Token | Jeff |
Token ID | 19139 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
rament | +6.214 |
cott | +6.064 |
utenberg | +6.059 |
peed | +5.908 |
annah | +5.760 |
Neg logit contributions | |
Jinping | -7.422 |
NOT | -7.101 |
Statements | -7.064 |
fore | -6.597 |
annot | -6.448 |
Token | Flake |
Token ID | 39727 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
rament | +4.850 |
utenberg | +4.735 |
cott | +4.498 |
peed | +4.405 |
annah | +4.252 |
Neg logit contributions | |
Jinping | -6.515 |
Statements | -6.266 |
NOT | -6.184 |
fore | -5.873 |
annot | -5.696 |
Token | ( |
Token ID | 7 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
utenberg | +1.977 |
rament | +1.961 |
peed | +1.850 |
cott | +1.839 |
annah | +1.752 |
Neg logit contributions | |
Jinping | -2.192 |
Statements | -2.114 |
NOT | -2.056 |
fore | -1.967 |
't | -1.931 |
Token | R |
Token ID | 49 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
rament | +1.444 |
utenberg | +1.442 |
peed | +1.349 |
cott | +1.349 |
ranging | +1.285 |
Neg logit contributions | |
Jinping | -2.095 |
Statements | -2.009 |
NOT | -1.936 |
fore | -1.871 |
't | -1.844 |
Token | - |
Token ID | 12 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
utenberg | +5.946 |
rament | +5.854 |
cott | +5.481 |
peed | +5.479 |
annah | +5.365 |
Neg logit contributions | |
Jinping | -8.531 |
Statements | -8.172 |
NOT | -7.810 |
fore | -7.495 |
annot | -7.389 |
Token | Ari |
Token ID | 26529 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
rament | +5.660 |
cott | +5.473 |
utenberg | +5.314 |
annah | +5.186 |
ranging | +5.175 |
Neg logit contributions | |
Jinping | -7.712 |
Statements | -7.238 |
fore | -6.459 |
annot | -6.451 |
NOT | -6.410 |
Token | z |
Token ID | 89 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
rament | +4.399 |
annah | +3.856 |
pan | +3.727 |
peed | +3.659 |
cott | +3.300 |
Neg logit contributions | |
Statements | -24.523 |
Jinping | -23.514 |
NOT | -23.403 |
fore | -23.331 |
Tycoon | -22.818 |
Token | .) |
Token ID | 2014 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
rament | +7.206 |
utenberg | +7.190 |
cott | +7.088 |
peed | +6.869 |
annah | +6.738 |
Neg logit contributions | |
Jinping | -14.081 |
Statements | -13.813 |
NOT | -13.120 |
fore | -12.561 |
annot | -12.461 |
Token | after |
Token ID | 706 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
utenberg | +1.298 |
rament | +1.293 |
peed | +1.231 |
cott | +1.214 |
ranging | +1.170 |
Neg logit contributions | |
Jinping | -1.234 |
Statements | -1.180 |
NOT | -1.144 |
fore | -1.094 |
't | -1.073 |
Token | he |
Token ID | 339 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
Jinping | +2.000 |
Statements | +1.920 |
NOT | +1.846 |
't | +1.772 |
fore | +1.772 |
Neg logit contributions | |
rament | -1.939 |
utenberg | -1.930 |
peed | -1.827 |
cott | -1.787 |
"$:/ | -1.747 |
Token | was |
Token ID | 373 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
utenberg | +0.447 |
rament | +0.447 |
peed | +0.422 |
cott | +0.418 |
ranging | +0.400 |
Neg logit contributions | |
Jinping | -0.501 |
Statements | -0.482 |
NOT | -0.468 |
fore | -0.449 |
't | -0.439 |
Token | that |
Token ID | 326 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
utenberg | +0.739 |
rament | +0.739 |
peed | +0.704 |
cott | +0.695 |
ranging | +0.675 |
Neg logit contributions | |
Jinping | -0.540 |
Statements | -0.510 |
NOT | -0.494 |
fore | -0.470 |
't | -0.461 |
Token | there |
Token ID | 612 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
rament | +2.287 |
utenberg | +2.258 |
peed | +2.166 |
cott | +2.130 |
"$:/ | +2.065 |
Neg logit contributions | |
Jinping | -2.562 |
Statements | -2.438 |
NOT | -2.376 |
fore | -2.281 |
annot | -2.220 |
Token | was |
Token ID | 373 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
rament | +1.967 |
utenberg | +1.965 |
peed | +1.876 |
cott | +1.866 |
ranging | +1.791 |
Neg logit contributions | |
Jinping | -1.680 |
Statements | -1.597 |
NOT | -1.538 |
fore | -1.451 |
't | -1.424 |
Token | a |
Token ID | 257 |
Feature activation | -6.09e-03 |
Pos logit contributions | |
rament | +2.394 |
utenberg | +2.385 |
peed | +2.268 |
cott | +2.234 |
ranging | +2.158 |
Neg logit contributions | |
Jinping | -2.512 |
Statements | -2.397 |
NOT | -2.310 |
fore | -2.219 |
't | -2.161 |
Token | " |
Token ID | 366 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
Jinping | +0.041 |
Statements | +0.039 |
NOT | +0.038 |
fore | +0.037 |
't | +0.036 |
Neg logit contributions | |
utenberg | -0.037 |
rament | -0.037 |
peed | -0.035 |
cott | -0.035 |
ranging | -0.033 |
Token | free |
Token ID | 5787 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
Jinping | +3.080 |
NOT | +2.989 |
Statements | +2.958 |
fore | +2.893 |
't | +2.799 |
Neg logit contributions | |
utenberg | -3.072 |
rament | -3.017 |
peed | -2.830 |
"$:/ | -2.725 |
cott | -2.720 |
Token | speech |
Token ID | 4046 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
cott | +11.128 |
acre | +10.906 |
rament | +10.767 |
minded | +10.760 |
utenberg | +10.715 |
Neg logit contributions | |
Jinping | -16.534 |
Statements | -15.976 |
NOT | -15.325 |
fore | -14.285 |
Tycoon | -14.190 |
Token | zone |
Token ID | 6516 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
rament | +4.422 |
utenberg | +4.319 |
cott | +4.005 |
peed | +3.976 |
acre | +3.670 |
Neg logit contributions | |
Jinping | -10.367 |
Statements | -10.207 |
NOT | -9.772 |
fore | -9.575 |
annot | -9.357 |
Token | " |
Token ID | 1 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
rament | +5.664 |
utenberg | +5.575 |
peed | +5.438 |
cott | +5.292 |
"$:/ | +5.124 |
Neg logit contributions | |
Jinping | -8.913 |
Statements | -8.785 |
NOT | -8.338 |
fore | -7.998 |
annot | -7.960 |
Token | at |
Token ID | 379 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
rament | +5.878 |
utenberg | +5.865 |
peed | +5.702 |
cott | +5.531 |
"$:/ | +5.408 |
Neg logit contributions | |
Jinping | -7.143 |
Statements | -6.957 |
NOT | -6.590 |
fore | -6.313 |
annot | -6.272 |
Token | the |
Token ID | 262 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
Jinping | +0.076 |
Statements | +0.072 |
NOT | +0.070 |
fore | +0.068 |
't | +0.067 |
Neg logit contributions | |
utenberg | -0.081 |
rament | -0.081 |
peed | -0.076 |
cott | -0.076 |
ranging | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
rament | +3.467 |
utenberg | +3.434 |
peed | +3.297 |
cott | +3.245 |
ranging | +3.158 |
Neg logit contributions | |
Jinping | -4.122 |
Statements | -3.979 |
NOT | -3.905 |
annot | -3.678 |
fore | -3.671 |
Token | seismic |
Token ID | 37463 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
utenberg | +1.864 |
rament | +1.861 |
peed | +1.770 |
cott | +1.750 |
ranging | +1.694 |
Neg logit contributions | |
Jinping | -1.618 |
Statements | -1.530 |
NOT | -1.496 |
fore | -1.416 |
't | -1.388 |
Token | eruption |
Token ID | 36176 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
rament | +1.531 |
utenberg | +1.529 |
peed | +1.452 |
cott | +1.425 |
ranging | +1.379 |
Neg logit contributions | |
Jinping | -1.592 |
Statements | -1.522 |
NOT | -1.484 |
fore | -1.419 |
't | -1.390 |
Token | that |
Token ID | 326 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
rament | +2.247 |
utenberg | +2.229 |
peed | +2.138 |
cott | +2.095 |
ranging | +2.037 |
Neg logit contributions | |
Jinping | -2.113 |
Statements | -2.027 |
NOT | -1.970 |
fore | -1.869 |
't | -1.831 |
Token | would |
Token ID | 561 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
rament | +6.020 |
utenberg | +5.851 |
cott | +5.660 |
peed | +5.567 |
ranging | +5.367 |
Neg logit contributions | |
Jinping | -6.172 |
Statements | -6.042 |
NOT | -6.009 |
fore | -5.551 |
annot | -5.521 |
Token | have |
Token ID | 423 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
peed | +9.379 |
utenberg | +9.276 |
rament | +9.241 |
"$:/ | +8.849 |
annah | +8.578 |
Neg logit contributions | |
Statements | -13.268 |
Jinping | -13.025 |
NOT | -12.154 |
annot | -11.979 |
fore | -11.823 |
Token | occurred |
Token ID | 5091 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
rament | +10.705 |
encomp | +10.684 |
minded | +10.624 |
peed | +10.524 |
ensued | +10.411 |
Neg logit contributions | |
Statements | -16.417 |
NOT | -15.443 |
Jinping | -15.056 |
annot | -14.535 |
Tycoon | -14.227 |
Token | if |
Token ID | 611 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
rament | +4.120 |
utenberg | +4.075 |
peed | +3.967 |
cott | +3.769 |
ranging | +3.638 |
Neg logit contributions | |
Jinping | -6.245 |
Statements | -6.171 |
NOT | -6.005 |
annot | -5.660 |
fore | -5.650 |
Token | the |
Token ID | 262 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
utenberg | +0.229 |
rament | +0.228 |
peed | +0.216 |
cott | +0.213 |
ranging | +0.205 |
Neg logit contributions | |
Jinping | -0.240 |
Statements | -0.229 |
NOT | -0.223 |
fore | -0.214 |
't | -0.210 |
Token | Ob |
Token ID | 1835 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
Jinping | +0.645 |
Statements | +0.608 |
NOT | +0.583 |
fore | +0.554 |
't | +0.542 |
Neg logit contributions | |
utenberg | -0.986 |
rament | -0.985 |
peed | -0.942 |
cott | -0.932 |
ranging | -0.905 |
Token | amas |
Token ID | 17485 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
Jinping | +5.597 |
Statements | +5.314 |
NOT | +5.226 |
't | +5.107 |
fore | +5.097 |
Neg logit contributions | |
utenberg | -5.317 |
rament | -5.220 |
peed | -5.033 |
cott | -4.963 |
"$:/ | -4.930 |
Token | its |
Token ID | 663 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
utenberg | +1.873 |
rament | +1.870 |
peed | +1.765 |
cott | +1.756 |
"$:/ | +1.685 |
Neg logit contributions | |
Jinping | -1.968 |
Statements | -1.884 |
NOT | -1.833 |
fore | -1.745 |
't | -1.715 |
Token | intellectual |
Token ID | 9028 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
rament | +3.021 |
utenberg | +3.019 |
peed | +2.827 |
cott | +2.817 |
ranging | +2.719 |
Neg logit contributions | |
Jinping | -3.055 |
Statements | -2.936 |
NOT | -2.848 |
fore | -2.708 |
't | -2.656 |
Token | protagonists |
Token ID | 39558 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
utenberg | +5.165 |
rament | +5.148 |
cott | +4.801 |
peed | +4.791 |
annah | +4.678 |
Neg logit contributions | |
Jinping | -4.929 |
Statements | -4.790 |
NOT | -4.646 |
fore | -4.353 |
annot | -4.290 |
Token | would |
Token ID | 561 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
utenberg | +0.619 |
rament | +0.619 |
peed | +0.585 |
cott | +0.576 |
ranging | +0.554 |
Neg logit contributions | |
Jinping | -0.673 |
Statements | -0.643 |
NOT | -0.625 |
fore | -0.602 |
't | -0.592 |
Token | not |
Token ID | 407 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
utenberg | +7.714 |
rament | +7.688 |
peed | +7.425 |
cott | +7.099 |
"$:/ | +7.089 |
Neg logit contributions | |
Jinping | -8.532 |
Statements | -8.341 |
NOT | -7.794 |
fore | -7.510 |
annot | -7.455 |
Token | have |
Token ID | 423 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
rament | +11.300 |
utenberg | +10.927 |
peed | +10.831 |
"$:/ | +10.577 |
cott | +10.343 |
Neg logit contributions | |
Jinping | -12.068 |
Statements | -11.968 |
NOT | -11.069 |
annot | -10.830 |
fore | -10.452 |
Token | had |
Token ID | 550 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
rament | +12.647 |
minded | +12.211 |
peed | +11.974 |
cott | +11.965 |
utenberg | +11.743 |
Neg logit contributions | |
Statements | -14.233 |
Jinping | -13.966 |
NOT | -13.440 |
annot | -12.976 |
Tycoon | -12.436 |
Token | a |
Token ID | 257 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
rament | +8.486 |
utenberg | +8.255 |
cott | +8.091 |
peed | +7.964 |
"$:/ | +7.719 |
Neg logit contributions | |
Jinping | -9.034 |
Statements | -8.807 |
NOT | -8.609 |
annot | -8.084 |
fore | -7.885 |
Token | tangible |
Token ID | 24607 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
utenberg | +2.333 |
rament | +2.326 |
peed | +2.190 |
cott | +2.175 |
ranging | +2.103 |
Neg logit contributions | |
Jinping | -2.478 |
Statements | -2.386 |
NOT | -2.312 |
fore | -2.186 |
annot | -2.160 |
Token |
|
Token ID | 198 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
utenberg | +6.396 |
rament | +6.312 |
peed | +5.994 |
cott | +5.884 |
ranging | +5.700 |
Neg logit contributions | |
Jinping | -7.901 |
Statements | -7.726 |
NOT | -7.401 |
annot | -7.035 |
fore | -6.987 |
Token |
|
Token ID | 198 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
rament | +1.829 |
utenberg | +1.823 |
peed | +1.753 |
cott | +1.730 |
ranging | +1.683 |
Neg logit contributions | |
Jinping | -1.891 |
Statements | -1.806 |
NOT | -1.716 |
fore | -1.644 |
annot | -1.609 |
Token | angler |
Token ID | 49910 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
Jinping | +3.954 |
Statements | +3.723 |
NOT | +3.668 |
't | +3.587 |
fore | +3.537 |
Neg logit contributions | |
rament | -3.507 |
utenberg | -3.453 |
peed | -3.320 |
"$:/ | -3.248 |
cott | -3.209 |
Token | Vine |
Token ID | 28879 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
Jinping | +1.310 |
Statements | +1.242 |
NOT | +1.202 |
fore | +1.146 |
't | +1.131 |
Neg logit contributions | |
utenberg | -1.732 |
rament | -1.714 |
peed | -1.642 |
cott | -1.607 |
"$:/ | -1.575 |
Token | yards |
Token ID | 33750 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
Jinping | +1.611 |
Statements | +1.527 |
NOT | +1.479 |
fore | +1.429 |
't | +1.408 |
Neg logit contributions | |
utenberg | -1.846 |
rament | -1.808 |
peed | -1.725 |
cott | -1.681 |
"$:/ | -1.679 |
Token | in |
Token ID | 287 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
Jinping | +0.323 |
Statements | +0.310 |
NOT | +0.302 |
fore | +0.289 |
't | +0.286 |
Neg logit contributions | |
utenberg | -0.292 |
rament | -0.289 |
peed | -0.273 |
cott | -0.269 |
"$:/ | -0.262 |
Token | the |
Token ID | 262 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
rament | +2.733 |
utenberg | +2.717 |
peed | +2.581 |
cott | +2.570 |
annah | +2.462 |
Neg logit contributions | |
Jinping | -2.578 |
Statements | -2.449 |
NOT | -2.390 |
fore | -2.280 |
't | -2.213 |
Token | U |
Token ID | 471 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
rament | +1.387 |
utenberg | +1.380 |
peed | +1.309 |
cott | +1.296 |
ranging | +1.245 |
Neg logit contributions | |
Jinping | -1.403 |
Statements | -1.336 |
NOT | -1.301 |
fore | -1.244 |
't | -1.222 |
Token | mp |
Token ID | 3149 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
rament | +13.916 |
cott | +13.305 |
utenberg | +12.731 |
annah | +12.413 |
peed | +11.983 |
Neg logit contributions | |
Jinping | -13.891 |
Statements | -13.447 |
NOT | -11.708 |
Tycoon | -11.232 |
� | -11.088 |
Token | qua |
Token ID | 39566 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
rament | +2.561 |
utenberg | +2.443 |
peed | +2.273 |
annah | +2.180 |
cott | +2.179 |
Neg logit contributions | |
Jinping | -8.702 |
Statements | -8.325 |
NOT | -8.235 |
fore | -7.862 |
annot | -7.766 |
Token | Valley |
Token ID | 6916 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
Jinping | +0.392 |
Statements | +0.376 |
NOT | +0.365 |
fore | +0.350 |
't | +0.346 |
Neg logit contributions | |
utenberg | -0.372 |
rament | -0.371 |
peed | -0.350 |
cott | -0.345 |
"$:/ | -0.334 |
Token | . |
Token ID | 13 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
Jinping | +0.239 |
Statements | +0.228 |
NOT | +0.222 |
fore | +0.213 |
't | +0.211 |
Neg logit contributions | |
utenberg | -0.235 |
rament | -0.234 |
peed | -0.222 |
cott | -0.218 |
"$:/ | -0.211 |
Token | � |
Token ID | 564 |
Feature activation | -8.75e-02 |
Pos logit contributions | |
utenberg | +3.698 |
rament | +3.669 |
peed | +3.482 |
cott | +3.478 |
ranging | +3.359 |
Neg logit contributions | |
Jinping | -3.485 |
Statements | -3.260 |
NOT | -3.196 |
fore | -3.068 |
annot | -2.986 |
Token | well |
Token ID | 880 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
utenberg | +1.529 |
rament | +1.528 |
peed | +1.428 |
cott | +1.386 |
"$:/ | +1.338 |
Neg logit contributions | |
Jinping | -2.444 |
Statements | -2.370 |
NOT | -2.298 |
fore | -2.229 |
annot | -2.173 |
Token | lead |
Token ID | 1085 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
utenberg | +2.362 |
rament | +2.349 |
peed | +2.221 |
cott | +2.149 |
"$:/ | +2.082 |
Neg logit contributions | |
Jinping | -3.405 |
Statements | -3.307 |
NOT | -3.193 |
fore | -3.085 |
annot | -3.016 |
Token | to |
Token ID | 284 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
Jinping | +1.530 |
Statements | +1.397 |
NOT | +1.346 |
't | +1.267 |
fore | +1.264 |
Neg logit contributions | |
rament | -2.599 |
utenberg | -2.596 |
peed | -2.515 |
cott | -2.470 |
annah | -2.402 |
Token | a |
Token ID | 257 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
rament | +2.463 |
utenberg | +2.452 |
peed | +2.303 |
cott | +2.278 |
ranging | +2.198 |
Neg logit contributions | |
Jinping | -2.921 |
Statements | -2.816 |
NOT | -2.755 |
fore | -2.642 |
't | -2.566 |
Token | ( |
Token ID | 357 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
rament | +2.122 |
utenberg | +2.115 |
peed | +1.977 |
cott | +1.957 |
"$:/ | +1.897 |
Neg logit contributions | |
Jinping | -2.454 |
Statements | -2.384 |
NOT | -2.325 |
fore | -2.220 |
annot | -2.166 |
Token | violent |
Token ID | 24498 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
Jinping | +1.991 |
Statements | +1.883 |
NOT | +1.869 |
fore | +1.807 |
't | +1.764 |
Neg logit contributions | |
utenberg | -1.910 |
rament | -1.909 |
peed | -1.803 |
cott | -1.753 |
"$:/ | -1.718 |
Token | ) |
Token ID | 8 |
Feature activation | +7.24e-01 |
Pos logit contributions | |
utenberg | +9.461 |
rament | +9.436 |
acre | +8.913 |
peed | +8.701 |
pan | +8.402 |
Neg logit contributions | |
Statements | -16.385 |
Jinping | -15.935 |
NOT | -15.625 |
fore | -15.039 |
annot | -14.611 |
Token | struggle |
Token ID | 6531 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
utenberg | +3.997 |
rament | +3.987 |
peed | +3.717 |
cott | +3.678 |
ranging | +3.551 |
Neg logit contributions | |
Jinping | -5.092 |
Statements | -4.966 |
NOT | -4.859 |
fore | -4.667 |
annot | -4.528 |
Token | , |
Token ID | 11 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
rament | +2.268 |
utenberg | +2.255 |
peed | +2.130 |
cott | +2.099 |
ranging | +2.020 |
Neg logit contributions | |
Jinping | -2.691 |
Statements | -2.596 |
NOT | -2.525 |
fore | -2.420 |
't | -2.371 |
Token | � |
Token ID | 447 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
utenberg | +2.099 |
rament | +2.088 |
peed | +1.971 |
cott | +1.945 |
ranging | +1.894 |
Neg logit contributions | |
Jinping | -2.323 |
Statements | -2.229 |
NOT | -2.170 |
fore | -2.081 |
't | -2.030 |
Token | � |
Token ID | 251 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
Jinping | +0.932 |
Statements | +0.885 |
NOT | +0.857 |
fore | +0.817 |
't | +0.804 |
Neg logit contributions | |
utenberg | -1.112 |
rament | -1.111 |
peed | -1.057 |
cott | -1.045 |
ranging | -1.009 |
Token | at |
Token ID | 379 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
Jinping | +0.169 |
Statements | +0.162 |
NOT | +0.158 |
fore | +0.152 |
't | +0.149 |
Neg logit contributions | |
rament | -0.159 |
utenberg | -0.159 |
peed | -0.150 |
cott | -0.147 |
"$:/ | -0.142 |
Token | the |
Token ID | 262 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
rament | +3.833 |
utenberg | +3.758 |
peed | +3.596 |
cott | +3.588 |
ranging | +3.429 |
Neg logit contributions | |
Jinping | -3.543 |
Statements | -3.362 |
NOT | -3.292 |
fore | -3.103 |
't | -3.036 |
Token | National |
Token ID | 2351 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
rament | +2.474 |
utenberg | +2.470 |
peed | +2.358 |
cott | +2.331 |
ranging | +2.243 |
Neg logit contributions | |
Jinping | -2.131 |
Statements | -2.015 |
NOT | -1.978 |
fore | -1.869 |
't | -1.829 |
Token | Museum |
Token ID | 9594 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
Jinping | +1.653 |
Statements | +1.592 |
NOT | +1.515 |
fore | +1.460 |
't | +1.450 |
Neg logit contributions | |
utenberg | -2.001 |
rament | -1.974 |
peed | -1.883 |
cott | -1.837 |
"$:/ | -1.806 |
Token | in |
Token ID | 287 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
rament | +3.171 |
utenberg | +3.159 |
peed | +3.010 |
cott | +2.974 |
ranging | +2.880 |
Neg logit contributions | |
Jinping | -2.745 |
Statements | -2.586 |
NOT | -2.525 |
fore | -2.407 |
't | -2.338 |
Token | U |
Token ID | 471 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
rament | +4.428 |
utenberg | +4.393 |
cott | +4.156 |
peed | +4.135 |
annah | +3.994 |
Neg logit contributions | |
Jinping | -4.289 |
Statements | -4.050 |
NOT | -3.996 |
fore | -3.772 |
annot | -3.695 |
Token | eno |
Token ID | 23397 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
rament | +12.981 |
utenberg | +12.375 |
cott | +12.164 |
pan | +11.529 |
annah | +11.459 |
Neg logit contributions | |
Jinping | -13.064 |
Statements | -12.942 |
NOT | -11.684 |
� | -10.777 |
fore | -10.750 |
Token | ( |
Token ID | 357 |
Feature activation | -1.55e-03 |
Pos logit contributions | |
Jinping | +1.550 |
Statements | +1.493 |
NOT | +1.441 |
fore | +1.399 |
't | +1.378 |
Neg logit contributions | |
utenberg | -1.273 |
rament | -1.272 |
peed | -1.208 |
cott | -1.186 |
"$:/ | -1.142 |
Token | End |
Token ID | 12915 |
Feature activation | -9.57e-01 |
Pos logit contributions | |
Jinping | +0.009 |
Statements | +0.009 |
NOT | +0.009 |
fore | +0.008 |
't | +0.008 |
Neg logit contributions | |
utenberg | -0.011 |
rament | -0.011 |
peed | -0.010 |
cott | -0.010 |
ranging | -0.010 |
Token | s |
Token ID | 82 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
Jinping | +5.498 |
Statements | +5.435 |
NOT | +5.229 |
't | +4.834 |
fore | +4.826 |
Neg logit contributions | |
utenberg | -6.491 |
rament | -6.412 |
peed | -6.244 |
cott | -6.030 |
"$:/ | -5.910 |
Token | May |
Token ID | 1737 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
utenberg | +0.336 |
rament | +0.336 |
peed | +0.322 |
cott | +0.319 |
ranging | +0.311 |
Neg logit contributions | |
Jinping | -0.186 |
Statements | -0.174 |
NOT | -0.167 |
fore | -0.157 |
't | -0.152 |
Token | in |
Token ID | 287 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
Jinping | +0.015 |
Statements | +0.014 |
NOT | +0.014 |
fore | +0.013 |
't | +0.013 |
Neg logit contributions | |
utenberg | -0.012 |
rament | -0.012 |
peed | -0.011 |
cott | -0.011 |
ranging | -0.011 |
Token | jail |
Token ID | 7356 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
rament | +2.444 |
utenberg | +2.432 |
cott | +2.260 |
peed | +2.259 |
annah | +2.127 |
Neg logit contributions | |
Jinping | -3.961 |
Statements | -3.813 |
NOT | -3.735 |
fore | -3.588 |
't | -3.522 |
Token | ? |
Token ID | 30 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
rament | +2.589 |
utenberg | +2.582 |
peed | +2.448 |
cott | +2.418 |
ranging | +2.320 |
Neg logit contributions | |
Jinping | -2.979 |
Statements | -2.870 |
NOT | -2.770 |
fore | -2.645 |
't | -2.600 |
Token | The |
Token ID | 383 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
Jinping | +3.489 |
Statements | +3.457 |
NOT | +3.226 |
't | +3.035 |
fore | +3.019 |
Neg logit contributions | |
rament | -5.222 |
utenberg | -5.190 |
peed | -4.926 |
cott | -4.837 |
owship | -4.698 |
Token | " |
Token ID | 366 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
Jinping | +2.404 |
Statements | +2.332 |
NOT | +2.259 |
fore | +2.194 |
't | +2.166 |
Neg logit contributions | |
utenberg | -1.703 |
rament | -1.685 |
peed | -1.596 |
cott | -1.550 |
ranging | -1.487 |
Token | free |
Token ID | 5787 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
Jinping | +1.357 |
Statements | +1.309 |
NOT | +1.298 |
fore | +1.248 |
't | +1.217 |
Neg logit contributions | |
utenberg | -1.382 |
rament | -1.370 |
peed | -1.294 |
cott | -1.249 |
"$:/ | -1.237 |
Token | speech |
Token ID | 4046 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
cott | +8.616 |
acre | +8.452 |
ranging | +8.204 |
rament | +8.180 |
minded | +8.154 |
Neg logit contributions | |
Jinping | -16.449 |
NOT | -15.387 |
Statements | -15.375 |
� | -14.500 |
fore | -14.448 |
Token | " |
Token ID | 1 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
rament | +2.687 |
utenberg | +2.682 |
cott | +2.507 |
peed | +2.489 |
annah | +2.360 |
Neg logit contributions | |
Jinping | -3.907 |
Statements | -3.763 |
NOT | -3.651 |
fore | -3.544 |
't | -3.435 |
Token | in |
Token ID | 287 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
Jinping | +2.441 |
Statements | +2.348 |
NOT | +2.282 |
fore | +2.185 |
't | +2.179 |
Neg logit contributions | |
rament | -1.871 |
utenberg | -1.869 |
peed | -1.756 |
cott | -1.713 |
ranging | -1.651 |
Token | Bac |
Token ID | 35583 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
Jinping | +1.238 |
Statements | +1.189 |
NOT | +1.139 |
fore | +1.093 |
't | +1.089 |
Neg logit contributions | |
utenberg | -1.382 |
rament | -1.371 |
peed | -1.315 |
cott | -1.275 |
ranging | -1.237 |
Token | ile |
Token ID | 576 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
Jinping | +1.873 |
Statements | +1.794 |
NOT | +1.750 |
fore | +1.690 |
't | +1.670 |
Neg logit contributions | |
rament | -1.402 |
utenberg | -1.402 |
peed | -1.313 |
cott | -1.293 |
ranging | -1.241 |
Token | all |
Token ID | 477 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
rament | +4.574 |
utenberg | +4.521 |
peed | +4.331 |
cott | +4.171 |
"$:/ | +4.110 |
Neg logit contributions | |
Jinping | -5.451 |
Statements | -5.170 |
NOT | -5.045 |
fore | -4.816 |
annot | -4.726 |
Token | this |
Token ID | 428 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
rament | +3.679 |
utenberg | +3.678 |
peed | +3.541 |
cott | +3.459 |
"$:/ | +3.326 |
Neg logit contributions | |
Jinping | -3.830 |
Statements | -3.633 |
NOT | -3.567 |
fore | -3.366 |
annot | -3.311 |
Token | goodness |
Token ID | 20437 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
rament | +4.850 |
utenberg | +4.841 |
peed | +4.653 |
cott | +4.513 |
ranging | +4.361 |
Neg logit contributions | |
Jinping | -4.610 |
Statements | -4.459 |
NOT | -4.299 |
fore | -4.072 |
annot | -3.961 |
Token | will |
Token ID | 481 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
utenberg | +4.602 |
rament | +4.587 |
peed | +4.422 |
cott | +4.321 |
"$:/ | +4.176 |
Neg logit contributions | |
Jinping | -3.868 |
Statements | -3.764 |
NOT | -3.587 |
fore | -3.399 |
annot | -3.295 |
Token | remain |
Token ID | 3520 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
utenberg | +6.334 |
rament | +6.300 |
peed | +6.141 |
annah | +5.699 |
"$:/ | +5.647 |
Neg logit contributions | |
Jinping | -7.640 |
Statements | -7.497 |
NOT | -6.883 |
fore | -6.851 |
annot | -6.782 |
Token | free |
Token ID | 1479 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
utenberg | +7.068 |
rament | +6.969 |
peed | +6.867 |
cott | +6.432 |
"$:/ | +6.235 |
Neg logit contributions | |
Jinping | -9.866 |
Statements | -9.762 |
NOT | -9.381 |
fore | -8.865 |
annot | -8.782 |
Token | for |
Token ID | 329 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
utenberg | +10.959 |
rament | +10.745 |
peed | +10.483 |
ranging | +10.094 |
cott | +9.917 |
Neg logit contributions | |
Jinping | -14.377 |
Statements | -13.755 |
NOT | -13.076 |
annot | -12.324 |
� | -12.200 |
Token | all |
Token ID | 477 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
utenberg | +3.173 |
rament | +3.158 |
peed | +3.004 |
cott | +2.909 |
"$:/ | +2.858 |
Neg logit contributions | |
Jinping | -4.092 |
Statements | -3.918 |
NOT | -3.806 |
fore | -3.634 |
annot | -3.571 |
Token | , |
Token ID | 11 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
rament | +5.713 |
utenberg | +5.677 |
peed | +5.448 |
cott | +5.378 |
"$:/ | +5.247 |
Neg logit contributions | |
Jinping | -6.284 |
Statements | -6.010 |
NOT | -5.887 |
annot | -5.588 |
fore | -5.530 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
utenberg | +2.622 |
rament | +2.589 |
peed | +2.481 |
cott | +2.410 |
"$:/ | +2.349 |
Neg logit contributions | |
Jinping | -2.723 |
Statements | -2.618 |
NOT | -2.527 |
fore | -2.416 |
annot | -2.362 |
Token | without |
Token ID | 1231 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
rament | +2.205 |
utenberg | +2.199 |
peed | +2.101 |
cott | +2.051 |
"$:/ | +1.987 |
Neg logit contributions | |
Jinping | -2.379 |
Statements | -2.288 |
NOT | -2.199 |
fore | -2.112 |
annot | -2.042 |
Token | talks |
Token ID | 6130 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
rament | +3.160 |
utenberg | +3.137 |
peed | +2.996 |
cott | +2.940 |
ranging | +2.858 |
Neg logit contributions | |
Jinping | -3.320 |
Statements | -3.130 |
NOT | -3.099 |
fore | -2.917 |
't | -2.851 |
Token | she |
Token ID | 673 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
rament | +6.843 |
utenberg | +6.818 |
peed | +6.740 |
cott | +6.411 |
ranging | +6.282 |
Neg logit contributions | |
Jinping | -9.334 |
Statements | -9.004 |
NOT | -8.762 |
fore | -8.208 |
annot | -8.096 |
Token | and |
Token ID | 290 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
rament | +8.492 |
utenberg | +8.320 |
cott | +8.150 |
peed | +8.095 |
"$:/ | +7.769 |
Neg logit contributions | |
Statements | -8.864 |
Jinping | -8.695 |
NOT | -8.465 |
fore | -7.818 |
annot | -7.589 |
Token | I |
Token ID | 314 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
utenberg | +1.282 |
rament | +1.281 |
peed | +1.214 |
cott | +1.198 |
ranging | +1.156 |
Neg logit contributions | |
Jinping | -1.279 |
Statements | -1.220 |
NOT | -1.184 |
fore | -1.135 |
't | -1.115 |
Token | would |
Token ID | 561 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
rament | +5.701 |
utenberg | +5.532 |
peed | +5.441 |
cott | +5.217 |
ranging | +5.088 |
Neg logit contributions | |
Jinping | -8.536 |
Statements | -8.424 |
NOT | -8.099 |
fore | -7.658 |
annot | -7.554 |
Token | have |
Token ID | 423 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
peed | +8.550 |
rament | +8.461 |
utenberg | +8.451 |
ranging | +8.082 |
cott | +7.813 |
Neg logit contributions | |
Jinping | -13.175 |
Statements | -12.888 |
NOT | -12.044 |
annot | -11.701 |
fore | -11.613 |
Token | . |
Token ID | 13 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
peed | +10.177 |
rament | +10.122 |
ranging | +9.892 |
minded | +9.783 |
cott | +9.675 |
Neg logit contributions | |
Jinping | -14.492 |
Statements | -14.346 |
NOT | -13.784 |
annot | -13.314 |
fore | -12.807 |
Token | They |
Token ID | 1119 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
rament | +0.615 |
utenberg | +0.615 |
peed | +0.587 |
cott | +0.581 |
ranging | +0.563 |
Neg logit contributions | |
Jinping | -0.524 |
Statements | -0.493 |
NOT | -0.480 |
fore | -0.458 |
't | -0.446 |
Token | were |
Token ID | 547 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
rament | +3.506 |
utenberg | +3.490 |
peed | +3.312 |
cott | +3.238 |
ranging | +3.168 |
Neg logit contributions | |
Jinping | -3.914 |
Statements | -3.776 |
NOT | -3.661 |
fore | -3.477 |
't | -3.384 |
Token | not |
Token ID | 407 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
rament | +2.481 |
utenberg | +2.465 |
peed | +2.344 |
cott | +2.304 |
ranging | +2.252 |
Neg logit contributions | |
Jinping | -2.490 |
Statements | -2.369 |
NOT | -2.302 |
fore | -2.196 |
annot | -2.135 |
Token | football |
Token ID | 4346 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
rament | +4.328 |
utenberg | +4.248 |
peed | +4.126 |
cott | +4.000 |
ranging | +3.985 |
Neg logit contributions | |
Jinping | -4.978 |
Statements | -4.724 |
NOT | -4.641 |
fore | -4.366 |
annot | -4.320 |
Token | App |
Token ID | 4677 |
Feature activation | -9.27e-01 |
Pos logit contributions | |
rament | +1.887 |
utenberg | +1.878 |
peed | +1.793 |
cott | +1.783 |
ranging | +1.721 |
Neg logit contributions | |
Jinping | -1.836 |
Statements | -1.744 |
NOT | -1.662 |
fore | -1.607 |
't | -1.566 |
Token | alach |
Token ID | 33786 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
Jinping | +5.326 |
Statements | +5.084 |
NOT | +5.002 |
fore | +4.846 |
't | +4.741 |
Neg logit contributions | |
utenberg | -6.254 |
rament | -6.150 |
peed | -5.862 |
cott | -5.716 |
owship | -5.660 |
Token | ia |
Token ID | 544 |
Feature activation | +9.83e-02 |
Pos logit contributions | |
Jinping | +1.841 |
Statements | +1.739 |
NOT | +1.674 |
fore | +1.656 |
't | +1.634 |
Neg logit contributions | |
utenberg | -1.811 |
rament | -1.808 |
peed | -1.731 |
cott | -1.670 |
"$:/ | -1.631 |
Token | and |
Token ID | 290 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
utenberg | +0.685 |
rament | +0.685 |
peed | +0.651 |
cott | +0.644 |
ranging | +0.622 |
Neg logit contributions | |
Jinping | -0.582 |
Statements | -0.553 |
NOT | -0.536 |
fore | -0.511 |
't | -0.501 |
Token | the |
Token ID | 262 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
rament | +1.599 |
utenberg | +1.596 |
peed | +1.511 |
cott | +1.495 |
ranging | +1.439 |
Neg logit contributions | |
Jinping | -1.659 |
Statements | -1.587 |
NOT | -1.543 |
fore | -1.476 |
't | -1.445 |
Token | U |
Token ID | 471 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
rament | +1.694 |
utenberg | +1.692 |
peed | +1.603 |
cott | +1.585 |
ranging | +1.530 |
Neg logit contributions | |
Jinping | -1.597 |
Statements | -1.518 |
NOT | -1.480 |
fore | -1.411 |
't | -1.384 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
rament | +10.869 |
utenberg | +10.603 |
cott | +10.331 |
annah | +10.198 |
peed | +9.581 |
Neg logit contributions | |
Jinping | -14.892 |
Statements | -14.300 |
NOT | -12.719 |
annot | -12.269 |
Tycoon | -12.236 |
Token | S |
Token ID | 50 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
utenberg | +6.677 |
rament | +6.488 |
cott | +6.219 |
peed | +6.007 |
annah | +5.928 |
Neg logit contributions | |
Jinping | -10.599 |
Statements | -10.043 |
fore | -9.662 |
NOT | -9.587 |
annot | -9.144 |
Token | . |
Token ID | 13 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
utenberg | +6.075 |
rament | +5.978 |
cott | +5.630 |
peed | +5.627 |
annah | +5.430 |
Neg logit contributions | |
Jinping | -7.804 |
Statements | -7.445 |
NOT | -7.303 |
fore | -7.009 |
't | -6.721 |
Token | southeast |
Token ID | 26015 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
rament | +3.589 |
utenberg | +3.583 |
peed | +3.403 |
cott | +3.374 |
annah | +3.242 |
Neg logit contributions | |
Jinping | -3.418 |
Statements | -3.231 |
NOT | -3.167 |
fore | -3.027 |
't | -2.954 |
Token | are |
Token ID | 389 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
rament | +3.737 |
utenberg | +3.709 |
peed | +3.540 |
cott | +3.529 |
ranging | +3.383 |
Neg logit contributions | |
Jinping | -4.023 |
Statements | -3.790 |
NOT | -3.714 |
fore | -3.524 |
't | -3.448 |
Token | research |
Token ID | 34033 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
Jinping | +7.134 |
fore | +7.018 |
NOT | +6.870 |
't | +6.606 |
annot | +6.568 |
Neg logit contributions | |
utenberg | -8.193 |
rament | -8.077 |
"$:/ | -7.512 |
peed | -7.477 |
owship | -7.322 |
Token | - |
Token ID | 12 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
rament | +3.194 |
utenberg | +3.177 |
peed | +3.029 |
cott | +2.983 |
ranging | +2.868 |
Neg logit contributions | |
Jinping | -3.795 |
Statements | -3.616 |
NOT | -3.524 |
fore | -3.350 |
't | -3.306 |
Token | threat |
Token ID | 19971 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
Jinping | +4.489 |
NOT | +4.284 |
Statements | +4.260 |
fore | +4.243 |
't | +4.071 |
Neg logit contributions | |
utenberg | -4.192 |
rament | -4.189 |
peed | -3.919 |
"$:/ | -3.848 |
cott | -3.790 |
Token | ens |
Token ID | 641 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
Jinping | +5.791 |
Statements | +5.644 |
NOT | +5.351 |
fore | +5.277 |
't | +5.212 |
Neg logit contributions | |
utenberg | -5.794 |
rament | -5.710 |
"$:/ | -5.341 |
peed | -5.296 |
cott | -5.061 |
Token | - |
Token ID | 12 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
Jinping | +0.657 |
Statements | +0.632 |
NOT | +0.611 |
fore | +0.592 |
't | +0.583 |
Neg logit contributions | |
utenberg | -0.667 |
rament | -0.662 |
peed | -0.625 |
cott | -0.616 |
"$:/ | -0.600 |
Token | free |
Token ID | 5787 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
Jinping | +3.640 |
Statements | +3.494 |
NOT | +3.487 |
fore | +3.420 |
't | +3.329 |
Neg logit contributions | |
utenberg | -3.603 |
rament | -3.571 |
peed | -3.353 |
cott | -3.240 |
"$:/ | -3.223 |
Token | - |
Token ID | 12 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
peed | +10.714 |
cott | +10.660 |
ranging | +10.646 |
rament | +10.554 |
acre | +10.263 |
Neg logit contributions | |
Jinping | -14.559 |
NOT | -13.226 |
Statements | -13.168 |
� | -12.537 |
Tycoon | -12.001 |
Token | speech |
Token ID | 45862 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
rament | +5.317 |
utenberg | +5.220 |
cott | +5.120 |
peed | +5.066 |
ranging | +5.042 |
Neg logit contributions | |
Jinping | -7.486 |
Statements | -7.114 |
NOT | -6.755 |
fore | -6.357 |
't | -6.334 |
Token | - |
Token ID | 12 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
utenberg | +4.748 |
rament | +4.745 |
peed | +4.534 |
cott | +4.497 |
ranging | +4.295 |
Neg logit contributions | |
Jinping | -5.786 |
Statements | -5.397 |
NOT | -5.318 |
fore | -5.111 |
't | -4.974 |
Token | and |
Token ID | 392 |
Feature activation | +8.24e-01 |
Pos logit contributions | |
Jinping | +3.128 |
Statements | +3.032 |
NOT | +3.013 |
fore | +2.940 |
't | +2.857 |
Neg logit contributions | |
utenberg | -3.037 |
rament | -3.016 |
peed | -2.842 |
"$:/ | -2.741 |
cott | -2.733 |
Token | - |
Token ID | 12 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
rament | +4.676 |
utenberg | +4.602 |
peed | +4.419 |
cott | +4.349 |
ranging | +4.212 |
Neg logit contributions | |
Jinping | -5.812 |
Statements | -5.527 |
NOT | -5.346 |
fore | -5.069 |
't | -4.981 |
Token | noting |
Token ID | 10820 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
utenberg | +2.400 |
rament | +2.386 |
peed | +2.232 |
cott | +2.212 |
ranging | +2.142 |
Neg logit contributions | |
Jinping | -3.270 |
Statements | -3.147 |
NOT | -3.069 |
fore | -2.956 |
't | -2.890 |
Token | that |
Token ID | 326 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
utenberg | +0.902 |
rament | +0.901 |
peed | +0.850 |
cott | +0.838 |
ranging | +0.805 |
Neg logit contributions | |
Jinping | -1.028 |
Statements | -0.983 |
NOT | -0.956 |
fore | -0.919 |
't | -0.906 |
Token | she |
Token ID | 673 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
rament | +3.476 |
utenberg | +3.458 |
peed | +3.272 |
cott | +3.230 |
ranging | +3.128 |
Neg logit contributions | |
Jinping | -4.065 |
Statements | -3.873 |
NOT | -3.786 |
fore | -3.630 |
annot | -3.541 |
Token | would |
Token ID | 561 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
rament | +2.790 |
utenberg | +2.779 |
cott | +2.612 |
peed | +2.605 |
annah | +2.504 |
Neg logit contributions | |
Jinping | -3.128 |
Statements | -3.061 |
NOT | -2.942 |
fore | -2.804 |
annot | -2.700 |
Token | not |
Token ID | 407 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
utenberg | +7.856 |
rament | +7.754 |
peed | +7.583 |
"$:/ | +7.487 |
cott | +7.243 |
Neg logit contributions | |
Jinping | -8.863 |
Statements | -8.651 |
NOT | -7.853 |
annot | -7.829 |
fore | -7.809 |
Token | have |
Token ID | 423 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
"$:/ | +8.922 |
utenberg | +8.877 |
rament | +8.853 |
peed | +8.732 |
ospons | +8.295 |
Neg logit contributions | |
Jinping | -12.049 |
Statements | -12.005 |
annot | -11.341 |
NOT | -11.222 |
fore | -10.900 |
Token | qualified |
Token ID | 10617 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
minded | +9.569 |
rament | +9.293 |
tolerated | +9.194 |
cott | +8.920 |
peed | +8.821 |
Neg logit contributions | |
Statements | -15.164 |
Jinping | -14.685 |
NOT | -14.597 |
annot | -14.187 |
fore | -13.694 |
Token | for |
Token ID | 329 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
Jinping | +2.677 |
Statements | +2.512 |
NOT | +2.401 |
fore | +2.351 |
't | +2.349 |
Neg logit contributions | |
utenberg | -3.484 |
rament | -3.448 |
peed | -3.278 |
cott | -3.232 |
ranging | -3.144 |
Token | the |
Token ID | 262 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
utenberg | +0.423 |
rament | +0.423 |
peed | +0.397 |
cott | +0.393 |
ranging | +0.379 |
Neg logit contributions | |
Jinping | -0.456 |
Statements | -0.437 |
NOT | -0.423 |
fore | -0.408 |
't | -0.404 |
Token | scholarship |
Token ID | 18652 |
Feature activation | -9.03e-02 |
Pos logit contributions | |
rament | +2.119 |
utenberg | +2.109 |
peed | +2.015 |
cott | +1.977 |
annah | +1.905 |
Neg logit contributions | |
Jinping | -2.198 |
Statements | -2.080 |
NOT | -2.034 |
fore | -1.941 |
't | -1.889 |
Token | under |
Token ID | 739 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Jinping | +0.615 |
Statements | +0.588 |
NOT | +0.571 |
fore | +0.552 |
't | +0.545 |
Neg logit contributions | |
utenberg | -0.555 |
rament | -0.552 |
peed | -0.519 |
cott | -0.513 |
ranging | -0.494 |
Token | .[ |
Token ID | 3693 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Jinping | +0.293 |
Statements | +0.279 |
NOT | +0.273 |
fore | +0.263 |
't | +0.260 |
Neg logit contributions | |
rament | -0.244 |
utenberg | -0.244 |
peed | -0.228 |
cott | -0.226 |
ranging | -0.217 |
Token | 7 |
Token ID | 22 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
Jinping | +1.828 |
Statements | +1.760 |
NOT | +1.740 |
fore | +1.659 |
't | +1.633 |
Neg logit contributions | |
utenberg | -2.016 |
rament | -2.009 |
peed | -1.891 |
cott | -1.857 |
"$:/ | -1.792 |
Token | ] |
Token ID | 60 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
utenberg | +2.946 |
rament | +2.865 |
peed | +2.659 |
cott | +2.566 |
ranging | +2.383 |
Neg logit contributions | |
Jinping | -7.906 |
Statements | -7.585 |
NOT | -7.435 |
fore | -7.248 |
't | -7.118 |
Token |
|
Token ID | 198 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
utenberg | +2.160 |
rament | +2.144 |
peed | +2.043 |
cott | +2.015 |
"$:/ | +1.944 |
Neg logit contributions | |
Jinping | -2.142 |
Statements | -2.034 |
NOT | -1.979 |
fore | -1.899 |
't | -1.862 |
Token |
|
Token ID | 198 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
utenberg | +0.508 |
rament | +0.507 |
peed | +0.485 |
cott | +0.479 |
ranging | +0.465 |
Neg logit contributions | |
Jinping | -0.489 |
Statements | -0.464 |
NOT | -0.445 |
fore | -0.428 |
annot | -0.414 |
Token | U |
Token ID | 52 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
utenberg | +2.832 |
rament | +2.822 |
peed | +2.697 |
cott | +2.671 |
ranging | +2.581 |
Neg logit contributions | |
Jinping | -2.698 |
Statements | -2.569 |
NOT | -2.457 |
fore | -2.353 |
't | -2.316 |
Token | ng |
Token ID | 782 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
rament | +13.238 |
utenberg | +12.773 |
cott | +12.699 |
peed | +12.049 |
ranging | +11.880 |
Neg logit contributions | |
Jinping | -11.807 |
Statements | -11.678 |
NOT | -10.285 |
fore | -9.990 |
annot | -9.571 |
Token | roup |
Token ID | 3233 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
rament | +5.059 |
utenberg | +4.999 |
cott | +4.684 |
peed | +4.639 |
annah | +4.515 |
Neg logit contributions | |
Jinping | -6.546 |
Statements | -6.278 |
NOT | -6.071 |
fore | -5.825 |
't | -5.650 |
Token | ed |
Token ID | 276 |
Feature activation | +8.94e-01 |
Pos logit contributions | |
utenberg | +2.405 |
rament | +2.394 |
peed | +2.279 |
cott | +2.246 |
"$:/ | +2.184 |
Neg logit contributions | |
Jinping | -2.363 |
Statements | -2.247 |
NOT | -2.199 |
fore | -2.088 |
't | -2.043 |
Token | [ |
Token ID | 685 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
utenberg | +5.754 |
rament | +5.622 |
peed | +5.440 |
cott | +5.332 |
"$:/ | +5.218 |
Neg logit contributions | |
Jinping | -5.452 |
Statements | -5.141 |
NOT | -5.090 |
fore | -4.825 |
't | -4.663 |
Token | edit |
Token ID | 4370 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
utenberg | +3.330 |
rament | +3.115 |
cott | +3.007 |
peed | +2.953 |
ranging | +2.876 |
Neg logit contributions | |
Jinping | -9.022 |
Statements | -8.695 |
NOT | -8.282 |
fore | -8.090 |
't | -7.939 |
Token | � |
Token ID | 251 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
Jinping | +0.152 |
Statements | +0.145 |
NOT | +0.141 |
fore | +0.135 |
't | +0.133 |
Neg logit contributions | |
utenberg | -0.163 |
rament | -0.163 |
peed | -0.154 |
cott | -0.152 |
ranging | -0.147 |
Token |
|
Token ID | 198 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
Jinping | +2.952 |
Statements | +2.915 |
NOT | +2.775 |
fore | +2.590 |
't | +2.587 |
Neg logit contributions | |
rament | -4.110 |
utenberg | -4.047 |
peed | -3.863 |
cott | -3.754 |
ranging | -3.688 |
Token |
|
Token ID | 198 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
Jinping | +0.311 |
Statements | +0.295 |
NOT | +0.285 |
fore | +0.273 |
't | +0.266 |
Neg logit contributions | |
utenberg | -0.350 |
rament | -0.350 |
peed | -0.333 |
cott | -0.329 |
ranging | -0.318 |
Token | http |
Token ID | 4023 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
rament | +1.465 |
utenberg | +1.464 |
peed | +1.394 |
cott | +1.381 |
ranging | +1.335 |
Neg logit contributions | |
Jinping | -1.284 |
Statements | -1.216 |
NOT | -1.171 |
fore | -1.124 |
't | -1.100 |
Token | :// |
Token ID | 1378 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
Jinping | +2.391 |
Statements | +2.295 |
NOT | +2.270 |
't | +2.190 |
fore | +2.182 |
Neg logit contributions | |
utenberg | -1.920 |
rament | -1.917 |
peed | -1.779 |
cott | -1.746 |
owship | -1.693 |
Token | free |
Token ID | 5787 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
Jinping | +3.883 |
NOT | +3.755 |
Statements | +3.696 |
fore | +3.664 |
't | +3.575 |
Neg logit contributions | |
utenberg | -4.524 |
rament | -4.523 |
peed | -4.236 |
cott | -4.210 |
"$:/ | -4.153 |
Token | be |
Token ID | 1350 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
peed | +12.749 |
rament | +12.749 |
ranging | +12.393 |
cott | +12.297 |
pan | +12.238 |
Neg logit contributions | |
Jinping | -12.454 |
Statements | -11.811 |
NOT | -11.449 |
Tycoon | -10.061 |
� | -9.900 |
Token | acon |
Token ID | 7807 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
utenberg | +2.685 |
rament | +2.666 |
peed | +2.567 |
cott | +2.508 |
annah | +2.433 |
Neg logit contributions | |
Jinping | -2.659 |
Statements | -2.537 |
NOT | -2.458 |
fore | -2.317 |
't | -2.301 |
Token | . |
Token ID | 13 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
utenberg | +0.822 |
rament | +0.822 |
peed | +0.778 |
cott | +0.768 |
ranging | +0.740 |
Neg logit contributions | |
Jinping | -0.831 |
Statements | -0.793 |
NOT | -0.769 |
fore | -0.736 |
't | -0.724 |
Token | com |
Token ID | 785 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
utenberg | +4.436 |
rament | +4.361 |
cott | +4.089 |
peed | +4.084 |
annah | +3.926 |
Neg logit contributions | |
Jinping | -6.766 |
Statements | -6.478 |
NOT | -6.223 |
fore | -5.943 |
't | -5.853 |
Token | / |
Token ID | 14 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
Jinping | +0.160 |
Statements | +0.155 |
NOT | +0.150 |
fore | +0.146 |
't | +0.144 |
Neg logit contributions | |
utenberg | -0.113 |
rament | -0.112 |
peed | -0.104 |
cott | -0.103 |
ranging | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
utenberg | +2.781 |
rament | +2.766 |
peed | +2.624 |
cott | +2.561 |
ranging | +2.492 |
Neg logit contributions | |
Jinping | -3.304 |
Statements | -3.171 |
NOT | -3.079 |
fore | -2.938 |
't | -2.891 |
Token | and |
Token ID | 290 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
utenberg | +0.363 |
rament | +0.363 |
peed | +0.343 |
cott | +0.338 |
ranging | +0.327 |
Neg logit contributions | |
Jinping | -0.397 |
Statements | -0.379 |
NOT | -0.369 |
fore | -0.354 |
't | -0.347 |
Token | what |
Token ID | 644 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
Jinping | +0.897 |
Statements | +0.854 |
NOT | +0.826 |
fore | +0.795 |
't | +0.784 |
Neg logit contributions | |
utenberg | -0.986 |
rament | -0.982 |
peed | -0.931 |
cott | -0.922 |
"$:/ | -0.888 |
Token | would |
Token ID | 561 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
rament | +2.255 |
utenberg | +2.238 |
peed | +2.141 |
cott | +2.098 |
ranging | +2.034 |
Neg logit contributions | |
Jinping | -2.373 |
Statements | -2.258 |
NOT | -2.193 |
fore | -2.085 |
annot | -2.039 |
Token | you |
Token ID | 345 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
rament | +5.038 |
utenberg | +5.023 |
peed | +4.775 |
cott | +4.591 |
"$:/ | +4.576 |
Neg logit contributions | |
Jinping | -6.716 |
Statements | -6.360 |
NOT | -6.152 |
fore | -5.895 |
annot | -5.871 |
Token | have |
Token ID | 423 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
utenberg | +8.971 |
rament | +8.667 |
peed | +8.515 |
ranging | +8.329 |
annah | +8.226 |
Neg logit contributions | |
Jinping | -9.821 |
Statements | -9.586 |
annot | -8.927 |
NOT | -8.872 |
� | -8.433 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
minded | +8.890 |
rament | +8.454 |
peed | +8.236 |
utenberg | +8.078 |
ranging | +7.992 |
Neg logit contributions | |
Jinping | -15.451 |
Statements | -15.305 |
annot | -14.279 |
NOT | -14.270 |
FORE | -13.548 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
utenberg | +8.499 |
rament | +8.453 |
peed | +8.109 |
cott | +8.037 |
ranging | +7.894 |
Neg logit contributions | |
Jinping | -7.732 |
Statements | -7.144 |
NOT | -6.917 |
annot | -6.534 |
fore | -6.498 |
Token | have |
Token ID | 423 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
utenberg | +6.679 |
rament | +6.478 |
peed | +6.295 |
"$:/ | +6.069 |
annah | +5.977 |
Neg logit contributions | |
Jinping | -6.836 |
Statements | -6.541 |
NOT | -6.163 |
annot | -5.969 |
fore | -5.963 |
Token | done |
Token ID | 1760 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
utenberg | +7.196 |
rament | +7.054 |
peed | +6.850 |
"$:/ | +6.623 |
minded | +6.461 |
Neg logit contributions | |
Statements | -12.613 |
Jinping | -12.601 |
NOT | -12.017 |
annot | -11.628 |
fore | -11.157 |
Token | differently |
Token ID | 10338 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
rament | +6.662 |
utenberg | +6.616 |
peed | +6.400 |
"$:/ | +6.045 |
cott | +5.992 |
Neg logit contributions | |
Jinping | -9.177 |
Statements | -8.885 |
NOT | -8.642 |
annot | -8.288 |
fore | -7.928 |
Token | by |
Token ID | 416 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
utenberg | +5.158 |
rament | +5.105 |
peed | +4.988 |
cott | +4.811 |
ranging | +4.748 |
Neg logit contributions | |
Jinping | -5.166 |
Statements | -4.986 |
NOT | -4.841 |
fore | -4.579 |
annot | -4.455 |
Token | the |
Token ID | 262 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
rament | +1.712 |
utenberg | +1.707 |
peed | +1.609 |
cott | +1.591 |
ranging | +1.534 |
Neg logit contributions | |
Jinping | -1.883 |
Statements | -1.793 |
NOT | -1.743 |
fore | -1.675 |
't | -1.633 |
Token | voters |
Token ID | 4446 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
rament | +3.327 |
utenberg | +3.314 |
cott | +3.108 |
peed | +3.103 |
"$:/ | +2.967 |
Neg logit contributions | |
Jinping | -3.573 |
Statements | -3.382 |
NOT | -3.283 |
fore | -3.169 |
't | -3.080 |
Token | that |
Token ID | 326 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
utenberg | +2.988 |
rament | +2.955 |
peed | +2.848 |
cott | +2.786 |
ranging | +2.702 |
Neg logit contributions | |
Jinping | -2.909 |
Statements | -2.796 |
NOT | -2.699 |
fore | -2.582 |
't | -2.504 |
Token | would |
Token ID | 561 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
rament | +3.106 |
utenberg | +3.084 |
peed | +2.935 |
cott | +2.886 |
ranging | +2.774 |
Neg logit contributions | |
Jinping | -3.498 |
Statements | -3.338 |
NOT | -3.252 |
fore | -3.126 |
't | -3.029 |
Token | have |
Token ID | 423 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
utenberg | +7.515 |
rament | +7.402 |
peed | +7.283 |
"$:/ | +6.985 |
annah | +6.822 |
Neg logit contributions | |
Jinping | -8.994 |
Statements | -8.810 |
NOT | -8.062 |
fore | -8.010 |
annot | -7.848 |
Token | their |
Token ID | 511 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
rament | +10.312 |
minded | +10.287 |
peed | +10.157 |
utenberg | +10.066 |
cott | +9.940 |
Neg logit contributions | |
Statements | -13.340 |
Jinping | -13.165 |
NOT | -12.354 |
fore | -11.983 |
annot | -11.949 |
Token | ballot |
Token ID | 11100 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
rament | +3.304 |
utenberg | +3.303 |
cott | +3.134 |
peed | +3.123 |
ranging | +3.015 |
Neg logit contributions | |
Jinping | -2.767 |
Statements | -2.587 |
NOT | -2.491 |
fore | -2.401 |
't | -2.298 |
Token | D |
Token ID | 360 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
Jinping | +2.254 |
Statements | +2.159 |
NOT | +2.114 |
fore | +2.017 |
't | +2.014 |
Neg logit contributions | |
rament | -2.137 |
utenberg | -2.134 |
peed | -2.017 |
cott | -1.962 |
ranging | -1.914 |
Token | . |
Token ID | 13 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
Jinping | +3.931 |
NOT | +3.730 |
Statements | +3.696 |
't | +3.455 |
annot | +3.386 |
Neg logit contributions | |
rament | -5.039 |
utenberg | -4.999 |
peed | -4.893 |
cott | -4.720 |
ranging | -4.663 |
Token | Q |
Token ID | 48 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
utenberg | +1.046 |
rament | +1.046 |
peed | +0.991 |
cott | +0.979 |
ranging | +0.945 |
Neg logit contributions | |
Jinping | -0.985 |
Statements | -0.938 |
NOT | -0.908 |
fore | -0.871 |
't | -0.854 |
Token | Ag |
Token ID | 2449 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
Jinping | +0.067 |
Statements | +0.064 |
NOT | +0.062 |
fore | +0.059 |
't | +0.058 |
Neg logit contributions | |
utenberg | -0.075 |
rament | -0.075 |
peed | -0.071 |
cott | -0.070 |
ranging | -0.068 |
Token | or |
Token ID | 273 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
utenberg | +5.679 |
rament | +5.336 |
cott | +5.222 |
peed | +5.121 |
ranging | +5.098 |
Neg logit contributions | |
Jinping | -5.358 |
Statements | -5.190 |
NOT | -4.962 |
fore | -4.594 |
� | -4.557 |
Token | ist |
Token ID | 396 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
rament | +2.133 |
utenberg | +2.127 |
peed | +2.018 |
cott | +2.006 |
annah | +1.928 |
Neg logit contributions | |
Jinping | -2.107 |
Statements | -2.001 |
NOT | -1.936 |
fore | -1.860 |
't | -1.813 |
Token | of |
Token ID | 286 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
utenberg | +1.586 |
rament | +1.582 |
peed | +1.502 |
cott | +1.487 |
ranging | +1.431 |
Neg logit contributions | |
Jinping | -1.615 |
Statements | -1.538 |
NOT | -1.494 |
fore | -1.430 |
't | -1.404 |
Token | the |
Token ID | 262 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
utenberg | +0.506 |
rament | +0.506 |
peed | +0.480 |
cott | +0.474 |
ranging | +0.458 |
Neg logit contributions | |
Jinping | -0.469 |
Statements | -0.447 |
NOT | -0.433 |
fore | -0.415 |
't | -0.407 |
Token | Free |
Token ID | 3232 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
rament | +3.206 |
utenberg | +3.189 |
cott | +3.028 |
peed | +3.020 |
annah | +2.917 |
Neg logit contributions | |
Jinping | -2.935 |
Statements | -2.748 |
NOT | -2.706 |
fore | -2.579 |
't | -2.509 |
Token | Thought |
Token ID | 27522 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
rament | +9.797 |
cott | +9.771 |
ranging | +9.494 |
utenberg | +9.432 |
peed | +9.204 |
Neg logit contributions | |
Jinping | -14.054 |
NOT | -13.246 |
fore | -12.471 |
� | -12.433 |
Statements | -12.322 |
Token | Project |
Token ID | 4935 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
rament | +4.241 |
utenberg | +4.164 |
peed | +3.987 |
cott | +3.899 |
ranging | +3.829 |
Neg logit contributions | |
Jinping | -4.755 |
Statements | -4.353 |
NOT | -4.247 |
fore | -4.214 |
annot | -4.153 |
Token | bas |
Token ID | 1615 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
rament | +3.411 |
utenberg | +3.410 |
peed | +3.276 |
cott | +3.235 |
annah | +3.107 |
Neg logit contributions | |
Jinping | -3.202 |
Statements | -3.033 |
NOT | -2.955 |
fore | -2.818 |
't | -2.741 |
Token | hed |
Token ID | 704 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Jinping | +3.507 |
Statements | +3.288 |
NOT | +3.237 |
't | +3.224 |
fore | +3.202 |
Neg logit contributions | |
utenberg | -2.719 |
rament | -2.704 |
peed | -2.498 |
"$:/ | -2.446 |
cott | -2.429 |
Token | the |
Token ID | 262 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
utenberg | +5.976 |
rament | +5.877 |
cott | +5.654 |
peed | +5.528 |
annah | +5.344 |
Neg logit contributions | |
Jinping | -6.678 |
Statements | -6.411 |
NOT | -6.385 |
fore | -6.113 |
annot | -5.893 |
Token | opathy |
Token ID | 27189 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
rament | +2.058 |
utenberg | +2.048 |
peed | +1.930 |
cott | +1.910 |
ranging | +1.834 |
Neg logit contributions | |
Jinping | -2.887 |
Statements | -2.750 |
NOT | -2.701 |
fore | -2.584 |
't | -2.561 |
Token | advocated |
Token ID | 25828 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
utenberg | +1.304 |
rament | +1.303 |
peed | +1.225 |
cott | +1.206 |
ranging | +1.157 |
Neg logit contributions | |
Jinping | -1.659 |
Statements | -1.590 |
NOT | -1.549 |
fore | -1.492 |
't | -1.470 |
Token | a |
Token ID | 257 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
rament | +4.727 |
utenberg | +4.664 |
peed | +4.470 |
cott | +4.386 |
ranging | +4.280 |
Neg logit contributions | |
Jinping | -5.848 |
Statements | -5.654 |
NOT | -5.514 |
annot | -5.254 |
fore | -5.250 |
Token | � |
Token ID | 564 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
rament | +3.514 |
utenberg | +3.494 |
peed | +3.299 |
cott | +3.267 |
"$:/ | +3.170 |
Neg logit contributions | |
Jinping | -3.953 |
Statements | -3.793 |
NOT | -3.719 |
fore | -3.504 |
annot | -3.464 |
Token | � |
Token ID | 250 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
rament | +2.955 |
utenberg | +2.950 |
peed | +2.813 |
cott | +2.776 |
ranging | +2.682 |
Neg logit contributions | |
Jinping | -2.611 |
Statements | -2.494 |
NOT | -2.398 |
fore | -2.301 |
't | -2.245 |
Token | like |
Token ID | 2339 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
utenberg | +2.738 |
rament | +2.737 |
peed | +2.600 |
cott | +2.582 |
ranging | +2.495 |
Neg logit contributions | |
Jinping | -2.634 |
Statements | -2.518 |
NOT | -2.413 |
fore | -2.300 |
't | -2.266 |
Token | cures |
Token ID | 50019 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
rament | +11.907 |
cott | +11.471 |
peed | +11.452 |
utenberg | +11.170 |
pan | +10.796 |
Neg logit contributions | |
Jinping | -12.697 |
Statements | -12.062 |
NOT | -11.400 |
fore | -10.792 |
annot | -10.753 |
Token | like |
Token ID | 588 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
peed | +10.591 |
rament | +10.478 |
cott | +10.264 |
utenberg | +10.176 |
pan | +9.857 |
Neg logit contributions | |
Jinping | -13.410 |
Statements | -12.931 |
NOT | -12.287 |
fore | -11.479 |
annot | -11.349 |
Token | � |
Token ID | 447 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
rament | +6.676 |
utenberg | +6.475 |
peed | +6.312 |
cott | +6.287 |
ranging | +6.034 |
Neg logit contributions | |
Jinping | -6.496 |
Statements | -6.295 |
NOT | -6.108 |
fore | -5.784 |
annot | -5.741 |
Token | � |
Token ID | 251 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
rament | +2.383 |
utenberg | +2.380 |
peed | +2.269 |
cott | +2.229 |
ranging | +2.165 |
Neg logit contributions | |
Jinping | -2.017 |
Statements | -1.915 |
NOT | -1.852 |
fore | -1.774 |
't | -1.725 |
Token | approach |
Token ID | 3164 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
utenberg | +0.343 |
rament | +0.342 |
peed | +0.324 |
cott | +0.319 |
ranging | +0.307 |
Neg logit contributions | |
Jinping | -0.360 |
Statements | -0.344 |
NOT | -0.334 |
fore | -0.321 |
't | -0.316 |
Token | you |
Token ID | 345 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
Jinping | +0.976 |
Statements | +0.934 |
NOT | +0.910 |
't | +0.875 |
fore | +0.875 |
Neg logit contributions | |
utenberg | -0.989 |
rament | -0.984 |
peed | -0.931 |
cott | -0.921 |
"$:/ | -0.889 |
Token | ever |
Token ID | 1683 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
rament | +1.916 |
utenberg | +1.910 |
peed | +1.830 |
cott | +1.786 |
annah | +1.743 |
Neg logit contributions | |
Jinping | -1.720 |
Statements | -1.646 |
NOT | -1.575 |
fore | -1.508 |
annot | -1.465 |
Token | wonder |
Token ID | 4240 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
Jinping | +1.347 |
Statements | +1.275 |
NOT | +1.241 |
fore | +1.187 |
't | +1.173 |
Neg logit contributions | |
utenberg | -1.540 |
rament | -1.536 |
peed | -1.453 |
cott | -1.444 |
ranging | -1.393 |
Token | what |
Token ID | 644 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
utenberg | +0.980 |
rament | +0.980 |
peed | +0.935 |
cott | +0.922 |
ranging | +0.893 |
Neg logit contributions | |
Jinping | -0.801 |
Statements | -0.761 |
NOT | -0.736 |
fore | -0.701 |
't | -0.686 |
Token | would |
Token ID | 561 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
rament | +3.401 |
utenberg | +3.400 |
peed | +3.234 |
cott | +3.173 |
ranging | +3.072 |
Neg logit contributions | |
Jinping | -3.559 |
Statements | -3.359 |
NOT | -3.301 |
fore | -3.123 |
annot | -3.059 |
Token | have |
Token ID | 423 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
utenberg | +5.812 |
rament | +5.782 |
peed | +5.524 |
cott | +5.196 |
"$:/ | +5.154 |
Neg logit contributions | |
Jinping | -11.454 |
Statements | -10.947 |
NOT | -10.551 |
fore | -10.406 |
annot | -10.231 |
Token | happened |
Token ID | 3022 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
rament | +8.119 |
peed | +8.080 |
utenberg | +7.869 |
cott | +7.836 |
minded | +7.834 |
Neg logit contributions | |
Statements | -15.816 |
Jinping | -15.491 |
NOT | -14.708 |
fore | -14.389 |
annot | -14.271 |
Token | if |
Token ID | 611 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
utenberg | +4.196 |
rament | +4.161 |
peed | +3.991 |
cott | +3.836 |
"$:/ | +3.714 |
Neg logit contributions | |
Jinping | -6.485 |
Statements | -6.347 |
NOT | -6.133 |
fore | -5.829 |
annot | -5.811 |
Token | you |
Token ID | 345 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
rament | +1.440 |
utenberg | +1.440 |
peed | +1.363 |
cott | +1.340 |
ranging | +1.293 |
Neg logit contributions | |
Jinping | -1.579 |
Statements | -1.507 |
NOT | -1.466 |
fore | -1.407 |
't | -1.380 |
Token | had |
Token ID | 550 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
rament | +3.521 |
utenberg | +3.468 |
peed | +3.343 |
cott | +3.212 |
ranging | +3.133 |
Neg logit contributions | |
Jinping | -4.594 |
Statements | -4.476 |
NOT | -4.304 |
fore | -4.136 |
annot | -4.002 |
Token | played |
Token ID | 2826 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
rament | +5.895 |
utenberg | +5.804 |
peed | +5.586 |
cott | +5.544 |
ranging | +5.431 |
Neg logit contributions | |
Jinping | -6.274 |
Statements | -5.957 |
NOT | -5.777 |
fore | -5.514 |
annot | -5.318 |
Token | out |
Token ID | 503 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
rament | +2.162 |
utenberg | +2.159 |
peed | +2.047 |
cott | +2.007 |
ranging | +1.933 |
Neg logit contributions | |
Jinping | -2.532 |
Statements | -2.425 |
NOT | -2.361 |
fore | -2.262 |
't | -2.219 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Jinping | +1.922 |
Statements | +1.827 |
NOT | +1.768 |
fore | +1.732 |
't | +1.728 |
Neg logit contributions | |
utenberg | -2.023 |
rament | -2.012 |
cott | -1.884 |
peed | -1.877 |
"$:/ | -1.819 |
Token | But |
Token ID | 887 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
rament | +0.807 |
utenberg | +0.802 |
peed | +0.774 |
cott | +0.766 |
ranging | +0.743 |
Neg logit contributions | |
Jinping | -0.687 |
Statements | -0.637 |
NOT | -0.626 |
fore | -0.600 |
annot | -0.585 |
Token | I |
Token ID | 314 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
Jinping | +0.220 |
Statements | +0.210 |
NOT | +0.204 |
fore | +0.195 |
't | +0.192 |
Neg logit contributions | |
rament | -0.207 |
utenberg | -0.207 |
peed | -0.196 |
cott | -0.193 |
ranging | -0.186 |
Token | just |
Token ID | 655 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
Jinping | +2.961 |
Statements | +2.786 |
NOT | +2.744 |
't | +2.646 |
fore | +2.637 |
Neg logit contributions | |
utenberg | -3.035 |
rament | -3.021 |
peed | -2.830 |
cott | -2.815 |
"$:/ | -2.722 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
Jinping | +0.319 |
Statements | +0.305 |
NOT | +0.296 |
fore | +0.284 |
't | +0.278 |
Neg logit contributions | |
utenberg | -0.311 |
rament | -0.311 |
peed | -0.295 |
cott | -0.291 |
ranging | -0.280 |
Token | 't |
Token ID | 470 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
't | +11.888 |
NOT | +7.606 |
fore | +5.923 |
Jinping | +5.476 |
Statements | +4.442 |
Neg logit contributions | |
proport | -30.732 |
encount | -28.929 |
exha | -28.571 |
vertisement | -27.900 |
ailability | -27.783 |
Token | bear |
Token ID | 6842 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
Jinping | +2.878 |
Statements | +2.712 |
NOT | +2.677 |
't | +2.602 |
fore | +2.548 |
Neg logit contributions | |
utenberg | -3.290 |
rament | -3.280 |
cott | -3.100 |
peed | -3.055 |
ranging | -2.960 |
Token | it |
Token ID | 340 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
rament | +3.070 |
utenberg | +3.045 |
peed | +2.913 |
cott | +2.878 |
ranging | +2.802 |
Neg logit contributions | |
Jinping | -2.865 |
Statements | -2.737 |
NOT | -2.636 |
fore | -2.524 |
annot | -2.440 |
Token | . |
Token ID | 13 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
rament | +3.650 |
utenberg | +3.640 |
peed | +3.473 |
cott | +3.404 |
annah | +3.315 |
Neg logit contributions | |
Jinping | -3.895 |
Statements | -3.779 |
NOT | -3.634 |
fore | -3.411 |
annot | -3.367 |
Token | I |
Token ID | 314 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
Jinping | +1.852 |
Statements | +1.779 |
NOT | +1.703 |
fore | +1.618 |
't | +1.604 |
Neg logit contributions | |
utenberg | -2.408 |
rament | -2.398 |
peed | -2.277 |
cott | -2.249 |
ranging | -2.177 |
Token | 199 |
Token ID | 19104 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
rament | +3.165 |
utenberg | +3.138 |
peed | +3.033 |
cott | +2.997 |
ranging | +2.901 |
Neg logit contributions | |
Jinping | -3.865 |
Statements | -3.682 |
NOT | -3.532 |
fore | -3.414 |
annot | -3.338 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
rament | +5.571 |
utenberg | +5.525 |
peed | +5.391 |
cott | +5.245 |
ranging | +5.059 |
Neg logit contributions | |
Jinping | -6.729 |
Statements | -6.511 |
NOT | -6.166 |
fore | -5.960 |
annot | -5.887 |
Token | 99 |
Token ID | 2079 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
rament | +6.039 |
utenberg | +6.008 |
peed | +5.972 |
cott | +5.855 |
ranging | +5.698 |
Neg logit contributions | |
Jinping | -6.345 |
Statements | -5.812 |
NOT | -5.605 |
fore | -5.520 |
annot | -5.480 |
Token | and |
Token ID | 290 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
utenberg | +3.291 |
rament | +3.290 |
peed | +3.152 |
cott | +3.091 |
ranging | +2.979 |
Neg logit contributions | |
Jinping | -3.817 |
Statements | -3.651 |
NOT | -3.517 |
fore | -3.381 |
annot | -3.307 |
Token | I |
Token ID | 314 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
utenberg | +0.599 |
rament | +0.599 |
peed | +0.566 |
cott | +0.558 |
ranging | +0.538 |
Neg logit contributions | |
Jinping | -0.649 |
Statements | -0.620 |
NOT | -0.603 |
fore | -0.579 |
't | -0.569 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
Jinping | +2.898 |
Statements | +2.727 |
NOT | +2.711 |
fore | +2.604 |
't | +2.603 |
Neg logit contributions | |
utenberg | -3.006 |
rament | -2.997 |
peed | -2.788 |
cott | -2.772 |
"$:/ | -2.687 |
Token | 't |
Token ID | 470 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
't | +11.583 |
NOT | +7.902 |
fore | +6.149 |
Jinping | +5.749 |
Statements | +5.175 |
Neg logit contributions | |
proport | -28.516 |
encount | -26.937 |
exha | -26.458 |
utenberg | -26.336 |
ailability | -26.127 |
Token | be |
Token ID | 307 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
Jinping | +6.521 |
NOT | +6.337 |
't | +6.316 |
Statements | +6.116 |
fore | +5.889 |
Neg logit contributions | |
rament | -8.296 |
utenberg | -8.163 |
cott | -7.809 |
ranging | -7.478 |
peed | -7.463 |
Token | more |
Token ID | 517 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
Jinping | +5.069 |
NOT | +4.616 |
Statements | +4.567 |
't | +4.551 |
fore | +4.385 |
Neg logit contributions | |
utenberg | -8.322 |
rament | -8.257 |
peed | -7.672 |
cott | -7.503 |
ranging | -7.479 |
Token | happier |
Token ID | 23030 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
Jinping | +5.131 |
Statements | +4.894 |
NOT | +4.842 |
't | +4.686 |
fore | +4.649 |
Neg logit contributions | |
utenberg | -4.493 |
rament | -4.405 |
peed | -4.011 |
cott | -3.989 |
ranging | -3.972 |
Token | . |
Token ID | 13 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
Jinping | +0.220 |
Statements | +0.210 |
NOT | +0.205 |
fore | +0.197 |
't | +0.195 |
Neg logit contributions | |
utenberg | -0.211 |
rament | -0.210 |
peed | -0.199 |
cott | -0.197 |
ranging | -0.189 |
Token | first |
Token ID | 717 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
Jinping | +4.887 |
Statements | +4.554 |
NOT | +4.466 |
fore | +4.384 |
't | +4.380 |
Neg logit contributions | |
utenberg | -4.890 |
rament | -4.844 |
cott | -4.681 |
peed | -4.625 |
ranging | -4.475 |
Token | real |
Token ID | 1103 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
Jinping | +3.015 |
Statements | +2.877 |
NOT | +2.818 |
fore | +2.722 |
't | +2.679 |
Neg logit contributions | |
utenberg | -2.889 |
rament | -2.884 |
peed | -2.736 |
cott | -2.729 |
ranging | -2.598 |
Token | relationship |
Token ID | 2776 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
Jinping | +5.426 |
Statements | +5.151 |
NOT | +5.079 |
fore | +4.920 |
't | +4.904 |
Neg logit contributions | |
rament | -5.027 |
utenberg | -5.009 |
cott | -4.781 |
peed | -4.747 |
ranging | -4.480 |
Token | and |
Token ID | 290 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
Jinping | +5.161 |
NOT | +4.891 |
Statements | +4.884 |
fore | +4.723 |
't | +4.712 |
Neg logit contributions | |
utenberg | -5.090 |
rament | -5.074 |
cott | -4.766 |
peed | -4.759 |
ranging | -4.560 |
Token | I |
Token ID | 314 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
Jinping | +2.998 |
Statements | +2.851 |
NOT | +2.773 |
fore | +2.694 |
't | +2.672 |
Neg logit contributions | |
utenberg | -3.407 |
rament | -3.365 |
cott | -3.208 |
peed | -3.189 |
ranging | -3.060 |
Token | don |
Token ID | 836 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
Jinping | +4.908 |
NOT | +4.548 |
't | +4.503 |
Statements | +4.481 |
fore | +4.442 |
Neg logit contributions | |
utenberg | -5.967 |
rament | -5.816 |
cott | -5.513 |
peed | -5.508 |
ranging | -5.272 |
Token | 't |
Token ID | 470 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
't | +11.527 |
NOT | +7.597 |
fore | +6.331 |
Jinping | +5.779 |
Statements | +5.140 |
Neg logit contributions | |
proport | -29.082 |
encount | -27.749 |
exha | -27.134 |
ailability | -26.478 |
vertisement | -26.392 |
Token | want |
Token ID | 765 |
Feature activation | -8.02e-02 |
Pos logit contributions | |
Jinping | +2.323 |
Statements | +2.190 |
NOT | +2.162 |
't | +2.088 |
fore | +2.074 |
Neg logit contributions | |
utenberg | -2.410 |
rament | -2.393 |
cott | -2.260 |
peed | -2.240 |
ranging | -2.164 |
Token | to |
Token ID | 284 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
Jinping | +0.466 |
Statements | +0.443 |
NOT | +0.429 |
fore | +0.411 |
't | +0.404 |
Neg logit contributions | |
utenberg | -0.570 |
rament | -0.567 |
peed | -0.541 |
cott | -0.534 |
ranging | -0.517 |
Token | screw |
Token ID | 9580 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
rament | +1.635 |
utenberg | +1.632 |
peed | +1.555 |
cott | +1.513 |
annah | +1.482 |
Neg logit contributions | |
Jinping | -1.652 |
Statements | -1.581 |
NOT | -1.515 |
fore | -1.456 |
annot | -1.427 |
Token | it |
Token ID | 340 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
Jinping | +0.770 |
Statements | +0.740 |
NOT | +0.720 |
fore | +0.696 |
't | +0.690 |
Neg logit contributions | |
utenberg | -0.630 |
rament | -0.628 |
peed | -0.590 |
cott | -0.582 |
ranging | -0.561 |
Token | include |
Token ID | 2291 |
Feature activation | +7.85e-01 |
Pos logit contributions | |
utenberg | +5.754 |
rament | +5.684 |
peed | +5.508 |
"$:/ | +5.199 |
annah | +5.160 |
Neg logit contributions | |
Jinping | -6.380 |
Statements | -6.167 |
NOT | -5.791 |
fore | -5.672 |
annot | -5.651 |
Token | during |
Token ID | 1141 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
rament | +4.364 |
utenberg | +4.354 |
peed | +4.186 |
cott | +4.010 |
ranging | +3.920 |
Neg logit contributions | |
Jinping | -5.571 |
Statements | -5.368 |
NOT | -5.217 |
fore | -4.999 |
annot | -4.913 |
Token | the |
Token ID | 262 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
utenberg | +2.234 |
rament | +2.234 |
peed | +2.106 |
cott | +2.056 |
ranging | +1.977 |
Neg logit contributions | |
Jinping | -3.044 |
Statements | -2.916 |
NOT | -2.855 |
fore | -2.738 |
't | -2.690 |
Token | development |
Token ID | 2478 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
rament | +2.378 |
utenberg | +2.373 |
peed | +2.238 |
cott | +2.172 |
"$:/ | +2.097 |
Neg logit contributions | |
Jinping | -3.538 |
Statements | -3.402 |
NOT | -3.337 |
fore | -3.194 |
annot | -3.158 |
Token | but |
Token ID | 475 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
Jinping | +1.560 |
Statements | +1.486 |
NOT | +1.443 |
fore | +1.401 |
't | +1.395 |
Neg logit contributions | |
utenberg | -1.492 |
rament | -1.487 |
cott | -1.388 |
peed | -1.385 |
annah | -1.327 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
rament | +0.350 |
utenberg | +0.349 |
peed | +0.332 |
cott | +0.327 |
ranging | +0.317 |
Neg logit contributions | |
Jinping | -0.333 |
Statements | -0.317 |
NOT | -0.308 |
fore | -0.294 |
't | -0.288 |
Token | 't |
Token ID | 470 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
't | +10.366 |
NOT | +6.674 |
fore | +5.280 |
Jinping | +4.857 |
Statements | +3.774 |
Neg logit contributions | |
proport | -29.037 |
encount | -27.288 |
rament | -27.252 |
utenberg | -27.176 |
exha | -26.739 |
Token | fit |
Token ID | 4197 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
utenberg | +0.643 |
rament | +0.642 |
peed | +0.607 |
cott | +0.600 |
ranging | +0.578 |
Neg logit contributions | |
Jinping | -0.654 |
Statements | -0.623 |
NOT | -0.606 |
fore | -0.582 |
't | -0.573 |
Token | in |
Token ID | 287 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
utenberg | +0.922 |
rament | +0.921 |
peed | +0.870 |
cott | +0.858 |
ranging | +0.826 |
Neg logit contributions | |
Jinping | -1.003 |
Statements | -0.959 |
NOT | -0.932 |
fore | -0.895 |
't | -0.880 |
Token | - |
Token ID | 532 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
rament | +1.813 |
utenberg | +1.809 |
peed | +1.715 |
cott | +1.679 |
ranging | +1.625 |
Neg logit contributions | |
Jinping | -2.078 |
Statements | -1.984 |
NOT | -1.941 |
fore | -1.855 |
't | -1.812 |
Token | those |
Token ID | 883 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
Jinping | +1.636 |
Statements | +1.567 |
NOT | +1.523 |
fore | +1.464 |
't | +1.451 |
Neg logit contributions | |
utenberg | -1.502 |
rament | -1.502 |
peed | -1.409 |
cott | -1.402 |
"$:/ | -1.342 |
Token | home |
Token ID | 1363 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
rament | +1.417 |
utenberg | +1.416 |
peed | +1.351 |
cott | +1.334 |
ranging | +1.292 |
Neg logit contributions | |
Jinping | -1.169 |
Statements | -1.114 |
NOT | -1.077 |
fore | -1.024 |
't | -0.995 |
Token | for |
Token ID | 329 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
rament | +1.780 |
utenberg | +1.769 |
peed | +1.682 |
cott | +1.653 |
ranging | +1.597 |
Neg logit contributions | |
Jinping | -1.972 |
Statements | -1.897 |
NOT | -1.842 |
fore | -1.762 |
't | -1.724 |
Token | Christmas |
Token ID | 6786 |
Feature activation | -9.35e-02 |
Pos logit contributions | |
Jinping | +1.127 |
Statements | +1.071 |
NOT | +1.033 |
fore | +0.998 |
't | +0.989 |
Neg logit contributions | |
utenberg | -1.156 |
rament | -1.153 |
peed | -1.087 |
cott | -1.071 |
ranging | -1.037 |
Token | and |
Token ID | 290 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
Jinping | +0.640 |
Statements | +0.609 |
NOT | +0.590 |
fore | +0.571 |
't | +0.564 |
Neg logit contributions | |
utenberg | -0.575 |
rament | -0.566 |
peed | -0.537 |
cott | -0.530 |
"$:/ | -0.510 |
Token | I |
Token ID | 314 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
Jinping | +3.293 |
Statements | +3.099 |
NOT | +3.002 |
't | +2.903 |
fore | +2.901 |
Neg logit contributions | |
utenberg | -3.391 |
rament | -3.332 |
peed | -3.153 |
cott | -3.120 |
ranging | -3.033 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
Jinping | +4.554 |
Statements | +4.177 |
NOT | +4.175 |
't | +4.136 |
fore | +4.019 |
Neg logit contributions | |
utenberg | -4.757 |
rament | -4.718 |
peed | -4.422 |
cott | -4.381 |
owship | -4.270 |
Token | 't |
Token ID | 470 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
't | +10.478 |
NOT | +6.795 |
fore | +5.204 |
Jinping | +4.907 |
Statements | +4.151 |
Neg logit contributions | |
proport | -29.241 |
encount | -27.669 |
exha | -27.072 |
utenberg | -26.608 |
ailability | -26.559 |
Token | find |
Token ID | 1064 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
Jinping | +4.233 |
NOT | +3.957 |
Statements | +3.919 |
't | +3.898 |
fore | +3.725 |
Neg logit contributions | |
utenberg | -4.900 |
rament | -4.893 |
cott | -4.639 |
peed | -4.520 |
ranging | -4.441 |
Token | a |
Token ID | 257 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
Jinping | +1.071 |
Statements | +1.021 |
NOT | +0.994 |
fore | +0.958 |
't | +0.948 |
Neg logit contributions | |
utenberg | -0.925 |
rament | -0.917 |
peed | -0.862 |
cott | -0.850 |
ranging | -0.823 |
Token | place |
Token ID | 1295 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
Jinping | +2.161 |
Statements | +2.059 |
NOT | +2.012 |
fore | +1.943 |
't | +1.918 |
Neg logit contributions | |
utenberg | -1.748 |
rament | -1.730 |
peed | -1.629 |
cott | -1.605 |
ranging | -1.548 |
Token | to |
Token ID | 284 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
Jinping | +2.267 |
Statements | +2.088 |
NOT | +2.072 |
fore | +2.012 |
't | +1.994 |
Neg logit contributions | |
utenberg | -3.217 |
rament | -3.134 |
peed | -2.962 |
cott | -2.958 |
ranging | -2.887 |
Token | in |
Token ID | 287 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
rament | +4.970 |
utenberg | +4.968 |
peed | +4.760 |
cott | +4.685 |
ranging | +4.548 |
Neg logit contributions | |
Jinping | -3.867 |
Statements | -3.642 |
NOT | -3.506 |
fore | -3.333 |
annot | -3.221 |
Token | there |
Token ID | 612 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
rament | +1.844 |
utenberg | +1.842 |
peed | +1.742 |
cott | +1.712 |
ranging | +1.651 |
Neg logit contributions | |
Jinping | -2.065 |
Statements | -1.973 |
NOT | -1.923 |
fore | -1.846 |
't | -1.807 |
Token | , |
Token ID | 11 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
utenberg | +0.766 |
rament | +0.765 |
peed | +0.721 |
cott | +0.713 |
ranging | +0.685 |
Neg logit contributions | |
Jinping | -0.824 |
Statements | -0.786 |
NOT | -0.765 |
fore | -0.737 |
't | -0.726 |
Token | but |
Token ID | 475 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
utenberg | +2.751 |
rament | +2.731 |
peed | +2.627 |
cott | +2.536 |
ranging | +2.487 |
Neg logit contributions | |
Jinping | -3.445 |
Statements | -3.318 |
NOT | -3.219 |
fore | -3.093 |
annot | -3.025 |
Token | unfortunately |
Token ID | 12716 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
rament | +2.019 |
utenberg | +2.004 |
peed | +1.940 |
cott | +1.868 |
ranging | +1.828 |
Neg logit contributions | |
Jinping | -2.229 |
Statements | -2.141 |
NOT | -2.072 |
fore | -1.982 |
annot | -1.939 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
Jinping | +1.353 |
Statements | +1.291 |
NOT | +1.256 |
fore | +1.207 |
't | +1.200 |
Neg logit contributions | |
utenberg | -1.300 |
rament | -1.287 |
peed | -1.207 |
cott | -1.206 |
"$:/ | -1.155 |
Token | 't |
Token ID | 470 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
't | +9.806 |
NOT | +6.172 |
fore | +4.625 |
Jinping | +4.371 |
annot | +3.271 |
Neg logit contributions | |
proport | -29.044 |
encount | -27.356 |
utenberg | -27.225 |
rament | -26.978 |
exha | -26.761 |
Token | . |
Token ID | 13 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
utenberg | +0.870 |
rament | +0.870 |
peed | +0.818 |
cott | +0.805 |
ranging | +0.773 |
Neg logit contributions | |
Jinping | -1.093 |
Statements | -1.048 |
NOT | -1.020 |
fore | -0.983 |
't | -0.966 |
Token | The |
Token ID | 383 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
rament | +2.301 |
utenberg | +2.297 |
peed | +2.228 |
cott | +2.181 |
ranging | +2.131 |
Neg logit contributions | |
Jinping | -1.967 |
Statements | -1.824 |
NOT | -1.799 |
fore | -1.722 |
annot | -1.688 |
Token | other |
Token ID | 584 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
rament | +2.162 |
utenberg | +2.155 |
peed | +2.052 |
cott | +2.015 |
ranging | +1.944 |
Neg logit contributions | |
Jinping | -2.295 |
Statements | -2.187 |
NOT | -2.129 |
fore | -2.037 |
't | -1.992 |
Token | thing |
Token ID | 1517 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
rament | +3.559 |
utenberg | +3.505 |
peed | +3.347 |
cott | +3.261 |
annah | +3.123 |
Neg logit contributions | |
Jinping | -5.356 |
Statements | -5.143 |
NOT | -5.007 |
fore | -4.807 |
annot | -4.751 |
Token | good |
Token ID | 922 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
Jinping | +0.865 |
Statements | +0.822 |
NOT | +0.802 |
fore | +0.771 |
't | +0.763 |
Neg logit contributions | |
utenberg | -0.799 |
rament | -0.797 |
peed | -0.751 |
cott | -0.742 |
ranging | -0.714 |
Token | enough |
Token ID | 1576 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
rament | +1.721 |
utenberg | +1.719 |
peed | +1.604 |
cott | +1.576 |
ranging | +1.508 |
Neg logit contributions | |
Jinping | -2.735 |
Statements | -2.633 |
NOT | -2.565 |
fore | -2.477 |
't | -2.434 |
Token | for |
Token ID | 329 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
utenberg | +0.927 |
rament | +0.926 |
peed | +0.876 |
cott | +0.864 |
ranging | +0.832 |
Neg logit contributions | |
Jinping | -0.973 |
Statements | -0.930 |
NOT | -0.903 |
fore | -0.867 |
't | -0.854 |
Token | those |
Token ID | 883 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
rament | +1.809 |
utenberg | +1.807 |
peed | +1.707 |
cott | +1.677 |
ranging | +1.616 |
Neg logit contributions | |
Jinping | -2.245 |
Statements | -2.155 |
NOT | -2.098 |
fore | -2.018 |
't | -1.973 |
Token | who |
Token ID | 508 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
Jinping | +1.000 |
Statements | +0.950 |
NOT | +0.922 |
fore | +0.882 |
't | +0.872 |
Neg logit contributions | |
utenberg | -1.088 |
rament | -1.087 |
peed | -1.027 |
cott | -1.017 |
ranging | -0.979 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
Jinping | +2.982 |
Statements | +2.808 |
NOT | +2.791 |
't | +2.715 |
fore | +2.690 |
Neg logit contributions | |
utenberg | -2.686 |
rament | -2.664 |
peed | -2.494 |
cott | -2.455 |
ranging | -2.390 |
Token | 't |
Token ID | 470 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
't | +12.100 |
NOT | +8.936 |
fore | +7.467 |
Jinping | +6.900 |
Statements | +6.638 |
Neg logit contributions | |
proport | -25.830 |
encount | -24.705 |
exha | -24.223 |
utenberg | -24.060 |
rament | -23.939 |
Token | get |
Token ID | 651 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
Jinping | +4.748 |
't | +4.529 |
NOT | +4.504 |
Statements | +4.395 |
fore | +4.261 |
Neg logit contributions | |
rament | -5.644 |
utenberg | -5.588 |
cott | -5.163 |
peed | -5.160 |
ranging | -5.023 |
Token | coverage |
Token ID | 5197 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
Jinping | +1.944 |
Statements | +1.858 |
NOT | +1.800 |
't | +1.753 |
fore | +1.743 |
Neg logit contributions | |
utenberg | -1.636 |
rament | -1.624 |
peed | -1.502 |
cott | -1.497 |
ranging | -1.440 |
Token | through |
Token ID | 832 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
Jinping | +1.228 |
Statements | +1.163 |
NOT | +1.132 |
't | +1.102 |
fore | +1.097 |
Neg logit contributions | |
rament | -1.148 |
utenberg | -1.145 |
cott | -1.066 |
peed | -1.063 |
annah | -1.024 |
Token | their |
Token ID | 511 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
Jinping | +3.425 |
Statements | +3.289 |
NOT | +3.176 |
't | +3.103 |
fore | +3.072 |
Neg logit contributions | |
utenberg | -3.611 |
rament | -3.597 |
cott | -3.332 |
peed | -3.320 |
"$:/ | -3.196 |
Token |
|
Token ID | 198 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
Jinping | +1.618 |
Statements | +1.555 |
NOT | +1.494 |
fore | +1.418 |
't | +1.408 |
Neg logit contributions | |
rament | -2.012 |
utenberg | -2.010 |
peed | -1.903 |
cott | -1.881 |
ranging | -1.825 |
Token |
|
Token ID | 198 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
Jinping | +1.557 |
Statements | +1.480 |
NOT | +1.439 |
fore | +1.372 |
't | +1.355 |
Neg logit contributions | |
utenberg | -1.877 |
rament | -1.875 |
peed | -1.779 |
cott | -1.758 |
ranging | -1.699 |
Token | No |
Token ID | 2949 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
Jinping | +3.713 |
NOT | +3.682 |
Statements | +3.581 |
fore | +3.438 |
't | +3.398 |
Neg logit contributions | |
utenberg | -4.537 |
rament | -4.534 |
peed | -4.234 |
cott | -4.120 |
"$:/ | -4.114 |
Token | , |
Token ID | 11 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Jinping | +1.804 |
Statements | +1.729 |
NOT | +1.678 |
fore | +1.628 |
't | +1.610 |
Neg logit contributions | |
utenberg | -1.491 |
rament | -1.483 |
peed | -1.395 |
cott | -1.376 |
"$:/ | -1.323 |
Token | I |
Token ID | 314 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
Jinping | +7.036 |
Statements | +6.656 |
NOT | +6.566 |
't | +6.396 |
fore | +6.098 |
Neg logit contributions | |
utenberg | -7.615 |
rament | -7.541 |
peed | -7.227 |
cott | -7.189 |
annah | -6.961 |
Token | don |
Token ID | 836 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
't | +9.384 |
Jinping | +9.310 |
NOT | +8.841 |
fore | +8.217 |
Statements | +8.087 |
Neg logit contributions | |
utenberg | -14.154 |
rament | -14.081 |
owship | -13.501 |
cott | -13.038 |
"$:/ | -12.837 |
Token | 't |
Token ID | 470 |
Feature activation | -3.85e-01 |
Pos logit contributions | |
't | +9.831 |
NOT | +6.239 |
fore | +4.564 |
Jinping | +4.383 |
Statements | +3.418 |
Neg logit contributions | |
proport | -29.473 |
encount | -28.069 |
exha | -27.628 |
utenberg | -26.991 |
neighb | -26.899 |
Token | know |
Token ID | 760 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
Jinping | +2.235 |
Statements | +2.094 |
NOT | +2.068 |
't | +1.970 |
fore | +1.961 |
Neg logit contributions | |
utenberg | -2.738 |
rament | -2.732 |
peed | -2.578 |
cott | -2.575 |
ranging | -2.490 |
Token | the |
Token ID | 262 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
rament | +0.329 |
utenberg | +0.329 |
peed | +0.311 |
cott | +0.307 |
ranging | +0.296 |
Neg logit contributions | |
Jinping | -0.332 |
Statements | -0.317 |
NOT | -0.308 |
fore | -0.295 |
't | -0.290 |
Token | OG |
Token ID | 34498 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
Jinping | +1.775 |
Statements | +1.684 |
NOT | +1.638 |
fore | +1.582 |
't | +1.571 |
Neg logit contributions | |
utenberg | -1.552 |
rament | -1.545 |
peed | -1.448 |
cott | -1.443 |
annah | -1.388 |
Token | . |
Token ID | 13 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
Jinping | +4.911 |
NOT | +4.692 |
Statements | +4.633 |
fore | +4.421 |
't | +4.395 |
Neg logit contributions | |
utenberg | -5.225 |
rament | -5.117 |
peed | -4.885 |
cott | -4.820 |
owship | -4.754 |
Token | in |
Token ID | 287 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
Jinping | +1.868 |
Statements | +1.758 |
NOT | +1.690 |
fore | +1.653 |
't | +1.635 |
Neg logit contributions | |
rament | -2.015 |
utenberg | -2.014 |
peed | -1.907 |
cott | -1.877 |
ranging | -1.833 |
Token | real |
Token ID | 1103 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
rament | +2.863 |
utenberg | +2.855 |
peed | +2.716 |
cott | +2.667 |
ranging | +2.562 |
Neg logit contributions | |
Jinping | -3.396 |
Statements | -3.203 |
NOT | -3.157 |
fore | -3.016 |
't | -2.921 |
Token | life |
Token ID | 1204 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
Jinping | +2.232 |
Statements | +2.149 |
NOT | +2.054 |
fore | +2.015 |
't | +1.990 |
Neg logit contributions | |
utenberg | -3.241 |
rament | -3.214 |
cott | -3.060 |
peed | -3.040 |
"$:/ | -2.948 |
Token | except |
Token ID | 2845 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
Jinping | +1.412 |
Statements | +1.340 |
NOT | +1.296 |
fore | +1.257 |
't | +1.239 |
Neg logit contributions | |
rament | -1.654 |
utenberg | -1.654 |
peed | -1.563 |
cott | -1.554 |
"$:/ | -1.494 |
Token | he |
Token ID | 339 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
rament | +0.443 |
utenberg | +0.443 |
peed | +0.419 |
cott | +0.412 |
ranging | +0.397 |
Neg logit contributions | |
Jinping | -0.504 |
Statements | -0.482 |
NOT | -0.468 |
fore | -0.450 |
't | -0.442 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
rament | +0.172 |
utenberg | +0.172 |
peed | +0.163 |
cott | +0.161 |
ranging | +0.155 |
Neg logit contributions | |
Jinping | -0.172 |
Statements | -0.164 |
NOT | -0.159 |
fore | -0.152 |
't | -0.149 |
Token | 't |
Token ID | 470 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
't | +10.753 |
NOT | +7.143 |
fore | +5.812 |
Jinping | +5.334 |
Statements | +4.974 |
Neg logit contributions | |
proport | -28.037 |
encount | -26.523 |
exha | -25.932 |
rament | -25.376 |
ailability | -25.366 |
Token | play |
Token ID | 711 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
Jinping | +3.574 |
Statements | +3.414 |
NOT | +3.385 |
't | +3.315 |
fore | +3.225 |
Neg logit contributions | |
utenberg | -3.435 |
rament | -3.415 |
cott | -3.223 |
peed | -3.184 |
ranging | -3.078 |
Token | the |
Token ID | 262 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
Jinping | +1.194 |
Statements | +1.145 |
NOT | +1.112 |
fore | +1.073 |
't | +1.058 |
Neg logit contributions | |
utenberg | -1.062 |
rament | -1.054 |
peed | -0.996 |
cott | -0.983 |
ranging | -0.946 |
Token | piano |
Token ID | 19132 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
Jinping | +1.980 |
Statements | +1.901 |
NOT | +1.854 |
fore | +1.797 |
't | +1.774 |
Neg logit contributions | |
utenberg | -1.481 |
rament | -1.458 |
peed | -1.379 |
cott | -1.357 |
ranging | -1.296 |
Token | as |
Token ID | 355 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
Jinping | +2.923 |
Statements | +2.741 |
NOT | +2.676 |
fore | +2.623 |
't | +2.587 |
Neg logit contributions | |
utenberg | -3.289 |
rament | -3.219 |
peed | -3.043 |
cott | -3.041 |
annah | -2.941 |
Token | Bay |
Token ID | 4696 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
utenberg | +0.158 |
rament | +0.157 |
peed | +0.150 |
cott | +0.147 |
"$:/ | +0.141 |
Neg logit contributions | |
Jinping | -0.186 |
Statements | -0.178 |
NOT | -0.173 |
fore | -0.166 |
't | -0.165 |
Token | Times |
Token ID | 3782 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
rament | +2.454 |
utenberg | +2.449 |
peed | +2.360 |
cott | +2.340 |
ranging | +2.291 |
Neg logit contributions | |
Jinping | -0.817 |
Statements | -0.735 |
NOT | -0.691 |
fore | -0.632 |
't | -0.600 |
Token | he |
Token ID | 339 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
rament | +2.086 |
utenberg | +2.083 |
peed | +1.967 |
cott | +1.938 |
ranging | +1.861 |
Neg logit contributions | |
Jinping | -2.407 |
Statements | -2.301 |
NOT | -2.239 |
fore | -2.151 |
't | -2.112 |
Token | " |
Token ID | 366 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
Jinping | +0.361 |
Statements | +0.345 |
NOT | +0.335 |
fore | +0.321 |
't | +0.314 |
Neg logit contributions | |
rament | -0.356 |
utenberg | -0.356 |
peed | -0.337 |
cott | -0.332 |
ranging | -0.320 |
Token | could |
Token ID | 24089 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
Jinping | +4.359 |
NOT | +4.221 |
fore | +4.132 |
Statements | +4.104 |
't | +4.000 |
Neg logit contributions | |
utenberg | -4.751 |
rament | -4.711 |
peed | -4.450 |
"$:/ | -4.376 |
cott | -4.243 |
Token | n |
Token ID | 77 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
Jinping | +3.197 |
NOT | +3.096 |
Statements | +2.973 |
't | +2.955 |
fore | +2.834 |
Neg logit contributions | |
rament | -4.861 |
utenberg | -4.785 |
peed | -4.629 |
cott | -4.574 |
ranging | -4.399 |
Token | 't |
Token ID | 470 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
't | +11.690 |
NOT | +8.839 |
fore | +7.273 |
Jinping | +7.053 |
Statements | +6.294 |
Neg logit contributions | |
proport | -25.096 |
rament | -23.897 |
utenberg | -23.862 |
encount | -23.766 |
anmar | -23.238 |
Token | stand |
Token ID | 1302 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
Jinping | +2.528 |
Statements | +2.358 |
NOT | +2.335 |
't | +2.250 |
fore | +2.199 |
Neg logit contributions | |
rament | -3.886 |
utenberg | -3.878 |
cott | -3.690 |
peed | -3.647 |
ranging | -3.569 |
Token | " |
Token ID | 1 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
rament | +4.924 |
utenberg | +4.723 |
peed | +4.563 |
cott | +4.536 |
ranging | +4.361 |
Neg logit contributions | |
Jinping | -5.592 |
Statements | -5.340 |
NOT | -5.161 |
fore | -4.916 |
annot | -4.743 |
Token | to |
Token ID | 284 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
rament | +2.846 |
utenberg | +2.833 |
peed | +2.672 |
cott | +2.654 |
ranging | +2.574 |
Neg logit contributions | |
Jinping | -3.142 |
Statements | -3.014 |
NOT | -2.925 |
fore | -2.792 |
't | -2.717 |
Token | bury |
Token ID | 30033 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
rament | +3.997 |
utenberg | +3.984 |
peed | +3.833 |
cott | +3.757 |
"$:/ | +3.733 |
Neg logit contributions | |
Jinping | -3.044 |
Statements | -2.871 |
NOT | -2.685 |
fore | -2.602 |
annot | -2.475 |
Token | the |
Token ID | 262 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
utenberg | +1.321 |
rament | +1.319 |
peed | +1.248 |
cott | +1.231 |
ranging | +1.186 |
Neg logit contributions | |
Jinping | -1.376 |
Statements | -1.314 |
NOT | -1.275 |
fore | -1.225 |
't | -1.204 |
Token | yard |
Token ID | 12699 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
utenberg | +1.105 |
rament | +1.105 |
peed | +1.042 |
cott | +1.027 |
ranging | +0.987 |
Neg logit contributions | |
Jinping | -1.297 |
Statements | -1.240 |
NOT | -1.208 |
fore | -1.161 |
't | -1.142 |
Token | , |
Token ID | 11 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
rament | +2.225 |
utenberg | +2.221 |
peed | +2.106 |
cott | +2.081 |
ranging | +2.007 |
Neg logit contributions | |
Jinping | -2.509 |
Statements | -2.408 |
NOT | -2.338 |
fore | -2.232 |
't | -2.185 |
Token | though |
Token ID | 996 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
utenberg | +2.060 |
rament | +2.048 |
peed | +1.928 |
cott | +1.909 |
ranging | +1.846 |
Neg logit contributions | |
Jinping | -2.649 |
Statements | -2.543 |
NOT | -2.481 |
fore | -2.383 |
't | -2.333 |
Token | he |
Token ID | 339 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
Jinping | +1.295 |
Statements | +1.239 |
NOT | +1.203 |
fore | +1.156 |
't | +1.150 |
Neg logit contributions | |
utenberg | -1.193 |
rament | -1.185 |
peed | -1.122 |
cott | -1.103 |
"$:/ | -1.060 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
Jinping | +1.397 |
Statements | +1.314 |
NOT | +1.277 |
fore | +1.232 |
't | +1.224 |
Neg logit contributions | |
utenberg | -1.582 |
rament | -1.576 |
peed | -1.493 |
cott | -1.475 |
ranging | -1.428 |
Token | � |
Token ID | 447 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
't | +10.828 |
NOT | +7.492 |
fore | +6.184 |
Jinping | +5.758 |
Statements | +5.377 |
Neg logit contributions | |
proport | -26.817 |
utenberg | -25.346 |
encount | -25.343 |
rament | -25.107 |
anmar | -24.795 |
Token | � |
Token ID | 247 |
Feature activation | -8.78e-03 |
Pos logit contributions | |
rament | +1.758 |
utenberg | +1.751 |
peed | +1.665 |
cott | +1.635 |
ranging | +1.566 |
Neg logit contributions | |
Jinping | -2.292 |
Statements | -2.203 |
NOT | -2.133 |
fore | -2.048 |
't | -2.007 |
Token | t |
Token ID | 83 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
Jinping | +0.048 |
Statements | +0.046 |
NOT | +0.044 |
fore | +0.042 |
't | +0.041 |
Neg logit contributions | |
rament | -0.065 |
utenberg | -0.065 |
peed | -0.062 |
cott | -0.061 |
ranging | -0.059 |
Token | remember |
Token ID | 3505 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
Jinping | +3.498 |
Statements | +3.280 |
NOT | +3.218 |
't | +3.134 |
fore | +3.097 |
Neg logit contributions | |
rament | -5.287 |
utenberg | -5.250 |
cott | -4.994 |
peed | -4.966 |
ranging | -4.797 |
Token | what |
Token ID | 644 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
rament | +3.508 |
utenberg | +3.431 |
peed | +3.298 |
cott | +3.288 |
ranging | +3.179 |
Neg logit contributions | |
Jinping | -3.821 |
Statements | -3.699 |
NOT | -3.625 |
fore | -3.445 |
annot | -3.372 |
Token | bankruptcy |
Token ID | 17235 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Jinping | +2.298 |
Statements | +2.187 |
NOT | +2.130 |
fore | +2.061 |
't | +2.030 |
Neg logit contributions | |
utenberg | -2.216 |
rament | -2.197 |
peed | -2.076 |
cott | -2.047 |
"$:/ | -1.974 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Jinping | +5.996 |
Statements | +5.662 |
fore | +5.493 |
NOT | +5.478 |
't | +5.384 |
Neg logit contributions | |
utenberg | -7.015 |
rament | -6.779 |
cott | -6.530 |
peed | -6.522 |
"$:/ | -6.303 |
Token | " |
Token ID | 366 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
utenberg | +1.284 |
rament | +1.284 |
peed | +1.223 |
cott | +1.210 |
ranging | +1.171 |
Neg logit contributions | |
Jinping | -1.153 |
Statements | -1.086 |
NOT | -1.060 |
fore | -1.017 |
't | -0.989 |
Token | I |
Token ID | 40 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
Jinping | +0.829 |
Statements | +0.795 |
NOT | +0.777 |
fore | +0.739 |
't | +0.727 |
Neg logit contributions | |
utenberg | -0.956 |
rament | -0.952 |
peed | -0.902 |
cott | -0.889 |
"$:/ | -0.861 |
Token | just |
Token ID | 655 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Jinping | +2.688 |
Statements | +2.521 |
NOT | +2.486 |
fore | +2.390 |
't | +2.389 |
Neg logit contributions | |
utenberg | -2.784 |
rament | -2.766 |
peed | -2.615 |
cott | -2.584 |
"$:/ | -2.507 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
rament | +0.755 |
utenberg | +0.755 |
peed | +0.717 |
cott | +0.707 |
annah | +0.679 |
Neg logit contributions | |
Jinping | -0.820 |
Statements | -0.788 |
NOT | -0.760 |
fore | -0.729 |
't | -0.710 |
Token | 't |
Token ID | 470 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
't | +9.925 |
NOT | +6.336 |
fore | +4.906 |
Jinping | +4.664 |
Statements | +3.805 |
Neg logit contributions | |
proport | -28.589 |
encount | -26.999 |
exha | -26.489 |
utenberg | -25.943 |
rament | -25.925 |
Token | stand |
Token ID | 1302 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
Jinping | +2.714 |
Statements | +2.553 |
NOT | +2.525 |
't | +2.446 |
fore | +2.401 |
Neg logit contributions | |
utenberg | -2.985 |
rament | -2.975 |
cott | -2.817 |
peed | -2.781 |
ranging | -2.699 |
Token | being |
Token ID | 852 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
rament | +2.550 |
utenberg | +2.496 |
peed | +2.384 |
cott | +2.360 |
annah | +2.280 |
Neg logit contributions | |
Jinping | -3.023 |
Statements | -2.899 |
NOT | -2.816 |
fore | -2.697 |
annot | -2.625 |
Token | a |
Token ID | 257 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
rament | +1.028 |
utenberg | +1.023 |
peed | +0.970 |
cott | +0.955 |
ranging | +0.913 |
Neg logit contributions | |
Jinping | -1.253 |
Statements | -1.204 |
NOT | -1.169 |
fore | -1.125 |
't | -1.101 |
Token | lawyer |
Token ID | 6853 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
Jinping | +3.216 |
Statements | +3.046 |
NOT | +2.972 |
fore | +2.889 |
't | +2.845 |
Neg logit contributions | |
utenberg | -3.540 |
rament | -3.490 |
peed | -3.346 |
cott | -3.281 |
"$:/ | -3.197 |
Token | past |
Token ID | 1613 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
Jinping | +3.064 |
Statements | +2.937 |
NOT | +2.767 |
fore | +2.708 |
't | +2.676 |
Neg logit contributions | |
utenberg | -3.657 |
rament | -3.612 |
peed | -3.403 |
cott | -3.399 |
"$:/ | -3.281 |
Token | — |
Token ID | 851 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
Jinping | +5.128 |
Statements | +4.787 |
fore | +4.630 |
't | +4.614 |
NOT | +4.602 |
Neg logit contributions | |
utenberg | -5.587 |
rament | -5.483 |
peed | -5.166 |
cott | -5.096 |
"$:/ | -5.019 |
Token | patients |
Token ID | 3871 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Jinping | +3.564 |
Statements | +3.390 |
NOT | +3.309 |
fore | +3.209 |
't | +3.185 |
Neg logit contributions | |
utenberg | -3.340 |
rament | -3.336 |
peed | -3.126 |
cott | -3.093 |
"$:/ | -2.966 |
Token | who |
Token ID | 508 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
utenberg | +0.159 |
rament | +0.159 |
peed | +0.151 |
cott | +0.149 |
ranging | +0.144 |
Neg logit contributions | |
Jinping | -0.139 |
Statements | -0.131 |
NOT | -0.127 |
fore | -0.122 |
't | -0.120 |
Token | either |
Token ID | 2035 |
Feature activation | -6.99e-02 |
Pos logit contributions | |
Jinping | +3.555 |
Statements | +3.354 |
NOT | +3.287 |
't | +3.236 |
fore | +3.178 |
Neg logit contributions | |
utenberg | -3.656 |
rament | -3.611 |
peed | -3.408 |
cott | -3.383 |
ranging | -3.272 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
Jinping | +0.429 |
Statements | +0.407 |
NOT | +0.395 |
fore | +0.378 |
't | +0.373 |
Neg logit contributions | |
utenberg | -0.474 |
rament | -0.473 |
peed | -0.448 |
cott | -0.443 |
ranging | -0.429 |
Token | 't |
Token ID | 470 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
't | +9.754 |
NOT | +6.443 |
fore | +5.204 |
Jinping | +4.689 |
Statements | +4.023 |
Neg logit contributions | |
proport | -27.262 |
encount | -25.986 |
exha | -25.508 |
utenberg | -25.329 |
rament | -25.253 |
Token | afford |
Token ID | 5368 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
Jinping | +4.528 |
NOT | +4.288 |
't | +4.255 |
Statements | +4.247 |
fore | +4.071 |
Neg logit contributions | |
rament | -5.149 |
utenberg | -5.122 |
cott | -4.756 |
peed | -4.719 |
ranging | -4.607 |
Token | insurance |
Token ID | 5096 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
Jinping | +1.421 |
Statements | +1.352 |
NOT | +1.312 |
fore | +1.262 |
't | +1.257 |
Neg logit contributions | |
utenberg | -1.438 |
rament | -1.426 |
peed | -1.340 |
cott | -1.325 |
annah | -1.280 |
Token | before |
Token ID | 878 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
Jinping | +1.121 |
Statements | +1.074 |
NOT | +1.042 |
fore | +1.010 |
't | +1.009 |
Neg logit contributions | |
utenberg | -0.892 |
rament | -0.883 |
peed | -0.822 |
cott | -0.817 |
annah | -0.785 |
Token | the |
Token ID | 262 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
Jinping | +2.159 |
Statements | +2.062 |
NOT | +2.005 |
't | +1.938 |
fore | +1.932 |
Neg logit contributions | |
utenberg | -1.826 |
rament | -1.806 |
peed | -1.686 |
cott | -1.664 |
annah | -1.602 |
Token | young |
Token ID | 1862 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
Jinping | +2.149 |
Statements | +2.040 |
NOT | +1.991 |
fore | +1.915 |
't | +1.905 |
Neg logit contributions | |
utenberg | -1.961 |
rament | -1.947 |
peed | -1.822 |
cott | -1.804 |
"$:/ | -1.732 |
Token | and |
Token ID | 290 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
Jinping | +2.944 |
Statements | +2.754 |
NOT | +2.699 |
fore | +2.629 |
't | +2.624 |
Neg logit contributions | |
utenberg | -2.939 |
rament | -2.917 |
peed | -2.732 |
cott | -2.726 |
"$:/ | -2.639 |
Token | vulnerable |
Token ID | 8826 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
Jinping | +3.671 |
Statements | +3.442 |
NOT | +3.397 |
't | +3.291 |
fore | +3.261 |
Neg logit contributions | |
utenberg | -3.708 |
rament | -3.660 |
peed | -3.392 |
cott | -3.390 |
ranging | -3.233 |
Token | , |
Token ID | 11 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
Jinping | +3.275 |
Statements | +3.046 |
NOT | +2.963 |
't | +2.896 |
fore | +2.879 |
Neg logit contributions | |
utenberg | -3.787 |
rament | -3.758 |
cott | -3.576 |
peed | -3.521 |
"$:/ | -3.405 |
Token | and |
Token ID | 290 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
Jinping | +2.923 |
Statements | +2.757 |
NOT | +2.717 |
fore | +2.618 |
't | +2.608 |
Neg logit contributions | |
rament | -2.721 |
utenberg | -2.686 |
peed | -2.521 |
cott | -2.510 |
"$:/ | -2.380 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
Jinping | +2.021 |
Statements | +1.925 |
NOT | +1.872 |
fore | +1.801 |
't | +1.784 |
Neg logit contributions | |
utenberg | -1.956 |
rament | -1.948 |
peed | -1.834 |
cott | -1.815 |
ranging | -1.744 |
Token | 't |
Token ID | 470 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
't | +9.199 |
NOT | +5.790 |
fore | +4.419 |
Jinping | +4.288 |
Statements | +3.464 |
Neg logit contributions | |
proport | -28.438 |
encount | -27.015 |
exha | -26.376 |
utenberg | -26.149 |
rament | -26.011 |
Token | have |
Token ID | 423 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
rament | +0.690 |
utenberg | +0.689 |
peed | +0.655 |
cott | +0.642 |
annah | +0.620 |
Neg logit contributions | |
Jinping | -0.770 |
Statements | -0.740 |
NOT | -0.715 |
fore | -0.687 |
't | -0.669 |
Token | anywhere |
Token ID | 6609 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
rament | +1.498 |
utenberg | +1.485 |
peed | +1.423 |
cott | +1.395 |
annah | +1.347 |
Neg logit contributions | |
Jinping | -1.905 |
Statements | -1.838 |
NOT | -1.803 |
fore | -1.715 |
annot | -1.693 |
Token | else |
Token ID | 2073 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
Jinping | +0.787 |
Statements | +0.737 |
NOT | +0.718 |
fore | +0.690 |
't | +0.674 |
Neg logit contributions | |
utenberg | -1.109 |
rament | -1.099 |
peed | -1.042 |
cott | -1.030 |
ranging | -1.002 |
Token | to |
Token ID | 284 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
Jinping | +0.348 |
Statements | +0.325 |
NOT | +0.315 |
fore | +0.301 |
't | +0.296 |
Neg logit contributions | |
utenberg | -0.557 |
rament | -0.554 |
peed | -0.528 |
cott | -0.524 |
ranging | -0.509 |
Token | , |
Token ID | 11 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
utenberg | +4.577 |
rament | +4.514 |
peed | +4.462 |
cott | +4.228 |
ranging | +4.100 |
Neg logit contributions | |
Jinping | -5.495 |
Statements | -5.306 |
NOT | -5.103 |
fore | -4.878 |
annot | -4.796 |
Token | and |
Token ID | 290 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
utenberg | +4.120 |
rament | +4.037 |
peed | +3.915 |
cott | +3.762 |
ranging | +3.738 |
Neg logit contributions | |
Jinping | -5.007 |
Statements | -4.779 |
NOT | -4.578 |
fore | -4.407 |
annot | -4.359 |
Token | of |
Token ID | 286 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
Jinping | +0.279 |
Statements | +0.267 |
NOT | +0.259 |
fore | +0.249 |
't | +0.245 |
Neg logit contributions | |
utenberg | -0.268 |
rament | -0.267 |
peed | -0.253 |
cott | -0.249 |
ranging | -0.240 |
Token | course |
Token ID | 1781 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
utenberg | +0.745 |
rament | +0.745 |
peed | +0.700 |
cott | +0.690 |
ranging | +0.660 |
Neg logit contributions | |
Jinping | -0.921 |
Statements | -0.884 |
NOT | -0.863 |
fore | -0.830 |
't | -0.822 |
Token | I |
Token ID | 314 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
Jinping | +2.977 |
Statements | +2.875 |
NOT | +2.809 |
't | +2.692 |
fore | +2.689 |
Neg logit contributions | |
utenberg | -2.884 |
rament | -2.873 |
peed | -2.670 |
cott | -2.670 |
ranging | -2.560 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
Jinping | +3.789 |
Statements | +3.568 |
NOT | +3.561 |
't | +3.449 |
fore | +3.392 |
Neg logit contributions | |
utenberg | -4.001 |
rament | -3.993 |
cott | -3.723 |
peed | -3.707 |
owship | -3.599 |
Token | 't |
Token ID | 470 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
't | +9.536 |
NOT | +6.300 |
fore | +5.075 |
Jinping | +4.420 |
Statements | +3.904 |
Neg logit contributions | |
proport | -27.342 |
encount | -25.977 |
exha | -25.592 |
rament | -25.332 |
utenberg | -25.121 |
Token | count |
Token ID | 954 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Jinping | +1.492 |
Statements | +1.422 |
NOT | +1.393 |
fore | +1.331 |
't | +1.324 |
Neg logit contributions | |
rament | -1.539 |
utenberg | -1.533 |
peed | -1.448 |
cott | -1.443 |
ranging | -1.384 |
Token | every |
Token ID | 790 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
rament | +1.474 |
utenberg | +1.472 |
peed | +1.395 |
cott | +1.371 |
ranging | +1.316 |
Neg logit contributions | |
Jinping | -1.690 |
Statements | -1.616 |
NOT | -1.570 |
fore | -1.509 |
't | -1.472 |
Token | vote |
Token ID | 3015 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
utenberg | +3.290 |
rament | +3.274 |
peed | +3.101 |
cott | +3.043 |
annah | +2.929 |
Neg logit contributions | |
Jinping | -3.852 |
Statements | -3.685 |
NOT | -3.573 |
fore | -3.421 |
't | -3.340 |
Token | that |
Token ID | 326 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
rament | +3.997 |
utenberg | +3.990 |
peed | +3.858 |
cott | +3.716 |
ranging | +3.581 |
Neg logit contributions | |
Jinping | -4.680 |
Statements | -4.489 |
NOT | -4.316 |
fore | -4.144 |
annot | -4.066 |
Token | her |
Token ID | 607 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
Jinping | +2.823 |
Statements | +2.715 |
NOT | +2.586 |
't | +2.520 |
fore | +2.504 |
Neg logit contributions | |
utenberg | -2.942 |
rament | -2.910 |
peed | -2.763 |
cott | -2.731 |
"$:/ | -2.634 |
Token | husband |
Token ID | 5229 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
Jinping | +0.424 |
Statements | +0.403 |
NOT | +0.392 |
fore | +0.375 |
't | +0.372 |
Neg logit contributions | |
utenberg | -0.437 |
rament | -0.434 |
peed | -0.412 |
cott | -0.405 |
ranging | -0.391 |
Token | , |
Token ID | 11 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
Jinping | +1.374 |
Statements | +1.304 |
NOT | +1.272 |
't | +1.225 |
fore | +1.221 |
Neg logit contributions | |
utenberg | -1.282 |
rament | -1.273 |
peed | -1.207 |
cott | -1.181 |
"$:/ | -1.145 |
Token | " |
Token ID | 366 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
Jinping | +0.327 |
Statements | +0.312 |
NOT | +0.304 |
fore | +0.292 |
't | +0.288 |
Neg logit contributions | |
utenberg | -0.294 |
rament | -0.293 |
peed | -0.277 |
cott | -0.273 |
"$:/ | -0.262 |
Token | Did |
Token ID | 11633 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
Jinping | +3.014 |
NOT | +2.926 |
Statements | +2.902 |
fore | +2.735 |
't | +2.711 |
Neg logit contributions | |
utenberg | -4.353 |
rament | -4.293 |
peed | -4.070 |
cott | -3.975 |
"$:/ | -3.959 |
Token | n |
Token ID | 77 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
Jinping | +3.068 |
NOT | +2.900 |
Statements | +2.866 |
't | +2.860 |
fore | +2.693 |
Neg logit contributions | |
utenberg | -4.746 |
rament | -4.741 |
peed | -4.530 |
cott | -4.497 |
"$:/ | -4.349 |
Token | 't |
Token ID | 470 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
't | +10.114 |
NOT | +7.331 |
Jinping | +5.941 |
fore | +5.591 |
Statements | +4.764 |
Neg logit contributions | |
proport | -25.158 |
rament | -23.971 |
utenberg | -23.936 |
encount | -23.556 |
exha | -23.034 |
Token | I |
Token ID | 314 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
rament | +2.483 |
utenberg | +2.480 |
peed | +2.357 |
cott | +2.320 |
annah | +2.243 |
Neg logit contributions | |
Jinping | -2.398 |
Statements | -2.287 |
NOT | -2.215 |
fore | -2.123 |
't | -2.073 |
Token | tell |
Token ID | 1560 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
Jinping | +0.880 |
Statements | +0.790 |
NOT | +0.769 |
't | +0.714 |
fore | +0.710 |
Neg logit contributions | |
rament | -2.273 |
utenberg | -2.270 |
cott | -2.176 |
peed | -2.169 |
"$:/ | -2.116 |
Token | you |
Token ID | 345 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
rament | +1.676 |
utenberg | +1.654 |
peed | +1.585 |
cott | +1.557 |
annah | +1.500 |
Neg logit contributions | |
Jinping | -2.086 |
Statements | -2.022 |
NOT | -1.959 |
fore | -1.875 |
annot | -1.840 |
Token | I |
Token ID | 314 |
Feature activation | +6.39e-03 |
Pos logit contributions | |
rament | +5.013 |
utenberg | +4.826 |
peed | +4.719 |
cott | +4.622 |
annah | +4.519 |
Neg logit contributions | |
Jinping | -5.593 |
Statements | -5.343 |
NOT | -5.126 |
annot | -4.946 |
fore | -4.905 |
Token | straight |
Token ID | 3892 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
Jinping | +4.141 |
Statements | +3.976 |
NOT | +3.833 |
't | +3.715 |
fore | +3.695 |
Neg logit contributions | |
utenberg | -4.266 |
rament | -4.238 |
peed | -4.022 |
cott | -3.975 |
"$:/ | -3.880 |
Token | facts |
Token ID | 6419 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
utenberg | +0.399 |
rament | +0.398 |
peed | +0.377 |
cott | +0.372 |
ranging | +0.358 |
Neg logit contributions | |
Jinping | -0.410 |
Statements | -0.391 |
NOT | -0.380 |
fore | -0.365 |
't | -0.358 |
Token | . |
Token ID | 13 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
Jinping | +2.685 |
Statements | +2.555 |
NOT | +2.473 |
't | +2.414 |
fore | +2.399 |
Neg logit contributions | |
utenberg | -2.890 |
rament | -2.875 |
peed | -2.680 |
cott | -2.675 |
"$:/ | -2.595 |
Token | He |
Token ID | 679 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
Jinping | +3.089 |
Statements | +3.018 |
NOT | +2.847 |
't | +2.707 |
fore | +2.679 |
Neg logit contributions | |
utenberg | -4.410 |
rament | -4.390 |
peed | -4.141 |
cott | -4.095 |
"$:/ | -3.979 |
Token | just |
Token ID | 655 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
Jinping | +2.958 |
Statements | +2.763 |
NOT | +2.686 |
't | +2.603 |
fore | +2.588 |
Neg logit contributions | |
rament | -3.592 |
utenberg | -3.586 |
peed | -3.402 |
cott | -3.354 |
ranging | -3.282 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
Jinping | +2.793 |
Statements | +2.635 |
NOT | +2.551 |
't | +2.470 |
fore | +2.455 |
Neg logit contributions | |
rament | -3.132 |
utenberg | -3.120 |
peed | -2.941 |
cott | -2.908 |
ranging | -2.839 |
Token | 't |
Token ID | 470 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
't | +9.840 |
NOT | +6.667 |
fore | +5.171 |
Jinping | +5.152 |
Statements | +4.453 |
Neg logit contributions | |
proport | -26.860 |
encount | -25.217 |
utenberg | -24.992 |
rament | -24.887 |
exha | -24.780 |
Token | handle |
Token ID | 5412 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
Jinping | +5.063 |
Statements | +4.787 |
NOT | +4.754 |
't | +4.747 |
fore | +4.499 |
Neg logit contributions | |
rament | -5.398 |
utenberg | -5.382 |
cott | -5.138 |
peed | -4.943 |
ranging | -4.892 |
Token | the |
Token ID | 262 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
Jinping | +0.433 |
Statements | +0.415 |
NOT | +0.403 |
fore | +0.388 |
't | +0.384 |
Neg logit contributions | |
utenberg | -0.358 |
rament | -0.357 |
peed | -0.336 |
cott | -0.331 |
ranging | -0.318 |
Token | truth |
Token ID | 3872 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
rament | +1.304 |
utenberg | +1.299 |
peed | +1.234 |
cott | +1.202 |
annah | +1.160 |
Neg logit contributions | |
Jinping | -1.516 |
Statements | -1.445 |
NOT | -1.415 |
fore | -1.359 |
't | -1.321 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
Jinping | +1.043 |
Statements | +0.991 |
NOT | +0.962 |
fore | +0.929 |
't | +0.923 |
Neg logit contributions | |
utenberg | -1.118 |
rament | -1.111 |
peed | -1.050 |
cott | -1.040 |
"$:/ | -1.007 |
Token | our |
Token ID | 674 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
utenberg | +1.737 |
rament | +1.736 |
peed | +1.644 |
cott | +1.618 |
ranging | +1.560 |
Neg logit contributions | |
Jinping | -1.860 |
Statements | -1.780 |
NOT | -1.725 |
fore | -1.653 |
't | -1.620 |
Token | finale |
Token ID | 19523 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
rament | +2.008 |
utenberg | +2.003 |
peed | +1.899 |
cott | +1.870 |
annah | +1.805 |
Neg logit contributions | |
Jinping | -2.059 |
Statements | -1.967 |
NOT | -1.912 |
fore | -1.830 |
't | -1.789 |
Token | , |
Token ID | 11 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
Jinping | +4.254 |
Statements | +4.047 |
NOT | +3.928 |
fore | +3.802 |
't | +3.788 |
Neg logit contributions | |
utenberg | -4.549 |
rament | -4.517 |
cott | -4.304 |
peed | -4.282 |
ranging | -4.080 |
Token | because |
Token ID | 780 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
Jinping | +0.254 |
Statements | +0.242 |
NOT | +0.236 |
fore | +0.228 |
't | +0.224 |
Neg logit contributions | |
rament | -0.218 |
utenberg | -0.217 |
peed | -0.205 |
cott | -0.202 |
ranging | -0.194 |
Token | I |
Token ID | 314 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
Jinping | +1.142 |
Statements | +1.089 |
NOT | +1.057 |
fore | +1.018 |
't | +1.004 |
Neg logit contributions | |
utenberg | -1.110 |
rament | -1.108 |
peed | -1.047 |
cott | -1.037 |
ranging | -0.996 |
Token | don |
Token ID | 836 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
Jinping | +4.135 |
NOT | +3.807 |
Statements | +3.798 |
't | +3.758 |
fore | +3.677 |
Neg logit contributions | |
utenberg | -5.259 |
rament | -5.203 |
cott | -4.906 |
peed | -4.882 |
"$:/ | -4.712 |
Token | � |
Token ID | 447 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
't | +10.072 |
NOT | +6.688 |
fore | +5.480 |
Jinping | +5.382 |
Statements | +4.754 |
Neg logit contributions | |
proport | -26.627 |
encount | -25.263 |
exha | -24.652 |
utenberg | -24.647 |
rament | -24.242 |
Token | � |
Token ID | 247 |
Feature activation | +7.39e-02 |
Pos logit contributions | |
Jinping | +1.588 |
Statements | +1.509 |
NOT | +1.477 |
fore | +1.412 |
't | +1.396 |
Neg logit contributions | |
rament | -1.970 |
utenberg | -1.964 |
peed | -1.848 |
cott | -1.828 |
ranging | -1.775 |
Token | t |
Token ID | 83 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
utenberg | +0.496 |
rament | +0.496 |
peed | +0.470 |
cott | +0.462 |
ranging | +0.447 |
Neg logit contributions | |
Jinping | -0.460 |
Statements | -0.437 |
NOT | -0.426 |
fore | -0.408 |
't | -0.403 |
Token | know |
Token ID | 760 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
Jinping | +2.137 |
Statements | +2.009 |
NOT | +1.974 |
't | +1.898 |
fore | +1.892 |
Neg logit contributions | |
utenberg | -2.537 |
rament | -2.526 |
cott | -2.388 |
peed | -2.365 |
ranging | -2.292 |
Token | what |
Token ID | 644 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
rament | +2.551 |
utenberg | +2.536 |
peed | +2.438 |
cott | +2.399 |
annah | +2.309 |
Neg logit contributions | |
Jinping | -2.463 |
Statements | -2.368 |
NOT | -2.294 |
fore | -2.176 |
't | -2.122 |
Token | crazy |
Token ID | 50112 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
Jinping | +5.124 |
NOT | +4.981 |
Statements | +4.897 |
fore | +4.858 |
't | +4.709 |
Neg logit contributions | |
utenberg | -5.311 |
rament | -5.109 |
"$:/ | -4.801 |
peed | -4.781 |
cott | -4.652 |
Token | " |
Token ID | 1 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
rament | +2.677 |
utenberg | +2.626 |
peed | +2.518 |
cott | +2.487 |
ranging | +2.385 |
Neg logit contributions | |
Jinping | -3.215 |
Statements | -3.102 |
NOT | -3.023 |
fore | -2.904 |
annot | -2.817 |
Token | Russian |
Token ID | 3394 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
Jinping | +1.197 |
Statements | +1.146 |
NOT | +1.111 |
fore | +1.073 |
't | +1.066 |
Neg logit contributions | |
utenberg | -0.912 |
rament | -0.906 |
peed | -0.850 |
cott | -0.836 |
"$:/ | -0.801 |
Token | leader |
Token ID | 3554 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
Jinping | +2.898 |
Statements | +2.774 |
NOT | +2.679 |
fore | +2.608 |
't | +2.590 |
Neg logit contributions | |
utenberg | -2.089 |
rament | -2.063 |
peed | -1.934 |
cott | -1.888 |
"$:/ | -1.839 |
Token | who |
Token ID | 508 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
Jinping | +3.379 |
Statements | +3.057 |
NOT | +2.917 |
fore | +2.837 |
't | +2.831 |
Neg logit contributions | |
utenberg | -5.887 |
rament | -5.808 |
peed | -5.593 |
cott | -5.445 |
annah | -5.389 |
Token | doesn |
Token ID | 1595 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
Jinping | +3.777 |
Statements | +3.471 |
NOT | +3.360 |
fore | +3.283 |
't | +3.266 |
Neg logit contributions | |
utenberg | -4.327 |
rament | -4.266 |
peed | -4.082 |
cott | -3.959 |
ranging | -3.916 |
Token | � |
Token ID | 447 |
Feature activation | -9.90e-02 |
Pos logit contributions | |
't | +12.536 |
NOT | +9.871 |
Jinping | +8.734 |
fore | +8.565 |
Statements | +8.100 |
Neg logit contributions | |
proport | -22.338 |
utenberg | -21.074 |
encount | -21.042 |
rament | -20.886 |
owship | -20.514 |
Token | � |
Token ID | 247 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
Jinping | +0.472 |
Statements | +0.440 |
NOT | +0.426 |
fore | +0.398 |
't | +0.393 |
Neg logit contributions | |
utenberg | -0.810 |
rament | -0.808 |
peed | -0.772 |
cott | -0.767 |
"$:/ | -0.748 |
Token | t |
Token ID | 83 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
rament | +3.851 |
utenberg | +3.849 |
peed | +3.684 |
cott | +3.682 |
annah | +3.536 |
Neg logit contributions | |
Jinping | -3.189 |
Statements | -3.121 |
NOT | -2.934 |
fore | -2.778 |
't | -2.675 |
Token | know |
Token ID | 760 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
Jinping | +1.241 |
Statements | +1.169 |
NOT | +1.144 |
fore | +1.091 |
't | +1.088 |
Neg logit contributions | |
utenberg | -1.301 |
rament | -1.295 |
peed | -1.223 |
cott | -1.214 |
ranging | -1.172 |
Token | how |
Token ID | 703 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
rament | +1.393 |
utenberg | +1.381 |
peed | +1.324 |
cott | +1.309 |
ranging | +1.261 |
Neg logit contributions | |
Jinping | -1.266 |
Statements | -1.217 |
NOT | -1.177 |
fore | -1.125 |
't | -1.094 |
Token | lose |
Token ID | 4425 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
Jinping | +1.162 |
Statements | +1.107 |
NOT | +1.075 |
fore | +1.031 |
't | +1.018 |
Neg logit contributions | |
utenberg | -1.171 |
rament | -1.170 |
peed | -1.105 |
cott | -1.091 |
ranging | -1.053 |
Token | control |
Token ID | 1630 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
Jinping | +0.769 |
Statements | +0.735 |
NOT | +0.714 |
fore | +0.693 |
't | +0.689 |
Neg logit contributions | |
utenberg | -0.639 |
rament | -0.638 |
peed | -0.594 |
cott | -0.588 |
ranging | -0.567 |
Token | , |
Token ID | 11 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
rament | +5.057 |
utenberg | +5.036 |
peed | +4.908 |
cott | +4.700 |
ranging | +4.575 |
Neg logit contributions | |
Jinping | -5.528 |
Statements | -5.356 |
NOT | -5.162 |
fore | -4.856 |
annot | -4.834 |
Token | but |
Token ID | 475 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
utenberg | +3.234 |
rament | +3.219 |
peed | +3.053 |
cott | +2.976 |
ranging | +2.916 |
Neg logit contributions | |
Jinping | -4.407 |
Statements | -4.244 |
NOT | -4.106 |
fore | -3.934 |
annot | -3.856 |
Token | he |
Token ID | 339 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
rament | +2.083 |
utenberg | +2.081 |
peed | +1.982 |
cott | +1.929 |
"$:/ | +1.873 |
Neg logit contributions | |
Jinping | -2.410 |
Statements | -2.306 |
NOT | -2.240 |
fore | -2.157 |
't | -2.090 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
rament | +2.889 |
utenberg | +2.881 |
peed | +2.783 |
cott | +2.709 |
annah | +2.628 |
Neg logit contributions | |
Jinping | -3.067 |
Statements | -2.998 |
NOT | -2.883 |
fore | -2.722 |
annot | -2.645 |
Token | � |
Token ID | 447 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
't | +12.518 |
NOT | +10.024 |
fore | +8.609 |
Jinping | +8.143 |
Statements | +7.956 |
Neg logit contributions | |
proport | -22.593 |
utenberg | -21.350 |
encount | -21.100 |
rament | -20.869 |
owship | -20.714 |
Token | � |
Token ID | 247 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
rament | +4.217 |
utenberg | +4.203 |
peed | +4.029 |
cott | +3.947 |
ranging | +3.868 |
Neg logit contributions | |
Jinping | -3.006 |
Statements | -2.847 |
NOT | -2.719 |
fore | -2.602 |
annot | -2.493 |
Token | t |
Token ID | 83 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
utenberg | +3.721 |
rament | +3.683 |
peed | +3.547 |
cott | +3.494 |
ranging | +3.410 |
Neg logit contributions | |
Jinping | -3.026 |
Statements | -2.919 |
NOT | -2.760 |
fore | -2.613 |
annot | -2.519 |
Token | go |
Token ID | 467 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
Jinping | +2.477 |
Statements | +2.337 |
NOT | +2.314 |
't | +2.255 |
fore | +2.223 |
Neg logit contributions | |
rament | -2.575 |
utenberg | -2.570 |
cott | -2.435 |
peed | -2.381 |
ranging | -2.302 |
Token | back |
Token ID | 736 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
utenberg | +4.906 |
rament | +4.884 |
peed | +4.707 |
cott | +4.590 |
"$:/ | +4.493 |
Neg logit contributions | |
Jinping | -6.966 |
Statements | -6.607 |
NOT | -6.483 |
annot | -6.116 |
fore | -6.069 |
Token | chaos |
Token ID | 11918 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
rament | +5.120 |
utenberg | +5.087 |
peed | +4.906 |
cott | +4.752 |
ranging | +4.605 |
Neg logit contributions | |
Jinping | -5.343 |
Statements | -5.074 |
NOT | -4.905 |
fore | -4.682 |
annot | -4.598 |
Token | with |
Token ID | 351 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
rament | +4.510 |
utenberg | +4.461 |
peed | +4.306 |
cott | +4.152 |
ranging | +4.004 |
Neg logit contributions | |
Jinping | -5.286 |
Statements | -5.099 |
NOT | -4.930 |
fore | -4.708 |
't | -4.619 |
Token | a |
Token ID | 257 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
utenberg | +3.211 |
rament | +3.202 |
peed | +3.054 |
cott | +2.980 |
ranging | +2.895 |
Neg logit contributions | |
Jinping | -3.224 |
Statements | -3.068 |
NOT | -2.986 |
fore | -2.870 |
't | -2.762 |
Token | whopping |
Token ID | 27833 |
Feature activation | +9.58e-01 |
Pos logit contributions | |
rament | +1.373 |
utenberg | +1.367 |
peed | +1.303 |
cott | +1.282 |
ranging | +1.236 |
Neg logit contributions | |
Jinping | -1.375 |
Statements | -1.313 |
NOT | -1.272 |
fore | -1.216 |
't | -1.191 |
Token | 12 |
Token ID | 1105 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
rament | +5.472 |
utenberg | +5.417 |
peed | +5.334 |
cott | +5.090 |
ranging | +4.987 |
Neg logit contributions | |
Jinping | -6.412 |
Statements | -6.094 |
NOT | -5.961 |
fore | -5.667 |
annot | -5.566 |
Token | attack |
Token ID | 1368 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
utenberg | +2.485 |
rament | +2.480 |
peed | +2.361 |
cott | +2.301 |
ranging | +2.220 |
Neg logit contributions | |
Jinping | -3.150 |
Statements | -3.015 |
NOT | -2.929 |
fore | -2.819 |
't | -2.765 |
Token | , |
Token ID | 11 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
rament | +7.702 |
peed | +7.685 |
utenberg | +7.611 |
cott | +6.948 |
ranging | +6.834 |
Neg logit contributions | |
Jinping | -8.893 |
Statements | -8.362 |
NOT | -8.174 |
fore | -7.755 |
annot | -7.602 |
Token | Master |
Token ID | 5599 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
utenberg | +4.327 |
rament | +4.316 |
peed | +4.159 |
cott | +3.974 |
ranging | +3.959 |
Neg logit contributions | |
Jinping | -5.482 |
Statements | -5.262 |
NOT | -5.079 |
fore | -4.920 |
annot | -4.797 |
Token | mind |
Token ID | 10155 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
rament | +4.087 |
utenberg | +4.066 |
peed | +3.835 |
cott | +3.707 |
ranging | +3.601 |
Neg logit contributions | |
Jinping | -7.064 |
Statements | -6.808 |
NOT | -6.652 |
fore | -6.333 |
't | -6.256 |
Token | that |
Token ID | 326 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
rament | +2.776 |
utenberg | +2.767 |
peed | +2.637 |
cott | +2.579 |
ranging | +2.490 |
Neg logit contributions | |
Jinping | -3.176 |
Statements | -3.038 |
NOT | -2.964 |
fore | -2.842 |
't | -2.776 |
Token | can |
Token ID | 460 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
rament | +5.628 |
utenberg | +5.562 |
peed | +5.452 |
cott | +5.335 |
annah | +5.150 |
Neg logit contributions | |
Jinping | -4.734 |
Statements | -4.526 |
NOT | -4.413 |
fore | -4.115 |
annot | -3.995 |
Token | would |
Token ID | 561 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
Jinping | +2.384 |
Statements | +2.200 |
NOT | +2.153 |
't | +2.101 |
fore | +2.101 |
Neg logit contributions | |
utenberg | -2.447 |
rament | -2.436 |
peed | -2.329 |
cott | -2.274 |
ranging | -2.205 |
Token | be |
Token ID | 307 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
rament | +3.336 |
utenberg | +3.334 |
peed | +3.177 |
cott | +3.043 |
"$:/ | +3.000 |
Neg logit contributions | |
Jinping | -4.205 |
Statements | -4.102 |
NOT | -3.914 |
fore | -3.800 |
annot | -3.720 |
Token | impossible |
Token ID | 5340 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
Jinping | +0.969 |
Statements | +0.922 |
NOT | +0.896 |
fore | +0.862 |
't | +0.850 |
Neg logit contributions | |
utenberg | -0.930 |
rament | -0.927 |
peed | -0.878 |
cott | -0.864 |
ranging | -0.833 |
Token | for |
Token ID | 329 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
Jinping | +4.224 |
NOT | +3.782 |
Statements | +3.697 |
fore | +3.656 |
't | +3.626 |
Neg logit contributions | |
rament | -6.864 |
utenberg | -6.853 |
peed | -6.511 |
cott | -6.479 |
ranging | -6.295 |
Token | the |
Token ID | 262 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
Jinping | +3.359 |
Statements | +3.134 |
NOT | +3.009 |
fore | +2.947 |
't | +2.934 |
Neg logit contributions | |
utenberg | -3.619 |
rament | -3.578 |
peed | -3.404 |
cott | -3.326 |
"$:/ | -3.256 |
Token | United |
Token ID | 1578 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Jinping | +1.155 |
Statements | +1.092 |
NOT | +1.054 |
fore | +1.012 |
't | +0.997 |
Neg logit contributions | |
utenberg | -1.377 |
rament | -1.368 |
peed | -1.307 |
cott | -1.287 |
"$:/ | -1.244 |
Token | States |
Token ID | 1829 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
rament | +6.279 |
utenberg | +5.959 |
cott | +5.936 |
peed | +5.736 |
annah | +5.554 |
Neg logit contributions | |
Jinping | -9.556 |
NOT | -8.977 |
Statements | -8.902 |
fore | -8.688 |
annot | -8.305 |
Token | to |
Token ID | 284 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
Jinping | +0.361 |
Statements | +0.335 |
NOT | +0.327 |
't | +0.308 |
fore | +0.308 |
Neg logit contributions | |
rament | -0.517 |
utenberg | -0.516 |
peed | -0.488 |
cott | -0.480 |
ranging | -0.471 |
Token | contemplate |
Token ID | 39248 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
Jinping | +4.162 |
NOT | +3.911 |
Statements | +3.899 |
't | +3.802 |
fore | +3.701 |
Neg logit contributions | |
utenberg | -3.744 |
rament | -3.724 |
cott | -3.465 |
peed | -3.456 |
ranging | -3.356 |
Token | the |
Token ID | 262 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
rament | +2.785 |
utenberg | +2.733 |
peed | +2.622 |
cott | +2.581 |
ranging | +2.478 |
Neg logit contributions | |
Jinping | -3.624 |
Statements | -3.519 |
NOT | -3.409 |
fore | -3.297 |
annot | -3.236 |
Token | continued |
Token ID | 3767 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
rament | +1.475 |
utenberg | +1.465 |
peed | +1.378 |
cott | +1.357 |
ranging | +1.298 |
Neg logit contributions | |
Jinping | -2.062 |
Statements | -1.987 |
NOT | -1.935 |
fore | -1.866 |
't | -1.831 |
Token | both |
Token ID | 1111 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
rament | +1.518 |
utenberg | +1.513 |
peed | +1.438 |
cott | +1.420 |
ranging | +1.373 |
Neg logit contributions | |
Jinping | -1.590 |
Statements | -1.526 |
NOT | -1.483 |
fore | -1.414 |
't | -1.388 |
Token | national |
Token ID | 2260 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
rament | +2.779 |
utenberg | +2.738 |
peed | +2.640 |
cott | +2.638 |
ranging | +2.540 |
Neg logit contributions | |
Jinping | -2.702 |
Statements | -2.608 |
NOT | -2.538 |
fore | -2.384 |
annot | -2.344 |
Token | governments |
Token ID | 6905 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
Jinping | +3.370 |
Statements | +3.179 |
NOT | +3.046 |
fore | +3.028 |
't | +2.973 |
Neg logit contributions | |
utenberg | -3.988 |
rament | -3.947 |
peed | -3.736 |
cott | -3.646 |
"$:/ | -3.577 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
utenberg | +1.309 |
rament | +1.307 |
peed | +1.234 |
cott | +1.218 |
ranging | +1.172 |
Neg logit contributions | |
Jinping | -1.388 |
Statements | -1.324 |
NOT | -1.287 |
fore | -1.236 |
't | -1.216 |
Token | international |
Token ID | 3230 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
rament | +0.693 |
utenberg | +0.691 |
peed | +0.656 |
cott | +0.648 |
ranging | +0.624 |
Neg logit contributions | |
Jinping | -0.722 |
Statements | -0.692 |
NOT | -0.672 |
fore | -0.642 |
't | -0.630 |
Token | co |
Token ID | 763 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
utenberg | +0.422 |
rament | +0.422 |
peed | +0.401 |
cott | +0.396 |
ranging | +0.383 |
Neg logit contributions | |
Jinping | -0.361 |
Statements | -0.343 |
NOT | -0.332 |
fore | -0.317 |
't | -0.311 |
Token | - |
Token ID | 12 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
rament | +7.157 |
utenberg | +6.847 |
cott | +6.831 |
peed | +6.786 |
ranging | +6.580 |
Neg logit contributions | |
Jinping | -9.720 |
Statements | -9.417 |
NOT | -9.276 |
't | -8.351 |
annot | -8.335 |
Token | operation |
Token ID | 27184 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
rament | +3.910 |
utenberg | +3.789 |
cott | +3.715 |
peed | +3.634 |
ranging | +3.536 |
Neg logit contributions | |
Jinping | -5.608 |
Statements | -5.379 |
NOT | -5.235 |
fore | -4.914 |
't | -4.904 |
Token | . |
Token ID | 13 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
rament | +1.993 |
utenberg | +1.987 |
peed | +1.905 |
cott | +1.879 |
ranging | +1.821 |
Neg logit contributions | |
Jinping | -1.798 |
Statements | -1.711 |
NOT | -1.666 |
fore | -1.582 |
't | -1.546 |
Token |
|
Token ID | 198 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
rament | +1.507 |
utenberg | +1.499 |
peed | +1.430 |
cott | +1.422 |
ranging | +1.378 |
Neg logit contributions | |
Jinping | -1.292 |
Statements | -1.219 |
NOT | -1.190 |
fore | -1.138 |
't | -1.108 |
Token |
|
Token ID | 198 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
rament | +0.102 |
utenberg | +0.102 |
peed | +0.097 |
cott | +0.096 |
ranging | +0.093 |
Neg logit contributions | |
Jinping | -0.096 |
Statements | -0.091 |
NOT | -0.088 |
fore | -0.084 |
annot | -0.081 |
Token | projects |
Token ID | 4493 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
utenberg | +5.077 |
rament | +5.041 |
cott | +4.886 |
peed | +4.831 |
annah | +4.627 |
Neg logit contributions | |
Jinping | -5.252 |
Statements | -4.819 |
NOT | -4.807 |
fore | -4.593 |
annot | -4.449 |
Token | . |
Token ID | 13 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
utenberg | +3.745 |
rament | +3.743 |
peed | +3.595 |
cott | +3.493 |
ranging | +3.402 |
Neg logit contributions | |
Jinping | -4.221 |
Statements | -4.054 |
NOT | -3.943 |
fore | -3.764 |
annot | -3.664 |
Token | I |
Token ID | 314 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
utenberg | +3.441 |
rament | +3.429 |
peed | +3.293 |
cott | +3.279 |
ranging | +3.197 |
Neg logit contributions | |
Jinping | -3.011 |
Statements | -2.748 |
NOT | -2.731 |
fore | -2.628 |
annot | -2.573 |
Token | thought |
Token ID | 1807 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Jinping | +0.857 |
Statements | +0.814 |
NOT | +0.796 |
fore | +0.764 |
't | +0.756 |
Neg logit contributions | |
utenberg | -0.854 |
rament | -0.853 |
peed | -0.808 |
cott | -0.794 |
"$:/ | -0.767 |
Token | I |
Token ID | 314 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
rament | +6.248 |
utenberg | +6.241 |
cott | +5.947 |
peed | +5.935 |
ranging | +5.738 |
Neg logit contributions | |
Jinping | -7.858 |
Statements | -7.289 |
NOT | -7.095 |
fore | -6.842 |
annot | -6.802 |
Token | would |
Token ID | 561 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
rament | +4.349 |
utenberg | +4.327 |
cott | +4.119 |
peed | +4.065 |
ranging | +3.895 |
Neg logit contributions | |
Jinping | -5.529 |
Statements | -5.340 |
NOT | -5.114 |
fore | -4.923 |
't | -4.758 |
Token | help |
Token ID | 1037 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
utenberg | +7.531 |
rament | +7.389 |
peed | +7.172 |
cott | +6.805 |
annah | +6.743 |
Neg logit contributions | |
Jinping | -9.101 |
Statements | -8.766 |
NOT | -8.219 |
fore | -8.124 |
annot | -8.015 |
Token | by |
Token ID | 416 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
rament | +6.010 |
utenberg | +5.981 |
peed | +5.749 |
cott | +5.570 |
"$:/ | +5.379 |
Neg logit contributions | |
Jinping | -7.579 |
Statements | -7.074 |
NOT | -7.035 |
fore | -6.752 |
annot | -6.647 |
Token | organizing |
Token ID | 16924 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
rament | +4.296 |
utenberg | +4.252 |
peed | +3.964 |
cott | +3.914 |
ranging | +3.834 |
Neg logit contributions | |
Jinping | -7.232 |
Statements | -7.014 |
NOT | -6.752 |
fore | -6.572 |
annot | -6.438 |
Token | and |
Token ID | 290 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
rament | +5.942 |
utenberg | +5.914 |
cott | +5.708 |
peed | +5.695 |
ranging | +5.397 |
Neg logit contributions | |
Jinping | -6.853 |
NOT | -6.303 |
Statements | -6.274 |
fore | -6.019 |
annot | -5.887 |
Token | writing |
Token ID | 3597 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
rament | +1.454 |
utenberg | +1.453 |
peed | +1.353 |
cott | +1.332 |
ranging | +1.269 |
Neg logit contributions | |
Jinping | -2.449 |
Statements | -2.363 |
NOT | -2.305 |
fore | -2.230 |
't | -2.189 |
Token | i |
Token ID | 72 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
utenberg | +2.980 |
rament | +2.956 |
peed | +2.806 |
cott | +2.791 |
ranging | +2.691 |
Neg logit contributions | |
Jinping | -4.290 |
Statements | -4.165 |
NOT | -3.991 |
fore | -3.826 |
annot | -3.703 |
Token | and |
Token ID | 290 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
rament | +0.097 |
utenberg | +0.097 |
peed | +0.092 |
cott | +0.091 |
ranging | +0.088 |
Neg logit contributions | |
Jinping | -0.098 |
Statements | -0.093 |
NOT | -0.090 |
fore | -0.087 |
't | -0.086 |
Token | Sh |
Token ID | 911 |
Feature activation | -7.34e-01 |
Pos logit contributions | |
Jinping | +1.145 |
Statements | +1.078 |
NOT | +1.017 |
't | +0.972 |
fore | +0.971 |
Neg logit contributions | |
utenberg | -1.780 |
rament | -1.774 |
peed | -1.705 |
cott | -1.679 |
"$:/ | -1.632 |
Token | us |
Token ID | 385 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Jinping | +3.804 |
Statements | +3.423 |
NOT | +3.364 |
't | +3.333 |
fore | +3.277 |
Neg logit contributions | |
rament | -5.506 |
utenberg | -5.500 |
peed | -5.318 |
cott | -5.238 |
"$:/ | -5.217 |
Token | aku |
Token ID | 8719 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
rament | +2.558 |
utenberg | +2.527 |
cott | +2.378 |
peed | +2.225 |
annah | +2.160 |
Neg logit contributions | |
Statements | -10.660 |
Jinping | -10.579 |
NOT | -10.223 |
fore | -9.797 |
� | -9.550 |
Token | U |
Token ID | 471 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
Jinping | +0.541 |
Statements | +0.497 |
NOT | +0.474 |
't | +0.451 |
fore | +0.449 |
Neg logit contributions | |
rament | -0.956 |
utenberg | -0.947 |
peed | -0.916 |
cott | -0.909 |
ranging | -0.882 |
Token | ch |
Token ID | 354 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
cott | +5.867 |
rament | +5.731 |
pan | +5.208 |
peed | +5.167 |
ranging | +5.080 |
Neg logit contributions | |
Statements | -15.238 |
Jinping | -15.094 |
NOT | -14.119 |
Tycoon | -13.783 |
IDs | -13.218 |
Token | iy |
Token ID | 7745 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
Jinping | +2.621 |
Statements | +2.464 |
NOT | +2.407 |
fore | +2.321 |
't | +2.312 |
Neg logit contributions | |
utenberg | -2.871 |
rament | -2.865 |
peed | -2.729 |
cott | -2.687 |
"$:/ | -2.600 |
Token | ama |
Token ID | 1689 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
rament | +1.664 |
utenberg | +1.650 |
peed | +1.526 |
cott | +1.526 |
annah | +1.441 |
Neg logit contributions | |
Jinping | -2.804 |
Statements | -2.723 |
NOT | -2.640 |
fore | -2.539 |
't | -2.475 |
Token | struck |
Token ID | 7425 |
Feature activation | +7.02e-01 |
Pos logit contributions | |
Jinping | +1.169 |
Statements | +1.113 |
NOT | +1.088 |
fore | +1.048 |
't | +1.038 |
Neg logit contributions | |
rament | -1.169 |
utenberg | -1.165 |
peed | -1.103 |
cott | -1.089 |
ranging | -1.051 |
Token | gold |
Token ID | 3869 |
Feature activation | -6.47e-01 |
Pos logit contributions | |
rament | +3.780 |
utenberg | +3.766 |
peed | +3.606 |
cott | +3.513 |
ranging | +3.344 |
Neg logit contributions | |
Jinping | -5.087 |
Statements | -4.863 |
NOT | -4.744 |
fore | -4.593 |
annot | -4.451 |
Token | is |
Token ID | 318 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
Jinping | +5.255 |
Statements | +4.944 |
fore | +4.894 |
NOT | +4.843 |
't | +4.729 |
Neg logit contributions | |
utenberg | -7.879 |
rament | -7.820 |
cott | -7.541 |
peed | -7.489 |
annah | -7.299 |
Token | teaching |
Token ID | 7743 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
Jinping | +1.974 |
Statements | +1.884 |
NOT | +1.830 |
fore | +1.765 |
't | +1.750 |
Neg logit contributions | |
rament | -1.926 |
utenberg | -1.923 |
peed | -1.824 |
cott | -1.791 |
annah | -1.733 |
Token | " |
Token ID | 366 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
utenberg | +0.137 |
rament | +0.137 |
peed | +0.129 |
cott | +0.127 |
ranging | +0.122 |
Neg logit contributions | |
Jinping | -0.156 |
Statements | -0.149 |
NOT | -0.145 |
fore | -0.140 |
't | -0.138 |
Token | fair |
Token ID | 22043 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
Jinping | +3.687 |
NOT | +3.544 |
Statements | +3.542 |
fore | +3.462 |
't | +3.346 |
Neg logit contributions | |
utenberg | -3.841 |
rament | -3.838 |
peed | -3.610 |
"$:/ | -3.504 |
cott | -3.443 |
Token | ness |
Token ID | 1108 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
rament | +3.965 |
utenberg | +3.956 |
cott | +3.741 |
peed | +3.705 |
ranging | +3.566 |
Neg logit contributions | |
Jinping | -6.697 |
Statements | -6.339 |
NOT | -6.231 |
fore | -6.023 |
annot | -5.854 |
Token | ." |
Token ID | 526 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
Jinping | +2.495 |
Statements | +2.398 |
NOT | +2.318 |
fore | +2.237 |
't | +2.226 |
Neg logit contributions | |
rament | -2.253 |
utenberg | -2.252 |
peed | -2.113 |
cott | -2.074 |
"$:/ | -2.016 |
Token | Dug |
Token ID | 27436 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
Jinping | +0.644 |
Statements | +0.612 |
NOT | +0.591 |
fore | +0.563 |
't | +0.553 |
Neg logit contributions | |
utenberg | -0.788 |
rament | -0.787 |
peed | -0.749 |
cott | -0.740 |
ranging | -0.716 |
Token | gan |
Token ID | 1030 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
utenberg | +1.177 |
rament | +1.165 |
peed | +1.077 |
cott | +1.049 |
annah | +1.000 |
Neg logit contributions | |
Jinping | -2.556 |
Statements | -2.498 |
NOT | -2.429 |
fore | -2.341 |
't | -2.289 |
Token | Middle |
Token ID | 6046 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
Jinping | +2.445 |
Statements | +2.324 |
NOT | +2.232 |
fore | +2.155 |
't | +2.123 |
Neg logit contributions | |
rament | -3.159 |
utenberg | -3.154 |
peed | -3.022 |
cott | -2.956 |
ranging | -2.894 |
Token | School |
Token ID | 3961 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
rament | +2.090 |
utenberg | +2.087 |
peed | +1.972 |
cott | +1.963 |
ranging | +1.882 |
Neg logit contributions | |
Jinping | -2.043 |
Statements | -1.935 |
NOT | -1.885 |
fore | -1.800 |
't | -1.753 |
Token | 's |
Token ID | 338 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Jinping | +2.253 |
Statements | +2.146 |
NOT | +2.063 |
fore | +1.977 |
't | +1.961 |
Neg logit contributions | |
utenberg | -2.720 |
rament | -2.714 |
peed | -2.608 |
cott | -2.550 |
ranging | -2.494 |
Token | � |
Token ID | 447 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
rament | +0.299 |
utenberg | +0.299 |
peed | +0.285 |
cott | +0.281 |
ranging | +0.272 |
Neg logit contributions | |
Jinping | -0.251 |
Statements | -0.237 |
NOT | -0.230 |
fore | -0.220 |
't | -0.215 |
Token | � |
Token ID | 251 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Jinping | +0.125 |
Statements | +0.119 |
NOT | +0.115 |
fore | +0.110 |
't | +0.108 |
Neg logit contributions | |
utenberg | -0.154 |
rament | -0.153 |
peed | -0.146 |
cott | -0.145 |
ranging | -0.140 |
Token | Cl |
Token ID | 2601 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
Statements | +6.224 |
Jinping | +6.220 |
NOT | +6.162 |
fore | +5.746 |
't | +5.731 |
Neg logit contributions | |
utenberg | -7.201 |
rament | -6.888 |
peed | -6.613 |
cott | -6.401 |
"$:/ | -6.297 |
Token | arks |
Token ID | 5558 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
Jinping | +2.197 |
Statements | +2.087 |
NOT | +2.077 |
fore | +1.963 |
't | +1.925 |
Neg logit contributions | |
utenberg | -3.995 |
rament | -3.963 |
peed | -3.815 |
cott | -3.786 |
ranging | -3.677 |
Token | on |
Token ID | 261 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
rament | +2.900 |
utenberg | +2.822 |
peed | +2.784 |
cott | +2.730 |
ranging | +2.497 |
Neg logit contributions | |
Jinping | -5.281 |
Statements | -5.064 |
NOT | -4.924 |
fore | -4.690 |
't | -4.685 |
Token | added |
Token ID | 2087 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
Jinping | +1.272 |
Statements | +1.215 |
NOT | +1.187 |
fore | +1.141 |
't | +1.124 |
Neg logit contributions | |
utenberg | -1.204 |
rament | -1.199 |
peed | -1.135 |
cott | -1.115 |
ranging | -1.078 |
Token | : |
Token ID | 25 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
utenberg | +1.403 |
rament | +1.397 |
peed | +1.317 |
cott | +1.297 |
ranging | +1.245 |
Neg logit contributions | |
Jinping | -1.711 |
Statements | -1.640 |
NOT | -1.599 |
fore | -1.539 |
't | -1.517 |
Token | � |
Token ID | 564 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
rament | +3.371 |
utenberg | +3.326 |
peed | +3.126 |
cott | +3.097 |
ranging | +2.982 |
Neg logit contributions | |
Jinping | -4.536 |
Statements | -4.298 |
NOT | -4.171 |
fore | -4.043 |
't | -3.962 |
Token | � |
Token ID | 250 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
rament | +0.197 |
utenberg | +0.196 |
peed | +0.188 |
cott | +0.184 |
ranging | +0.179 |
Neg logit contributions | |
Jinping | -0.182 |
Statements | -0.173 |
NOT | -0.167 |
fore | -0.161 |
't | -0.156 |
Token | He |
Token ID | 1544 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
Jinping | +1.780 |
Statements | +1.715 |
NOT | +1.692 |
fore | +1.624 |
't | +1.590 |
Neg logit contributions | |
utenberg | -1.976 |
rament | -1.952 |
peed | -1.849 |
cott | -1.814 |
"$:/ | -1.775 |
Token | will |
Token ID | 481 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
rament | +3.145 |
utenberg | +3.137 |
cott | +2.985 |
peed | +2.976 |
annah | +2.867 |
Neg logit contributions | |
Jinping | -2.213 |
Statements | -2.117 |
NOT | -2.023 |
fore | -1.899 |
't | -1.824 |
Token | , |
Token ID | 11 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
utenberg | +2.488 |
rament | +2.485 |
peed | +2.378 |
cott | +2.329 |
ranging | +2.244 |
Neg logit contributions | |
Jinping | -2.686 |
Statements | -2.554 |
NOT | -2.499 |
fore | -2.396 |
't | -2.331 |
Token | high |
Token ID | 1029 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
Jinping | +2.746 |
Statements | +2.596 |
NOT | +2.544 |
fore | +2.459 |
't | +2.434 |
Neg logit contributions | |
utenberg | -2.638 |
rament | -2.627 |
peed | -2.470 |
cott | -2.448 |
"$:/ | -2.374 |
Token | - |
Token ID | 12 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
utenberg | +1.407 |
rament | +1.405 |
peed | +1.327 |
cott | +1.311 |
ranging | +1.263 |
Neg logit contributions | |
Jinping | -1.480 |
Statements | -1.416 |
NOT | -1.375 |
fore | -1.321 |
't | -1.301 |
Token | output |
Token ID | 22915 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Jinping | +0.243 |
Statements | +0.230 |
NOT | +0.225 |
fore | +0.214 |
't | +0.210 |
Neg logit contributions | |
utenberg | -0.342 |
rament | -0.342 |
peed | -0.324 |
cott | -0.320 |
"$:/ | -0.315 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
utenberg | +0.938 |
rament | +0.938 |
peed | +0.888 |
cott | +0.872 |
ranging | +0.839 |
Neg logit contributions | |
Jinping | -1.077 |
Statements | -1.029 |
NOT | -1.003 |
fore | -0.964 |
't | -0.945 |
Token | . |
Token ID | 13 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
Jinping | +0.928 |
Statements | +0.899 |
NOT | +0.860 |
't | +0.820 |
fore | +0.819 |
Neg logit contributions | |
rament | -1.002 |
utenberg | -0.991 |
peed | -0.930 |
cott | -0.930 |
"$:/ | -0.917 |
Token | 5 |
Token ID | 20 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
rament | +2.876 |
utenberg | +2.849 |
peed | +2.774 |
cott | +2.729 |
ranging | +2.637 |
Neg logit contributions | |
Jinping | -3.453 |
Statements | -3.283 |
NOT | -3.142 |
fore | -3.043 |
annot | -2.981 |
Token | - |
Token ID | 12 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
utenberg | +0.969 |
rament | +0.966 |
peed | +0.916 |
cott | +0.903 |
ranging | +0.869 |
Neg logit contributions | |
Jinping | -1.052 |
Statements | -1.003 |
NOT | -0.978 |
fore | -0.939 |
't | -0.921 |
Token | liter |
Token ID | 17201 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
Jinping | +1.304 |
Statements | +1.249 |
NOT | +1.204 |
fore | +1.151 |
't | +1.131 |
Neg logit contributions | |
rament | -1.604 |
utenberg | -1.588 |
peed | -1.501 |
cott | -1.483 |
"$:/ | -1.469 |
Token | Eco |
Token ID | 38719 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
Jinping | +1.710 |
Statements | +1.630 |
NOT | +1.574 |
fore | +1.504 |
't | +1.498 |
Neg logit contributions | |
utenberg | -2.013 |
rament | -2.012 |
peed | -1.897 |
cott | -1.886 |
"$:/ | -1.826 |
Token | Boost |
Token ID | 45686 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
rament | +1.012 |
utenberg | +1.007 |
peed | +0.960 |
cott | +0.947 |
ranging | +0.913 |
Neg logit contributions | |
Jinping | -1.030 |
Statements | -0.977 |
NOT | -0.949 |
fore | -0.911 |
't | -0.893 |
Token | of |
Token ID | 286 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
Jinping | +4.213 |
Statements | +3.898 |
NOT | +3.790 |
fore | +3.746 |
't | +3.708 |
Neg logit contributions | |
utenberg | -4.198 |
rament | -4.147 |
cott | -3.900 |
peed | -3.861 |
annah | -3.710 |
Token | my |
Token ID | 616 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
Jinping | +1.295 |
Statements | +1.222 |
NOT | +1.183 |
fore | +1.146 |
't | +1.139 |
Neg logit contributions | |
utenberg | -1.284 |
rament | -1.272 |
peed | -1.213 |
cott | -1.191 |
"$:/ | -1.150 |
Token | family |
Token ID | 1641 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
utenberg | +1.962 |
rament | +1.962 |
peed | +1.842 |
cott | +1.825 |
ranging | +1.767 |
Neg logit contributions | |
Jinping | -2.146 |
Statements | -2.048 |
NOT | -1.999 |
fore | -1.912 |
't | -1.873 |
Token | history |
Token ID | 2106 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
Jinping | +2.662 |
Statements | +2.525 |
NOT | +2.403 |
fore | +2.365 |
't | +2.342 |
Neg logit contributions | |
utenberg | -3.372 |
rament | -3.334 |
peed | -3.225 |
cott | -3.180 |
"$:/ | -3.066 |
Token | and |
Token ID | 290 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
Jinping | +2.099 |
Statements | +2.013 |
NOT | +1.931 |
fore | +1.885 |
't | +1.865 |
Neg logit contributions | |
utenberg | -2.059 |
rament | -2.045 |
peed | -1.939 |
cott | -1.913 |
"$:/ | -1.842 |
Token | of |
Token ID | 286 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
Jinping | +2.197 |
Statements | +2.101 |
NOT | +2.027 |
fore | +1.969 |
't | +1.948 |
Neg logit contributions | |
utenberg | -2.089 |
rament | -2.064 |
peed | -1.959 |
cott | -1.929 |
"$:/ | -1.856 |
Token | which |
Token ID | 543 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Jinping | +1.093 |
Statements | +1.040 |
NOT | +1.008 |
fore | +0.977 |
't | +0.964 |
Neg logit contributions | |
utenberg | -1.032 |
rament | -1.026 |
peed | -0.976 |
cott | -0.958 |
"$:/ | -0.921 |
Token | I |
Token ID | 314 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
Jinping | +3.582 |
Statements | +3.385 |
NOT | +3.237 |
fore | +3.156 |
't | +3.139 |
Neg logit contributions | |
utenberg | -4.070 |
rament | -4.025 |
peed | -3.853 |
cott | -3.795 |
"$:/ | -3.654 |
Token | � |
Token ID | 447 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Jinping | +5.685 |
NOT | +5.122 |
Statements | +5.122 |
't | +5.107 |
fore | +4.970 |
Neg logit contributions | |
utenberg | -6.831 |
rament | -6.623 |
peed | -6.521 |
cott | -6.348 |
"$:/ | -6.186 |
Token | � |
Token ID | 247 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
Jinping | +0.224 |
Statements | +0.213 |
NOT | +0.207 |
fore | +0.199 |
't | +0.195 |
Neg logit contributions | |
utenberg | -0.227 |
rament | -0.227 |
peed | -0.215 |
cott | -0.212 |
ranging | -0.204 |
Token | m |
Token ID | 76 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
utenberg | +0.401 |
rament | +0.400 |
peed | +0.379 |
cott | +0.373 |
ranging | +0.360 |
Neg logit contributions | |
Jinping | -0.408 |
Statements | -0.389 |
NOT | -0.378 |
fore | -0.363 |
't | -0.356 |
Token | ALL |
Token ID | 7036 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
Jinping | +2.008 |
NOT | +1.924 |
Statements | +1.907 |
fore | +1.745 |
't | +1.735 |
Neg logit contributions | |
utenberg | -2.845 |
rament | -2.764 |
peed | -2.660 |
cott | -2.632 |
ranging | -2.547 |
Token | , |
Token ID | 11 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
utenberg | +1.591 |
rament | +1.589 |
peed | +1.491 |
cott | +1.470 |
ranging | +1.404 |
Neg logit contributions | |
Jinping | -2.231 |
Statements | -2.142 |
NOT | -2.079 |
fore | -2.017 |
't | -1.980 |
Token | A |
Token ID | 317 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
rament | +2.361 |
utenberg | +2.350 |
peed | +2.238 |
cott | +2.213 |
"$:/ | +2.134 |
Neg logit contributions | |
Jinping | -2.137 |
Statements | -2.026 |
NOT | -1.944 |
fore | -1.880 |
't | -1.826 |
Token | UB |
Token ID | 10526 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
rament | +0.801 |
utenberg | +0.800 |
peed | +0.761 |
cott | +0.754 |
ranging | +0.731 |
Neg logit contributions | |
Jinping | -0.565 |
Statements | -0.533 |
NOT | -0.509 |
fore | -0.484 |
't | -0.473 |
Token | URN |
Token ID | 27064 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
rament | +2.945 |
utenberg | +2.808 |
peed | +2.624 |
cott | +2.609 |
annah | +2.477 |
Neg logit contributions | |
Jinping | -5.586 |
Statements | -5.285 |
NOT | -5.166 |
fore | -5.110 |
annot | -4.951 |
Token |
|
Token ID | 198 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
utenberg | +1.094 |
rament | +1.090 |
peed | +1.038 |
cott | +1.026 |
ranging | +0.995 |
Neg logit contributions | |
Jinping | -0.900 |
Statements | -0.855 |
NOT | -0.832 |
fore | -0.790 |
't | -0.778 |
Token |
|
Token ID | 198 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
utenberg | +0.311 |
rament | +0.311 |
peed | +0.297 |
cott | +0.293 |
ranging | +0.284 |
Neg logit contributions | |
Jinping | -0.293 |
Statements | -0.279 |
NOT | -0.267 |
fore | -0.256 |
annot | -0.248 |
Token | Y |
Token ID | 56 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
Jinping | +0.942 |
Statements | +0.887 |
NOT | +0.865 |
fore | +0.800 |
't | +0.786 |
Neg logit contributions | |
rament | -1.797 |
utenberg | -1.795 |
peed | -1.715 |
cott | -1.699 |
"$:/ | -1.660 |
Token | ards |
Token ID | 1371 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
Jinping | +4.237 |
Statements | +4.003 |
NOT | +3.824 |
fore | +3.669 |
't | +3.648 |
Neg logit contributions | |
rament | -6.950 |
utenberg | -6.924 |
peed | -6.569 |
cott | -6.508 |
"$:/ | -6.425 |
Token | : |
Token ID | 25 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
peed | +4.831 |
rament | +4.821 |
utenberg | +4.800 |
cott | +4.638 |
ranging | +4.517 |
Neg logit contributions | |
Jinping | -9.273 |
Statements | -8.934 |
NOT | -8.563 |
fore | -8.273 |
annot | -8.185 |
Token | 307 |
Token ID | 38369 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
rament | +2.736 |
utenberg | +2.735 |
peed | +2.600 |
cott | +2.568 |
ranging | +2.465 |
Neg logit contributions | |
Jinping | -3.230 |
Statements | -3.047 |
NOT | -2.983 |
fore | -2.889 |
annot | -2.819 |
Token | Star |
Token ID | 2907 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
rament | +1.401 |
utenberg | +1.400 |
peed | +1.327 |
cott | +1.310 |
ranging | +1.263 |
Neg logit contributions | |
Jinping | -1.412 |
Statements | -1.349 |
NOT | -1.310 |
fore | -1.255 |
't | -1.230 |
Token | Wars |
Token ID | 6176 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
rament | +2.439 |
utenberg | +2.428 |
peed | +2.312 |
cott | +2.292 |
ranging | +2.212 |
Neg logit contributions | |
Jinping | -2.113 |
Statements | -2.003 |
NOT | -1.942 |
fore | -1.857 |
't | -1.808 |
Token | movie |
Token ID | 3807 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
Jinping | +1.377 |
Statements | +1.310 |
NOT | +1.272 |
fore | +1.222 |
't | +1.217 |
Neg logit contributions | |
utenberg | -1.331 |
rament | -1.325 |
peed | -1.234 |
cott | -1.224 |
ranging | -1.190 |
Token | in |
Token ID | 287 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
Jinping | +2.185 |
Statements | +2.054 |
NOT | +2.011 |
't | +1.928 |
fore | +1.927 |
Neg logit contributions | |
utenberg | -2.258 |
rament | -2.247 |
peed | -2.104 |
cott | -2.083 |
ranging | -2.012 |
Token | the |
Token ID | 262 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Jinping | +1.183 |
Statements | +1.122 |
NOT | +1.083 |
fore | +1.045 |
't | +1.037 |
Neg logit contributions | |
rament | -1.162 |
utenberg | -1.162 |
peed | -1.098 |
cott | -1.080 |
"$:/ | -1.043 |
Token | age |
Token ID | 2479 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
Jinping | +1.034 |
Statements | +0.981 |
NOT | +0.949 |
fore | +0.909 |
't | +0.900 |
Neg logit contributions | |
rament | -1.098 |
utenberg | -1.096 |
peed | -1.041 |
cott | -1.029 |
ranging | -0.992 |
Token | of |
Token ID | 286 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
Jinping | +2.064 |
Statements | +1.925 |
NOT | +1.896 |
fore | +1.807 |
't | +1.794 |
Neg logit contributions | |
rament | -2.941 |
utenberg | -2.921 |
peed | -2.787 |
cott | -2.755 |
"$:/ | -2.661 |
Token | social |
Token ID | 1919 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
Jinping | +2.363 |
Statements | +2.238 |
NOT | +2.164 |
fore | +2.076 |
't | +2.076 |
Neg logit contributions | |
rament | -2.351 |
utenberg | -2.324 |
peed | -2.200 |
cott | -2.168 |
ranging | -2.092 |
Token | media |
Token ID | 2056 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
Jinping | +1.024 |
Statements | +0.974 |
NOT | +0.943 |
't | +0.894 |
fore | +0.892 |
Neg logit contributions | |
rament | -1.174 |
utenberg | -1.174 |
peed | -1.114 |
cott | -1.101 |
ranging | -1.065 |
Token | , |
Token ID | 11 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
utenberg | +2.224 |
rament | +2.223 |
peed | +2.106 |
cott | +2.071 |
ranging | +1.995 |
Neg logit contributions | |
Jinping | -2.462 |
Statements | -2.362 |
NOT | -2.302 |
fore | -2.207 |
't | -2.156 |
Token | and |
Token ID | 290 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
utenberg | +0.607 |
rament | +0.604 |
peed | +0.571 |
cott | +0.561 |
ranging | +0.541 |
Neg logit contributions | |
Jinping | -0.730 |
Statements | -0.701 |
NOT | -0.681 |
fore | -0.656 |
't | -0.644 |
Token | is |
Token ID | 318 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
Jinping | +3.572 |
Statements | +3.128 |
fore | +3.054 |
't | +3.031 |
NOT | +3.028 |
Neg logit contributions | |
utenberg | -6.148 |
rament | -6.146 |
peed | -5.936 |
cott | -5.761 |
ranging | -5.714 |
Token | up |
Token ID | 510 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Jinping | +3.563 |
Statements | +3.360 |
NOT | +3.301 |
't | +3.190 |
fore | +3.143 |
Neg logit contributions | |
utenberg | -3.697 |
rament | -3.667 |
peed | -3.461 |
cott | -3.413 |
"$:/ | -3.339 |
Token | to |
Token ID | 284 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
Jinping | +4.363 |
fore | +3.979 |
NOT | +3.973 |
Statements | +3.942 |
't | +3.912 |
Neg logit contributions | |
utenberg | -9.630 |
rament | -9.426 |
peed | -9.144 |
cott | -8.876 |
"$:/ | -8.703 |
Token | them |
Token ID | 606 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
Jinping | +3.437 |
Statements | +3.240 |
NOT | +3.115 |
fore | +3.043 |
't | +3.032 |
Neg logit contributions | |
utenberg | -3.273 |
rament | -3.201 |
peed | -3.044 |
cott | -3.026 |
ranging | -2.918 |
Token | how |
Token ID | 703 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
Jinping | +3.099 |
Statements | +2.829 |
NOT | +2.770 |
't | +2.678 |
fore | +2.675 |
Neg logit contributions | |
utenberg | -4.243 |
rament | -4.183 |
peed | -4.009 |
cott | -3.947 |
ranging | -3.822 |
Token | to |
Token ID | 284 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
Jinping | +1.005 |
Statements | +0.957 |
NOT | +0.928 |
fore | +0.892 |
't | +0.880 |
Neg logit contributions | |
utenberg | -1.072 |
rament | -1.062 |
peed | -1.007 |
cott | -0.992 |
ranging | -0.961 |
Token | rule |
Token ID | 3896 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
Jinping | +3.152 |
Statements | +2.974 |
NOT | +2.933 |
't | +2.833 |
fore | +2.806 |
Neg logit contributions | |
rament | -2.869 |
utenberg | -2.861 |
cott | -2.687 |
peed | -2.685 |
ranging | -2.567 |
Token | it |
Token ID | 340 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
rament | +3.868 |
utenberg | +3.804 |
peed | +3.646 |
cott | +3.585 |
ranging | +3.432 |
Neg logit contributions | |
Jinping | -4.489 |
Statements | -4.346 |
NOT | -4.243 |
fore | -4.041 |
annot | -3.970 |
Token | after |
Token ID | 706 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
utenberg | +0.907 |
rament | +0.906 |
peed | +0.858 |
cott | +0.846 |
ranging | +0.815 |
Neg logit contributions | |
Jinping | -0.958 |
Statements | -0.917 |
NOT | -0.891 |
fore | -0.853 |
't | -0.838 |
Token | ." |
Token ID | 526 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Jinping | +2.966 |
Statements | +2.813 |
NOT | +2.711 |
fore | +2.639 |
't | +2.610 |
Neg logit contributions | |
utenberg | -2.839 |
rament | -2.807 |
peed | -2.677 |
cott | -2.614 |
ranging | -2.514 |
Token |
|
Token ID | 198 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
rament | +2.019 |
utenberg | +2.018 |
peed | +1.919 |
cott | +1.897 |
ranging | +1.834 |
Neg logit contributions | |
Jinping | -1.769 |
Statements | -1.676 |
NOT | -1.627 |
fore | -1.557 |
't | -1.523 |
Token | , |
Token ID | 11 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
utenberg | +0.502 |
rament | +0.501 |
peed | +0.467 |
cott | +0.466 |
annah | +0.442 |
Neg logit contributions | |
Jinping | -0.685 |
Statements | -0.655 |
NOT | -0.638 |
fore | -0.618 |
't | -0.611 |
Token | my |
Token ID | 616 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
Jinping | +2.450 |
Statements | +2.327 |
NOT | +2.223 |
fore | +2.135 |
't | +2.116 |
Neg logit contributions | |
utenberg | -2.956 |
rament | -2.945 |
peed | -2.820 |
cott | -2.806 |
annah | -2.689 |
Token | wife |
Token ID | 3656 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
Jinping | +2.219 |
Statements | +2.153 |
NOT | +2.068 |
fore | +2.007 |
't | +1.980 |
Neg logit contributions | |
utenberg | -1.900 |
rament | -1.897 |
peed | -1.781 |
cott | -1.762 |
ranging | -1.675 |
Token | has |
Token ID | 468 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
Jinping | +5.554 |
Statements | +5.129 |
NOT | +4.943 |
't | +4.825 |
fore | +4.825 |
Neg logit contributions | |
utenberg | -5.544 |
rament | -5.424 |
peed | -5.174 |
cott | -5.041 |
ranging | -4.986 |
Token | convinced |
Token ID | 9431 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
rament | +1.100 |
utenberg | +1.099 |
peed | +1.037 |
cott | +1.020 |
ranging | +0.980 |
Neg logit contributions | |
Jinping | -1.363 |
Statements | -1.307 |
NOT | -1.275 |
fore | -1.227 |
't | -1.198 |
Token | me |
Token ID | 502 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
Jinping | +1.860 |
Statements | +1.772 |
NOT | +1.718 |
't | +1.662 |
fore | +1.660 |
Neg logit contributions | |
utenberg | -1.588 |
rament | -1.566 |
peed | -1.475 |
cott | -1.461 |
ranging | -1.400 |
Token | I |
Token ID | 314 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
Jinping | +2.917 |
Statements | +2.768 |
NOT | +2.714 |
't | +2.588 |
fore | +2.564 |
Neg logit contributions | |
utenberg | -3.069 |
rament | -3.040 |
peed | -2.882 |
cott | -2.842 |
ranging | -2.747 |
Token | just |
Token ID | 655 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Jinping | +3.267 |
Statements | +3.018 |
NOT | +3.009 |
't | +2.903 |
fore | +2.847 |
Neg logit contributions | |
utenberg | -4.234 |
rament | -4.171 |
peed | -3.941 |
cott | -3.915 |
ranging | -3.833 |
Token | can |
Token ID | 460 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Jinping | +0.765 |
Statements | +0.725 |
NOT | +0.705 |
fore | +0.672 |
't | +0.666 |
Neg logit contributions | |
rament | -0.883 |
utenberg | -0.882 |
peed | -0.837 |
cott | -0.825 |
ranging | -0.801 |
Token | � |
Token ID | 447 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
't | +6.038 |
NOT | +5.500 |
Jinping | +5.254 |
Statements | +4.913 |
fore | +4.298 |
Neg logit contributions | |
rament | -13.368 |
utenberg | -13.001 |
cott | -12.514 |
peed | -12.413 |
owship | -12.163 |
Token | � |
Token ID | 247 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
rament | +1.367 |
utenberg | +1.360 |
peed | +1.299 |
cott | +1.272 |
ranging | +1.236 |
Neg logit contributions | |
Jinping | -1.346 |
Statements | -1.289 |
NOT | -1.243 |
fore | -1.199 |
't | -1.160 |
Token | : |
Token ID | 25 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
Jinping | +3.289 |
Statements | +3.094 |
NOT | +2.927 |
't | +2.845 |
fore | +2.824 |
Neg logit contributions | |
rament | -4.144 |
utenberg | -4.111 |
cott | -3.868 |
peed | -3.836 |
"$:/ | -3.724 |
Token | Stan |
Token ID | 7299 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
Jinping | +2.397 |
Statements | +2.279 |
NOT | +2.168 |
fore | +2.063 |
't | +2.055 |
Neg logit contributions | |
rament | -3.300 |
utenberg | -3.281 |
peed | -3.112 |
cott | -3.079 |
"$:/ | -3.002 |
Token | Walker |
Token ID | 10120 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
utenberg | +2.469 |
rament | +2.468 |
peed | +2.346 |
cott | +2.339 |
ranging | +2.243 |
Neg logit contributions | |
Jinping | -2.234 |
Statements | -2.130 |
NOT | -2.067 |
fore | -1.962 |
't | -1.919 |
Token | , |
Token ID | 11 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
utenberg | +3.136 |
rament | +3.093 |
peed | +2.956 |
cott | +2.951 |
ranging | +2.788 |
Neg logit contributions | |
Jinping | -4.317 |
Statements | -4.161 |
NOT | -4.063 |
fore | -3.886 |
't | -3.806 |
Token | Natalie |
Token ID | 35571 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
Jinping | +3.677 |
Statements | +3.457 |
NOT | +3.292 |
fore | +3.193 |
't | +3.189 |
Neg logit contributions | |
rament | -4.085 |
utenberg | -4.071 |
peed | -3.824 |
cott | -3.769 |
"$:/ | -3.686 |
Token | B |
Token ID | 347 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
Jinping | +4.715 |
Statements | +4.495 |
NOT | +4.226 |
't | +4.209 |
fore | +4.131 |
Neg logit contributions | |
rament | -5.161 |
utenberg | -5.149 |
peed | -4.821 |
cott | -4.667 |
ranging | -4.598 |
Token | assing |
Token ID | 19696 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Jinping | +2.401 |
Statements | +2.304 |
NOT | +2.248 |
fore | +2.190 |
't | +2.170 |
Neg logit contributions | |
rament | -1.999 |
utenberg | -1.991 |
peed | -1.870 |
cott | -1.847 |
"$:/ | -1.794 |
Token | th |
Token ID | 400 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
rament | +0.882 |
utenberg | +0.880 |
peed | +0.824 |
cott | +0.807 |
ranging | +0.774 |
Neg logit contributions | |
Jinping | -1.253 |
Statements | -1.205 |
NOT | -1.174 |
fore | -1.138 |
't | -1.120 |
Token | wa |
Token ID | 10247 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
Jinping | +3.046 |
Statements | +2.910 |
NOT | +2.856 |
fore | +2.832 |
't | +2.769 |
Neg logit contributions | |
rament | -2.626 |
utenberg | -2.601 |
peed | -2.496 |
"$:/ | -2.421 |
cott | -2.361 |
Token | ight |
Token ID | 432 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
Jinping | +2.107 |
Statements | +1.939 |
NOT | +1.858 |
fore | +1.758 |
't | +1.753 |
Neg logit contributions | |
rament | -4.699 |
utenberg | -4.653 |
peed | -4.513 |
cott | -4.443 |
ranging | -4.379 |
Token | e |
Token ID | 68 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
rament | +0.977 |
utenberg | +0.969 |
peed | +0.910 |
cott | +0.891 |
"$:/ | +0.858 |
Neg logit contributions | |
Jinping | -1.314 |
Statements | -1.263 |
NOT | -1.231 |
fore | -1.192 |
't | -1.174 |
Token | goals |
Token ID | 4661 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
utenberg | +0.305 |
rament | +0.304 |
peed | +0.290 |
cott | +0.286 |
ranging | +0.277 |
Neg logit contributions | |
Jinping | -0.257 |
Statements | -0.245 |
NOT | -0.236 |
fore | -0.226 |
't | -0.222 |
Token | are |
Token ID | 389 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
Jinping | +1.940 |
Statements | +1.871 |
NOT | +1.782 |
fore | +1.725 |
't | +1.712 |
Neg logit contributions | |
utenberg | -2.232 |
rament | -2.208 |
cott | -2.093 |
peed | -2.089 |
"$:/ | -2.021 |
Token | focused |
Token ID | 5670 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
Jinping | +4.135 |
Statements | +3.958 |
NOT | +3.836 |
't | +3.744 |
fore | +3.685 |
Neg logit contributions | |
utenberg | -3.934 |
rament | -3.913 |
peed | -3.715 |
cott | -3.652 |
annah | -3.524 |
Token | on |
Token ID | 319 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
utenberg | +0.987 |
rament | +0.986 |
peed | +0.928 |
cott | +0.916 |
ranging | +0.880 |
Neg logit contributions | |
Jinping | -1.176 |
Statements | -1.126 |
NOT | -1.096 |
fore | -1.055 |
't | -1.039 |
Token | expanding |
Token ID | 11581 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
utenberg | +0.195 |
rament | +0.194 |
peed | +0.183 |
cott | +0.180 |
ranging | +0.172 |
Neg logit contributions | |
Jinping | -0.258 |
Statements | -0.248 |
NOT | -0.241 |
fore | -0.233 |
't | -0.230 |
Token | tree |
Token ID | 5509 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
rament | +5.000 |
utenberg | +4.932 |
peed | +4.690 |
cott | +4.610 |
ranging | +4.444 |
Neg logit contributions | |
Jinping | -6.012 |
Statements | -5.708 |
NOT | -5.547 |
fore | -5.327 |
annot | -5.192 |
Token | canopy |
Token ID | 39418 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
Jinping | +2.481 |
Statements | +2.396 |
NOT | +2.346 |
fore | +2.291 |
't | +2.265 |
Neg logit contributions | |
utenberg | -2.228 |
rament | -2.185 |
peed | -2.047 |
cott | -2.014 |
"$:/ | -1.969 |
Token | cover |
Token ID | 3002 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
utenberg | +0.515 |
rament | +0.513 |
peed | +0.480 |
cott | +0.475 |
"$:/ | +0.455 |
Neg logit contributions | |
Jinping | -0.656 |
Statements | -0.630 |
NOT | -0.614 |
fore | -0.593 |
't | -0.585 |
Token | while |
Token ID | 981 |
Feature activation | +8.94e-01 |
Pos logit contributions | |
utenberg | +3.883 |
rament | +3.880 |
peed | +3.701 |
cott | +3.596 |
ranging | +3.497 |
Neg logit contributions | |
Jinping | -4.866 |
Statements | -4.636 |
NOT | -4.519 |
fore | -4.281 |
annot | -4.232 |
Token | repairing |
Token ID | 39005 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
rament | +4.505 |
utenberg | +4.335 |
peed | +4.192 |
cott | +4.011 |
ranging | +3.953 |
Neg logit contributions | |
Jinping | -6.661 |
Statements | -6.486 |
NOT | -6.280 |
fore | -6.038 |
annot | -5.990 |
Token | the |
Token ID | 262 |
Feature activation | +6.95e-02 |
Pos logit contributions | |
utenberg | +0.205 |
rament | +0.205 |
peed | +0.194 |
cott | +0.191 |
ranging | +0.184 |
Neg logit contributions | |
Jinping | -0.215 |
Statements | -0.205 |
NOT | -0.199 |
fore | -0.192 |
't | -0.189 |
Token | and |
Token ID | 290 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
Jinping | +2.403 |
Statements | +2.289 |
NOT | +2.234 |
fore | +2.152 |
't | +2.131 |
Neg logit contributions | |
rament | -2.365 |
utenberg | -2.344 |
peed | -2.232 |
cott | -2.192 |
"$:/ | -2.123 |
Token | I |
Token ID | 314 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
Jinping | +3.055 |
Statements | +2.905 |
NOT | +2.830 |
fore | +2.725 |
't | +2.714 |
Neg logit contributions | |
utenberg | -3.101 |
rament | -3.086 |
peed | -2.898 |
cott | -2.869 |
"$:/ | -2.774 |
Token | bet |
Token ID | 731 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
Jinping | +3.213 |
Statements | +2.993 |
NOT | +2.960 |
't | +2.859 |
fore | +2.839 |
Neg logit contributions | |
utenberg | -4.167 |
rament | -4.138 |
peed | -3.890 |
cott | -3.882 |
"$:/ | -3.764 |
Token | the |
Token ID | 262 |
Feature activation | -5.15e-02 |
Pos logit contributions | |
Jinping | +2.796 |
Statements | +2.664 |
NOT | +2.589 |
fore | +2.501 |
't | +2.495 |
Neg logit contributions | |
utenberg | -2.579 |
rament | -2.560 |
peed | -2.412 |
cott | -2.373 |
"$:/ | -2.314 |
Token | mayor |
Token ID | 9591 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
Jinping | +0.337 |
Statements | +0.322 |
NOT | +0.312 |
fore | +0.301 |
't | +0.296 |
Neg logit contributions | |
utenberg | -0.328 |
rament | -0.327 |
peed | -0.310 |
cott | -0.305 |
ranging | -0.295 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
Jinping | +2.065 |
Statements | +1.938 |
NOT | +1.888 |
fore | +1.829 |
't | +1.815 |
Neg logit contributions | |
utenberg | -2.210 |
rament | -2.207 |
peed | -2.083 |
cott | -2.050 |
ranging | -1.998 |
Token | 't |
Token ID | 470 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
't | +7.108 |
NOT | +4.903 |
Jinping | +4.226 |
fore | +3.917 |
Statements | +3.533 |
Neg logit contributions | |
proport | -22.566 |
utenberg | -21.789 |
rament | -21.494 |
encount | -21.112 |
owship | -21.064 |
Token | like |
Token ID | 588 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
utenberg | +1.488 |
rament | +1.487 |
peed | +1.421 |
cott | +1.389 |
annah | +1.344 |
Neg logit contributions | |
Jinping | -1.475 |
Statements | -1.416 |
NOT | -1.363 |
fore | -1.305 |
annot | -1.269 |
Token | it |
Token ID | 340 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
rament | +3.508 |
utenberg | +3.446 |
peed | +3.284 |
cott | +3.276 |
ranging | +3.120 |
Neg logit contributions | |
Jinping | -4.648 |
Statements | -4.504 |
NOT | -4.351 |
fore | -4.188 |
annot | -4.075 |
Token | one |
Token ID | 530 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
rament | +3.430 |
utenberg | +3.426 |
peed | +3.274 |
cott | +3.222 |
annah | +3.102 |
Neg logit contributions | |
Jinping | -3.813 |
Statements | -3.713 |
NOT | -3.565 |
fore | -3.378 |
annot | -3.305 |
Token | bit |
Token ID | 1643 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
Jinping | +2.011 |
Statements | +1.843 |
NOT | +1.775 |
't | +1.741 |
fore | +1.727 |
Neg logit contributions | |
rament | -3.327 |
utenberg | -3.314 |
peed | -3.140 |
cott | -3.081 |
ranging | -3.035 |
Token | Greece |
Token ID | 10315 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
rament | +1.510 |
utenberg | +1.497 |
peed | +1.421 |
cott | +1.402 |
annah | +1.362 |
Neg logit contributions | |
Jinping | -2.065 |
Statements | -1.968 |
NOT | -1.952 |
fore | -1.861 |
annot | -1.816 |
Token | in |
Token ID | 287 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Jinping | +5.254 |
Statements | +4.761 |
NOT | +4.643 |
fore | +4.633 |
't | +4.602 |
Neg logit contributions | |
utenberg | -5.299 |
rament | -5.211 |
peed | -4.879 |
cott | -4.837 |
owship | -4.650 |
Token | a |
Token ID | 257 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
rament | +1.099 |
utenberg | +1.099 |
peed | +1.036 |
cott | +1.025 |
ranging | +0.986 |
Neg logit contributions | |
Jinping | -1.223 |
Statements | -1.168 |
NOT | -1.138 |
fore | -1.092 |
't | -1.071 |
Token | way |
Token ID | 835 |
Feature activation | -7.14e-01 |
Pos logit contributions | |
Jinping | +2.812 |
Statements | +2.670 |
NOT | +2.592 |
fore | +2.522 |
't | +2.479 |
Neg logit contributions | |
utenberg | -2.607 |
rament | -2.597 |
peed | -2.466 |
cott | -2.406 |
ranging | -2.322 |
Token | it |
Token ID | 340 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
Jinping | +4.556 |
Statements | +4.284 |
NOT | +4.225 |
't | +4.089 |
fore | +4.064 |
Neg logit contributions | |
utenberg | -4.713 |
rament | -4.621 |
peed | -4.348 |
cott | -4.262 |
annah | -4.094 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
Jinping | +2.829 |
Statements | +2.606 |
NOT | +2.559 |
fore | +2.479 |
't | +2.476 |
Neg logit contributions | |
utenberg | -3.416 |
rament | -3.386 |
peed | -3.238 |
cott | -3.164 |
ranging | -3.077 |
Token | 't |
Token ID | 470 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
't | +4.281 |
NOT | +2.860 |
Jinping | +2.210 |
fore | +1.928 |
Statements | +1.621 |
Neg logit contributions | |
utenberg | -21.094 |
rament | -20.933 |
proport | -20.706 |
owship | -20.311 |
anmar | -20.182 |
Token | in |
Token ID | 287 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
rament | +3.599 |
utenberg | +3.597 |
peed | +3.412 |
cott | +3.348 |
ranging | +3.235 |
Neg logit contributions | |
Jinping | -4.036 |
Statements | -3.935 |
NOT | -3.777 |
fore | -3.567 |
annot | -3.544 |
Token | other |
Token ID | 584 |
Feature activation | +5.38e-03 |
Pos logit contributions | |
Jinping | +0.684 |
Statements | +0.654 |
NOT | +0.632 |
fore | +0.610 |
't | +0.603 |
Neg logit contributions | |
utenberg | -0.660 |
rament | -0.660 |
peed | -0.625 |
cott | -0.613 |
ranging | -0.593 |
Token | cultures |
Token ID | 13817 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
utenberg | +0.034 |
rament | +0.034 |
peed | +0.033 |
cott | +0.032 |
ranging | +0.031 |
Neg logit contributions | |
Jinping | -0.035 |
Statements | -0.033 |
NOT | -0.032 |
fore | -0.031 |
't | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
Jinping | +1.882 |
Statements | +1.767 |
NOT | +1.715 |
fore | +1.676 |
't | +1.662 |
Neg logit contributions | |
utenberg | -2.055 |
rament | -2.044 |
peed | -1.917 |
cott | -1.906 |
"$:/ | -1.852 |
Token | ub |
Token ID | 549 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
rament | +2.828 |
utenberg | +2.798 |
peed | +2.644 |
cott | +2.531 |
ranging | +2.494 |
Neg logit contributions | |
Jinping | -3.223 |
Statements | -3.196 |
NOT | -3.021 |
fore | -2.842 |
annot | -2.754 |
Token | said |
Token ID | 531 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
Jinping | +2.002 |
Statements | +1.890 |
NOT | +1.827 |
fore | +1.797 |
't | +1.771 |
Neg logit contributions | |
utenberg | -2.170 |
rament | -2.169 |
peed | -2.074 |
cott | -2.052 |
"$:/ | -1.998 |
Token | . |
Token ID | 13 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
utenberg | +1.457 |
rament | +1.456 |
peed | +1.381 |
cott | +1.362 |
ranging | +1.315 |
Neg logit contributions | |
Jinping | -1.422 |
Statements | -1.357 |
NOT | -1.317 |
fore | -1.259 |
't | -1.236 |
Token | "... |
Token ID | 27896 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
utenberg | +4.355 |
rament | +4.264 |
peed | +4.013 |
cott | +3.957 |
"$:/ | +3.858 |
Neg logit contributions | |
Jinping | -5.678 |
Statements | -5.337 |
NOT | -5.255 |
fore | -5.057 |
't | -4.939 |
Token | I |
Token ID | 40 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
Jinping | +2.388 |
Statements | +2.323 |
NOT | +2.236 |
fore | +2.127 |
't | +2.093 |
Neg logit contributions | |
utenberg | -3.149 |
rament | -3.121 |
peed | -2.970 |
cott | -2.934 |
ranging | -2.829 |
Token | don |
Token ID | 836 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
Jinping | +2.610 |
Statements | +2.425 |
NOT | +2.388 |
fore | +2.301 |
't | +2.300 |
Neg logit contributions | |
utenberg | -3.637 |
rament | -3.590 |
peed | -3.422 |
cott | -3.389 |
"$:/ | -3.306 |
Token | 't |
Token ID | 470 |
Feature activation | +9.44e-02 |
Pos logit contributions | |
't | +7.606 |
NOT | +6.033 |
Jinping | +5.148 |
Statements | +5.032 |
fore | +5.011 |
Neg logit contributions | |
proport | -19.343 |
utenberg | -19.292 |
rament | -18.905 |
owship | -18.557 |
anmar | -18.415 |
Token | really |
Token ID | 1107 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
rament | +0.688 |
utenberg | +0.688 |
peed | +0.657 |
cott | +0.648 |
ranging | +0.628 |
Neg logit contributions | |
Jinping | -0.528 |
Statements | -0.501 |
NOT | -0.482 |
fore | -0.459 |
't | -0.447 |
Token | care |
Token ID | 1337 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
Jinping | +1.861 |
Statements | +1.733 |
NOT | +1.697 |
't | +1.628 |
fore | +1.619 |
Neg logit contributions | |
utenberg | -2.661 |
rament | -2.639 |
cott | -2.506 |
peed | -2.493 |
ranging | -2.415 |
Token | , |
Token ID | 11 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
rament | +2.559 |
utenberg | +2.545 |
peed | +2.445 |
cott | +2.396 |
ranging | +2.309 |
Neg logit contributions | |
Jinping | -2.613 |
Statements | -2.531 |
NOT | -2.449 |
fore | -2.320 |
annot | -2.275 |
Token | to |
Token ID | 284 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
Jinping | +1.521 |
Statements | +1.456 |
NOT | +1.414 |
fore | +1.361 |
't | +1.345 |
Neg logit contributions | |
utenberg | -1.366 |
rament | -1.362 |
peed | -1.286 |
cott | -1.272 |
ranging | -1.220 |
Token | . |
Token ID | 13 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
Jinping | +0.263 |
Statements | +0.249 |
NOT | +0.242 |
fore | +0.233 |
't | +0.231 |
Neg logit contributions | |
utenberg | -0.276 |
rament | -0.275 |
peed | -0.261 |
cott | -0.258 |
"$:/ | -0.250 |
Token | � |
Token ID | 564 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
utenberg | +3.771 |
rament | +3.760 |
peed | +3.569 |
cott | +3.538 |
"$:/ | +3.432 |
Neg logit contributions | |
Jinping | -3.683 |
Statements | -3.454 |
NOT | -3.384 |
fore | -3.261 |
't | -3.169 |
Token | � |
Token ID | 250 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
utenberg | +0.210 |
rament | +0.210 |
peed | +0.200 |
cott | +0.197 |
ranging | +0.191 |
Neg logit contributions | |
Jinping | -0.179 |
Statements | -0.170 |
NOT | -0.165 |
fore | -0.157 |
't | -0.154 |
Token | I |
Token ID | 40 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
Jinping | +2.265 |
Statements | +2.194 |
NOT | +2.158 |
fore | +2.053 |
't | +2.029 |
Neg logit contributions | |
utenberg | -2.565 |
rament | -2.550 |
peed | -2.410 |
cott | -2.342 |
"$:/ | -2.292 |
Token | personally |
Token ID | 7620 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Jinping | +0.827 |
Statements | +0.779 |
NOT | +0.760 |
fore | +0.725 |
't | +0.718 |
Neg logit contributions | |
utenberg | -1.090 |
rament | -1.082 |
peed | -1.031 |
cott | -1.018 |
"$:/ | -0.988 |
Token | don |
Token ID | 836 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
utenberg | +0.890 |
rament | +0.889 |
peed | +0.848 |
cott | +0.838 |
ranging | +0.812 |
Neg logit contributions | |
Jinping | -0.688 |
Statements | -0.652 |
NOT | -0.630 |
fore | -0.600 |
't | -0.588 |
Token | � |
Token ID | 447 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
't | +10.477 |
NOT | +9.084 |
Jinping | +8.465 |
Statements | +8.280 |
fore | +8.151 |
Neg logit contributions | |
utenberg | -15.756 |
proport | -15.510 |
rament | -15.495 |
owship | -14.986 |
anmar | -14.788 |
Token | � |
Token ID | 247 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
rament | +1.819 |
utenberg | +1.818 |
peed | +1.742 |
cott | +1.720 |
ranging | +1.673 |
Neg logit contributions | |
Jinping | -1.242 |
Statements | -1.176 |
NOT | -1.128 |
fore | -1.074 |
't | -1.039 |
Token | t |
Token ID | 83 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
utenberg | +3.816 |
rament | +3.800 |
peed | +3.666 |
cott | +3.653 |
ranging | +3.513 |
Neg logit contributions | |
Jinping | -3.200 |
Statements | -3.110 |
NOT | -2.956 |
fore | -2.785 |
annot | -2.695 |
Token | agree |
Token ID | 4236 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
utenberg | +0.752 |
rament | +0.751 |
peed | +0.713 |
cott | +0.704 |
ranging | +0.680 |
Neg logit contributions | |
Jinping | -0.700 |
Statements | -0.667 |
NOT | -0.646 |
fore | -0.619 |
't | -0.607 |
Token | with |
Token ID | 351 |
Feature activation | +9.60e-01 |
Pos logit contributions | |
rament | +4.984 |
utenberg | +4.971 |
peed | +4.737 |
cott | +4.643 |
"$:/ | +4.433 |
Neg logit contributions | |
Jinping | -6.567 |
Statements | -6.279 |
NOT | -6.111 |
fore | -5.916 |
annot | -5.783 |
Token | over |
Token ID | 625 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
utenberg | +0.488 |
rament | +0.486 |
peed | +0.461 |
cott | +0.453 |
ranging | +0.437 |
Neg logit contributions | |
Jinping | -0.531 |
Statements | -0.507 |
NOT | -0.493 |
fore | -0.472 |
't | -0.464 |
Token | some |
Token ID | 617 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
Jinping | +5.028 |
Statements | +4.771 |
NOT | +4.562 |
fore | +4.452 |
't | +4.409 |
Neg logit contributions | |
utenberg | -5.209 |
rament | -5.202 |
cott | -4.882 |
peed | -4.875 |
"$:/ | -4.702 |
Token | basic |
Token ID | 4096 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
Jinping | +5.314 |
Statements | +5.065 |
NOT | +4.865 |
fore | +4.669 |
't | +4.615 |
Neg logit contributions | |
utenberg | -6.913 |
rament | -6.831 |
peed | -6.420 |
cott | -6.396 |
annah | -6.211 |
Token | dos |
Token ID | 23430 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
Jinping | +4.891 |
Statements | +4.696 |
NOT | +4.470 |
't | +4.241 |
fore | +4.210 |
Neg logit contributions | |
utenberg | -6.778 |
rament | -6.729 |
peed | -6.293 |
cott | -6.288 |
annah | -6.146 |
Token | and |
Token ID | 290 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
utenberg | +1.377 |
rament | +1.376 |
peed | +1.294 |
cott | +1.274 |
ranging | +1.223 |
Neg logit contributions | |
Jinping | -1.743 |
Statements | -1.671 |
NOT | -1.627 |
fore | -1.568 |
't | -1.543 |
Token | don |
Token ID | 836 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
Jinping | +2.880 |
't | +2.837 |
NOT | +2.381 |
Statements | +2.278 |
fore | +2.244 |
Neg logit contributions | |
utenberg | -13.204 |
rament | -13.018 |
peed | -12.533 |
"$:/ | -12.446 |
cott | -12.338 |
Token | 't |
Token ID | 470 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
't | +7.812 |
Jinping | +5.694 |
NOT | +5.681 |
fore | +5.344 |
Statements | +5.059 |
Neg logit contributions | |
rament | -21.098 |
utenberg | -20.602 |
owship | -19.425 |
proport | -19.416 |
"$:/ | -19.117 |
Token | s |
Token ID | 82 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
Jinping | +2.246 |
Statements | +2.143 |
NOT | +2.061 |
fore | +1.979 |
't | +1.952 |
Neg logit contributions | |
utenberg | -2.772 |
rament | -2.761 |
peed | -2.635 |
cott | -2.587 |
ranging | -2.516 |
Token | , |
Token ID | 11 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
Jinping | +4.834 |
Statements | +4.617 |
NOT | +4.424 |
't | +4.291 |
fore | +4.290 |
Neg logit contributions | |
utenberg | -5.576 |
rament | -5.495 |
cott | -5.170 |
peed | -5.090 |
"$:/ | -4.988 |
Token | using |
Token ID | 1262 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Jinping | +2.274 |
Statements | +2.171 |
NOT | +2.111 |
fore | +2.033 |
't | +2.012 |
Neg logit contributions | |
rament | -2.065 |
utenberg | -2.053 |
peed | -1.934 |
cott | -1.913 |
annah | -1.835 |
Token | some |
Token ID | 617 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Jinping | +6.599 |
Statements | +6.457 |
NOT | +6.145 |
't | +5.973 |
fore | +5.915 |
Neg logit contributions | |
rament | -6.889 |
utenberg | -6.713 |
cott | -6.334 |
peed | -6.325 |
ranging | -6.081 |