Token | repeatedly |
Token ID | 7830 |
Feature activation | -8.38e-02 |
Pos logit contributions | |
ibaba | +0.447 |
emale | +0.435 |
aimon | +0.339 |
mustache | +0.334 |
WRITE | +0.334 |
Neg logit contributions | |
meg | -0.317 |
star | -0.306 |
� | -0.301 |
power | -0.297 |
mania | -0.296 |
Token | criticized |
Token ID | 12318 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
ibaba | +0.362 |
emale | +0.353 |
aimon | +0.274 |
ijing | +0.272 |
mustache | +0.271 |
Neg logit contributions | |
meg | -0.249 |
star | -0.239 |
� | -0.237 |
power | -0.232 |
pres | -0.232 |
Token | President |
Token ID | 1992 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
ibaba | +0.978 |
emale | +0.952 |
aimon | +0.735 |
ijing | +0.725 |
WRITE | +0.718 |
Neg logit contributions | |
meg | -0.715 |
star | -0.698 |
� | -0.683 |
pres | -0.674 |
power | -0.672 |
Token | Trump |
Token ID | 1301 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
ibaba | +0.232 |
emale | +0.225 |
ijing | +0.167 |
aimon | +0.167 |
WRITE | +0.164 |
Neg logit contributions | |
meg | -0.207 |
star | -0.201 |
� | -0.199 |
rovers | -0.194 |
mania | -0.194 |
Token | Donald |
Token ID | 3759 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
ibaba | +0.174 |
emale | +0.170 |
ijing | +0.134 |
aimon | +0.134 |
mustache | +0.133 |
Neg logit contributions | |
meg | -0.097 |
star | -0.092 |
� | -0.091 |
pres | -0.089 |
power | -0.088 |
Token | John |
Token ID | 1757 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
star | +1.832 |
meg | +1.830 |
mania | +1.772 |
power | +1.718 |
pres | +1.676 |
Neg logit contributions | |
ibaba | -4.305 |
emale | -4.178 |
ijing | -3.362 |
aimon | -3.361 |
irtual | -3.282 |
Token | Trump |
Token ID | 1301 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
meg | +3.701 |
power | +3.684 |
star | +3.627 |
pres | +3.568 |
rovers | +3.513 |
Neg logit contributions | |
ibaba | -6.499 |
emale | -6.491 |
aimon | -5.071 |
ijing | -4.954 |
irtual | -4.811 |
Token | House |
Token ID | 18102 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
meg | +1.913 |
star | +1.871 |
power | +1.789 |
pres | +1.778 |
� | +1.769 |
Neg logit contributions | |
ibaba | -3.964 |
emale | -3.886 |
aimon | -3.119 |
ijing | -3.069 |
WRITE | -3.047 |
Token | committee |
Token ID | 5583 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
meg | +1.353 |
star | +1.341 |
power | +1.311 |
mania | +1.276 |
pres | +1.265 |
Neg logit contributions | |
ibaba | -2.252 |
emale | -2.196 |
aimon | -1.743 |
ijing | -1.720 |
WRITE | -1.671 |
Token | believes |
Token ID | 5804 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
meg | +0.086 |
star | +0.083 |
� | +0.079 |
mania | +0.078 |
pres | +0.077 |
Neg logit contributions | |
ibaba | -0.337 |
emale | -0.331 |
aimon | -0.275 |
ijing | -0.272 |
mustache | -0.271 |
Token | it |
Token ID | 340 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
ibaba | +1.815 |
emale | +1.775 |
aimon | +1.357 |
ijing | +1.331 |
WRITE | +1.309 |
Neg logit contributions | |
meg | -1.406 |
star | -1.358 |
pres | -1.315 |
� | -1.311 |
power | -1.303 |
Token | s |
Token ID | 82 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
meg | +0.320 |
star | +0.306 |
� | +0.295 |
mania | +0.289 |
power | +0.289 |
Neg logit contributions | |
ibaba | -0.881 |
emale | -0.861 |
aimon | -0.706 |
ijing | -0.699 |
WRITE | -0.693 |
Token | " |
Token ID | 1 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ibaba | +0.411 |
emale | +0.401 |
aimon | +0.330 |
ijing | +0.327 |
WRITE | +0.324 |
Neg logit contributions | |
meg | -0.149 |
star | -0.142 |
� | -0.138 |
mania | -0.135 |
power | -0.134 |
Token |
|
Token ID | 198 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
ibaba | +0.682 |
emale | +0.661 |
aimon | +0.534 |
ijing | +0.527 |
WRITE | +0.524 |
Neg logit contributions | |
meg | -0.327 |
star | -0.316 |
� | -0.307 |
mania | -0.301 |
power | -0.300 |
Token |
|
Token ID | 198 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
ibaba | +1.672 |
emale | +1.624 |
aimon | +1.325 |
WRITE | +1.306 |
ijing | +1.302 |
Neg logit contributions | |
meg | -0.751 |
star | -0.724 |
� | -0.702 |
pres | -0.688 |
rovers | -0.687 |
Token | Ent |
Token ID | 14539 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
ibaba | +2.182 |
emale | +2.129 |
aimon | +1.707 |
ijing | +1.691 |
WRITE | +1.668 |
Neg logit contributions | |
meg | -1.176 |
star | -1.143 |
� | -1.119 |
rovers | -1.104 |
mania | -1.092 |
Token | ire |
Token ID | 557 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
meg | +0.406 |
star | +0.396 |
� | +0.394 |
mania | +0.383 |
power | +0.382 |
Neg logit contributions | |
ibaba | -0.580 |
emale | -0.563 |
aimon | -0.439 |
ijing | -0.428 |
WRITE | -0.419 |
Token | classrooms |
Token ID | 32579 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
meg | +4.088 |
star | +4.047 |
power | +3.895 |
mania | +3.892 |
pres | +3.852 |
Neg logit contributions | |
ibaba | -5.701 |
emale | -5.536 |
aimon | -4.320 |
WRITE | -4.218 |
ijing | -4.183 |
Token | of |
Token ID | 286 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
meg | +0.077 |
star | +0.075 |
� | +0.074 |
pres | +0.072 |
mania | +0.072 |
Neg logit contributions | |
ibaba | -0.130 |
emale | -0.127 |
aimon | -0.100 |
ijing | -0.099 |
WRITE | -0.098 |
Token | Somali |
Token ID | 30886 |
Feature activation | -5.72e-03 |
Pos logit contributions | |
meg | +0.587 |
star | +0.573 |
� | +0.560 |
mania | +0.554 |
pres | +0.553 |
Neg logit contributions | |
ibaba | -0.791 |
emale | -0.770 |
aimon | -0.593 |
ijing | -0.583 |
WRITE | -0.577 |
Token | children |
Token ID | 1751 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
ibaba | +0.024 |
emale | +0.023 |
aimon | +0.017 |
ijing | +0.017 |
mustache | +0.017 |
Neg logit contributions | |
meg | -0.018 |
star | -0.018 |
� | -0.017 |
rovers | -0.017 |
power | -0.017 |
Token | - |
Token ID | 532 |
Feature activation | -8.47e-02 |
Pos logit contributions | |
ibaba | +0.107 |
emale | +0.105 |
aimon | +0.083 |
ijing | +0.082 |
mustache | +0.081 |
Neg logit contributions | |
meg | -0.065 |
star | -0.063 |
� | -0.062 |
pres | -0.061 |
rovers | -0.060 |
Token | airport |
Token ID | 9003 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
meg | +0.844 |
star | +0.825 |
� | +0.800 |
power | +0.796 |
mania | +0.796 |
Neg logit contributions | |
ibaba | -1.132 |
emale | -1.104 |
aimon | -0.846 |
ijing | -0.831 |
WRITE | -0.822 |
Token | hold |
Token ID | 1745 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
meg | +0.152 |
star | +0.149 |
� | +0.145 |
mania | +0.143 |
power | +0.143 |
Neg logit contributions | |
ibaba | -0.237 |
emale | -0.231 |
aimon | -0.181 |
ijing | -0.178 |
WRITE | -0.176 |
Token | about |
Token ID | 546 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
meg | +0.696 |
star | +0.677 |
� | +0.652 |
power | +0.651 |
pres | +0.651 |
Neg logit contributions | |
ibaba | -1.130 |
emale | -1.106 |
aimon | -0.868 |
ijing | -0.853 |
WRITE | -0.844 |
Token | 34 |
Token ID | 4974 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
meg | +0.361 |
star | +0.353 |
power | +0.342 |
pres | +0.340 |
mania | +0.340 |
Neg logit contributions | |
ibaba | -0.472 |
emale | -0.463 |
aimon | -0.353 |
ijing | -0.346 |
WRITE | -0.341 |
Token | million |
Token ID | 1510 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
meg | +0.789 |
star | +0.767 |
� | +0.740 |
power | +0.735 |
pres | +0.734 |
Neg logit contributions | |
ibaba | -1.441 |
emale | -1.408 |
aimon | -1.119 |
ijing | -1.102 |
WRITE | -1.092 |
Token | gallons |
Token ID | 23239 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
meg | +1.716 |
star | +1.667 |
� | +1.610 |
pres | +1.601 |
power | +1.600 |
Neg logit contributions | |
ibaba | -2.856 |
emale | -2.791 |
aimon | -2.194 |
ijing | -2.160 |
WRITE | -2.138 |
Token | total |
Token ID | 2472 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
meg | +3.194 |
star | +3.087 |
power | +2.967 |
� | +2.954 |
pres | +2.944 |
Neg logit contributions | |
ibaba | -6.505 |
emale | -6.407 |
aimon | -5.132 |
ijing | -5.027 |
WRITE | -4.997 |
Token | , |
Token ID | 11 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
meg | +1.721 |
star | +1.670 |
� | +1.604 |
power | +1.594 |
pres | +1.593 |
Neg logit contributions | |
ibaba | -3.522 |
emale | -3.446 |
aimon | -2.765 |
ijing | -2.725 |
WRITE | -2.701 |
Token | according |
Token ID | 1864 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
meg | +1.202 |
star | +1.170 |
� | +1.130 |
power | +1.123 |
pres | +1.122 |
Neg logit contributions | |
ibaba | -2.061 |
emale | -2.012 |
aimon | -1.589 |
ijing | -1.565 |
WRITE | -1.550 |
Token | to |
Token ID | 284 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
meg | +0.265 |
star | +0.258 |
� | +0.246 |
power | +0.246 |
pres | +0.245 |
Neg logit contributions | |
ibaba | -0.564 |
emale | -0.555 |
aimon | -0.444 |
ijing | -0.439 |
WRITE | -0.433 |
Token | Gale |
Token ID | 36483 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
meg | +0.388 |
star | +0.381 |
� | +0.371 |
pres | +0.370 |
mania | +0.368 |
Neg logit contributions | |
ibaba | -0.536 |
emale | -0.519 |
aimon | -0.399 |
ijing | -0.394 |
WRITE | -0.391 |
Token | Im |
Token ID | 1846 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
meg | +0.831 |
star | +0.813 |
� | +0.787 |
power | +0.783 |
pres | +0.783 |
Neg logit contributions | |
ibaba | -1.156 |
emale | -1.127 |
aimon | -0.869 |
ijing | -0.854 |
WRITE | -0.844 |
Token | more |
Token ID | 3549 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
meg | +1.258 |
star | +1.226 |
� | +1.185 |
power | +1.177 |
pres | +1.177 |
Neg logit contributions | |
ibaba | -2.073 |
emale | -2.024 |
aimon | -1.591 |
ijing | -1.567 |
WRITE | -1.552 |
Token | n |
Token ID | 77 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
ibaba | +0.769 |
emale | +0.752 |
aimon | +0.597 |
ijing | +0.587 |
WRITE | +0.577 |
Neg logit contributions | |
meg | -0.432 |
star | -0.422 |
� | -0.409 |
pres | -0.405 |
power | -0.405 |
Token | are |
Token ID | 389 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
ibaba | +1.479 |
emale | +1.448 |
aimon | +1.142 |
ijing | +1.128 |
WRITE | +1.112 |
Neg logit contributions | |
meg | -0.839 |
star | -0.815 |
� | -0.795 |
mania | -0.789 |
rovers | -0.784 |
Token | made |
Token ID | 925 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
meg | +0.332 |
star | +0.324 |
� | +0.314 |
power | +0.311 |
mania | +0.311 |
Neg logit contributions | |
ibaba | -0.537 |
emale | -0.525 |
aimon | -0.412 |
ijing | -0.405 |
WRITE | -0.401 |
Token | up |
Token ID | 510 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
meg | +0.340 |
star | +0.329 |
� | +0.324 |
mania | +0.318 |
power | +0.317 |
Neg logit contributions | |
ibaba | -0.607 |
emale | -0.595 |
aimon | -0.470 |
ijing | -0.467 |
WRITE | -0.460 |
Token | of |
Token ID | 286 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
meg | +2.937 |
star | +2.821 |
� | +2.694 |
pres | +2.675 |
power | +2.674 |
Neg logit contributions | |
ibaba | -6.774 |
emale | -6.556 |
aimon | -5.356 |
ijing | -5.270 |
WRITE | -5.216 |
Token | a |
Token ID | 257 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
meg | +1.656 |
star | +1.617 |
� | +1.561 |
power | +1.555 |
pres | +1.552 |
Neg logit contributions | |
ibaba | -2.486 |
emale | -2.417 |
aimon | -1.884 |
ijing | -1.853 |
WRITE | -1.832 |
Token | mot |
Token ID | 2369 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
meg | +0.073 |
star | +0.071 |
� | +0.069 |
mania | +0.069 |
power | +0.068 |
Neg logit contributions | |
ibaba | -0.115 |
emale | -0.112 |
aimon | -0.088 |
ijing | -0.086 |
mustache | -0.086 |
Token | ley |
Token ID | 1636 |
Feature activation | +9.77e-02 |
Pos logit contributions | |
meg | +1.109 |
star | +1.075 |
mania | +1.073 |
� | +1.068 |
power | +1.051 |
Neg logit contributions | |
ibaba | -1.858 |
emale | -1.808 |
aimon | -1.416 |
ijing | -1.398 |
mustache | -1.378 |
Token | crew |
Token ID | 5462 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
meg | +0.279 |
star | +0.271 |
� | +0.263 |
mania | +0.261 |
power | +0.261 |
Neg logit contributions | |
ibaba | -0.437 |
emale | -0.427 |
aimon | -0.334 |
ijing | -0.328 |
WRITE | -0.325 |
Token | Democratic |
Token ID | 33939 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
meg | +0.352 |
star | +0.342 |
� | +0.334 |
mania | +0.329 |
pres | +0.328 |
Neg logit contributions | |
ibaba | -0.681 |
emale | -0.664 |
aimon | -0.533 |
ijing | -0.525 |
WRITE | -0.519 |
Token | presidential |
Token ID | 4787 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
ibaba | +0.209 |
emale | +0.205 |
aimon | +0.162 |
ijing | +0.162 |
WRITE | +0.161 |
Neg logit contributions | |
meg | -0.117 |
star | -0.112 |
� | -0.110 |
pres | -0.108 |
power | -0.108 |
Token | nominee |
Token ID | 10429 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
ibaba | +1.827 |
emale | +1.764 |
aimon | +1.314 |
margins | +1.312 |
WRITE | +1.310 |
Neg logit contributions | |
meg | -1.603 |
� | -1.547 |
star | -1.544 |
rovers | -1.504 |
mania | -1.500 |
Token | Hillary |
Token ID | 5073 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
ibaba | +0.866 |
emale | +0.832 |
WRITE | +0.606 |
ijing | +0.603 |
aimon | +0.600 |
Neg logit contributions | |
meg | -0.935 |
star | -0.914 |
� | -0.908 |
rovers | -0.893 |
mania | -0.891 |
Token | Clinton |
Token ID | 2605 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
meg | +1.758 |
star | +1.726 |
� | +1.683 |
power | +1.676 |
pres | +1.676 |
Neg logit contributions | |
ibaba | -1.619 |
emale | -1.569 |
aimon | -1.131 |
ijing | -1.106 |
WRITE | -1.090 |
Token | Hillary |
Token ID | 5073 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ibaba | +0.840 |
emale | +0.816 |
aimon | +0.642 |
ijing | +0.641 |
WRITE | +0.639 |
Neg logit contributions | |
meg | -0.542 |
star | -0.521 |
� | -0.515 |
power | -0.504 |
rovers | -0.504 |
Token | Diane |
Token ID | 26542 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
star | +2.847 |
meg | +2.836 |
mania | +2.653 |
pres | +2.630 |
power | +2.582 |
Neg logit contributions | |
ibaba | -6.596 |
emale | -6.396 |
ijing | -5.249 |
aimon | -5.204 |
mustache | -5.140 |
Token | Rodham |
Token ID | 41848 |
Feature activation | +8.26e-01 |
Pos logit contributions | |
star | +2.599 |
meg | +2.596 |
pres | +2.514 |
mania | +2.487 |
� | +2.469 |
Neg logit contributions | |
ibaba | -3.340 |
emale | -3.248 |
ijing | -2.472 |
aimon | -2.403 |
mustache | -2.386 |
Token | Clinton |
Token ID | 2605 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
meg | +2.925 |
star | +2.913 |
power | +2.823 |
pres | +2.808 |
mania | +2.797 |
Neg logit contributions | |
ibaba | -3.090 |
emale | -3.057 |
aimon | -2.222 |
ijing | -2.205 |
mustache | -2.140 |
Token | Sanders |
Token ID | 26747 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
meg | +1.512 |
star | +1.466 |
� | +1.404 |
power | +1.394 |
pres | +1.394 |
Neg logit contributions | |
ibaba | -3.376 |
emale | -3.303 |
aimon | -2.669 |
ijing | -2.632 |
WRITE | -2.609 |
Token | : |
Token ID | 25 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
meg | +1.305 |
star | +1.272 |
mania | +1.226 |
power | +1.214 |
� | +1.213 |
Neg logit contributions | |
ibaba | -2.602 |
emale | -2.548 |
aimon | -2.030 |
ijing | -2.003 |
WRITE | -1.989 |
Token | that |
Token ID | 326 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
ibaba | +0.055 |
emale | +0.053 |
aimon | +0.042 |
ijing | +0.042 |
WRITE | +0.041 |
Neg logit contributions | |
meg | -0.031 |
star | -0.030 |
� | -0.030 |
power | -0.029 |
mania | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
meg | +0.747 |
star | +0.726 |
power | +0.694 |
pres | +0.693 |
mania | +0.693 |
Neg logit contributions | |
ibaba | -1.497 |
emale | -1.470 |
aimon | -1.177 |
ijing | -1.156 |
WRITE | -1.140 |
Token | traits |
Token ID | 12796 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
meg | +0.247 |
star | +0.242 |
� | +0.234 |
pres | +0.234 |
power | +0.233 |
Neg logit contributions | |
ibaba | -0.316 |
emale | -0.308 |
aimon | -0.235 |
ijing | -0.230 |
WRITE | -0.227 |
Token | / |
Token ID | 14 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
meg | +0.547 |
star | +0.533 |
� | +0.512 |
power | +0.510 |
pres | +0.510 |
Neg logit contributions | |
ibaba | -0.992 |
emale | -0.969 |
aimon | -0.770 |
ijing | -0.757 |
WRITE | -0.750 |
Token | ent |
Token ID | 298 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
meg | +0.402 |
star | +0.391 |
� | +0.377 |
power | +0.375 |
pres | +0.374 |
Neg logit contributions | |
ibaba | -0.729 |
emale | -0.712 |
aimon | -0.565 |
ijing | -0.557 |
WRITE | -0.552 |
Token | itle |
Token ID | 2578 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
meg | +0.877 |
star | +0.856 |
� | +0.827 |
pres | +0.823 |
power | +0.823 |
Neg logit contributions | |
ibaba | -1.392 |
emale | -1.358 |
aimon | -1.064 |
ijing | -1.047 |
WRITE | -1.037 |
Token | ments |
Token ID | 902 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
meg | +1.653 |
star | +1.528 |
power | +1.456 |
pres | +1.438 |
mania | +1.399 |
Neg logit contributions | |
ibaba | -7.882 |
emale | -7.731 |
aimon | -6.505 |
ijing | -6.401 |
WRITE | -6.390 |
Token | we |
Token ID | 356 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
meg | +0.673 |
star | +0.655 |
� | +0.632 |
power | +0.628 |
pres | +0.628 |
Neg logit contributions | |
ibaba | -1.212 |
emale | -1.184 |
aimon | -0.939 |
ijing | -0.925 |
WRITE | -0.916 |
Token | normally |
Token ID | 7685 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
ibaba | +0.981 |
emale | +0.955 |
aimon | +0.745 |
ijing | +0.734 |
WRITE | +0.728 |
Neg logit contributions | |
meg | -0.643 |
star | -0.626 |
� | -0.609 |
power | -0.603 |
pres | -0.602 |
Token | associate |
Token ID | 11602 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
ibaba | +1.361 |
emale | +1.314 |
aimon | +0.999 |
ijing | +0.987 |
WRITE | +0.985 |
Neg logit contributions | |
meg | -1.102 |
star | -1.070 |
� | -1.058 |
power | -1.039 |
pres | -1.038 |
Token | with |
Token ID | 351 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
ibaba | +1.061 |
emale | +1.030 |
aimon | +0.806 |
ijing | +0.798 |
WRITE | +0.788 |
Neg logit contributions | |
meg | -0.669 |
star | -0.655 |
� | -0.634 |
pres | -0.632 |
power | -0.629 |
Token | us |
Token ID | 385 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
meg | +0.534 |
star | +0.518 |
� | +0.508 |
pres | +0.494 |
mania | +0.494 |
Neg logit contributions | |
ibaba | -1.207 |
emale | -1.175 |
aimon | -0.955 |
ijing | -0.942 |
WRITE | -0.929 |
Token | / |
Token ID | 14 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
ibaba | +0.695 |
emale | +0.679 |
aimon | +0.542 |
ijing | +0.537 |
WRITE | +0.528 |
Neg logit contributions | |
meg | -0.353 |
star | -0.345 |
� | -0.343 |
pres | -0.332 |
mania | -0.330 |
Token | wiki |
Token ID | 15466 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
ibaba | +2.382 |
emale | +2.327 |
aimon | +1.861 |
ijing | +1.840 |
WRITE | +1.821 |
Neg logit contributions | |
meg | -1.215 |
star | -1.187 |
� | -1.162 |
mania | -1.143 |
pres | -1.135 |
Token | / |
Token ID | 14 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
ibaba | +1.783 |
emale | +1.748 |
aimon | +1.387 |
ijing | +1.379 |
WRITE | +1.372 |
Neg logit contributions | |
meg | -0.979 |
star | -0.948 |
� | -0.935 |
mania | -0.917 |
pres | -0.916 |
Token | trans |
Token ID | 7645 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
meg | +1.185 |
star | +1.152 |
� | +1.107 |
power | +1.104 |
pres | +1.102 |
Neg logit contributions | |
ibaba | -2.182 |
emale | -2.132 |
aimon | -1.695 |
ijing | -1.670 |
WRITE | -1.653 |
Token | cript |
Token ID | 6519 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
meg | +2.390 |
star | +2.344 |
power | +2.282 |
pres | +2.270 |
mania | +2.258 |
Neg logit contributions | |
ibaba | -2.791 |
emale | -2.700 |
aimon | -2.020 |
ijing | -1.994 |
WRITE | -1.977 |
Token | s |
Token ID | 82 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
meg | +3.427 |
star | +3.322 |
power | +3.207 |
pres | +3.180 |
mania | +3.171 |
Neg logit contributions | |
ibaba | -6.001 |
emale | -5.839 |
aimon | -4.674 |
ijing | -4.556 |
WRITE | -4.495 |
Token | / |
Token ID | 14 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
meg | +0.649 |
star | +0.635 |
� | +0.616 |
pres | +0.609 |
mania | +0.609 |
Neg logit contributions | |
ibaba | -1.043 |
emale | -1.019 |
aimon | -0.798 |
ijing | -0.787 |
WRITE | -0.780 |
Token | sc |
Token ID | 1416 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
meg | +0.769 |
star | +0.748 |
� | +0.722 |
mania | +0.716 |
pres | +0.715 |
Neg logit contributions | |
ibaba | -1.456 |
emale | -1.423 |
aimon | -1.134 |
ijing | -1.118 |
WRITE | -1.108 |
Token | aling |
Token ID | 4272 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
meg | +0.907 |
star | +0.885 |
� | +0.858 |
power | +0.850 |
pres | +0.850 |
Neg logit contributions | |
ibaba | -1.460 |
emale | -1.424 |
aimon | -1.117 |
ijing | -1.100 |
WRITE | -1.088 |
Token | bitcoin |
Token ID | 35395 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
meg | +0.817 |
star | +0.795 |
� | +0.767 |
mania | +0.758 |
pres | +0.756 |
Neg logit contributions | |
ibaba | -1.792 |
emale | -1.754 |
aimon | -1.412 |
ijing | -1.396 |
WRITE | -1.384 |
Token | elegant |
Token ID | 19992 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
meg | +0.075 |
star | +0.073 |
� | +0.071 |
power | +0.070 |
mania | +0.070 |
Neg logit contributions | |
ibaba | -0.108 |
emale | -0.106 |
aimon | -0.082 |
ijing | -0.081 |
WRITE | -0.080 |
Token | it |
Token ID | 340 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
meg | +0.114 |
star | +0.111 |
� | +0.107 |
pres | +0.106 |
power | +0.106 |
Neg logit contributions | |
ibaba | -0.206 |
emale | -0.201 |
aimon | -0.160 |
ijing | -0.158 |
WRITE | -0.156 |
Token | seems |
Token ID | 2331 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
meg | +1.281 |
star | +1.245 |
� | +1.203 |
power | +1.194 |
pres | +1.193 |
Neg logit contributions | |
ibaba | -2.280 |
emale | -2.226 |
aimon | -1.765 |
ijing | -1.740 |
WRITE | -1.723 |
Token | out |
Token ID | 503 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
meg | +0.883 |
star | +0.855 |
� | +0.827 |
mania | +0.823 |
power | +0.820 |
Neg logit contributions | |
ibaba | -1.591 |
emale | -1.551 |
aimon | -1.232 |
ijing | -1.220 |
WRITE | -1.206 |
Token | of |
Token ID | 286 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
ibaba | +0.681 |
emale | +0.668 |
aimon | +0.529 |
ijing | +0.523 |
WRITE | +0.517 |
Neg logit contributions | |
meg | -0.373 |
star | -0.361 |
� | -0.359 |
mania | -0.350 |
pres | -0.346 |
Token | place |
Token ID | 1295 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
meg | +2.132 |
star | +2.091 |
� | +2.019 |
pres | +2.017 |
power | +2.017 |
Neg logit contributions | |
ibaba | -2.741 |
emale | -2.676 |
aimon | -2.041 |
ijing | -2.002 |
WRITE | -1.975 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
meg | +2.901 |
star | +2.839 |
power | +2.692 |
pres | +2.689 |
� | +2.685 |
Neg logit contributions | |
ibaba | -6.221 |
emale | -6.113 |
aimon | -4.873 |
ijing | -4.796 |
WRITE | -4.734 |
Token | At |
Token ID | 1629 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
ibaba | +1.119 |
emale | +1.091 |
aimon | +0.864 |
ijing | +0.850 |
WRITE | +0.849 |
Neg logit contributions | |
meg | -0.654 |
star | -0.634 |
� | -0.615 |
power | -0.610 |
pres | -0.609 |
Token | the |
Token ID | 262 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
ibaba | +0.229 |
emale | +0.222 |
aimon | +0.174 |
ijing | +0.172 |
WRITE | +0.169 |
Neg logit contributions | |
meg | -0.150 |
star | -0.146 |
� | -0.143 |
power | -0.141 |
pres | -0.141 |
Token | menu |
Token ID | 6859 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
ibaba | +0.664 |
emale | +0.642 |
aimon | +0.492 |
ijing | +0.485 |
WRITE | +0.480 |
Neg logit contributions | |
meg | -0.535 |
star | -0.519 |
� | -0.514 |
mania | -0.507 |
power | -0.506 |
Token | � |
Token ID | 447 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
ibaba | +1.005 |
emale | +0.971 |
aimon | +0.771 |
ijing | +0.759 |
WRITE | +0.749 |
Neg logit contributions | |
meg | -0.606 |
star | -0.582 |
� | -0.578 |
mania | -0.567 |
pres | -0.565 |
Token | the |
Token ID | 262 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
ibaba | +1.948 |
emale | +1.887 |
aimon | +1.467 |
ijing | +1.441 |
WRITE | +1.438 |
Neg logit contributions | |
meg | -1.415 |
star | -1.379 |
� | -1.357 |
mania | -1.340 |
pres | -1.328 |
Token | book |
Token ID | 1492 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ibaba | +0.179 |
emale | +0.174 |
aimon | +0.134 |
ijing | +0.132 |
WRITE | +0.131 |
Neg logit contributions | |
meg | -0.132 |
star | -0.128 |
� | -0.126 |
mania | -0.124 |
pres | -0.124 |
Token | he |
Token ID | 339 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
meg | +0.621 |
star | +0.604 |
� | +0.587 |
mania | +0.580 |
pres | +0.579 |
Neg logit contributions | |
ibaba | -1.036 |
emale | -1.011 |
aimon | -0.798 |
ijing | -0.786 |
WRITE | -0.780 |
Token | learned |
Token ID | 4499 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
meg | +0.047 |
star | +0.045 |
� | +0.044 |
power | +0.043 |
pres | +0.043 |
Neg logit contributions | |
ibaba | -0.084 |
emale | -0.082 |
aimon | -0.065 |
ijing | -0.064 |
WRITE | -0.064 |
Token | that |
Token ID | 326 |
Feature activation | +7.71e-02 |
Pos logit contributions | |
ibaba | +0.376 |
emale | +0.363 |
aimon | +0.290 |
ijing | +0.284 |
WRITE | +0.283 |
Neg logit contributions | |
meg | -0.230 |
star | -0.222 |
� | -0.216 |
mania | -0.214 |
pres | -0.214 |
Token | Georg |
Token ID | 6850 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
meg | +0.221 |
star | +0.215 |
� | +0.210 |
mania | +0.207 |
pres | +0.206 |
Neg logit contributions | |
ibaba | -0.343 |
emale | -0.334 |
aimon | -0.262 |
ijing | -0.257 |
WRITE | -0.257 |
Token | es |
Token ID | 274 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
meg | +3.091 |
star | +2.997 |
pres | +2.874 |
mania | +2.869 |
power | +2.868 |
Neg logit contributions | |
ibaba | -6.049 |
emale | -5.918 |
aimon | -4.731 |
ijing | -4.664 |
WRITE | -4.622 |
Token | Mé |
Token ID | 43579 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
meg | +1.296 |
star | +1.262 |
� | +1.228 |
power | +1.214 |
mania | +1.210 |
Neg logit contributions | |
ibaba | -2.006 |
emale | -1.960 |
aimon | -1.530 |
ijing | -1.509 |
WRITE | -1.496 |
Token | li |
Token ID | 4528 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
ibaba | +1.383 |
emale | +1.349 |
aimon | +1.040 |
ijing | +1.026 |
WRITE | +1.013 |
Neg logit contributions | |
meg | -0.963 |
star | -0.932 |
� | -0.923 |
mania | -0.904 |
pres | -0.901 |
Token | è |
Token ID | 14064 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
meg | +1.107 |
star | +1.052 |
� | +1.040 |
mania | +1.030 |
rovers | +1.007 |
Neg logit contributions | |
ibaba | -2.206 |
emale | -2.167 |
ijing | -1.728 |
aimon | -1.719 |
WRITE | -1.683 |
Token | s |
Token ID | 82 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
meg | +0.247 |
star | +0.236 |
pres | +0.216 |
power | +0.214 |
mania | +0.213 |
Neg logit contributions | |
ibaba | -1.094 |
emale | -1.074 |
aimon | -0.901 |
ijing | -0.889 |
WRITE | -0.883 |
Token | responded |
Token ID | 7082 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
meg | +0.165 |
star | +0.159 |
� | +0.156 |
power | +0.153 |
pres | +0.153 |
Neg logit contributions | |
ibaba | -0.280 |
emale | -0.273 |
aimon | -0.216 |
ijing | -0.213 |
mustache | -0.212 |
Token | by |
Token ID | 416 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
meg | +0.808 |
star | +0.782 |
� | +0.758 |
mania | +0.751 |
pres | +0.746 |
Neg logit contributions | |
ibaba | -1.516 |
emale | -1.484 |
aimon | -1.188 |
ijing | -1.171 |
WRITE | -1.155 |
Token | signing |
Token ID | 8415 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
ibaba | +2.558 |
emale | +2.475 |
aimon | +1.895 |
mustache | +1.863 |
ijing | +1.851 |
Neg logit contributions | |
meg | -2.178 |
star | -2.124 |
� | -2.107 |
rovers | -2.076 |
mania | -2.062 |
Token | the |
Token ID | 262 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
meg | +1.137 |
star | +1.109 |
� | +1.078 |
mania | +1.069 |
power | +1.068 |
Neg logit contributions | |
ibaba | -1.626 |
emale | -1.585 |
aimon | -1.228 |
ijing | -1.208 |
WRITE | -1.199 |
Token | " |
Token ID | 366 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
meg | +0.772 |
star | +0.754 |
� | +0.731 |
power | +0.727 |
pres | +0.727 |
Neg logit contributions | |
ibaba | -1.101 |
emale | -1.072 |
aimon | -0.830 |
ijing | -0.816 |
WRITE | -0.807 |
Token | Direct |
Token ID | 13470 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
meg | +1.044 |
star | +1.012 |
� | +0.977 |
mania | +0.969 |
power | +0.967 |
Neg logit contributions | |
ibaba | -2.069 |
emale | -2.024 |
aimon | -1.620 |
ijing | -1.597 |
WRITE | -1.583 |
Token | ive |
Token ID | 425 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
meg | +3.719 |
star | +3.673 |
power | +3.582 |
pres | +3.549 |
sim | +3.482 |
Neg logit contributions | |
ibaba | -5.366 |
emale | -5.204 |
aimon | -4.093 |
ijing | -3.938 |
WRITE | -3.917 |
Token | of |
Token ID | 286 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
meg | +0.837 |
star | +0.812 |
� | +0.787 |
mania | +0.777 |
power | +0.775 |
Neg logit contributions | |
ibaba | -1.580 |
emale | -1.546 |
aimon | -1.233 |
ijing | -1.217 |
WRITE | -1.207 |
Token | Peoples |
Token ID | 41770 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
meg | +0.918 |
star | +0.896 |
� | +0.869 |
power | +0.864 |
pres | +0.863 |
Neg logit contributions | |
ibaba | -1.332 |
emale | -1.298 |
aimon | -1.006 |
ijing | -0.990 |
WRITE | -0.980 |
Token | ' |
Token ID | 6 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
ibaba | +1.056 |
emale | +1.027 |
WRITE | +0.823 |
aimon | +0.820 |
ijing | +0.816 |
Neg logit contributions | |
meg | -0.545 |
star | -0.523 |
� | -0.523 |
rovers | -0.504 |
mania | -0.501 |
Token | Comm |
Token ID | 1520 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
ibaba | +2.086 |
emale | +2.036 |
WRITE | +1.609 |
aimon | +1.596 |
ijing | +1.588 |
Neg logit contributions | |
meg | -1.213 |
� | -1.170 |
star | -1.159 |
rovers | -1.126 |
mania | -1.125 |
Token |
|
Token ID | 198 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
meg | +0.928 |
star | +0.901 |
power | +0.855 |
mania | +0.852 |
pres | +0.851 |
Neg logit contributions | |
ibaba | -2.374 |
emale | -2.336 |
aimon | -1.893 |
ijing | -1.874 |
WRITE | -1.846 |
Token | History |
Token ID | 18122 |
Feature activation | +9.46e-03 |
Pos logit contributions | |
meg | +1.195 |
star | +1.161 |
� | +1.117 |
pres | +1.106 |
mania | +1.106 |
Neg logit contributions | |
ibaba | -2.408 |
emale | -2.354 |
aimon | -1.888 |
ijing | -1.861 |
WRITE | -1.845 |
Token | [ |
Token ID | 685 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
meg | +0.026 |
star | +0.025 |
� | +0.025 |
pres | +0.024 |
mania | +0.024 |
Neg logit contributions | |
ibaba | -0.043 |
emale | -0.042 |
aimon | -0.033 |
ijing | -0.033 |
WRITE | -0.032 |
Token | edit |
Token ID | 4370 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
meg | +0.715 |
star | +0.699 |
power | +0.671 |
pres | +0.669 |
mania | +0.667 |
Neg logit contributions | |
ibaba | -1.285 |
emale | -1.259 |
aimon | -0.994 |
ijing | -0.980 |
WRITE | -0.971 |
Token | ] |
Token ID | 2361 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
meg | +0.053 |
star | +0.051 |
� | +0.050 |
power | +0.050 |
pres | +0.050 |
Neg logit contributions | |
ibaba | -0.075 |
emale | -0.073 |
aimon | -0.056 |
ijing | -0.055 |
WRITE | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
meg | +1.825 |
star | +1.801 |
power | +1.712 |
mania | +1.707 |
pres | +1.686 |
Neg logit contributions | |
ibaba | -5.400 |
emale | -5.366 |
aimon | -4.334 |
ijing | -4.311 |
mustache | -4.237 |
Token |
|
Token ID | 198 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
meg | +2.166 |
star | +2.146 |
power | +2.032 |
mania | +2.014 |
pres | +2.000 |
Neg logit contributions | |
emale | -6.897 |
ibaba | -6.878 |
ijing | -5.533 |
aimon | -5.522 |
ortunately | -5.499 |
Token | WE |
Token ID | 8845 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
meg | +2.763 |
star | +2.679 |
power | +2.580 |
pres | +2.572 |
� | +2.560 |
Neg logit contributions | |
ibaba | -6.061 |
emale | -5.920 |
aimon | -4.757 |
ijing | -4.695 |
mustache | -4.669 |
Token | P |
Token ID | 47 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
meg | +0.334 |
star | +0.320 |
pres | +0.305 |
power | +0.305 |
� | +0.304 |
Neg logit contributions | |
ibaba | -0.998 |
emale | -0.981 |
aimon | -0.804 |
ijing | -0.793 |
mustache | -0.785 |
Token | was |
Token ID | 373 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
meg | +0.300 |
star | +0.291 |
� | +0.282 |
pres | +0.277 |
power | +0.277 |
Neg logit contributions | |
ibaba | -0.589 |
emale | -0.575 |
aimon | -0.461 |
ijing | -0.455 |
WRITE | -0.451 |
Token | ratified |
Token ID | 39067 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
meg | +0.390 |
star | +0.378 |
� | +0.364 |
power | +0.360 |
pres | +0.359 |
Neg logit contributions | |
ibaba | -0.851 |
emale | -0.832 |
aimon | -0.672 |
ijing | -0.663 |
WRITE | -0.657 |
Token | matchup |
Token ID | 18406 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
meg | +0.666 |
star | +0.652 |
� | +0.632 |
power | +0.629 |
pres | +0.629 |
Neg logit contributions | |
ibaba | -0.851 |
emale | -0.829 |
aimon | -0.632 |
ijing | -0.621 |
WRITE | -0.613 |
Token | between |
Token ID | 1022 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
meg | +0.114 |
star | +0.111 |
� | +0.110 |
power | +0.107 |
rovers | +0.106 |
Neg logit contributions | |
ibaba | -0.162 |
emale | -0.157 |
aimon | -0.121 |
ijing | -0.121 |
WRITE | -0.120 |
Token | Donald |
Token ID | 3759 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ibaba | +1.266 |
emale | +1.235 |
aimon | +0.938 |
WRITE | +0.930 |
ijing | +0.928 |
Neg logit contributions | |
meg | -0.989 |
star | -0.960 |
� | -0.951 |
power | -0.935 |
rovers | -0.928 |
Token | Trump |
Token ID | 1301 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
meg | +0.588 |
star | +0.567 |
� | +0.566 |
pres | +0.554 |
power | +0.552 |
Neg logit contributions | |
ibaba | -0.651 |
emale | -0.636 |
aimon | -0.472 |
ijing | -0.468 |
WRITE | -0.464 |
Token | Donald |
Token ID | 3759 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ibaba | +0.097 |
emale | +0.095 |
ijing | +0.075 |
aimon | +0.074 |
mustache | +0.074 |
Neg logit contributions | |
meg | -0.056 |
star | -0.054 |
� | -0.053 |
power | -0.052 |
rovers | -0.052 |
Token | John |
Token ID | 1757 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
meg | +2.229 |
star | +2.209 |
mania | +2.138 |
power | +2.055 |
sim | +1.983 |
Neg logit contributions | |
ibaba | -5.276 |
emale | -5.116 |
ijing | -4.149 |
aimon | -4.148 |
irtual | -4.069 |
Token | Trump |
Token ID | 1301 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
meg | +3.216 |
power | +3.178 |
star | +3.171 |
pres | +3.090 |
rovers | +3.050 |
Neg logit contributions | |
emale | -5.665 |
ibaba | -5.652 |
aimon | -4.461 |
ijing | -4.346 |
irtual | -4.204 |
Token | House |
Token ID | 18102 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
meg | +1.853 |
star | +1.815 |
power | +1.736 |
pres | +1.726 |
� | +1.708 |
Neg logit contributions | |
ibaba | -3.842 |
emale | -3.771 |
aimon | -3.032 |
ijing | -2.982 |
WRITE | -2.948 |
Token | committee |
Token ID | 5583 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
meg | +1.520 |
star | +1.508 |
power | +1.469 |
mania | +1.434 |
pres | +1.409 |
Neg logit contributions | |
ibaba | -2.510 |
emale | -2.421 |
aimon | -1.953 |
ijing | -1.920 |
irtual | -1.841 |
Token | believes |
Token ID | 5804 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
meg | +0.175 |
star | +0.167 |
� | +0.158 |
mania | +0.156 |
pres | +0.154 |
Neg logit contributions | |
ibaba | -0.706 |
emale | -0.693 |
aimon | -0.578 |
ijing | -0.572 |
WRITE | -0.569 |
Token | it |
Token ID | 340 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
ibaba | +1.128 |
emale | +1.096 |
aimon | +0.835 |
ijing | +0.820 |
WRITE | +0.802 |
Neg logit contributions | |
meg | -0.907 |
star | -0.865 |
pres | -0.838 |
rovers | -0.832 |
power | -0.832 |
Token | that |
Token ID | 326 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
meg | +0.784 |
star | +0.762 |
� | +0.738 |
pres | +0.732 |
rovers | +0.730 |
Neg logit contributions | |
ibaba | -1.309 |
emale | -1.273 |
aimon | -1.004 |
ijing | -0.995 |
WRITE | -0.989 |
Token | make |
Token ID | 787 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
meg | +0.893 |
star | +0.872 |
� | +0.842 |
power | +0.839 |
pres | +0.838 |
Neg logit contributions | |
ibaba | -1.386 |
emale | -1.351 |
aimon | -1.056 |
ijing | -1.038 |
WRITE | -1.026 |
Token | US |
Token ID | 1294 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
meg | +1.078 |
star | +1.052 |
� | +1.020 |
pres | +1.012 |
power | +1.012 |
Neg logit contributions | |
ibaba | -1.606 |
emale | -1.566 |
aimon | -1.217 |
ijing | -1.200 |
WRITE | -1.189 |
Token | generals |
Token ID | 26770 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
meg | +0.445 |
star | +0.435 |
� | +0.423 |
mania | +0.418 |
power | +0.417 |
Neg logit contributions | |
ibaba | -0.620 |
emale | -0.605 |
aimon | -0.465 |
ijing | -0.459 |
WRITE | -0.455 |
Token | sit |
Token ID | 1650 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
meg | +0.479 |
star | +0.463 |
� | +0.453 |
pres | +0.447 |
power | +0.446 |
Neg logit contributions | |
ibaba | -0.764 |
emale | -0.743 |
aimon | -0.582 |
ijing | -0.579 |
mustache | -0.575 |
Token | up |
Token ID | 510 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
meg | +0.590 |
star | +0.570 |
� | +0.559 |
pres | +0.549 |
power | +0.547 |
Neg logit contributions | |
ibaba | -0.992 |
emale | -0.967 |
aimon | -0.761 |
ijing | -0.754 |
WRITE | -0.744 |
Token | and |
Token ID | 290 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
meg | +2.769 |
star | +2.697 |
power | +2.619 |
mania | +2.575 |
� | +2.572 |
Neg logit contributions | |
ibaba | -5.982 |
emale | -5.896 |
aimon | -4.736 |
ijing | -4.666 |
WRITE | -4.574 |
Token | listen |
Token ID | 6004 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
meg | +0.639 |
star | +0.622 |
� | +0.603 |
mania | +0.597 |
pres | +0.597 |
Neg logit contributions | |
ibaba | -1.060 |
emale | -1.034 |
aimon | -0.813 |
ijing | -0.801 |
WRITE | -0.799 |
Token | . |
Token ID | 13 |
Feature activation | +7.51e-02 |
Pos logit contributions | |
meg | +0.768 |
star | +0.742 |
� | +0.711 |
mania | +0.706 |
pres | +0.703 |
Neg logit contributions | |
ibaba | -1.652 |
emale | -1.612 |
aimon | -1.304 |
ijing | -1.293 |
WRITE | -1.283 |
Token | The |
Token ID | 383 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
meg | +0.200 |
star | +0.194 |
� | +0.188 |
power | +0.186 |
pres | +0.186 |
Neg logit contributions | |
ibaba | -0.350 |
emale | -0.342 |
aimon | -0.270 |
ijing | -0.266 |
WRITE | -0.265 |
Token | stakes |
Token ID | 21147 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
meg | +0.616 |
star | +0.602 |
� | +0.583 |
power | +0.580 |
pres | +0.580 |
Neg logit contributions | |
ibaba | -0.863 |
emale | -0.841 |
aimon | -0.649 |
ijing | -0.638 |
WRITE | -0.631 |
Token | History |
Token ID | 18122 |
Feature activation | +9.46e-03 |
Pos logit contributions | |
meg | +1.195 |
star | +1.161 |
� | +1.117 |
pres | +1.106 |
mania | +1.106 |
Neg logit contributions | |
ibaba | -2.408 |
emale | -2.354 |
aimon | -1.888 |
ijing | -1.861 |
WRITE | -1.845 |
Token | [ |
Token ID | 685 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
meg | +0.026 |
star | +0.025 |
� | +0.025 |
pres | +0.024 |
mania | +0.024 |
Neg logit contributions | |
ibaba | -0.043 |
emale | -0.042 |
aimon | -0.033 |
ijing | -0.033 |
WRITE | -0.032 |
Token | edit |
Token ID | 4370 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
meg | +0.715 |
star | +0.699 |
power | +0.671 |
pres | +0.669 |
mania | +0.667 |
Neg logit contributions | |
ibaba | -1.285 |
emale | -1.259 |
aimon | -0.994 |
ijing | -0.980 |
WRITE | -0.971 |
Token | ] |
Token ID | 2361 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
meg | +0.053 |
star | +0.051 |
� | +0.050 |
power | +0.050 |
pres | +0.050 |
Neg logit contributions | |
ibaba | -0.075 |
emale | -0.073 |
aimon | -0.056 |
ijing | -0.055 |
WRITE | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
meg | +1.825 |
star | +1.801 |
power | +1.712 |
mania | +1.707 |
pres | +1.686 |
Neg logit contributions | |
ibaba | -5.400 |
emale | -5.366 |
aimon | -4.334 |
ijing | -4.311 |
mustache | -4.237 |
Token |
|
Token ID | 198 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
meg | +2.166 |
star | +2.146 |
power | +2.032 |
mania | +2.014 |
pres | +2.000 |
Neg logit contributions | |
emale | -6.897 |
ibaba | -6.878 |
ijing | -5.533 |
aimon | -5.522 |
ortunately | -5.499 |
Token | WE |
Token ID | 8845 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
meg | +2.763 |
star | +2.679 |
power | +2.580 |
pres | +2.572 |
� | +2.560 |
Neg logit contributions | |
ibaba | -6.061 |
emale | -5.920 |
aimon | -4.757 |
ijing | -4.695 |
mustache | -4.669 |
Token | P |
Token ID | 47 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
meg | +0.334 |
star | +0.320 |
pres | +0.305 |
power | +0.305 |
� | +0.304 |
Neg logit contributions | |
ibaba | -0.998 |
emale | -0.981 |
aimon | -0.804 |
ijing | -0.793 |
mustache | -0.785 |
Token | was |
Token ID | 373 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
meg | +0.300 |
star | +0.291 |
� | +0.282 |
pres | +0.277 |
power | +0.277 |
Neg logit contributions | |
ibaba | -0.589 |
emale | -0.575 |
aimon | -0.461 |
ijing | -0.455 |
WRITE | -0.451 |
Token | ratified |
Token ID | 39067 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
meg | +0.390 |
star | +0.378 |
� | +0.364 |
power | +0.360 |
pres | +0.359 |
Neg logit contributions | |
ibaba | -0.851 |
emale | -0.832 |
aimon | -0.672 |
ijing | -0.663 |
WRITE | -0.657 |
Token | as |
Token ID | 355 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
meg | +0.261 |
star | +0.253 |
� | +0.245 |
power | +0.241 |
pres | +0.241 |
Neg logit contributions | |
ibaba | -0.484 |
emale | -0.475 |
aimon | -0.379 |
ijing | -0.374 |
WRITE | -0.371 |
Token | Republican |
Token ID | 3415 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
meg | +0.326 |
star | +0.319 |
� | +0.312 |
power | +0.309 |
mania | +0.308 |
Neg logit contributions | |
ibaba | -0.440 |
emale | -0.429 |
aimon | -0.329 |
ijing | -0.325 |
WRITE | -0.321 |
Token | nominee |
Token ID | 10429 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
meg | +0.164 |
star | +0.159 |
� | +0.158 |
power | +0.154 |
pres | +0.154 |
Neg logit contributions | |
ibaba | -0.224 |
emale | -0.219 |
aimon | -0.168 |
ijing | -0.167 |
WRITE | -0.167 |
Token | Donald |
Token ID | 3759 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
ibaba | +0.309 |
emale | +0.301 |
aimon | +0.229 |
WRITE | +0.228 |
ijing | +0.228 |
Neg logit contributions | |
meg | -0.247 |
� | -0.240 |
star | -0.239 |
rovers | -0.232 |
power | -0.232 |
Token | Trump |
Token ID | 1301 |
Feature activation | -3.38e-04 |
Pos logit contributions | |
ibaba | +0.135 |
emale | +0.131 |
aimon | +0.100 |
ijing | +0.098 |
WRITE | +0.097 |
Neg logit contributions | |
meg | -0.108 |
star | -0.106 |
� | -0.103 |
power | -0.102 |
pres | -0.102 |
Token | Donald |
Token ID | 3759 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ibaba | +0.001 |
emale | +0.001 |
aimon | +0.001 |
ijing | +0.001 |
WRITE | +0.001 |
Neg logit contributions | |
meg | -0.001 |
star | -0.001 |
� | -0.001 |
power | -0.001 |
pres | -0.001 |
Token | John |
Token ID | 1757 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
star | +2.548 |
meg | +2.536 |
mania | +2.431 |
power | +2.359 |
sim | +2.263 |
Neg logit contributions | |
ibaba | -6.152 |
emale | -5.954 |
aimon | -4.837 |
ijing | -4.832 |
irtual | -4.755 |
Token | Trump |
Token ID | 1301 |
Feature activation | +9.80e-01 |
Pos logit contributions | |
meg | +3.213 |
power | +3.188 |
star | +3.168 |
pres | +3.096 |
rovers | +3.056 |
Neg logit contributions | |
emale | -5.561 |
ibaba | -5.522 |
aimon | -4.322 |
ijing | -4.242 |
irtual | -4.104 |
Token | House |
Token ID | 18102 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
meg | +2.330 |
star | +2.286 |
power | +2.192 |
pres | +2.173 |
� | +2.146 |
Neg logit contributions | |
ibaba | -4.858 |
emale | -4.762 |
aimon | -3.827 |
ijing | -3.766 |
WRITE | -3.728 |
Token | committee |
Token ID | 5583 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
meg | +1.555 |
star | +1.549 |
power | +1.514 |
mania | +1.465 |
pres | +1.450 |
Neg logit contributions | |
ibaba | -2.574 |
emale | -2.487 |
aimon | -2.002 |
ijing | -1.964 |
irtual | -1.888 |
Token | believes |
Token ID | 5804 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
meg | +0.122 |
star | +0.116 |
� | +0.110 |
mania | +0.108 |
pres | +0.107 |
Neg logit contributions | |
ibaba | -0.491 |
emale | -0.483 |
aimon | -0.402 |
ijing | -0.398 |
WRITE | -0.396 |
Token | it |
Token ID | 340 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
ibaba | +1.441 |
emale | +1.406 |
aimon | +1.064 |
ijing | +1.042 |
WRITE | +1.018 |
Neg logit contributions | |
meg | -1.217 |
star | -1.168 |
pres | -1.134 |
� | -1.134 |
power | -1.127 |
Token | us |
Token ID | 385 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
meg | +0.613 |
star | +0.587 |
power | +0.546 |
pres | +0.545 |
� | +0.544 |
Neg logit contributions | |
ibaba | -2.226 |
emale | -2.189 |
aimon | -1.814 |
ijing | -1.794 |
WRITE | -1.784 |
Token | / |
Token ID | 14 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
ibaba | +0.153 |
emale | +0.150 |
aimon | +0.119 |
ijing | +0.118 |
WRITE | +0.117 |
Neg logit contributions | |
meg | -0.077 |
star | -0.075 |
� | -0.073 |
pres | -0.072 |
mania | -0.072 |
Token | wiki |
Token ID | 15466 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
ibaba | +1.494 |
emale | +1.452 |
aimon | +1.149 |
ijing | +1.132 |
WRITE | +1.123 |
Neg logit contributions | |
meg | -0.893 |
star | -0.871 |
� | -0.855 |
mania | -0.843 |
rovers | -0.840 |
Token | / |
Token ID | 14 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
ibaba | +1.320 |
emale | +1.285 |
aimon | +1.037 |
ijing | +1.021 |
WRITE | +1.016 |
Neg logit contributions | |
meg | -0.642 |
star | -0.620 |
� | -0.605 |
mania | -0.593 |
rovers | -0.592 |
Token | trans |
Token ID | 7645 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
meg | +1.346 |
star | +1.308 |
� | +1.262 |
mania | +1.251 |
pres | +1.250 |
Neg logit contributions | |
ibaba | -2.589 |
emale | -2.530 |
aimon | -2.020 |
ijing | -1.991 |
WRITE | -1.973 |
Token | cript |
Token ID | 6519 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
meg | +1.328 |
star | +1.311 |
pres | +1.261 |
power | +1.253 |
sim | +1.235 |
Neg logit contributions | |
ibaba | -2.548 |
emale | -2.515 |
aimon | -1.990 |
WRITE | -1.964 |
ijing | -1.960 |
Token | s |
Token ID | 82 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
meg | +1.234 |
star | +1.184 |
pres | +1.082 |
power | +1.062 |
mania | +1.035 |
Neg logit contributions | |
ibaba | -7.514 |
emale | -7.499 |
aimon | -6.294 |
ijing | -6.237 |
mustache | -6.185 |
Token | / |
Token ID | 14 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
meg | +0.543 |
star | +0.528 |
� | +0.510 |
power | +0.505 |
mania | +0.504 |
Neg logit contributions | |
ibaba | -1.077 |
emale | -1.054 |
aimon | -0.843 |
ijing | -0.833 |
WRITE | -0.828 |
Token | sc |
Token ID | 1416 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
meg | +0.921 |
star | +0.889 |
� | +0.862 |
mania | +0.852 |
pres | +0.843 |
Neg logit contributions | |
ibaba | -2.230 |
emale | -2.184 |
aimon | -1.776 |
ijing | -1.751 |
WRITE | -1.736 |
Token | aling |
Token ID | 4272 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
meg | +0.933 |
star | +0.897 |
power | +0.879 |
pres | +0.869 |
mania | +0.868 |
Neg logit contributions | |
ibaba | -1.428 |
emale | -1.401 |
aimon | -1.084 |
mustache | -1.076 |
WRITE | -1.071 |
Token | bitcoin |
Token ID | 35395 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
meg | +1.145 |
star | +1.125 |
� | +1.108 |
pres | +1.090 |
mania | +1.088 |
Neg logit contributions | |
ibaba | -1.341 |
emale | -1.299 |
aimon | -0.976 |
ijing | -0.965 |
WRITE | -0.937 |
Token | 158 |
Token ID | 21273 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
meg | +0.171 |
star | +0.163 |
� | +0.159 |
mania | +0.159 |
power | +0.158 |
Neg logit contributions | |
ibaba | -0.296 |
emale | -0.293 |
aimon | -0.229 |
ijing | -0.227 |
WRITE | -0.225 |
Token | ] |
Token ID | 60 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
meg | +1.266 |
star | +1.234 |
� | +1.194 |
mania | +1.186 |
pres | +1.185 |
Neg logit contributions | |
ibaba | -1.919 |
emale | -1.885 |
aimon | -1.456 |
ijing | -1.439 |
WRITE | -1.426 |
Token |
|
Token ID | 198 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
meg | +1.426 |
star | +1.390 |
� | +1.334 |
mania | +1.329 |
power | +1.329 |
Neg logit contributions | |
ibaba | -2.697 |
emale | -2.634 |
aimon | -2.099 |
ijing | -2.070 |
WRITE | -2.045 |
Token |
|
Token ID | 198 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
meg | +1.205 |
star | +1.177 |
power | +1.115 |
mania | +1.109 |
pres | +1.107 |
Neg logit contributions | |
ibaba | -3.261 |
emale | -3.223 |
aimon | -2.603 |
ijing | -2.587 |
mustache | -2.537 |
Token | See |
Token ID | 6214 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
meg | +1.476 |
star | +1.430 |
� | +1.374 |
power | +1.361 |
pres | +1.361 |
Neg logit contributions | |
ibaba | -3.234 |
emale | -3.164 |
aimon | -2.553 |
ijing | -2.518 |
WRITE | -2.497 |
Token | also |
Token ID | 635 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
meg | +1.910 |
star | +1.867 |
power | +1.771 |
mania | +1.768 |
� | +1.759 |
Neg logit contributions | |
ibaba | -4.155 |
emale | -4.085 |
aimon | -3.275 |
ijing | -3.239 |
WRITE | -3.194 |
Token | [ |
Token ID | 685 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
meg | +3.221 |
star | +3.150 |
mania | +3.027 |
� | +3.022 |
power | +3.013 |
Neg logit contributions | |
ibaba | -5.502 |
emale | -5.392 |
aimon | -4.263 |
ijing | -4.186 |
WRITE | -4.132 |
Token | edit |
Token ID | 4370 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
meg | +0.875 |
star | +0.860 |
power | +0.821 |
pres | +0.820 |
mania | +0.817 |
Neg logit contributions | |
ibaba | -1.601 |
emale | -1.572 |
aimon | -1.240 |
ijing | -1.223 |
WRITE | -1.209 |
Token | ] |
Token ID | 2361 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
meg | +0.273 |
star | +0.266 |
� | +0.259 |
power | +0.256 |
pres | +0.256 |
Neg logit contributions | |
ibaba | -0.382 |
emale | -0.372 |
aimon | -0.288 |
ijing | -0.283 |
WRITE | -0.280 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
meg | +2.338 |
star | +2.270 |
� | +2.176 |
power | +2.162 |
mania | +2.160 |
Neg logit contributions | |
ibaba | -5.024 |
emale | -4.915 |
aimon | -3.960 |
ijing | -3.905 |
WRITE | -3.870 |
Token | I |
Token ID | 40 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
meg | +0.506 |
star | +0.491 |
� | +0.480 |
mania | +0.473 |
pres | +0.471 |
Neg logit contributions | |
ibaba | -0.976 |
emale | -0.951 |
aimon | -0.763 |
ijing | -0.752 |
WRITE | -0.744 |
Token | on |
Token ID | 319 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
meg | +0.812 |
star | +0.793 |
� | +0.762 |
mania | +0.761 |
power | +0.760 |
Neg logit contributions | |
ibaba | -1.294 |
emale | -1.263 |
aimon | -0.988 |
ijing | -0.974 |
WRITE | -0.963 |
Token | a |
Token ID | 257 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
meg | +0.394 |
star | +0.385 |
� | +0.373 |
power | +0.372 |
pres | +0.371 |
Neg logit contributions | |
ibaba | -0.522 |
emale | -0.508 |
aimon | -0.389 |
ijing | -0.383 |
WRITE | -0.379 |
Token | media |
Token ID | 2056 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
meg | +0.148 |
star | +0.144 |
� | +0.141 |
mania | +0.140 |
pres | +0.140 |
Neg logit contributions | |
ibaba | -0.175 |
emale | -0.170 |
aimon | -0.128 |
ijing | -0.126 |
WRITE | -0.125 |
Token | bus |
Token ID | 1323 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ibaba | +0.103 |
emale | +0.100 |
aimon | +0.075 |
ijing | +0.074 |
WRITE | +0.073 |
Neg logit contributions | |
meg | -0.087 |
star | -0.085 |
� | -0.083 |
power | -0.082 |
mania | -0.082 |
Token | ) |
Token ID | 8 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
meg | +0.409 |
star | +0.399 |
� | +0.386 |
power | +0.384 |
pres | +0.384 |
Neg logit contributions | |
ibaba | -0.601 |
emale | -0.586 |
aimon | -0.455 |
ijing | -0.448 |
WRITE | -0.443 |
Token | pic |
Token ID | 8301 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
meg | +0.623 |
star | +0.603 |
� | +0.584 |
power | +0.576 |
mania | +0.576 |
Neg logit contributions | |
ibaba | -1.245 |
emale | -1.218 |
aimon | -0.975 |
ijing | -0.962 |
WRITE | -0.956 |
Token | . |
Token ID | 13 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
meg | +1.711 |
star | +1.704 |
mania | +1.538 |
power | +1.536 |
pres | +1.529 |
Neg logit contributions | |
ibaba | -7.112 |
emale | -7.075 |
aimon | -5.780 |
ijing | -5.776 |
mustache | -5.660 |
Token | twitter |
Token ID | 6956 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
meg | +0.689 |
star | +0.664 |
� | +0.630 |
power | +0.625 |
pres | +0.624 |
Neg logit contributions | |
ibaba | -1.988 |
emale | -1.949 |
aimon | -1.600 |
ijing | -1.582 |
WRITE | -1.568 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
meg | +0.736 |
star | +0.711 |
power | +0.662 |
mania | +0.660 |
� | +0.660 |
Neg logit contributions | |
ibaba | -2.542 |
emale | -2.502 |
aimon | -2.064 |
ijing | -2.046 |
WRITE | -2.019 |
Token | com |
Token ID | 785 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
meg | +0.356 |
star | +0.344 |
� | +0.327 |
power | +0.325 |
pres | +0.325 |
Neg logit contributions | |
ibaba | -0.918 |
emale | -0.901 |
aimon | -0.733 |
ijing | -0.725 |
WRITE | -0.718 |
Token | / |
Token ID | 14 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
ibaba | +1.509 |
emale | +1.469 |
aimon | +1.186 |
ijing | +1.167 |
WRITE | +1.165 |
Neg logit contributions | |
meg | -0.731 |
star | -0.705 |
� | -0.683 |
pres | -0.674 |
mania | -0.674 |
Token | with |
Token ID | 351 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
ibaba | +0.507 |
emale | +0.493 |
aimon | +0.389 |
ijing | +0.387 |
WRITE | +0.382 |
Neg logit contributions | |
meg | -0.307 |
star | -0.297 |
� | -0.293 |
power | -0.287 |
pres | -0.285 |
Token | Republican |
Token ID | 3415 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
meg | +0.326 |
star | +0.319 |
� | +0.312 |
power | +0.309 |
mania | +0.308 |
Neg logit contributions | |
ibaba | -0.440 |
emale | -0.429 |
aimon | -0.329 |
ijing | -0.325 |
WRITE | -0.321 |
Token | nominee |
Token ID | 10429 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
meg | +0.164 |
star | +0.159 |
� | +0.158 |
power | +0.154 |
pres | +0.154 |
Neg logit contributions | |
ibaba | -0.224 |
emale | -0.219 |
aimon | -0.168 |
ijing | -0.167 |
WRITE | -0.167 |
Token | Donald |
Token ID | 3759 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
ibaba | +0.309 |
emale | +0.301 |
aimon | +0.229 |
WRITE | +0.228 |
ijing | +0.228 |
Neg logit contributions | |
meg | -0.247 |
� | -0.240 |
star | -0.239 |
rovers | -0.232 |
power | -0.232 |
Token | Trump |
Token ID | 1301 |
Feature activation | -3.38e-04 |
Pos logit contributions | |
ibaba | +0.135 |
emale | +0.131 |
aimon | +0.100 |
ijing | +0.098 |
WRITE | +0.097 |
Neg logit contributions | |
meg | -0.108 |
star | -0.106 |
� | -0.103 |
power | -0.102 |
pres | -0.102 |
Token | Donald |
Token ID | 3759 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ibaba | +0.001 |
emale | +0.001 |
aimon | +0.001 |
ijing | +0.001 |
WRITE | +0.001 |
Neg logit contributions | |
meg | -0.001 |
star | -0.001 |
� | -0.001 |
power | -0.001 |
pres | -0.001 |
Token | John |
Token ID | 1757 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
star | +2.548 |
meg | +2.536 |
mania | +2.431 |
power | +2.359 |
sim | +2.263 |
Neg logit contributions | |
ibaba | -6.152 |
emale | -5.954 |
aimon | -4.837 |
ijing | -4.832 |
irtual | -4.755 |
Token | Trump |
Token ID | 1301 |
Feature activation | +9.80e-01 |
Pos logit contributions | |
meg | +3.213 |
power | +3.188 |
star | +3.168 |
pres | +3.096 |
rovers | +3.056 |
Neg logit contributions | |
emale | -5.561 |
ibaba | -5.522 |
aimon | -4.322 |
ijing | -4.242 |
irtual | -4.104 |
Token | House |
Token ID | 18102 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
meg | +2.330 |
star | +2.286 |
power | +2.192 |
pres | +2.173 |
� | +2.146 |
Neg logit contributions | |
ibaba | -4.858 |
emale | -4.762 |
aimon | -3.827 |
ijing | -3.766 |
WRITE | -3.728 |
Token | committee |
Token ID | 5583 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
meg | +1.555 |
star | +1.549 |
power | +1.514 |
mania | +1.465 |
pres | +1.450 |
Neg logit contributions | |
ibaba | -2.574 |
emale | -2.487 |
aimon | -2.002 |
ijing | -1.964 |
irtual | -1.888 |
Token | believes |
Token ID | 5804 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
meg | +0.122 |
star | +0.116 |
� | +0.110 |
mania | +0.108 |
pres | +0.107 |
Neg logit contributions | |
ibaba | -0.491 |
emale | -0.483 |
aimon | -0.402 |
ijing | -0.398 |
WRITE | -0.396 |
Token | prison |
Token ID | 3770 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
meg | +0.786 |
star | +0.774 |
pres | +0.747 |
power | +0.746 |
mania | +0.745 |
Neg logit contributions | |
ibaba | -0.933 |
emale | -0.913 |
aimon | -0.688 |
ijing | -0.668 |
WRITE | -0.664 |
Token | MORE |
Token ID | 12011 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
meg | +1.562 |
star | +1.519 |
power | +1.477 |
pres | +1.476 |
� | +1.469 |
Neg logit contributions | |
ibaba | -2.171 |
emale | -2.124 |
aimon | -1.641 |
ijing | -1.610 |
mustache | -1.572 |
Token | and |
Token ID | 290 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
meg | +0.877 |
star | +0.844 |
� | +0.821 |
pres | +0.806 |
power | +0.805 |
Neg logit contributions | |
ibaba | -1.633 |
emale | -1.593 |
aimon | -1.261 |
ijing | -1.255 |
WRITE | -1.244 |
Token | Hillary |
Token ID | 5073 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
meg | +1.174 |
star | +1.151 |
� | +1.117 |
pres | +1.113 |
power | +1.112 |
Neg logit contributions | |
ibaba | -1.379 |
emale | -1.341 |
aimon | -1.010 |
ijing | -0.990 |
WRITE | -0.977 |
Token | Clinton |
Token ID | 2605 |
Feature activation | +1.88e-03 |
Pos logit contributions | |
meg | +1.442 |
star | +1.424 |
mania | +1.385 |
pres | +1.384 |
power | +1.381 |
Neg logit contributions | |
ibaba | -1.425 |
emale | -1.392 |
aimon | -1.017 |
ijing | -0.987 |
WRITE | -0.969 |
Token | Hillary |
Token ID | 5073 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
meg | +0.007 |
star | +0.007 |
� | +0.007 |
power | +0.007 |
pres | +0.007 |
Neg logit contributions | |
ibaba | -0.007 |
emale | -0.006 |
aimon | -0.005 |
WRITE | -0.005 |
ijing | -0.005 |
Token | Diane |
Token ID | 26542 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
star | +2.565 |
meg | +2.554 |
mania | +2.390 |
power | +2.367 |
pres | +2.347 |
Neg logit contributions | |
ibaba | -6.088 |
emale | -5.869 |
ijing | -4.812 |
aimon | -4.739 |
mustache | -4.700 |
Token | Rodham |
Token ID | 41848 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
star | +2.662 |
meg | +2.654 |
pres | +2.579 |
mania | +2.555 |
� | +2.547 |
Neg logit contributions | |
ibaba | -3.473 |
emale | -3.375 |
ijing | -2.544 |
aimon | -2.492 |
mustache | -2.468 |
Token | Clinton |
Token ID | 2605 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
meg | +2.797 |
star | +2.780 |
power | +2.704 |
pres | +2.691 |
mania | +2.677 |
Neg logit contributions | |
ibaba | -3.020 |
emale | -3.012 |
aimon | -2.176 |
ijing | -2.176 |
mustache | -2.106 |
Token | Sanders |
Token ID | 26747 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
meg | +1.548 |
star | +1.503 |
� | +1.431 |
pres | +1.426 |
power | +1.425 |
Neg logit contributions | |
ibaba | -3.646 |
emale | -3.569 |
aimon | -2.895 |
ijing | -2.856 |
WRITE | -2.831 |
Token | : |
Token ID | 25 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
meg | +1.234 |
star | +1.201 |
mania | +1.162 |
power | +1.149 |
� | +1.142 |
Neg logit contributions | |
ibaba | -2.482 |
emale | -2.431 |
aimon | -1.937 |
ijing | -1.913 |
WRITE | -1.898 |
Token | conscience |
Token ID | 18346 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
meg | +0.460 |
star | +0.449 |
� | +0.439 |
mania | +0.433 |
power | +0.433 |
Neg logit contributions | |
ibaba | -0.611 |
emale | -0.591 |
aimon | -0.457 |
ijing | -0.447 |
WRITE | -0.445 |
Token | or |
Token ID | 393 |
Feature activation | -6.89e-02 |
Pos logit contributions | |
ibaba | +0.224 |
emale | +0.216 |
aimon | +0.177 |
mustache | +0.175 |
ijing | +0.175 |
Neg logit contributions | |
meg | -0.111 |
star | -0.107 |
� | -0.106 |
rovers | -0.103 |
mania | -0.102 |
Token | acknowledgment |
Token ID | 48182 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
ibaba | +0.285 |
emale | +0.274 |
aimon | +0.213 |
ijing | +0.208 |
mustache | +0.207 |
Neg logit contributions | |
meg | -0.219 |
star | -0.213 |
� | -0.210 |
rovers | -0.206 |
mania | -0.206 |
Token | of |
Token ID | 286 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ibaba | +0.992 |
emale | +0.949 |
aimon | +0.767 |
ijing | +0.759 |
WRITE | +0.754 |
Neg logit contributions | |
meg | -0.585 |
star | -0.565 |
� | -0.552 |
rovers | -0.541 |
mania | -0.540 |
Token | mis |
Token ID | 2984 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
meg | +0.481 |
star | +0.469 |
� | +0.458 |
mania | +0.452 |
pres | +0.451 |
Neg logit contributions | |
ibaba | -0.699 |
emale | -0.678 |
aimon | -0.530 |
ijing | -0.520 |
WRITE | -0.516 |
Token | d |
Token ID | 67 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
meg | +1.235 |
star | +1.202 |
pres | +1.154 |
� | +1.153 |
power | +1.153 |
Neg logit contributions | |
ibaba | -2.279 |
emale | -2.227 |
aimon | -1.768 |
ijing | -1.746 |
WRITE | -1.730 |
Token | eeds |
Token ID | 39642 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
ibaba | +3.956 |
emale | +3.780 |
aimon | +2.614 |
ijing | +2.216 |
WRITE | +2.122 |
Neg logit contributions | |
meg | -6.644 |
� | -6.606 |
star | -6.574 |
pres | -6.406 |
mania | -6.373 |
Token | that |
Token ID | 326 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
ibaba | +1.927 |
emale | +1.852 |
aimon | +1.493 |
ijing | +1.492 |
mustache | +1.480 |
Neg logit contributions | |
meg | -1.065 |
star | -1.028 |
� | -1.017 |
rovers | -0.996 |
mania | -0.979 |
Token | destroyed |
Token ID | 6572 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
meg | +0.050 |
star | +0.049 |
� | +0.047 |
mania | +0.047 |
pres | +0.046 |
Neg logit contributions | |
ibaba | -0.090 |
emale | -0.088 |
aimon | -0.070 |
ijing | -0.069 |
WRITE | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
ibaba | +0.916 |
emale | +0.888 |
aimon | +0.686 |
ijing | +0.675 |
WRITE | +0.675 |
Neg logit contributions | |
meg | -0.653 |
star | -0.637 |
� | -0.625 |
mania | -0.616 |
pres | -0.615 |
Token | financial |
Token ID | 3176 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
ibaba | +0.412 |
emale | +0.401 |
aimon | +0.303 |
mustache | +0.299 |
ijing | +0.299 |
Neg logit contributions | |
meg | -0.330 |
star | -0.321 |
� | -0.315 |
mania | -0.312 |
pres | -0.311 |
Token | , |
Token ID | 11 |
Feature activation | +8.74e-02 |
Pos logit contributions | |
ibaba | +0.170 |
emale | +0.165 |
aimon | +0.132 |
ijing | +0.130 |
WRITE | +0.129 |
Neg logit contributions | |
meg | -0.090 |
star | -0.087 |
� | -0.085 |
mania | -0.084 |
pres | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
meg | +0.237 |
star | +0.230 |
� | +0.224 |
mania | +0.221 |
power | +0.221 |
Neg logit contributions | |
ibaba | -0.403 |
emale | -0.391 |
aimon | -0.310 |
ijing | -0.305 |
WRITE | -0.303 |
Token | he |
Token ID | 339 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
meg | +0.342 |
star | +0.332 |
� | +0.321 |
power | +0.318 |
pres | +0.318 |
Neg logit contributions | |
ibaba | -0.615 |
emale | -0.599 |
aimon | -0.477 |
ijing | -0.469 |
WRITE | -0.466 |
Token | pointed |
Token ID | 6235 |
Feature activation | +8.50e-02 |
Pos logit contributions | |
ibaba | +0.297 |
emale | +0.288 |
aimon | +0.230 |
ijing | +0.227 |
WRITE | +0.226 |
Neg logit contributions | |
meg | -0.164 |
star | -0.158 |
� | -0.154 |
power | -0.152 |
pres | -0.152 |
Token | his |
Token ID | 465 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
meg | +0.233 |
star | +0.226 |
� | +0.219 |
mania | +0.218 |
power | +0.217 |
Neg logit contributions | |
ibaba | -0.387 |
emale | -0.377 |
aimon | -0.298 |
ijing | -0.294 |
WRITE | -0.292 |
Token | finger |
Token ID | 7660 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ibaba | +1.919 |
emale | +1.873 |
aimon | +1.397 |
mustache | +1.382 |
ijing | +1.355 |
Neg logit contributions | |
meg | -1.843 |
star | -1.802 |
� | -1.762 |
mania | -1.761 |
rovers | -1.754 |
Token | at |
Token ID | 379 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
ibaba | +6.068 |
emale | +5.851 |
mustache | +4.730 |
aimon | +4.712 |
ijing | +4.704 |
Neg logit contributions | |
meg | -3.869 |
star | -3.771 |
rovers | -3.713 |
� | -3.707 |
mania | -3.632 |
Token | me |
Token ID | 502 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
ibaba | +0.688 |
emale | +0.666 |
aimon | +0.519 |
ijing | +0.509 |
WRITE | +0.508 |
Neg logit contributions | |
meg | -0.485 |
star | -0.472 |
� | -0.462 |
mania | -0.457 |
pres | -0.456 |
Token | , |
Token ID | 11 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
meg | +0.494 |
star | +0.478 |
� | +0.463 |
mania | +0.460 |
power | +0.457 |
Neg logit contributions | |
ibaba | -0.984 |
emale | -0.962 |
aimon | -0.772 |
ijing | -0.762 |
WRITE | -0.754 |
Token | and |
Token ID | 290 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
ibaba | +0.563 |
emale | +0.547 |
aimon | +0.428 |
ijing | +0.421 |
mustache | +0.420 |
Neg logit contributions | |
meg | -0.368 |
star | -0.358 |
� | -0.349 |
mania | -0.346 |
power | -0.344 |
Token | the |
Token ID | 262 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
ibaba | +0.465 |
emale | +0.452 |
aimon | +0.357 |
WRITE | +0.351 |
ijing | +0.350 |
Neg logit contributions | |
meg | -0.281 |
star | -0.273 |
� | -0.266 |
mania | -0.263 |
power | -0.263 |
Token | knee |
Token ID | 10329 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
ibaba | +0.095 |
emale | +0.093 |
aimon | +0.073 |
ijing | +0.072 |
WRITE | +0.072 |
Neg logit contributions | |
meg | -0.059 |
star | -0.057 |
� | -0.056 |
power | -0.055 |
mania | -0.055 |
Token | surgery |
Token ID | 8185 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
ibaba | +1.216 |
emale | +1.169 |
aimon | +0.889 |
mustache | +0.878 |
ijing | +0.872 |
Neg logit contributions | |
meg | -1.018 |
star | -0.988 |
� | -0.976 |
pres | -0.961 |
power | -0.956 |
Token | , |
Token ID | 11 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
ibaba | +2.997 |
emale | +2.904 |
aimon | +2.296 |
mustache | +2.291 |
margins | +2.285 |
Neg logit contributions | |
meg | -1.886 |
� | -1.843 |
star | -1.818 |
power | -1.752 |
pres | -1.750 |
Token | a |
Token ID | 257 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
ibaba | +0.138 |
emale | +0.134 |
aimon | +0.103 |
mustache | +0.101 |
ijing | +0.101 |
Neg logit contributions | |
meg | -0.105 |
star | -0.102 |
� | -0.100 |
pres | -0.098 |
power | -0.098 |
Token | broken |
Token ID | 5445 |
Feature activation | -7.86e-03 |
Pos logit contributions | |
ibaba | +0.375 |
emale | +0.364 |
aimon | +0.278 |
ijing | +0.272 |
WRITE | +0.270 |
Neg logit contributions | |
meg | -0.301 |
star | -0.293 |
� | -0.287 |
power | -0.284 |
pres | -0.284 |
Token | finger |
Token ID | 7660 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ibaba | +0.027 |
emale | +0.026 |
aimon | +0.019 |
ijing | +0.018 |
WRITE | +0.018 |
Neg logit contributions | |
meg | -0.030 |
star | -0.030 |
� | -0.029 |
power | -0.029 |
mania | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
ibaba | +6.024 |
emale | +5.825 |
mustache | +4.690 |
aimon | +4.666 |
buckle | +4.570 |
Neg logit contributions | |
meg | -3.984 |
� | -3.849 |
star | -3.838 |
rovers | -3.730 |
pres | -3.700 |
Token | his |
Token ID | 465 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
ibaba | +0.067 |
emale | +0.065 |
aimon | +0.051 |
mustache | +0.050 |
ijing | +0.049 |
Neg logit contributions | |
meg | -0.050 |
star | -0.049 |
� | -0.048 |
pres | -0.047 |
power | -0.047 |
Token | manager |
Token ID | 4706 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
ibaba | +1.023 |
emale | +1.000 |
aimon | +0.773 |
mustache | +0.756 |
ijing | +0.754 |
Neg logit contributions | |
meg | -0.752 |
star | -0.730 |
� | -0.716 |
power | -0.711 |
pres | -0.707 |
Token | and |
Token ID | 290 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
meg | +0.846 |
star | +0.822 |
� | +0.797 |
power | +0.789 |
pres | +0.789 |
Neg logit contributions | |
ibaba | -1.427 |
emale | -1.392 |
aimon | -1.100 |
ijing | -1.081 |
WRITE | -1.074 |
Token | the |
Token ID | 262 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
meg | +0.716 |
star | +0.699 |
� | +0.678 |
power | +0.674 |
pres | +0.673 |
Neg logit contributions | |
ibaba | -1.050 |
emale | -1.023 |
aimon | -0.794 |
ijing | -0.781 |
WRITE | -0.773 |
Token |
|
Token ID | 198 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
meg | +0.028 |
star | +0.027 |
� | +0.026 |
power | +0.026 |
mania | +0.026 |
Neg logit contributions | |
ibaba | -0.054 |
emale | -0.053 |
aimon | -0.042 |
ijing | -0.042 |
WRITE | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
meg | +0.265 |
star | +0.256 |
� | +0.246 |
pres | +0.243 |
power | +0.243 |
Neg logit contributions | |
ibaba | -0.631 |
emale | -0.617 |
aimon | -0.502 |
ijing | -0.495 |
WRITE | -0.492 |
Token | And |
Token ID | 1870 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
meg | +0.271 |
star | +0.264 |
� | +0.253 |
power | +0.252 |
pres | +0.251 |
Neg logit contributions | |
ibaba | -0.564 |
emale | -0.551 |
aimon | -0.443 |
ijing | -0.437 |
WRITE | -0.432 |
Token | don |
Token ID | 836 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
ibaba | +2.325 |
emale | +2.265 |
aimon | +1.800 |
ijing | +1.770 |
WRITE | +1.766 |
Neg logit contributions | |
meg | -1.314 |
star | -1.278 |
� | -1.254 |
mania | -1.231 |
pres | -1.229 |
Token | 't |
Token ID | 470 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
ibaba | +0.196 |
emale | +0.192 |
aimon | +0.157 |
ijing | +0.155 |
WRITE | +0.154 |
Neg logit contributions | |
meg | -0.077 |
star | -0.074 |
� | -0.071 |
pres | -0.070 |
power | -0.070 |
Token | worry |
Token ID | 5490 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
meg | +1.009 |
star | +0.987 |
� | +0.951 |
power | +0.949 |
mania | +0.947 |
Neg logit contributions | |
ibaba | -1.580 |
emale | -1.541 |
aimon | -1.203 |
ijing | -1.185 |
WRITE | -1.168 |
Token | – |
Token ID | 784 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
ibaba | +5.827 |
emale | +5.644 |
WRITE | +4.519 |
aimon | +4.518 |
mustache | +4.451 |
Neg logit contributions | |
meg | -3.856 |
rovers | -3.736 |
star | -3.730 |
� | -3.717 |
pres | -3.662 |
Token | you |
Token ID | 345 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
ibaba | +0.364 |
emale | +0.356 |
aimon | +0.282 |
ijing | +0.277 |
WRITE | +0.275 |
Neg logit contributions | |
meg | -0.207 |
star | -0.202 |
� | -0.195 |
power | -0.193 |
pres | -0.193 |
Token | will |
Token ID | 481 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
ibaba | +0.407 |
emale | +0.398 |
aimon | +0.315 |
ijing | +0.311 |
WRITE | +0.308 |
Neg logit contributions | |
meg | -0.228 |
star | -0.222 |
� | -0.215 |
power | -0.213 |
pres | -0.213 |
Token | once |
Token ID | 1752 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
meg | +0.567 |
star | +0.554 |
power | +0.533 |
pres | +0.531 |
mania | +0.530 |
Neg logit contributions | |
ibaba | -0.958 |
emale | -0.935 |
aimon | -0.736 |
ijing | -0.725 |
WRITE | -0.711 |
Token | again |
Token ID | 757 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
meg | +1.120 |
star | +1.088 |
power | +1.042 |
� | +1.041 |
mania | +1.039 |
Neg logit contributions | |
ibaba | -2.209 |
emale | -2.162 |
aimon | -1.730 |
ijing | -1.702 |
WRITE | -1.684 |
Token | HA |
Token ID | 7801 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
meg | +0.455 |
star | +0.442 |
� | +0.432 |
mania | +0.426 |
pres | +0.424 |
Neg logit contributions | |
ibaba | -0.881 |
emale | -0.858 |
aimon | -0.689 |
ijing | -0.678 |
WRITE | -0.671 |
Token | V |
Token ID | 53 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
ibaba | +0.085 |
emale | +0.083 |
aimon | +0.066 |
ijing | +0.066 |
WRITE | +0.065 |
Neg logit contributions | |
meg | -0.043 |
star | -0.042 |
� | -0.040 |
power | -0.040 |
pres | -0.040 |
Token | ANA |
Token ID | 31574 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
ibaba | +1.740 |
emale | +1.665 |
aimon | +1.180 |
WRITE | +1.171 |
ijing | +1.165 |
Neg logit contributions | |
meg | -2.102 |
star | -2.049 |
� | -2.038 |
pres | -2.002 |
power | -1.993 |
Token | ( |
Token ID | 357 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
ibaba | +2.856 |
emale | +2.795 |
aimon | +2.210 |
WRITE | +2.202 |
ijing | +2.173 |
Neg logit contributions | |
meg | -1.709 |
star | -1.648 |
� | -1.601 |
rovers | -1.585 |
power | -1.578 |
Token | Reuters |
Token ID | 12637 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
ibaba | +3.268 |
emale | +3.160 |
WRITE | +2.417 |
aimon | +2.412 |
ijing | +2.387 |
Neg logit contributions | |
meg | -2.723 |
star | -2.650 |
� | -2.593 |
rovers | -2.589 |
power | -2.567 |
Token | ) |
Token ID | 8 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ibaba | +1.504 |
emale | +1.464 |
aimon | +1.115 |
ijing | +1.095 |
WRITE | +1.083 |
Neg logit contributions | |
meg | -1.188 |
star | -1.162 |
� | -1.129 |
power | -1.123 |
pres | -1.122 |
Token | - |
Token ID | 532 |
Feature activation | -8.66e-01 |
Pos logit contributions | |
ibaba | +4.955 |
emale | +4.737 |
WRITE | +3.648 |
aimon | +3.576 |
ijing | +3.479 |
Neg logit contributions | |
meg | -4.578 |
star | -4.463 |
rovers | -4.391 |
� | -4.386 |
pres | -4.331 |
Token | Latin |
Token ID | 9133 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
ibaba | +3.362 |
emale | +3.237 |
WRITE | +2.453 |
aimon | +2.400 |
ijing | +2.394 |
Neg logit contributions | |
meg | -2.881 |
star | -2.811 |
� | -2.784 |
rovers | -2.723 |
pres | -2.715 |
Token | American |
Token ID | 1605 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
ibaba | +0.296 |
emale | +0.286 |
ijing | +0.210 |
aimon | +0.209 |
mustache | +0.204 |
Neg logit contributions | |
meg | -0.277 |
star | -0.269 |
� | -0.268 |
pres | -0.265 |
rovers | -0.263 |
Token | and |
Token ID | 290 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
meg | +0.364 |
star | +0.355 |
� | +0.345 |
pres | +0.343 |
power | +0.342 |
Neg logit contributions | |
ibaba | -0.514 |
emale | -0.501 |
aimon | -0.387 |
ijing | -0.381 |
WRITE | -0.377 |
Token | Caribbean |
Token ID | 18020 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ibaba | +0.927 |
emale | +0.896 |
aimon | +0.684 |
ijing | +0.673 |
mustache | +0.667 |
Neg logit contributions | |
meg | -0.747 |
star | -0.731 |
� | -0.725 |
pres | -0.714 |
mania | -0.711 |
Token | tour |
Token ID | 4205 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
meg | +0.103 |
star | +0.100 |
� | +0.098 |
power | +0.097 |
mania | +0.097 |
Neg logit contributions | |
ibaba | -0.170 |
emale | -0.166 |
aimon | -0.131 |
ijing | -0.129 |
WRITE | -0.128 |
Token | , |
Token ID | 11 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
meg | +0.387 |
star | +0.376 |
� | +0.364 |
power | +0.361 |
pres | +0.360 |
Neg logit contributions | |
ibaba | -0.699 |
emale | -0.683 |
aimon | -0.543 |
ijing | -0.535 |
WRITE | -0.529 |
Token | plenty |
Token ID | 6088 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
meg | +0.249 |
star | +0.242 |
� | +0.236 |
mania | +0.233 |
power | +0.233 |
Neg logit contributions | |
ibaba | -0.418 |
emale | -0.408 |
aimon | -0.321 |
ijing | -0.316 |
WRITE | -0.315 |
Token | of |
Token ID | 286 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
ibaba | +3.415 |
emale | +3.360 |
aimon | +2.607 |
mustache | +2.589 |
ijing | +2.550 |
Neg logit contributions | |
meg | -2.104 |
star | -2.037 |
� | -2.021 |
pres | -2.014 |
mania | -2.005 |
Token | body |
Token ID | 1767 |
Feature activation | +6.71e-02 |
Pos logit contributions | |
meg | +0.147 |
star | +0.143 |
� | +0.140 |
power | +0.138 |
mania | +0.138 |
Neg logit contributions | |
ibaba | -0.224 |
emale | -0.219 |
aimon | -0.171 |
ijing | -0.168 |
WRITE | -0.167 |
Token | parts |
Token ID | 3354 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
meg | +0.191 |
star | +0.186 |
� | +0.183 |
mania | +0.181 |
rovers | +0.179 |
Neg logit contributions | |
ibaba | -0.296 |
emale | -0.291 |
aimon | -0.226 |
ijing | -0.223 |
mustache | -0.221 |
Token | were |
Token ID | 547 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
ibaba | +5.566 |
emale | +5.487 |
mustache | +4.286 |
aimon | +4.274 |
buckle | +4.189 |
Neg logit contributions | |
meg | -3.652 |
� | -3.603 |
star | -3.573 |
rovers | -3.506 |
mania | -3.456 |
Token | already |
Token ID | 1541 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
ibaba | +0.695 |
emale | +0.678 |
aimon | +0.536 |
ijing | +0.528 |
mustache | +0.526 |
Neg logit contributions | |
meg | -0.425 |
star | -0.413 |
� | -0.406 |
power | -0.398 |
mania | -0.398 |
Token | set |
Token ID | 900 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
ibaba | +0.622 |
emale | +0.605 |
aimon | +0.475 |
ijing | +0.468 |
WRITE | +0.464 |
Neg logit contributions | |
meg | -0.396 |
star | -0.385 |
� | -0.376 |
power | -0.371 |
mania | -0.370 |
Token | up |
Token ID | 510 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
ibaba | +0.251 |
emale | +0.244 |
aimon | +0.194 |
ijing | +0.192 |
WRITE | +0.190 |
Neg logit contributions | |
meg | -0.136 |
star | -0.132 |
� | -0.128 |
power | -0.127 |
mania | -0.127 |
Token | , |
Token ID | 11 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
meg | +1.361 |
star | +1.321 |
� | +1.263 |
pres | +1.260 |
power | +1.256 |
Neg logit contributions | |
ibaba | -2.849 |
emale | -2.788 |
aimon | -2.238 |
ijing | -2.206 |
WRITE | -2.186 |
Token | - |
Token ID | 12 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
ibaba | +1.708 |
emale | +1.637 |
aimon | +1.192 |
ijing | +1.149 |
mustache | +1.147 |
Neg logit contributions | |
meg | -1.957 |
star | -1.924 |
� | -1.889 |
pres | -1.873 |
mania | -1.870 |
Token | level |
Token ID | 5715 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
ibaba | +0.097 |
emale | +0.095 |
aimon | +0.076 |
ijing | +0.075 |
WRITE | +0.074 |
Neg logit contributions | |
meg | -0.050 |
star | -0.049 |
� | -0.048 |
mania | -0.047 |
pres | -0.047 |
Token | of |
Token ID | 286 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
ibaba | +1.409 |
emale | +1.361 |
aimon | +1.057 |
mustache | +1.038 |
ijing | +1.036 |
Neg logit contributions | |
meg | -1.064 |
star | -1.036 |
� | -1.011 |
rovers | -0.997 |
pres | -0.995 |
Token | intellectual |
Token ID | 9028 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
ibaba | +1.246 |
emale | +1.201 |
aimon | +0.948 |
mustache | +0.927 |
ijing | +0.927 |
Neg logit contributions | |
meg | -0.889 |
star | -0.863 |
� | -0.848 |
rovers | -0.828 |
mania | -0.827 |
Token | rig |
Token ID | 7805 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
ibaba | +0.157 |
emale | +0.151 |
aimon | +0.118 |
ijing | +0.116 |
WRITE | +0.115 |
Neg logit contributions | |
meg | -0.109 |
star | -0.106 |
� | -0.104 |
pres | -0.102 |
mania | -0.102 |
Token | or |
Token ID | 273 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
meg | +0.583 |
star | +0.568 |
� | +0.559 |
mania | +0.557 |
pres | +0.548 |
Neg logit contributions | |
ibaba | -0.882 |
emale | -0.854 |
aimon | -0.670 |
ijing | -0.662 |
WRITE | -0.645 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
ibaba | +5.686 |
emale | +5.394 |
aimon | +4.391 |
mustache | +4.384 |
ijing | +4.372 |
Neg logit contributions | |
meg | -3.329 |
star | -3.212 |
� | -3.191 |
rovers | -3.127 |
mania | -3.117 |
Token | But |
Token ID | 887 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
meg | +0.310 |
star | +0.302 |
� | +0.292 |
power | +0.290 |
mania | +0.290 |
Neg logit contributions | |
ibaba | -0.508 |
emale | -0.496 |
aimon | -0.390 |
ijing | -0.384 |
WRITE | -0.381 |
Token | they |
Token ID | 484 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
ibaba | +0.503 |
emale | +0.488 |
aimon | +0.386 |
ijing | +0.380 |
WRITE | +0.379 |
Neg logit contributions | |
meg | -0.307 |
star | -0.299 |
� | -0.290 |
mania | -0.287 |
power | -0.286 |
Token | surrounded |
Token ID | 11191 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
ibaba | +0.122 |
emale | +0.119 |
aimon | +0.094 |
ijing | +0.093 |
WRITE | +0.092 |
Neg logit contributions | |
meg | -0.072 |
star | -0.070 |
� | -0.068 |
mania | -0.067 |
power | -0.067 |
Token | themselves |
Token ID | 2405 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
meg | +0.097 |
star | +0.094 |
� | +0.091 |
mania | +0.090 |
power | +0.090 |
Neg logit contributions | |
ibaba | -0.151 |
emale | -0.146 |
aimon | -0.115 |
ijing | -0.113 |
WRITE | -0.112 |
Token | ( |
Token ID | 357 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
ibaba | +0.175 |
emale | +0.170 |
aimon | +0.134 |
ijing | +0.132 |
WRITE | +0.131 |
Neg logit contributions | |
meg | -0.103 |
star | -0.101 |
� | -0.098 |
mania | -0.097 |
power | -0.097 |
Token | the |
Token ID | 1169 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
ibaba | +0.239 |
emale | +0.233 |
aimon | +0.188 |
ijing | +0.186 |
WRITE | +0.184 |
Neg logit contributions | |
meg | -0.109 |
star | -0.106 |
� | -0.103 |
mania | -0.101 |
power | -0.101 |
Token | proposal |
Token ID | 6961 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
ibaba | +0.501 |
emale | +0.489 |
aimon | +0.379 |
ijing | +0.373 |
WRITE | +0.370 |
Neg logit contributions | |
meg | -0.346 |
star | -0.338 |
� | -0.328 |
mania | -0.326 |
power | -0.326 |
Token | by |
Token ID | 416 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
ibaba | +1.700 |
emale | +1.652 |
aimon | +1.313 |
ijing | +1.299 |
WRITE | +1.280 |
Neg logit contributions | |
meg | -0.989 |
star | -0.960 |
� | -0.931 |
power | -0.925 |
mania | -0.924 |
Token | Andy |
Token ID | 12382 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
ibaba | +0.311 |
emale | +0.303 |
aimon | +0.234 |
ijing | +0.231 |
WRITE | +0.228 |
Neg logit contributions | |
meg | -0.220 |
star | -0.215 |
� | -0.209 |
mania | -0.207 |
pres | -0.207 |
Token | Stern |
Token ID | 22990 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ibaba | +1.047 |
emale | +1.013 |
aimon | +0.799 |
ijing | +0.786 |
WRITE | +0.780 |
Neg logit contributions | |
meg | -0.706 |
star | -0.676 |
� | -0.669 |
pres | -0.668 |
power | -0.662 |
Token | in |
Token ID | 287 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
ibaba | +5.860 |
emale | +5.750 |
aimon | +4.606 |
ijing | +4.580 |
mustache | +4.454 |
Neg logit contributions | |
meg | -3.185 |
� | -3.044 |
star | -3.003 |
mania | -2.948 |
power | -2.942 |
Token | Ra |
Token ID | 7567 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
ibaba | +0.375 |
emale | +0.366 |
aimon | +0.285 |
ijing | +0.281 |
WRITE | +0.278 |
Neg logit contributions | |
meg | -0.254 |
star | -0.247 |
� | -0.240 |
pres | -0.238 |
mania | -0.238 |
Token | ising |
Token ID | 1710 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
meg | +0.620 |
star | +0.607 |
� | +0.593 |
mania | +0.588 |
power | +0.587 |
Neg logit contributions | |
ibaba | -0.725 |
emale | -0.705 |
aimon | -0.531 |
ijing | -0.521 |
WRITE | -0.514 |
Token | the |
Token ID | 262 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
meg | +0.176 |
star | +0.172 |
� | +0.167 |
power | +0.166 |
pres | +0.166 |
Neg logit contributions | |
ibaba | -0.241 |
emale | -0.235 |
aimon | -0.181 |
ijing | -0.178 |
WRITE | -0.176 |
Token | Floor |
Token ID | 22343 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
ibaba | +1.536 |
emale | +1.503 |
aimon | +1.113 |
ijing | +1.113 |
WRITE | +1.110 |
Neg logit contributions | |
meg | -1.293 |
star | -1.257 |
� | -1.230 |
mania | -1.230 |
pres | -1.224 |
Token | Frank |
Token ID | 17439 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
meg | +0.066 |
� | +0.064 |
star | +0.064 |
pres | +0.062 |
rovers | +0.062 |
Neg logit contributions | |
ibaba | -0.088 |
emale | -0.087 |
ijing | -0.066 |
aimon | -0.066 |
WRITE | -0.065 |
Token | iv |
Token ID | 452 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
meg | +0.044 |
star | +0.042 |
� | +0.040 |
mania | +0.040 |
power | +0.039 |
Neg logit contributions | |
ibaba | -0.138 |
emale | -0.134 |
aimon | -0.111 |
ijing | -0.110 |
WRITE | -0.108 |
Token | sk |
Token ID | 8135 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
ibaba | +0.121 |
emale | +0.114 |
ijing | +0.095 |
aimon | +0.095 |
irtual | +0.093 |
Neg logit contributions | |
meg | -0.054 |
� | -0.054 |
star | -0.051 |
mania | -0.051 |
power | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
meg | +0.703 |
star | +0.685 |
� | +0.661 |
mania | +0.655 |
power | +0.655 |
Neg logit contributions | |
ibaba | -1.357 |
emale | -1.324 |
aimon | -1.057 |
ijing | -1.046 |
WRITE | -1.033 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
ibaba | +1.191 |
emale | +1.160 |
aimon | +0.903 |
WRITE | +0.896 |
ijing | +0.889 |
Neg logit contributions | |
meg | -0.773 |
star | -0.753 |
� | -0.735 |
pres | -0.723 |
power | -0.720 |
Token | [ |
Token ID | 58 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
meg | +0.051 |
star | +0.049 |
� | +0.048 |
mania | +0.047 |
pres | +0.047 |
Neg logit contributions | |
ibaba | -0.098 |
emale | -0.095 |
aimon | -0.077 |
ijing | -0.075 |
WRITE | -0.075 |
Token | Ex |
Token ID | 3109 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
ibaba | +5.912 |
emale | +5.692 |
aimon | +4.566 |
ijing | +4.551 |
WRITE | +4.544 |
Neg logit contributions | |
meg | -3.131 |
� | -3.001 |
star | -2.991 |
rovers | -2.908 |
mania | -2.900 |
Token | clusive |
Token ID | 5731 |
Feature activation | -9.12e-03 |
Pos logit contributions | |
meg | +0.144 |
star | +0.140 |
� | +0.138 |
mania | +0.135 |
power | +0.135 |
Neg logit contributions | |
ibaba | -0.233 |
emale | -0.229 |
aimon | -0.178 |
ijing | -0.176 |
WRITE | -0.173 |
Token | ] |
Token ID | 60 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
ibaba | +0.040 |
emale | +0.039 |
aimon | +0.031 |
ijing | +0.030 |
WRITE | +0.030 |
Neg logit contributions | |
meg | -0.026 |
star | -0.026 |
� | -0.025 |
mania | -0.025 |
pres | -0.025 |
Token | Best |
Token ID | 6705 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
meg | +0.593 |
star | +0.578 |
� | +0.564 |
pres | +0.559 |
power | +0.559 |
Neg logit contributions | |
ibaba | -0.799 |
emale | -0.778 |
aimon | -0.597 |
ijing | -0.588 |
WRITE | -0.584 |
Token | Buy |
Token ID | 11763 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
meg | +0.413 |
star | +0.403 |
� | +0.393 |
power | +0.390 |
mania | +0.390 |
Neg logit contributions | |
ibaba | -0.553 |
emale | -0.539 |
aimon | -0.413 |
ijing | -0.407 |
WRITE | -0.402 |
Token | , |
Token ID | 11 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
ibaba | +0.207 |
emale | +0.202 |
aimon | +0.158 |
ijing | +0.156 |
WRITE | +0.156 |
Neg logit contributions | |
meg | -0.125 |
star | -0.121 |
� | -0.118 |
mania | -0.116 |
power | -0.116 |
Token | I |
Token ID | 314 |
Feature activation | -5.61e-03 |
Pos logit contributions | |
ibaba | +0.496 |
emale | +0.484 |
aimon | +0.383 |
ijing | +0.377 |
WRITE | +0.375 |
Neg logit contributions | |
meg | -0.285 |
star | -0.277 |
� | -0.269 |
power | -0.265 |
mania | -0.265 |
Token | 'm |
Token ID | 1101 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
ibaba | +0.026 |
emale | +0.026 |
aimon | +0.020 |
ijing | +0.020 |
WRITE | +0.020 |
Neg logit contributions | |
meg | -0.015 |
star | -0.014 |
� | -0.014 |
pres | -0.014 |
power | -0.014 |
Token | not |
Token ID | 407 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
meg | +0.041 |
star | +0.040 |
� | +0.038 |
power | +0.038 |
pres | +0.038 |
Neg logit contributions | |
ibaba | -0.064 |
emale | -0.062 |
aimon | -0.049 |
ijing | -0.048 |
WRITE | -0.047 |
Token | looking |
Token ID | 2045 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
ibaba | +0.754 |
emale | +0.735 |
aimon | +0.571 |
ijing | +0.566 |
WRITE | +0.560 |
Neg logit contributions | |
meg | -0.523 |
star | -0.506 |
� | -0.496 |
power | -0.488 |
pres | -0.488 |
Token | ahead |
Token ID | 4058 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
meg | +0.539 |
star | +0.518 |
� | +0.511 |
mania | +0.499 |
pres | +0.499 |
Neg logit contributions | |
ibaba | -0.996 |
emale | -0.969 |
aimon | -0.773 |
ijing | -0.769 |
WRITE | -0.761 |
Token | , |
Token ID | 11 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
ibaba | +5.596 |
emale | +5.448 |
aimon | +4.371 |
ijing | +4.369 |
WRITE | +4.364 |
Neg logit contributions | |
meg | -3.289 |
� | -3.179 |
star | -3.137 |
rovers | -3.093 |
power | -3.022 |
Token | I |
Token ID | 314 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ibaba | +0.561 |
emale | +0.548 |
aimon | +0.431 |
ijing | +0.425 |
WRITE | +0.423 |
Neg logit contributions | |
meg | -0.332 |
star | -0.323 |
� | -0.314 |
power | -0.310 |
mania | -0.310 |
Token | 'm |
Token ID | 1101 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ibaba | +0.489 |
emale | +0.478 |
aimon | +0.383 |
ijing | +0.379 |
WRITE | +0.378 |
Neg logit contributions | |
meg | -0.250 |
star | -0.241 |
� | -0.236 |
pres | -0.231 |
power | -0.230 |
Token | not |
Token ID | 407 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ibaba | +0.568 |
emale | +0.551 |
aimon | +0.426 |
ijing | +0.420 |
WRITE | +0.415 |
Neg logit contributions | |
meg | -0.427 |
star | -0.414 |
� | -0.406 |
rovers | -0.400 |
pres | -0.400 |
Token | looking |
Token ID | 2045 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
ibaba | +0.519 |
emale | +0.504 |
aimon | +0.385 |
ijing | +0.379 |
WRITE | +0.376 |
Neg logit contributions | |
meg | -0.416 |
star | -0.404 |
� | -0.396 |
power | -0.391 |
pres | -0.391 |
Token | 2015 |
Token ID | 1853 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
meg | +0.101 |
star | +0.098 |
� | +0.096 |
mania | +0.095 |
pres | +0.095 |
Neg logit contributions | |
ibaba | -0.184 |
emale | -0.179 |
aimon | -0.143 |
ijing | -0.141 |
WRITE | -0.140 |
Token |
|
Token ID | 198 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
ibaba | +0.123 |
emale | +0.120 |
aimon | +0.096 |
ijing | +0.095 |
WRITE | +0.095 |
Neg logit contributions | |
meg | -0.064 |
star | -0.062 |
� | -0.061 |
mania | -0.060 |
pres | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ibaba | +0.871 |
emale | +0.851 |
aimon | +0.686 |
ijing | +0.676 |
WRITE | +0.673 |
Neg logit contributions | |
meg | -0.436 |
star | -0.422 |
� | -0.412 |
pres | -0.405 |
mania | -0.405 |
Token | FR |
Token ID | 10913 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
ibaba | +0.651 |
emale | +0.636 |
aimon | +0.511 |
ijing | +0.503 |
WRITE | +0.499 |
Neg logit contributions | |
meg | -0.334 |
star | -0.323 |
� | -0.317 |
mania | -0.311 |
rovers | -0.311 |
Token | ES |
Token ID | 1546 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
ibaba | +1.416 |
emale | +1.382 |
aimon | +1.078 |
WRITE | +1.066 |
ijing | +1.064 |
Neg logit contributions | |
meg | -0.967 |
star | -0.943 |
� | -0.923 |
power | -0.905 |
rovers | -0.902 |
Token | NO |
Token ID | 15285 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ibaba | +1.541 |
emale | +1.523 |
ijing | +1.184 |
aimon | +1.182 |
WRITE | +1.176 |
Neg logit contributions | |
meg | -0.899 |
star | -0.884 |
� | -0.853 |
power | -0.849 |
rovers | -0.844 |
Token | , |
Token ID | 11 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
emale | +5.213 |
ibaba | +5.192 |
WRITE | +4.099 |
aimon | +4.045 |
ijing | +3.980 |
Neg logit contributions | |
meg | -3.699 |
star | -3.580 |
� | -3.491 |
rovers | -3.465 |
mania | -3.446 |
Token | Calif |
Token ID | 3192 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
ibaba | +0.799 |
emale | +0.779 |
aimon | +0.586 |
WRITE | +0.574 |
ijing | +0.574 |
Neg logit contributions | |
meg | -0.697 |
star | -0.680 |
� | -0.665 |
power | -0.660 |
mania | -0.658 |
Token | . |
Token ID | 13 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
ibaba | +2.210 |
emale | +2.169 |
aimon | +1.762 |
ijing | +1.750 |
WRITE | +1.744 |
Neg logit contributions | |
meg | -0.858 |
� | -0.817 |
star | -0.816 |
mania | -0.793 |
power | -0.778 |
Token | ( |
Token ID | 357 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
ibaba | +1.428 |
emale | +1.394 |
aimon | +1.086 |
WRITE | +1.086 |
ijing | +1.072 |
Neg logit contributions | |
meg | -0.938 |
star | -0.910 |
� | -0.888 |
mania | -0.879 |
power | -0.874 |
Token | K |
Token ID | 42 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
ibaba | +2.690 |
emale | +2.632 |
aimon | +2.139 |
WRITE | +2.124 |
ijing | +2.095 |
Neg logit contributions | |
meg | -1.279 |
star | -1.235 |
� | -1.202 |
rovers | -1.186 |
mania | -1.186 |
Token | ( |
Token ID | 357 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
meg | +0.245 |
star | +0.237 |
� | +0.229 |
power | +0.226 |
pres | +0.225 |
Neg logit contributions | |
ibaba | -0.514 |
emale | -0.504 |
aimon | -0.405 |
ijing | -0.400 |
WRITE | -0.397 |
Token | Figure |
Token ID | 11337 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
ibaba | +1.896 |
emale | +1.861 |
aimon | +1.486 |
ijing | +1.465 |
WRITE | +1.449 |
Neg logit contributions | |
meg | -0.958 |
star | -0.929 |
� | -0.904 |
rovers | -0.887 |
mania | -0.886 |
Token | 2 |
Token ID | 362 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
ibaba | +1.753 |
emale | +1.704 |
aimon | +1.262 |
ijing | +1.254 |
WRITE | +1.239 |
Neg logit contributions | |
meg | -1.572 |
star | -1.532 |
� | -1.500 |
power | -1.484 |
pres | -1.482 |
Token | A |
Token ID | 32 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
ibaba | +1.620 |
emale | +1.586 |
aimon | +1.211 |
ijing | +1.204 |
WRITE | +1.191 |
Neg logit contributions | |
meg | -1.153 |
star | -1.125 |
� | -1.101 |
mania | -1.087 |
power | -1.085 |
Token | , |
Token ID | 11 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
ibaba | +0.193 |
emale | +0.188 |
aimon | +0.140 |
ijing | +0.138 |
WRITE | +0.137 |
Neg logit contributions | |
meg | -0.167 |
star | -0.163 |
� | -0.159 |
mania | -0.158 |
power | -0.158 |
Token | [ |
Token ID | 685 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ibaba | +0.635 |
emale | +0.620 |
aimon | +0.486 |
ijing | +0.480 |
WRITE | +0.475 |
Neg logit contributions | |
meg | -0.388 |
star | -0.377 |
� | -0.366 |
mania | -0.363 |
power | -0.363 |
Token | Action |
Token ID | 12502 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
ibaba | +5.830 |
emale | +5.689 |
aimon | +4.520 |
ijing | +4.495 |
WRITE | +4.433 |
Neg logit contributions | |
meg | -3.101 |
star | -2.968 |
� | -2.967 |
rovers | -2.930 |
mania | -2.892 |
Token | Junior |
Token ID | 20000 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
ibaba | +0.149 |
emale | +0.145 |
aimon | +0.111 |
ijing | +0.110 |
WRITE | +0.109 |
Neg logit contributions | |
meg | -0.106 |
star | -0.103 |
� | -0.100 |
pres | -0.099 |
power | -0.099 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
meg | +0.787 |
star | +0.771 |
� | +0.748 |
power | +0.744 |
mania | +0.744 |
Neg logit contributions | |
ibaba | -1.045 |
emale | -1.015 |
aimon | -0.779 |
ijing | -0.766 |
WRITE | -0.755 |
Token | , |
Token ID | 11 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ibaba | +0.637 |
emale | +0.626 |
aimon | +0.488 |
ijing | +0.483 |
WRITE | +0.479 |
Neg logit contributions | |
meg | -0.394 |
star | -0.382 |
� | -0.371 |
mania | -0.368 |
pres | -0.367 |
Token | Inv |
Token ID | 10001 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
ibaba | +0.551 |
emale | +0.538 |
aimon | +0.411 |
ijing | +0.403 |
WRITE | +0.400 |
Neg logit contributions | |
meg | -0.425 |
star | -0.415 |
� | -0.403 |
mania | -0.401 |
pres | -0.401 |
Token | Y |
Token ID | 575 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
ibaba | +0.004 |
emale | +0.004 |
aimon | +0.002 |
ijing | +0.002 |
WRITE | +0.002 |
Neg logit contributions | |
meg | -0.005 |
star | -0.005 |
� | -0.005 |
power | -0.005 |
pres | -0.005 |
Token | ip |
Token ID | 541 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
meg | +0.252 |
star | +0.246 |
� | +0.243 |
mania | +0.237 |
pres | +0.235 |
Neg logit contributions | |
ibaba | -0.381 |
emale | -0.370 |
aimon | -0.288 |
ijing | -0.287 |
WRITE | -0.280 |
Token |
|
Token ID | 198 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
meg | +0.742 |
star | +0.724 |
� | +0.699 |
power | +0.695 |
mania | +0.694 |
Neg logit contributions | |
ibaba | -1.220 |
emale | -1.191 |
aimon | -0.938 |
ijing | -0.921 |
WRITE | -0.912 |
Token |
|
Token ID | 198 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
ibaba | +2.972 |
emale | +2.849 |
aimon | +2.339 |
WRITE | +2.325 |
ijing | +2.283 |
Neg logit contributions | |
meg | -1.450 |
star | -1.391 |
� | -1.367 |
rovers | -1.350 |
pres | -1.332 |
Token | H |
Token ID | 39 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
ibaba | +1.833 |
emale | +1.790 |
aimon | +1.385 |
ijing | +1.381 |
WRITE | +1.363 |
Neg logit contributions | |
meg | -1.226 |
star | -1.197 |
� | -1.162 |
pres | -1.157 |
rovers | -1.154 |
Token | ONG |
Token ID | 18494 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ibaba | +0.941 |
emale | +0.906 |
aimon | +0.645 |
ijing | +0.642 |
WRITE | +0.618 |
Neg logit contributions | |
meg | -1.071 |
star | -1.057 |
� | -1.036 |
mania | -1.027 |
pres | -1.025 |
Token | K |
Token ID | 509 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
ibaba | +4.783 |
emale | +4.670 |
WRITE | +3.827 |
PLA | +3.756 |
ijing | +3.668 |
Neg logit contributions | |
meg | -3.975 |
rovers | -3.811 |
star | -3.791 |
power | -3.642 |
pres | -3.629 |
Token | ONG |
Token ID | 18494 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
ibaba | +1.149 |
emale | +1.119 |
aimon | +0.632 |
ijing | +0.627 |
WRITE | +0.597 |
Neg logit contributions | |
meg | -2.635 |
star | -2.587 |
rovers | -2.540 |
� | -2.537 |
mania | -2.529 |
Token | ( |
Token ID | 357 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
ibaba | +4.054 |
emale | +3.999 |
ijing | +3.113 |
WRITE | +3.108 |
aimon | +3.064 |
Neg logit contributions | |
meg | -2.573 |
star | -2.497 |
rovers | -2.416 |
pres | -2.399 |
� | -2.391 |
Token | Reuters |
Token ID | 12637 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
ibaba | +2.910 |
emale | +2.792 |
ijing | +2.129 |
WRITE | +2.119 |
aimon | +2.108 |
Neg logit contributions | |
meg | -2.446 |
star | -2.400 |
� | -2.344 |
rovers | -2.330 |
pres | -2.320 |
Token | ) |
Token ID | 8 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ibaba | +1.315 |
emale | +1.280 |
aimon | +0.975 |
ijing | +0.959 |
WRITE | +0.945 |
Neg logit contributions | |
meg | -1.020 |
star | -0.999 |
� | -0.969 |
power | -0.965 |
mania | -0.964 |
Token | suffered |
Token ID | 6989 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
ibaba | +1.958 |
emale | +1.912 |
aimon | +1.515 |
mustache | +1.494 |
WRITE | +1.491 |
Neg logit contributions | |
meg | -1.171 |
star | -1.131 |
� | -1.113 |
rovers | -1.089 |
pres | -1.088 |
Token | a |
Token ID | 257 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
ibaba | +0.940 |
emale | +0.907 |
aimon | +0.701 |
mustache | +0.691 |
WRITE | +0.689 |
Neg logit contributions | |
meg | -0.690 |
star | -0.674 |
� | -0.661 |
pres | -0.655 |
power | -0.653 |
Token | rupt |
Token ID | 31580 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
ibaba | +1.227 |
emale | +1.183 |
aimon | +0.896 |
mustache | +0.879 |
ijing | +0.876 |
Neg logit contributions | |
meg | -1.036 |
star | -1.013 |
� | -0.996 |
pres | -0.982 |
mania | -0.979 |
Token | ured |
Token ID | 1522 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
ibaba | +2.320 |
emale | +2.251 |
aimon | +1.604 |
ijing | +1.552 |
WRITE | +1.516 |
Neg logit contributions | |
meg | -2.708 |
� | -2.680 |
star | -2.637 |
mania | -2.598 |
pres | -2.567 |
Token | Achilles |
Token ID | 39561 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
ibaba | +0.307 |
emale | +0.295 |
aimon | +0.206 |
ijing | +0.201 |
WRITE | +0.200 |
Neg logit contributions | |
meg | -0.384 |
star | -0.377 |
� | -0.370 |
pres | -0.368 |
mania | -0.367 |
Token | tendon |
Token ID | 42205 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ibaba | +2.004 |
emale | +1.948 |
aimon | +1.343 |
mustache | +1.320 |
WRITE | +1.302 |
Neg logit contributions | |
meg | -2.716 |
star | -2.670 |
� | -2.636 |
pres | -2.613 |
mania | -2.592 |
Token | in |
Token ID | 287 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
ibaba | +5.580 |
emale | +5.430 |
aimon | +4.368 |
mustache | +4.361 |
WRITE | +4.293 |
Neg logit contributions | |
meg | -3.146 |
star | -3.063 |
� | -3.019 |
rovers | -2.950 |
pres | -2.941 |
Token | Thursday |
Token ID | 3635 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
ibaba | +0.566 |
emale | +0.549 |
aimon | +0.436 |
ijing | +0.427 |
WRITE | +0.424 |
Neg logit contributions | |
meg | -0.345 |
star | -0.336 |
� | -0.328 |
pres | -0.323 |
power | -0.323 |
Token | 's |
Token ID | 338 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
ibaba | +0.395 |
emale | +0.383 |
aimon | +0.313 |
ijing | +0.307 |
mustache | +0.307 |
Neg logit contributions | |
meg | -0.178 |
� | -0.169 |
star | -0.168 |
rovers | -0.165 |
power | -0.162 |
Token | game |
Token ID | 983 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
ibaba | +0.641 |
emale | +0.621 |
aimon | +0.473 |
ijing | +0.462 |
WRITE | +0.461 |
Neg logit contributions | |
meg | -0.526 |
star | -0.515 |
� | -0.504 |
pres | -0.498 |
power | -0.498 |
Token | against |
Token ID | 1028 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
ibaba | +1.038 |
emale | +1.015 |
aimon | +0.819 |
ijing | +0.809 |
WRITE | +0.807 |
Neg logit contributions | |
meg | -0.470 |
star | -0.453 |
� | -0.442 |
power | -0.432 |
mania | -0.430 |
Token | Media |
Token ID | 6343 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
ibaba | +0.497 |
emale | +0.486 |
aimon | +0.372 |
ijing | +0.368 |
WRITE | +0.366 |
Neg logit contributions | |
meg | -0.364 |
star | -0.354 |
� | -0.347 |
pres | -0.343 |
mania | -0.343 |
Token | ite |
Token ID | 578 |
Feature activation | +8.92e-02 |
Pos logit contributions | |
ibaba | +1.036 |
emale | +1.014 |
aimon | +0.813 |
ijing | +0.806 |
WRITE | +0.806 |
Neg logit contributions | |
meg | -0.495 |
star | -0.475 |
� | -0.466 |
pres | -0.456 |
rovers | -0.454 |
Token | : |
Token ID | 25 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
meg | +0.220 |
star | +0.213 |
� | +0.206 |
pres | +0.204 |
mania | +0.204 |
Neg logit contributions | |
ibaba | -0.432 |
emale | -0.422 |
aimon | -0.337 |
ijing | -0.335 |
WRITE | -0.331 |
Token |
|
Token ID | 198 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
ibaba | +0.975 |
emale | +0.951 |
aimon | +0.761 |
ijing | +0.757 |
WRITE | +0.755 |
Neg logit contributions | |
meg | -0.480 |
star | -0.462 |
� | -0.447 |
mania | -0.444 |
pres | -0.440 |
Token |
|
Token ID | 198 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
ibaba | +1.728 |
emale | +1.663 |
aimon | +1.364 |
WRITE | +1.350 |
ijing | +1.334 |
Neg logit contributions | |
meg | -0.829 |
star | -0.793 |
� | -0.780 |
rovers | -0.767 |
pres | -0.756 |
Token | [ |
Token ID | 58 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ibaba | +2.023 |
emale | +1.989 |
aimon | +1.597 |
ijing | +1.584 |
WRITE | +1.574 |
Neg logit contributions | |
meg | -0.992 |
star | -0.959 |
� | -0.934 |
rovers | -0.921 |
mania | -0.919 |
Token | Chuck |
Token ID | 44324 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
ibaba | +5.565 |
emale | +5.473 |
ijing | +4.333 |
aimon | +4.333 |
WRITE | +4.304 |
Neg logit contributions | |
meg | -3.056 |
star | -2.936 |
� | -2.912 |
rovers | -2.831 |
mania | -2.803 |
Token | Todd |
Token ID | 14377 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
ibaba | +0.669 |
emale | +0.656 |
aimon | +0.501 |
ijing | +0.499 |
WRITE | +0.497 |
Neg logit contributions | |
meg | -0.488 |
� | -0.470 |
star | -0.465 |
pres | -0.463 |
rovers | -0.454 |
Token | :] |
Token ID | 47715 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
meg | +1.048 |
star | +1.019 |
� | +0.991 |
pres | +0.986 |
power | +0.982 |
Neg logit contributions | |
ibaba | -1.616 |
emale | -1.588 |
aimon | -1.234 |
ijing | -1.220 |
WRITE | -1.211 |
Token | � |
Token ID | 564 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
ibaba | +1.713 |
emale | +1.672 |
aimon | +1.295 |
WRITE | +1.286 |
ijing | +1.282 |
Neg logit contributions | |
meg | -1.152 |
star | -1.121 |
� | -1.093 |
mania | -1.079 |
power | -1.077 |
Token | � |
Token ID | 250 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
meg | +0.527 |
star | +0.516 |
� | +0.504 |
mania | +0.501 |
pres | +0.500 |
Neg logit contributions | |
ibaba | -0.561 |
emale | -0.542 |
aimon | -0.399 |
ijing | -0.396 |
WRITE | -0.392 |
Token | ONG |
Token ID | 18494 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ibaba | +0.941 |
emale | +0.906 |
aimon | +0.645 |
ijing | +0.642 |
WRITE | +0.618 |
Neg logit contributions | |
meg | -1.071 |
star | -1.057 |
� | -1.036 |
mania | -1.027 |
pres | -1.025 |
Token | K |
Token ID | 509 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
ibaba | +4.783 |
emale | +4.670 |
WRITE | +3.827 |
PLA | +3.756 |
ijing | +3.668 |
Neg logit contributions | |
meg | -3.975 |
rovers | -3.811 |
star | -3.791 |
power | -3.642 |
pres | -3.629 |
Token | ONG |
Token ID | 18494 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
ibaba | +1.149 |
emale | +1.119 |
aimon | +0.632 |
ijing | +0.627 |
WRITE | +0.597 |
Neg logit contributions | |
meg | -2.635 |
star | -2.587 |
rovers | -2.540 |
� | -2.537 |
mania | -2.529 |
Token | ( |
Token ID | 357 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
ibaba | +4.054 |
emale | +3.999 |
ijing | +3.113 |
WRITE | +3.108 |
aimon | +3.064 |
Neg logit contributions | |
meg | -2.573 |
star | -2.497 |
rovers | -2.416 |
pres | -2.399 |
� | -2.391 |
Token | Reuters |
Token ID | 12637 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
ibaba | +2.910 |
emale | +2.792 |
ijing | +2.129 |
WRITE | +2.119 |
aimon | +2.108 |
Neg logit contributions | |
meg | -2.446 |
star | -2.400 |
� | -2.344 |
rovers | -2.330 |
pres | -2.320 |
Token | ) |
Token ID | 8 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ibaba | +1.315 |
emale | +1.280 |
aimon | +0.975 |
ijing | +0.959 |
WRITE | +0.945 |
Neg logit contributions | |
meg | -1.020 |
star | -0.999 |
� | -0.969 |
power | -0.965 |
mania | -0.964 |
Token | - |
Token ID | 532 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
ibaba | +4.463 |
emale | +4.297 |
WRITE | +3.331 |
aimon | +3.220 |
ijing | +3.157 |
Neg logit contributions | |
meg | -4.238 |
star | -4.116 |
rovers | -4.063 |
� | -4.032 |
pres | -4.009 |
Token | One |
Token ID | 1881 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
ibaba | +2.902 |
emale | +2.796 |
ijing | +2.100 |
WRITE | +2.095 |
aimon | +2.090 |
Neg logit contributions | |
meg | -2.434 |
star | -2.385 |
� | -2.337 |
pres | -2.302 |
power | -2.299 |
Token | needs |
Token ID | 2476 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
ibaba | +2.015 |
emale | +1.938 |
ijing | +1.538 |
aimon | +1.533 |
mustache | +1.508 |
Neg logit contributions | |
meg | -1.151 |
star | -1.126 |
� | -1.097 |
pres | -1.092 |
rovers | -1.089 |
Token | to |
Token ID | 284 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
ibaba | +0.807 |
emale | +0.781 |
aimon | +0.609 |
ijing | +0.606 |
WRITE | +0.598 |
Neg logit contributions | |
meg | -0.527 |
star | -0.512 |
� | -0.500 |
mania | -0.498 |
power | -0.492 |
Token | look |
Token ID | 804 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
ibaba | +0.711 |
emale | +0.690 |
aimon | +0.540 |
ijing | +0.534 |
WRITE | +0.531 |
Neg logit contributions | |
meg | -0.463 |
star | -0.451 |
� | -0.441 |
mania | -0.435 |
power | -0.434 |
Token | re |
Token ID | 260 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
ibaba | +0.438 |
emale | +0.431 |
aimon | +0.346 |
ijing | +0.340 |
WRITE | +0.336 |
Neg logit contributions | |
meg | -0.220 |
� | -0.210 |
star | -0.209 |
mania | -0.206 |
rovers | -0.204 |
Token | leasing |
Token ID | 48764 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
ibaba | +0.207 |
emale | +0.202 |
aimon | +0.152 |
ijing | +0.148 |
WRITE | +0.144 |
Neg logit contributions | |
meg | -0.185 |
star | -0.181 |
� | -0.180 |
rovers | -0.178 |
mania | -0.177 |
Token | factor |
Token ID | 5766 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
meg | +0.148 |
star | +0.144 |
� | +0.143 |
mania | +0.142 |
pres | +0.142 |
Neg logit contributions | |
ibaba | -0.121 |
emale | -0.117 |
aimon | -0.082 |
ijing | -0.080 |
WRITE | -0.079 |
Token | ( |
Token ID | 357 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
meg | +0.614 |
star | +0.597 |
� | +0.580 |
power | +0.575 |
pres | +0.572 |
Neg logit contributions | |
ibaba | -1.153 |
emale | -1.123 |
aimon | -0.892 |
ijing | -0.884 |
WRITE | -0.878 |
Token | CR |
Token ID | 9419 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
ibaba | +1.287 |
emale | +1.257 |
aimon | +0.983 |
ijing | +0.979 |
WRITE | +0.970 |
Neg logit contributions | |
meg | -0.804 |
star | -0.790 |
� | -0.778 |
pres | -0.762 |
mania | -0.762 |
Token | F |
Token ID | 37 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ibaba | +0.656 |
emale | +0.649 |
aimon | +0.514 |
ijing | +0.508 |
WRITE | +0.505 |
Neg logit contributions | |
meg | -0.344 |
� | -0.339 |
star | -0.332 |
rovers | -0.327 |
power | -0.323 |
Token | ) |
Token ID | 8 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
ibaba | +4.544 |
emale | +4.446 |
aimon | +3.306 |
ijing | +3.284 |
mustache | +3.206 |
Neg logit contributions | |
� | -3.864 |
meg | -3.820 |
rovers | -3.758 |
power | -3.730 |
star | -3.706 |
Token | is |
Token ID | 318 |
Feature activation | -9.26e-03 |
Pos logit contributions | |
ibaba | +2.491 |
emale | +2.396 |
aimon | +1.887 |
ijing | +1.874 |
WRITE | +1.864 |
Neg logit contributions | |
meg | -1.518 |
star | -1.483 |
� | -1.465 |
power | -1.437 |
mania | -1.433 |
Token | a |
Token ID | 257 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
ibaba | +0.040 |
emale | +0.039 |
aimon | +0.030 |
ijing | +0.030 |
WRITE | +0.029 |
Neg logit contributions | |
meg | -0.028 |
star | -0.027 |
� | -0.027 |
mania | -0.026 |
power | -0.026 |
Token | hormone |
Token ID | 17770 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
meg | +0.253 |
star | +0.247 |
� | +0.243 |
mania | +0.240 |
pres | +0.239 |
Neg logit contributions | |
ibaba | -0.332 |
emale | -0.323 |
aimon | -0.247 |
ijing | -0.243 |
WRITE | -0.241 |
Token | secret |
Token ID | 3200 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
meg | +1.125 |
star | +1.093 |
� | +1.059 |
power | +1.050 |
mania | +1.049 |
Neg logit contributions | |
ibaba | -2.054 |
emale | -2.004 |
aimon | -1.594 |
ijing | -1.573 |
WRITE | -1.558 |
Token | finish |
Token ID | 5461 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
ibaba | +1.573 |
emale | +1.520 |
aimon | +1.169 |
WRITE | +1.153 |
mustache | +1.145 |
Neg logit contributions | |
meg | -1.199 |
star | -1.169 |
� | -1.154 |
pres | -1.136 |
mania | -1.132 |
Token | the |
Token ID | 262 |
Feature activation | -9.57e-02 |
Pos logit contributions | |
meg | +0.485 |
star | +0.473 |
� | +0.457 |
power | +0.454 |
pres | +0.454 |
Neg logit contributions | |
ibaba | -0.801 |
emale | -0.782 |
aimon | -0.615 |
ijing | -0.606 |
WRITE | -0.600 |
Token | season |
Token ID | 1622 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
ibaba | +0.460 |
emale | +0.448 |
aimon | +0.360 |
ijing | +0.353 |
WRITE | +0.353 |
Neg logit contributions | |
meg | -0.238 |
star | -0.230 |
� | -0.225 |
pres | -0.222 |
mania | -0.222 |
Token | with |
Token ID | 351 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
ibaba | +0.680 |
emale | +0.663 |
aimon | +0.521 |
ijing | +0.513 |
WRITE | +0.509 |
Neg logit contributions | |
meg | -0.423 |
star | -0.411 |
� | -0.399 |
pres | -0.396 |
power | -0.395 |
Token | 10 |
Token ID | 838 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ibaba | +0.934 |
emale | +0.906 |
aimon | +0.683 |
ijing | +0.670 |
WRITE | +0.665 |
Neg logit contributions | |
meg | -0.797 |
star | -0.781 |
� | -0.763 |
pres | -0.757 |
mania | -0.755 |
Token | 26 |
Token ID | 2075 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ibaba | +0.710 |
emale | +0.684 |
aimon | +0.531 |
ijing | +0.525 |
WRITE | +0.519 |
Neg logit contributions | |
meg | -0.501 |
star | -0.489 |
� | -0.477 |
pres | -0.474 |
power | -0.470 |
Token | yards |
Token ID | 5695 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
ibaba | +4.667 |
emale | +4.423 |
aimon | +3.404 |
mustache | +3.390 |
margins | +3.331 |
Neg logit contributions | |
meg | -3.927 |
star | -3.817 |
� | -3.793 |
rovers | -3.760 |
pres | -3.686 |
Token | at |
Token ID | 379 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ibaba | +1.459 |
emale | +1.405 |
aimon | +1.125 |
WRITE | +1.105 |
mustache | +1.101 |
Neg logit contributions | |
meg | -0.880 |
star | -0.850 |
� | -0.830 |
rovers | -0.821 |
pres | -0.820 |
Token | 5 |
Token ID | 642 |
Feature activation | +9.93e-02 |
Pos logit contributions | |
meg | +0.573 |
star | +0.558 |
� | +0.546 |
pres | +0.543 |
mania | +0.541 |
Neg logit contributions | |
ibaba | -0.726 |
emale | -0.706 |
aimon | -0.540 |
ijing | -0.528 |
WRITE | -0.524 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
meg | +0.200 |
star | +0.192 |
� | +0.185 |
pres | +0.183 |
power | +0.183 |
Neg logit contributions | |
ibaba | -0.526 |
emale | -0.514 |
aimon | -0.421 |
ijing | -0.415 |
WRITE | -0.412 |
Token | 4 |
Token ID | 19 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
ibaba | +0.655 |
emale | +0.639 |
aimon | +0.505 |
ijing | +0.496 |
WRITE | +0.493 |
Neg logit contributions | |
meg | -0.389 |
star | -0.378 |
� | -0.367 |
pres | -0.364 |
power | -0.363 |
Token | example |
Token ID | 1672 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
emale | +2.496 |
ibaba | +2.488 |
aimon | +1.868 |
margins | +1.835 |
mustache | +1.822 |
Neg logit contributions | |
meg | -2.133 |
star | -2.057 |
� | -2.045 |
mania | -2.019 |
pres | -2.015 |
Token | with |
Token ID | 351 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
ibaba | +2.270 |
emale | +2.190 |
aimon | +1.750 |
ijing | +1.748 |
WRITE | +1.727 |
Neg logit contributions | |
meg | -1.310 |
star | -1.256 |
� | -1.240 |
mania | -1.221 |
rovers | -1.220 |
Token | hope |
Token ID | 2911 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
ibaba | +1.605 |
emale | +1.569 |
aimon | +1.214 |
ijing | +1.202 |
WRITE | +1.195 |
Neg logit contributions | |
meg | -1.155 |
star | -1.116 |
� | -1.103 |
mania | -1.083 |
rovers | -1.081 |
Token | , |
Token ID | 11 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
ibaba | +2.388 |
emale | +2.294 |
aimon | +1.843 |
ijing | +1.824 |
WRITE | +1.815 |
Neg logit contributions | |
meg | -1.374 |
star | -1.316 |
rovers | -1.291 |
� | -1.288 |
mania | -1.272 |
Token | never |
Token ID | 1239 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
meg | +0.226 |
star | +0.219 |
� | +0.216 |
mania | +0.212 |
pres | +0.211 |
Neg logit contributions | |
ibaba | -0.312 |
emale | -0.303 |
aimon | -0.235 |
ijing | -0.230 |
WRITE | -0.229 |
Token | fear |
Token ID | 3252 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ibaba | +0.124 |
emale | +0.121 |
aimon | +0.095 |
ijing | +0.094 |
WRITE | +0.094 |
Neg logit contributions | |
meg | -0.075 |
star | -0.073 |
� | -0.071 |
mania | -0.070 |
pres | -0.070 |
Token | ' |
Token ID | 6 |
Feature activation | -2.83e-03 |
Pos logit contributions | |
ibaba | +5.192 |
emale | +4.980 |
aimon | +3.975 |
mustache | +3.892 |
WRITE | +3.889 |
Neg logit contributions | |
meg | -3.330 |
� | -3.224 |
star | -3.203 |
rovers | -3.167 |
pres | -3.143 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
ibaba | +0.014 |
emale | +0.014 |
aimon | +0.011 |
ijing | +0.011 |
WRITE | +0.011 |
Neg logit contributions | |
meg | -0.007 |
star | -0.006 |
� | -0.006 |
mania | -0.006 |
pres | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -3.17e-03 |
Pos logit contributions | |
meg | +0.275 |
star | +0.265 |
� | +0.255 |
pres | +0.252 |
mania | +0.252 |
Neg logit contributions | |
ibaba | -0.652 |
emale | -0.637 |
aimon | -0.519 |
ijing | -0.511 |
WRITE | -0.508 |
Token | Mrs |
Token ID | 27034 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
ibaba | +0.015 |
emale | +0.015 |
aimon | +0.012 |
ijing | +0.012 |
WRITE | +0.012 |
Neg logit contributions | |
meg | -0.008 |
star | -0.007 |
� | -0.007 |
mania | -0.007 |
rovers | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
meg | +0.349 |
star | +0.336 |
� | +0.315 |
power | +0.314 |
mania | +0.313 |
Neg logit contributions | |
ibaba | -1.140 |
emale | -1.119 |
aimon | -0.924 |
ijing | -0.913 |
WRITE | -0.904 |
Token | , |
Token ID | 11 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
ibaba | +0.129 |
emale | +0.126 |
aimon | +0.100 |
ijing | +0.099 |
WRITE | +0.098 |
Neg logit contributions | |
meg | -0.066 |
star | -0.064 |
� | -0.062 |
mania | -0.061 |
pres | -0.061 |
Token | 000 |
Token ID | 830 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
meg | +0.879 |
star | +0.858 |
� | +0.828 |
power | +0.824 |
pres | +0.822 |
Neg logit contributions | |
ibaba | -1.361 |
emale | -1.329 |
aimon | -1.038 |
ijing | -1.021 |
WRITE | -1.012 |
Token | won |
Token ID | 1839 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
meg | +0.320 |
star | +0.312 |
� | +0.301 |
power | +0.300 |
pres | +0.300 |
Neg logit contributions | |
ibaba | -0.501 |
emale | -0.488 |
aimon | -0.383 |
ijing | -0.375 |
WRITE | -0.372 |
Token | ($ |
Token ID | 7198 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
meg | +1.924 |
star | +1.885 |
power | +1.824 |
� | +1.816 |
pres | +1.814 |
Neg logit contributions | |
ibaba | -2.776 |
emale | -2.729 |
aimon | -2.108 |
ijing | -2.052 |
WRITE | -2.033 |
Token | 8 |
Token ID | 23 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
meg | +0.534 |
star | +0.521 |
� | +0.503 |
power | +0.499 |
pres | +0.498 |
Neg logit contributions | |
ibaba | -0.927 |
emale | -0.904 |
aimon | -0.716 |
ijing | -0.705 |
mustache | -0.696 |
Token | 52 |
Token ID | 4309 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
ibaba | +0.921 |
emale | +0.899 |
aimon | +0.710 |
ijing | +0.700 |
WRITE | +0.693 |
Neg logit contributions | |
meg | -0.533 |
star | -0.520 |
� | -0.502 |
power | -0.498 |
pres | -0.498 |
Token | . |
Token ID | 13 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
ibaba | +4.896 |
emale | +4.697 |
ijing | +3.608 |
aimon | +3.582 |
WRITE | +3.542 |
Neg logit contributions | |
meg | -3.657 |
star | -3.626 |
mania | -3.528 |
rovers | -3.524 |
� | -3.514 |
Token | 54 |
Token ID | 4051 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
ibaba | +0.747 |
emale | +0.729 |
aimon | +0.574 |
ijing | +0.566 |
WRITE | +0.559 |
Neg logit contributions | |
meg | -0.446 |
star | -0.435 |
� | -0.420 |
power | -0.417 |
pres | -0.417 |
Token | ) |
Token ID | 8 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
ibaba | +2.136 |
emale | +2.059 |
aimon | +1.540 |
ijing | +1.538 |
WRITE | +1.521 |
Neg logit contributions | |
meg | -1.824 |
star | -1.797 |
� | -1.752 |
rovers | -1.749 |
mania | -1.745 |
Token | smartphone |
Token ID | 11745 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
meg | +1.383 |
star | +1.355 |
power | +1.311 |
pres | +1.305 |
� | +1.302 |
Neg logit contributions | |
ibaba | -1.891 |
emale | -1.859 |
aimon | -1.428 |
ijing | -1.393 |
WRITE | -1.376 |
Token | . |
Token ID | 13 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
ibaba | +0.992 |
emale | +0.966 |
aimon | +0.775 |
ijing | +0.767 |
WRITE | +0.762 |
Neg logit contributions | |
meg | -0.486 |
star | -0.471 |
� | -0.456 |
pres | -0.450 |
mania | -0.449 |
Token | Ol |
Token ID | 6544 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
ibaba | +0.379 |
emale | +0.369 |
aimon | +0.289 |
ijing | +0.284 |
WRITE | +0.281 |
Neg logit contributions | |
meg | -0.244 |
star | -0.238 |
� | -0.230 |
power | -0.229 |
pres | -0.229 |
Token | af |
Token ID | 1878 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
ibaba | +1.402 |
emale | +1.368 |
aimon | +1.125 |
ijing | +1.099 |
WRITE | +1.087 |
Neg logit contributions | |
meg | -0.542 |
� | -0.530 |
star | -0.525 |
power | -0.509 |
mania | -0.507 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
ibaba | +2.552 |
emale | +2.499 |
aimon | +2.068 |
mustache | +2.025 |
ijing | +2.023 |
Neg logit contributions | |
meg | -0.933 |
star | -0.875 |
� | -0.873 |
rovers | -0.852 |
power | -0.845 |
Token | I |
Token ID | 314 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
meg | +0.842 |
star | +0.819 |
� | +0.791 |
power | +0.785 |
mania | +0.784 |
Neg logit contributions | |
ibaba | -1.518 |
emale | -1.482 |
aimon | -1.176 |
ijing | -1.158 |
WRITE | -1.149 |
Token | stood |
Token ID | 6204 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
meg | +1.141 |
star | +1.115 |
� | +1.071 |
power | +1.069 |
pres | +1.068 |
Neg logit contributions | |
ibaba | -1.906 |
emale | -1.863 |
aimon | -1.467 |
ijing | -1.444 |
WRITE | -1.424 |
Token | still |
Token ID | 991 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
meg | +0.806 |
star | +0.782 |
� | +0.765 |
mania | +0.752 |
pres | +0.752 |
Neg logit contributions | |
ibaba | -1.415 |
emale | -1.381 |
aimon | -1.094 |
ijing | -1.079 |
WRITE | -1.073 |
Token | . |
Token ID | 13 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
meg | +2.042 |
star | +1.996 |
power | +1.908 |
mania | +1.904 |
pres | +1.900 |
Neg logit contributions | |
ibaba | -4.114 |
emale | -4.031 |
aimon | -3.225 |
ijing | -3.178 |
WRITE | -3.133 |
Token |
|
Token ID | 198 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
meg | +0.565 |
star | +0.551 |
� | +0.532 |
power | +0.528 |
pres | +0.528 |
Neg logit contributions | |
ibaba | -0.978 |
emale | -0.955 |
aimon | -0.755 |
ijing | -0.743 |
WRITE | -0.736 |
Token |
|
Token ID | 198 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
meg | +0.737 |
star | +0.716 |
power | +0.678 |
� | +0.677 |
mania | +0.677 |
Neg logit contributions | |
ibaba | -1.866 |
emale | -1.834 |
aimon | -1.487 |
ijing | -1.472 |
WRITE | -1.452 |
Token | My |
Token ID | 3666 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
meg | +0.568 |
star | +0.544 |
� | +0.523 |
mania | +0.512 |
rovers | +0.510 |
Neg logit contributions | |
ibaba | -1.594 |
emale | -1.561 |
aimon | -1.286 |
ijing | -1.266 |
WRITE | -1.260 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
ibaba | +0.409 |
emale | +0.397 |
aimon | +0.302 |
ijing | +0.295 |
mustache | +0.293 |
Neg logit contributions | |
meg | -0.340 |
star | -0.332 |
� | -0.325 |
pres | -0.322 |
mania | -0.322 |
Token | election |
Token ID | 3071 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
ibaba | +0.658 |
emale | +0.638 |
aimon | +0.492 |
ijing | +0.487 |
WRITE | +0.483 |
Neg logit contributions | |
meg | -0.472 |
star | -0.460 |
� | -0.451 |
pres | -0.445 |
mania | -0.444 |
Token | approaches |
Token ID | 10581 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
meg | +0.175 |
star | +0.171 |
� | +0.166 |
power | +0.165 |
mania | +0.165 |
Neg logit contributions | |
ibaba | -0.235 |
emale | -0.229 |
aimon | -0.175 |
ijing | -0.172 |
WRITE | -0.170 |
Token | . |
Token ID | 13 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
ibaba | +4.309 |
emale | +4.195 |
aimon | +3.416 |
ijing | +3.391 |
WRITE | +3.381 |
Neg logit contributions | |
meg | -1.869 |
star | -1.821 |
� | -1.792 |
power | -1.737 |
mania | -1.737 |
Token | For |
Token ID | 1114 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
ibaba | +0.161 |
emale | +0.157 |
aimon | +0.124 |
ijing | +0.122 |
WRITE | +0.122 |
Neg logit contributions | |
meg | -0.095 |
star | -0.092 |
� | -0.089 |
power | -0.088 |
pres | -0.088 |
Token | a |
Token ID | 257 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
meg | +0.901 |
star | +0.880 |
� | +0.853 |
power | +0.846 |
pres | +0.846 |
Neg logit contributions | |
ibaba | -1.375 |
emale | -1.341 |
aimon | -1.045 |
ijing | -1.029 |
WRITE | -1.019 |
Token | few |
Token ID | 1178 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
meg | +2.508 |
star | +2.445 |
power | +2.343 |
pres | +2.338 |
mania | +2.328 |
Neg logit contributions | |
ibaba | -4.510 |
emale | -4.410 |
aimon | -3.514 |
ijing | -3.442 |
WRITE | -3.413 |
Token | years |
Token ID | 812 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
meg | +2.279 |
star | +2.218 |
� | +2.123 |
pres | +2.121 |
power | +2.117 |
Neg logit contributions | |
ibaba | -4.291 |
emale | -4.190 |
aimon | -3.350 |
ijing | -3.277 |
WRITE | -3.259 |
Token | , |
Token ID | 11 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
meg | +2.032 |
star | +1.975 |
� | +1.906 |
power | +1.889 |
mania | +1.887 |
Neg logit contributions | |
ibaba | -3.873 |
emale | -3.781 |
aimon | -3.019 |
ijing | -2.979 |
WRITE | -2.950 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
meg | +0.772 |
star | +0.755 |
� | +0.733 |
power | +0.727 |
mania | +0.726 |
Neg logit contributions | |
ibaba | -1.105 |
emale | -1.076 |
aimon | -0.833 |
ijing | -0.821 |
WRITE | -0.813 |
Token | economy |
Token ID | 3773 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
meg | +0.033 |
star | +0.032 |
� | +0.031 |
power | +0.031 |
mania | +0.031 |
Neg logit contributions | |
ibaba | -0.044 |
emale | -0.043 |
aimon | -0.033 |
ijing | -0.032 |
WRITE | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
ibaba | +1.206 |
emale | +1.169 |
aimon | +0.924 |
ijing | +0.920 |
mustache | +0.919 |
Neg logit contributions | |
meg | -0.723 |
star | -0.703 |
� | -0.695 |
rovers | -0.679 |
pres | -0.675 |
Token | I |
Token ID | 40 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ibaba | +0.777 |
emale | +0.761 |
aimon | +0.608 |
ijing | +0.599 |
WRITE | +0.595 |
Neg logit contributions | |
meg | -0.409 |
star | -0.395 |
� | -0.386 |
power | -0.379 |
rovers | -0.379 |
Token | spent |
Token ID | 3377 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
meg | +0.400 |
star | +0.388 |
� | +0.374 |
power | +0.371 |
pres | +0.371 |
Neg logit contributions | |
ibaba | -0.779 |
emale | -0.762 |
aimon | -0.609 |
ijing | -0.600 |
WRITE | -0.595 |
Token | a |
Token ID | 257 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
meg | +0.777 |
star | +0.758 |
� | +0.730 |
power | +0.728 |
mania | +0.727 |
Neg logit contributions | |
ibaba | -1.292 |
emale | -1.261 |
aimon | -0.992 |
ijing | -0.977 |
WRITE | -0.966 |
Token | long |
Token ID | 890 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
meg | +1.254 |
star | +1.220 |
� | +1.179 |
power | +1.170 |
pres | +1.169 |
Neg logit contributions | |
ibaba | -2.219 |
emale | -2.167 |
aimon | -1.717 |
ijing | -1.690 |
WRITE | -1.675 |
Token | time |
Token ID | 640 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
meg | +1.048 |
star | +1.013 |
� | +0.971 |
power | +0.961 |
mania | +0.961 |
Neg logit contributions | |
ibaba | -2.531 |
emale | -2.478 |
aimon | -2.013 |
ijing | -1.987 |
WRITE | -1.971 |
Token | looking |
Token ID | 2045 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
meg | +0.738 |
star | +0.720 |
� | +0.698 |
power | +0.694 |
pres | +0.693 |
Neg logit contributions | |
ibaba | -1.075 |
emale | -1.047 |
aimon | -0.813 |
ijing | -0.800 |
WRITE | -0.792 |
Token | for |
Token ID | 329 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
meg | +2.020 |
star | +1.956 |
� | +1.889 |
mania | +1.868 |
pres | +1.867 |
Neg logit contributions | |
ibaba | -4.045 |
emale | -3.949 |
aimon | -3.164 |
ijing | -3.128 |
WRITE | -3.099 |
Token | what |
Token ID | 644 |
Feature activation | -8.82e-02 |
Pos logit contributions | |
ibaba | +1.772 |
emale | +1.721 |
aimon | +1.320 |
ijing | +1.308 |
mustache | +1.304 |
Neg logit contributions | |
meg | -1.299 |
star | -1.263 |
� | -1.250 |
pres | -1.225 |
power | -1.221 |
Token | used |
Token ID | 973 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
ibaba | +0.403 |
emale | +0.392 |
aimon | +0.309 |
ijing | +0.306 |
mustache | +0.303 |
Neg logit contributions | |
meg | -0.243 |
star | -0.235 |
� | -0.230 |
pres | -0.226 |
power | -0.225 |
Token | to |
Token ID | 284 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ibaba | +0.545 |
emale | +0.533 |
aimon | +0.428 |
ijing | +0.421 |
WRITE | +0.418 |
Neg logit contributions | |
meg | -0.270 |
star | -0.262 |
� | -0.252 |
power | -0.250 |
mania | -0.250 |
Token | be |
Token ID | 307 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
meg | +0.457 |
star | +0.447 |
� | +0.442 |
mania | +0.435 |
power | +0.431 |
Neg logit contributions | |
ibaba | -0.605 |
emale | -0.592 |
aimon | -0.452 |
ijing | -0.445 |
mustache | -0.444 |
Token |
|
Token ID | 198 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
ibaba | +1.085 |
emale | +1.049 |
aimon | +0.860 |
WRITE | +0.848 |
ijing | +0.841 |
Neg logit contributions | |
meg | -0.498 |
star | -0.478 |
� | -0.468 |
rovers | -0.458 |
pres | -0.455 |
Token | The |
Token ID | 464 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
ibaba | +1.227 |
emale | +1.197 |
aimon | +0.972 |
WRITE | +0.946 |
ijing | +0.945 |
Neg logit contributions | |
meg | -0.614 |
star | -0.590 |
� | -0.579 |
rovers | -0.570 |
mania | -0.566 |
Token | trade |
Token ID | 3292 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
ibaba | +0.711 |
emale | +0.689 |
aimon | +0.525 |
ijing | +0.513 |
WRITE | +0.512 |
Neg logit contributions | |
meg | -0.580 |
star | -0.565 |
� | -0.555 |
pres | -0.549 |
mania | -0.548 |
Token | also |
Token ID | 635 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
meg | +0.166 |
star | +0.161 |
� | +0.157 |
pres | +0.155 |
power | +0.154 |
Neg logit contributions | |
ibaba | -0.276 |
emale | -0.268 |
aimon | -0.213 |
ijing | -0.208 |
WRITE | -0.208 |
Token | reun |
Token ID | 15398 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
meg | +0.669 |
star | +0.653 |
� | +0.629 |
power | +0.627 |
pres | +0.627 |
Neg logit contributions | |
ibaba | -1.090 |
emale | -1.065 |
aimon | -0.836 |
ijing | -0.823 |
WRITE | -0.814 |
Token | ites |
Token ID | 2737 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
meg | +1.082 |
star | +1.050 |
� | +0.994 |
power | +0.994 |
pres | +0.990 |
Neg logit contributions | |
ibaba | -2.663 |
emale | -2.607 |
aimon | -2.120 |
ijing | -2.093 |
WRITE | -2.077 |
Token | Fel |
Token ID | 13937 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
meg | +2.136 |
star | +2.097 |
power | +2.016 |
mania | +2.013 |
� | +2.012 |
Neg logit contributions | |
ibaba | -3.052 |
emale | -2.978 |
aimon | -2.296 |
ijing | -2.266 |
WRITE | -2.227 |
Token | ton |
Token ID | 1122 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
meg | +0.508 |
star | +0.490 |
power | +0.460 |
pres | +0.460 |
mania | +0.459 |
Neg logit contributions | |
ibaba | -1.460 |
emale | -1.432 |
aimon | -1.174 |
ijing | -1.162 |
WRITE | -1.153 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ibaba | +1.038 |
emale | +1.009 |
aimon | +0.801 |
ijing | +0.781 |
WRITE | +0.773 |
Neg logit contributions | |
meg | -0.624 |
star | -0.602 |
� | -0.592 |
power | -0.584 |
rovers | -0.583 |
Token | Thomas |
Token ID | 5658 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
meg | +0.434 |
star | +0.421 |
� | +0.412 |
pres | +0.407 |
power | +0.406 |
Neg logit contributions | |
ibaba | -0.633 |
emale | -0.616 |
aimon | -0.481 |
ijing | -0.470 |
WRITE | -0.469 |
Token | with |
Token ID | 351 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
ibaba | +0.520 |
emale | +0.504 |
aimon | +0.403 |
ijing | +0.394 |
WRITE | +0.392 |
Neg logit contributions | |
meg | -0.292 |
star | -0.284 |
� | -0.275 |
pres | -0.272 |
rovers | -0.272 |
Token | go |
Token ID | 467 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
meg | +0.635 |
star | +0.618 |
� | +0.601 |
mania | +0.592 |
power | +0.591 |
Neg logit contributions | |
ibaba | -1.170 |
emale | -1.142 |
aimon | -0.909 |
ijing | -0.895 |
WRITE | -0.891 |
Token | into |
Token ID | 656 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
meg | +0.122 |
star | +0.119 |
� | +0.115 |
pres | +0.114 |
mania | +0.113 |
Neg logit contributions | |
ibaba | -0.225 |
emale | -0.219 |
aimon | -0.175 |
ijing | -0.173 |
WRITE | -0.171 |
Token | the |
Token ID | 262 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
meg | +0.965 |
star | +0.943 |
� | +0.912 |
pres | +0.903 |
power | +0.900 |
Neg logit contributions | |
ibaba | -1.675 |
emale | -1.631 |
aimon | -1.293 |
ijing | -1.273 |
WRITE | -1.260 |
Token | details |
Token ID | 3307 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
meg | +0.110 |
star | +0.108 |
� | +0.105 |
pres | +0.104 |
mania | +0.104 |
Neg logit contributions | |
ibaba | -0.160 |
emale | -0.156 |
aimon | -0.121 |
ijing | -0.119 |
WRITE | -0.118 |
Token | of |
Token ID | 286 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
meg | +0.849 |
star | +0.826 |
� | +0.794 |
mania | +0.789 |
pres | +0.789 |
Neg logit contributions | |
ibaba | -1.565 |
emale | -1.523 |
aimon | -1.217 |
ijing | -1.203 |
WRITE | -1.191 |
Token | sound |
Token ID | 2128 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
ibaba | +0.457 |
emale | +0.443 |
aimon | +0.345 |
ijing | +0.339 |
WRITE | +0.336 |
Neg logit contributions | |
meg | -0.317 |
star | -0.312 |
� | -0.306 |
pres | -0.300 |
rovers | -0.300 |
Token | commercial |
Token ID | 5068 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
meg | +2.092 |
star | +2.044 |
� | +1.976 |
power | +1.969 |
mania | +1.967 |
Neg logit contributions | |
ibaba | -3.102 |
emale | -3.026 |
aimon | -2.352 |
ijing | -2.310 |
WRITE | -2.291 |
Token | banking |
Token ID | 11754 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
meg | +0.343 |
star | +0.332 |
� | +0.329 |
pres | +0.322 |
rovers | +0.322 |
Neg logit contributions | |
ibaba | -0.496 |
emale | -0.478 |
aimon | -0.371 |
ijing | -0.368 |
margins | -0.364 |
Token | , |
Token ID | 11 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
meg | +0.573 |
star | +0.556 |
� | +0.542 |
pres | +0.528 |
power | +0.526 |
Neg logit contributions | |
ibaba | -1.075 |
emale | -1.045 |
aimon | -0.836 |
ijing | -0.829 |
WRITE | -0.817 |
Token | how |
Token ID | 703 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
ibaba | +0.869 |
emale | +0.843 |
aimon | +0.659 |
ijing | +0.653 |
mustache | +0.646 |
Neg logit contributions | |
meg | -0.551 |
star | -0.537 |
� | -0.528 |
mania | -0.517 |
pres | -0.516 |
Token | it |
Token ID | 340 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
ibaba | +1.585 |
emale | +1.525 |
aimon | +1.201 |
ijing | +1.196 |
WRITE | +1.188 |
Neg logit contributions | |
meg | -0.995 |
star | -0.960 |
� | -0.959 |
rovers | -0.934 |
pres | -0.928 |
Token | Center |
Token ID | 3337 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
meg | +0.711 |
star | +0.690 |
� | +0.677 |
mania | +0.668 |
pres | +0.667 |
Neg logit contributions | |
ibaba | -0.901 |
emale | -0.878 |
aimon | -0.664 |
ijing | -0.656 |
WRITE | -0.651 |
Token | History |
Token ID | 7443 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
meg | +0.041 |
star | +0.040 |
� | +0.039 |
pres | +0.039 |
mania | +0.038 |
Neg logit contributions | |
ibaba | -0.066 |
emale | -0.065 |
aimon | -0.051 |
ijing | -0.050 |
WRITE | -0.050 |
Token | Office |
Token ID | 4452 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
ibaba | +0.140 |
emale | +0.136 |
aimon | +0.106 |
ijing | +0.105 |
WRITE | +0.105 |
Neg logit contributions | |
meg | -0.093 |
star | -0.090 |
� | -0.088 |
mania | -0.087 |
pres | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
meg | +0.880 |
star | +0.852 |
� | +0.829 |
pres | +0.822 |
mania | +0.822 |
Neg logit contributions | |
ibaba | -1.483 |
emale | -1.449 |
aimon | -1.141 |
ijing | -1.127 |
WRITE | -1.120 |
Token | 1997 |
Token ID | 8309 |
Feature activation | +4.85e-03 |
Pos logit contributions | |
meg | +0.849 |
star | +0.829 |
� | +0.794 |
power | +0.790 |
pres | +0.789 |
Neg logit contributions | |
ibaba | -1.669 |
emale | -1.630 |
aimon | -1.303 |
ijing | -1.284 |
WRITE | -1.268 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
meg | +0.010 |
star | +0.009 |
� | +0.009 |
mania | +0.009 |
pres | +0.009 |
Neg logit contributions | |
ibaba | -0.026 |
emale | -0.025 |
aimon | -0.021 |
ijing | -0.020 |
WRITE | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
meg | +0.377 |
star | +0.362 |
� | +0.351 |
mania | +0.350 |
pres | +0.348 |
Neg logit contributions | |
ibaba | -0.769 |
emale | -0.750 |
aimon | -0.601 |
ijing | -0.594 |
WRITE | -0.593 |
Token |
|
Token ID | 198 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
ibaba | +1.272 |
emale | +1.228 |
aimon | +1.006 |
WRITE | +0.994 |
ijing | +0.985 |
Neg logit contributions | |
meg | -0.595 |
star | -0.571 |
� | -0.563 |
rovers | -0.549 |
pres | -0.545 |
Token | � |
Token ID | 447 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
ibaba | +1.479 |
emale | +1.453 |
aimon | +1.160 |
ijing | +1.147 |
WRITE | +1.139 |
Neg logit contributions | |
meg | -0.795 |
star | -0.759 |
� | -0.750 |
mania | -0.744 |
rovers | -0.735 |
Token | � |
Token ID | 250 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
ibaba | +1.348 |
emale | +1.289 |
ijing | +0.952 |
aimon | +0.950 |
WRITE | +0.934 |
Neg logit contributions | |
meg | -1.287 |
star | -1.257 |
� | -1.226 |
mania | -1.225 |
pres | -1.221 |
Token | B |
Token ID | 33 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
meg | +0.088 |
star | +0.084 |
� | +0.082 |
mania | +0.081 |
power | +0.080 |
Neg logit contributions | |
ibaba | -0.173 |
emale | -0.169 |
aimon | -0.135 |
ijing | -0.134 |
WRITE | -0.132 |
Token | ent |
Token ID | 298 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
meg | +0.524 |
star | +0.510 |
� | +0.491 |
mania | +0.487 |
pres | +0.486 |
Neg logit contributions | |
ibaba | -1.053 |
emale | -1.030 |
aimon | -0.825 |
ijing | -0.813 |
WRITE | -0.806 |
Token | access |
Token ID | 1895 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
meg | +0.053 |
star | +0.052 |
� | +0.050 |
power | +0.050 |
pres | +0.050 |
Neg logit contributions | |
ibaba | -0.067 |
emale | -0.066 |
aimon | -0.050 |
ijing | -0.049 |
WRITE | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
ibaba | +0.902 |
emale | +0.869 |
aimon | +0.701 |
ijing | +0.694 |
WRITE | +0.680 |
Neg logit contributions | |
meg | -0.474 |
star | -0.468 |
� | -0.455 |
mania | -0.447 |
power | -0.442 |
Token | this |
Token ID | 428 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
ibaba | +1.081 |
emale | +1.046 |
aimon | +0.802 |
ijing | +0.798 |
WRITE | +0.794 |
Neg logit contributions | |
meg | -0.810 |
star | -0.796 |
� | -0.784 |
rovers | -0.771 |
mania | -0.768 |
Token | illegal |
Token ID | 5293 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
ibaba | +0.840 |
emale | +0.804 |
aimon | +0.608 |
ijing | +0.599 |
WRITE | +0.595 |
Neg logit contributions | |
meg | -0.719 |
star | -0.707 |
� | -0.694 |
mania | -0.685 |
rovers | -0.683 |
Token | content |
Token ID | 2695 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
meg | +1.461 |
star | +1.432 |
� | +1.389 |
power | +1.389 |
pres | +1.386 |
Neg logit contributions | |
ibaba | -1.499 |
emale | -1.463 |
aimon | -1.076 |
ijing | -1.053 |
WRITE | -1.035 |
Token | . |
Token ID | 13 |
Feature activation | -9.32e-02 |
Pos logit contributions | |
meg | +0.127 |
star | +0.123 |
� | +0.118 |
mania | +0.117 |
power | +0.116 |
Neg logit contributions | |
ibaba | -0.322 |
emale | -0.312 |
aimon | -0.257 |
ijing | -0.255 |
WRITE | -0.251 |
Token | � |
Token ID | 447 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
ibaba | +0.439 |
emale | +0.428 |
aimon | +0.340 |
WRITE | +0.337 |
ijing | +0.336 |
Neg logit contributions | |
meg | -0.239 |
star | -0.233 |
� | -0.225 |
mania | -0.223 |
power | -0.222 |
Token | � |
Token ID | 251 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
meg | +0.462 |
star | +0.452 |
mania | +0.440 |
� | +0.439 |
pres | +0.439 |
Neg logit contributions | |
ibaba | -0.496 |
emale | -0.476 |
aimon | -0.354 |
ijing | -0.349 |
WRITE | -0.348 |
Token |
|
Token ID | 198 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
ibaba | +0.740 |
emale | +0.719 |
aimon | +0.576 |
WRITE | +0.573 |
ijing | +0.571 |
Neg logit contributions | |
meg | -0.375 |
star | -0.364 |
� | -0.351 |
mania | -0.349 |
rovers | -0.347 |
Token |
|
Token ID | 198 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
ibaba | +1.151 |
emale | +1.114 |
aimon | +0.912 |
WRITE | +0.901 |
ijing | +0.892 |
Neg logit contributions | |
meg | -0.534 |
star | -0.512 |
� | -0.503 |
rovers | -0.492 |
pres | -0.489 |
Token | much |
Token ID | 881 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
ibaba | +0.203 |
emale | +0.197 |
aimon | +0.154 |
mustache | +0.151 |
ijing | +0.151 |
Neg logit contributions | |
meg | -0.138 |
star | -0.135 |
� | -0.132 |
pres | -0.131 |
power | -0.130 |
Token | as |
Token ID | 355 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
ibaba | +0.460 |
emale | +0.447 |
aimon | +0.362 |
ijing | +0.356 |
mustache | +0.355 |
Neg logit contributions | |
meg | -0.223 |
star | -0.215 |
� | -0.211 |
pres | -0.209 |
power | -0.207 |
Token | I |
Token ID | 314 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
ibaba | +1.312 |
emale | +1.275 |
aimon | +1.022 |
ijing | +1.007 |
WRITE | +1.006 |
Neg logit contributions | |
meg | -0.702 |
star | -0.681 |
� | -0.665 |
pres | -0.655 |
power | -0.651 |
Token | do |
Token ID | 466 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
ibaba | +0.718 |
emale | +0.691 |
aimon | +0.523 |
ijing | +0.517 |
WRITE | +0.514 |
Neg logit contributions | |
meg | -0.612 |
star | -0.597 |
� | -0.588 |
pres | -0.579 |
mania | -0.575 |
Token | , |
Token ID | 11 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
ibaba | +0.937 |
emale | +0.914 |
aimon | +0.742 |
ijing | +0.734 |
WRITE | +0.730 |
Neg logit contributions | |
meg | -0.404 |
star | -0.390 |
� | -0.382 |
pres | -0.378 |
power | -0.373 |
Token | but |
Token ID | 475 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
meg | +0.482 |
star | +0.470 |
� | +0.455 |
power | +0.451 |
mania | +0.451 |
Neg logit contributions | |
ibaba | -0.795 |
emale | -0.775 |
aimon | -0.610 |
ijing | -0.600 |
WRITE | -0.595 |
Token | I |
Token ID | 314 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
meg | +0.175 |
star | +0.170 |
� | +0.164 |
pres | +0.163 |
power | +0.163 |
Neg logit contributions | |
ibaba | -0.316 |
emale | -0.309 |
aimon | -0.245 |
ijing | -0.241 |
WRITE | -0.239 |
Token | think |
Token ID | 892 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
meg | +0.253 |
star | +0.246 |
� | +0.239 |
pres | +0.236 |
mania | +0.236 |
Neg logit contributions | |
ibaba | -0.461 |
emale | -0.449 |
aimon | -0.357 |
ijing | -0.352 |
WRITE | -0.350 |
Token | we |
Token ID | 356 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
ibaba | +1.028 |
emale | +1.000 |
aimon | +0.795 |
ijing | +0.783 |
mustache | +0.776 |
Neg logit contributions | |
meg | -0.577 |
star | -0.559 |
� | -0.545 |
pres | -0.540 |
mania | -0.538 |
Token | can |
Token ID | 460 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
ibaba | +0.144 |
emale | +0.140 |
aimon | +0.111 |
ijing | +0.109 |
WRITE | +0.108 |
Neg logit contributions | |
meg | -0.084 |
star | -0.082 |
� | -0.080 |
pres | -0.079 |
mania | -0.079 |
Token | win |
Token ID | 1592 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
ibaba | +0.263 |
emale | +0.255 |
aimon | +0.195 |
ijing | +0.193 |
WRITE | +0.191 |
Neg logit contributions | |
meg | -0.197 |
star | -0.193 |
� | -0.188 |
pres | -0.186 |
mania | -0.186 |
Token | can |
Token ID | 460 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
meg | +0.374 |
star | +0.364 |
� | +0.351 |
power | +0.349 |
pres | +0.349 |
Neg logit contributions | |
ibaba | -0.677 |
emale | -0.662 |
aimon | -0.525 |
ijing | -0.518 |
WRITE | -0.513 |
Token | use |
Token ID | 779 |
Feature activation | +9.70e-03 |
Pos logit contributions | |
meg | +0.043 |
star | +0.041 |
� | +0.040 |
mania | +0.040 |
pres | +0.040 |
Neg logit contributions | |
ibaba | -0.077 |
emale | -0.075 |
aimon | -0.059 |
WRITE | -0.059 |
ijing | -0.059 |
Token | this |
Token ID | 428 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
meg | +0.026 |
star | +0.025 |
� | +0.025 |
mania | +0.024 |
pres | +0.024 |
Neg logit contributions | |
ibaba | -0.045 |
emale | -0.043 |
aimon | -0.034 |
WRITE | -0.034 |
ijing | -0.034 |
Token | on |
Token ID | 319 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
meg | +0.140 |
star | +0.137 |
� | +0.133 |
mania | +0.131 |
pres | +0.130 |
Neg logit contributions | |
ibaba | -0.259 |
emale | -0.252 |
aimon | -0.202 |
WRITE | -0.199 |
ijing | -0.198 |
Token | your |
Token ID | 534 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
meg | +0.483 |
star | +0.471 |
� | +0.459 |
mania | +0.455 |
pres | +0.454 |
Neg logit contributions | |
ibaba | -0.660 |
emale | -0.642 |
aimon | -0.494 |
ijing | -0.486 |
WRITE | -0.483 |
Token | opponent |
Token ID | 6125 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
meg | +0.469 |
star | +0.459 |
� | +0.449 |
mania | +0.445 |
power | +0.445 |
Neg logit contributions | |
ibaba | -0.479 |
emale | -0.465 |
aimon | -0.343 |
ijing | -0.336 |
WRITE | -0.333 |
Token | � |
Token ID | 447 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
meg | +1.424 |
star | +1.386 |
� | +1.327 |
power | +1.324 |
pres | +1.320 |
Neg logit contributions | |
ibaba | -2.892 |
emale | -2.831 |
aimon | -2.270 |
ijing | -2.236 |
WRITE | -2.216 |
Token | � |
Token ID | 247 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
ibaba | +0.878 |
emale | +0.854 |
aimon | +0.665 |
ijing | +0.655 |
WRITE | +0.648 |
Neg logit contributions | |
meg | -0.588 |
star | -0.574 |
� | -0.556 |
mania | -0.553 |
pres | -0.552 |
Token | s |
Token ID | 82 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
ibaba | +0.240 |
emale | +0.236 |
aimon | +0.196 |
ijing | +0.194 |
WRITE | +0.192 |
Neg logit contributions | |
meg | -0.060 |
star | -0.056 |
� | -0.055 |
mania | -0.052 |
rovers | -0.052 |
Token | high |
Token ID | 1029 |
Feature activation | +6.84e-02 |
Pos logit contributions | |
ibaba | +0.103 |
emale | +0.101 |
aimon | +0.080 |
ijing | +0.078 |
WRITE | +0.078 |
Neg logit contributions | |
meg | -0.059 |
star | -0.058 |
� | -0.056 |
pres | -0.055 |
mania | -0.055 |
Token | Level |
Token ID | 5684 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
meg | +0.230 |
star | +0.225 |
� | +0.219 |
pres | +0.218 |
mania | +0.218 |
Neg logit contributions | |
ibaba | -0.269 |
emale | -0.262 |
aimon | -0.198 |
ijing | -0.194 |
WRITE | -0.193 |
Token | have |
Token ID | 423 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
meg | +0.476 |
star | +0.464 |
� | +0.447 |
power | +0.447 |
mania | +0.445 |
Neg logit contributions | |
ibaba | -0.770 |
emale | -0.752 |
aimon | -0.589 |
ijing | -0.580 |
WRITE | -0.573 |
Token | to |
Token ID | 284 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
meg | +0.874 |
star | +0.851 |
� | +0.824 |
pres | +0.817 |
power | +0.816 |
Neg logit contributions | |
ibaba | -1.463 |
emale | -1.426 |
aimon | -1.125 |
ijing | -1.108 |
WRITE | -1.097 |
Token | learn |
Token ID | 2193 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
meg | +0.679 |
star | +0.664 |
� | +0.641 |
power | +0.641 |
pres | +0.640 |
Neg logit contributions | |
ibaba | -0.980 |
emale | -0.956 |
aimon | -0.740 |
ijing | -0.727 |
WRITE | -0.718 |
Token | to |
Token ID | 284 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
meg | +0.611 |
star | +0.595 |
� | +0.575 |
power | +0.572 |
pres | +0.571 |
Neg logit contributions | |
ibaba | -1.033 |
emale | -1.010 |
aimon | -0.796 |
ijing | -0.783 |
WRITE | -0.776 |
Token | overtake |
Token ID | 48155 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
ibaba | +1.680 |
emale | +1.626 |
aimon | +1.256 |
ijing | +1.235 |
WRITE | +1.234 |
Neg logit contributions | |
meg | -1.235 |
star | -1.202 |
� | -1.182 |
mania | -1.159 |
pres | -1.159 |
Token | . |
Token ID | 13 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
meg | +0.311 |
star | +0.301 |
� | +0.293 |
pres | +0.286 |
mania | +0.286 |
Neg logit contributions | |
ibaba | -0.626 |
emale | -0.609 |
aimon | -0.491 |
ijing | -0.485 |
WRITE | -0.481 |
Token | The |
Token ID | 383 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
meg | +0.810 |
star | +0.797 |
mania | +0.768 |
power | +0.768 |
pres | +0.761 |
Neg logit contributions | |
ibaba | -1.359 |
emale | -1.334 |
aimon | -1.043 |
ijing | -1.031 |
mustache | -1.010 |
Token | race |
Token ID | 3234 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
meg | +0.491 |
star | +0.481 |
power | +0.465 |
mania | +0.464 |
� | +0.463 |
Neg logit contributions | |
ibaba | -0.671 |
emale | -0.655 |
aimon | -0.503 |
ijing | -0.493 |
WRITE | -0.487 |
Token | may |
Token ID | 743 |
Feature activation | +8.27e-02 |
Pos logit contributions | |
ibaba | +1.216 |
emale | +1.171 |
aimon | +0.912 |
ijing | +0.896 |
WRITE | +0.886 |
Neg logit contributions | |
meg | -0.881 |
star | -0.853 |
� | -0.847 |
rovers | -0.826 |
pres | -0.825 |
Token | only |
Token ID | 691 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
meg | +0.239 |
star | +0.232 |
� | +0.228 |
pres | +0.224 |
mania | +0.223 |
Neg logit contributions | |
ibaba | -0.366 |
emale | -0.356 |
aimon | -0.279 |
ijing | -0.274 |
WRITE | -0.273 |
Token | be |
Token ID | 307 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
ibaba | +0.802 |
emale | +0.782 |
aimon | +0.608 |
ijing | +0.597 |
WRITE | +0.592 |
Neg logit contributions | |
meg | -0.548 |
star | -0.535 |
� | -0.521 |
pres | -0.515 |
mania | -0.515 |
Token | my |
Token ID | 616 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
ibaba | +1.644 |
emale | +1.590 |
aimon | +1.228 |
WRITE | +1.208 |
ijing | +1.207 |
Neg logit contributions | |
meg | -1.171 |
star | -1.149 |
� | -1.131 |
pres | -1.113 |
rovers | -1.109 |
Token | first |
Token ID | 717 |
Feature activation | +8.06e-02 |
Pos logit contributions | |
ibaba | +0.711 |
emale | +0.695 |
aimon | +0.547 |
WRITE | +0.538 |
ijing | +0.538 |
Neg logit contributions | |
meg | -0.433 |
star | -0.420 |
� | -0.412 |
power | -0.408 |
pres | -0.407 |
Token | real |
Token ID | 1103 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
meg | +0.224 |
star | +0.218 |
� | +0.213 |
pres | +0.210 |
power | +0.210 |
Neg logit contributions | |
ibaba | -0.366 |
emale | -0.357 |
aimon | -0.280 |
ijing | -0.276 |
WRITE | -0.275 |
Token | session |
Token ID | 6246 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
ibaba | +0.563 |
emale | +0.548 |
aimon | +0.427 |
ijing | +0.420 |
WRITE | +0.417 |
Neg logit contributions | |
meg | -0.381 |
star | -0.371 |
� | -0.362 |
pres | -0.359 |
power | -0.359 |
Token | with |
Token ID | 351 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
ibaba | +1.272 |
emale | +1.237 |
aimon | +0.984 |
ijing | +0.971 |
WRITE | +0.967 |
Neg logit contributions | |
meg | -0.720 |
star | -0.695 |
� | -0.684 |
pres | -0.673 |
rovers | -0.668 |
Token | it |
Token ID | 340 |
Feature activation | -8.21e-02 |
Pos logit contributions | |
ibaba | +0.902 |
emale | +0.880 |
aimon | +0.686 |
ijing | +0.676 |
WRITE | +0.671 |
Neg logit contributions | |
meg | -0.584 |
star | -0.569 |
� | -0.552 |
pres | -0.549 |
power | -0.548 |
Token | was |
Token ID | 373 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
ibaba | +0.391 |
emale | +0.381 |
aimon | +0.304 |
ijing | +0.300 |
WRITE | +0.298 |
Neg logit contributions | |
meg | -0.208 |
star | -0.201 |
� | -0.197 |
pres | -0.193 |
mania | -0.193 |
Token | � |
Token ID | 564 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
ibaba | +1.033 |
emale | +1.003 |
aimon | +0.788 |
ijing | +0.781 |
WRITE | +0.778 |
Neg logit contributions | |
meg | -0.646 |
star | -0.628 |
� | -0.609 |
mania | -0.607 |
pres | -0.606 |
Token | � |
Token ID | 250 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
ibaba | +1.262 |
emale | +1.220 |
aimon | +0.891 |
ijing | +0.890 |
WRITE | +0.875 |
Neg logit contributions | |
meg | -1.197 |
star | -1.176 |
� | -1.144 |
mania | -1.143 |
pres | -1.140 |
Token | dec |
Token ID | 12501 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ibaba | +1.386 |
emale | +1.359 |
aimon | +1.082 |
ijing | +1.071 |
WRITE | +1.058 |
Neg logit contributions | |
meg | -0.679 |
star | -0.656 |
� | -0.641 |
mania | -0.629 |
rovers | -0.625 |
Token | ided |
Token ID | 1384 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
ibaba | +0.533 |
emale | +0.519 |
aimon | +0.397 |
ijing | +0.390 |
WRITE | +0.386 |
Neg logit contributions | |
meg | -0.408 |
star | -0.399 |
� | -0.388 |
power | -0.385 |
pres | -0.385 |
Token | that |
Token ID | 326 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
meg | +0.430 |
star | +0.420 |
� | +0.406 |
power | +0.403 |
pres | +0.403 |
Neg logit contributions | |
ibaba | -0.694 |
emale | -0.678 |
aimon | -0.531 |
ijing | -0.523 |
WRITE | -0.518 |
Token | self |
Token ID | 2116 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
meg | +0.281 |
star | +0.274 |
� | +0.265 |
mania | +0.263 |
power | +0.263 |
Neg logit contributions | |
ibaba | -0.461 |
emale | -0.449 |
aimon | -0.354 |
ijing | -0.348 |
WRITE | -0.345 |
Token | - |
Token ID | 12 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
meg | +0.393 |
star | +0.366 |
� | +0.365 |
rovers | +0.352 |
mania | +0.351 |
Neg logit contributions | |
ibaba | -0.945 |
emale | -0.921 |
aimon | -0.752 |
mustache | -0.748 |
ijing | -0.742 |
Token | enc |
Token ID | 12685 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
meg | +0.028 |
star | +0.027 |
� | +0.027 |
mania | +0.026 |
power | +0.025 |
Neg logit contributions | |
ibaba | -0.063 |
emale | -0.062 |
aimon | -0.050 |
ijing | -0.049 |
mustache | -0.049 |
Token | our |
Token ID | 454 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
ibaba | +1.788 |
emale | +1.751 |
aimon | +1.379 |
ijing | +1.330 |
WRITE | +1.304 |
Neg logit contributions | |
meg | -1.209 |
� | -1.183 |
star | -1.163 |
pres | -1.125 |
power | -1.124 |
Token | agement |
Token ID | 5082 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
ibaba | +0.871 |
emale | +0.853 |
aimon | +0.665 |
ijing | +0.653 |
mustache | +0.644 |
Neg logit contributions | |
meg | -0.579 |
star | -0.560 |
� | -0.554 |
mania | -0.546 |
pres | -0.545 |
Token | , |
Token ID | 11 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
ibaba | +0.983 |
emale | +0.955 |
aimon | +0.755 |
ijing | +0.739 |
WRITE | +0.736 |
Neg logit contributions | |
meg | -0.617 |
star | -0.602 |
� | -0.583 |
mania | -0.579 |
power | -0.579 |
Token | he |
Token ID | 339 |
Feature activation | -2.04e-03 |
Pos logit contributions | |
ibaba | +0.679 |
emale | +0.662 |
aimon | +0.520 |
WRITE | +0.510 |
ijing | +0.509 |
Neg logit contributions | |
meg | -0.424 |
star | -0.410 |
� | -0.400 |
mania | -0.396 |
power | -0.395 |
Token | managed |
Token ID | 5257 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
ibaba | +0.009 |
emale | +0.009 |
aimon | +0.007 |
ijing | +0.007 |
WRITE | +0.007 |
Neg logit contributions | |
meg | -0.006 |
star | -0.006 |
� | -0.006 |
power | -0.006 |
pres | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
meg | +0.603 |
star | +0.586 |
mania | +0.564 |
� | +0.564 |
power | +0.561 |
Neg logit contributions | |
ibaba | -1.171 |
emale | -1.143 |
aimon | -0.913 |
ijing | -0.899 |
WRITE | -0.894 |
Token | control |
Token ID | 1630 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
meg | +0.666 |
star | +0.652 |
� | +0.630 |
pres | +0.630 |
power | +0.630 |
Neg logit contributions | |
ibaba | -0.855 |
emale | -0.834 |
aimon | -0.636 |
ijing | -0.625 |
WRITE | -0.612 |
Token | to |
Token ID | 284 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
meg | +2.009 |
star | +1.949 |
� | +1.878 |
power | +1.864 |
mania | +1.864 |
Neg logit contributions | |
ibaba | -3.897 |
emale | -3.806 |
aimon | -3.051 |
ijing | -3.005 |
WRITE | -2.971 |
Token | schizophrenia |
Token ID | 22794 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
meg | +0.385 |
star | +0.375 |
� | +0.364 |
power | +0.359 |
pres | +0.359 |
Neg logit contributions | |
ibaba | -0.659 |
emale | -0.643 |
aimon | -0.511 |
ijing | -0.502 |
WRITE | -0.498 |
Token | , |
Token ID | 11 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
meg | +0.712 |
star | +0.692 |
� | +0.676 |
pres | +0.663 |
rovers | +0.663 |
Neg logit contributions | |
ibaba | -1.153 |
emale | -1.119 |
aimon | -0.884 |
ijing | -0.874 |
WRITE | -0.863 |
Token | they |
Token ID | 484 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
ibaba | +0.763 |
emale | +0.741 |
aimon | +0.588 |
WRITE | +0.576 |
ijing | +0.576 |
Neg logit contributions | |
meg | -0.469 |
star | -0.455 |
� | -0.448 |
power | -0.436 |
mania | -0.435 |
Token | disagree |
Token ID | 12546 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
ibaba | +0.151 |
emale | +0.147 |
aimon | +0.116 |
ijing | +0.115 |
WRITE | +0.114 |
Neg logit contributions | |
meg | -0.093 |
star | -0.090 |
� | -0.089 |
mania | -0.087 |
power | -0.086 |
Token | about |
Token ID | 546 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
meg | +0.157 |
star | +0.153 |
� | +0.149 |
mania | +0.147 |
power | +0.145 |
Neg logit contributions | |
ibaba | -0.265 |
emale | -0.260 |
aimon | -0.207 |
ijing | -0.205 |
mustache | -0.203 |
Token | exactly |
Token ID | 3446 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
ibaba | +1.310 |
emale | +1.265 |
aimon | +0.984 |
ijing | +0.962 |
mustache | +0.957 |
Neg logit contributions | |
meg | -1.006 |
star | -0.985 |
� | -0.963 |
mania | -0.949 |
power | -0.946 |
Token | how |
Token ID | 703 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
meg | +0.700 |
star | +0.680 |
� | +0.659 |
mania | +0.654 |
power | +0.651 |
Neg logit contributions | |
ibaba | -1.168 |
emale | -1.134 |
aimon | -0.897 |
ijing | -0.884 |
WRITE | -0.878 |
Token | — |
Token ID | 960 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
ibaba | +0.348 |
emale | +0.334 |
aimon | +0.270 |
mustache | +0.266 |
ijing | +0.266 |
Neg logit contributions | |
meg | -0.208 |
star | -0.201 |
� | -0.199 |
rovers | -0.193 |
power | -0.192 |
Token | it |
Token ID | 270 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
meg | +1.002 |
star | +0.974 |
� | +0.946 |
mania | +0.936 |
power | +0.933 |
Neg logit contributions | |
ibaba | -1.726 |
emale | -1.685 |
aimon | -1.331 |
ijing | -1.313 |
WRITE | -1.298 |
Token | 's |
Token ID | 338 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ibaba | +0.195 |
emale | +0.187 |
ijing | +0.151 |
aimon | +0.151 |
mustache | +0.148 |
Neg logit contributions | |
meg | -0.109 |
star | -0.104 |
� | -0.102 |
rovers | -0.100 |
mania | -0.099 |
Token | common |
Token ID | 2219 |
Feature activation | -7.69e-02 |
Pos logit contributions | |
ibaba | +0.821 |
emale | +0.792 |
aimon | +0.622 |
ijing | +0.612 |
WRITE | +0.611 |
Neg logit contributions | |
meg | -0.542 |
star | -0.529 |
� | -0.515 |
pres | -0.510 |
mania | -0.509 |
Token | sense |
Token ID | 2565 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ibaba | +0.282 |
emale | +0.273 |
aimon | +0.200 |
ijing | +0.196 |
WRITE | +0.195 |
Neg logit contributions | |
meg | -0.281 |
star | -0.275 |
� | -0.268 |
mania | -0.267 |
pres | -0.267 |
Token | enough |
Token ID | 1576 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
meg | +0.262 |
star | +0.255 |
� | +0.245 |
pres | +0.243 |
mania | +0.243 |
Neg logit contributions | |
ibaba | -0.551 |
emale | -0.537 |
aimon | -0.434 |
ijing | -0.428 |
WRITE | -0.426 |
Token | to |
Token ID | 284 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
ibaba | +1.915 |
emale | +1.841 |
aimon | +1.494 |
WRITE | +1.478 |
ijing | +1.472 |
Neg logit contributions | |
meg | -1.035 |
star | -0.999 |
� | -0.969 |
mania | -0.967 |
pres | -0.959 |
Token | get |
Token ID | 651 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
meg | +0.929 |
star | +0.907 |
power | +0.876 |
pres | +0.875 |
� | +0.873 |
Neg logit contributions | |
ibaba | -1.415 |
emale | -1.384 |
aimon | -1.076 |
ijing | -1.059 |
WRITE | -1.038 |
Token | around |
Token ID | 1088 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
ibaba | +0.320 |
emale | +0.311 |
aimon | +0.247 |
ijing | +0.243 |
WRITE | +0.242 |
Neg logit contributions | |
meg | -0.185 |
star | -0.180 |
� | -0.175 |
power | -0.173 |
pres | -0.173 |
Token | our |
Token ID | 674 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
ibaba | +0.999 |
emale | +0.970 |
aimon | +0.791 |
ijing | +0.777 |
WRITE | +0.773 |
Neg logit contributions | |
meg | -0.480 |
star | -0.467 |
� | -0.451 |
power | -0.446 |
pres | -0.445 |
Token | own |
Token ID | 898 |
Feature activation | -8.26e-02 |
Pos logit contributions | |
ibaba | +0.869 |
emale | +0.846 |
aimon | +0.660 |
ijing | +0.648 |
WRITE | +0.646 |
Neg logit contributions | |
meg | -0.620 |
star | -0.606 |
� | -0.590 |
mania | -0.586 |
power | -0.583 |
Token | space |
Token ID | 2272 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
ibaba | +0.342 |
emale | +0.332 |
aimon | +0.258 |
ijing | +0.254 |
WRITE | +0.253 |
Neg logit contributions | |
meg | -0.257 |
star | -0.251 |
� | -0.246 |
mania | -0.243 |
pres | -0.242 |
Token | without |
Token ID | 1231 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
meg | +0.237 |
star | +0.230 |
� | +0.222 |
power | +0.219 |
mania | +0.218 |
Neg logit contributions | |
ibaba | -0.548 |
emale | -0.534 |
aimon | -0.435 |
ijing | -0.428 |
WRITE | -0.427 |
Token | man |
Token ID | 582 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
ibaba | +0.138 |
emale | +0.134 |
aimon | +0.104 |
ijing | +0.102 |
WRITE | +0.101 |
Neg logit contributions | |
meg | -0.104 |
star | -0.102 |
� | -0.099 |
pres | -0.098 |
mania | -0.098 |
Token | for |
Token ID | 329 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
meg | +0.601 |
star | +0.583 |
� | +0.562 |
mania | +0.560 |
pres | +0.557 |
Neg logit contributions | |
ibaba | -1.358 |
emale | -1.318 |
aimon | -1.063 |
ijing | -1.058 |
WRITE | -1.048 |
Token | an |
Token ID | 281 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
ibaba | +0.958 |
emale | +0.923 |
aimon | +0.698 |
ijing | +0.691 |
WRITE | +0.686 |
Neg logit contributions | |
meg | -0.798 |
star | -0.780 |
mania | -0.760 |
� | -0.760 |
pres | -0.754 |
Token | equitable |
Token ID | 37164 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
meg | +0.074 |
star | +0.073 |
� | +0.071 |
pres | +0.071 |
power | +0.070 |
Neg logit contributions | |
ibaba | -0.072 |
emale | -0.070 |
aimon | -0.051 |
ijing | -0.050 |
WRITE | -0.049 |
Token | future |
Token ID | 2003 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
meg | +0.102 |
star | +0.099 |
� | +0.096 |
power | +0.096 |
pres | +0.096 |
Neg logit contributions | |
ibaba | -0.136 |
emale | -0.133 |
aimon | -0.102 |
ijing | -0.100 |
WRITE | -0.099 |
Token | – |
Token ID | 784 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
ibaba | +0.102 |
emale | +0.099 |
aimon | +0.081 |
ijing | +0.080 |
WRITE | +0.080 |
Neg logit contributions | |
meg | -0.042 |
star | -0.041 |
� | -0.039 |
mania | -0.039 |
pres | -0.038 |
Token | they |
Token ID | 484 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
meg | +0.399 |
star | +0.387 |
� | +0.372 |
mania | +0.369 |
power | +0.369 |
Neg logit contributions | |
ibaba | -0.822 |
emale | -0.802 |
aimon | -0.644 |
ijing | -0.636 |
WRITE | -0.630 |
Token | are |
Token ID | 389 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
ibaba | +1.566 |
emale | +1.514 |
ijing | +1.174 |
aimon | +1.165 |
WRITE | +1.159 |
Neg logit contributions | |
meg | -1.024 |
star | -0.997 |
� | -0.969 |
mania | -0.959 |
rovers | -0.953 |
Token | the |
Token ID | 262 |
Feature activation | +7.28e-03 |
Pos logit contributions | |
ibaba | +1.033 |
emale | +1.001 |
aimon | +0.775 |
ijing | +0.768 |
WRITE | +0.755 |
Neg logit contributions | |
meg | -0.725 |
star | -0.706 |
� | -0.688 |
mania | -0.683 |
pres | -0.680 |
Token | enemy |
Token ID | 4472 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
meg | +0.023 |
star | +0.022 |
� | +0.021 |
power | +0.021 |
pres | +0.021 |
Neg logit contributions | |
ibaba | -0.031 |
emale | -0.030 |
aimon | -0.023 |
ijing | -0.023 |
WRITE | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +6.86e-02 |
Pos logit contributions | |
meg | +0.439 |
star | +0.424 |
� | +0.410 |
mania | +0.407 |
pres | +0.405 |
Neg logit contributions | |
ibaba | -1.003 |
emale | -0.972 |
aimon | -0.791 |
ijing | -0.782 |
WRITE | -0.775 |
Token |
|
Token ID | 198 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
meg | +0.171 |
star | +0.166 |
� | +0.160 |
mania | +0.159 |
power | +0.159 |
Neg logit contributions | |
ibaba | -0.331 |
emale | -0.323 |
aimon | -0.258 |
ijing | -0.254 |
WRITE | -0.253 |
Token | home |
Token ID | 1363 |
Feature activation | -9.63e-02 |
Pos logit contributions | |
ibaba | +1.905 |
emale | +1.866 |
aimon | +1.450 |
mustache | +1.429 |
ijing | +1.427 |
Neg logit contributions | |
meg | -1.286 |
star | -1.254 |
� | -1.231 |
pres | -1.210 |
mania | -1.209 |
Token | - |
Token ID | 12 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
ibaba | +0.472 |
emale | +0.460 |
aimon | +0.371 |
ijing | +0.367 |
WRITE | +0.361 |
Neg logit contributions | |
meg | -0.229 |
star | -0.222 |
� | -0.216 |
mania | -0.214 |
pres | -0.212 |
Token | grown |
Token ID | 22377 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
ibaba | +2.509 |
emale | +2.449 |
aimon | +1.956 |
ijing | +1.919 |
WRITE | +1.881 |
Neg logit contributions | |
meg | -1.409 |
star | -1.368 |
� | -1.361 |
mania | -1.339 |
rovers | -1.316 |
Token | Pent |
Token ID | 9696 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
ibaba | +1.311 |
emale | +1.281 |
aimon | +0.993 |
mustache | +0.976 |
ijing | +0.974 |
Neg logit contributions | |
meg | -0.930 |
star | -0.909 |
� | -0.893 |
pres | -0.880 |
mania | -0.878 |
Token | ec |
Token ID | 721 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
meg | +0.073 |
star | +0.071 |
� | +0.069 |
mania | +0.069 |
power | +0.069 |
Neg logit contributions | |
ibaba | -0.104 |
emale | -0.102 |
aimon | -0.079 |
ijing | -0.078 |
WRITE | -0.077 |
Token | ost |
Token ID | 455 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
ibaba | +2.642 |
emale | +2.595 |
aimon | +2.143 |
ijing | +2.100 |
WRITE | +2.082 |
Neg logit contributions | |
meg | -0.916 |
� | -0.877 |
star | -0.861 |
mania | -0.833 |
pres | -0.826 |
Token | al |
Token ID | 282 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
ibaba | +3.845 |
emale | +3.807 |
aimon | +3.039 |
ijing | +2.956 |
mustache | +2.870 |
Neg logit contributions | |
meg | -2.389 |
star | -2.305 |
� | -2.291 |
pres | -2.272 |
mania | -2.254 |
Token | group |
Token ID | 1448 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
meg | +0.054 |
star | +0.053 |
� | +0.052 |
pres | +0.051 |
mania | +0.051 |
Neg logit contributions | |
ibaba | -0.067 |
emale | -0.066 |
aimon | -0.050 |
ijing | -0.049 |
WRITE | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
meg | +0.278 |
star | +0.272 |
� | +0.263 |
mania | +0.260 |
pres | +0.260 |
Neg logit contributions | |
ibaba | -0.538 |
emale | -0.525 |
aimon | -0.421 |
ijing | -0.415 |
WRITE | -0.410 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
ibaba | +0.661 |
emale | +0.646 |
aimon | +0.500 |
ijing | +0.492 |
WRITE | +0.489 |
Neg logit contributions | |
meg | -0.460 |
star | -0.449 |
� | -0.436 |
power | -0.434 |
mania | -0.433 |
Token | new |
Token ID | 649 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
ibaba | +0.444 |
emale | +0.434 |
aimon | +0.335 |
ijing | +0.330 |
WRITE | +0.327 |
Neg logit contributions | |
meg | -0.315 |
star | -0.308 |
� | -0.300 |
power | -0.297 |
mania | -0.297 |
Token | career |
Token ID | 3451 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
meg | +0.329 |
star | +0.322 |
� | +0.312 |
pres | +0.310 |
power | +0.310 |
Neg logit contributions | |
ibaba | -0.452 |
emale | -0.441 |
aimon | -0.339 |
ijing | -0.334 |
WRITE | -0.329 |
Token | out |
Token ID | 503 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
ibaba | +1.469 |
emale | +1.427 |
aimon | +1.129 |
mustache | +1.119 |
ijing | +1.110 |
Neg logit contributions | |
meg | -0.906 |
star | -0.873 |
� | -0.864 |
pres | -0.849 |
power | -0.846 |
Token | of |
Token ID | 286 |
Feature activation | +6.70e-01 |
Pos logit contributions | |
meg | +0.420 |
star | +0.408 |
� | +0.393 |
power | +0.389 |
pres | +0.389 |
Neg logit contributions | |
ibaba | -0.844 |
emale | -0.826 |
aimon | -0.661 |
ijing | -0.652 |
WRITE | -0.646 |
Token | playing |
Token ID | 2712 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
meg | +2.053 |
star | +2.010 |
pres | +1.936 |
power | +1.933 |
� | +1.932 |
Neg logit contributions | |
ibaba | -2.869 |
emale | -2.788 |
aimon | -2.147 |
ijing | -2.114 |
WRITE | -2.076 |
Token | un |
Token ID | 555 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
ibaba | +0.403 |
emale | +0.393 |
aimon | +0.303 |
WRITE | +0.300 |
ijing | +0.299 |
Neg logit contributions | |
meg | -0.283 |
star | -0.276 |
� | -0.271 |
mania | -0.267 |
pres | -0.267 |
Token | hing |
Token ID | 722 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
meg | +0.125 |
star | +0.120 |
� | +0.119 |
mania | +0.116 |
power | +0.115 |
Neg logit contributions | |
ibaba | -0.252 |
emale | -0.248 |
aimon | -0.199 |
ijing | -0.196 |
WRITE | -0.192 |
Token | ed |
Token ID | 276 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
ibaba | +4.197 |
emale | +4.111 |
aimon | +3.340 |
ijing | +3.312 |
WRITE | +3.281 |
Neg logit contributions | |
meg | -1.674 |
star | -1.613 |
� | -1.562 |
mania | -1.542 |
power | -1.529 |
Token | lun |
Token ID | 14678 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
ibaba | +1.465 |
emale | +1.435 |
aimon | +1.109 |
mustache | +1.097 |
WRITE | +1.096 |
Neg logit contributions | |
meg | -1.029 |
star | -1.001 |
� | -0.984 |
mania | -0.971 |
pres | -0.968 |
Token | atics |
Token ID | 23372 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
meg | +0.167 |
star | +0.160 |
� | +0.159 |
mania | +0.157 |
pres | +0.155 |
Neg logit contributions | |
ibaba | -0.335 |
emale | -0.327 |
aimon | -0.265 |
ijing | -0.260 |
WRITE | -0.256 |
Token | . |
Token ID | 13 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
meg | +0.464 |
star | +0.451 |
� | +0.437 |
pres | +0.432 |
mania | +0.430 |
Neg logit contributions | |
ibaba | -0.917 |
emale | -0.895 |
aimon | -0.723 |
ijing | -0.713 |
mustache | -0.708 |
Token | Maybe |
Token ID | 6674 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
ibaba | +1.149 |
emale | +1.122 |
aimon | +0.879 |
WRITE | +0.871 |
ijing | +0.865 |
Neg logit contributions | |
meg | -0.718 |
star | -0.694 |
� | -0.678 |
power | -0.670 |
mania | -0.670 |
Token | agh |
Token ID | 10471 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
ibaba | +1.204 |
emale | +1.154 |
aimon | +0.884 |
ijing | +0.845 |
WRITE | +0.835 |
Neg logit contributions | |
meg | -1.093 |
� | -1.070 |
pres | -1.040 |
star | -1.038 |
mania | -1.030 |
Token | Kil |
Token ID | 19902 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
meg | +0.484 |
star | +0.469 |
� | +0.460 |
pres | +0.455 |
power | +0.453 |
Neg logit contributions | |
ibaba | -0.803 |
emale | -0.788 |
aimon | -0.615 |
WRITE | -0.609 |
ijing | -0.604 |
Token | c |
Token ID | 66 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ibaba | +1.898 |
emale | +1.829 |
aimon | +1.478 |
ijing | +1.452 |
WRITE | +1.433 |
Neg logit contributions | |
meg | -1.030 |
� | -1.010 |
star | -0.991 |
rovers | -0.965 |
mania | -0.962 |
Token | oy |
Token ID | 726 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
ibaba | +0.072 |
emale | +0.070 |
aimon | +0.054 |
ijing | +0.053 |
WRITE | +0.052 |
Neg logit contributions | |
meg | -0.058 |
star | -0.056 |
� | -0.055 |
pres | -0.054 |
mania | -0.054 |
Token | ne |
Token ID | 710 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
ibaba | +0.270 |
emale | +0.265 |
aimon | +0.214 |
ijing | +0.211 |
WRITE | +0.210 |
Neg logit contributions | |
meg | -0.115 |
star | -0.112 |
� | -0.108 |
mania | -0.107 |
pres | -0.107 |
Token |
|
Token ID | 198 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
meg | +0.121 |
star | +0.118 |
� | +0.114 |
pres | +0.112 |
mania | +0.112 |
Neg logit contributions | |
ibaba | -0.249 |
emale | -0.243 |
aimon | -0.195 |
ijing | -0.193 |
WRITE | -0.192 |
Token |
|
Token ID | 198 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
ibaba | +4.043 |
emale | +3.814 |
aimon | +3.167 |
WRITE | +3.159 |
margins | +3.063 |
Neg logit contributions | |
meg | -2.150 |
� | -2.064 |
star | -2.054 |
rovers | -2.030 |
pres | -1.973 |
Token | A |
Token ID | 32 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
ibaba | +3.257 |
emale | +3.172 |
aimon | +2.558 |
ijing | +2.536 |
WRITE | +2.514 |
Neg logit contributions | |
meg | -1.712 |
star | -1.661 |
� | -1.627 |
rovers | -1.617 |
mania | -1.591 |
Token | swath |
Token ID | 50176 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
ibaba | +0.978 |
emale | +0.948 |
aimon | +0.732 |
ijing | +0.726 |
WRITE | +0.716 |
Neg logit contributions | |
meg | -0.674 |
star | -0.657 |
� | -0.649 |
mania | -0.640 |
pres | -0.635 |
Token | of |
Token ID | 286 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
ibaba | +3.014 |
emale | +2.924 |
ijing | +2.297 |
mustache | +2.285 |
aimon | +2.277 |
Neg logit contributions | |
meg | -1.847 |
� | -1.791 |
star | -1.788 |
pres | -1.746 |
sim | -1.734 |
Token | secondary |
Token ID | 9233 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
meg | +0.080 |
star | +0.078 |
� | +0.076 |
pres | +0.075 |
power | +0.075 |
Neg logit contributions | |
ibaba | -0.100 |
emale | -0.097 |
aimon | -0.074 |
ijing | -0.073 |
WRITE | -0.072 |
Token | ." |
Token ID | 526 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
ibaba | +1.427 |
emale | +1.397 |
aimon | +1.153 |
WRITE | +1.132 |
ijing | +1.126 |
Neg logit contributions | |
meg | -0.572 |
� | -0.543 |
star | -0.538 |
rovers | -0.534 |
mania | -0.528 |
Token | My |
Token ID | 2011 |
Feature activation | -7.55e-02 |
Pos logit contributions | |
ibaba | +0.390 |
emale | +0.380 |
aimon | +0.302 |
WRITE | +0.298 |
ijing | +0.296 |
Neg logit contributions | |
meg | -0.217 |
star | -0.209 |
� | -0.205 |
mania | -0.201 |
power | -0.201 |
Token | eyes |
Token ID | 2951 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
ibaba | +0.291 |
emale | +0.282 |
aimon | +0.212 |
ijing | +0.207 |
WRITE | +0.206 |
Neg logit contributions | |
meg | -0.260 |
star | -0.255 |
� | -0.249 |
mania | -0.247 |
power | -0.247 |
Token | are |
Token ID | 389 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
ibaba | +1.493 |
emale | +1.445 |
aimon | +1.105 |
WRITE | +1.099 |
mustache | +1.096 |
Neg logit contributions | |
meg | -1.272 |
� | -1.234 |
star | -1.227 |
pres | -1.219 |
mania | -1.207 |
Token | still |
Token ID | 991 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
ibaba | +2.043 |
emale | +1.980 |
aimon | +1.514 |
mustache | +1.495 |
ijing | +1.494 |
Neg logit contributions | |
meg | -1.742 |
star | -1.674 |
� | -1.662 |
pres | -1.655 |
rovers | -1.653 |
Token | wide |
Token ID | 3094 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
ibaba | +1.487 |
emale | +1.446 |
aimon | +1.079 |
ijing | +1.060 |
mustache | +1.058 |
Neg logit contributions | |
meg | -1.462 |
� | -1.408 |
star | -1.400 |
pres | -1.396 |
mania | -1.394 |
Token | open |
Token ID | 1280 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
ibaba | +3.003 |
emale | +2.934 |
mustache | +2.282 |
WRITE | +2.249 |
aimon | +2.235 |
Neg logit contributions | |
meg | -2.559 |
� | -2.507 |
rovers | -2.465 |
mania | -2.440 |
pres | -2.429 |
Token | . |
Token ID | 13 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
ibaba | +1.319 |
emale | +1.278 |
aimon | +1.035 |
mustache | +1.033 |
WRITE | +1.025 |
Neg logit contributions | |
meg | -0.673 |
star | -0.646 |
� | -0.643 |
rovers | -0.631 |
mania | -0.628 |
Token | I |
Token ID | 314 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
ibaba | +0.306 |
emale | +0.299 |
aimon | +0.237 |
WRITE | +0.233 |
ijing | +0.232 |
Neg logit contributions | |
meg | -0.181 |
star | -0.175 |
� | -0.171 |
mania | -0.168 |
power | -0.168 |
Token | wasn |
Token ID | 2492 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
meg | +0.142 |
star | +0.138 |
� | +0.134 |
pres | +0.133 |
mania | +0.133 |
Neg logit contributions | |
ibaba | -0.242 |
emale | -0.236 |
aimon | -0.187 |
ijing | -0.184 |
WRITE | -0.183 |
Token | 't |
Token ID | 470 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
ibaba | +0.098 |
emale | +0.095 |
aimon | +0.078 |
ijing | +0.077 |
WRITE | +0.077 |
Neg logit contributions | |
meg | -0.039 |
star | -0.038 |
� | -0.036 |
pres | -0.036 |
power | -0.036 |
Token | who |
Token ID | 508 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
ibaba | +2.318 |
emale | +2.257 |
aimon | +1.807 |
ijing | +1.784 |
mustache | +1.779 |
Neg logit contributions | |
meg | -1.331 |
star | -1.280 |
� | -1.261 |
rovers | -1.240 |
power | -1.236 |
Token | was |
Token ID | 373 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
ibaba | +1.185 |
emale | +1.158 |
aimon | +0.911 |
ijing | +0.902 |
mustache | +0.902 |
Neg logit contributions | |
meg | -0.728 |
star | -0.706 |
� | -0.690 |
power | -0.682 |
rovers | -0.677 |
Token | hit |
Token ID | 2277 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
ibaba | +1.296 |
emale | +1.266 |
aimon | +0.979 |
ijing | +0.965 |
WRITE | +0.954 |
Neg logit contributions | |
meg | -0.910 |
star | -0.885 |
� | -0.864 |
power | -0.854 |
mania | -0.852 |
Token | by |
Token ID | 416 |
Feature activation | -7.27e-02 |
Pos logit contributions | |
ibaba | +1.582 |
emale | +1.557 |
aimon | +1.226 |
ijing | +1.211 |
WRITE | +1.211 |
Neg logit contributions | |
meg | -0.892 |
star | -0.857 |
� | -0.851 |
rovers | -0.831 |
pres | -0.828 |
Token | sh |
Token ID | 427 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
ibaba | +0.306 |
emale | +0.297 |
aimon | +0.230 |
ijing | +0.225 |
WRITE | +0.224 |
Neg logit contributions | |
meg | -0.225 |
star | -0.220 |
� | -0.216 |
power | -0.213 |
pres | -0.212 |
Token | rapnel |
Token ID | 48766 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
meg | +0.309 |
star | +0.304 |
� | +0.299 |
pres | +0.294 |
mania | +0.293 |
Neg logit contributions | |
ibaba | -0.329 |
emale | -0.319 |
aimon | -0.237 |
ijing | -0.232 |
WRITE | -0.227 |
Token | in |
Token ID | 287 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
ibaba | +4.226 |
emale | +4.101 |
mustache | +3.381 |
aimon | +3.349 |
buckle | +3.320 |
Neg logit contributions | |
meg | -2.301 |
� | -2.241 |
star | -2.222 |
rovers | -2.136 |
pres | -2.110 |
Token | World |
Token ID | 2159 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
meg | +0.145 |
star | +0.141 |
� | +0.137 |
power | +0.136 |
pres | +0.136 |
Neg logit contributions | |
ibaba | -0.230 |
emale | -0.223 |
aimon | -0.175 |
ijing | -0.173 |
WRITE | -0.171 |
Token | War |
Token ID | 1810 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
ibaba | +1.233 |
emale | +1.204 |
aimon | +1.015 |
ijing | +1.004 |
WRITE | +0.995 |
Neg logit contributions | |
meg | -0.269 |
star | -0.251 |
� | -0.241 |
pres | -0.231 |
mania | -0.230 |
Token | II |
Token ID | 2873 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
ibaba | +2.765 |
emale | +2.671 |
aimon | +2.162 |
ijing | +2.148 |
WRITE | +2.132 |
Neg logit contributions | |
meg | -1.254 |
star | -1.206 |
� | -1.193 |
pres | -1.151 |
power | -1.146 |
Token | ; |
Token ID | 26 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
ibaba | +1.518 |
emale | +1.473 |
aimon | +1.187 |
mustache | +1.183 |
ijing | +1.176 |
Neg logit contributions | |
meg | -0.739 |
star | -0.706 |
� | -0.696 |
mania | -0.678 |
power | -0.675 |