Token | Board |
Token ID | 5926 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
atti | +0.860 |
etti | +0.834 |
karma | +0.833 |
latex | +0.827 |
Neighborhood | +0.785 |
Neg logit contributions | |
GBT | -0.859 |
Catal | -0.742 |
Bah | -0.736 |
Franç | -0.728 |
ンジ | -0.700 |
Token | { |
Token ID | 1391 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
atti | +1.031 |
etti | +0.996 |
karma | +0.989 |
latex | +0.984 |
strap | +0.924 |
Neg logit contributions | |
GBT | -1.445 |
Catal | -1.278 |
Bah | -1.269 |
Franç | -1.269 |
ンジ | -1.221 |
Token | Board |
Token ID | 5926 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
GBT | +0.495 |
Catal | +0.443 |
Bah | +0.439 |
Franç | +0.434 |
ンジ | +0.421 |
Neg logit contributions | |
atti | -0.323 |
etti | -0.310 |
karma | -0.308 |
latex | -0.306 |
strap | -0.285 |
Token | { |
Token ID | 1391 |
Feature activation | -9.31e-01 |
Pos logit contributions | |
atti | +1.592 |
etti | +1.543 |
karma | +1.536 |
latex | +1.527 |
strap | +1.446 |
Neg logit contributions | |
GBT | -1.754 |
Catal | -1.529 |
Franç | -1.518 |
Bah | -1.516 |
ンジ | -1.456 |
Token | turn |
Token ID | 1210 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
GBT | +0.819 |
Catal | +0.726 |
Bah | +0.718 |
Franç | +0.710 |
ンジ | +0.685 |
Neg logit contributions | |
atti | -0.641 |
etti | -0.618 |
karma | -0.614 |
latex | -0.610 |
strap | -0.572 |
Token | : |
Token ID | 1058 |
Feature activation | +6.01e+00 |
Pos logit contributions | |
atti | +1.718 |
etti | +1.661 |
karma | +1.641 |
latex | +1.627 |
strap | +1.545 |
Neg logit contributions | |
GBT | -2.109 |
Catal | -1.860 |
Bah | -1.834 |
Franç | -1.834 |
ンジ | -1.759 |
Token | self |
Token ID | 2116 |
Feature activation | +8.21e-01 |
Pos logit contributions | |
atti | +3.935 |
karma | +3.893 |
latex | +3.812 |
etti | +3.795 |
Neighborhood | +3.654 |
Neg logit contributions | |
GBT | -5.246 |
Catal | -4.675 |
Bah | -4.605 |
Franç | -4.590 |
ンジ | -4.401 |
Token | . |
Token ID | 764 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
atti | +0.627 |
etti | +0.606 |
karma | +0.602 |
latex | +0.598 |
strap | +0.565 |
Neg logit contributions | |
GBT | -0.660 |
Catal | -0.578 |
Bah | -0.570 |
Franç | -0.563 |
ンジ | -0.542 |
Token | clone |
Token ID | 21018 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
atti | +1.462 |
karma | +1.423 |
etti | +1.419 |
latex | +1.406 |
strap | +1.342 |
Neg logit contributions | |
GBT | -1.592 |
Catal | -1.402 |
Franç | -1.387 |
Bah | -1.379 |
ンジ | -1.323 |
Token | (), |
Token ID | 29994 |
Feature activation | +9.31e-01 |
Pos logit contributions | |
atti | +2.458 |
etti | +2.396 |
karma | +2.364 |
latex | +2.350 |
strap | +2.250 |
Neg logit contributions | |
GBT | -3.039 |
Catal | -2.665 |
Franç | -2.653 |
Bah | -2.633 |
ンジ | -2.530 |
Token | pieces |
Token ID | 5207 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
atti | +0.645 |
etti | +0.627 |
karma | +0.625 |
latex | +0.618 |
strap | +0.578 |
Neg logit contributions | |
GBT | -0.812 |
Catal | -0.717 |
Bah | -0.710 |
Franç | -0.701 |
ンジ | -0.679 |
Token |
|
Token ID | 198 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
GBT | +0.179 |
Catal | +0.157 |
Bah | +0.155 |
Franç | +0.153 |
ンジ | +0.148 |
Neg logit contributions | |
atti | -0.158 |
etti | -0.153 |
karma | -0.152 |
latex | -0.150 |
strap | -0.142 |
Token | This |
Token ID | 1212 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
atti | +0.141 |
etti | +0.136 |
karma | +0.135 |
latex | +0.134 |
strap | +0.127 |
Neg logit contributions | |
GBT | -0.151 |
Catal | -0.132 |
Bah | -0.131 |
Franç | -0.130 |
ンジ | -0.125 |
Token | fact |
Token ID | 1109 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
GBT | +0.277 |
Catal | +0.246 |
Bah | +0.243 |
Franç | +0.240 |
ンジ | +0.232 |
Neg logit contributions | |
atti | -0.215 |
etti | -0.206 |
karma | -0.204 |
latex | -0.203 |
strap | -0.191 |
Token | did |
Token ID | 750 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
GBT | +0.469 |
Catal | +0.416 |
Bah | +0.411 |
Franç | +0.407 |
ンジ | +0.393 |
Neg logit contributions | |
atti | -0.366 |
etti | -0.353 |
karma | -0.349 |
latex | -0.347 |
strap | -0.326 |
Token | not |
Token ID | 407 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
GBT | +0.854 |
Catal | +0.747 |
Bah | +0.738 |
Franç | +0.726 |
ンジ | +0.699 |
Neg logit contributions | |
atti | -0.801 |
etti | -0.772 |
karma | -0.764 |
latex | -0.761 |
strap | -0.719 |
Token | escape |
Token ID | 6654 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
GBT | +0.110 |
Catal | +0.096 |
Bah | +0.095 |
Franç | +0.094 |
ンジ | +0.091 |
Neg logit contributions | |
atti | -0.100 |
etti | -0.097 |
karma | -0.097 |
latex | -0.096 |
strap | -0.090 |
Token | government |
Token ID | 1230 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
atti | +3.686 |
karma | +3.619 |
etti | +3.612 |
latex | +3.587 |
Neighborhood | +3.346 |
Neg logit contributions | |
GBT | -5.299 |
Catal | -4.709 |
Bah | -4.681 |
Franç | -4.523 |
ンジ | -4.512 |
Token | officials |
Token ID | 2828 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
atti | +1.108 |
etti | +1.072 |
karma | +1.072 |
latex | +1.065 |
strap | +0.991 |
Neg logit contributions | |
GBT | -1.428 |
Catal | -1.264 |
Bah | -1.246 |
Franç | -1.227 |
ンジ | -1.200 |
Token | , |
Token ID | 11 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
atti | +1.187 |
etti | +1.156 |
karma | +1.148 |
latex | +1.145 |
strap | +1.075 |
Neg logit contributions | |
GBT | -1.280 |
Catal | -1.114 |
Bah | -1.100 |
Franç | -1.074 |
ンジ | -1.054 |
Token | and |
Token ID | 290 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
GBT | +0.262 |
Catal | +0.230 |
Bah | +0.227 |
Franç | +0.225 |
ンジ | +0.215 |
Neg logit contributions | |
atti | -0.240 |
etti | -0.231 |
karma | -0.229 |
latex | -0.227 |
strap | -0.215 |
Token | when |
Token ID | 618 |
Feature activation | +2.21e-03 |
Pos logit contributions | |
atti | +0.529 |
etti | +0.509 |
karma | +0.506 |
latex | +0.502 |
strap | +0.471 |
Neg logit contributions | |
GBT | -0.669 |
Catal | -0.592 |
Bah | -0.586 |
Franç | -0.578 |
ンジ | -0.559 |
Token | the |
Token ID | 262 |
Feature activation | -7.93e-01 |
Pos logit contributions | |
GBT | +0.224 |
Catal | +0.198 |
Bah | +0.196 |
Franç | +0.193 |
ンジ | +0.188 |
Neg logit contributions | |
atti | -0.175 |
etti | -0.169 |
karma | -0.168 |
latex | -0.167 |
strap | -0.156 |
Token | extent |
Token ID | 6287 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
GBT | +0.711 |
Catal | +0.631 |
Bah | +0.624 |
Franç | +0.616 |
ンジ | +0.597 |
Neg logit contributions | |
atti | -0.534 |
etti | -0.513 |
karma | -0.509 |
latex | -0.506 |
strap | -0.473 |
Token | of |
Token ID | 286 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
GBT | +0.655 |
Catal | +0.576 |
Bah | +0.569 |
Franç | +0.562 |
ンジ | +0.542 |
Neg logit contributions | |
atti | -0.565 |
etti | -0.545 |
karma | -0.541 |
latex | -0.538 |
strap | -0.506 |
Token | the |
Token ID | 262 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
atti | +0.353 |
etti | +0.340 |
karma | +0.339 |
latex | +0.337 |
strap | +0.313 |
Neg logit contributions | |
GBT | -0.480 |
Catal | -0.425 |
Bah | -0.420 |
Franç | -0.415 |
ンジ | -0.404 |
Token | rubbish |
Token ID | 35022 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
atti | +0.268 |
etti | +0.259 |
karma | +0.259 |
latex | +0.258 |
Neighborhood | +0.238 |
Neg logit contributions | |
GBT | -0.360 |
Catal | -0.318 |
Bah | -0.313 |
Franç | -0.309 |
ンジ | -0.302 |
Token |
|
Token ID | 198 |
Feature activation | +5.80e+00 |
Pos logit contributions | |
atti | +0.225 |
etti | +0.216 |
karma | +0.215 |
latex | +0.214 |
strap | +0.201 |
Neg logit contributions | |
GBT | -0.264 |
Catal | -0.232 |
Bah | -0.229 |
Franç | -0.226 |
ンジ | -0.219 |
Token |
|
Token ID | 198 |
Feature activation | +5.44e+00 |
Pos logit contributions | |
atti | +3.545 |
etti | +3.385 |
karma | +3.370 |
latex | +3.185 |
Dustin | +3.139 |
Neg logit contributions | |
GBT | -5.467 |
Franç | -4.688 |
ンジ | -4.687 |
Catal | -4.664 |
Bah | -4.607 |
Token | The |
Token ID | 464 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
atti | +3.973 |
etti | +3.795 |
karma | +3.641 |
latex | +3.586 |
strap | +3.536 |
Neg logit contributions | |
GBT | -4.383 |
Franç | -3.900 |
ンジ | -3.697 |
Goku | -3.653 |
Catal | -3.619 |
Token | cost |
Token ID | 1575 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
atti | +0.781 |
etti | +0.753 |
karma | +0.750 |
latex | +0.746 |
Neighborhood | +0.695 |
Neg logit contributions | |
GBT | -0.922 |
Catal | -0.808 |
Bah | -0.801 |
Franç | -0.789 |
ンジ | -0.765 |
Token | of |
Token ID | 286 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
atti | +1.022 |
etti | +0.981 |
karma | +0.975 |
latex | +0.971 |
strap | +0.912 |
Neg logit contributions | |
GBT | -1.224 |
Catal | -1.075 |
Bah | -1.061 |
Franç | -1.047 |
ンジ | -1.015 |
Token | a |
Token ID | 257 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
atti | +0.151 |
karma | +0.145 |
etti | +0.145 |
latex | +0.144 |
strap | +0.133 |
Neg logit contributions | |
GBT | -0.223 |
Catal | -0.198 |
Bah | -0.196 |
Franç | -0.193 |
ンジ | -0.188 |
Token | Clemson |
Token ID | 27801 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
GBT | +1.061 |
Catal | +0.939 |
Bah | +0.932 |
Franç | +0.924 |
Goku | +0.888 |
Neg logit contributions | |
atti | -0.811 |
etti | -0.774 |
karma | -0.767 |
latex | -0.763 |
strap | -0.725 |
Token | Game |
Token ID | 3776 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
GBT | +0.887 |
Catal | +0.791 |
Bah | +0.789 |
Franç | +0.780 |
Goku | +0.754 |
Neg logit contributions | |
atti | -0.560 |
etti | -0.531 |
karma | -0.528 |
latex | -0.525 |
strap | -0.490 |
Token | in |
Token ID | 287 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
atti | +0.401 |
etti | +0.388 |
karma | +0.385 |
latex | +0.382 |
strap | +0.362 |
Neg logit contributions | |
GBT | -0.414 |
Catal | -0.361 |
Bah | -0.356 |
Franç | -0.351 |
ンジ | -0.339 |
Token | 2016 |
Token ID | 1584 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
atti | +0.012 |
etti | +0.012 |
karma | +0.012 |
latex | +0.011 |
strap | +0.011 |
Neg logit contributions | |
GBT | -0.013 |
Catal | -0.012 |
Bah | -0.012 |
Franç | -0.012 |
ンジ | -0.011 |
Token | ( |
Token ID | 357 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
GBT | +0.581 |
Catal | +0.503 |
Bah | +0.497 |
Franç | +0.490 |
ンジ | +0.469 |
Neg logit contributions | |
atti | -0.635 |
etti | -0.614 |
karma | -0.610 |
latex | -0.607 |
strap | -0.576 |
Token | Photo |
Token ID | 6191 |
Feature activation | +5.71e+00 |
Pos logit contributions | |
GBT | +0.574 |
Catal | +0.516 |
Bah | +0.511 |
Franç | +0.504 |
ンジ | +0.490 |
Neg logit contributions | |
atti | -0.336 |
etti | -0.321 |
karma | -0.318 |
latex | -0.316 |
strap | -0.292 |
Token | : |
Token ID | 25 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
atti | +3.622 |
etti | +3.465 |
karma | +3.460 |
latex | +3.381 |
Dustin | +3.255 |
Neg logit contributions | |
GBT | -5.379 |
Catal | -4.617 |
Bah | -4.578 |
ンジ | -4.536 |
Franç | -4.508 |
Token | Wayne |
Token ID | 13329 |
Feature activation | -9.92e-01 |
Pos logit contributions | |
atti | +0.506 |
etti | +0.488 |
karma | +0.484 |
latex | +0.481 |
strap | +0.454 |
Neg logit contributions | |
GBT | -0.550 |
Catal | -0.482 |
Bah | -0.476 |
Franç | -0.468 |
ンジ | -0.453 |
Token | McG |
Token ID | 11130 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
GBT | +0.823 |
Catal | +0.723 |
Bah | +0.714 |
Franç | +0.703 |
ンジ | +0.682 |
Neg logit contributions | |
atti | -0.733 |
etti | -0.707 |
karma | -0.701 |
latex | -0.695 |
strap | -0.657 |
Token | a |
Token ID | 64 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
GBT | +0.291 |
Catal | +0.256 |
Bah | +0.253 |
Franç | +0.250 |
ンジ | +0.242 |
Neg logit contributions | |
atti | -0.253 |
etti | -0.245 |
karma | -0.241 |
latex | -0.240 |
strap | -0.227 |
Token | hee |
Token ID | 21067 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
atti | +1.387 |
etti | +1.322 |
karma | +1.284 |
latex | +1.278 |
strap | +1.210 |
Neg logit contributions | |
GBT | -2.003 |
Catal | -1.792 |
Bah | -1.764 |
Franç | -1.739 |
ンジ | -1.701 |
Token | this |
Token ID | 428 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
atti | +0.677 |
etti | +0.653 |
karma | +0.648 |
latex | +0.643 |
strap | +0.603 |
Neg logit contributions | |
GBT | -0.888 |
Catal | -0.787 |
Bah | -0.779 |
Franç | -0.768 |
ンジ | -0.744 |
Token | post |
Token ID | 1281 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
atti | +0.365 |
etti | +0.350 |
karma | +0.348 |
latex | +0.346 |
strap | +0.322 |
Neg logit contributions | |
GBT | -0.509 |
Catal | -0.453 |
Bah | -0.448 |
Franç | -0.442 |
ンジ | -0.429 |
Token | I |
Token ID | 314 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
atti | +0.734 |
etti | +0.710 |
karma | +0.706 |
latex | +0.701 |
strap | +0.662 |
Neg logit contributions | |
GBT | -0.756 |
Catal | -0.660 |
Bah | -0.651 |
Franç | -0.641 |
ンジ | -0.619 |
Token | 'll |
Token ID | 1183 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
atti | +1.487 |
etti | +1.425 |
karma | +1.411 |
latex | +1.389 |
strap | +1.307 |
Neg logit contributions | |
GBT | -2.273 |
Catal | -2.030 |
Bah | -2.012 |
Franç | -1.993 |
ンジ | -1.936 |
Token | round |
Token ID | 2835 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
atti | +0.953 |
karma | +0.927 |
etti | +0.925 |
latex | +0.919 |
strap | +0.857 |
Neg logit contributions | |
GBT | -1.269 |
Catal | -1.124 |
Bah | -1.116 |
Franç | -1.104 |
ンジ | -1.071 |
Token | up |
Token ID | 510 |
Feature activation | +5.68e+00 |
Pos logit contributions | |
atti | +1.734 |
karma | +1.699 |
etti | +1.683 |
latex | +1.682 |
strap | +1.607 |
Neg logit contributions | |
GBT | -1.581 |
Catal | -1.356 |
Bah | -1.332 |
Franç | -1.311 |
ンジ | -1.264 |
Token | some |
Token ID | 617 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
atti | +3.009 |
karma | +2.973 |
latex | +2.929 |
etti | +2.892 |
strap | +2.688 |
Neg logit contributions | |
GBT | -5.775 |
Catal | -5.166 |
Bah | -5.140 |
Franç | -5.068 |
ンジ | -5.002 |
Token | got |
Token ID | 1392 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
atti | +1.233 |
karma | +1.186 |
etti | +1.158 |
latex | +1.152 |
strap | +1.044 |
Neg logit contributions | |
GBT | -2.372 |
Catal | -2.148 |
Bah | -2.126 |
Franç | -2.108 |
ンジ | -2.050 |
Token | cha |
Token ID | 11693 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
atti | +0.892 |
karma | +0.853 |
etti | +0.846 |
latex | +0.826 |
strap | +0.778 |
Neg logit contributions | |
GBT | -1.501 |
Catal | -1.348 |
Bah | -1.322 |
Franç | -1.320 |
ンジ | -1.273 |
Token | 's |
Token ID | 338 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
atti | +0.465 |
karma | +0.452 |
etti | +0.452 |
latex | +0.445 |
strap | +0.422 |
Neg logit contributions | |
GBT | -0.420 |
Catal | -0.360 |
Bah | -0.356 |
Franç | -0.351 |
ンジ | -0.338 |
Token | I |
Token ID | 314 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
atti | +2.416 |
etti | +2.333 |
karma | +2.331 |
latex | +2.300 |
strap | +2.165 |
Neg logit contributions | |
GBT | -2.671 |
Catal | -2.319 |
Bah | -2.285 |
Franç | -2.254 |
ンジ | -2.197 |
Token | government |
Token ID | 1230 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
atti | +0.573 |
etti | +0.552 |
latex | +0.549 |
karma | +0.549 |
Neighborhood | +0.518 |
Neg logit contributions | |
GBT | -0.576 |
Catal | -0.498 |
Bah | -0.489 |
Franç | -0.481 |
ンジ | -0.475 |
Token | and |
Token ID | 290 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
atti | +0.109 |
etti | +0.105 |
karma | +0.104 |
latex | +0.104 |
strap | +0.098 |
Neg logit contributions | |
GBT | -0.116 |
Catal | -0.101 |
Bah | -0.100 |
Franç | -0.098 |
ンジ | -0.095 |
Token | militants |
Token ID | 13162 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
atti | +0.841 |
etti | +0.810 |
karma | +0.810 |
latex | +0.808 |
Neighborhood | +0.761 |
Neg logit contributions | |
GBT | -0.903 |
Catal | -0.784 |
Bah | -0.774 |
Franç | -0.761 |
ンジ | -0.753 |
Token | have |
Token ID | 423 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
GBT | +0.943 |
Catal | +0.836 |
Bah | +0.827 |
Franç | +0.816 |
ンジ | +0.793 |
Neg logit contributions | |
atti | -0.703 |
etti | -0.677 |
karma | -0.671 |
latex | -0.666 |
strap | -0.623 |
Token | stalled |
Token ID | 29759 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
GBT | +0.608 |
Catal | +0.540 |
Bah | +0.534 |
Franç | +0.527 |
ンジ | +0.511 |
Neg logit contributions | |
atti | -0.440 |
etti | -0.423 |
karma | -0.419 |
latex | -0.416 |
strap | -0.389 |
Token |
|
Token ID | 198 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
GBT | +1.244 |
Catal | +1.097 |
Bah | +1.083 |
Franç | +1.070 |
ンジ | +1.024 |
Neg logit contributions | |
atti | -1.162 |
etti | -1.120 |
karma | -1.115 |
latex | -1.106 |
strap | -1.045 |
Token |
|
Token ID | 198 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
atti | +3.416 |
etti | +3.247 |
karma | +3.229 |
latex | +3.029 |
Dustin | +3.005 |
Neg logit contributions | |
GBT | -5.233 |
ンジ | -4.499 |
Franç | -4.498 |
Catal | -4.457 |
Bah | -4.412 |
Token | Pak |
Token ID | 29675 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
atti | +3.830 |
etti | +3.637 |
karma | +3.482 |
toe | +3.418 |
latex | +3.415 |
Neg logit contributions | |
GBT | -3.979 |
Franç | -3.539 |
ンジ | -3.386 |
Goku | -3.322 |
Catal | -3.280 |
Token | istani |
Token ID | 16688 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
atti | +0.668 |
etti | +0.649 |
karma | +0.643 |
latex | +0.639 |
strap | +0.609 |
Neg logit contributions | |
GBT | -0.578 |
Catal | -0.497 |
Bah | -0.489 |
Franç | -0.484 |
ンジ | -0.462 |
Token | fighter |
Token ID | 10543 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
GBT | +0.164 |
Catal | +0.145 |
Bah | +0.144 |
Franç | +0.142 |
ンジ | +0.137 |
Neg logit contributions | |
atti | -0.125 |
etti | -0.120 |
karma | -0.119 |
latex | -0.118 |
strap | -0.111 |
Token | jets |
Token ID | 20792 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
atti | +0.460 |
etti | +0.438 |
karma | +0.433 |
latex | +0.429 |
strap | +0.391 |
Neg logit contributions | |
GBT | -0.991 |
Catal | -0.897 |
Bah | -0.885 |
Franç | -0.875 |
ンジ | -0.859 |
Token | above |
Token ID | 2029 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
atti | +0.681 |
etti | +0.659 |
karma | +0.655 |
latex | +0.652 |
strap | +0.617 |
Neg logit contributions | |
GBT | -0.671 |
Catal | -0.586 |
Bah | -0.578 |
Franç | -0.569 |
ンジ | -0.548 |
Token | the |
Token ID | 262 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
atti | +0.219 |
etti | +0.210 |
karma | +0.208 |
latex | +0.208 |
strap | +0.193 |
Neg logit contributions | |
GBT | -0.303 |
Catal | -0.269 |
Bah | -0.267 |
Franç | -0.263 |
ンジ | -0.256 |
Token | Earth |
Token ID | 3668 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
GBT | +0.371 |
Catal | +0.327 |
Bah | +0.323 |
Franç | +0.319 |
ンジ | +0.310 |
Neg logit contributions | |
atti | -0.295 |
etti | -0.284 |
karma | -0.283 |
latex | -0.282 |
strap | -0.263 |
Token | 's |
Token ID | 338 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
GBT | +0.246 |
Catal | +0.214 |
Bah | +0.211 |
Franç | +0.208 |
ンジ | +0.202 |
Neg logit contributions | |
atti | -0.247 |
etti | -0.239 |
karma | -0.237 |
latex | -0.236 |
strap | -0.223 |
Token | surface |
Token ID | 4417 |
Feature activation | +8.65e-01 |
Pos logit contributions | |
atti | +0.227 |
etti | +0.212 |
karma | +0.211 |
latex | +0.209 |
strap | +0.186 |
Neg logit contributions | |
GBT | -0.611 |
Catal | -0.557 |
Bah | -0.552 |
Franç | -0.547 |
ンジ | -0.534 |
Token |
|
Token ID | 198 |
Feature activation | +5.45e+00 |
Pos logit contributions | |
atti | +0.630 |
etti | +0.608 |
karma | +0.606 |
latex | +0.603 |
strap | +0.567 |
Neg logit contributions | |
GBT | -0.727 |
Catal | -0.637 |
Bah | -0.630 |
Franç | -0.620 |
ンジ | -0.603 |
Token |
|
Token ID | 198 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
atti | +3.424 |
etti | +3.286 |
karma | +3.247 |
latex | +3.083 |
strap | +3.031 |
Neg logit contributions | |
GBT | -5.038 |
Franç | -4.338 |
ンジ | -4.322 |
Catal | -4.313 |
Bah | -4.265 |
Token | The |
Token ID | 464 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
atti | +3.899 |
etti | +3.768 |
karma | +3.578 |
MIT | +3.560 |
strap | +3.516 |
Neg logit contributions | |
GBT | -3.993 |
Franç | -3.554 |
ンジ | -3.419 |
Catal | -3.335 |
Goku | -3.331 |
Token | European |
Token ID | 3427 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
atti | +0.957 |
etti | +0.922 |
karma | +0.917 |
latex | +0.914 |
strap | +0.852 |
Neg logit contributions | |
GBT | -1.089 |
Catal | -0.955 |
Bah | -0.945 |
Franç | -0.928 |
ンジ | -0.906 |
Token | Space |
Token ID | 4687 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
atti | +1.149 |
etti | +1.100 |
karma | +1.092 |
latex | +1.086 |
strap | +1.004 |
Neg logit contributions | |
GBT | -1.802 |
Catal | -1.612 |
Bah | -1.598 |
Franç | -1.574 |
ンジ | -1.532 |
Token | Agency |
Token ID | 7732 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
atti | +0.979 |
etti | +0.948 |
karma | +0.941 |
latex | +0.935 |
strap | +0.890 |
Neg logit contributions | |
GBT | -0.825 |
Catal | -0.708 |
Bah | -0.696 |
Franç | -0.681 |
ンジ | -0.659 |
Token | extent |
Token ID | 6287 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
GBT | +0.711 |
Catal | +0.631 |
Bah | +0.624 |
Franç | +0.616 |
ンジ | +0.597 |
Neg logit contributions | |
atti | -0.534 |
etti | -0.513 |
karma | -0.509 |
latex | -0.506 |
strap | -0.473 |
Token | of |
Token ID | 286 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
GBT | +0.655 |
Catal | +0.576 |
Bah | +0.569 |
Franç | +0.562 |
ンジ | +0.542 |
Neg logit contributions | |
atti | -0.565 |
etti | -0.545 |
karma | -0.541 |
latex | -0.538 |
strap | -0.506 |
Token | the |
Token ID | 262 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
atti | +0.353 |
etti | +0.340 |
karma | +0.339 |
latex | +0.337 |
strap | +0.313 |
Neg logit contributions | |
GBT | -0.480 |
Catal | -0.425 |
Bah | -0.420 |
Franç | -0.415 |
ンジ | -0.404 |
Token | rubbish |
Token ID | 35022 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
atti | +0.268 |
etti | +0.259 |
karma | +0.259 |
latex | +0.258 |
Neighborhood | +0.238 |
Neg logit contributions | |
GBT | -0.360 |
Catal | -0.318 |
Bah | -0.313 |
Franç | -0.309 |
ンジ | -0.302 |
Token |
|
Token ID | 198 |
Feature activation | +5.80e+00 |
Pos logit contributions | |
atti | +0.225 |
etti | +0.216 |
karma | +0.215 |
latex | +0.214 |
strap | +0.201 |
Neg logit contributions | |
GBT | -0.264 |
Catal | -0.232 |
Bah | -0.229 |
Franç | -0.226 |
ンジ | -0.219 |
Token |
|
Token ID | 198 |
Feature activation | +5.44e+00 |
Pos logit contributions | |
atti | +3.545 |
etti | +3.385 |
karma | +3.370 |
latex | +3.185 |
Dustin | +3.139 |
Neg logit contributions | |
GBT | -5.467 |
Franç | -4.688 |
ンジ | -4.687 |
Catal | -4.664 |
Bah | -4.607 |
Token | The |
Token ID | 464 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
atti | +3.973 |
etti | +3.795 |
karma | +3.641 |
latex | +3.586 |
strap | +3.536 |
Neg logit contributions | |
GBT | -4.383 |
Franç | -3.900 |
ンジ | -3.697 |
Goku | -3.653 |
Catal | -3.619 |
Token | cost |
Token ID | 1575 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
atti | +0.781 |
etti | +0.753 |
karma | +0.750 |
latex | +0.746 |
Neighborhood | +0.695 |
Neg logit contributions | |
GBT | -0.922 |
Catal | -0.808 |
Bah | -0.801 |
Franç | -0.789 |
ンジ | -0.765 |
Token | of |
Token ID | 286 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
atti | +1.022 |
etti | +0.981 |
karma | +0.975 |
latex | +0.971 |
strap | +0.912 |
Neg logit contributions | |
GBT | -1.224 |
Catal | -1.075 |
Bah | -1.061 |
Franç | -1.047 |
ンジ | -1.015 |
Token | a |
Token ID | 257 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
atti | +0.151 |
karma | +0.145 |
etti | +0.145 |
latex | +0.144 |
strap | +0.133 |
Neg logit contributions | |
GBT | -0.223 |
Catal | -0.198 |
Bah | -0.196 |
Franç | -0.193 |
ンジ | -0.188 |
Token | clean |
Token ID | 3424 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
atti | +0.662 |
etti | +0.635 |
karma | +0.634 |
latex | +0.631 |
strap | +0.588 |
Neg logit contributions | |
GBT | -0.835 |
Catal | -0.738 |
Bah | -0.729 |
Franç | -0.719 |
ンジ | -0.697 |
Token | this |
Token ID | 428 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
GBT | +0.124 |
Catal | +0.109 |
Bah | +0.108 |
Franç | +0.107 |
ンジ | +0.103 |
Neg logit contributions | |
atti | -0.103 |
etti | -0.099 |
karma | -0.098 |
latex | -0.098 |
strap | -0.092 |
Token | series |
Token ID | 2168 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
GBT | +0.291 |
Catal | +0.256 |
Bah | +0.253 |
Franç | +0.249 |
ンジ | +0.242 |
Neg logit contributions | |
atti | -0.252 |
etti | -0.243 |
karma | -0.242 |
latex | -0.240 |
strap | -0.226 |
Token | does |
Token ID | 857 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
GBT | +0.198 |
Catal | +0.175 |
Bah | +0.172 |
Franç | +0.170 |
ンジ | +0.164 |
Neg logit contributions | |
atti | -0.165 |
etti | -0.159 |
karma | -0.158 |
latex | -0.157 |
strap | -0.147 |
Token | not |
Token ID | 407 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
GBT | +0.269 |
Catal | +0.238 |
Bah | +0.235 |
Franç | +0.232 |
ンジ | +0.224 |
Neg logit contributions | |
atti | -0.221 |
etti | -0.213 |
karma | -0.211 |
latex | -0.210 |
strap | -0.197 |
Token | shy |
Token ID | 15800 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
atti | +0.042 |
etti | +0.040 |
karma | +0.040 |
latex | +0.040 |
strap | +0.037 |
Neg logit contributions | |
GBT | -0.060 |
Catal | -0.053 |
Bah | -0.053 |
Franç | -0.052 |
ンジ | -0.050 |
Token | away |
Token ID | 1497 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
atti | +1.135 |
etti | +1.093 |
karma | +1.078 |
latex | +1.072 |
strap | +1.005 |
Neg logit contributions | |
GBT | -1.814 |
Catal | -1.621 |
Bah | -1.602 |
Franç | -1.581 |
ンジ | -1.541 |
Token | from |
Token ID | 422 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
atti | +3.180 |
etti | +3.123 |
latex | +3.103 |
karma | +3.083 |
strap | +2.852 |
Neg logit contributions | |
GBT | -5.066 |
Catal | -4.490 |
Bah | -4.398 |
Franç | -4.324 |
ンジ | -4.218 |
Token | the |
Token ID | 262 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
atti | +1.593 |
karma | +1.549 |
latex | +1.537 |
etti | +1.535 |
Neighborhood | +1.408 |
Neg logit contributions | |
GBT | -2.219 |
Catal | -1.984 |
Bah | -1.950 |
Franç | -1.907 |
ンジ | -1.877 |
Token | brutality |
Token ID | 24557 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
atti | +0.448 |
etti | +0.431 |
karma | +0.430 |
latex | +0.427 |
strap | +0.396 |
Neg logit contributions | |
GBT | -0.607 |
Catal | -0.542 |
Bah | -0.534 |
Franç | -0.526 |
ンジ | -0.512 |
Token | of |
Token ID | 286 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
GBT | +0.809 |
Catal | +0.710 |
Bah | +0.701 |
Franç | +0.691 |
ンジ | +0.667 |
Neg logit contributions | |
atti | -0.734 |
etti | -0.709 |
karma | -0.704 |
latex | -0.700 |
strap | -0.660 |
Token | nature |
Token ID | 3450 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
atti | +0.735 |
etti | +0.709 |
karma | +0.707 |
latex | +0.702 |
strap | +0.653 |
Neg logit contributions | |
GBT | -0.959 |
Catal | -0.852 |
Bah | -0.842 |
Franç | -0.827 |
ンジ | -0.804 |
Token | battle |
Token ID | 3344 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
GBT | +0.104 |
Catal | +0.092 |
Bah | +0.091 |
Franç | +0.090 |
ンジ | +0.087 |
Neg logit contributions | |
atti | -0.073 |
etti | -0.070 |
karma | -0.069 |
latex | -0.069 |
strap | -0.064 |
Token | - |
Token ID | 12 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
atti | +0.602 |
etti | +0.585 |
karma | +0.581 |
latex | +0.576 |
strap | +0.550 |
Neg logit contributions | |
GBT | -0.479 |
Catal | -0.410 |
Bah | -0.403 |
Franç | -0.395 |
ンジ | -0.379 |
Token | front |
Token ID | 8534 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
GBT | +0.201 |
Catal | +0.178 |
Bah | +0.176 |
Franç | +0.174 |
ンジ | +0.168 |
Neg logit contributions | |
atti | -0.164 |
etti | -0.158 |
karma | -0.157 |
latex | -0.157 |
strap | -0.146 |
Token | s |
Token ID | 82 |
Feature activation | +8.98e-01 |
Pos logit contributions | |
GBT | +0.180 |
Catal | +0.158 |
Bah | +0.156 |
Franç | +0.154 |
ンジ | +0.149 |
Neg logit contributions | |
atti | -0.154 |
etti | -0.149 |
karma | -0.148 |
latex | -0.147 |
strap | -0.138 |
Token | " |
Token ID | 1 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
atti | +0.621 |
etti | +0.600 |
karma | +0.594 |
latex | +0.589 |
strap | +0.553 |
Neg logit contributions | |
GBT | -0.789 |
Catal | -0.697 |
Bah | -0.688 |
Franç | -0.675 |
ンジ | -0.658 |
Token |
|
Token ID | 198 |
Feature activation | +5.32e+00 |
Pos logit contributions | |
GBT | +0.332 |
Catal | +0.291 |
Bah | +0.288 |
Franç | +0.284 |
ンジ | +0.274 |
Neg logit contributions | |
atti | -0.298 |
etti | -0.287 |
karma | -0.285 |
latex | -0.284 |
strap | -0.267 |
Token |
|
Token ID | 198 |
Feature activation | +5.05e+00 |
Pos logit contributions | |
atti | +3.325 |
etti | +3.182 |
karma | +3.163 |
latex | +2.992 |
Dustin | +2.937 |
Neg logit contributions | |
GBT | -4.937 |
Franç | -4.248 |
ンジ | -4.230 |
Catal | -4.219 |
Bah | -4.182 |
Token | Ent |
Token ID | 14539 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
atti | +3.698 |
etti | +3.524 |
karma | +3.394 |
latex | +3.314 |
MIT | +3.283 |
Neg logit contributions | |
GBT | -4.034 |
Franç | -3.590 |
ンジ | -3.428 |
Goku | -3.368 |
Catal | -3.335 |
Token | ire |
Token ID | 557 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
atti | +0.731 |
etti | +0.703 |
karma | +0.696 |
latex | +0.691 |
strap | +0.646 |
Neg logit contributions | |
GBT | -1.010 |
Catal | -0.898 |
Bah | -0.889 |
Franç | -0.877 |
ンジ | -0.850 |
Token | classrooms |
Token ID | 32579 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
GBT | +0.574 |
Catal | +0.506 |
Bah | +0.500 |
Franç | +0.493 |
ンジ | +0.476 |
Neg logit contributions | |
atti | -0.488 |
etti | -0.470 |
karma | -0.466 |
latex | -0.463 |
strap | -0.436 |
Token | of |
Token ID | 286 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
atti | +0.228 |
etti | +0.220 |
karma | +0.218 |
latex | +0.217 |
strap | +0.204 |
Neg logit contributions | |
GBT | -0.266 |
Catal | -0.235 |
Bah | -0.232 |
Franç | -0.228 |
ンジ | -0.221 |
Token | says |
Token ID | 1139 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
GBT | +0.402 |
Catal | +0.357 |
Bah | +0.353 |
Franç | +0.350 |
Goku | +0.337 |
Neg logit contributions | |
atti | -0.302 |
etti | -0.290 |
karma | -0.286 |
latex | -0.283 |
strap | -0.267 |
Token | curator |
Token ID | 46132 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
GBT | +1.069 |
Catal | +0.946 |
Bah | +0.931 |
Franç | +0.928 |
Goku | +0.888 |
Neg logit contributions | |
atti | -0.925 |
etti | -0.886 |
karma | -0.878 |
latex | -0.870 |
strap | -0.827 |
Token | Dr |
Token ID | 1583 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
GBT | +0.196 |
Catal | +0.173 |
Bah | +0.170 |
Franç | +0.169 |
ンジ | +0.161 |
Neg logit contributions | |
atti | -0.183 |
etti | -0.176 |
karma | -0.175 |
latex | -0.174 |
strap | -0.165 |
Token | Ian |
Token ID | 12930 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
GBT | +0.769 |
Catal | +0.678 |
Bah | +0.669 |
Franç | +0.663 |
ンジ | +0.634 |
Neg logit contributions | |
atti | -0.693 |
etti | -0.667 |
karma | -0.664 |
latex | -0.660 |
strap | -0.622 |
Token | Jenkins |
Token ID | 21835 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
GBT | +0.888 |
Catal | +0.780 |
Bah | +0.771 |
Franç | +0.761 |
ンジ | +0.732 |
Neg logit contributions | |
atti | -0.808 |
etti | -0.779 |
karma | -0.774 |
latex | -0.770 |
strap | -0.726 |
Token |
|
Token ID | 198 |
Feature activation | +5.32e+00 |
Pos logit contributions | |
GBT | +0.044 |
Catal | +0.039 |
Bah | +0.038 |
Franç | +0.038 |
ンジ | +0.036 |
Neg logit contributions | |
atti | -0.040 |
etti | -0.039 |
karma | -0.038 |
latex | -0.038 |
strap | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
atti | +3.331 |
etti | +3.202 |
karma | +3.173 |
latex | +3.006 |
strap | +2.950 |
Neg logit contributions | |
GBT | -4.922 |
Franç | -4.237 |
ンジ | -4.221 |
Catal | -4.195 |
Bah | -4.160 |
Token | The |
Token ID | 464 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
atti | +3.787 |
etti | +3.640 |
karma | +3.496 |
latex | +3.443 |
MIT | +3.391 |
Neg logit contributions | |
GBT | -4.152 |
Franç | -3.667 |
ンジ | -3.577 |
Goku | -3.486 |
Catal | -3.448 |
Token | art |
Token ID | 1242 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
atti | +1.078 |
etti | +1.040 |
karma | +1.035 |
latex | +1.032 |
strap | +0.963 |
Neg logit contributions | |
GBT | -1.235 |
Catal | -1.084 |
Bah | -1.073 |
Franç | -1.055 |
ンジ | -1.027 |
Token | of |
Token ID | 286 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
atti | +0.038 |
etti | +0.037 |
karma | +0.037 |
latex | +0.036 |
strap | +0.034 |
Neg logit contributions | |
GBT | -0.043 |
Catal | -0.037 |
Bah | -0.037 |
Franç | -0.036 |
ンジ | -0.035 |
Token | Ancient |
Token ID | 13406 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
atti | +0.154 |
etti | +0.148 |
karma | +0.146 |
latex | +0.145 |
strap | +0.135 |
Neg logit contributions | |
GBT | -0.229 |
Catal | -0.205 |
Bah | -0.202 |
Franç | -0.200 |
ンジ | -0.194 |
Token | the |
Token ID | 262 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
atti | +0.201 |
etti | +0.194 |
karma | +0.193 |
latex | +0.191 |
strap | +0.180 |
Neg logit contributions | |
GBT | -0.238 |
Catal | -0.210 |
Bah | -0.207 |
Franç | -0.204 |
ンジ | -0.198 |
Token | public |
Token ID | 1171 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
GBT | +0.506 |
Catal | +0.444 |
Bah | +0.439 |
Franç | +0.433 |
ンジ | +0.418 |
Neg logit contributions | |
atti | -0.453 |
etti | -0.438 |
karma | -0.435 |
latex | -0.432 |
strap | -0.406 |
Token | will |
Token ID | 481 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
GBT | +0.401 |
Catal | +0.356 |
Bah | +0.352 |
Franç | +0.347 |
ンジ | +0.337 |
Neg logit contributions | |
atti | -0.304 |
etti | -0.292 |
karma | -0.291 |
latex | -0.288 |
strap | -0.269 |
Token | say |
Token ID | 910 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
GBT | +0.210 |
Catal | +0.187 |
Bah | +0.184 |
Franç | +0.182 |
ンジ | +0.176 |
Neg logit contributions | |
atti | -0.154 |
etti | -0.148 |
karma | -0.147 |
latex | -0.146 |
strap | -0.136 |
Token | " |
Token ID | 1 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
GBT | +0.802 |
Catal | +0.708 |
Bah | +0.699 |
Franç | +0.691 |
ンジ | +0.664 |
Neg logit contributions | |
atti | -0.690 |
etti | -0.664 |
karma | -0.657 |
latex | -0.655 |
strap | -0.617 |
Token |
|
Token ID | 198 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
atti | +0.165 |
etti | +0.159 |
karma | +0.158 |
latex | +0.156 |
Neighborhood | +0.147 |
Neg logit contributions | |
GBT | -0.195 |
Catal | -0.171 |
Bah | -0.169 |
Franç | -0.167 |
ンジ | -0.162 |
Token |
|
Token ID | 198 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
atti | +3.312 |
etti | +3.160 |
karma | +3.144 |
latex | +2.974 |
Dustin | +2.922 |
Neg logit contributions | |
GBT | -4.963 |
Franç | -4.257 |
ンジ | -4.244 |
Catal | -4.235 |
Bah | -4.204 |
Token | Lord |
Token ID | 22438 |
Feature activation | -6.43e-01 |
Pos logit contributions | |
atti | +3.693 |
etti | +3.541 |
karma | +3.419 |
toe | +3.337 |
latex | +3.323 |
Neg logit contributions | |
GBT | -3.549 |
Franç | -3.131 |
Goku | -2.991 |
Bah | -2.982 |
ンジ | -2.977 |
Token | Hes |
Token ID | 44773 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
GBT | +0.579 |
Catal | +0.515 |
Bah | +0.509 |
Franç | +0.503 |
ンジ | +0.487 |
Neg logit contributions | |
atti | -0.429 |
etti | -0.412 |
karma | -0.409 |
latex | -0.406 |
strap | -0.380 |
Token | elt |
Token ID | 2120 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
atti | +0.281 |
karma | +0.274 |
latex | +0.273 |
etti | +0.273 |
strap | +0.259 |
Neg logit contributions | |
GBT | -0.180 |
Catal | -0.151 |
Bah | -0.149 |
Franç | -0.146 |
Goku | -0.137 |
Token | ine |
Token ID | 500 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
atti | +0.899 |
etti | +0.859 |
karma | +0.840 |
latex | +0.835 |
strap | +0.775 |
Neg logit contributions | |
GBT | -1.630 |
Catal | -1.463 |
Bah | -1.446 |
Franç | -1.429 |
ンジ | -1.395 |
Token | { |
Token ID | 1391 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
atti | +1.663 |
etti | +1.595 |
latex | +1.582 |
karma | +1.581 |
strap | +1.481 |
Neg logit contributions | |
GBT | -2.543 |
Catal | -2.242 |
Franç | -2.233 |
Bah | -2.228 |
ンジ | -2.158 |
Token | ... |
Token ID | 2644 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
GBT | +1.588 |
Catal | +1.399 |
Bah | +1.382 |
Franç | +1.362 |
ンジ | +1.317 |
Neg logit contributions | |
atti | -1.354 |
etti | -1.301 |
karma | -1.287 |
latex | -1.277 |
strap | -1.204 |
Token | prev |
Token ID | 8654 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
atti | +0.286 |
etti | +0.277 |
karma | +0.275 |
latex | +0.273 |
strap | +0.258 |
Neg logit contributions | |
GBT | -0.343 |
Catal | -0.303 |
Bah | -0.299 |
Franç | -0.296 |
ンジ | -0.285 |
Token | _ |
Token ID | 62 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
atti | +2.463 |
etti | +2.420 |
karma | +2.402 |
latex | +2.380 |
strap | +2.238 |
Neg logit contributions | |
GBT | -2.778 |
Catal | -2.440 |
Franç | -2.424 |
Bah | -2.383 |
Goku | -2.308 |
Token | state |
Token ID | 5219 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
atti | +1.172 |
etti | +1.130 |
karma | +1.113 |
latex | +1.097 |
strap | +1.047 |
Neg logit contributions | |
GBT | -1.893 |
Catal | -1.690 |
Franç | -1.670 |
Bah | -1.665 |
ンジ | -1.605 |
Token | : |
Token ID | 1058 |
Feature activation | +5.26e+00 |
Pos logit contributions | |
atti | +1.941 |
etti | +1.893 |
karma | +1.879 |
latex | +1.870 |
strap | +1.750 |
Neg logit contributions | |
GBT | -2.648 |
Catal | -2.348 |
Bah | -2.323 |
Franç | -2.321 |
ンジ | -2.233 |
Token | Option |
Token ID | 16018 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
atti | +2.871 |
karma | +2.785 |
etti | +2.773 |
latex | +2.762 |
Neighborhood | +2.645 |
Neg logit contributions | |
GBT | -5.202 |
Catal | -4.608 |
Bah | -4.590 |
Franç | -4.570 |
ンジ | -4.423 |
Token | < |
Token ID | 1279 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
atti | +3.057 |
etti | +2.980 |
karma | +2.949 |
latex | +2.949 |
strap | +2.813 |
Neg logit contributions | |
GBT | -3.082 |
Catal | -2.650 |
Franç | -2.650 |
Bah | -2.620 |
ンジ | -2.527 |
Token | Arc |
Token ID | 10173 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
atti | +2.428 |
etti | +2.363 |
karma | +2.327 |
latex | +2.314 |
Neighborhood | +2.227 |
Neg logit contributions | |
GBT | -2.145 |
Catal | -1.821 |
Franç | -1.816 |
Bah | -1.797 |
ンジ | -1.717 |
Token | < |
Token ID | 1279 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
atti | +1.339 |
etti | +1.306 |
latex | +1.290 |
karma | +1.289 |
strap | +1.229 |
Neg logit contributions | |
GBT | -1.015 |
Catal | -0.856 |
Bah | -0.849 |
Franç | -0.844 |
ンジ | -0.802 |
Token | Board |
Token ID | 5926 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
atti | +0.394 |
etti | +0.381 |
karma | +0.376 |
latex | +0.376 |
strap | +0.356 |
Neg logit contributions | |
GBT | -0.417 |
Catal | -0.363 |
Bah | -0.359 |
Franç | -0.356 |
ンジ | -0.343 |
Token | curator |
Token ID | 46132 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
GBT | +1.069 |
Catal | +0.946 |
Bah | +0.931 |
Franç | +0.928 |
Goku | +0.888 |
Neg logit contributions | |
atti | -0.925 |
etti | -0.886 |
karma | -0.878 |
latex | -0.870 |
strap | -0.827 |
Token | Dr |
Token ID | 1583 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
GBT | +0.196 |
Catal | +0.173 |
Bah | +0.170 |
Franç | +0.169 |
ンジ | +0.161 |
Neg logit contributions | |
atti | -0.183 |
etti | -0.176 |
karma | -0.175 |
latex | -0.174 |
strap | -0.165 |
Token | Ian |
Token ID | 12930 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
GBT | +0.769 |
Catal | +0.678 |
Bah | +0.669 |
Franç | +0.663 |
ンジ | +0.634 |
Neg logit contributions | |
atti | -0.693 |
etti | -0.667 |
karma | -0.664 |
latex | -0.660 |
strap | -0.622 |
Token | Jenkins |
Token ID | 21835 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
GBT | +0.888 |
Catal | +0.780 |
Bah | +0.771 |
Franç | +0.761 |
ンジ | +0.732 |
Neg logit contributions | |
atti | -0.808 |
etti | -0.779 |
karma | -0.774 |
latex | -0.770 |
strap | -0.726 |
Token |
|
Token ID | 198 |
Feature activation | +5.32e+00 |
Pos logit contributions | |
GBT | +0.044 |
Catal | +0.039 |
Bah | +0.038 |
Franç | +0.038 |
ンジ | +0.036 |
Neg logit contributions | |
atti | -0.040 |
etti | -0.039 |
karma | -0.038 |
latex | -0.038 |
strap | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
atti | +3.331 |
etti | +3.202 |
karma | +3.173 |
latex | +3.006 |
strap | +2.950 |
Neg logit contributions | |
GBT | -4.922 |
Franç | -4.237 |
ンジ | -4.221 |
Catal | -4.195 |
Bah | -4.160 |
Token | The |
Token ID | 464 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
atti | +3.787 |
etti | +3.640 |
karma | +3.496 |
latex | +3.443 |
MIT | +3.391 |
Neg logit contributions | |
GBT | -4.152 |
Franç | -3.667 |
ンジ | -3.577 |
Goku | -3.486 |
Catal | -3.448 |
Token | art |
Token ID | 1242 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
atti | +1.078 |
etti | +1.040 |
karma | +1.035 |
latex | +1.032 |
strap | +0.963 |
Neg logit contributions | |
GBT | -1.235 |
Catal | -1.084 |
Bah | -1.073 |
Franç | -1.055 |
ンジ | -1.027 |
Token | of |
Token ID | 286 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
atti | +0.038 |
etti | +0.037 |
karma | +0.037 |
latex | +0.036 |
strap | +0.034 |
Neg logit contributions | |
GBT | -0.043 |
Catal | -0.037 |
Bah | -0.037 |
Franç | -0.036 |
ンジ | -0.035 |
Token | Ancient |
Token ID | 13406 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
atti | +0.154 |
etti | +0.148 |
karma | +0.146 |
latex | +0.145 |
strap | +0.135 |
Neg logit contributions | |
GBT | -0.229 |
Catal | -0.205 |
Bah | -0.202 |
Franç | -0.200 |
ンジ | -0.194 |
Token | Greece |
Token ID | 10315 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
atti | +1.151 |
karma | +1.113 |
etti | +1.110 |
latex | +1.108 |
strap | +1.048 |
Neg logit contributions | |
GBT | -0.944 |
Catal | -0.802 |
Bah | -0.795 |
Franç | -0.776 |
ンジ | -0.751 |
Token | the |
Token ID | 262 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
atti | +0.219 |
etti | +0.210 |
karma | +0.208 |
latex | +0.208 |
strap | +0.193 |
Neg logit contributions | |
GBT | -0.303 |
Catal | -0.269 |
Bah | -0.267 |
Franç | -0.263 |
ンジ | -0.256 |
Token | Earth |
Token ID | 3668 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
GBT | +0.371 |
Catal | +0.327 |
Bah | +0.323 |
Franç | +0.319 |
ンジ | +0.310 |
Neg logit contributions | |
atti | -0.295 |
etti | -0.284 |
karma | -0.283 |
latex | -0.282 |
strap | -0.263 |
Token | 's |
Token ID | 338 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
GBT | +0.246 |
Catal | +0.214 |
Bah | +0.211 |
Franç | +0.208 |
ンジ | +0.202 |
Neg logit contributions | |
atti | -0.247 |
etti | -0.239 |
karma | -0.237 |
latex | -0.236 |
strap | -0.223 |
Token | surface |
Token ID | 4417 |
Feature activation | +8.65e-01 |
Pos logit contributions | |
atti | +0.227 |
etti | +0.212 |
karma | +0.211 |
latex | +0.209 |
strap | +0.186 |
Neg logit contributions | |
GBT | -0.611 |
Catal | -0.557 |
Bah | -0.552 |
Franç | -0.547 |
ンジ | -0.534 |
Token |
|
Token ID | 198 |
Feature activation | +5.45e+00 |
Pos logit contributions | |
atti | +0.630 |
etti | +0.608 |
karma | +0.606 |
latex | +0.603 |
strap | +0.567 |
Neg logit contributions | |
GBT | -0.727 |
Catal | -0.637 |
Bah | -0.630 |
Franç | -0.620 |
ンジ | -0.603 |
Token |
|
Token ID | 198 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
atti | +3.424 |
etti | +3.286 |
karma | +3.247 |
latex | +3.083 |
strap | +3.031 |
Neg logit contributions | |
GBT | -5.038 |
Franç | -4.338 |
ンジ | -4.322 |
Catal | -4.313 |
Bah | -4.265 |
Token | The |
Token ID | 464 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
atti | +3.899 |
etti | +3.768 |
karma | +3.578 |
MIT | +3.560 |
strap | +3.516 |
Neg logit contributions | |
GBT | -3.993 |
Franç | -3.554 |
ンジ | -3.419 |
Catal | -3.335 |
Goku | -3.331 |
Token | European |
Token ID | 3427 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
atti | +0.957 |
etti | +0.922 |
karma | +0.917 |
latex | +0.914 |
strap | +0.852 |
Neg logit contributions | |
GBT | -1.089 |
Catal | -0.955 |
Bah | -0.945 |
Franç | -0.928 |
ンジ | -0.906 |
Token | Space |
Token ID | 4687 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
atti | +1.149 |
etti | +1.100 |
karma | +1.092 |
latex | +1.086 |
strap | +1.004 |
Neg logit contributions | |
GBT | -1.802 |
Catal | -1.612 |
Bah | -1.598 |
Franç | -1.574 |
ンジ | -1.532 |
Token | Agency |
Token ID | 7732 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
atti | +0.979 |
etti | +0.948 |
karma | +0.941 |
latex | +0.935 |
strap | +0.890 |
Neg logit contributions | |
GBT | -0.825 |
Catal | -0.708 |
Bah | -0.696 |
Franç | -0.681 |
ンジ | -0.659 |
Token | ( |
Token ID | 357 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
atti | +1.684 |
etti | +1.625 |
latex | +1.604 |
karma | +1.600 |
strap | +1.514 |
Neg logit contributions | |
GBT | -1.980 |
Catal | -1.757 |
Bah | -1.729 |
Franç | -1.692 |
ンジ | -1.677 |
Token | download |
Token ID | 4321 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
atti | +1.228 |
etti | +1.184 |
karma | +1.175 |
latex | +1.168 |
strap | +1.100 |
Neg logit contributions | |
GBT | -1.408 |
Catal | -1.240 |
Bah | -1.225 |
Franç | -1.209 |
ンジ | -1.166 |
Token | . |
Token ID | 13 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
atti | +2.183 |
karma | +2.129 |
etti | +2.103 |
latex | +2.063 |
strap | +1.998 |
Neg logit contributions | |
GBT | -2.424 |
Catal | -2.110 |
Bah | -2.068 |
ンジ | -2.008 |
Franç | -2.004 |
Token | Includes |
Token ID | 29581 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
atti | +0.538 |
latex | +0.531 |
etti | +0.524 |
karma | +0.523 |
Neighborhood | +0.495 |
Neg logit contributions | |
GBT | -0.731 |
Catal | -0.637 |
Bah | -0.612 |
ンジ | -0.610 |
Franç | -0.599 |
Token | the |
Token ID | 262 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
atti | +1.429 |
etti | +1.372 |
karma | +1.368 |
latex | +1.357 |
strap | +1.262 |
Neg logit contributions | |
GBT | -2.156 |
Catal | -1.921 |
Bah | -1.901 |
Franç | -1.875 |
ンジ | -1.830 |
Token | Catholic |
Token ID | 7835 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
atti | +2.124 |
latex | +2.037 |
etti | +2.028 |
karma | +2.019 |
Neighborhood | +1.899 |
Neg logit contributions | |
GBT | -1.762 |
Catal | -1.484 |
Bah | -1.469 |
Franç | -1.460 |
ンジ | -1.404 |
Token | Encyclopedia |
Token ID | 36732 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
atti | +1.610 |
latex | +1.585 |
etti | +1.570 |
karma | +1.561 |
strap | +1.493 |
Neg logit contributions | |
GBT | -1.456 |
Catal | -1.235 |
Bah | -1.214 |
Franç | -1.193 |
ンジ | -1.174 |
Token | , |
Token ID | 11 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
atti | +3.714 |
karma | +3.586 |
latex | +3.553 |
etti | +3.519 |
strap | +3.281 |
Neg logit contributions | |
GBT | -4.446 |
Catal | -3.734 |
Bah | -3.714 |
ンジ | -3.648 |
Franç | -3.612 |
Token | Church |
Token ID | 4564 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
atti | +2.160 |
etti | +2.071 |
latex | +2.065 |
karma | +2.050 |
strap | +1.945 |
Neg logit contributions | |
GBT | -2.083 |
ンジ | -1.770 |
Bah | -1.725 |
Catal | -1.704 |
Franç | -1.649 |
Token | Fathers |
Token ID | 30969 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
atti | +1.260 |
etti | +1.194 |
latex | +1.182 |
karma | +1.182 |
strap | +1.082 |
Neg logit contributions | |
GBT | -2.780 |
Catal | -2.508 |
Bah | -2.486 |
Franç | -2.453 |
ンジ | -2.408 |
Token | , |
Token ID | 11 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
atti | +1.231 |
etti | +1.189 |
karma | +1.179 |
latex | +1.171 |
strap | +1.110 |
Neg logit contributions | |
GBT | -1.054 |
Catal | -0.917 |
Bah | -0.903 |
Franç | -0.885 |
ンジ | -0.852 |
Token | Sum |
Token ID | 5060 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
atti | +0.831 |
etti | +0.801 |
karma | +0.800 |
latex | +0.796 |
strap | +0.746 |
Neg logit contributions | |
GBT | -0.933 |
Catal | -0.813 |
Bah | -0.803 |
Franç | -0.791 |
ンジ | -0.773 |
Token | image |
Token ID | 2939 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
atti | +2.036 |
latex | +1.976 |
karma | +1.971 |
etti | +1.968 |
strap | +1.881 |
Neg logit contributions | |
GBT | -1.740 |
Catal | -1.487 |
Bah | -1.464 |
Franç | -1.441 |
ンジ | -1.414 |
Token |
|
Token ID | 198 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
atti | +1.601 |
etti | +1.539 |
karma | +1.531 |
latex | +1.527 |
strap | +1.434 |
Neg logit contributions | |
GBT | -1.874 |
Catal | -1.639 |
Bah | -1.631 |
Franç | -1.597 |
ンジ | -1.556 |
Token |
|
Token ID | 198 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
atti | +1.236 |
etti | +1.190 |
karma | +1.180 |
latex | +1.156 |
strap | +1.111 |
Neg logit contributions | |
GBT | -1.540 |
Catal | -1.344 |
Bah | -1.326 |
Franç | -1.323 |
ンジ | -1.290 |
Token | larg |
Token ID | 15521 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
atti | +1.066 |
etti | +1.012 |
karma | +0.992 |
latex | +0.978 |
strap | +0.929 |
Neg logit contributions | |
GBT | -2.121 |
Franç | -1.899 |
Catal | -1.897 |
Bah | -1.895 |
ンジ | -1.839 |
Token | er |
Token ID | 263 |
Feature activation | +4.64e+00 |
Pos logit contributions | |
atti | +1.600 |
etti | +1.554 |
karma | +1.532 |
latex | +1.520 |
strap | +1.437 |
Neg logit contributions | |
GBT | -1.813 |
Catal | -1.591 |
Bah | -1.574 |
Franç | -1.566 |
ンジ | -1.509 |
Token | image |
Token ID | 2939 |
Feature activation | +5.12e+00 |
Pos logit contributions | |
atti | +3.737 |
latex | +3.647 |
karma | +3.638 |
etti | +3.630 |
strap | +3.444 |
Neg logit contributions | |
GBT | -3.507 |
Catal | -2.998 |
Bah | -2.950 |
Franç | -2.938 |
ンジ | -2.886 |
Token | T |
Token ID | 309 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
atti | +3.035 |
etti | +2.906 |
latex | +2.890 |
karma | +2.869 |
Dustin | +2.650 |
Neg logit contributions | |
GBT | -4.958 |
Catal | -4.437 |
Bah | -4.429 |
Franç | -4.304 |
ンジ | -4.252 |
Token | IFF |
Token ID | 29267 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
atti | +2.060 |
etti | +1.983 |
karma | +1.951 |
latex | +1.941 |
strap | +1.853 |
Neg logit contributions | |
GBT | -1.743 |
Catal | -1.513 |
Bah | -1.487 |
Franç | -1.448 |
ンジ | -1.400 |
Token | original |
Token ID | 2656 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
GBT | +0.327 |
Catal | +0.291 |
Bah | +0.288 |
Franç | +0.288 |
ンジ | +0.274 |
Neg logit contributions | |
atti | -0.269 |
etti | -0.259 |
karma | -0.255 |
latex | -0.253 |
strap | -0.238 |
Token | image |
Token ID | 2939 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
atti | +0.974 |
etti | +0.946 |
karma | +0.942 |
latex | +0.939 |
strap | +0.894 |
Neg logit contributions | |
GBT | -0.756 |
Catal | -0.644 |
Bah | -0.635 |
Franç | -0.621 |
ンジ | -0.597 |
Token | Download |
Token ID | 10472 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
atti | +2.418 |
etti | +2.336 |
karma | +2.334 |
latex | +2.318 |
strap | +2.207 |
Neg logit contributions | |
GBT | -2.170 |
Bah | -1.857 |
Catal | -1.856 |
Franç | -1.790 |
ンジ | -1.755 |
Token | and |
Token ID | 290 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
atti | +0.109 |
etti | +0.105 |
karma | +0.104 |
latex | +0.104 |
strap | +0.098 |
Neg logit contributions | |
GBT | -0.116 |
Catal | -0.101 |
Bah | -0.100 |
Franç | -0.098 |
ンジ | -0.095 |
Token | militants |
Token ID | 13162 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
atti | +0.841 |
etti | +0.810 |
karma | +0.810 |
latex | +0.808 |
Neighborhood | +0.761 |
Neg logit contributions | |
GBT | -0.903 |
Catal | -0.784 |
Bah | -0.774 |
Franç | -0.761 |
ンジ | -0.753 |
Token | have |
Token ID | 423 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
GBT | +0.943 |
Catal | +0.836 |
Bah | +0.827 |
Franç | +0.816 |
ンジ | +0.793 |
Neg logit contributions | |
atti | -0.703 |
etti | -0.677 |
karma | -0.671 |
latex | -0.666 |
strap | -0.623 |
Token | stalled |
Token ID | 29759 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
GBT | +0.608 |
Catal | +0.540 |
Bah | +0.534 |
Franç | +0.527 |
ンジ | +0.511 |
Neg logit contributions | |
atti | -0.440 |
etti | -0.423 |
karma | -0.419 |
latex | -0.416 |
strap | -0.389 |
Token |
|
Token ID | 198 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
GBT | +1.244 |
Catal | +1.097 |
Bah | +1.083 |
Franç | +1.070 |
ンジ | +1.024 |
Neg logit contributions | |
atti | -1.162 |
etti | -1.120 |
karma | -1.115 |
latex | -1.106 |
strap | -1.045 |
Token |
|
Token ID | 198 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
atti | +3.416 |
etti | +3.247 |
karma | +3.229 |
latex | +3.029 |
Dustin | +3.005 |
Neg logit contributions | |
GBT | -5.233 |
ンジ | -4.499 |
Franç | -4.498 |
Catal | -4.457 |
Bah | -4.412 |
Token | Pak |
Token ID | 29675 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
atti | +3.830 |
etti | +3.637 |
karma | +3.482 |
toe | +3.418 |
latex | +3.415 |
Neg logit contributions | |
GBT | -3.979 |
Franç | -3.539 |
ンジ | -3.386 |
Goku | -3.322 |
Catal | -3.280 |
Token | istani |
Token ID | 16688 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
atti | +0.668 |
etti | +0.649 |
karma | +0.643 |
latex | +0.639 |
strap | +0.609 |
Neg logit contributions | |
GBT | -0.578 |
Catal | -0.497 |
Bah | -0.489 |
Franç | -0.484 |
ンジ | -0.462 |
Token | fighter |
Token ID | 10543 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
GBT | +0.164 |
Catal | +0.145 |
Bah | +0.144 |
Franç | +0.142 |
ンジ | +0.137 |
Neg logit contributions | |
atti | -0.125 |
etti | -0.120 |
karma | -0.119 |
latex | -0.118 |
strap | -0.111 |
Token | jets |
Token ID | 20792 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
atti | +0.460 |
etti | +0.438 |
karma | +0.433 |
latex | +0.429 |
strap | +0.391 |
Neg logit contributions | |
GBT | -0.991 |
Catal | -0.897 |
Bah | -0.885 |
Franç | -0.875 |
ンジ | -0.859 |
Token | have |
Token ID | 423 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
atti | +0.436 |
etti | +0.417 |
karma | +0.413 |
latex | +0.410 |
strap | +0.381 |
Neg logit contributions | |
GBT | -0.709 |
Catal | -0.634 |
Bah | -0.626 |
Franç | -0.618 |
ンジ | -0.603 |
Token | FX |
Token ID | 17213 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
GBT | +0.028 |
Catal | +0.024 |
Bah | +0.024 |
Franç | +0.024 |
ンジ | +0.023 |
Neg logit contributions | |
atti | -0.026 |
etti | -0.025 |
karma | -0.025 |
latex | -0.025 |
strap | -0.023 |
Token | O |
Token ID | 46 |
Feature activation | -7.18e-02 |
Pos logit contributions | |
atti | +1.164 |
etti | +1.124 |
karma | +1.120 |
latex | +1.113 |
Neighborhood | +1.041 |
Neg logit contributions | |
GBT | -1.300 |
Catal | -1.149 |
Bah | -1.118 |
Franç | -1.110 |
ンジ | -1.080 |
Token | Korea |
Token ID | 4969 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
GBT | +0.059 |
Catal | +0.052 |
Bah | +0.051 |
Franç | +0.051 |
ンジ | +0.049 |
Neg logit contributions | |
atti | -0.053 |
etti | -0.052 |
karma | -0.051 |
latex | -0.051 |
strap | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
atti | +0.271 |
etti | +0.262 |
karma | +0.260 |
latex | +0.259 |
strap | +0.243 |
Neg logit contributions | |
GBT | -0.307 |
Catal | -0.270 |
Bah | -0.266 |
Franç | -0.262 |
ンジ | -0.254 |
Token |
|
Token ID | 198 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
GBT | +0.294 |
Catal | +0.258 |
Bah | +0.255 |
Franç | +0.252 |
ンジ | +0.243 |
Neg logit contributions | |
atti | -0.264 |
etti | -0.255 |
karma | -0.253 |
latex | -0.251 |
strap | -0.237 |
Token | as |
Token ID | 292 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
atti | +0.123 |
etti | +0.119 |
karma | +0.117 |
latex | +0.116 |
strap | +0.110 |
Neg logit contributions | |
GBT | -0.134 |
Catal | -0.118 |
Bah | -0.116 |
Franç | -0.115 |
ンジ | -0.111 |
Token | d |
Token ID | 67 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
atti | +3.476 |
karma | +3.391 |
etti | +3.378 |
latex | +3.320 |
strap | +3.107 |
Neg logit contributions | |
GBT | -4.346 |
Catal | -3.866 |
Bah | -3.773 |
Franç | -3.710 |
ンジ | -3.648 |
Token |
|
Token ID | 198 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
GBT | +0.204 |
Catal | +0.180 |
Bah | +0.178 |
Franç | +0.175 |
ンジ | +0.170 |
Neg logit contributions | |
atti | -0.179 |
etti | -0.173 |
karma | -0.171 |
latex | -0.170 |
strap | -0.160 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
atti | +0.948 |
etti | +0.915 |
karma | +0.908 |
latex | +0.891 |
strap | +0.852 |
Neg logit contributions | |
GBT | -1.152 |
Catal | -1.009 |
Bah | -0.993 |
Franç | -0.990 |
ンジ | -0.964 |
Token | Gu |
Token ID | 8205 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
atti | +1.065 |
etti | +1.033 |
karma | +1.018 |
latex | +1.008 |
strap | +0.962 |
Neg logit contributions | |
GBT | -1.167 |
Catal | -1.025 |
Franç | -1.007 |
Bah | -1.005 |
ンジ | -0.972 |
Token | M |
Token ID | 44 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
atti | +1.043 |
etti | +1.006 |
karma | +0.996 |
latex | +0.984 |
strap | +0.930 |
Neg logit contributions | |
GBT | -1.240 |
Catal | -1.097 |
Bah | -1.080 |
Franç | -1.069 |
ンジ | -1.035 |
Token | - |
Token ID | 12 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
atti | +0.602 |
etti | +0.585 |
karma | +0.581 |
latex | +0.576 |
strap | +0.550 |
Neg logit contributions | |
GBT | -0.479 |
Catal | -0.410 |
Bah | -0.403 |
Franç | -0.395 |
ンジ | -0.379 |
Token | front |
Token ID | 8534 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
GBT | +0.201 |
Catal | +0.178 |
Bah | +0.176 |
Franç | +0.174 |
ンジ | +0.168 |
Neg logit contributions | |
atti | -0.164 |
etti | -0.158 |
karma | -0.157 |
latex | -0.157 |
strap | -0.146 |
Token | s |
Token ID | 82 |
Feature activation | +8.98e-01 |
Pos logit contributions | |
GBT | +0.180 |
Catal | +0.158 |
Bah | +0.156 |
Franç | +0.154 |
ンジ | +0.149 |
Neg logit contributions | |
atti | -0.154 |
etti | -0.149 |
karma | -0.148 |
latex | -0.147 |
strap | -0.138 |
Token | " |
Token ID | 1 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
atti | +0.621 |
etti | +0.600 |
karma | +0.594 |
latex | +0.589 |
strap | +0.553 |
Neg logit contributions | |
GBT | -0.789 |
Catal | -0.697 |
Bah | -0.688 |
Franç | -0.675 |
ンジ | -0.658 |
Token |
|
Token ID | 198 |
Feature activation | +5.32e+00 |
Pos logit contributions | |
GBT | +0.332 |
Catal | +0.291 |
Bah | +0.288 |
Franç | +0.284 |
ンジ | +0.274 |
Neg logit contributions | |
atti | -0.298 |
etti | -0.287 |
karma | -0.285 |
latex | -0.284 |
strap | -0.267 |
Token |
|
Token ID | 198 |
Feature activation | +5.05e+00 |
Pos logit contributions | |
atti | +3.325 |
etti | +3.182 |
karma | +3.163 |
latex | +2.992 |
Dustin | +2.937 |
Neg logit contributions | |
GBT | -4.937 |
Franç | -4.248 |
ンジ | -4.230 |
Catal | -4.219 |
Bah | -4.182 |
Token | Ent |
Token ID | 14539 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
atti | +3.698 |
etti | +3.524 |
karma | +3.394 |
latex | +3.314 |
MIT | +3.283 |
Neg logit contributions | |
GBT | -4.034 |
Franç | -3.590 |
ンジ | -3.428 |
Goku | -3.368 |
Catal | -3.335 |
Token | ire |
Token ID | 557 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
atti | +0.731 |
etti | +0.703 |
karma | +0.696 |
latex | +0.691 |
strap | +0.646 |
Neg logit contributions | |
GBT | -1.010 |
Catal | -0.898 |
Bah | -0.889 |
Franç | -0.877 |
ンジ | -0.850 |
Token | classrooms |
Token ID | 32579 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
GBT | +0.574 |
Catal | +0.506 |
Bah | +0.500 |
Franç | +0.493 |
ンジ | +0.476 |
Neg logit contributions | |
atti | -0.488 |
etti | -0.470 |
karma | -0.466 |
latex | -0.463 |
strap | -0.436 |
Token | of |
Token ID | 286 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
atti | +0.228 |
etti | +0.220 |
karma | +0.218 |
latex | +0.217 |
strap | +0.204 |
Neg logit contributions | |
GBT | -0.266 |
Catal | -0.235 |
Bah | -0.232 |
Franç | -0.228 |
ンジ | -0.221 |
Token | Somali |
Token ID | 30886 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
atti | +0.454 |
etti | +0.438 |
karma | +0.436 |
latex | +0.433 |
strap | +0.407 |
Neg logit contributions | |
GBT | -0.525 |
Catal | -0.461 |
Bah | -0.456 |
Franç | -0.449 |
ンジ | -0.435 |
Token | , |
Token ID | 837 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
atti | +0.734 |
etti | +0.708 |
karma | +0.704 |
latex | +0.700 |
strap | +0.661 |
Neg logit contributions | |
GBT | -0.774 |
Catal | -0.675 |
Bah | -0.667 |
Franç | -0.658 |
ンジ | -0.637 |
Token | r |
Token ID | 374 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
atti | +1.123 |
karma | +1.084 |
etti | +1.083 |
latex | +1.078 |
strap | +1.010 |
Neg logit contributions | |
GBT | -1.293 |
Catal | -1.139 |
Bah | -1.127 |
Franç | -1.108 |
ンジ | -1.076 |
Token | ' |
Token ID | 705 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
atti | +0.509 |
etti | +0.490 |
karma | +0.487 |
latex | +0.485 |
strap | +0.457 |
Neg logit contributions | |
GBT | -0.580 |
Catal | -0.511 |
Bah | -0.504 |
Franç | -0.498 |
ンジ | -0.481 |
Token | \ |
Token ID | 59 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
GBT | +1.043 |
Catal | +0.911 |
Bah | +0.897 |
Franç | +0.884 |
ンジ | +0.847 |
Neg logit contributions | |
atti | -1.070 |
etti | -1.034 |
karma | -1.027 |
latex | -1.019 |
strap | -0.964 |
Token | [ |
Token ID | 58 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
GBT | +0.805 |
Catal | +0.710 |
Bah | +0.703 |
Franç | +0.689 |
Goku | +0.665 |
Neg logit contributions | |
atti | -0.688 |
etti | -0.664 |
karma | -0.663 |
latex | -0.657 |
strap | -0.615 |
Token | +( |
Token ID | 33747 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
GBT | +3.775 |
Catal | +3.274 |
Bah | +3.219 |
Franç | +3.167 |
ンジ | +3.051 |
Neg logit contributions | |
atti | -3.567 |
etti | -3.426 |
karma | -3.403 |
latex | -3.351 |
strap | -3.154 |
Token | \ |
Token ID | 59 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
GBT | +4.858 |
Catal | +4.264 |
Bah | +4.189 |
Franç | +4.160 |
ンジ | +3.979 |
Neg logit contributions | |
atti | -4.587 |
etti | -4.418 |
karma | -4.386 |
latex | -4.310 |
Dustin | -4.083 |
Token | S |
Token ID | 50 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
GBT | +1.763 |
Catal | +1.556 |
Bah | +1.541 |
Franç | +1.506 |
Goku | +1.455 |
Neg logit contributions | |
atti | -1.581 |
etti | -1.538 |
karma | -1.530 |
latex | -1.509 |
Neighborhood | -1.418 |
Token | + |
Token ID | 10 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
GBT | +1.711 |
Catal | +1.473 |
Bah | +1.457 |
Franç | +1.437 |
ンジ | +1.380 |
Neg logit contributions | |
atti | -1.772 |
etti | -1.710 |
karma | -1.710 |
latex | -1.696 |
strap | -1.607 |
Token | )\ |
Token ID | 19415 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
GBT | +2.323 |
Catal | +2.017 |
Bah | +2.002 |
Franç | +1.965 |
Goku | +1.888 |
Neg logit contributions | |
atti | -2.380 |
etti | -2.309 |
karma | -2.302 |
latex | -2.294 |
strap | -2.170 |
Token | s |
Token ID | 82 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
GBT | +0.655 |
Catal | +0.576 |
Bah | +0.568 |
Franç | +0.558 |
ンジ | +0.536 |
Neg logit contributions | |
atti | -0.604 |
etti | -0.585 |
karma | -0.582 |
latex | -0.577 |
strap | -0.542 |
Token | - |
Token ID | 12 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
GBT | +0.571 |
Catal | +0.496 |
Bah | +0.488 |
Franç | +0.481 |
ンジ | +0.463 |
Neg logit contributions | |
atti | -0.614 |
etti | -0.593 |
karma | -0.589 |
latex | -0.586 |
strap | -0.555 |
Token | a |
Token ID | 64 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
GBT | +0.282 |
Catal | +0.249 |
Bah | +0.247 |
Franç | +0.243 |
ンジ | +0.235 |
Neg logit contributions | |
atti | -0.230 |
etti | -0.222 |
karma | -0.220 |
latex | -0.218 |
strap | -0.205 |
Token | ... |
Token ID | 986 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
GBT | +0.894 |
Catal | +0.777 |
Bah | +0.767 |
Franç | +0.755 |
ンジ | +0.726 |
Neg logit contributions | |
atti | -0.936 |
etti | -0.906 |
karma | -0.900 |
latex | -0.895 |
strap | -0.848 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
atti | +1.169 |
etti | +1.129 |
karma | +1.120 |
latex | +1.115 |
strap | +1.045 |
Neg logit contributions | |
GBT | -1.357 |
Catal | -1.192 |
Bah | -1.177 |
Franç | -1.155 |
ンジ | -1.125 |
Token | Buy |
Token ID | 14518 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
GBT | +0.425 |
Catal | +0.372 |
Bah | +0.367 |
Franç | +0.363 |
ンジ | +0.350 |
Neg logit contributions | |
atti | -0.396 |
etti | -0.383 |
karma | -0.380 |
latex | -0.377 |
strap | -0.357 |
Token | Photo |
Token ID | 5555 |
Feature activation | -5.72e+00 |
Pos logit contributions | |
GBT | +0.424 |
Catal | +0.378 |
Bah | +0.374 |
Franç | +0.370 |
ンジ | +0.358 |
Neg logit contributions | |
atti | -0.288 |
etti | -0.277 |
karma | -0.274 |
latex | -0.272 |
strap | -0.254 |
Token | F |
Token ID | 376 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
GBT | +4.625 |
Catal | +4.147 |
Franç | +4.092 |
Bah | +4.066 |
Goku | +3.946 |
Neg logit contributions | |
atti | -4.147 |
etti | -4.007 |
karma | -3.965 |
latex | -3.944 |
strap | -3.738 |
Token | SU |
Token ID | 12564 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
GBT | +1.015 |
Catal | +0.900 |
Bah | +0.890 |
Franç | +0.878 |
ンジ | +0.850 |
Neg logit contributions | |
atti | -0.795 |
etti | -0.765 |
karma | -0.758 |
latex | -0.753 |
strap | -0.706 |
Token | target |
Token ID | 2496 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
GBT | +0.791 |
Catal | +0.704 |
Bah | +0.699 |
Franç | +0.690 |
ンジ | +0.667 |
Neg logit contributions | |
atti | -0.560 |
etti | -0.536 |
karma | -0.533 |
latex | -0.529 |
strap | -0.494 |
Token | Marvin |
Token ID | 35105 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
GBT | +2.325 |
Catal | +2.058 |
Bah | +2.044 |
Franç | +2.024 |
Goku | +1.954 |
Neg logit contributions | |
atti | -1.902 |
etti | -1.824 |
karma | -1.804 |
latex | -1.802 |
strap | -1.687 |
Token | Wilson |
Token ID | 8127 |
Feature activation | -8.80e-01 |
Pos logit contributions | |
GBT | +1.235 |
Catal | +1.090 |
Bah | +1.079 |
Franç | +1.070 |
Goku | +1.029 |
Neg logit contributions | |
atti | -1.092 |
etti | -1.051 |
karma | -1.048 |
latex | -1.043 |
strap | -0.975 |
Token | b |
Token ID | 65 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
atti | +0.010 |
etti | +0.010 |
karma | +0.010 |
latex | +0.010 |
strap | +0.009 |
Neg logit contributions | |
GBT | -0.016 |
Catal | -0.014 |
Bah | -0.014 |
Franç | -0.014 |
ンジ | -0.014 |
Token | ] |
Token ID | 60 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
atti | +1.613 |
etti | +1.554 |
karma | +1.507 |
latex | +1.477 |
strap | +1.409 |
Neg logit contributions | |
GBT | -2.654 |
Catal | -2.347 |
Bah | -2.330 |
Franç | -2.319 |
ンジ | -2.278 |
Token | Personal |
Token ID | 30228 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
GBT | +0.338 |
Catal | +0.298 |
Bah | +0.294 |
Franç | +0.287 |
ンジ | +0.278 |
Neg logit contributions | |
atti | -0.319 |
etti | -0.309 |
karma | -0.307 |
latex | -0.306 |
strap | -0.287 |
Token | best |
Token ID | 1266 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
GBT | +2.579 |
Catal | +2.346 |
Bah | +2.333 |
Franç | +2.298 |
Goku | +2.250 |
Neg logit contributions | |
atti | -1.073 |
etti | -1.004 |
karma | -0.978 |
latex | -0.974 |
strap | -0.891 |
Token | : |
Token ID | 25 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
GBT | +0.283 |
Catal | +0.249 |
Bah | +0.247 |
Franç | +0.243 |
Goku | +0.233 |
Neg logit contributions | |
atti | -0.261 |
etti | -0.252 |
karma | -0.249 |
latex | -0.248 |
strap | -0.233 |
Token | [/ |
Token ID | 13412 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
atti | +0.190 |
etti | +0.180 |
karma | +0.178 |
latex | +0.177 |
strap | +0.161 |
Neg logit contributions | |
GBT | -0.402 |
Catal | -0.364 |
Bah | -0.361 |
Franç | -0.357 |
ンジ | -0.347 |
Token | b |
Token ID | 65 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
GBT | +4.798 |
Bah | +4.323 |
Catal | +4.310 |
Franç | +4.085 |
Goku | +4.008 |
Neg logit contributions | |
atti | -3.822 |
latex | -3.750 |
etti | -3.722 |
karma | -3.682 |
Neighborhood | -3.474 |
Token | ] |
Token ID | 60 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
atti | +1.094 |
etti | +1.048 |
karma | +1.030 |
latex | +1.008 |
strap | +0.957 |
Neg logit contributions | |
GBT | -1.745 |
Catal | -1.552 |
Bah | -1.537 |
Franç | -1.522 |
ンジ | -1.493 |
Token | 00 |
Token ID | 3571 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
GBT | +0.164 |
Catal | +0.141 |
Bah | +0.138 |
Franç | +0.136 |
ンジ | +0.130 |
Neg logit contributions | |
atti | -0.207 |
etti | -0.201 |
karma | -0.199 |
latex | -0.198 |
strap | -0.189 |
Token | : |
Token ID | 25 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
GBT | +0.291 |
Catal | +0.256 |
Bah | +0.253 |
Franç | +0.250 |
ンジ | +0.241 |
Neg logit contributions | |
atti | -0.260 |
etti | -0.251 |
karma | -0.249 |
latex | -0.248 |
strap | -0.233 |
Token | 47 |
Token ID | 2857 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
atti | +0.443 |
etti | +0.428 |
karma | +0.424 |
latex | +0.418 |
strap | +0.401 |
Neg logit contributions | |
GBT | -0.418 |
Catal | -0.359 |
Bah | -0.354 |
Franç | -0.351 |
ンジ | -0.341 |
Token | eness |
Token ID | 9449 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
GBT | +0.194 |
Catal | +0.170 |
Bah | +0.168 |
Franç | +0.165 |
ンジ | +0.159 |
Neg logit contributions | |
atti | -0.188 |
etti | -0.182 |
karma | -0.181 |
latex | -0.180 |
strap | -0.169 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
GBT | +0.748 |
Catal | +0.646 |
Bah | +0.637 |
Franç | +0.627 |
ンジ | +0.603 |
Neg logit contributions | |
atti | -0.831 |
etti | -0.806 |
karma | -0.800 |
latex | -0.795 |
strap | -0.754 |
Token | � |
Token ID | 447 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
GBT | +0.018 |
Catal | +0.015 |
Bah | +0.015 |
Franç | +0.015 |
ンジ | +0.014 |
Neg logit contributions | |
atti | -0.019 |
etti | -0.018 |
karma | -0.018 |
latex | -0.018 |
strap | -0.017 |
Token | � |
Token ID | 251 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
GBT | +0.246 |
Catal | +0.218 |
Bah | +0.216 |
Franç | +0.213 |
ンジ | +0.207 |
Neg logit contributions | |
atti | -0.188 |
etti | -0.182 |
karma | -0.180 |
latex | -0.179 |
strap | -0.167 |
Token | No |
Token ID | 1400 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
GBT | +1.033 |
Catal | +0.906 |
Bah | +0.896 |
Franç | +0.883 |
ンジ | +0.848 |
Neg logit contributions | |
atti | -0.969 |
etti | -0.934 |
karma | -0.929 |
latex | -0.925 |
strap | -0.871 |
Token | sooner |
Token ID | 14556 |
Feature activation | -5.44e+00 |
Pos logit contributions | |
atti | +1.384 |
etti | +1.348 |
karma | +1.342 |
latex | +1.321 |
Neighborhood | +1.217 |
Neg logit contributions | |
GBT | -1.964 |
Catal | -1.746 |
Bah | -1.709 |
Franç | -1.690 |
ンジ | -1.657 |
Token | had |
Token ID | 550 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
GBT | +4.272 |
Franç | +3.712 |
Catal | +3.694 |
Bah | +3.675 |
Goku | +3.637 |
Neg logit contributions | |
atti | -4.061 |
etti | -3.866 |
karma | -3.860 |
latex | -3.853 |
™: | -3.682 |
Token | Donna |
Token ID | 29216 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
GBT | +0.429 |
Catal | +0.373 |
Bah | +0.369 |
Franç | +0.364 |
ンジ | +0.348 |
Neg logit contributions | |
atti | -0.447 |
etti | -0.431 |
karma | -0.428 |
latex | -0.426 |
strap | -0.404 |
Token | and |
Token ID | 290 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
GBT | +1.804 |
Catal | +1.605 |
Bah | +1.588 |
Franç | +1.574 |
Goku | +1.516 |
Neg logit contributions | |
atti | -1.312 |
etti | -1.255 |
karma | -1.245 |
latex | -1.237 |
strap | -1.157 |
Token | her |
Token ID | 607 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
GBT | +1.347 |
Catal | +1.181 |
Bah | +1.166 |
Franç | +1.159 |
Goku | +1.107 |
Neg logit contributions | |
atti | -1.324 |
etti | -1.273 |
karma | -1.260 |
latex | -1.256 |
strap | -1.190 |
Token | family |
Token ID | 1641 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
GBT | +0.245 |
Catal | +0.218 |
Bah | +0.216 |
Franç | +0.214 |
ンジ | +0.206 |
Neg logit contributions | |
atti | -0.175 |
etti | -0.168 |
karma | -0.166 |
latex | -0.164 |
strap | -0.154 |
Token | b |
Token ID | 65 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
atti | +0.059 |
etti | +0.057 |
karma | +0.057 |
latex | +0.056 |
strap | +0.054 |
Neg logit contributions | |
GBT | -0.057 |
Catal | -0.049 |
Bah | -0.049 |
Franç | -0.048 |
ンジ | -0.046 |
Token | ] |
Token ID | 60 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
atti | +1.196 |
etti | +1.155 |
karma | +1.134 |
latex | +1.127 |
strap | +1.049 |
Neg logit contributions | |
GBT | -1.749 |
Catal | -1.556 |
Bah | -1.545 |
Franç | -1.523 |
ンジ | -1.476 |
Token | Personal |
Token ID | 30228 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
GBT | +1.243 |
Catal | +1.094 |
Bah | +1.086 |
Franç | +1.059 |
ンジ | +1.024 |
Neg logit contributions | |
atti | -1.107 |
etti | -1.065 |
karma | -1.059 |
latex | -1.054 |
strap | -0.987 |
Token | best |
Token ID | 1266 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
GBT | +1.373 |
Catal | +1.212 |
Bah | +1.204 |
Franç | +1.185 |
Goku | +1.142 |
Neg logit contributions | |
atti | -1.121 |
etti | -1.077 |
karma | -1.060 |
latex | -1.060 |
strap | -0.998 |
Token | : |
Token ID | 25 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
GBT | +0.742 |
Catal | +0.657 |
Bah | +0.652 |
Franç | +0.642 |
Goku | +0.620 |
Neg logit contributions | |
atti | -0.596 |
etti | -0.573 |
karma | -0.564 |
latex | -0.562 |
strap | -0.527 |
Token | [/ |
Token ID | 13412 |
Feature activation | -5.43e+00 |
Pos logit contributions | |
GBT | +0.501 |
Catal | +0.453 |
Bah | +0.450 |
Franç | +0.445 |
ンジ | +0.432 |
Neg logit contributions | |
atti | -0.247 |
etti | -0.234 |
karma | -0.231 |
latex | -0.229 |
strap | -0.210 |
Token | b |
Token ID | 65 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
GBT | +4.644 |
Catal | +4.155 |
Bah | +4.154 |
Franç | +3.968 |
Goku | +3.876 |
Neg logit contributions | |
atti | -3.656 |
latex | -3.577 |
etti | -3.555 |
karma | -3.538 |
Neighborhood | -3.339 |
Token | ] |
Token ID | 60 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
atti | +0.833 |
etti | +0.799 |
karma | +0.790 |
latex | +0.782 |
strap | +0.727 |
Neg logit contributions | |
GBT | -1.289 |
Catal | -1.153 |
Bah | -1.143 |
Franç | -1.127 |
ンジ | -1.096 |
Token | 1 |
Token ID | 352 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
GBT | +0.484 |
Catal | +0.422 |
Bah | +0.418 |
Franç | +0.412 |
ンジ | +0.396 |
Neg logit contributions | |
atti | -0.472 |
etti | -0.456 |
karma | -0.452 |
latex | -0.450 |
strap | -0.426 |
Token | : |
Token ID | 25 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
GBT | +0.641 |
Catal | +0.564 |
Bah | +0.557 |
Franç | +0.549 |
ンジ | +0.530 |
Neg logit contributions | |
atti | -0.579 |
etti | -0.558 |
karma | -0.554 |
latex | -0.551 |
strap | -0.519 |
Token | 07 |
Token ID | 2998 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
GBT | +0.513 |
Catal | +0.444 |
Bah | +0.438 |
Franç | +0.431 |
ンジ | +0.414 |
Neg logit contributions | |
atti | -0.560 |
etti | -0.542 |
karma | -0.539 |
latex | -0.536 |
strap | -0.508 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
GBT | +0.090 |
Catal | +0.078 |
Bah | +0.077 |
Franç | +0.076 |
ンジ | +0.073 |
Neg logit contributions | |
atti | -0.098 |
etti | -0.095 |
karma | -0.094 |
latex | -0.094 |
strap | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
GBT | +1.271 |
Catal | +1.106 |
Bah | +1.093 |
Franç | +1.066 |
Goku | +1.021 |
Neg logit contributions | |
atti | -1.491 |
etti | -1.445 |
latex | -1.437 |
karma | -1.434 |
strap | -1.352 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
GBT | +0.644 |
Catal | +0.559 |
Bah | +0.546 |
Franç | +0.539 |
ンジ | +0.516 |
Neg logit contributions | |
atti | -0.737 |
etti | -0.715 |
karma | -0.712 |
latex | -0.709 |
strap | -0.671 |
Token |
|
Token ID | 198 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
GBT | +0.544 |
Catal | +0.477 |
Bah | +0.472 |
Franç | +0.465 |
ンジ | +0.448 |
Neg logit contributions | |
atti | -0.495 |
etti | -0.478 |
karma | -0.474 |
latex | -0.471 |
strap | -0.444 |
Token |
|
Token ID | 198 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
atti | +1.583 |
etti | +1.523 |
karma | +1.523 |
latex | +1.477 |
strap | +1.417 |
Neg logit contributions | |
GBT | -2.028 |
Catal | -1.776 |
Franç | -1.747 |
Bah | -1.745 |
ンジ | -1.708 |
Token | Subject |
Token ID | 19776 |
Feature activation | -5.38e+00 |
Pos logit contributions | |
atti | +1.578 |
etti | +1.493 |
karma | +1.487 |
latex | +1.409 |
strap | +1.386 |
Neg logit contributions | |
GBT | -2.443 |
Franç | -2.200 |
Catal | -2.165 |
Bah | -2.137 |
ンジ | -2.102 |
Token | : |
Token ID | 25 |
Feature activation | +9.33e-01 |
Pos logit contributions | |
GBT | +3.741 |
Catal | +3.397 |
Bah | +3.396 |
Franç | +3.200 |
Goku | +3.120 |
Neg logit contributions | |
atti | -4.402 |
latex | -4.348 |
etti | -4.279 |
karma | -4.242 |
strap | -3.962 |
Token | Que |
Token ID | 4670 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
atti | +0.681 |
etti | +0.658 |
karma | +0.655 |
latex | +0.651 |
strap | +0.609 |
Neg logit contributions | |
GBT | -0.782 |
Catal | -0.688 |
Bah | -0.678 |
Franç | -0.668 |
ンジ | -0.649 |
Token | er |
Token ID | 263 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
atti | +0.695 |
etti | +0.671 |
karma | +0.664 |
latex | +0.659 |
strap | +0.619 |
Neg logit contributions | |
GBT | -0.794 |
Catal | -0.702 |
Bah | -0.693 |
Franç | -0.682 |
ンジ | -0.663 |
Token | magnet |
Token ID | 19972 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
GBT | +0.589 |
Catal | +0.518 |
Bah | +0.513 |
Franç | +0.506 |
ンジ | +0.488 |
Neg logit contributions | |
atti | -0.515 |
etti | -0.497 |
karma | -0.492 |
latex | -0.489 |
strap | -0.462 |
Token | ? |
Token ID | 30 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
GBT | +1.169 |
Catal | +1.028 |
Bah | +1.016 |
Franç | +1.003 |
ンジ | +0.964 |
Neg logit contributions | |
atti | -1.074 |
etti | -1.035 |
karma | -1.027 |
latex | -1.021 |
strap | -0.966 |
Token | [ |
Token ID | 685 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
atti | +1.279 |
etti | +1.227 |
karma | +1.219 |
latex | +1.207 |
strap | +1.121 |
Neg logit contributions | |
GBT | -2.010 |
Catal | -1.799 |
Bah | -1.774 |
Franç | -1.759 |
ンジ | -1.710 |
Token | b |
Token ID | 65 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
GBT | +0.382 |
Catal | +0.342 |
Bah | +0.338 |
Franç | +0.335 |
ンジ | +0.325 |
Neg logit contributions | |
atti | -0.247 |
etti | -0.237 |
karma | -0.235 |
latex | -0.233 |
strap | -0.217 |
Token | ] |
Token ID | 60 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
atti | +1.554 |
etti | +1.495 |
karma | +1.453 |
latex | +1.418 |
strap | +1.359 |
Neg logit contributions | |
GBT | -2.554 |
Catal | -2.262 |
Bah | -2.243 |
Franç | -2.235 |
ンジ | -2.194 |
Token | Console |
Token ID | 47581 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
GBT | +0.231 |
Catal | +0.201 |
Bah | +0.198 |
Franç | +0.195 |
ンジ | +0.187 |
Neg logit contributions | |
atti | -0.240 |
etti | -0.233 |
karma | -0.231 |
latex | -0.230 |
strap | -0.218 |
Token | : |
Token ID | 25 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
GBT | +0.053 |
Catal | +0.047 |
Bah | +0.046 |
Franç | +0.046 |
ンジ | +0.044 |
Neg logit contributions | |
atti | -0.048 |
etti | -0.047 |
karma | -0.046 |
latex | -0.046 |
strap | -0.043 |
Token | [/ |
Token ID | 13412 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
GBT | +0.147 |
Catal | +0.134 |
Bah | +0.132 |
Franç | +0.131 |
ンジ | +0.127 |
Neg logit contributions | |
atti | -0.070 |
etti | -0.067 |
karma | -0.066 |
latex | -0.066 |
strap | -0.060 |
Token | b |
Token ID | 65 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
GBT | +4.571 |
Bah | +4.122 |
Catal | +4.119 |
Franç | +3.893 |
Goku | +3.817 |
Neg logit contributions | |
atti | -3.575 |
latex | -3.497 |
etti | -3.495 |
karma | -3.451 |
Neighborhood | -3.243 |
Token | ] |
Token ID | 60 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
atti | +1.182 |
etti | +1.133 |
karma | +1.110 |
latex | +1.087 |
strap | +1.035 |
Neg logit contributions | |
GBT | -1.899 |
Catal | -1.687 |
Bah | -1.671 |
Franç | -1.659 |
ンジ | -1.627 |
Token | NES |
Token ID | 31925 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
GBT | +0.233 |
Catal | +0.203 |
Bah | +0.200 |
Franç | +0.198 |
ンジ | +0.190 |
Neg logit contributions | |
atti | -0.236 |
etti | -0.228 |
karma | -0.226 |
latex | -0.225 |
strap | -0.213 |
Token | ( |
Token ID | 357 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
GBT | +0.113 |
Catal | +0.098 |
Bah | +0.097 |
Franç | +0.095 |
ンジ | +0.092 |
Neg logit contributions | |
atti | -0.115 |
etti | -0.111 |
karma | -0.110 |
latex | -0.109 |
strap | -0.104 |
Token | P |
Token ID | 47 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
atti | +0.977 |
etti | +0.935 |
karma | +0.922 |
latex | +0.917 |
strap | +0.864 |
Neg logit contributions | |
GBT | -1.406 |
Catal | -1.248 |
Bah | -1.231 |
Franç | -1.228 |
ンジ | -1.192 |
Token | 13 |
Token ID | 1511 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
GBT | +0.986 |
Catal | +0.859 |
Bah | +0.847 |
Franç | +0.836 |
Goku | +0.798 |
Neg logit contributions | |
atti | -1.085 |
etti | -1.050 |
karma | -1.044 |
latex | -1.039 |
strap | -0.984 |
Token | , |
Token ID | 11 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
GBT | +0.899 |
Catal | +0.783 |
Bah | +0.773 |
Franç | +0.754 |
ンジ | +0.721 |
Neg logit contributions | |
atti | -1.071 |
etti | -1.035 |
latex | -1.029 |
karma | -1.027 |
strap | -0.973 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
GBT | +1.316 |
Catal | +1.142 |
Bah | +1.125 |
Franç | +1.110 |
ンジ | +1.065 |
Neg logit contributions | |
atti | -1.403 |
etti | -1.358 |
karma | -1.350 |
latex | -1.341 |
strap | -1.270 |
Token |
|
Token ID | 198 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
GBT | +0.468 |
Catal | +0.411 |
Bah | +0.406 |
Franç | +0.401 |
ンジ | +0.387 |
Neg logit contributions | |
atti | -0.423 |
etti | -0.409 |
karma | -0.406 |
latex | -0.403 |
strap | -0.380 |
Token |
|
Token ID | 198 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
atti | +1.468 |
etti | +1.411 |
karma | +1.409 |
latex | +1.368 |
strap | +1.311 |
Neg logit contributions | |
GBT | -1.863 |
Catal | -1.628 |
Franç | -1.606 |
Bah | -1.600 |
ンジ | -1.568 |
Token | Subject |
Token ID | 19776 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
atti | +1.406 |
etti | +1.343 |
karma | +1.325 |
latex | +1.267 |
strap | +1.226 |
Neg logit contributions | |
GBT | -2.196 |
Franç | -1.975 |
Catal | -1.949 |
Bah | -1.916 |
ンジ | -1.896 |
Token | : |
Token ID | 25 |
Feature activation | +9.08e-01 |
Pos logit contributions | |
GBT | +3.686 |
Bah | +3.330 |
Catal | +3.325 |
Franç | +3.151 |
Goku | +3.074 |
Neg logit contributions | |
atti | -4.219 |
latex | -4.132 |
etti | -4.098 |
karma | -4.057 |
strap | -3.815 |
Token | boys |
Token ID | 6510 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
atti | +0.658 |
etti | +0.635 |
karma | +0.633 |
latex | +0.628 |
strap | +0.588 |
Neg logit contributions | |
GBT | -0.765 |
Catal | -0.674 |
Bah | -0.665 |
Franç | -0.656 |
ンジ | -0.636 |
Token | beware |
Token ID | 40600 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
atti | +1.517 |
etti | +1.460 |
karma | +1.453 |
latex | +1.441 |
strap | +1.351 |
Neg logit contributions | |
GBT | -1.926 |
Catal | -1.717 |
Bah | -1.687 |
Franç | -1.666 |
ンジ | -1.618 |
Token | This |
Token ID | 770 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
atti | +1.104 |
etti | +1.069 |
karma | +1.065 |
latex | +1.057 |
strap | +0.993 |
Neg logit contributions | |
GBT | -1.208 |
Catal | -1.061 |
Bah | -1.046 |
Franç | -1.032 |
ンジ | -0.998 |
Token | film |
Token ID | 2646 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
atti | +1.511 |
etti | +1.456 |
karma | +1.453 |
latex | +1.448 |
Neighborhood | +1.340 |
Neg logit contributions | |
GBT | -1.991 |
Catal | -1.769 |
Bah | -1.747 |
Franç | -1.720 |
ンジ | -1.677 |
Token | 9 |
Token ID | 860 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
GBT | +0.732 |
Catal | +0.635 |
Bah | +0.626 |
Franç | +0.617 |
ンジ | +0.591 |
Neg logit contributions | |
atti | -0.796 |
etti | -0.770 |
karma | -0.765 |
latex | -0.761 |
strap | -0.721 |
Token | , |
Token ID | 11 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
GBT | +0.854 |
Catal | +0.742 |
Bah | +0.733 |
Franç | +0.717 |
ンジ | +0.687 |
Neg logit contributions | |
atti | -0.976 |
etti | -0.946 |
karma | -0.939 |
latex | -0.938 |
strap | -0.885 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
GBT | +0.646 |
Catal | +0.562 |
Bah | +0.548 |
Franç | +0.542 |
ンジ | +0.518 |
Neg logit contributions | |
atti | -0.723 |
etti | -0.702 |
karma | -0.699 |
latex | -0.697 |
strap | -0.657 |
Token |
|
Token ID | 198 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
GBT | +0.820 |
Catal | +0.721 |
Bah | +0.713 |
Franç | +0.701 |
ンジ | +0.674 |
Neg logit contributions | |
atti | -0.762 |
etti | -0.735 |
karma | -0.730 |
latex | -0.728 |
strap | -0.684 |
Token |
|
Token ID | 198 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
atti | +1.234 |
etti | +1.188 |
karma | +1.186 |
latex | +1.155 |
strap | +1.105 |
Neg logit contributions | |
GBT | -1.540 |
Catal | -1.349 |
Bah | -1.326 |
Franç | -1.325 |
ンジ | -1.293 |
Token | Subject |
Token ID | 19776 |
Feature activation | -5.14e+00 |
Pos logit contributions | |
atti | +1.398 |
etti | +1.322 |
karma | +1.311 |
latex | +1.246 |
strap | +1.227 |
Neg logit contributions | |
GBT | -2.154 |
Franç | -1.936 |
Catal | -1.908 |
Bah | -1.884 |
ンジ | -1.854 |
Token | : |
Token ID | 25 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
GBT | +3.623 |
Catal | +3.283 |
Bah | +3.280 |
Franç | +3.098 |
Goku | +3.021 |
Neg logit contributions | |
atti | -4.184 |
latex | -4.127 |
etti | -4.069 |
karma | -4.034 |
strap | -3.769 |
Token | R |
Token ID | 371 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
atti | +0.664 |
etti | +0.641 |
karma | +0.639 |
latex | +0.635 |
strap | +0.594 |
Neg logit contributions | |
GBT | -0.772 |
Catal | -0.681 |
Bah | -0.670 |
Franç | -0.661 |
ンジ | -0.642 |
Token | ipe |
Token ID | 3757 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
atti | +0.190 |
etti | +0.184 |
karma | +0.183 |
latex | +0.181 |
strap | +0.172 |
Neg logit contributions | |
GBT | -0.175 |
Catal | -0.151 |
Bah | -0.149 |
Franç | -0.147 |
ンジ | -0.142 |
Token | for |
Token ID | 329 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
atti | +0.719 |
etti | +0.694 |
karma | +0.690 |
latex | +0.686 |
strap | +0.646 |
Neg logit contributions | |
GBT | -0.792 |
Catal | -0.696 |
Bah | -0.687 |
Franç | -0.678 |
ンジ | -0.652 |
Token | M |
Token ID | 337 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
GBT | +0.247 |
Catal | +0.218 |
Bah | +0.216 |
Franç | +0.214 |
Goku | +0.206 |
Neg logit contributions | |
atti | -0.199 |
etti | -0.191 |
karma | -0.188 |
latex | -0.187 |
strap | -0.177 |
Token | mentioned |
Token ID | 4750 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
GBT | +0.379 |
Catal | +0.336 |
Bah | +0.333 |
Franç | +0.329 |
Goku | +0.317 |
Neg logit contributions | |
atti | -0.277 |
etti | -0.267 |
karma | -0.264 |
latex | -0.262 |
strap | -0.246 |
Token | . |
Token ID | 13 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
GBT | +0.547 |
Catal | +0.475 |
Bah | +0.469 |
Franç | +0.462 |
ンジ | +0.446 |
Neg logit contributions | |
atti | -0.562 |
etti | -0.544 |
karma | -0.541 |
latex | -0.537 |
strap | -0.509 |
Token |
|
Token ID | 198 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
atti | +0.599 |
etti | +0.578 |
karma | +0.573 |
latex | +0.567 |
strap | +0.535 |
Neg logit contributions | |
GBT | -0.712 |
Catal | -0.624 |
Bah | -0.616 |
Franç | -0.611 |
ンジ | -0.592 |
Token |
|
Token ID | 198 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
GBT | +0.049 |
Catal | +0.043 |
Bah | +0.042 |
Franç | +0.042 |
ンジ | +0.040 |
Neg logit contributions | |
atti | -0.043 |
etti | -0.041 |
karma | -0.041 |
latex | -0.041 |
strap | -0.038 |
Token | No |
Token ID | 2949 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
atti | +0.056 |
etti | +0.054 |
karma | +0.053 |
latex | +0.053 |
strap | +0.050 |
Neg logit contributions | |
GBT | -0.060 |
Catal | -0.052 |
Bah | -0.051 |
Franç | -0.051 |
ンジ | -0.049 |
Token | word |
Token ID | 1573 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
atti | +1.173 |
etti | +1.129 |
karma | +1.121 |
latex | +1.115 |
strap | +1.037 |
Neg logit contributions | |
GBT | -1.522 |
Catal | -1.348 |
Bah | -1.327 |
Franç | -1.314 |
ンジ | -1.275 |
Token | if |
Token ID | 611 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
GBT | +5.131 |
Catal | +4.660 |
Bah | +4.660 |
Franç | +4.577 |
Goku | +4.529 |
Neg logit contributions | |
atti | -2.703 |
etti | -2.535 |
latex | -2.481 |
karma | -2.456 |
strap | -2.321 |
Token | the |
Token ID | 262 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
GBT | +1.137 |
Catal | +1.008 |
Bah | +0.999 |
Franç | +0.985 |
Goku | +0.952 |
Neg logit contributions | |
atti | -0.895 |
etti | -0.860 |
karma | -0.849 |
latex | -0.846 |
strap | -0.800 |
Token | Str |
Token ID | 4285 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
atti | +0.034 |
etti | +0.033 |
karma | +0.032 |
latex | +0.032 |
strap | +0.030 |
Neg logit contributions | |
GBT | -0.047 |
Catal | -0.042 |
Bah | -0.041 |
Franç | -0.041 |
Goku | -0.039 |
Token | and |
Token ID | 392 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
GBT | +1.020 |
Catal | +0.930 |
Bah | +0.923 |
Franç | +0.899 |
Goku | +0.871 |
Neg logit contributions | |
atti | -0.586 |
latex | -0.584 |
karma | -0.582 |
etti | -0.569 |
Neighborhood | -0.525 |
Token | Craft |
Token ID | 15745 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
GBT | +1.130 |
Catal | +0.992 |
Bah | +0.978 |
Franç | +0.967 |
Goku | +0.928 |
Neg logit contributions | |
atti | -1.071 |
etti | -1.039 |
latex | -1.030 |
karma | -1.030 |
strap | -0.972 |
Token | 4 |
Token ID | 604 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
atti | +0.418 |
etti | +0.404 |
karma | +0.402 |
latex | +0.398 |
strap | +0.378 |
Neg logit contributions | |
GBT | -0.422 |
Catal | -0.366 |
Bah | -0.361 |
Franç | -0.355 |
ンジ | -0.346 |
Token | , |
Token ID | 11 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
GBT | +0.706 |
Catal | +0.612 |
Bah | +0.604 |
Franç | +0.594 |
ンジ | +0.569 |
Neg logit contributions | |
atti | -0.788 |
etti | -0.763 |
karma | -0.758 |
latex | -0.755 |
strap | -0.714 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
GBT | +0.165 |
Catal | +0.143 |
Bah | +0.139 |
Franç | +0.138 |
ンジ | +0.133 |
Neg logit contributions | |
atti | -0.180 |
karma | -0.175 |
etti | -0.175 |
latex | -0.174 |
Neighborhood | -0.165 |
Token |
|
Token ID | 198 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
GBT | +0.366 |
Catal | +0.322 |
Bah | +0.317 |
Franç | +0.314 |
ンジ | +0.303 |
Neg logit contributions | |
atti | -0.329 |
etti | -0.318 |
karma | -0.316 |
latex | -0.313 |
strap | -0.296 |
Token |
|
Token ID | 198 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
atti | +1.472 |
etti | +1.417 |
karma | +1.416 |
latex | +1.375 |
strap | +1.319 |
Neg logit contributions | |
GBT | -1.873 |
Catal | -1.640 |
Bah | -1.612 |
Franç | -1.612 |
ンジ | -1.575 |
Token | Subject |
Token ID | 19776 |
Feature activation | -5.04e+00 |
Pos logit contributions | |
atti | +1.493 |
etti | +1.412 |
karma | +1.407 |
latex | +1.328 |
strap | +1.318 |
Neg logit contributions | |
GBT | -2.321 |
Franç | -2.086 |
Catal | -2.052 |
Bah | -2.022 |
ンジ | -1.989 |
Token | : |
Token ID | 25 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
GBT | +3.542 |
Catal | +3.206 |
Bah | +3.203 |
Franç | +3.033 |
Goku | +2.953 |
Neg logit contributions | |
atti | -4.105 |
latex | -4.043 |
etti | -3.989 |
karma | -3.949 |
Neighborhood | -3.692 |
Token | User |
Token ID | 11787 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
atti | +0.634 |
etti | +0.611 |
karma | +0.610 |
latex | +0.605 |
strap | +0.567 |
Neg logit contributions | |
GBT | -0.725 |
Catal | -0.638 |
Bah | -0.629 |
Franç | -0.620 |
ンジ | -0.602 |
Token | reviews |
Token ID | 8088 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
GBT | +0.118 |
Catal | +0.104 |
Bah | +0.103 |
Franç | +0.101 |
ンジ | +0.098 |
Neg logit contributions | |
atti | -0.102 |
etti | -0.099 |
karma | -0.098 |
latex | -0.098 |
strap | -0.092 |
Token | are |
Token ID | 389 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
atti | +1.081 |
etti | +1.041 |
karma | +1.039 |
latex | +1.030 |
strap | +0.967 |
Neg logit contributions | |
GBT | -1.216 |
Catal | -1.067 |
Bah | -1.052 |
Franç | -1.036 |
ンジ | -1.007 |
Token | a |
Token ID | 257 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
atti | +0.919 |
karma | +0.885 |
etti | +0.882 |
latex | +0.875 |
strap | +0.811 |
Neg logit contributions | |
GBT | -1.217 |
Catal | -1.080 |
Bah | -1.067 |
Franç | -1.053 |
ンジ | -1.024 |
Token | 6 |
Token ID | 718 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
GBT | +0.542 |
Catal | +0.469 |
Bah | +0.463 |
Franç | +0.456 |
ンジ | +0.438 |
Neg logit contributions | |
atti | -0.596 |
etti | -0.578 |
karma | -0.574 |
latex | -0.570 |
strap | -0.541 |
Token | , |
Token ID | 11 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
GBT | +0.410 |
Catal | +0.355 |
Bah | +0.350 |
Franç | +0.345 |
ンジ | +0.331 |
Neg logit contributions | |
atti | -0.452 |
etti | -0.438 |
karma | -0.435 |
latex | -0.433 |
strap | -0.410 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
GBT | +0.323 |
Catal | +0.282 |
Bah | +0.275 |
Franç | +0.271 |
Goku | +0.259 |
Neg logit contributions | |
atti | -0.355 |
etti | -0.345 |
karma | -0.343 |
latex | -0.342 |
strap | -0.323 |
Token |
|
Token ID | 198 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
GBT | +0.694 |
Catal | +0.610 |
Bah | +0.603 |
Franç | +0.594 |
ンジ | +0.572 |
Neg logit contributions | |
atti | -0.634 |
etti | -0.612 |
karma | -0.608 |
latex | -0.604 |
strap | -0.570 |
Token |
|
Token ID | 198 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
atti | +1.406 |
etti | +1.353 |
karma | +1.351 |
latex | +1.314 |
strap | +1.258 |
Neg logit contributions | |
GBT | -1.782 |
Catal | -1.558 |
Franç | -1.534 |
Bah | -1.532 |
ンジ | -1.498 |
Token | Subject |
Token ID | 19776 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
atti | +1.443 |
etti | +1.364 |
karma | +1.349 |
latex | +1.267 |
strap | +1.260 |
Neg logit contributions | |
GBT | -2.344 |
Franç | -2.123 |
Catal | -2.075 |
Bah | -2.045 |
ンジ | -2.028 |
Token | : |
Token ID | 25 |
Feature activation | +9.72e-01 |
Pos logit contributions | |
GBT | +3.566 |
Catal | +3.226 |
Bah | +3.225 |
Franç | +3.048 |
Goku | +2.970 |
Neg logit contributions | |
atti | -4.084 |
latex | -4.028 |
etti | -3.973 |
karma | -3.942 |
strap | -3.681 |
Token | R |
Token ID | 371 |
Feature activation | -6.99e-01 |
Pos logit contributions | |
atti | +0.701 |
etti | +0.676 |
karma | +0.673 |
latex | +0.669 |
strap | +0.625 |
Neg logit contributions | |
GBT | -0.824 |
Catal | -0.727 |
Bah | -0.716 |
Franç | -0.706 |
ンジ | -0.686 |
Token | iff |
Token ID | 733 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
GBT | +0.577 |
Catal | +0.507 |
Bah | +0.501 |
Franç | +0.495 |
ンジ | +0.477 |
Neg logit contributions | |
atti | -0.518 |
etti | -0.500 |
karma | -0.496 |
latex | -0.493 |
strap | -0.465 |
Token | Tra |
Token ID | 15721 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
atti | +1.617 |
etti | +1.555 |
karma | +1.544 |
latex | +1.535 |
strap | +1.450 |
Neg logit contributions | |
GBT | -1.722 |
Catal | -1.496 |
Bah | -1.475 |
Franç | -1.464 |
ンジ | -1.415 |
Token | x |
Token ID | 87 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
GBT | +1.264 |
Catal | +1.079 |
Bah | +1.072 |
Franç | +1.045 |
ンジ | +1.002 |
Neg logit contributions | |
atti | -1.694 |
etti | -1.642 |
karma | -1.638 |
latex | -1.630 |
strap | -1.555 |
Token | Amount |
Token ID | 31264 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
GBT | +1.086 |
Catal | +0.952 |
Bah | +0.940 |
Franç | +0.924 |
ンジ | +0.889 |
Neg logit contributions | |
atti | -1.037 |
etti | -1.001 |
karma | -0.994 |
latex | -0.988 |
strap | -0.932 |
Token | ; |
Token ID | 26 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
GBT | +0.015 |
Catal | +0.013 |
Bah | +0.013 |
Franç | +0.012 |
ンジ | +0.012 |
Neg logit contributions | |
atti | -0.015 |
etti | -0.014 |
karma | -0.014 |
latex | -0.014 |
strap | -0.013 |
Token | } |
Token ID | 1782 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
GBT | +1.521 |
Catal | +1.329 |
Bah | +1.315 |
Franç | +1.292 |
ンジ | +1.245 |
Neg logit contributions | |
atti | -1.454 |
etti | -1.394 |
karma | -1.380 |
latex | -1.370 |
strap | -1.304 |
Token | 1 |
Token ID | 352 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
atti | +0.399 |
etti | +0.388 |
karma | +0.386 |
latex | +0.384 |
strap | +0.362 |
Neg logit contributions | |
GBT | -0.397 |
Catal | -0.346 |
Bah | -0.341 |
Franç | -0.337 |
ンジ | -0.324 |
Token | 2 |
Token ID | 362 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
GBT | +0.645 |
Catal | +0.558 |
Bah | +0.550 |
Franç | +0.541 |
ンジ | +0.519 |
Neg logit contributions | |
atti | -0.711 |
etti | -0.688 |
karma | -0.683 |
latex | -0.678 |
strap | -0.645 |
Token | 3 |
Token ID | 513 |
Feature activation | -5.01e+00 |
Pos logit contributions | |
GBT | +3.213 |
Catal | +2.806 |
Bah | +2.774 |
Goku | +2.693 |
Franç | +2.679 |
Neg logit contributions | |
atti | -3.462 |
etti | -3.302 |
karma | -3.246 |
latex | -3.210 |
strap | -3.096 |
Token | 4 |
Token ID | 604 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
GBT | +3.456 |
Bah | +2.957 |
Catal | +2.952 |
ンジ | +2.802 |
Franç | +2.801 |
Neg logit contributions | |
atti | -4.305 |
etti | -4.179 |
karma | -4.055 |
latex | -3.976 |
strap | -3.875 |
Token | 5 |
Token ID | 642 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
GBT | +3.110 |
Bah | +2.638 |
Catal | +2.626 |
Franç | +2.527 |
Goku | +2.494 |
Neg logit contributions | |
atti | -4.198 |
etti | -4.051 |
karma | -3.925 |
latex | -3.896 |
strap | -3.775 |
Token | 6 |
Token ID | 718 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
GBT | +2.745 |
Catal | +2.327 |
Bah | +2.292 |
Franç | +2.203 |
ンジ | +2.192 |
Neg logit contributions | |
atti | -3.977 |
etti | -3.825 |
karma | -3.727 |
latex | -3.683 |
strap | -3.575 |
Token | 7 |
Token ID | 767 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
GBT | +2.622 |
Catal | +2.253 |
Bah | +2.243 |
Franç | +2.161 |
ンジ | +2.150 |
Neg logit contributions | |
atti | -3.119 |
etti | -2.982 |
karma | -2.895 |
latex | -2.862 |
strap | -2.743 |
Token | 8 |
Token ID | 807 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
GBT | +2.579 |
Catal | +2.210 |
Bah | +2.191 |
Franç | +2.105 |
ンジ | +2.077 |
Neg logit contributions | |
atti | -3.255 |
etti | -3.131 |
karma | -3.084 |
latex | -3.029 |
strap | -2.908 |
Token | 26 |
Token ID | 2608 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
GBT | +0.625 |
Catal | +0.542 |
Bah | +0.534 |
Franç | +0.526 |
ンジ | +0.503 |
Neg logit contributions | |
atti | -0.697 |
etti | -0.675 |
karma | -0.671 |
latex | -0.667 |
strap | -0.633 |
Token | , |
Token ID | 11 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
GBT | +0.810 |
Catal | +0.703 |
Bah | +0.694 |
Franç | +0.680 |
ンジ | +0.651 |
Neg logit contributions | |
atti | -0.924 |
etti | -0.895 |
karma | -0.889 |
latex | -0.888 |
strap | -0.838 |
Token | 2009 |
Token ID | 3717 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
GBT | +0.367 |
Catal | +0.318 |
Bah | +0.310 |
Franç | +0.305 |
ンジ | +0.293 |
Neg logit contributions | |
atti | -0.408 |
etti | -0.396 |
karma | -0.395 |
latex | -0.394 |
strap | -0.372 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
GBT | +0.055 |
Catal | +0.048 |
Bah | +0.048 |
Franç | +0.047 |
ンジ | +0.046 |
Neg logit contributions | |
atti | -0.049 |
etti | -0.048 |
karma | -0.047 |
latex | -0.047 |
strap | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
atti | +1.257 |
etti | +1.210 |
karma | +1.208 |
latex | +1.176 |
strap | +1.127 |
Neg logit contributions | |
GBT | -1.575 |
Catal | -1.379 |
Bah | -1.356 |
Franç | -1.355 |
ンジ | -1.323 |
Token | Subject |
Token ID | 19776 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
atti | +1.388 |
etti | +1.316 |
karma | +1.311 |
latex | +1.249 |
strap | +1.225 |
Neg logit contributions | |
GBT | -2.092 |
Franç | -1.866 |
Catal | -1.843 |
Bah | -1.821 |
ンジ | -1.785 |
Token | : |
Token ID | 25 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
GBT | +3.502 |
Bah | +3.165 |
Catal | +3.164 |
Franç | +2.998 |
Goku | +2.914 |
Neg logit contributions | |
atti | -4.036 |
latex | -3.973 |
etti | -3.924 |
karma | -3.889 |
Neighborhood | -3.630 |
Token | THE |
Token ID | 3336 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
atti | +0.841 |
etti | +0.811 |
karma | +0.810 |
latex | +0.805 |
strap | +0.752 |
Neg logit contributions | |
GBT | -0.953 |
Catal | -0.838 |
Bah | -0.826 |
Franç | -0.813 |
ンジ | -0.791 |
Token | IR |
Token ID | 14826 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
atti | +0.219 |
etti | +0.212 |
karma | +0.210 |
latex | +0.209 |
strap | +0.197 |
Neg logit contributions | |
GBT | -0.219 |
Catal | -0.191 |
Bah | -0.189 |
Franç | -0.186 |
ンジ | -0.179 |
Token | ONY |
Token ID | 40508 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
GBT | +0.277 |
Catal | +0.240 |
Bah | +0.236 |
Franç | +0.234 |
ンジ | +0.224 |
Neg logit contributions | |
atti | -0.317 |
etti | -0.306 |
karma | -0.303 |
latex | -0.303 |
strap | -0.287 |
Token | OF |
Token ID | 3963 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
atti | +1.255 |
etti | +1.215 |
karma | +1.208 |
latex | +1.200 |
Neighborhood | +1.135 |
Neg logit contributions | |
GBT | -1.267 |
Catal | -1.100 |
Bah | -1.087 |
Franç | -1.066 |
ンジ | -1.037 |
Token | 5 |
Token ID | 642 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
GBT | +0.812 |
Catal | +0.705 |
Bah | +0.696 |
Franç | +0.687 |
Goku | +0.655 |
Neg logit contributions | |
atti | -0.904 |
etti | -0.875 |
karma | -0.870 |
latex | -0.866 |
strap | -0.820 |
Token | , |
Token ID | 11 |
Feature activation | -9.61e-02 |
Pos logit contributions | |
GBT | +1.005 |
Catal | +0.872 |
Bah | +0.861 |
Franç | +0.845 |
ンジ | +0.809 |
Neg logit contributions | |
atti | -1.144 |
etti | -1.108 |
karma | -1.100 |
latex | -1.098 |
strap | -1.037 |
Token | 2009 |
Token ID | 3717 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
GBT | +0.072 |
Catal | +0.063 |
Bah | +0.061 |
Franç | +0.060 |
ンジ | +0.058 |
Neg logit contributions | |
atti | -0.079 |
etti | -0.076 |
karma | -0.076 |
latex | -0.076 |
strap | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
GBT | +0.146 |
Catal | +0.128 |
Bah | +0.126 |
Franç | +0.125 |
ンジ | +0.120 |
Neg logit contributions | |
atti | -0.131 |
etti | -0.126 |
karma | -0.125 |
latex | -0.124 |
strap | -0.117 |
Token |
|
Token ID | 198 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
atti | +1.581 |
etti | +1.521 |
karma | +1.520 |
latex | +1.474 |
strap | +1.416 |
Neg logit contributions | |
GBT | -2.026 |
Catal | -1.774 |
Franç | -1.744 |
Bah | -1.743 |
ンジ | -1.706 |
Token | Subject |
Token ID | 19776 |
Feature activation | -4.96e+00 |
Pos logit contributions | |
atti | +1.649 |
etti | +1.558 |
karma | +1.558 |
latex | +1.475 |
strap | +1.457 |
Neg logit contributions | |
GBT | -2.546 |
Franç | -2.276 |
Catal | -2.244 |
Bah | -2.216 |
ンジ | -2.175 |
Token | : |
Token ID | 25 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
GBT | +3.500 |
Catal | +3.159 |
Bah | +3.156 |
Franç | +2.992 |
Goku | +2.911 |
Neg logit contributions | |
atti | -4.028 |
latex | -3.964 |
etti | -3.914 |
karma | -3.875 |
Neighborhood | -3.628 |
Token | Que |
Token ID | 4670 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
atti | +0.709 |
etti | +0.683 |
karma | +0.683 |
latex | +0.677 |
strap | +0.634 |
Neg logit contributions | |
GBT | -0.810 |
Catal | -0.713 |
Bah | -0.702 |
Franç | -0.692 |
ンジ | -0.673 |
Token | ers |
Token ID | 364 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
atti | +0.707 |
etti | +0.683 |
karma | +0.678 |
latex | +0.671 |
strap | +0.632 |
Neg logit contributions | |
GBT | -0.793 |
Catal | -0.699 |
Bah | -0.691 |
Franç | -0.680 |
ンジ | -0.659 |
Token | are |
Token ID | 389 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
GBT | +0.702 |
Catal | +0.617 |
Bah | +0.612 |
Franç | +0.604 |
ンジ | +0.581 |
Neg logit contributions | |
atti | -0.596 |
etti | -0.574 |
karma | -0.568 |
latex | -0.565 |
strap | -0.533 |
Token | Here |
Token ID | 3423 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
atti | +1.006 |
karma | +0.968 |
etti | +0.968 |
latex | +0.962 |
strap | +0.891 |
Neg logit contributions | |
GBT | -1.312 |
Catal | -1.169 |
Bah | -1.153 |
Franç | -1.136 |
ンジ | -1.104 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
GBT | +0.509 |
Catal | +0.449 |
Bah | +0.444 |
Franç | +0.438 |
ンジ | +0.423 |
Neg logit contributions | |
atti | -0.427 |
etti | -0.411 |
karma | -0.407 |
latex | -0.404 |
strap | -0.381 |
Token | game |
Token ID | 983 |
Feature activation | -8.61e-02 |
Pos logit contributions | |
atti | +0.693 |
etti | +0.667 |
karma | +0.666 |
latex | +0.661 |
strap | +0.612 |
Neg logit contributions | |
GBT | -0.956 |
Catal | -0.851 |
Bah | -0.841 |
Franç | -0.827 |
ンジ | -0.807 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
GBT | +0.066 |
Catal | +0.058 |
Bah | +0.057 |
Franç | +0.056 |
ンジ | +0.054 |
Neg logit contributions | |
atti | -0.069 |
etti | -0.067 |
karma | -0.066 |
latex | -0.066 |
strap | -0.062 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
GBT | +0.140 |
Catal | +0.123 |
Bah | +0.122 |
Franç | +0.120 |
ンジ | +0.116 |
Neg logit contributions | |
atti | -0.121 |
etti | -0.117 |
karma | -0.116 |
latex | -0.115 |
strap | -0.108 |
Token | Share |
Token ID | 11649 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
GBT | +0.220 |
Catal | +0.193 |
Bah | +0.190 |
Franç | +0.188 |
ンジ | +0.182 |
Neg logit contributions | |
atti | -0.205 |
etti | -0.198 |
karma | -0.196 |
latex | -0.195 |
strap | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
GBT | +2.495 |
Catal | +2.185 |
Bah | +2.154 |
Franç | +2.140 |
Goku | +2.048 |
Neg logit contributions | |
atti | -2.602 |
etti | -2.528 |
karma | -2.497 |
latex | -2.494 |
strap | -2.357 |
Token | Two |
Token ID | 4930 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
GBT | +4.022 |
Catal | +3.582 |
Bah | +3.577 |
Franç | +3.547 |
Goku | +3.472 |
Neg logit contributions | |
atti | -3.555 |
latex | -3.413 |
etti | -3.403 |
karma | -3.375 |
strap | -3.205 |
Token | iconic |
Token ID | 14133 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
GBT | +0.249 |
Catal | +0.220 |
Bah | +0.218 |
Franç | +0.215 |
ンジ | +0.208 |
Neg logit contributions | |
atti | -0.195 |
etti | -0.188 |
karma | -0.186 |
latex | -0.185 |
strap | -0.174 |
Token | horror |
Token ID | 9961 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
atti | +0.333 |
etti | +0.322 |
karma | +0.321 |
latex | +0.320 |
strap | +0.298 |
Neg logit contributions | |
GBT | -0.395 |
Catal | -0.348 |
Bah | -0.343 |
Franç | -0.338 |
ンジ | -0.328 |
Token | titles |
Token ID | 8714 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
GBT | +1.123 |
Catal | +0.991 |
Bah | +0.981 |
Franç | +0.968 |
ンジ | +0.935 |
Neg logit contributions | |
atti | -0.913 |
etti | -0.881 |
karma | -0.872 |
latex | -0.866 |
strap | -0.815 |
Token | are |
Token ID | 389 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
atti | +0.734 |
etti | +0.709 |
karma | +0.704 |
latex | +0.697 |
strap | +0.656 |
Neg logit contributions | |
GBT | -0.923 |
Catal | -0.817 |
Bah | -0.807 |
Franç | -0.796 |
ンジ | -0.773 |
Token | Legends |
Token ID | 14270 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
atti | +0.335 |
etti | +0.324 |
karma | +0.322 |
latex | +0.319 |
strap | +0.300 |
Neg logit contributions | |
GBT | -0.395 |
Catal | -0.348 |
Bah | -0.344 |
Franç | -0.339 |
ンジ | -0.328 |
Token | of |
Token ID | 286 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
atti | +0.771 |
etti | +0.748 |
latex | +0.738 |
karma | +0.735 |
Neighborhood | +0.700 |
Neg logit contributions | |
GBT | -0.969 |
Catal | -0.849 |
Bah | -0.840 |
Franç | -0.815 |
ンジ | -0.803 |
Token | Tomorrow |
Token ID | 25939 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
GBT | +0.102 |
Catal | +0.086 |
Bah | +0.084 |
Franç | +0.083 |
ンジ | +0.078 |
Neg logit contributions | |
atti | -0.155 |
etti | -0.151 |
karma | -0.150 |
latex | -0.149 |
strap | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
atti | +1.158 |
etti | +1.128 |
karma | +1.115 |
latex | +1.109 |
™: | +1.049 |
Neg logit contributions | |
GBT | -1.142 |
Catal | -0.987 |
Bah | -0.980 |
Franç | -0.954 |
ンジ | -0.930 |
Token | No |
Token ID | 1400 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
atti | +0.493 |
etti | +0.477 |
karma | +0.472 |
latex | +0.466 |
Neighborhood | +0.442 |
Neg logit contributions | |
GBT | -0.594 |
Catal | -0.521 |
Bah | -0.513 |
Franç | -0.505 |
ンジ | -0.494 |
Token | word |
Token ID | 1573 |
Feature activation | -4.88e+00 |
Pos logit contributions | |
atti | +1.378 |
etti | +1.332 |
karma | +1.330 |
latex | +1.317 |
Neighborhood | +1.224 |
Neg logit contributions | |
GBT | -1.715 |
Catal | -1.517 |
Bah | -1.487 |
Franç | -1.461 |
ンジ | -1.430 |
Token | on |
Token ID | 319 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
GBT | +5.256 |
Catal | +4.821 |
Bah | +4.812 |
Franç | +4.741 |
Goku | +4.665 |
Neg logit contributions | |
atti | -2.246 |
etti | -2.073 |
latex | -2.043 |
karma | -2.025 |
strap | -1.867 |
Token | what |
Token ID | 644 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
atti | +0.061 |
etti | +0.058 |
karma | +0.058 |
latex | +0.058 |
strap | +0.054 |
Neg logit contributions | |
GBT | -0.075 |
Catal | -0.067 |
Bah | -0.066 |
Franç | -0.065 |
ンジ | -0.063 |
Token | this |
Token ID | 428 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
GBT | +0.668 |
Catal | +0.591 |
Bah | +0.586 |
Franç | +0.578 |
Goku | +0.557 |
Neg logit contributions | |
atti | -0.526 |
etti | -0.505 |
karma | -0.500 |
latex | -0.498 |
strap | -0.469 |
Token | collaboration |
Token ID | 12438 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
GBT | +0.220 |
Catal | +0.196 |
Bah | +0.194 |
Franç | +0.191 |
ンジ | +0.185 |
Neg logit contributions | |
atti | -0.165 |
etti | -0.159 |
karma | -0.157 |
latex | -0.156 |
strap | -0.146 |
Token | will |
Token ID | 481 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
GBT | +0.731 |
Catal | +0.647 |
Bah | +0.639 |
Franç | +0.631 |
ンジ | +0.611 |
Neg logit contributions | |
atti | -0.569 |
etti | -0.547 |
karma | -0.544 |
latex | -0.540 |
strap | -0.505 |
Token | be |
Token ID | 1350 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
GBT | +0.035 |
Catal | +0.031 |
Bah | +0.030 |
Franç | +0.030 |
ンジ | +0.029 |
Neg logit contributions | |
atti | -0.033 |
etti | -0.031 |
karma | -0.031 |
latex | -0.031 |
strap | -0.029 |
Token | ek |
Token ID | 988 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
atti | +0.377 |
etti | +0.365 |
karma | +0.361 |
latex | +0.358 |
strap | +0.340 |
Neg logit contributions | |
GBT | -0.373 |
Catal | -0.324 |
Bah | -0.320 |
Franç | -0.315 |
ンジ | -0.304 |
Token | . |
Token ID | 13 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
atti | +0.054 |
etti | +0.052 |
karma | +0.052 |
latex | +0.051 |
strap | +0.049 |
Neg logit contributions | |
GBT | -0.051 |
Catal | -0.045 |
Bah | -0.044 |
Franç | -0.043 |
ンジ | -0.042 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
GBT | +0.677 |
Catal | +0.598 |
Bah | +0.591 |
Franç | +0.583 |
ンジ | +0.563 |
Neg logit contributions | |
atti | -0.573 |
etti | -0.552 |
karma | -0.548 |
latex | -0.545 |
strap | -0.512 |
Token | Share |
Token ID | 11649 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
GBT | +0.499 |
Catal | +0.437 |
Bah | +0.431 |
Franç | +0.426 |
ンジ | +0.410 |
Neg logit contributions | |
atti | -0.465 |
etti | -0.449 |
karma | -0.446 |
latex | -0.443 |
strap | -0.419 |
Token | . |
Token ID | 13 |
Feature activation | -4.77e+00 |
Pos logit contributions | |
GBT | +2.638 |
Catal | +2.311 |
Bah | +2.278 |
Franç | +2.264 |
Goku | +2.168 |
Neg logit contributions | |
atti | -2.745 |
etti | -2.669 |
karma | -2.633 |
latex | -2.632 |
strap | -2.485 |
Token | We |
Token ID | 775 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
GBT | +3.904 |
Catal | +3.477 |
Bah | +3.472 |
Franç | +3.445 |
Goku | +3.366 |
Neg logit contributions | |
atti | -3.451 |
latex | -3.313 |
etti | -3.302 |
karma | -3.277 |
strap | -3.107 |
Token | found |
Token ID | 1043 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
GBT | +0.916 |
Catal | +0.815 |
Bah | +0.812 |
Franç | +0.799 |
Goku | +0.775 |
Neg logit contributions | |
atti | -0.621 |
etti | -0.596 |
karma | -0.593 |
latex | -0.589 |
strap | -0.549 |
Token | a |
Token ID | 257 |
Feature activation | -7.30e-01 |
Pos logit contributions | |
GBT | +0.524 |
Catal | +0.460 |
Bah | +0.454 |
Franç | +0.448 |
ンジ | +0.432 |
Neg logit contributions | |
atti | -0.480 |
etti | -0.463 |
karma | -0.459 |
latex | -0.456 |
strap | -0.431 |
Token | home |
Token ID | 1363 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
GBT | +0.649 |
Catal | +0.576 |
Bah | +0.570 |
Franç | +0.563 |
ンジ | +0.544 |
Neg logit contributions | |
atti | -0.495 |
etti | -0.476 |
karma | -0.471 |
latex | -0.468 |
strap | -0.439 |
Token | for |
Token ID | 329 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
GBT | +2.256 |
Catal | +1.971 |
Bah | +1.955 |
Franç | +1.924 |
Goku | +1.853 |
Neg logit contributions | |
atti | -2.131 |
etti | -2.059 |
karma | -2.034 |
latex | -2.026 |
strap | -1.912 |
Token | [ |
Token ID | 685 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
atti | +1.025 |
etti | +0.987 |
karma | +0.983 |
latex | +0.974 |
strap | +0.909 |
Neg logit contributions | |
GBT | -1.413 |
Catal | -1.257 |
Bah | -1.239 |
Franç | -1.228 |
ンジ | -1.188 |
Token | b |
Token ID | 65 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
atti | +0.287 |
etti | +0.275 |
karma | +0.272 |
latex | +0.270 |
strap | +0.253 |
Neg logit contributions | |
GBT | -0.433 |
Catal | -0.387 |
Bah | -0.382 |
Franç | -0.380 |
ンジ | -0.367 |
Token | ] |
Token ID | 60 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
atti | +1.448 |
etti | +1.393 |
karma | +1.359 |
latex | +1.335 |
strap | +1.263 |
Neg logit contributions | |
GBT | -2.292 |
Catal | -2.039 |
Bah | -2.020 |
Franç | -1.996 |
ンジ | -1.964 |
Token | Console |
Token ID | 47581 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
GBT | +1.214 |
Catal | +1.071 |
Bah | +1.062 |
Franç | +1.031 |
ンジ | +1.000 |
Neg logit contributions | |
atti | -1.063 |
etti | -1.026 |
karma | -1.020 |
latex | -1.018 |
strap | -0.952 |
Token | : |
Token ID | 25 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
GBT | +0.581 |
Catal | +0.525 |
Bah | +0.522 |
Franç | +0.515 |
ンジ | +0.500 |
Neg logit contributions | |
atti | -0.299 |
etti | -0.284 |
karma | -0.279 |
latex | -0.277 |
strap | -0.255 |
Token | [/ |
Token ID | 13412 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
GBT | +0.577 |
Catal | +0.523 |
Bah | +0.519 |
Franç | +0.514 |
ンジ | +0.500 |
Neg logit contributions | |
atti | -0.280 |
etti | -0.265 |
karma | -0.262 |
latex | -0.260 |
strap | -0.238 |
Token | b |
Token ID | 65 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
GBT | +4.142 |
Bah | +3.711 |
Catal | +3.709 |
Franç | +3.545 |
Goku | +3.456 |
Neg logit contributions | |
atti | -3.143 |
etti | -3.058 |
latex | -3.054 |
karma | -3.041 |
Neighborhood | -2.849 |
Token | ] |
Token ID | 60 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
atti | +0.955 |
etti | +0.915 |
karma | +0.904 |
latex | +0.895 |
strap | +0.833 |
Neg logit contributions | |
GBT | -1.479 |
Catal | -1.323 |
Bah | -1.311 |
Franç | -1.292 |
ンジ | -1.258 |
Token | Xbox |
Token ID | 9445 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
GBT | +0.943 |
Catal | +0.824 |
Bah | +0.816 |
Franç | +0.805 |
ンジ | +0.772 |
Neg logit contributions | |
atti | -0.921 |
etti | -0.889 |
karma | -0.883 |
latex | -0.877 |
strap | -0.830 |
Token | 360 |
Token ID | 11470 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
atti | +2.169 |
etti | +2.087 |
karma | +2.079 |
latex | +2.048 |
strap | +1.944 |
Neg logit contributions | |
GBT | -2.747 |
Catal | -2.408 |
Bah | -2.357 |
Franç | -2.339 |
ンジ | -2.286 |
Token | ( |
Token ID | 357 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
atti | +1.199 |
etti | +1.154 |
karma | +1.149 |
latex | +1.141 |
strap | +1.076 |
Neg logit contributions | |
GBT | -1.371 |
Catal | -1.204 |
Bah | -1.183 |
Franç | -1.173 |
ンジ | -1.135 |
Token | ; |
Token ID | 26 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
GBT | +0.015 |
Catal | +0.013 |
Bah | +0.013 |
Franç | +0.012 |
ンジ | +0.012 |
Neg logit contributions | |
atti | -0.015 |
etti | -0.014 |
karma | -0.014 |
latex | -0.014 |
strap | -0.013 |
Token | } |
Token ID | 1782 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
GBT | +1.521 |
Catal | +1.329 |
Bah | +1.315 |
Franç | +1.292 |
ンジ | +1.245 |
Neg logit contributions | |
atti | -1.454 |
etti | -1.394 |
karma | -1.380 |
latex | -1.370 |
strap | -1.304 |
Token | 1 |
Token ID | 352 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
atti | +0.399 |
etti | +0.388 |
karma | +0.386 |
latex | +0.384 |
strap | +0.362 |
Neg logit contributions | |
GBT | -0.397 |
Catal | -0.346 |
Bah | -0.341 |
Franç | -0.337 |
ンジ | -0.324 |
Token | 2 |
Token ID | 362 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
GBT | +0.645 |
Catal | +0.558 |
Bah | +0.550 |
Franç | +0.541 |
ンジ | +0.519 |
Neg logit contributions | |
atti | -0.711 |
etti | -0.688 |
karma | -0.683 |
latex | -0.678 |
strap | -0.645 |
Token | 3 |
Token ID | 513 |
Feature activation | -5.01e+00 |
Pos logit contributions | |
GBT | +3.213 |
Catal | +2.806 |
Bah | +2.774 |
Goku | +2.693 |
Franç | +2.679 |
Neg logit contributions | |
atti | -3.462 |
etti | -3.302 |
karma | -3.246 |
latex | -3.210 |
strap | -3.096 |
Token | 4 |
Token ID | 604 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
GBT | +3.456 |
Bah | +2.957 |
Catal | +2.952 |
ンジ | +2.802 |
Franç | +2.801 |
Neg logit contributions | |
atti | -4.305 |
etti | -4.179 |
karma | -4.055 |
latex | -3.976 |
strap | -3.875 |
Token | 5 |
Token ID | 642 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
GBT | +3.110 |
Bah | +2.638 |
Catal | +2.626 |
Franç | +2.527 |
Goku | +2.494 |
Neg logit contributions | |
atti | -4.198 |
etti | -4.051 |
karma | -3.925 |
latex | -3.896 |
strap | -3.775 |
Token | 6 |
Token ID | 718 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
GBT | +2.745 |
Catal | +2.327 |
Bah | +2.292 |
Franç | +2.203 |
ンジ | +2.192 |
Neg logit contributions | |
atti | -3.977 |
etti | -3.825 |
karma | -3.727 |
latex | -3.683 |
strap | -3.575 |
Token | 7 |
Token ID | 767 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
GBT | +2.622 |
Catal | +2.253 |
Bah | +2.243 |
Franç | +2.161 |
ンジ | +2.150 |
Neg logit contributions | |
atti | -3.119 |
etti | -2.982 |
karma | -2.895 |
latex | -2.862 |
strap | -2.743 |
Token | 8 |
Token ID | 807 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
GBT | +2.579 |
Catal | +2.210 |
Bah | +2.191 |
Franç | +2.105 |
ンジ | +2.077 |
Neg logit contributions | |
atti | -3.255 |
etti | -3.131 |
karma | -3.084 |
latex | -3.029 |
strap | -2.908 |
Token | void |
Token ID | 7951 |
Feature activation | -7.60e-02 |
Pos logit contributions | |
GBT | +2.534 |
Bah | +2.212 |
Catal | +2.211 |
Franç | +2.136 |
ンジ | +2.093 |
Neg logit contributions | |
atti | -2.551 |
etti | -2.453 |
karma | -2.411 |
latex | -2.381 |
strap | -2.257 |
Token | , |
Token ID | 11 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
GBT | +0.298 |
Catal | +0.259 |
Bah | +0.256 |
Franç | +0.252 |
ンジ | +0.242 |
Neg logit contributions | |
atti | -0.308 |
etti | -0.297 |
karma | -0.295 |
latex | -0.294 |
strap | -0.278 |
Token | and |
Token ID | 290 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
atti | +0.637 |
etti | +0.618 |
karma | +0.617 |
latex | +0.612 |
strap | +0.580 |
Neg logit contributions | |
GBT | -0.543 |
Catal | -0.466 |
Bah | -0.459 |
Franç | -0.451 |
ンジ | -0.434 |
Token | because |
Token ID | 780 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
atti | +0.409 |
etti | +0.395 |
karma | +0.392 |
latex | +0.389 |
strap | +0.366 |
Neg logit contributions | |
GBT | -0.490 |
Catal | -0.432 |
Bah | -0.427 |
Franç | -0.421 |
ンジ | -0.408 |
Token | of |
Token ID | 286 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
GBT | +1.361 |
Catal | +1.199 |
Bah | +1.191 |
Franç | +1.178 |
Goku | +1.136 |
Neg logit contributions | |
atti | -1.247 |
etti | -1.199 |
latex | -1.186 |
karma | -1.184 |
strap | -1.116 |
Token | that |
Token ID | 326 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
GBT | +0.689 |
Catal | +0.610 |
Bah | +0.603 |
Franç | +0.595 |
ンジ | +0.574 |
Neg logit contributions | |
atti | -0.563 |
etti | -0.542 |
karma | -0.536 |
latex | -0.533 |
strap | -0.501 |
Token | I |
Token ID | 314 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
atti | +0.317 |
etti | +0.307 |
karma | +0.305 |
latex | +0.303 |
strap | +0.286 |
Neg logit contributions | |
GBT | -0.322 |
Catal | -0.281 |
Bah | -0.277 |
Franç | -0.273 |
ンジ | -0.263 |
Token | don |
Token ID | 836 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
atti | +2.170 |
etti | +2.110 |
karma | +2.081 |
latex | +2.061 |
strap | +1.955 |
Neg logit contributions | |
GBT | -2.602 |
Catal | -2.303 |
Bah | -2.266 |
Franç | -2.246 |
ンジ | -2.182 |
Token | � |
Token ID | 447 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
atti | +0.422 |
etti | +0.413 |
karma | +0.408 |
latex | +0.405 |
strap | +0.391 |
Neg logit contributions | |
GBT | -0.216 |
Catal | -0.173 |
Bah | -0.169 |
Franç | -0.167 |
ンジ | -0.157 |
Token | � |
Token ID | 247 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
atti | +0.081 |
etti | +0.078 |
karma | +0.077 |
latex | +0.077 |
strap | +0.072 |
Neg logit contributions | |
GBT | -0.084 |
Catal | -0.074 |
Bah | -0.073 |
Franç | -0.072 |
ンジ | -0.069 |
Token | t |
Token ID | 83 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
GBT | +0.013 |
Catal | +0.012 |
Bah | +0.011 |
Franç | +0.011 |
ンジ | +0.011 |
Neg logit contributions | |
atti | -0.010 |
etti | -0.009 |
karma | -0.009 |
latex | -0.009 |
strap | -0.009 |
Token | throw |
Token ID | 3714 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
GBT | +0.485 |
Catal | +0.430 |
Bah | +0.426 |
Franç | +0.422 |
Goku | +0.409 |
Neg logit contributions | |
atti | -0.351 |
etti | -0.335 |
karma | -0.333 |
latex | -0.331 |
strap | -0.310 |
Token | regulations |
Token ID | 6647 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
GBT | +0.556 |
Catal | +0.484 |
Bah | +0.478 |
Franç | +0.471 |
ンジ | +0.452 |
Neg logit contributions | |
atti | -0.574 |
etti | -0.556 |
karma | -0.551 |
latex | -0.548 |
strap | -0.519 |
Token | as |
Token ID | 355 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
atti | +0.183 |
etti | +0.177 |
karma | +0.175 |
latex | +0.175 |
strap | +0.164 |
Neg logit contributions | |
GBT | -0.190 |
Catal | -0.166 |
Bah | -0.164 |
Franç | -0.161 |
ンジ | -0.156 |
Token | another |
Token ID | 1194 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
atti | +0.768 |
etti | +0.742 |
karma | +0.738 |
latex | +0.731 |
strap | +0.683 |
Neg logit contributions | |
GBT | -1.002 |
Catal | -0.889 |
Bah | -0.879 |
Franç | -0.864 |
ンジ | -0.843 |
Token | government |
Token ID | 1230 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
atti | +0.599 |
etti | +0.577 |
karma | +0.576 |
latex | +0.573 |
strap | +0.531 |
Neg logit contributions | |
GBT | -0.767 |
Catal | -0.679 |
Bah | -0.673 |
Franç | -0.658 |
ンジ | -0.643 |
Token | did |
Token ID | 750 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
GBT | +0.129 |
Catal | +0.114 |
Bah | +0.112 |
Franç | +0.111 |
ンジ | +0.107 |
Neg logit contributions | |
atti | -0.110 |
etti | -0.106 |
karma | -0.106 |
latex | -0.105 |
strap | -0.099 |
Token | almost |
Token ID | 2048 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
GBT | +1.305 |
Catal | +1.145 |
Bah | +1.129 |
Franç | +1.119 |
ンジ | +1.069 |
Neg logit contributions | |
atti | -1.266 |
etti | -1.220 |
karma | -1.211 |
latex | -1.204 |
strap | -1.138 |
Token | 25 |
Token ID | 1679 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
atti | +2.444 |
karma | +2.367 |
etti | +2.351 |
latex | +2.342 |
strap | +2.176 |
Neg logit contributions | |
GBT | -3.036 |
Catal | -2.673 |
Bah | -2.649 |
Franç | -2.582 |
ンジ | -2.552 |
Token | years |
Token ID | 812 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
atti | +0.169 |
etti | +0.161 |
karma | +0.160 |
latex | +0.159 |
strap | +0.146 |
Neg logit contributions | |
GBT | -0.322 |
Catal | -0.290 |
Bah | -0.287 |
Franç | -0.284 |
ンジ | -0.277 |
Token | ago |
Token ID | 2084 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
atti | +0.168 |
etti | +0.163 |
karma | +0.162 |
latex | +0.161 |
strap | +0.153 |
Neg logit contributions | |
GBT | -0.134 |
Catal | -0.115 |
Bah | -0.113 |
Franç | -0.111 |
ンジ | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
atti | +0.475 |
etti | +0.460 |
karma | +0.457 |
latex | +0.454 |
strap | +0.430 |
Neg logit contributions | |
GBT | -0.472 |
Catal | -0.411 |
Bah | -0.406 |
Franç | -0.398 |
ンジ | -0.385 |
Token | Some |
Token ID | 2773 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
atti | +0.344 |
etti | +0.334 |
karma | +0.330 |
latex | +0.326 |
Neighborhood | +0.310 |
Neg logit contributions | |
GBT | -0.423 |
Catal | -0.371 |
Bah | -0.367 |
Franç | -0.360 |
ンジ | -0.351 |
Token | Function |
Token ID | 15553 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
atti | +0.664 |
etti | +0.642 |
karma | +0.641 |
latex | +0.635 |
strap | +0.599 |
Neg logit contributions | |
GBT | -0.730 |
Catal | -0.638 |
Bah | -0.632 |
Franç | -0.621 |
ンジ | -0.603 |
Token |
|
Token ID | 198 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
GBT | +0.405 |
Catal | +0.357 |
Bah | +0.352 |
Franç | +0.348 |
ンジ | +0.334 |
Neg logit contributions | |
atti | -0.366 |
etti | -0.352 |
karma | -0.349 |
latex | -0.347 |
strap | -0.328 |
Token |
|
Token ID | 198 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
atti | +1.081 |
etti | +1.044 |
karma | +1.037 |
latex | +1.021 |
strap | +0.973 |
Neg logit contributions | |
GBT | -1.286 |
Catal | -1.125 |
Bah | -1.111 |
Franç | -1.104 |
ンジ | -1.074 |
Token | List |
Token ID | 8053 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
atti | +1.108 |
etti | +1.077 |
karma | +1.061 |
latex | +1.044 |
strap | +1.009 |
Neg logit contributions | |
GBT | -1.239 |
Catal | -1.079 |
Franç | -1.074 |
Bah | -1.068 |
ンジ | -1.034 |
Token | function |
Token ID | 2163 |
Feature activation | +4.41e-01 |
Pos logit contributions | |
atti | +0.867 |
karma | +0.843 |
etti | +0.838 |
latex | +0.832 |
strap | +0.782 |
Neg logit contributions | |
GBT | -1.019 |
Catal | -0.892 |
Bah | -0.884 |
Franç | -0.869 |
ンジ | -0.849 |
Token | should |
Token ID | 815 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
atti | +0.316 |
etti | +0.305 |
karma | +0.304 |
latex | +0.301 |
strap | +0.283 |
Neg logit contributions | |
GBT | -0.375 |
Catal | -0.331 |
Bah | -0.327 |
Franç | -0.322 |
ンジ | -0.312 |
Token | read |
Token ID | 1100 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
atti | +1.990 |
etti | +1.939 |
karma | +1.938 |
latex | +1.911 |
strap | +1.789 |
Neg logit contributions | |
GBT | -2.944 |
Catal | -2.622 |
Bah | -2.609 |
Franç | -2.568 |
ンジ | -2.499 |
Token | from |
Token ID | 422 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
atti | +2.211 |
karma | +2.188 |
etti | +2.161 |
latex | +2.143 |
strap | +2.005 |
Neg logit contributions | |
GBT | -2.993 |
Catal | -2.643 |
Bah | -2.635 |
Franç | -2.569 |
ンジ | -2.497 |
Token | the |
Token ID | 262 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
atti | +1.743 |
karma | +1.694 |
etti | +1.690 |
latex | +1.677 |
strap | +1.563 |
Neg logit contributions | |
GBT | -2.365 |
Catal | -2.098 |
Bah | -2.085 |
Franç | -2.039 |
ンジ | -1.996 |
Token | JSON |
Token ID | 19449 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
atti | +0.902 |
etti | +0.869 |
karma | +0.869 |
latex | +0.859 |
strap | +0.791 |
Neg logit contributions | |
GBT | -1.473 |
Catal | -1.321 |
Bah | -1.311 |
Franç | -1.289 |
ンジ | -1.259 |
Token | data |
Token ID | 1366 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
atti | +1.345 |
etti | +1.302 |
karma | +1.299 |
latex | +1.294 |
strap | +1.198 |
Neg logit contributions | |
GBT | -1.904 |
Catal | -1.692 |
Bah | -1.678 |
Franç | -1.649 |
ンジ | -1.617 |
Token |
|
Token ID | 198 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
atti | +0.671 |
etti | +0.646 |
karma | +0.642 |
latex | +0.632 |
strap | +0.602 |
Neg logit contributions | |
GBT | -0.798 |
Catal | -0.696 |
Bah | -0.688 |
Franç | -0.684 |
ンジ | -0.664 |
Token | Get |
Token ID | 3855 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
atti | +0.788 |
etti | +0.758 |
karma | +0.747 |
latex | +0.740 |
strap | +0.709 |
Neg logit contributions | |
GBT | -0.862 |
Catal | -0.749 |
Franç | -0.746 |
Bah | -0.738 |
ンジ | -0.718 |
Token | our |
Token ID | 674 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
GBT | +1.200 |
Catal | +1.078 |
Bah | +1.067 |
Franç | +1.055 |
ンジ | +1.025 |
Neg logit contributions | |
atti | -0.710 |
etti | -0.678 |
karma | -0.671 |
latex | -0.665 |
strap | -0.616 |
Token | daily |
Token ID | 4445 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
GBT | +0.378 |
Catal | +0.332 |
Bah | +0.327 |
Franç | +0.323 |
ンジ | +0.311 |
Neg logit contributions | |
atti | -0.359 |
etti | -0.346 |
karma | -0.343 |
latex | -0.341 |
strap | -0.323 |
Token | newsletter |
Token ID | 13129 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
atti | +0.173 |
etti | +0.169 |
karma | +0.168 |
latex | +0.167 |
strap | +0.157 |
Neg logit contributions | |
GBT | -0.161 |
Catal | -0.139 |
Bah | -0.138 |
Franç | -0.135 |
ンジ | -0.130 |
Token | Upgrade |
Token ID | 24236 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
atti | +1.181 |
etti | +1.149 |
karma | +1.143 |
latex | +1.131 |
Neighborhood | +1.049 |
Neg logit contributions | |
GBT | -1.808 |
Bah | -1.596 |
Catal | -1.568 |
Franç | -1.540 |
ンジ | -1.512 |
Token | your |
Token ID | 534 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
atti | +1.893 |
karma | +1.847 |
etti | +1.806 |
latex | +1.780 |
strap | +1.705 |
Neg logit contributions | |
GBT | -3.371 |
Catal | -2.989 |
Bah | -2.967 |
Franç | -2.937 |
ンジ | -2.872 |
Token | inbox |
Token ID | 13734 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
atti | +0.460 |
etti | +0.440 |
karma | +0.438 |
latex | +0.433 |
strap | +0.399 |
Neg logit contributions | |
GBT | -0.811 |
Catal | -0.728 |
Bah | -0.720 |
Franç | -0.714 |
ンジ | -0.693 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
atti | +0.130 |
etti | +0.126 |
karma | +0.126 |
latex | +0.124 |
strap | +0.118 |
Neg logit contributions | |
GBT | -0.141 |
Catal | -0.123 |
Bah | -0.122 |
Franç | -0.120 |
ンジ | -0.116 |
Token | get |
Token ID | 651 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
atti | +1.579 |
etti | +1.517 |
karma | +1.506 |
latex | +1.495 |
strap | +1.402 |
Neg logit contributions | |
GBT | -2.070 |
Catal | -1.837 |
Bah | -1.816 |
Franç | -1.794 |
ンジ | -1.736 |
Token | our |
Token ID | 674 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
atti | +0.606 |
etti | +0.578 |
karma | +0.575 |
latex | +0.569 |
strap | +0.526 |
Neg logit contributions | |
GBT | -1.071 |
Catal | -0.963 |
Bah | -0.954 |
Franç | -0.945 |
ンジ | -0.915 |
Token | In |
Token ID | 818 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
atti | +0.257 |
etti | +0.248 |
karma | +0.245 |
latex | +0.244 |
strap | +0.232 |
Neg logit contributions | |
GBT | -0.284 |
Catal | -0.249 |
Bah | -0.246 |
Franç | -0.245 |
ンジ | -0.235 |
Token | the |
Token ID | 262 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
atti | +0.046 |
etti | +0.044 |
karma | +0.044 |
latex | +0.044 |
strap | +0.041 |
Neg logit contributions | |
GBT | -0.059 |
Catal | -0.052 |
Bah | -0.052 |
Franç | -0.051 |
ンジ | -0.049 |
Token | code |
Token ID | 2438 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
GBT | +0.777 |
Catal | +0.684 |
Bah | +0.676 |
Franç | +0.671 |
Goku | +0.644 |
Neg logit contributions | |
atti | -0.627 |
etti | -0.599 |
karma | -0.593 |
latex | -0.587 |
strap | -0.556 |
Token | , |
Token ID | 11 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
GBT | +1.751 |
Catal | +1.515 |
Bah | +1.499 |
Franç | +1.492 |
Goku | +1.421 |
Neg logit contributions | |
atti | -1.811 |
etti | -1.750 |
karma | -1.732 |
latex | -1.710 |
strap | -1.629 |
Token | V |
Token ID | 569 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
GBT | +1.458 |
Catal | +1.293 |
Bah | +1.274 |
Franç | +1.267 |
Goku | +1.226 |
Neg logit contributions | |
atti | -1.068 |
etti | -1.011 |
karma | -1.006 |
latex | -0.998 |
strap | -0.944 |
Token | ort |
Token ID | 419 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
atti | +0.327 |
etti | +0.312 |
karma | +0.305 |
latex | +0.302 |
strap | +0.277 |
Neg logit contributions | |
GBT | -0.709 |
Catal | -0.644 |
Bah | -0.638 |
Franç | -0.632 |
ンジ | -0.616 |
Token | on |
Token ID | 261 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
atti | +2.299 |
etti | +2.233 |
karma | +2.203 |
latex | +2.174 |
strap | +2.078 |
Neg logit contributions | |
GBT | -2.387 |
Catal | -2.083 |
Bah | -2.060 |
Franç | -2.047 |
ンジ | -1.957 |
Token | Sim |
Token ID | 8890 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
atti | +2.027 |
etti | +1.979 |
karma | +1.967 |
latex | +1.948 |
strap | +1.840 |
Neg logit contributions | |
GBT | -2.046 |
Catal | -1.781 |
Bah | -1.761 |
Franç | -1.753 |
ンジ | -1.677 |
Token | :: |
Token ID | 3712 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
GBT | +0.320 |
Catal | +0.280 |
Bah | +0.276 |
Franç | +0.272 |
ンジ | +0.262 |
Neg logit contributions | |
atti | -0.315 |
etti | -0.305 |
karma | -0.303 |
latex | -0.301 |
strap | -0.285 |
Token | Pop |
Token ID | 16979 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
atti | +1.488 |
etti | +1.444 |
karma | +1.423 |
latex | +1.415 |
strap | +1.350 |
Neg logit contributions | |
GBT | -1.700 |
Catal | -1.489 |
Franç | -1.479 |
Bah | -1.473 |
ンジ | -1.411 |
Token | ulate |
Token ID | 5039 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
atti | +0.180 |
etti | +0.174 |
karma | +0.172 |
latex | +0.171 |
strap | +0.161 |
Neg logit contributions | |
GBT | -0.225 |
Catal | -0.199 |
Bah | -0.196 |
Franç | -0.194 |
ンジ | -0.187 |
Token | ( |
Token ID | 357 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
atti | +1.580 |
etti | +1.534 |
karma | +1.532 |
latex | +1.523 |
strap | +1.440 |
Neg logit contributions | |
GBT | -1.500 |
Catal | -1.293 |
Bah | -1.277 |
Franç | -1.256 |
ンジ | -1.212 |
Token | ch |
Token ID | 354 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
GBT | +0.425 |
Catal | +0.374 |
Bah | +0.369 |
Franç | +0.364 |
ンジ | +0.351 |
Neg logit contributions | |
atti | -0.374 |
etti | -0.361 |
karma | -0.358 |
latex | -0.356 |
strap | -0.335 |
Token | ag |
Token ID | 363 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
GBT | +0.266 |
Catal | +0.231 |
Bah | +0.227 |
Franç | +0.223 |
ンジ | +0.216 |
Neg logit contributions | |
atti | -0.282 |
etti | -0.272 |
karma | -0.269 |
latex | -0.267 |
strap | -0.254 |
Token | ha |
Token ID | 3099 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
GBT | +0.295 |
Catal | +0.259 |
Bah | +0.255 |
Franç | +0.251 |
ンジ | +0.243 |
Neg logit contributions | |
atti | -0.262 |
etti | -0.253 |
karma | -0.250 |
latex | -0.248 |
strap | -0.234 |
Token | � |
Token ID | 447 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
GBT | +1.360 |
Catal | +1.186 |
Bah | +1.177 |
Franç | +1.161 |
ンジ | +1.114 |
Neg logit contributions | |
atti | -1.326 |
etti | -1.285 |
karma | -1.273 |
latex | -1.270 |
strap | -1.203 |
Token | � |
Token ID | 247 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
GBT | +0.966 |
Catal | +0.840 |
Bah | +0.829 |
Franç | +0.817 |
ンジ | +0.787 |
Neg logit contributions | |
atti | -1.002 |
etti | -0.970 |
karma | -0.963 |
latex | -0.957 |
strap | -0.907 |
Token | an |
Token ID | 272 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
GBT | +0.027 |
Catal | +0.024 |
Bah | +0.023 |
Franç | +0.023 |
ンジ | +0.022 |
Neg logit contributions | |
atti | -0.025 |
etti | -0.024 |
karma | -0.024 |
latex | -0.024 |
strap | -0.023 |
Token | ), |
Token ID | 828 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
GBT | +0.960 |
Catal | +0.840 |
Bah | +0.829 |
Franç | +0.818 |
ンジ | +0.787 |
Neg logit contributions | |
atti | -0.928 |
etti | -0.897 |
karma | -0.891 |
latex | -0.886 |
strap | -0.837 |
Token | good |
Token ID | 922 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
atti | +0.181 |
etti | +0.175 |
karma | +0.174 |
latex | +0.173 |
strap | +0.163 |
Neg logit contributions | |
GBT | -0.201 |
Catal | -0.177 |
Bah | -0.175 |
Franç | -0.172 |
ンジ | -0.166 |
Token | health |
Token ID | 1535 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
atti | +0.448 |
karma | +0.432 |
etti | +0.428 |
latex | +0.422 |
strap | +0.391 |
Neg logit contributions | |
GBT | -0.704 |
Catal | -0.629 |
Bah | -0.620 |
Franç | -0.614 |
ンジ | -0.596 |
Token | ( |
Token ID | 357 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
GBT | +0.558 |
Catal | +0.482 |
Bah | +0.475 |
Franç | +0.468 |
ンジ | +0.449 |
Neg logit contributions | |
atti | -0.633 |
etti | -0.613 |
karma | -0.609 |
latex | -0.606 |
strap | -0.575 |
Token | at |
Token ID | 379 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
GBT | +0.271 |
Catal | +0.236 |
Bah | +0.233 |
Franç | +0.231 |
ンジ | +0.223 |
Neg logit contributions | |
atti | -0.253 |
etti | -0.243 |
karma | -0.242 |
latex | -0.240 |
strap | -0.227 |
Token | Emmanuel |
Token ID | 32390 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
atti | +0.279 |
etti | +0.270 |
karma | +0.268 |
latex | +0.267 |
strap | +0.252 |
Neg logit contributions | |
GBT | -0.284 |
Catal | -0.248 |
Bah | -0.245 |
Franç | -0.241 |
ンジ | -0.232 |
Token | College |
Token ID | 5535 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
atti | +0.125 |
etti | +0.120 |
karma | +0.119 |
latex | +0.118 |
strap | +0.110 |
Neg logit contributions | |
GBT | -0.179 |
Catal | -0.160 |
Bah | -0.158 |
Franç | -0.156 |
ンジ | -0.151 |
Token | in |
Token ID | 287 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
atti | +0.840 |
karma | +0.811 |
etti | +0.811 |
latex | +0.802 |
strap | +0.760 |
Neg logit contributions | |
GBT | -0.893 |
Catal | -0.775 |
Bah | -0.764 |
Franç | -0.753 |
ンジ | -0.733 |
Token | Boston |
Token ID | 6182 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
atti | +1.190 |
etti | +1.150 |
karma | +1.137 |
latex | +1.125 |
strap | +1.066 |
Neg logit contributions | |
GBT | -1.292 |
Catal | -1.134 |
Bah | -1.113 |
Franç | -1.102 |
ンジ | -1.065 |
Token | ( |
Token ID | 357 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
atti | +1.944 |
etti | +1.882 |
karma | +1.873 |
latex | +1.854 |
Neighborhood | +1.772 |
Neg logit contributions | |
GBT | -1.877 |
Catal | -1.605 |
Bah | -1.588 |
Franç | -1.557 |
ンジ | -1.522 |
Token | pictured |
Token ID | 28852 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
atti | +1.227 |
etti | +1.173 |
karma | +1.163 |
latex | +1.151 |
strap | +1.086 |
Neg logit contributions | |
GBT | -1.543 |
Catal | -1.355 |
Bah | -1.342 |
Franç | -1.339 |
ンジ | -1.291 |
Token | ), |
Token ID | 828 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
GBT | +0.682 |
Catal | +0.595 |
Bah | +0.589 |
Franç | +0.581 |
ンジ | +0.563 |
Neg logit contributions | |
atti | -0.609 |
etti | -0.585 |
karma | -0.582 |
latex | -0.578 |
strap | -0.545 |
Token | the |
Token ID | 262 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
atti | +0.709 |
etti | +0.681 |
karma | +0.675 |
latex | +0.670 |
strap | +0.626 |
Neg logit contributions | |
GBT | -1.005 |
Catal | -0.895 |
Bah | -0.885 |
Franç | -0.875 |
ンジ | -0.847 |
Token | Latin |
Token ID | 9133 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
atti | +0.140 |
etti | +0.134 |
karma | +0.133 |
latex | +0.132 |
strap | +0.124 |
Neg logit contributions | |
GBT | -0.195 |
Catal | -0.173 |
Bah | -0.171 |
Franç | -0.169 |
ンジ | -0.164 |
Token | Academy |
Token ID | 8581 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
atti | +2.033 |
etti | +1.931 |
karma | +1.927 |
latex | +1.893 |
Neighborhood | +1.839 |
Neg logit contributions | |
GBT | -1.853 |
Catal | -1.563 |
Bah | -1.534 |
Franç | -1.514 |
ンジ | -1.486 |
Token | : |
Token ID | 25 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
atti | +2.540 |
karma | +2.476 |
etti | +2.474 |
latex | +2.422 |
™: | +2.340 |
Neg logit contributions | |
GBT | -2.826 |
Catal | -2.424 |
Bah | -2.407 |
Franç | -2.378 |
ンジ | -2.325 |
Token | He |
Token ID | 679 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
atti | +2.249 |
karma | +2.199 |
etti | +2.185 |
latex | +2.140 |
Neighborhood | +2.035 |
Neg logit contributions | |
GBT | -2.382 |
Catal | -2.044 |
Bah | -2.034 |
Franç | -1.983 |
ンジ | -1.927 |
Token | had |
Token ID | 550 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
atti | +1.411 |
etti | +1.371 |
karma | +1.359 |
latex | +1.335 |
strap | +1.254 |
Neg logit contributions | |
GBT | -2.025 |
Catal | -1.801 |
Bah | -1.782 |
Franç | -1.737 |
ンジ | -1.699 |
Token | built |
Token ID | 3170 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
atti | +0.708 |
karma | +0.683 |
etti | +0.682 |
latex | +0.673 |
strap | +0.628 |
Neg logit contributions | |
GBT | -0.975 |
Catal | -0.866 |
Bah | -0.856 |
Franç | -0.841 |
ンジ | -0.819 |
Token | a |
Token ID | 257 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
GBT | +0.391 |
Catal | +0.346 |
Bah | +0.341 |
Franç | +0.337 |
ンジ | +0.326 |
Neg logit contributions | |
atti | -0.316 |
etti | -0.304 |
karma | -0.304 |
latex | -0.300 |
strap | -0.282 |
Token | reservoir |
Token ID | 22115 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
GBT | +0.689 |
Catal | +0.612 |
Bah | +0.604 |
Franç | +0.596 |
ンジ | +0.578 |
Neg logit contributions | |
atti | -0.525 |
etti | -0.505 |
karma | -0.503 |
latex | -0.498 |
strap | -0.467 |
Token | of |
Token ID | 286 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
atti | +0.268 |
karma | +0.258 |
etti | +0.258 |
latex | +0.255 |
strap | +0.240 |
Neg logit contributions | |
GBT | -0.330 |
Catal | -0.290 |
Bah | -0.287 |
Franç | -0.281 |
ンジ | -0.273 |
Token | the |
Token ID | 262 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
atti | +0.260 |
etti | +0.251 |
karma | +0.250 |
latex | +0.247 |
strap | +0.231 |
Neg logit contributions | |
GBT | -0.352 |
Catal | -0.313 |
Bah | -0.309 |
Franç | -0.305 |
ンジ | -0.296 |
Token | trust |
Token ID | 3774 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
atti | +0.729 |
karma | +0.708 |
etti | +0.701 |
latex | +0.693 |
Neighborhood | +0.641 |
Neg logit contributions | |
GBT | -1.043 |
Catal | -0.923 |
Bah | -0.913 |
Franç | -0.899 |
ンジ | -0.875 |
Token | that |
Token ID | 326 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
GBT | +0.171 |
Catal | +0.150 |
Bah | +0.148 |
Franç | +0.146 |
ンジ | +0.141 |
Neg logit contributions | |
atti | -0.149 |
etti | -0.144 |
karma | -0.143 |
latex | -0.142 |
strap | -0.134 |
Token | is |
Token ID | 318 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
atti | +0.096 |
etti | +0.092 |
karma | +0.092 |
latex | +0.091 |
strap | +0.086 |
Neg logit contributions | |
GBT | -0.120 |
Catal | -0.106 |
Bah | -0.105 |
Franç | -0.103 |
ンジ | -0.100 |
Token | a |
Token ID | 257 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
atti | +0.285 |
etti | +0.274 |
karma | +0.272 |
latex | +0.270 |
strap | +0.252 |
Neg logit contributions | |
GBT | -0.367 |
Catal | -0.324 |
Bah | -0.320 |
Franç | -0.316 |
ンジ | -0.308 |
Token | statement |
Token ID | 2643 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
atti | +0.022 |
etti | +0.021 |
karma | +0.021 |
latex | +0.020 |
strap | +0.019 |
Neg logit contributions | |
GBT | -0.030 |
Catal | -0.027 |
Bah | -0.027 |
Franç | -0.026 |
ンジ | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
GBT | +1.028 |
Catal | +0.901 |
Bah | +0.890 |
Franç | +0.878 |
ンジ | +0.844 |
Neg logit contributions | |
atti | -0.980 |
etti | -0.946 |
karma | -0.939 |
latex | -0.934 |
strap | -0.883 |
Token |
|
Token ID | 198 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
GBT | +0.130 |
Catal | +0.115 |
Bah | +0.114 |
Franç | +0.113 |
ンジ | +0.109 |
Neg logit contributions | |
atti | -0.102 |
etti | -0.098 |
karma | -0.097 |
latex | -0.096 |
strap | -0.091 |
Token |
|
Token ID | 198 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
GBT | +0.482 |
Catal | +0.424 |
Bah | +0.418 |
Franç | +0.412 |
ンジ | +0.397 |
Neg logit contributions | |
atti | -0.441 |
etti | -0.426 |
karma | -0.423 |
latex | -0.420 |
strap | -0.396 |
Token | Trump |
Token ID | 6170 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
GBT | +0.271 |
Catal | +0.236 |
Bah | +0.233 |
Franç | +0.233 |
ンジ | +0.224 |
Neg logit contributions | |
atti | -0.256 |
etti | -0.247 |
karma | -0.244 |
latex | -0.241 |
strap | -0.230 |
Token | says |
Token ID | 1139 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
atti | +0.413 |
etti | +0.400 |
karma | +0.396 |
latex | +0.391 |
strap | +0.366 |
Neg logit contributions | |
GBT | -0.540 |
Catal | -0.479 |
Bah | -0.472 |
Franç | -0.465 |
ンジ | -0.454 |
Token | he |
Token ID | 339 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
GBT | +0.486 |
Catal | +0.421 |
Bah | +0.416 |
Franç | +0.409 |
ンジ | +0.391 |
Neg logit contributions | |
atti | -0.536 |
etti | -0.519 |
karma | -0.515 |
latex | -0.513 |
strap | -0.486 |
Token | WILL |
Token ID | 17682 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
GBT | +0.128 |
Catal | +0.114 |
Bah | +0.112 |
Franç | +0.111 |
ンジ | +0.107 |
Neg logit contributions | |
atti | -0.103 |
etti | -0.099 |
karma | -0.098 |
latex | -0.097 |
strap | -0.091 |
Token | NOT |
Token ID | 5626 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
GBT | +0.234 |
Catal | +0.207 |
Bah | +0.205 |
Franç | +0.202 |
ンジ | +0.195 |
Neg logit contributions | |
atti | -0.190 |
etti | -0.183 |
karma | -0.182 |
latex | -0.181 |
strap | -0.170 |
Token | debate |
Token ID | 4384 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
GBT | +0.327 |
Catal | +0.290 |
Bah | +0.287 |
Franç | +0.283 |
ンジ | +0.274 |
Neg logit contributions | |
atti | -0.256 |
etti | -0.247 |
karma | -0.245 |
latex | -0.243 |
strap | -0.228 |
Token | pay |
Token ID | 1414 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
GBT | +0.782 |
Catal | +0.700 |
Bah | +0.694 |
Franç | +0.686 |
ンジ | +0.665 |
Neg logit contributions | |
atti | -0.510 |
etti | -0.490 |
karma | -0.481 |
latex | -0.480 |
strap | -0.448 |
Token | prospects |
Token ID | 13285 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
GBT | +0.024 |
Catal | +0.021 |
Bah | +0.021 |
Franç | +0.021 |
ンジ | +0.020 |
Neg logit contributions | |
atti | -0.022 |
etti | -0.021 |
karma | -0.021 |
latex | -0.021 |
strap | -0.019 |
Token | ". |
Token ID | 1911 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
GBT | +1.415 |
Catal | +1.246 |
Bah | +1.234 |
Franç | +1.224 |
Goku | +1.174 |
Neg logit contributions | |
atti | -1.316 |
etti | -1.261 |
karma | -1.250 |
latex | -1.246 |
strap | -1.181 |
Token | " |
Token ID | 366 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
GBT | +0.515 |
Catal | +0.454 |
Bah | +0.449 |
Franç | +0.443 |
ンジ | +0.429 |
Neg logit contributions | |
atti | -0.419 |
etti | -0.404 |
karma | -0.401 |
latex | -0.398 |
strap | -0.374 |
Token | Against |
Token ID | 39276 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
GBT | +0.562 |
Catal | +0.494 |
Bah | +0.488 |
Franç | +0.483 |
ンジ | +0.465 |
Neg logit contributions | |
atti | -0.497 |
etti | -0.479 |
karma | -0.475 |
latex | -0.472 |
strap | -0.445 |
Token | this |
Token ID | 428 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
GBT | +1.023 |
Catal | +0.907 |
Bah | +0.896 |
Franç | +0.887 |
ンジ | +0.857 |
Neg logit contributions | |
atti | -0.789 |
etti | -0.758 |
karma | -0.750 |
latex | -0.746 |
strap | -0.701 |
Token | backdrop |
Token ID | 26373 |
Feature activation | -6.43e-01 |
Pos logit contributions | |
atti | +0.151 |
etti | +0.145 |
karma | +0.144 |
latex | +0.143 |
strap | +0.133 |
Neg logit contributions | |
GBT | -0.226 |
Catal | -0.202 |
Bah | -0.200 |
Franç | -0.197 |
ンジ | -0.192 |
Token | , |
Token ID | 11 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
GBT | +0.494 |
Catal | +0.431 |
Bah | +0.425 |
Franç | +0.420 |
ンジ | +0.401 |
Neg logit contributions | |
atti | -0.513 |
etti | -0.495 |
karma | -0.491 |
latex | -0.489 |
strap | -0.463 |
Token | employers |
Token ID | 11390 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
atti | +0.036 |
etti | +0.035 |
karma | +0.035 |
latex | +0.034 |
strap | +0.032 |
Neg logit contributions | |
GBT | -0.049 |
Catal | -0.043 |
Bah | -0.043 |
Franç | -0.042 |
ンジ | -0.041 |
Token | will |
Token ID | 481 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
GBT | +0.106 |
Catal | +0.094 |
Bah | +0.093 |
Franç | +0.093 |
ンジ | +0.089 |
Neg logit contributions | |
atti | -0.069 |
etti | -0.066 |
karma | -0.066 |
latex | -0.065 |
strap | -0.061 |
Token | need |
Token ID | 761 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
atti | +0.021 |
etti | +0.020 |
karma | +0.020 |
latex | +0.020 |
strap | +0.018 |
Neg logit contributions | |
GBT | -0.032 |
Catal | -0.029 |
Bah | -0.028 |
Franç | -0.028 |
ンジ | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
atti | +0.176 |
etti | +0.168 |
karma | +0.168 |
latex | +0.167 |
strap | +0.156 |
Neg logit contributions | |
GBT | -0.228 |
Catal | -0.202 |
Bah | -0.200 |
Franç | -0.197 |
ンジ | -0.191 |
Token | last |
Token ID | 938 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
GBT | +0.366 |
Catal | +0.325 |
Bah | +0.321 |
Franç | +0.318 |
ンジ | +0.307 |
Neg logit contributions | |
atti | -0.276 |
etti | -0.266 |
karma | -0.263 |
latex | -0.261 |
strap | -0.245 |
Token | possible |
Token ID | 1744 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
GBT | +1.585 |
Catal | +1.412 |
Bah | +1.397 |
Franç | +1.380 |
ンジ | +1.333 |
Neg logit contributions | |
atti | -1.131 |
etti | -1.087 |
karma | -1.073 |
latex | -1.065 |
strap | -0.998 |
Token | moment |
Token ID | 2589 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
atti | +0.470 |
karma | +0.446 |
etti | +0.446 |
latex | +0.443 |
strap | +0.404 |
Neg logit contributions | |
GBT | -0.907 |
Catal | -0.818 |
Bah | -0.810 |
Franç | -0.800 |
ンジ | -0.784 |
Token | before |
Token ID | 878 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
atti | +1.501 |
karma | +1.460 |
etti | +1.455 |
latex | +1.445 |
strap | +1.370 |
Neg logit contributions | |
GBT | -1.633 |
Catal | -1.426 |
Bah | -1.402 |
Franç | -1.373 |
ンジ | -1.349 |
Token | the |
Token ID | 262 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
GBT | +0.647 |
Catal | +0.572 |
Bah | +0.568 |
Franç | +0.563 |
Goku | +0.545 |
Neg logit contributions | |
atti | -0.543 |
etti | -0.520 |
latex | -0.511 |
karma | -0.511 |
strap | -0.482 |
Token | final |
Token ID | 2457 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
GBT | +0.184 |
Catal | +0.163 |
Bah | +0.161 |
Franç | +0.159 |
ンジ | +0.154 |
Neg logit contributions | |
atti | -0.141 |
etti | -0.136 |
karma | -0.135 |
latex | -0.134 |
strap | -0.126 |
Token | verdict |
Token ID | 15593 |
Feature activation | +9.74e-01 |
Pos logit contributions | |
GBT | +1.308 |
Catal | +1.170 |
Bah | +1.159 |
Franç | +1.146 |
ンジ | +1.108 |
Neg logit contributions | |
atti | -0.841 |
etti | -0.805 |
karma | -0.795 |
latex | -0.790 |
strap | -0.736 |
Token | was |
Token ID | 373 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
atti | +0.696 |
etti | +0.671 |
karma | +0.669 |
latex | +0.663 |
strap | +0.624 |
Neg logit contributions | |
GBT | -0.834 |
Catal | -0.738 |
Bah | -0.725 |
Franç | -0.713 |
ンジ | -0.698 |
Token | written |
Token ID | 3194 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
GBT | +0.723 |
Catal | +0.642 |
Bah | +0.637 |
Franç | +0.631 |
Goku | +0.610 |
Neg logit contributions | |
atti | -0.511 |
etti | -0.491 |
karma | -0.484 |
latex | -0.481 |
strap | -0.451 |
Token | : |
Token ID | 25 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
GBT | +1.450 |
Catal | +1.251 |
Bah | +1.239 |
Franç | +1.225 |
Goku | +1.176 |
Neg logit contributions | |
atti | -1.682 |
etti | -1.620 |
latex | -1.602 |
karma | -1.601 |
strap | -1.521 |
Token | WHO |
Token ID | 19494 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
GBT | +1.110 |
Catal | +0.983 |
Bah | +0.972 |
Franç | +0.961 |
ンジ | +0.926 |
Neg logit contributions | |
atti | -0.920 |
etti | -0.885 |
karma | -0.879 |
latex | -0.872 |
strap | -0.820 |
Token | spokesman |
Token ID | 6523 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
GBT | +0.881 |
Catal | +0.777 |
Bah | +0.770 |
Franç | +0.760 |
ンジ | +0.732 |
Neg logit contributions | |
atti | -0.735 |
etti | -0.707 |
karma | -0.702 |
latex | -0.695 |
strap | -0.656 |
Token | Gregory |
Token ID | 20653 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
GBT | +0.276 |
Catal | +0.241 |
Bah | +0.238 |
Franç | +0.233 |
ンジ | +0.226 |
Neg logit contributions | |
atti | -0.267 |
etti | -0.259 |
karma | -0.257 |
latex | -0.255 |
strap | -0.241 |
Token | Hart |
Token ID | 11345 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
atti | +0.147 |
etti | +0.142 |
karma | +0.141 |
latex | +0.140 |
strap | +0.132 |
Neg logit contributions | |
GBT | -0.163 |
Catal | -0.143 |
Bah | -0.141 |
Franç | -0.139 |
ンジ | -0.135 |
Token | l |
Token ID | 75 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
GBT | +0.051 |
Catal | +0.045 |
Bah | +0.044 |
Franç | +0.044 |
ンジ | +0.042 |
Neg logit contributions | |
atti | -0.039 |
etti | -0.038 |
karma | -0.037 |
latex | -0.037 |
strap | -0.035 |
Token | said |
Token ID | 531 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
atti | +0.115 |
etti | +0.111 |
karma | +0.109 |
latex | +0.108 |
strap | +0.101 |
Neg logit contributions | |
GBT | -0.181 |
Catal | -0.162 |
Bah | -0.160 |
Franç | -0.158 |
ンジ | -0.154 |
Token | it |
Token ID | 340 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
GBT | +0.412 |
Catal | +0.364 |
Bah | +0.361 |
Franç | +0.356 |
ンジ | +0.342 |
Neg logit contributions | |
atti | -0.350 |
etti | -0.336 |
karma | -0.333 |
latex | -0.330 |
strap | -0.312 |
Token | had |
Token ID | 550 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
atti | +0.440 |
etti | +0.423 |
karma | +0.421 |
latex | +0.419 |
strap | +0.390 |
Neg logit contributions | |
GBT | -0.590 |
Catal | -0.525 |
Bah | -0.518 |
Franç | -0.509 |
ンジ | -0.497 |
Token | been |
Token ID | 587 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
GBT | +0.489 |
Catal | +0.435 |
Bah | +0.430 |
Franç | +0.426 |
ンジ | +0.411 |
Neg logit contributions | |
atti | -0.350 |
etti | -0.337 |
karma | -0.333 |
latex | -0.330 |
strap | -0.309 |
Token | expecting |
Token ID | 12451 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
atti | +0.350 |
etti | +0.336 |
karma | +0.335 |
latex | +0.332 |
strap | +0.311 |
Neg logit contributions | |
GBT | -0.471 |
Catal | -0.419 |
Bah | -0.415 |
Franç | -0.409 |
ンジ | -0.397 |
Token | something |
Token ID | 1223 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
GBT | +0.522 |
Catal | +0.461 |
Bah | +0.457 |
Franç | +0.452 |
ンジ | +0.434 |
Neg logit contributions | |
atti | -0.435 |
etti | -0.418 |
karma | -0.414 |
latex | -0.410 |
strap | -0.387 |
Token | before |
Token ID | 878 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
atti | +0.208 |
etti | +0.201 |
karma | +0.200 |
latex | +0.198 |
strap | +0.186 |
Neg logit contributions | |
GBT | -0.255 |
Catal | -0.226 |
Bah | -0.223 |
Franç | -0.220 |
ンジ | -0.213 |
Token | . |
Token ID | 13 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
GBT | +0.579 |
Catal | +0.503 |
Bah | +0.496 |
Franç | +0.489 |
ンジ | +0.468 |
Neg logit contributions | |
atti | -0.619 |
etti | -0.598 |
karma | -0.594 |
latex | -0.591 |
strap | -0.560 |
Token | They |
Token ID | 1119 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
atti | +0.283 |
etti | +0.273 |
karma | +0.271 |
latex | +0.268 |
strap | +0.253 |
Neg logit contributions | |
GBT | -0.350 |
Catal | -0.308 |
Bah | -0.304 |
Franç | -0.299 |
ンジ | -0.292 |
Token | have |
Token ID | 423 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
GBT | +0.627 |
Catal | +0.560 |
Bah | +0.554 |
Franç | +0.548 |
ンジ | +0.531 |
Neg logit contributions | |
atti | -0.421 |
etti | -0.403 |
karma | -0.400 |
latex | -0.397 |
strap | -0.370 |
Token | not |
Token ID | 407 |
Feature activation | -8.99e-01 |
Pos logit contributions | |
GBT | +1.272 |
Catal | +1.138 |
Bah | +1.126 |
Franç | +1.118 |
ンジ | +1.077 |
Neg logit contributions | |
atti | -0.840 |
etti | -0.803 |
karma | -0.791 |
latex | -0.787 |
strap | -0.734 |
Token | been |
Token ID | 587 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
GBT | +0.853 |
Catal | +0.764 |
Bah | +0.757 |
Franç | +0.751 |
ンジ | +0.725 |
Neg logit contributions | |
atti | -0.555 |
etti | -0.529 |
karma | -0.523 |
latex | -0.521 |
strap | -0.485 |
Token | lull |
Token ID | 42858 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
GBT | +1.592 |
Catal | +1.416 |
Bah | +1.401 |
Franç | +1.390 |
Goku | +1.346 |
Neg logit contributions | |
atti | -1.115 |
etti | -1.070 |
karma | -1.052 |
latex | -1.047 |
strap | -0.978 |
Token | ed |
Token ID | 276 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
GBT | +0.385 |
Catal | +0.337 |
Bah | +0.333 |
Franç | +0.328 |
ンジ | +0.316 |
Neg logit contributions | |
atti | -0.366 |
etti | -0.353 |
karma | -0.351 |
latex | -0.349 |
strap | -0.329 |
Token | into |
Token ID | 656 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
GBT | +0.545 |
Catal | +0.486 |
Bah | +0.481 |
Franç | +0.475 |
ンジ | +0.460 |
Neg logit contributions | |
atti | -0.386 |
etti | -0.370 |
karma | -0.367 |
latex | -0.365 |
strap | -0.341 |
Token | compl |
Token ID | 2299 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
atti | +0.874 |
etti | +0.839 |
karma | +0.838 |
latex | +0.827 |
strap | +0.768 |
Neg logit contributions | |
GBT | -1.264 |
Catal | -1.124 |
Bah | -1.109 |
Franç | -1.094 |
ンジ | -1.071 |
Token | ac |
Token ID | 330 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
GBT | +0.188 |
Catal | +0.159 |
Bah | +0.157 |
Franç | +0.154 |
ンジ | +0.147 |
Neg logit contributions | |
atti | -0.262 |
etti | -0.255 |
karma | -0.253 |
latex | -0.252 |
strap | -0.240 |
Token | the |
Token ID | 262 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
atti | +1.725 |
etti | +1.677 |
karma | +1.675 |
latex | +1.663 |
Neighborhood | +1.547 |
Neg logit contributions | |
GBT | -2.009 |
Catal | -1.777 |
Bah | -1.744 |
Franç | -1.721 |
ンジ | -1.674 |
Token | next |
Token ID | 1306 |
Feature activation | -9.71e-01 |
Pos logit contributions | |
atti | +0.387 |
etti | +0.373 |
karma | +0.372 |
latex | +0.370 |
Neighborhood | +0.344 |
Neg logit contributions | |
GBT | -0.491 |
Catal | -0.435 |
Bah | -0.429 |
Franç | -0.423 |
ンジ | -0.410 |
Token | time |
Token ID | 640 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
GBT | +0.866 |
Catal | +0.769 |
Bah | +0.760 |
Franç | +0.751 |
ンジ | +0.727 |
Neg logit contributions | |
atti | -0.656 |
etti | -0.631 |
karma | -0.626 |
latex | -0.622 |
strap | -0.582 |
Token | around |
Token ID | 1088 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
atti | +0.481 |
etti | +0.464 |
karma | +0.463 |
latex | +0.460 |
strap | +0.433 |
Neg logit contributions | |
GBT | -0.522 |
Catal | -0.458 |
Bah | -0.452 |
Franç | -0.442 |
ンジ | -0.429 |
Token | . |
Token ID | 13 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
atti | +0.389 |
etti | +0.377 |
karma | +0.375 |
latex | +0.372 |
strap | +0.352 |
Neg logit contributions | |
GBT | -0.383 |
Catal | -0.333 |
Bah | -0.329 |
Franç | -0.323 |
ンジ | -0.313 |
Token | So |
Token ID | 1406 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
atti | +0.098 |
etti | +0.095 |
karma | +0.094 |
latex | +0.093 |
strap | +0.088 |
Neg logit contributions | |
GBT | -0.116 |
Catal | -0.102 |
Bah | -0.101 |
Franç | -0.100 |
ンジ | -0.096 |
Token | certainly |
Token ID | 3729 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
GBT | +0.675 |
Catal | +0.594 |
Bah | +0.587 |
Franç | +0.580 |
ンジ | +0.557 |
Neg logit contributions | |
atti | -0.595 |
etti | -0.573 |
karma | -0.568 |
latex | -0.565 |
strap | -0.533 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
GBT | +1.526 |
Catal | +1.347 |
Bah | +1.331 |
Franç | +1.318 |
Goku | +1.269 |
Neg logit contributions | |
atti | -1.348 |
etti | -1.298 |
karma | -1.280 |
latex | -1.278 |
strap | -1.207 |
Token | this |
Token ID | 428 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
GBT | +1.045 |
Catal | +0.924 |
Bah | +0.913 |
Franç | +0.904 |
Goku | +0.871 |
Neg logit contributions | |
atti | -0.880 |
etti | -0.846 |
karma | -0.835 |
latex | -0.834 |
strap | -0.786 |
Token | is |
Token ID | 318 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
GBT | +0.674 |
Catal | +0.593 |
Bah | +0.586 |
Franç | +0.579 |
ンジ | +0.559 |
Neg logit contributions | |
atti | -0.578 |
etti | -0.557 |
karma | -0.553 |
latex | -0.550 |
strap | -0.517 |
Token | not |
Token ID | 407 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
GBT | +0.880 |
Catal | +0.779 |
Bah | +0.770 |
Franç | +0.761 |
Goku | +0.732 |
Neg logit contributions | |
atti | -0.716 |
etti | -0.690 |
karma | -0.681 |
latex | -0.678 |
strap | -0.639 |
Token | á |
Token ID | 6557 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
GBT | +0.517 |
Catal | +0.453 |
Bah | +0.448 |
Franç | +0.442 |
ンジ | +0.426 |
Neg logit contributions | |
atti | -0.473 |
etti | -0.457 |
karma | -0.454 |
latex | -0.452 |
strap | -0.426 |
Token | z |
Token ID | 89 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
GBT | +1.555 |
Catal | +1.375 |
Bah | +1.362 |
Franç | +1.345 |
ンジ | +1.292 |
Neg logit contributions | |
atti | -1.303 |
etti | -1.256 |
karma | -1.253 |
latex | -1.247 |
strap | -1.165 |
Token | ( |
Token ID | 357 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
GBT | +1.496 |
Catal | +1.305 |
Bah | +1.288 |
Franç | +1.270 |
ンジ | +1.221 |
Neg logit contributions | |
atti | -1.490 |
etti | -1.441 |
karma | -1.432 |
latex | -1.423 |
strap | -1.346 |
Token | Mich |
Token ID | 11180 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
GBT | +0.966 |
Catal | +0.849 |
Bah | +0.839 |
Franç | +0.827 |
ンジ | +0.798 |
Neg logit contributions | |
atti | -0.868 |
etti | -0.837 |
karma | -0.832 |
latex | -0.826 |
strap | -0.779 |
Token | l |
Token ID | 75 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
GBT | +0.997 |
Catal | +0.858 |
Bah | +0.846 |
Franç | +0.832 |
ンジ | +0.799 |
Neg logit contributions | |
atti | -1.164 |
etti | -1.128 |
karma | -1.120 |
latex | -1.113 |
strap | -1.058 |
Token | í |
Token ID | 8836 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
GBT | +1.512 |
Catal | +1.312 |
Bah | +1.298 |
Franç | +1.278 |
ンジ | +1.223 |
Neg logit contributions | |
atti | -1.621 |
etti | -1.572 |
karma | -1.571 |
latex | -1.564 |
strap | -1.478 |
Token | k |
Token ID | 74 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
GBT | +0.126 |
Catal | +0.111 |
Bah | +0.110 |
Franç | +0.108 |
ンジ | +0.105 |
Neg logit contributions | |
atti | -0.109 |
etti | -0.105 |
karma | -0.104 |
latex | -0.103 |
strap | -0.097 |
Token | 41 |
Token ID | 6073 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
atti | +0.174 |
etti | +0.168 |
karma | +0.166 |
latex | +0.165 |
strap | +0.157 |
Neg logit contributions | |
GBT | -0.156 |
Catal | -0.135 |
Bah | -0.132 |
Franç | -0.130 |
ンジ | -0.127 |
Token | ), |
Token ID | 828 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
GBT | +0.281 |
Catal | +0.246 |
Bah | +0.243 |
Franç | +0.239 |
ンジ | +0.232 |
Neg logit contributions | |
atti | -0.259 |
etti | -0.250 |
karma | -0.249 |
latex | -0.247 |
strap | -0.233 |
Token | Kr |
Token ID | 13685 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
GBT | +0.533 |
Catal | +0.467 |
Bah | +0.461 |
Franç | +0.454 |
ンジ | +0.439 |
Neg logit contributions | |
atti | -0.491 |
etti | -0.474 |
karma | -0.470 |
latex | -0.467 |
strap | -0.441 |
Token | ?, |
Token ID | 21747 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
GBT | +0.683 |
Catal | +0.592 |
Bah | +0.583 |
Franç | +0.576 |
ンジ | +0.555 |
Neg logit contributions | |
atti | -0.728 |
etti | -0.704 |
karma | -0.695 |
latex | -0.693 |
strap | -0.658 |
Token | doesn |
Token ID | 1595 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
GBT | +0.268 |
Catal | +0.236 |
Bah | +0.233 |
Franç | +0.230 |
ンジ | +0.222 |
Neg logit contributions | |
atti | -0.237 |
etti | -0.229 |
karma | -0.227 |
latex | -0.225 |
strap | -0.212 |
Token | 't |
Token ID | 470 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
atti | +0.290 |
etti | +0.280 |
karma | +0.278 |
latex | +0.276 |
strap | +0.260 |
Neg logit contributions | |
GBT | -0.328 |
Catal | -0.288 |
Bah | -0.285 |
Franç | -0.281 |
ンジ | -0.271 |
Token | need |
Token ID | 761 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
GBT | +0.059 |
Catal | +0.052 |
Bah | +0.052 |
Franç | +0.051 |
ンジ | +0.049 |
Neg logit contributions | |
atti | -0.042 |
etti | -0.041 |
karma | -0.040 |
latex | -0.040 |
strap | -0.037 |
Token | any |
Token ID | 597 |
Feature activation | +5.95e-02 |
Pos logit contributions | |
atti | +0.414 |
etti | +0.400 |
karma | +0.397 |
latex | +0.396 |
strap | +0.371 |
Neg logit contributions | |
GBT | -0.489 |
Catal | -0.432 |
Bah | -0.427 |
Franç | -0.421 |
ンジ | -0.408 |
Token | further |
Token ID | 2252 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
atti | +0.038 |
etti | +0.036 |
karma | +0.036 |
latex | +0.036 |
strap | +0.033 |
Neg logit contributions | |
GBT | -0.055 |
Catal | -0.049 |
Bah | -0.049 |
Franç | -0.048 |
ンジ | -0.047 |
Token | verification |
Token ID | 19637 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
GBT | +1.474 |
Catal | +1.301 |
Bah | +1.295 |
Franç | +1.289 |
Goku | +1.248 |
Neg logit contributions | |
atti | -1.028 |
etti | -0.982 |
karma | -0.958 |
latex | -0.952 |
strap | -0.903 |
Token | . |
Token ID | 13 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
GBT | +2.740 |
Catal | +2.389 |
Bah | +2.378 |
Franç | +2.333 |
Goku | +2.262 |
Neg logit contributions | |
atti | -3.027 |
etti | -2.926 |
latex | -2.883 |
karma | -2.881 |
strap | -2.719 |
Token |
|
Token ID | 198 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
GBT | +0.830 |
Catal | +0.731 |
Bah | +0.723 |
Franç | +0.713 |
ンジ | +0.687 |
Neg logit contributions | |
atti | -0.723 |
etti | -0.697 |
karma | -0.692 |
latex | -0.688 |
strap | -0.647 |
Token |
|
Token ID | 198 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
GBT | +1.075 |
Catal | +0.946 |
Bah | +0.935 |
Franç | +0.916 |
ンジ | +0.884 |
Neg logit contributions | |
atti | -1.003 |
etti | -0.970 |
karma | -0.963 |
latex | -0.959 |
strap | -0.900 |
Token | ** |
Token ID | 1174 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
GBT | +0.826 |
Catal | +0.725 |
Bah | +0.716 |
Franç | +0.705 |
ンジ | +0.680 |
Neg logit contributions | |
atti | -0.762 |
etti | -0.736 |
karma | -0.731 |
latex | -0.726 |
strap | -0.685 |
Token | In |
Token ID | 554 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
atti | +0.042 |
etti | +0.040 |
karma | +0.040 |
latex | +0.040 |
strap | +0.038 |
Neg logit contributions | |
GBT | -0.047 |
Catal | -0.041 |
Bah | -0.040 |
Franç | -0.040 |
ンジ | -0.039 |
Token | Ash |
Token ID | 7844 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
atti | +0.203 |
etti | +0.197 |
karma | +0.195 |
latex | +0.194 |
strap | +0.183 |
Neg logit contributions | |
GBT | -0.221 |
Catal | -0.193 |
Bah | -0.191 |
Franç | -0.188 |
ンジ | -0.182 |
Token | raf |
Token ID | 32188 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
GBT | +0.440 |
Catal | +0.380 |
Bah | +0.372 |
Franç | +0.367 |
ンジ | +0.353 |
Neg logit contributions | |
atti | -0.516 |
etti | -0.501 |
karma | -0.497 |
latex | -0.492 |
strap | -0.471 |
Token | Gh |
Token ID | 11972 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
GBT | +0.402 |
Catal | +0.350 |
Bah | +0.344 |
Franç | +0.339 |
ンジ | +0.327 |
Neg logit contributions | |
atti | -0.394 |
etti | -0.381 |
karma | -0.378 |
latex | -0.374 |
strap | -0.355 |
Token | ani |
Token ID | 3216 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
atti | +0.102 |
etti | +0.098 |
karma | +0.095 |
latex | +0.094 |
strap | +0.090 |
Neg logit contributions | |
GBT | -0.140 |
Catal | -0.124 |
Franç | -0.121 |
Bah | -0.121 |
ンジ | -0.118 |
Token | came |
Token ID | 1625 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
GBT | +1.255 |
Catal | +1.107 |
Bah | +1.099 |
Franç | +1.084 |
ンジ | +1.041 |
Neg logit contributions | |
atti | -1.057 |
etti | -1.016 |
karma | -1.011 |
latex | -1.006 |
strap | -0.946 |
Token | second |
Token ID | 1218 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
GBT | +0.847 |
Catal | +0.736 |
Bah | +0.728 |
Franç | +0.715 |
ンジ | +0.689 |
Neg logit contributions | |
atti | -0.858 |
etti | -0.830 |
karma | -0.824 |
latex | -0.820 |
strap | -0.776 |
Token | with |
Token ID | 351 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
GBT | +3.008 |
Bah | +2.621 |
Catal | +2.613 |
Franç | +2.599 |
Goku | +2.495 |
Neg logit contributions | |
atti | -3.354 |
etti | -3.249 |
karma | -3.172 |
latex | -3.167 |
strap | -3.027 |
Token | 31 |
Token ID | 3261 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
GBT | +0.888 |
Catal | +0.782 |
Bah | +0.774 |
Franç | +0.764 |
ンジ | +0.734 |
Neg logit contributions | |
atti | -0.778 |
etti | -0.751 |
karma | -0.742 |
latex | -0.738 |
strap | -0.697 |
Token | . |
Token ID | 13 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
GBT | +1.311 |
Catal | +1.138 |
Bah | +1.125 |
Franç | +1.110 |
ンジ | +1.061 |
Neg logit contributions | |
atti | -1.392 |
etti | -1.344 |
karma | -1.330 |
latex | -1.324 |
strap | -1.254 |
Token | 5 |
Token ID | 20 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
GBT | +0.603 |
Catal | +0.527 |
Bah | +0.521 |
Franç | +0.513 |
ンジ | +0.494 |
Neg logit contributions | |
atti | -0.590 |
etti | -0.570 |
karma | -0.566 |
latex | -0.564 |
strap | -0.532 |
Token | %. |
Token ID | 7225 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
atti | +0.664 |
etti | +0.642 |
karma | +0.642 |
latex | +0.636 |
strap | +0.604 |
Neg logit contributions | |
GBT | -0.646 |
Catal | -0.563 |
Bah | -0.552 |
Franç | -0.545 |
ンジ | -0.526 |
Token | or |
Token ID | 393 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
GBT | +0.962 |
Catal | +0.841 |
Bah | +0.830 |
Franç | +0.819 |
ンジ | +0.789 |
Neg logit contributions | |
atti | -0.935 |
etti | -0.903 |
karma | -0.898 |
latex | -0.892 |
strap | -0.843 |
Token | D |
Token ID | 360 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
atti | +0.197 |
etti | +0.191 |
karma | +0.190 |
latex | +0.188 |
strap | +0.178 |
Neg logit contributions | |
GBT | -0.190 |
Catal | -0.165 |
Bah | -0.163 |
Franç | -0.161 |
ンジ | -0.155 |
Token | ). |
Token ID | 737 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
GBT | +0.284 |
Catal | +0.249 |
Bah | +0.245 |
Franç | +0.243 |
ンジ | +0.234 |
Neg logit contributions | |
atti | -0.272 |
etti | -0.262 |
karma | -0.261 |
latex | -0.258 |
strap | -0.244 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
GBT | +0.440 |
Catal | +0.386 |
Bah | +0.381 |
Franç | +0.376 |
ンジ | +0.364 |
Neg logit contributions | |
atti | -0.386 |
etti | -0.372 |
karma | -0.369 |
latex | -0.367 |
strap | -0.345 |
Token | The |
Token ID | 464 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
GBT | +0.384 |
Catal | +0.336 |
Bah | +0.332 |
Franç | +0.328 |
ンジ | +0.316 |
Neg logit contributions | |
atti | -0.359 |
etti | -0.346 |
karma | -0.343 |
latex | -0.341 |
strap | -0.323 |
Token | explanations |
Token ID | 18681 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
GBT | +1.134 |
Catal | +1.005 |
Bah | +0.992 |
Franç | +0.980 |
ンジ | +0.944 |
Neg logit contributions | |
atti | -0.931 |
etti | -0.897 |
karma | -0.887 |
latex | -0.881 |
strap | -0.833 |
Token | in |
Token ID | 287 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
GBT | +2.726 |
Catal | +2.439 |
Bah | +2.401 |
Franç | +2.390 |
Goku | +2.313 |
Neg logit contributions | |
atti | -2.143 |
etti | -2.063 |
karma | -2.026 |
latex | -2.007 |
strap | -1.903 |
Token | each |
Token ID | 1123 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
GBT | +0.070 |
Catal | +0.062 |
Bah | +0.061 |
Franç | +0.060 |
ンジ | +0.058 |
Neg logit contributions | |
atti | -0.054 |
etti | -0.052 |
karma | -0.052 |
latex | -0.052 |
strap | -0.048 |
Token | case |
Token ID | 1339 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
GBT | +1.290 |
Catal | +1.136 |
Bah | +1.121 |
Franç | +1.109 |
Goku | +1.066 |
Neg logit contributions | |
atti | -1.140 |
etti | -1.101 |
karma | -1.086 |
latex | -1.078 |
strap | -1.021 |
Token | are |
Token ID | 389 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
atti | +0.393 |
etti | +0.377 |
karma | +0.375 |
latex | +0.372 |
strap | +0.347 |
Neg logit contributions | |
GBT | -0.556 |
Catal | -0.495 |
Bah | -0.489 |
Franç | -0.483 |
ンジ | -0.470 |
Token | slightly |
Token ID | 4622 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
GBT | +0.284 |
Catal | +0.252 |
Bah | +0.249 |
Franç | +0.246 |
ンジ | +0.238 |
Neg logit contributions | |
atti | -0.216 |
etti | -0.208 |
karma | -0.205 |
latex | -0.204 |
strap | -0.192 |
Token | Of |
Token ID | 5189 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
GBT | +0.498 |
Catal | +0.437 |
Bah | +0.431 |
Franç | +0.426 |
ンジ | +0.411 |
Neg logit contributions | |
atti | -0.452 |
etti | -0.436 |
karma | -0.432 |
latex | -0.430 |
strap | -0.405 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
GBT | +1.060 |
Catal | +0.904 |
Bah | +0.892 |
Franç | +0.880 |
ンジ | +0.837 |
Neg logit contributions | |
atti | -1.345 |
etti | -1.308 |
karma | -1.299 |
latex | -1.293 |
strap | -1.235 |
Token | investigation |
Token ID | 3645 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
GBT | +0.948 |
Catal | +0.840 |
Bah | +0.831 |
Franç | +0.821 |
ンジ | +0.792 |
Neg logit contributions | |
atti | -0.745 |
etti | -0.717 |
karma | -0.709 |
latex | -0.705 |
strap | -0.664 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
atti | +0.355 |
etti | +0.342 |
karma | +0.340 |
latex | +0.338 |
strap | +0.318 |
Neg logit contributions | |
GBT | -0.404 |
Catal | -0.355 |
Bah | -0.351 |
Franç | -0.346 |
ンジ | -0.335 |
Token | The |
Token ID | 383 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
atti | +0.146 |
etti | +0.139 |
karma | +0.138 |
latex | +0.137 |
strap | +0.128 |
Neg logit contributions | |
GBT | -0.224 |
Catal | -0.200 |
Bah | -0.198 |
Franç | -0.197 |
ンジ | -0.190 |
Token | Daily |
Token ID | 6714 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
GBT | +0.628 |
Catal | +0.557 |
Bah | +0.551 |
Franç | +0.544 |
ンジ | +0.525 |
Neg logit contributions | |
atti | -0.496 |
etti | -0.477 |
karma | -0.473 |
latex | -0.470 |
strap | -0.441 |
Token | Caller |
Token ID | 10244 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
GBT | +3.109 |
Catal | +2.781 |
Bah | +2.735 |
Franç | +2.688 |
ンジ | +2.596 |
Neg logit contributions | |
atti | -2.789 |
etti | -2.680 |
karma | -2.660 |
latex | -2.636 |
strap | -2.483 |
Token | co |
Token ID | 763 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
GBT | +0.197 |
Catal | +0.175 |
Bah | +0.173 |
Franç | +0.171 |
ンジ | +0.165 |
Neg logit contributions | |
atti | -0.150 |
etti | -0.144 |
karma | -0.143 |
latex | -0.142 |
strap | -0.133 |
Token | - |
Token ID | 12 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
GBT | +0.125 |
Catal | +0.109 |
Bah | +0.107 |
Franç | +0.106 |
ンジ | +0.102 |
Neg logit contributions | |
atti | -0.128 |
etti | -0.124 |
karma | -0.123 |
latex | -0.123 |
strap | -0.116 |
Token | founder |
Token ID | 15454 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
atti | +1.022 |
etti | +0.974 |
karma | +0.964 |
latex | +0.954 |
strap | +0.885 |
Neg logit contributions | |
GBT | -1.784 |
Catal | -1.600 |
Bah | -1.585 |
Franç | -1.571 |
ンジ | -1.525 |
Token | said |
Token ID | 531 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
atti | +0.964 |
etti | +0.929 |
karma | +0.921 |
latex | +0.915 |
strap | +0.858 |
Neg logit contributions | |
GBT | -1.207 |
Catal | -1.067 |
Bah | -1.056 |
Franç | -1.041 |
ンジ | -1.008 |
Token | be |
Token ID | 307 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
GBT | +0.455 |
Catal | +0.405 |
Bah | +0.401 |
Franç | +0.396 |
ンジ | +0.383 |
Neg logit contributions | |
atti | -0.323 |
etti | -0.310 |
karma | -0.307 |
latex | -0.305 |
strap | -0.285 |
Token | others |
Token ID | 1854 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
GBT | +0.800 |
Catal | +0.709 |
Bah | +0.701 |
Franç | +0.692 |
ンジ | +0.669 |
Neg logit contributions | |
atti | -0.626 |
etti | -0.602 |
karma | -0.597 |
latex | -0.593 |
strap | -0.557 |
Token | who |
Token ID | 508 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
GBT | +0.425 |
Catal | +0.372 |
Bah | +0.368 |
Franç | +0.362 |
ンジ | +0.351 |
Neg logit contributions | |
atti | -0.391 |
etti | -0.378 |
karma | -0.376 |
latex | -0.373 |
strap | -0.352 |
Token | think |
Token ID | 892 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
atti | +0.767 |
etti | +0.740 |
karma | +0.736 |
latex | +0.730 |
strap | +0.678 |
Neg logit contributions | |
GBT | -1.122 |
Catal | -0.999 |
Bah | -0.986 |
Franç | -0.971 |
ンジ | -0.956 |
Token | that |
Token ID | 326 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
atti | +0.120 |
etti | +0.116 |
karma | +0.115 |
latex | +0.114 |
strap | +0.107 |
Neg logit contributions | |
GBT | -0.154 |
Catal | -0.136 |
Bah | -0.135 |
Franç | -0.133 |
ンジ | -0.129 |
Token | Lord |
Token ID | 4453 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
GBT | +0.011 |
Catal | +0.010 |
Bah | +0.010 |
Franç | +0.010 |
ンジ | +0.010 |
Neg logit contributions | |
atti | -0.009 |
etti | -0.009 |
karma | -0.009 |
latex | -0.009 |
strap | -0.008 |
Token | Hes |
Token ID | 44773 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
GBT | +2.864 |
Catal | +2.531 |
Franç | +2.507 |
Bah | +2.490 |
Goku | +2.417 |
Neg logit contributions | |
atti | -2.347 |
karma | -2.229 |
etti | -2.228 |
latex | -2.227 |
strap | -2.095 |
Token | elt |
Token ID | 2120 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
GBT | +1.603 |
Catal | +1.345 |
Bah | +1.325 |
Franç | +1.290 |
Goku | +1.204 |
Neg logit contributions | |
atti | -2.837 |
karma | -2.788 |
latex | -2.784 |
etti | -2.734 |
Neighborhood | -2.653 |
Token | ine |
Token ID | 500 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
GBT | +1.409 |
Catal | +1.271 |
Bah | +1.260 |
Franç | +1.250 |
Goku | +1.202 |
Neg logit contributions | |
atti | -0.896 |
karma | -0.861 |
etti | -0.854 |
latex | -0.853 |
Neighborhood | -0.785 |
Token | has |
Token ID | 468 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
GBT | +1.075 |
Catal | +0.939 |
Bah | +0.928 |
Franç | +0.916 |
ンジ | +0.881 |
Neg logit contributions | |
atti | -1.036 |
etti | -1.000 |
karma | -0.993 |
latex | -0.987 |
strap | -0.933 |
Token | a |
Token ID | 257 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
GBT | +0.632 |
Catal | +0.562 |
Bah | +0.556 |
Franç | +0.549 |
ンジ | +0.531 |
Neg logit contributions | |
atti | -0.467 |
etti | -0.448 |
karma | -0.444 |
latex | -0.441 |
strap | -0.413 |