.444 |
Token | Chronicle |
Token ID | 14160 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Roose | +1.407 |
ibaba | +1.126 |
muzzle | +1.042 |
rils | +0.893 |
smart | +0.892 |
Neg logit contributions | |
�� | -5.411 |
aur | -5.264 |
� | -4.976 |
aires | -4.845 |
�� | -4.727 |
Token | 12 |
Token ID | 1105 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Roose | +7.638 |
Volume | +6.518 |
muzzle | +6.152 |
Alm | +6.011 |
azines | +5.892 |
Neg logit contributions | |
�� | -17.438 |
aur | -14.802 |
� | -14.673 |
�� | -14.180 |
aires | -13.823 |
Token | of |
Token ID | 286 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
aur | +0.848 |
�� | +0.807 |
� | +0.730 |
aires | +0.723 |
ional | +0.671 |
Neg logit contributions | |
Roose | -1.008 |
ibaba | -0.958 |
mosqu | -0.911 |
muzzle | -0.891 |
rils | -0.881 |
Token | 15 |
Token ID | 1315 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
aur | +1.989 |
�� | +1.910 |
� | +1.742 |
aires | +1.695 |
� | +1.527 |
Neg logit contributions | |
Roose | -2.816 |
ibaba | -2.709 |
rils | -2.577 |
mosqu | -2.557 |
muzzle | -2.496 |
Token | Photo |
Token ID | 5555 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
Roose | +0.344 |
ibaba | +0.327 |
muzzle | +0.317 |
rils | +0.310 |
azines | +0.305 |
Neg logit contributions | |
aur | -0.172 |
�� | -0.170 |
� | -0.143 |
aires | -0.138 |
�� | -0.123 |
Token | : |
Token ID | 25 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Roose | +3.550 |
ibaba | +3.198 |
muzzle | +3.147 |
ICLE | +2.972 |
smart | +2.958 |
Neg logit contributions | |
�� | -4.002 |
aur | -3.804 |
� | -3.502 |
aires | -3.351 |
�� | -3.234 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Roose | +7.527 |
Volume | +6.181 |
smart | +6.094 |
muzzle | +6.072 |
Alm | +5.776 |
Neg logit contributions | |
�� | -15.528 |
aur | -13.106 |
� | -13.012 |
�� | -12.954 |
�� | -12.393 |
Token | Twitter |
Token ID | 14254 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
Roose | +6.179 |
smart | +5.775 |
ICLE | +5.289 |
rils | +5.255 |
ibaba | +5.242 |
Neg logit contributions | |
�� | -11.709 |
aur | -9.757 |
�� | -9.744 |
� | -9.562 |
aires | -9.387 |
Token |
|
Token ID | 198 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Roose | +1.838 |
ibaba | +1.624 |
muzzle | +1.574 |
smart | +1.535 |
Volume | +1.525 |
Neg logit contributions | |
�� | -3.121 |
aur | -2.793 |
� | -2.649 |
�� | -2.589 |
aires | -2.571 |
Token |
|
Token ID | 198 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Roose | +6.948 |
muzzle | +5.834 |
Volume | +5.728 |
smart | +5.691 |
ibaba | +5.562 |
Neg logit contributions | |
�� | -12.973 |
aur | -11.287 |
� | -11.041 |
�� | -10.776 |
�� | -10.282 |
Token | More |
Token ID | 5167 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
Roose | +6.353 |
smart | +5.955 |
ICLE | +5.793 |
ibaba | +5.650 |
rils | +5.554 |
Neg logit contributions | |
�� | -11.519 |
aur | -9.849 |
�� | -9.681 |
� | -9.528 |
aires | -9.273 |
Token |
|
Token ID | 198 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Roose | +4.746 |
ibaba | +4.191 |
muzzle | +4.130 |
smart | +3.942 |
ICLE | +3.906 |
Neg logit contributions | |
�� | -6.654 |
aur | -6.103 |
� | -5.779 |
aires | -5.517 |
�� | -5.426 |
Token |
|
Token ID | 198 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
Roose | +7.710 |
Volume | +6.342 |
smart | +6.277 |
muzzle | +6.231 |
ICLE | +5.983 |
Neg logit contributions | |
�� | -17.098 |
� | -14.458 |
aur | -14.457 |
�� | -14.284 |
�� | -13.827 |
Token | Tumblr |
Token ID | 39415 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
Roose | +7.032 |
smart | +6.566 |
ICLE | +6.267 |
ibaba | +6.144 |
rils | +6.067 |
Neg logit contributions | |
�� | -13.424 |
aur | -11.285 |
� | -11.084 |
�� | -11.049 |
equival | -10.594 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Roose | +4.661 |
ibaba | +4.055 |
muzzle | +4.038 |
ICLE | +3.843 |
smart | +3.839 |
Neg logit contributions | |
�� | -6.500 |
aur | -5.997 |
� | -5.631 |
aires | -5.386 |
�� | -5.347 |
Token |
|
Token ID | 198 |
Feature activation | -8.22e-01 |
Pos logit contributions | |
Roose | +7.205 |
muzzle | +6.019 |
Volume | +5.915 |
smart | +5.888 |
ibaba | +5.749 |
Neg logit contributions | |
�� | -13.435 |
aur | -11.674 |
� | -11.440 |
�� | -11.140 |
�� | -10.630 |
Token | Pinterest |
Token ID | 35767 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
Roose | +5.393 |
smart | +4.736 |
ibaba | +4.706 |
ICLE | +4.578 |
muzzle | +4.541 |
Neg logit contributions | |
�� | -8.533 |
aur | -7.516 |
� | -7.171 |
�� | -6.999 |
aires | -6.943 |
Token | Le |
Token ID | 1004 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
Roose | +0.739 |
ibaba | +0.629 |
muzzle | +0.625 |
azines | +0.583 |
Volume | +0.581 |
Neg logit contributions | |
�� | -1.125 |
aur | -1.049 |
� | -0.980 |
aires | -0.945 |
�� | -0.926 |
Token | a |
Token ID | 64 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
Roose | +1.360 |
ibaba | +1.206 |
muzzle | +1.188 |
ICLE | +1.124 |
rils | +1.098 |
Neg logit contributions | |
�� | -1.842 |
aur | -1.829 |
� | -1.663 |
aires | -1.611 |
�� | -1.520 |
Token | Suzuki |
Token ID | 35807 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
aur | +0.753 |
�� | +0.750 |
� | +0.679 |
aires | +0.663 |
�� | +0.597 |
Neg logit contributions | |
Roose | -0.822 |
ibaba | -0.763 |
muzzle | -0.760 |
rils | -0.731 |
ICLE | -0.715 |
Token | , |
Token ID | 11 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
Roose | +1.275 |
ibaba | +1.194 |
muzzle | +1.148 |
rils | +1.108 |
ICLE | +1.100 |
Neg logit contributions | |
�� | -1.027 |
aur | -1.013 |
� | -0.903 |
aires | -0.865 |
ional | -0.806 |
Token | The |
Token ID | 383 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
Roose | +0.699 |
ibaba | +0.654 |
muzzle | +0.622 |
rils | +0.599 |
azines | +0.588 |
Neg logit contributions | |
�� | -0.885 |
aur | -0.871 |
� | -0.802 |
aires | -0.770 |
�� | -0.728 |
Token | Chronicle |
Token ID | 14160 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Roose | +1.379 |
ibaba | +1.105 |
muzzle | +1.022 |
rils | +0.877 |
smart | +0.875 |
Neg logit contributions | |
�� | -5.282 |
aur | -5.140 |
� | -4.857 |
aires | -4.730 |
�� | -4.614 |
Token | 13 |
Token ID | 1511 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
Roose | +7.682 |
muzzle | +6.578 |
Volume | +6.526 |
smart | +6.467 |
bang | +6.197 |
Neg logit contributions | |
�� | -16.098 |
aur | -13.683 |
� | -13.577 |
�� | -12.963 |
�� | -12.739 |
Token | of |
Token ID | 286 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
aur | +0.094 |
�� | +0.091 |
� | +0.082 |
aires | +0.081 |
ional | +0.075 |
Neg logit contributions | |
Roose | -0.107 |
ibaba | -0.101 |
muzzle | -0.094 |
mosqu | -0.094 |
rils | -0.092 |
Token | 15 |
Token ID | 1315 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
aur | +2.570 |
�� | +2.430 |
� | +2.231 |
aires | +2.168 |
� | +1.968 |
Neg logit contributions | |
Roose | -3.725 |
ibaba | -3.597 |
rils | -3.434 |
mosqu | -3.421 |
muzzle | -3.297 |
Token | Photo |
Token ID | 5555 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
Roose | +0.584 |
ibaba | +0.554 |
muzzle | +0.538 |
rils | +0.525 |
azines | +0.517 |
Neg logit contributions | |
aur | -0.297 |
�� | -0.296 |
� | -0.248 |
aires | -0.239 |
�� | -0.214 |
Token | : |
Token ID | 25 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
Roose | +3.606 |
ibaba | +3.244 |
muzzle | +3.195 |
ICLE | +3.017 |
smart | +3.004 |
Neg logit contributions | |
�� | -4.087 |
aur | -3.880 |
� | -3.574 |
aires | -3.419 |
�� | -3.304 |
Token | i |
Token ID | 72 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
Roose | +1.795 |
ibaba | +1.623 |
muzzle | +1.608 |
ICLE | +1.532 |
smart | +1.530 |
Neg logit contributions | |
�� | -1.779 |
aur | -1.657 |
� | -1.528 |
aires | -1.453 |
�� | -1.415 |
Token | ] |
Token ID | 60 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Roose | +2.219 |
muzzle | +1.990 |
ibaba | +1.987 |
smart | +1.895 |
Volume | +1.886 |
Neg logit contributions | |
�� | -2.734 |
aur | -2.507 |
� | -2.332 |
aires | -2.251 |
�� | -2.204 |
Token | and |
Token ID | 290 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
�� | +1.116 |
aur | +1.057 |
� | +0.976 |
aires | +0.932 |
�� | +0.902 |
Neg logit contributions | |
Roose | -0.983 |
ibaba | -0.881 |
muzzle | -0.870 |
ICLE | -0.822 |
smart | -0.818 |
Token | 33 |
Token ID | 4747 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
�� | +0.346 |
aur | +0.335 |
� | +0.299 |
aires | +0.284 |
�� | +0.267 |
Neg logit contributions | |
Roose | -0.484 |
ibaba | -0.456 |
muzzle | -0.439 |
rils | -0.428 |
ICLE | -0.428 |
Token | percent |
Token ID | 1411 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
Roose | +2.701 |
muzzle | +2.429 |
ibaba | +2.297 |
disappro | +2.191 |
ICLE | +2.166 |
Neg logit contributions | |
�� | -5.830 |
aur | -5.406 |
� | -4.932 |
aires | -4.863 |
�� | -4.811 |
Token | [ |
Token ID | 58 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Roose | +2.153 |
muzzle | +1.920 |
ibaba | +1.899 |
smart | +1.792 |
ICLE | +1.788 |
Neg logit contributions | |
�� | -2.862 |
aur | -2.666 |
� | -2.503 |
aires | -2.400 |
�� | -2.347 |
Token | ii |
Token ID | 4178 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
Roose | +6.407 |
smart | +6.144 |
ibaba | +5.984 |
iera | +5.800 |
Volume | +5.534 |
Neg logit contributions | |
�� | -14.876 |
aur | -13.343 |
� | -12.845 |
aires | -12.715 |
�� | -12.661 |
Token | ] |
Token ID | 60 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Roose | +1.847 |
ibaba | +1.660 |
muzzle | +1.648 |
smart | +1.576 |
Volume | +1.559 |
Neg logit contributions | |
�� | -2.327 |
aur | -2.122 |
� | -1.973 |
aires | -1.926 |
�� | -1.888 |
Token | depending |
Token ID | 6906 |
Feature activation | +9.51e-03 |
Pos logit contributions | |
Roose | +1.224 |
ibaba | +1.097 |
muzzle | +1.089 |
smart | +1.027 |
ICLE | +1.026 |
Neg logit contributions | |
�� | -1.375 |
aur | -1.291 |
� | -1.192 |
aires | -1.139 |
�� | -1.107 |
Token | on |
Token ID | 319 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
�� | +0.089 |
aur | +0.082 |
� | +0.076 |
aires | +0.074 |
�� | +0.072 |
Neg logit contributions | |
Roose | -0.075 |
muzzle | -0.068 |
ibaba | -0.067 |
smart | -0.064 |
Volume | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Roose | +0.339 |
ibaba | +0.304 |
muzzle | +0.300 |
ICLE | +0.283 |
smart | +0.282 |
Neg logit contributions | |
�� | -0.399 |
aur | -0.378 |
� | -0.349 |
aires | -0.334 |
�� | -0.324 |
Token | should |
Token ID | 815 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
�� | +2.792 |
aur | +2.692 |
� | +2.460 |
aires | +2.352 |
�� | +2.252 |
Neg logit contributions | |
Roose | -2.789 |
ibaba | -2.632 |
muzzle | -2.510 |
ICLE | -2.421 |
azines | -2.419 |
Token | now |
Token ID | 783 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
�� | +1.630 |
aur | +1.580 |
� | +1.446 |
aires | +1.392 |
ional | +1.309 |
Neg logit contributions | |
Roose | -1.540 |
ibaba | -1.388 |
muzzle | -1.320 |
ICLE | -1.299 |
azines | -1.292 |
Token | decide |
Token ID | 5409 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
�� | +1.644 |
aur | +1.600 |
� | +1.457 |
aires | +1.413 |
ional | +1.328 |
Neg logit contributions | |
Roose | -1.564 |
ibaba | -1.415 |
ICLE | -1.337 |
muzzle | -1.332 |
azines | -1.319 |
Token | once |
Token ID | 1752 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
�� | +2.288 |
aur | +2.220 |
� | +2.025 |
aires | +1.941 |
ional | +1.866 |
Neg logit contributions | |
Roose | -2.349 |
ibaba | -2.195 |
muzzle | -2.072 |
azines | -2.007 |
ICLE | -1.998 |
Token | and |
Token ID | 290 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
Roose | +2.196 |
ibaba | +1.936 |
muzzle | +1.925 |
smart | +1.800 |
ICLE | +1.799 |
Neg logit contributions | |
�� | -2.898 |
aur | -2.729 |
� | -2.540 |
aires | -2.426 |
�� | -2.370 |
Token | for |
Token ID | 329 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Roose | +2.229 |
ibaba | +1.998 |
muzzle | +1.975 |
ICLE | +1.864 |
smart | +1.858 |
Neg logit contributions | |
�� | -2.550 |
aur | -2.411 |
� | -2.226 |
aires | -2.126 |
�� | -2.062 |
Token | all |
Token ID | 477 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
Roose | +7.758 |
muzzle | +7.312 |
bang | +7.133 |
smart | +6.766 |
ibaba | +6.678 |
Neg logit contributions | |
�� | -15.380 |
aur | -13.370 |
� | -13.214 |
�� | -12.741 |
aires | -12.558 |
Token | whether |
Token ID | 1771 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
Roose | +2.722 |
muzzle | +2.420 |
ibaba | +2.349 |
smart | +2.231 |
ICLE | +2.228 |
Neg logit contributions | |
�� | -3.610 |
aur | -3.362 |
� | -3.132 |
aires | -2.988 |
�� | -2.933 |
Token | it |
Token ID | 340 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
Roose | +2.389 |
muzzle | +2.103 |
ibaba | +2.077 |
smart | +1.957 |
ICLE | +1.948 |
Neg logit contributions | |
�� | -3.145 |
aur | -2.938 |
� | -2.727 |
aires | -2.607 |
�� | -2.562 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
�� | +0.183 |
aur | +0.174 |
� | +0.160 |
aires | +0.154 |
�� | +0.148 |
Neg logit contributions | |
Roose | -0.167 |
ibaba | -0.151 |
muzzle | -0.147 |
ICLE | -0.141 |
smart | -0.140 |
Token | to |
Token ID | 284 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
Roose | +0.220 |
ibaba | +0.197 |
muzzle | +0.195 |
ICLE | +0.184 |
smart | +0.183 |
Neg logit contributions | |
�� | -0.260 |
aur | -0.245 |
� | -0.226 |
aires | -0.217 |
�� | -0.211 |
Token | 203 |
Token ID | 22416 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Roose | +2.475 |
ibaba | +2.209 |
muzzle | +2.201 |
ICLE | +2.103 |
smart | +2.102 |
Neg logit contributions | |
�� | -2.517 |
aur | -2.311 |
� | -2.131 |
aires | -2.009 |
�� | -1.976 |
Token | x |
Token ID | 2124 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
aur | +1.617 |
�� | +1.611 |
� | +1.462 |
aires | +1.432 |
ional | +1.336 |
Neg logit contributions | |
Roose | -1.419 |
ibaba | -1.325 |
muzzle | -1.222 |
ICLE | -1.212 |
azines | -1.205 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Roose | +3.841 |
ibaba | +3.473 |
muzzle | +3.441 |
ICLE | +3.269 |
smart | +3.261 |
Neg logit contributions | |
�� | -3.699 |
aur | -3.462 |
� | -3.179 |
aires | -3.015 |
�� | -2.923 |
Token | x |
Token ID | 87 |
Feature activation | -9.22e-01 |
Pos logit contributions | |
Roose | +1.684 |
ibaba | +1.534 |
muzzle | +1.507 |
ICLE | +1.430 |
rils | +1.418 |
Neg logit contributions | |
�� | -1.463 |
aur | -1.434 |
� | -1.265 |
aires | -1.234 |
�� | -1.141 |
Token | . |
Token ID | 13 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
Roose | +6.833 |
muzzle | +6.062 |
ibaba | +5.929 |
smart | +5.926 |
ICLE | +5.783 |
Neg logit contributions | |
�� | -9.021 |
aur | -7.791 |
� | -7.600 |
�� | -7.242 |
aires | -6.950 |
Token | x |
Token ID | 87 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
Roose | +5.792 |
ibaba | +5.221 |
muzzle | +5.179 |
smart | +5.026 |
ICLE | +4.903 |
Neg logit contributions | |
�� | -6.900 |
aur | -6.264 |
� | -5.969 |
aires | -5.593 |
�� | -5.587 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
Roose | +9.121 |
smart | +8.082 |
muzzle | +8.081 |
ibaba | +8.005 |
Volume | +7.693 |
Neg logit contributions | |
�� | -14.103 |
aur | -11.764 |
� | -11.642 |
�� | -11.350 |
�� | -11.157 |
Token | 112 |
Token ID | 14686 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
Roose | +6.539 |
muzzle | +5.823 |
ibaba | +5.817 |
smart | +5.616 |
rils | +5.602 |
Neg logit contributions | |
�� | -10.920 |
aur | -9.864 |
� | -9.457 |
aires | -9.394 |
�� | -9.044 |
Token | 10 |
Token ID | 838 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
Roose | +1.524 |
ibaba | +1.386 |
muzzle | +1.370 |
ICLE | +1.303 |
smart | +1.299 |
Neg logit contributions | |
�� | -1.386 |
aur | -1.302 |
� | -1.189 |
aires | -1.129 |
�� | -1.089 |
Token | 25 |
Token ID | 1495 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
�� | +0.492 |
aur | +0.476 |
� | +0.426 |
aires | +0.412 |
�� | +0.386 |
Neg logit contributions | |
Roose | -0.580 |
ibaba | -0.537 |
muzzle | -0.526 |
rils | -0.497 |
azines | -0.496 |
Token | 151 |
Token ID | 24309 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
Roose | +1.101 |
ibaba | +0.991 |
muzzle | +0.983 |
smart | +0.936 |
ICLE | +0.936 |
Neg logit contributions | |
�� | -1.100 |
aur | -1.021 |
� | -0.939 |
aires | -0.888 |
�� | -0.868 |
Token | . |
Token ID | 13 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
Roose | +2.076 |
muzzle | +1.858 |
ibaba | +1.853 |
smart | +1.751 |
ICLE | +1.748 |
Neg logit contributions | |
�� | -2.291 |
aur | -2.130 |
� | -1.975 |
aires | -1.882 |
�� | -1.840 |
Token | If |
Token ID | 1002 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
Roose | +0.515 |
ibaba | +0.468 |
muzzle | +0.457 |
ICLE | +0.436 |
rils | +0.433 |
Neg logit contributions | |
�� | -0.561 |
aur | -0.533 |
� | -0.490 |
aires | -0.468 |
�� | -0.451 |
Token | ' |
Token ID | 705 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
Roose | +0.490 |
ibaba | +0.441 |
muzzle | +0.435 |
ICLE | +0.413 |
smart | +0.410 |
Neg logit contributions | |
�� | -0.526 |
aur | -0.498 |
� | -0.459 |
aires | -0.438 |
�� | -0.423 |
Token | mod |
Token ID | 4666 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
Roose | +4.036 |
smart | +3.664 |
ibaba | +3.662 |
muzzle | +3.645 |
azines | +3.511 |
Neg logit contributions | |
�� | -3.719 |
aur | -3.339 |
� | -3.119 |
aires | -2.911 |
�� | -2.903 |
Token | eline |
Token ID | 4470 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
Roose | +0.417 |
ibaba | +0.381 |
muzzle | +0.377 |
rils | +0.360 |
ICLE | +0.360 |
Neg logit contributions | |
�� | -0.404 |
aur | -0.374 |
� | -0.347 |
aires | -0.324 |
�� | -0.322 |
Token | ' |
Token ID | 6 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Roose | +3.849 |
muzzle | +3.371 |
ibaba | +3.361 |
smart | +3.316 |
azines | +3.244 |
Neg logit contributions | |
�� | -6.113 |
aur | -5.632 |
� | -5.306 |
�� | -5.078 |
aires | -5.061 |
Token | is |
Token ID | 318 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
Roose | +7.943 |
muzzle | +7.509 |
bang | +7.300 |
Volume | +7.207 |
smart | +6.983 |
Neg logit contributions | |
�� | -14.042 |
aur | -12.237 |
�� | -11.683 |
� | -11.674 |
�� | -11.389 |
Token | off |
Token ID | 572 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
Roose | +5.975 |
muzzle | +5.420 |
Volume | +5.083 |
bang | +5.078 |
smart | +4.945 |
Neg logit contributions | |
�� | -9.259 |
aur | -8.293 |
� | -7.817 |
�� | -7.684 |
aires | -7.596 |
Token | or |
Token ID | 393 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
Roose | +5.229 |
muzzle | +4.773 |
ibaba | +4.734 |
smart | +4.573 |
bang | +4.534 |
Neg logit contributions | |
�� | -6.774 |
aur | -6.228 |
� | -5.801 |
aires | -5.656 |
�� | -5.636 |
Token | ' |
Token ID | 705 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
Roose | +2.524 |
muzzle | +2.264 |
ibaba | +2.235 |
smart | +2.114 |
ICLE | +2.109 |
Neg logit contributions | |
�� | -3.016 |
aur | -2.825 |
� | -2.607 |
aires | -2.498 |
�� | -2.450 |
Token | model |
Token ID | 19849 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
Roose | +2.643 |
ibaba | +2.346 |
muzzle | +2.334 |
smart | +2.308 |
azines | +2.235 |
Neg logit contributions | |
�� | -3.262 |
aur | -2.997 |
� | -2.822 |
aires | -2.663 |
�� | -2.657 |
Token | the |
Token ID | 262 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
Roose | +3.431 |
muzzle | +3.047 |
ibaba | +2.998 |
smart | +2.884 |
azines | +2.868 |
Neg logit contributions | |
�� | -4.262 |
aur | -4.041 |
� | -3.730 |
aires | -3.559 |
�� | -3.484 |
Token | industry |
Token ID | 2831 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
Roose | +2.654 |
muzzle | +2.352 |
ibaba | +2.345 |
azines | +2.228 |
smart | +2.220 |
Neg logit contributions | |
�� | -2.990 |
aur | -2.846 |
� | -2.616 |
aires | -2.502 |
�� | -2.425 |
Token | , |
Token ID | 11 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
Roose | +3.778 |
muzzle | +3.488 |
ibaba | +3.395 |
smart | +3.298 |
bang | +3.276 |
Neg logit contributions | |
�� | -4.796 |
aur | -4.389 |
� | -4.122 |
aires | -3.894 |
�� | -3.841 |
Token | took |
Token ID | 1718 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
Roose | +1.223 |
ibaba | +1.097 |
muzzle | +1.083 |
ICLE | +1.023 |
smart | +1.019 |
Neg logit contributions | |
�� | -1.398 |
aur | -1.321 |
� | -1.220 |
aires | -1.165 |
�� | -1.130 |
Token | a |
Token ID | 257 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
Roose | +2.173 |
muzzle | +1.926 |
ibaba | +1.926 |
smart | +1.799 |
Volume | +1.795 |
Neg logit contributions | |
�� | -2.629 |
aur | -2.472 |
� | -2.285 |
aires | -2.192 |
�� | -2.137 |
Token | much |
Token ID | 881 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Roose | +4.721 |
muzzle | +4.288 |
ibaba | +4.177 |
bang | +4.060 |
smart | +3.944 |
Neg logit contributions | |
�� | -5.965 |
aur | -5.566 |
� | -5.185 |
aires | -4.889 |
�� | -4.829 |
Token | needed |
Token ID | 2622 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
Roose | +7.956 |
muzzle | +7.046 |
bang | +6.937 |
disappro | +6.778 |
shel | +6.563 |
Neg logit contributions | |
�� | -14.385 |
aur | -12.894 |
� | -12.228 |
�� | -11.915 |
aires | -11.703 |
Token | break |
Token ID | 2270 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
Roose | +1.456 |
muzzle | +1.290 |
ibaba | +1.288 |
smart | +1.202 |
ICLE | +1.196 |
Neg logit contributions | |
�� | -1.911 |
aur | -1.801 |
� | -1.675 |
aires | -1.600 |
�� | -1.567 |
Token | from |
Token ID | 422 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
Roose | +0.635 |
ibaba | +0.572 |
muzzle | +0.565 |
ICLE | +0.535 |
smart | +0.533 |
Neg logit contributions | |
�� | -0.683 |
aur | -0.644 |
� | -0.593 |
aires | -0.566 |
�� | -0.548 |
Token | their |
Token ID | 511 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Roose | +1.196 |
ibaba | +1.071 |
muzzle | +1.063 |
ICLE | +1.001 |
smart | +1.001 |
Neg logit contributions | |
�� | -1.322 |
aur | -1.248 |
� | -1.151 |
aires | -1.099 |
�� | -1.065 |
Token | summer |
Token ID | 3931 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
�� | +1.876 |
aur | +1.779 |
� | +1.626 |
aires | +1.553 |
�� | +1.480 |
Neg logit contributions | |
Roose | -1.807 |
ibaba | -1.669 |
muzzle | -1.588 |
ICLE | -1.541 |
rils | -1.521 |
Token | title |
Token ID | 3670 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
�� | +4.121 |
aur | +3.950 |
� | +3.606 |
aires | +3.500 |
ional | +3.338 |
Neg logit contributions | |
Roose | -3.842 |
ibaba | -3.550 |
muzzle | -3.374 |
rils | -3.321 |
azines | -3.305 |
Token | made |
Token ID | 925 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
�� | +1.948 |
aur | +1.851 |
� | +1.697 |
aires | +1.633 |
�� | +1.563 |
Neg logit contributions | |
Roose | -1.801 |
ibaba | -1.646 |
muzzle | -1.605 |
ICLE | -1.539 |
rils | -1.535 |
Token | not |
Token ID | 407 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
Roose | +2.156 |
ibaba | +1.911 |
muzzle | +1.909 |
smart | +1.780 |
ICLE | +1.778 |
Neg logit contributions | |
�� | -2.506 |
aur | -2.355 |
� | -2.168 |
aires | -2.086 |
�� | -2.018 |
Token | one |
Token ID | 530 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
Roose | +4.424 |
ibaba | +3.977 |
muzzle | +3.922 |
bang | +3.623 |
ICLE | +3.595 |
Neg logit contributions | |
�� | -6.696 |
aur | -6.246 |
� | -5.760 |
aires | -5.595 |
�� | -5.432 |
Token | , |
Token ID | 11 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
�� | +0.170 |
aur | +0.161 |
� | +0.148 |
aires | +0.142 |
�� | +0.137 |
Neg logit contributions | |
Roose | -0.149 |
ibaba | -0.134 |
muzzle | -0.132 |
ICLE | -0.125 |
smart | -0.124 |
Token | but |
Token ID | 475 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Roose | +4.010 |
muzzle | +3.658 |
ibaba | +3.575 |
bang | +3.392 |
smart | +3.346 |
Neg logit contributions | |
�� | -6.330 |
aur | -5.879 |
� | -5.480 |
aires | -5.366 |
�� | -5.192 |
Token | two |
Token ID | 734 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
Roose | +8.026 |
muzzle | +7.270 |
ibaba | +7.220 |
Volume | +7.013 |
bang | +6.829 |
Neg logit contributions | |
�� | -13.746 |
aur | -12.530 |
aires | -11.769 |
� | -11.637 |
�� | -11.125 |
Token | appearances |
Token ID | 11057 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
�� | +4.160 |
aur | +4.047 |
� | +3.707 |
aires | +3.583 |
�� | +3.426 |
Neg logit contributions | |
Roose | -3.719 |
ibaba | -3.351 |
muzzle | -3.200 |
ICLE | -3.180 |
rils | -3.167 |
Token | during |
Token ID | 1141 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
Roose | +0.607 |
ibaba | +0.549 |
muzzle | +0.545 |
ICLE | +0.514 |
smart | +0.513 |
Neg logit contributions | |
�� | -0.611 |
aur | -0.572 |
� | -0.526 |
aires | -0.500 |
�� | -0.485 |
Token | the |
Token ID | 262 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
�� | +0.232 |
aur | +0.219 |
� | +0.202 |
aires | +0.192 |
�� | +0.186 |
Neg logit contributions | |
Roose | -0.220 |
ibaba | -0.199 |
muzzle | -0.196 |
ICLE | -0.186 |
smart | -0.185 |
Token | pre |
Token ID | 662 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
�� | +2.522 |
aur | +2.393 |
� | +2.213 |
aires | +2.106 |
�� | +2.034 |
Neg logit contributions | |
Roose | -2.317 |
ibaba | -2.103 |
muzzle | -2.035 |
ICLE | -1.961 |
smart | -1.952 |
Token | informed |
Token ID | 35698 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
�� | +1.552 |
aur | +1.499 |
� | +1.368 |
aires | +1.314 |
�� | +1.263 |
Neg logit contributions | |
Roose | -1.312 |
ibaba | -1.185 |
muzzle | -1.162 |
ICLE | -1.104 |
rils | -1.088 |
Token | non |
Token ID | 1729 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
�� | +3.391 |
aur | +3.308 |
� | +2.983 |
aires | +2.857 |
�� | +2.738 |
Neg logit contributions | |
Roose | -3.537 |
ibaba | -3.256 |
muzzle | -3.125 |
ICLE | -3.048 |
rils | -3.021 |
Token | |
Token ID | 3907 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
�� | +2.108 |
aur | +2.000 |
� | +1.854 |
aires | +1.774 |
�� | +1.721 |
Neg logit contributions | |
Roose | -1.669 |
ibaba | -1.488 |
muzzle | -1.467 |
ICLE | -1.382 |
smart | -1.373 |
Token | int |
Token ID | 600 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
�� | +8.456 |
aur | +8.363 |
� | +7.666 |
aires | +7.376 |
ional | +7.012 |
Neg logit contributions | |
Roose | -6.679 |
ibaba | -5.870 |
muzzle | -5.713 |
rils | -5.540 |
ICLE | -5.388 |
Token | ellectual |
Token ID | 29706 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
�� | +1.710 |
aur | +1.619 |
� | +1.445 |
aires | +1.365 |
�� | +1.300 |
Neg logit contributions | |
Roose | -2.419 |
ibaba | -2.206 |
muzzle | -2.191 |
ICLE | -2.094 |
azines | -2.078 |
Token | I |
Token ID | 314 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
�� | +3.596 |
aur | +3.491 |
� | +3.158 |
aires | +3.049 |
�� | +2.888 |
Neg logit contributions | |
Roose | -3.480 |
ibaba | -3.146 |
muzzle | -3.045 |
ICLE | -2.952 |
rils | -2.931 |
Token | � |
Token ID | 447 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
aur | +6.826 |
�� | +6.602 |
� | +6.159 |
aires | +5.517 |
�� | +5.436 |
Neg logit contributions | |
Roose | -8.159 |
ibaba | -7.610 |
muzzle | -7.126 |
ICLE | -7.015 |
azines | -6.966 |
Token | � |
Token ID | 247 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
�� | +0.632 |
aur | +0.579 |
� | +0.504 |
aires | +0.470 |
�� | +0.442 |
Neg logit contributions | |
Roose | -1.214 |
ibaba | -1.125 |
muzzle | -1.116 |
ICLE | -1.074 |
smart | -1.070 |
Token | ve |
Token ID | 303 |
Feature activation | +7.67e-01 |
Pos logit contributions | |
�� | +3.088 |
aur | +3.028 |
� | +2.740 |
aires | +2.576 |
�� | +2.494 |
Neg logit contributions | |
Roose | -3.310 |
ibaba | -2.971 |
muzzle | -2.920 |
ICLE | -2.773 |
azines | -2.763 |
Token | ever |
Token ID | 1683 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
aur | +5.623 |
�� | +5.532 |
� | +5.177 |
�� | +4.622 |
aires | +4.618 |
Neg logit contributions | |
Roose | -6.944 |
ibaba | -6.488 |
muzzle | -6.070 |
azines | -5.964 |
ICLE | -5.900 |
Token | met |
Token ID | 1138 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
�� | +3.983 |
aur | +3.961 |
� | +3.658 |
aires | +3.424 |
�� | +3.339 |
Neg logit contributions | |
Roose | -3.727 |
ibaba | -3.387 |
muzzle | -3.210 |
azines | -3.117 |
rils | -3.109 |
Token | cattle |
Token ID | 17025 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Roose | +3.755 |
muzzle | +3.308 |
ibaba | +3.238 |
smart | +3.038 |
bang | +3.018 |
Neg logit contributions | |
�� | -5.554 |
aur | -5.200 |
� | -4.854 |
aires | -4.660 |
�� | -4.580 |
Token | , |
Token ID | 11 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
�� | +1.400 |
aur | +1.339 |
� | +1.225 |
aires | +1.166 |
�� | +1.120 |
Neg logit contributions | |
Roose | -1.444 |
ibaba | -1.315 |
muzzle | -1.284 |
ICLE | -1.235 |
rils | -1.222 |
Token | and |
Token ID | 290 |
Feature activation | -7.99e-02 |
Pos logit contributions | |
�� | +0.975 |
aur | +0.929 |
� | +0.854 |
aires | +0.815 |
�� | +0.787 |
Neg logit contributions | |
Roose | -0.895 |
ibaba | -0.803 |
muzzle | -0.789 |
ICLE | -0.755 |
rils | -0.747 |
Token | all |
Token ID | 477 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
Roose | +0.652 |
ibaba | +0.585 |
muzzle | +0.579 |
ICLE | +0.545 |
smart | +0.545 |
Neg logit contributions | |
�� | -0.724 |
aur | -0.684 |
� | -0.630 |
aires | -0.602 |
�� | -0.583 |
Token | that |
Token ID | 326 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
Roose | +1.631 |
ibaba | +1.461 |
muzzle | +1.453 |
smart | +1.362 |
ICLE | +1.357 |
Neg logit contributions | |
�� | -1.843 |
aur | -1.728 |
� | -1.593 |
aires | -1.526 |
�� | -1.478 |
Token | is |
Token ID | 318 |
Feature activation | +9.04e-01 |
Pos logit contributions | |
�� | +3.981 |
aur | +3.905 |
� | +3.543 |
aires | +3.357 |
�� | +3.264 |
Neg logit contributions | |
Roose | -3.727 |
ibaba | -3.342 |
muzzle | -3.235 |
ICLE | -3.192 |
smart | -3.125 |
Token | in |
Token ID | 287 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
�� | +7.110 |
aur | +7.065 |
� | +6.365 |
aires | +5.877 |
�� | +5.824 |
Neg logit contributions | |
Roose | -7.400 |
ibaba | -6.783 |
ICLE | -6.438 |
azines | -6.428 |
rils | -6.427 |
Token | the |
Token ID | 262 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
�� | +0.542 |
aur | +0.512 |
� | +0.472 |
aires | +0.450 |
�� | +0.436 |
Neg logit contributions | |
Roose | -0.499 |
ibaba | -0.449 |
muzzle | -0.443 |
ICLE | -0.421 |
smart | -0.419 |
Token | city |
Token ID | 1748 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
Roose | +0.858 |
ibaba | +0.761 |
muzzle | +0.749 |
ICLE | +0.704 |
smart | +0.702 |
Neg logit contributions | |
�� | -1.160 |
aur | -1.102 |
� | -1.024 |
aires | -0.982 |
�� | -0.955 |
Token | , |
Token ID | 11 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
Roose | +0.425 |
ibaba | +0.383 |
muzzle | +0.380 |
ICLE | +0.359 |
smart | +0.358 |
Neg logit contributions | |
�� | -0.443 |
aur | -0.417 |
� | -0.384 |
aires | -0.366 |
�� | -0.354 |
Token | even |
Token ID | 772 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
�� | +0.719 |
aur | +0.687 |
� | +0.630 |
aires | +0.600 |
�� | +0.578 |
Neg logit contributions | |
Roose | -0.685 |
ibaba | -0.616 |
muzzle | -0.603 |
ICLE | -0.580 |
rils | -0.574 |
Token | number |
Token ID | 1271 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
Roose | +0.683 |
ibaba | +0.603 |
muzzle | +0.596 |
ICLE | +0.557 |
smart | +0.555 |
Neg logit contributions | |
�� | -0.958 |
aur | -0.910 |
� | -0.846 |
aires | -0.812 |
�� | -0.791 |
Token | of |
Token ID | 286 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
�� | +3.573 |
aur | +3.381 |
� | +3.123 |
aires | +2.985 |
�� | +2.894 |
Neg logit contributions | |
Roose | -3.071 |
ibaba | -2.751 |
muzzle | -2.717 |
ICLE | -2.565 |
smart | -2.555 |
Token | injured |
Token ID | 6686 |
Feature activation | +6.80e-01 |
Pos logit contributions | |
�� | +1.016 |
aur | +0.971 |
� | +0.897 |
aires | +0.864 |
�� | +0.823 |
Neg logit contributions | |
Roose | -0.852 |
ibaba | -0.773 |
muzzle | -0.749 |
smart | -0.717 |
ICLE | -0.717 |
Token | has |
Token ID | 468 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
�� | +5.108 |
aur | +5.104 |
� | +4.622 |
aires | +4.518 |
ional | +4.161 |
Neg logit contributions | |
Roose | -5.896 |
ibaba | -5.548 |
muzzle | -5.244 |
ICLE | -5.184 |
azines | -5.126 |
Token | reached |
Token ID | 4251 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
�� | +3.457 |
aur | +3.374 |
� | +3.139 |
aires | +2.946 |
�� | +2.817 |
Neg logit contributions | |
Roose | -3.285 |
ibaba | -3.116 |
muzzle | -2.898 |
ICLE | -2.837 |
smart | -2.812 |
Token | 132 |
Token ID | 21761 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
aur | +5.214 |
�� | +5.183 |
� | +4.817 |
aires | +4.628 |
equival | +4.414 |
Neg logit contributions | |
Roose | -6.515 |
ibaba | -6.200 |
ICLE | -5.870 |
azines | -5.847 |
muzzle | -5.813 |
Token | .] |
Token ID | 8183 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
�� | +7.449 |
aur | +7.273 |
� | +6.759 |
aires | +6.379 |
�� | +5.801 |
Neg logit contributions | |
Roose | -8.806 |
ibaba | -8.385 |
muzzle | -7.967 |
azines | -7.686 |
smart | -7.549 |
Token |
|
Token ID | 198 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
Roose | +0.577 |
ibaba | +0.496 |
muzzle | +0.492 |
Volume | +0.473 |
ICLE | +0.467 |
Neg logit contributions | |
�� | -0.792 |
aur | -0.742 |
� | -0.688 |
aires | -0.660 |
�� | -0.656 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
aur | +0.276 |
�� | +0.266 |
� | +0.240 |
aires | +0.239 |
ional | +0.220 |
Neg logit contributions | |
Roose | -0.311 |
ibaba | -0.292 |
mosqu | -0.278 |
muzzle | -0.278 |
rils | -0.266 |
Token | [ |
Token ID | 58 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
�� | +0.160 |
aur | +0.152 |
� | +0.140 |
aires | +0.133 |
�� | +0.129 |
Neg logit contributions | |
Roose | -0.144 |
ibaba | -0.129 |
muzzle | -0.127 |
ICLE | -0.120 |
smart | -0.120 |
Token | UPDATE |
Token ID | 16977 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
�� | +2.707 |
aur | +2.567 |
� | +2.380 |
aires | +2.278 |
�� | +2.211 |
Neg logit contributions | |
Roose | -2.137 |
ibaba | -1.905 |
muzzle | -1.879 |
ICLE | -1.765 |
smart | -1.759 |
Token | display |
Token ID | 3359 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
�� | +0.221 |
aur | +0.208 |
� | +0.190 |
aires | +0.181 |
�� | +0.175 |
Neg logit contributions | |
Roose | -0.237 |
ibaba | -0.214 |
muzzle | -0.212 |
ICLE | -0.201 |
smart | -0.201 |
Token | at |
Token ID | 379 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
Roose | +2.060 |
muzzle | +1.857 |
ibaba | +1.850 |
smart | +1.749 |
ICLE | +1.739 |
Neg logit contributions | |
�� | -2.084 |
aur | -1.924 |
� | -1.778 |
aires | -1.695 |
�� | -1.638 |
Token | the |
Token ID | 262 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
Roose | +2.565 |
muzzle | +2.278 |
ibaba | +2.238 |
Volume | +2.128 |
smart | +2.106 |
Neg logit contributions | |
�� | -3.046 |
aur | -2.847 |
� | -2.640 |
aires | -2.530 |
�� | -2.470 |
Token | National |
Token ID | 2351 |
Feature activation | -6.33e-01 |
Pos logit contributions | |
Roose | +3.250 |
muzzle | +2.904 |
ibaba | +2.794 |
Volume | +2.692 |
ICLE | +2.670 |
Neg logit contributions | |
�� | -3.701 |
aur | -3.453 |
� | -3.201 |
aires | -3.063 |
�� | -2.992 |
Token | Museum |
Token ID | 9594 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
Roose | +4.314 |
muzzle | +3.789 |
Volume | +3.596 |
ibaba | +3.545 |
smart | +3.438 |
Neg logit contributions | |
�� | -6.541 |
aur | -6.019 |
� | -5.662 |
�� | -5.487 |
aires | -5.348 |
Token | of |
Token ID | 286 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
�� | +0.776 |
aur | +0.734 |
� | +0.678 |
aires | +0.648 |
�� | +0.628 |
Neg logit contributions | |
Roose | -0.675 |
ibaba | -0.606 |
muzzle | -0.598 |
ICLE | -0.565 |
smart | -0.563 |
Token | the |
Token ID | 262 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
�� | +8.005 |
aur | +7.882 |
� | +6.978 |
aires | +6.754 |
ional | +6.302 |
Neg logit contributions | |
Roose | -8.551 |
ibaba | -8.269 |
rils | -8.009 |
muzzle | -7.619 |
azines | -7.610 |
Token | United |
Token ID | 1578 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
�� | +4.073 |
aur | +3.954 |
� | +3.622 |
aires | +3.453 |
�� | +3.284 |
Neg logit contributions | |
Roose | -3.813 |
ibaba | -3.632 |
muzzle | -3.479 |
rils | -3.430 |
azines | -3.356 |
Token | States |
Token ID | 1829 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
aur | +2.998 |
�� | +2.912 |
aires | +2.605 |
� | +2.593 |
�� | +2.380 |
Neg logit contributions | |
Roose | -3.802 |
ibaba | -3.712 |
muzzle | -3.489 |
rils | -3.429 |
ICLE | -3.414 |
Token | Air |
Token ID | 3701 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
�� | +1.451 |
aur | +1.390 |
� | +1.264 |
aires | +1.209 |
�� | +1.166 |
Neg logit contributions | |
Roose | -1.434 |
ibaba | -1.312 |
muzzle | -1.283 |
ICLE | -1.221 |
rils | -1.221 |
Token | Force |
Token ID | 5221 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
�� | +1.427 |
aur | +1.323 |
� | +1.187 |
aires | +1.115 |
�� | +1.068 |
Neg logit contributions | |
Roose | -2.078 |
ibaba | -1.908 |
muzzle | -1.891 |
ICLE | -1.810 |
smart | -1.807 |
Token | 4 |
Token ID | 604 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
�� | +3.650 |
aur | +3.554 |
� | +3.175 |
aires | +2.991 |
�� | +2.814 |
Neg logit contributions | |
Roose | -5.108 |
ibaba | -4.827 |
muzzle | -4.705 |
rils | -4.507 |
azines | -4.481 |
Token | Pierce |
Token ID | 23023 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
�� | +0.118 |
aur | +0.112 |
� | +0.103 |
aires | +0.099 |
�� | +0.096 |
Neg logit contributions | |
Roose | -0.097 |
ibaba | -0.087 |
muzzle | -0.086 |
ICLE | -0.081 |
smart | -0.080 |
Token | Bros |
Token ID | 14266 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
Roose | +2.085 |
ibaba | +1.905 |
muzzle | +1.900 |
smart | +1.806 |
azines | +1.805 |
Neg logit contributions | |
�� | -1.470 |
aur | -1.357 |
� | -1.218 |
aires | -1.146 |
�� | -1.103 |
Token | nan |
Token ID | 12647 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
Roose | +1.038 |
ibaba | +0.922 |
muzzle | +0.910 |
ICLE | +0.862 |
smart | +0.851 |
Neg logit contributions | |
�� | -1.194 |
aur | -1.174 |
� | -1.067 |
aires | -1.027 |
�� | -0.970 |
Token | films |
Token ID | 7328 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
�� | +2.981 |
aur | +2.849 |
� | +2.646 |
aires | +2.549 |
�� | +2.460 |
Neg logit contributions | |
Roose | -2.120 |
ibaba | -1.880 |
muzzle | -1.842 |
ICLE | -1.732 |
smart | -1.723 |
Token | , |
Token ID | 11 |
Feature activation | +9.07e-01 |
Pos logit contributions | |
�� | +2.261 |
aur | +2.118 |
� | +1.952 |
aires | +1.871 |
�� | +1.805 |
Neg logit contributions | |
Roose | -2.147 |
muzzle | -1.920 |
ibaba | -1.913 |
azines | -1.811 |
ICLE | -1.809 |
Token | 1995 |
Token ID | 8735 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
aur | +5.870 |
�� | +5.795 |
� | +5.058 |
aires | +4.908 |
equival | +4.512 |
Neg logit contributions | |
Roose | -8.504 |
ibaba | -8.044 |
muzzle | -7.589 |
ICLE | -7.530 |
rils | -7.520 |
Token | - |
Token ID | 12 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
Roose | +1.883 |
muzzle | +1.605 |
ibaba | +1.584 |
Volume | +1.531 |
azines | +1.529 |
Neg logit contributions | |
�� | -2.882 |
aur | -2.715 |
� | -2.530 |
aires | -2.432 |
�� | -2.413 |
Token | 2002 |
Token ID | 16942 |
Feature activation | -6.70e-03 |
Pos logit contributions | |
�� | +4.487 |
aur | +4.415 |
� | +4.023 |
aires | +3.810 |
�� | +3.529 |
Neg logit contributions | |
Roose | -4.829 |
ibaba | -4.587 |
muzzle | -4.380 |
rils | -4.137 |
azines | -4.127 |
Token | - |
Token ID | 12 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
Roose | +0.049 |
ibaba | +0.043 |
muzzle | +0.043 |
ICLE | +0.040 |
smart | +0.040 |
Neg logit contributions | |
�� | -0.066 |
aur | -0.062 |
� | -0.058 |
aires | -0.056 |
�� | -0.055 |
Token | 3 |
Token ID | 513 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
�� | +2.341 |
aur | +2.281 |
� | +2.043 |
aires | +1.938 |
�� | +1.810 |
Neg logit contributions | |
Roose | -3.186 |
ibaba | -3.019 |
muzzle | -2.934 |
rils | -2.808 |
azines | -2.797 |
Token | Right |
Token ID | 11028 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Roose | +0.651 |
ibaba | +0.589 |
muzzle | +0.580 |
rils | +0.543 |
ICLE | +0.543 |
Neg logit contributions | |
�� | -0.722 |
aur | -0.686 |
� | -0.631 |
aires | -0.604 |
�� | -0.581 |
Token | . |
Token ID | 13 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
�� | +3.680 |
aur | +3.524 |
� | +3.219 |
aires | +3.063 |
�� | +2.916 |
Neg logit contributions | |
Roose | -4.091 |
ibaba | -3.742 |
muzzle | -3.631 |
ICLE | -3.516 |
rils | -3.508 |
Token | Now |
Token ID | 2735 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
Roose | +1.613 |
ibaba | +1.451 |
muzzle | +1.432 |
ICLE | +1.351 |
smart | +1.346 |
Neg logit contributions | |
�� | -1.854 |
aur | -1.755 |
� | -1.620 |
aires | -1.549 |
�� | -1.499 |
Token | imagine |
Token ID | 5967 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
�� | +0.571 |
aur | +0.542 |
� | +0.496 |
aires | +0.476 |
�� | +0.456 |
Neg logit contributions | |
Roose | -0.548 |
ibaba | -0.497 |
muzzle | -0.487 |
ICLE | -0.467 |
rils | -0.462 |
Token | the |
Token ID | 262 |
Feature activation | -9.68e-02 |
Pos logit contributions | |
�� | +2.405 |
aur | +2.310 |
� | +2.096 |
aires | +2.036 |
ional | +1.922 |
Neg logit contributions | |
Roose | -2.439 |
ibaba | -2.324 |
muzzle | -2.205 |
ICLE | -2.157 |
azines | -2.150 |
Token | seismic |
Token ID | 37463 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
Roose | +0.753 |
ibaba | +0.680 |
muzzle | +0.666 |
ICLE | +0.632 |
smart | +0.628 |
Neg logit contributions | |
�� | -0.908 |
aur | -0.862 |
� | -0.797 |
aires | -0.764 |
�� | -0.737 |
Token | eruption |
Token ID | 36176 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
�� | +3.928 |
aur | +3.842 |
� | +3.437 |
aires | +3.386 |
ional | +3.172 |
Neg logit contributions | |
Roose | -3.655 |
ibaba | -3.454 |
ICLE | -3.216 |
muzzle | -3.206 |
rils | -3.194 |
Token | that |
Token ID | 326 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
�� | +2.520 |
aur | +2.434 |
� | +2.175 |
aires | +2.146 |
ional | +2.000 |
Neg logit contributions | |
Roose | -2.550 |
ibaba | -2.387 |
muzzle | -2.268 |
ICLE | -2.224 |
rils | -2.208 |
Token | would |
Token ID | 561 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
�� | +2.560 |
aur | +2.422 |
� | +2.192 |
aires | +2.136 |
�� | +1.993 |
Neg logit contributions | |
Roose | -2.598 |
ibaba | -2.450 |
muzzle | -2.339 |
ICLE | -2.292 |
rils | -2.278 |
Token | have |
Token ID | 423 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
�� | +1.567 |
aur | +1.510 |
� | +1.376 |
aires | +1.327 |
�� | +1.263 |
Neg logit contributions | |
Roose | -1.249 |
ibaba | -1.136 |
muzzle | -1.077 |
ICLE | -1.055 |
azines | -1.047 |
Token | occurred |
Token ID | 5091 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
Roose | +2.232 |
muzzle | +1.949 |
ibaba | +1.944 |
ICLE | +1.805 |
smart | +1.800 |
Neg logit contributions | |
�� | -3.131 |
aur | -2.966 |
� | -2.764 |
aires | -2.654 |
�� | -2.589 |
Token | local |
Token ID | 1957 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
Roose | +0.558 |
ibaba | +0.489 |
muzzle | +0.480 |
ICLE | +0.446 |
smart | +0.445 |
Neg logit contributions | |
�� | -0.913 |
aur | -0.871 |
� | -0.813 |
aires | -0.783 |
�� | -0.763 |
Token | residents |
Token ID | 5085 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
Roose | +1.507 |
muzzle | +1.260 |
ibaba | +1.251 |
ICLE | +1.138 |
smart | +1.120 |
Neg logit contributions | |
�� | -3.358 |
aur | -3.191 |
� | -3.018 |
aires | -2.895 |
�� | -2.858 |
Token | of |
Token ID | 286 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
Roose | +1.771 |
muzzle | +1.592 |
ibaba | +1.576 |
ICLE | +1.503 |
smart | +1.492 |
Neg logit contributions | |
�� | -1.884 |
aur | -1.745 |
� | -1.628 |
aires | -1.544 |
�� | -1.499 |
Token | noise |
Token ID | 7838 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
�� | +0.402 |
aur | +0.382 |
� | +0.353 |
aires | +0.338 |
�� | +0.326 |
Neg logit contributions | |
Roose | -0.336 |
ibaba | -0.302 |
muzzle | -0.295 |
ICLE | -0.280 |
smart | -0.280 |
Token | and |
Token ID | 290 |
Feature activation | -7.66e-04 |
Pos logit contributions | |
Roose | +2.514 |
muzzle | +2.264 |
ibaba | +2.198 |
bang | +2.126 |
Volume | +2.092 |
Neg logit contributions | |
�� | -3.056 |
aur | -2.865 |
� | -2.651 |
aires | -2.547 |
�� | -2.501 |
Token | trash |
Token ID | 13913 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
Roose | +0.006 |
ibaba | +0.005 |
muzzle | +0.005 |
ICLE | +0.005 |
smart | +0.005 |
Neg logit contributions | |
�� | -0.008 |
aur | -0.007 |
� | -0.007 |
aires | -0.007 |
�� | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
�� | +0.241 |
aur | +0.227 |
� | +0.209 |
aires | +0.199 |
�� | +0.192 |
Neg logit contributions | |
Roose | -0.232 |
ibaba | -0.210 |
muzzle | -0.207 |
ICLE | -0.196 |
smart | -0.195 |
Token | Est |
Token ID | 10062 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
Roose | +4.803 |
muzzle | +4.181 |
ibaba | +4.164 |
Volume | +4.028 |
ICLE | +4.010 |
Neg logit contributions | |
�� | -6.019 |
aur | -5.441 |
� | -5.153 |
aires | -4.890 |
�� | -4.824 |
Token | eb |
Token ID | 1765 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
Roose | +4.101 |
ibaba | +3.718 |
muzzle | +3.570 |
rils | +3.485 |
azines | +3.451 |
Neg logit contributions | |
�� | -5.541 |
aur | -4.880 |
� | -4.738 |
�� | -4.520 |
aires | -4.370 |
Token | an |
Token ID | 272 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
�� | +2.620 |
aur | +2.578 |
� | +2.278 |
aires | +2.166 |
�� | +2.060 |
Neg logit contributions | |
Roose | -3.387 |
ibaba | -3.082 |
muzzle | -3.060 |
ICLE | -2.951 |
azines | -2.910 |
Token | Par |
Token ID | 2547 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
�� | +1.652 |
aur | +1.592 |
� | +1.466 |
aires | +1.415 |
�� | +1.348 |
Neg logit contributions | |
Roose | -1.324 |
ibaba | -1.201 |
muzzle | -1.186 |
ICLE | -1.116 |
smart | -1.109 |
Token | ; |
Token ID | 26 |
Feature activation | -7.70e-02 |
Pos logit contributions | |
�� | +2.275 |
aur | +2.140 |
� | +1.963 |
aires | +1.867 |
�� | +1.808 |
Neg logit contributions | |
Roose | -2.290 |
ibaba | -2.067 |
muzzle | -2.049 |
ICLE | -1.944 |
smart | -1.937 |
Token | Poll |
Token ID | 12868 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
Roose | +0.614 |
ibaba | +0.550 |
muzzle | +0.543 |
ICLE | +0.513 |
smart | +0.511 |
Neg logit contributions | |
�� | -0.712 |
aur | -0.673 |
� | -0.622 |
aires | -0.594 |
�� | -0.576 |
Token | ock |
Token ID | 735 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
aur | +3.308 |
�� | +3.030 |
aires | +2.854 |
� | +2.811 |
ional | +2.694 |
Neg logit contributions | |
Roose | -2.704 |
ibaba | -2.398 |
mosqu | -2.348 |
muzzle | -2.312 |
ICLE | -2.212 |
Token | 's |
Token ID | 338 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
Roose | +2.140 |
ibaba | +1.899 |
muzzle | +1.897 |
ICLE | +1.782 |
smart | +1.779 |
Neg logit contributions | |
�� | -2.445 |
aur | -2.297 |
� | -2.124 |
aires | -2.023 |
�� | -1.974 |
Token | letters |
Token ID | 7475 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
Roose | +2.285 |
muzzle | +2.052 |
ibaba | +2.041 |
ICLE | +1.933 |
smart | +1.923 |
Neg logit contributions | |
�� | -2.225 |
aur | -2.070 |
� | -1.909 |
aires | -1.809 |
�� | -1.759 |
Token | were |
Token ID | 547 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Roose | +0.670 |
ibaba | +0.605 |
muzzle | +0.594 |
ICLE | +0.563 |
smart | +0.561 |
Neg logit contributions | |
�� | -0.753 |
aur | -0.716 |
� | -0.659 |
aires | -0.630 |
�� | -0.608 |
Token | read |
Token ID | 1100 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
�� | +1.064 |
aur | +1.035 |
� | +0.946 |
aires | +0.910 |
�� | +0.870 |
Neg logit contributions | |
Roose | -0.843 |
ibaba | -0.792 |
muzzle | -0.733 |
rils | -0.711 |
ICLE | -0.710 |
Token | and |
Token ID | 290 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Roose | +3.510 |
muzzle | +3.167 |
ibaba | +3.087 |
smart | +2.987 |
ICLE | +2.981 |
Neg logit contributions | |
�� | -4.146 |
aur | -3.774 |
� | -3.520 |
aires | -3.372 |
�� | -3.360 |
Token | apparently |
Token ID | 5729 |
Feature activation | -8.12e-03 |
Pos logit contributions | |
Roose | +1.625 |
ibaba | +1.408 |
muzzle | +1.395 |
ICLE | +1.291 |
smart | +1.283 |
Neg logit contributions | |
�� | -2.733 |
aur | -2.601 |
� | -2.434 |
aires | -2.345 |
�� | -2.285 |
Token | influenced |
Token ID | 12824 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Roose | +0.054 |
ibaba | +0.049 |
muzzle | +0.047 |
ICLE | +0.045 |
rils | +0.044 |
Neg logit contributions | |
�� | -0.084 |
aur | -0.081 |
� | -0.075 |
aires | -0.072 |
�� | -0.070 |
Token | proceedings |
Token ID | 13106 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
Roose | +1.410 |
muzzle | +1.245 |
ibaba | +1.233 |
ICLE | +1.168 |
Volume | +1.156 |
Neg logit contributions | |
�� | -1.770 |
aur | -1.660 |
� | -1.538 |
aires | -1.477 |
�� | -1.434 |
Token | � |
Token ID | 247 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
Roose | +1.050 |
ibaba | +0.978 |
muzzle | +0.971 |
ICLE | +0.934 |
smart | +0.934 |
Neg logit contributions | |
�� | -0.513 |
aur | -0.463 |
� | -0.431 |
�� | -0.367 |
aires | -0.365 |
Token | s |
Token ID | 82 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
�� | +0.250 |
aur | +0.236 |
� | +0.219 |
aires | +0.209 |
�� | +0.205 |
Neg logit contributions | |
Roose | -0.193 |
ibaba | -0.170 |
muzzle | -0.168 |
smart | -0.158 |
ICLE | -0.158 |
Token | wife |
Token ID | 3656 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
�� | +1.427 |
aur | +1.360 |
� | +1.259 |
aires | +1.210 |
�� | +1.161 |
Neg logit contributions | |
Roose | -1.115 |
ibaba | -1.023 |
muzzle | -0.986 |
ICLE | -0.941 |
smart | -0.935 |
Token | , |
Token ID | 11 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
�� | +0.947 |
aur | +0.893 |
� | +0.821 |
aires | +0.783 |
�� | +0.757 |
Neg logit contributions | |
Roose | -0.907 |
ibaba | -0.818 |
muzzle | -0.808 |
ICLE | -0.766 |
smart | -0.763 |
Token | Isabel |
Token ID | 41838 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Roose | +0.445 |
ibaba | +0.393 |
muzzle | +0.390 |
ICLE | +0.366 |
smart | +0.365 |
Neg logit contributions | |
�� | -0.568 |
aur | -0.537 |
� | -0.498 |
aires | -0.477 |
�� | -0.464 |
Token | le |
Token ID | 293 |
Feature activation | +9.31e-03 |
Pos logit contributions | |
Roose | +0.854 |
ibaba | +0.751 |
muzzle | +0.745 |
smart | +0.709 |
ICLE | +0.707 |
Neg logit contributions | |
�� | -0.977 |
aur | -0.905 |
� | -0.837 |
aires | -0.797 |
�� | -0.786 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
�� | +0.083 |
aur | +0.078 |
� | +0.072 |
aires | +0.068 |
�� | +0.067 |
Neg logit contributions | |
Roose | -0.078 |
ibaba | -0.070 |
muzzle | -0.069 |
smart | -0.065 |
ICLE | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
Roose | +0.332 |
ibaba | +0.297 |
muzzle | +0.293 |
ICLE | +0.276 |
smart | +0.275 |
Neg logit contributions | |
�� | -0.396 |
aur | -0.374 |
� | -0.346 |
aires | -0.331 |
�� | -0.321 |
Token |
|
Token ID | 198 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
aur | +0.515 |
�� | +0.489 |
� | +0.443 |
aires | +0.442 |
ional | +0.409 |
Neg logit contributions | |
Roose | -0.588 |
ibaba | -0.553 |
mosqu | -0.527 |
muzzle | -0.525 |
rils | -0.506 |
Token | For |
Token ID | 1890 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Roose | +0.395 |
ibaba | +0.356 |
muzzle | +0.352 |
ICLE | +0.332 |
smart | +0.330 |
Neg logit contributions | |
�� | -0.432 |
aur | -0.409 |
� | -0.377 |
aires | -0.360 |
�� | -0.348 |
Token | Buffalo |
Token ID | 14905 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
�� | +1.877 |
aur | +1.788 |
� | +1.651 |
aires | +1.578 |
�� | +1.517 |
Neg logit contributions | |
Roose | -1.600 |
ibaba | -1.460 |
muzzle | -1.427 |
ICLE | -1.367 |
rils | -1.359 |
Token | other |
Token ID | 584 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
�� | +1.277 |
aur | +1.208 |
� | +1.112 |
aires | +1.049 |
�� | +1.020 |
Neg logit contributions | |
Roose | -1.178 |
ibaba | -1.084 |
muzzle | -1.049 |
smart | -1.002 |
rils | -1.000 |
Token | court |
Token ID | 2184 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
�� | +0.997 |
aur | +0.951 |
� | +0.877 |
aires | +0.835 |
�� | +0.809 |
Neg logit contributions | |
Roose | -0.808 |
ibaba | -0.731 |
muzzle | -0.706 |
ICLE | -0.670 |
smart | -0.667 |
Token | decisions |
Token ID | 5370 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
Roose | +3.450 |
muzzle | +3.012 |
ibaba | +2.933 |
disappro | +2.773 |
smart | +2.769 |
Neg logit contributions | |
�� | -5.853 |
aur | -5.418 |
� | -5.147 |
aires | -4.970 |
�� | -4.925 |
Token | , |
Token ID | 11 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
Roose | +1.833 |
muzzle | +1.656 |
ibaba | +1.637 |
smart | +1.553 |
ICLE | +1.548 |
Neg logit contributions | |
�� | -2.003 |
aur | -1.849 |
� | -1.710 |
aires | -1.663 |
�� | -1.594 |
Token | but |
Token ID | 475 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
Roose | +3.452 |
muzzle | +3.087 |
ibaba | +3.034 |
ICLE | +2.829 |
bang | +2.824 |
Neg logit contributions | |
�� | -4.998 |
aur | -4.643 |
� | -4.285 |
aires | -4.236 |
�� | -4.123 |
Token | also |
Token ID | 635 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Roose | +1.998 |
ibaba | +1.774 |
muzzle | +1.771 |
ICLE | +1.649 |
smart | +1.643 |
Neg logit contributions | |
�� | -2.471 |
aur | -2.338 |
� | -2.151 |
aires | -2.073 |
�� | -2.015 |
Token | through |
Token ID | 832 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
�� | +1.064 |
aur | +1.005 |
� | +0.926 |
aires | +0.878 |
�� | +0.848 |
Neg logit contributions | |
Roose | -1.040 |
ibaba | -0.943 |
muzzle | -0.924 |
ICLE | -0.881 |
smart | -0.879 |
Token | state |
Token ID | 1181 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Roose | +0.500 |
ibaba | +0.448 |
muzzle | +0.444 |
ICLE | +0.419 |
smart | +0.417 |
Neg logit contributions | |
�� | -0.567 |
aur | -0.536 |
� | -0.494 |
aires | -0.473 |
�� | -0.458 |
Token | laws |
Token ID | 3657 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
�� | +2.241 |
aur | +2.134 |
� | +1.967 |
aires | +1.870 |
�� | +1.799 |
Neg logit contributions | |
Roose | -1.984 |
ibaba | -1.794 |
muzzle | -1.748 |
rils | -1.660 |
ICLE | -1.640 |
Token | that |
Token ID | 326 |
Feature activation | +3.93e-03 |
Pos logit contributions | |
�� | +1.147 |
aur | +1.092 |
� | +1.002 |
aires | +0.946 |
�� | +0.922 |
Neg logit contributions | |
Roose | -1.106 |
ibaba | -1.002 |
muzzle | -0.977 |
ICLE | -0.931 |
azines | -0.928 |
Token | criminal |
Token ID | 4301 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
�� | +0.037 |
aur | +0.035 |
� | +0.032 |
aires | +0.031 |
�� | +0.030 |
Neg logit contributions | |
Roose | -0.031 |
ibaba | -0.028 |
muzzle | -0.027 |
ICLE | -0.026 |
smart | -0.026 |
Token | considered |
Token ID | 3177 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
Roose | +2.100 |
ibaba | +1.822 |
muzzle | +1.819 |
smart | +1.689 |
ICLE | +1.675 |
Neg logit contributions | |
�� | -3.239 |
aur | -3.055 |
� | -2.863 |
aires | -2.754 |
�� | -2.683 |
Token | to |
Token ID | 284 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
Roose | +1.469 |
muzzle | +1.288 |
ibaba | +1.284 |
smart | +1.231 |
ICLE | +1.196 |
Neg logit contributions | |
�� | -1.854 |
aur | -1.745 |
� | -1.605 |
aires | -1.569 |
�� | -1.520 |
Token | be |
Token ID | 307 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
Roose | +1.055 |
muzzle | +0.957 |
ibaba | +0.923 |
smart | +0.886 |
bang | +0.884 |
Neg logit contributions | |
�� | -1.312 |
aur | -1.220 |
� | -1.117 |
aires | -1.087 |
�� | -1.064 |
Token | a |
Token ID | 257 |
Feature activation | +6.22e-03 |
Pos logit contributions | |
Roose | +1.495 |
ibaba | +1.330 |
muzzle | +1.322 |
smart | +1.248 |
ICLE | +1.242 |
Neg logit contributions | |
�� | -1.718 |
aur | -1.619 |
� | -1.495 |
aires | -1.433 |
�� | -1.387 |
Token | contender |
Token ID | 24190 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
�� | +0.059 |
aur | +0.056 |
� | +0.052 |
aires | +0.049 |
�� | +0.048 |
Neg logit contributions | |
Roose | -0.048 |
ibaba | -0.043 |
muzzle | -0.042 |
ICLE | -0.040 |
rils | -0.040 |
Token | for |
Token ID | 329 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
Roose | +2.781 |
muzzle | +2.450 |
ibaba | +2.435 |
smart | +2.314 |
bang | +2.268 |
Neg logit contributions | |
�� | -3.738 |
aur | -3.460 |
� | -3.242 |
aires | -3.121 |
�� | -3.049 |
Token | the |
Token ID | 262 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
Roose | +2.578 |
muzzle | +2.259 |
ibaba | +2.243 |
Volume | +2.111 |
smart | +2.110 |
Neg logit contributions | |
�� | -3.163 |
aur | -3.001 |
� | -2.738 |
aires | -2.637 |
�� | -2.575 |
Token | presidential |
Token ID | 4787 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
Roose | +1.674 |
ibaba | +1.532 |
muzzle | +1.515 |
ICLE | +1.447 |
smart | +1.443 |
Neg logit contributions | |
�� | -1.316 |
aur | -1.229 |
� | -1.114 |
aires | -1.051 |
�� | -1.010 |
Token | race |
Token ID | 3234 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
Roose | +2.977 |
muzzle | +2.684 |
ibaba | +2.666 |
Volume | +2.524 |
smart | +2.521 |
Neg logit contributions | |
�� | -2.728 |
aur | -2.543 |
� | -2.300 |
aires | -2.169 |
�� | -2.123 |
Token | in |
Token ID | 287 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Roose | +2.528 |
ibaba | +2.275 |
muzzle | +2.272 |
smart | +2.149 |
rils | +2.127 |
Neg logit contributions | |
�� | -2.762 |
aur | -2.545 |
� | -2.364 |
aires | -2.257 |
�� | -2.183 |
Token | 2016 |
Token ID | 1584 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
Roose | +0.900 |
ibaba | +0.810 |
muzzle | +0.801 |
ICLE | +0.758 |
smart | +0.756 |
Neg logit contributions | |
�� | -0.945 |
aur | -0.891 |
� | -0.819 |
aires | -0.781 |
�� | -0.756 |
Token | that |
Token ID | 326 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
�� | +2.480 |
aur | +2.360 |
� | +2.170 |
aires | +2.065 |
�� | +1.982 |
Neg logit contributions | |
Roose | -2.512 |
ibaba | -2.322 |
muzzle | -2.247 |
ICLE | -2.143 |
smart | -2.139 |
Token | � |
Token ID | 564 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
Roose | +0.862 |
muzzle | +0.768 |
ibaba | +0.767 |
smart | +0.727 |
ICLE | +0.725 |
Neg logit contributions | |
�� | -0.864 |
aur | -0.812 |
� | -0.741 |
aires | -0.707 |
�� | -0.683 |
Token | � |
Token ID | 250 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
�� | +1.069 |
aur | +0.947 |
� | +0.935 |
�� | +0.746 |
� | +0.704 |
Neg logit contributions | |
Roose | -2.480 |
ibaba | -2.331 |
muzzle | -2.304 |
azines | -2.227 |
ICLE | -2.223 |
Token | John |
Token ID | 7554 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
Roose | +0.164 |
ibaba | +0.146 |
muzzle | +0.144 |
ICLE | +0.136 |
smart | +0.135 |
Neg logit contributions | |
�� | -0.200 |
aur | -0.190 |
� | -0.176 |
aires | -0.168 |
�� | -0.163 |
Token | McCain |
Token ID | 14264 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
Roose | +1.998 |
ibaba | +1.817 |
muzzle | +1.813 |
smart | +1.752 |
azines | +1.740 |
Neg logit contributions | |
�� | -1.312 |
aur | -1.195 |
� | -1.055 |
aires | -0.998 |
�� | -0.965 |
Token | is |
Token ID | 318 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
�� | +0.123 |
aur | +0.116 |
� | +0.106 |
aires | +0.101 |
�� | +0.097 |
Neg logit contributions | |
Roose | -0.131 |
ibaba | -0.119 |
muzzle | -0.117 |
ICLE | -0.112 |
smart | -0.111 |
Token | an |
Token ID | 281 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
Roose | +1.459 |
ibaba | +1.300 |
muzzle | +1.291 |
smart | +1.225 |
ICLE | +1.214 |
Neg logit contributions | |
�� | -1.710 |
aur | -1.609 |
� | -1.481 |
aires | -1.427 |
�� | -1.375 |
Token | American |
Token ID | 1605 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
Roose | +1.316 |
ibaba | +1.180 |
muzzle | +1.175 |
smart | +1.120 |
ICLE | +1.108 |
Neg logit contributions | |
�� | -1.299 |
aur | -1.211 |
� | -1.105 |
aires | -1.061 |
�� | -1.024 |
Token | hero |
Token ID | 4293 |
Feature activation | +3.70e-03 |
Pos logit contributions | |
Roose | +3.088 |
muzzle | +2.727 |
ibaba | +2.677 |
ICLE | +2.563 |
smart | +2.552 |
Neg logit contributions | |
�� | -4.094 |
aur | -3.832 |
� | -3.567 |
aires | -3.463 |
�� | -3.329 |
Token | & |
Token ID | 1222 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
�� | +0.029 |
aur | +0.028 |
� | +0.025 |
aires | +0.024 |
�� | +0.023 |
Neg logit contributions | |
Roose | -0.034 |
ibaba | -0.031 |
muzzle | -0.031 |
ICLE | -0.029 |
smart | -0.029 |
Token | one |
Token ID | 530 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
Roose | +0.339 |
ibaba | +0.303 |
muzzle | +0.300 |
ICLE | +0.283 |
smart | +0.283 |
Neg logit contributions | |
�� | -0.391 |
aur | -0.368 |
� | -0.340 |
aires | -0.325 |
�� | -0.315 |
Token | line |
Token ID | 1627 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
�� | +1.081 |
aur | +1.028 |
� | +0.942 |
aires | +0.899 |
�� | +0.864 |
Neg logit contributions | |
Roose | -1.065 |
ibaba | -0.965 |
muzzle | -0.948 |
ICLE | -0.904 |
smart | -0.899 |
Token | , |
Token ID | 11 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
�� | +1.787 |
aur | +1.716 |
� | +1.551 |
aires | +1.489 |
�� | +1.413 |
Neg logit contributions | |
Roose | -1.929 |
ibaba | -1.758 |
muzzle | -1.726 |
ICLE | -1.652 |
rils | -1.635 |
Token | but |
Token ID | 475 |
Feature activation | -8.20e-02 |
Pos logit contributions | |
Roose | +0.837 |
ibaba | +0.750 |
muzzle | +0.739 |
ICLE | +0.699 |
smart | +0.695 |
Neg logit contributions | |
�� | -0.991 |
aur | -0.941 |
� | -0.869 |
aires | -0.831 |
�� | -0.805 |
Token | in |
Token ID | 287 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
Roose | +0.691 |
ibaba | +0.627 |
muzzle | +0.617 |
ICLE | +0.588 |
rils | +0.585 |
Neg logit contributions | |
�� | -0.716 |
aur | -0.679 |
� | -0.623 |
aires | -0.595 |
�� | -0.573 |
Token | the |
Token ID | 262 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
�� | +3.679 |
aur | +3.514 |
� | +3.181 |
aires | +3.048 |
�� | +2.895 |
Neg logit contributions | |
Roose | -3.854 |
ibaba | -3.577 |
muzzle | -3.451 |
rils | -3.367 |
ICLE | -3.364 |
Token | 21 |
Token ID | 2310 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
�� | +1.211 |
aur | +1.150 |
� | +1.062 |
aires | +1.022 |
�� | +0.984 |
Neg logit contributions | |
Roose | -0.993 |
ibaba | -0.907 |
muzzle | -0.873 |
rils | -0.836 |
ICLE | -0.834 |
Token | st |
Token ID | 301 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
Roose | +1.057 |
ibaba | +0.938 |
muzzle | +0.929 |
ICLE | +0.879 |
smart | +0.876 |
Neg logit contributions | |
�� | -1.311 |
aur | -1.237 |
� | -1.146 |
aires | -1.092 |
�� | -1.066 |
Token | century |
Token ID | 4289 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
Roose | +1.587 |
muzzle | +1.290 |
ibaba | +1.280 |
Volume | +1.180 |
smart | +1.164 |
Neg logit contributions | |
�� | -3.761 |
aur | -3.563 |
� | -3.364 |
aires | -3.253 |
�� | -3.204 |
Token | , |
Token ID | 11 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
Roose | +3.370 |
muzzle | +3.001 |
ibaba | +2.999 |
smart | +2.845 |
bang | +2.818 |
Neg logit contributions | |
�� | -3.793 |
aur | -3.523 |
� | -3.279 |
aires | -3.096 |
�� | -3.062 |
Token | you |
Token ID | 345 |
Feature activation | +8.07e-03 |
Pos logit contributions | |
Roose | +1.822 |
ibaba | +1.635 |
muzzle | +1.619 |
ICLE | +1.526 |
smart | +1.526 |
Neg logit contributions | |
�� | -1.983 |
aur | -1.868 |
� | -1.722 |
aires | -1.639 |
�� | -1.594 |
Token | can |
Token ID | 460 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
�� | +0.069 |
aur | +0.066 |
� | +0.061 |
aires | +0.058 |
�� | +0.055 |
Neg logit contributions | |
Roose | -0.069 |
ibaba | -0.062 |
muzzle | -0.061 |
ICLE | -0.059 |
azines | -0.058 |
Token | facing |
Token ID | 6476 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
Roose | +0.887 |
ibaba | +0.794 |
muzzle | +0.783 |
ICLE | +0.740 |
smart | +0.736 |
Neg logit contributions | |
�� | -1.063 |
aur | -1.007 |
� | -0.932 |
aires | -0.891 |
�� | -0.864 |
Token | a |
Token ID | 257 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
�� | +0.237 |
aur | +0.225 |
� | +0.209 |
aires | +0.199 |
�� | +0.192 |
Neg logit contributions | |
Roose | -0.206 |
ibaba | -0.184 |
muzzle | -0.181 |
ICLE | -0.173 |
smart | -0.171 |
Token | declining |
Token ID | 18080 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
�� | +1.498 |
aur | +1.431 |
� | +1.313 |
aires | +1.257 |
�� | +1.208 |
Neg logit contributions | |
Roose | -1.376 |
ibaba | -1.228 |
muzzle | -1.195 |
ICLE | -1.170 |
azines | -1.166 |
Token | consumer |
Token ID | 7172 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
Roose | +1.949 |
ibaba | +1.755 |
muzzle | +1.734 |
ICLE | +1.643 |
smart | +1.637 |
Neg logit contributions | |
�� | -2.060 |
aur | -1.944 |
� | -1.788 |
aires | -1.706 |
�� | -1.650 |
Token | base |
Token ID | 2779 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
Roose | +1.008 |
ibaba | +0.911 |
muzzle | +0.899 |
ICLE | +0.857 |
smart | +0.854 |
Neg logit contributions | |
�� | -0.953 |
aur | -0.900 |
� | -0.823 |
aires | -0.783 |
�� | -0.754 |
Token | for |
Token ID | 329 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
�� | +0.560 |
aur | +0.530 |
� | +0.485 |
aires | +0.462 |
�� | +0.446 |
Neg logit contributions | |
Roose | -0.564 |
ibaba | -0.509 |
muzzle | -0.503 |
ICLE | -0.478 |
smart | -0.476 |
Token | their |
Token ID | 511 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
Roose | +3.048 |
muzzle | +2.721 |
ibaba | +2.682 |
smart | +2.525 |
bang | +2.506 |
Neg logit contributions | |
�� | -4.202 |
aur | -3.970 |
� | -3.679 |
aires | -3.541 |
�� | -3.452 |
Token | products |
Token ID | 3186 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Roose | +3.226 |
muzzle | +2.802 |
ibaba | +2.705 |
bang | +2.533 |
smart | +2.515 |
Neg logit contributions | |
�� | -6.725 |
aur | -6.381 |
� | -6.000 |
aires | -5.819 |
�� | -5.695 |
Token | . |
Token ID | 13 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
Roose | +0.877 |
ibaba | +0.789 |
muzzle | +0.789 |
smart | +0.746 |
ICLE | +0.739 |
Neg logit contributions | |
�� | -0.965 |
aur | -0.891 |
� | -0.829 |
aires | -0.798 |
�� | -0.771 |
Token |
|
Token ID | 198 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
Roose | +0.427 |
ibaba | +0.387 |
muzzle | +0.380 |
ICLE | +0.359 |
rils | +0.357 |
Neg logit contributions | |
�� | -0.501 |
aur | -0.479 |
� | -0.442 |
aires | -0.423 |
�� | -0.407 |
Token |
|
Token ID | 198 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
aur | +0.388 |
�� | +0.374 |
� | +0.337 |
aires | +0.335 |
ional | +0.310 |
Neg logit contributions | |
Roose | -0.441 |
ibaba | -0.414 |
muzzle | -0.396 |
mosqu | -0.395 |
rils | -0.380 |
Token | morning |
Token ID | 43911 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Roose | +0.868 |
ibaba | +0.777 |
muzzle | +0.768 |
ICLE | +0.723 |
smart | +0.723 |
Neg logit contributions | |
�� | -1.035 |
aur | -0.978 |
� | -0.904 |
aires | -0.866 |
�� | -0.841 |
Token | we |
Token ID | 356 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
�� | +1.361 |
aur | +1.302 |
� | +1.193 |
aires | +1.143 |
�� | +1.088 |
Neg logit contributions | |
Roose | -1.279 |
ibaba | -1.168 |
muzzle | -1.122 |
ICLE | -1.098 |
rils | -1.086 |
Token | came |
Token ID | 1625 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
�� | +0.409 |
aur | +0.393 |
� | +0.366 |
aires | +0.347 |
�� | +0.332 |
Neg logit contributions | |
Roose | -0.352 |
ibaba | -0.323 |
muzzle | -0.308 |
ICLE | -0.300 |
azines | -0.299 |
Token | across |
Token ID | 1973 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
Roose | +2.978 |
muzzle | +2.653 |
ibaba | +2.641 |
bang | +2.473 |
smart | +2.451 |
Neg logit contributions | |
�� | -3.545 |
aur | -3.303 |
� | -3.061 |
aires | -2.931 |
�� | -2.871 |
Token | a |
Token ID | 257 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
Roose | +1.571 |
ibaba | +1.395 |
muzzle | +1.395 |
smart | +1.298 |
ICLE | +1.298 |
Neg logit contributions | |
�� | -1.832 |
aur | -1.719 |
� | -1.592 |
aires | -1.523 |
�� | -1.480 |
Token | black |
Token ID | 2042 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
Roose | +0.622 |
ibaba | +0.557 |
muzzle | +0.548 |
ICLE | +0.519 |
smart | +0.517 |
Neg logit contributions | |
�� | -0.747 |
aur | -0.708 |
� | -0.654 |
aires | -0.625 |
�� | -0.606 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
Roose | +2.364 |
muzzle | +2.137 |
ibaba | +2.126 |
smart | +1.965 |
bang | +1.961 |
Neg logit contributions | |
�� | -2.962 |
aur | -2.794 |
� | -2.593 |
aires | -2.494 |
�� | -2.425 |
Token | and |
Token ID | 290 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
�� | +2.968 |
aur | +2.910 |
� | +2.617 |
aires | +2.484 |
�� | +2.384 |
Neg logit contributions | |
Roose | -3.105 |
ibaba | -2.816 |
muzzle | -2.678 |
azines | -2.638 |
ICLE | -2.625 |
Token | her |
Token ID | 607 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
�� | +2.001 |
aur | +1.938 |
� | +1.795 |
aires | +1.696 |
�� | +1.629 |
Neg logit contributions | |
Roose | -1.785 |
ibaba | -1.646 |
muzzle | -1.542 |
azines | -1.528 |
ICLE | -1.522 |
Token | two |
Token ID | 734 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
�� | +1.343 |
aur | +1.286 |
� | +1.186 |
aires | +1.128 |
�� | +1.089 |
Neg logit contributions | |
Roose | -1.134 |
ibaba | -1.025 |
muzzle | -0.977 |
ICLE | -0.948 |
azines | -0.939 |
Token | cub |
Token ID | 13617 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
�� | +2.670 |
aur | +2.583 |
� | +2.386 |
aires | +2.291 |
�� | +2.163 |
Neg logit contributions | |
Roose | -2.234 |
ibaba | -2.094 |
ICLE | -1.942 |
muzzle | -1.924 |
azines | -1.899 |
Token | This |
Token ID | 770 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Roose | +0.963 |
ibaba | +0.862 |
muzzle | +0.851 |
ICLE | +0.802 |
smart | +0.799 |
Neg logit contributions | |
�� | -1.158 |
aur | -1.096 |
� | -1.014 |
aires | -0.970 |
�� | -0.941 |
Token | is |
Token ID | 318 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
Roose | +1.070 |
ibaba | +0.962 |
muzzle | +0.949 |
ICLE | +0.899 |
smart | +0.896 |
Neg logit contributions | |
�� | -1.179 |
aur | -1.115 |
� | -1.028 |
aires | -0.980 |
�� | -0.950 |
Token | also |
Token ID | 635 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
Roose | +4.908 |
muzzle | +4.275 |
ibaba | +4.213 |
bang | +4.017 |
Volume | +4.002 |
Neg logit contributions | |
�� | -6.307 |
aur | -5.890 |
� | -5.465 |
aires | -5.340 |
�� | -5.139 |
Token | mentioned |
Token ID | 4750 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
Roose | +3.004 |
ibaba | +2.627 |
muzzle | +2.626 |
ICLE | +2.448 |
bang | +2.434 |
Neg logit contributions | |
�� | -3.973 |
aur | -3.728 |
� | -3.466 |
aires | -3.356 |
�� | -3.249 |
Token | in |
Token ID | 287 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
Roose | +3.350 |
muzzle | +2.956 |
ibaba | +2.955 |
ICLE | +2.767 |
Volume | +2.761 |
Neg logit contributions | |
�� | -4.037 |
aur | -3.754 |
� | -3.503 |
aires | -3.380 |
�� | -3.265 |
Token | Sunni |
Token ID | 18592 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
Roose | +1.401 |
ibaba | +1.248 |
muzzle | +1.238 |
Volume | +1.163 |
ICLE | +1.162 |
Neg logit contributions | |
�� | -1.667 |
aur | -1.567 |
� | -1.450 |
aires | -1.392 |
�� | -1.345 |
Token | Had |
Token ID | 11161 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
Roose | +6.868 |
muzzle | +6.298 |
Volume | +6.128 |
bang | +5.981 |
ibaba | +5.965 |
Neg logit contributions | |
�� | -8.459 |
aur | -7.625 |
� | -7.112 |
aires | -7.081 |
�� | -6.717 |
Token | ith |
Token ID | 342 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
Roose | +3.582 |
ibaba | +3.226 |
Volume | +2.905 |
muzzle | +2.829 |
smart | +2.774 |
Neg logit contributions | |
�� | -9.644 |
aur | -8.849 |
� | -8.361 |
�� | -8.130 |
aires | -8.107 |
Token | collections |
Token ID | 17268 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
Roose | +3.056 |
ibaba | +2.766 |
muzzle | +2.726 |
Volume | +2.633 |
ICLE | +2.600 |
Neg logit contributions | |
�� | -3.568 |
aur | -3.289 |
� | -3.044 |
aires | -2.949 |
�� | -2.816 |
Token | : |
Token ID | 25 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
�� | +0.485 |
aur | +0.459 |
� | +0.421 |
aires | +0.401 |
�� | +0.388 |
Neg logit contributions | |
Roose | -0.471 |
ibaba | -0.427 |
muzzle | -0.420 |
ICLE | -0.399 |
smart | -0.397 |
Token |
|
Token ID | 198 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Roose | +2.212 |
ibaba | +1.940 |
muzzle | +1.929 |
Volume | +1.858 |
smart | +1.791 |
Neg logit contributions | |
�� | -3.029 |
aur | -2.843 |
� | -2.650 |
aires | -2.550 |
�� | -2.479 |
Token | ersed |
Token ID | 20204 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
Roose | +1.687 |
ibaba | +1.505 |
muzzle | +1.493 |
smart | +1.400 |
azines | +1.399 |
Neg logit contributions | |
�� | -2.386 |
aur | -2.207 |
� | -2.074 |
aires | -1.968 |
�� | -1.955 |
Token | with |
Token ID | 351 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
�� | +0.241 |
aur | +0.223 |
� | +0.202 |
aires | +0.194 |
�� | +0.186 |
Neg logit contributions | |
Roose | -0.269 |
muzzle | -0.245 |
ibaba | -0.242 |
bang | -0.230 |
smart | -0.229 |
Token | red |
Token ID | 2266 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
Roose | +0.417 |
ibaba | +0.374 |
muzzle | +0.368 |
ICLE | +0.347 |
smart | +0.345 |
Neg logit contributions | |
�� | -0.521 |
aur | -0.494 |
� | -0.457 |
aires | -0.438 |
�� | -0.425 |
Token | - |
Token ID | 12 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
Roose | +4.555 |
muzzle | +4.072 |
ibaba | +3.964 |
bang | +3.805 |
smart | +3.739 |
Neg logit contributions | |
�� | -6.104 |
aur | -5.726 |
� | -5.356 |
aires | -5.095 |
�� | -5.054 |
Token | painted |
Token ID | 47351 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
Roose | +0.290 |
ibaba | +0.260 |
muzzle | +0.257 |
ICLE | +0.243 |
smart | +0.241 |
Neg logit contributions | |
�� | -0.327 |
aur | -0.310 |
� | -0.285 |
aires | -0.273 |
�� | -0.264 |
Token | wood |
Token ID | 4898 |
Feature activation | -9.52e-01 |
Pos logit contributions | |
Roose | +2.162 |
muzzle | +1.896 |
ibaba | +1.877 |
ICLE | +1.747 |
bang | +1.746 |
Neg logit contributions | |
�� | -2.977 |
aur | -2.813 |
� | -2.622 |
aires | -2.504 |
�� | -2.453 |
Token | stripes |
Token ID | 28806 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
Roose | +6.737 |
muzzle | +6.221 |
shel | +5.933 |
bang | +5.901 |
smart | +5.812 |
Neg logit contributions | |
�� | -9.045 |
aur | -8.267 |
� | -7.795 |
aires | -7.365 |
�� | -7.328 |
Token | ( |
Token ID | 357 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
�� | +1.750 |
aur | +1.656 |
� | +1.517 |
aires | +1.440 |
�� | +1.389 |
Neg logit contributions | |
Roose | -1.806 |
ibaba | -1.645 |
muzzle | -1.613 |
ICLE | -1.543 |
smart | -1.529 |
Token | from |
Token ID | 6738 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
Roose | +0.464 |
ibaba | +0.416 |
muzzle | +0.411 |
ICLE | +0.387 |
smart | +0.385 |
Neg logit contributions | |
�� | -0.549 |
aur | -0.520 |
� | -0.481 |
aires | -0.459 |
�� | -0.445 |
Token | $ |
Token ID | 720 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
�� | +2.598 |
aur | +2.472 |
� | +2.277 |
aires | +2.180 |
�� | +2.102 |
Neg logit contributions | |
Roose | -2.223 |
ibaba | -2.036 |
muzzle | -1.984 |
ICLE | -1.899 |
rils | -1.890 |
Token | 399 |
Token ID | 28771 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
�� | +0.372 |
aur | +0.359 |
� | +0.325 |
aires | +0.310 |
�� | +0.292 |
Neg logit contributions | |
Roose | -0.430 |
ibaba | -0.400 |
muzzle | -0.391 |
azines | -0.371 |
rils | -0.370 |
Token | does |
Token ID | 857 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
Roose | +1.406 |
ibaba | +1.267 |
muzzle | +1.252 |
ICLE | +1.184 |
smart | +1.179 |
Neg logit contributions | |
�� | -1.515 |
aur | -1.432 |
� | -1.321 |
aires | -1.259 |
�� | -1.218 |
Token | this |
Token ID | 428 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
Roose | +2.777 |
ibaba | +2.501 |
muzzle | +2.467 |
smart | +2.329 |
Volume | +2.321 |
Neg logit contributions | |
�� | -3.063 |
aur | -2.873 |
� | -2.643 |
aires | -2.524 |
�� | -2.454 |
Token | mean |
Token ID | 1612 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
�� | +2.415 |
aur | +2.303 |
� | +2.081 |
aires | +1.961 |
�� | +1.895 |
Neg logit contributions | |
Roose | -2.753 |
ibaba | -2.493 |
muzzle | -2.430 |
azines | -2.368 |
ICLE | -2.359 |
Token | ? |
Token ID | 30 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
Roose | +1.672 |
ibaba | +1.503 |
muzzle | +1.483 |
smart | +1.415 |
ICLE | +1.401 |
Neg logit contributions | |
�� | -1.879 |
aur | -1.754 |
� | -1.625 |
aires | -1.557 |
�� | -1.516 |
Token | A |
Token ID | 317 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
Roose | +2.645 |
ibaba | +2.344 |
muzzle | +2.305 |
smart | +2.184 |
ICLE | +2.182 |
Neg logit contributions | |
�� | -3.256 |
aur | -3.062 |
� | -2.838 |
aires | -2.710 |
�� | -2.651 |
Token | lot |
Token ID | 1256 |
Feature activation | -9.33e-01 |
Pos logit contributions | |
�� | +0.108 |
aur | +0.102 |
� | +0.094 |
aires | +0.090 |
�� | +0.087 |
Neg logit contributions | |
Roose | -0.095 |
ibaba | -0.086 |
muzzle | -0.084 |
ICLE | -0.080 |
azines | -0.079 |
Token | of |
Token ID | 286 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
Roose | +6.477 |
muzzle | +5.892 |
bang | +5.672 |
smart | +5.573 |
ICLE | +5.456 |
Neg logit contributions | |
�� | -9.030 |
aur | -8.031 |
� | -7.653 |
aires | -7.379 |
�� | -7.264 |
Token | messages |
Token ID | 6218 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
Roose | +0.653 |
ibaba | +0.587 |
muzzle | +0.578 |
ICLE | +0.548 |
smart | +0.543 |
Neg logit contributions | |
�� | -0.747 |
aur | -0.709 |
� | -0.654 |
aires | -0.625 |
�� | -0.605 |
Token | will |
Token ID | 481 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
�� | +1.893 |
aur | +1.842 |
� | +1.679 |
aires | +1.601 |
�� | +1.514 |
Neg logit contributions | |
Roose | -1.822 |
ibaba | -1.677 |
muzzle | -1.616 |
ICLE | -1.552 |
azines | -1.537 |
Token | be |
Token ID | 307 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
�� | +1.297 |
aur | +1.241 |
� | +1.139 |
aires | +1.089 |
�� | +1.036 |
Neg logit contributions | |
Roose | -1.171 |
ibaba | -1.065 |
muzzle | -1.028 |
ICLE | -0.989 |
azines | -0.980 |
Token | shared |
Token ID | 4888 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
�� | +3.515 |
aur | +3.410 |
� | +3.115 |
aires | +2.996 |
�� | +2.816 |
Neg logit contributions | |
Roose | -3.343 |
ibaba | -3.156 |
muzzle | -2.970 |
ICLE | -2.900 |
azines | -2.887 |
Token | girls |
Token ID | 4813 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
Roose | +0.288 |
ibaba | +0.259 |
muzzle | +0.253 |
ICLE | +0.241 |
smart | +0.238 |
Neg logit contributions | |
�� | -0.358 |
aur | -0.340 |
� | -0.315 |
aires | -0.302 |
�� | -0.291 |
Token | as |
Token ID | 355 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
�� | +0.717 |
aur | +0.692 |
� | +0.625 |
aires | +0.599 |
�� | +0.573 |
Neg logit contributions | |
Roose | -0.770 |
ibaba | -0.711 |
muzzle | -0.684 |
rils | -0.661 |
ICLE | -0.660 |
Token | young |
Token ID | 1862 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
�� | +1.630 |
aur | +1.556 |
� | +1.428 |
aires | +1.362 |
�� | +1.310 |
Neg logit contributions | |
Roose | -1.457 |
ibaba | -1.349 |
muzzle | -1.289 |
ICLE | -1.245 |
azines | -1.236 |
Token | as |
Token ID | 355 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
�� | +2.744 |
aur | +2.739 |
� | +2.368 |
aires | +2.345 |
equival | +2.160 |
Neg logit contributions | |
Roose | -4.972 |
ibaba | -4.753 |
ICLE | -4.602 |
rils | -4.445 |
muzzle | -4.433 |
Token | 10 |
Token ID | 838 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
�� | +0.482 |
aur | +0.461 |
� | +0.417 |
aires | +0.399 |
�� | +0.380 |
Neg logit contributions | |
Roose | -0.546 |
ibaba | -0.505 |
muzzle | -0.488 |
ICLE | -0.473 |
rils | -0.468 |
Token | or |
Token ID | 393 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
�� | +1.361 |
aur | +1.293 |
� | +1.182 |
aires | +1.123 |
�� | +1.084 |
Neg logit contributions | |
Roose | -1.348 |
ibaba | -1.221 |
muzzle | -1.196 |
ICLE | -1.137 |
rils | -1.131 |
Token | 11 |
Token ID | 1367 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
Roose | +6.794 |
muzzle | +6.113 |
bang | +5.843 |
ibaba | +5.721 |
Volume | +5.524 |
Neg logit contributions | |
�� | -10.026 |
aur | -9.332 |
� | -8.744 |
�� | -8.147 |
aires | -8.135 |
Token | years |
Token ID | 812 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
�� | +3.132 |
aur | +2.985 |
� | +2.706 |
aires | +2.568 |
�� | +2.464 |
Neg logit contributions | |
Roose | -3.358 |
ibaba | -3.065 |
muzzle | -2.973 |
ICLE | -2.848 |
rils | -2.832 |
Token | old |
Token ID | 1468 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
�� | +5.179 |
aur | +4.902 |
� | +4.472 |
aires | +4.228 |
�� | +4.027 |
Neg logit contributions | |
Roose | -5.867 |
ibaba | -5.382 |
muzzle | -5.256 |
ICLE | -5.027 |
rils | -5.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
�� | +3.904 |
aur | +3.732 |
� | +3.360 |
aires | +3.191 |
�� | +3.072 |
Neg logit contributions | |
Roose | -4.564 |
ibaba | -4.158 |
muzzle | -4.056 |
ICLE | -3.894 |
rils | -3.892 |
Token |
|
Token ID | 198 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
Roose | +0.948 |
ibaba | +0.838 |
muzzle | +0.829 |
ICLE | +0.783 |
smart | +0.780 |
Neg logit contributions | |
�� | -1.178 |
aur | -1.111 |
� | -1.029 |
aires | -0.985 |
�� | -0.960 |
Token | , |
Token ID | 11 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
�� | +0.257 |
aur | +0.241 |
� | +0.222 |
aires | +0.211 |
�� | +0.204 |
Neg logit contributions | |
Roose | -0.259 |
ibaba | -0.234 |
muzzle | -0.232 |
ICLE | -0.220 |
smart | -0.220 |
Token | 2017 |
Token ID | 2177 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Roose | +1.346 |
ibaba | +1.219 |
muzzle | +1.197 |
ICLE | +1.134 |
smart | +1.130 |
Neg logit contributions | |
�� | -1.435 |
aur | -1.362 |
� | -1.250 |
aires | -1.195 |
�� | -1.150 |
Token | . |
Token ID | 13 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
�� | +0.917 |
aur | +0.858 |
� | +0.790 |
aires | +0.754 |
�� | +0.729 |
Neg logit contributions | |
Roose | -0.909 |
ibaba | -0.818 |
muzzle | -0.814 |
ICLE | -0.771 |
smart | -0.769 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
Roose | +1.919 |
ibaba | +1.682 |
muzzle | +1.671 |
Volume | +1.607 |
ICLE | +1.569 |
Neg logit contributions | |
�� | -2.603 |
aur | -2.422 |
� | -2.262 |
aires | -2.196 |
�� | -2.139 |
Token |
|
Token ID | 198 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
Roose | +1.232 |
ibaba | +1.136 |
muzzle | +1.095 |
rils | +1.041 |
ICLE | +1.040 |
Neg logit contributions | |
�� | -1.211 |
aur | -1.208 |
� | -1.078 |
aires | -1.052 |
ional | -0.978 |
Token | [ |
Token ID | 58 |
Feature activation | -9.80e-01 |
Pos logit contributions | |
Roose | +2.959 |
ibaba | +2.622 |
muzzle | +2.593 |
ICLE | +2.466 |
smart | +2.456 |
Neg logit contributions | |
�� | -3.889 |
aur | -3.645 |
� | -3.399 |
aires | -3.263 |
�� | -3.178 |
Token | 22 |
Token ID | 1828 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
Roose | +7.166 |
muzzle | +6.361 |
smart | +6.335 |
ibaba | +6.241 |
Volume | +6.140 |
Neg logit contributions | |
�� | -9.441 |
aur | -8.279 |
� | -7.815 |
aires | -7.726 |
�� | -7.555 |
Token | ] |
Token ID | 60 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
aur | +1.064 |
�� | +1.004 |
� | +0.915 |
aires | +0.883 |
ional | +0.806 |
Neg logit contributions | |
Roose | -1.628 |
ibaba | -1.565 |
mosqu | -1.501 |
muzzle | -1.482 |
rils | -1.432 |
Token | Literally |
Token ID | 48414 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
Roose | +0.431 |
ibaba | +0.386 |
muzzle | +0.381 |
ICLE | +0.359 |
smart | +0.358 |
Neg logit contributions | |
�� | -0.513 |
aur | -0.487 |
� | -0.450 |
aires | -0.430 |
�� | -0.417 |
Token | , |
Token ID | 11 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
Roose | +0.575 |
ibaba | +0.516 |
muzzle | +0.511 |
ICLE | +0.481 |
smart | +0.480 |
Neg logit contributions | |
�� | -0.661 |
aur | -0.622 |
� | -0.576 |
aires | -0.551 |
�� | -0.534 |
Token | R |
Token ID | 371 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
Roose | +1.935 |
ibaba | +1.750 |
muzzle | +1.746 |
ICLE | +1.646 |
Volume | +1.637 |
Neg logit contributions | |
�� | -1.865 |
aur | -1.737 |
� | -1.596 |
aires | -1.528 |
�� | -1.474 |