Token | The |
Token ID | 464 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
workers | +0.447 |
trainers | +0.438 |
tubing | +0.409 |
HQ | +0.399 |
enta | +0.396 |
Neg logit contributions | |
Crimes | -0.467 |
certific | -0.448 |
witz | -0.443 |
Gutenberg | -0.441 |
unlaw | -0.424 |
Token | Globe |
Token ID | 18260 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
Crimes | +1.378 |
certific | +1.311 |
Gutenberg | +1.297 |
witz | +1.290 |
Vienna | +1.243 |
Neg logit contributions | |
workers | -1.370 |
trainers | -1.318 |
tubing | -1.235 |
HQ | -1.231 |
enta | -1.212 |
Token | 's |
Token ID | 338 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
Crimes | +1.623 |
certific | +1.518 |
Gutenberg | +1.498 |
witz | +1.492 |
unlaw | +1.430 |
Neg logit contributions | |
workers | -1.856 |
trainers | -1.812 |
tubing | -1.702 |
HQ | -1.667 |
enta | -1.666 |
Token | new |
Token ID | 649 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Crimes | +1.945 |
certific | +1.810 |
Gutenberg | +1.785 |
witz | +1.770 |
Vienna | +1.726 |
Neg logit contributions | |
workers | -2.000 |
trainers | -1.916 |
tubing | -1.809 |
HQ | -1.795 |
enta | -1.776 |
Token | Real |
Token ID | 6416 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
workers | +0.308 |
trainers | +0.299 |
tubing | +0.281 |
HQ | +0.277 |
enta | +0.274 |
Neg logit contributions | |
Crimes | -0.296 |
certific | -0.281 |
witz | -0.277 |
Gutenberg | -0.276 |
unlaw | -0.265 |
Token | Estate |
Token ID | 23015 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
Crimes | +0.634 |
certific | +0.593 |
witz | +0.587 |
Gutenberg | +0.584 |
unlaw | +0.556 |
Neg logit contributions | |
workers | -0.835 |
trainers | -0.819 |
tubing | -0.775 |
HQ | -0.758 |
enta | -0.753 |
Token | Beat |
Token ID | 12568 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
Crimes | +7.477 |
certific | +6.607 |
Gutenberg | +6.572 |
witz | +6.433 |
Vienna | +6.378 |
Neg logit contributions | |
workers | -7.305 |
trainers | -7.212 |
tubing | -6.945 |
enta | -6.732 |
joining | -6.690 |
Token | offers |
Token ID | 4394 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
workers | +0.897 |
trainers | +0.875 |
tubing | +0.819 |
HQ | +0.808 |
enta | +0.797 |
Neg logit contributions | |
Crimes | -0.841 |
certific | -0.807 |
witz | -0.793 |
Gutenberg | -0.789 |
unlaw | -0.761 |
Token | news |
Token ID | 1705 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
Crimes | +0.322 |
certific | +0.309 |
witz | +0.303 |
Gutenberg | +0.302 |
Vienna | +0.290 |
Neg logit contributions | |
workers | -0.340 |
trainers | -0.329 |
tubing | -0.309 |
HQ | -0.305 |
enta | -0.302 |
Token | and |
Token ID | 290 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
Crimes | +1.244 |
certific | +1.178 |
witz | +1.166 |
Gutenberg | +1.147 |
Vienna | +1.110 |
Neg logit contributions | |
workers | -1.347 |
trainers | -1.319 |
tubing | -1.230 |
enta | -1.202 |
HQ | -1.201 |
Token | analysis |
Token ID | 3781 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
Crimes | +0.077 |
certific | +0.073 |
witz | +0.072 |
Gutenberg | +0.072 |
unlaw | +0.069 |
Neg logit contributions | |
workers | -0.088 |
trainers | -0.086 |
tubing | -0.081 |
HQ | -0.080 |
enta | -0.079 |
Token | play |
Token ID | 711 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
Crimes | +1.249 |
certific | +1.194 |
witz | +1.177 |
Gutenberg | +1.162 |
Vienna | +1.113 |
Neg logit contributions | |
workers | -1.541 |
trainers | -1.494 |
tubing | -1.407 |
HQ | -1.401 |
enta | -1.378 |
Token | because |
Token ID | 780 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
Crimes | +2.277 |
certific | +2.161 |
witz | +2.134 |
Gutenberg | +2.120 |
Vienna | +2.050 |
Neg logit contributions | |
workers | -2.545 |
trainers | -2.451 |
tubing | -2.320 |
HQ | -2.296 |
enta | -2.272 |
Token | of |
Token ID | 286 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Crimes | +2.242 |
certific | +2.128 |
witz | +2.081 |
Gutenberg | +2.075 |
Vienna | +2.023 |
Neg logit contributions | |
workers | -2.752 |
trainers | -2.633 |
tubing | -2.538 |
enta | -2.499 |
HQ | -2.498 |
Token | the |
Token ID | 262 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
Crimes | +0.585 |
certific | +0.563 |
witz | +0.550 |
Gutenberg | +0.549 |
unlaw | +0.530 |
Neg logit contributions | |
workers | -0.563 |
trainers | -0.541 |
tubing | -0.510 |
HQ | -0.502 |
enta | -0.499 |
Token | Champions |
Token ID | 6662 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
workers | +0.182 |
trainers | +0.176 |
tubing | +0.165 |
HQ | +0.163 |
enta | +0.161 |
Neg logit contributions | |
Crimes | -0.184 |
certific | -0.177 |
witz | -0.174 |
Gutenberg | -0.173 |
unlaw | -0.167 |
Token | League |
Token ID | 4041 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
Crimes | +2.791 |
certific | +2.553 |
Vienna | +2.535 |
Gutenberg | +2.512 |
witz | +2.500 |
Neg logit contributions | |
workers | -4.041 |
trainers | -3.916 |
tubing | -3.712 |
enta | -3.661 |
agen | -3.642 |
Token | , |
Token ID | 11 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Crimes | +5.928 |
certific | +5.529 |
witz | +5.527 |
Gutenberg | +5.412 |
Vienna | +5.409 |
Neg logit contributions | |
workers | -7.204 |
trainers | -6.828 |
agen | -6.542 |
enta | -6.518 |
HQ | -6.514 |
Token | and |
Token ID | 290 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Crimes | +0.426 |
certific | +0.408 |
witz | +0.401 |
Gutenberg | +0.399 |
Vienna | +0.384 |
Neg logit contributions | |
workers | -0.443 |
trainers | -0.430 |
tubing | -0.404 |
HQ | -0.398 |
enta | -0.395 |
Token | we |
Token ID | 356 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Crimes | +0.620 |
certific | +0.593 |
witz | +0.583 |
Gutenberg | +0.580 |
Vienna | +0.559 |
Neg logit contributions | |
workers | -0.647 |
trainers | -0.625 |
tubing | -0.589 |
HQ | -0.582 |
enta | -0.577 |
Token | rarely |
Token ID | 8365 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
Crimes | +0.046 |
certific | +0.045 |
witz | +0.044 |
Gutenberg | +0.044 |
unlaw | +0.042 |
Neg logit contributions | |
workers | -0.044 |
trainers | -0.043 |
tubing | -0.040 |
HQ | -0.039 |
enta | -0.039 |
Token | kept |
Token ID | 4030 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
Crimes | +0.233 |
certific | +0.224 |
witz | +0.220 |
Gutenberg | +0.219 |
unlaw | +0.211 |
Neg logit contributions | |
workers | -0.240 |
trainers | -0.234 |
tubing | -0.219 |
HQ | -0.216 |
enta | -0.213 |
Token | fist |
Token ID | 18606 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
workers | +0.465 |
trainers | +0.456 |
tubing | +0.422 |
HQ | +0.413 |
enta | +0.408 |
Neg logit contributions | |
Crimes | -0.561 |
certific | -0.540 |
witz | -0.532 |
Gutenberg | -0.530 |
unlaw | -0.514 |
Token | while |
Token ID | 981 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
workers | +0.393 |
trainers | +0.384 |
tubing | +0.359 |
HQ | +0.354 |
enta | +0.349 |
Neg logit contributions | |
Crimes | -0.370 |
certific | -0.355 |
witz | -0.349 |
Gutenberg | -0.347 |
unlaw | -0.336 |
Token | an |
Token ID | 281 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
workers | +1.839 |
trainers | +1.795 |
tubing | +1.667 |
HQ | +1.628 |
enta | +1.607 |
Neg logit contributions | |
Crimes | -2.184 |
certific | -2.104 |
witz | -2.068 |
Gutenberg | -2.062 |
unlaw | -2.004 |
Token | ambulance |
Token ID | 22536 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
workers | +0.106 |
trainers | +0.102 |
tubing | +0.095 |
HQ | +0.093 |
enta | +0.092 |
Neg logit contributions | |
Crimes | -0.123 |
certific | -0.120 |
witz | -0.117 |
Gutenberg | -0.117 |
unlaw | -0.113 |
Token | s |
Token ID | 264 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
Crimes | +1.618 |
certific | +1.578 |
witz | +1.533 |
Gutenberg | +1.516 |
Vienna | +1.476 |
Neg logit contributions | |
workers | -1.626 |
trainers | -1.555 |
HQ | -1.476 |
tubing | -1.445 |
enta | -1.436 |
Token | iren |
Token ID | 24080 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
Crimes | +1.638 |
witz | +1.569 |
certific | +1.554 |
Gutenberg | +1.520 |
Vienna | +1.456 |
Neg logit contributions | |
workers | -2.590 |
trainers | -2.514 |
HQ | -2.387 |
tubing | -2.382 |
enta | -2.296 |
Token | w |
Token ID | 266 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
Crimes | +5.858 |
certific | +5.781 |
witz | +5.578 |
Gutenberg | +5.473 |
Vienna | +5.374 |
Neg logit contributions | |
workers | -6.234 |
HQ | -5.826 |
trainers | -5.760 |
agen | -5.534 |
enta | -5.486 |
Token | ailed |
Token ID | 6255 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Crimes | +0.821 |
certific | +0.781 |
witz | +0.769 |
Gutenberg | +0.762 |
Vienna | +0.719 |
Neg logit contributions | |
workers | -1.363 |
trainers | -1.330 |
tubing | -1.265 |
HQ | -1.256 |
enta | -1.222 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
Crimes | +2.722 |
certific | +2.613 |
witz | +2.560 |
Gutenberg | +2.526 |
Vienna | +2.474 |
Neg logit contributions | |
workers | -2.977 |
trainers | -2.824 |
HQ | -2.673 |
tubing | -2.639 |
enta | -2.636 |
Token | her |
Token ID | 607 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
Crimes | +0.721 |
certific | +0.687 |
witz | +0.678 |
Gutenberg | +0.672 |
Vienna | +0.646 |
Neg logit contributions | |
workers | -0.829 |
trainers | -0.805 |
tubing | -0.753 |
HQ | -0.750 |
enta | -0.739 |
Token | window |
Token ID | 4324 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
workers | +0.030 |
trainers | +0.029 |
tubing | +0.027 |
HQ | +0.027 |
enta | +0.026 |
Neg logit contributions | |
Crimes | -0.036 |
certific | -0.035 |
Gutenberg | -0.034 |
witz | -0.034 |
Vienna | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
Crimes | +2.603 |
witz | +2.505 |
certific | +2.496 |
Gutenberg | +2.455 |
Vienna | +2.369 |
Neg logit contributions | |
workers | -2.786 |
trainers | -2.706 |
HQ | -2.521 |
tubing | -2.505 |
enta | -2.473 |
Token | Just |
Token ID | 2329 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
Crimes | +0.634 |
certific | +0.603 |
witz | +0.595 |
Gutenberg | +0.594 |
unlaw | +0.568 |
Neg logit contributions | |
workers | -0.724 |
trainers | -0.708 |
tubing | -0.665 |
HQ | -0.655 |
enta | -0.647 |
Token | five |
Token ID | 1936 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
Crimes | +1.429 |
certific | +1.372 |
witz | +1.354 |
Gutenberg | +1.347 |
unlaw | +1.296 |
Neg logit contributions | |
workers | -1.321 |
trainers | -1.281 |
tubing | -1.196 |
HQ | -1.183 |
enta | -1.163 |
Token | shots |
Token ID | 6934 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
Crimes | +1.154 |
certific | +1.107 |
witz | +1.086 |
Gutenberg | +1.079 |
unlaw | +1.043 |
Neg logit contributions | |
workers | -1.213 |
trainers | -1.177 |
tubing | -1.105 |
HQ | -1.094 |
enta | -1.077 |
Token | on |
Token ID | 319 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
Crimes | +2.181 |
certific | +2.086 |
witz | +2.070 |
Gutenberg | +2.056 |
Vienna | +1.987 |
Neg logit contributions | |
workers | -2.036 |
trainers | -1.971 |
HQ | -1.841 |
tubing | -1.832 |
enta | -1.802 |
Token | goal |
Token ID | 3061 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
Crimes | +2.236 |
Gutenberg | +2.131 |
certific | +2.130 |
witz | +2.086 |
Vienna | +2.018 |
Neg logit contributions | |
workers | -2.725 |
trainers | -2.655 |
HQ | -2.497 |
tubing | -2.490 |
enta | -2.469 |
Token | were |
Token ID | 547 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
Crimes | +5.892 |
witz | +5.629 |
certific | +5.606 |
Gutenberg | +5.525 |
Vienna | +5.515 |
Neg logit contributions | |
workers | -6.091 |
trainers | -5.905 |
enta | -5.521 |
HQ | -5.510 |
agen | -5.423 |
Token | recorded |
Token ID | 6264 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
Crimes | +0.646 |
certific | +0.620 |
witz | +0.608 |
Gutenberg | +0.607 |
unlaw | +0.584 |
Neg logit contributions | |
workers | -0.653 |
trainers | -0.633 |
tubing | -0.593 |
HQ | -0.588 |
enta | -0.577 |
Token | in |
Token ID | 287 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
Crimes | +1.039 |
certific | +0.996 |
witz | +0.981 |
Gutenberg | +0.977 |
unlaw | +0.942 |
Neg logit contributions | |
workers | -1.012 |
trainers | -0.987 |
tubing | -0.918 |
HQ | -0.910 |
enta | -0.897 |
Token | the |
Token ID | 262 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
Crimes | +1.136 |
certific | +1.088 |
witz | +1.071 |
Gutenberg | +1.068 |
Vienna | +1.028 |
Neg logit contributions | |
workers | -1.172 |
trainers | -1.138 |
tubing | -1.062 |
HQ | -1.056 |
enta | -1.039 |
Token | extra |
Token ID | 3131 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Crimes | +0.902 |
certific | +0.863 |
witz | +0.857 |
Gutenberg | +0.853 |
Vienna | +0.827 |
Neg logit contributions | |
workers | -0.779 |
trainers | -0.750 |
tubing | -0.696 |
HQ | -0.694 |
enta | -0.680 |
Token | Islamist |
Token ID | 17780 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
workers | +0.005 |
trainers | +0.005 |
tubing | +0.005 |
HQ | +0.005 |
enta | +0.005 |
Neg logit contributions | |
Crimes | -0.006 |
certific | -0.006 |
Gutenberg | -0.006 |
witz | -0.006 |
Vienna | -0.005 |
Token | militants |
Token ID | 13162 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
Crimes | +0.977 |
certific | +0.941 |
Gutenberg | +0.910 |
witz | +0.906 |
unlaw | +0.889 |
Neg logit contributions | |
workers | -0.924 |
trainers | -0.881 |
tubing | -0.832 |
HQ | -0.824 |
enta | -0.819 |
Token | , |
Token ID | 11 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
workers | +0.098 |
trainers | +0.094 |
tubing | +0.088 |
HQ | +0.088 |
enta | +0.086 |
Neg logit contributions | |
Crimes | -0.102 |
certific | -0.098 |
witz | -0.096 |
Gutenberg | -0.096 |
unlaw | -0.092 |
Token | a |
Token ID | 257 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
workers | +1.201 |
trainers | +1.173 |
tubing | +1.095 |
HQ | +1.079 |
enta | +1.064 |
Neg logit contributions | |
Crimes | -1.162 |
certific | -1.117 |
witz | -1.099 |
Gutenberg | -1.093 |
unlaw | -1.055 |
Token | Human |
Token ID | 5524 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
workers | +0.736 |
trainers | +0.720 |
tubing | +0.673 |
HQ | +0.664 |
enta | +0.654 |
Neg logit contributions | |
Crimes | -0.687 |
certific | -0.660 |
witz | -0.649 |
Gutenberg | -0.646 |
unlaw | -0.623 |
Token | Rights |
Token ID | 6923 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
workers | +0.130 |
trainers | +0.128 |
tubing | +0.122 |
HQ | +0.120 |
enta | +0.119 |
Neg logit contributions | |
Crimes | -0.064 |
certific | -0.060 |
witz | -0.059 |
Gutenberg | -0.058 |
unlaw | -0.055 |
Token | Watch |
Token ID | 6305 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
Crimes | +3.450 |
certific | +2.696 |
Vienna | +2.695 |
Gutenberg | +2.682 |
witz | +2.583 |
Neg logit contributions | |
workers | -8.659 |
trainers | -8.521 |
tubing | -8.268 |
joining | -8.240 |
issan | -8.155 |
Token | ( |
Token ID | 357 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
Crimes | +0.399 |
certific | +0.373 |
witz | +0.364 |
Gutenberg | +0.360 |
Vienna | +0.349 |
Neg logit contributions | |
workers | -0.718 |
trainers | -0.698 |
tubing | -0.669 |
HQ | -0.658 |
enta | -0.656 |
Token | HR |
Token ID | 17184 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Crimes | +0.174 |
certific | +0.167 |
witz | +0.163 |
Gutenberg | +0.162 |
unlaw | +0.155 |
Neg logit contributions | |
workers | -0.263 |
trainers | -0.256 |
tubing | -0.242 |
HQ | -0.242 |
enta | -0.238 |
Token | W |
Token ID | 54 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
Crimes | +0.552 |
certific | +0.531 |
witz | +0.522 |
Gutenberg | +0.520 |
unlaw | +0.503 |
Neg logit contributions | |
workers | -0.514 |
trainers | -0.500 |
tubing | -0.466 |
HQ | -0.459 |
enta | -0.453 |
Token | ) |
Token ID | 8 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
Crimes | +0.510 |
certific | +0.478 |
witz | +0.472 |
Gutenberg | +0.468 |
Vienna | +0.456 |
Neg logit contributions | |
workers | -0.649 |
trainers | -0.635 |
tubing | -0.602 |
enta | -0.584 |
HQ | -0.583 |
Token | to |
Token ID | 284 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
Crimes | +1.058 |
certific | +1.016 |
Gutenberg | +1.008 |
witz | +1.008 |
Vienna | +0.979 |
Neg logit contributions | |
workers | -0.900 |
trainers | -0.866 |
HQ | -0.800 |
tubing | -0.797 |
enta | -0.788 |
Token | the |
Token ID | 262 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
Crimes | +1.115 |
certific | +1.064 |
Gutenberg | +1.054 |
witz | +1.048 |
Vienna | +1.032 |
Neg logit contributions | |
workers | -1.163 |
trainers | -1.118 |
HQ | -1.053 |
tubing | -1.046 |
enta | -1.034 |
Token | Museum |
Token ID | 9594 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
Crimes | +0.618 |
certific | +0.590 |
Gutenberg | +0.584 |
witz | +0.581 |
Vienna | +0.565 |
Neg logit contributions | |
workers | -0.611 |
trainers | -0.589 |
HQ | -0.549 |
tubing | -0.549 |
enta | -0.540 |
Token | of |
Token ID | 286 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
workers | +0.458 |
trainers | +0.448 |
tubing | +0.421 |
HQ | +0.416 |
enta | +0.411 |
Neg logit contributions | |
Crimes | -0.366 |
certific | -0.349 |
witz | -0.343 |
Gutenberg | -0.343 |
unlaw | -0.328 |
Token | Fine |
Token ID | 17867 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
Crimes | +0.512 |
certific | +0.487 |
Gutenberg | +0.480 |
witz | +0.480 |
Vienna | +0.461 |
Neg logit contributions | |
workers | -0.563 |
trainers | -0.548 |
tubing | -0.514 |
HQ | -0.509 |
enta | -0.501 |
Token | Arts |
Token ID | 11536 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
Crimes | +2.337 |
witz | +2.126 |
certific | +2.104 |
Vienna | +2.102 |
Gutenberg | +2.093 |
Neg logit contributions | |
workers | -2.856 |
trainers | -2.824 |
HQ | -2.654 |
tubing | -2.652 |
enta | -2.578 |
Token | , |
Token ID | 11 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
Crimes | +5.231 |
Vienna | +5.073 |
Gutenberg | +4.937 |
certific | +4.846 |
witz | +4.743 |
Neg logit contributions | |
workers | -6.327 |
trainers | -6.111 |
tubing | -5.757 |
HQ | -5.717 |
ox | -5.673 |
Token | the |
Token ID | 262 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Crimes | +1.431 |
certific | +1.377 |
Gutenberg | +1.365 |
witz | +1.341 |
Vienna | +1.326 |
Neg logit contributions | |
workers | -1.445 |
trainers | -1.383 |
HQ | -1.305 |
enta | -1.291 |
tubing | -1.285 |
Token | National |
Token ID | 2351 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
Crimes | +0.320 |
certific | +0.305 |
Gutenberg | +0.302 |
witz | +0.300 |
Vienna | +0.290 |
Neg logit contributions | |
workers | -0.338 |
trainers | -0.327 |
tubing | -0.306 |
HQ | -0.304 |
enta | -0.300 |
Token | Gallery |
Token ID | 12917 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
Crimes | +0.673 |
certific | +0.624 |
Gutenberg | +0.620 |
witz | +0.614 |
Vienna | +0.593 |
Neg logit contributions | |
workers | -0.984 |
trainers | -0.967 |
tubing | -0.916 |
HQ | -0.905 |
enta | -0.890 |
Token | and |
Token ID | 290 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
workers | +1.368 |
trainers | +1.335 |
tubing | +1.253 |
HQ | +1.237 |
enta | +1.221 |
Neg logit contributions | |
Crimes | -1.192 |
certific | -1.142 |
witz | -1.121 |
Gutenberg | -1.116 |
unlaw | -1.075 |
Token | N |
Token ID | 399 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
workers | +0.449 |
trainers | +0.439 |
tubing | +0.415 |
HQ | +0.409 |
enta | +0.405 |
Neg logit contributions | |
Crimes | -0.339 |
certific | -0.322 |
witz | -0.316 |
Gutenberg | -0.315 |
unlaw | -0.300 |
Token | ATIONS |
Token ID | 18421 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
workers | +0.005 |
trainers | +0.005 |
tubing | +0.005 |
HQ | +0.004 |
enta | +0.004 |
Neg logit contributions | |
Crimes | -0.008 |
certific | -0.008 |
witz | -0.008 |
Gutenberg | -0.008 |
unlaw | -0.007 |
Token | ( |
Token ID | 357 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Crimes | +0.828 |
certific | +0.779 |
witz | +0.765 |
Gutenberg | +0.763 |
Vienna | +0.735 |
Neg logit contributions | |
workers | -1.005 |
trainers | -0.980 |
tubing | -0.926 |
HQ | -0.913 |
enta | -0.902 |
Token | Reuters |
Token ID | 12637 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
Crimes | +0.535 |
witz | +0.500 |
certific | +0.499 |
Gutenberg | +0.495 |
Vienna | +0.472 |
Neg logit contributions | |
workers | -0.726 |
trainers | -0.724 |
tubing | -0.680 |
enta | -0.662 |
HQ | -0.657 |
Token | ) |
Token ID | 8 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
workers | +0.761 |
trainers | +0.733 |
HQ | +0.693 |
tubing | +0.683 |
ox | +0.677 |
Neg logit contributions | |
Crimes | -0.688 |
certific | -0.682 |
witz | -0.654 |
Gutenberg | -0.654 |
unlaw | -0.644 |
Token | - |
Token ID | 532 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
Crimes | +3.371 |
witz | +3.117 |
Gutenberg | +3.102 |
Vienna | +3.068 |
certific | +2.860 |
Neg logit contributions | |
trainers | -3.956 |
workers | -3.950 |
tubing | -3.737 |
enta | -3.616 |
airs | -3.577 |
Token | An |
Token ID | 1052 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
Crimes | +5.408 |
Vienna | +4.934 |
Gutenberg | +4.872 |
witz | +4.805 |
Fraud | +4.382 |
Neg logit contributions | |
trainers | -6.348 |
workers | -6.234 |
tubing | -5.998 |
enta | -5.827 |
HQ | -5.763 |
Token | army |
Token ID | 5428 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
Crimes | +0.356 |
certific | +0.340 |
witz | +0.334 |
Gutenberg | +0.333 |
unlaw | +0.321 |
Neg logit contributions | |
workers | -0.400 |
trainers | -0.390 |
tubing | -0.364 |
HQ | -0.360 |
enta | -0.356 |
Token | battalion |
Token ID | 36328 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
Crimes | +0.524 |
certific | +0.503 |
witz | +0.495 |
Gutenberg | +0.491 |
unlaw | +0.476 |
Neg logit contributions | |
workers | -0.537 |
trainers | -0.520 |
tubing | -0.488 |
HQ | -0.481 |
enta | -0.476 |
Token | in |
Token ID | 287 |
Feature activation | +7.20e-02 |
Pos logit contributions | |
Crimes | +0.991 |
certific | +0.954 |
witz | +0.939 |
Gutenberg | +0.927 |
unlaw | +0.899 |
Neg logit contributions | |
workers | -0.991 |
trainers | -0.956 |
tubing | -0.899 |
HQ | -0.892 |
enta | -0.881 |
Token | the |
Token ID | 262 |
Feature activation | -8.80e-02 |
Pos logit contributions | |
Crimes | +0.189 |
certific | +0.180 |
witz | +0.177 |
Gutenberg | +0.176 |
Vienna | +0.170 |
Neg logit contributions | |
workers | -0.216 |
trainers | -0.209 |
tubing | -0.197 |
HQ | -0.196 |
enta | -0.193 |
Token | V |
Token ID | 53 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
workers | +0.648 |
trainers | +0.632 |
tubing | +0.589 |
HQ | +0.584 |
enta | +0.572 |
Neg logit contributions | |
Crimes | -0.658 |
certific | -0.633 |
witz | -0.621 |
Gutenberg | -0.620 |
unlaw | -0.599 |
Token | ANA |
Token ID | 31574 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
Crimes | +3.445 |
witz | +3.301 |
certific | +3.282 |
Gutenberg | +3.221 |
Vienna | +3.189 |
Neg logit contributions | |
workers | -2.317 |
trainers | -2.249 |
tubing | -2.030 |
enta | -1.974 |
HQ | -1.970 |
Token | ( |
Token ID | 357 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
Crimes | +1.083 |
certific | +1.032 |
witz | +1.005 |
Gutenberg | +1.002 |
Vienna | +0.973 |
Neg logit contributions | |
workers | -1.227 |
trainers | -1.194 |
tubing | -1.123 |
HQ | -1.099 |
enta | -1.096 |
Token | Reuters |
Token ID | 12637 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
Crimes | +0.535 |
certific | +0.503 |
witz | +0.502 |
Gutenberg | +0.491 |
Vienna | +0.472 |
Neg logit contributions | |
workers | -0.729 |
trainers | -0.725 |
tubing | -0.682 |
enta | -0.663 |
HQ | -0.660 |
Token | ) |
Token ID | 8 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Crimes | +0.377 |
certific | +0.363 |
witz | +0.355 |
Gutenberg | +0.354 |
unlaw | +0.341 |
Neg logit contributions | |
workers | -0.445 |
trainers | -0.434 |
tubing | -0.407 |
HQ | -0.404 |
enta | -0.398 |
Token | - |
Token ID | 532 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
Crimes | +3.248 |
witz | +3.013 |
Gutenberg | +3.001 |
Vienna | +2.962 |
certific | +2.768 |
Neg logit contributions | |
trainers | -3.792 |
workers | -3.788 |
tubing | -3.580 |
enta | -3.463 |
HQ | -3.424 |
Token | Latin |
Token ID | 9133 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
Crimes | +5.795 |
Vienna | +5.415 |
Gutenberg | +5.210 |
witz | +5.108 |
certific | +4.841 |
Neg logit contributions | |
trainers | -5.819 |
workers | -5.719 |
tubing | -5.423 |
HQ | -5.348 |
enta | -5.287 |
Token | American |
Token ID | 1605 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
Crimes | +0.643 |
certific | +0.615 |
witz | +0.605 |
Gutenberg | +0.605 |
Vienna | +0.581 |
Neg logit contributions | |
workers | -0.612 |
trainers | -0.596 |
tubing | -0.559 |
HQ | -0.553 |
enta | -0.547 |
Token | and |
Token ID | 290 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
Crimes | +1.386 |
certific | +1.316 |
Gutenberg | +1.297 |
witz | +1.283 |
unlaw | +1.249 |
Neg logit contributions | |
workers | -1.307 |
trainers | -1.258 |
HQ | -1.195 |
tubing | -1.188 |
enta | -1.165 |
Token | Caribbean |
Token ID | 18020 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
workers | +1.369 |
trainers | +1.341 |
tubing | +1.264 |
HQ | +1.245 |
enta | +1.233 |
Neg logit contributions | |
Crimes | -0.987 |
certific | -0.942 |
witz | -0.927 |
Gutenberg | -0.916 |
unlaw | -0.881 |
Token | leaders |
Token ID | 2766 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
Crimes | +3.746 |
certific | +3.570 |
Gutenberg | +3.468 |
witz | +3.390 |
Vienna | +3.383 |
Neg logit contributions | |
workers | -4.015 |
trainers | -3.806 |
HQ | -3.734 |
enta | -3.671 |
tubing | -3.669 |
Token | USS |
Token ID | 32835 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
workers | +0.296 |
trainers | +0.289 |
tubing | +0.271 |
HQ | +0.266 |
enta | +0.263 |
Neg logit contributions | |
Crimes | -0.283 |
certific | -0.271 |
witz | -0.266 |
Gutenberg | -0.265 |
unlaw | -0.256 |
Token | ELS |
Token ID | 37142 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
Crimes | +0.808 |
certific | +0.755 |
witz | +0.747 |
Gutenberg | +0.744 |
Vienna | +0.725 |
Neg logit contributions | |
workers | -1.117 |
trainers | -1.073 |
tubing | -1.023 |
enta | -1.011 |
agen | -0.991 |
Token | ( |
Token ID | 357 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
Crimes | +1.362 |
certific | +1.272 |
witz | +1.261 |
Gutenberg | +1.252 |
Vienna | +1.215 |
Neg logit contributions | |
workers | -1.723 |
trainers | -1.684 |
tubing | -1.584 |
HQ | -1.561 |
enta | -1.540 |
Token | Reuters |
Token ID | 12637 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
Crimes | +0.570 |
certific | +0.534 |
witz | +0.533 |
Gutenberg | +0.525 |
Vienna | +0.503 |
Neg logit contributions | |
workers | -0.792 |
trainers | -0.787 |
tubing | -0.740 |
enta | -0.721 |
HQ | -0.717 |
Token | ) |
Token ID | 8 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Crimes | +0.120 |
certific | +0.116 |
witz | +0.113 |
Gutenberg | +0.113 |
unlaw | +0.109 |
Neg logit contributions | |
workers | -0.140 |
trainers | -0.136 |
tubing | -0.127 |
HQ | -0.127 |
enta | -0.125 |
Token | - |
Token ID | 532 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
Crimes | +3.331 |
witz | +3.093 |
Gutenberg | +3.081 |
Vienna | +3.040 |
certific | +2.822 |
Neg logit contributions | |
trainers | -3.913 |
workers | -3.907 |
tubing | -3.693 |
enta | -3.571 |
HQ | -3.535 |
Token | The |
Token ID | 383 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
Crimes | +5.236 |
Vienna | +4.777 |
Gutenberg | +4.708 |
witz | +4.623 |
Fraud | +4.261 |
Neg logit contributions | |
trainers | -6.376 |
workers | -6.261 |
tubing | -5.971 |
HQ | -5.926 |
enta | -5.835 |
Token | world |
Token ID | 995 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Crimes | +1.021 |
certific | +0.960 |
witz | +0.948 |
Gutenberg | +0.947 |
Vienna | +0.917 |
Neg logit contributions | |
workers | -1.280 |
trainers | -1.239 |
HQ | -1.176 |
tubing | -1.168 |
enta | -1.153 |
Token | � |
Token ID | 447 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Crimes | +0.344 |
certific | +0.330 |
Gutenberg | +0.327 |
witz | +0.324 |
Vienna | +0.314 |
Neg logit contributions | |
workers | -0.308 |
trainers | -0.297 |
tubing | -0.277 |
HQ | -0.276 |
enta | -0.269 |
Token | � |
Token ID | 247 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
Crimes | +0.936 |
certific | +0.895 |
witz | +0.880 |
Gutenberg | +0.876 |
unlaw | +0.841 |
Neg logit contributions | |
workers | -1.037 |
trainers | -1.012 |
tubing | -0.947 |
HQ | -0.934 |
enta | -0.923 |
Token | s |
Token ID | 82 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Crimes | +0.117 |
certific | +0.113 |
witz | +0.111 |
Gutenberg | +0.110 |
unlaw | +0.106 |
Neg logit contributions | |
workers | -0.106 |
trainers | -0.104 |
tubing | -0.096 |
HQ | -0.095 |
enta | -0.093 |
Token | the |
Token ID | 262 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
Crimes | +0.516 |
certific | +0.490 |
Gutenberg | +0.480 |
witz | +0.479 |
Vienna | +0.464 |
Neg logit contributions | |
workers | -0.512 |
trainers | -0.497 |
tubing | -0.469 |
HQ | -0.467 |
enta | -0.457 |
Token | family |
Token ID | 1641 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
Crimes | +0.778 |
certific | +0.742 |
Gutenberg | +0.728 |
witz | +0.725 |
unlaw | +0.704 |
Neg logit contributions | |
workers | -0.709 |
trainers | -0.687 |
tubing | -0.643 |
HQ | -0.639 |
enta | -0.628 |
Token | of |
Token ID | 286 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
Crimes | +0.793 |
certific | +0.749 |
Gutenberg | +0.731 |
witz | +0.729 |
Vienna | +0.707 |
Neg logit contributions | |
workers | -1.048 |
trainers | -1.021 |
tubing | -0.970 |
HQ | -0.965 |
enta | -0.946 |
Token | a |
Token ID | 257 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
workers | +0.709 |
trainers | +0.691 |
tubing | +0.647 |
HQ | +0.641 |
enta | +0.631 |
Neg logit contributions | |
Crimes | -0.678 |
certific | -0.649 |
witz | -0.637 |
Gutenberg | -0.635 |
unlaw | -0.613 |
Token | real |
Token ID | 1103 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
workers | +1.494 |
trainers | +1.454 |
tubing | +1.355 |
HQ | +1.330 |
enta | +1.312 |
Neg logit contributions | |
Crimes | -1.626 |
certific | -1.562 |
witz | -1.542 |
Gutenberg | -1.536 |
unlaw | -1.485 |
Token | estate |
Token ID | 7964 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
Crimes | +0.610 |
certific | +0.578 |
witz | +0.567 |
Gutenberg | +0.562 |
unlaw | +0.541 |
Neg logit contributions | |
workers | -0.710 |
trainers | -0.699 |
HQ | -0.656 |
tubing | -0.653 |
airs | -0.635 |
Token | agent |
Token ID | 5797 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
Crimes | +5.360 |
certific | +5.069 |
witz | +4.952 |
Gutenberg | +4.859 |
Vienna | +4.850 |
Neg logit contributions | |
workers | -5.933 |
trainers | -5.904 |
tubing | -5.440 |
airs | -5.427 |
HQ | -5.412 |
Token | murdered |
Token ID | 12864 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Crimes | +1.375 |
certific | +1.326 |
witz | +1.290 |
Gutenberg | +1.286 |
Vienna | +1.255 |
Neg logit contributions | |
workers | -1.308 |
trainers | -1.275 |
HQ | -1.187 |
tubing | -1.185 |
enta | -1.169 |
Token | by |
Token ID | 416 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
Crimes | +0.491 |
certific | +0.470 |
witz | +0.461 |
Gutenberg | +0.458 |
Vienna | +0.444 |
Neg logit contributions | |
workers | -0.487 |
trainers | -0.475 |
tubing | -0.442 |
HQ | -0.442 |
enta | -0.437 |
Token | members |
Token ID | 1866 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
Crimes | +1.507 |
certific | +1.444 |
Gutenberg | +1.408 |
witz | +1.395 |
unlaw | +1.369 |
Neg logit contributions | |
workers | -1.501 |
trainers | -1.458 |
HQ | -1.384 |
tubing | -1.370 |
enta | -1.358 |
Token | of |
Token ID | 286 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
workers | +1.403 |
trainers | +1.372 |
tubing | +1.291 |
HQ | +1.277 |
enta | +1.260 |
Neg logit contributions | |
Crimes | -1.114 |
certific | -1.064 |
witz | -1.044 |
Gutenberg | -1.040 |
unlaw | -0.999 |
Token | S |
Token ID | 50 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
workers | +0.474 |
trainers | +0.455 |
tubing | +0.414 |
HQ | +0.414 |
enta | +0.402 |
Neg logit contributions | |
Crimes | -0.747 |
certific | -0.730 |
witz | -0.716 |
Gutenberg | -0.713 |
unlaw | -0.700 |
Token | . |
Token ID | 13 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
workers | +1.347 |
trainers | +1.312 |
tubing | +1.219 |
HQ | +1.214 |
enta | +1.202 |
Neg logit contributions | |
Crimes | -1.240 |
certific | -1.209 |
witz | -1.178 |
Gutenberg | -1.168 |
unlaw | -1.152 |
Token | President |
Token ID | 1992 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
Crimes | +0.188 |
certific | +0.180 |
witz | +0.177 |
Gutenberg | +0.177 |
Vienna | +0.171 |
Neg logit contributions | |
workers | -0.150 |
trainers | -0.146 |
tubing | -0.136 |
HQ | -0.133 |
enta | -0.131 |
Token | Donald |
Token ID | 3759 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
workers | +0.979 |
trainers | +0.956 |
tubing | +0.897 |
HQ | +0.885 |
enta | +0.873 |
Neg logit contributions | |
Crimes | -0.872 |
certific | -0.840 |
witz | -0.823 |
Gutenberg | -0.815 |
unlaw | -0.789 |
Token | Trump |
Token ID | 1301 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
Crimes | +2.913 |
Gutenberg | +2.736 |
witz | +2.723 |
certific | +2.715 |
Vienna | +2.645 |
Neg logit contributions | |
workers | -2.359 |
trainers | -2.295 |
tubing | -2.140 |
enta | -2.076 |
HQ | -2.072 |
Token | waves |
Token ID | 9813 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
workers | +0.295 |
trainers | +0.288 |
tubing | +0.270 |
HQ | +0.267 |
enta | +0.263 |
Neg logit contributions | |
Crimes | -0.249 |
certific | -0.238 |
witz | -0.234 |
Gutenberg | -0.232 |
unlaw | -0.224 |
Token | as |
Token ID | 355 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
Crimes | +5.362 |
Gutenberg | +5.123 |
witz | +5.122 |
certific | +5.095 |
Vienna | +4.963 |
Neg logit contributions | |
workers | -5.738 |
trainers | -5.676 |
enta | -5.389 |
HQ | -5.310 |
tubing | -5.300 |
Token | he |
Token ID | 339 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
Crimes | +0.252 |
certific | +0.244 |
witz | +0.239 |
Gutenberg | +0.239 |
Vienna | +0.231 |
Neg logit contributions | |
workers | -0.205 |
trainers | -0.200 |
tubing | -0.186 |
HQ | -0.185 |
enta | -0.180 |
Token | steps |
Token ID | 4831 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
Crimes | +0.711 |
certific | +0.679 |
witz | +0.666 |
Gutenberg | +0.661 |
unlaw | +0.634 |
Neg logit contributions | |
workers | -0.897 |
trainers | -0.875 |
tubing | -0.824 |
HQ | -0.818 |
enta | -0.806 |
Token | out |
Token ID | 503 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Crimes | +0.324 |
certific | +0.310 |
witz | +0.307 |
Gutenberg | +0.303 |
unlaw | +0.292 |
Neg logit contributions | |
workers | -0.365 |
trainers | -0.355 |
tubing | -0.332 |
HQ | -0.330 |
enta | -0.326 |
Token | from |
Token ID | 422 |
Feature activation | +4.99e-03 |
Pos logit contributions | |
Crimes | +0.473 |
certific | +0.450 |
witz | +0.445 |
Gutenberg | +0.442 |
Vienna | +0.425 |
Neg logit contributions | |
workers | -0.553 |
trainers | -0.540 |
tubing | -0.507 |
HQ | -0.503 |
enta | -0.497 |
Token | period |
Token ID | 2278 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
workers | +0.224 |
trainers | +0.218 |
tubing | +0.203 |
HQ | +0.201 |
enta | +0.198 |
Neg logit contributions | |
Crimes | -0.227 |
certific | -0.218 |
witz | -0.215 |
Gutenberg | -0.214 |
unlaw | -0.207 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
Crimes | +0.756 |
certific | +0.728 |
witz | +0.721 |
Gutenberg | +0.717 |
Vienna | +0.692 |
Neg logit contributions | |
workers | -0.727 |
trainers | -0.700 |
HQ | -0.654 |
tubing | -0.654 |
enta | -0.648 |
Token | 25 |
Token ID | 1679 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
workers | +0.563 |
trainers | +0.549 |
tubing | +0.511 |
HQ | +0.504 |
enta | +0.497 |
Neg logit contributions | |
Crimes | -0.589 |
certific | -0.567 |
witz | -0.557 |
Gutenberg | -0.555 |
unlaw | -0.537 |
Token | shots |
Token ID | 6934 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
Crimes | +0.278 |
certific | +0.267 |
witz | +0.263 |
Gutenberg | +0.262 |
unlaw | +0.253 |
Neg logit contributions | |
workers | -0.278 |
trainers | -0.271 |
tubing | -0.253 |
HQ | -0.250 |
enta | -0.246 |
Token | on |
Token ID | 319 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
Crimes | +1.901 |
certific | +1.827 |
witz | +1.802 |
Gutenberg | +1.795 |
Vienna | +1.744 |
Neg logit contributions | |
workers | -1.769 |
trainers | -1.719 |
HQ | -1.602 |
tubing | -1.599 |
enta | -1.564 |
Token | goal |
Token ID | 3061 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
Crimes | +1.983 |
certific | +1.896 |
Gutenberg | +1.882 |
witz | +1.863 |
unlaw | +1.796 |
Neg logit contributions | |
workers | -2.212 |
trainers | -2.140 |
tubing | -2.004 |
HQ | -2.002 |
enta | -1.975 |
Token | fired |
Token ID | 6294 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
Crimes | +5.360 |
certific | +5.157 |
witz | +5.076 |
Gutenberg | +5.027 |
Vienna | +5.001 |
Neg logit contributions | |
workers | -5.800 |
trainers | -5.655 |
HQ | -5.270 |
agen | -5.265 |
tubing | -5.233 |
Token | off |
Token ID | 572 |
Feature activation | +6.90e-01 |
Pos logit contributions | |
Crimes | +1.494 |
certific | +1.429 |
witz | +1.415 |
Gutenberg | +1.400 |
Vienna | +1.353 |
Neg logit contributions | |
workers | -1.470 |
trainers | -1.429 |
tubing | -1.331 |
HQ | -1.324 |
enta | -1.307 |
Token | between |
Token ID | 1022 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
Crimes | +1.851 |
certific | +1.764 |
witz | +1.742 |
Gutenberg | +1.733 |
Vienna | +1.669 |
Neg logit contributions | |
workers | -2.004 |
trainers | -1.949 |
tubing | -1.837 |
HQ | -1.824 |
enta | -1.788 |
Token | the |
Token ID | 262 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
Crimes | +1.049 |
certific | +1.003 |
witz | +0.989 |
Gutenberg | +0.983 |
unlaw | +0.947 |
Neg logit contributions | |
workers | -1.108 |
trainers | -1.078 |
tubing | -1.011 |
HQ | -0.999 |
enta | -0.983 |
Token | two |
Token ID | 734 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
workers | +0.968 |
trainers | +0.950 |
tubing | +0.883 |
HQ | +0.864 |
enta | +0.858 |
Neg logit contributions | |
Crimes | -0.995 |
certific | -0.959 |
witz | -0.943 |
Gutenberg | -0.939 |
unlaw | -0.908 |
Token | The |
Token ID | 383 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
Crimes | +0.413 |
certific | +0.392 |
witz | +0.387 |
Gutenberg | +0.386 |
unlaw | +0.370 |
Neg logit contributions | |
workers | -0.456 |
trainers | -0.446 |
tubing | -0.418 |
HQ | -0.411 |
enta | -0.407 |
Token | number |
Token ID | 1271 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
Crimes | +0.702 |
certific | +0.671 |
witz | +0.661 |
Gutenberg | +0.657 |
unlaw | +0.634 |
Neg logit contributions | |
workers | -0.782 |
trainers | -0.761 |
tubing | -0.714 |
HQ | -0.707 |
enta | -0.698 |
Token | of |
Token ID | 286 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
Crimes | +0.153 |
certific | +0.146 |
witz | +0.144 |
Gutenberg | +0.143 |
unlaw | +0.138 |
Neg logit contributions | |
workers | -0.183 |
trainers | -0.178 |
tubing | -0.168 |
HQ | -0.166 |
enta | -0.164 |
Token | attacks |
Token ID | 3434 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
workers | +1.209 |
trainers | +1.186 |
tubing | +1.103 |
HQ | +1.088 |
enta | +1.067 |
Neg logit contributions | |
Crimes | -1.083 |
certific | -1.034 |
witz | -1.021 |
Gutenberg | -1.017 |
unlaw | -0.976 |
Token | on |
Token ID | 319 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
workers | +1.309 |
trainers | +1.277 |
tubing | +1.181 |
HQ | +1.151 |
ox | +1.134 |
Neg logit contributions | |
Crimes | -1.528 |
certific | -1.483 |
witz | -1.460 |
Gutenberg | -1.453 |
unlaw | -1.417 |
Token | refugee |
Token ID | 12540 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
Crimes | +0.587 |
certific | +0.563 |
witz | +0.553 |
Gutenberg | +0.550 |
unlaw | +0.530 |
Neg logit contributions | |
workers | -0.657 |
trainers | -0.641 |
tubing | -0.601 |
HQ | -0.594 |
enta | -0.586 |
Token | shelters |
Token ID | 23629 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Crimes | +4.746 |
certific | +4.589 |
Vienna | +4.399 |
witz | +4.371 |
Gutenberg | +4.317 |
Neg logit contributions | |
workers | -6.351 |
trainers | -6.220 |
enta | -5.969 |
ox | -5.920 |
joining | -5.883 |
Token | has |
Token ID | 468 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
workers | +0.096 |
trainers | +0.094 |
tubing | +0.087 |
HQ | +0.086 |
enta | +0.085 |
Neg logit contributions | |
Crimes | -0.100 |
certific | -0.096 |
witz | -0.095 |
Gutenberg | -0.094 |
unlaw | -0.092 |
Token | more |
Token ID | 517 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
workers | +0.406 |
trainers | +0.395 |
tubing | +0.368 |
HQ | +0.364 |
enta | +0.358 |
Neg logit contributions | |
Crimes | -0.429 |
certific | -0.413 |
witz | -0.406 |
Gutenberg | -0.404 |
unlaw | -0.390 |
Token | than |
Token ID | 621 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
Crimes | +0.240 |
certific | +0.228 |
witz | +0.222 |
Gutenberg | +0.221 |
unlaw | +0.210 |
Neg logit contributions | |
workers | -0.385 |
trainers | -0.376 |
tubing | -0.357 |
HQ | -0.354 |
enta | -0.351 |
Token | tripled |
Token ID | 43761 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
workers | +0.957 |
trainers | +0.923 |
tubing | +0.834 |
HQ | +0.814 |
enta | +0.800 |
Neg logit contributions | |
Crimes | -1.801 |
certific | -1.745 |
witz | -1.723 |
Gutenberg | -1.719 |
unlaw | -1.677 |
Token | 25 |
Token ID | 1679 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
workers | +1.833 |
trainers | +1.771 |
tubing | +1.652 |
HQ | +1.617 |
enta | +1.605 |
Neg logit contributions | |
Crimes | -1.945 |
certific | -1.838 |
witz | -1.827 |
Gutenberg | -1.815 |
unlaw | -1.769 |
Token | to |
Token ID | 284 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
workers | +2.284 |
trainers | +2.227 |
tubing | +2.092 |
HQ | +2.063 |
enta | +2.042 |
Neg logit contributions | |
Crimes | -1.937 |
certific | -1.848 |
witz | -1.819 |
Gutenberg | -1.812 |
unlaw | -1.750 |
Token | 49 |
Token ID | 5125 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
workers | +0.674 |
trainers | +0.658 |
tubing | +0.615 |
HQ | +0.608 |
enta | +0.602 |
Neg logit contributions | |
Crimes | -0.612 |
certific | -0.585 |
witz | -0.577 |
Gutenberg | -0.572 |
unlaw | -0.556 |
Token | with |
Token ID | 351 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
Crimes | +0.552 |
certific | +0.537 |
witz | +0.524 |
Gutenberg | +0.522 |
Vienna | +0.500 |
Neg logit contributions | |
workers | -0.540 |
trainers | -0.528 |
tubing | -0.493 |
HQ | -0.487 |
ox | -0.476 |
Token | a |
Token ID | 257 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
workers | +2.183 |
trainers | +2.131 |
tubing | +1.970 |
HQ | +1.942 |
enta | +1.913 |
Neg logit contributions | |
Crimes | -2.330 |
certific | -2.224 |
witz | -2.205 |
Gutenberg | -2.186 |
unlaw | -2.129 |
Token | bachelor |
Token ID | 29384 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
workers | +0.609 |
trainers | +0.592 |
tubing | +0.550 |
HQ | +0.542 |
enta | +0.534 |
Neg logit contributions | |
Crimes | -0.704 |
certific | -0.678 |
witz | -0.668 |
Gutenberg | -0.665 |
unlaw | -0.645 |
Token | � |
Token ID | 447 |
Feature activation | -9.71e-03 |
Pos logit contributions | |
Crimes | +5.114 |
certific | +5.021 |
witz | +4.724 |
Gutenberg | +4.720 |
Vienna | +4.689 |
Neg logit contributions | |
workers | -5.897 |
trainers | -5.693 |
tubing | -5.422 |
HQ | -5.279 |
joining | -5.202 |
Token | � |
Token ID | 247 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
workers | +0.027 |
trainers | +0.026 |
tubing | +0.024 |
HQ | +0.024 |
enta | +0.024 |
Neg logit contributions | |
Crimes | -0.028 |
certific | -0.027 |
witz | -0.026 |
Gutenberg | -0.026 |
unlaw | -0.025 |
Token | s |
Token ID | 82 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Crimes | +0.312 |
certific | +0.302 |
witz | +0.296 |
Gutenberg | +0.293 |
unlaw | +0.282 |
Neg logit contributions | |
workers | -0.315 |
trainers | -0.308 |
tubing | -0.288 |
HQ | -0.284 |
enta | -0.279 |
Token | degree |
Token ID | 4922 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
workers | +0.408 |
trainers | +0.399 |
tubing | +0.379 |
HQ | +0.375 |
enta | +0.371 |
Neg logit contributions | |
Crimes | -0.224 |
certific | -0.212 |
witz | -0.207 |
Gutenberg | -0.206 |
unlaw | -0.195 |
Token | or |
Token ID | 393 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
Crimes | +3.194 |
certific | +3.145 |
witz | +2.994 |
Gutenberg | +2.981 |
Vienna | +2.847 |
Neg logit contributions | |
workers | -3.781 |
trainers | -3.686 |
tubing | -3.461 |
HQ | -3.435 |
agen | -3.388 |
Token | у |
Token ID | 35072 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
workers | +1.001 |
trainers | +0.975 |
tubing | +0.915 |
HQ | +0.906 |
enta | +0.892 |
Neg logit contributions | |
Crimes | -0.874 |
certific | -0.837 |
witz | -0.822 |
Gutenberg | -0.818 |
unlaw | -0.788 |
Token | � |
Token ID | 140 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
workers | +0.435 |
trainers | +0.426 |
tubing | +0.401 |
HQ | +0.396 |
enta | +0.391 |
Neg logit contributions | |
Crimes | -0.340 |
certific | -0.325 |
witz | -0.319 |
Gutenberg | -0.317 |
unlaw | -0.305 |
Token | � |
Token ID | 115 |
Feature activation | +9.49e-01 |
Pos logit contributions | |
Crimes | +2.566 |
witz | +2.415 |
certific | +2.410 |
Gutenberg | +2.385 |
Vienna | +2.314 |
Neg logit contributions | |
workers | -3.390 |
trainers | -3.313 |
tubing | -3.103 |
HQ | -3.077 |
enta | -3.019 |
Token | ь |
Token ID | 45367 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
Crimes | +2.386 |
certific | +2.246 |
witz | +2.229 |
Gutenberg | +2.206 |
Vienna | +2.121 |
Neg logit contributions | |
workers | -2.911 |
trainers | -2.863 |
tubing | -2.673 |
HQ | -2.628 |
enta | -2.619 |
Token | я |
Token ID | 40623 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
Crimes | +1.080 |
certific | +1.019 |
witz | +1.007 |
Gutenberg | +0.998 |
Vienna | +0.954 |
Neg logit contributions | |
workers | -1.439 |
trainers | -1.409 |
tubing | -1.329 |
HQ | -1.305 |
enta | -1.297 |
Token | м |
Token ID | 43108 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
Crimes | +1.306 |
certific | +1.239 |
witz | +1.225 |
Gutenberg | +1.215 |
unlaw | +1.163 |
Neg logit contributions | |
workers | -1.611 |
trainers | -1.578 |
tubing | -1.479 |
HQ | -1.452 |
enta | -1.445 |
Token | и |
Token ID | 18849 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
Crimes | +4.585 |
witz | +4.252 |
certific | +4.250 |
Gutenberg | +4.156 |
Vienna | +4.071 |
Neg logit contributions | |
workers | -6.604 |
trainers | -6.493 |
tubing | -6.080 |
joining | -5.971 |
enta | -5.959 |
Token | � |
Token ID | 12466 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
Crimes | +1.974 |
certific | +1.857 |
witz | +1.822 |
Gutenberg | +1.794 |
Vienna | +1.726 |
Neg logit contributions | |
workers | -2.964 |
trainers | -2.896 |
tubing | -2.723 |
HQ | -2.689 |
enta | -2.658 |
Token | � |
Token ID | 123 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
Crimes | +1.044 |
certific | +0.990 |
witz | +0.976 |
Gutenberg | +0.970 |
unlaw | +0.930 |
Neg logit contributions | |
workers | -1.252 |
trainers | -1.231 |
tubing | -1.151 |
HQ | -1.134 |
enta | -1.116 |
Token | о� |
Token ID | 25443 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
workers | +0.076 |
trainers | +0.074 |
tubing | +0.069 |
HQ | +0.068 |
enta | +0.067 |
Neg logit contributions | |
Crimes | -0.077 |
certific | -0.074 |
witz | -0.073 |
Gutenberg | -0.072 |
unlaw | -0.070 |
Token | � |
Token ID | 111 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
workers | +0.069 |
trainers | +0.067 |
tubing | +0.063 |
HQ | +0.062 |
enta | +0.062 |
Neg logit contributions | |
Crimes | -0.054 |
certific | -0.051 |
witz | -0.051 |
Gutenberg | -0.050 |
unlaw | -0.048 |
Token | owns |
Token ID | 12216 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
Crimes | +1.355 |
certific | +1.316 |
Gutenberg | +1.272 |
witz | +1.271 |
Vienna | +1.221 |
Neg logit contributions | |
workers | -1.450 |
trainers | -1.396 |
tubing | -1.317 |
HQ | -1.304 |
enta | -1.287 |
Token | four |
Token ID | 1440 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
Crimes | +1.250 |
certific | +1.206 |
Gutenberg | +1.176 |
witz | +1.166 |
Vienna | +1.131 |
Neg logit contributions | |
workers | -1.307 |
trainers | -1.260 |
HQ | -1.186 |
tubing | -1.185 |
enta | -1.167 |
Token | numbered |
Token ID | 25840 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
Crimes | +1.437 |
certific | +1.392 |
Gutenberg | +1.354 |
witz | +1.342 |
Vienna | +1.304 |
Neg logit contributions | |
workers | -1.537 |
trainers | -1.480 |
tubing | -1.389 |
HQ | -1.387 |
enta | -1.364 |
Token | Ontario |
Token ID | 10553 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
Crimes | +0.122 |
certific | +0.117 |
Gutenberg | +0.115 |
witz | +0.114 |
unlaw | +0.110 |
Neg logit contributions | |
workers | -0.122 |
trainers | -0.118 |
tubing | -0.111 |
HQ | -0.110 |
enta | -0.108 |
Token | real |
Token ID | 1103 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
Crimes | +1.067 |
certific | +1.024 |
Gutenberg | +0.997 |
witz | +0.996 |
unlaw | +0.957 |
Neg logit contributions | |
workers | -1.213 |
trainers | -1.178 |
tubing | -1.105 |
HQ | -1.098 |
enta | -1.083 |
Token | estate |
Token ID | 7964 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Crimes | +1.314 |
certific | +1.238 |
witz | +1.205 |
Gutenberg | +1.199 |
unlaw | +1.141 |
Neg logit contributions | |
workers | -2.109 |
trainers | -2.065 |
tubing | -1.962 |
HQ | -1.957 |
enta | -1.915 |
Token | holding |
Token ID | 4769 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
Crimes | +4.947 |
certific | +4.815 |
Gutenberg | +4.543 |
Vienna | +4.487 |
unlaw | +4.467 |
Neg logit contributions | |
workers | -6.220 |
trainers | -5.859 |
agen | -5.657 |
ox | -5.630 |
enta | -5.623 |
Token | companies |
Token ID | 2706 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
workers | +0.288 |
trainers | +0.282 |
tubing | +0.267 |
HQ | +0.265 |
enta | +0.262 |
Neg logit contributions | |
Crimes | -0.180 |
certific | -0.171 |
witz | -0.166 |
Gutenberg | -0.166 |
unlaw | -0.158 |
Token | with |
Token ID | 351 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
Crimes | +0.531 |
certific | +0.513 |
Gutenberg | +0.500 |
witz | +0.498 |
unlaw | +0.479 |
Neg logit contributions | |
workers | -0.565 |
trainers | -0.548 |
tubing | -0.514 |
HQ | -0.508 |
enta | -0.501 |
Token | his |
Token ID | 465 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
workers | +0.613 |
trainers | +0.598 |
tubing | +0.559 |
HQ | +0.552 |
enta | +0.545 |
Neg logit contributions | |
Crimes | -0.583 |
certific | -0.560 |
witz | -0.550 |
Gutenberg | -0.547 |
unlaw | -0.528 |
Token | siblings |
Token ID | 20569 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
workers | +0.202 |
trainers | +0.196 |
tubing | +0.182 |
HQ | +0.179 |
enta | +0.177 |
Neg logit contributions | |
Crimes | -0.237 |
certific | -0.229 |
Gutenberg | -0.224 |
witz | -0.224 |
unlaw | -0.216 |
Token | firmly |
Token ID | 14245 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
workers | +0.322 |
trainers | +0.311 |
tubing | +0.290 |
HQ | +0.287 |
enta | +0.283 |
Neg logit contributions | |
Crimes | -0.352 |
certific | -0.339 |
witz | -0.333 |
Gutenberg | -0.331 |
unlaw | -0.321 |
Token | on |
Token ID | 319 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
Crimes | +1.098 |
certific | +1.069 |
witz | +1.041 |
Gutenberg | +1.025 |
unlaw | +1.005 |
Neg logit contributions | |
workers | -1.097 |
trainers | -1.048 |
tubing | -0.989 |
HQ | -0.982 |
enta | -0.967 |
Token | course |
Token ID | 1781 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
Crimes | +1.757 |
certific | +1.687 |
Gutenberg | +1.658 |
witz | +1.657 |
unlaw | +1.607 |
Neg logit contributions | |
workers | -1.570 |
trainers | -1.499 |
tubing | -1.412 |
HQ | -1.396 |
enta | -1.376 |
Token | for |
Token ID | 329 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
workers | +0.122 |
trainers | +0.118 |
tubing | +0.109 |
HQ | +0.108 |
enta | +0.106 |
Neg logit contributions | |
Crimes | -0.155 |
certific | -0.149 |
witz | -0.147 |
Gutenberg | -0.146 |
unlaw | -0.142 |
Token | Champions |
Token ID | 6662 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
workers | +1.155 |
trainers | +1.134 |
tubing | +1.047 |
HQ | +1.033 |
enta | +1.016 |
Neg logit contributions | |
Crimes | -1.249 |
certific | -1.207 |
witz | -1.184 |
Gutenberg | -1.173 |
unlaw | -1.143 |
Token | League |
Token ID | 4041 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Crimes | +2.987 |
Vienna | +2.777 |
certific | +2.729 |
Gutenberg | +2.722 |
witz | +2.706 |
Neg logit contributions | |
workers | -4.361 |
trainers | -4.284 |
tubing | -4.067 |
enta | -3.960 |
agen | -3.903 |
Token | qualification |
Token ID | 28587 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
Crimes | +4.863 |
certific | +4.563 |
Gutenberg | +4.526 |
Vienna | +4.435 |
witz | +4.427 |
Neg logit contributions | |
workers | -6.358 |
trainers | -6.074 |
enta | -5.800 |
HQ | -5.735 |
tubing | -5.721 |
Token | following |
Token ID | 1708 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
Crimes | +1.363 |
certific | +1.299 |
witz | +1.284 |
Gutenberg | +1.283 |
Vienna | +1.236 |
Neg logit contributions | |
workers | -1.343 |
trainers | -1.301 |
tubing | -1.215 |
HQ | -1.208 |
enta | -1.196 |
Token | important |
Token ID | 1593 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
workers | +0.562 |
trainers | +0.550 |
tubing | +0.511 |
HQ | +0.503 |
enta | +0.495 |
Neg logit contributions | |
Crimes | -0.598 |
certific | -0.577 |
witz | -0.566 |
Gutenberg | -0.563 |
unlaw | -0.547 |
Token | victories |
Token ID | 19017 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
workers | +2.447 |
trainers | +2.424 |
tubing | +2.251 |
HQ | +2.204 |
enta | +2.167 |
Neg logit contributions | |
Crimes | -2.521 |
certific | -2.433 |
witz | -2.407 |
Gutenberg | -2.384 |
unlaw | -2.318 |
Token | at |
Token ID | 379 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
Crimes | +0.344 |
certific | +0.329 |
witz | +0.324 |
Gutenberg | +0.322 |
unlaw | +0.310 |
Neg logit contributions | |
workers | -0.399 |
trainers | -0.390 |
tubing | -0.366 |
HQ | -0.361 |
enta | -0.356 |
Token | a |
Token ID | 257 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
workers | +0.018 |
trainers | +0.018 |
tubing | +0.017 |
HQ | +0.016 |
enta | +0.016 |
Neg logit contributions | |
Crimes | -0.020 |
certific | -0.019 |
witz | -0.019 |
Gutenberg | -0.019 |
unlaw | -0.018 |
Token | really |
Token ID | 1107 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
Crimes | +0.511 |
certific | +0.491 |
witz | +0.482 |
Gutenberg | +0.480 |
unlaw | +0.461 |
Neg logit contributions | |
workers | -0.530 |
trainers | -0.517 |
tubing | -0.483 |
HQ | -0.478 |
enta | -0.473 |
Token | bad |
Token ID | 2089 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
Crimes | +0.040 |
certific | +0.039 |
witz | +0.038 |
Gutenberg | +0.038 |
unlaw | +0.037 |
Neg logit contributions | |
workers | -0.040 |
trainers | -0.038 |
tubing | -0.036 |
HQ | -0.035 |
enta | -0.035 |
Token | mistake |
Token ID | 7457 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
workers | +0.523 |
trainers | +0.512 |
tubing | +0.480 |
HQ | +0.474 |
enta | +0.468 |
Neg logit contributions | |
Crimes | -0.449 |
certific | -0.429 |
witz | -0.421 |
Gutenberg | -0.419 |
unlaw | -0.404 |
Token | to |
Token ID | 284 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
workers | +0.912 |
trainers | +0.886 |
tubing | +0.819 |
HQ | +0.806 |
enta | +0.793 |
Neg logit contributions | |
Crimes | -1.178 |
certific | -1.137 |
witz | -1.120 |
Gutenberg | -1.116 |
unlaw | -1.083 |
Token | wave |
Token ID | 6769 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
Crimes | +1.432 |
certific | +1.377 |
witz | +1.343 |
Gutenberg | +1.335 |
unlaw | +1.290 |
Neg logit contributions | |
workers | -1.620 |
trainers | -1.573 |
HQ | -1.473 |
tubing | -1.468 |
enta | -1.438 |
Token | away |
Token ID | 1497 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
Crimes | +5.280 |
certific | +5.130 |
Gutenberg | +5.124 |
witz | +5.086 |
Vienna | +4.914 |
Neg logit contributions | |
workers | -5.574 |
trainers | -5.474 |
tubing | -5.154 |
HQ | -5.078 |
enta | -5.008 |
Token | policy |
Token ID | 2450 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Crimes | +0.217 |
certific | +0.210 |
witz | +0.206 |
Gutenberg | +0.206 |
unlaw | +0.198 |
Neg logit contributions | |
workers | -0.215 |
trainers | -0.210 |
tubing | -0.196 |
HQ | -0.194 |
enta | -0.191 |
Token | sill |
Token ID | 49276 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
workers | +0.393 |
trainers | +0.384 |
tubing | +0.361 |
HQ | +0.357 |
enta | +0.353 |
Neg logit contributions | |
Crimes | -0.325 |
certific | -0.312 |
witz | -0.306 |
Gutenberg | -0.304 |
unlaw | -0.292 |
Token | iness |
Token ID | 1272 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
Crimes | +0.119 |
certific | +0.115 |
witz | +0.115 |
Gutenberg | +0.113 |
unlaw | +0.110 |
Neg logit contributions | |
workers | -0.057 |
trainers | -0.055 |
tubing | -0.049 |
HQ | -0.048 |
enta | -0.047 |
Token | with |
Token ID | 351 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
Crimes | +1.362 |
certific | +1.308 |
witz | +1.287 |
Gutenberg | +1.275 |
Vienna | +1.224 |
Neg logit contributions | |
workers | -1.422 |
trainers | -1.380 |
tubing | -1.293 |
HQ | -1.279 |
enta | -1.267 |
Token | HR |
Token ID | 15172 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
workers | +1.075 |
trainers | +1.046 |
tubing | +0.967 |
HQ | +0.952 |
enta | +0.936 |
Neg logit contributions | |
Crimes | -1.264 |
certific | -1.213 |
witz | -1.201 |
Gutenberg | -1.195 |
unlaw | -1.158 |
Token | ? |
Token ID | 30 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
Crimes | +0.630 |
certific | +0.607 |
witz | +0.592 |
Gutenberg | +0.587 |
unlaw | +0.567 |
Neg logit contributions | |
workers | -0.621 |
trainers | -0.605 |
tubing | -0.568 |
HQ | -0.557 |
enta | -0.555 |
Token |
|
Token ID | 198 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Crimes | +0.252 |
certific | +0.239 |
witz | +0.236 |
Gutenberg | +0.235 |
Vienna | +0.224 |
Neg logit contributions | |
workers | -0.294 |
trainers | -0.288 |
tubing | -0.271 |
HQ | -0.266 |
enta | -0.265 |
Token |
|
Token ID | 198 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Crimes | +2.636 |
witz | +2.479 |
Gutenberg | +2.418 |
Vienna | +2.341 |
certific | +2.328 |
Neg logit contributions | |
trainers | -3.715 |
workers | -3.708 |
tubing | -3.546 |
enta | -3.438 |
HQ | -3.370 |
Token | A |
Token ID | 32 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
Crimes | +2.356 |
witz | +2.276 |
certific | +2.269 |
Gutenberg | +2.208 |
Vienna | +2.083 |
Neg logit contributions | |
workers | -3.495 |
trainers | -3.484 |
tubing | -3.362 |
enta | -3.215 |
agen | -3.155 |
Token | : |
Token ID | 25 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
Crimes | +3.559 |
witz | +3.304 |
Gutenberg | +3.226 |
Vienna | +3.106 |
certific | +3.080 |
Neg logit contributions | |
trainers | -6.086 |
workers | -6.069 |
tubing | -5.877 |
enta | -5.606 |
airs | -5.575 |
Token | Yes |
Token ID | 5297 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
Crimes | +4.974 |
witz | +4.513 |
Gutenberg | +4.435 |
Vienna | +4.400 |
certific | +4.375 |
Neg logit contributions | |
workers | -6.026 |
trainers | -5.953 |
tubing | -5.897 |
enta | -5.769 |
agen | -5.593 |
Token | . |
Token ID | 13 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
Crimes | +4.441 |
certific | +4.274 |
witz | +4.143 |
Gutenberg | +3.984 |
Vienna | +3.911 |
Neg logit contributions | |
workers | -5.988 |
trainers | -5.690 |
tubing | -5.579 |
enta | -5.514 |
joining | -5.461 |
Token | There |
Token ID | 1318 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
Crimes | +1.234 |
certific | +1.152 |
witz | +1.149 |
Gutenberg | +1.144 |
Vienna | +1.099 |
Neg logit contributions | |
workers | -1.401 |
trainers | -1.378 |
tubing | -1.305 |
enta | -1.268 |
HQ | -1.263 |
Token | are |
Token ID | 389 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
Crimes | +1.894 |
certific | +1.791 |
witz | +1.779 |
Gutenberg | +1.730 |
Vienna | +1.666 |
Neg logit contributions | |
workers | -2.432 |
trainers | -2.363 |
tubing | -2.252 |
HQ | -2.224 |
enta | -2.197 |
Token | new |
Token ID | 649 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
workers | +0.040 |
trainers | +0.038 |
tubing | +0.036 |
HQ | +0.035 |
enta | +0.035 |
Neg logit contributions | |
Crimes | -0.047 |
certific | -0.045 |
witz | -0.045 |
Gutenberg | -0.044 |
unlaw | -0.043 |
Token | TA |
Token ID | 5603 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
Crimes | +2.736 |
certific | +2.572 |
witz | +2.530 |
Gutenberg | +2.509 |
Vienna | +2.447 |
Neg logit contributions | |
workers | -2.657 |
trainers | -2.626 |
tubing | -2.403 |
agen | -2.329 |
HQ | -2.329 |
Token | WA |
Token ID | 15543 |
Feature activation | -7.69e-02 |
Pos logit contributions | |
workers | +0.155 |
trainers | +0.152 |
HQ | +0.144 |
tubing | +0.143 |
ox | +0.139 |
Neg logit contributions | |
Crimes | -0.120 |
certific | -0.118 |
witz | -0.116 |
Gutenberg | -0.115 |
unlaw | -0.111 |
Token | ( |
Token ID | 357 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
workers | +0.223 |
trainers | +0.218 |
tubing | +0.204 |
HQ | +0.201 |
enta | +0.198 |
Neg logit contributions | |
Crimes | -0.209 |
certific | -0.200 |
witz | -0.197 |
Gutenberg | -0.196 |
unlaw | -0.189 |
Token | Reuters |
Token ID | 12637 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
Crimes | +0.291 |
certific | +0.273 |
witz | +0.271 |
Gutenberg | +0.269 |
Vienna | +0.255 |
Neg logit contributions | |
workers | -0.374 |
trainers | -0.369 |
tubing | -0.347 |
enta | -0.338 |
HQ | -0.338 |
Token | ) |
Token ID | 8 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
workers | +0.036 |
trainers | +0.035 |
HQ | +0.033 |
tubing | +0.033 |
enta | +0.032 |
Neg logit contributions | |
Crimes | -0.031 |
certific | -0.031 |
witz | -0.030 |
Gutenberg | -0.030 |
unlaw | -0.029 |
Token | - |
Token ID | 532 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
Crimes | +3.123 |
witz | +2.899 |
Gutenberg | +2.887 |
Vienna | +2.846 |
certific | +2.658 |
Neg logit contributions | |
trainers | -3.650 |
workers | -3.646 |
tubing | -3.438 |
enta | -3.336 |
HQ | -3.311 |
Token | Canada |
Token ID | 3340 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
Crimes | +5.189 |
Gutenberg | +4.648 |
witz | +4.571 |
Vienna | +4.564 |
Fraud | +4.302 |
Neg logit contributions | |
trainers | -5.943 |
workers | -5.847 |
tubing | -5.648 |
HQ | -5.461 |
enta | -5.439 |
Token | � |
Token ID | 447 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
workers | +0.921 |
trainers | +0.895 |
tubing | +0.831 |
HQ | +0.820 |
enta | +0.807 |
Neg logit contributions | |
Crimes | -1.061 |
certific | -1.020 |
witz | -1.003 |
Gutenberg | -1.000 |
unlaw | -0.966 |
Token | � |
Token ID | 247 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Crimes | +0.808 |
certific | +0.769 |
witz | +0.758 |
Gutenberg | +0.753 |
unlaw | +0.722 |
Neg logit contributions | |
workers | -0.868 |
trainers | -0.847 |
tubing | -0.791 |
HQ | -0.781 |
enta | -0.772 |
Token | s |
Token ID | 82 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
Crimes | +0.337 |
certific | +0.322 |
witz | +0.318 |
Gutenberg | +0.315 |
unlaw | +0.303 |
Neg logit contributions | |
workers | -0.323 |
trainers | -0.317 |
tubing | -0.295 |
HQ | -0.291 |
enta | -0.286 |
Token | top |
Token ID | 1353 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
workers | +0.531 |
trainers | +0.514 |
tubing | +0.480 |
HQ | +0.477 |
enta | +0.469 |
Neg logit contributions | |
Crimes | -0.562 |
certific | -0.538 |
Gutenberg | -0.528 |
witz | -0.527 |
unlaw | -0.509 |
Token | � |
Token ID | 447 |
Feature activation | -9.98e-02 |
Pos logit contributions | |
workers | +0.918 |
trainers | +0.896 |
tubing | +0.840 |
HQ | +0.830 |
enta | +0.819 |
Neg logit contributions | |
Crimes | -0.802 |
certific | -0.769 |
witz | -0.755 |
Gutenberg | -0.751 |
unlaw | -0.723 |
Token | � |
Token ID | 251 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
workers | +0.258 |
trainers | +0.250 |
tubing | +0.231 |
HQ | +0.228 |
enta | +0.225 |
Neg logit contributions | |
Crimes | -0.305 |
certific | -0.292 |
witz | -0.289 |
Gutenberg | -0.287 |
unlaw | -0.278 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
workers | +0.678 |
trainers | +0.663 |
tubing | +0.624 |
HQ | +0.613 |
enta | +0.607 |
Neg logit contributions | |
Crimes | -0.575 |
certific | -0.539 |
witz | -0.539 |
Gutenberg | -0.536 |
Vienna | -0.510 |
Token | Personal |
Token ID | 30228 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
workers | +0.826 |
trainers | +0.807 |
tubing | +0.753 |
HQ | +0.738 |
enta | +0.730 |
Neg logit contributions | |
Crimes | -0.865 |
certific | -0.831 |
witz | -0.820 |
Gutenberg | -0.816 |
unlaw | -0.787 |
Token | Liberty |
Token ID | 14734 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
workers | +2.302 |
trainers | +2.278 |
tubing | +2.131 |
HQ | +2.106 |
enta | +2.078 |
Neg logit contributions | |
Crimes | -1.532 |
certific | -1.472 |
witz | -1.452 |
Gutenberg | -1.424 |
unlaw | -1.373 |
Token | Poll |
Token ID | 12868 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
workers | +1.840 |
trainers | +1.810 |
tubing | +1.693 |
HQ | +1.674 |
enta | +1.645 |
Neg logit contributions | |
Crimes | -1.486 |
certific | -1.439 |
witz | -1.414 |
Gutenberg | -1.399 |
unlaw | -1.352 |
Token | Exercise |
Token ID | 32900 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
workers | +6.272 |
trainers | +6.063 |
ox | +5.663 |
tubing | +5.657 |
HQ | +5.637 |
Neg logit contributions | |
certific | -5.638 |
Crimes | -5.600 |
witz | -5.332 |
Gutenberg | -5.295 |
unlaw | -5.292 |
Token | your |
Token ID | 534 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
workers | +3.000 |
trainers | +2.970 |
tubing | +2.769 |
HQ | +2.730 |
ox | +2.686 |
Neg logit contributions | |
Crimes | -2.644 |
certific | -2.563 |
witz | -2.510 |
Gutenberg | -2.484 |
unlaw | -2.434 |
Token | right |
Token ID | 826 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
workers | +0.967 |
trainers | +0.947 |
tubing | +0.893 |
HQ | +0.883 |
enta | +0.871 |
Neg logit contributions | |
Crimes | -0.687 |
certific | -0.655 |
witz | -0.642 |
Gutenberg | -0.638 |
unlaw | -0.612 |
Token | to |
Token ID | 284 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
Crimes | +0.407 |
certific | +0.391 |
witz | +0.385 |
Gutenberg | +0.383 |
unlaw | +0.374 |
Neg logit contributions | |
workers | -0.326 |
trainers | -0.316 |
tubing | -0.293 |
HQ | -0.289 |
enta | -0.286 |
Token | vote |
Token ID | 3015 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
Crimes | +1.049 |
certific | +1.007 |
Gutenberg | +0.980 |
witz | +0.974 |
unlaw | +0.953 |
Neg logit contributions | |
workers | -1.074 |
trainers | -1.037 |
tubing | -0.973 |
HQ | -0.966 |
enta | -0.951 |
Token | first |
Token ID | 717 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
workers | +0.568 |
trainers | +0.553 |
tubing | +0.514 |
HQ | +0.506 |
enta | +0.499 |
Neg logit contributions | |
Crimes | -0.648 |
certific | -0.622 |
witz | -0.613 |
Gutenberg | -0.611 |
unlaw | -0.589 |
Token | Ghostbusters |
Token ID | 50037 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
workers | +0.840 |
trainers | +0.823 |
tubing | +0.773 |
HQ | +0.761 |
enta | +0.754 |
Neg logit contributions | |
Crimes | -0.709 |
certific | -0.671 |
Gutenberg | -0.664 |
witz | -0.662 |
unlaw | -0.632 |
Token | , |
Token ID | 11 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
Crimes | +0.049 |
certific | +0.047 |
witz | +0.047 |
Gutenberg | +0.046 |
unlaw | +0.045 |
Neg logit contributions | |
workers | -0.049 |
trainers | -0.048 |
tubing | -0.045 |
HQ | -0.044 |
enta | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
workers | +0.055 |
trainers | +0.053 |
tubing | +0.050 |
HQ | +0.049 |
ox | +0.048 |
Neg logit contributions | |
Crimes | -0.058 |
certific | -0.056 |
witz | -0.055 |
Gutenberg | -0.054 |
unlaw | -0.053 |
Token | ne |
Token ID | 710 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
workers | +0.051 |
trainers | +0.050 |
tubing | +0.046 |
HQ | +0.045 |
enta | +0.045 |
Neg logit contributions | |
Crimes | -0.058 |
certific | -0.055 |
Gutenberg | -0.055 |
witz | -0.055 |
Vienna | -0.052 |
Token | igh |
Token ID | 394 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
Crimes | +2.124 |
witz | +2.000 |
Gutenberg | +1.981 |
certific | +1.981 |
Vienna | +1.936 |
Neg logit contributions | |
workers | -2.219 |
trainers | -2.187 |
tubing | -2.025 |
HQ | -1.984 |
enta | -1.973 |
Token | bor |
Token ID | 2865 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
workers | +3.343 |
trainers | +3.123 |
tubing | +2.808 |
agen | +2.705 |
ox | +2.679 |
Neg logit contributions | |
Crimes | -8.255 |
certific | -8.147 |
Gutenberg | -7.969 |
unlaw | -7.778 |
witz | -7.757 |
Token | hood |
Token ID | 2894 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
Crimes | +0.438 |
certific | +0.406 |
witz | +0.402 |
Gutenberg | +0.400 |
Vienna | +0.380 |
Neg logit contributions | |
workers | -0.715 |
trainers | -0.704 |
tubing | -0.664 |
HQ | -0.658 |
enta | -0.652 |
Token | in |
Token ID | 287 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
Crimes | +1.373 |
certific | +1.291 |
witz | +1.287 |
Gutenberg | +1.283 |
Vienna | +1.256 |
Neg logit contributions | |
workers | -1.364 |
trainers | -1.338 |
tubing | -1.245 |
HQ | -1.221 |
enta | -1.210 |
Token | New |
Token ID | 968 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
Crimes | +0.139 |
certific | +0.131 |
witz | +0.130 |
Gutenberg | +0.129 |
Vienna | +0.122 |
Neg logit contributions | |
workers | -0.206 |
trainers | -0.203 |
tubing | -0.191 |
HQ | -0.189 |
enta | -0.186 |
Token | York |
Token ID | 1971 |
Feature activation | +7.86e-03 |
Pos logit contributions | |
workers | +0.493 |
trainers | +0.477 |
tubing | +0.444 |
HQ | +0.443 |
enta | +0.434 |
Neg logit contributions | |
Crimes | -0.508 |
certific | -0.496 |
witz | -0.481 |
Gutenberg | -0.479 |
unlaw | -0.470 |
Token | south |
Token ID | 5366 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
workers | +0.324 |
trainers | +0.316 |
tubing | +0.296 |
HQ | +0.292 |
enta | +0.288 |
Neg logit contributions | |
Crimes | -0.306 |
certific | -0.294 |
witz | -0.288 |
Gutenberg | -0.287 |
unlaw | -0.277 |
Token | as |
Token ID | 355 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
Crimes | +1.295 |
certific | +1.227 |
witz | +1.221 |
Gutenberg | +1.210 |
Vienna | +1.163 |
Neg logit contributions | |
workers | -1.332 |
trainers | -1.300 |
tubing | -1.216 |
HQ | -1.201 |
enta | -1.177 |
Token | the |
Token ID | 262 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
workers | +0.554 |
trainers | +0.540 |
tubing | +0.504 |
HQ | +0.499 |
enta | +0.491 |
Neg logit contributions | |
Crimes | -0.554 |
certific | -0.530 |
witz | -0.523 |
Gutenberg | -0.521 |
unlaw | -0.502 |
Token | city |
Token ID | 1748 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
workers | +0.178 |
trainers | +0.172 |
tubing | +0.160 |
HQ | +0.159 |
enta | +0.156 |
Neg logit contributions | |
Crimes | -0.190 |
certific | -0.182 |
witz | -0.180 |
Gutenberg | -0.179 |
unlaw | -0.173 |
Token | of |
Token ID | 286 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
workers | +1.836 |
trainers | +1.793 |
tubing | +1.683 |
HQ | +1.661 |
enta | +1.641 |
Neg logit contributions | |
Crimes | -1.578 |
certific | -1.513 |
witz | -1.483 |
Gutenberg | -1.476 |
unlaw | -1.422 |
Token | Zin |
Token ID | 36971 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
workers | +1.756 |
trainers | +1.713 |
tubing | +1.595 |
HQ | +1.569 |
enta | +1.552 |
Neg logit contributions | |
Crimes | -1.810 |
certific | -1.758 |
witz | -1.721 |
Gutenberg | -1.702 |
unlaw | -1.655 |
Token | , |
Token ID | 11 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
workers | +5.503 |
trainers | +5.374 |
ox | +4.980 |
HQ | +4.950 |
enta | +4.947 |
Neg logit contributions | |
Crimes | -6.060 |
certific | -5.948 |
Gutenberg | -5.672 |
unlaw | -5.668 |
witz | -5.632 |
Token | and |
Token ID | 290 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
workers | +0.559 |
trainers | +0.544 |
tubing | +0.507 |
HQ | +0.502 |
enta | +0.496 |
Neg logit contributions | |
Crimes | -0.577 |
certific | -0.552 |
witz | -0.544 |
Gutenberg | -0.542 |
unlaw | -0.522 |
Token | has |
Token ID | 468 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
Crimes | +0.525 |
certific | +0.497 |
Gutenberg | +0.493 |
witz | +0.492 |
unlaw | +0.472 |
Neg logit contributions | |
workers | -0.547 |
trainers | -0.529 |
HQ | -0.496 |
tubing | -0.496 |
enta | -0.488 |
Token | even |
Token ID | 772 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
Crimes | +0.191 |
certific | +0.184 |
witz | +0.181 |
Gutenberg | +0.180 |
unlaw | +0.174 |
Neg logit contributions | |
workers | -0.183 |
trainers | -0.177 |
tubing | -0.165 |
HQ | -0.164 |
enta | -0.161 |
Token | landed |
Token ID | 11406 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
Crimes | +0.409 |
certific | +0.394 |
witz | +0.386 |
Gutenberg | +0.384 |
unlaw | +0.371 |
Neg logit contributions | |
workers | -0.409 |
trainers | -0.397 |
tubing | -0.370 |
HQ | -0.369 |
enta | -0.361 |
Token | 10 |
Token ID | 940 |
Feature activation | -6.06e-02 |
Pos logit contributions | |
workers | +0.838 |
trainers | +0.795 |
tubing | +0.717 |
HQ | +0.717 |
ox | +0.698 |
Neg logit contributions | |
Crimes | -1.579 |
certific | -1.544 |
Gutenberg | -1.509 |
witz | -1.506 |
unlaw | -1.482 |
Token | : |
Token ID | 25 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
workers | +0.193 |
trainers | +0.187 |
tubing | +0.176 |
HQ | +0.175 |
enta | +0.172 |
Neg logit contributions | |
Crimes | -0.148 |
certific | -0.144 |
witz | -0.139 |
Gutenberg | -0.138 |
unlaw | -0.135 |
Token | 47 |
Token ID | 2857 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
workers | +0.558 |
trainers | +0.542 |
tubing | +0.500 |
HQ | +0.492 |
enta | +0.484 |
Neg logit contributions | |
Crimes | -0.740 |
certific | -0.714 |
witz | -0.704 |
Gutenberg | -0.700 |
unlaw | -0.680 |
Token | ] |
Token ID | 60 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
workers | +0.861 |
trainers | +0.831 |
HQ | +0.781 |
tubing | +0.773 |
ox | +0.768 |
Neg logit contributions | |
Crimes | -0.800 |
certific | -0.791 |
witz | -0.754 |
Gutenberg | -0.753 |
unlaw | -0.749 |
Token | < |
Token ID | 1279 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
workers | +0.573 |
trainers | +0.561 |
tubing | +0.531 |
HQ | +0.526 |
enta | +0.520 |
Neg logit contributions | |
Crimes | -0.348 |
certific | -0.330 |
witz | -0.322 |
Gutenberg | -0.321 |
unlaw | -0.306 |
Token | sych |
Token ID | 2924 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
workers | +0.475 |
trainers | +0.464 |
tubing | +0.430 |
HQ | +0.422 |
enta | +0.416 |
Neg logit contributions | |
Crimes | -0.563 |
certific | -0.543 |
witz | -0.536 |
Gutenberg | -0.533 |
unlaw | -0.516 |
Token | omen |
Token ID | 3674 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
workers | +5.468 |
trainers | +5.216 |
ox | +4.982 |
enta | +4.866 |
agen | +4.857 |
Neg logit contributions | |
Crimes | -6.235 |
certific | -5.994 |
Gutenberg | -5.902 |
witz | -5.768 |
unlaw | -5.664 |
Token | > |
Token ID | 29 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
Crimes | +0.186 |
certific | +0.179 |
witz | +0.177 |
Gutenberg | +0.176 |
unlaw | +0.170 |
Neg logit contributions | |
workers | -0.166 |
trainers | -0.162 |
tubing | -0.151 |
HQ | -0.148 |
enta | -0.146 |
Token | GU |
Token ID | 19348 |
Feature activation | -8.37e-01 |
Pos logit contributions | |
workers | +0.082 |
trainers | +0.080 |
tubing | +0.074 |
HQ | +0.074 |
enta | +0.073 |
Neg logit contributions | |
Crimes | -0.080 |
certific | -0.076 |
witz | -0.075 |
Gutenberg | -0.075 |
unlaw | -0.072 |
Token | YS |
Token ID | 16309 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
workers | +2.479 |
trainers | +2.425 |
HQ | +2.270 |
tubing | +2.256 |
ox | +2.225 |
Neg logit contributions | |
Crimes | -2.198 |
certific | -2.094 |
Gutenberg | -2.062 |
witz | -2.059 |
unlaw | -1.992 |
Token | STA |
Token ID | 47550 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
workers | +1.342 |
trainers | +1.316 |
tubing | +1.238 |
HQ | +1.226 |
enta | +1.207 |
Neg logit contributions | |
Crimes | -1.061 |
certific | -1.015 |
witz | -0.995 |
Gutenberg | -0.991 |
unlaw | -0.952 |
Token | that |
Token ID | 326 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
Crimes | +1.010 |
certific | +0.978 |
Gutenberg | +0.959 |
witz | +0.952 |
Vienna | +0.926 |
Neg logit contributions | |
workers | -0.909 |
trainers | -0.885 |
tubing | -0.826 |
HQ | -0.814 |
enta | -0.800 |
Token | Republican |
Token ID | 3415 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
Crimes | +1.230 |
certific | +1.172 |
Gutenberg | +1.154 |
witz | +1.138 |
Vienna | +1.110 |
Neg logit contributions | |
workers | -1.240 |
trainers | -1.201 |
HQ | -1.132 |
tubing | -1.123 |
enta | -1.101 |
Token | Rep |
Token ID | 1432 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
Crimes | +0.836 |
certific | +0.804 |
Gutenberg | +0.789 |
witz | +0.777 |
Vienna | +0.762 |
Neg logit contributions | |
workers | -0.672 |
trainers | -0.641 |
tubing | -0.604 |
HQ | -0.599 |
enta | -0.591 |
Token | . |
Token ID | 13 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
workers | +1.721 |
trainers | +1.661 |
HQ | +1.552 |
tubing | +1.539 |
ox | +1.530 |
Neg logit contributions | |
Crimes | -1.634 |
certific | -1.613 |
witz | -1.546 |
unlaw | -1.527 |
Gutenberg | -1.527 |
Token | Ryan |
Token ID | 6047 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
workers | +0.194 |
trainers | +0.189 |
tubing | +0.178 |
HQ | +0.175 |
enta | +0.174 |
Neg logit contributions | |
Crimes | -0.180 |
certific | -0.171 |
Gutenberg | -0.169 |
witz | -0.169 |
unlaw | -0.161 |
Token | Zin |
Token ID | 36971 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
Crimes | +0.267 |
certific | +0.254 |
witz | +0.254 |
Gutenberg | +0.252 |
Vienna | +0.243 |
Neg logit contributions | |
workers | -0.205 |
trainers | -0.198 |
tubing | -0.185 |
HQ | -0.183 |
enta | -0.181 |
Token | ke |
Token ID | 365 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
workers | +3.754 |
trainers | +3.365 |
ox | +3.301 |
enta | +3.185 |
electric | +3.123 |
Neg logit contributions | |
Crimes | -7.631 |
unlaw | -7.558 |
Gutenberg | -7.466 |
certific | -7.410 |
witz | -7.075 |
Token | won |
Token ID | 1839 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
Crimes | +0.353 |
certific | +0.337 |
witz | +0.333 |
Gutenberg | +0.329 |
unlaw | +0.315 |
Neg logit contributions | |
workers | -0.394 |
trainers | -0.383 |
tubing | -0.361 |
HQ | -0.356 |
enta | -0.350 |
Token | by |
Token ID | 416 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
workers | +0.567 |
trainers | +0.553 |
tubing | +0.516 |
HQ | +0.508 |
enta | +0.502 |
Neg logit contributions | |
Crimes | -0.577 |
certific | -0.554 |
witz | -0.546 |
Gutenberg | -0.543 |
unlaw | -0.522 |
Token | 15 |
Token ID | 1315 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
Crimes | +0.580 |
certific | +0.558 |
Gutenberg | +0.548 |
witz | +0.545 |
unlaw | +0.524 |
Neg logit contributions | |
workers | -0.581 |
trainers | -0.567 |
HQ | -0.528 |
tubing | -0.528 |
enta | -0.519 |
Token | points |
Token ID | 2173 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
workers | +1.768 |
trainers | +1.725 |
tubing | +1.613 |
HQ | +1.592 |
enta | +1.571 |
Neg logit contributions | |
Crimes | -1.639 |
certific | -1.571 |
witz | -1.547 |
Gutenberg | -1.536 |
unlaw | -1.484 |
Token | , |
Token ID | 11 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
workers | +2.158 |
trainers | +2.093 |
tubing | +1.961 |
HQ | +1.944 |
enta | +1.915 |
Neg logit contributions | |
Crimes | -2.097 |
certific | -2.031 |
witz | -1.984 |
Gutenberg | -1.977 |
unlaw | -1.919 |
Token | according |
Token ID | 1864 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
workers | +0.988 |
trainers | +0.961 |
tubing | +0.899 |
HQ | +0.887 |
enta | +0.874 |
Neg logit contributions | |
Crimes | -0.985 |
certific | -0.950 |
witz | -0.934 |
Gutenberg | -0.928 |
unlaw | -0.902 |
Token | to |
Token ID | 284 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
workers | +0.926 |
trainers | +0.892 |
HQ | +0.820 |
tubing | +0.814 |
ox | +0.796 |
Neg logit contributions | |
Crimes | -1.389 |
certific | -1.378 |
Gutenberg | -1.328 |
witz | -1.326 |
unlaw | -1.326 |
Token | a |
Token ID | 257 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
workers | +0.725 |
trainers | +0.710 |
tubing | +0.663 |
HQ | +0.656 |
enta | +0.647 |
Neg logit contributions | |
Crimes | -0.671 |
certific | -0.646 |
witz | -0.633 |
Gutenberg | -0.630 |
unlaw | -0.610 |
Token | new |
Token ID | 649 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
workers | +0.865 |
trainers | +0.844 |
tubing | +0.787 |
HQ | +0.779 |
enta | +0.765 |
Neg logit contributions | |
Crimes | -0.868 |
certific | -0.837 |
witz | -0.820 |
Gutenberg | -0.816 |
unlaw | -0.794 |
Token | poll |
Token ID | 3278 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
workers | +1.856 |
trainers | +1.820 |
tubing | +1.700 |
HQ | +1.692 |
enta | +1.656 |
Neg logit contributions | |
Crimes | -1.711 |
certific | -1.661 |
witz | -1.609 |
Gutenberg | -1.600 |
unlaw | -1.570 |
Token | released |
Token ID | 2716 |
Feature activation | -9.97e-01 |
Pos logit contributions | |
workers | +5.197 |
trainers | +5.068 |
tubing | +4.694 |
HQ | +4.679 |
ox | +4.609 |
Neg logit contributions | |
Crimes | -5.625 |
certific | -5.506 |
unlaw | -5.343 |
Gutenberg | -5.312 |
witz | -5.242 |
Token | Thursday |
Token ID | 3635 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
workers | +2.878 |
trainers | +2.819 |
tubing | +2.645 |
HQ | +2.615 |
ox | +2.584 |
Neg logit contributions | |
Crimes | -2.631 |
certific | -2.576 |
Gutenberg | -2.462 |
witz | -2.456 |
unlaw | -2.437 |
Token | by |
Token ID | 416 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
workers | +0.046 |
trainers | +0.045 |
tubing | +0.042 |
HQ | +0.042 |
enta | +0.041 |
Neg logit contributions | |
Crimes | -0.040 |
certific | -0.038 |
witz | -0.037 |
Gutenberg | -0.037 |
unlaw | -0.036 |
Token | Public |
Token ID | 5094 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
Crimes | +0.267 |
certific | +0.254 |
witz | +0.251 |
Gutenberg | +0.250 |
Vienna | +0.239 |
Neg logit contributions | |
workers | -0.309 |
trainers | -0.300 |
tubing | -0.283 |
HQ | -0.278 |
enta | -0.275 |
Token | Policy |
Token ID | 7820 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
workers | +2.287 |
trainers | +2.244 |
tubing | +2.129 |
HQ | +2.113 |
enta | +2.089 |
Neg logit contributions | |
Crimes | -1.173 |
certific | -1.110 |
witz | -1.081 |
Gutenberg | -1.075 |
unlaw | -1.018 |
Token | Reuters |
Token ID | 8428 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
workers | +0.181 |
trainers | +0.177 |
tubing | +0.165 |
HQ | +0.163 |
enta | +0.160 |
Neg logit contributions | |
Crimes | -0.189 |
certific | -0.181 |
witz | -0.179 |
Gutenberg | -0.178 |
unlaw | -0.172 |
Token | / |
Token ID | 14 |
Feature activation | -5.72e-02 |
Pos logit contributions | |
workers | +1.448 |
trainers | +1.411 |
tubing | +1.316 |
HQ | +1.304 |
enta | +1.280 |
Neg logit contributions | |
Crimes | -1.436 |
certific | -1.386 |
witz | -1.359 |
Gutenberg | -1.352 |
unlaw | -1.311 |
Token | I |
Token ID | 40 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
workers | +0.178 |
trainers | +0.174 |
tubing | +0.164 |
HQ | +0.161 |
enta | +0.160 |
Neg logit contributions | |
Crimes | -0.143 |
certific | -0.137 |
witz | -0.134 |
Gutenberg | -0.134 |
unlaw | -0.128 |
Token | ps |
Token ID | 862 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
workers | +0.511 |
trainers | +0.486 |
tubing | +0.434 |
HQ | +0.428 |
enta | +0.417 |
Neg logit contributions | |
Crimes | -1.109 |
certific | -1.080 |
witz | -1.066 |
Gutenberg | -1.062 |
unlaw | -1.040 |
Token | os |
Token ID | 418 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
workers | +0.466 |
trainers | +0.453 |
tubing | +0.432 |
enta | +0.412 |
HQ | +0.411 |
Neg logit contributions | |
Crimes | -0.458 |
certific | -0.441 |
witz | -0.434 |
Gutenberg | -0.429 |
unlaw | -0.419 |
Token | poll |
Token ID | 3278 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
workers | +0.561 |
trainers | +0.546 |
tubing | +0.510 |
HQ | +0.500 |
enta | +0.494 |
Neg logit contributions | |
Crimes | -0.627 |
certific | -0.601 |
witz | -0.593 |
Gutenberg | -0.593 |
unlaw | -0.569 |
Token | , |
Token ID | 11 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
workers | +5.044 |
trainers | +4.871 |
tubing | +4.503 |
enta | +4.413 |
ox | +4.395 |
Neg logit contributions | |
Crimes | -5.748 |
certific | -5.668 |
unlaw | -5.501 |
Gutenberg | -5.478 |
witz | -5.383 |
Token | half |
Token ID | 2063 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
workers | +1.116 |
trainers | +1.090 |
tubing | +1.014 |
HQ | +1.000 |
enta | +0.984 |
Neg logit contributions | |
Crimes | -1.116 |
certific | -1.073 |
witz | -1.056 |
Gutenberg | -1.049 |
unlaw | -1.017 |
Token | believe |
Token ID | 1975 |
Feature activation | -8.58e-01 |
Pos logit contributions | |
workers | +1.457 |
trainers | +1.421 |
tubing | +1.333 |
HQ | +1.306 |
enta | +1.298 |
Neg logit contributions | |
Crimes | -1.342 |
certific | -1.280 |
witz | -1.260 |
Gutenberg | -1.258 |
unlaw | -1.219 |
Token | the |
Token ID | 262 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
workers | +2.492 |
trainers | +2.442 |
tubing | +2.276 |
HQ | +2.246 |
ox | +2.210 |
Neg logit contributions | |
Crimes | -2.302 |
certific | -2.224 |
witz | -2.182 |
Gutenberg | -2.151 |
unlaw | -2.095 |
Token | software |
Token ID | 3788 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
Crimes | +0.045 |
certific | +0.043 |
witz | +0.042 |
Gutenberg | +0.042 |
unlaw | +0.040 |
Neg logit contributions | |
workers | -0.057 |
trainers | -0.056 |
tubing | -0.053 |
HQ | -0.052 |
enta | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
workers | +0.824 |
trainers | +0.801 |
tubing | +0.748 |
HQ | +0.737 |
enta | +0.730 |
Neg logit contributions | |
Crimes | -0.850 |
certific | -0.808 |
witz | -0.797 |
Gutenberg | -0.795 |
unlaw | -0.766 |
Token | case |
Token ID | 1339 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
workers | +0.651 |
trainers | +0.636 |
tubing | +0.600 |
HQ | +0.591 |
enta | +0.589 |
Neg logit contributions | |
Crimes | -0.476 |
certific | -0.448 |
witz | -0.440 |
Gutenberg | -0.440 |
unlaw | -0.420 |
Token | . |
Token ID | 13 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
Crimes | +1.926 |
witz | +1.791 |
certific | +1.784 |
Gutenberg | +1.769 |
Vienna | +1.698 |
Neg logit contributions | |
workers | -2.155 |
trainers | -2.117 |
tubing | -1.987 |
enta | -1.967 |
HQ | -1.965 |
Token | In |
Token ID | 554 |
Feature activation | -6.55e-02 |
Pos logit contributions | |
workers | +0.657 |
trainers | +0.644 |
tubing | +0.606 |
HQ | +0.591 |
enta | +0.588 |
Neg logit contributions | |
Crimes | -0.614 |
certific | -0.573 |
Gutenberg | -0.570 |
witz | -0.570 |
Vienna | -0.545 |
Token | an |
Token ID | 281 |
Feature activation | -6.02e-03 |
Pos logit contributions | |
workers | +0.181 |
trainers | +0.176 |
tubing | +0.164 |
HQ | +0.163 |
enta | +0.162 |
Neg logit contributions | |
Crimes | -0.188 |
certific | -0.179 |
Gutenberg | -0.176 |
witz | -0.176 |
Vienna | -0.170 |
Token | interview |
Token ID | 2720 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
workers | +0.017 |
trainers | +0.016 |
tubing | +0.015 |
HQ | +0.015 |
enta | +0.015 |
Neg logit contributions | |
Crimes | -0.017 |
certific | -0.017 |
Gutenberg | -0.016 |
witz | -0.016 |
unlaw | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
workers | +4.938 |
trainers | +4.789 |
tubing | +4.402 |
ox | +4.299 |
HQ | +4.277 |
Neg logit contributions | |
Crimes | -5.840 |
certific | -5.723 |
unlaw | -5.552 |
witz | -5.550 |
Gutenberg | -5.523 |
Token | he |
Token ID | 339 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
workers | +1.454 |
trainers | +1.413 |
tubing | +1.317 |
HQ | +1.297 |
enta | +1.281 |
Neg logit contributions | |
Crimes | -1.625 |
certific | -1.563 |
witz | -1.534 |
Gutenberg | -1.531 |
unlaw | -1.480 |
Token | said |
Token ID | 531 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
workers | +0.482 |
trainers | +0.469 |
tubing | +0.439 |
HQ | +0.435 |
enta | +0.429 |
Neg logit contributions | |
Crimes | -0.457 |
certific | -0.439 |
witz | -0.431 |
Gutenberg | -0.429 |
unlaw | -0.413 |
Token | that |
Token ID | 326 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
workers | +1.822 |
trainers | +1.771 |
tubing | +1.651 |
HQ | +1.628 |
enta | +1.609 |
Neg logit contributions | |
Crimes | -2.022 |
certific | -1.940 |
witz | -1.910 |
Gutenberg | -1.904 |
unlaw | -1.840 |
Token | the |
Token ID | 262 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
workers | +1.391 |
trainers | +1.350 |
tubing | +1.260 |
HQ | +1.240 |
enta | +1.231 |
Neg logit contributions | |
Crimes | -1.564 |
certific | -1.497 |
witz | -1.474 |
Gutenberg | -1.472 |
unlaw | -1.418 |
Token | ve |
Token ID | 1569 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
workers | +1.212 |
trainers | +1.176 |
tubing | +1.085 |
HQ | +1.068 |
enta | +1.051 |
Neg logit contributions | |
Crimes | -1.609 |
certific | -1.554 |
witz | -1.530 |
Gutenberg | -1.525 |
unlaw | -1.480 |
Token | gans |
Token ID | 39352 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
workers | +0.752 |
trainers | +0.723 |
tubing | +0.654 |
HQ | +0.643 |
enta | +0.631 |
Neg logit contributions | |
Crimes | -1.388 |
certific | -1.348 |
witz | -1.328 |
Gutenberg | -1.325 |
unlaw | -1.295 |
Token | , |
Token ID | 11 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
workers | +0.200 |
trainers | +0.195 |
tubing | +0.183 |
HQ | +0.181 |
enta | +0.179 |
Neg logit contributions | |
Crimes | -0.173 |
certific | -0.165 |
witz | -0.162 |
Gutenberg | -0.162 |
unlaw | -0.155 |
Token | according |
Token ID | 1864 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Crimes | +0.139 |
certific | +0.131 |
Gutenberg | +0.129 |
witz | +0.128 |
Vienna | +0.123 |
Neg logit contributions | |
workers | -0.197 |
trainers | -0.192 |
tubing | -0.181 |
HQ | -0.180 |
enta | -0.178 |
Token | to |
Token ID | 284 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
workers | +0.351 |
trainers | +0.340 |
tubing | +0.311 |
HQ | +0.309 |
ox | +0.301 |
Neg logit contributions | |
Crimes | -0.497 |
certific | -0.487 |
witz | -0.475 |
Gutenberg | -0.473 |
unlaw | -0.466 |
Token | Gallup |
Token ID | 29183 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
workers | +0.069 |
trainers | +0.067 |
tubing | +0.063 |
HQ | +0.062 |
enta | +0.062 |
Neg logit contributions | |
Crimes | -0.065 |
certific | -0.062 |
witz | -0.061 |
Gutenberg | -0.061 |
unlaw | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
workers | +5.284 |
trainers | +5.209 |
HQ | +4.873 |
tubing | +4.850 |
agen | +4.768 |
Neg logit contributions | |
Crimes | -5.318 |
certific | -5.286 |
Gutenberg | -5.089 |
witz | -5.064 |
unlaw | -4.985 |
Token |
|
Token ID | 198 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
Crimes | +0.139 |
Gutenberg | +0.129 |
witz | +0.129 |
certific | +0.128 |
Vienna | +0.124 |
Neg logit contributions | |
workers | -0.163 |
trainers | -0.160 |
tubing | -0.151 |
HQ | -0.147 |
enta | -0.146 |
Token |
|
Token ID | 198 |
Feature activation | -9.20e-02 |
Pos logit contributions | |
Crimes | +0.327 |
certific | +0.315 |
witz | +0.308 |
Gutenberg | +0.307 |
unlaw | +0.297 |
Neg logit contributions | |
workers | -0.367 |
trainers | -0.358 |
tubing | -0.335 |
HQ | -0.332 |
enta | -0.327 |
Token | Follow |
Token ID | 7155 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
workers | +0.249 |
trainers | +0.245 |
tubing | +0.230 |
HQ | +0.220 |
enta | +0.220 |
Neg logit contributions | |
Crimes | -0.264 |
certific | -0.252 |
witz | -0.252 |
Gutenberg | -0.250 |
unlaw | -0.239 |
Token | @ |
Token ID | 2488 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
Crimes | +1.720 |
certific | +1.623 |
witz | +1.603 |
Gutenberg | +1.591 |
Vienna | +1.529 |
Neg logit contributions | |
workers | -1.918 |
trainers | -1.858 |
tubing | -1.752 |
HQ | -1.729 |
enta | -1.704 |
Token | / |
Token ID | 14 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
workers | +1.412 |
trainers | +1.376 |
tubing | +1.285 |
HQ | +1.268 |
enta | +1.250 |
Neg logit contributions | |
Crimes | -1.417 |
certific | -1.361 |
witz | -1.339 |
Gutenberg | -1.333 |
unlaw | -1.287 |
Token | OR |
Token ID | 1581 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
workers | +2.402 |
trainers | +2.326 |
HQ | +2.144 |
tubing | +2.120 |
enta | +2.066 |
Neg logit contributions | |
Crimes | -3.289 |
certific | -3.215 |
witz | -3.132 |
Gutenberg | -3.124 |
unlaw | -3.049 |
Token | G |
Token ID | 38 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
Crimes | +0.249 |
witz | +0.236 |
certific | +0.235 |
Gutenberg | +0.231 |
Vienna | +0.223 |
Neg logit contributions | |
workers | -0.214 |
trainers | -0.208 |
tubing | -0.193 |
HQ | -0.188 |
enta | -0.188 |
Token | released |
Token ID | 2716 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
Crimes | +1.173 |
certific | +1.107 |
witz | +1.094 |
Gutenberg | +1.089 |
Vienna | +1.054 |
Neg logit contributions | |
workers | -1.118 |
trainers | -1.090 |
tubing | -1.015 |
HQ | -0.984 |
enta | -0.981 |
Token | a |
Token ID | 257 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
workers | +0.684 |
trainers | +0.666 |
tubing | +0.626 |
HQ | +0.616 |
enta | +0.610 |
Neg logit contributions | |
Crimes | -0.604 |
certific | -0.575 |
witz | -0.569 |
Gutenberg | -0.567 |
unlaw | -0.543 |
Token | poll |
Token ID | 3278 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
Crimes | +0.942 |
certific | +0.902 |
witz | +0.892 |
Gutenberg | +0.891 |
Vienna | +0.851 |
Neg logit contributions | |
workers | -0.930 |
trainers | -0.901 |
tubing | -0.841 |
HQ | -0.830 |
enta | -0.826 |
Token | showing |
Token ID | 4478 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
workers | +4.741 |
trainers | +4.612 |
tubing | +4.297 |
HQ | +4.205 |
agen | +4.151 |
Neg logit contributions | |
Crimes | -5.784 |
certific | -5.613 |
Gutenberg | -5.521 |
witz | -5.416 |
unlaw | -5.382 |
Token | that |
Token ID | 326 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
workers | +1.488 |
trainers | +1.448 |
tubing | +1.344 |
HQ | +1.324 |
enta | +1.304 |
Neg logit contributions | |
Crimes | -1.722 |
certific | -1.660 |
witz | -1.634 |
Gutenberg | -1.625 |
unlaw | -1.576 |
Token | 24 |
Token ID | 1987 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
workers | +0.344 |
trainers | +0.334 |
tubing | +0.312 |
HQ | +0.308 |
enta | +0.304 |
Neg logit contributions | |
Crimes | -0.368 |
certific | -0.354 |
witz | -0.348 |
Gutenberg | -0.348 |
unlaw | -0.335 |
Token | percent |
Token ID | 1411 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
workers | +1.235 |
trainers | +1.208 |
tubing | +1.140 |
HQ | +1.127 |
enta | +1.114 |
Neg logit contributions | |
Crimes | -0.882 |
certific | -0.840 |
witz | -0.823 |
Gutenberg | -0.819 |
unlaw | -0.785 |
Token | of |
Token ID | 286 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
workers | +1.260 |
trainers | +1.231 |
tubing | +1.158 |
HQ | +1.144 |
enta | +1.130 |
Neg logit contributions | |
Crimes | -1.013 |
certific | -0.968 |
witz | -0.950 |
Gutenberg | -0.945 |
unlaw | -0.908 |
Token | NS |
Token ID | 8035 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Crimes | +2.062 |
certific | +1.916 |
witz | +1.911 |
Gutenberg | +1.859 |
unlaw | +1.794 |
Neg logit contributions | |
workers | -2.834 |
trainers | -2.779 |
tubing | -2.623 |
HQ | -2.537 |
enta | -2.519 |
Token | via |
Token ID | 2884 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
Crimes | +2.624 |
Gutenberg | +2.437 |
certific | +2.428 |
witz | +2.401 |
Vienna | +2.308 |
Neg logit contributions | |
workers | -3.301 |
trainers | -3.243 |
tubing | -3.051 |
HQ | -2.940 |
airs | -2.936 |
Token | Getty |
Token ID | 7136 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
Crimes | +0.705 |
certific | +0.667 |
witz | +0.653 |
Gutenberg | +0.649 |
unlaw | +0.617 |
Neg logit contributions | |
workers | -1.140 |
trainers | -1.116 |
tubing | -1.056 |
HQ | -1.044 |
enta | -1.033 |
Token | Images |
Token ID | 5382 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
workers | +3.814 |
trainers | +3.692 |
HQ | +3.431 |
tubing | +3.370 |
ox | +3.313 |
Neg logit contributions | |
Crimes | -5.012 |
certific | -4.980 |
witz | -4.810 |
Gutenberg | -4.785 |
unlaw | -4.765 |
Token | ) |
Token ID | 8 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
workers | +2.506 |
trainers | +2.460 |
HQ | +2.295 |
tubing | +2.283 |
enta | +2.251 |
Neg logit contributions | |
Crimes | -2.353 |
certific | -2.292 |
witz | -2.224 |
Gutenberg | -2.203 |
unlaw | -2.169 |
Token | Indiana |
Token ID | 10278 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
workers | +1.499 |
trainers | +1.462 |
tubing | +1.365 |
HQ | +1.351 |
enta | +1.324 |
Neg logit contributions | |
Crimes | -1.473 |
certific | -1.436 |
witz | -1.401 |
Gutenberg | -1.386 |
unlaw | -1.354 |
Token | Gov |
Token ID | 10964 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
workers | +3.252 |
trainers | +3.191 |
tubing | +2.822 |
enta | +2.679 |
HQ | +2.667 |
Neg logit contributions | |
Crimes | -7.178 |
certific | -7.000 |
witz | -6.953 |
Gutenberg | -6.942 |
unlaw | -6.895 |
Token | . |
Token ID | 13 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
workers | +4.002 |
trainers | +3.963 |
tubing | +3.687 |
ox | +3.651 |
enta | +3.648 |
Neg logit contributions | |
Crimes | -4.117 |
certific | -4.112 |
witz | -3.984 |
Gutenberg | -3.969 |
unlaw | -3.871 |
Token | Mike |
Token ID | 4995 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
workers | +1.577 |
trainers | +1.533 |
tubing | +1.429 |
HQ | +1.410 |
enta | +1.389 |
Neg logit contributions | |
Crimes | -1.693 |
certific | -1.622 |
witz | -1.596 |
Gutenberg | -1.590 |
unlaw | -1.536 |
Token | Pence |
Token ID | 17339 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Crimes | +1.697 |
certific | +1.621 |
witz | +1.592 |
Gutenberg | +1.585 |
unlaw | +1.524 |
Neg logit contributions | |
workers | -2.017 |
trainers | -1.970 |
tubing | -1.851 |
HQ | -1.828 |
enta | -1.805 |
Token | Pence |
Token ID | 17339 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
workers | +2.854 |
trainers | +2.832 |
tubing | +2.620 |
HQ | +2.571 |
enta | +2.539 |
Neg logit contributions | |
Crimes | -2.819 |
certific | -2.747 |
witz | -2.681 |
Gutenberg | -2.672 |
unlaw | -2.636 |
Token | the |
Token ID | 262 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
workers | +1.425 |
trainers | +1.389 |
tubing | +1.289 |
HQ | +1.266 |
enta | +1.249 |
Neg logit contributions | |
Crimes | -1.592 |
certific | -1.532 |
witz | -1.512 |
Gutenberg | -1.500 |
unlaw | -1.452 |
Token | white |
Token ID | 2330 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
workers | +0.399 |
trainers | +0.387 |
tubing | +0.357 |
HQ | +0.351 |
enta | +0.345 |
Neg logit contributions | |
Crimes | -0.536 |
certific | -0.518 |
witz | -0.510 |
Gutenberg | -0.508 |
unlaw | -0.493 |
Token | minority |
Token ID | 9137 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
workers | +0.779 |
trainers | +0.759 |
tubing | +0.707 |
HQ | +0.697 |
enta | +0.687 |
Neg logit contributions | |
Crimes | -0.810 |
certific | -0.778 |
witz | -0.766 |
Gutenberg | -0.762 |
unlaw | -0.737 |
Token | would |
Token ID | 561 |
Feature activation | -8.87e-01 |
Pos logit contributions | |
Crimes | +0.154 |
certific | +0.149 |
Gutenberg | +0.146 |
witz | +0.146 |
Vienna | +0.141 |
Neg logit contributions | |
workers | -0.155 |
trainers | -0.150 |
tubing | -0.141 |
HQ | -0.140 |
enta | -0.139 |
Token | be |
Token ID | 307 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
workers | +2.406 |
trainers | +2.355 |
tubing | +2.199 |
HQ | +2.130 |
enta | +2.120 |
Neg logit contributions | |
Crimes | -2.572 |
certific | -2.471 |
witz | -2.430 |
Gutenberg | -2.430 |
unlaw | -2.340 |
Token | crucial |
Token ID | 8780 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
workers | +0.651 |
trainers | +0.629 |
tubing | +0.592 |
HQ | +0.587 |
enta | +0.577 |
Neg logit contributions | |
Crimes | -0.650 |
certific | -0.627 |
witz | -0.613 |
Gutenberg | -0.612 |
Vienna | -0.591 |
Token | to |
Token ID | 284 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
workers | +4.427 |
trainers | +4.361 |
tubing | +3.992 |
HQ | +3.848 |
ox | +3.798 |
Neg logit contributions | |
Crimes | -6.090 |
certific | -5.953 |
witz | -5.801 |
Gutenberg | -5.770 |
unlaw | -5.652 |
Token | securing |
Token ID | 19732 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
workers | +3.270 |
trainers | +3.210 |
tubing | +2.960 |
HQ | +2.900 |
enta | +2.887 |
Neg logit contributions | |
Crimes | -3.659 |
certific | -3.538 |
witz | -3.511 |
Gutenberg | -3.413 |
unlaw | -3.342 |
Token | Western |
Token ID | 4885 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
workers | +1.875 |
trainers | +1.830 |
tubing | +1.702 |
HQ | +1.673 |
enta | +1.650 |
Neg logit contributions | |
Crimes | -1.989 |
certific | -1.912 |
witz | -1.891 |
Gutenberg | -1.870 |
unlaw | -1.811 |
Token | investment |
Token ID | 4896 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
Crimes | +0.408 |
certific | +0.392 |
Gutenberg | +0.384 |
witz | +0.381 |
Vienna | +0.368 |
Neg logit contributions | |
workers | -0.434 |
trainers | -0.420 |
tubing | -0.397 |
HQ | -0.393 |
enta | -0.386 |
Token | in |
Token ID | 287 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Crimes | +1.303 |
certific | +1.248 |
Gutenberg | +1.220 |
witz | +1.219 |
unlaw | +1.179 |
Neg logit contributions | |
workers | -1.359 |
trainers | -1.317 |
tubing | -1.235 |
HQ | -1.235 |
enta | -1.220 |
Token | 1 |
Token ID | 16 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
workers | +1.082 |
trainers | +1.048 |
tubing | +0.968 |
HQ | +0.954 |
enta | +0.939 |
Neg logit contributions | |
Crimes | -1.522 |
certific | -1.473 |
witz | -1.448 |
Gutenberg | -1.441 |
Vienna | -1.396 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
workers | +2.347 |
trainers | +2.293 |
tubing | +2.123 |
HQ | +2.102 |
enta | +2.089 |
Neg logit contributions | |
Crimes | -2.159 |
certific | -2.081 |
Gutenberg | -2.052 |
witz | -2.051 |
unlaw | -1.992 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
Crimes | +0.388 |
certific | +0.365 |
witz | +0.364 |
Gutenberg | +0.362 |
Vienna | +0.347 |
Neg logit contributions | |
workers | -0.439 |
trainers | -0.429 |
tubing | -0.404 |
HQ | -0.395 |
enta | -0.392 |
Token | A |
Token ID | 32 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
workers | +0.257 |
trainers | +0.253 |
tubing | +0.236 |
HQ | +0.229 |
enta | +0.228 |
Neg logit contributions | |
Crimes | -0.268 |
certific | -0.257 |
witz | -0.255 |
Gutenberg | -0.253 |
unlaw | -0.243 |
Token | new |
Token ID | 649 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
Crimes | +2.225 |
certific | +2.155 |
witz | +2.109 |
Gutenberg | +2.104 |
Vienna | +2.031 |
Neg logit contributions | |
workers | -2.075 |
trainers | -2.028 |
tubing | -1.877 |
HQ | -1.857 |
enta | -1.830 |
Token | survey |
Token ID | 5526 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
workers | +2.040 |
trainers | +2.007 |
tubing | +1.884 |
HQ | +1.851 |
enta | +1.825 |
Neg logit contributions | |
Crimes | -1.785 |
certific | -1.719 |
witz | -1.687 |
Gutenberg | -1.672 |
unlaw | -1.623 |
Token | looked |
Token ID | 3114 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
workers | +5.020 |
trainers | +4.924 |
tubing | +4.557 |
HQ | +4.413 |
ox | +4.383 |
Neg logit contributions | |
Crimes | -5.362 |
certific | -5.221 |
Gutenberg | -5.110 |
witz | -5.053 |
unlaw | -4.961 |
Token | into |
Token ID | 656 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
workers | +1.138 |
trainers | +1.107 |
tubing | +1.030 |
HQ | +1.015 |
enta | +1.002 |
Neg logit contributions | |
Crimes | -1.251 |
certific | -1.202 |
witz | -1.184 |
Gutenberg | -1.179 |
unlaw | -1.140 |
Token | the |
Token ID | 262 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
Crimes | +0.542 |
certific | +0.520 |
Gutenberg | +0.511 |
witz | +0.508 |
unlaw | +0.492 |
Neg logit contributions | |
workers | -0.528 |
trainers | -0.507 |
tubing | -0.479 |
HQ | -0.476 |
enta | -0.472 |
Token | nation |
Token ID | 3277 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
workers | +0.960 |
trainers | +0.935 |
tubing | +0.875 |
HQ | +0.865 |
enta | +0.853 |
Neg logit contributions | |
Crimes | -0.910 |
certific | -0.873 |
witz | -0.857 |
Gutenberg | -0.855 |
unlaw | -0.824 |
Token | � |
Token ID | 447 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
workers | +0.647 |
trainers | +0.628 |
tubing | +0.583 |
HQ | +0.576 |
enta | +0.565 |
Neg logit contributions | |
Crimes | -0.775 |
certific | -0.746 |
witz | -0.737 |
Gutenberg | -0.733 |
unlaw | -0.707 |
Token | , |
Token ID | 11 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
workers | +0.737 |
trainers | +0.717 |
tubing | +0.671 |
HQ | +0.662 |
enta | +0.653 |
Neg logit contributions | |
Crimes | -0.723 |
certific | -0.694 |
witz | -0.683 |
Gutenberg | -0.680 |
unlaw | -0.656 |
Token | he |
Token ID | 339 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
Crimes | +1.137 |
certific | +1.081 |
Gutenberg | +1.072 |
witz | +1.071 |
Vienna | +1.022 |
Neg logit contributions | |
workers | -1.089 |
trainers | -1.055 |
tubing | -0.992 |
HQ | -0.977 |
enta | -0.965 |
Token | said |
Token ID | 531 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
Crimes | +0.098 |
certific | +0.094 |
witz | +0.093 |
Gutenberg | +0.092 |
unlaw | +0.088 |
Neg logit contributions | |
workers | -0.099 |
trainers | -0.096 |
tubing | -0.090 |
HQ | -0.090 |
enta | -0.088 |
Token | in |
Token ID | 287 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
workers | +1.177 |
trainers | +1.146 |
tubing | +1.067 |
HQ | +1.052 |
enta | +1.037 |
Neg logit contributions | |
Crimes | -1.274 |
certific | -1.226 |
witz | -1.207 |
Gutenberg | -1.202 |
unlaw | -1.162 |
Token | an |
Token ID | 281 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
workers | +0.317 |
trainers | +0.309 |
tubing | +0.290 |
HQ | +0.286 |
enta | +0.282 |
Neg logit contributions | |
Crimes | -0.283 |
certific | -0.271 |
witz | -0.266 |
Gutenberg | -0.265 |
unlaw | -0.256 |
Token | interview |
Token ID | 2720 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
workers | +0.527 |
trainers | +0.514 |
tubing | +0.477 |
HQ | +0.471 |
enta | +0.463 |
Neg logit contributions | |
Crimes | -0.593 |
certific | -0.572 |
witz | -0.563 |
Gutenberg | -0.559 |
unlaw | -0.543 |
Token | with |
Token ID | 351 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
workers | +4.682 |
trainers | +4.544 |
tubing | +4.188 |
ox | +4.132 |
HQ | +4.097 |
Neg logit contributions | |
Crimes | -5.709 |
certific | -5.573 |
witz | -5.399 |
unlaw | -5.370 |
Gutenberg | -5.341 |
Token | the |
Token ID | 262 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
workers | +1.831 |
trainers | +1.792 |
tubing | +1.673 |
HQ | +1.653 |
enta | +1.635 |
Neg logit contributions | |
Crimes | -1.647 |
certific | -1.598 |
witz | -1.565 |
Gutenberg | -1.559 |
unlaw | -1.509 |
Token | Daily |
Token ID | 6714 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
workers | +1.847 |
trainers | +1.811 |
tubing | +1.701 |
HQ | +1.685 |
enta | +1.667 |
Neg logit contributions | |
Crimes | -1.388 |
certific | -1.334 |
witz | -1.307 |
Gutenberg | -1.293 |
unlaw | -1.251 |
Token | in |
Token ID | 287 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
Crimes | +0.136 |
certific | +0.130 |
witz | +0.127 |
Gutenberg | +0.127 |
unlaw | +0.122 |
Neg logit contributions | |
workers | -0.176 |
trainers | -0.172 |
tubing | -0.162 |
HQ | -0.160 |
enta | -0.158 |
Token | November |
Token ID | 3389 |
Feature activation | +7.32e-03 |
Pos logit contributions | |
workers | +0.608 |
trainers | +0.593 |
tubing | +0.554 |
HQ | +0.547 |
enta | +0.540 |
Neg logit contributions | |
Crimes | -0.580 |
certific | -0.557 |
witz | -0.547 |
Gutenberg | -0.545 |
unlaw | -0.526 |
Token | ati |
Token ID | 7246 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
workers | +0.226 |
trainers | +0.218 |
tubing | +0.199 |
HQ | +0.196 |
enta | +0.190 |
Neg logit contributions | |
Crimes | -0.373 |
certific | -0.361 |
witz | -0.357 |
Gutenberg | -0.355 |
unlaw | -0.345 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
workers | +0.828 |
trainers | +0.808 |
tubing | +0.757 |
HQ | +0.746 |
enta | +0.736 |
Neg logit contributions | |
Crimes | -0.773 |
certific | -0.744 |
witz | -0.731 |
Gutenberg | -0.727 |
unlaw | -0.702 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
Crimes | +0.314 |
certific | +0.297 |
witz | +0.294 |
Gutenberg | +0.294 |
unlaw | +0.280 |
Neg logit contributions | |
workers | -0.353 |
trainers | -0.345 |
tubing | -0.324 |
HQ | -0.318 |
enta | -0.315 |
Token | In |
Token ID | 818 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
workers | +0.526 |
trainers | +0.515 |
tubing | +0.481 |
HQ | +0.470 |
enta | +0.466 |
Neg logit contributions | |
Crimes | -0.549 |
certific | -0.527 |
witz | -0.521 |
Gutenberg | -0.518 |
unlaw | -0.498 |
Token | an |
Token ID | 281 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
workers | +0.037 |
trainers | +0.036 |
tubing | +0.034 |
HQ | +0.033 |
enta | +0.033 |
Neg logit contributions | |
Crimes | -0.036 |
certific | -0.035 |
witz | -0.034 |
Gutenberg | -0.034 |
Vienna | -0.033 |
Token | interview |
Token ID | 2720 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
workers | +0.064 |
trainers | +0.062 |
tubing | +0.058 |
HQ | +0.057 |
enta | +0.056 |
Neg logit contributions | |
Crimes | -0.062 |
certific | -0.060 |
witz | -0.059 |
Gutenberg | -0.059 |
unlaw | -0.056 |
Token | over |
Token ID | 625 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
workers | +4.437 |
trainers | +4.335 |
tubing | +4.009 |
ox | +3.890 |
HQ | +3.842 |
Neg logit contributions | |
Crimes | -5.854 |
certific | -5.692 |
witz | -5.539 |
unlaw | -5.535 |
Gutenberg | -5.495 |
Token | the |
Token ID | 262 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
workers | +1.506 |
trainers | +1.467 |
tubing | +1.368 |
HQ | +1.349 |
enta | +1.326 |
Neg logit contributions | |
Crimes | -1.570 |
certific | -1.513 |
witz | -1.488 |
Gutenberg | -1.480 |
unlaw | -1.433 |
Token | weekend |
Token ID | 5041 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
workers | +1.887 |
trainers | +1.847 |
tubing | +1.717 |
HQ | +1.675 |
enta | +1.649 |
Neg logit contributions | |
Crimes | -2.222 |
certific | -2.157 |
witz | -2.125 |
Gutenberg | -2.101 |
unlaw | -2.036 |
Token | with |
Token ID | 351 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
Crimes | +0.007 |
certific | +0.007 |
witz | +0.007 |
Gutenberg | +0.007 |
unlaw | +0.006 |
Neg logit contributions | |
workers | -0.007 |
trainers | -0.007 |
tubing | -0.006 |
HQ | -0.006 |
enta | -0.006 |
Token | Sal |
Token ID | 4849 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
workers | +2.772 |
trainers | +2.726 |
tubing | +2.543 |
HQ | +2.499 |
enta | +2.472 |
Neg logit contributions | |
Crimes | -2.503 |
certific | -2.437 |
witz | -2.383 |
Gutenberg | -2.381 |
unlaw | -2.304 |
Token | they |
Token ID | 484 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
workers | +0.840 |
trainers | +0.815 |
tubing | +0.760 |
HQ | +0.752 |
enta | +0.741 |
Neg logit contributions | |
Crimes | -0.956 |
certific | -0.921 |
witz | -0.905 |
Gutenberg | -0.904 |
unlaw | -0.874 |
Token | still |
Token ID | 991 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
workers | +0.541 |
trainers | +0.527 |
tubing | +0.495 |
HQ | +0.491 |
enta | +0.484 |
Neg logit contributions | |
Crimes | -0.452 |
certific | -0.434 |
witz | -0.425 |
Gutenberg | -0.421 |
unlaw | -0.405 |
Token | believed |
Token ID | 4762 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
workers | +1.835 |
trainers | +1.791 |
tubing | +1.681 |
HQ | +1.661 |
enta | +1.639 |
Neg logit contributions | |
Crimes | -1.583 |
certific | -1.517 |
witz | -1.489 |
Gutenberg | -1.482 |
unlaw | -1.427 |
Token | it |
Token ID | 340 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
workers | +1.438 |
trainers | +1.400 |
tubing | +1.303 |
HQ | +1.285 |
enta | +1.266 |
Neg logit contributions | |
Crimes | -1.568 |
certific | -1.509 |
witz | -1.485 |
Gutenberg | -1.478 |
unlaw | -1.430 |
Token | was |
Token ID | 373 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
Crimes | +0.139 |
certific | +0.133 |
witz | +0.132 |
Gutenberg | +0.130 |
Vienna | +0.125 |
Neg logit contributions | |
workers | -0.146 |
trainers | -0.142 |
tubing | -0.133 |
HQ | -0.131 |
enta | -0.131 |
Token | crucial |
Token ID | 8780 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
workers | +1.097 |
trainers | +1.068 |
tubing | +0.997 |
HQ | +0.986 |
enta | +0.975 |
Neg logit contributions | |
Crimes | -1.099 |
certific | -1.055 |
witz | -1.037 |
Gutenberg | -1.034 |
unlaw | -0.999 |
Token | that |
Token ID | 326 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
workers | +4.436 |
trainers | +4.317 |
tubing | +3.985 |
HQ | +3.918 |
ox | +3.868 |
Neg logit contributions | |
Crimes | -5.922 |
certific | -5.783 |
witz | -5.652 |
Gutenberg | -5.644 |
unlaw | -5.565 |
Token | Democrats |
Token ID | 4956 |
Feature activation | -4.13e-02 |
Pos logit contributions | |
workers | +0.749 |
trainers | +0.728 |
tubing | +0.677 |
HQ | +0.668 |
enta | +0.658 |
Neg logit contributions | |
Crimes | -0.863 |
certific | -0.832 |
witz | -0.817 |
Gutenberg | -0.816 |
unlaw | -0.788 |
Token | pass |
Token ID | 1208 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
workers | +0.114 |
trainers | +0.111 |
tubing | +0.104 |
HQ | +0.102 |
enta | +0.101 |
Neg logit contributions | |
Crimes | -0.118 |
certific | -0.113 |
witz | -0.111 |
Gutenberg | -0.111 |
unlaw | -0.107 |
Token | a |
Token ID | 257 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
Crimes | +1.562 |
certific | +1.497 |
Gutenberg | +1.462 |
witz | +1.458 |
Vienna | +1.411 |
Neg logit contributions | |
workers | -1.730 |
trainers | -1.690 |
tubing | -1.594 |
HQ | -1.575 |
enta | -1.571 |
Token | plan |
Token ID | 1410 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Crimes | +0.574 |
certific | +0.546 |
witz | +0.533 |
Gutenberg | +0.532 |
Vienna | +0.511 |
Neg logit contributions | |
workers | -0.801 |
trainers | -0.786 |
tubing | -0.742 |
HQ | -0.735 |
enta | -0.730 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
Crimes | +0.566 |
certific | +0.539 |
witz | +0.532 |
Gutenberg | +0.530 |
unlaw | +0.508 |
Neg logit contributions | |
workers | -0.603 |
trainers | -0.588 |
tubing | -0.554 |
HQ | -0.548 |
enta | -0.541 |
Token | The |
Token ID | 383 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
Crimes | +0.372 |
certific | +0.349 |
Gutenberg | +0.347 |
witz | +0.346 |
unlaw | +0.330 |
Neg logit contributions | |
workers | -0.442 |
trainers | -0.433 |
tubing | -0.408 |
HQ | -0.401 |
enta | -0.397 |
Token | notion |
Token ID | 9495 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
Crimes | +0.890 |
certific | +0.851 |
witz | +0.836 |
Gutenberg | +0.836 |
unlaw | +0.805 |
Neg logit contributions | |
workers | -0.963 |
trainers | -0.937 |
tubing | -0.880 |
HQ | -0.872 |
enta | -0.859 |
Token | of |
Token ID | 286 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
Crimes | +0.665 |
certific | +0.634 |
witz | +0.627 |
Gutenberg | +0.623 |
unlaw | +0.597 |
Neg logit contributions | |
workers | -0.745 |
trainers | -0.726 |
tubing | -0.683 |
HQ | -0.678 |
enta | -0.666 |
Token | the |
Token ID | 262 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
workers | +1.080 |
trainers | +1.056 |
tubing | +0.985 |
HQ | +0.967 |
enta | +0.956 |
Neg logit contributions | |
Crimes | -1.088 |
certific | -1.047 |
witz | -1.029 |
Gutenberg | -1.022 |
unlaw | -0.986 |
Token | useful |
Token ID | 4465 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
workers | +0.927 |
trainers | +0.906 |
tubing | +0.846 |
HQ | +0.833 |
enta | +0.823 |
Neg logit contributions | |
Crimes | -0.900 |
certific | -0.865 |
witz | -0.851 |
Gutenberg | -0.845 |
unlaw | -0.815 |
Token | as |
Token ID | 355 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
workers | +5.193 |
trainers | +5.074 |
tubing | +4.725 |
HQ | +4.644 |
agen | +4.578 |
Neg logit contributions | |
Crimes | -5.176 |
certific | -4.996 |
witz | -4.893 |
Gutenberg | -4.835 |
unlaw | -4.678 |
Token | good |
Token ID | 922 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
workers | +0.296 |
trainers | +0.288 |
tubing | +0.270 |
HQ | +0.266 |
enta | +0.262 |
Neg logit contributions | |
Crimes | -0.290 |
certific | -0.279 |
witz | -0.274 |
Gutenberg | -0.274 |
unlaw | -0.264 |
Token | - |
Token ID | 12 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
workers | +2.717 |
trainers | +2.659 |
tubing | +2.465 |
HQ | +2.420 |
enta | +2.394 |
Neg logit contributions | |
Crimes | -2.583 |
certific | -2.499 |
Gutenberg | -2.446 |
witz | -2.443 |
unlaw | -2.362 |
Token | for |
Token ID | 1640 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
Crimes | +1.064 |
certific | +1.020 |
witz | +1.007 |
Gutenberg | +0.999 |
unlaw | +0.966 |
Neg logit contributions | |
workers | -0.879 |
trainers | -0.871 |
tubing | -0.812 |
HQ | -0.796 |
enta | -0.787 |
Token | - |
Token ID | 12 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
workers | +1.386 |
trainers | +1.356 |
tubing | +1.278 |
HQ | +1.259 |
enta | +1.248 |
Neg logit contributions | |
Crimes | -0.994 |
certific | -0.950 |
witz | -0.930 |
Gutenberg | -0.924 |
unlaw | -0.884 |
Token |
|
Token ID | 198 |
Feature activation | -9.04e-02 |
Pos logit contributions | |
Crimes | +0.215 |
certific | +0.206 |
witz | +0.202 |
Gutenberg | +0.201 |
unlaw | +0.194 |
Neg logit contributions | |
workers | -0.244 |
trainers | -0.239 |
tubing | -0.224 |
HQ | -0.221 |
enta | -0.218 |
Token | In |
Token ID | 818 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
workers | +0.242 |
trainers | +0.239 |
tubing | +0.224 |
enta | +0.214 |
HQ | +0.213 |
Neg logit contributions | |
Crimes | -0.263 |
witz | -0.251 |
certific | -0.250 |
Gutenberg | -0.249 |
unlaw | -0.237 |
Token | July |
Token ID | 2901 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
Crimes | +0.357 |
certific | +0.339 |
witz | +0.337 |
Gutenberg | +0.336 |
Vienna | +0.322 |
Neg logit contributions | |
workers | -0.364 |
trainers | -0.353 |
tubing | -0.331 |
HQ | -0.327 |
enta | -0.321 |
Token | 2005 |
Token ID | 5075 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
Crimes | +2.210 |
Gutenberg | +2.089 |
witz | +2.082 |
certific | +2.081 |
Vienna | +1.999 |
Neg logit contributions | |
workers | -2.503 |
trainers | -2.448 |
tubing | -2.290 |
HQ | -2.269 |
enta | -2.238 |
Token | , |
Token ID | 11 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
Crimes | +0.292 |
certific | +0.276 |
Gutenberg | +0.275 |
witz | +0.275 |
Vienna | +0.262 |
Neg logit contributions | |
workers | -0.313 |
trainers | -0.305 |
tubing | -0.287 |
HQ | -0.282 |
enta | -0.278 |
Token | Gallup |
Token ID | 29183 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Crimes | +0.730 |
certific | +0.692 |
Gutenberg | +0.692 |
witz | +0.687 |
Vienna | +0.658 |
Neg logit contributions | |
workers | -0.768 |
trainers | -0.747 |
tubing | -0.704 |
HQ | -0.692 |
enta | -0.687 |
Token | found |
Token ID | 1043 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
workers | +5.107 |
trainers | +4.979 |
HQ | +4.665 |
tubing | +4.660 |
agen | +4.556 |
Neg logit contributions | |
Crimes | -5.138 |
certific | -4.975 |
witz | -4.918 |
Gutenberg | -4.889 |
unlaw | -4.774 |
Token | that |
Token ID | 326 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
Crimes | +0.115 |
certific | +0.109 |
Gutenberg | +0.109 |
witz | +0.109 |
Vienna | +0.104 |
Neg logit contributions | |
workers | -0.107 |
trainers | -0.104 |
tubing | -0.098 |
HQ | -0.096 |
enta | -0.095 |
Token | 25 |
Token ID | 1679 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Crimes | +0.368 |
certific | +0.352 |
Gutenberg | +0.350 |
witz | +0.347 |
Vienna | +0.333 |
Neg logit contributions | |
workers | -0.371 |
trainers | -0.359 |
tubing | -0.337 |
HQ | -0.333 |
enta | -0.330 |
Token | percent |
Token ID | 1411 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
workers | +0.703 |
trainers | +0.687 |
tubing | +0.649 |
HQ | +0.642 |
enta | +0.634 |
Neg logit contributions | |
Crimes | -0.487 |
certific | -0.463 |
witz | -0.454 |
Gutenberg | -0.451 |
unlaw | -0.432 |
Token | of |
Token ID | 286 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
workers | +0.555 |
trainers | +0.542 |
tubing | +0.510 |
HQ | +0.504 |
enta | +0.498 |
Neg logit contributions | |
Crimes | -0.439 |
certific | -0.419 |
witz | -0.412 |
Gutenberg | -0.409 |
unlaw | -0.393 |
Token | In |
Token ID | 818 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
Crimes | +0.886 |
witz | +0.847 |
Gutenberg | +0.836 |
certific | +0.836 |
Vienna | +0.796 |
Neg logit contributions | |
workers | -0.869 |
trainers | -0.865 |
tubing | -0.814 |
enta | -0.774 |
HQ | -0.765 |
Token | a |
Token ID | 257 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
Crimes | +0.784 |
certific | +0.744 |
witz | +0.736 |
Gutenberg | +0.729 |
Vienna | +0.705 |
Neg logit contributions | |
workers | -0.890 |
trainers | -0.863 |
tubing | -0.810 |
HQ | -0.808 |
enta | -0.791 |
Token | New |
Token ID | 968 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
Crimes | +1.138 |
certific | +1.088 |
witz | +1.076 |
Gutenberg | +1.075 |
Vienna | +1.030 |
Neg logit contributions | |
workers | -1.133 |
trainers | -1.097 |
tubing | -1.026 |
HQ | -1.015 |
enta | -1.003 |
Token | York |
Token ID | 1971 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
workers | +0.494 |
trainers | +0.484 |
tubing | +0.454 |
HQ | +0.445 |
enta | +0.440 |
Neg logit contributions | |
Crimes | -0.463 |
certific | -0.440 |
witz | -0.435 |
Gutenberg | -0.433 |
unlaw | -0.415 |
Token | State |
Token ID | 1812 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
Crimes | +1.185 |
certific | +1.131 |
witz | +1.109 |
Gutenberg | +1.105 |
Vienna | +1.052 |
Neg logit contributions | |
workers | -1.416 |
trainers | -1.375 |
tubing | -1.292 |
HQ | -1.275 |
enta | -1.261 |
Token | poll |
Token ID | 3278 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
workers | +0.376 |
trainers | +0.366 |
tubing | +0.342 |
HQ | +0.337 |
enta | +0.333 |
Neg logit contributions | |
Crimes | -0.376 |
certific | -0.360 |
witz | -0.354 |
Gutenberg | -0.353 |
unlaw | -0.340 |
Token | of |
Token ID | 286 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
workers | +5.399 |
trainers | +5.258 |
tubing | +4.904 |
HQ | +4.858 |
enta | +4.831 |
Neg logit contributions | |
Crimes | -4.922 |
certific | -4.751 |
witz | -4.638 |
Gutenberg | -4.633 |
unlaw | -4.562 |
Token | 1 |
Token ID | 352 |
Feature activation | -8.26e-01 |
Pos logit contributions | |
workers | +1.630 |
trainers | +1.589 |
tubing | +1.481 |
HQ | +1.459 |
enta | +1.439 |
Neg logit contributions | |
Crimes | -1.693 |
certific | -1.628 |
witz | -1.602 |
Gutenberg | -1.595 |
unlaw | -1.543 |
Token | , |
Token ID | 11 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
workers | +2.199 |
trainers | +2.137 |
tubing | +1.991 |
HQ | +1.975 |
ox | +1.936 |
Neg logit contributions | |
Crimes | -2.411 |
certific | -2.340 |
witz | -2.293 |
Gutenberg | -2.273 |
unlaw | -2.221 |
Token | 302 |
Token ID | 22709 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
workers | +1.826 |
trainers | +1.774 |
HQ | +1.651 |
tubing | +1.641 |
ox | +1.618 |
Neg logit contributions | |
Crimes | -1.919 |
certific | -1.882 |
witz | -1.825 |
Gutenberg | -1.806 |
unlaw | -1.778 |
Token | voters |
Token ID | 4446 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
workers | +0.602 |
trainers | +0.581 |
tubing | +0.539 |
HQ | +0.532 |
enta | +0.522 |
Neg logit contributions | |
Crimes | -0.767 |
certific | -0.743 |
witz | -0.727 |
Gutenberg | -0.726 |
Vienna | -0.700 |
Token | , |
Token ID | 11 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
workers | +0.306 |
trainers | +0.296 |
tubing | +0.279 |
HQ | +0.276 |
enta | +0.272 |
Neg logit contributions | |
Crimes | -0.289 |
certific | -0.278 |
witz | -0.273 |
Gutenberg | -0.271 |
unlaw | -0.262 |
Token | a |
Token ID | 257 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
Crimes | +0.641 |
certific | +0.620 |
witz | +0.606 |
Gutenberg | +0.606 |
Vienna | +0.585 |
Neg logit contributions | |
workers | -0.597 |
trainers | -0.568 |
tubing | -0.539 |
HQ | -0.536 |
enta | -0.527 |
Token | National |
Token ID | 2351 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
Crimes | +0.717 |
certific | +0.696 |
witz | +0.683 |
Gutenberg | +0.678 |
Vienna | +0.655 |
Neg logit contributions | |
workers | -0.681 |
trainers | -0.660 |
tubing | -0.615 |
HQ | -0.610 |
enta | -0.601 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Crimes | +1.237 |
certific | +1.151 |
Gutenberg | +1.133 |
witz | +1.121 |
Vienna | +1.089 |
Neg logit contributions | |
workers | -1.534 |
trainers | -1.486 |
tubing | -1.413 |
HQ | -1.392 |
enta | -1.379 |
Token | Foundation |
Token ID | 5693 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
workers | +2.205 |
trainers | +2.153 |
HQ | +2.038 |
tubing | +2.029 |
ox | +2.009 |
Neg logit contributions | |
Crimes | -1.137 |
certific | -1.081 |
witz | -1.059 |
Gutenberg | -1.041 |
unlaw | -1.000 |
Token | poll |
Token ID | 3278 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Crimes | +2.006 |
certific | +1.922 |
witz | +1.881 |
Gutenberg | +1.856 |
Vienna | +1.808 |
Neg logit contributions | |
workers | -2.507 |
trainers | -2.412 |
tubing | -2.298 |
HQ | -2.274 |
joining | -2.245 |
Token | found |
Token ID | 1043 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
workers | +4.814 |
trainers | +4.664 |
tubing | +4.298 |
agen | +4.210 |
HQ | +4.201 |
Neg logit contributions | |
Crimes | -5.492 |
certific | -5.353 |
Gutenberg | -5.239 |
unlaw | -5.149 |
witz | -5.137 |
Token | that |
Token ID | 326 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
workers | +1.187 |
trainers | +1.152 |
tubing | +1.071 |
HQ | +1.056 |
enta | +1.039 |
Neg logit contributions | |
Crimes | -1.418 |
certific | -1.367 |
witz | -1.346 |
Gutenberg | -1.341 |
unlaw | -1.299 |
Token | 59 |
Token ID | 7863 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
workers | +0.517 |
trainers | +0.501 |
tubing | +0.468 |
HQ | +0.462 |
enta | +0.455 |
Neg logit contributions | |
Crimes | -0.575 |
certific | -0.555 |
witz | -0.545 |
Gutenberg | -0.543 |
unlaw | -0.525 |
Token | percent |
Token ID | 1411 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
workers | +1.948 |
trainers | +1.913 |
tubing | +1.796 |
HQ | +1.774 |
enta | +1.756 |
Neg logit contributions | |
Crimes | -1.443 |
certific | -1.375 |
witz | -1.355 |
Gutenberg | -1.352 |
unlaw | -1.291 |
Token | of |
Token ID | 286 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
workers | +1.511 |
trainers | +1.480 |
tubing | +1.387 |
HQ | +1.368 |
enta | +1.354 |
Neg logit contributions | |
Crimes | -1.248 |
certific | -1.191 |
witz | -1.172 |
Gutenberg | -1.169 |
unlaw | -1.122 |
Token | to |
Token ID | 284 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
Crimes | +1.406 |
witz | +1.341 |
certific | +1.330 |
Gutenberg | +1.315 |
Vienna | +1.270 |
Neg logit contributions | |
workers | -1.314 |
trainers | -1.264 |
tubing | -1.186 |
HQ | -1.177 |
enta | -1.165 |
Token | win |
Token ID | 1592 |
Feature activation | -4.86e-03 |
Pos logit contributions | |
workers | +0.032 |
trainers | +0.031 |
tubing | +0.029 |
HQ | +0.029 |
enta | +0.028 |
Neg logit contributions | |
Crimes | -0.034 |
certific | -0.033 |
witz | -0.032 |
Gutenberg | -0.032 |
unlaw | -0.031 |
Token | their |
Token ID | 511 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
workers | +0.014 |
trainers | +0.013 |
tubing | +0.012 |
HQ | +0.012 |
enta | +0.012 |
Neg logit contributions | |
Crimes | -0.014 |
certific | -0.013 |
witz | -0.013 |
Gutenberg | -0.013 |
unlaw | -0.012 |
Token | fourth |
Token ID | 5544 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
workers | +1.177 |
trainers | +1.149 |
tubing | +1.064 |
HQ | +1.050 |
enta | +1.028 |
Neg logit contributions | |
Crimes | -1.336 |
certific | -1.290 |
witz | -1.267 |
Gutenberg | -1.264 |
unlaw | -1.226 |
Token | Stanley |
Token ID | 14167 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
Crimes | +1.118 |
certific | +1.068 |
Gutenberg | +1.054 |
witz | +1.048 |
Vienna | +1.009 |
Neg logit contributions | |
workers | -1.228 |
trainers | -1.190 |
tubing | -1.113 |
HQ | -1.102 |
enta | -1.099 |
Token | Cup |
Token ID | 5454 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Crimes | +3.304 |
witz | +3.053 |
certific | +3.039 |
Gutenberg | +3.016 |
Vienna | +2.913 |
Neg logit contributions | |
workers | -4.086 |
trainers | -3.987 |
tubing | -3.752 |
HQ | -3.729 |
enta | -3.708 |
Token | in |
Token ID | 287 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Crimes | +4.371 |
certific | +4.101 |
Gutenberg | +4.045 |
witz | +4.040 |
Vienna | +3.896 |
Neg logit contributions | |
workers | -5.043 |
trainers | -4.949 |
enta | -4.619 |
tubing | -4.564 |
HQ | -4.541 |
Token | seven |
Token ID | 3598 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
Crimes | +0.423 |
certific | +0.406 |
witz | +0.399 |
Gutenberg | +0.397 |
unlaw | +0.384 |
Neg logit contributions | |
workers | -0.438 |
trainers | -0.428 |
tubing | -0.400 |
HQ | -0.394 |
enta | -0.389 |
Token | seasons |
Token ID | 7028 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
Crimes | +0.747 |
certific | +0.723 |
witz | +0.711 |
Gutenberg | +0.708 |
unlaw | +0.688 |
Neg logit contributions | |
workers | -0.541 |
trainers | -0.523 |
tubing | -0.482 |
HQ | -0.476 |
enta | -0.469 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
Crimes | +0.927 |
certific | +0.878 |
witz | +0.867 |
Gutenberg | +0.858 |
Vienna | +0.823 |
Neg logit contributions | |
workers | -1.135 |
trainers | -1.106 |
tubing | -1.039 |
HQ | -1.032 |
enta | -1.019 |
Token |
|
Token ID | 198 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
workers | +0.128 |
trainers | +0.125 |
tubing | +0.118 |
HQ | +0.115 |
enta | +0.115 |
Neg logit contributions | |
Crimes | -0.109 |
certific | -0.103 |
witz | -0.102 |
Gutenberg | -0.102 |
Vienna | -0.097 |
Token | add |
Token ID | 751 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
Crimes | +1.107 |
certific | +1.065 |
witz | +1.036 |
Gutenberg | +1.034 |
unlaw | +0.994 |
Neg logit contributions | |
workers | -1.165 |
trainers | -1.136 |
tubing | -1.057 |
HQ | -1.055 |
enta | -1.038 |
Token | � |
Token ID | 564 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
Crimes | +1.159 |
certific | +1.106 |
witz | +1.082 |
Gutenberg | +1.082 |
unlaw | +1.044 |
Neg logit contributions | |
workers | -1.136 |
trainers | -1.105 |
tubing | -1.029 |
HQ | -1.021 |
enta | -1.013 |
Token | � |
Token ID | 250 |
Feature activation | -8.20e-02 |
Pos logit contributions | |
workers | +0.451 |
trainers | +0.439 |
tubing | +0.413 |
HQ | +0.408 |
enta | +0.403 |
Neg logit contributions | |
Crimes | -0.406 |
certific | -0.393 |
witz | -0.384 |
Gutenberg | -0.383 |
unlaw | -0.370 |
Token | IR |
Token ID | 4663 |
Feature activation | +8.28e-02 |
Pos logit contributions | |
workers | +0.227 |
trainers | +0.221 |
tubing | +0.206 |
HQ | +0.204 |
enta | +0.201 |
Neg logit contributions | |
Crimes | -0.234 |
certific | -0.225 |
witz | -0.221 |
Gutenberg | -0.220 |
unlaw | -0.212 |
Token | S |
Token ID | 50 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
Crimes | +0.282 |
certific | +0.274 |
witz | +0.270 |
Gutenberg | +0.269 |
unlaw | +0.262 |
Neg logit contributions | |
workers | -0.183 |
trainers | -0.176 |
tubing | -0.162 |
HQ | -0.160 |
enta | -0.156 |
Token | Telephone |
Token ID | 44735 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Crimes | +2.559 |
certific | +2.390 |
Gutenberg | +2.329 |
witz | +2.318 |
Vienna | +2.247 |
Neg logit contributions | |
workers | -2.773 |
trainers | -2.684 |
tubing | -2.530 |
enta | -2.508 |
HQ | -2.473 |
Token | Sc |
Token ID | 1446 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
Crimes | +4.731 |
certific | +4.266 |
Gutenberg | +4.211 |
witz | +4.207 |
Vienna | +4.055 |
Neg logit contributions | |
workers | -4.918 |
trainers | -4.871 |
tubing | -4.565 |
enta | -4.562 |
agen | -4.466 |
Token | am |
Token ID | 321 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
Crimes | +0.539 |
certific | +0.513 |
witz | +0.507 |
Gutenberg | +0.505 |
unlaw | +0.485 |
Neg logit contributions | |
workers | -0.597 |
trainers | -0.585 |
tubing | -0.550 |
HQ | -0.543 |
enta | -0.535 |
Token | � |
Token ID | 447 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
Crimes | +0.438 |
certific | +0.401 |
witz | +0.390 |
Gutenberg | +0.387 |
unlaw | +0.360 |
Neg logit contributions | |
workers | -1.153 |
trainers | -1.134 |
tubing | -1.083 |
HQ | -1.071 |
enta | -1.065 |
Token | � |
Token ID | 251 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
Crimes | +1.929 |
certific | +1.834 |
witz | +1.830 |
Gutenberg | +1.799 |
unlaw | +1.757 |
Neg logit contributions | |
workers | -1.522 |
trainers | -1.472 |
tubing | -1.342 |
HQ | -1.329 |
enta | -1.319 |
Token | in |
Token ID | 287 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
Crimes | +1.068 |
certific | +1.023 |
witz | +1.010 |
Gutenberg | +1.003 |
unlaw | +0.971 |
Neg logit contributions | |
workers | -0.892 |
trainers | -0.866 |
tubing | -0.801 |
HQ | -0.792 |
enta | -0.783 |
Token | a |
Token ID | 257 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
workers | +0.713 |
trainers | +0.694 |
tubing | +0.648 |
HQ | +0.639 |
enta | +0.630 |
Neg logit contributions | |
Crimes | -0.745 |
certific | -0.716 |
witz | -0.704 |
Gutenberg | -0.701 |
unlaw | -0.677 |
Token | professor |
Token ID | 6240 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
Crimes | +0.408 |
certific | +0.395 |
witz | +0.387 |
Gutenberg | +0.386 |
Vienna | +0.373 |
Neg logit contributions | |
workers | -0.385 |
trainers | -0.374 |
tubing | -0.348 |
HQ | -0.347 |
enta | -0.342 |
Token | at |
Token ID | 379 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
workers | +1.688 |
trainers | +1.646 |
tubing | +1.535 |
HQ | +1.511 |
enta | +1.492 |
Neg logit contributions | |
Crimes | -1.660 |
certific | -1.592 |
witz | -1.569 |
Gutenberg | -1.561 |
unlaw | -1.512 |
Token | the |
Token ID | 262 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
workers | +1.641 |
trainers | +1.604 |
tubing | +1.496 |
HQ | +1.475 |
enta | +1.455 |
Neg logit contributions | |
Crimes | -1.583 |
certific | -1.529 |
witz | -1.497 |
Gutenberg | -1.488 |
unlaw | -1.442 |
Token | Civil |
Token ID | 7511 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
workers | +1.551 |
trainers | +1.515 |
tubing | +1.418 |
HQ | +1.400 |
enta | +1.381 |
Neg logit contributions | |
Crimes | -1.401 |
certific | -1.347 |
witz | -1.322 |
Gutenberg | -1.315 |
unlaw | -1.271 |
Token | Aviation |
Token ID | 25186 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Crimes | +2.619 |
certific | +2.431 |
witz | +2.389 |
Gutenberg | +2.381 |
Vienna | +2.283 |
Neg logit contributions | |
workers | -3.842 |
trainers | -3.780 |
tubing | -3.590 |
HQ | -3.536 |
enta | -3.486 |
Token | University |
Token ID | 2059 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Crimes | +3.931 |
certific | +3.569 |
Gutenberg | +3.531 |
witz | +3.485 |
Vienna | +3.472 |
Neg logit contributions | |
workers | -5.401 |
trainers | -5.327 |
tubing | -5.101 |
HQ | -4.963 |
joining | -4.908 |
Token | of |
Token ID | 286 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
workers | +0.384 |
trainers | +0.375 |
tubing | +0.354 |
HQ | +0.349 |
enta | +0.344 |
Neg logit contributions | |
Crimes | -0.331 |
certific | -0.317 |
witz | -0.310 |
Gutenberg | -0.310 |
Vienna | -0.297 |
Token | China |
Token ID | 2807 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
workers | +2.186 |
trainers | +2.131 |
tubing | +1.993 |
HQ | +1.974 |
ox | +1.951 |
Neg logit contributions | |
Crimes | -1.914 |
certific | -1.864 |
witz | -1.818 |
Gutenberg | -1.801 |
unlaw | -1.755 |
Token | . |
Token ID | 13 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
workers | +1.750 |
trainers | +1.704 |
tubing | +1.589 |
HQ | +1.570 |
enta | +1.550 |
Neg logit contributions | |
Crimes | -1.709 |
certific | -1.647 |
witz | -1.618 |
Gutenberg | -1.611 |
unlaw | -1.564 |
Token | " |
Token ID | 366 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
workers | +2.154 |
trainers | +2.099 |
tubing | +1.962 |
HQ | +1.936 |
enta | +1.910 |
Neg logit contributions | |
Crimes | -2.095 |
certific | -2.016 |
witz | -1.980 |
Gutenberg | -1.970 |
unlaw | -1.904 |
Token | am |
Token ID | 321 |
Feature activation | +6.86e-02 |
Pos logit contributions | |
Crimes | +0.668 |
certific | +0.639 |
witz | +0.631 |
Gutenberg | +0.627 |
unlaw | +0.602 |
Neg logit contributions | |
workers | -0.627 |
trainers | -0.609 |
tubing | -0.567 |
HQ | -0.563 |
enta | -0.549 |
Token | ( |
Token ID | 357 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
Crimes | +0.186 |
certific | +0.177 |
witz | +0.174 |
Gutenberg | +0.174 |
unlaw | +0.166 |
Neg logit contributions | |
workers | -0.199 |
trainers | -0.195 |
tubing | -0.183 |
HQ | -0.181 |
enta | -0.178 |
Token | 2013 |
Token ID | 6390 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
workers | +0.121 |
trainers | +0.119 |
tubing | +0.110 |
HQ | +0.107 |
enta | +0.106 |
Neg logit contributions | |
Crimes | -0.155 |
certific | -0.148 |
witz | -0.147 |
Gutenberg | -0.146 |
unlaw | -0.141 |
Token | ) |
Token ID | 8 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
workers | +0.380 |
trainers | +0.370 |
tubing | +0.348 |
HQ | +0.343 |
enta | +0.340 |
Neg logit contributions | |
Crimes | -0.320 |
certific | -0.305 |
witz | -0.300 |
Gutenberg | -0.298 |
unlaw | -0.287 |
Token | was |
Token ID | 373 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
Crimes | +0.179 |
certific | +0.170 |
witz | +0.168 |
Gutenberg | +0.167 |
Vienna | +0.160 |
Neg logit contributions | |
workers | -0.194 |
trainers | -0.188 |
tubing | -0.177 |
HQ | -0.174 |
enta | -0.174 |
Token | similarly |
Token ID | 12470 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Crimes | +1.241 |
certific | +1.196 |
Gutenberg | +1.163 |
witz | +1.161 |
Vienna | +1.122 |
Neg logit contributions | |
workers | -1.260 |
trainers | -1.212 |
tubing | -1.135 |
HQ | -1.123 |
enta | -1.111 |
Token | sought |
Token ID | 7194 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
Crimes | +4.632 |
certific | +4.573 |
Gutenberg | +4.345 |
witz | +4.338 |
unlaw | +4.267 |
Neg logit contributions | |
workers | -4.812 |
trainers | -4.553 |
HQ | -4.302 |
airs | -4.236 |
enta | -4.223 |
Token | to |
Token ID | 284 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
Crimes | +1.340 |
certific | +1.274 |
witz | +1.259 |
Gutenberg | +1.253 |
unlaw | +1.207 |
Neg logit contributions | |
workers | -1.415 |
trainers | -1.369 |
tubing | -1.278 |
HQ | -1.264 |
enta | -1.259 |
Token | be |
Token ID | 307 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
Crimes | +1.584 |
certific | +1.500 |
witz | +1.473 |
Gutenberg | +1.466 |
unlaw | +1.421 |
Neg logit contributions | |
workers | -1.736 |
trainers | -1.667 |
tubing | -1.568 |
HQ | -1.564 |
enta | -1.552 |
Token | banned |
Token ID | 9301 |
Feature activation | +7.29e-02 |
Pos logit contributions | |
Crimes | +1.668 |
certific | +1.606 |
Gutenberg | +1.568 |
witz | +1.567 |
unlaw | +1.511 |
Neg logit contributions | |
workers | -1.659 |
trainers | -1.599 |
tubing | -1.497 |
HQ | -1.476 |
enta | -1.466 |
Token | by |
Token ID | 416 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
Crimes | +0.205 |
certific | +0.196 |
witz | +0.193 |
Gutenberg | +0.192 |
unlaw | +0.185 |
Neg logit contributions | |
workers | -0.205 |
trainers | -0.200 |
tubing | -0.187 |
HQ | -0.184 |
enta | -0.182 |
Token | scored |
Token ID | 7781 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
workers | +1.853 |
trainers | +1.827 |
tubing | +1.695 |
HQ | +1.663 |
enta | +1.642 |
Neg logit contributions | |
Crimes | -1.936 |
certific | -1.873 |
witz | -1.834 |
Gutenberg | -1.829 |
unlaw | -1.776 |
Token | his |
Token ID | 465 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
Crimes | +2.400 |
certific | +2.286 |
witz | +2.270 |
Gutenberg | +2.261 |
Vienna | +2.215 |
Neg logit contributions | |
workers | -2.368 |
trainers | -2.274 |
tubing | -2.160 |
HQ | -2.126 |
enta | -2.112 |
Token | 100 |
Token ID | 1802 |
Feature activation | +7.92e-01 |
Pos logit contributions | |
Crimes | +1.439 |
certific | +1.370 |
witz | +1.350 |
Gutenberg | +1.345 |
Vienna | +1.303 |
Neg logit contributions | |
workers | -1.436 |
trainers | -1.381 |
tubing | -1.307 |
HQ | -1.286 |
enta | -1.277 |
Token | th |
Token ID | 400 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
Crimes | +2.010 |
certific | +1.873 |
Gutenberg | +1.841 |
witz | +1.839 |
Vienna | +1.781 |
Neg logit contributions | |
workers | -2.434 |
trainers | -2.373 |
tubing | -2.253 |
HQ | -2.199 |
enta | -2.185 |
Token | Premier |
Token ID | 9952 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Crimes | +1.077 |
certific | +1.037 |
Gutenberg | +1.004 |
witz | +0.999 |
Vienna | +0.976 |
Neg logit contributions | |
workers | -1.200 |
trainers | -1.162 |
tubing | -1.096 |
HQ | -1.080 |
enta | -1.069 |
Token | League |
Token ID | 4041 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
Crimes | +3.067 |
Gutenberg | +2.889 |
certific | +2.819 |
witz | +2.795 |
Vienna | +2.761 |
Neg logit contributions | |
workers | -4.430 |
trainers | -4.319 |
tubing | -4.173 |
agen | -3.957 |
enta | -3.954 |
Token | goal |
Token ID | 3061 |
Feature activation | +9.29e-01 |
Pos logit contributions | |
Crimes | +4.076 |
certific | +3.904 |
Gutenberg | +3.842 |
Vienna | +3.688 |
witz | +3.611 |
Neg logit contributions | |
workers | -5.779 |
trainers | -5.582 |
tubing | -5.306 |
HQ | -5.298 |
enta | -5.234 |
Token | at |
Token ID | 379 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
Crimes | +2.548 |
certific | +2.407 |
witz | +2.401 |
Gutenberg | +2.389 |
Vienna | +2.322 |
Neg logit contributions | |
workers | -2.656 |
trainers | -2.585 |
tubing | -2.408 |
HQ | -2.399 |
enta | -2.349 |
Token | Newcastle |
Token ID | 22410 |
Feature activation | -8.56e-02 |
Pos logit contributions | |
Crimes | +1.368 |
certific | +1.293 |
witz | +1.292 |
Gutenberg | +1.281 |
Vienna | +1.244 |
Neg logit contributions | |
workers | -1.525 |
trainers | -1.467 |
tubing | -1.398 |
HQ | -1.377 |
enta | -1.350 |
Token | on |
Token ID | 319 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
workers | +0.238 |
trainers | +0.231 |
tubing | +0.217 |
HQ | +0.213 |
enta | +0.209 |
Neg logit contributions | |
Crimes | -0.243 |
certific | -0.233 |
witz | -0.229 |
Gutenberg | -0.229 |
unlaw | -0.221 |
Token | Saturday |
Token ID | 3909 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
Crimes | +0.043 |
certific | +0.041 |
witz | +0.041 |
Gutenberg | +0.040 |
unlaw | +0.039 |
Neg logit contributions | |
workers | -0.055 |
trainers | -0.054 |
tubing | -0.051 |
HQ | -0.050 |
enta | -0.050 |
Token | � |
Token ID | 247 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
workers | +0.235 |
trainers | +0.228 |
tubing | +0.212 |
HQ | +0.209 |
enta | +0.206 |
Neg logit contributions | |
Crimes | -0.271 |
certific | -0.262 |
witz | -0.257 |
Gutenberg | -0.257 |
unlaw | -0.249 |
Token | life |
Token ID | 1204 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
Crimes | +0.410 |
certific | +0.391 |
witz | +0.385 |
Gutenberg | +0.383 |
unlaw | +0.365 |
Neg logit contributions | |
workers | -0.449 |
trainers | -0.439 |
tubing | -0.412 |
HQ | -0.408 |
enta | -0.401 |
Token | choices |
Token ID | 7747 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
Crimes | +1.176 |
certific | +1.130 |
witz | +1.110 |
Gutenberg | +1.105 |
unlaw | +1.066 |
Neg logit contributions | |
workers | -1.215 |
trainers | -1.182 |
tubing | -1.108 |
HQ | -1.094 |
enta | -1.078 |
Token | and |
Token ID | 290 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
workers | +1.993 |
trainers | +1.949 |
tubing | +1.816 |
HQ | +1.793 |
enta | +1.771 |
Neg logit contributions | |
Crimes | -1.926 |
certific | -1.851 |
witz | -1.821 |
Gutenberg | -1.812 |
unlaw | -1.759 |
Token | their |
Token ID | 511 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
Crimes | +0.969 |
certific | +0.937 |
witz | +0.912 |
Gutenberg | +0.911 |
unlaw | +0.878 |
Neg logit contributions | |
workers | -0.923 |
trainers | -0.891 |
tubing | -0.840 |
HQ | -0.833 |
enta | -0.821 |
Token | professional |
Token ID | 4708 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
workers | +0.315 |
trainers | +0.306 |
tubing | +0.285 |
HQ | +0.282 |
enta | +0.277 |
Neg logit contributions | |
Crimes | -0.344 |
certific | -0.331 |
witz | -0.326 |
Gutenberg | -0.324 |
unlaw | -0.314 |
Token | choices |
Token ID | 7747 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
Crimes | +1.990 |
certific | +1.912 |
Gutenberg | +1.854 |
witz | +1.840 |
unlaw | +1.783 |
Neg logit contributions | |
workers | -2.381 |
trainers | -2.278 |
tubing | -2.185 |
enta | -2.162 |
HQ | -2.158 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
workers | +0.765 |
trainers | +0.747 |
tubing | +0.700 |
HQ | +0.691 |
enta | +0.682 |
Neg logit contributions | |
Crimes | -0.684 |
certific | -0.655 |
witz | -0.644 |
Gutenberg | -0.641 |
unlaw | -0.617 |
Token | � |
Token ID | 447 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
Crimes | +0.500 |
certific | +0.465 |
witz | +0.464 |
Gutenberg | +0.463 |
Vienna | +0.442 |
Neg logit contributions | |
workers | -0.570 |
trainers | -0.560 |
tubing | -0.529 |
enta | -0.514 |
HQ | -0.514 |
Token | � |
Token ID | 251 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
Crimes | +1.265 |
certific | +1.193 |
witz | +1.184 |
Gutenberg | +1.177 |
unlaw | +1.129 |
Neg logit contributions | |
workers | -1.316 |
trainers | -1.286 |
tubing | -1.190 |
HQ | -1.172 |
enta | -1.165 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Crimes | +0.086 |
certific | +0.080 |
witz | +0.080 |
Gutenberg | +0.080 |
Vienna | +0.076 |
Neg logit contributions | |
workers | -0.096 |
trainers | -0.095 |
tubing | -0.089 |
HQ | -0.087 |
enta | -0.087 |
Token | that |
Token ID | 326 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Crimes | +2.432 |
certific | +2.325 |
witz | +2.316 |
Gutenberg | +2.273 |
Vienna | +2.205 |
Neg logit contributions | |
workers | -2.450 |
trainers | -2.349 |
tubing | -2.213 |
enta | -2.176 |
HQ | -2.163 |
Token | of |
Token ID | 286 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Crimes | +0.479 |
certific | +0.462 |
witz | +0.450 |
Gutenberg | +0.449 |
unlaw | +0.430 |
Neg logit contributions | |
workers | -0.621 |
trainers | -0.604 |
tubing | -0.569 |
HQ | -0.564 |
enta | -0.560 |
Token | Belgium |
Token ID | 15664 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
workers | +0.365 |
trainers | +0.356 |
tubing | +0.332 |
HQ | +0.328 |
enta | +0.324 |
Neg logit contributions | |
Crimes | -0.369 |
certific | -0.356 |
witz | -0.349 |
Gutenberg | -0.348 |
unlaw | -0.335 |
Token | , |
Token ID | 11 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Crimes | +0.203 |
certific | +0.194 |
witz | +0.192 |
Gutenberg | +0.190 |
Vienna | +0.183 |
Neg logit contributions | |
workers | -0.203 |
trainers | -0.199 |
tubing | -0.185 |
HQ | -0.183 |
enta | -0.180 |
Token | France |
Token ID | 4881 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
Crimes | +0.384 |
certific | +0.368 |
witz | +0.362 |
Gutenberg | +0.361 |
Vienna | +0.348 |
Neg logit contributions | |
workers | -0.396 |
trainers | -0.388 |
tubing | -0.363 |
HQ | -0.357 |
enta | -0.355 |
Token | and |
Token ID | 290 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
Crimes | +0.127 |
certific | +0.121 |
witz | +0.120 |
Gutenberg | +0.118 |
Vienna | +0.114 |
Neg logit contributions | |
workers | -0.143 |
trainers | -0.140 |
tubing | -0.131 |
HQ | -0.129 |
enta | -0.127 |
Token | Austria |
Token ID | 17322 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
Crimes | +0.847 |
certific | +0.811 |
witz | +0.798 |
Gutenberg | +0.794 |
Vienna | +0.775 |
Neg logit contributions | |
workers | -1.002 |
trainers | -0.978 |
tubing | -0.919 |
enta | -0.916 |
HQ | -0.910 |
Token | to |
Token ID | 284 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
Crimes | +1.377 |
certific | +1.313 |
witz | +1.303 |
Gutenberg | +1.282 |
Vienna | +1.249 |
Neg logit contributions | |
workers | -1.393 |
trainers | -1.362 |
tubing | -1.269 |
HQ | -1.252 |
enta | -1.232 |
Token | 79 |
Token ID | 9225 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
Crimes | +0.046 |
certific | +0.044 |
witz | +0.044 |
Gutenberg | +0.044 |
unlaw | +0.042 |
Neg logit contributions | |
workers | -0.045 |
trainers | -0.043 |
tubing | -0.041 |
HQ | -0.040 |
enta | -0.040 |
Token | percent |
Token ID | 1411 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
workers | +0.472 |
trainers | +0.461 |
tubing | +0.433 |
HQ | +0.428 |
enta | +0.422 |
Neg logit contributions | |
Crimes | -0.401 |
certific | -0.384 |
witz | -0.377 |
Gutenberg | -0.375 |
unlaw | -0.361 |
Token | of |
Token ID | 286 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
workers | +0.131 |
trainers | +0.128 |
tubing | +0.120 |
HQ | +0.118 |
enta | +0.117 |
Neg logit contributions | |
Crimes | -0.107 |
certific | -0.102 |
witz | -0.100 |
Gutenberg | -0.100 |
unlaw | -0.096 |
Token | well |
Token ID | 880 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
Crimes | +0.995 |
certific | +0.954 |
Gutenberg | +0.939 |
witz | +0.937 |
Vienna | +0.898 |
Neg logit contributions | |
workers | -1.032 |
trainers | -0.987 |
tubing | -0.939 |
HQ | -0.922 |
enta | -0.913 |
Token | he |
Token ID | 339 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
workers | +1.280 |
trainers | +1.248 |
tubing | +1.163 |
HQ | +1.146 |
enta | +1.130 |
Neg logit contributions | |
Crimes | -1.344 |
certific | -1.293 |
witz | -1.271 |
Gutenberg | -1.267 |
unlaw | -1.225 |
Token | could |
Token ID | 714 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
Crimes | +0.163 |
certific | +0.157 |
witz | +0.155 |
Gutenberg | +0.153 |
unlaw | +0.148 |
Neg logit contributions | |
workers | -0.156 |
trainers | -0.152 |
tubing | -0.141 |
HQ | -0.140 |
enta | -0.138 |
Token | just |
Token ID | 655 |
Feature activation | +7.29e-01 |
Pos logit contributions | |
Crimes | +0.341 |
certific | +0.329 |
witz | +0.323 |
Gutenberg | +0.320 |
unlaw | +0.309 |
Neg logit contributions | |
workers | -0.332 |
trainers | -0.321 |
tubing | -0.299 |
HQ | -0.298 |
enta | -0.293 |
Token | hang |
Token ID | 8181 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
Crimes | +2.045 |
certific | +1.968 |
witz | +1.929 |
Gutenberg | +1.919 |
Vienna | +1.852 |
Neg logit contributions | |
workers | -2.033 |
trainers | -1.957 |
HQ | -1.846 |
tubing | -1.814 |
enta | -1.801 |
Token | the |
Token ID | 262 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
Crimes | +1.312 |
certific | +1.250 |
witz | +1.247 |
Gutenberg | +1.227 |
Vienna | +1.187 |
Neg logit contributions | |
workers | -1.379 |
trainers | -1.335 |
HQ | -1.248 |
tubing | -1.248 |
enta | -1.237 |
Token | m |
Token ID | 285 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
Crimes | +0.837 |
certific | +0.806 |
witz | +0.794 |
Gutenberg | +0.794 |
Vienna | +0.766 |
Neg logit contributions | |
workers | -0.715 |
trainers | -0.687 |
tubing | -0.637 |
HQ | -0.634 |
enta | -0.631 |
Token | ics |
Token ID | 873 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
Crimes | +0.263 |
certific | +0.252 |
witz | +0.249 |
Gutenberg | +0.247 |
unlaw | +0.238 |
Neg logit contributions | |
workers | -0.254 |
trainers | -0.248 |
tubing | -0.231 |
HQ | -0.228 |
enta | -0.224 |
Token | in |
Token ID | 287 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
Crimes | +1.591 |
certific | +1.526 |
witz | +1.517 |
Gutenberg | +1.503 |
Vienna | +1.449 |
Neg logit contributions | |
workers | -1.532 |
trainers | -1.482 |
HQ | -1.392 |
enta | -1.368 |
tubing | -1.367 |
Token | certain |
Token ID | 1728 |
Feature activation | -9.21e-02 |
Pos logit contributions | |
Crimes | +1.199 |
certific | +1.156 |
witz | +1.132 |
Gutenberg | +1.130 |
Vienna | +1.095 |
Neg logit contributions | |
workers | -1.217 |
trainers | -1.169 |
enta | -1.100 |
HQ | -1.100 |
tubing | -1.094 |
Token | areas |
Token ID | 3006 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
workers | +0.259 |
trainers | +0.252 |
tubing | +0.235 |
HQ | +0.233 |
enta | +0.230 |
Neg logit contributions | |
Crimes | -0.258 |
certific | -0.249 |
witz | -0.244 |
Gutenberg | -0.243 |
unlaw | -0.234 |
Token | It |
Token ID | 632 |
Feature activation | +8.70e-02 |
Pos logit contributions | |
Crimes | +0.191 |
certific | +0.181 |
witz | +0.179 |
Gutenberg | +0.178 |
unlaw | +0.171 |
Neg logit contributions | |
workers | -0.209 |
trainers | -0.204 |
tubing | -0.192 |
HQ | -0.188 |
enta | -0.186 |
Token | wasn |
Token ID | 2492 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Crimes | +0.244 |
certific | +0.235 |
witz | +0.232 |
Gutenberg | +0.230 |
unlaw | +0.221 |
Neg logit contributions | |
workers | -0.244 |
trainers | -0.237 |
tubing | -0.221 |
HQ | -0.221 |
enta | -0.217 |
Token | 't |
Token ID | 470 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
workers | +0.279 |
trainers | +0.267 |
tubing | +0.242 |
HQ | +0.241 |
enta | +0.234 |
Neg logit contributions | |
Crimes | -0.483 |
certific | -0.471 |
witz | -0.462 |
Gutenberg | -0.461 |
unlaw | -0.452 |
Token | exactly |
Token ID | 3446 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
workers | +0.695 |
trainers | +0.675 |
tubing | +0.623 |
HQ | +0.613 |
enta | +0.603 |
Neg logit contributions | |
Crimes | -0.908 |
certific | -0.876 |
witz | -0.864 |
Gutenberg | -0.860 |
unlaw | -0.834 |
Token | the |
Token ID | 262 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Crimes | +0.506 |
certific | +0.486 |
witz | +0.476 |
Gutenberg | +0.475 |
unlaw | +0.458 |
Neg logit contributions | |
workers | -0.540 |
trainers | -0.526 |
tubing | -0.491 |
HQ | -0.487 |
enta | -0.481 |
Token | w |
Token ID | 266 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
workers | +0.771 |
trainers | +0.752 |
tubing | +0.699 |
HQ | +0.685 |
enta | +0.674 |
Neg logit contributions | |
Crimes | -0.908 |
certific | -0.876 |
witz | -0.862 |
Gutenberg | -0.858 |
unlaw | -0.831 |
Token | is |
Token ID | 271 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
Crimes | +1.223 |
certific | +1.160 |
witz | +1.157 |
Gutenberg | +1.150 |
unlaw | +1.096 |
Neg logit contributions | |
workers | -1.426 |
trainers | -1.391 |
tubing | -1.309 |
HQ | -1.279 |
enta | -1.251 |
Token | est |
Token ID | 395 |
Feature activation | -8.58e-01 |
Pos logit contributions | |
workers | +0.360 |
trainers | +0.337 |
tubing | +0.278 |
HQ | +0.267 |
enta | +0.255 |
Neg logit contributions | |
Crimes | -1.458 |
certific | -1.421 |
witz | -1.407 |
Gutenberg | -1.403 |
unlaw | -1.373 |
Token | choice |
Token ID | 3572 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
workers | +2.539 |
trainers | +2.504 |
tubing | +2.364 |
HQ | +2.307 |
enta | +2.285 |
Neg logit contributions | |
Crimes | -2.250 |
certific | -2.137 |
witz | -2.109 |
Gutenberg | -2.106 |
unlaw | -2.011 |
Token | of |
Token ID | 286 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
workers | +0.317 |
trainers | +0.309 |
tubing | +0.288 |
HQ | +0.284 |
enta | +0.280 |
Neg logit contributions | |
Crimes | -0.325 |
certific | -0.312 |
witz | -0.307 |
Gutenberg | -0.306 |
unlaw | -0.295 |
Token | words |
Token ID | 2456 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
Crimes | +0.977 |
certific | +0.934 |
witz | +0.917 |
Gutenberg | +0.912 |
unlaw | +0.888 |
Neg logit contributions | |
workers | -1.101 |
trainers | -1.057 |
HQ | -0.995 |
tubing | -0.989 |
enta | -0.976 |
Token | to |
Token ID | 284 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
Crimes | +1.900 |
certific | +1.807 |
witz | +1.788 |
Gutenberg | +1.774 |
unlaw | +1.751 |
Neg logit contributions | |
workers | -1.714 |
trainers | -1.630 |
HQ | -1.499 |
tubing | -1.495 |
enta | -1.494 |
Token | use |
Token ID | 779 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
Crimes | +1.476 |
certific | +1.405 |
witz | +1.378 |
Gutenberg | +1.367 |
unlaw | +1.338 |
Neg logit contributions | |
workers | -1.509 |
trainers | -1.449 |
HQ | -1.354 |
tubing | -1.345 |
enta | -1.333 |
Token | a |
Token ID | 257 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
workers | +1.209 |
trainers | +1.181 |
tubing | +1.102 |
HQ | +1.087 |
enta | +1.071 |
Neg logit contributions | |
Crimes | -1.203 |
certific | -1.157 |
witz | -1.139 |
Gutenberg | -1.133 |
unlaw | -1.092 |
Token | humanitarian |
Token ID | 15706 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
workers | +0.237 |
trainers | +0.231 |
tubing | +0.216 |
HQ | +0.214 |
enta | +0.211 |
Neg logit contributions | |
Crimes | -0.205 |
certific | -0.197 |
witz | -0.193 |
Gutenberg | -0.192 |
unlaw | -0.185 |
Token | or |
Token ID | 393 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
Crimes | +2.465 |
certific | +2.332 |
witz | +2.273 |
Gutenberg | +2.226 |
Vienna | +2.203 |
Neg logit contributions | |
workers | -3.408 |
trainers | -3.301 |
HQ | -3.117 |
enta | -3.105 |
tubing | -3.088 |
Token | peace |
Token ID | 4167 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
Crimes | +0.804 |
certific | +0.772 |
witz | +0.751 |
Gutenberg | +0.746 |
unlaw | +0.728 |
Neg logit contributions | |
workers | -0.994 |
trainers | -0.961 |
tubing | -0.908 |
HQ | -0.901 |
enta | -0.892 |
Token | keeping |
Token ID | 19934 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
Crimes | +2.291 |
certific | +2.210 |
witz | +2.169 |
Gutenberg | +2.129 |
unlaw | +2.099 |
Neg logit contributions | |
workers | -1.911 |
trainers | -1.877 |
tubing | -1.728 |
HQ | -1.714 |
enta | -1.690 |
Token | operation |
Token ID | 4905 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
workers | +0.057 |
trainers | +0.056 |
tubing | +0.052 |
HQ | +0.052 |
enta | +0.051 |
Neg logit contributions | |
Crimes | -0.055 |
certific | -0.053 |
witz | -0.052 |
Gutenberg | -0.052 |
unlaw | -0.050 |
Token | as |
Token ID | 355 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
Crimes | +0.255 |
certific | +0.246 |
witz | +0.242 |
Gutenberg | +0.241 |
unlaw | +0.233 |
Neg logit contributions | |
workers | -0.223 |
trainers | -0.217 |
tubing | -0.201 |
HQ | -0.199 |
enta | -0.196 |
Token | a |
Token ID | 257 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
Crimes | +0.532 |
certific | +0.509 |
witz | +0.498 |
Gutenberg | +0.498 |
unlaw | +0.479 |
Neg logit contributions | |
workers | -0.639 |
trainers | -0.618 |
tubing | -0.583 |
HQ | -0.580 |
enta | -0.574 |
Token | pret |
Token ID | 2181 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
workers | +0.562 |
trainers | +0.546 |
tubing | +0.514 |
HQ | +0.510 |
enta | +0.506 |
Neg logit contributions | |
Crimes | -0.425 |
certific | -0.405 |
witz | -0.396 |
Gutenberg | -0.395 |
unlaw | -0.380 |
Token | these |
Token ID | 777 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
workers | +0.434 |
trainers | +0.421 |
tubing | +0.393 |
HQ | +0.389 |
enta | +0.384 |
Neg logit contributions | |
Crimes | -0.439 |
certific | -0.422 |
witz | -0.414 |
Gutenberg | -0.413 |
unlaw | -0.398 |
Token | platforms |
Token ID | 9554 |
Feature activation | -8.80e-01 |
Pos logit contributions | |
Crimes | +0.112 |
certific | +0.108 |
Gutenberg | +0.106 |
witz | +0.106 |
unlaw | +0.102 |
Neg logit contributions | |
workers | -0.104 |
trainers | -0.101 |
tubing | -0.094 |
HQ | -0.093 |
enta | -0.093 |
Token | that |
Token ID | 326 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
workers | +2.373 |
trainers | +2.333 |
tubing | +2.164 |
HQ | +2.121 |
enta | +2.090 |
Neg logit contributions | |
Crimes | -2.541 |
certific | -2.441 |
witz | -2.404 |
Gutenberg | -2.383 |
unlaw | -2.347 |
Token | enable |
Token ID | 7139 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
workers | +1.116 |
trainers | +1.091 |
tubing | +1.018 |
HQ | +1.003 |
enta | +0.989 |
Neg logit contributions | |
Crimes | -1.081 |
certific | -1.037 |
witz | -1.019 |
Gutenberg | -1.014 |
unlaw | -0.983 |
Token | so |
Token ID | 523 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
workers | +1.537 |
trainers | +1.507 |
tubing | +1.401 |
HQ | +1.377 |
enta | +1.353 |
Neg logit contributions | |
Crimes | -1.645 |
certific | -1.579 |
witz | -1.555 |
Gutenberg | -1.543 |
unlaw | -1.502 |
Token | many |
Token ID | 867 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
workers | +0.166 |
trainers | +0.160 |
tubing | +0.146 |
HQ | +0.142 |
enta | +0.139 |
Neg logit contributions | |
Crimes | -0.304 |
certific | -0.295 |
witz | -0.292 |
Gutenberg | -0.291 |
unlaw | -0.282 |
Token | people |
Token ID | 661 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
workers | +0.259 |
trainers | +0.253 |
tubing | +0.236 |
HQ | +0.233 |
enta | +0.229 |
Neg logit contributions | |
Crimes | -0.263 |
certific | -0.253 |
witz | -0.249 |
Gutenberg | -0.248 |
unlaw | -0.239 |
Token | use |
Token ID | 779 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
workers | +0.901 |
trainers | +0.874 |
tubing | +0.807 |
HQ | +0.796 |
enta | +0.784 |
Neg logit contributions | |
Crimes | -1.143 |
certific | -1.104 |
witz | -1.087 |
Gutenberg | -1.084 |
unlaw | -1.047 |
Token | to |
Token ID | 284 |
Feature activation | -7.71e-01 |
Pos logit contributions | |
workers | +2.337 |
trainers | +2.283 |
tubing | +2.128 |
HQ | +2.096 |
enta | +2.058 |
Neg logit contributions | |
Crimes | -2.528 |
certific | -2.428 |
witz | -2.392 |
Gutenberg | -2.375 |
unlaw | -2.321 |
Token | build |
Token ID | 1382 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
workers | +2.207 |
trainers | +2.153 |
tubing | +2.016 |
HQ | +1.988 |
enta | +1.961 |
Neg logit contributions | |
Crimes | -2.123 |
certific | -2.038 |
witz | -1.996 |
Gutenberg | -1.991 |
unlaw | -1.935 |
Token | such |
Token ID | 884 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
Crimes | +0.389 |
certific | +0.377 |
witz | +0.369 |
Gutenberg | +0.369 |
Vienna | +0.354 |
Neg logit contributions | |
workers | -0.368 |
trainers | -0.355 |
tubing | -0.331 |
HQ | -0.328 |
enta | -0.326 |
Token | el |
Token ID | 417 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
workers | +2.600 |
trainers | +2.500 |
enta | +2.306 |
tubing | +2.301 |
HQ | +2.296 |
Neg logit contributions | |
Crimes | -3.375 |
certific | -3.269 |
Gutenberg | -3.183 |
witz | -3.178 |
unlaw | -3.128 |
Token | iving |
Token ID | 1412 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
Crimes | +0.720 |
certific | +0.689 |
witz | +0.688 |
Gutenberg | +0.677 |
unlaw | +0.656 |
Neg logit contributions | |
workers | -0.590 |
trainers | -0.582 |
tubing | -0.538 |
HQ | -0.531 |
enta | -0.515 |
Token | / |
Token ID | 14 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
Crimes | +0.553 |
certific | +0.527 |
witz | +0.524 |
Gutenberg | +0.517 |
unlaw | +0.500 |
Neg logit contributions | |
workers | -0.569 |
trainers | -0.562 |
tubing | -0.528 |
HQ | -0.513 |
enta | -0.509 |
Token | su |
Token ID | 2385 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
workers | +0.505 |
trainers | +0.493 |
tubing | +0.458 |
HQ | +0.449 |
enta | +0.442 |
Neg logit contributions | |
Crimes | -0.606 |
certific | -0.583 |
witz | -0.576 |
Gutenberg | -0.571 |
unlaw | -0.554 |
Token | icide |
Token ID | 5285 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
workers | +1.373 |
trainers | +1.322 |
HQ | +1.220 |
tubing | +1.219 |
enta | +1.203 |
Neg logit contributions | |
Crimes | -1.690 |
certific | -1.642 |
Gutenberg | -1.612 |
witz | -1.598 |
unlaw | -1.559 |
Token | pre |
Token ID | 3866 |
Feature activation | -9.44e-02 |
Pos logit contributions | |
Crimes | +0.031 |
certific | +0.030 |
witz | +0.029 |
Gutenberg | +0.029 |
unlaw | +0.028 |
Neg logit contributions | |
workers | -0.031 |
trainers | -0.030 |
tubing | -0.028 |
HQ | -0.028 |
enta | -0.027 |
Token | vention |
Token ID | 4018 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
workers | +0.299 |
trainers | +0.293 |
tubing | +0.276 |
HQ | +0.272 |
enta | +0.269 |
Neg logit contributions | |
Crimes | -0.231 |
certific | -0.220 |
witz | -0.216 |
Gutenberg | -0.215 |
unlaw | -0.206 |
Token | / |
Token ID | 14 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
Crimes | +0.245 |
certific | +0.235 |
witz | +0.233 |
Gutenberg | +0.230 |
unlaw | +0.222 |
Neg logit contributions | |
workers | -0.246 |
trainers | -0.242 |
tubing | -0.226 |
HQ | -0.221 |
enta | -0.219 |
Token | doc |
Token ID | 15390 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
workers | +0.732 |
trainers | +0.712 |
tubing | +0.662 |
HQ | +0.652 |
enta | +0.642 |
Neg logit contributions | |
Crimes | -0.830 |
certific | -0.799 |
witz | -0.787 |
Gutenberg | -0.783 |
unlaw | -0.758 |
Token | uments |
Token ID | 2886 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
workers | +1.239 |
trainers | +1.199 |
tubing | +1.099 |
HQ | +1.084 |
enta | +1.063 |
Neg logit contributions | |
Crimes | -1.833 |
certific | -1.773 |
witz | -1.746 |
Gutenberg | -1.741 |
unlaw | -1.692 |
Token | / |
Token ID | 14 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
workers | +0.346 |
trainers | +0.338 |
tubing | +0.317 |
HQ | +0.311 |
enta | +0.308 |
Neg logit contributions | |
Crimes | -0.317 |
certific | -0.304 |
witz | -0.299 |
Gutenberg | -0.298 |
unlaw | -0.287 |
Token | ( |
Token ID | 357 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
Crimes | +0.344 |
certific | +0.319 |
witz | +0.318 |
Gutenberg | +0.317 |
Vienna | +0.303 |
Neg logit contributions | |
workers | -0.406 |
trainers | -0.400 |
tubing | -0.377 |
enta | -0.367 |
HQ | -0.366 |
Token | In |
Token ID | 818 |
Feature activation | -1.72e-03 |
Pos logit contributions | |
workers | +1.939 |
trainers | +1.874 |
tubing | +1.741 |
HQ | +1.736 |
enta | +1.697 |
Neg logit contributions | |
Crimes | -2.106 |
certific | -2.035 |
witz | -1.998 |
Gutenberg | -1.994 |
unlaw | -1.933 |
Token | my |
Token ID | 616 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
workers | +0.005 |
trainers | +0.005 |
tubing | +0.004 |
HQ | +0.004 |
enta | +0.004 |
Neg logit contributions | |
Crimes | -0.005 |
certific | -0.005 |
witz | -0.005 |
Gutenberg | -0.005 |
Vienna | -0.004 |
Token | neighborhood |
Token ID | 6232 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
workers | +0.326 |
trainers | +0.317 |
tubing | +0.295 |
HQ | +0.290 |
enta | +0.287 |
Neg logit contributions | |
Crimes | -0.360 |
certific | -0.346 |
witz | -0.340 |
Gutenberg | -0.339 |
unlaw | -0.327 |
Token | , |
Token ID | 11 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
workers | +0.361 |
trainers | +0.353 |
tubing | +0.330 |
HQ | +0.325 |
enta | +0.321 |
Neg logit contributions | |
Crimes | -0.355 |
certific | -0.339 |
witz | -0.334 |
Gutenberg | -0.332 |
unlaw | -0.320 |
Token | I |
Token ID | 314 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
workers | +1.182 |
trainers | +1.152 |
tubing | +1.076 |
HQ | +1.063 |
enta | +1.049 |
Neg logit contributions | |
Crimes | -1.155 |
certific | -1.108 |
witz | -1.090 |
Gutenberg | -1.085 |
unlaw | -1.046 |
Token | know |
Token ID | 760 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
workers | +0.351 |
trainers | +0.342 |
tubing | +0.318 |
HQ | +0.314 |
enta | +0.310 |
Neg logit contributions | |
Crimes | -0.388 |
certific | -0.373 |
witz | -0.365 |
Gutenberg | -0.362 |
unlaw | -0.350 |
Token | of |
Token ID | 286 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
workers | +0.942 |
trainers | +0.916 |
tubing | +0.855 |
HQ | +0.845 |
enta | +0.832 |
Neg logit contributions | |
Crimes | -1.014 |
certific | -0.973 |
witz | -0.958 |
Gutenberg | -0.951 |
unlaw | -0.920 |
Token | no |
Token ID | 645 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
workers | +1.111 |
trainers | +1.080 |
tubing | +1.001 |
HQ | +0.986 |
enta | +0.971 |
Neg logit contributions | |
Crimes | -1.325 |
certific | -1.278 |
witz | -1.258 |
Gutenberg | -1.253 |
unlaw | -1.214 |
Token | sidewalks |
Token ID | 38282 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
workers | +0.992 |
trainers | +0.969 |
tubing | +0.903 |
HQ | +0.890 |
enta | +0.876 |
Neg logit contributions | |
Crimes | -1.042 |
certific | -1.003 |
witz | -0.989 |
Gutenberg | -0.984 |
unlaw | -0.952 |
Token | that |
Token ID | 326 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
workers | +1.468 |
trainers | +1.428 |
tubing | +1.329 |
HQ | +1.307 |
enta | +1.284 |
Neg logit contributions | |
Crimes | -1.716 |
certific | -1.650 |
witz | -1.630 |
Gutenberg | -1.626 |
unlaw | -1.574 |
Token | maze |
Token ID | 31237 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
workers | +0.812 |
trainers | +0.788 |
tubing | +0.734 |
HQ | +0.726 |
enta | +0.718 |
Neg logit contributions | |
Crimes | -0.843 |
certific | -0.807 |
witz | -0.794 |
Gutenberg | -0.792 |
unlaw | -0.764 |
Token | , |
Token ID | 11 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
workers | +0.790 |
trainers | +0.765 |
tubing | +0.711 |
HQ | +0.703 |
enta | +0.697 |
Neg logit contributions | |
Crimes | -0.853 |
certific | -0.818 |
witz | -0.807 |
Gutenberg | -0.804 |
unlaw | -0.771 |
Token | they |
Token ID | 484 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
workers | +1.270 |
trainers | +1.236 |
tubing | +1.150 |
HQ | +1.133 |
enta | +1.117 |
Neg logit contributions | |
Crimes | -1.404 |
certific | -1.352 |
witz | -1.330 |
Gutenberg | -1.325 |
unlaw | -1.282 |
Token | � |
Token ID | 447 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
Crimes | +0.090 |
certific | +0.087 |
witz | +0.085 |
Gutenberg | +0.084 |
Vienna | +0.081 |
Neg logit contributions | |
workers | -0.092 |
trainers | -0.089 |
HQ | -0.083 |
tubing | -0.083 |
enta | -0.081 |
Token | � |
Token ID | 247 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
workers | +0.464 |
trainers | +0.452 |
tubing | +0.423 |
HQ | +0.420 |
enta | +0.411 |
Neg logit contributions | |
Crimes | -0.484 |
certific | -0.469 |
Gutenberg | -0.460 |
witz | -0.459 |
unlaw | -0.447 |
Token | ll |
Token ID | 297 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
workers | +0.081 |
trainers | +0.079 |
tubing | +0.073 |
HQ | +0.071 |
enta | +0.070 |
Neg logit contributions | |
Crimes | -0.101 |
certific | -0.098 |
witz | -0.097 |
Gutenberg | -0.096 |
unlaw | -0.093 |
Token | see |
Token ID | 766 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
workers | +2.076 |
trainers | +2.023 |
tubing | +1.885 |
HQ | +1.851 |
enta | +1.826 |
Neg logit contributions | |
Crimes | -2.261 |
certific | -2.176 |
witz | -2.140 |
Gutenberg | -2.135 |
unlaw | -2.066 |
Token | , |
Token ID | 11 |
Feature activation | -3.19e-03 |
Pos logit contributions | |
workers | +0.761 |
trainers | +0.737 |
tubing | +0.686 |
HQ | +0.681 |
enta | +0.672 |
Neg logit contributions | |
Crimes | -0.824 |
certific | -0.790 |
witz | -0.778 |
Gutenberg | -0.775 |
unlaw | -0.747 |
Token | � |
Token ID | 564 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
workers | +0.009 |
trainers | +0.008 |
HQ | +0.008 |
tubing | +0.008 |
enta | +0.008 |
Neg logit contributions | |
Crimes | -0.009 |
certific | -0.009 |
witz | -0.009 |
Gutenberg | -0.009 |
Vienna | -0.009 |
Token | � |
Token ID | 250 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
Crimes | +0.555 |
certific | +0.527 |
witz | +0.519 |
Gutenberg | +0.516 |
unlaw | +0.497 |
Neg logit contributions | |
workers | -0.634 |
trainers | -0.620 |
tubing | -0.579 |
HQ | -0.574 |
enta | -0.566 |
Token | The |
Token ID | 464 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
workers | +0.080 |
trainers | +0.079 |
tubing | +0.073 |
HQ | +0.072 |
enta | +0.071 |
Neg logit contributions | |
Crimes | -0.095 |
certific | -0.091 |
witz | -0.090 |
Gutenberg | -0.089 |
unlaw | -0.086 |
Token | Moore |
Token ID | 8877 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
Crimes | +1.670 |
certific | +1.579 |
Gutenberg | +1.572 |
witz | +1.553 |
Vienna | +1.498 |
Neg logit contributions | |
workers | -1.594 |
trainers | -1.541 |
tubing | -1.438 |
HQ | -1.433 |
enta | -1.404 |
Token | doubles |
Token ID | 21938 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
workers | +1.564 |
trainers | +1.531 |
tubing | +1.439 |
HQ | +1.419 |
enta | +1.403 |
Neg logit contributions | |
Crimes | -1.283 |
certific | -1.232 |
witz | -1.203 |
Gutenberg | -1.201 |
unlaw | -1.158 |
Token | peak |
Token ID | 36729 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Crimes | +1.745 |
certific | +1.620 |
witz | +1.603 |
Gutenberg | +1.557 |
unlaw | +1.454 |
Neg logit contributions | |
workers | -4.198 |
trainers | -4.112 |
HQ | -3.890 |
tubing | -3.884 |
agen | -3.819 |
Token | , |
Token ID | 11 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
Crimes | +0.503 |
certific | +0.481 |
witz | +0.475 |
Gutenberg | +0.472 |
unlaw | +0.454 |
Neg logit contributions | |
workers | -0.528 |
trainers | -0.514 |
tubing | -0.481 |
HQ | -0.476 |
enta | -0.469 |
Token | everyone |
Token ID | 2506 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
workers | +0.225 |
trainers | +0.218 |
tubing | +0.203 |
HQ | +0.201 |
enta | +0.198 |
Neg logit contributions | |
Crimes | -0.249 |
certific | -0.239 |
witz | -0.235 |
Gutenberg | -0.235 |
unlaw | -0.227 |
Token | keeps |
Token ID | 7622 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
workers | +0.904 |
trainers | +0.881 |
tubing | +0.824 |
HQ | +0.811 |
enta | +0.802 |
Neg logit contributions | |
Crimes | -0.883 |
certific | -0.849 |
witz | -0.833 |
Gutenberg | -0.831 |
unlaw | -0.804 |
Token | referring |
Token ID | 9759 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
Crimes | +0.789 |
certific | +0.759 |
witz | +0.748 |
Gutenberg | +0.746 |
Vienna | +0.718 |
Neg logit contributions | |
workers | -0.695 |
trainers | -0.672 |
tubing | -0.624 |
HQ | -0.622 |
enta | -0.612 |
Token | to |
Token ID | 284 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
Crimes | +2.603 |
certific | +2.471 |
Gutenberg | +2.450 |
witz | +2.446 |
Vienna | +2.356 |
Neg logit contributions | |
workers | -2.273 |
trainers | -2.203 |
tubing | -2.046 |
HQ | -2.034 |
enta | -2.000 |
Token | him |
Token ID | 683 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
Crimes | +1.927 |
certific | +1.817 |
Gutenberg | +1.815 |
witz | +1.780 |
Vienna | +1.731 |
Neg logit contributions | |
workers | -1.931 |
trainers | -1.860 |
tubing | -1.751 |
HQ | -1.743 |
enta | -1.707 |
Token | as |
Token ID | 355 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
Crimes | +0.408 |
certific | +0.389 |
witz | +0.384 |
Gutenberg | +0.382 |
unlaw | +0.367 |
Neg logit contributions | |
workers | -0.417 |
trainers | -0.405 |
tubing | -0.380 |
HQ | -0.375 |
enta | -0.370 |
Token | � |
Token ID | 564 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Crimes | +0.733 |
certific | +0.695 |
Gutenberg | +0.688 |
witz | +0.686 |
unlaw | +0.657 |
Neg logit contributions | |
workers | -0.747 |
trainers | -0.724 |
tubing | -0.678 |
HQ | -0.673 |
enta | -0.661 |
Token | World |
Token ID | 2159 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
Crimes | +0.350 |
certific | +0.338 |
witz | +0.331 |
Gutenberg | +0.331 |
unlaw | +0.317 |
Neg logit contributions | |
workers | -0.328 |
trainers | -0.317 |
tubing | -0.295 |
HQ | -0.292 |
enta | -0.290 |
Token | War |
Token ID | 1810 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
Crimes | +0.144 |
certific | +0.137 |
Gutenberg | +0.133 |
witz | +0.133 |
unlaw | +0.126 |
Neg logit contributions | |
workers | -0.214 |
trainers | -0.210 |
tubing | -0.198 |
HQ | -0.196 |
enta | -0.194 |
Token | Two |
Token ID | 4930 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
Crimes | +1.275 |
witz | +1.197 |
certific | +1.191 |
Gutenberg | +1.162 |
Vienna | +1.117 |
Neg logit contributions | |
workers | -2.057 |
trainers | -2.017 |
tubing | -1.890 |
HQ | -1.882 |
enta | -1.867 |
Token | ( |
Token ID | 357 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Crimes | +1.619 |
certific | +1.569 |
witz | +1.523 |
Gutenberg | +1.522 |
Vienna | +1.469 |
Neg logit contributions | |
workers | -1.728 |
trainers | -1.654 |
tubing | -1.550 |
HQ | -1.538 |
enta | -1.535 |
Token | circ |
Token ID | 21170 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
workers | +0.562 |
trainers | +0.547 |
tubing | +0.503 |
HQ | +0.490 |
enta | +0.485 |
Neg logit contributions | |
Crimes | -0.797 |
certific | -0.770 |
witz | -0.761 |
Gutenberg | -0.755 |
unlaw | -0.733 |
Token | a |
Token ID | 64 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
workers | +0.945 |
trainers | +0.921 |
tubing | +0.860 |
HQ | +0.847 |
enta | +0.836 |
Neg logit contributions | |
Crimes | -0.970 |
certific | -0.933 |
witz | -0.917 |
Gutenberg | -0.915 |
unlaw | -0.883 |
Token | 1941 |
Token ID | 23234 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
workers | +2.946 |
trainers | +2.867 |
tubing | +2.680 |
HQ | +2.663 |
enta | +2.602 |
Neg logit contributions | |
Crimes | -3.108 |
certific | -3.027 |
Gutenberg | -2.954 |
witz | -2.938 |
unlaw | -2.887 |
Token | ) |
Token ID | 8 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
workers | +1.652 |
trainers | +1.615 |
tubing | +1.517 |
HQ | +1.500 |
enta | +1.478 |
Neg logit contributions | |
Crimes | -1.378 |
certific | -1.320 |
witz | -1.296 |
Gutenberg | -1.291 |
unlaw | -1.242 |
Token | improvement |
Token ID | 9025 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
Crimes | +0.121 |
certific | +0.116 |
witz | +0.114 |
Gutenberg | +0.114 |
unlaw | +0.109 |
Neg logit contributions | |
workers | -0.136 |
trainers | -0.133 |
tubing | -0.124 |
HQ | -0.123 |
enta | -0.122 |
Token | on |
Token ID | 319 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
Crimes | +0.196 |
certific | +0.188 |
witz | +0.185 |
Gutenberg | +0.183 |
unlaw | +0.177 |
Neg logit contributions | |
workers | -0.204 |
trainers | -0.199 |
tubing | -0.186 |
HQ | -0.184 |
enta | -0.182 |
Token | the |
Token ID | 262 |
Feature activation | +2.15e-03 |
Pos logit contributions | |
Crimes | +1.325 |
certific | +1.275 |
witz | +1.256 |
Gutenberg | +1.253 |
Vienna | +1.201 |
Neg logit contributions | |
workers | -1.430 |
trainers | -1.382 |
tubing | -1.293 |
HQ | -1.287 |
enta | -1.272 |
Token | bubbles |
Token ID | 25037 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
Crimes | +2.091 |
certific | +2.006 |
witz | +1.966 |
Gutenberg | +1.958 |
Vienna | +1.871 |
Neg logit contributions | |
workers | -2.208 |
trainers | -2.130 |
HQ | -2.004 |
tubing | -1.990 |
enta | -1.965 |
Token | through |
Token ID | 832 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
Crimes | +0.756 |
certific | +0.727 |
witz | +0.716 |
Gutenberg | +0.714 |
unlaw | +0.688 |
Neg logit contributions | |
workers | -0.686 |
trainers | -0.665 |
tubing | -0.617 |
HQ | -0.610 |
enta | -0.603 |
Token | the |
Token ID | 262 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
workers | +1.254 |
trainers | +1.223 |
tubing | +1.138 |
HQ | +1.121 |
enta | +1.104 |
Neg logit contributions | |
Crimes | -1.351 |
certific | -1.300 |
witz | -1.280 |
Gutenberg | -1.274 |
unlaw | -1.233 |
Token | speakers |
Token ID | 11636 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
workers | +0.708 |
trainers | +0.685 |
tubing | +0.636 |
HQ | +0.629 |
enta | +0.621 |
Neg logit contributions | |
Crimes | -0.836 |
certific | -0.804 |
witz | -0.791 |
Gutenberg | -0.789 |
unlaw | -0.764 |
Token | is |
Token ID | 318 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
Crimes | +0.347 |
certific | +0.333 |
witz | +0.328 |
Gutenberg | +0.326 |
unlaw | +0.313 |
Neg logit contributions | |
workers | -0.358 |
trainers | -0.348 |
tubing | -0.325 |
HQ | -0.323 |
enta | -0.318 |
Token | hard |
Token ID | 1327 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
workers | +0.665 |
trainers | +0.646 |
tubing | +0.601 |
HQ | +0.593 |
enta | +0.584 |
Neg logit contributions | |
Crimes | -0.747 |
certific | -0.719 |
witz | -0.708 |
Gutenberg | -0.705 |
unlaw | -0.681 |
Token | to |
Token ID | 284 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
workers | +2.541 |
trainers | +2.452 |
tubing | +2.272 |
HQ | +2.205 |
ox | +2.162 |
Neg logit contributions | |
Crimes | -3.551 |
certific | -3.444 |
Gutenberg | -3.405 |
witz | -3.381 |
unlaw | -3.284 |
Token | over |
Token ID | 625 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
workers | +0.737 |
trainers | +0.719 |
tubing | +0.671 |
HQ | +0.659 |
enta | +0.651 |
Neg logit contributions | |
Crimes | -0.774 |
certific | -0.744 |
witz | -0.732 |
Gutenberg | -0.729 |
unlaw | -0.705 |
Token | state |
Token ID | 5219 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
Crimes | +0.872 |
certific | +0.845 |
witz | +0.831 |
Gutenberg | +0.829 |
Vienna | +0.792 |
Neg logit contributions | |
workers | -0.723 |
trainers | -0.705 |
tubing | -0.655 |
HQ | -0.649 |
enta | -0.632 |
Token | . |
Token ID | 13 |
Feature activation | +8.92e-03 |
Pos logit contributions | |
Crimes | +2.003 |
certific | +1.891 |
witz | +1.871 |
Gutenberg | +1.832 |
Vienna | +1.772 |
Neg logit contributions | |
workers | -2.496 |
trainers | -2.432 |
tubing | -2.278 |
enta | -2.259 |
HQ | -2.255 |
Token | Capcom |
Token ID | 35936 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
Crimes | +0.024 |
certific | +0.023 |
witz | +0.023 |
Gutenberg | +0.022 |
unlaw | +0.022 |
Neg logit contributions | |
workers | -0.026 |
trainers | -0.026 |
tubing | -0.024 |
HQ | -0.024 |
enta | -0.023 |
Token | she |
Token ID | 673 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
workers | +0.970 |
trainers | +0.944 |
tubing | +0.883 |
HQ | +0.872 |
enta | +0.859 |
Neg logit contributions | |
Crimes | -0.993 |
certific | -0.952 |
witz | -0.938 |
Gutenberg | -0.936 |
unlaw | -0.900 |
Token | said |
Token ID | 531 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
workers | +1.046 |
trainers | +1.019 |
tubing | +0.953 |
HQ | +0.941 |
enta | +0.927 |
Neg logit contributions | |
Crimes | -1.037 |
certific | -0.996 |
witz | -0.980 |
Gutenberg | -0.975 |
unlaw | -0.940 |
Token | . |
Token ID | 13 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
workers | +1.988 |
trainers | +1.942 |
tubing | +1.819 |
HQ | +1.794 |
enta | +1.773 |
Neg logit contributions | |
Crimes | -1.739 |
certific | -1.671 |
witz | -1.635 |
Gutenberg | -1.631 |
unlaw | -1.574 |
Token |
|
Token ID | 198 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
workers | +1.385 |
trainers | +1.351 |
tubing | +1.267 |
HQ | +1.249 |
enta | +1.234 |
Neg logit contributions | |
Crimes | -1.261 |
certific | -1.205 |
witz | -1.186 |
Gutenberg | -1.182 |
unlaw | -1.136 |
Token |
|
Token ID | 198 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
workers | +0.832 |
trainers | +0.806 |
HQ | +0.752 |
tubing | +0.750 |
ox | +0.737 |
Neg logit contributions | |
Crimes | -0.775 |
certific | -0.757 |
witz | -0.730 |
Gutenberg | -0.729 |
unlaw | -0.716 |
Token | Though |
Token ID | 10915 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
workers | +1.213 |
trainers | +1.184 |
tubing | +1.103 |
HQ | +1.083 |
enta | +1.070 |
Neg logit contributions | |
Crimes | -1.321 |
certific | -1.269 |
witz | -1.252 |
Gutenberg | -1.246 |
unlaw | -1.203 |
Token | she |
Token ID | 673 |
Feature activation | +2.34e-03 |
Pos logit contributions | |
workers | +3.239 |
trainers | +3.168 |
tubing | +2.941 |
HQ | +2.890 |
ox | +2.846 |
Neg logit contributions | |
Crimes | -3.362 |
certific | -3.250 |
witz | -3.181 |
Gutenberg | -3.169 |
unlaw | -3.085 |
Token | misses |
Token ID | 18297 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
Crimes | +0.007 |
certific | +0.007 |
witz | +0.006 |
Gutenberg | +0.006 |
unlaw | +0.006 |
Neg logit contributions | |
workers | -0.006 |
trainers | -0.006 |
HQ | -0.006 |
tubing | -0.006 |
enta | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
workers | +2.288 |
trainers | +2.244 |
tubing | +2.084 |
HQ | +2.045 |
enta | +2.013 |
Neg logit contributions | |
Crimes | -2.486 |
certific | -2.397 |
witz | -2.361 |
Gutenberg | -2.343 |
unlaw | -2.275 |
Token | " |
Token ID | 366 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Crimes | +0.163 |
certific | +0.157 |
Gutenberg | +0.154 |
witz | +0.154 |
Vienna | +0.149 |
Neg logit contributions | |
workers | -0.149 |
trainers | -0.143 |
tubing | -0.133 |
HQ | -0.133 |
enta | -0.131 |
Token | int |
Token ID | 600 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
Crimes | +0.512 |
certific | +0.491 |
witz | +0.489 |
Gutenberg | +0.483 |
unlaw | +0.467 |
Neg logit contributions | |
workers | -0.418 |
trainers | -0.413 |
tubing | -0.382 |
HQ | -0.371 |
enta | -0.370 |
Token | , |
Token ID | 11 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
Crimes | +1.345 |
certific | +1.282 |
witz | +1.257 |
Gutenberg | +1.245 |
Vienna | +1.190 |
Neg logit contributions | |
workers | -2.086 |
trainers | -2.031 |
tubing | -1.917 |
HQ | -1.897 |
enta | -1.883 |
Token | greed |
Token ID | 25474 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
workers | +0.772 |
trainers | +0.751 |
tubing | +0.703 |
HQ | +0.693 |
enta | +0.682 |
Neg logit contributions | |
Crimes | -0.765 |
certific | -0.733 |
witz | -0.722 |
Gutenberg | -0.716 |
unlaw | -0.694 |
Token | - |
Token ID | 12 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
Crimes | +0.882 |
certific | +0.840 |
witz | +0.823 |
Gutenberg | +0.819 |
unlaw | +0.786 |
Neg logit contributions | |
workers | -1.192 |
trainers | -1.165 |
tubing | -1.098 |
HQ | -1.085 |
enta | -1.073 |
Token | is |
Token ID | 271 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
Crimes | +1.002 |
certific | +0.954 |
witz | +0.943 |
Gutenberg | +0.933 |
unlaw | +0.896 |
Neg logit contributions | |
workers | -1.195 |
trainers | -1.172 |
tubing | -1.101 |
HQ | -1.087 |
enta | -1.074 |
Token | - |
Token ID | 12 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
workers | +0.234 |
trainers | +0.228 |
tubing | +0.216 |
HQ | +0.213 |
enta | +0.211 |
Neg logit contributions | |
Crimes | -0.166 |
certific | -0.158 |
witz | -0.155 |
Gutenberg | -0.154 |
unlaw | -0.148 |
Token | good |
Token ID | 11274 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
Crimes | +0.599 |
witz | +0.570 |
certific | +0.569 |
Gutenberg | +0.560 |
unlaw | +0.546 |
Neg logit contributions | |
trainers | -0.514 |
workers | -0.513 |
tubing | -0.475 |
HQ | -0.461 |
enta | -0.457 |
Token | orthodoxy |
Token ID | 49371 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
workers | +3.348 |
trainers | +3.288 |
tubing | +3.093 |
HQ | +3.072 |
enta | +3.013 |
Neg logit contributions | |
Crimes | -2.618 |
certific | -2.529 |
Gutenberg | -2.475 |
witz | -2.445 |
unlaw | -2.395 |
Token | . |
Token ID | 13 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
Crimes | +0.651 |
certific | +0.624 |
witz | +0.612 |
Gutenberg | +0.606 |
unlaw | +0.581 |
Neg logit contributions | |
workers | -0.763 |
trainers | -0.744 |
tubing | -0.699 |
HQ | -0.691 |
enta | -0.682 |
Token |
|
Token ID | 198 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Crimes | +0.663 |
Gutenberg | +0.619 |
witz | +0.615 |
certific | +0.614 |
Vienna | +0.589 |
Neg logit contributions | |
workers | -0.779 |
trainers | -0.765 |
tubing | -0.719 |
HQ | -0.699 |
enta | -0.698 |
Token |
|
Token ID | 198 |
Feature activation | +1.10e-03 |
Pos logit contributions | |
Crimes | +0.464 |
certific | +0.447 |
witz | +0.436 |
Gutenberg | +0.435 |
unlaw | +0.422 |
Neg logit contributions | |
workers | -0.514 |
trainers | -0.500 |
tubing | -0.468 |
HQ | -0.464 |
enta | -0.456 |
Token | To |
Token ID | 2514 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
Crimes | +0.003 |
certific | +0.003 |
witz | +0.003 |
Gutenberg | +0.003 |
unlaw | +0.003 |
Neg logit contributions | |
workers | -0.003 |
trainers | -0.003 |
tubing | -0.003 |
enta | -0.003 |
HQ | -0.003 |
Token | ... |
Token ID | 2644 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
workers | +0.921 |
trainers | +0.902 |
tubing | +0.854 |
HQ | +0.847 |
enta | +0.838 |
Neg logit contributions | |
Crimes | -0.547 |
certific | -0.519 |
witz | -0.505 |
Gutenberg | -0.501 |
unlaw | -0.481 |
Token | } |
Token ID | 1782 |
Feature activation | -5.10e-02 |
Pos logit contributions | |
workers | +0.838 |
trainers | +0.816 |
tubing | +0.763 |
HQ | +0.754 |
enta | +0.744 |
Neg logit contributions | |
Crimes | -0.815 |
certific | -0.782 |
witz | -0.768 |
Gutenberg | -0.764 |
unlaw | -0.737 |
Token |
|
Token ID | 198 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
workers | +0.130 |
trainers | +0.126 |
tubing | +0.116 |
HQ | +0.116 |
enta | +0.114 |
Neg logit contributions | |
Crimes | -0.157 |
certific | -0.151 |
witz | -0.148 |
Gutenberg | -0.147 |
unlaw | -0.143 |
Token |
|
Token ID | 198 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
workers | +1.017 |
trainers | +0.977 |
HQ | +0.915 |
tubing | +0.909 |
ox | +0.898 |
Neg logit contributions | |
Crimes | -0.996 |
certific | -0.984 |
Gutenberg | -0.941 |
witz | -0.940 |
unlaw | -0.934 |
Token | The |
Token ID | 464 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
workers | +1.410 |
trainers | +1.364 |
tubing | +1.269 |
HQ | +1.262 |
enta | +1.239 |
Neg logit contributions | |
Crimes | -1.496 |
certific | -1.440 |
Gutenberg | -1.413 |
witz | -1.411 |
unlaw | -1.363 |
Token | Shared |
Token ID | 39403 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Crimes | +1.410 |
certific | +1.355 |
witz | +1.323 |
Gutenberg | +1.317 |
unlaw | +1.283 |
Neg logit contributions | |
workers | -1.305 |
trainers | -1.266 |
tubing | -1.189 |
HQ | -1.174 |
enta | -1.164 |
Token | < |
Token ID | 27 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
workers | +3.677 |
trainers | +3.550 |
tubing | +3.284 |
HQ | +3.236 |
ox | +3.192 |
Neg logit contributions | |
Crimes | -4.162 |
certific | -3.996 |
witz | -3.943 |
Gutenberg | -3.919 |
unlaw | -3.806 |
Token | ' |
Token ID | 6 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
Crimes | +1.260 |
certific | +1.212 |
witz | +1.185 |
Gutenberg | +1.175 |
Vienna | +1.157 |
Neg logit contributions | |
workers | -0.716 |
trainers | -0.708 |
tubing | -0.639 |
HQ | -0.620 |
enta | -0.611 |
Token | a |
Token ID | 64 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
workers | +1.450 |
trainers | +1.382 |
tubing | +1.282 |
HQ | +1.281 |
ox | +1.279 |
Neg logit contributions | |
Crimes | -1.650 |
certific | -1.618 |
Gutenberg | -1.580 |
witz | -1.566 |
unlaw | -1.548 |
Token | , |
Token ID | 11 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
workers | +1.321 |
trainers | +1.290 |
tubing | +1.203 |
HQ | +1.190 |
enta | +1.168 |
Neg logit contributions | |
Crimes | -1.295 |
certific | -1.241 |
witz | -1.218 |
Gutenberg | -1.212 |
unlaw | -1.175 |
Token | T |
Token ID | 309 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
Crimes | +1.486 |
certific | +1.409 |
witz | +1.381 |
Gutenberg | +1.361 |
Vienna | +1.346 |
Neg logit contributions | |
workers | -1.375 |
trainers | -1.347 |
tubing | -1.253 |
HQ | -1.249 |
agen | -1.223 |