Token | - |
Token ID | 12 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
zin | +4.459 |
umat | +4.355 |
fuelled | +4.288 |
andals | +4.156 |
piring | +4.013 |
Neg logit contributions | |
confines | -5.957 |
slightest | -5.408 |
bounds | -4.961 |
assumption | -4.883 |
margins | -4.536 |
Token | readable |
Token ID | 46155 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
zin | +5.845 |
umat | +5.665 |
fuelled | +5.491 |
andals | +5.370 |
piring | +5.225 |
Neg logit contributions | |
confines | -8.262 |
slightest | -7.546 |
bounds | -6.926 |
assumption | -6.829 |
margins | -6.416 |
Token | zones |
Token ID | 14123 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
zin | +8.309 |
umat | +8.049 |
fuelled | +7.958 |
andals | +7.824 |
piring | +7.552 |
Neg logit contributions | |
confines | -6.192 |
slightest | -5.506 |
bounds | -4.760 |
assumption | -4.758 |
margins | -4.171 |
Token | in |
Token ID | 287 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
zin | +6.807 |
umat | +6.576 |
fuelled | +6.331 |
andals | +6.163 |
ijing | +5.886 |
Neg logit contributions | |
confines | -11.876 |
slightest | -10.960 |
bounds | -9.949 |
assumption | -9.867 |
margins | -9.252 |
Token | passport |
Token ID | 17981 |
Feature activation | +9.74e-01 |
Pos logit contributions | |
zin | +5.481 |
umat | +5.433 |
fuelled | +5.294 |
andals | +5.224 |
piring | +5.085 |
Neg logit contributions | |
confines | -4.447 |
slightest | -3.917 |
bounds | -3.492 |
assumption | -3.424 |
margins | -3.100 |
Token | images |
Token ID | 4263 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
zin | +3.614 |
umat | +3.530 |
fuelled | +3.429 |
andals | +3.383 |
ijing | +3.247 |
Neg logit contributions | |
confines | -4.399 |
slightest | -3.977 |
bounds | -3.626 |
assumption | -3.591 |
margins | -3.275 |
Token | # |
Token ID | 1303 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
zin | +15.573 |
umat | +15.320 |
fuelled | +14.799 |
andals | +14.510 |
af | +14.252 |
Neg logit contributions | |
confines | -16.068 |
slightest | -14.586 |
bounds | -12.479 |
assumption | -12.206 |
margins | -11.329 |
Token | import |
Token ID | 1330 |
Feature activation | +9.80e-01 |
Pos logit contributions | |
zin | +9.840 |
umat | +9.716 |
fuelled | +9.349 |
andals | +9.167 |
ijing | +8.895 |
Neg logit contributions | |
confines | -11.228 |
slightest | -10.107 |
bounds | -9.071 |
assumption | -8.784 |
margins | -8.321 |
Token | the |
Token ID | 262 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
zin | +3.685 |
umat | +3.573 |
fuelled | +3.476 |
andals | +3.401 |
ijing | +3.282 |
Neg logit contributions | |
confines | -4.414 |
slightest | -4.012 |
bounds | -3.597 |
assumption | -3.538 |
margins | -3.292 |
Token | necessary |
Token ID | 3306 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
zin | +1.812 |
umat | +1.771 |
fuelled | +1.736 |
andals | +1.710 |
ijing | +1.661 |
Neg logit contributions | |
confines | -1.368 |
slightest | -1.194 |
bounds | -1.066 |
assumption | -1.038 |
margins | -0.947 |
Token | packages |
Token ID | 10392 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
zin | +1.967 |
umat | +1.926 |
fuelled | +1.866 |
andals | +1.848 |
ijing | +1.752 |
Neg logit contributions | |
confines | -2.573 |
slightest | -2.342 |
bounds | -2.146 |
assumption | -2.113 |
margins | -1.967 |
Token | health |
Token ID | 1535 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
zin | +0.045 |
umat | +0.044 |
andals | +0.043 |
fuelled | +0.043 |
ijing | +0.041 |
Neg logit contributions | |
confines | -0.041 |
slightest | -0.037 |
bounds | -0.034 |
assumption | -0.033 |
margins | -0.030 |
Token | resources |
Token ID | 4133 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
zin | +1.325 |
umat | +1.305 |
andals | +1.262 |
fuelled | +1.259 |
ijing | +1.208 |
Neg logit contributions | |
confines | -1.256 |
slightest | -1.108 |
bounds | -1.019 |
assumption | -1.000 |
margins | -0.923 |
Token | for |
Token ID | 329 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
zin | +3.502 |
umat | +3.442 |
fuelled | +3.332 |
andals | +3.305 |
ijing | +3.162 |
Neg logit contributions | |
confines | -3.907 |
slightest | -3.515 |
bounds | -3.224 |
assumption | -3.162 |
margins | -2.933 |
Token | injury |
Token ID | 5095 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
zin | +1.653 |
umat | +1.623 |
fuelled | +1.574 |
andals | +1.568 |
ijing | +1.500 |
Neg logit contributions | |
confines | -1.811 |
slightest | -1.627 |
bounds | -1.496 |
assumption | -1.468 |
margins | -1.359 |
Token | and |
Token ID | 290 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
umat | +8.829 |
zin | +8.811 |
fuelled | +8.468 |
andals | +8.140 |
ijing | +7.807 |
Neg logit contributions | |
confines | -12.064 |
slightest | -11.058 |
bounds | -9.986 |
assumption | -9.807 |
margins | -9.137 |
Token | violence |
Token ID | 3685 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
zin | +5.525 |
umat | +5.479 |
fuelled | +5.312 |
andals | +5.206 |
piring | +4.966 |
Neg logit contributions | |
confines | -6.856 |
slightest | -6.201 |
bounds | -5.710 |
assumption | -5.577 |
margins | -5.208 |
Token | prevention |
Token ID | 14196 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
umat | +14.330 |
zin | +14.285 |
fuelled | +13.863 |
utics | +12.987 |
andals | +12.972 |
Neg logit contributions | |
confines | -16.804 |
slightest | -15.632 |
bounds | -13.670 |
assumption | -13.311 |
margins | -12.356 |
Token | will |
Token ID | 481 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
zin | +9.638 |
umat | +9.461 |
fuelled | +9.397 |
andals | +8.990 |
piring | +8.672 |
Neg logit contributions | |
confines | -10.713 |
slightest | -9.782 |
bounds | -8.689 |
assumption | -8.463 |
margins | -7.847 |
Token | grow |
Token ID | 1663 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
zin | +1.289 |
umat | +1.268 |
fuelled | +1.230 |
andals | +1.221 |
ijing | +1.172 |
Neg logit contributions | |
confines | -1.256 |
slightest | -1.121 |
bounds | -1.021 |
assumption | -1.000 |
margins | -0.921 |
Token | more |
Token ID | 517 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
zin | +2.058 |
umat | +2.022 |
fuelled | +1.936 |
andals | +1.929 |
ijing | +1.832 |
Neg logit contributions | |
confines | -2.816 |
slightest | -2.558 |
bounds | -2.369 |
assumption | -2.330 |
margins | -2.181 |
Token | pronounced |
Token ID | 16293 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
zin | +3.987 |
umat | +3.916 |
fuelled | +3.802 |
andals | +3.726 |
piring | +3.557 |
Neg logit contributions | |
confines | -5.519 |
slightest | -5.013 |
bounds | -4.628 |
assumption | -4.550 |
margins | -4.250 |
Token | on |
Token ID | 319 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
zin | +6.962 |
umat | +6.923 |
fuelled | +6.796 |
andals | +6.609 |
piring | +6.398 |
Neg logit contributions | |
confines | -8.782 |
slightest | -8.045 |
bounds | -7.321 |
assumption | -7.191 |
margins | -6.728 |
Token | campus |
Token ID | 7611 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
confines | +1.451 |
slightest | +1.238 |
bounds | +1.143 |
assumption | +1.111 |
margins | +1.015 |
Neg logit contributions | |
zin | -1.885 |
umat | -1.837 |
fuelled | -1.795 |
andals | -1.779 |
ijing | -1.736 |
Token | after |
Token ID | 706 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
zin | +1.725 |
umat | +1.679 |
fuelled | +1.604 |
andals | +1.590 |
ijing | +1.508 |
Neg logit contributions | |
confines | -3.121 |
slightest | -2.851 |
bounds | -2.674 |
assumption | -2.636 |
margins | -2.483 |
Token | report |
Token ID | 989 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
zin | +6.462 |
umat | +6.365 |
fuelled | +6.211 |
andals | +6.173 |
piring | +5.943 |
Neg logit contributions | |
confines | -6.566 |
slightest | -5.884 |
bounds | -5.377 |
assumption | -5.224 |
margins | -4.861 |
Token | of |
Token ID | 286 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
zin | +7.816 |
umat | +7.690 |
andals | +7.471 |
fuelled | +7.447 |
ijing | +6.947 |
Neg logit contributions | |
confines | -10.836 |
slightest | -9.948 |
bounds | -9.105 |
assumption | -8.977 |
margins | -8.430 |
Token | sexual |
Token ID | 3206 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
zin | +9.191 |
umat | +9.120 |
fuelled | +9.023 |
andals | +8.906 |
piring | +8.590 |
Neg logit contributions | |
confines | -11.672 |
slightest | -10.517 |
bounds | -9.714 |
assumption | -9.445 |
margins | -8.967 |
Token | assault |
Token ID | 4641 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
zin | +13.006 |
umat | +12.826 |
fuelled | +12.732 |
andals | +12.405 |
piring | +11.892 |
Neg logit contributions | |
confines | -16.912 |
slightest | -15.174 |
bounds | -14.194 |
assumption | -13.644 |
margins | -13.055 |
Token |
|
Token ID | 198 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
zin | +14.395 |
umat | +14.393 |
fuelled | +14.162 |
andals | +14.148 |
piring | +13.146 |
Neg logit contributions | |
confines | -15.946 |
slightest | -14.594 |
bounds | -13.017 |
assumption | -12.766 |
margins | -12.092 |
Token |
|
Token ID | 198 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
confines | +3.729 |
slightest | +3.346 |
bounds | +3.239 |
assumption | +3.146 |
margins | +2.969 |
Neg logit contributions | |
zin | -2.090 |
umat | -2.078 |
fuelled | -1.985 |
andals | -1.981 |
ijing | -1.902 |
Token | NEWS |
Token ID | 49597 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
zin | +0.081 |
umat | +0.079 |
fuelled | +0.077 |
andals | +0.075 |
ijing | +0.072 |
Neg logit contributions | |
confines | -0.135 |
slightest | -0.123 |
bounds | -0.116 |
assumption | -0.114 |
margins | -0.107 |
Token | LE |
Token ID | 2538 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
zin | +7.790 |
umat | +7.703 |
fuelled | +7.379 |
andals | +7.203 |
piring | +6.879 |
Neg logit contributions | |
confines | -12.557 |
slightest | -11.561 |
bounds | -10.631 |
assumption | -10.502 |
margins | -9.798 |
Token | - |
Token ID | 12 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
zin | +6.130 |
umat | +5.785 |
fuelled | +5.684 |
andals | +5.455 |
piring | +5.307 |
Neg logit contributions | |
confines | -8.870 |
slightest | -8.122 |
bounds | -7.340 |
assumption | -7.243 |
margins | -6.839 |
Token | i |
Token ID | 72 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
zin | +7.342 |
umat | +6.710 |
fuelled | +6.167 |
andals | +6.041 |
ijing | +5.884 |
Neg logit contributions | |
confines | -17.623 |
slightest | -16.267 |
assumption | -14.839 |
bounds | -14.705 |
margins | -14.071 |
Token | ", |
Token ID | 1600 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
zin | +4.644 |
umat | +4.449 |
fuelled | +4.216 |
andals | +4.072 |
ijing | +3.867 |
Neg logit contributions | |
confines | -11.609 |
slightest | -10.929 |
assumption | -9.888 |
bounds | -9.854 |
margins | -9.381 |
Token | " |
Token ID | 366 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
zin | +0.469 |
umat | +0.459 |
fuelled | +0.443 |
andals | +0.439 |
ijing | +0.417 |
Neg logit contributions | |
confines | -0.652 |
slightest | -0.593 |
bounds | -0.548 |
assumption | -0.539 |
margins | -0.504 |
Token | -- |
Token ID | 438 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
zin | +7.435 |
umat | +7.130 |
fuelled | +6.843 |
andals | +6.812 |
piring | +6.509 |
Neg logit contributions | |
confines | -11.017 |
slightest | -10.179 |
bounds | -9.110 |
assumption | -9.102 |
margins | -8.466 |
Token | images |
Token ID | 17566 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
zin | +6.712 |
umat | +6.450 |
andals | +5.985 |
fuelled | +5.735 |
piring | +5.560 |
Neg logit contributions | |
confines | -15.782 |
slightest | -14.762 |
assumption | -13.604 |
bounds | -13.442 |
margins | -12.472 |
Token | ", |
Token ID | 1600 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
zin | +7.309 |
umat | +6.974 |
fuelled | +6.590 |
andals | +6.027 |
af | +6.025 |
Neg logit contributions | |
confines | -23.536 |
slightest | -22.537 |
assumption | -20.051 |
bounds | -19.698 |
margins | -19.122 |
Token | required |
Token ID | 2672 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
zin | +6.470 |
umat | +6.316 |
fuelled | +6.219 |
andals | +6.046 |
piring | +5.867 |
Neg logit contributions | |
confines | -6.744 |
slightest | -6.039 |
bounds | -5.435 |
assumption | -5.353 |
margins | -4.929 |
Token | = |
Token ID | 28 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
zin | +2.494 |
umat | +2.394 |
fuelled | +2.289 |
andals | +2.267 |
piring | +2.142 |
Neg logit contributions | |
confines | -4.881 |
slightest | -4.567 |
bounds | -4.123 |
assumption | -4.098 |
margins | -3.876 |
Token | True |
Token ID | 17821 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
zin | +2.286 |
umat | +2.230 |
fuelled | +2.148 |
andals | +2.127 |
piring | +2.053 |
Neg logit contributions | |
confines | -3.027 |
slightest | -2.763 |
bounds | -2.505 |
assumption | -2.477 |
margins | -2.317 |
Token | , |
Token ID | 11 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
zin | +5.147 |
umat | +4.841 |
fuelled | +4.718 |
andals | +4.515 |
piring | +4.386 |
Neg logit contributions | |
confines | -10.099 |
slightest | -9.664 |
assumption | -8.541 |
bounds | -8.512 |
margins | -8.064 |
Token | Organization |
Token ID | 12275 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
zin | +3.826 |
umat | +3.759 |
fuelled | +3.646 |
andals | +3.633 |
ijing | +3.466 |
Neg logit contributions | |
confines | -3.969 |
slightest | -3.535 |
bounds | -3.247 |
assumption | -3.186 |
margins | -2.948 |
Token | . |
Token ID | 13 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
confines | +0.838 |
slightest | +0.749 |
bounds | +0.706 |
assumption | +0.697 |
margins | +0.649 |
Neg logit contributions | |
zin | -0.600 |
umat | -0.587 |
andals | -0.563 |
fuelled | -0.563 |
ijing | -0.534 |
Token | In |
Token ID | 554 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
zin | +5.532 |
umat | +5.445 |
fuelled | +5.254 |
andals | +5.211 |
piring | +4.966 |
Neg logit contributions | |
confines | -6.821 |
slightest | -6.228 |
bounds | -5.672 |
assumption | -5.585 |
margins | -5.192 |
Token | juries |
Token ID | 47496 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
zin | +5.573 |
umat | +5.475 |
fuelled | +5.236 |
andals | +5.207 |
piring | +4.971 |
Neg logit contributions | |
confines | -7.725 |
slightest | -7.053 |
bounds | -6.484 |
assumption | -6.384 |
margins | -5.973 |
Token | and |
Token ID | 290 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
zin | +9.046 |
umat | +8.977 |
fuelled | +8.794 |
andals | +8.399 |
ijing | +8.124 |
Neg logit contributions | |
confines | -11.986 |
slightest | -10.986 |
bounds | -9.987 |
assumption | -9.744 |
margins | -9.081 |
Token | violence |
Token ID | 3685 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
zin | +6.570 |
umat | +6.516 |
fuelled | +6.312 |
andals | +6.211 |
ijing | +5.920 |
Neg logit contributions | |
confines | -8.333 |
slightest | -7.585 |
bounds | -6.954 |
assumption | -6.825 |
margins | -6.329 |
Token | : |
Token ID | 25 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
zin | +15.240 |
umat | +14.909 |
fuelled | +14.863 |
andals | +14.022 |
ijing | +13.685 |
Neg logit contributions | |
confines | -15.675 |
slightest | -14.595 |
bounds | -12.760 |
assumption | -12.367 |
margins | -11.362 |
Token | the |
Token ID | 262 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
zin | +8.100 |
umat | +8.007 |
fuelled | +7.857 |
andals | +7.670 |
ijing | +7.369 |
Neg logit contributions | |
confines | -8.465 |
slightest | -7.678 |
bounds | -6.891 |
assumption | -6.732 |
margins | -6.187 |
Token | facts |
Token ID | 6419 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
zin | +2.542 |
umat | +2.505 |
fuelled | +2.430 |
andals | +2.408 |
ijing | +2.317 |
Neg logit contributions | |
confines | -2.376 |
slightest | -2.117 |
bounds | -1.922 |
assumption | -1.879 |
margins | -1.727 |
Token | . |
Token ID | 13 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
zin | +5.326 |
umat | +5.228 |
fuelled | +5.056 |
andals | +4.993 |
ijing | +4.776 |
Neg logit contributions | |
confines | -6.742 |
slightest | -6.116 |
bounds | -5.626 |
assumption | -5.518 |
margins | -5.150 |
Token | Geneva |
Token ID | 20552 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
zin | +2.087 |
umat | +2.057 |
fuelled | +1.999 |
andals | +1.987 |
piring | +1.907 |
Neg logit contributions | |
confines | -1.790 |
slightest | -1.561 |
bounds | -1.431 |
assumption | -1.395 |
margins | -1.279 |
Token | campus |
Token ID | 7611 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
confines | +1.451 |
slightest | +1.238 |
bounds | +1.143 |
assumption | +1.111 |
margins | +1.015 |
Neg logit contributions | |
zin | -1.885 |
umat | -1.837 |
fuelled | -1.795 |
andals | -1.779 |
ijing | -1.736 |
Token | after |
Token ID | 706 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
zin | +1.725 |
umat | +1.679 |
fuelled | +1.604 |
andals | +1.590 |
ijing | +1.508 |
Neg logit contributions | |
confines | -3.121 |
slightest | -2.851 |
bounds | -2.674 |
assumption | -2.636 |
margins | -2.483 |
Token | report |
Token ID | 989 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
zin | +6.462 |
umat | +6.365 |
fuelled | +6.211 |
andals | +6.173 |
piring | +5.943 |
Neg logit contributions | |
confines | -6.566 |
slightest | -5.884 |
bounds | -5.377 |
assumption | -5.224 |
margins | -4.861 |
Token | of |
Token ID | 286 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
zin | +7.816 |
umat | +7.690 |
andals | +7.471 |
fuelled | +7.447 |
ijing | +6.947 |
Neg logit contributions | |
confines | -10.836 |
slightest | -9.948 |
bounds | -9.105 |
assumption | -8.977 |
margins | -8.430 |
Token | sexual |
Token ID | 3206 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
zin | +9.191 |
umat | +9.120 |
fuelled | +9.023 |
andals | +8.906 |
piring | +8.590 |
Neg logit contributions | |
confines | -11.672 |
slightest | -10.517 |
bounds | -9.714 |
assumption | -9.445 |
margins | -8.967 |
Token | assault |
Token ID | 4641 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
zin | +13.006 |
umat | +12.826 |
fuelled | +12.732 |
andals | +12.405 |
piring | +11.892 |
Neg logit contributions | |
confines | -16.912 |
slightest | -15.174 |
bounds | -14.194 |
assumption | -13.644 |
margins | -13.055 |
Token |
|
Token ID | 198 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
zin | +14.395 |
umat | +14.393 |
fuelled | +14.162 |
andals | +14.148 |
piring | +13.146 |
Neg logit contributions | |
confines | -15.946 |
slightest | -14.594 |
bounds | -13.017 |
assumption | -12.766 |
margins | -12.092 |
Token |
|
Token ID | 198 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
confines | +3.729 |
slightest | +3.346 |
bounds | +3.239 |
assumption | +3.146 |
margins | +2.969 |
Neg logit contributions | |
zin | -2.090 |
umat | -2.078 |
fuelled | -1.985 |
andals | -1.981 |
ijing | -1.902 |
Token | NEWS |
Token ID | 49597 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
zin | +0.081 |
umat | +0.079 |
fuelled | +0.077 |
andals | +0.075 |
ijing | +0.072 |
Neg logit contributions | |
confines | -0.135 |
slightest | -0.123 |
bounds | -0.116 |
assumption | -0.114 |
margins | -0.107 |
Token | LE |
Token ID | 2538 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
zin | +7.790 |
umat | +7.703 |
fuelled | +7.379 |
andals | +7.203 |
piring | +6.879 |
Neg logit contributions | |
confines | -12.557 |
slightest | -11.561 |
bounds | -10.631 |
assumption | -10.502 |
margins | -9.798 |
Token | TT |
Token ID | 15751 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
confines | +0.296 |
slightest | +0.268 |
bounds | +0.253 |
assumption | +0.250 |
margins | +0.234 |
Neg logit contributions | |
zin | -0.177 |
umat | -0.174 |
fuelled | -0.168 |
andals | -0.165 |
ijing | -0.155 |
Token | river |
Token ID | 38291 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
zin | +1.480 |
umat | +1.452 |
fuelled | +1.395 |
andals | +1.381 |
ijing | +1.309 |
Neg logit contributions | |
confines | -2.294 |
slightest | -2.094 |
bounds | -1.950 |
assumption | -1.915 |
margins | -1.800 |
Token | Award |
Token ID | 11289 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
zin | +0.374 |
umat | +0.366 |
fuelled | +0.353 |
andals | +0.350 |
ijing | +0.331 |
Neg logit contributions | |
confines | -0.565 |
slightest | -0.517 |
bounds | -0.480 |
assumption | -0.474 |
margins | -0.443 |
Token | for |
Token ID | 329 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
zin | +3.855 |
umat | +3.792 |
fuelled | +3.687 |
andals | +3.650 |
piring | +3.511 |
Neg logit contributions | |
confines | -3.887 |
slightest | -3.456 |
bounds | -3.158 |
assumption | -3.072 |
margins | -2.847 |
Token | Dist |
Token ID | 4307 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
zin | +5.910 |
umat | +5.846 |
fuelled | +5.574 |
andals | +5.556 |
piring | +5.315 |
Neg logit contributions | |
confines | -8.842 |
slightest | -8.040 |
bounds | -7.498 |
assumption | -7.335 |
margins | -6.862 |
Token | inguished |
Token ID | 46709 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
zin | +4.684 |
umat | +4.559 |
andals | +4.106 |
reet | +4.089 |
piring | +3.916 |
Neg logit contributions | |
confines | -21.029 |
slightest | -19.801 |
assumption | -18.586 |
bounds | -18.394 |
margins | -17.502 |
Token | Human |
Token ID | 5524 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
zin | +6.777 |
umat | +6.599 |
fuelled | +6.196 |
andals | +6.097 |
piring | +5.992 |
Neg logit contributions | |
confines | -12.251 |
slightest | -11.369 |
bounds | -10.454 |
assumption | -10.280 |
margins | -9.756 |
Token | itarian |
Token ID | 8353 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
umat | +8.477 |
zin | +8.264 |
piring | +7.383 |
fuelled | +7.179 |
af | +7.103 |
Neg logit contributions | |
confines | -22.856 |
slightest | -21.176 |
bounds | -19.561 |
assumption | -19.395 |
margins | -18.533 |
Token | Service |
Token ID | 4809 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
zin | +8.705 |
umat | +8.008 |
Started | +7.777 |
fuelled | +7.744 |
piring | +7.600 |
Neg logit contributions | |
confines | -16.772 |
slightest | -15.799 |
bounds | -14.163 |
assumption | -13.794 |
margins | -13.283 |
Token | winner |
Token ID | 8464 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
zin | +6.761 |
umat | +6.656 |
fuelled | +6.407 |
andals | +6.321 |
piring | +6.034 |
Neg logit contributions | |
confines | -8.929 |
slightest | -8.098 |
bounds | -7.458 |
assumption | -7.274 |
margins | -6.835 |
Token | was |
Token ID | 373 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
confines | +1.411 |
slightest | +1.274 |
bounds | +1.197 |
assumption | +1.172 |
margins | +1.100 |
Neg logit contributions | |
zin | -0.926 |
umat | -0.903 |
andals | -0.872 |
fuelled | -0.860 |
ijing | -0.822 |
Token | Ar |
Token ID | 943 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
zin | +5.113 |
umat | +5.024 |
fuelled | +4.859 |
andals | +4.812 |
ijing | +4.602 |
Neg logit contributions | |
confines | -6.105 |
slightest | -5.515 |
bounds | -5.071 |
assumption | -4.979 |
margins | -4.634 |
Token |
|
Token ID | 198 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
confines | +2.969 |
slightest | +2.692 |
bounds | +2.524 |
assumption | +2.483 |
margins | +2.308 |
Neg logit contributions | |
zin | -2.004 |
umat | -1.982 |
andals | -1.877 |
fuelled | -1.829 |
ijing | -1.790 |
Token |
|
Token ID | 198 |
Feature activation | +7.53e-03 |
Pos logit contributions | |
confines | +4.124 |
slightest | +3.704 |
bounds | +3.580 |
assumption | +3.482 |
margins | +3.293 |
Neg logit contributions | |
zin | -2.299 |
umat | -2.285 |
fuelled | -2.171 |
andals | -2.163 |
ijing | -2.091 |
Token | St |
Token ID | 1273 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
zin | +0.023 |
umat | +0.023 |
fuelled | +0.022 |
andals | +0.022 |
ijing | +0.021 |
Neg logit contributions | |
confines | -0.039 |
slightest | -0.036 |
bounds | -0.033 |
assumption | -0.033 |
margins | -0.031 |
Token | ri |
Token ID | 380 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
zin | +3.460 |
umat | +3.381 |
fuelled | +3.222 |
andals | +3.205 |
piring | +2.993 |
Neg logit contributions | |
confines | -6.719 |
slightest | -6.179 |
bounds | -5.777 |
assumption | -5.697 |
margins | -5.377 |
Token | king |
Token ID | 3364 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
zin | +6.867 |
umat | +6.728 |
andals | +6.514 |
fuelled | +6.437 |
piring | +6.224 |
Neg logit contributions | |
confines | -7.728 |
slightest | -6.956 |
bounds | -6.377 |
assumption | -6.266 |
margins | -5.804 |
Token | images |
Token ID | 4263 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
zin | +6.376 |
umat | +6.285 |
fuelled | +6.100 |
andals | +6.074 |
piring | +5.783 |
Neg logit contributions | |
confines | -7.758 |
slightest | -7.018 |
bounds | -6.419 |
assumption | -6.319 |
margins | -5.837 |
Token | capture |
Token ID | 8006 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
umat | +13.989 |
zin | +13.896 |
fuelled | +13.891 |
andals | +13.108 |
piring | +12.976 |
Neg logit contributions | |
confines | -15.931 |
slightest | -14.546 |
bounds | -12.881 |
assumption | -12.590 |
margins | -11.911 |
Token | the |
Token ID | 262 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
zin | +9.224 |
umat | +9.128 |
andals | +9.058 |
fuelled | +9.028 |
piring | +8.556 |
Neg logit contributions | |
confines | -12.385 |
slightest | -11.143 |
bounds | -10.221 |
assumption | -10.124 |
margins | -9.255 |
Token | moment |
Token ID | 2589 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
zin | +4.435 |
umat | +4.388 |
fuelled | +4.266 |
andals | +4.217 |
piring | +4.076 |
Neg logit contributions | |
confines | -3.761 |
slightest | -3.314 |
bounds | -3.004 |
assumption | -2.939 |
margins | -2.679 |
Token | more |
Token ID | 517 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
zin | +7.589 |
fuelled | +7.520 |
andals | +7.409 |
umat | +7.343 |
piring | +6.999 |
Neg logit contributions | |
confines | -12.489 |
slightest | -11.428 |
bounds | -10.463 |
assumption | -10.376 |
margins | -9.612 |
Token | than |
Token ID | 621 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
zin | +6.648 |
fuelled | +6.332 |
umat | +6.285 |
andals | +5.976 |
piring | +5.842 |
Neg logit contributions | |
confines | -8.582 |
slightest | -7.902 |
bounds | -6.963 |
assumption | -6.827 |
margins | -6.508 |
Token | � |
Token ID | 239 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
zin | +5.056 |
umat | +4.994 |
fuelled | +4.752 |
andals | +4.677 |
ijing | +4.430 |
Neg logit contributions | |
confines | -9.327 |
slightest | -8.577 |
bounds | -8.015 |
assumption | -7.872 |
margins | -7.445 |
Token | м |
Token ID | 43108 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
zin | +1.908 |
umat | +1.848 |
fuelled | +1.733 |
andals | +1.711 |
ijing | +1.566 |
Neg logit contributions | |
confines | -5.610 |
slightest | -5.208 |
bounds | -4.918 |
assumption | -4.853 |
margins | -4.622 |
Token | н |
Token ID | 22177 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
zin | +1.890 |
umat | +1.841 |
fuelled | +1.743 |
andals | +1.724 |
ijing | +1.602 |
Neg logit contributions | |
confines | -4.426 |
slightest | -4.090 |
bounds | -3.844 |
assumption | -3.791 |
margins | -3.596 |
Token | и |
Token ID | 18849 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
zin | +4.823 |
umat | +4.639 |
fuelled | +4.476 |
andals | +4.379 |
ijing | +4.152 |
Neg logit contributions | |
confines | -10.279 |
slightest | -9.498 |
bounds | -8.894 |
assumption | -8.757 |
margins | -8.303 |
Token | к |
Token ID | 31583 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
zin | +3.110 |
umat | +2.981 |
fuelled | +2.829 |
andals | +2.772 |
ijing | +2.569 |
Neg logit contributions | |
confines | -8.757 |
slightest | -8.122 |
bounds | -7.667 |
assumption | -7.553 |
margins | -7.192 |
Token | о� |
Token ID | 25443 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
zin | +3.698 |
umat | +3.597 |
fuelled | +3.441 |
andals | +3.397 |
ijing | +3.186 |
Neg logit contributions | |
confines | -7.410 |
slightest | -6.824 |
bounds | -6.387 |
assumption | -6.297 |
margins | -5.949 |
Token | � |
Token ID | 110 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
zin | +7.961 |
umat | +7.921 |
fuelled | +7.298 |
andals | +6.986 |
ijing | +6.818 |
Neg logit contributions | |
confines | -21.479 |
slightest | -19.876 |
bounds | -18.855 |
assumption | -18.601 |
margins | -17.393 |
Token | |
Token ID | 220 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
zin | +9.673 |
umat | +9.322 |
fuelled | +9.145 |
andals | +8.816 |
ijing | +8.537 |
Neg logit contributions | |
confines | -14.303 |
slightest | -13.170 |
bounds | -12.094 |
assumption | -11.892 |
margins | -11.164 |
Token | р |
Token ID | 21169 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
zin | +7.076 |
umat | +7.003 |
fuelled | +6.541 |
andals | +6.383 |
piring | +6.096 |
Neg logit contributions | |
confines | -16.634 |
slightest | -15.482 |
bounds | -14.401 |
assumption | -14.358 |
margins | -13.550 |
Token | а |
Token ID | 16142 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
zin | +4.137 |
umat | +4.022 |
fuelled | +3.813 |
andals | +3.768 |
ijing | +3.536 |
Neg logit contributions | |
confines | -9.426 |
slightest | -8.705 |
bounds | -8.153 |
assumption | -8.043 |
margins | -7.640 |
Token | н |
Token ID | 22177 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
zin | +4.838 |
umat | +4.679 |
fuelled | +4.424 |
andals | +4.337 |
ijing | +4.031 |
Neg logit contributions | |
confines | -12.516 |
slightest | -11.649 |
bounds | -10.949 |
assumption | -10.800 |
margins | -10.258 |
Token | 1 |
Token ID | 352 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
zin | +2.005 |
umat | +1.967 |
fuelled | +1.902 |
andals | +1.886 |
ijing | +1.802 |
Neg logit contributions | |
confines | -2.426 |
slightest | -2.191 |
bounds | -2.018 |
assumption | -1.978 |
margins | -1.839 |
Token | I |
Token ID | 314 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
zin | +0.469 |
umat | +0.461 |
fuelled | +0.443 |
andals | +0.441 |
ijing | +0.420 |
Neg logit contributions | |
confines | -0.584 |
slightest | -0.528 |
bounds | -0.488 |
assumption | -0.480 |
margins | -0.447 |
Token | think |
Token ID | 892 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
zin | +6.089 |
umat | +5.924 |
fuelled | +5.847 |
andals | +5.653 |
ijing | +5.440 |
Neg logit contributions | |
confines | -6.992 |
slightest | -6.284 |
bounds | -5.751 |
assumption | -5.570 |
margins | -5.261 |
Token | for |
Token ID | 329 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
zin | +2.867 |
umat | +2.824 |
fuelled | +2.746 |
andals | +2.712 |
piring | +2.594 |
Neg logit contributions | |
confines | -3.267 |
slightest | -2.939 |
bounds | -2.689 |
assumption | -2.635 |
margins | -2.439 |
Token | quite |
Token ID | 2407 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
zin | +4.972 |
umat | +4.860 |
fuelled | +4.755 |
andals | +4.670 |
piring | +4.517 |
Neg logit contributions | |
confines | -4.729 |
slightest | -4.196 |
bounds | -3.799 |
assumption | -3.708 |
margins | -3.430 |
Token | a |
Token ID | 257 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
zin | +8.998 |
umat | +8.621 |
fuelled | +8.561 |
af | +8.168 |
andals | +7.981 |
Neg logit contributions | |
confines | -15.999 |
slightest | -14.544 |
bounds | -13.120 |
assumption | -12.674 |
margins | -12.239 |
Token | while |
Token ID | 981 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
zin | +6.697 |
umat | +6.267 |
ilst | +6.255 |
fuelled | +6.222 |
andals | +5.798 |
Neg logit contributions | |
confines | -21.824 |
slightest | -19.945 |
bounds | -18.973 |
assumption | -17.944 |
margins | -17.917 |
Token | will |
Token ID | 481 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
zin | +7.209 |
umat | +6.980 |
fuelled | +6.905 |
andals | +6.598 |
ijing | +6.374 |
Neg logit contributions | |
confines | -9.902 |
slightest | -9.156 |
bounds | -8.234 |
assumption | -8.048 |
margins | -7.497 |
Token | always |
Token ID | 1464 |
Feature activation | +9.80e-01 |
Pos logit contributions | |
zin | +5.378 |
umat | +5.302 |
fuelled | +5.169 |
andals | +5.090 |
piring | +4.897 |
Neg logit contributions | |
confines | -4.445 |
slightest | -3.953 |
bounds | -3.530 |
assumption | -3.378 |
margins | -3.127 |
Token | be |
Token ID | 307 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
zin | +3.999 |
umat | +3.909 |
fuelled | +3.816 |
andals | +3.758 |
piring | +3.617 |
Neg logit contributions | |
confines | -4.056 |
slightest | -3.691 |
bounds | -3.287 |
assumption | -3.193 |
margins | -2.991 |
Token | visible |
Token ID | 7424 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
zin | +5.539 |
umat | +5.359 |
fuelled | +5.260 |
andals | +5.100 |
piring | +4.967 |
Neg logit contributions | |
confines | -8.283 |
slightest | -7.533 |
bounds | -6.940 |
assumption | -6.802 |
margins | -6.436 |
Token | Occupy |
Token ID | 25551 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
zin | +2.716 |
umat | +2.662 |
fuelled | +2.570 |
andals | +2.547 |
ijing | +2.425 |
Neg logit contributions | |
confines | -3.606 |
slightest | -3.270 |
bounds | -3.022 |
assumption | -2.971 |
margins | -2.774 |
Token | demonstrators |
Token ID | 25016 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
zin | +4.410 |
umat | +4.334 |
fuelled | +4.189 |
andals | +4.156 |
ijing | +3.975 |
Neg logit contributions | |
confines | -5.085 |
slightest | -4.581 |
bounds | -4.211 |
assumption | -4.130 |
margins | -3.838 |
Token | whose |
Token ID | 3025 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
zin | +8.262 |
umat | +7.995 |
fuelled | +7.896 |
andals | +7.683 |
piring | +7.310 |
Neg logit contributions | |
confines | -11.648 |
slightest | -10.654 |
bounds | -9.819 |
assumption | -9.794 |
margins | -9.110 |
Token | protests |
Token ID | 8536 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
zin | +8.342 |
umat | +8.212 |
fuelled | +8.071 |
andals | +7.988 |
piring | +7.598 |
Neg logit contributions | |
confines | -7.748 |
slightest | -6.870 |
bounds | -6.218 |
assumption | -6.051 |
margins | -5.589 |
Token | are |
Token ID | 389 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
fuelled | +12.478 |
zin | +12.145 |
umat | +12.104 |
andals | +11.941 |
piring | +11.178 |
Neg logit contributions | |
confines | -14.518 |
slightest | -13.287 |
assumption | -11.883 |
bounds | -11.880 |
margins | -10.769 |
Token | being |
Token ID | 852 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
zin | +7.879 |
fuelled | +7.764 |
umat | +7.751 |
andals | +7.460 |
piring | +7.145 |
Neg logit contributions | |
confines | -9.550 |
slightest | -8.615 |
bounds | -7.913 |
assumption | -7.860 |
margins | -7.264 |
Token | disbanded |
Token ID | 47302 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
fuelled | +13.804 |
umat | +13.173 |
zin | +13.052 |
andals | +12.188 |
propelled | +11.899 |
Neg logit contributions | |
confines | -16.031 |
slightest | -14.373 |
assumption | -13.411 |
bounds | -13.150 |
margins | -12.302 |
Token | -- |
Token ID | 1377 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
zin | +1.536 |
umat | +1.514 |
fuelled | +1.450 |
andals | +1.438 |
ijing | +1.378 |
Neg logit contributions | |
confines | -2.128 |
slightest | -1.926 |
bounds | -1.792 |
assumption | -1.756 |
margins | -1.645 |
Token | sometimes |
Token ID | 3360 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
zin | +4.538 |
umat | +4.443 |
fuelled | +4.340 |
andals | +4.287 |
piring | +4.091 |
Neg logit contributions | |
confines | -4.889 |
slightest | -4.390 |
bounds | -4.010 |
assumption | -3.929 |
margins | -3.629 |
Token | violently |
Token ID | 25856 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
zin | +2.296 |
umat | +2.257 |
fuelled | +2.171 |
andals | +2.161 |
ijing | +2.067 |
Neg logit contributions | |
confines | -2.805 |
slightest | -2.534 |
bounds | -2.336 |
assumption | -2.293 |
margins | -2.135 |
Token | -- |
Token ID | 1377 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
zin | +7.052 |
umat | +6.903 |
fuelled | +6.715 |
andals | +6.605 |
piring | +6.267 |
Neg logit contributions | |
confines | -9.040 |
slightest | -8.233 |
bounds | -7.578 |
assumption | -7.460 |
margins | -6.948 |
Token | C |
Token ID | 327 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
zin | +6.315 |
umat | +6.141 |
fuelled | +5.975 |
andals | +5.892 |
ijing | +5.686 |
Neg logit contributions | |
confines | -7.505 |
slightest | -6.824 |
bounds | -6.241 |
assumption | -6.141 |
margins | -5.653 |
Token | aj |
Token ID | 1228 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
zin | +3.597 |
umat | +3.513 |
andals | +3.198 |
fuelled | +3.129 |
ijing | +2.937 |
Neg logit contributions | |
confines | -9.906 |
slightest | -9.299 |
bounds | -8.606 |
assumption | -8.573 |
margins | -8.117 |
Token | un |
Token ID | 403 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
zin | +3.058 |
umat | +3.007 |
andals | +2.760 |
fuelled | +2.739 |
ijing | +2.583 |
Neg logit contributions | |
confines | -6.832 |
slightest | -6.358 |
bounds | -5.890 |
assumption | -5.848 |
margins | -5.534 |
Token | Kitchen |
Token ID | 22541 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
zin | +2.462 |
umat | +2.423 |
fuelled | +2.361 |
andals | +2.346 |
ijing | +2.271 |
Neg logit contributions | |
confines | -1.698 |
slightest | -1.482 |
bounds | -1.317 |
assumption | -1.282 |
margins | -1.152 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
zin | +3.687 |
umat | +3.599 |
fuelled | +3.451 |
andals | +3.412 |
ijing | +3.222 |
Neg logit contributions | |
confines | -6.670 |
slightest | -6.133 |
bounds | -5.713 |
assumption | -5.629 |
margins | -5.300 |
Token | Recently |
Token ID | 24661 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
zin | +0.543 |
umat | +0.532 |
fuelled | +0.516 |
andals | +0.507 |
ijing | +0.478 |
Neg logit contributions | |
confines | -0.893 |
slightest | -0.815 |
bounds | -0.763 |
assumption | -0.752 |
margins | -0.707 |
Token | , |
Token ID | 11 |
Feature activation | +9.24e-01 |
Pos logit contributions | |
zin | +13.799 |
umat | +13.675 |
fuelled | +13.475 |
andals | +13.084 |
ijing | +12.691 |
Neg logit contributions | |
confines | -15.648 |
slightest | -14.251 |
bounds | -12.953 |
assumption | -12.542 |
margins | -11.686 |
Token | there |
Token ID | 612 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
zin | +3.612 |
umat | +3.550 |
fuelled | +3.432 |
andals | +3.407 |
ijing | +3.258 |
Neg logit contributions | |
confines | -4.099 |
slightest | -3.690 |
bounds | -3.389 |
assumption | -3.324 |
margins | -3.086 |
Token | � |
Token ID | 447 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
zin | +4.106 |
umat | +4.022 |
fuelled | +3.866 |
andals | +3.831 |
ijing | +3.631 |
Neg logit contributions | |
confines | -6.279 |
slightest | -5.729 |
bounds | -5.322 |
assumption | -5.235 |
margins | -4.914 |
Token | � |
Token ID | 247 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
confines | +4.150 |
slightest | +3.751 |
bounds | +3.497 |
assumption | +3.470 |
margins | +3.218 |
Neg logit contributions | |
zin | -2.986 |
umat | -2.844 |
andals | -2.788 |
fuelled | -2.780 |
ilst | -2.619 |
Token | s |
Token ID | 82 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
zin | +1.963 |
umat | +1.920 |
fuelled | +1.835 |
andals | +1.822 |
ijing | +1.724 |
Neg logit contributions | |
confines | -3.150 |
slightest | -2.895 |
bounds | -2.689 |
assumption | -2.657 |
margins | -2.490 |
Token | � |
Token ID | 140 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
zin | +3.307 |
umat | +3.185 |
fuelled | +3.008 |
andals | +2.957 |
ijing | +2.721 |
Neg logit contributions | |
confines | -9.179 |
slightest | -8.527 |
bounds | -8.019 |
assumption | -7.924 |
margins | -7.535 |
Token | � |
Token ID | 111 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
zin | +4.807 |
umat | +4.721 |
fuelled | +4.481 |
andals | +4.402 |
ijing | +4.184 |
Neg logit contributions | |
confines | -9.354 |
slightest | -8.588 |
bounds | -8.076 |
assumption | -7.903 |
margins | -7.497 |
Token | н |
Token ID | 22177 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
zin | +4.832 |
umat | +4.693 |
fuelled | +4.465 |
andals | +4.345 |
ijing | +4.083 |
Neg logit contributions | |
confines | -10.589 |
slightest | -9.769 |
bounds | -9.161 |
assumption | -9.029 |
margins | -8.570 |
Token | е |
Token ID | 16843 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
zin | +5.964 |
umat | +5.806 |
fuelled | +5.560 |
andals | +5.343 |
ijing | +5.092 |
Neg logit contributions | |
confines | -13.374 |
slightest | -12.381 |
bounds | -11.517 |
assumption | -11.401 |
margins | -10.888 |
Token | р |
Token ID | 21169 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
zin | +4.594 |
umat | +4.389 |
fuelled | +4.197 |
andals | +4.101 |
ijing | +3.817 |
Neg logit contributions | |
confines | -11.545 |
slightest | -10.692 |
bounds | -10.054 |
assumption | -9.909 |
margins | -9.426 |
Token | о� |
Token ID | 25443 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
zin | +6.042 |
umat | +5.826 |
fuelled | +5.640 |
andals | +5.484 |
ijing | +5.188 |
Neg logit contributions | |
confines | -11.840 |
slightest | -10.928 |
bounds | -10.159 |
assumption | -10.038 |
margins | -9.486 |
Token | � |
Token ID | 110 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
zin | +8.467 |
umat | +8.419 |
fuelled | +7.757 |
andals | +7.541 |
ijing | +7.398 |
Neg logit contributions | |
confines | -20.952 |
slightest | -19.422 |
bounds | -18.357 |
assumption | -18.022 |
margins | -16.990 |
Token | с |
Token ID | 21727 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
zin | +7.469 |
umat | +7.281 |
fuelled | +6.994 |
andals | +6.763 |
ijing | +6.472 |
Neg logit contributions | |
confines | -14.755 |
slightest | -13.735 |
bounds | -12.740 |
assumption | -12.535 |
margins | -11.914 |
Token | к |
Token ID | 31583 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
zin | +3.938 |
umat | +3.825 |
fuelled | +3.617 |
andals | +3.554 |
ijing | +3.313 |
Neg logit contributions | |
confines | -9.798 |
slightest | -9.076 |
bounds | -8.537 |
assumption | -8.413 |
margins | -7.995 |
Token | а |
Token ID | 16142 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
zin | +4.677 |
umat | +4.551 |
fuelled | +4.356 |
andals | +4.258 |
ijing | +4.026 |
Neg logit contributions | |
confines | -9.641 |
slightest | -8.880 |
bounds | -8.318 |
assumption | -8.195 |
margins | -7.782 |
Token | я |
Token ID | 40623 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
zin | +6.019 |
umat | +5.792 |
fuelled | +5.646 |
andals | +5.430 |
ijing | +5.114 |
Neg logit contributions | |
confines | -13.293 |
slightest | -12.362 |
bounds | -11.535 |
assumption | -11.361 |
margins | -10.793 |
Token | he |
Token ID | 339 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
confines | +5.497 |
slightest | +4.893 |
bounds | +4.603 |
assumption | +4.481 |
margins | +4.104 |
Neg logit contributions | |
umat | -4.851 |
zin | -4.817 |
andals | -4.642 |
ijing | -4.516 |
fuelled | -4.510 |
Token | added |
Token ID | 2087 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
confines | +0.286 |
slightest | +0.256 |
bounds | +0.236 |
assumption | +0.230 |
margins | +0.213 |
Neg logit contributions | |
zin | -0.262 |
umat | -0.259 |
andals | -0.250 |
fuelled | -0.247 |
ijing | -0.240 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
confines | +0.732 |
slightest | +0.666 |
bounds | +0.627 |
assumption | +0.615 |
margins | +0.579 |
Neg logit contributions | |
zin | -0.435 |
umat | -0.432 |
andals | -0.410 |
fuelled | -0.405 |
ijing | -0.390 |
Token |
|
Token ID | 198 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
zin | +3.054 |
umat | +2.957 |
fuelled | +2.812 |
andals | +2.798 |
piring | +2.586 |
Neg logit contributions | |
confines | -6.456 |
slightest | -5.976 |
bounds | -5.575 |
assumption | -5.512 |
margins | -5.210 |
Token |
|
Token ID | 198 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
confines | +5.851 |
slightest | +5.187 |
bounds | +5.100 |
assumption | +4.920 |
margins | +4.643 |
Neg logit contributions | |
umat | -3.502 |
zin | -3.494 |
andals | -3.330 |
fuelled | -3.321 |
ijing | -3.249 |
Token | Thousands |
Token ID | 37482 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
confines | +2.403 |
slightest | +2.157 |
bounds | +2.047 |
assumption | +2.014 |
margins | +1.879 |
Neg logit contributions | |
zin | -1.570 |
umat | -1.558 |
fuelled | -1.510 |
andals | -1.487 |
ijing | -1.433 |
Token | of |
Token ID | 286 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
zin | +12.160 |
umat | +11.752 |
fuelled | +11.521 |
andals | +10.863 |
piring | +10.796 |
Neg logit contributions | |
confines | -17.820 |
slightest | -16.201 |
bounds | -14.694 |
assumption | -14.438 |
margins | -13.530 |
Token | emails |
Token ID | 7237 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
zin | +11.180 |
umat | +10.910 |
fuelled | +10.630 |
andals | +10.429 |
piring | +10.227 |
Neg logit contributions | |
confines | -11.891 |
slightest | -10.585 |
bounds | -9.581 |
assumption | -9.480 |
margins | -8.696 |
Token | were |
Token ID | 547 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
zin | +6.552 |
umat | +6.431 |
fuelled | +6.295 |
andals | +6.179 |
piring | +5.844 |
Neg logit contributions | |
confines | -9.864 |
slightest | -9.020 |
bounds | -8.287 |
assumption | -8.185 |
margins | -7.748 |
Token | leaked |
Token ID | 14109 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
zin | +7.939 |
umat | +7.815 |
fuelled | +7.803 |
andals | +7.605 |
piring | +7.158 |
Neg logit contributions | |
confines | -13.127 |
slightest | -11.889 |
bounds | -11.062 |
assumption | -10.955 |
margins | -10.467 |
Token | to |
Token ID | 284 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
zin | +8.874 |
fuelled | +8.597 |
umat | +8.575 |
andals | +8.340 |
piring | +7.915 |
Neg logit contributions | |
confines | -12.998 |
slightest | -11.931 |
bounds | -10.959 |
assumption | -10.663 |
margins | -10.157 |
Token | by |
Token ID | 416 |
Feature activation | +9.89e-01 |
Pos logit contributions | |
zin | +6.414 |
umat | +6.325 |
fuelled | +6.151 |
andals | +5.956 |
ijing | +5.688 |
Neg logit contributions | |
confines | -9.047 |
slightest | -8.278 |
bounds | -7.609 |
assumption | -7.498 |
margins | -6.960 |
Token | the |
Token ID | 262 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
zin | +3.708 |
umat | +3.638 |
fuelled | +3.523 |
andals | +3.485 |
piring | +3.327 |
Neg logit contributions | |
confines | -4.504 |
slightest | -4.081 |
bounds | -3.748 |
assumption | -3.680 |
margins | -3.426 |
Token | Zika |
Token ID | 30861 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
confines | +0.101 |
slightest | +0.092 |
bounds | +0.085 |
assumption | +0.084 |
margins | +0.079 |
Neg logit contributions | |
zin | -0.068 |
umat | -0.067 |
fuelled | -0.064 |
andals | -0.064 |
ijing | -0.061 |
Token | virus |
Token ID | 9471 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
zin | +2.539 |
umat | +2.501 |
andals | +2.403 |
fuelled | +2.399 |
piring | +2.281 |
Neg logit contributions | |
confines | -3.366 |
slightest | -3.047 |
bounds | -2.822 |
assumption | -2.776 |
margins | -2.588 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
zin | +3.739 |
umat | +3.660 |
fuelled | +3.496 |
andals | +3.479 |
ijing | +3.283 |
Neg logit contributions | |
confines | -6.243 |
slightest | -5.706 |
bounds | -5.324 |
assumption | -5.240 |
margins | -4.930 |
Token | Thousands |
Token ID | 30405 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
zin | +4.794 |
umat | +4.685 |
fuelled | +4.494 |
andals | +4.470 |
piring | +4.227 |
Neg logit contributions | |
confines | -7.354 |
slightest | -6.797 |
bounds | -6.231 |
assumption | -6.139 |
margins | -5.750 |
Token | of |
Token ID | 286 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
zin | +11.646 |
umat | +11.636 |
fuelled | +11.299 |
andals | +10.365 |
piring | +10.343 |
Neg logit contributions | |
confines | -17.868 |
slightest | -16.314 |
bounds | -14.912 |
assumption | -14.404 |
margins | -13.656 |
Token | our |
Token ID | 674 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
umat | +10.140 |
zin | +10.066 |
fuelled | +9.725 |
andals | +9.331 |
piring | +9.256 |
Neg logit contributions | |
confines | -12.704 |
slightest | -11.515 |
bounds | -10.518 |
assumption | -10.181 |
margins | -9.496 |
Token | citizens |
Token ID | 4290 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
zin | +4.257 |
umat | +4.232 |
fuelled | +4.073 |
andals | +4.031 |
piring | +3.900 |
Neg logit contributions | |
confines | -4.290 |
slightest | -3.826 |
bounds | -3.512 |
assumption | -3.406 |
margins | -3.153 |
Token | are |
Token ID | 389 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
zin | +1.123 |
umat | +1.099 |
andals | +1.059 |
fuelled | +1.049 |
ijing | +0.999 |
Neg logit contributions | |
confines | -1.654 |
slightest | -1.508 |
bounds | -1.403 |
assumption | -1.380 |
margins | -1.291 |
Token | exposed |
Token ID | 7362 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
zin | +4.666 |
umat | +4.628 |
fuelled | +4.511 |
andals | +4.353 |
piring | +4.236 |
Neg logit contributions | |
confines | -5.697 |
slightest | -5.138 |
bounds | -4.733 |
assumption | -4.632 |
margins | -4.326 |
Token | of |
Token ID | 286 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
fuelled | +10.984 |
zin | +10.727 |
umat | +10.375 |
andals | +10.154 |
piring | +9.753 |
Neg logit contributions | |
confines | -16.316 |
slightest | -15.050 |
bounds | -13.599 |
assumption | -13.442 |
margins | -12.748 |
Token | rape |
Token ID | 7262 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
fuelled | +9.537 |
zin | +9.507 |
umat | +9.332 |
andals | +8.864 |
piring | +8.842 |
Neg logit contributions | |
confines | -11.993 |
slightest | -10.763 |
bounds | -9.830 |
assumption | -9.635 |
margins | -9.020 |
Token | and |
Token ID | 290 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
zin | +9.456 |
fuelled | +9.224 |
umat | +9.191 |
andals | +8.972 |
piring | +8.478 |
Neg logit contributions | |
confines | -13.092 |
slightest | -11.906 |
bounds | -10.950 |
assumption | -10.737 |
margins | -10.049 |
Token | sexual |
Token ID | 3206 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
zin | +6.069 |
umat | +5.990 |
fuelled | +5.946 |
andals | +5.771 |
piring | +5.489 |
Neg logit contributions | |
confines | -7.429 |
slightest | -6.738 |
bounds | -6.189 |
assumption | -6.051 |
margins | -5.637 |
Token | mol |
Token ID | 18605 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
zin | +9.565 |
fuelled | +9.409 |
umat | +9.337 |
andals | +8.844 |
piring | +8.316 |
Neg logit contributions | |
confines | -15.077 |
slightest | -13.773 |
bounds | -12.799 |
assumption | -12.466 |
margins | -11.692 |
Token | estation |
Token ID | 27364 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
zin | +5.076 |
umat | +4.912 |
andals | +4.520 |
fuelled | +4.415 |
piring | +3.972 |
Neg logit contributions | |
confines | -16.528 |
slightest | -15.468 |
bounds | -14.526 |
assumption | -14.285 |
margins | -13.663 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
fuelled | +12.707 |
zin | +12.493 |
andals | +12.082 |
umat | +12.022 |
piring | +11.354 |
Neg logit contributions | |
confines | -16.660 |
slightest | -15.341 |
bounds | -13.870 |
assumption | -13.677 |
margins | -12.660 |
Token | The |
Token ID | 383 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
zin | +2.915 |
umat | +2.845 |
fuelled | +2.713 |
andals | +2.682 |
ijing | +2.515 |
Neg logit contributions | |
confines | -5.818 |
slightest | -5.350 |
bounds | -5.013 |
assumption | -4.939 |
margins | -4.670 |
Token | 39 |
Token ID | 5014 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
confines | +0.678 |
slightest | +0.599 |
bounds | +0.541 |
assumption | +0.531 |
margins | +0.484 |
Neg logit contributions | |
zin | -0.813 |
umat | -0.798 |
fuelled | -0.774 |
andals | -0.771 |
ijing | -0.742 |
Token | - |
Token ID | 12 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
zin | +0.627 |
umat | +0.611 |
fuelled | +0.593 |
andals | +0.590 |
ijing | +0.561 |
Neg logit contributions | |
confines | -0.718 |
slightest | -0.647 |
bounds | -0.592 |
assumption | -0.582 |
margins | -0.543 |
Token | year |
Token ID | 1941 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
zin | +0.040 |
umat | +0.038 |
fuelled | +0.037 |
andals | +0.036 |
ijing | +0.034 |
Neg logit contributions | |
confines | -0.082 |
slightest | -0.076 |
bounds | -0.071 |
assumption | -0.070 |
margins | -0.066 |
Token | suspects |
Token ID | 11826 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
zin | +3.534 |
umat | +3.473 |
fuelled | +3.384 |
andals | +3.342 |
piring | +3.208 |
Neg logit contributions | |
confines | -4.069 |
slightest | -3.664 |
bounds | -3.363 |
assumption | -3.281 |
margins | -3.053 |
Token | in |
Token ID | 287 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
zin | +6.244 |
umat | +6.128 |
fuelled | +5.954 |
andals | +5.831 |
piring | +5.569 |
Neg logit contributions | |
confines | -8.547 |
slightest | -7.788 |
bounds | -7.182 |
assumption | -7.051 |
margins | -6.622 |
Token | connection |
Token ID | 4637 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
zin | +4.197 |
umat | +4.112 |
fuelled | +4.010 |
andals | +3.960 |
piring | +3.813 |
Neg logit contributions | |
confines | -4.113 |
slightest | -3.680 |
bounds | -3.339 |
assumption | -3.263 |
margins | -3.016 |
Token | with |
Token ID | 351 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
zin | +4.322 |
umat | +4.196 |
fuelled | +4.085 |
andals | +4.018 |
piring | +3.811 |
Neg logit contributions | |
confines | -6.750 |
slightest | -6.170 |
bounds | -5.724 |
assumption | -5.627 |
margins | -5.268 |
Token | the |
Token ID | 262 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
zin | +3.825 |
umat | +3.759 |
fuelled | +3.637 |
andals | +3.608 |
ijing | +3.449 |
Neg logit contributions | |
confines | -4.346 |
slightest | -3.912 |
bounds | -3.593 |
assumption | -3.523 |
margins | -3.271 |
Token | mass |
Token ID | 2347 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
zin | +1.692 |
umat | +1.658 |
fuelled | +1.595 |
andals | +1.581 |
ijing | +1.510 |
Neg logit contributions | |
confines | -2.403 |
slightest | -2.189 |
bounds | -2.024 |
assumption | -1.995 |
margins | -1.865 |
Token | rape |
Token ID | 7262 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
zin | +13.327 |
umat | +12.982 |
fuelled | +12.941 |
andals | +12.700 |
piring | +12.111 |
Neg logit contributions | |
confines | -14.991 |
slightest | -13.817 |
bounds | -12.557 |
assumption | -11.954 |
margins | -11.166 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
zin | +12.156 |
fuelled | +11.953 |
umat | +11.880 |
andals | +11.678 |
piring | +10.950 |
Neg logit contributions | |
confines | -13.781 |
slightest | -12.450 |
bounds | -11.387 |
assumption | -10.859 |
margins | -10.267 |
Token |
|
Token ID | 198 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
zin | +2.985 |
umat | +2.915 |
fuelled | +2.789 |
andals | +2.755 |
ijing | +2.595 |
Neg logit contributions | |
confines | -5.612 |
slightest | -5.147 |
bounds | -4.822 |
assumption | -4.748 |
margins | -4.483 |
Token |
|
Token ID | 198 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
confines | +0.623 |
slightest | +0.568 |
bounds | +0.539 |
assumption | +0.529 |
margins | +0.500 |
Neg logit contributions | |
zin | -0.317 |
umat | -0.312 |
fuelled | -0.296 |
andals | -0.295 |
ijing | -0.279 |
Token | � |
Token ID | 447 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
zin | +1.089 |
umat | +1.067 |
fuelled | +1.031 |
andals | +1.015 |
ijing | +0.960 |
Neg logit contributions | |
confines | -1.878 |
slightest | -1.712 |
bounds | -1.609 |
assumption | -1.585 |
margins | -1.491 |
Token | 24 |
Token ID | 1987 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
zin | +3.573 |
umat | +3.509 |
fuelled | +3.397 |
andals | +3.369 |
ijing | +3.223 |
Neg logit contributions | |
confines | -4.023 |
slightest | -3.621 |
bounds | -3.319 |
assumption | -3.257 |
margins | -3.022 |
Token | Arrest |
Token ID | 35080 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
zin | +2.994 |
umat | +2.940 |
fuelled | +2.833 |
andals | +2.818 |
ijing | +2.679 |
Neg logit contributions | |
confines | -3.962 |
slightest | -3.594 |
bounds | -3.322 |
assumption | -3.263 |
margins | -3.054 |
Token | ed |
Token ID | 276 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
zin | +8.417 |
umat | +8.254 |
andals | +8.029 |
fuelled | +7.977 |
piring | +7.595 |
Neg logit contributions | |
confines | -10.093 |
slightest | -9.138 |
bounds | -8.335 |
assumption | -8.218 |
margins | -7.608 |
Token | at |
Token ID | 379 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
zin | +8.957 |
umat | +8.664 |
andals | +8.432 |
fuelled | +8.335 |
piring | +8.013 |
Neg logit contributions | |
confines | -12.255 |
slightest | -11.158 |
bounds | -10.254 |
assumption | -10.090 |
margins | -9.507 |
Token | Prot |
Token ID | 5038 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
zin | +2.683 |
umat | +2.636 |
fuelled | +2.551 |
andals | +2.529 |
ijing | +2.417 |
Neg logit contributions | |
confines | -3.120 |
slightest | -2.812 |
bounds | -2.587 |
assumption | -2.537 |
margins | -2.357 |
Token | ests |
Token ID | 3558 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
zin | +7.814 |
umat | +7.402 |
andals | +7.218 |
fuelled | +6.891 |
piring | +6.746 |
Neg logit contributions | |
confines | -14.194 |
slightest | -13.165 |
bounds | -12.261 |
assumption | -12.009 |
margins | -11.453 |
Token | & |
Token ID | 1222 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
zin | +13.186 |
umat | +12.536 |
andals | +12.389 |
fuelled | +12.379 |
piring | +11.843 |
Neg logit contributions | |
confines | -15.902 |
slightest | -14.789 |
bounds | -13.214 |
assumption | -13.016 |
margins | -12.276 |
Token | Jill |
Token ID | 22772 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
zin | +5.678 |
umat | +5.559 |
fuelled | +5.373 |
andals | +5.359 |
piring | +5.085 |
Neg logit contributions | |
confines | -7.092 |
slightest | -6.438 |
bounds | -5.909 |
assumption | -5.809 |
margins | -5.414 |
Token | Stein |
Token ID | 15215 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
zin | +2.304 |
umat | +2.274 |
fuelled | +2.210 |
andals | +2.195 |
ijing | +2.111 |
Neg logit contributions | |
confines | -1.989 |
slightest | -1.758 |
bounds | -1.598 |
assumption | -1.556 |
margins | -1.428 |
Token | Esc |
Token ID | 16319 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
zin | +3.130 |
umat | +3.071 |
fuelled | +2.960 |
andals | +2.936 |
ijing | +2.796 |
Neg logit contributions | |
confines | -4.163 |
slightest | -3.776 |
bounds | -3.492 |
assumption | -3.430 |
margins | -3.205 |
Token | orted |
Token ID | 9741 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
zin | +4.084 |
umat | +3.988 |
fuelled | +3.766 |
andals | +3.754 |
ijing | +3.468 |
Neg logit contributions | |
confines | -9.521 |
slightest | -8.810 |
bounds | -8.265 |
assumption | -8.160 |
margins | -7.739 |
Token | ? |
Token ID | 30 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
confines | +1.663 |
slightest | +1.502 |
bounds | +1.401 |
assumption | +1.381 |
margins | +1.298 |
Neg logit contributions | |
zin | -1.132 |
umat | -1.120 |
andals | -1.078 |
fuelled | -1.056 |
ijing | -1.024 |
Token | � |
Token ID | 447 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
confines | +1.706 |
slightest | +1.516 |
bounds | +1.454 |
assumption | +1.418 |
margins | +1.343 |
Neg logit contributions | |
zin | -1.060 |
umat | -1.048 |
fuelled | -1.010 |
andals | -0.993 |
ijing | -0.953 |
Token | � |
Token ID | 251 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
zin | +0.203 |
umat | +0.197 |
fuelled | +0.189 |
andals | +0.188 |
ijing | +0.176 |
Neg logit contributions | |
confines | -0.406 |
slightest | -0.371 |
bounds | -0.351 |
assumption | -0.344 |
margins | -0.325 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
zin | +1.732 |
umat | +1.697 |
fuelled | +1.620 |
andals | +1.605 |
ijing | +1.512 |
Neg logit contributions | |
confines | -3.347 |
slightest | -3.064 |
bounds | -2.878 |
assumption | -2.834 |
margins | -2.678 |
Token |
|
Token ID | 198 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
confines | +6.137 |
slightest | +5.384 |
bounds | +5.337 |
assumption | +5.137 |
margins | +4.838 |
Neg logit contributions | |
umat | -3.771 |
zin | -3.746 |
andals | -3.594 |
fuelled | -3.592 |
ijing | -3.490 |
Token | Inf |
Token ID | 18943 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
confines | +3.963 |
slightest | +3.524 |
bounds | +3.373 |
assumption | +3.302 |
margins | +3.081 |
Neg logit contributions | |
zin | -2.735 |
umat | -2.735 |
fuelled | -2.645 |
andals | -2.612 |
ijing | -2.500 |
Token | am |
Token ID | 321 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
umat | +9.239 |
zin | +9.203 |
andals | +8.517 |
fuelled | +8.151 |
ijing | +7.865 |
Neg logit contributions | |
confines | -20.610 |
slightest | -18.992 |
bounds | -18.094 |
assumption | -17.738 |
margins | -16.592 |
Token | ously |
Token ID | 3481 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
zin | +5.490 |
umat | +5.411 |
fuelled | +5.077 |
andals | +4.979 |
ijing | +4.589 |
Neg logit contributions | |
confines | -15.711 |
slightest | -14.519 |
bounds | -13.735 |
assumption | -13.535 |
margins | -12.858 |
Token | , |
Token ID | 11 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
zin | +6.457 |
umat | +6.300 |
fuelled | +6.188 |
andals | +5.898 |
piring | +5.722 |
Neg logit contributions | |
confines | -9.316 |
slightest | -8.505 |
bounds | -7.854 |
assumption | -7.705 |
margins | -7.259 |
Token | Al |
Token ID | 978 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
zin | +0.301 |
umat | +0.298 |
andals | +0.286 |
fuelled | +0.283 |
ijing | +0.274 |
Neg logit contributions | |
confines | -0.342 |
slightest | -0.307 |
bounds | -0.283 |
assumption | -0.278 |
margins | -0.258 |
Token | bright |
Token ID | 29199 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
confines | +2.121 |
slightest | +1.928 |
bounds | +1.807 |
assumption | +1.774 |
margins | +1.659 |
Neg logit contributions | |
zin | -1.367 |
umat | -1.357 |
andals | -1.301 |
fuelled | -1.301 |
ijing | -1.223 |
Token | is |
Token ID | 318 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
zin | +3.641 |
umat | +3.578 |
fuelled | +3.463 |
andals | +3.436 |
ijing | +3.281 |
Neg logit contributions | |
confines | -4.164 |
slightest | -3.753 |
bounds | -3.443 |
assumption | -3.380 |
margins | -3.139 |
Token | partially |
Token ID | 12387 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
zin | +4.542 |
umat | +4.500 |
fuelled | +4.388 |
andals | +4.248 |
piring | +4.037 |
Neg logit contributions | |
confines | -6.915 |
slightest | -6.323 |
bounds | -5.835 |
assumption | -5.790 |
margins | -5.417 |
Token | submerged |
Token ID | 37930 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
zin | +2.431 |
umat | +2.396 |
fuelled | +2.326 |
andals | +2.265 |
ijing | +2.147 |
Neg logit contributions | |
confines | -4.008 |
slightest | -3.677 |
bounds | -3.402 |
assumption | -3.365 |
margins | -3.155 |
Token | in |
Token ID | 287 |
Feature activation | +7.62e-02 |
Pos logit contributions | |
zin | +0.227 |
umat | +0.223 |
andals | +0.212 |
fuelled | +0.210 |
ijing | +0.203 |
Neg logit contributions | |
confines | -0.315 |
slightest | -0.283 |
bounds | -0.266 |
assumption | -0.258 |
margins | -0.244 |
Token | flood |
Token ID | 6947 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
zin | +0.307 |
umat | +0.300 |
fuelled | +0.290 |
andals | +0.289 |
ijing | +0.277 |
Neg logit contributions | |
confines | -0.332 |
slightest | -0.297 |
bounds | -0.274 |
assumption | -0.268 |
margins | -0.249 |
Token | waters |
Token ID | 10150 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
zin | +14.104 |
fuelled | +13.983 |
umat | +13.912 |
andals | +13.063 |
ijing | +13.001 |
Neg logit contributions | |
confines | -11.617 |
slightest | -10.523 |
assumption | -9.258 |
bounds | -9.115 |
margins | -8.138 |
Token | of |
Token ID | 286 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
zin | +12.271 |
fuelled | +12.236 |
umat | +11.753 |
andals | +11.545 |
piring | +10.948 |
Neg logit contributions | |
confines | -15.844 |
slightest | -14.907 |
assumption | -13.288 |
bounds | -13.186 |
margins | -12.096 |
Token | an |
Token ID | 281 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
zin | +5.970 |
umat | +5.863 |
fuelled | +5.684 |
andals | +5.628 |
piring | +5.373 |
Neg logit contributions | |
confines | -7.093 |
slightest | -6.453 |
bounds | -5.898 |
assumption | -5.809 |
margins | -5.382 |
Token | over |
Token ID | 625 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
zin | +8.841 |
umat | +8.829 |
fuelled | +8.677 |
andals | +8.496 |
ijing | +8.232 |
Neg logit contributions | |
confines | -7.132 |
slightest | -6.373 |
bounds | -5.709 |
assumption | -5.475 |
margins | -5.037 |
Token | - |
Token ID | 12 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
zin | +4.394 |
umat | +4.306 |
fuelled | +4.163 |
andals | +4.078 |
ijing | +3.912 |
Neg logit contributions | |
confines | -6.561 |
slightest | -5.980 |
bounds | -5.539 |
assumption | -5.442 |
margins | -5.118 |
Token | fl |
Token ID | 2704 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
zin | +4.645 |
umat | +4.524 |
fuelled | +4.370 |
andals | +4.237 |
piring | +4.049 |
Neg logit contributions | |
confines | -8.322 |
slightest | -7.634 |
bounds | -7.106 |
assumption | -7.002 |
margins | -6.619 |
Token | ninety |
Token ID | 37989 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
confines | +0.348 |
slightest | +0.312 |
bounds | +0.286 |
assumption | +0.281 |
margins | +0.258 |
Neg logit contributions | |
zin | -0.350 |
umat | -0.346 |
andals | -0.335 |
fuelled | -0.331 |
ijing | -0.319 |
Token | seconds |
Token ID | 4201 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
zin | +0.217 |
umat | +0.216 |
andals | +0.208 |
fuelled | +0.207 |
ijing | +0.199 |
Neg logit contributions | |
confines | -0.213 |
slightest | -0.191 |
bounds | -0.174 |
assumption | -0.171 |
margins | -0.157 |
Token | . |
Token ID | 13 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
confines | +1.693 |
slightest | +1.521 |
bounds | +1.422 |
assumption | +1.393 |
margins | +1.296 |
Neg logit contributions | |
umat | -1.342 |
zin | -1.342 |
andals | -1.288 |
fuelled | -1.257 |
ijing | -1.215 |
Token | The |
Token ID | 383 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
zin | +1.064 |
umat | +1.047 |
fuelled | +0.989 |
andals | +0.979 |
ijing | +0.927 |
Neg logit contributions | |
confines | -2.241 |
slightest | -2.051 |
bounds | -1.940 |
assumption | -1.909 |
margins | -1.806 |
Token | P |
Token ID | 350 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
confines | +0.986 |
slightest | +0.878 |
bounds | +0.796 |
assumption | +0.785 |
margins | +0.716 |
Neg logit contributions | |
zin | -1.082 |
umat | -1.071 |
andals | -1.030 |
fuelled | -1.029 |
ijing | -0.995 |
Token | chain |
Token ID | 6333 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
confines | +3.782 |
slightest | +3.382 |
bounds | +3.088 |
assumption | +3.073 |
margins | +2.727 |
Neg logit contributions | |
zin | -4.022 |
umat | -3.971 |
andals | -3.876 |
fuelled | -3.823 |
piring | -3.706 |
Token | then |
Token ID | 788 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
confines | +18.051 |
slightest | +15.838 |
assumption | +15.509 |
bounds | +15.203 |
ones | +14.002 |
Neg logit contributions | |
zin | -14.460 |
umat | -14.316 |
andals | -13.841 |
ijing | -13.571 |
piring | -13.262 |
Token | would |
Token ID | 561 |
Feature activation | -9.28e-01 |
Pos logit contributions | |
confines | +7.237 |
slightest | +6.500 |
bounds | +6.175 |
assumption | +5.991 |
margins | +5.549 |
Neg logit contributions | |
zin | -5.018 |
umat | -4.880 |
andals | -4.723 |
ijing | -4.538 |
fuelled | -4.444 |
Token | take |
Token ID | 1011 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
confines | +3.660 |
slightest | +3.243 |
bounds | +2.989 |
assumption | +2.922 |
margins | +2.640 |
Neg logit contributions | |
zin | -4.039 |
umat | -3.978 |
andals | -3.865 |
fuelled | -3.810 |
ijing | -3.740 |
Token | its |
Token ID | 663 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
confines | +2.087 |
slightest | +1.864 |
bounds | +1.713 |
assumption | +1.680 |
margins | +1.536 |
Neg logit contributions | |
zin | -2.157 |
umat | -2.126 |
andals | -2.065 |
fuelled | -2.038 |
ijing | -1.960 |
Token | turn |
Token ID | 1210 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
confines | +1.216 |
slightest | +1.073 |
bounds | +0.967 |
assumption | +0.945 |
margins | +0.852 |
Neg logit contributions | |
zin | -1.545 |
umat | -1.524 |
andals | -1.481 |
fuelled | -1.465 |
ijing | -1.422 |
Token | efforts |
Token ID | 4040 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
zin | +3.462 |
umat | +3.398 |
fuelled | +3.312 |
andals | +3.285 |
piring | +3.168 |
Neg logit contributions | |
confines | -3.096 |
slightest | -2.739 |
bounds | -2.482 |
assumption | -2.431 |
margins | -2.225 |
Token | , |
Token ID | 11 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
zin | +3.353 |
umat | +3.291 |
fuelled | +3.174 |
andals | +3.150 |
ijing | +3.003 |
Neg logit contributions | |
confines | -4.298 |
slightest | -3.892 |
bounds | -3.594 |
assumption | -3.529 |
margins | -3.293 |
Token | effectively |
Token ID | 6840 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
confines | +1.915 |
slightest | +1.719 |
bounds | +1.569 |
assumption | +1.539 |
margins | +1.419 |
Neg logit contributions | |
zin | -1.940 |
umat | -1.916 |
andals | -1.844 |
fuelled | -1.826 |
ijing | -1.773 |
Token | keeping |
Token ID | 5291 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
confines | +1.140 |
slightest | +1.006 |
bounds | +0.918 |
assumption | +0.895 |
margins | +0.827 |
Neg logit contributions | |
zin | -1.277 |
umat | -1.261 |
andals | -1.217 |
fuelled | -1.210 |
ijing | -1.174 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
confines | +5.248 |
slightest | +4.665 |
bounds | +4.332 |
assumption | +4.172 |
margins | +3.917 |
Neg logit contributions | |
zin | -5.616 |
umat | -5.503 |
fuelled | -5.341 |
andals | -5.311 |
ijing | -5.234 |
Token | Senate |
Token ID | 3845 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
confines | +5.863 |
slightest | +5.227 |
bounds | +4.749 |
assumption | +4.632 |
margins | +4.243 |
Neg logit contributions | |
zin | -6.692 |
umat | -6.518 |
fuelled | -6.382 |
andals | -6.308 |
ijing | -6.196 |
Token | from |
Token ID | 422 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
confines | +18.131 |
slightest | +15.906 |
bounds | +15.557 |
assumption | +15.074 |
ones | +14.041 |
Neg logit contributions | |
zin | -14.319 |
umat | -14.209 |
andals | -13.621 |
ijing | -13.485 |
utics | -13.029 |
Token | considering |
Token ID | 6402 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
confines | +8.067 |
slightest | +7.153 |
bounds | +6.612 |
assumption | +6.364 |
margins | +5.907 |
Neg logit contributions | |
zin | -8.607 |
umat | -8.407 |
andals | -8.280 |
ijing | -8.070 |
fuelled | -8.029 |
Token | the |
Token ID | 262 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
confines | +2.188 |
slightest | +1.946 |
bounds | +1.768 |
assumption | +1.713 |
margins | +1.580 |
Neg logit contributions | |
zin | -2.575 |
umat | -2.533 |
fuelled | -2.457 |
andals | -2.449 |
ijing | -2.370 |
Token | legislation |
Token ID | 5520 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
confines | +4.617 |
slightest | +4.164 |
bounds | +3.748 |
assumption | +3.650 |
margins | +3.353 |
Neg logit contributions | |
zin | -5.198 |
umat | -5.085 |
fuelled | -4.964 |
andals | -4.914 |
ijing | -4.783 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
confines | +0.765 |
slightest | +0.696 |
bounds | +0.649 |
assumption | +0.635 |
margins | +0.597 |
Neg logit contributions | |
zin | -0.521 |
umat | -0.510 |
andals | -0.487 |
fuelled | -0.486 |
ijing | -0.464 |
Token | and |
Token ID | 290 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
confines | +2.761 |
slightest | +2.470 |
bounds | +2.296 |
assumption | +2.252 |
margins | +2.085 |
Neg logit contributions | |
zin | -2.406 |
umat | -2.386 |
andals | -2.301 |
fuelled | -2.269 |
piring | -2.185 |
Token | the |
Token ID | 262 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
confines | +1.674 |
slightest | +1.491 |
bounds | +1.388 |
assumption | +1.353 |
margins | +1.255 |
Neg logit contributions | |
zin | -1.477 |
umat | -1.454 |
andals | -1.401 |
fuelled | -1.396 |
ijing | -1.334 |
Token | App |
Token ID | 2034 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
confines | +4.797 |
slightest | +4.249 |
bounds | +3.907 |
assumption | +3.850 |
margins | +3.472 |
Neg logit contributions | |
zin | -5.157 |
umat | -5.105 |
fuelled | -4.951 |
andals | -4.922 |
ijing | -4.724 |
Token | oint |
Token ID | 1563 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
zin | +0.328 |
umat | +0.323 |
fuelled | +0.312 |
andals | +0.309 |
piring | +0.292 |
Neg logit contributions | |
confines | -0.508 |
slightest | -0.458 |
bounds | -0.429 |
assumption | -0.423 |
margins | -0.397 |
Token | ments |
Token ID | 902 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
confines | +5.148 |
slightest | +4.567 |
bounds | +4.315 |
assumption | +4.151 |
margins | +3.852 |
Neg logit contributions | |
umat | -4.688 |
zin | -4.584 |
fuelled | -4.558 |
andals | -4.429 |
ijing | -4.313 |
Token | Clause |
Token ID | 28081 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
confines | +1.059 |
slightest | +0.925 |
bounds | +0.850 |
assumption | +0.833 |
margins | +0.750 |
Neg logit contributions | |
zin | -1.318 |
umat | -1.310 |
andals | -1.257 |
fuelled | -1.257 |
ijing | -1.206 |
Token | , |
Token ID | 11 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
confines | +16.921 |
bounds | +14.227 |
slightest | +13.615 |
assumption | +13.546 |
ones | +12.413 |
Neg logit contributions | |
umat | -14.227 |
zin | -13.971 |
andals | -13.578 |
fuelled | -13.304 |
ilst | -13.037 |
Token | the |
Token ID | 262 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
confines | +8.095 |
slightest | +7.007 |
bounds | +6.663 |
assumption | +6.422 |
ones | +5.827 |
Neg logit contributions | |
zin | -9.313 |
umat | -9.219 |
andals | -9.063 |
fuelled | -8.684 |
ilst | -8.580 |
Token | relevant |
Token ID | 5981 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
confines | +5.710 |
slightest | +5.074 |
bounds | +4.656 |
assumption | +4.609 |
margins | +4.137 |
Neg logit contributions | |
zin | -5.974 |
umat | -5.910 |
andals | -5.738 |
fuelled | -5.685 |
ijing | -5.418 |
Token | portion |
Token ID | 6903 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
confines | +6.828 |
slightest | +5.843 |
bounds | +5.459 |
assumption | +5.408 |
margins | +4.786 |
Neg logit contributions | |
zin | -8.563 |
umat | -8.562 |
andals | -8.260 |
fuelled | -8.158 |
utics | -7.935 |
Token | of |
Token ID | 286 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
confines | +3.517 |
slightest | +3.082 |
bounds | +2.952 |
assumption | +2.864 |
margins | +2.660 |
Neg logit contributions | |
zin | -2.793 |
umat | -2.739 |
andals | -2.620 |
fuelled | -2.598 |
ilst | -2.471 |
Token | House |
Token ID | 2097 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
confines | +1.426 |
slightest | +1.264 |
bounds | +1.138 |
assumption | +1.115 |
margins | +1.019 |
Neg logit contributions | |
zin | -1.728 |
umat | -1.697 |
fuelled | -1.649 |
andals | -1.639 |
ijing | -1.581 |
Token | would |
Token ID | 561 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
confines | +13.104 |
slightest | +11.792 |
assumption | +11.139 |
bounds | +10.937 |
margins | +10.199 |
Neg logit contributions | |
zin | -11.476 |
umat | -11.184 |
andals | -10.867 |
ijing | -10.502 |
utics | -10.477 |
Token | not |
Token ID | 407 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
confines | +4.733 |
slightest | +4.206 |
bounds | +3.865 |
assumption | +3.792 |
margins | +3.488 |
Neg logit contributions | |
zin | -4.904 |
umat | -4.849 |
andals | -4.709 |
fuelled | -4.674 |
ijing | -4.569 |
Token | accept |
Token ID | 2453 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
confines | +7.513 |
slightest | +6.649 |
bounds | +6.072 |
assumption | +5.931 |
margins | +5.421 |
Neg logit contributions | |
zin | -8.803 |
umat | -8.640 |
andals | -8.497 |
fuelled | -8.385 |
ijing | -8.292 |
Token | the |
Token ID | 262 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
confines | +7.439 |
slightest | +6.700 |
bounds | +6.121 |
assumption | +6.005 |
margins | +5.592 |
Neg logit contributions | |
zin | -8.287 |
umat | -8.092 |
andals | -7.779 |
fuelled | -7.744 |
ijing | -7.467 |
Token | Senate |
Token ID | 3845 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
confines | +4.861 |
slightest | +4.326 |
bounds | +3.886 |
assumption | +3.835 |
margins | +3.492 |
Neg logit contributions | |
zin | -5.958 |
umat | -5.833 |
fuelled | -5.673 |
andals | -5.672 |
ijing | -5.434 |
Token | - |
Token ID | 12 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
confines | +15.400 |
slightest | +13.514 |
bounds | +12.946 |
assumption | +12.946 |
margins | +11.442 |
Neg logit contributions | |
zin | -16.373 |
umat | -16.366 |
andals | -15.616 |
utics | -15.228 |
fuelled | -15.177 |
Token | pass |
Token ID | 6603 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
confines | +3.644 |
slightest | +3.333 |
bounds | +3.109 |
assumption | +3.054 |
margins | +2.897 |
Neg logit contributions | |
zin | -2.247 |
umat | -2.209 |
fuelled | -2.092 |
andals | -2.077 |
ijing | -1.966 |
Token | ed |
Token ID | 276 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
confines | +4.082 |
slightest | +3.565 |
bounds | +3.336 |
assumption | +3.283 |
margins | +2.996 |
Neg logit contributions | |
zin | -4.654 |
umat | -4.597 |
fuelled | -4.441 |
andals | -4.430 |
utics | -4.246 |
Token | plan |
Token ID | 1410 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
confines | +2.902 |
slightest | +2.504 |
bounds | +2.256 |
assumption | +2.205 |
margins | +1.976 |
Neg logit contributions | |
zin | -4.310 |
umat | -4.251 |
andals | -4.135 |
fuelled | -4.130 |
ijing | -3.977 |
Token | . |
Token ID | 13 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
confines | +1.696 |
slightest | +1.523 |
bounds | +1.419 |
assumption | +1.395 |
margins | +1.296 |
Neg logit contributions | |
zin | -1.341 |
umat | -1.315 |
fuelled | -1.259 |
andals | -1.259 |
piring | -1.202 |
Token | right |
Token ID | 826 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
confines | +0.281 |
slightest | +0.242 |
bounds | +0.214 |
assumption | +0.208 |
margins | +0.185 |
Neg logit contributions | |
zin | -0.445 |
umat | -0.439 |
fuelled | -0.428 |
andals | -0.426 |
piring | -0.411 |
Token | hand |
Token ID | 1021 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
confines | +8.882 |
slightest | +7.779 |
bounds | +6.592 |
assumption | +6.467 |
margins | +5.832 |
Neg logit contributions | |
zin | -16.394 |
umat | -15.962 |
andals | -15.879 |
utics | -15.569 |
fuelled | -15.474 |
Token | . |
Token ID | 13 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
confines | +8.825 |
slightest | +7.971 |
bounds | +7.492 |
assumption | +7.266 |
margins | +6.828 |
Neg logit contributions | |
zin | -6.626 |
umat | -6.476 |
andals | -6.365 |
fuelled | -6.039 |
ijing | -6.036 |
Token | With |
Token ID | 2080 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
confines | +0.189 |
slightest | +0.172 |
bounds | +0.162 |
assumption | +0.159 |
margins | +0.150 |
Neg logit contributions | |
zin | -0.108 |
umat | -0.105 |
fuelled | -0.101 |
andals | -0.100 |
ijing | -0.095 |
Token | your |
Token ID | 534 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
confines | +2.130 |
slightest | +1.906 |
bounds | +1.729 |
assumption | +1.695 |
margins | +1.558 |
Neg logit contributions | |
zin | -2.350 |
umat | -2.290 |
fuelled | -2.236 |
andals | -2.230 |
ijing | -2.138 |
Token | left |
Token ID | 1364 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
confines | +2.809 |
slightest | +2.475 |
bounds | +2.210 |
assumption | +2.161 |
margins | +1.952 |
Neg logit contributions | |
zin | -3.844 |
umat | -3.771 |
fuelled | -3.667 |
andals | -3.667 |
ijing | -3.546 |
Token | hand |
Token ID | 1021 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
confines | +14.968 |
slightest | +13.359 |
bounds | +12.282 |
assumption | +12.029 |
ones | +11.194 |
Neg logit contributions | |
zin | -16.449 |
andals | -15.994 |
umat | -15.965 |
utics | -15.468 |
ijing | -15.257 |
Token | hold |
Token ID | 1745 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
confines | +9.272 |
slightest | +8.175 |
bounds | +7.714 |
assumption | +7.393 |
margins | +6.968 |
Neg logit contributions | |
zin | -8.941 |
umat | -8.649 |
andals | -8.533 |
ijing | -8.307 |
fuelled | -8.249 |
Token | onto |
Token ID | 4291 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
confines | +6.393 |
slightest | +5.682 |
bounds | +5.271 |
assumption | +5.046 |
margins | +4.695 |
Neg logit contributions | |
zin | -7.182 |
umat | -7.026 |
andals | -6.843 |
fuelled | -6.735 |
ijing | -6.515 |
Token | the |
Token ID | 262 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
confines | +6.583 |
slightest | +5.824 |
bounds | +5.347 |
assumption | +5.168 |
margins | +4.816 |
Neg logit contributions | |
zin | -7.401 |
umat | -7.152 |
andals | -6.962 |
fuelled | -6.921 |
ijing | -6.731 |
Token | top |
Token ID | 1353 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
confines | +5.284 |
slightest | +4.648 |
bounds | +4.125 |
assumption | +3.999 |
margins | +3.626 |
Neg logit contributions | |
zin | -7.419 |
umat | -7.279 |
andals | -7.085 |
fuelled | -7.006 |
ijing | -6.887 |
Token | Increases |
Token ID | 28544 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
confines | +3.505 |
slightest | +3.105 |
bounds | +2.961 |
assumption | +2.882 |
margins | +2.700 |
Neg logit contributions | |
zin | -2.473 |
umat | -2.387 |
fuelled | -2.340 |
andals | -2.323 |
ijing | -2.211 |
Token | growth |
Token ID | 3349 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
confines | +0.689 |
slightest | +0.613 |
bounds | +0.562 |
assumption | +0.548 |
margins | +0.506 |
Neg logit contributions | |
zin | -0.738 |
umat | -0.724 |
andals | -0.705 |
fuelled | -0.695 |
ijing | -0.675 |
Token | hormone |
Token ID | 17770 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
zin | +6.094 |
umat | +5.965 |
fuelled | +5.841 |
andals | +5.720 |
piring | +5.499 |
Neg logit contributions | |
confines | -6.434 |
slightest | -5.804 |
bounds | -5.272 |
assumption | -5.145 |
margins | -4.753 |
Token | ( |
Token ID | 357 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
confines | +1.704 |
slightest | +1.526 |
bounds | +1.424 |
assumption | +1.405 |
margins | +1.290 |
Neg logit contributions | |
umat | -1.557 |
zin | -1.555 |
andals | -1.518 |
fuelled | -1.459 |
piring | -1.432 |
Token | the |
Token ID | 1169 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
confines | +1.260 |
slightest | +1.144 |
bounds | +1.069 |
assumption | +1.045 |
margins | +0.968 |
Neg logit contributions | |
zin | -0.894 |
umat | -0.887 |
andals | -0.857 |
fuelled | -0.842 |
piring | -0.808 |
Token | opposite |
Token ID | 6697 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
confines | +5.365 |
slightest | +4.723 |
assumption | +4.144 |
bounds | +4.054 |
ones | +3.421 |
Neg logit contributions | |
zin | -8.967 |
umat | -8.934 |
andals | -8.855 |
fuelled | -8.661 |
piring | -8.460 |
Token | of |
Token ID | 286 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
confines | +14.486 |
assumption | +12.332 |
bounds | +12.211 |
slightest | +11.812 |
ones | +10.839 |
Neg logit contributions | |
umat | -16.667 |
andals | -16.630 |
zin | -16.362 |
ijing | -16.158 |
ilst | -15.879 |
Token | insulin |
Token ID | 14588 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
confines | +5.181 |
slightest | +4.584 |
bounds | +4.231 |
assumption | +4.166 |
margins | +3.772 |
Neg logit contributions | |
zin | -5.577 |
umat | -5.447 |
andals | -5.446 |
fuelled | -5.177 |
ijing | -5.156 |
Token | ). |
Token ID | 737 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
confines | +1.478 |
slightest | +1.337 |
bounds | +1.237 |
assumption | +1.224 |
margins | +1.136 |
Neg logit contributions | |
zin | -1.159 |
umat | -1.152 |
andals | -1.120 |
fuelled | -1.095 |
ijing | -1.063 |
Token |
|
Token ID | 198 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
zin | +0.270 |
umat | +0.262 |
fuelled | +0.251 |
andals | +0.249 |
ijing | +0.235 |
Neg logit contributions | |
confines | -0.532 |
slightest | -0.487 |
bounds | -0.458 |
assumption | -0.451 |
margins | -0.426 |
Token |
|
Token ID | 198 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
confines | +3.585 |
slightest | +3.205 |
bounds | +3.109 |
assumption | +3.021 |
margins | +2.850 |
Neg logit contributions | |
zin | -2.014 |
umat | -2.003 |
fuelled | -1.914 |
andals | -1.910 |
ijing | -1.834 |
Token | � |
Token ID | 447 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
confines | +4.411 |
slightest | +3.761 |
bounds | +3.645 |
assumption | +3.465 |
margins | +3.194 |
Neg logit contributions | |
umat | -5.034 |
zin | -5.028 |
fuelled | -4.872 |
andals | -4.862 |
ijing | -4.709 |
Token | � |
Token ID | 247 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
confines | +6.641 |
slightest | +5.999 |
bounds | +5.624 |
assumption | +5.534 |
margins | +5.171 |
Neg logit contributions | |
zin | -4.701 |
umat | -4.476 |
andals | -4.464 |
fuelled | -4.410 |
ilst | -4.107 |
Token | t |
Token ID | 83 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
zin | +1.087 |
umat | +1.063 |
fuelled | +1.017 |
andals | +1.002 |
piring | +0.945 |
Neg logit contributions | |
confines | -1.913 |
slightest | -1.754 |
bounds | -1.632 |
assumption | -1.608 |
margins | -1.516 |
Token | affect |
Token ID | 2689 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
confines | +7.097 |
slightest | +6.425 |
bounds | +5.895 |
assumption | +5.683 |
margins | +5.272 |
Neg logit contributions | |
zin | -7.674 |
umat | -7.469 |
andals | -7.342 |
ijing | -7.136 |
fuelled | -7.122 |
Token | the |
Token ID | 262 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
confines | +7.174 |
slightest | +6.520 |
bounds | +5.908 |
assumption | +5.676 |
margins | +5.370 |
Neg logit contributions | |
zin | -8.486 |
umat | -8.415 |
andals | -8.085 |
fuelled | -7.883 |
ijing | -7.697 |
Token | console |
Token ID | 8624 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
confines | +8.383 |
slightest | +7.664 |
bounds | +6.816 |
assumption | +6.630 |
margins | +6.177 |
Neg logit contributions | |
zin | -9.039 |
umat | -9.000 |
andals | -8.621 |
fuelled | -8.382 |
ijing | -8.234 |
Token | release |
Token ID | 2650 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
confines | +16.461 |
slightest | +15.018 |
bounds | +13.877 |
assumption | +13.208 |
margins | +12.472 |
Neg logit contributions | |
umat | -15.158 |
zin | -14.691 |
andals | -14.173 |
utics | -13.897 |
piring | -13.545 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
confines | +8.215 |
slightest | +7.521 |
bounds | +6.934 |
assumption | +6.858 |
margins | +6.458 |
Neg logit contributions | |
zin | -6.815 |
umat | -6.771 |
andals | -6.424 |
fuelled | -6.207 |
piring | -6.122 |
Token | but |
Token ID | 475 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
confines | +4.979 |
slightest | +4.412 |
bounds | +4.038 |
assumption | +3.931 |
margins | +3.641 |
Neg logit contributions | |
zin | -5.771 |
umat | -5.685 |
andals | -5.480 |
fuelled | -5.325 |
ijing | -5.275 |
Token | the |
Token ID | 262 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
confines | +3.417 |
slightest | +3.066 |
bounds | +2.824 |
assumption | +2.768 |
margins | +2.581 |
Neg logit contributions | |
zin | -3.114 |
umat | -3.081 |
andals | -2.943 |
fuelled | -2.900 |
ijing | -2.838 |
Token | transition |
Token ID | 6801 |
Feature activation | -8.94e-01 |
Pos logit contributions | |
confines | +3.758 |
slightest | +3.366 |
bounds | +3.027 |
assumption | +2.981 |
margins | +2.742 |
Neg logit contributions | |
zin | -4.085 |
umat | -4.040 |
andals | -3.887 |
fuelled | -3.860 |
ijing | -3.745 |
Token | has |
Token ID | 468 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
zin | +6.276 |
umat | +6.143 |
fuelled | +6.019 |
andals | +5.894 |
piring | +5.587 |
Neg logit contributions | |
confines | -7.805 |
slightest | -7.113 |
bounds | -6.499 |
assumption | -6.355 |
margins | -5.935 |
Token | no |
Token ID | 645 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
zin | +1.430 |
umat | +1.407 |
fuelled | +1.348 |
andals | +1.348 |
ijing | +1.285 |
Neg logit contributions | |
confines | -1.793 |
slightest | -1.618 |
bounds | -1.499 |
assumption | -1.472 |
margins | -1.376 |
Token | single |
Token ID | 2060 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
confines | +7.820 |
slightest | +7.008 |
bounds | +6.638 |
assumption | +6.422 |
margins | +5.950 |
Neg logit contributions | |
zin | -7.489 |
umat | -7.410 |
andals | -7.250 |
fuelled | -7.076 |
piring | -6.904 |
Token | � |
Token ID | 564 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
confines | +8.010 |
slightest | +7.160 |
bounds | +6.620 |
assumption | +6.602 |
margins | +5.852 |
Neg logit contributions | |
umat | -8.113 |
zin | -8.081 |
andals | -7.942 |
fuelled | -7.770 |
piring | -7.637 |
Token | � |
Token ID | 250 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
zin | +0.617 |
umat | +0.597 |
fuelled | +0.575 |
andals | +0.570 |
ijing | +0.535 |
Neg logit contributions | |
confines | -1.066 |
slightest | -0.978 |
bounds | -0.910 |
assumption | -0.898 |
margins | -0.844 |
Token | Silver |
Token ID | 26766 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
confines | +2.860 |
slightest | +2.627 |
bounds | +2.429 |
assumption | +2.414 |
margins | +2.248 |
Neg logit contributions | |
zin | -1.805 |
umat | -1.771 |
fuelled | -1.730 |
andals | -1.707 |
piring | -1.626 |
Token | Bullet |
Token ID | 18003 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
confines | +12.814 |
slightest | +11.524 |
bounds | +10.504 |
assumption | +10.394 |
ones | +8.955 |
Neg logit contributions | |
umat | -17.162 |
andals | -16.927 |
zin | -16.696 |
fuelled | -16.186 |
piring | -16.028 |
Token | � |
Token ID | 447 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
confines | +6.545 |
slightest | +5.850 |
assumption | +5.531 |
bounds | +5.484 |
margins | +5.022 |
Neg logit contributions | |
zin | -5.080 |
umat | -5.036 |
andals | -4.837 |
fuelled | -4.761 |
piring | -4.672 |
Token | � |
Token ID | 251 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
confines | +9.617 |
slightest | +8.625 |
bounds | +8.079 |
assumption | +8.068 |
margins | +7.489 |
Neg logit contributions | |
zin | -6.158 |
umat | -5.932 |
fuelled | -5.703 |
andals | -5.644 |
ilst | -5.437 |
Token | but |
Token ID | 475 |
Feature activation | -9.49e-01 |
Pos logit contributions | |
confines | +3.649 |
slightest | +3.316 |
bounds | +3.072 |
assumption | +3.068 |
margins | +2.826 |
Neg logit contributions | |
zin | -2.603 |
umat | -2.569 |
andals | -2.455 |
fuelled | -2.451 |
piring | -2.334 |
Token | rather |
Token ID | 2138 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
confines | +4.029 |
slightest | +3.509 |
bounds | +3.297 |
assumption | +3.214 |
margins | +2.954 |
Neg logit contributions | |
zin | -3.863 |
umat | -3.838 |
andals | -3.687 |
fuelled | -3.669 |
piring | -3.557 |
Token | () |
Token ID | 3419 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
confines | +3.182 |
slightest | +2.776 |
bounds | +2.596 |
assumption | +2.498 |
margins | +2.262 |
Neg logit contributions | |
zin | -3.883 |
umat | -3.860 |
andals | -3.752 |
fuelled | -3.698 |
piring | -3.589 |
Token | , |
Token ID | 837 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
confines | +3.191 |
slightest | +2.887 |
bounds | +2.740 |
assumption | +2.668 |
margins | +2.501 |
Neg logit contributions | |
zin | -2.075 |
umat | -2.052 |
andals | -1.980 |
fuelled | -1.930 |
piring | -1.868 |
Token | not |
Token ID | 407 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
confines | +2.111 |
slightest | +1.878 |
bounds | +1.735 |
assumption | +1.687 |
margins | +1.551 |
Neg logit contributions | |
zin | -2.191 |
umat | -2.159 |
andals | -2.094 |
fuelled | -2.065 |
ijing | -2.004 |
Token | to |
Token ID | 284 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
confines | +8.242 |
slightest | +7.324 |
bounds | +6.867 |
assumption | +6.605 |
margins | +6.132 |
Neg logit contributions | |
zin | -7.994 |
umat | -7.921 |
andals | -7.754 |
fuelled | -7.512 |
ijing | -7.335 |
Token | the |
Token ID | 262 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
confines | +5.591 |
slightest | +4.979 |
bounds | +4.612 |
assumption | +4.425 |
margins | +4.069 |
Neg logit contributions | |
zin | -6.077 |
umat | -6.001 |
andals | -5.826 |
fuelled | -5.762 |
piring | -5.585 |
Token | bottom |
Token ID | 4220 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
confines | +7.615 |
slightest | +6.698 |
bounds | +6.115 |
assumption | +5.858 |
margins | +5.355 |
Neg logit contributions | |
zin | -9.899 |
umat | -9.803 |
andals | -9.680 |
fuelled | -9.419 |
ijing | -9.262 |
Token | of |
Token ID | 286 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
confines | +16.792 |
slightest | +14.860 |
bounds | +14.286 |
assumption | +13.712 |
margins | +12.918 |
Neg logit contributions | |
umat | -13.811 |
andals | -13.588 |
zin | -13.458 |
fuelled | -12.818 |
ijing | -12.720 |
Token | system |
Token ID | 1080 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
confines | +6.994 |
slightest | +6.184 |
bounds | +5.852 |
assumption | +5.577 |
margins | +5.106 |
Neg logit contributions | |
zin | -7.717 |
umat | -7.565 |
andals | -7.457 |
fuelled | -7.257 |
ijing | -7.104 |
Token | () |
Token ID | 3419 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
confines | +6.483 |
slightest | +5.634 |
bounds | +5.222 |
assumption | +4.979 |
margins | +4.527 |
Neg logit contributions | |
zin | -8.871 |
umat | -8.784 |
andals | -8.601 |
fuelled | -8.484 |
piring | -8.243 |
Token | ' |
Token ID | 705 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
confines | +7.551 |
slightest | +6.798 |
bounds | +6.486 |
assumption | +6.228 |
margins | +5.916 |
Neg logit contributions | |
zin | -4.996 |
umat | -4.941 |
andals | -4.876 |
fuelled | -4.654 |
piring | -4.556 |
Token | s |
Token ID | 82 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
confines | +3.256 |
slightest | +2.897 |
bounds | +2.721 |
assumption | +2.645 |
margins | +2.458 |
Neg logit contributions | |
zin | -2.973 |
umat | -2.955 |
andals | -2.840 |
fuelled | -2.834 |
ijing | -2.737 |
Token | President |
Token ID | 1992 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
zin | +0.603 |
umat | +0.595 |
fuelled | +0.575 |
andals | +0.574 |
ijing | +0.549 |
Neg logit contributions | |
confines | -0.565 |
slightest | -0.502 |
bounds | -0.459 |
assumption | -0.449 |
margins | -0.414 |
Token | Barack |
Token ID | 8732 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
zin | +3.581 |
umat | +3.525 |
fuelled | +3.415 |
andals | +3.397 |
ijing | +3.251 |
Neg logit contributions | |
confines | -3.638 |
slightest | -3.249 |
bounds | -2.974 |
assumption | -2.910 |
margins | -2.689 |
Token | Obama |
Token ID | 2486 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
zin | +4.835 |
umat | +4.711 |
fuelled | +4.599 |
andals | +4.513 |
piring | +4.387 |
Neg logit contributions | |
confines | -4.686 |
slightest | -4.296 |
bounds | -3.766 |
assumption | -3.705 |
margins | -3.449 |
Token | . |
Token ID | 13 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
zin | +0.397 |
umat | +0.390 |
andals | +0.373 |
fuelled | +0.369 |
piring | +0.352 |
Neg logit contributions | |
confines | -0.584 |
slightest | -0.529 |
bounds | -0.496 |
assumption | -0.486 |
margins | -0.457 |
Token | The |
Token ID | 383 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
zin | +2.847 |
umat | +2.786 |
fuelled | +2.672 |
andals | +2.646 |
ijing | +2.502 |
Neg logit contributions | |
confines | -4.667 |
slightest | -4.264 |
bounds | -3.975 |
assumption | -3.910 |
margins | -3.680 |
Token | Democrat |
Token ID | 9755 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
confines | +0.832 |
slightest | +0.737 |
bounds | +0.666 |
assumption | +0.653 |
margins | +0.596 |
Neg logit contributions | |
zin | -0.980 |
umat | -0.962 |
fuelled | -0.934 |
andals | -0.929 |
ijing | -0.893 |
Token | - |
Token ID | 12 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
confines | +17.131 |
slightest | +14.925 |
bounds | +14.312 |
assumption | +14.263 |
margins | +12.987 |
Neg logit contributions | |
zin | -13.260 |
umat | -13.002 |
andals | -12.486 |
utics | -12.460 |
utic | -11.796 |
Token | controlled |
Token ID | 14401 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
confines | +3.290 |
slightest | +2.982 |
bounds | +2.765 |
assumption | +2.700 |
margins | +2.553 |
Neg logit contributions | |
zin | -2.418 |
umat | -2.380 |
andals | -2.301 |
fuelled | -2.244 |
piring | -2.170 |
Token | Senate |
Token ID | 3845 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
confines | +12.044 |
slightest | +9.921 |
bounds | +9.337 |
assumption | +8.775 |
margins | +8.410 |
Neg logit contributions | |
zin | -17.724 |
umat | -17.390 |
andals | -16.808 |
piring | -16.329 |
fuelled | -16.276 |
Token | has |
Token ID | 468 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
confines | +16.172 |
slightest | +14.455 |
bounds | +13.691 |
assumption | +13.395 |
margins | +12.440 |
Neg logit contributions | |
zin | -11.730 |
umat | -11.319 |
andals | -11.177 |
piring | -10.757 |
ijing | -10.744 |
Token | repeatedly |
Token ID | 7830 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
confines | +2.894 |
slightest | +2.622 |
bounds | +2.432 |
assumption | +2.377 |
margins | +2.232 |
Neg logit contributions | |
zin | -2.231 |
umat | -2.194 |
andals | -2.106 |
fuelled | -2.035 |
ijing | -2.010 |
Token | concern |
Token ID | 2328 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
zin | +0.431 |
umat | +0.424 |
fuelled | +0.408 |
andals | +0.404 |
ijing | +0.394 |
Neg logit contributions | |
confines | -0.445 |
slightest | -0.401 |
bounds | -0.368 |
assumption | -0.360 |
margins | -0.335 |
Token | for |
Token ID | 329 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
confines | +1.240 |
slightest | +1.109 |
bounds | +1.048 |
assumption | +1.014 |
margins | +0.956 |
Neg logit contributions | |
zin | -1.031 |
umat | -1.015 |
fuelled | -0.969 |
andals | -0.969 |
ijing | -0.938 |
Token | Hispanics |
Token ID | 30714 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
confines | +0.815 |
slightest | +0.731 |
bounds | +0.670 |
assumption | +0.657 |
margins | +0.613 |
Neg logit contributions | |
zin | -0.787 |
umat | -0.775 |
fuelled | -0.748 |
andals | -0.747 |
ijing | -0.718 |
Token | than |
Token ID | 621 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
confines | +1.393 |
slightest | +1.233 |
bounds | +1.137 |
assumption | +1.104 |
margins | +1.039 |
Neg logit contributions | |
zin | -1.428 |
umat | -1.403 |
andals | -1.356 |
fuelled | -1.350 |
ijing | -1.309 |
Token | the |
Token ID | 262 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
confines | +4.168 |
slightest | +3.747 |
bounds | +3.427 |
assumption | +3.351 |
margins | +3.128 |
Neg logit contributions | |
zin | -3.990 |
umat | -3.879 |
fuelled | -3.729 |
andals | -3.719 |
ijing | -3.612 |
Token | Democratic |
Token ID | 4390 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
confines | +5.302 |
slightest | +4.670 |
bounds | +4.191 |
assumption | +4.112 |
margins | +3.784 |
Neg logit contributions | |
zin | -6.552 |
umat | -6.440 |
fuelled | -6.292 |
andals | -6.163 |
ijing | -6.006 |
Token | Party |
Token ID | 3615 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
confines | +12.962 |
slightest | +11.259 |
assumption | +10.198 |
bounds | +9.894 |
ones | +9.793 |
Neg logit contributions | |
umat | -17.328 |
zin | -17.238 |
andals | -16.782 |
utics | -16.725 |
fuelled | -16.720 |
Token | . |
Token ID | 13 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
confines | +12.948 |
slightest | +11.642 |
bounds | +10.738 |
assumption | +10.717 |
margins | +10.355 |
Neg logit contributions | |
zin | -9.591 |
umat | -9.560 |
andals | -9.091 |
fuelled | -8.968 |
piring | -8.831 |
Token |
|
Token ID | 198 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
confines | +1.679 |
slightest | +1.509 |
bounds | +1.439 |
assumption | +1.417 |
margins | +1.354 |
Neg logit contributions | |
zin | -0.897 |
umat | -0.891 |
fuelled | -0.847 |
andals | -0.830 |
ijing | -0.801 |
Token |
|
Token ID | 198 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
confines | +4.708 |
slightest | +4.197 |
bounds | +4.103 |
assumption | +3.969 |
margins | +3.743 |
Neg logit contributions | |
zin | -2.724 |
umat | -2.722 |
andals | -2.593 |
fuelled | -2.591 |
ijing | -2.496 |
Token | Lat |
Token ID | 24220 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
confines | +1.996 |
slightest | +1.804 |
bounds | +1.710 |
assumption | +1.689 |
margins | +1.597 |
Neg logit contributions | |
zin | -1.186 |
umat | -1.173 |
fuelled | -1.138 |
andals | -1.116 |
ijing | -1.060 |
Token | an |
Token ID | 281 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
confines | +4.787 |
slightest | +4.315 |
bounds | +3.984 |
assumption | +3.922 |
margins | +3.664 |
Neg logit contributions | |
zin | -4.286 |
umat | -4.237 |
andals | -4.012 |
fuelled | -3.972 |
ijing | -3.862 |
Token | uphill |
Token ID | 35256 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
zin | +2.218 |
umat | +2.181 |
fuelled | +2.113 |
andals | +2.094 |
ijing | +2.002 |
Neg logit contributions | |
confines | -2.479 |
slightest | -2.228 |
bounds | -2.048 |
assumption | -2.001 |
margins | -1.858 |
Token | battle |
Token ID | 3344 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
confines | +4.535 |
slightest | +4.050 |
assumption | +3.788 |
bounds | +3.706 |
margins | +3.373 |
Neg logit contributions | |
zin | -4.905 |
umat | -4.857 |
andals | -4.739 |
fuelled | -4.640 |
ijing | -4.519 |
Token | in |
Token ID | 287 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
confines | +1.740 |
slightest | +1.544 |
bounds | +1.460 |
assumption | +1.434 |
margins | +1.346 |
Neg logit contributions | |
umat | -1.283 |
zin | -1.280 |
andals | -1.221 |
fuelled | -1.201 |
ijing | -1.167 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
confines | +5.375 |
slightest | +4.707 |
bounds | +4.358 |
assumption | +4.248 |
margins | +3.966 |
Neg logit contributions | |
zin | -5.964 |
umat | -5.888 |
andals | -5.686 |
fuelled | -5.677 |
ijing | -5.415 |
Token | Senate |
Token ID | 3845 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
confines | +4.350 |
slightest | +3.734 |
bounds | +3.298 |
assumption | +3.183 |
margins | +2.938 |
Neg logit contributions | |
zin | -6.597 |
umat | -6.487 |
fuelled | -6.358 |
andals | -6.353 |
ijing | -6.134 |
Token | . |
Token ID | 13 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
confines | +16.289 |
bounds | +13.547 |
slightest | +13.542 |
assumption | +13.141 |
margins | +12.275 |
Neg logit contributions | |
umat | -13.252 |
zin | -12.867 |
andals | -12.595 |
ijing | -12.387 |
piring | -12.311 |
Token | An |
Token ID | 1052 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
confines | +2.729 |
slightest | +2.415 |
bounds | +2.330 |
assumption | +2.287 |
margins | +2.175 |
Neg logit contributions | |
zin | -1.578 |
umat | -1.574 |
fuelled | -1.508 |
andals | -1.463 |
ijing | -1.408 |
Token | amendment |
Token ID | 11326 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
zin | +5.904 |
umat | +5.817 |
fuelled | +5.647 |
andals | +5.595 |
ijing | +5.424 |
Neg logit contributions | |
confines | -4.607 |
slightest | -4.027 |
bounds | -3.660 |
assumption | -3.448 |
margins | -3.180 |
Token | by |
Token ID | 416 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
confines | +6.305 |
slightest | +5.572 |
bounds | +5.248 |
assumption | +5.153 |
margins | +4.795 |
Neg logit contributions | |
zin | -5.072 |
umat | -4.990 |
andals | -4.750 |
fuelled | -4.631 |
ijing | -4.532 |
Token | Dodd |
Token ID | 32145 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
confines | +2.614 |
slightest | +2.334 |
bounds | +2.131 |
assumption | +2.096 |
margins | +1.940 |
Neg logit contributions | |
zin | -2.639 |
umat | -2.624 |
fuelled | -2.537 |
andals | -2.525 |
ijing | -2.405 |
Token | binary |
Token ID | 13934 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
confines | +5.679 |
slightest | +5.135 |
bounds | +4.793 |
assumption | +4.740 |
margins | +4.285 |
Neg logit contributions | |
umat | -5.044 |
zin | -4.931 |
fuelled | -4.756 |
andals | -4.712 |
piring | -4.616 |
Token | in |
Token ID | 287 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
confines | +4.444 |
slightest | +4.001 |
bounds | +3.735 |
assumption | +3.689 |
margins | +3.358 |
Neg logit contributions | |
umat | -3.715 |
zin | -3.689 |
andals | -3.509 |
fuelled | -3.465 |
piring | -3.364 |
Token | partition |
Token ID | 18398 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
confines | +5.974 |
slightest | +5.305 |
bounds | +4.974 |
assumption | +4.836 |
margins | +4.419 |
Neg logit contributions | |
umat | -6.144 |
zin | -6.131 |
andals | -5.874 |
fuelled | -5.860 |
piring | -5.662 |
Token | # |
Token ID | 1303 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
confines | +6.812 |
slightest | +6.075 |
bounds | +5.715 |
assumption | +5.567 |
margins | +5.203 |
Neg logit contributions | |
zin | -5.903 |
umat | -5.816 |
andals | -5.559 |
fuelled | -5.508 |
piring | -5.366 |
Token | 25 |
Token ID | 1495 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
confines | +3.074 |
slightest | +2.818 |
bounds | +2.657 |
assumption | +2.603 |
margins | +2.456 |
Neg logit contributions | |
zin | -1.738 |
umat | -1.702 |
andals | -1.630 |
fuelled | -1.627 |
ijing | -1.534 |
Token | instead |
Token ID | 2427 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
confines | +3.452 |
slightest | +3.096 |
bounds | +2.918 |
assumption | +2.872 |
margins | +2.656 |
Neg logit contributions | |
umat | -2.598 |
zin | -2.586 |
andals | -2.479 |
fuelled | -2.461 |
ijing | -2.355 |
Token | of |
Token ID | 286 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
confines | +15.327 |
slightest | +13.362 |
assumption | +13.026 |
bounds | +12.747 |
ones | +11.721 |
Neg logit contributions | |
zin | -14.050 |
umat | -13.967 |
andals | -13.690 |
fuelled | -13.344 |
ijing | -13.188 |
Token | # |
Token ID | 1303 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
confines | +13.960 |
slightest | +12.593 |
bounds | +11.893 |
assumption | +11.751 |
ones | +10.716 |
Neg logit contributions | |
zin | -13.150 |
umat | -12.892 |
andals | -12.610 |
utics | -12.179 |
ijing | -12.124 |
Token | 19 |
Token ID | 1129 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
confines | +8.286 |
slightest | +7.615 |
bounds | +7.204 |
assumption | +7.068 |
margins | +6.673 |
Neg logit contributions | |
zin | -4.883 |
umat | -4.734 |
andals | -4.594 |
fuelled | -4.588 |
ijing | -4.379 |
Token | . |
Token ID | 13 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
confines | +4.268 |
slightest | +3.838 |
bounds | +3.609 |
assumption | +3.571 |
margins | +3.326 |
Neg logit contributions | |
zin | -2.898 |
umat | -2.884 |
andals | -2.778 |
fuelled | -2.772 |
ijing | -2.651 |
Token | As |
Token ID | 1081 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
confines | +2.705 |
slightest | +2.480 |
bounds | +2.343 |
assumption | +2.311 |
margins | +2.170 |
Neg logit contributions | |
zin | -1.348 |
umat | -1.320 |
fuelled | -1.256 |
andals | -1.237 |
ijing | -1.183 |
Token | glacier |
Token ID | 44539 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
confines | +3.783 |
slightest | +3.303 |
bounds | +2.941 |
assumption | +2.853 |
margins | +2.619 |
Neg logit contributions | |
zin | -5.229 |
umat | -5.183 |
andals | -5.001 |
fuelled | -4.952 |
ijing | -4.845 |
Token | , |
Token ID | 11 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
confines | +5.444 |
slightest | +4.843 |
bounds | +4.536 |
assumption | +4.378 |
margins | +4.199 |
Neg logit contributions | |
zin | -4.830 |
umat | -4.787 |
andals | -4.634 |
fuelled | -4.497 |
ijing | -4.412 |
Token | even |
Token ID | 772 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
confines | +0.318 |
slightest | +0.284 |
bounds | +0.260 |
assumption | +0.254 |
margins | +0.236 |
Neg logit contributions | |
zin | -0.310 |
umat | -0.304 |
andals | -0.294 |
fuelled | -0.292 |
ijing | -0.282 |
Token | as |
Token ID | 355 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
zin | +0.263 |
umat | +0.259 |
andals | +0.249 |
fuelled | +0.249 |
ijing | +0.239 |
Neg logit contributions | |
confines | -0.271 |
slightest | -0.243 |
bounds | -0.222 |
assumption | -0.217 |
margins | -0.201 |
Token | its |
Token ID | 663 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
zin | +0.355 |
umat | +0.351 |
fuelled | +0.338 |
andals | +0.337 |
ijing | +0.324 |
Neg logit contributions | |
confines | -0.346 |
slightest | -0.309 |
bounds | -0.282 |
assumption | -0.275 |
margins | -0.254 |
Token | bottom |
Token ID | 4220 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
confines | +0.543 |
slightest | +0.470 |
bounds | +0.417 |
assumption | +0.405 |
margins | +0.365 |
Neg logit contributions | |
zin | -0.821 |
umat | -0.811 |
fuelled | -0.787 |
andals | -0.786 |
ijing | -0.760 |
Token | ( |
Token ID | 357 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
confines | +16.716 |
slightest | +14.380 |
bounds | +14.180 |
assumption | +13.040 |
halves | +12.915 |
Neg logit contributions | |
umat | -13.949 |
zin | -13.544 |
andals | -13.495 |
fuelled | -12.669 |
ijing | -12.658 |
Token | below |
Token ID | 35993 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
confines | +7.764 |
slightest | +7.003 |
bounds | +6.405 |
assumption | +6.061 |
margins | +5.847 |
Neg logit contributions | |
umat | -7.783 |
zin | -7.706 |
fuelled | -7.566 |
andals | -7.482 |
ijing | -7.149 |
Token | water |
Token ID | 1660 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
confines | +11.581 |
slightest | +10.190 |
bounds | +9.663 |
assumption | +9.050 |
margins | +8.390 |
Neg logit contributions | |
umat | -13.115 |
zin | -13.038 |
andals | -12.657 |
fuelled | -12.329 |
utics | -12.171 |
Token | ) |
Token ID | 8 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
confines | +6.801 |
slightest | +6.052 |
bounds | +5.704 |
assumption | +5.481 |
margins | +5.245 |
Neg logit contributions | |
umat | -5.949 |
zin | -5.885 |
andals | -5.690 |
fuelled | -5.621 |
ijing | -5.419 |
Token | rises |
Token ID | 16736 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
confines | +7.347 |
slightest | +6.424 |
bounds | +6.165 |
assumption | +5.765 |
margins | +5.530 |
Neg logit contributions | |
zin | -6.198 |
umat | -6.129 |
andals | -5.900 |
fuelled | -5.735 |
ijing | -5.704 |
Token | but |
Token ID | 475 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
confines | +6.186 |
slightest | +5.466 |
bounds | +5.028 |
assumption | +5.002 |
margins | +4.507 |
Neg logit contributions | |
zin | -6.989 |
umat | -6.869 |
andals | -6.655 |
fuelled | -6.480 |
ijing | -6.393 |
Token | my |
Token ID | 616 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
confines | +1.331 |
slightest | +1.193 |
bounds | +1.096 |
assumption | +1.080 |
margins | +0.994 |
Neg logit contributions | |
zin | -1.228 |
umat | -1.212 |
fuelled | -1.164 |
andals | -1.163 |
ijing | -1.113 |
Token | advice |
Token ID | 5608 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
confines | +0.886 |
slightest | +0.776 |
bounds | +0.695 |
assumption | +0.688 |
margins | +0.614 |
Neg logit contributions | |
zin | -1.192 |
umat | -1.178 |
fuelled | -1.143 |
andals | -1.140 |
piring | -1.099 |
Token | is |
Token ID | 318 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
confines | +4.650 |
slightest | +4.075 |
bounds | +3.893 |
assumption | +3.804 |
margins | +3.523 |
Neg logit contributions | |
zin | -3.937 |
umat | -3.886 |
andals | -3.797 |
fuelled | -3.718 |
ijing | -3.579 |
Token | to |
Token ID | 284 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
confines | +5.581 |
slightest | +4.948 |
bounds | +4.610 |
assumption | +4.609 |
margins | +4.160 |
Neg logit contributions | |
umat | -5.315 |
zin | -5.299 |
andals | -5.154 |
fuelled | -5.006 |
piring | -4.941 |
Token | skip |
Token ID | 14267 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
confines | +1.867 |
slightest | +1.653 |
bounds | +1.513 |
assumption | +1.483 |
margins | +1.356 |
Neg logit contributions | |
zin | -2.024 |
umat | -1.997 |
andals | -1.928 |
fuelled | -1.928 |
ijing | -1.851 |
Token | it |
Token ID | 340 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
confines | +14.841 |
slightest | +13.485 |
assumption | +12.753 |
bounds | +12.633 |
margins | +11.573 |
Neg logit contributions | |
zin | -14.431 |
umat | -14.209 |
andals | -13.873 |
fuelled | -13.292 |
piring | -13.211 |
Token | only |
Token ID | 691 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
confines | +6.991 |
slightest | +6.370 |
bounds | +5.977 |
assumption | +5.858 |
margins | +5.420 |
Neg logit contributions | |
zin | -5.448 |
umat | -5.377 |
andals | -5.194 |
fuelled | -5.115 |
piring | -4.953 |
Token | if |
Token ID | 611 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
confines | +10.806 |
slightest | +9.729 |
bounds | +8.957 |
assumption | +8.908 |
margins | +8.112 |
Neg logit contributions | |
zin | -10.471 |
umat | -10.265 |
andals | -9.869 |
fuelled | -9.551 |
utics | -9.485 |
Token | you |
Token ID | 345 |
Feature activation | +6.42e-02 |
Pos logit contributions | |
confines | +4.233 |
slightest | +3.766 |
bounds | +3.447 |
assumption | +3.398 |
margins | +3.089 |
Neg logit contributions | |
zin | -4.638 |
umat | -4.613 |
andals | -4.451 |
fuelled | -4.383 |
piring | -4.245 |
Token | are |
Token ID | 389 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
zin | +0.236 |
umat | +0.232 |
fuelled | +0.223 |
andals | +0.222 |
ijing | +0.212 |
Neg logit contributions | |
confines | -0.300 |
slightest | -0.272 |
bounds | -0.251 |
assumption | -0.247 |
margins | -0.230 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
confines | +3.865 |
slightest | +3.405 |
bounds | +3.187 |
assumption | +3.130 |
margins | +2.890 |
Neg logit contributions | |
zin | -3.737 |
umat | -3.633 |
fuelled | -3.517 |
andals | -3.510 |
utics | -3.337 |
Token | new |
Token ID | 649 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
confines | +7.287 |
slightest | +6.610 |
bounds | +6.130 |
assumption | +6.105 |
margins | +5.654 |
Neg logit contributions | |
zin | -5.057 |
umat | -4.811 |
andals | -4.750 |
fuelled | -4.687 |
utics | -4.475 |
Token | website |
Token ID | 3052 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
confines | +4.221 |
slightest | +3.696 |
bounds | +3.397 |
assumption | +3.360 |
margins | +3.042 |
Neg logit contributions | |
zin | -4.554 |
umat | -4.433 |
andals | -4.348 |
fuelled | -4.317 |
utics | -4.133 |
Token | as |
Token ID | 355 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
confines | +4.510 |
slightest | +4.061 |
bounds | +3.820 |
assumption | +3.785 |
margins | +3.515 |
Neg logit contributions | |
zin | -3.047 |
umat | -2.960 |
andals | -2.823 |
fuelled | -2.813 |
piring | -2.673 |
Token | the |
Token ID | 262 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
confines | +1.115 |
slightest | +0.997 |
bounds | +0.925 |
assumption | +0.917 |
margins | +0.850 |
Neg logit contributions | |
zin | -0.962 |
umat | -0.941 |
fuelled | -0.910 |
andals | -0.907 |
ijing | -0.862 |
Token | default |
Token ID | 4277 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
confines | +3.542 |
slightest | +3.184 |
bounds | +2.915 |
assumption | +2.890 |
margins | +2.652 |
Neg logit contributions | |
zin | -3.314 |
umat | -3.237 |
fuelled | -3.133 |
andals | -3.123 |
ijing | -2.999 |
Token | Nice |
Token ID | 18460 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
confines | +16.458 |
slightest | +14.301 |
assumption | +13.985 |
bounds | +13.780 |
ones | +12.928 |
Neg logit contributions | |
zin | -13.581 |
umat | -13.197 |
andals | -12.923 |
utics | -12.422 |
fuelled | -12.132 |
Token | Hash |
Token ID | 26257 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
confines | +3.997 |
slightest | +3.595 |
bounds | +3.353 |
assumption | +3.294 |
margins | +3.096 |
Neg logit contributions | |
zin | -2.951 |
umat | -2.939 |
fuelled | -2.823 |
andals | -2.789 |
piring | -2.641 |
Token | website |
Token ID | 3052 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
confines | +0.096 |
slightest | +0.086 |
bounds | +0.079 |
assumption | +0.077 |
margins | +0.071 |
Neg logit contributions | |
zin | -0.091 |
umat | -0.089 |
fuelled | -0.086 |
andals | -0.086 |
ijing | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +7.90e-01 |
Pos logit contributions | |
confines | +2.552 |
slightest | +2.297 |
bounds | +2.157 |
assumption | +2.132 |
margins | +1.978 |
Neg logit contributions | |
zin | -1.761 |
umat | -1.713 |
andals | -1.637 |
fuelled | -1.636 |
piring | -1.567 |
Token | The |
Token ID | 383 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
zin | +2.098 |
umat | +2.044 |
fuelled | +1.944 |
andals | +1.918 |
ijing | +1.794 |
Neg logit contributions | |
confines | -4.512 |
slightest | -4.154 |
bounds | -3.905 |
assumption | -3.849 |
margins | -3.642 |
Token | House |
Token ID | 2097 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
confines | +1.586 |
slightest | +1.400 |
bounds | +1.266 |
assumption | +1.238 |
margins | +1.133 |
Neg logit contributions | |
zin | -1.941 |
umat | -1.919 |
fuelled | -1.859 |
andals | -1.853 |
ijing | -1.776 |
Token | of |
Token ID | 286 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
confines | +10.717 |
slightest | +9.332 |
bounds | +8.997 |
assumption | +8.728 |
margins | +8.183 |
Neg logit contributions | |
zin | -9.304 |
umat | -9.248 |
andals | -8.968 |
fuelled | -8.544 |
piring | -8.403 |
Token | Representatives |
Token ID | 17132 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
confines | +1.759 |
slightest | +1.567 |
bounds | +1.468 |
assumption | +1.422 |
margins | +1.311 |
Neg logit contributions | |
zin | -1.586 |
umat | -1.567 |
fuelled | -1.514 |
andals | -1.504 |
ijing | -1.437 |
Token | ( |
Token ID | 357 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
confines | +9.792 |
slightest | +8.457 |
bounds | +8.240 |
assumption | +7.901 |
margins | +7.432 |
Neg logit contributions | |
umat | -8.150 |
zin | -8.137 |
andals | -7.875 |
fuelled | -7.486 |
ijing | -7.429 |
Token | the |
Token ID | 1169 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
confines | +6.825 |
slightest | +6.156 |
bounds | +5.875 |
assumption | +5.723 |
margins | +5.405 |
Neg logit contributions | |
zin | -3.961 |
umat | -3.941 |
fuelled | -3.712 |
andals | -3.700 |
utics | -3.478 |
Token | lower |
Token ID | 2793 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
confines | +5.950 |
slightest | +5.226 |
bounds | +4.794 |
assumption | +4.682 |
margins | +4.314 |
Neg logit contributions | |
zin | -6.736 |
umat | -6.644 |
andals | -6.485 |
fuelled | -6.428 |
utics | -6.167 |
Token | house |
Token ID | 2156 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
confines | +10.960 |
bounds | +8.588 |
slightest | +8.513 |
assumption | +8.011 |
halves | +6.979 |
Neg logit contributions | |
zin | -19.429 |
umat | -19.030 |
andals | -19.015 |
piring | -18.688 |
fuelled | -18.280 |
Token | ) |
Token ID | 8 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
confines | +5.897 |
slightest | +5.185 |
bounds | +4.900 |
assumption | +4.832 |
margins | +4.490 |
Neg logit contributions | |
zin | -4.868 |
umat | -4.803 |
andals | -4.653 |
fuelled | -4.575 |
piring | -4.412 |
Token | has |
Token ID | 468 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
confines | +7.017 |
slightest | +6.104 |
bounds | +5.877 |
assumption | +5.631 |
margins | +5.315 |
Neg logit contributions | |
zin | -5.854 |
umat | -5.813 |
andals | -5.577 |
fuelled | -5.435 |
piring | -5.284 |
Token | 150 |
Token ID | 6640 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
confines | +4.622 |
slightest | +4.133 |
bounds | +3.870 |
assumption | +3.787 |
margins | +3.567 |
Neg logit contributions | |
zin | -3.950 |
umat | -3.942 |
andals | -3.756 |
fuelled | -3.665 |
ijing | -3.569 |
Token | members |
Token ID | 1866 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
confines | +1.313 |
slightest | +1.151 |
bounds | +1.053 |
assumption | +1.019 |
margins | +0.935 |
Neg logit contributions | |
zin | -1.545 |
umat | -1.531 |
fuelled | -1.487 |
andals | -1.486 |
ijing | -1.421 |
Token | / |
Token ID | 14 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
confines | +2.226 |
slightest | +2.031 |
bounds | +1.896 |
assumption | +1.860 |
margins | +1.752 |
Neg logit contributions | |
zin | -1.606 |
umat | -1.579 |
andals | -1.503 |
fuelled | -1.491 |
ijing | -1.417 |
Token | below |
Token ID | 35993 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
confines | +7.221 |
slightest | +6.478 |
bounds | +5.852 |
assumption | +5.597 |
margins | +5.348 |
Neg logit contributions | |
umat | -8.171 |
fuelled | -8.153 |
zin | -8.072 |
andals | -7.912 |
ijing | -7.618 |
Token | , |
Token ID | 11 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
confines | +8.704 |
slightest | +7.817 |
bounds | +7.472 |
assumption | +7.217 |
margins | +6.798 |
Neg logit contributions | |
umat | -7.588 |
zin | -7.583 |
andals | -7.133 |
fuelled | -6.914 |
piring | -6.805 |
Token | or |
Token ID | 393 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
confines | +3.119 |
slightest | +2.807 |
bounds | +2.590 |
assumption | +2.521 |
margins | +2.361 |
Neg logit contributions | |
zin | -3.052 |
umat | -3.011 |
andals | -2.912 |
fuelled | -2.879 |
ijing | -2.763 |
Token | left |
Token ID | 1364 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
confines | +2.189 |
slightest | +1.960 |
bounds | +1.805 |
assumption | +1.753 |
margins | +1.634 |
Neg logit contributions | |
zin | -2.186 |
umat | -2.163 |
andals | -2.074 |
fuelled | -2.059 |
ijing | -1.987 |
Token | / |
Token ID | 14 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
confines | +8.254 |
slightest | +7.261 |
bounds | +6.834 |
assumption | +6.601 |
margins | +6.256 |
Neg logit contributions | |
zin | -8.213 |
umat | -8.166 |
andals | -7.880 |
fuelled | -7.722 |
piring | -7.579 |
Token | right |
Token ID | 3506 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
confines | +12.197 |
slightest | +10.582 |
bounds | +9.634 |
assumption | +9.512 |
margins | +8.929 |
Neg logit contributions | |
umat | -15.920 |
andals | -15.846 |
zin | -15.577 |
fuelled | -15.574 |
ijing | -14.898 |
Token | . |
Token ID | 13 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
confines | +6.203 |
slightest | +5.536 |
bounds | +5.251 |
assumption | +5.077 |
margins | +4.830 |
Neg logit contributions | |
umat | -5.154 |
zin | -5.105 |
andals | -4.888 |
fuelled | -4.695 |
piring | -4.626 |
Token |
|
Token ID | 198 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
zin | +1.974 |
umat | +1.924 |
fuelled | +1.837 |
andals | +1.822 |
ijing | +1.714 |
Neg logit contributions | |
confines | -3.952 |
slightest | -3.624 |
bounds | -3.407 |
assumption | -3.357 |
margins | -3.169 |
Token |
|
Token ID | 198 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
confines | +3.650 |
slightest | +3.268 |
bounds | +3.169 |
assumption | +3.079 |
margins | +2.903 |
Neg logit contributions | |
zin | -2.056 |
umat | -2.048 |
andals | -1.953 |
fuelled | -1.945 |
ijing | -1.876 |
Token | With |
Token ID | 3152 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
confines | +0.918 |
slightest | +0.833 |
bounds | +0.788 |
assumption | +0.778 |
margins | +0.728 |
Neg logit contributions | |
zin | -0.519 |
umat | -0.507 |
fuelled | -0.492 |
andals | -0.491 |
ijing | -0.462 |
Token | ruled |
Token ID | 8879 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
confines | +7.367 |
slightest | +6.444 |
bounds | +6.091 |
assumption | +5.866 |
margins | +5.597 |
Neg logit contributions | |
zin | -6.460 |
umat | -6.286 |
andals | -6.193 |
ijing | -5.888 |
fuelled | -5.861 |
Token | last |
Token ID | 938 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
confines | +0.748 |
slightest | +0.670 |
bounds | +0.622 |
assumption | +0.612 |
margins | +0.573 |
Neg logit contributions | |
zin | -0.609 |
umat | -0.598 |
andals | -0.574 |
fuelled | -0.573 |
ijing | -0.551 |
Token | week |
Token ID | 1285 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
zin | +1.442 |
umat | +1.412 |
fuelled | +1.353 |
andals | +1.345 |
ijing | +1.273 |
Neg logit contributions | |
confines | -2.254 |
slightest | -2.057 |
bounds | -1.914 |
assumption | -1.884 |
margins | -1.769 |
Token | that |
Token ID | 326 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
zin | +2.788 |
umat | +2.738 |
fuelled | +2.645 |
andals | +2.622 |
ijing | +2.501 |
Neg logit contributions | |
confines | -3.442 |
slightest | -3.116 |
bounds | -2.868 |
assumption | -2.816 |
margins | -2.622 |
Token | the |
Token ID | 262 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
zin | +1.972 |
umat | +1.938 |
fuelled | +1.876 |
andals | +1.861 |
ijing | +1.778 |
Neg logit contributions | |
confines | -2.235 |
slightest | -2.014 |
bounds | -1.848 |
assumption | -1.812 |
margins | -1.682 |
Token | lower |
Token ID | 2793 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
confines | +1.040 |
slightest | +0.919 |
bounds | +0.832 |
assumption | +0.814 |
margins | +0.743 |
Neg logit contributions | |
zin | -1.218 |
umat | -1.199 |
fuelled | -1.163 |
andals | -1.158 |
ijing | -1.115 |
Token | court |
Token ID | 2184 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
confines | +13.261 |
bounds | +10.974 |
slightest | +10.602 |
assumption | +10.066 |
margins | +9.194 |
Neg logit contributions | |
zin | -17.102 |
andals | -16.564 |
umat | -16.347 |
utics | -15.717 |
ijing | -15.709 |
Token | judge |
Token ID | 5052 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
confines | +10.528 |
slightest | +9.255 |
bounds | +8.808 |
assumption | +8.753 |
margins | +7.997 |
Neg logit contributions | |
zin | -8.561 |
umat | -8.487 |
andals | -8.214 |
ijing | -7.977 |
piring | -7.912 |
Token | was |
Token ID | 373 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
confines | +9.906 |
slightest | +8.662 |
bounds | +8.348 |
assumption | +8.128 |
margins | +7.555 |
Neg logit contributions | |
zin | -7.705 |
umat | -7.602 |
andals | -7.485 |
fuelled | -7.191 |
ijing | -7.149 |
Token | wrong |
Token ID | 2642 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
confines | +1.852 |
slightest | +1.604 |
bounds | +1.467 |
assumption | +1.411 |
margins | +1.262 |
Neg logit contributions | |
zin | -2.539 |
umat | -2.518 |
andals | -2.434 |
fuelled | -2.402 |
piring | -2.326 |
Token | . |
Token ID | 13 |
Feature activation | +3.20e-04 |
Pos logit contributions | |
confines | +7.100 |
slightest | +6.184 |
bounds | +6.136 |
assumption | +5.952 |
margins | +5.538 |
Neg logit contributions | |
zin | -5.491 |
umat | -5.384 |
andals | -5.253 |
fuelled | -5.186 |
piring | -4.974 |
Token | quality |
Token ID | 3081 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
confines | +0.666 |
slightest | +0.599 |
bounds | +0.548 |
assumption | +0.538 |
margins | +0.496 |
Neg logit contributions | |
zin | -0.631 |
umat | -0.623 |
fuelled | -0.600 |
andals | -0.599 |
ijing | -0.573 |
Token | cable |
Token ID | 7862 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
confines | +2.324 |
slightest | +2.065 |
bounds | +1.884 |
assumption | +1.856 |
margins | +1.696 |
Neg logit contributions | |
zin | -2.441 |
umat | -2.417 |
andals | -2.324 |
fuelled | -2.314 |
piring | -2.220 |
Token | . |
Token ID | 13 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
confines | +9.948 |
slightest | +8.965 |
assumption | +8.368 |
bounds | +8.243 |
ones | +7.534 |
Neg logit contributions | |
umat | -8.646 |
zin | -8.495 |
andals | -8.151 |
fuelled | -7.807 |
utics | -7.743 |
Token | The |
Token ID | 383 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
zin | +0.060 |
umat | +0.059 |
fuelled | +0.057 |
andals | +0.056 |
ijing | +0.053 |
Neg logit contributions | |
confines | -0.114 |
slightest | -0.104 |
bounds | -0.098 |
assumption | -0.096 |
margins | -0.091 |
Token | free |
Token ID | 1479 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
confines | +2.536 |
slightest | +2.255 |
bounds | +2.017 |
assumption | +1.994 |
margins | +1.817 |
Neg logit contributions | |
zin | -3.011 |
umat | -2.972 |
fuelled | -2.881 |
andals | -2.867 |
ijing | -2.761 |
Token | cable |
Token ID | 7862 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
confines | +9.766 |
slightest | +8.588 |
assumption | +8.030 |
bounds | +7.902 |
margins | +7.328 |
Neg logit contributions | |
umat | -9.005 |
zin | -8.901 |
andals | -8.798 |
fuelled | -8.705 |
utics | -8.403 |
Token | did |
Token ID | 750 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
confines | +16.070 |
slightest | +13.875 |
bounds | +13.240 |
assumption | +13.145 |
ones | +12.201 |
Neg logit contributions | |
zin | -14.054 |
umat | -13.895 |
andals | -13.280 |
utics | -12.769 |
ijing | -12.632 |
Token | not |
Token ID | 407 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
confines | +6.879 |
slightest | +6.172 |
bounds | +5.703 |
assumption | +5.563 |
margins | +5.155 |
Neg logit contributions | |
zin | -6.939 |
umat | -6.819 |
andals | -6.630 |
fuelled | -6.519 |
ijing | -6.460 |
Token | do |
Token ID | 466 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
confines | +8.667 |
slightest | +7.821 |
bounds | +7.099 |
assumption | +6.942 |
margins | +6.434 |
Neg logit contributions | |
zin | -8.930 |
umat | -8.724 |
andals | -8.377 |
fuelled | -8.293 |
ijing | -8.282 |
Token | the |
Token ID | 262 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
confines | +6.577 |
slightest | +5.954 |
bounds | +5.436 |
assumption | +5.234 |
margins | +4.888 |
Neg logit contributions | |
zin | -7.447 |
umat | -7.299 |
andals | -7.169 |
fuelled | -7.080 |
ijing | -6.917 |
Token | sub |
Token ID | 850 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
confines | +7.411 |
slightest | +6.864 |
bounds | +5.844 |
assumption | +5.784 |
margins | +5.149 |
Neg logit contributions | |
zin | -11.016 |
umat | -10.911 |
andals | -10.545 |
fuelled | -10.529 |
ijing | -10.375 |
Token | � |
Token ID | 447 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
confines | +4.549 |
slightest | +4.063 |
bounds | +3.862 |
assumption | +3.685 |
margins | +3.502 |
Neg logit contributions | |
zin | -3.310 |
umat | -3.256 |
andals | -3.120 |
fuelled | -3.090 |
ijing | -3.001 |
Token | � |
Token ID | 247 |
Feature activation | +9.14e-01 |
Pos logit contributions | |
zin | +0.198 |
umat | +0.193 |
fuelled | +0.186 |
andals | +0.184 |
ijing | +0.174 |
Neg logit contributions | |
confines | -0.290 |
slightest | -0.264 |
bounds | -0.244 |
assumption | -0.241 |
margins | -0.226 |
Token | s |
Token ID | 82 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
zin | +3.272 |
umat | +3.209 |
fuelled | +3.094 |
andals | +3.069 |
ijing | +2.921 |
Neg logit contributions | |
confines | -4.353 |
slightest | -3.949 |
bounds | -3.650 |
assumption | -3.587 |
margins | -3.351 |
Token | 2013 |
Token ID | 2211 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
confines | +0.413 |
slightest | +0.367 |
bounds | +0.333 |
assumption | +0.327 |
margins | +0.300 |
Neg logit contributions | |
zin | -0.478 |
umat | -0.471 |
fuelled | -0.457 |
andals | -0.452 |
ijing | -0.440 |
Token | bullying |
Token ID | 20714 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
zin | +2.123 |
umat | +2.091 |
fuelled | +2.032 |
andals | +2.020 |
ijing | +1.946 |
Neg logit contributions | |
confines | -1.759 |
slightest | -1.553 |
bounds | -1.402 |
assumption | -1.370 |
margins | -1.250 |
Token | scandal |
Token ID | 10731 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
zin | +5.065 |
umat | +4.989 |
fuelled | +4.872 |
andals | +4.829 |
piring | +4.621 |
Neg logit contributions | |
confines | -4.776 |
slightest | -4.269 |
bounds | -3.866 |
assumption | -3.778 |
margins | -3.472 |
Token | in |
Token ID | 287 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
umat | +8.467 |
zin | +8.429 |
fuelled | +8.292 |
andals | +8.065 |
piring | +7.599 |
Neg logit contributions | |
confines | -10.411 |
slightest | -9.577 |
bounds | -8.658 |
assumption | -8.609 |
margins | -7.919 |
Token | Miami |
Token ID | 8437 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
zin | +1.068 |
umat | +1.052 |
fuelled | +1.020 |
andals | +1.013 |
ijing | +0.973 |
Neg logit contributions | |
confines | -1.014 |
slightest | -0.903 |
bounds | -0.823 |
assumption | -0.804 |
margins | -0.739 |
Token | kept |
Token ID | 4030 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
zin | +1.993 |
umat | +1.944 |
fuelled | +1.861 |
andals | +1.856 |
ijing | +1.768 |
Neg logit contributions | |
confines | -2.825 |
slightest | -2.556 |
bounds | -2.379 |
assumption | -2.331 |
margins | -2.188 |
Token | him |
Token ID | 683 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
confines | +1.197 |
slightest | +1.061 |
bounds | +0.968 |
assumption | +0.945 |
margins | +0.872 |
Neg logit contributions | |
zin | -1.314 |
umat | -1.297 |
andals | -1.247 |
fuelled | -1.245 |
ijing | -1.199 |
Token | out |
Token ID | 503 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
confines | +2.095 |
slightest | +1.879 |
bounds | +1.752 |
assumption | +1.708 |
margins | +1.591 |
Neg logit contributions | |
zin | -1.704 |
umat | -1.659 |
andals | -1.605 |
fuelled | -1.571 |
ijing | -1.533 |
Token | This |
Token ID | 770 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
zin | +2.969 |
umat | +2.907 |
fuelled | +2.783 |
andals | +2.754 |
ijing | +2.600 |
Neg logit contributions | |
confines | -5.114 |
slightest | -4.681 |
bounds | -4.370 |
assumption | -4.301 |
margins | -4.052 |
Token | was |
Token ID | 373 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
zin | +8.106 |
umat | +7.865 |
fuelled | +7.857 |
andals | +7.623 |
piring | +7.358 |
Neg logit contributions | |
confines | -8.064 |
slightest | -7.188 |
bounds | -6.599 |
assumption | -6.322 |
margins | -5.953 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
zin | +3.106 |
umat | +3.054 |
fuelled | +2.948 |
andals | +2.929 |
ijing | +2.801 |
Neg logit contributions | |
confines | -3.549 |
slightest | -3.196 |
bounds | -2.937 |
assumption | -2.881 |
margins | -2.676 |
Token | beginning |
Token ID | 3726 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
confines | +0.487 |
slightest | +0.436 |
bounds | +0.395 |
assumption | +0.389 |
margins | +0.355 |
Neg logit contributions | |
zin | -0.527 |
umat | -0.520 |
fuelled | -0.502 |
andals | -0.502 |
ijing | -0.481 |
Token | of |
Token ID | 286 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
zin | +3.768 |
umat | +3.669 |
fuelled | +3.558 |
andals | +3.504 |
ijing | +3.333 |
Neg logit contributions | |
confines | -5.772 |
slightest | -5.290 |
bounds | -4.889 |
assumption | -4.776 |
margins | -4.509 |
Token | an |
Token ID | 281 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
zin | +1.933 |
umat | +1.908 |
fuelled | +1.843 |
andals | +1.832 |
ijing | +1.754 |
Neg logit contributions | |
confines | -1.970 |
slightest | -1.761 |
bounds | -1.613 |
assumption | -1.581 |
margins | -1.460 |
Token | exodus |
Token ID | 39524 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
zin | +8.906 |
umat | +8.696 |
fuelled | +8.568 |
andals | +8.317 |
piring | +8.109 |
Neg logit contributions | |
confines | -8.880 |
slightest | -7.832 |
bounds | -7.211 |
assumption | -6.728 |
margins | -6.520 |
Token | of |
Token ID | 286 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
zin | +3.130 |
umat | +3.083 |
andals | +2.941 |
fuelled | +2.940 |
ijing | +2.788 |
Neg logit contributions | |
confines | -4.484 |
slightest | -4.076 |
bounds | -3.788 |
assumption | -3.730 |
margins | -3.488 |
Token | would |
Token ID | 561 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
zin | +4.762 |
umat | +4.662 |
fuelled | +4.552 |
andals | +4.486 |
piring | +4.328 |
Neg logit contributions | |
confines | -4.781 |
slightest | -4.270 |
bounds | -3.900 |
assumption | -3.810 |
margins | -3.513 |
Token | - |
Token ID | 12 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
zin | +2.025 |
umat | +1.983 |
fuelled | +1.917 |
andals | +1.897 |
ijing | +1.811 |
Neg logit contributions | |
confines | -2.499 |
slightest | -2.253 |
bounds | -2.079 |
assumption | -2.038 |
margins | -1.901 |
Token | be |
Token ID | 1350 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
confines | +2.839 |
slightest | +2.603 |
assumption | +2.421 |
bounds | +2.420 |
margins | +2.247 |
Neg logit contributions | |
umat | -1.635 |
zin | -1.625 |
fuelled | -1.563 |
andals | -1.523 |
ijing | -1.440 |
Token | A |
Token ID | 32 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
confines | +1.175 |
slightest | +1.063 |
bounds | +1.001 |
assumption | +0.988 |
margins | +0.929 |
Neg logit contributions | |
zin | -0.742 |
umat | -0.725 |
fuelled | -0.704 |
andals | -0.690 |
ijing | -0.655 |
Token | Brexit |
Token ID | 11435 |
Feature activation | +9.93e-01 |
Pos logit contributions | |
zin | +5.490 |
umat | +5.416 |
fuelled | +5.248 |
andals | +5.213 |
piring | +5.019 |
Neg logit contributions | |
confines | -4.706 |
slightest | -4.163 |
bounds | -3.759 |
assumption | -3.664 |
margins | -3.360 |
Token | betrayal |
Token ID | 29355 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
zin | +4.589 |
umat | +4.528 |
fuelled | +4.411 |
andals | +4.355 |
ijing | +4.212 |
Neg logit contributions | |
confines | -3.662 |
slightest | -3.225 |
bounds | -2.900 |
assumption | -2.814 |
margins | -2.569 |
Token | is |
Token ID | 318 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
zin | +5.709 |
umat | +5.629 |
fuelled | +5.506 |
andals | +5.368 |
piring | +5.132 |
Neg logit contributions | |
confines | -7.199 |
slightest | -6.545 |
bounds | -5.985 |
assumption | -5.867 |
margins | -5.482 |
Token | coming |
Token ID | 2406 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
zin | +3.151 |
umat | +3.096 |
fuelled | +2.986 |
andals | +2.975 |
ijing | +2.842 |
Neg logit contributions | |
confines | -3.542 |
slightest | -3.187 |
bounds | -2.928 |
assumption | -2.870 |
margins | -2.665 |
Token | – |
Token ID | 784 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
zin | +9.236 |
umat | +9.115 |
fuelled | +8.951 |
andals | +8.654 |
ijing | +8.316 |
Neg logit contributions | |
confines | -12.144 |
slightest | -11.067 |
bounds | -10.111 |
assumption | -9.967 |
margins | -9.320 |
Token | but |
Token ID | 475 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
zin | +10.422 |
fuelled | +10.274 |
umat | +10.254 |
andals | +9.806 |
piring | +9.565 |
Neg logit contributions | |
confines | -10.126 |
slightest | -9.060 |
bounds | -8.142 |
assumption | -7.878 |
margins | -7.318 |
Token | who |
Token ID | 508 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
zin | +1.820 |
umat | +1.786 |
fuelled | +1.738 |
andals | +1.718 |
piring | +1.646 |
Neg logit contributions | |
confines | -1.954 |
slightest | -1.756 |
bounds | -1.606 |
assumption | -1.574 |
margins | -1.456 |
Token | will |
Token ID | 481 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
zin | +4.383 |
umat | +4.317 |
fuelled | +4.195 |
andals | +4.129 |
ijing | +3.945 |
Neg logit contributions | |
confines | -5.012 |
slightest | -4.527 |
bounds | -4.136 |
assumption | -4.090 |
margins | -3.774 |
Token | get |
Token ID | 651 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
zin | +1.432 |
umat | +1.413 |
fuelled | +1.374 |
andals | +1.359 |
ijing | +1.304 |
Neg logit contributions | |
confines | -1.332 |
slightest | -1.186 |
bounds | -1.075 |
assumption | -1.051 |
margins | -0.966 |
Token | the |
Token ID | 262 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
confines | +0.808 |
slightest | +0.724 |
bounds | +0.665 |
assumption | +0.650 |
margins | +0.603 |
Neg logit contributions | |
zin | -0.781 |
umat | -0.764 |
andals | -0.741 |
fuelled | -0.735 |
ijing | -0.707 |
Token | retion |
Token ID | 12307 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
zin | +1.454 |
umat | +1.414 |
fuelled | +1.372 |
andals | +1.331 |
ijing | +1.296 |
Neg logit contributions | |
confines | -2.501 |
slightest | -2.284 |
bounds | -2.140 |
assumption | -2.095 |
margins | -1.982 |
Token | � |
Token ID | 447 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
confines | +9.596 |
bounds | +8.324 |
slightest | +8.221 |
assumption | +7.829 |
margins | +7.539 |
Neg logit contributions | |
zin | -7.825 |
umat | -7.647 |
andals | -7.486 |
fuelled | -7.295 |
ijing | -7.106 |
Token | � |
Token ID | 251 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
confines | +7.595 |
slightest | +6.712 |
bounds | +6.571 |
assumption | +6.352 |
margins | +5.957 |
Neg logit contributions | |
zin | -4.596 |
umat | -4.443 |
andals | -4.350 |
fuelled | -4.287 |
ilst | -4.069 |
Token | to |
Token ID | 284 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
confines | +0.724 |
slightest | +0.644 |
bounds | +0.613 |
assumption | +0.595 |
margins | +0.559 |
Neg logit contributions | |
zin | -0.555 |
umat | -0.551 |
andals | -0.525 |
fuelled | -0.525 |
ijing | -0.501 |
Token | handle |
Token ID | 5412 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
confines | +2.782 |
slightest | +2.464 |
bounds | +2.272 |
assumption | +2.214 |
margins | +2.049 |
Neg logit contributions | |
zin | -2.898 |
umat | -2.839 |
andals | -2.757 |
fuelled | -2.757 |
ijing | -2.653 |
Token | complaints |
Token ID | 9687 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
confines | +1.618 |
slightest | +1.463 |
bounds | +1.350 |
assumption | +1.336 |
margins | +1.235 |
Neg logit contributions | |
zin | -1.442 |
umat | -1.422 |
fuelled | -1.374 |
andals | -1.368 |
ijing | -1.317 |
Token | . |
Token ID | 13 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
zin | +7.775 |
umat | +7.601 |
fuelled | +7.514 |
andals | +7.228 |
piring | +6.939 |
Neg logit contributions | |
confines | -10.855 |
slightest | -9.980 |
bounds | -9.111 |
assumption | -8.990 |
margins | -8.412 |
Token | Under |
Token ID | 4698 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
zin | +1.359 |
umat | +1.332 |
fuelled | +1.283 |
andals | +1.251 |
ijing | +1.192 |
Neg logit contributions | |
confines | -2.625 |
slightest | -2.394 |
bounds | -2.262 |
assumption | -2.225 |
margins | -2.103 |
Token | existing |
Token ID | 4683 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
confines | +0.851 |
slightest | +0.760 |
bounds | +0.699 |
assumption | +0.689 |
margins | +0.632 |
Neg logit contributions | |
zin | -0.850 |
umat | -0.836 |
fuelled | -0.819 |
andals | -0.806 |
ijing | -0.776 |
Token | regulations |
Token ID | 6647 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
confines | +4.332 |
slightest | +3.704 |
bounds | +3.458 |
assumption | +3.405 |
margins | +3.042 |
Neg logit contributions | |
zin | -5.494 |
umat | -5.464 |
fuelled | -5.311 |
andals | -5.279 |
ijing | -5.085 |
Token | , |
Token ID | 11 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
confines | +3.387 |
slightest | +2.993 |
bounds | +2.828 |
assumption | +2.777 |
margins | +2.585 |
Neg logit contributions | |
zin | -2.654 |
umat | -2.614 |
andals | -2.526 |
fuelled | -2.499 |
ijing | -2.396 |
Token | new |
Token ID | 649 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
zin | +0.826 |
umat | +0.812 |
fuelled | +0.781 |
andals | +0.778 |
ijing | +0.746 |
Neg logit contributions | |
confines | -0.887 |
slightest | -0.795 |
bounds | -0.728 |
assumption | -0.715 |
margins | -0.662 |
Token | garden |
Token ID | 11376 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
zin | +0.598 |
umat | +0.589 |
fuelled | +0.569 |
andals | +0.567 |
ijing | +0.547 |
Neg logit contributions | |
confines | -0.483 |
slightest | -0.423 |
bounds | -0.382 |
assumption | -0.374 |
margins | -0.340 |
Token | centre |
Token ID | 7372 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
zin | +2.186 |
umat | +2.152 |
fuelled | +2.079 |
andals | +2.072 |
ijing | +1.992 |
Neg logit contributions | |
confines | -1.947 |
slightest | -1.722 |
bounds | -1.564 |
assumption | -1.531 |
margins | -1.405 |
Token | ." |
Token ID | 526 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
confines | +1.405 |
slightest | +1.261 |
bounds | +1.168 |
assumption | +1.153 |
margins | +1.075 |
Neg logit contributions | |
zin | -1.118 |
umat | -1.099 |
andals | -1.041 |
fuelled | -1.032 |
ijing | -1.001 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
zin | +1.067 |
umat | +1.038 |
fuelled | +0.986 |
andals | +0.976 |
piring | +0.902 |
Neg logit contributions | |
confines | -2.547 |
slightest | -2.358 |
bounds | -2.213 |
assumption | -2.184 |
margins | -2.072 |
Token | Five |
Token ID | 20029 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
zin | +0.835 |
umat | +0.818 |
fuelled | +0.792 |
andals | +0.780 |
ijing | +0.735 |
Neg logit contributions | |
confines | -1.377 |
slightest | -1.257 |
bounds | -1.175 |
assumption | -1.159 |
margins | -1.089 |
Token | years |
Token ID | 812 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
zin | +9.320 |
umat | +9.113 |
fuelled | +8.847 |
andals | +8.802 |
piring | +8.438 |
Neg logit contributions | |
confines | -9.351 |
slightest | -8.338 |
bounds | -7.595 |
assumption | -7.437 |
margins | -6.852 |
Token | ago |
Token ID | 2084 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
zin | +4.379 |
umat | +4.286 |
fuelled | +4.121 |
andals | +4.075 |
ijing | +3.852 |
Neg logit contributions | |
confines | -7.097 |
slightest | -6.495 |
bounds | -6.038 |
assumption | -5.940 |
margins | -5.591 |
Token | , |
Token ID | 11 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
zin | +3.556 |
umat | +3.490 |
fuelled | +3.366 |
andals | +3.339 |
ijing | +3.182 |
Neg logit contributions | |
confines | -4.597 |
slightest | -4.164 |
bounds | -3.846 |
assumption | -3.778 |
margins | -3.526 |
Token | I |
Token ID | 314 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
zin | +1.851 |
umat | +1.822 |
fuelled | +1.752 |
andals | +1.744 |
ijing | +1.670 |
Neg logit contributions | |
confines | -2.073 |
slightest | -1.863 |
bounds | -1.712 |
assumption | -1.681 |
margins | -1.561 |
Token | moved |
Token ID | 3888 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
zin | +3.330 |
umat | +3.265 |
fuelled | +3.145 |
andals | +3.106 |
ijing | +2.946 |
Neg logit contributions | |
confines | -5.006 |
slightest | -4.563 |
bounds | -4.236 |
assumption | -4.169 |
margins | -3.914 |
Token | whose |
Token ID | 3025 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
confines | +2.533 |
slightest | +2.275 |
bounds | +2.103 |
assumption | +2.059 |
margins | +1.913 |
Neg logit contributions | |
zin | -2.148 |
umat | -2.116 |
andals | -2.027 |
fuelled | -1.999 |
ijing | -1.945 |
Token | clubhouse |
Token ID | 48018 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
confines | +1.202 |
slightest | +1.066 |
bounds | +0.965 |
assumption | +0.950 |
margins | +0.869 |
Neg logit contributions | |
zin | -1.350 |
umat | -1.326 |
fuelled | -1.280 |
andals | -1.278 |
ijing | -1.234 |
Token | chemistry |
Token ID | 16585 |
Feature activation | +9.61e-02 |
Pos logit contributions | |
confines | +7.841 |
slightest | +6.849 |
bounds | +6.469 |
assumption | +6.325 |
margins | +5.951 |
Neg logit contributions | |
zin | -6.182 |
umat | -6.064 |
andals | -5.704 |
ijing | -5.580 |
fuelled | -5.527 |
Token | could |
Token ID | 714 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
zin | +0.342 |
umat | +0.335 |
andals | +0.320 |
fuelled | +0.318 |
ijing | +0.305 |
Neg logit contributions | |
confines | -0.467 |
slightest | -0.421 |
bounds | -0.392 |
assumption | -0.384 |
margins | -0.361 |
Token | � |
Token ID | 447 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
confines | +1.934 |
slightest | +1.721 |
bounds | +1.577 |
assumption | +1.534 |
margins | +1.419 |
Neg logit contributions | |
zin | -1.989 |
umat | -1.966 |
andals | -1.900 |
fuelled | -1.893 |
ijing | -1.827 |
Token | � |
Token ID | 247 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
confines | +4.359 |
slightest | +3.993 |
bounds | +3.730 |
assumption | +3.688 |
margins | +3.476 |
Neg logit contributions | |
zin | -2.542 |
umat | -2.421 |
andals | -2.372 |
fuelled | -2.357 |
ilst | -2.197 |
Token | ve |
Token ID | 303 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
zin | +1.319 |
umat | +1.290 |
fuelled | +1.237 |
andals | +1.226 |
ijing | +1.156 |
Neg logit contributions | |
confines | -2.258 |
slightest | -2.068 |
bounds | -1.929 |
assumption | -1.899 |
margins | -1.789 |
Token | been |
Token ID | 587 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
confines | +1.896 |
slightest | +1.737 |
bounds | +1.610 |
assumption | +1.561 |
margins | +1.494 |
Neg logit contributions | |
zin | -1.305 |
umat | -1.264 |
andals | -1.241 |
fuelled | -1.229 |
piring | -1.172 |
Token | better |
Token ID | 1365 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
confines | +1.180 |
slightest | +1.071 |
bounds | +0.979 |
assumption | +0.955 |
margins | +0.888 |
Neg logit contributions | |
zin | -1.049 |
umat | -1.029 |
andals | -0.995 |
fuelled | -0.980 |
ijing | -0.948 |
Token | last |
Token ID | 938 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
confines | +3.757 |
slightest | +3.382 |
bounds | +3.143 |
assumption | +3.062 |
margins | +2.873 |
Neg logit contributions | |
zin | -2.923 |
umat | -2.887 |
andals | -2.774 |
fuelled | -2.708 |
ijing | -2.674 |
Token | season |
Token ID | 1622 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
confines | +0.602 |
slightest | +0.544 |
bounds | +0.503 |
assumption | +0.497 |
margins | +0.462 |
Neg logit contributions | |
zin | -0.465 |
umat | -0.456 |
andals | -0.436 |
fuelled | -0.433 |
ijing | -0.417 |
Token | sounds |
Token ID | 5238 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
zin | +8.739 |
umat | +8.552 |
fuelled | +8.452 |
andals | +8.317 |
ijing | +8.116 |
Neg logit contributions | |
confines | -6.998 |
slightest | -6.144 |
bounds | -5.492 |
assumption | -5.379 |
margins | -4.834 |
Token | , |
Token ID | 11 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
zin | +6.965 |
umat | +6.780 |
fuelled | +6.615 |
andals | +6.468 |
ijing | +6.244 |
Neg logit contributions | |
confines | -9.559 |
slightest | -8.687 |
bounds | -8.013 |
assumption | -7.874 |
margins | -7.345 |
Token | such |
Token ID | 884 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
zin | +1.178 |
umat | +1.158 |
fuelled | +1.115 |
andals | +1.109 |
ijing | +1.056 |
Neg logit contributions | |
confines | -1.491 |
slightest | -1.351 |
bounds | -1.247 |
assumption | -1.225 |
margins | -1.143 |
Token | as |
Token ID | 355 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
zin | +3.679 |
umat | +3.548 |
fuelled | +3.498 |
andals | +3.415 |
piring | +3.263 |
Neg logit contributions | |
confines | -4.384 |
slightest | -4.007 |
bounds | -3.583 |
assumption | -3.553 |
margins | -3.264 |
Token | � |
Token ID | 564 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
confines | +1.318 |
slightest | +1.188 |
bounds | +1.086 |
assumption | +1.057 |
margins | +0.991 |
Neg logit contributions | |
zin | -1.277 |
umat | -1.265 |
fuelled | -1.212 |
andals | -1.208 |
ijing | -1.156 |
Token | � |
Token ID | 250 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
zin | +5.534 |
umat | +5.478 |
fuelled | +5.104 |
andals | +5.073 |
piring | +4.795 |
Neg logit contributions | |
confines | -12.204 |
slightest | -11.230 |
bounds | -10.584 |
assumption | -10.420 |
margins | -9.857 |
Token | bah |
Token ID | 47041 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
zin | +4.104 |
umat | +3.983 |
fuelled | +3.807 |
andals | +3.781 |
ijing | +3.577 |
Neg logit contributions | |
confines | -7.029 |
slightest | -6.408 |
bounds | -5.993 |
assumption | -5.907 |
margins | -5.547 |
Token | , |
Token ID | 11 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
zin | +1.276 |
umat | +1.258 |
fuelled | +1.218 |
andals | +1.210 |
piring | +1.146 |
Neg logit contributions | |
confines | -1.786 |
slightest | -1.628 |
bounds | -1.511 |
assumption | -1.482 |
margins | -1.392 |
Token | � |
Token ID | 447 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
confines | +2.983 |
slightest | +2.729 |
bounds | +2.512 |
assumption | +2.462 |
margins | +2.297 |
Neg logit contributions | |
zin | -2.465 |
umat | -2.426 |
andals | -2.342 |
fuelled | -2.333 |
ijing | -2.225 |
Token | � |
Token ID | 251 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
confines | +4.706 |
slightest | +4.298 |
bounds | +4.035 |
assumption | +3.980 |
margins | +3.737 |
Neg logit contributions | |
zin | -2.595 |
umat | -2.479 |
andals | -2.402 |
fuelled | -2.395 |
ilst | -2.263 |
Token | � |
Token ID | 564 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
zin | +1.882 |
umat | +1.853 |
fuelled | +1.789 |
andals | +1.779 |
ijing | +1.698 |
Neg logit contributions | |
confines | -2.126 |
slightest | -1.919 |
bounds | -1.764 |
assumption | -1.727 |
margins | -1.606 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
confines | +2.020 |
slightest | +1.832 |
bounds | +1.750 |
assumption | +1.711 |
margins | +1.617 |
Neg logit contributions | |
zin | -1.061 |
umat | -1.048 |
fuelled | -0.998 |
andals | -0.994 |
ijing | -0.945 |
Token | Me |
Token ID | 5308 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
confines | +0.715 |
slightest | +0.650 |
bounds | +0.615 |
assumption | +0.604 |
margins | +0.570 |
Neg logit contributions | |
zin | -0.398 |
umat | -0.392 |
fuelled | -0.382 |
andals | -0.371 |
ijing | -0.354 |
Token | and |
Token ID | 290 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
zin | +3.610 |
umat | +3.526 |
fuelled | +3.375 |
andals | +3.332 |
piring | +3.120 |
Neg logit contributions | |
confines | -7.164 |
slightest | -6.586 |
bounds | -6.167 |
assumption | -6.078 |
margins | -5.750 |
Token | you |
Token ID | 345 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
confines | +6.553 |
slightest | +5.804 |
bounds | +5.336 |
assumption | +5.318 |
margins | +4.849 |
Neg logit contributions | |
zin | -6.863 |
umat | -6.857 |
andals | -6.707 |
fuelled | -6.707 |
ijing | -6.464 |
Token | . |
Token ID | 13 |
Feature activation | +9.17e-02 |
Pos logit contributions | |
zin | +2.203 |
umat | +2.169 |
fuelled | +2.079 |
andals | +2.076 |
ijing | +1.973 |
Neg logit contributions | |
confines | -3.036 |
slightest | -2.757 |
bounds | -2.556 |
assumption | -2.514 |
margins | -2.350 |
Token | Our |
Token ID | 3954 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
zin | +0.263 |
umat | +0.257 |
fuelled | +0.247 |
andals | +0.244 |
ijing | +0.231 |
Neg logit contributions | |
confines | -0.502 |
slightest | -0.456 |
bounds | -0.432 |
assumption | -0.424 |
margins | -0.401 |
Token | sisters |
Token ID | 15153 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
zin | +0.174 |
umat | +0.172 |
fuelled | +0.166 |
andals | +0.166 |
ijing | +0.159 |
Neg logit contributions | |
confines | -0.160 |
slightest | -0.142 |
bounds | -0.129 |
assumption | -0.127 |
margins | -0.116 |
Token | and |
Token ID | 290 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
confines | +3.173 |
slightest | +2.835 |
bounds | +2.645 |
assumption | +2.590 |
margins | +2.413 |
Neg logit contributions | |
zin | -2.637 |
umat | -2.632 |
andals | -2.574 |
fuelled | -2.474 |
ijing | -2.414 |
Token | brothers |
Token ID | 9397 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
confines | +4.011 |
slightest | +3.521 |
bounds | +3.232 |
assumption | +3.098 |
margins | +2.833 |
Neg logit contributions | |
zin | -5.080 |
umat | -5.030 |
fuelled | -4.837 |
andals | -4.826 |
ijing | -4.791 |
Token | . |
Token ID | 13 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
zin | +2.781 |
umat | +2.735 |
andals | +2.633 |
fuelled | +2.617 |
ijing | +2.487 |
Neg logit contributions | |
confines | -3.858 |
slightest | -3.492 |
bounds | -3.249 |
assumption | -3.188 |
margins | -2.987 |
Token | The |
Token ID | 383 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
zin | +1.194 |
umat | +1.166 |
fuelled | +1.114 |
andals | +1.095 |
ijing | +1.036 |
Neg logit contributions | |
confines | -2.461 |
slightest | -2.248 |
bounds | -2.126 |
assumption | -2.090 |
margins | -1.980 |
Token | air |
Token ID | 1633 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
confines | +2.473 |
slightest | +2.184 |
bounds | +1.957 |
assumption | +1.922 |
margins | +1.760 |
Neg logit contributions | |
zin | -3.161 |
umat | -3.138 |
andals | -3.034 |
fuelled | -2.987 |
ijing | -2.909 |
Token | by |
Token ID | 416 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
zin | +0.062 |
umat | +0.061 |
andals | +0.059 |
fuelled | +0.058 |
ijing | +0.056 |
Neg logit contributions | |
confines | -0.081 |
slightest | -0.073 |
bounds | -0.068 |
assumption | -0.067 |
margins | -0.063 |
Token | CO |
Token ID | 7375 |
Feature activation | -5.10e-02 |
Pos logit contributions | |
zin | +4.311 |
umat | +4.216 |
fuelled | +4.142 |
andals | +4.065 |
piring | +3.916 |
Neg logit contributions | |
confines | -4.302 |
slightest | -3.838 |
bounds | -3.493 |
assumption | -3.425 |
margins | -3.147 |
Token | 2 |
Token ID | 362 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
confines | +0.214 |
slightest | +0.189 |
bounds | +0.174 |
assumption | +0.171 |
margins | +0.159 |
Neg logit contributions | |
zin | -0.214 |
umat | -0.212 |
andals | -0.202 |
fuelled | -0.201 |
piring | -0.195 |
Token | : |
Token ID | 1058 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
zin | +3.693 |
umat | +3.617 |
fuelled | +3.516 |
andals | +3.476 |
ijing | +3.322 |
Neg logit contributions | |
confines | -4.242 |
slightest | -3.824 |
bounds | -3.511 |
assumption | -3.443 |
margins | -3.198 |
Token | more |
Token ID | 517 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
zin | +1.242 |
umat | +1.223 |
fuelled | +1.177 |
andals | +1.174 |
piring | +1.120 |
Neg logit contributions | |
confines | -1.482 |
slightest | -1.337 |
bounds | -1.233 |
assumption | -1.212 |
margins | -1.126 |
Token | food |
Token ID | 2057 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
zin | +2.430 |
umat | +2.391 |
fuelled | +2.316 |
andals | +2.300 |
ijing | +2.206 |
Neg logit contributions | |
confines | -2.473 |
slightest | -2.213 |
bounds | -2.021 |
assumption | -1.980 |
margins | -1.829 |
Token | , |
Token ID | 11 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
zin | +1.970 |
umat | +1.943 |
andals | +1.864 |
fuelled | +1.858 |
ijing | +1.782 |
Neg logit contributions | |
confines | -2.292 |
slightest | -2.060 |
bounds | -1.900 |
assumption | -1.867 |
margins | -1.737 |
Token | bigger |
Token ID | 5749 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
confines | +1.020 |
slightest | +0.915 |
bounds | +0.836 |
assumption | +0.821 |
margins | +0.761 |
Neg logit contributions | |
zin | -1.001 |
umat | -0.988 |
andals | -0.951 |
fuelled | -0.939 |
ijing | -0.912 |
Token | plants |
Token ID | 6134 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
confines | +4.468 |
slightest | +3.972 |
bounds | +3.625 |
assumption | +3.539 |
margins | +3.333 |
Neg logit contributions | |
zin | -5.004 |
umat | -4.973 |
andals | -4.790 |
fuelled | -4.659 |
ijing | -4.583 |
Token | and |
Token ID | 290 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
zin | +4.065 |
umat | +3.989 |
fuelled | +3.863 |
andals | +3.809 |
piring | +3.627 |
Neg logit contributions | |
confines | -5.466 |
slightest | -4.967 |
bounds | -4.585 |
assumption | -4.505 |
margins | -4.204 |
Token | thinking |
Token ID | 3612 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
zin | +0.217 |
umat | +0.215 |
andals | +0.208 |
fuelled | +0.208 |
ijing | +0.200 |
Neg logit contributions | |
confines | -0.199 |
slightest | -0.178 |
bounds | -0.161 |
assumption | -0.158 |
margins | -0.145 |
Token | about |
Token ID | 546 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
zin | +4.871 |
umat | +4.782 |
fuelled | +4.647 |
andals | +4.543 |
piring | +4.334 |
Neg logit contributions | |
confines | -7.080 |
slightest | -6.448 |
bounds | -5.953 |
assumption | -5.860 |
margins | -5.454 |
Token | these |
Token ID | 777 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
zin | +3.077 |
umat | +3.025 |
fuelled | +2.939 |
andals | +2.906 |
piring | +2.789 |
Neg logit contributions | |
confines | -3.184 |
slightest | -2.853 |
bounds | -2.599 |
assumption | -2.548 |
margins | -2.347 |
Token | things |
Token ID | 1243 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
zin | +7.770 |
umat | +7.613 |
fuelled | +7.495 |
andals | +7.308 |
piring | +7.041 |
Neg logit contributions | |
confines | -7.744 |
slightest | -6.928 |
bounds | -6.294 |
assumption | -6.171 |
margins | -5.618 |
Token | - |
Token ID | 532 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
zin | +2.849 |
umat | +2.800 |
fuelled | +2.683 |
andals | +2.682 |
ijing | +2.556 |
Neg logit contributions | |
confines | -3.693 |
slightest | -3.349 |
bounds | -3.101 |
assumption | -3.047 |
margins | -2.845 |
Token | there |
Token ID | 612 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
zin | +4.855 |
umat | +4.769 |
fuelled | +4.694 |
andals | +4.575 |
piring | +4.407 |
Neg logit contributions | |
confines | -5.188 |
slightest | -4.650 |
bounds | -4.229 |
assumption | -4.137 |
margins | -3.813 |
Token | were |
Token ID | 547 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
zin | +1.753 |
umat | +1.712 |
fuelled | +1.640 |
andals | +1.633 |
ijing | +1.559 |
Neg logit contributions | |
confines | -2.387 |
slightest | -2.164 |
bounds | -2.014 |
assumption | -1.974 |
margins | -1.852 |
Token | a |
Token ID | 257 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
zin | +3.007 |
umat | +2.950 |
fuelled | +2.869 |
andals | +2.838 |
piring | +2.718 |
Neg logit contributions | |
confines | -3.224 |
slightest | -2.895 |
bounds | -2.643 |
assumption | -2.595 |
margins | -2.397 |
Token | few |
Token ID | 1178 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
zin | +5.489 |
umat | +5.379 |
fuelled | +5.242 |
andals | +5.165 |
piring | +4.957 |
Neg logit contributions | |
confines | -5.769 |
slightest | -5.149 |
bounds | -4.697 |
assumption | -4.602 |
margins | -4.271 |
Token | things |
Token ID | 1243 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
zin | +2.130 |
umat | +2.098 |
fuelled | +2.024 |
andals | +2.017 |
ijing | +1.937 |
Neg logit contributions | |
confines | -2.252 |
slightest | -2.020 |
bounds | -1.851 |
assumption | -1.813 |
margins | -1.681 |
Token | we |
Token ID | 356 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
zin | +1.302 |
umat | +1.282 |
andals | +1.231 |
fuelled | +1.224 |
ijing | +1.178 |
Neg logit contributions | |
confines | -1.559 |
slightest | -1.407 |
bounds | -1.304 |
assumption | -1.275 |
margins | -1.189 |
Token | appeared |
Token ID | 4120 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
zin | +1.085 |
umat | +1.049 |
andals | +0.987 |
fuelled | +0.985 |
ijing | +0.944 |
Neg logit contributions | |
confines | -1.877 |
slightest | -1.719 |
bounds | -1.617 |
assumption | -1.583 |
margins | -1.495 |
Token | on |
Token ID | 319 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
zin | +2.090 |
umat | +2.048 |
andals | +1.942 |
fuelled | +1.941 |
ijing | +1.873 |
Neg logit contributions | |
confines | -2.688 |
slightest | -2.430 |
bounds | -2.268 |
assumption | -2.208 |
margins | -2.067 |
Token | social |
Token ID | 1919 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
confines | +0.557 |
slightest | +0.503 |
bounds | +0.465 |
assumption | +0.458 |
margins | +0.427 |
Neg logit contributions | |
zin | -0.470 |
umat | -0.458 |
fuelled | -0.442 |
andals | -0.436 |
ijing | -0.418 |
Token | media |
Token ID | 2056 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
zin | +1.809 |
umat | +1.783 |
andals | +1.717 |
fuelled | +1.701 |
ijing | +1.646 |
Neg logit contributions | |
confines | -1.662 |
slightest | -1.478 |
bounds | -1.343 |
assumption | -1.326 |
margins | -1.224 |
Token | , |
Token ID | 11 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
zin | +1.661 |
umat | +1.634 |
andals | +1.551 |
fuelled | +1.543 |
ijing | +1.491 |
Neg logit contributions | |
confines | -2.065 |
slightest | -1.854 |
bounds | -1.724 |
assumption | -1.695 |
margins | -1.588 |
Token | administrators |
Token ID | 18618 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
confines | +0.047 |
slightest | +0.043 |
bounds | +0.039 |
assumption | +0.038 |
margins | +0.035 |
Neg logit contributions | |
zin | -0.047 |
umat | -0.046 |
fuelled | -0.044 |
andals | -0.044 |
ijing | -0.042 |
Token | at |
Token ID | 379 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
confines | +6.691 |
slightest | +6.104 |
bounds | +5.693 |
assumption | +5.617 |
margins | +5.211 |
Neg logit contributions | |
zin | -5.151 |
umat | -4.921 |
andals | -4.827 |
utics | -4.623 |
piring | -4.610 |
Token | the |
Token ID | 262 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
confines | +1.076 |
slightest | +0.953 |
bounds | +0.879 |
assumption | +0.858 |
margins | +0.787 |
Neg logit contributions | |
zin | -1.152 |
umat | -1.134 |
fuelled | -1.104 |
andals | -1.092 |
ijing | -1.049 |
Token | college |
Token ID | 4152 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
confines | +2.984 |
slightest | +2.653 |
bounds | +2.433 |
assumption | +2.383 |
margins | +2.196 |
Neg logit contributions | |
zin | -3.039 |
umat | -2.974 |
fuelled | -2.904 |
andals | -2.889 |
ijing | -2.760 |
Token | immediately |
Token ID | 3393 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
confines | +9.700 |
slightest | +8.820 |
bounds | +8.216 |
assumption | +8.182 |
margins | +7.523 |
Neg logit contributions | |
zin | -6.988 |
umat | -6.669 |
andals | -6.480 |
piring | -6.338 |
utics | -6.257 |
Token | sent |
Token ID | 1908 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
confines | +2.879 |
slightest | +2.603 |
bounds | +2.430 |
assumption | +2.384 |
margins | +2.222 |
Neg logit contributions | |
zin | -2.166 |
umat | -2.083 |
andals | -2.031 |
fuelled | -1.965 |
piring | -1.923 |
Token | These |
Token ID | 2312 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
zin | +4.163 |
umat | +4.058 |
fuelled | +3.885 |
andals | +3.843 |
ijing | +3.622 |
Neg logit contributions | |
confines | -7.604 |
slightest | -6.983 |
bounds | -6.519 |
assumption | -6.422 |
margins | -6.056 |
Token | powers |
Token ID | 5635 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
zin | +1.662 |
umat | +1.628 |
fuelled | +1.588 |
andals | +1.567 |
piring | +1.500 |
Neg logit contributions | |
confines | -1.818 |
slightest | -1.635 |
bounds | -1.495 |
assumption | -1.464 |
margins | -1.355 |
Token | are |
Token ID | 389 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
confines | +1.646 |
slightest | +1.475 |
bounds | +1.387 |
assumption | +1.360 |
margins | +1.266 |
Neg logit contributions | |
zin | -1.258 |
umat | -1.227 |
andals | -1.180 |
fuelled | -1.152 |
ijing | -1.126 |
Token | never |
Token ID | 1239 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
confines | +1.806 |
slightest | +1.633 |
bounds | +1.509 |
assumption | +1.473 |
margins | +1.362 |
Neg logit contributions | |
zin | -1.527 |
umat | -1.508 |
andals | -1.445 |
fuelled | -1.399 |
piring | -1.375 |
Token | explained |
Token ID | 4893 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
confines | +6.066 |
slightest | +5.536 |
bounds | +5.080 |
assumption | +4.957 |
ones | +4.570 |
Neg logit contributions | |
zin | -4.921 |
umat | -4.836 |
andals | -4.699 |
utics | -4.440 |
fuelled | -4.413 |
Token | . |
Token ID | 13 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
confines | +8.773 |
slightest | +7.929 |
bounds | +7.390 |
assumption | +7.176 |
margins | +6.735 |
Neg logit contributions | |
zin | -6.869 |
umat | -6.864 |
andals | -6.663 |
fuelled | -6.428 |
piring | -6.344 |
Token |
|
Token ID | 198 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
confines | +3.169 |
slightest | +2.839 |
bounds | +2.736 |
assumption | +2.681 |
margins | +2.522 |
Neg logit contributions | |
zin | -1.645 |
umat | -1.643 |
fuelled | -1.551 |
andals | -1.519 |
ijing | -1.467 |
Token |
|
Token ID | 198 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
confines | +7.278 |
bounds | +6.343 |
slightest | +6.313 |
assumption | +6.094 |
margins | +5.713 |
Neg logit contributions | |
umat | -4.983 |
zin | -4.858 |
andals | -4.740 |
fuelled | -4.733 |
ijing | -4.524 |
Token | Next |
Token ID | 10019 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
confines | +5.014 |
slightest | +4.529 |
bounds | +4.300 |
assumption | +4.236 |
margins | +3.950 |
Neg logit contributions | |
zin | -3.106 |
umat | -3.083 |
fuelled | -3.019 |
andals | -2.942 |
ijing | -2.803 |
Token | Page |
Token ID | 7873 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
confines | +0.138 |
slightest | +0.125 |
bounds | +0.116 |
assumption | +0.115 |
margins | +0.106 |
Neg logit contributions | |
zin | -0.112 |
umat | -0.110 |
andals | -0.106 |
fuelled | -0.106 |
ijing | -0.100 |
Token | — |
Token ID | 851 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
zin | +1.264 |
umat | +1.248 |
fuelled | +1.191 |
andals | +1.171 |
ijing | +1.123 |
Neg logit contributions | |
confines | -1.820 |
slightest | -1.647 |
bounds | -1.535 |
assumption | -1.513 |
margins | -1.423 |
Token | , |
Token ID | 11 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
umat | +1.581 |
zin | +1.580 |
fuelled | +1.520 |
andals | +1.493 |
ijing | +1.416 |
Neg logit contributions | |
confines | -2.516 |
slightest | -2.282 |
bounds | -2.128 |
assumption | -2.099 |
margins | -1.958 |
Token | head |
Token ID | 1182 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
confines | +2.236 |
slightest | +1.978 |
bounds | +1.831 |
assumption | +1.800 |
margins | +1.639 |
Neg logit contributions | |
zin | -2.384 |
umat | -2.341 |
andals | -2.275 |
fuelled | -2.244 |
piring | -2.154 |
Token | of |
Token ID | 286 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
confines | +1.140 |
slightest | +1.033 |
bounds | +0.964 |
assumption | +0.947 |
margins | +0.885 |
Neg logit contributions | |
zin | -0.814 |
umat | -0.802 |
fuelled | -0.769 |
andals | -0.767 |
ijing | -0.725 |
Token | Nuclear |
Token ID | 19229 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
zin | +1.841 |
umat | +1.808 |
fuelled | +1.750 |
andals | +1.741 |
piring | +1.665 |
Neg logit contributions | |
confines | -1.967 |
slightest | -1.762 |
bounds | -1.620 |
assumption | -1.588 |
margins | -1.469 |
Token | Engineering |
Token ID | 14044 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
zin | +2.458 |
umat | +2.408 |
fuelled | +2.325 |
andals | +2.316 |
piring | +2.219 |
Neg logit contributions | |
confines | -2.974 |
slightest | -2.671 |
bounds | -2.471 |
assumption | -2.431 |
margins | -2.266 |
Token | at |
Token ID | 379 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
zin | +1.846 |
umat | +1.805 |
fuelled | +1.739 |
andals | +1.736 |
piring | +1.661 |
Neg logit contributions | |
confines | -2.341 |
slightest | -2.106 |
bounds | -1.960 |
assumption | -1.927 |
margins | -1.799 |
Token | the |
Token ID | 262 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
confines | +1.362 |
slightest | +1.207 |
bounds | +1.113 |
assumption | +1.097 |
margins | +1.011 |
Neg logit contributions | |
zin | -1.359 |
umat | -1.339 |
fuelled | -1.294 |
andals | -1.283 |
piring | -1.233 |
Token | Poly |
Token ID | 12280 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
confines | +3.567 |
slightest | +3.162 |
bounds | +2.888 |
assumption | +2.880 |
margins | +2.634 |
Neg logit contributions | |
zin | -3.645 |
umat | -3.566 |
andals | -3.471 |
fuelled | -3.462 |
piring | -3.325 |
Token | techn |
Token ID | 23873 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
confines | +0.559 |
slightest | +0.508 |
bounds | +0.474 |
assumption | +0.467 |
margins | +0.440 |
Neg logit contributions | |
umat | -0.355 |
zin | -0.354 |
andals | -0.339 |
fuelled | -0.338 |
piring | -0.316 |
Token | ic |
Token ID | 291 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
confines | +4.244 |
slightest | +3.771 |
bounds | +3.534 |
assumption | +3.382 |
margins | +3.098 |
Neg logit contributions | |
zin | -3.953 |
fuelled | -3.924 |
umat | -3.858 |
andals | -3.811 |
piring | -3.660 |
Token | University |
Token ID | 2059 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
zin | +0.224 |
umat | +0.220 |
fuelled | +0.215 |
andals | +0.214 |
piring | +0.207 |
Neg logit contributions | |
confines | -0.188 |
slightest | -0.163 |
bounds | -0.150 |
assumption | -0.146 |
margins | -0.132 |
Token | upset |
Token ID | 9247 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
zin | +3.687 |
umat | +3.619 |
fuelled | +3.526 |
andals | +3.462 |
piring | +3.322 |
Neg logit contributions | |
confines | -4.467 |
slightest | -4.030 |
bounds | -3.716 |
assumption | -3.643 |
margins | -3.390 |
Token | about |
Token ID | 546 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
zin | +0.613 |
umat | +0.603 |
andals | +0.576 |
fuelled | +0.571 |
ijing | +0.547 |
Neg logit contributions | |
confines | -0.807 |
slightest | -0.729 |
bounds | -0.678 |
assumption | -0.666 |
margins | -0.624 |
Token | this |
Token ID | 428 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
zin | +2.571 |
umat | +2.528 |
fuelled | +2.446 |
andals | +2.429 |
ijing | +2.326 |
Neg logit contributions | |
confines | -2.777 |
slightest | -2.493 |
bounds | -2.285 |
assumption | -2.240 |
margins | -2.075 |
Token | flyer |
Token ID | 42464 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
zin | +8.129 |
umat | +7.939 |
fuelled | +7.875 |
andals | +7.707 |
piring | +7.419 |
Neg logit contributions | |
confines | -8.365 |
slightest | -7.432 |
bounds | -6.872 |
assumption | -6.575 |
margins | -6.199 |
Token | ," |
Token ID | 553 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
confines | +0.780 |
slightest | +0.710 |
bounds | +0.656 |
assumption | +0.651 |
margins | +0.603 |
Neg logit contributions | |
zin | -0.574 |
umat | -0.566 |
andals | -0.538 |
fuelled | -0.528 |
ijing | -0.514 |
Token | Hobby |
Token ID | 44327 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
confines | +2.519 |
slightest | +2.247 |
bounds | +2.103 |
assumption | +2.039 |
margins | +1.901 |
Neg logit contributions | |
zin | -2.094 |
umat | -2.047 |
andals | -1.966 |
fuelled | -1.949 |
ijing | -1.892 |
Token | said |
Token ID | 531 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
confines | +8.039 |
slightest | +6.937 |
bounds | +6.694 |
assumption | +6.330 |
margins | +6.007 |
Neg logit contributions | |
zin | -6.270 |
umat | -6.247 |
andals | -6.091 |
utics | -5.863 |
ijing | -5.851 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
confines | +2.068 |
slightest | +1.849 |
bounds | +1.751 |
assumption | +1.712 |
margins | +1.608 |
Neg logit contributions | |
umat | -1.331 |
zin | -1.327 |
andals | -1.251 |
fuelled | -1.234 |
piring | -1.200 |
Token | " |
Token ID | 366 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
zin | +3.056 |
umat | +2.976 |
fuelled | +2.862 |
andals | +2.836 |
piring | +2.642 |
Neg logit contributions | |
confines | -5.425 |
slightest | -4.996 |
bounds | -4.638 |
assumption | -4.578 |
margins | -4.307 |
Token | They |
Token ID | 2990 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
zin | +3.654 |
umat | +3.557 |
fuelled | +3.374 |
andals | +3.366 |
piring | +3.126 |
Neg logit contributions | |
confines | -8.367 |
slightest | -7.789 |
bounds | -7.235 |
assumption | -7.133 |
margins | -6.762 |
Token | don |
Token ID | 836 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
zin | +4.561 |
umat | +4.489 |
fuelled | +4.396 |
andals | +4.262 |
ijing | +4.074 |
Neg logit contributions | |
confines | -5.768 |
slightest | -5.212 |
bounds | -4.801 |
assumption | -4.700 |
margins | -4.396 |
Token | a |
Token ID | 257 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
confines | +0.051 |
slightest | +0.046 |
bounds | +0.043 |
assumption | +0.042 |
margins | +0.039 |
Neg logit contributions | |
zin | -0.043 |
umat | -0.042 |
andals | -0.041 |
fuelled | -0.041 |
piring | -0.039 |
Token | little |
Token ID | 1310 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
umat | +0.810 |
zin | +0.810 |
andals | +0.781 |
fuelled | +0.779 |
ijing | +0.744 |
Neg logit contributions | |
confines | -0.902 |
slightest | -0.815 |
bounds | -0.751 |
assumption | -0.738 |
margins | -0.681 |
Token | bit |
Token ID | 1643 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
confines | +3.524 |
slightest | +3.153 |
bounds | +2.867 |
assumption | +2.789 |
margins | +2.599 |
Neg logit contributions | |
umat | -4.250 |
zin | -4.216 |
andals | -4.131 |
fuelled | -4.053 |
piring | -3.905 |
Token | , |
Token ID | 11 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
zin | +0.108 |
umat | +0.108 |
andals | +0.104 |
fuelled | +0.103 |
ijing | +0.098 |
Neg logit contributions | |
confines | -0.115 |
slightest | -0.104 |
bounds | -0.095 |
assumption | -0.093 |
margins | -0.087 |
Token | and |
Token ID | 290 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
confines | +1.187 |
slightest | +1.060 |
bounds | +0.973 |
assumption | +0.951 |
margins | +0.881 |
Neg logit contributions | |
zin | -1.160 |
umat | -1.147 |
andals | -1.105 |
fuelled | -1.101 |
piring | -1.054 |
Token | then |
Token ID | 788 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
confines | +0.692 |
slightest | +0.623 |
bounds | +0.575 |
assumption | +0.564 |
margins | +0.526 |
Neg logit contributions | |
zin | -0.567 |
umat | -0.559 |
fuelled | -0.537 |
andals | -0.535 |
piring | -0.510 |
Token | she |
Token ID | 673 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
confines | +2.746 |
slightest | +2.458 |
bounds | +2.281 |
assumption | +2.223 |
margins | +2.086 |
Neg logit contributions | |
zin | -2.264 |
umat | -2.258 |
andals | -2.149 |
fuelled | -2.145 |
piring | -2.046 |
Token | can |
Token ID | 460 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
confines | +0.953 |
slightest | +0.855 |
bounds | +0.789 |
assumption | +0.767 |
margins | +0.716 |
Neg logit contributions | |
zin | -0.837 |
umat | -0.825 |
fuelled | -0.799 |
andals | -0.793 |
piring | -0.762 |
Token | also |
Token ID | 635 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
confines | +1.951 |
slightest | +1.737 |
bounds | +1.592 |
assumption | +1.556 |
margins | +1.437 |
Neg logit contributions | |
zin | -1.982 |
umat | -1.962 |
fuelled | -1.898 |
andals | -1.887 |
piring | -1.811 |
Token | seamlessly |
Token ID | 33681 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
confines | +3.599 |
slightest | +3.199 |
bounds | +2.936 |
assumption | +2.875 |
margins | +2.648 |
Neg logit contributions | |
zin | -3.632 |
umat | -3.612 |
fuelled | -3.479 |
andals | -3.465 |
piring | -3.319 |
Token | weave |
Token ID | 37982 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
confines | +2.494 |
slightest | +2.217 |
bounds | +2.046 |
assumption | +2.004 |
margins | +1.854 |
Neg logit contributions | |
zin | -2.388 |
umat | -2.370 |
fuelled | -2.283 |
andals | -2.277 |
ijing | -2.180 |
Token | as |
Token ID | 355 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
zin | +3.426 |
umat | +3.316 |
fuelled | +3.231 |
andals | +3.192 |
ijing | +3.085 |
Neg logit contributions | |
confines | -3.848 |
slightest | -3.469 |
bounds | -3.158 |
assumption | -3.092 |
margins | -2.853 |
Token | I |
Token ID | 314 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
zin | +1.470 |
umat | +1.443 |
fuelled | +1.400 |
andals | +1.384 |
ijing | +1.324 |
Neg logit contributions | |
confines | -1.758 |
slightest | -1.586 |
bounds | -1.459 |
assumption | -1.430 |
margins | -1.331 |
Token | understand |
Token ID | 1833 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
zin | +2.848 |
umat | +2.800 |
fuelled | +2.707 |
andals | +2.678 |
ijing | +2.561 |
Neg logit contributions | |
confines | -3.394 |
slightest | -3.059 |
bounds | -2.818 |
assumption | -2.763 |
margins | -2.578 |
Token | , |
Token ID | 11 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
confines | +6.439 |
slightest | +5.768 |
bounds | +5.338 |
assumption | +5.333 |
margins | +4.963 |
Neg logit contributions | |
umat | -5.420 |
zin | -5.398 |
andals | -5.221 |
fuelled | -5.097 |
piring | -4.925 |
Token | the |
Token ID | 262 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
confines | +1.723 |
slightest | +1.544 |
bounds | +1.417 |
assumption | +1.398 |
margins | +1.287 |
Neg logit contributions | |
zin | -1.656 |
umat | -1.649 |
andals | -1.572 |
fuelled | -1.564 |
ijing | -1.498 |
Token | Ministry |
Token ID | 9475 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
confines | +4.037 |
slightest | +3.635 |
bounds | +3.278 |
assumption | +3.255 |
margins | +2.956 |
Neg logit contributions | |
zin | -4.247 |
umat | -4.191 |
fuelled | -4.043 |
andals | -4.025 |
utics | -3.861 |
Token | of |
Token ID | 286 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
confines | +10.571 |
slightest | +9.469 |
bounds | +8.753 |
assumption | +8.738 |
margins | +7.968 |
Neg logit contributions | |
zin | -9.051 |
umat | -9.004 |
andals | -8.641 |
fuelled | -8.473 |
piring | -8.352 |
Token | State |
Token ID | 1812 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
zin | +0.584 |
umat | +0.573 |
fuelled | +0.555 |
andals | +0.549 |
piring | +0.525 |
Neg logit contributions | |
confines | -0.635 |
slightest | -0.570 |
bounds | -0.525 |
assumption | -0.515 |
margins | -0.477 |
Token | Security |
Token ID | 4765 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
confines | +3.501 |
slightest | +3.129 |
bounds | +2.863 |
assumption | +2.841 |
margins | +2.638 |
Neg logit contributions | |
zin | -3.375 |
umat | -3.336 |
andals | -3.208 |
fuelled | -3.143 |
ijing | -3.053 |
Token | would |
Token ID | 561 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
confines | +5.573 |
slightest | +5.003 |
bounds | +4.648 |
assumption | +4.586 |
margins | +4.241 |
Neg logit contributions | |
zin | -4.395 |
umat | -4.297 |
andals | -4.163 |
fuelled | -4.076 |
piring | -3.975 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
confines | +5.467 |
slightest | +4.889 |
bounds | +4.491 |
assumption | +4.443 |
margins | +4.014 |
Neg logit contributions | |
zin | -5.468 |
umat | -5.392 |
andals | -5.249 |
fuelled | -5.172 |
ijing | -5.024 |
Token | just |
Token ID | 655 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
zin | +1.428 |
umat | +1.404 |
fuelled | +1.357 |
andals | +1.338 |
ijing | +1.276 |
Neg logit contributions | |
confines | -1.864 |
slightest | -1.690 |
bounds | -1.559 |
assumption | -1.533 |
margins | -1.430 |
Token | took |
Token ID | 1718 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
confines | +2.855 |
slightest | +2.577 |
bounds | +2.404 |
assumption | +2.356 |
margins | +2.213 |
Neg logit contributions | |
zin | -2.073 |
umat | -2.028 |
andals | -1.961 |
fuelled | -1.926 |
ijing | -1.860 |
Token | away |
Token ID | 1497 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
confines | +2.914 |
slightest | +2.585 |
bounds | +2.413 |
assumption | +2.362 |
margins | +2.180 |
Neg logit contributions | |
zin | -2.936 |
umat | -2.875 |
fuelled | -2.783 |
andals | -2.772 |
ijing | -2.670 |
Token | the |
Token ID | 262 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
confines | +7.797 |
slightest | +6.817 |
bounds | +6.486 |
assumption | +6.381 |
margins | +5.828 |
Neg logit contributions | |
zin | -8.364 |
umat | -7.965 |
andals | -7.826 |
fuelled | -7.724 |
ijing | -7.547 |
Token | 1 |
Token ID | 352 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
confines | +9.011 |
slightest | +8.066 |
assumption | +7.477 |
bounds | +7.369 |
margins | +6.648 |
Neg logit contributions | |
zin | -9.603 |
umat | -9.264 |
andals | -9.026 |
fuelled | -8.990 |
ijing | -8.736 |
Token | mana |
Token ID | 13149 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
confines | +5.505 |
slightest | +4.846 |
bounds | +4.474 |
assumption | +4.463 |
margins | +3.893 |
Neg logit contributions | |
zin | -6.497 |
umat | -6.363 |
andals | -6.262 |
fuelled | -6.211 |
ijing | -6.012 |
Token | ability |
Token ID | 2694 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
confines | +10.412 |
slightest | +9.111 |
assumption | +8.814 |
bounds | +8.777 |
margins | +7.952 |
Neg logit contributions | |
zin | -9.101 |
umat | -8.914 |
andals | -8.620 |
fuelled | -8.248 |
piring | -8.135 |
Token | on |
Token ID | 319 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
confines | +9.437 |
slightest | +8.420 |
bounds | +8.120 |
assumption | +8.069 |
margins | +7.412 |
Neg logit contributions | |
zin | -7.555 |
umat | -7.455 |
andals | -7.212 |
fuelled | -6.880 |
ijing | -6.862 |
Token | that |
Token ID | 326 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
confines | +5.391 |
slightest | +4.755 |
bounds | +4.404 |
assumption | +4.375 |
margins | +3.942 |
Neg logit contributions | |
zin | -6.096 |
umat | -5.868 |
andals | -5.755 |
fuelled | -5.694 |
ijing | -5.533 |
Token | potion |
Token ID | 26572 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
zin | +0.188 |
umat | +0.185 |
fuelled | +0.179 |
andals | +0.178 |
ijing | +0.172 |
Neg logit contributions | |
confines | -0.181 |
slightest | -0.162 |
bounds | -0.147 |
assumption | -0.146 |
margins | -0.133 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
zin | +1.156 |
umat | +1.133 |
fuelled | +1.088 |
andals | +1.087 |
ijing | +1.034 |
Neg logit contributions | |
confines | -1.508 |
slightest | -1.367 |
bounds | -1.266 |
assumption | -1.246 |
margins | -1.161 |
Token | . |
Token ID | 13 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
confines | +0.281 |
slightest | +0.253 |
bounds | +0.241 |
assumption | +0.233 |
margins | +0.219 |
Neg logit contributions | |
zin | -0.172 |
umat | -0.172 |
andals | -0.166 |
fuelled | -0.164 |
ijing | -0.158 |
Token | Fro |
Token ID | 9734 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
zin | +3.194 |
umat | +3.131 |
fuelled | +3.013 |
andals | +2.985 |
ijing | +2.842 |
Neg logit contributions | |
confines | -4.734 |
slightest | -4.301 |
bounds | -4.006 |
assumption | -3.935 |
margins | -3.689 |
Token | do |
Token ID | 4598 |
Feature activation | -9.64e-01 |
Pos logit contributions | |
confines | +4.043 |
slightest | +3.596 |
bounds | +3.446 |
assumption | +3.354 |
margins | +3.048 |
Neg logit contributions | |
umat | -3.293 |
zin | -3.234 |
fuelled | -3.186 |
andals | -3.171 |
ijing | -3.089 |
Token | falls |
Token ID | 8953 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
confines | +4.347 |
slightest | +3.812 |
bounds | +3.609 |
assumption | +3.452 |
margins | +3.204 |
Neg logit contributions | |
zin | -3.718 |
umat | -3.662 |
andals | -3.532 |
fuelled | -3.500 |
ijing | -3.437 |
Token | to |
Token ID | 284 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
confines | +3.118 |
slightest | +2.827 |
bounds | +2.656 |
assumption | +2.561 |
margins | +2.370 |
Neg logit contributions | |
zin | -2.380 |
umat | -2.320 |
andals | -2.222 |
fuelled | -2.208 |
ijing | -2.166 |
Token | Shel |
Token ID | 15325 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
confines | +1.645 |
slightest | +1.462 |
bounds | +1.354 |
assumption | +1.320 |
margins | +1.206 |
Neg logit contributions | |
zin | -1.640 |
umat | -1.604 |
fuelled | -1.569 |
andals | -1.556 |
ijing | -1.515 |
Token | ob |
Token ID | 672 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
confines | +11.371 |
bounds | +10.039 |
slightest | +9.756 |
assumption | +9.251 |
ones | +8.669 |
Neg logit contributions | |
fuelled | -8.765 |
andals | -8.554 |
zin | -8.511 |
umat | -8.456 |
ijing | -8.343 |
Token | 's |
Token ID | 338 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
confines | +6.341 |
slightest | +5.479 |
bounds | +5.247 |
assumption | +5.010 |
margins | +4.672 |
Neg logit contributions | |
zin | -5.596 |
umat | -5.559 |
andals | -5.413 |
fuelled | -5.326 |
ijing | -5.251 |
Token | sting |
Token ID | 24276 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
confines | +4.501 |
slightest | +3.984 |
bounds | +3.652 |
assumption | +3.575 |
margins | +3.254 |
Neg logit contributions | |
zin | -4.668 |
umat | -4.491 |
fuelled | -4.452 |
andals | -4.432 |
ijing | -4.301 |
Token | . |
Token ID | 13 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
confines | +0.979 |
slightest | +0.883 |
bounds | +0.814 |
assumption | +0.801 |
margins | +0.750 |
Neg logit contributions | |
zin | -0.754 |
umat | -0.753 |
andals | -0.729 |
fuelled | -0.717 |
ijing | -0.694 |
Token | But |
Token ID | 887 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
zin | +2.453 |
umat | +2.406 |
fuelled | +2.316 |
andals | +2.292 |
ijing | +2.193 |
Neg logit contributions | |
confines | -3.748 |
slightest | -3.393 |
bounds | -3.179 |
assumption | -3.119 |
margins | -2.928 |
Token | within |
Token ID | 1626 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
confines | +5.713 |
slightest | +5.166 |
bounds | +4.806 |
assumption | +4.675 |
margins | +4.377 |
Neg logit contributions | |
zin | -4.618 |
umat | -4.601 |
fuelled | -4.370 |
andals | -4.363 |
ijing | -4.177 |
Token | the |
Token ID | 262 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
confines | +5.157 |
slightest | +4.519 |
bounds | +4.240 |
assumption | +4.075 |
margins | +3.805 |
Neg logit contributions | |
zin | -5.061 |
umat | -5.018 |
fuelled | -4.829 |
andals | -4.773 |
ijing | -4.592 |
Token | core |
Token ID | 4755 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
confines | +7.803 |
slightest | +6.884 |
bounds | +6.397 |
assumption | +6.237 |
margins | +5.778 |
Neg logit contributions | |
zin | -7.167 |
umat | -7.072 |
fuelled | -6.812 |
andals | -6.810 |
ijing | -6.516 |
Token | library |
Token ID | 5888 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
confines | +8.614 |
slightest | +7.444 |
bounds | +7.103 |
assumption | +7.002 |
margins | +6.399 |
Neg logit contributions | |
zin | -7.081 |
umat | -7.014 |
fuelled | -6.820 |
andals | -6.644 |
ijing | -6.416 |
Token | , |
Token ID | 11 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
confines | +4.892 |
slightest | +4.338 |
bounds | +4.123 |
assumption | +4.048 |
margins | +3.783 |
Neg logit contributions | |
zin | -3.423 |
umat | -3.398 |
andals | -3.241 |
fuelled | -3.236 |
piring | -3.089 |
Token | without |
Token ID | 1231 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
confines | +1.891 |
slightest | +1.693 |
bounds | +1.558 |
assumption | +1.524 |
margins | +1.407 |
Neg logit contributions | |
zin | -1.847 |
umat | -1.824 |
andals | -1.759 |
fuelled | -1.746 |
ijing | -1.686 |
Token | much |
Token ID | 881 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
confines | +9.689 |
slightest | +8.930 |
bounds | +8.168 |
assumption | +7.967 |
margins | +7.189 |
Neg logit contributions | |
umat | -10.008 |
zin | -9.998 |
andals | -9.539 |
fuelled | -9.491 |
ijing | -9.208 |
Token | in |
Token ID | 287 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
confines | +0.906 |
slightest | +0.811 |
bounds | +0.738 |
assumption | +0.725 |
margins | +0.661 |
Neg logit contributions | |
zin | -0.987 |
umat | -0.975 |
fuelled | -0.941 |
andals | -0.940 |
ijing | -0.906 |
Token | the |
Token ID | 262 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
confines | +6.978 |
slightest | +6.151 |
bounds | +5.806 |
assumption | +5.545 |
margins | +5.099 |
Neg logit contributions | |
umat | -8.027 |
zin | -7.968 |
fuelled | -7.669 |
andals | -7.525 |
ilst | -7.308 |
Token | way |
Token ID | 835 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
confines | +3.786 |
slightest | +3.179 |
bounds | +2.584 |
assumption | +2.388 |
margins | +2.032 |
Neg logit contributions | |
umat | -10.164 |
zin | -10.142 |
fuelled | -9.749 |
andals | -9.713 |
piring | -9.464 |
Token | of |
Token ID | 286 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
confines | +6.781 |
slightest | +6.007 |
bounds | +5.764 |
assumption | +5.621 |
margins | +5.256 |
Neg logit contributions | |
umat | -5.908 |
zin | -5.830 |
andals | -5.568 |
fuelled | -5.545 |
ijing | -5.370 |
Token | case |
Token ID | 1339 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
confines | +3.346 |
slightest | +2.902 |
bounds | +2.669 |
assumption | +2.606 |
margins | +2.392 |
Neg logit contributions | |
zin | -4.131 |
umat | -4.084 |
fuelled | -3.957 |
andals | -3.954 |
piring | -3.830 |
Token | (" |
Token ID | 5855 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
confines | +3.649 |
slightest | +3.133 |
bounds | +3.020 |
assumption | +3.003 |
margins | +2.722 |
Neg logit contributions | |
umat | -3.619 |
zin | -3.613 |
andals | -3.521 |
fuelled | -3.470 |
piring | -3.451 |
Token | Snap |
Token ID | 43826 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
confines | +0.369 |
slightest | +0.336 |
bounds | +0.314 |
assumption | +0.309 |
margins | +0.289 |
Neg logit contributions | |
zin | -0.243 |
umat | -0.240 |
fuelled | -0.233 |
andals | -0.230 |
ijing | -0.218 |
Token | "), |
Token ID | 12340 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
confines | +0.558 |
slightest | +0.504 |
bounds | +0.467 |
assumption | +0.459 |
margins | +0.428 |
Neg logit contributions | |
zin | -0.429 |
umat | -0.426 |
fuelled | -0.413 |
andals | -0.410 |
ijing | -0.388 |
Token | the |
Token ID | 262 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
confines | +6.977 |
slightest | +6.227 |
bounds | +5.821 |
assumption | +5.764 |
margins | +5.292 |
Neg logit contributions | |
zin | -5.949 |
umat | -5.887 |
andals | -5.677 |
fuelled | -5.590 |
piring | -5.436 |
Token | command |
Token ID | 3141 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
confines | +5.343 |
slightest | +4.706 |
bounds | +4.264 |
assumption | +4.260 |
margins | +3.799 |
Neg logit contributions | |
zin | -6.135 |
umat | -6.112 |
andals | -5.955 |
fuelled | -5.932 |
piring | -5.752 |
Token | like |
Token ID | 588 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
confines | +8.610 |
slightest | +7.291 |
bounds | +6.753 |
assumption | +6.697 |
margins | +6.016 |
Neg logit contributions | |
umat | -12.122 |
zin | -12.012 |
andals | -11.813 |
fuelled | -11.387 |
piring | -11.365 |
Token | utility |
Token ID | 10361 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
confines | +5.641 |
slightest | +5.006 |
assumption | +4.604 |
bounds | +4.602 |
margins | +4.220 |
Neg logit contributions | |
zin | -5.223 |
umat | -5.182 |
fuelled | -5.055 |
andals | -5.004 |
piring | -4.771 |
Token | will |
Token ID | 481 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
confines | +2.683 |
slightest | +2.381 |
bounds | +2.209 |
assumption | +2.172 |
margins | +2.014 |
Neg logit contributions | |
zin | -2.431 |
umat | -2.400 |
andals | -2.324 |
fuelled | -2.312 |
ijing | -2.200 |
Token | be |
Token ID | 307 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
confines | +6.163 |
slightest | +5.302 |
bounds | +4.925 |
assumption | +4.878 |
margins | +4.308 |
Neg logit contributions | |
zin | -7.728 |
umat | -7.634 |
andals | -7.420 |
fuelled | -7.380 |
ijing | -7.105 |
Token | referenced |
Token ID | 20717 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
confines | +5.492 |
slightest | +4.942 |
bounds | +4.599 |
assumption | +4.537 |
margins | +4.188 |
Neg logit contributions | |
zin | -4.301 |
umat | -4.196 |
andals | -4.062 |
fuelled | -3.943 |
piring | -3.845 |