Token | t |
Token ID | 83 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Surgery | +0.310 |
Broadcasting | +0.303 |
scope | +0.286 |
hra | +0.278 |
aviour | +0.267 |
Neg logit contributions | |
ishes | -0.588 |
ery | -0.583 |
adoes | -0.567 |
WI | -0.563 |
suscept | -0.556 |
Token | I |
Token ID | 314 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
Surgery | +0.776 |
Broadcasting | +0.760 |
scope | +0.730 |
hra | +0.715 |
aviour | +0.696 |
Neg logit contributions | |
ishes | -1.008 |
ery | -0.997 |
adoes | -0.968 |
WI | -0.961 |
suscept | -0.947 |
Token | pray |
Token ID | 12472 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ishes | +0.875 |
ery | +0.864 |
adoes | +0.838 |
WI | +0.830 |
suscept | +0.814 |
Neg logit contributions | |
Surgery | -0.789 |
Broadcasting | -0.774 |
scope | -0.746 |
hra | -0.730 |
aviour | -0.715 |
Token | for |
Token ID | 329 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
Surgery | +1.577 |
Broadcasting | +1.549 |
scope | +1.495 |
hra | +1.468 |
aviour | +1.427 |
Neg logit contributions | |
ishes | -1.647 |
ery | -1.623 |
adoes | -1.581 |
WI | -1.558 |
suscept | -1.540 |
Token | a |
Token ID | 257 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
ishes | +0.102 |
ery | +0.100 |
adoes | +0.098 |
WI | +0.096 |
suscept | +0.095 |
Neg logit contributions | |
Surgery | -0.092 |
Broadcasting | -0.090 |
scope | -0.087 |
hra | -0.085 |
aviour | -0.083 |
Token | miracle |
Token ID | 20820 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
Surgery | +0.423 |
Broadcasting | +0.416 |
scope | +0.400 |
hra | +0.392 |
aviour | +0.380 |
Neg logit contributions | |
ishes | -0.516 |
ery | -0.510 |
adoes | -0.495 |
WI | -0.490 |
suscept | -0.484 |
Token | too |
Token ID | 1165 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Surgery | +7.114 |
Broadcasting | +6.918 |
scope | +6.737 |
hra | +6.625 |
aviour | +6.327 |
Neg logit contributions | |
ishes | -7.696 |
adoes | -7.454 |
ery | -7.335 |
WI | -7.324 |
suscept | -7.237 |
Token | ? |
Token ID | 30 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
ishes | +0.911 |
ery | +0.901 |
adoes | +0.873 |
WI | +0.863 |
suscept | +0.846 |
Neg logit contributions | |
Surgery | -0.990 |
Broadcasting | -0.975 |
scope | -0.943 |
hra | -0.928 |
aviour | -0.903 |
Token |
|
Token ID | 198 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
Surgery | +1.428 |
Broadcasting | +1.397 |
scope | +1.336 |
hra | +1.305 |
aviour | +1.262 |
Neg logit contributions | |
ishes | -1.869 |
ery | -1.843 |
adoes | -1.803 |
WI | -1.773 |
suscept | -1.763 |
Token |
|
Token ID | 198 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
ishes | +0.150 |
ery | +0.149 |
WI | +0.143 |
adoes | +0.143 |
iors | +0.140 |
Neg logit contributions | |
Surgery | -0.116 |
Broadcasting | -0.113 |
scope | -0.108 |
hra | -0.106 |
aviour | -0.105 |
Token | Someone |
Token ID | 28211 |
Feature activation | -8.62e-01 |
Pos logit contributions | |
ishes | +0.254 |
ery | +0.251 |
adoes | +0.243 |
WI | +0.241 |
suscept | +0.235 |
Neg logit contributions | |
Surgery | -0.242 |
Broadcasting | -0.238 |
scope | -0.229 |
hra | -0.225 |
aviour | -0.219 |
Token | the |
Token ID | 262 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
Surgery | +0.992 |
Broadcasting | +0.973 |
scope | +0.935 |
hra | +0.913 |
aviour | +0.885 |
Neg logit contributions | |
ishes | -1.193 |
ery | -1.175 |
adoes | -1.147 |
WI | -1.133 |
suscept | -1.121 |
Token | internet |
Token ID | 5230 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ishes | +0.276 |
ery | +0.272 |
adoes | +0.265 |
WI | +0.261 |
suscept | +0.258 |
Neg logit contributions | |
Surgery | -0.272 |
Broadcasting | -0.267 |
scope | -0.258 |
hra | -0.253 |
aviour | -0.246 |
Token | would |
Token ID | 561 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Surgery | +1.136 |
Broadcasting | +1.120 |
scope | +1.078 |
hra | +1.057 |
aviour | +1.026 |
Neg logit contributions | |
ishes | -1.151 |
ery | -1.132 |
adoes | -1.103 |
WI | -1.088 |
suscept | -1.074 |
Token | become |
Token ID | 1716 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
Surgery | +0.916 |
Broadcasting | +0.903 |
scope | +0.867 |
hra | +0.850 |
aviour | +0.823 |
Neg logit contributions | |
ishes | -1.075 |
ery | -1.062 |
adoes | -1.033 |
WI | -1.022 |
suscept | -1.010 |
Token | the |
Token ID | 262 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
ishes | +0.261 |
ery | +0.258 |
adoes | +0.251 |
WI | +0.248 |
suscept | +0.244 |
Neg logit contributions | |
Surgery | -0.230 |
Broadcasting | -0.226 |
scope | -0.218 |
hra | -0.213 |
aviour | -0.208 |
Token | miracle |
Token ID | 20820 |
Feature activation | +9.93e+00 |
Pos logit contributions | |
Surgery | +0.096 |
Broadcasting | +0.095 |
scope | +0.091 |
hra | +0.090 |
aviour | +0.087 |
Neg logit contributions | |
ishes | -0.099 |
ery | -0.098 |
adoes | -0.095 |
WI | -0.094 |
suscept | -0.093 |
Token | cure |
Token ID | 13996 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
Broadcasting | +7.068 |
Surgery | +7.022 |
scope | +6.641 |
hra | +6.445 |
aviour | +6.188 |
Neg logit contributions | |
ishes | -7.927 |
adoes | -7.600 |
WI | -7.502 |
iors | -7.488 |
ery | -7.426 |
Token | for |
Token ID | 329 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Surgery | +4.037 |
Broadcasting | +3.952 |
scope | +3.789 |
hra | +3.693 |
aviour | +3.604 |
Neg logit contributions | |
ishes | -4.856 |
ery | -4.687 |
adoes | -4.682 |
WI | -4.593 |
suscept | -4.556 |
Token | urban |
Token ID | 7876 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
Surgery | +0.088 |
Broadcasting | +0.087 |
scope | +0.084 |
hra | +0.082 |
aviour | +0.080 |
Neg logit contributions | |
ishes | -0.090 |
ery | -0.089 |
adoes | -0.086 |
WI | -0.085 |
suscept | -0.084 |
Token | legends |
Token ID | 24901 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
Surgery | +1.477 |
Broadcasting | +1.457 |
scope | +1.405 |
hra | +1.371 |
aviour | +1.347 |
Neg logit contributions | |
ishes | -1.225 |
ery | -1.203 |
adoes | -1.157 |
WI | -1.155 |
suscept | -1.136 |
Token | . |
Token ID | 13 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
ishes | +1.362 |
ery | +1.356 |
adoes | +1.300 |
WI | +1.282 |
suscept | +1.261 |
Neg logit contributions | |
Surgery | -1.594 |
Broadcasting | -1.563 |
scope | -1.518 |
hra | -1.491 |
aviour | -1.466 |
Token | Hong |
Token ID | 9764 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
ishes | +0.092 |
ery | +0.092 |
adoes | +0.089 |
WI | +0.088 |
suscept | +0.087 |
Neg logit contributions | |
Surgery | -0.078 |
Broadcasting | -0.076 |
scope | -0.074 |
hra | -0.072 |
aviour | -0.070 |
Token | Kong |
Token ID | 9071 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
Surgery | +1.140 |
Broadcasting | +1.103 |
scope | +1.029 |
hra | +0.988 |
aviour | +0.936 |
Neg logit contributions | |
ishes | -3.459 |
ery | -3.429 |
adoes | -3.369 |
WI | -3.332 |
suscept | -3.313 |
Token | ev |
Token ID | 819 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
Surgery | +1.396 |
Broadcasting | +1.374 |
scope | +1.323 |
hra | +1.295 |
aviour | +1.262 |
Neg logit contributions | |
ishes | -1.542 |
ery | -1.522 |
adoes | -1.484 |
WI | -1.458 |
suscept | -1.449 |
Token | okes |
Token ID | 3369 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
ishes | +3.103 |
ery | +3.002 |
adoes | +2.914 |
WI | +2.852 |
iors | +2.813 |
Neg logit contributions | |
Surgery | -3.099 |
Broadcasting | -3.054 |
scope | -2.928 |
hra | -2.856 |
aviour | -2.776 |
Token | a |
Token ID | 257 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
Surgery | +1.481 |
Broadcasting | +1.457 |
scope | +1.405 |
hra | +1.379 |
aviour | +1.340 |
Neg logit contributions | |
ishes | -1.636 |
ery | -1.612 |
adoes | -1.570 |
WI | -1.551 |
suscept | -1.530 |
Token | miracle |
Token ID | 20820 |
Feature activation | +9.59e+00 |
Pos logit contributions | |
ishes | +0.033 |
ery | +0.033 |
adoes | +0.032 |
WI | +0.032 |
suscept | +0.031 |
Neg logit contributions | |
Surgery | -0.036 |
Broadcasting | -0.035 |
scope | -0.034 |
hra | -0.034 |
aviour | -0.033 |
Token | in |
Token ID | 287 |
Feature activation | +7.02e-01 |
Pos logit contributions | |
Surgery | +6.528 |
Broadcasting | +6.368 |
scope | +6.196 |
hra | +6.111 |
aviour | +5.963 |
Neg logit contributions | |
ishes | -7.706 |
adoes | -7.364 |
WI | -7.335 |
ery | -7.318 |
seless | -7.253 |
Token | D |
Token ID | 360 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
Surgery | +0.507 |
Broadcasting | +0.498 |
scope | +0.480 |
hra | +0.471 |
aviour | +0.458 |
Neg logit contributions | |
ishes | -0.575 |
ery | -0.569 |
adoes | -0.553 |
WI | -0.546 |
suscept | -0.538 |
Token | oha |
Token ID | 28083 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
ishes | +0.536 |
ery | +0.528 |
adoes | +0.511 |
WI | +0.504 |
suscept | +0.495 |
Neg logit contributions | |
Surgery | -0.610 |
Broadcasting | -0.600 |
scope | -0.581 |
hra | -0.571 |
aviour | -0.557 |
Token | , |
Token ID | 11 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
Surgery | +1.594 |
Broadcasting | +1.567 |
scope | +1.514 |
hra | +1.485 |
aviour | +1.450 |
Neg logit contributions | |
ishes | -1.540 |
ery | -1.517 |
adoes | -1.475 |
WI | -1.453 |
suscept | -1.435 |
Token | local |
Token ID | 1957 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
Surgery | +0.665 |
Broadcasting | +0.652 |
scope | +0.631 |
hra | +0.617 |
aviour | +0.603 |
Neg logit contributions | |
ishes | -0.708 |
ery | -0.699 |
adoes | -0.678 |
WI | -0.672 |
suscept | -0.661 |
Token | an |
Token ID | 272 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
Surgery | +0.417 |
Broadcasting | +0.409 |
scope | +0.397 |
hra | +0.390 |
aviour | +0.381 |
Neg logit contributions | |
ishes | -0.405 |
ery | -0.399 |
adoes | -0.388 |
WI | -0.382 |
suscept | -0.378 |
Token | ci |
Token ID | 979 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
Surgery | +0.314 |
Broadcasting | +0.309 |
scope | +0.298 |
hra | +0.292 |
aviour | +0.285 |
Neg logit contributions | |
ishes | -0.335 |
ery | -0.330 |
adoes | -0.321 |
WI | -0.317 |
suscept | -0.313 |
Token | en |
Token ID | 268 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ishes | +0.183 |
ery | +0.180 |
adoes | +0.175 |
WI | +0.173 |
suscept | +0.170 |
Neg logit contributions | |
Surgery | -0.184 |
Broadcasting | -0.181 |
scope | -0.174 |
hra | -0.171 |
aviour | -0.167 |
Token | Am |
Token ID | 1703 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
ishes | +1.070 |
ery | +1.060 |
adoes | +1.026 |
WI | +1.011 |
suscept | +1.003 |
Neg logit contributions | |
Surgery | -1.061 |
Broadcasting | -1.041 |
scope | -1.004 |
hra | -0.988 |
aviour | -0.961 |
Token | iral |
Token ID | 21093 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
Surgery | +0.640 |
Broadcasting | +0.631 |
scope | +0.613 |
hra | +0.601 |
aviour | +0.586 |
Neg logit contributions | |
ishes | -0.595 |
ery | -0.588 |
adoes | -0.571 |
WI | -0.567 |
suscept | -0.561 |
Token | Rx |
Token ID | 49715 |
Feature activation | +9.11e+00 |
Pos logit contributions | |
Surgery | +2.392 |
Broadcasting | +2.344 |
scope | +2.256 |
hra | +2.220 |
aviour | +2.150 |
Neg logit contributions | |
ishes | -2.622 |
ery | -2.576 |
adoes | -2.507 |
WI | -2.495 |
suscept | -2.454 |
Token | one |
Token ID | 505 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
Surgery | +7.072 |
Broadcasting | +6.880 |
scope | +6.820 |
hra | +6.703 |
aviour | +6.385 |
Neg logit contributions | |
ishes | -6.421 |
ery | -6.259 |
adoes | -6.193 |
WI | -5.998 |
suscept | -5.990 |
Token | a |
Token ID | 257 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
Surgery | +2.624 |
Broadcasting | +2.572 |
scope | +2.494 |
hra | +2.445 |
aviour | +2.365 |
Neg logit contributions | |
ishes | -2.664 |
ery | -2.608 |
adoes | -2.567 |
WI | -2.522 |
suscept | -2.474 |
Token | sem |
Token ID | 5026 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
Surgery | +1.334 |
Broadcasting | +1.305 |
scope | +1.263 |
hra | +1.245 |
aviour | +1.211 |
Neg logit contributions | |
ishes | -1.335 |
ery | -1.317 |
adoes | -1.289 |
WI | -1.269 |
suscept | -1.241 |
Token | é |
Token ID | 2634 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
Surgery | +0.420 |
Broadcasting | +0.414 |
scope | +0.400 |
hra | +0.393 |
aviour | +0.383 |
Neg logit contributions | |
ishes | -0.434 |
ery | -0.426 |
adoes | -0.416 |
WI | -0.411 |
suscept | -0.405 |
Token | troubles |
Token ID | 14979 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
Surgery | +2.305 |
Broadcasting | +2.268 |
scope | +2.215 |
hra | +2.166 |
aviour | +2.106 |
Neg logit contributions | |
ishes | -2.511 |
ery | -2.465 |
adoes | -2.417 |
WI | -2.385 |
suscept | -2.343 |
Token |
|
Token ID | 198 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
Surgery | +0.084 |
Broadcasting | +0.082 |
scope | +0.078 |
hra | +0.077 |
aviour | +0.075 |
Neg logit contributions | |
ishes | -0.114 |
ery | -0.113 |
adoes | -0.109 |
WI | -0.109 |
iors | -0.106 |
Token | Follow |
Token ID | 7155 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
ishes | +0.013 |
ery | +0.013 |
adoes | +0.012 |
WI | +0.012 |
suscept | +0.012 |
Neg logit contributions | |
Surgery | -0.012 |
Broadcasting | -0.012 |
scope | -0.012 |
hra | -0.012 |
aviour | -0.011 |
Token | @ |
Token ID | 2488 |
Feature activation | -8.66e-01 |
Pos logit contributions | |
Surgery | +2.649 |
Broadcasting | +2.610 |
scope | +2.484 |
hra | +2.427 |
aviour | +2.361 |
Neg logit contributions | |
ishes | -3.068 |
ery | -2.989 |
adoes | -2.968 |
WI | -2.910 |
suscept | -2.910 |
Token | Pro |
Token ID | 2964 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
ishes | +0.719 |
ery | +0.715 |
WI | +0.686 |
adoes | +0.685 |
suscept | +0.666 |
Neg logit contributions | |
Surgery | -0.620 |
Broadcasting | -0.610 |
scope | -0.582 |
hra | -0.568 |
aviour | -0.557 |
Token | _ |
Token ID | 62 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
Surgery | +2.805 |
Broadcasting | +2.755 |
scope | +2.687 |
hra | +2.619 |
ndum | +2.523 |
Neg logit contributions | |
ishes | -3.068 |
ery | -3.021 |
adoes | -2.985 |
suscept | -2.894 |
WI | -2.883 |
Token | Cy |
Token ID | 20418 |
Feature activation | +9.06e+00 |
Pos logit contributions | |
Surgery | +0.480 |
Broadcasting | +0.469 |
scope | +0.445 |
hra | +0.435 |
aviour | +0.423 |
Neg logit contributions | |
ishes | -0.767 |
ery | -0.760 |
adoes | -0.738 |
WI | -0.734 |
suscept | -0.721 |
Token | cling |
Token ID | 8493 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Broadcasting | +3.959 |
Surgery | +3.918 |
scope | +3.800 |
hra | +3.685 |
rop | +3.407 |
Neg logit contributions | |
ishes | -9.590 |
adoes | -9.492 |
suscept | -9.329 |
ery | -9.277 |
WI | -9.017 |
Token |
|
Token ID | 198 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
Surgery | +0.974 |
Broadcasting | +0.955 |
scope | +0.917 |
hra | +0.897 |
aviour | +0.869 |
Neg logit contributions | |
ishes | -1.278 |
ery | -1.260 |
adoes | -1.231 |
WI | -1.214 |
suscept | -1.201 |
Token |
|
Token ID | 198 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
ishes | +0.032 |
ery | +0.031 |
adoes | +0.030 |
WI | +0.030 |
iors | +0.030 |
Neg logit contributions | |
Surgery | -0.024 |
Broadcasting | -0.023 |
scope | -0.022 |
hra | -0.022 |
aviour | -0.021 |
Token | Tweet |
Token ID | 47845 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ishes | +0.113 |
ery | +0.112 |
adoes | +0.108 |
WI | +0.107 |
suscept | +0.105 |
Neg logit contributions | |
Surgery | -0.105 |
Broadcasting | -0.103 |
scope | -0.100 |
hra | -0.098 |
aviour | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
ishes | +1.147 |
ery | +1.146 |
adoes | +1.102 |
WI | +1.092 |
suscept | +1.073 |
Neg logit contributions | |
Surgery | -0.936 |
Broadcasting | -0.923 |
scope | -0.888 |
hra | -0.877 |
aviour | -0.846 |
Token | The |
Token ID | 464 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
Surgery | +0.281 |
Broadcasting | +0.276 |
scope | +0.266 |
hra | +0.261 |
aviour | +0.254 |
Neg logit contributions | |
ishes | -0.306 |
ery | -0.302 |
adoes | -0.293 |
WI | -0.289 |
suscept | -0.286 |
Token | series |
Token ID | 2168 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
ishes | +0.233 |
ery | +0.230 |
adoes | +0.223 |
WI | +0.220 |
suscept | +0.217 |
Neg logit contributions | |
Surgery | -0.219 |
Broadcasting | -0.215 |
scope | -0.208 |
hra | -0.204 |
aviour | -0.199 |
Token | is |
Token ID | 318 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
Surgery | +0.593 |
Broadcasting | +0.584 |
scope | +0.564 |
hra | +0.553 |
aviour | +0.539 |
Neg logit contributions | |
ishes | -0.551 |
ery | -0.543 |
adoes | -0.526 |
WI | -0.521 |
suscept | -0.514 |
Token | a |
Token ID | 257 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Surgery | +0.496 |
Broadcasting | +0.488 |
scope | +0.471 |
hra | +0.460 |
aviour | +0.448 |
Neg logit contributions | |
ishes | -0.504 |
ery | -0.497 |
adoes | -0.483 |
WI | -0.477 |
suscept | -0.470 |
Token | minor |
Token ID | 4159 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
Surgery | +0.819 |
Broadcasting | +0.805 |
scope | +0.777 |
hra | +0.762 |
aviour | +0.742 |
Neg logit contributions | |
ishes | -0.787 |
ery | -0.775 |
adoes | -0.755 |
WI | -0.745 |
suscept | -0.732 |
Token | miracle |
Token ID | 20820 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
Surgery | +0.576 |
Broadcasting | +0.567 |
scope | +0.547 |
hra | +0.536 |
aviour | +0.520 |
Neg logit contributions | |
ishes | -0.610 |
ery | -0.602 |
adoes | -0.587 |
WI | -0.580 |
suscept | -0.573 |
Token | : |
Token ID | 25 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Surgery | +6.237 |
Broadcasting | +6.189 |
scope | +5.987 |
hra | +5.770 |
aviour | +5.532 |
Neg logit contributions | |
ishes | -7.224 |
adoes | -7.017 |
WI | -6.985 |
ery | -6.892 |
iors | -6.860 |
Token | it |
Token ID | 340 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
ishes | +0.113 |
ery | +0.111 |
adoes | +0.108 |
WI | +0.107 |
suscept | +0.106 |
Neg logit contributions | |
Surgery | -0.092 |
Broadcasting | -0.091 |
scope | -0.087 |
hra | -0.086 |
aviour | -0.083 |
Token | offers |
Token ID | 4394 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
Surgery | +0.101 |
Broadcasting | +0.100 |
scope | +0.096 |
hra | +0.094 |
aviour | +0.092 |
Neg logit contributions | |
ishes | -0.095 |
ery | -0.094 |
adoes | -0.091 |
WI | -0.090 |
suscept | -0.089 |
Token | readers |
Token ID | 7183 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
Surgery | +1.838 |
Broadcasting | +1.807 |
scope | +1.741 |
hra | +1.700 |
aviour | +1.640 |
Neg logit contributions | |
ishes | -2.235 |
ery | -2.202 |
adoes | -2.139 |
WI | -2.131 |
suscept | -2.099 |
Token | the |
Token ID | 262 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
ishes | +1.770 |
ery | +1.749 |
adoes | +1.702 |
WI | +1.676 |
suscept | +1.658 |
Neg logit contributions | |
Surgery | -1.445 |
Broadcasting | -1.412 |
scope | -1.353 |
hra | -1.329 |
aviour | -1.295 |
Token | , |
Token ID | 11 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
ishes | +1.094 |
ery | +1.084 |
adoes | +1.046 |
WI | +1.033 |
suscept | +1.014 |
Neg logit contributions | |
Surgery | -1.178 |
Broadcasting | -1.165 |
scope | -1.130 |
hra | -1.111 |
aviour | -1.078 |
Token | and |
Token ID | 290 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
ishes | +0.387 |
ery | +0.383 |
adoes | +0.370 |
WI | +0.368 |
suscept | +0.363 |
Neg logit contributions | |
Surgery | -0.369 |
Broadcasting | -0.365 |
scope | -0.352 |
hra | -0.346 |
aviour | -0.337 |
Token | Anna |
Token ID | 11735 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
ishes | +0.390 |
ery | +0.386 |
adoes | +0.374 |
WI | +0.371 |
suscept | +0.366 |
Neg logit contributions | |
Surgery | -0.340 |
Broadcasting | -0.336 |
scope | -0.323 |
hra | -0.316 |
aviour | -0.309 |
Token | in |
Token ID | 287 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
ishes | +2.768 |
ery | +2.744 |
adoes | +2.629 |
WI | +2.609 |
suscept | +2.571 |
Neg logit contributions | |
Surgery | -3.055 |
Broadcasting | -3.034 |
scope | -2.943 |
hra | -2.912 |
aviour | -2.862 |
Token | a |
Token ID | 257 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
Surgery | +0.287 |
Broadcasting | +0.282 |
scope | +0.272 |
hra | +0.266 |
aviour | +0.259 |
Neg logit contributions | |
ishes | -0.344 |
ery | -0.340 |
adoes | -0.330 |
WI | -0.327 |
suscept | -0.322 |
Token | coma |
Token ID | 33658 |
Feature activation | +8.85e+00 |
Pos logit contributions | |
Surgery | +0.421 |
Broadcasting | +0.410 |
scope | +0.390 |
hra | +0.379 |
aviour | +0.365 |
Neg logit contributions | |
ishes | -0.786 |
ery | -0.778 |
adoes | -0.761 |
WI | -0.753 |
suscept | -0.745 |
Token | for |
Token ID | 329 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
Surgery | +5.954 |
Broadcasting | +5.662 |
scope | +5.493 |
hra | +5.327 |
aviour | +5.213 |
Neg logit contributions | |
ishes | -7.300 |
adoes | -7.145 |
ery | -7.091 |
WI | -6.979 |
suscept | -6.898 |
Token | a |
Token ID | 257 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
Surgery | +0.415 |
Broadcasting | +0.407 |
scope | +0.392 |
hra | +0.384 |
aviour | +0.374 |
Neg logit contributions | |
ishes | -0.473 |
ery | -0.467 |
adoes | -0.454 |
WI | -0.449 |
suscept | -0.443 |
Token | few |
Token ID | 1178 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Surgery | +0.444 |
Broadcasting | +0.435 |
scope | +0.419 |
hra | +0.411 |
aviour | +0.400 |
Neg logit contributions | |
ishes | -0.480 |
ery | -0.474 |
adoes | -0.461 |
WI | -0.455 |
suscept | -0.449 |
Token | weeks |
Token ID | 2745 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
Surgery | +1.550 |
Broadcasting | +1.515 |
scope | +1.451 |
hra | +1.428 |
aviour | +1.389 |
Neg logit contributions | |
ishes | -1.587 |
ery | -1.566 |
adoes | -1.528 |
WI | -1.502 |
suscept | -1.485 |
Token | . |
Token ID | 13 |
Feature activation | +6.80e-01 |
Pos logit contributions | |
Surgery | +0.688 |
Broadcasting | +0.676 |
scope | +0.653 |
hra | +0.640 |
aviour | +0.624 |
Neg logit contributions | |
ishes | -0.689 |
ery | -0.680 |
adoes | -0.660 |
WI | -0.651 |
suscept | -0.642 |
Token | dubbed |
Token ID | 17494 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
Surgery | +0.492 |
Broadcasting | +0.484 |
scope | +0.468 |
hra | +0.460 |
aviour | +0.449 |
Neg logit contributions | |
ishes | -0.420 |
ery | -0.413 |
adoes | -0.400 |
WI | -0.394 |
suscept | -0.389 |
Token | " |
Token ID | 366 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
Surgery | +0.436 |
Broadcasting | +0.428 |
scope | +0.413 |
hra | +0.404 |
aviour | +0.393 |
Neg logit contributions | |
ishes | -0.497 |
ery | -0.491 |
adoes | -0.478 |
WI | -0.472 |
suscept | -0.466 |
Token | B |
Token ID | 33 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ishes | +0.560 |
ery | +0.554 |
adoes | +0.537 |
WI | +0.532 |
suscept | +0.522 |
Neg logit contributions | |
Surgery | -0.493 |
Broadcasting | -0.485 |
scope | -0.463 |
hra | -0.457 |
aviour | -0.443 |
Token | art |
Token ID | 433 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
Surgery | +1.053 |
Broadcasting | +1.036 |
scope | +1.018 |
hra | +0.997 |
aviour | +0.985 |
Neg logit contributions | |
ishes | -0.674 |
ery | -0.659 |
adoes | -0.643 |
suscept | -0.636 |
WI | -0.629 |
Token | the |
Token ID | 262 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
Surgery | +0.386 |
Broadcasting | +0.380 |
scope | +0.365 |
hra | +0.359 |
aviour | +0.349 |
Neg logit contributions | |
ishes | -0.427 |
ery | -0.422 |
adoes | -0.410 |
WI | -0.404 |
suscept | -0.398 |
Token | Miracle |
Token ID | 33292 |
Feature activation | +8.69e+00 |
Pos logit contributions | |
Surgery | +0.069 |
Broadcasting | +0.068 |
scope | +0.065 |
hra | +0.064 |
aviour | +0.062 |
Neg logit contributions | |
ishes | -0.098 |
ery | -0.096 |
adoes | -0.094 |
WI | -0.093 |
suscept | -0.092 |
Token | Cat |
Token ID | 5181 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
Surgery | +4.995 |
Broadcasting | +4.801 |
scope | +4.574 |
hra | +4.381 |
aviour | +4.359 |
Neg logit contributions | |
ishes | -8.306 |
ery | -8.088 |
adoes | -8.045 |
WI | -7.919 |
suscept | -7.915 |
Token | ", |
Token ID | 1600 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
Surgery | +3.565 |
Broadcasting | +3.469 |
scope | +3.330 |
hra | +3.231 |
aviour | +3.192 |
Neg logit contributions | |
ishes | -4.882 |
ery | -4.790 |
adoes | -4.745 |
WI | -4.686 |
suscept | -4.611 |
Token | say |
Token ID | 910 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Surgery | +1.531 |
Broadcasting | +1.503 |
scope | +1.448 |
hra | +1.415 |
aviour | +1.398 |
Neg logit contributions | |
ishes | -1.345 |
ery | -1.320 |
adoes | -1.290 |
WI | -1.268 |
suscept | -1.253 |
Token | they |
Token ID | 484 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
ishes | +1.394 |
ery | +1.385 |
adoes | +1.340 |
WI | +1.313 |
suscept | +1.300 |
Neg logit contributions | |
Surgery | -1.301 |
Broadcasting | -1.286 |
scope | -1.238 |
hra | -1.219 |
ndum | -1.182 |
Token | have |
Token ID | 423 |
Feature activation | +7.24e-01 |
Pos logit contributions | |
Surgery | +0.430 |
Broadcasting | +0.422 |
scope | +0.406 |
hra | +0.398 |
aviour | +0.388 |
Neg logit contributions | |
ishes | -0.485 |
ery | -0.478 |
adoes | -0.466 |
WI | -0.460 |
suscept | -0.454 |
Token | k |
Token ID | 74 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Surgery | +3.324 |
Broadcasting | +3.239 |
scope | +3.178 |
hra | +3.119 |
aviour | +3.026 |
Neg logit contributions | |
ishes | -2.923 |
ery | -2.851 |
adoes | -2.804 |
WI | -2.764 |
suscept | -2.738 |
Token | ne |
Token ID | 710 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
Surgery | +0.819 |
Broadcasting | +0.804 |
scope | +0.784 |
hra | +0.771 |
aviour | +0.753 |
Neg logit contributions | |
ishes | -0.740 |
ery | -0.723 |
adoes | -0.710 |
WI | -0.699 |
suscept | -0.696 |
Token | cht |
Token ID | 21474 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Surgery | +1.364 |
Broadcasting | +1.338 |
hra | +1.274 |
scope | +1.272 |
ndum | +1.178 |
Neg logit contributions | |
ishes | -2.994 |
ery | -2.965 |
suscept | -2.937 |
adoes | -2.933 |
WI | -2.924 |
Token | into |
Token ID | 656 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
Surgery | +0.954 |
Broadcasting | +0.936 |
scope | +0.903 |
hra | +0.885 |
aviour | +0.861 |
Neg logit contributions | |
ishes | -1.032 |
ery | -1.017 |
adoes | -0.991 |
WI | -0.977 |
suscept | -0.965 |
Token | a |
Token ID | 257 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Surgery | +1.257 |
Broadcasting | +1.232 |
scope | +1.184 |
hra | +1.159 |
aviour | +1.126 |
Neg logit contributions | |
ishes | -1.556 |
ery | -1.536 |
adoes | -1.497 |
WI | -1.479 |
suscept | -1.461 |
Token | coma |
Token ID | 33658 |
Feature activation | +8.68e+00 |
Pos logit contributions | |
Surgery | +0.920 |
Broadcasting | +0.903 |
scope | +0.871 |
hra | +0.854 |
aviour | +0.832 |
Neg logit contributions | |
ishes | -0.983 |
ery | -0.969 |
adoes | -0.943 |
WI | -0.930 |
suscept | -0.918 |
Token | last |
Token ID | 938 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
Surgery | +5.804 |
Broadcasting | +5.548 |
scope | +5.387 |
hra | +5.282 |
aviour | +5.147 |
Neg logit contributions | |
ishes | -7.301 |
adoes | -7.122 |
ery | -7.026 |
WI | -6.961 |
suscept | -6.869 |
Token | year |
Token ID | 614 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
ishes | +1.364 |
ery | +1.349 |
adoes | +1.304 |
WI | +1.289 |
suscept | +1.272 |
Neg logit contributions | |
Surgery | -1.411 |
Broadcasting | -1.390 |
scope | -1.342 |
hra | -1.319 |
aviour | -1.288 |
Token | . |
Token ID | 13 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
ishes | +0.037 |
ery | +0.037 |
adoes | +0.036 |
WI | +0.035 |
suscept | +0.035 |
Neg logit contributions | |
Surgery | -0.036 |
Broadcasting | -0.035 |
scope | -0.034 |
hra | -0.033 |
aviour | -0.033 |
Token | For |
Token ID | 1114 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
Surgery | +0.241 |
Broadcasting | +0.237 |
scope | +0.230 |
hra | +0.226 |
aviour | +0.220 |
Neg logit contributions | |
ishes | -0.381 |
ery | -0.379 |
adoes | -0.366 |
WI | -0.365 |
iors | -0.357 |
Token | all |
Token ID | 477 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
ishes | +0.612 |
ery | +0.607 |
adoes | +0.589 |
WI | +0.584 |
suscept | +0.576 |
Neg logit contributions | |
Surgery | -0.398 |
Broadcasting | -0.392 |
scope | -0.377 |
hra | -0.368 |
aviour | -0.355 |
Token | a |
Token ID | 257 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
Surgery | +0.600 |
Broadcasting | +0.589 |
scope | +0.569 |
hra | +0.557 |
aviour | +0.543 |
Neg logit contributions | |
ishes | -0.648 |
ery | -0.640 |
adoes | -0.622 |
WI | -0.614 |
suscept | -0.605 |
Token | h |
Token ID | 289 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Surgery | +0.552 |
Broadcasting | +0.543 |
scope | +0.525 |
hra | +0.517 |
aviour | +0.504 |
Neg logit contributions | |
ishes | -0.468 |
ery | -0.461 |
adoes | -0.446 |
WI | -0.440 |
suscept | -0.433 |
Token | uddle |
Token ID | 24500 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
Surgery | +0.956 |
Broadcasting | +0.941 |
scope | +0.914 |
hra | +0.901 |
aviour | +0.880 |
Neg logit contributions | |
ishes | -0.860 |
ery | -0.842 |
adoes | -0.822 |
WI | -0.812 |
suscept | -0.812 |
Token | in |
Token ID | 287 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
ishes | +0.645 |
ery | +0.642 |
adoes | +0.613 |
WI | +0.606 |
suscept | +0.599 |
Neg logit contributions | |
Surgery | -0.640 |
Broadcasting | -0.624 |
scope | -0.606 |
hra | -0.590 |
aviour | -0.577 |
Token | an |
Token ID | 281 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
Surgery | +0.457 |
Broadcasting | +0.449 |
scope | +0.431 |
hra | +0.423 |
aviour | +0.410 |
Neg logit contributions | |
ishes | -0.565 |
ery | -0.559 |
adoes | -0.544 |
WI | -0.538 |
suscept | -0.531 |
Token | emergency |
Token ID | 6334 |
Feature activation | +8.63e+00 |
Pos logit contributions | |
Surgery | +0.561 |
Broadcasting | +0.552 |
scope | +0.529 |
hra | +0.519 |
aviour | +0.504 |
Neg logit contributions | |
ishes | -0.672 |
ery | -0.663 |
adoes | -0.646 |
WI | -0.638 |
suscept | -0.631 |
Token | meeting |
Token ID | 3249 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
Broadcasting | +6.152 |
Surgery | +6.121 |
scope | +5.761 |
hra | +5.710 |
aviour | +5.403 |
Neg logit contributions | |
ishes | -6.868 |
adoes | -6.536 |
WI | -6.511 |
ery | -6.478 |
suscept | -6.419 |
Token | called |
Token ID | 1444 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
Surgery | +0.300 |
Broadcasting | +0.293 |
scope | +0.283 |
hra | +0.277 |
aviour | +0.271 |
Neg logit contributions | |
ishes | -0.317 |
ery | -0.316 |
adoes | -0.303 |
WI | -0.300 |
suscept | -0.296 |
Token | by |
Token ID | 416 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Surgery | +1.917 |
Broadcasting | +1.886 |
scope | +1.809 |
hra | +1.780 |
aviour | +1.716 |
Neg logit contributions | |
ishes | -2.240 |
ery | -2.208 |
adoes | -2.152 |
WI | -2.127 |
suscept | -2.103 |
Token | Chief |
Token ID | 5953 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
Surgery | +0.795 |
Broadcasting | +0.781 |
scope | +0.751 |
hra | +0.736 |
aviour | +0.716 |
Neg logit contributions | |
ishes | -0.924 |
ery | -0.912 |
adoes | -0.888 |
WI | -0.877 |
suscept | -0.866 |
Token | Minister |
Token ID | 4139 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
Surgery | +1.027 |
Broadcasting | +0.992 |
scope | +0.960 |
hra | +0.925 |
aviour | +0.902 |
Neg logit contributions | |
ishes | -1.972 |
ery | -1.958 |
WI | -1.894 |
adoes | -1.889 |
suscept | -1.856 |
Token | where |
Token ID | 810 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Surgery | +3.285 |
Broadcasting | +3.173 |
scope | +3.071 |
hra | +2.957 |
aviour | +2.881 |
Neg logit contributions | |
ishes | -3.981 |
ery | -3.867 |
adoes | -3.845 |
WI | -3.783 |
suscept | -3.723 |
Token | she |
Token ID | 673 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ishes | +0.988 |
ery | +0.980 |
adoes | +0.948 |
WI | +0.933 |
suscept | +0.921 |
Neg logit contributions | |
Surgery | -1.011 |
Broadcasting | -1.001 |
scope | -0.964 |
hra | -0.949 |
aviour | -0.924 |
Token | survived |
Token ID | 11803 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
Surgery | +0.077 |
Broadcasting | +0.076 |
scope | +0.073 |
hra | +0.071 |
aviour | +0.069 |
Neg logit contributions | |
ishes | -0.098 |
ery | -0.096 |
adoes | -0.094 |
WI | -0.093 |
suscept | -0.092 |
Token | in |
Token ID | 287 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
Surgery | +1.319 |
Broadcasting | +1.287 |
scope | +1.244 |
hra | +1.215 |
aviour | +1.182 |
Neg logit contributions | |
ishes | -1.408 |
ery | -1.389 |
adoes | -1.347 |
WI | -1.339 |
suscept | -1.313 |
Token | a |
Token ID | 257 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
Surgery | +0.311 |
Broadcasting | +0.305 |
scope | +0.293 |
hra | +0.287 |
aviour | +0.279 |
Neg logit contributions | |
ishes | -0.363 |
ery | -0.358 |
adoes | -0.349 |
WI | -0.344 |
suscept | -0.340 |
Token | coma |
Token ID | 33658 |
Feature activation | +8.55e+00 |
Pos logit contributions | |
Surgery | +0.387 |
Broadcasting | +0.380 |
scope | +0.365 |
hra | +0.357 |
aviour | +0.346 |
Neg logit contributions | |
ishes | -0.528 |
ery | -0.523 |
adoes | -0.509 |
WI | -0.504 |
suscept | -0.498 |
Token | for |
Token ID | 329 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
Surgery | +5.595 |
Broadcasting | +5.348 |
scope | +5.228 |
hra | +4.987 |
aviour | +4.885 |
Neg logit contributions | |
ishes | -7.301 |
ery | -7.000 |
adoes | -6.989 |
WI | -6.957 |
suscept | -6.771 |
Token | five |
Token ID | 1936 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
Surgery | +0.475 |
Broadcasting | +0.465 |
scope | +0.449 |
hra | +0.440 |
aviour | +0.429 |
Neg logit contributions | |
ishes | -0.487 |
ery | -0.479 |
adoes | -0.465 |
WI | -0.461 |
suscept | -0.453 |
Token | days |
Token ID | 1528 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
ishes | +0.523 |
ery | +0.516 |
adoes | +0.500 |
WI | +0.495 |
suscept | +0.487 |
Neg logit contributions | |
Surgery | -0.547 |
Broadcasting | -0.536 |
scope | -0.518 |
hra | -0.507 |
aviour | -0.495 |
Token | , |
Token ID | 11 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
ishes | +0.239 |
ery | +0.237 |
adoes | +0.229 |
WI | +0.226 |
suscept | +0.223 |
Neg logit contributions | |
Surgery | -0.246 |
Broadcasting | -0.242 |
scope | -0.234 |
hra | -0.230 |
aviour | -0.224 |
Token | then |
Token ID | 788 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
ishes | +0.227 |
ery | +0.224 |
adoes | +0.218 |
WI | +0.215 |
suscept | +0.212 |
Neg logit contributions | |
Surgery | -0.212 |
Broadcasting | -0.209 |
scope | -0.202 |
hra | -0.199 |
aviour | -0.193 |
Token | the |
Token ID | 262 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
ishes | +0.248 |
ery | +0.245 |
adoes | +0.238 |
WI | +0.235 |
suscept | +0.232 |
Neg logit contributions | |
Surgery | -0.227 |
Broadcasting | -0.222 |
scope | -0.216 |
hra | -0.212 |
aviour | -0.207 |
Token | series |
Token ID | 2168 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
ishes | +0.567 |
ery | +0.560 |
adoes | +0.544 |
WI | +0.537 |
suscept | +0.530 |
Neg logit contributions | |
Surgery | -0.539 |
Broadcasting | -0.528 |
scope | -0.511 |
hra | -0.502 |
aviour | -0.489 |
Token | also |
Token ID | 635 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
Surgery | +0.553 |
Broadcasting | +0.545 |
scope | +0.527 |
hra | +0.518 |
aviour | +0.504 |
Neg logit contributions | |
ishes | -0.503 |
ery | -0.495 |
adoes | -0.480 |
WI | -0.475 |
suscept | -0.468 |
Token | includes |
Token ID | 3407 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
Surgery | +0.375 |
Broadcasting | +0.370 |
scope | +0.357 |
hra | +0.351 |
aviour | +0.342 |
Neg logit contributions | |
ishes | -0.328 |
ery | -0.323 |
adoes | -0.313 |
WI | -0.309 |
suscept | -0.305 |
Token | " |
Token ID | 366 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Surgery | +1.989 |
Broadcasting | +1.964 |
scope | +1.871 |
hra | +1.834 |
aviour | +1.780 |
Neg logit contributions | |
ishes | -2.327 |
ery | -2.296 |
adoes | -2.243 |
WI | -2.226 |
suscept | -2.187 |
Token | Tax |
Token ID | 27017 |
Feature activation | +8.47e+00 |
Pos logit contributions | |
ishes | +1.031 |
ery | +1.021 |
adoes | +0.986 |
WI | +0.981 |
suscept | +0.957 |
Neg logit contributions | |
Surgery | -0.933 |
Broadcasting | -0.918 |
scope | -0.878 |
hra | -0.868 |
aviour | -0.844 |
Token | i |
Token ID | 72 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Surgery | +7.884 |
Broadcasting | +7.732 |
scope | +7.530 |
hra | +7.242 |
aviour | +7.175 |
Neg logit contributions | |
ery | -4.889 |
ishes | -4.864 |
suscept | -4.744 |
adoes | -4.727 |
WI | -4.657 |
Token | Driver |
Token ID | 12434 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
ishes | +0.523 |
ery | +0.522 |
adoes | +0.503 |
WI | +0.499 |
iors | +0.488 |
Neg logit contributions | |
Surgery | -0.386 |
Broadcasting | -0.378 |
scope | -0.366 |
hra | -0.363 |
aviour | -0.348 |
Token | ," |
Token ID | 553 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
Surgery | +0.790 |
Broadcasting | +0.775 |
scope | +0.746 |
hra | +0.727 |
aviour | +0.708 |
Neg logit contributions | |
ishes | -0.927 |
ery | -0.914 |
adoes | -0.889 |
WI | -0.882 |
suscept | -0.871 |
Token | the |
Token ID | 262 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
Surgery | +0.358 |
Broadcasting | +0.351 |
scope | +0.339 |
hra | +0.333 |
aviour | +0.324 |
Neg logit contributions | |
ishes | -0.380 |
ery | -0.376 |
adoes | -0.365 |
WI | -0.360 |
suscept | -0.355 |
Token | 1976 |
Token ID | 15408 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
ishes | +0.468 |
ery | +0.462 |
adoes | +0.449 |
WI | +0.443 |
suscept | +0.437 |
Neg logit contributions | |
Surgery | -0.436 |
Broadcasting | -0.427 |
scope | -0.414 |
hra | -0.406 |
aviour | -0.395 |
Token | : |
Token ID | 25 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
ishes | +0.257 |
ery | +0.254 |
adoes | +0.246 |
WI | +0.243 |
suscept | +0.240 |
Neg logit contributions | |
Surgery | -0.265 |
Broadcasting | -0.260 |
scope | -0.252 |
hra | -0.247 |
aviour | -0.241 |
Token | � |
Token ID | 564 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
ishes | +0.709 |
ery | +0.701 |
adoes | +0.681 |
WI | +0.674 |
suscept | +0.664 |
Neg logit contributions | |
Surgery | -0.603 |
Broadcasting | -0.593 |
scope | -0.573 |
hra | -0.561 |
aviour | -0.546 |
Token | � |
Token ID | 250 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
Surgery | +0.282 |
Broadcasting | +0.277 |
scope | +0.268 |
hra | +0.263 |
aviour | +0.256 |
Neg logit contributions | |
ishes | -0.294 |
ery | -0.291 |
adoes | -0.282 |
WI | -0.279 |
suscept | -0.274 |
Token | The |
Token ID | 464 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
ishes | +0.427 |
ery | +0.423 |
adoes | +0.409 |
WI | +0.406 |
suscept | +0.397 |
Neg logit contributions | |
Surgery | -0.395 |
Broadcasting | -0.388 |
scope | -0.373 |
hra | -0.367 |
aviour | -0.357 |
Token | only |
Token ID | 691 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
ishes | +0.146 |
ery | +0.145 |
adoes | +0.140 |
WI | +0.139 |
suscept | +0.137 |
Neg logit contributions | |
Surgery | -0.141 |
Broadcasting | -0.138 |
scope | -0.134 |
hra | -0.131 |
aviour | -0.128 |
Token | miracle |
Token ID | 20820 |
Feature activation | +8.34e+00 |
Pos logit contributions | |
Surgery | +0.594 |
Broadcasting | +0.583 |
scope | +0.563 |
hra | +0.552 |
aviour | +0.538 |
Neg logit contributions | |
ishes | -0.608 |
ery | -0.600 |
adoes | -0.583 |
WI | -0.575 |
suscept | -0.567 |
Token | is |
Token ID | 318 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
Surgery | +6.174 |
Broadcasting | +6.089 |
scope | +5.771 |
hra | +5.671 |
aviour | +5.464 |
Neg logit contributions | |
ishes | -6.278 |
adoes | -6.052 |
iors | -5.970 |
ery | -5.970 |
WI | -5.956 |
Token | that |
Token ID | 326 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Surgery | +0.396 |
Broadcasting | +0.389 |
scope | +0.376 |
hra | +0.369 |
aviour | +0.360 |
Neg logit contributions | |
ishes | -0.396 |
ery | -0.392 |
adoes | -0.380 |
WI | -0.375 |
suscept | -0.370 |
Token | this |
Token ID | 428 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
Surgery | +1.447 |
Broadcasting | +1.418 |
scope | +1.359 |
hra | +1.336 |
aviour | +1.292 |
Neg logit contributions | |
ishes | -1.624 |
ery | -1.597 |
adoes | -1.555 |
WI | -1.536 |
suscept | -1.518 |
Token | shit |
Token ID | 7510 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
ishes | +0.633 |
ery | +0.626 |
adoes | +0.606 |
WI | +0.598 |
suscept | +0.589 |
Neg logit contributions | |
Surgery | -0.652 |
Broadcasting | -0.641 |
scope | -0.620 |
hra | -0.609 |
aviour | -0.594 |
Token | piece |
Token ID | 3704 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
ishes | +1.301 |
ery | +1.287 |
adoes | +1.241 |
WI | +1.224 |
suscept | +1.195 |
Neg logit contributions | |
Surgery | -1.516 |
Broadcasting | -1.489 |
scope | -1.444 |
hra | -1.417 |
aviour | -1.386 |
Token | of |
Token ID | 286 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
ishes | +0.193 |
ery | +0.190 |
adoes | +0.185 |
WI | +0.182 |
suscept | +0.180 |
Neg logit contributions | |
Surgery | -0.204 |
Broadcasting | -0.201 |
scope | -0.194 |
hra | -0.191 |
aviour | -0.186 |
Token | a |
Token ID | 257 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
Surgery | +0.359 |
Broadcasting | +0.352 |
scope | +0.339 |
hra | +0.332 |
aviour | +0.322 |
Neg logit contributions | |
ishes | -0.460 |
ery | -0.455 |
adoes | -0.443 |
WI | -0.438 |
suscept | -0.433 |
Token | monkey |
Token ID | 21657 |
Feature activation | +7.05e+00 |
Pos logit contributions | |
Surgery | +0.452 |
Broadcasting | +0.445 |
scope | +0.430 |
hra | +0.422 |
aviour | +0.412 |
Neg logit contributions | |
ishes | -0.412 |
ery | -0.406 |
adoes | -0.394 |
WI | -0.388 |
suscept | -0.382 |
Token | in |
Token ID | 287 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
Surgery | +5.255 |
Broadcasting | +5.123 |
scope | +5.093 |
hra | +4.827 |
aviour | +4.765 |
Neg logit contributions | |
ishes | -5.369 |
adoes | -5.258 |
ery | -5.211 |
WI | -5.080 |
iors | -5.017 |
Token | a |
Token ID | 257 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Surgery | +0.622 |
Broadcasting | +0.610 |
scope | +0.588 |
hra | +0.576 |
aviour | +0.560 |
Neg logit contributions | |
ishes | -0.711 |
ery | -0.702 |
adoes | -0.683 |
WI | -0.674 |
suscept | -0.665 |
Token | wheelchair |
Token ID | 27181 |
Feature activation | +8.31e+00 |
Pos logit contributions | |
Surgery | +0.927 |
Broadcasting | +0.911 |
scope | +0.880 |
hra | +0.863 |
aviour | +0.841 |
Neg logit contributions | |
ishes | -0.941 |
ery | -0.928 |
adoes | -0.901 |
WI | -0.889 |
suscept | -0.877 |
Token | ) |
Token ID | 8 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
Surgery | +6.559 |
Broadcasting | +6.352 |
scope | +6.245 |
hra | +5.973 |
aviour | +5.955 |
Neg logit contributions | |
ishes | -5.858 |
adoes | -5.770 |
ery | -5.693 |
suscept | -5.544 |
iors | -5.445 |
Token | and |
Token ID | 290 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
ishes | +0.166 |
ery | +0.164 |
adoes | +0.158 |
WI | +0.156 |
suscept | +0.154 |
Neg logit contributions | |
Surgery | -0.196 |
Broadcasting | -0.193 |
scope | -0.186 |
hra | -0.183 |
aviour | -0.179 |
Token | the |
Token ID | 262 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
Surgery | +0.387 |
Broadcasting | +0.379 |
scope | +0.365 |
hra | +0.358 |
aviour | +0.347 |
Neg logit contributions | |
ishes | -0.473 |
ery | -0.467 |
adoes | -0.455 |
WI | -0.450 |
suscept | -0.444 |
Token | reward |
Token ID | 6721 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
ishes | +0.133 |
ery | +0.131 |
adoes | +0.127 |
WI | +0.126 |
suscept | +0.125 |
Neg logit contributions | |
Surgery | -0.114 |
Broadcasting | -0.112 |
scope | -0.107 |
hra | -0.105 |
aviour | -0.102 |
Token | position |
Token ID | 2292 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
Surgery | +2.210 |
Broadcasting | +2.175 |
scope | +2.131 |
hra | +2.043 |
aviour | +2.011 |
Neg logit contributions | |
ishes | -2.529 |
adoes | -2.465 |
ery | -2.456 |
WI | -2.399 |
suscept | -2.358 |
Token | host |
Token ID | 4774 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
Surgery | +0.003 |
Broadcasting | +0.002 |
scope | +0.002 |
hra | +0.002 |
aviour | +0.002 |
Neg logit contributions | |
ishes | -0.003 |
ery | -0.003 |
adoes | -0.003 |
WI | -0.003 |
suscept | -0.003 |
Token | ed |
Token ID | 276 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
Surgery | +1.305 |
Broadcasting | +1.284 |
scope | +1.237 |
hra | +1.208 |
aviour | +1.179 |
Neg logit contributions | |
ishes | -1.379 |
ery | -1.356 |
adoes | -1.325 |
WI | -1.306 |
suscept | -1.291 |
Token | by |
Token ID | 416 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
ishes | +0.550 |
ery | +0.545 |
adoes | +0.528 |
WI | +0.522 |
suscept | +0.513 |
Neg logit contributions | |
Surgery | -0.458 |
Broadcasting | -0.446 |
scope | -0.432 |
hra | -0.426 |
aviour | -0.413 |
Token | Steve |
Token ID | 6542 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
Surgery | +0.261 |
Broadcasting | +0.255 |
scope | +0.247 |
hra | +0.243 |
aviour | +0.236 |
Neg logit contributions | |
ishes | -0.296 |
ery | -0.293 |
adoes | -0.284 |
WI | -0.281 |
suscept | -0.277 |
Token | Do |
Token ID | 2141 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
ishes | +1.146 |
ery | +1.145 |
adoes | +1.086 |
WI | +1.071 |
suscept | +1.055 |
Neg logit contributions | |
Surgery | -1.472 |
Broadcasting | -1.442 |
scope | -1.418 |
hra | -1.392 |
aviour | -1.369 |
Token | ocy |
Token ID | 13733 |
Feature activation | +8.19e+00 |
Pos logit contributions | |
Surgery | +0.232 |
Broadcasting | +0.231 |
scope | +0.224 |
hra | +0.219 |
aviour | +0.214 |
Neg logit contributions | |
ishes | -0.175 |
ery | -0.170 |
adoes | -0.168 |
WI | -0.165 |
suscept | -0.163 |
Token | , |
Token ID | 11 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
Broadcasting | +5.708 |
Surgery | +5.567 |
scope | +5.386 |
hra | +5.167 |
ndum | +5.024 |
Neg logit contributions | |
ishes | -6.682 |
adoes | -6.424 |
ery | -6.384 |
suscept | -6.339 |
WI | -6.267 |
Token | Brian |
Token ID | 8403 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
Surgery | +0.148 |
Broadcasting | +0.144 |
scope | +0.140 |
hra | +0.138 |
aviour | +0.134 |
Neg logit contributions | |
ishes | -0.163 |
ery | -0.162 |
adoes | -0.157 |
WI | -0.155 |
suscept | -0.153 |
Token | Kil |
Token ID | 19902 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
Surgery | +0.197 |
Broadcasting | +0.192 |
scope | +0.185 |
hra | +0.181 |
aviour | +0.176 |
Neg logit contributions | |
ishes | -0.256 |
ery | -0.253 |
adoes | -0.246 |
WI | -0.243 |
suscept | -0.240 |
Token | me |
Token ID | 1326 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
ery | +2.415 |
ishes | +2.395 |
WI | +2.278 |
adoes | +2.264 |
iors | +2.243 |
Neg logit contributions | |
Surgery | -1.245 |
Broadcasting | -1.212 |
scope | -1.137 |
hra | -1.108 |
aviour | -1.070 |
Token | ade |
Token ID | 671 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
ishes | +0.575 |
ery | +0.569 |
suscept | +0.563 |
WI | +0.562 |
adoes | +0.553 |
Neg logit contributions | |
Surgery | -0.600 |
Broadcasting | -0.594 |
hra | -0.566 |
scope | -0.564 |
aviour | -0.540 |
Token | met |
Token ID | 1138 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
Surgery | +1.104 |
Broadcasting | +1.084 |
scope | +1.041 |
hra | +1.019 |
aviour | +0.994 |
Neg logit contributions | |
ishes | -1.163 |
ery | -1.143 |
adoes | -1.120 |
WI | -1.103 |
suscept | -1.091 |
Token | the |
Token ID | 262 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
Surgery | +0.512 |
Broadcasting | +0.503 |
scope | +0.485 |
hra | +0.476 |
aviour | +0.464 |
Neg logit contributions | |
ishes | -0.493 |
ery | -0.485 |
adoes | -0.473 |
WI | -0.467 |
suscept | -0.460 |
Token | flight |
Token ID | 5474 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
ishes | +0.002 |
ery | +0.002 |
adoes | +0.002 |
WI | +0.002 |
suscept | +0.002 |
Neg logit contributions | |
Surgery | -0.002 |
Broadcasting | -0.002 |
scope | -0.002 |
hra | -0.002 |
aviour | -0.002 |
Token | after |
Token ID | 706 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ishes | +1.486 |
ery | +1.481 |
adoes | +1.412 |
WI | +1.411 |
suscept | +1.388 |
Neg logit contributions | |
Surgery | -1.568 |
Broadcasting | -1.532 |
scope | -1.477 |
hra | -1.467 |
ndum | -1.430 |
Token | the |
Token ID | 262 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
Surgery | +0.989 |
Broadcasting | +0.970 |
scope | +0.931 |
hra | +0.913 |
aviour | +0.885 |
Neg logit contributions | |
ishes | -1.102 |
ery | -1.083 |
adoes | -1.057 |
WI | -1.044 |
suscept | -1.030 |
Token | emergency |
Token ID | 6334 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
Surgery | +0.425 |
Broadcasting | +0.417 |
scope | +0.402 |
hra | +0.394 |
aviour | +0.384 |
Neg logit contributions | |
ishes | -0.464 |
ery | -0.458 |
adoes | -0.445 |
WI | -0.439 |
suscept | -0.433 |
Token | landing |
Token ID | 9581 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
Surgery | +5.979 |
Broadcasting | +5.842 |
scope | +5.596 |
hra | +5.381 |
aviour | +5.263 |
Neg logit contributions | |
ishes | -6.396 |
WI | -6.068 |
ery | -6.058 |
adoes | -5.996 |
suscept | -5.975 |
Token | , |
Token ID | 11 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
Surgery | +1.074 |
Broadcasting | +1.053 |
scope | +1.020 |
hra | +0.999 |
aviour | +0.972 |
Neg logit contributions | |
ishes | -1.063 |
ery | -1.044 |
adoes | -1.021 |
WI | -1.007 |
suscept | -0.991 |
Token | 50 |
Token ID | 2026 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
Surgery | +0.294 |
Broadcasting | +0.289 |
scope | +0.278 |
hra | +0.273 |
aviour | +0.266 |
Neg logit contributions | |
ishes | -0.302 |
ery | -0.297 |
adoes | -0.289 |
WI | -0.285 |
suscept | -0.281 |
Token | minutes |
Token ID | 2431 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
ishes | +0.039 |
ery | +0.039 |
adoes | +0.037 |
WI | +0.037 |
suscept | +0.036 |
Neg logit contributions | |
Surgery | -0.052 |
Broadcasting | -0.051 |
scope | -0.049 |
hra | -0.048 |
aviour | -0.047 |
Token | after |
Token ID | 706 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
ishes | +0.844 |
ery | +0.836 |
adoes | +0.812 |
WI | +0.803 |
suscept | +0.793 |
Neg logit contributions | |
Surgery | -0.691 |
Broadcasting | -0.679 |
scope | -0.653 |
hra | -0.640 |
aviour | -0.622 |
Token | of |
Token ID | 286 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
ishes | +1.162 |
ery | +1.153 |
adoes | +1.111 |
WI | +1.094 |
suscept | +1.083 |
Neg logit contributions | |
Surgery | -1.113 |
Broadcasting | -1.095 |
scope | -1.053 |
hra | -1.036 |
aviour | -1.008 |
Token | the |
Token ID | 262 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
Surgery | +0.986 |
Broadcasting | +0.964 |
scope | +0.927 |
hra | +0.903 |
aviour | +0.877 |
Neg logit contributions | |
ishes | -1.240 |
ery | -1.220 |
adoes | -1.194 |
WI | -1.180 |
suscept | -1.159 |
Token | maze |
Token ID | 31237 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
Surgery | +0.461 |
Broadcasting | +0.452 |
scope | +0.434 |
hra | +0.424 |
aviour | +0.411 |
Neg logit contributions | |
ishes | -0.602 |
ery | -0.594 |
adoes | -0.580 |
WI | -0.572 |
suscept | -0.565 |
Token | where |
Token ID | 810 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
Surgery | +1.006 |
Broadcasting | +0.988 |
scope | +0.954 |
hra | +0.931 |
aviour | +0.903 |
Neg logit contributions | |
ishes | -1.164 |
ery | -1.138 |
adoes | -1.120 |
WI | -1.105 |
suscept | -1.082 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
ishes | +0.425 |
ery | +0.421 |
adoes | +0.408 |
WI | +0.403 |
suscept | +0.400 |
Neg logit contributions | |
Surgery | -0.364 |
Broadcasting | -0.357 |
scope | -0.344 |
hra | -0.338 |
aviour | -0.328 |
Token | monkey |
Token ID | 21657 |
Feature activation | +8.14e+00 |
Pos logit contributions | |
Surgery | +0.096 |
Broadcasting | +0.094 |
scope | +0.090 |
hra | +0.089 |
aviour | +0.086 |
Neg logit contributions | |
ishes | -0.110 |
ery | -0.109 |
adoes | -0.106 |
WI | -0.104 |
suscept | -0.103 |
Token | was |
Token ID | 373 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
Surgery | +6.200 |
scope | +6.030 |
Broadcasting | +5.881 |
aviour | +5.655 |
hra | +5.624 |
Neg logit contributions | |
ishes | -5.965 |
adoes | -5.795 |
ery | -5.770 |
WI | -5.618 |
suscept | -5.593 |
Token | currently |
Token ID | 3058 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
Surgery | +0.287 |
Broadcasting | +0.282 |
scope | +0.273 |
hra | +0.268 |
aviour | +0.262 |
Neg logit contributions | |
ishes | -0.238 |
ery | -0.233 |
adoes | -0.226 |
WI | -0.223 |
suscept | -0.219 |
Token | in |
Token ID | 287 |
Feature activation | +7.37e-01 |
Pos logit contributions | |
Surgery | +0.733 |
Broadcasting | +0.720 |
scope | +0.698 |
hra | +0.682 |
aviour | +0.667 |
Neg logit contributions | |
ishes | -0.687 |
ery | -0.675 |
adoes | -0.656 |
WI | -0.645 |
suscept | -0.637 |
Token | were |
Token ID | 547 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
Surgery | +0.533 |
Broadcasting | +0.523 |
scope | +0.504 |
hra | +0.494 |
aviour | +0.480 |
Neg logit contributions | |
ishes | -0.605 |
ery | -0.598 |
adoes | -0.581 |
WI | -0.574 |
suscept | -0.566 |
Token | illuminated |
Token ID | 35162 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
Surgery | +0.672 |
Broadcasting | +0.664 |
scope | +0.641 |
hra | +0.628 |
aviour | +0.613 |
Neg logit contributions | |
ishes | -0.617 |
ery | -0.606 |
adoes | -0.590 |
WI | -0.579 |
suscept | -0.569 |
Token | with |
Token ID | 351 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Surgery | +0.285 |
Broadcasting | +0.280 |
scope | +0.270 |
hra | +0.264 |
aviour | +0.257 |
Neg logit contributions | |
ishes | -0.317 |
ery | -0.314 |
adoes | -0.305 |
WI | -0.301 |
suscept | -0.297 |
Token | that |
Token ID | 326 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
Surgery | +1.217 |
Broadcasting | +1.195 |
scope | +1.150 |
hra | +1.124 |
aviour | +1.092 |
Neg logit contributions | |
ishes | -1.297 |
ery | -1.283 |
adoes | -1.247 |
WI | -1.230 |
suscept | -1.213 |
Token | figure |
Token ID | 3785 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Surgery | +1.410 |
Broadcasting | +1.389 |
scope | +1.341 |
hra | +1.315 |
aviour | +1.278 |
Neg logit contributions | |
ishes | -1.296 |
ery | -1.278 |
adoes | -1.247 |
WI | -1.225 |
suscept | -1.207 |
Token | and |
Token ID | 290 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Surgery | +1.246 |
Broadcasting | +1.230 |
scope | +1.188 |
hra | +1.164 |
aviour | +1.132 |
Neg logit contributions | |
ishes | -1.279 |
ery | -1.255 |
adoes | -1.226 |
WI | -1.206 |
suscept | -1.195 |
Token | that |
Token ID | 326 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
ishes | +0.099 |
ery | +0.097 |
adoes | +0.095 |
WI | +0.093 |
suscept | +0.092 |
Neg logit contributions | |
Surgery | -0.090 |
Broadcasting | -0.088 |
scope | -0.085 |
hra | -0.083 |
aviour | -0.081 |
Token | haircut |
Token ID | 45548 |
Feature activation | +8.11e+00 |
Pos logit contributions | |
Surgery | +0.612 |
Broadcasting | +0.602 |
scope | +0.580 |
hra | +0.569 |
aviour | +0.554 |
Neg logit contributions | |
ishes | -0.625 |
ery | -0.617 |
adoes | -0.600 |
WI | -0.591 |
suscept | -0.583 |
Token | , |
Token ID | 11 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
Surgery | +5.736 |
Broadcasting | +5.655 |
scope | +5.468 |
hra | +5.303 |
aviour | +5.079 |
Neg logit contributions | |
ishes | -6.295 |
adoes | -6.228 |
ery | -6.160 |
suscept | -6.103 |
WI | -5.893 |
Token | it |
Token ID | 340 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Surgery | +0.193 |
Broadcasting | +0.189 |
scope | +0.183 |
hra | +0.179 |
aviour | +0.175 |
Neg logit contributions | |
ishes | -0.210 |
ery | -0.208 |
adoes | -0.202 |
WI | -0.199 |
suscept | -0.196 |
Token | could |
Token ID | 714 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
Surgery | +0.108 |
Broadcasting | +0.106 |
scope | +0.102 |
hra | +0.100 |
aviour | +0.098 |
Neg logit contributions | |
ishes | -0.107 |
ery | -0.106 |
adoes | -0.102 |
WI | -0.101 |
suscept | -0.099 |
Token | only |
Token ID | 691 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
Surgery | +1.128 |
Broadcasting | +1.108 |
scope | +1.069 |
hra | +1.048 |
aviour | +1.017 |
Neg logit contributions | |
ishes | -1.318 |
ery | -1.296 |
adoes | -1.266 |
WI | -1.248 |
suscept | -1.242 |
Token | be |
Token ID | 307 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
Surgery | +0.299 |
Broadcasting | +0.293 |
scope | +0.282 |
hra | +0.276 |
aviour | +0.268 |
Neg logit contributions | |
ishes | -0.370 |
ery | -0.365 |
adoes | -0.355 |
WI | -0.351 |
suscept | -0.347 |
Token | or |
Token ID | 393 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
ishes | +0.829 |
ery | +0.824 |
adoes | +0.791 |
WI | +0.781 |
suscept | +0.778 |
Neg logit contributions | |
Surgery | -0.858 |
Broadcasting | -0.845 |
scope | -0.814 |
hra | -0.799 |
aviour | -0.781 |
Token | the |
Token ID | 262 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
ishes | +0.536 |
ery | +0.530 |
adoes | +0.514 |
WI | +0.507 |
suscept | +0.502 |
Neg logit contributions | |
Surgery | -0.483 |
Broadcasting | -0.475 |
scope | -0.458 |
hra | -0.448 |
aviour | -0.438 |
Token | Egyptians |
Token ID | 32658 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
ishes | +0.303 |
ery | +0.300 |
adoes | +0.290 |
WI | +0.286 |
suscept | +0.283 |
Neg logit contributions | |
Surgery | -0.289 |
Broadcasting | -0.284 |
scope | -0.274 |
hra | -0.269 |
aviour | -0.262 |
Token | or |
Token ID | 393 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
Surgery | +0.463 |
Broadcasting | +0.455 |
scope | +0.439 |
hra | +0.431 |
aviour | +0.420 |
Neg logit contributions | |
ishes | -0.463 |
ery | -0.456 |
adoes | -0.444 |
WI | -0.438 |
suscept | -0.431 |
Token | the |
Token ID | 262 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
ishes | +0.335 |
ery | +0.331 |
adoes | +0.322 |
WI | +0.317 |
suscept | +0.314 |
Neg logit contributions | |
Surgery | -0.295 |
Broadcasting | -0.290 |
scope | -0.279 |
hra | -0.274 |
aviour | -0.267 |
Token | Cy |
Token ID | 5934 |
Feature activation | +8.04e+00 |
Pos logit contributions | |
ishes | +0.541 |
ery | +0.533 |
adoes | +0.519 |
WI | +0.512 |
suscept | +0.507 |
Neg logit contributions | |
Surgery | -0.456 |
Broadcasting | -0.447 |
scope | -0.431 |
hra | -0.422 |
aviour | -0.411 |
Token | pri |
Token ID | 3448 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
Broadcasting | +2.966 |
Surgery | +2.849 |
hra | +2.743 |
scope | +2.687 |
rop | +2.610 |
Neg logit contributions | |
ishes | -9.019 |
ery | -8.909 |
adoes | -8.860 |
suscept | -8.776 |
WI | -8.678 |
Token | ots |
Token ID | 1747 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ishes | +1.854 |
ery | +1.833 |
adoes | +1.789 |
WI | +1.732 |
iors | +1.717 |
Neg logit contributions | |
Surgery | -1.546 |
Broadcasting | -1.486 |
scope | -1.443 |
hra | -1.407 |
aviour | -1.376 |
Token | they |
Token ID | 484 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
Surgery | +1.016 |
Broadcasting | +0.997 |
scope | +0.962 |
hra | +0.945 |
aviour | +0.918 |
Neg logit contributions | |
ishes | -1.090 |
ery | -1.073 |
adoes | -1.048 |
WI | -1.034 |
suscept | -1.018 |
Token | weren |
Token ID | 6304 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
Surgery | +0.590 |
Broadcasting | +0.579 |
scope | +0.559 |
hra | +0.549 |
aviour | +0.535 |
Neg logit contributions | |
ishes | -0.591 |
ery | -0.583 |
adoes | -0.567 |
WI | -0.560 |
suscept | -0.552 |
Token | 't |
Token ID | 470 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
ishes | +1.405 |
ery | +1.395 |
adoes | +1.325 |
WI | +1.322 |
iors | +1.287 |
Neg logit contributions | |
Surgery | -1.920 |
Broadcasting | -1.892 |
scope | -1.815 |
hra | -1.798 |
aviour | -1.773 |
Token | her |
Token ID | 607 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
ishes | +1.517 |
ery | +1.502 |
adoes | +1.452 |
WI | +1.438 |
suscept | +1.420 |
Neg logit contributions | |
Surgery | -1.244 |
Broadcasting | -1.222 |
scope | -1.177 |
hra | -1.165 |
aviour | -1.129 |
Token | little |
Token ID | 1310 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
ishes | +0.300 |
ery | +0.296 |
adoes | +0.287 |
WI | +0.283 |
suscept | +0.280 |
Neg logit contributions | |
Surgery | -0.306 |
Broadcasting | -0.300 |
scope | -0.291 |
hra | -0.286 |
aviour | -0.279 |
Token | fist |
Token ID | 18606 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
Surgery | +0.999 |
Broadcasting | +0.984 |
scope | +0.950 |
hra | +0.931 |
aviour | +0.909 |
Neg logit contributions | |
ishes | -0.868 |
ery | -0.854 |
adoes | -0.830 |
WI | -0.817 |
suscept | -0.804 |
Token | while |
Token ID | 981 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
Surgery | +1.601 |
Broadcasting | +1.574 |
scope | +1.533 |
hra | +1.488 |
aviour | +1.450 |
Neg logit contributions | |
ishes | -1.583 |
ery | -1.555 |
adoes | -1.525 |
WI | -1.500 |
suscept | -1.489 |
Token | an |
Token ID | 281 |
Feature activation | +9.33e-02 |
Pos logit contributions | |
Surgery | +2.389 |
Broadcasting | +2.351 |
scope | +2.265 |
hra | +2.204 |
aviour | +2.140 |
Neg logit contributions | |
ishes | -2.609 |
ery | -2.560 |
adoes | -2.521 |
WI | -2.476 |
suscept | -2.444 |
Token | ambulance |
Token ID | 22536 |
Feature activation | +8.02e+00 |
Pos logit contributions | |
Surgery | +0.074 |
Broadcasting | +0.073 |
scope | +0.071 |
hra | +0.069 |
aviour | +0.068 |
Neg logit contributions | |
ishes | -0.070 |
ery | -0.069 |
adoes | -0.067 |
WI | -0.066 |
suscept | -0.065 |
Token | s |
Token ID | 264 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
Surgery | +6.597 |
Broadcasting | +6.479 |
scope | +6.329 |
hra | +6.093 |
aviour | +5.963 |
Neg logit contributions | |
ishes | -5.234 |
adoes | -5.155 |
ery | -5.084 |
WI | -5.022 |
iors | -4.981 |
Token | iren |
Token ID | 24080 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
Surgery | +0.368 |
Broadcasting | +0.360 |
scope | +0.346 |
hra | +0.339 |
aviour | +0.328 |
Neg logit contributions | |
ishes | -0.468 |
ery | -0.463 |
adoes | -0.450 |
WI | -0.445 |
suscept | -0.438 |
Token | w |
Token ID | 266 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
Broadcasting | +3.671 |
Surgery | +3.661 |
scope | +3.521 |
hra | +3.456 |
aviour | +3.340 |
Neg logit contributions | |
ishes | -3.190 |
ery | -3.111 |
adoes | -3.070 |
suscept | -3.037 |
WI | -3.032 |
Token | ailed |
Token ID | 6255 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
ishes | +1.331 |
ery | +1.308 |
adoes | +1.236 |
WI | +1.209 |
iors | +1.182 |
Neg logit contributions | |
Surgery | -1.990 |
Broadcasting | -1.939 |
scope | -1.894 |
hra | -1.858 |
ndum | -1.820 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ishes | +0.638 |
ery | +0.630 |
adoes | +0.611 |
WI | +0.602 |
suscept | +0.593 |
Neg logit contributions | |
Surgery | -0.668 |
Broadcasting | -0.656 |
scope | -0.635 |
hra | -0.623 |
aviour | -0.608 |
Token | for |
Token ID | 329 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ishes | +0.181 |
ery | +0.179 |
adoes | +0.174 |
WI | +0.171 |
suscept | +0.169 |
Neg logit contributions | |
Surgery | -0.170 |
Broadcasting | -0.167 |
scope | -0.161 |
hra | -0.158 |
aviour | -0.154 |
Token | a |
Token ID | 257 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
ishes | +0.932 |
ery | +0.921 |
adoes | +0.894 |
WI | +0.883 |
suscept | +0.870 |
Neg logit contributions | |
Surgery | -0.887 |
Broadcasting | -0.872 |
scope | -0.843 |
hra | -0.828 |
aviour | -0.807 |
Token | few |
Token ID | 1178 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ishes | +0.176 |
ery | +0.174 |
adoes | +0.168 |
WI | +0.166 |
suscept | +0.164 |
Neg logit contributions | |
Surgery | -0.186 |
Broadcasting | -0.184 |
scope | -0.177 |
hra | -0.174 |
aviour | -0.170 |
Token | other |
Token ID | 584 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
Surgery | +1.432 |
Broadcasting | +1.409 |
scope | +1.357 |
hra | +1.330 |
ndum | +1.285 |
Neg logit contributions | |
ishes | -1.315 |
ery | -1.296 |
adoes | -1.255 |
WI | -1.238 |
suscept | -1.225 |
Token | fr |
Token ID | 1216 |
Feature activation | -7.37e+00 |
Pos logit contributions | |
ishes | +0.040 |
ery | +0.039 |
adoes | +0.038 |
WI | +0.037 |
suscept | +0.037 |
Neg logit contributions | |
Surgery | -0.044 |
Broadcasting | -0.043 |
scope | -0.042 |
hra | -0.041 |
aviour | -0.040 |
Token | atern |
Token ID | 9205 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
ishes | +3.019 |
ery | +2.640 |
adoes | +2.611 |
WI | +2.451 |
eries | +2.340 |
Neg logit contributions | |
Surgery | -8.501 |
Broadcasting | -8.294 |
hra | -7.962 |
scope | -7.847 |
ndum | -7.826 |
Token | ities |
Token ID | 871 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
ishes | +4.032 |
ery | +3.616 |
adoes | +3.352 |
ries | +3.054 |
iors | +2.996 |
Neg logit contributions | |
Surgery | -16.160 |
Broadcasting | -15.716 |
scope | -15.526 |
hra | -15.402 |
aviour | -15.309 |
Token | are |
Token ID | 389 |
Feature activation | -8.12e-01 |
Pos logit contributions | |
ishes | +2.730 |
ery | +2.722 |
adoes | +2.613 |
WI | +2.590 |
iors | +2.535 |
Neg logit contributions | |
Surgery | -2.692 |
Broadcasting | -2.636 |
scope | -2.553 |
hra | -2.502 |
aviour | -2.463 |
Token | also |
Token ID | 635 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
ishes | +0.658 |
ery | +0.651 |
adoes | +0.632 |
WI | +0.624 |
suscept | +0.617 |
Neg logit contributions | |
Surgery | -0.594 |
Broadcasting | -0.583 |
scope | -0.562 |
hra | -0.551 |
aviour | -0.537 |
Token | in |
Token ID | 287 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
Surgery | +0.471 |
Broadcasting | +0.463 |
scope | +0.446 |
hra | +0.437 |
aviour | +0.422 |
Neg logit contributions | |
ishes | -0.526 |
ery | -0.518 |
adoes | -0.505 |
WI | -0.499 |
suscept | -0.490 |
Token | a |
Token ID | 257 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
Surgery | +0.032 |
Broadcasting | +0.031 |
scope | +0.030 |
hra | +0.030 |
aviour | +0.029 |
Neg logit contributions | |
ishes | -0.038 |
ery | -0.037 |
adoes | -0.036 |
WI | -0.036 |
suscept | -0.036 |
Token | division |
Token ID | 7297 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ishes | +0.112 |
ery | +0.111 |
adoes | +0.108 |
WI | +0.107 |
suscept | +0.105 |
Neg logit contributions | |
Surgery | -0.106 |
Broadcasting | -0.104 |
scope | -0.101 |
hra | -0.099 |
aviour | -0.097 |
Token | housing |
Token ID | 5627 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
ishes | +1.395 |
ery | +1.383 |
adoes | +1.334 |
WI | +1.318 |
suscept | +1.297 |
Neg logit contributions | |
Surgery | -1.269 |
Broadcasting | -1.248 |
scope | -1.200 |
hra | -1.181 |
aviour | -1.151 |
Token | typical |
Token ID | 7226 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
ishes | +0.017 |
ery | +0.017 |
adoes | +0.016 |
WI | +0.016 |
suscept | +0.016 |
Neg logit contributions | |
Surgery | -0.016 |
Broadcasting | -0.016 |
scope | -0.015 |
hra | -0.015 |
aviour | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
Surgery | +0.493 |
Broadcasting | +0.484 |
scope | +0.466 |
hra | +0.457 |
aviour | +0.441 |
Neg logit contributions | |
ishes | -0.572 |
ery | -0.566 |
adoes | -0.552 |
WI | -0.544 |
suscept | -0.538 |
Token | fr |
Token ID | 1216 |
Feature activation | -6.30e+00 |
Pos logit contributions | |
ishes | +0.071 |
ery | +0.070 |
adoes | +0.068 |
WI | +0.067 |
suscept | +0.066 |
Neg logit contributions | |
Surgery | -0.073 |
Broadcasting | -0.072 |
scope | -0.069 |
hra | -0.068 |
aviour | -0.066 |
Token | atern |
Token ID | 9205 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
ishes | +2.400 |
ery | +2.234 |
adoes | +2.147 |
WI | +2.026 |
iors | +1.906 |
Neg logit contributions | |
Surgery | -7.347 |
Broadcasting | -7.229 |
hra | -6.899 |
scope | -6.875 |
ndum | -6.801 |
Token | ities |
Token ID | 871 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
ishes | +3.772 |
ery | +3.349 |
adoes | +3.095 |
iors | +2.955 |
ries | +2.942 |
Neg logit contributions | |
Surgery | -15.648 |
Broadcasting | -15.263 |
hra | -14.923 |
scope | -14.858 |
aviour | -14.739 |
Token | and |
Token ID | 290 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ishes | +2.037 |
ery | +2.030 |
WI | +1.931 |
adoes | +1.931 |
iors | +1.875 |
Neg logit contributions | |
Surgery | -2.640 |
Broadcasting | -2.597 |
scope | -2.521 |
hra | -2.478 |
aviour | -2.440 |
Token | sor |
Token ID | 25655 |
Feature activation | -9.82e+00 |
Pos logit contributions | |
ishes | +0.312 |
ery | +0.302 |
adoes | +0.279 |
WI | +0.269 |
suscept | +0.259 |
Neg logit contributions | |
Surgery | -1.298 |
Broadcasting | -1.285 |
scope | -1.258 |
hra | -1.243 |
aviour | -1.222 |
Token | or |
Token ID | 273 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
ishes | +5.325 |
adoes | +5.115 |
ery | +5.065 |
ries | +4.870 |
WI | +4.709 |
Neg logit contributions | |
Surgery | -9.842 |
Broadcasting | -9.598 |
scope | -9.091 |
aviour | -8.965 |
hra | -8.947 |
Token | ities |
Token ID | 871 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
ishes | +0.804 |
ery | +0.785 |
adoes | +0.745 |
WI | +0.727 |
suscept | +0.709 |
Neg logit contributions | |
Surgery | -1.997 |
Broadcasting | -1.973 |
scope | -1.926 |
hra | -1.902 |
aviour | -1.868 |
Token | to |
Token ID | 284 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
ishes | +1.105 |
ery | +1.093 |
adoes | +1.058 |
WI | +1.046 |
suscept | +1.031 |
Neg logit contributions | |
Surgery | -1.071 |
Broadcasting | -1.053 |
scope | -1.017 |
hra | -0.999 |
aviour | -0.974 |
Token | just |
Token ID | 655 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ishes | +0.091 |
ery | +0.090 |
adoes | +0.088 |
WI | +0.087 |
suscept | +0.085 |
Neg logit contributions | |
Surgery | -0.079 |
Broadcasting | -0.078 |
scope | -0.075 |
hra | -0.073 |
aviour | -0.071 |
Token | apply |
Token ID | 4174 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
ishes | +0.963 |
ery | +0.952 |
adoes | +0.924 |
WI | +0.913 |
suscept | +0.900 |
Neg logit contributions | |
Surgery | -0.820 |
Broadcasting | -0.807 |
scope | -0.777 |
hra | -0.756 |
aviour | -0.739 |
Token | to |
Token ID | 284 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
ishes | +0.160 |
ery | +0.158 |
adoes | +0.153 |
WI | +0.151 |
suscept | +0.149 |
Neg logit contributions | |
Surgery | -0.159 |
Broadcasting | -0.156 |
scope | -0.151 |
hra | -0.148 |
aviour | -0.144 |
Token | fr |
Token ID | 1216 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
ishes | +0.393 |
ery | +0.388 |
adoes | +0.377 |
WI | +0.372 |
suscept | +0.367 |
Neg logit contributions | |
Surgery | -0.387 |
Broadcasting | -0.380 |
scope | -0.367 |
hra | -0.360 |
aviour | -0.350 |
Token | atern |
Token ID | 9205 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
ishes | +2.606 |
ery | +2.454 |
adoes | +2.371 |
WI | +2.281 |
iors | +2.157 |
Neg logit contributions | |
Surgery | -6.840 |
Broadcasting | -6.753 |
scope | -6.476 |
hra | -6.451 |
ndum | -6.363 |
Token | ities |
Token ID | 871 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
ishes | +3.686 |
ery | +3.287 |
adoes | +3.024 |
iors | +3.009 |
ries | +2.870 |
Neg logit contributions | |
Surgery | -14.565 |
Broadcasting | -14.197 |
hra | -13.839 |
scope | -13.836 |
aviour | -13.720 |
Token | or |
Token ID | 393 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
ishes | +2.030 |
ery | +2.022 |
adoes | +1.937 |
WI | +1.928 |
suscept | +1.883 |
Neg logit contributions | |
Surgery | -2.203 |
Broadcasting | -2.160 |
scope | -2.087 |
hra | -2.050 |
aviour | -2.017 |
Token | just |
Token ID | 655 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
ishes | +0.317 |
ery | +0.309 |
adoes | +0.290 |
WI | +0.281 |
suscept | +0.274 |
Neg logit contributions | |
Surgery | -0.990 |
Broadcasting | -0.980 |
scope | -0.958 |
hra | -0.946 |
aviour | -0.928 |
Token | sor |
Token ID | 25655 |
Feature activation | -9.29e+00 |
Pos logit contributions | |
ishes | +1.661 |
ery | +1.636 |
adoes | +1.587 |
WI | +1.567 |
suscept | +1.541 |
Neg logit contributions | |
Surgery | -1.832 |
Broadcasting | -1.800 |
scope | -1.741 |
hra | -1.709 |
aviour | -1.674 |
Token | or |
Token ID | 273 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ishes | +5.071 |
ery | +4.964 |
adoes | +4.879 |
iors | +4.695 |
WI | +4.582 |
Neg logit contributions | |
Surgery | -9.067 |
Broadcasting | -8.881 |
scope | -8.468 |
hra | -8.367 |
aviour | -8.353 |
Token | of |
Token ID | 286 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
Surgery | +0.493 |
Broadcasting | +0.484 |
scope | +0.466 |
hra | +0.457 |
aviour | +0.441 |
Neg logit contributions | |
ishes | -0.572 |
ery | -0.566 |
adoes | -0.552 |
WI | -0.544 |
suscept | -0.538 |
Token | fr |
Token ID | 1216 |
Feature activation | -6.30e+00 |
Pos logit contributions | |
ishes | +0.071 |
ery | +0.070 |
adoes | +0.068 |
WI | +0.067 |
suscept | +0.066 |
Neg logit contributions | |
Surgery | -0.073 |
Broadcasting | -0.072 |
scope | -0.069 |
hra | -0.068 |
aviour | -0.066 |
Token | atern |
Token ID | 9205 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
ishes | +2.400 |
ery | +2.234 |
adoes | +2.147 |
WI | +2.026 |
iors | +1.906 |
Neg logit contributions | |
Surgery | -7.347 |
Broadcasting | -7.229 |
hra | -6.899 |
scope | -6.875 |
ndum | -6.801 |
Token | ities |
Token ID | 871 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
ishes | +3.772 |
ery | +3.349 |
adoes | +3.095 |
iors | +2.955 |
ries | +2.942 |
Neg logit contributions | |
Surgery | -15.648 |
Broadcasting | -15.263 |
hra | -14.923 |
scope | -14.858 |
aviour | -14.739 |
Token | and |
Token ID | 290 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ishes | +2.037 |
ery | +2.030 |
WI | +1.931 |
adoes | +1.931 |
iors | +1.875 |
Neg logit contributions | |
Surgery | -2.640 |
Broadcasting | -2.597 |
scope | -2.521 |
hra | -2.478 |
aviour | -2.440 |
Token | sor |
Token ID | 25655 |
Feature activation | -9.82e+00 |
Pos logit contributions | |
ishes | +0.312 |
ery | +0.302 |
adoes | +0.279 |
WI | +0.269 |
suscept | +0.259 |
Neg logit contributions | |
Surgery | -1.298 |
Broadcasting | -1.285 |
scope | -1.258 |
hra | -1.243 |
aviour | -1.222 |
Token | or |
Token ID | 273 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
ishes | +5.325 |
adoes | +5.115 |
ery | +5.065 |
ries | +4.870 |
WI | +4.709 |
Neg logit contributions | |
Surgery | -9.842 |
Broadcasting | -9.598 |
scope | -9.091 |
aviour | -8.965 |
hra | -8.947 |
Token | ities |
Token ID | 871 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
ishes | +0.804 |
ery | +0.785 |
adoes | +0.745 |
WI | +0.727 |
suscept | +0.709 |
Neg logit contributions | |
Surgery | -1.997 |
Broadcasting | -1.973 |
scope | -1.926 |
hra | -1.902 |
aviour | -1.868 |
Token | on |
Token ID | 319 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
ishes | +2.169 |
ery | +2.165 |
adoes | +2.052 |
WI | +2.048 |
iors | +1.989 |
Neg logit contributions | |
Surgery | -3.058 |
Broadcasting | -3.014 |
scope | -2.930 |
hra | -2.882 |
aviour | -2.842 |
Token | campus |
Token ID | 7611 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ishes | +1.914 |
ery | +1.895 |
adoes | +1.828 |
WI | +1.809 |
suscept | +1.775 |
Neg logit contributions | |
Surgery | -1.964 |
Broadcasting | -1.924 |
scope | -1.867 |
hra | -1.840 |
aviour | -1.793 |
Token | ), |
Token ID | 828 |
Feature activation | -3.50e-03 |
Pos logit contributions | |
ishes | +1.335 |
ery | +1.325 |
adoes | +1.266 |
WI | +1.256 |
suscept | +1.226 |
Neg logit contributions | |
Surgery | -1.829 |
Broadcasting | -1.798 |
scope | -1.749 |
hra | -1.723 |
aviour | -1.689 |
Token | fr |
Token ID | 1216 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
ishes | +0.393 |
ery | +0.388 |
adoes | +0.377 |
WI | +0.372 |
suscept | +0.367 |
Neg logit contributions | |
Surgery | -0.387 |
Broadcasting | -0.380 |
scope | -0.367 |
hra | -0.360 |
aviour | -0.350 |
Token | atern |
Token ID | 9205 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
ishes | +2.606 |
ery | +2.454 |
adoes | +2.371 |
WI | +2.281 |
iors | +2.157 |
Neg logit contributions | |
Surgery | -6.840 |
Broadcasting | -6.753 |
scope | -6.476 |
hra | -6.451 |
ndum | -6.363 |
Token | ities |
Token ID | 871 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
ishes | +3.686 |
ery | +3.287 |
adoes | +3.024 |
iors | +3.009 |
ries | +2.870 |
Neg logit contributions | |
Surgery | -14.565 |
Broadcasting | -14.197 |
hra | -13.839 |
scope | -13.836 |
aviour | -13.720 |
Token | or |
Token ID | 393 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
ishes | +2.030 |
ery | +2.022 |
adoes | +1.937 |
WI | +1.928 |
suscept | +1.883 |
Neg logit contributions | |
Surgery | -2.203 |
Broadcasting | -2.160 |
scope | -2.087 |
hra | -2.050 |
aviour | -2.017 |
Token | just |
Token ID | 655 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
ishes | +0.317 |
ery | +0.309 |
adoes | +0.290 |
WI | +0.281 |
suscept | +0.274 |
Neg logit contributions | |
Surgery | -0.990 |
Broadcasting | -0.980 |
scope | -0.958 |
hra | -0.946 |
aviour | -0.928 |
Token | sor |
Token ID | 25655 |
Feature activation | -9.29e+00 |
Pos logit contributions | |
ishes | +1.661 |
ery | +1.636 |
adoes | +1.587 |
WI | +1.567 |
suscept | +1.541 |
Neg logit contributions | |
Surgery | -1.832 |
Broadcasting | -1.800 |
scope | -1.741 |
hra | -1.709 |
aviour | -1.674 |
Token | or |
Token ID | 273 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ishes | +5.071 |
ery | +4.964 |
adoes | +4.879 |
iors | +4.695 |
WI | +4.582 |
Neg logit contributions | |
Surgery | -9.067 |
Broadcasting | -8.881 |
scope | -8.468 |
hra | -8.367 |
aviour | -8.353 |
Token | ities |
Token ID | 871 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
ishes | +0.566 |
ery | +0.552 |
adoes | +0.527 |
WI | +0.508 |
suscept | +0.492 |
Neg logit contributions | |
Surgery | -1.444 |
Broadcasting | -1.426 |
scope | -1.390 |
hra | -1.373 |
aviour | -1.352 |
Token | . |
Token ID | 13 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
ishes | +2.279 |
ery | +2.273 |
adoes | +2.171 |
WI | +2.162 |
suscept | +2.117 |
Neg logit contributions | |
Surgery | -2.474 |
Broadcasting | -2.426 |
scope | -2.345 |
hra | -2.308 |
aviour | -2.272 |
Token | This |
Token ID | 770 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
ishes | +0.306 |
ery | +0.303 |
adoes | +0.294 |
WI | +0.292 |
suscept | +0.286 |
Neg logit contributions | |
Surgery | -0.256 |
Broadcasting | -0.251 |
scope | -0.243 |
hra | -0.238 |
aviour | -0.233 |
Token | goes |
Token ID | 2925 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
ishes | +0.573 |
ery | +0.565 |
adoes | +0.549 |
WI | +0.541 |
suscept | +0.534 |
Neg logit contributions | |
Surgery | -0.626 |
Broadcasting | -0.616 |
scope | -0.595 |
hra | -0.585 |
aviour | -0.570 |
Token | will |
Token ID | 481 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
ishes | +0.082 |
ery | +0.081 |
adoes | +0.079 |
WI | +0.078 |
suscept | +0.077 |
Neg logit contributions | |
Surgery | -0.084 |
Broadcasting | -0.083 |
scope | -0.080 |
hra | -0.078 |
aviour | -0.076 |
Token | consider |
Token ID | 2074 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
Surgery | +0.530 |
Broadcasting | +0.522 |
scope | +0.504 |
hra | +0.492 |
aviour | +0.475 |
Neg logit contributions | |
ishes | -0.638 |
ery | -0.627 |
adoes | -0.612 |
WI | -0.606 |
suscept | -0.602 |
Token | the |
Token ID | 262 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
Surgery | +1.179 |
Broadcasting | +1.164 |
scope | +1.125 |
hra | +1.102 |
aviour | +1.064 |
Neg logit contributions | |
ishes | -1.239 |
ery | -1.217 |
adoes | -1.186 |
WI | -1.172 |
suscept | -1.154 |
Token | role |
Token ID | 2597 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ishes | +0.359 |
ery | +0.354 |
adoes | +0.344 |
WI | +0.339 |
suscept | +0.335 |
Neg logit contributions | |
Surgery | -0.344 |
Broadcasting | -0.338 |
scope | -0.326 |
hra | -0.321 |
aviour | -0.312 |
Token | of |
Token ID | 286 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
Surgery | +1.078 |
Broadcasting | +1.067 |
scope | +1.030 |
hra | +1.004 |
aviour | +0.969 |
Neg logit contributions | |
ishes | -1.317 |
ery | -1.290 |
adoes | -1.278 |
WI | -1.247 |
suscept | -1.243 |
Token | ment |
Token ID | 6229 |
Feature activation | -8.91e+00 |
Pos logit contributions | |
Surgery | +0.405 |
Broadcasting | +0.398 |
scope | +0.385 |
hra | +0.378 |
aviour | +0.368 |
Neg logit contributions | |
ishes | -0.383 |
ery | -0.377 |
adoes | -0.366 |
WI | -0.361 |
suscept | -0.355 |
Token | hol |
Token ID | 3937 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ery | +5.789 |
ishes | +5.639 |
iors | +5.355 |
adoes | +5.229 |
WI | +5.154 |
Neg logit contributions | |
Surgery | -7.931 |
Broadcasting | -7.684 |
ndum | -7.488 |
scope | -7.426 |
hra | -7.272 |
Token | and |
Token ID | 290 |
Feature activation | -8.57e-01 |
Pos logit contributions | |
Surgery | +0.909 |
Broadcasting | +0.894 |
scope | +0.866 |
hra | +0.842 |
aviour | +0.822 |
Neg logit contributions | |
ishes | -0.927 |
ery | -0.912 |
adoes | -0.889 |
WI | -0.881 |
suscept | -0.868 |
Token | other |
Token ID | 584 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
ishes | +0.666 |
ery | +0.657 |
adoes | +0.638 |
WI | +0.629 |
suscept | +0.620 |
Neg logit contributions | |
Surgery | -0.657 |
Broadcasting | -0.646 |
scope | -0.623 |
hra | -0.612 |
aviour | -0.596 |
Token | flavors |
Token ID | 17361 |
Feature activation | -6.29e+00 |
Pos logit contributions | |
ishes | +0.227 |
ery | +0.223 |
adoes | +0.216 |
WI | +0.213 |
suscept | +0.209 |
Neg logit contributions | |
Surgery | -0.282 |
Broadcasting | -0.278 |
scope | -0.269 |
hra | -0.265 |
aviour | -0.259 |
Token | in |
Token ID | 287 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ishes | +4.590 |
ery | +4.576 |
adoes | +4.323 |
WI | +4.264 |
suscept | +4.205 |
Neg logit contributions | |
Surgery | -5.007 |
Broadcasting | -4.928 |
scope | -4.727 |
hra | -4.646 |
ndum | -4.626 |
Token | racist |
Token ID | 9811 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
ishes | +0.234 |
ery | +0.231 |
adoes | +0.223 |
WI | +0.220 |
suscept | +0.217 |
Neg logit contributions | |
Surgery | -0.281 |
Broadcasting | -0.276 |
scope | -0.267 |
hra | -0.263 |
aviour | -0.257 |
Token | , |
Token ID | 11 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
ishes | +1.686 |
ery | +1.678 |
adoes | +1.615 |
WI | +1.593 |
suscept | +1.570 |
Neg logit contributions | |
Surgery | -1.801 |
Broadcasting | -1.769 |
scope | -1.722 |
hra | -1.694 |
aviour | -1.652 |
Token | class |
Token ID | 1398 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
ishes | +0.300 |
ery | +0.296 |
adoes | +0.288 |
WI | +0.284 |
suscept | +0.280 |
Neg logit contributions | |
Surgery | -0.311 |
Broadcasting | -0.305 |
scope | -0.295 |
hra | -0.290 |
aviour | -0.282 |
Token | ist |
Token ID | 396 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ishes | +1.343 |
ery | +1.326 |
adoes | +1.289 |
WI | +1.274 |
suscept | +1.257 |
Neg logit contributions | |
Surgery | -1.156 |
Broadcasting | -1.135 |
scope | -1.092 |
hra | -1.071 |
aviour | -1.041 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ishes | +0.769 |
ery | +0.759 |
adoes | +0.736 |
WI | +0.726 |
suscept | +0.716 |
Neg logit contributions | |
Surgery | -0.792 |
Broadcasting | -0.778 |
scope | -0.752 |
hra | -0.739 |
aviour | -0.720 |
Token | misogyn |
Token ID | 22795 |
Feature activation | -8.85e+00 |
Pos logit contributions | |
ishes | +0.904 |
ery | +0.891 |
adoes | +0.864 |
WI | +0.852 |
suscept | +0.840 |
Neg logit contributions | |
Surgery | -0.988 |
Broadcasting | -0.972 |
scope | -0.940 |
hra | -0.923 |
aviour | -0.901 |
Token | istic |
Token ID | 2569 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
ery | +6.133 |
ishes | +6.006 |
adoes | +5.698 |
WI | +5.513 |
seless | +5.363 |
Neg logit contributions | |
Surgery | -7.441 |
Broadcasting | -7.281 |
scope | -6.904 |
hra | -6.836 |
aviour | -6.579 |
Token | , |
Token ID | 11 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
ishes | +2.899 |
ery | +2.893 |
adoes | +2.774 |
WI | +2.733 |
suscept | +2.698 |
Neg logit contributions | |
Surgery | -3.175 |
Broadcasting | -3.102 |
scope | -3.030 |
hra | -2.989 |
aviour | -2.908 |
Token | heter |
Token ID | 14445 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
ishes | +1.115 |
ery | +1.100 |
adoes | +1.063 |
WI | +1.049 |
suscept | +1.034 |
Neg logit contributions | |
Surgery | -1.287 |
Broadcasting | -1.266 |
scope | -1.226 |
hra | -1.206 |
aviour | -1.175 |
Token | ose |
Token ID | 577 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
ishes | +1.468 |
ery | +1.449 |
adoes | +1.398 |
WI | +1.378 |
suscept | +1.352 |
Neg logit contributions | |
Surgery | -1.767 |
Broadcasting | -1.737 |
scope | -1.682 |
hra | -1.654 |
aviour | -1.611 |
Token | x |
Token ID | 87 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
ishes | +0.323 |
ery | +0.318 |
adoes | +0.311 |
WI | +0.306 |
suscept | +0.304 |
Neg logit contributions | |
Surgery | -0.322 |
Broadcasting | -0.317 |
scope | -0.308 |
hra | -0.302 |
aviour | -0.295 |
Token | , |
Token ID | 11 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
ishes | +0.608 |
ery | +0.599 |
adoes | +0.580 |
WI | +0.573 |
suscept | +0.563 |
Neg logit contributions | |
Surgery | -0.680 |
Broadcasting | -0.669 |
scope | -0.648 |
hra | -0.637 |
aviour | -0.622 |
Token | 000 |
Token ID | 830 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
ishes | +0.547 |
ery | +0.540 |
adoes | +0.524 |
WI | +0.517 |
suscept | +0.509 |
Neg logit contributions | |
Surgery | -0.566 |
Broadcasting | -0.557 |
scope | -0.538 |
hra | -0.528 |
aviour | -0.515 |
Token | ; |
Token ID | 26 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
ishes | +0.498 |
ery | +0.491 |
adoes | +0.476 |
WI | +0.470 |
suscept | +0.463 |
Neg logit contributions | |
Surgery | -0.532 |
Broadcasting | -0.523 |
scope | -0.506 |
hra | -0.497 |
aviour | -0.484 |
Token | field |
Token ID | 2214 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
ishes | +0.624 |
ery | +0.617 |
adoes | +0.598 |
WI | +0.591 |
suscept | +0.580 |
Neg logit contributions | |
Surgery | -0.624 |
Broadcasting | -0.612 |
scope | -0.593 |
hra | -0.582 |
aviour | -0.569 |
Token | 36 |
Token ID | 4570 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ishes | +2.304 |
ery | +2.278 |
adoes | +2.183 |
WI | +2.154 |
suscept | +2.116 |
Neg logit contributions | |
Surgery | -2.722 |
Broadcasting | -2.687 |
scope | -2.606 |
hra | -2.592 |
aviour | -2.517 |
Token | batt |
Token ID | 6274 |
Feature activation | -8.61e+00 |
Pos logit contributions | |
ishes | +1.089 |
ery | +1.078 |
adoes | +1.041 |
WI | +1.028 |
suscept | +1.007 |
Neg logit contributions | |
Surgery | -1.149 |
Broadcasting | -1.130 |
scope | -1.098 |
hra | -1.084 |
aviour | -1.056 |
Token | alions |
Token ID | 50022 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
ishes | +3.809 |
ery | +3.544 |
adoes | +3.185 |
eries | +3.046 |
iors | +3.002 |
Neg logit contributions | |
Surgery | -9.601 |
Broadcasting | -9.466 |
hra | -9.088 |
aviour | -9.040 |
scope | -8.968 |
Token | of |
Token ID | 286 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
ery | +3.501 |
ishes | +3.491 |
adoes | +3.363 |
WI | +3.320 |
suscept | +3.250 |
Neg logit contributions | |
Surgery | -3.559 |
Broadcasting | -3.489 |
scope | -3.377 |
hra | -3.347 |
aviour | -3.254 |
Token | Marines |
Token ID | 19586 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ishes | +0.659 |
ery | +0.650 |
adoes | +0.632 |
WI | +0.624 |
suscept | +0.615 |
Neg logit contributions | |
Surgery | -0.628 |
Broadcasting | -0.617 |
scope | -0.595 |
hra | -0.584 |
aviour | -0.568 |
Token | ; |
Token ID | 26 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ishes | +1.081 |
ery | +1.070 |
adoes | +1.038 |
WI | +1.025 |
suscept | +1.009 |
Neg logit contributions | |
Surgery | -1.046 |
Broadcasting | -1.029 |
scope | -0.995 |
hra | -0.979 |
aviour | -0.951 |
Token | set |
Token ID | 900 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
ishes | +0.977 |
ery | +0.966 |
adoes | +0.936 |
WI | +0.923 |
suscept | +0.910 |
Neg logit contributions | |
Surgery | -0.998 |
Broadcasting | -0.980 |
scope | -0.947 |
hra | -0.931 |
aviour | -0.908 |
Token | er |
Token ID | 263 |
Feature activation | +8.84e-01 |
Pos logit contributions | |
Surgery | +1.733 |
Broadcasting | +1.696 |
scope | +1.642 |
hra | +1.613 |
aviour | +1.563 |
Neg logit contributions | |
ishes | -1.973 |
ery | -1.929 |
adoes | -1.895 |
WI | -1.873 |
suscept | -1.847 |
Token |
|
Token ID | 198 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
Surgery | +0.567 |
Broadcasting | +0.554 |
scope | +0.531 |
hra | +0.519 |
aviour | +0.503 |
Neg logit contributions | |
ishes | -0.798 |
ery | -0.788 |
adoes | -0.770 |
WI | -0.761 |
suscept | -0.752 |
Token |
|
Token ID | 198 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
ishes | +0.049 |
ery | +0.049 |
adoes | +0.047 |
WI | +0.047 |
iors | +0.046 |
Neg logit contributions | |
Surgery | -0.038 |
Broadcasting | -0.037 |
scope | -0.035 |
hra | -0.034 |
aviour | -0.034 |
Token | G |
Token ID | 38 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
ishes | +0.120 |
ery | +0.119 |
adoes | +0.115 |
WI | +0.114 |
suscept | +0.113 |
Neg logit contributions | |
Surgery | -0.106 |
Broadcasting | -0.104 |
scope | -0.100 |
hra | -0.098 |
aviour | -0.095 |
Token | rim |
Token ID | 3036 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
Surgery | +0.261 |
Broadcasting | +0.256 |
scope | +0.247 |
hra | +0.243 |
aviour | +0.237 |
Neg logit contributions | |
ishes | -0.254 |
ery | -0.251 |
adoes | -0.243 |
WI | -0.240 |
suscept | -0.236 |
Token | Lav |
Token ID | 21438 |
Feature activation | -8.60e+00 |
Pos logit contributions | |
ishes | +1.489 |
ery | +1.483 |
adoes | +1.432 |
WI | +1.409 |
suscept | +1.385 |
Neg logit contributions | |
Surgery | -1.305 |
Broadcasting | -1.293 |
scope | -1.242 |
hra | -1.219 |
aviour | -1.197 |
Token | am |
Token ID | 321 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
ishes | +5.985 |
ery | +5.820 |
adoes | +5.532 |
iors | +5.315 |
WI | +5.234 |
Neg logit contributions | |
Surgery | -7.234 |
Broadcasting | -7.142 |
scope | -6.811 |
hra | -6.673 |
aviour | -6.553 |
Token | ancer |
Token ID | 8250 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
ishes | +0.338 |
ery | +0.334 |
adoes | +0.323 |
WI | +0.316 |
suscept | +0.312 |
Neg logit contributions | |
Surgery | -0.389 |
Broadcasting | -0.383 |
scope | -0.371 |
hra | -0.364 |
aviour | -0.355 |
Token | vs |
Token ID | 3691 |
Feature activation | -8.57e-01 |
Pos logit contributions | |
Surgery | +1.906 |
Broadcasting | +1.855 |
scope | +1.791 |
hra | +1.768 |
aviour | +1.720 |
Neg logit contributions | |
ishes | -1.778 |
ery | -1.731 |
adoes | -1.690 |
WI | -1.675 |
suscept | -1.640 |
Token | Ragnar |
Token ID | 32797 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
ishes | +0.743 |
ery | +0.735 |
adoes | +0.714 |
WI | +0.706 |
suscept | +0.697 |
Neg logit contributions | |
Surgery | -0.576 |
Broadcasting | -0.566 |
scope | -0.546 |
hra | -0.534 |
aviour | -0.520 |
Token |
|
Token ID | 198 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
Surgery | +1.772 |
Broadcasting | +1.744 |
scope | +1.690 |
hra | +1.652 |
aviour | +1.612 |
Neg logit contributions | |
ishes | -1.662 |
ery | -1.638 |
adoes | -1.593 |
WI | -1.573 |
suscept | -1.555 |
Token | is |
Token ID | 318 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
ishes | +2.174 |
ery | +2.152 |
adoes | +2.070 |
WI | +2.038 |
suscept | +1.999 |
Neg logit contributions | |
Surgery | -2.468 |
Broadcasting | -2.419 |
scope | -2.354 |
hra | -2.328 |
aviour | -2.265 |
Token | on |
Token ID | 319 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Surgery | +0.507 |
Broadcasting | +0.499 |
scope | +0.482 |
hra | +0.473 |
aviour | +0.458 |
Neg logit contributions | |
ishes | -0.544 |
ery | -0.536 |
adoes | -0.522 |
WI | -0.514 |
suscept | -0.508 |
Token | the |
Token ID | 262 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
ishes | +0.968 |
ery | +0.960 |
adoes | +0.930 |
WI | +0.921 |
suscept | +0.909 |
Neg logit contributions | |
Surgery | -0.796 |
Broadcasting | -0.780 |
scope | -0.750 |
hra | -0.737 |
aviour | -0.717 |
Token | edge |
Token ID | 5743 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Surgery | +0.307 |
Broadcasting | +0.302 |
scope | +0.291 |
hra | +0.285 |
aviour | +0.277 |
Neg logit contributions | |
ishes | -0.305 |
ery | -0.299 |
adoes | -0.291 |
WI | -0.286 |
suscept | -0.282 |
Token | of |
Token ID | 286 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ishes | +0.106 |
ery | +0.104 |
adoes | +0.101 |
WI | +0.100 |
suscept | +0.099 |
Neg logit contributions | |
Surgery | -0.090 |
Broadcasting | -0.088 |
scope | -0.085 |
hra | -0.083 |
aviour | -0.081 |
Token | vit |
Token ID | 9090 |
Feature activation | -8.57e+00 |
Pos logit contributions | |
Surgery | +1.017 |
Broadcasting | +0.999 |
scope | +0.962 |
hra | +0.942 |
aviour | +0.915 |
Neg logit contributions | |
ishes | -1.094 |
ery | -1.077 |
adoes | -1.048 |
WI | -1.037 |
suscept | -1.021 |
Token | icultural |
Token ID | 26823 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
ery | +2.777 |
ishes | +2.756 |
adoes | +2.480 |
WI | +2.198 |
iors | +2.187 |
Neg logit contributions | |
Surgery | -10.349 |
Broadcasting | -10.167 |
scope | -9.914 |
ndum | -9.753 |
hra | -9.725 |
Token | viability |
Token ID | 32141 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
ishes | +1.726 |
ery | +1.713 |
adoes | +1.651 |
WI | +1.632 |
suscept | +1.612 |
Neg logit contributions | |
Surgery | -1.734 |
Broadcasting | -1.706 |
scope | -1.647 |
hra | -1.625 |
aviour | -1.581 |
Token | , |
Token ID | 11 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
Surgery | +0.352 |
Broadcasting | +0.345 |
scope | +0.334 |
hra | +0.328 |
aviour | +0.319 |
Neg logit contributions | |
ishes | -0.337 |
ery | -0.332 |
adoes | -0.323 |
WI | -0.318 |
suscept | -0.314 |
Token | the |
Token ID | 262 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
ishes | +0.446 |
ery | +0.440 |
adoes | +0.428 |
WI | +0.423 |
suscept | +0.417 |
Neg logit contributions | |
Surgery | -0.399 |
Broadcasting | -0.392 |
scope | -0.377 |
hra | -0.370 |
aviour | -0.360 |
Token | math |
Token ID | 10688 |
Feature activation | -8.90e-01 |
Pos logit contributions | |
ishes | +0.593 |
ery | +0.585 |
adoes | +0.567 |
WI | +0.559 |
suscept | +0.551 |
Neg logit contributions | |
Surgery | -0.654 |
Broadcasting | -0.643 |
scope | -0.623 |
hra | -0.612 |
aviour | -0.597 |
Token | word |
Token ID | 1573 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ishes | +0.445 |
ery | +0.439 |
adoes | +0.427 |
WI | +0.421 |
suscept | +0.416 |
Neg logit contributions | |
Surgery | -0.434 |
Broadcasting | -0.426 |
scope | -0.411 |
hra | -0.404 |
aviour | -0.393 |
Token | of |
Token ID | 286 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
ishes | +0.875 |
ery | +0.864 |
adoes | +0.837 |
WI | +0.828 |
suscept | +0.816 |
Neg logit contributions | |
Surgery | -0.840 |
Broadcasting | -0.826 |
scope | -0.796 |
hra | -0.781 |
aviour | -0.762 |
Token | caution |
Token ID | 13041 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Surgery | +0.657 |
Broadcasting | +0.648 |
scope | +0.624 |
hra | +0.615 |
aviour | +0.600 |
Neg logit contributions | |
ishes | -0.464 |
ery | -0.451 |
adoes | -0.442 |
WI | -0.432 |
suscept | -0.425 |
Token | . |
Token ID | 13 |
Feature activation | +9.51e-03 |
Pos logit contributions | |
ishes | +0.129 |
ery | +0.127 |
adoes | +0.124 |
WI | +0.122 |
suscept | +0.121 |
Neg logit contributions | |
Surgery | -0.126 |
Broadcasting | -0.124 |
scope | -0.119 |
hra | -0.117 |
aviour | -0.114 |
Token | Some |
Token ID | 2773 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
Surgery | +0.006 |
Broadcasting | +0.006 |
scope | +0.006 |
hra | +0.006 |
aviour | +0.005 |
Neg logit contributions | |
ishes | -0.009 |
ery | -0.009 |
adoes | -0.008 |
WI | -0.008 |
suscept | -0.008 |
Token | flour |
Token ID | 10601 |
Feature activation | -8.51e+00 |
Pos logit contributions | |
Surgery | +1.943 |
Broadcasting | +1.915 |
scope | +1.842 |
hra | +1.830 |
aviour | +1.758 |
Neg logit contributions | |
ishes | -1.461 |
ery | -1.431 |
adoes | -1.405 |
WI | -1.377 |
iors | -1.352 |
Token | are |
Token ID | 389 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ishes | +6.094 |
ery | +5.855 |
adoes | +5.500 |
WI | +5.385 |
suscept | +5.353 |
Neg logit contributions | |
Surgery | -7.081 |
Broadcasting | -6.974 |
scope | -6.696 |
hra | -6.598 |
ndum | -6.509 |
Token | labeled |
Token ID | 15494 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ishes | +0.797 |
ery | +0.784 |
adoes | +0.763 |
WI | +0.752 |
suscept | +0.738 |
Neg logit contributions | |
Surgery | -0.882 |
Broadcasting | -0.867 |
scope | -0.839 |
hra | -0.824 |
aviour | -0.803 |
Token | as |
Token ID | 355 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
ishes | +0.880 |
ery | +0.868 |
adoes | +0.844 |
WI | +0.833 |
suscept | +0.821 |
Neg logit contributions | |
Surgery | -0.848 |
Broadcasting | -0.833 |
scope | -0.804 |
hra | -0.789 |
aviour | -0.768 |
Token | plant |
Token ID | 4618 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Surgery | +0.951 |
Broadcasting | +0.930 |
scope | +0.899 |
hra | +0.886 |
aviour | +0.856 |
Neg logit contributions | |
ishes | -0.894 |
ery | -0.879 |
adoes | -0.862 |
WI | -0.848 |
suscept | -0.830 |
Token | ain |
Token ID | 391 |
Feature activation | -6.86e-01 |
Pos logit contributions | |
Surgery | +1.172 |
Broadcasting | +1.161 |
scope | +1.120 |
hra | +1.114 |
aviour | +1.094 |
Neg logit contributions | |
ishes | -1.215 |
ery | -1.190 |
adoes | -1.178 |
WI | -1.177 |
suscept | -1.152 |
Token | women |
Token ID | 1466 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
Surgery | +0.285 |
Broadcasting | +0.279 |
scope | +0.269 |
hra | +0.263 |
aviour | +0.256 |
Neg logit contributions | |
ishes | -0.335 |
ery | -0.331 |
adoes | -0.322 |
WI | -0.318 |
suscept | -0.314 |
Token | suffering |
Token ID | 7195 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
ishes | +0.578 |
ery | +0.571 |
adoes | +0.551 |
WI | +0.545 |
suscept | +0.536 |
Neg logit contributions | |
Surgery | -0.626 |
Broadcasting | -0.613 |
scope | -0.593 |
hra | -0.582 |
aviour | -0.567 |
Token | under |
Token ID | 739 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
Surgery | +0.321 |
Broadcasting | +0.316 |
scope | +0.306 |
hra | +0.299 |
aviour | +0.292 |
Neg logit contributions | |
ishes | -0.310 |
ery | -0.306 |
adoes | -0.297 |
WI | -0.293 |
suscept | -0.289 |
Token | a |
Token ID | 257 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Surgery | +1.557 |
Broadcasting | +1.532 |
scope | +1.477 |
hra | +1.448 |
aviour | +1.408 |
Neg logit contributions | |
ishes | -1.662 |
ery | -1.637 |
adoes | -1.597 |
WI | -1.572 |
suscept | -1.559 |
Token | brutally |
Token ID | 30040 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Surgery | +0.083 |
Broadcasting | +0.082 |
scope | +0.079 |
hra | +0.078 |
aviour | +0.076 |
Neg logit contributions | |
ishes | -0.080 |
ery | -0.079 |
adoes | -0.076 |
WI | -0.076 |
suscept | -0.074 |
Token | misogyn |
Token ID | 22795 |
Feature activation | -8.50e+00 |
Pos logit contributions | |
ishes | +0.873 |
ery | +0.862 |
adoes | +0.834 |
WI | +0.823 |
suscept | +0.810 |
Neg logit contributions | |
Surgery | -0.979 |
Broadcasting | -0.964 |
scope | -0.932 |
hra | -0.918 |
aviour | -0.895 |
Token | istic |
Token ID | 2569 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
ery | +5.739 |
ishes | +5.610 |
adoes | +5.201 |
WI | +5.027 |
eries | +4.948 |
Neg logit contributions | |
Surgery | -7.280 |
Broadcasting | -7.116 |
scope | -6.815 |
hra | -6.615 |
edience | -6.402 |
Token | cler |
Token ID | 13451 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
ishes | +2.449 |
ery | +2.443 |
adoes | +2.346 |
WI | +2.332 |
suscept | +2.293 |
Neg logit contributions | |
Surgery | -2.089 |
Broadcasting | -2.031 |
scope | -1.977 |
hra | -1.941 |
aviour | -1.888 |
Token | ical |
Token ID | 605 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
Surgery | +2.075 |
Broadcasting | +2.049 |
scope | +1.976 |
hra | +1.954 |
aviour | +1.902 |
Neg logit contributions | |
ishes | -2.408 |
ery | -2.382 |
adoes | -2.337 |
WI | -2.309 |
suscept | -2.298 |
Token | culture |
Token ID | 3968 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
Surgery | +0.144 |
Broadcasting | +0.141 |
scope | +0.134 |
hra | +0.131 |
aviour | +0.127 |
Neg logit contributions | |
ishes | -0.222 |
ery | -0.220 |
adoes | -0.214 |
WI | -0.212 |
suscept | -0.209 |
Token | she |
Token ID | 673 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
ishes | +1.176 |
ery | +1.161 |
adoes | +1.121 |
WI | +1.107 |
suscept | +1.094 |
Neg logit contributions | |
Surgery | -1.312 |
Broadcasting | -1.289 |
scope | -1.245 |
hra | -1.222 |
aviour | -1.194 |
Token | our |
Token ID | 454 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
ishes | +1.305 |
ery | +1.268 |
adoes | +1.188 |
WI | +1.144 |
iors | +1.111 |
Neg logit contributions | |
Surgery | -3.089 |
Broadcasting | -3.053 |
scope | -2.951 |
hra | -2.918 |
ndum | -2.852 |
Token | — |
Token ID | 960 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ishes | +2.349 |
ery | +2.316 |
adoes | +2.245 |
WI | +2.211 |
suscept | +2.186 |
Neg logit contributions | |
Surgery | -1.869 |
Broadcasting | -1.848 |
scope | -1.778 |
hra | -1.747 |
aviour | -1.703 |
Token | Pl |
Token ID | 3646 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
Surgery | +0.138 |
Broadcasting | +0.135 |
scope | +0.130 |
hra | +0.128 |
aviour | +0.124 |
Neg logit contributions | |
ishes | -0.143 |
ery | -0.142 |
adoes | -0.138 |
WI | -0.136 |
suscept | -0.134 |
Token | ant |
Token ID | 415 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ishes | +0.318 |
ery | +0.312 |
adoes | +0.300 |
WI | +0.295 |
suscept | +0.289 |
Neg logit contributions | |
Surgery | -0.546 |
Broadcasting | -0.539 |
scope | -0.524 |
hra | -0.517 |
aviour | -0.506 |
Token | ain |
Token ID | 391 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
ishes | +0.932 |
ery | +0.919 |
adoes | +0.890 |
WI | +0.878 |
suscept | +0.866 |
Neg logit contributions | |
Surgery | -0.981 |
Broadcasting | -0.965 |
scope | -0.933 |
hra | -0.916 |
aviour | -0.893 |
Token | flour |
Token ID | 10601 |
Feature activation | -8.45e+00 |
Pos logit contributions | |
ishes | +0.591 |
ery | +0.582 |
adoes | +0.563 |
WI | +0.556 |
suscept | +0.547 |
Neg logit contributions | |
Surgery | -0.776 |
Broadcasting | -0.763 |
scope | -0.740 |
hra | -0.727 |
aviour | -0.711 |
Token | is |
Token ID | 318 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
ishes | +5.862 |
ery | +5.595 |
adoes | +5.367 |
WI | +5.209 |
suscept | +5.065 |
Neg logit contributions | |
Surgery | -7.133 |
Broadcasting | -7.044 |
scope | -6.785 |
hra | -6.764 |
aviour | -6.588 |
Token | quickly |
Token ID | 2952 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ishes | +0.541 |
ery | +0.534 |
adoes | +0.519 |
WI | +0.513 |
suscept | +0.506 |
Neg logit contributions | |
Surgery | -0.486 |
Broadcasting | -0.477 |
scope | -0.460 |
hra | -0.451 |
aviour | -0.438 |
Token | becoming |
Token ID | 5033 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
ishes | +1.170 |
ery | +1.154 |
adoes | +1.123 |
WI | +1.109 |
suscept | +1.091 |
Neg logit contributions | |
Surgery | -1.146 |
Broadcasting | -1.127 |
scope | -1.087 |
hra | -1.066 |
aviour | -1.038 |
Token | one |
Token ID | 530 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
ishes | +2.786 |
ery | +2.752 |
adoes | +2.668 |
WI | +2.626 |
suscept | +2.624 |
Neg logit contributions | |
Surgery | -2.383 |
Broadcasting | -2.351 |
scope | -2.272 |
hra | -2.215 |
aviour | -2.180 |
Token | of |
Token ID | 286 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
Surgery | +0.980 |
Broadcasting | +0.958 |
scope | +0.925 |
hra | +0.907 |
aviour | +0.872 |
Neg logit contributions | |
ishes | -1.237 |
ery | -1.223 |
adoes | -1.199 |
WI | -1.179 |
suscept | -1.169 |
Token | well |
Token ID | 880 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
Surgery | +3.321 |
Broadcasting | +3.282 |
scope | +3.180 |
hra | +3.107 |
aviour | +3.040 |
Neg logit contributions | |
ishes | -3.170 |
ery | -3.093 |
adoes | -3.071 |
WI | -3.016 |
suscept | -2.947 |
Token | - |
Token ID | 12 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
Surgery | +0.673 |
Broadcasting | +0.663 |
scope | +0.642 |
hra | +0.629 |
aviour | +0.616 |
Neg logit contributions | |
ishes | -0.623 |
ery | -0.612 |
adoes | -0.601 |
WI | -0.588 |
suscept | -0.581 |
Token | known |
Token ID | 4002 |
Feature activation | +9.30e-01 |
Pos logit contributions | |
Surgery | +0.132 |
Broadcasting | +0.129 |
scope | +0.123 |
hra | +0.121 |
aviour | +0.117 |
Neg logit contributions | |
ishes | -0.198 |
ery | -0.196 |
adoes | -0.191 |
WI | -0.189 |
suscept | -0.187 |
Token | type |
Token ID | 2099 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Surgery | +0.736 |
Broadcasting | +0.724 |
scope | +0.701 |
hra | +0.687 |
aviour | +0.670 |
Neg logit contributions | |
ishes | -0.697 |
ery | -0.687 |
adoes | -0.668 |
WI | -0.660 |
suscept | -0.648 |
Token | of |
Token ID | 286 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
ishes | +0.995 |
ery | +0.986 |
adoes | +0.953 |
WI | +0.941 |
suscept | +0.932 |
Neg logit contributions | |
Surgery | -0.896 |
Broadcasting | -0.882 |
scope | -0.845 |
hra | -0.832 |
aviour | -0.809 |
Token | ign |
Token ID | 3627 |
Feature activation | -8.44e+00 |
Pos logit contributions | |
ishes | +0.560 |
ery | +0.552 |
adoes | +0.534 |
WI | +0.526 |
suscept | +0.519 |
Neg logit contributions | |
Surgery | -0.642 |
Broadcasting | -0.632 |
scope | -0.612 |
hra | -0.602 |
aviour | -0.587 |
Token | eous |
Token ID | 15303 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
ery | +2.698 |
ishes | +2.585 |
adoes | +2.233 |
iors | +2.036 |
eries | +2.008 |
Neg logit contributions | |
Surgery | -10.254 |
Broadcasting | -10.148 |
hra | -9.713 |
scope | -9.660 |
ndum | -9.463 |
Token | rock |
Token ID | 3881 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ishes | +0.641 |
ery | +0.634 |
adoes | +0.609 |
WI | +0.600 |
suscept | +0.595 |
Neg logit contributions | |
Surgery | -0.721 |
Broadcasting | -0.709 |
scope | -0.684 |
hra | -0.675 |
aviour | -0.657 |
Token | found |
Token ID | 1043 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ishes | +0.964 |
ery | +0.955 |
adoes | +0.917 |
WI | +0.903 |
suscept | +0.896 |
Neg logit contributions | |
Surgery | -1.055 |
Broadcasting | -1.039 |
scope | -0.998 |
hra | -0.986 |
aviour | -0.962 |
Token | in |
Token ID | 287 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
Surgery | +0.765 |
Broadcasting | +0.751 |
scope | +0.725 |
hra | +0.711 |
aviour | +0.692 |
Neg logit contributions | |
ishes | -0.813 |
ery | -0.803 |
adoes | -0.780 |
WI | -0.770 |
suscept | -0.760 |
Token | many |
Token ID | 867 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
ishes | +0.155 |
ery | +0.153 |
adoes | +0.149 |
WI | +0.147 |
suscept | +0.145 |
Neg logit contributions | |
Surgery | -0.149 |
Broadcasting | -0.146 |
scope | -0.141 |
hra | -0.139 |
aviour | -0.135 |
Token | a |
Token ID | 257 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
ishes | +1.021 |
ery | +1.009 |
adoes | +0.985 |
WI | +0.966 |
suscept | +0.954 |
Neg logit contributions | |
Surgery | -0.939 |
Broadcasting | -0.921 |
scope | -0.894 |
hra | -0.874 |
aviour | -0.853 |
Token | history |
Token ID | 2106 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
ishes | +0.548 |
ery | +0.543 |
adoes | +0.525 |
WI | +0.518 |
suscept | +0.510 |
Neg logit contributions | |
Surgery | -0.568 |
Broadcasting | -0.558 |
scope | -0.540 |
hra | -0.530 |
aviour | -0.518 |
Token | of |
Token ID | 286 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
Surgery | +0.518 |
Broadcasting | +0.509 |
scope | +0.490 |
hra | +0.481 |
aviour | +0.465 |
Neg logit contributions | |
ishes | -0.611 |
ery | -0.600 |
adoes | -0.587 |
WI | -0.579 |
suscept | -0.572 |
Token | homophobic |
Token ID | 36034 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
Surgery | +0.283 |
Broadcasting | +0.278 |
scope | +0.269 |
hra | +0.265 |
aviour | +0.258 |
Neg logit contributions | |
ishes | -0.250 |
ery | -0.246 |
adoes | -0.239 |
WI | -0.235 |
suscept | -0.232 |
Token | , |
Token ID | 11 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Surgery | +0.487 |
Broadcasting | +0.481 |
scope | +0.463 |
hra | +0.455 |
aviour | +0.447 |
Neg logit contributions | |
ishes | -0.426 |
ery | -0.417 |
adoes | -0.407 |
WI | -0.403 |
suscept | -0.392 |
Token | misogyn |
Token ID | 22795 |
Feature activation | -8.42e+00 |
Pos logit contributions | |
ishes | +0.730 |
ery | +0.723 |
adoes | +0.695 |
WI | +0.686 |
suscept | +0.676 |
Neg logit contributions | |
Surgery | -0.924 |
Broadcasting | -0.909 |
scope | -0.881 |
hra | -0.867 |
aviour | -0.844 |
Token | ist |
Token ID | 396 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
ery | +5.537 |
ishes | +5.380 |
adoes | +5.075 |
WI | +4.926 |
iors | +4.835 |
Neg logit contributions | |
Surgery | -7.494 |
Broadcasting | -7.318 |
scope | -6.965 |
hra | -6.823 |
ndum | -6.628 |
Token | , |
Token ID | 11 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
ishes | +2.066 |
ery | +2.047 |
adoes | +1.978 |
WI | +1.942 |
suscept | +1.923 |
Neg logit contributions | |
Surgery | -2.384 |
Broadcasting | -2.339 |
scope | -2.270 |
hra | -2.230 |
aviour | -2.168 |
Token | and |
Token ID | 290 |
Feature activation | -9.42e-01 |
Pos logit contributions | |
ishes | +1.085 |
ery | +1.075 |
adoes | +1.036 |
WI | +1.024 |
suscept | +1.009 |
Neg logit contributions | |
Surgery | -1.247 |
Broadcasting | -1.225 |
scope | -1.185 |
hra | -1.169 |
aviour | -1.135 |
Token | racist |
Token ID | 9811 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
ishes | +0.636 |
ery | +0.628 |
adoes | +0.605 |
WI | +0.596 |
suscept | +0.587 |
Neg logit contributions | |
Surgery | -0.818 |
Broadcasting | -0.805 |
scope | -0.781 |
hra | -0.768 |
aviour | -0.750 |
Token | comments |
Token ID | 3651 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
ishes | +2.065 |
ery | +2.050 |
adoes | +1.969 |
suscept | +1.930 |
WI | +1.928 |
Neg logit contributions | |
Surgery | -2.542 |
Broadcasting | -2.493 |
scope | -2.434 |
hra | -2.387 |
aviour | -2.322 |
Token | almond |
Token ID | 26948 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
Surgery | +0.252 |
Broadcasting | +0.248 |
scope | +0.241 |
hra | +0.237 |
aviour | +0.232 |
Neg logit contributions | |
ishes | -0.173 |
ery | -0.169 |
adoes | -0.163 |
WI | -0.161 |
suscept | -0.158 |
Token | flour |
Token ID | 10601 |
Feature activation | -7.72e+00 |
Pos logit contributions | |
ishes | +0.234 |
ery | +0.228 |
adoes | +0.222 |
WI | +0.218 |
suscept | +0.214 |
Neg logit contributions | |
Surgery | -0.406 |
Broadcasting | -0.401 |
scope | -0.391 |
hra | -0.386 |
aviour | -0.376 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ishes | +5.362 |
ery | +5.322 |
adoes | +5.005 |
WI | +4.919 |
suscept | +4.832 |
Neg logit contributions | |
Surgery | -6.343 |
Broadcasting | -6.225 |
scope | -6.031 |
hra | -5.966 |
aviour | -5.911 |
Token | Pumpkin |
Token ID | 36724 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
ishes | +0.133 |
ery | +0.132 |
adoes | +0.128 |
WI | +0.127 |
suscept | +0.125 |
Neg logit contributions | |
Surgery | -0.088 |
Broadcasting | -0.087 |
scope | -0.084 |
hra | -0.082 |
aviour | -0.079 |
Token | seed |
Token ID | 9403 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
Surgery | +0.607 |
Broadcasting | +0.595 |
scope | +0.577 |
hra | +0.563 |
aviour | +0.548 |
Neg logit contributions | |
ishes | -0.669 |
ery | -0.660 |
adoes | -0.648 |
WI | -0.644 |
suscept | -0.637 |
Token | flour |
Token ID | 10601 |
Feature activation | -8.40e+00 |
Pos logit contributions | |
ishes | +1.440 |
ery | +1.413 |
adoes | +1.353 |
WI | +1.327 |
suscept | +1.300 |
Neg logit contributions | |
Surgery | -2.697 |
Broadcasting | -2.661 |
scope | -2.592 |
hra | -2.556 |
aviour | -2.506 |
Token | measures |
Token ID | 5260 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
ishes | +5.549 |
ery | +5.429 |
adoes | +5.116 |
WI | +4.978 |
suscept | +4.844 |
Neg logit contributions | |
Surgery | -7.302 |
Broadcasting | -7.181 |
scope | -6.923 |
hra | -6.819 |
aviour | -6.731 |
Token | more |
Token ID | 517 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
Surgery | +0.549 |
Broadcasting | +0.539 |
scope | +0.523 |
hra | +0.513 |
aviour | +0.499 |
Neg logit contributions | |
ishes | -0.513 |
ery | -0.506 |
adoes | -0.491 |
WI | -0.485 |
suscept | -0.476 |
Token | closely |
Token ID | 7173 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
ishes | +1.095 |
ery | +1.078 |
adoes | +1.038 |
WI | +1.022 |
suscept | +1.011 |
Neg logit contributions | |
Surgery | -1.453 |
Broadcasting | -1.431 |
scope | -1.383 |
hra | -1.363 |
aviour | -1.334 |
Token | to |
Token ID | 284 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
ishes | +1.623 |
ery | +1.595 |
adoes | +1.540 |
WI | +1.518 |
suscept | +1.496 |
Neg logit contributions | |
Surgery | -1.993 |
Broadcasting | -1.962 |
scope | -1.896 |
hra | -1.866 |
aviour | -1.825 |
Token | almond |
Token ID | 26948 |
Feature activation | -9.59e-01 |
Pos logit contributions | |
ishes | +0.100 |
ery | +0.098 |
adoes | +0.095 |
WI | +0.094 |
suscept | +0.092 |
Neg logit contributions | |
Surgery | -0.123 |
Broadcasting | -0.121 |
scope | -0.118 |
hra | -0.116 |
aviour | -0.113 |
Token | writing |
Token ID | 3597 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
ishes | +1.445 |
ery | +1.432 |
adoes | +1.379 |
WI | +1.357 |
suscept | +1.346 |
Neg logit contributions | |
Surgery | -1.536 |
Broadcasting | -1.505 |
scope | -1.455 |
hra | -1.433 |
aviour | -1.393 |
Token | vicious |
Token ID | 17686 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ishes | +3.610 |
ery | +3.565 |
adoes | +3.403 |
suscept | +3.341 |
iors | +3.325 |
Neg logit contributions | |
Surgery | -3.642 |
Broadcasting | -3.584 |
scope | -3.497 |
hra | -3.474 |
ndum | -3.348 |
Token | ly |
Token ID | 306 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
ishes | +0.973 |
ery | +0.958 |
adoes | +0.926 |
WI | +0.915 |
suscept | +0.897 |
Neg logit contributions | |
Surgery | -1.263 |
Broadcasting | -1.244 |
scope | -1.206 |
hra | -1.185 |
aviour | -1.160 |
Token | pornographic |
Token ID | 47642 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ishes | +1.814 |
ery | +1.788 |
adoes | +1.728 |
WI | +1.704 |
suscept | +1.681 |
Neg logit contributions | |
Surgery | -2.175 |
Broadcasting | -2.140 |
scope | -2.071 |
hra | -2.042 |
aviour | -1.986 |
Token | , |
Token ID | 11 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
ishes | +2.439 |
ery | +2.400 |
adoes | +2.318 |
WI | +2.266 |
suscept | +2.250 |
Neg logit contributions | |
Surgery | -2.963 |
Broadcasting | -2.910 |
scope | -2.838 |
hra | -2.795 |
aviour | -2.719 |
Token | misogyn |
Token ID | 22795 |
Feature activation | -8.38e+00 |
Pos logit contributions | |
ishes | +0.375 |
ery | +0.369 |
adoes | +0.358 |
WI | +0.353 |
suscept | +0.347 |
Neg logit contributions | |
Surgery | -0.463 |
Broadcasting | -0.455 |
scope | -0.441 |
hra | -0.434 |
aviour | -0.424 |
Token | istic |
Token ID | 2569 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
ery | +5.279 |
ishes | +5.197 |
adoes | +4.832 |
WI | +4.626 |
seless | +4.577 |
Neg logit contributions | |
Surgery | -7.608 |
Broadcasting | -7.428 |
scope | -7.096 |
hra | -7.021 |
ndum | -6.798 |
Token | stories |
Token ID | 3923 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
ishes | +2.748 |
ery | +2.718 |
adoes | +2.621 |
WI | +2.558 |
suscept | +2.546 |
Neg logit contributions | |
Surgery | -3.245 |
Broadcasting | -3.188 |
scope | -3.102 |
hra | -3.056 |
aviour | -2.965 |
Token | he |
Token ID | 339 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
ery | +2.672 |
ishes | +2.669 |
adoes | +2.529 |
WI | +2.481 |
suscept | +2.476 |
Neg logit contributions | |
Surgery | -3.109 |
Broadcasting | -3.057 |
scope | -2.962 |
hra | -2.923 |
ndum | -2.848 |
Token | would |
Token ID | 561 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
ishes | +0.589 |
ery | +0.581 |
adoes | +0.564 |
WI | +0.555 |
suscept | +0.547 |
Neg logit contributions | |
Surgery | -0.636 |
Broadcasting | -0.625 |
scope | -0.605 |
hra | -0.594 |
aviour | -0.579 |
Token | try |
Token ID | 1949 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
Surgery | +0.046 |
Broadcasting | +0.045 |
scope | +0.043 |
hra | +0.043 |
aviour | +0.042 |
Neg logit contributions | |
ishes | -0.050 |
ery | -0.049 |
adoes | -0.048 |
WI | -0.047 |
suscept | -0.046 |
Token | generally |
Token ID | 4143 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
ishes | +0.064 |
ery | +0.064 |
adoes | +0.062 |
WI | +0.061 |
suscept | +0.060 |
Neg logit contributions | |
Surgery | -0.064 |
Broadcasting | -0.063 |
scope | -0.061 |
hra | -0.059 |
aviour | -0.058 |
Token | can |
Token ID | 460 |
Feature activation | -8.08e-03 |
Pos logit contributions | |
ishes | +1.242 |
ery | +1.224 |
adoes | +1.181 |
WI | +1.166 |
suscept | +1.153 |
Neg logit contributions | |
Surgery | -1.335 |
Broadcasting | -1.311 |
scope | -1.266 |
hra | -1.239 |
aviour | -1.217 |
Token | be |
Token ID | 307 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ishes | +0.007 |
ery | +0.007 |
adoes | +0.007 |
WI | +0.007 |
suscept | +0.006 |
Neg logit contributions | |
Surgery | -0.006 |
Broadcasting | -0.005 |
scope | -0.005 |
hra | -0.005 |
aviour | -0.005 |
Token | substitute |
Token ID | 15373 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ishes | +0.749 |
ery | +0.736 |
adoes | +0.709 |
WI | +0.696 |
suscept | +0.683 |
Neg logit contributions | |
Surgery | -1.197 |
Broadcasting | -1.180 |
scope | -1.147 |
hra | -1.130 |
aviour | -1.106 |
Token | wheat |
Token ID | 17135 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
Surgery | +1.329 |
Broadcasting | +1.301 |
scope | +1.255 |
hra | +1.236 |
aviour | +1.193 |
Neg logit contributions | |
ishes | -1.479 |
ery | -1.461 |
adoes | -1.430 |
WI | -1.421 |
suscept | -1.397 |
Token | flour |
Token ID | 10601 |
Feature activation | -8.33e+00 |
Pos logit contributions | |
ishes | +2.439 |
ery | +2.385 |
adoes | +2.267 |
WI | +2.219 |
suscept | +2.188 |
Neg logit contributions | |
Surgery | -3.696 |
Broadcasting | -3.641 |
scope | -3.538 |
hra | -3.469 |
aviour | -3.416 |
Token | 1 |
Token ID | 352 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
ishes | +6.147 |
ery | +6.016 |
adoes | +5.614 |
WI | +5.555 |
suscept | +5.486 |
Neg logit contributions | |
Surgery | -6.534 |
Broadcasting | -6.415 |
scope | -6.250 |
hra | -6.076 |
aviour | -6.020 |
Token | : |
Token ID | 25 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
ishes | +0.251 |
ery | +0.247 |
adoes | +0.240 |
WI | +0.236 |
suscept | +0.233 |
Neg logit contributions | |
Surgery | -0.280 |
Broadcasting | -0.275 |
scope | -0.266 |
hra | -0.262 |
aviour | -0.255 |
Token | 1 |
Token ID | 16 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Surgery | +0.127 |
Broadcasting | +0.124 |
scope | +0.119 |
hra | +0.116 |
aviour | +0.113 |
Neg logit contributions | |
ishes | -0.173 |
ery | -0.171 |
adoes | -0.167 |
WI | -0.165 |
suscept | -0.163 |
Token | . |
Token ID | 13 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
ishes | +0.085 |
ery | +0.084 |
adoes | +0.082 |
WI | +0.080 |
suscept | +0.079 |
Neg logit contributions | |
Surgery | -0.081 |
Broadcasting | -0.080 |
scope | -0.077 |
hra | -0.076 |
aviour | -0.073 |
Token | It |
Token ID | 632 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
ishes | +0.242 |
ery | +0.240 |
adoes | +0.233 |
WI | +0.231 |
suscept | +0.228 |
Neg logit contributions | |
Surgery | -0.167 |
Broadcasting | -0.164 |
scope | -0.157 |
hra | -0.154 |
aviour | -0.150 |
Token | buy |
Token ID | 2822 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
ishes | +0.369 |
ery | +0.364 |
adoes | +0.353 |
WI | +0.350 |
suscept | +0.344 |
Neg logit contributions | |
Surgery | -0.352 |
Broadcasting | -0.345 |
scope | -0.333 |
hra | -0.327 |
aviour | -0.319 |
Token | Barry |
Token ID | 14488 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
ishes | +0.360 |
ery | +0.355 |
adoes | +0.345 |
WI | +0.341 |
suscept | +0.336 |
Neg logit contributions | |
Surgery | -0.347 |
Broadcasting | -0.341 |
scope | -0.329 |
hra | -0.323 |
aviour | -0.314 |
Token | Farm |
Token ID | 11272 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
Surgery | +2.370 |
Broadcasting | +2.328 |
scope | +2.256 |
hra | +2.189 |
aviour | +2.096 |
Neg logit contributions | |
ishes | -2.606 |
ery | -2.567 |
adoes | -2.513 |
WI | -2.484 |
suscept | -2.474 |
Token | plant |
Token ID | 4618 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
ishes | +0.351 |
ery | +0.346 |
adoes | +0.338 |
WI | +0.333 |
suscept | +0.329 |
Neg logit contributions | |
Surgery | -0.342 |
Broadcasting | -0.335 |
scope | -0.324 |
hra | -0.318 |
aviour | -0.309 |
Token | ain |
Token ID | 391 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Surgery | +0.198 |
Broadcasting | +0.196 |
scope | +0.190 |
hra | +0.188 |
aviour | +0.184 |
Neg logit contributions | |
ishes | -0.202 |
ery | -0.199 |
adoes | -0.196 |
WI | -0.195 |
suscept | -0.191 |
Token | flour |
Token ID | 10601 |
Feature activation | -8.33e+00 |
Pos logit contributions | |
ishes | +0.991 |
ery | +0.976 |
adoes | +0.944 |
WI | +0.930 |
suscept | +0.916 |
Neg logit contributions | |
Surgery | -1.227 |
Broadcasting | -1.208 |
scope | -1.171 |
hra | -1.151 |
aviour | -1.125 |
Token | from |
Token ID | 422 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
ishes | +5.923 |
ery | +5.801 |
adoes | +5.494 |
WI | +5.440 |
suscept | +5.315 |
Neg logit contributions | |
Surgery | -6.703 |
Broadcasting | -6.567 |
scope | -6.337 |
hra | -6.322 |
aviour | -6.262 |
Token | am |
Token ID | 716 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ishes | +0.645 |
ery | +0.637 |
adoes | +0.618 |
WI | +0.612 |
suscept | +0.603 |
Neg logit contributions | |
Surgery | -0.559 |
Broadcasting | -0.549 |
scope | -0.529 |
hra | -0.519 |
aviour | -0.506 |
Token | azon |
Token ID | 5168 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
ishes | +0.686 |
ery | +0.676 |
adoes | +0.647 |
WI | +0.634 |
suscept | +0.621 |
Neg logit contributions | |
Surgery | -1.107 |
Broadcasting | -1.091 |
scope | -1.059 |
hra | -1.043 |
aviour | -1.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
Surgery | +0.163 |
Broadcasting | +0.160 |
scope | +0.155 |
hra | +0.152 |
aviour | +0.148 |
Neg logit contributions | |
ishes | -0.166 |
ery | -0.163 |
adoes | -0.159 |
WI | -0.157 |
suscept | -0.155 |
Token |
|
Token ID | 198 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
ishes | +0.142 |
ery | +0.141 |
adoes | +0.136 |
WI | +0.135 |
suscept | +0.133 |
Neg logit contributions | |
Surgery | -0.109 |
Broadcasting | -0.107 |
scope | -0.102 |
hra | -0.100 |
aviour | -0.098 |
Token | with |
Token ID | 351 |
Feature activation | -8.47e-01 |
Pos logit contributions | |
ishes | +1.364 |
ery | +1.353 |
adoes | +1.293 |
WI | +1.280 |
suscept | +1.249 |
Neg logit contributions | |
Surgery | -1.662 |
Broadcasting | -1.636 |
scope | -1.593 |
hra | -1.555 |
aviour | -1.528 |
Token | the |
Token ID | 262 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
ishes | +0.677 |
ery | +0.669 |
adoes | +0.649 |
WI | +0.642 |
suscept | +0.632 |
Neg logit contributions | |
Surgery | -0.629 |
Broadcasting | -0.616 |
scope | -0.596 |
hra | -0.585 |
aviour | -0.569 |
Token | Russian |
Token ID | 3394 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ishes | +0.421 |
ery | +0.415 |
adoes | +0.403 |
WI | +0.398 |
suscept | +0.392 |
Neg logit contributions | |
Surgery | -0.428 |
Broadcasting | -0.420 |
scope | -0.406 |
hra | -0.399 |
aviour | -0.389 |
Token | government |
Token ID | 1230 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
ishes | +1.615 |
ery | +1.597 |
adoes | +1.538 |
WI | +1.526 |
suscept | +1.498 |
Neg logit contributions | |
Surgery | -1.776 |
Broadcasting | -1.735 |
scope | -1.691 |
hra | -1.667 |
aviour | -1.624 |
Token | , |
Token ID | 11 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
Surgery | +0.047 |
Broadcasting | +0.046 |
scope | +0.045 |
hra | +0.044 |
aviour | +0.043 |
Neg logit contributions | |
ishes | -0.045 |
ery | -0.044 |
adoes | -0.043 |
WI | -0.042 |
suscept | -0.041 |
Token | misogyn |
Token ID | 22795 |
Feature activation | -8.27e+00 |
Pos logit contributions | |
ishes | +0.293 |
ery | +0.290 |
adoes | +0.281 |
WI | +0.280 |
suscept | +0.275 |
Neg logit contributions | |
Surgery | -0.231 |
Broadcasting | -0.227 |
scope | -0.219 |
hra | -0.214 |
aviour | -0.208 |
Token | istic |
Token ID | 2569 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ery | +5.827 |
ishes | +5.813 |
adoes | +5.428 |
WI | +5.303 |
iors | +5.252 |
Neg logit contributions | |
Surgery | -6.836 |
Broadcasting | -6.689 |
scope | -6.457 |
hra | -6.285 |
aviour | -6.115 |
Token | and |
Token ID | 290 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ishes | +2.495 |
ery | +2.471 |
adoes | +2.373 |
WI | +2.329 |
suscept | +2.311 |
Neg logit contributions | |
Surgery | -2.891 |
Broadcasting | -2.817 |
scope | -2.773 |
hra | -2.714 |
aviour | -2.637 |
Token | hateful |
Token ID | 30367 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
ishes | +0.856 |
ery | +0.844 |
adoes | +0.813 |
WI | +0.802 |
suscept | +0.789 |
Neg logit contributions | |
Surgery | -1.116 |
Broadcasting | -1.098 |
scope | -1.064 |
hra | -1.047 |
aviour | -1.023 |
Token | rhetoric |
Token ID | 13214 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
ishes | +1.922 |
ery | +1.904 |
adoes | +1.833 |
WI | +1.807 |
suscept | +1.785 |
Neg logit contributions | |
Surgery | -2.204 |
Broadcasting | -2.155 |
scope | -2.110 |
hra | -2.060 |
aviour | -2.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
ishes | +2.682 |
ery | +2.651 |
adoes | +2.553 |
WI | +2.527 |
suscept | +2.482 |
Neg logit contributions | |
Surgery | -3.067 |
Broadcasting | -3.013 |
scope | -2.941 |
hra | -2.859 |
aviour | -2.793 |
Token | interview |
Token ID | 2720 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
Surgery | +1.687 |
Broadcasting | +1.661 |
scope | +1.599 |
hra | +1.567 |
aviour | +1.531 |
Neg logit contributions | |
ishes | -1.745 |
ery | -1.713 |
adoes | -1.672 |
WI | -1.648 |
suscept | -1.629 |
Token | . |
Token ID | 13 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
Surgery | +2.592 |
Broadcasting | +2.542 |
scope | +2.442 |
hra | +2.408 |
aviour | +2.344 |
Neg logit contributions | |
ishes | -2.620 |
ery | -2.560 |
adoes | -2.532 |
WI | -2.471 |
suscept | -2.438 |
Token |
|
Token ID | 198 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
Surgery | +0.389 |
Broadcasting | +0.382 |
scope | +0.370 |
hra | +0.362 |
aviour | +0.352 |
Neg logit contributions | |
ishes | -0.551 |
ery | -0.546 |
adoes | -0.529 |
WI | -0.527 |
suscept | -0.517 |
Token |
|
Token ID | 198 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Surgery | +0.288 |
Broadcasting | +0.281 |
scope | +0.268 |
hra | +0.263 |
aviour | +0.257 |
Neg logit contributions | |
ishes | -0.390 |
ery | -0.388 |
adoes | -0.375 |
WI | -0.373 |
iors | -0.365 |
Token | A |
Token ID | 32 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Surgery | +0.124 |
Broadcasting | +0.122 |
scope | +0.117 |
hra | +0.115 |
aviour | +0.112 |
Neg logit contributions | |
ishes | -0.143 |
ery | -0.141 |
adoes | -0.137 |
WI | -0.136 |
suscept | -0.133 |
Token | Norwegian |
Token ID | 22158 |
Feature activation | +4.80e+00 |
Pos logit contributions | |
ishes | +1.223 |
ery | +1.215 |
adoes | +1.165 |
WI | +1.162 |
suscept | +1.139 |
Neg logit contributions | |
Surgery | -1.195 |
Broadcasting | -1.178 |
scope | -1.136 |
hra | -1.112 |
aviour | -1.083 |
Token | team |
Token ID | 1074 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
Surgery | +3.190 |
Broadcasting | +3.186 |
scope | +2.985 |
hra | +2.877 |
aviour | +2.823 |
Neg logit contributions | |
ishes | -4.156 |
ery | -4.087 |
adoes | -4.028 |
WI | -3.942 |
suscept | -3.895 |
Token | will |
Token ID | 481 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
ishes | +0.435 |
ery | +0.430 |
adoes | +0.416 |
WI | +0.411 |
suscept | +0.405 |
Neg logit contributions | |
Surgery | -0.459 |
Broadcasting | -0.451 |
scope | -0.436 |
hra | -0.428 |
aviour | -0.417 |
Token | have |
Token ID | 423 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
Surgery | +0.536 |
Broadcasting | +0.526 |
scope | +0.506 |
hra | +0.496 |
aviour | +0.482 |
Neg logit contributions | |
ishes | -0.639 |
ery | -0.631 |
adoes | -0.614 |
WI | -0.607 |
suscept | -0.599 |
Token | a |
Token ID | 257 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
Surgery | +0.878 |
Broadcasting | +0.862 |
scope | +0.830 |
hra | +0.814 |
aviour | +0.791 |
Neg logit contributions | |
ishes | -0.978 |
ery | -0.965 |
adoes | -0.939 |
WI | -0.927 |
suscept | -0.915 |
Token | chance |
Token ID | 2863 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
Surgery | +0.281 |
Broadcasting | +0.276 |
scope | +0.267 |
hra | +0.263 |
aviour | +0.256 |
Neg logit contributions | |
ishes | -0.252 |
ery | -0.249 |
adoes | -0.241 |
WI | -0.238 |
suscept | -0.234 |
Token | when |
Token ID | 618 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Surgery | +2.966 |
Broadcasting | +2.926 |
scope | +2.796 |
hra | +2.773 |
aviour | +2.657 |
Neg logit contributions | |
ishes | -3.315 |
ery | -3.231 |
adoes | -3.179 |
WI | -3.139 |
suscept | -3.101 |
Token | the |
Token ID | 262 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
ishes | +0.833 |
ery | +0.825 |
adoes | +0.800 |
WI | +0.787 |
suscept | +0.778 |
Neg logit contributions | |
Surgery | -0.786 |
Broadcasting | -0.772 |
scope | -0.745 |
hra | -0.731 |
aviour | -0.714 |
Token | sports |
Token ID | 5701 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
ishes | +0.419 |
ery | +0.413 |
adoes | +0.400 |
WI | +0.392 |
suscept | +0.387 |
Neg logit contributions | |
Surgery | -0.472 |
Broadcasting | -0.464 |
scope | -0.450 |
hra | -0.444 |
aviour | -0.434 |
Token | teams |
Token ID | 3466 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
Surgery | +2.660 |
Broadcasting | +2.635 |
scope | +2.542 |
hra | +2.465 |
aviour | +2.399 |
Neg logit contributions | |
ishes | -2.742 |
ery | -2.683 |
adoes | -2.635 |
WI | -2.605 |
suscept | -2.575 |
Token | were |
Token ID | 547 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
ery | +1.690 |
ishes | +1.690 |
adoes | +1.625 |
WI | +1.601 |
suscept | +1.569 |
Neg logit contributions | |
Surgery | -1.832 |
Broadcasting | -1.789 |
scope | -1.748 |
hra | -1.716 |
aviour | -1.683 |
Token | eliminated |
Token ID | 15254 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
ishes | +0.652 |
ery | +0.640 |
adoes | +0.619 |
WI | +0.611 |
suscept | +0.601 |
Neg logit contributions | |
Surgery | -0.854 |
Broadcasting | -0.843 |
scope | -0.816 |
hra | -0.803 |
aviour | -0.784 |
Token | . |
Token ID | 13 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
Surgery | +3.102 |
Broadcasting | +3.043 |
scope | +2.927 |
hra | +2.901 |
aviour | +2.787 |
Neg logit contributions | |
ishes | -3.307 |
ery | -3.217 |
adoes | -3.159 |
WI | -3.137 |
suscept | -3.097 |
Token | They |
Token ID | 1119 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
ishes | +0.366 |
ery | +0.363 |
adoes | +0.352 |
WI | +0.349 |
iors | +0.343 |
Neg logit contributions | |
Surgery | -0.265 |
Broadcasting | -0.260 |
scope | -0.253 |
hra | -0.248 |
aviour | -0.243 |
Token | could |
Token ID | 714 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
ishes | +0.045 |
ery | +0.044 |
adoes | +0.043 |
WI | +0.042 |
suscept | +0.041 |
Neg logit contributions | |
Surgery | -0.050 |
Broadcasting | -0.049 |
scope | -0.047 |
hra | -0.046 |
aviour | -0.045 |
Token | send |
Token ID | 3758 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
Surgery | +0.251 |
Broadcasting | +0.246 |
scope | +0.237 |
hra | +0.232 |
aviour | +0.225 |
Neg logit contributions | |
ishes | -0.311 |
ery | -0.307 |
adoes | -0.299 |
WI | -0.296 |
suscept | -0.292 |
Token | their |
Token ID | 511 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
ishes | +1.408 |
ery | +1.395 |
adoes | +1.349 |
WI | +1.323 |
suscept | +1.303 |
Neg logit contributions | |
Surgery | -1.663 |
Broadcasting | -1.633 |
scope | -1.584 |
hra | -1.561 |
aviour | -1.528 |
Token | case |
Token ID | 1339 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
Surgery | +1.142 |
Broadcasting | +1.111 |
scope | +1.072 |
hra | +1.049 |
aviour | +1.008 |
Neg logit contributions | |
ishes | -1.646 |
ery | -1.633 |
adoes | -1.597 |
WI | -1.584 |
suscept | -1.566 |
Token | who |
Token ID | 508 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ishes | +0.884 |
ery | +0.875 |
adoes | +0.844 |
WI | +0.833 |
suscept | +0.822 |
Neg logit contributions | |
Surgery | -0.895 |
Broadcasting | -0.881 |
scope | -0.850 |
hra | -0.835 |
aviour | -0.812 |
Token | took |
Token ID | 1718 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
ishes | +1.264 |
ery | +1.246 |
adoes | +1.208 |
WI | +1.187 |
suscept | +1.169 |
Neg logit contributions | |
Surgery | -1.407 |
Broadcasting | -1.387 |
scope | -1.341 |
hra | -1.316 |
aviour | -1.286 |
Token | over |
Token ID | 625 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
Surgery | +0.056 |
Broadcasting | +0.055 |
scope | +0.053 |
hra | +0.052 |
aviour | +0.050 |
Neg logit contributions | |
ishes | -0.062 |
ery | -0.061 |
adoes | -0.060 |
WI | -0.059 |
suscept | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
Surgery | +2.444 |
Broadcasting | +2.376 |
scope | +2.266 |
hra | +2.229 |
aviour | +2.157 |
Neg logit contributions | |
ishes | -2.894 |
ery | -2.843 |
adoes | -2.795 |
WI | -2.770 |
suscept | -2.715 |
Token | asylum |
Token ID | 13142 |
Feature activation | +4.97e+00 |
Pos logit contributions | |
Surgery | +0.237 |
Broadcasting | +0.232 |
scope | +0.222 |
hra | +0.218 |
aviour | +0.211 |
Neg logit contributions | |
ishes | -0.314 |
ery | -0.311 |
adoes | -0.303 |
WI | -0.299 |
suscept | -0.296 |
Token | and |
Token ID | 290 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
Surgery | +3.658 |
Broadcasting | +3.578 |
scope | +3.439 |
hra | +3.393 |
aviour | +3.293 |
Neg logit contributions | |
ishes | -3.952 |
ery | -3.837 |
adoes | -3.808 |
WI | -3.751 |
suscept | -3.681 |
Token | is |
Token ID | 318 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
Surgery | +0.142 |
Broadcasting | +0.140 |
scope | +0.135 |
hra | +0.133 |
aviour | +0.130 |
Neg logit contributions | |
ishes | -0.121 |
ery | -0.119 |
adoes | -0.116 |
WI | -0.114 |
suscept | -0.112 |
Token | waiting |
Token ID | 4953 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Surgery | +0.513 |
Broadcasting | +0.504 |
scope | +0.487 |
hra | +0.478 |
aviour | +0.466 |
Neg logit contributions | |
ishes | -0.494 |
ery | -0.486 |
adoes | -0.474 |
WI | -0.466 |
suscept | -0.459 |
Token | to |
Token ID | 284 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
ishes | +0.855 |
ery | +0.846 |
adoes | +0.820 |
WI | +0.810 |
suscept | +0.799 |
Neg logit contributions | |
Surgery | -0.794 |
Broadcasting | -0.780 |
scope | -0.752 |
hra | -0.737 |
aviour | -0.718 |
Token | kill |
Token ID | 1494 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Surgery | +0.121 |
Broadcasting | +0.118 |
scope | +0.114 |
hra | +0.112 |
aviour | +0.109 |
Neg logit contributions | |
ishes | -0.132 |
ery | -0.131 |
adoes | -0.127 |
WI | -0.125 |
suscept | -0.124 |
Token | do |
Token ID | 466 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Surgery | +0.483 |
Broadcasting | +0.474 |
scope | +0.456 |
hra | +0.447 |
aviour | +0.434 |
Neg logit contributions | |
ishes | -0.564 |
ery | -0.557 |
adoes | -0.542 |
WI | -0.535 |
suscept | -0.528 |
Token | their |
Token ID | 511 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
Surgery | +1.082 |
Broadcasting | +1.063 |
scope | +1.021 |
hra | +1.002 |
aviour | +0.971 |
Neg logit contributions | |
ishes | -1.186 |
ery | -1.168 |
adoes | -1.141 |
WI | -1.130 |
suscept | -1.111 |
Token | characters |
Token ID | 3435 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
ishes | +0.419 |
ery | +0.414 |
adoes | +0.403 |
WI | +0.398 |
suscept | +0.392 |
Neg logit contributions | |
Surgery | -0.383 |
Broadcasting | -0.376 |
scope | -0.362 |
hra | -0.355 |
aviour | -0.346 |
Token | with |
Token ID | 351 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
ishes | +0.547 |
ery | +0.541 |
adoes | +0.522 |
WI | +0.516 |
suscept | +0.508 |
Neg logit contributions | |
Surgery | -0.534 |
Broadcasting | -0.525 |
scope | -0.507 |
hra | -0.495 |
aviour | -0.485 |
Token | their |
Token ID | 511 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
Surgery | +1.519 |
Broadcasting | +1.495 |
scope | +1.439 |
hra | +1.410 |
aviour | +1.364 |
Neg logit contributions | |
ishes | -1.688 |
ery | -1.660 |
adoes | -1.621 |
WI | -1.599 |
suscept | -1.576 |
Token | eyes |
Token ID | 2951 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
ishes | +0.384 |
ery | +0.379 |
adoes | +0.367 |
WI | +0.362 |
suscept | +0.357 |
Neg logit contributions | |
Surgery | -0.416 |
Broadcasting | -0.409 |
scope | -0.395 |
hra | -0.389 |
aviour | -0.379 |
Token | closed |
Token ID | 4838 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Broadcasting | +3.479 |
Surgery | +3.477 |
scope | +3.360 |
hra | +3.280 |
aviour | +3.162 |
Neg logit contributions | |
ishes | -3.580 |
ery | -3.504 |
adoes | -3.451 |
suscept | -3.385 |
WI | -3.358 |
Token | . |
Token ID | 13 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
Surgery | +1.244 |
Broadcasting | +1.224 |
scope | +1.182 |
hra | +1.160 |
aviour | +1.129 |
Neg logit contributions | |
ishes | -1.219 |
ery | -1.199 |
adoes | -1.167 |
WI | -1.152 |
suscept | -1.136 |
Token | So |
Token ID | 1406 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
Surgery | +0.498 |
Broadcasting | +0.487 |
scope | +0.470 |
hra | +0.460 |
aviour | +0.448 |
Neg logit contributions | |
ishes | -0.745 |
ery | -0.739 |
adoes | -0.717 |
WI | -0.714 |
suscept | -0.699 |
Token | all |
Token ID | 477 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ishes | +0.002 |
ery | +0.002 |
adoes | +0.002 |
WI | +0.002 |
suscept | +0.002 |
Neg logit contributions | |
Surgery | -0.002 |
Broadcasting | -0.002 |
scope | -0.002 |
hra | -0.002 |
aviour | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
Surgery | +0.824 |
Broadcasting | +0.811 |
scope | +0.779 |
hra | +0.764 |
aviour | +0.738 |
Neg logit contributions | |
ishes | -0.988 |
ery | -0.973 |
adoes | -0.950 |
WI | -0.939 |
suscept | -0.928 |
Token | Mr |
Token ID | 5246 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ishes | +0.446 |
ery | +0.441 |
adoes | +0.426 |
WI | +0.423 |
suscept | +0.414 |
Neg logit contributions | |
Surgery | -0.453 |
Broadcasting | -0.445 |
scope | -0.429 |
hra | -0.422 |
aviour | -0.411 |
Token | Cameron |
Token ID | 10821 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
ishes | +1.454 |
ery | +1.440 |
adoes | +1.399 |
WI | +1.385 |
suscept | +1.364 |
Neg logit contributions | |
Surgery | -1.084 |
Broadcasting | -1.060 |
scope | -1.024 |
hra | -1.002 |
aviour | -0.968 |
Token | is |
Token ID | 318 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
ishes | +0.037 |
ery | +0.036 |
adoes | +0.035 |
WI | +0.035 |
suscept | +0.034 |
Neg logit contributions | |
Surgery | -0.041 |
Broadcasting | -0.041 |
scope | -0.039 |
hra | -0.039 |
aviour | -0.038 |
Token | expected |
Token ID | 2938 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Surgery | +0.372 |
Broadcasting | +0.365 |
scope | +0.352 |
hra | +0.346 |
aviour | +0.336 |
Neg logit contributions | |
ishes | -0.403 |
ery | -0.399 |
adoes | -0.387 |
WI | -0.383 |
suscept | -0.378 |
Token | to |
Token ID | 284 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
Surgery | +0.833 |
Broadcasting | +0.819 |
scope | +0.791 |
hra | +0.777 |
aviour | +0.756 |
Neg logit contributions | |
ishes | -0.793 |
ery | -0.785 |
adoes | -0.758 |
WI | -0.748 |
suscept | -0.735 |
Token | announce |
Token ID | 5453 |
Feature activation | +4.97e+00 |
Pos logit contributions | |
Surgery | +0.306 |
Broadcasting | +0.300 |
scope | +0.288 |
hra | +0.282 |
aviour | +0.274 |
Neg logit contributions | |
ishes | -0.396 |
ery | -0.392 |
adoes | -0.382 |
WI | -0.377 |
suscept | -0.372 |
Token | this |
Token ID | 428 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
Surgery | +3.432 |
Broadcasting | +3.391 |
scope | +3.244 |
hra | +3.173 |
aviour | +3.083 |
Neg logit contributions | |
ishes | -4.189 |
ery | -4.114 |
adoes | -4.063 |
WI | -4.023 |
suscept | -3.947 |
Token | month |
Token ID | 1227 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ishes | +0.379 |
ery | +0.375 |
adoes | +0.360 |
WI | +0.355 |
suscept | +0.351 |
Neg logit contributions | |
Surgery | -0.477 |
Broadcasting | -0.469 |
scope | -0.456 |
hra | -0.448 |
aviour | -0.438 |
Token | that |
Token ID | 326 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
ishes | +0.954 |
ery | +0.953 |
adoes | +0.911 |
WI | +0.893 |
suscept | +0.889 |
Neg logit contributions | |
Surgery | -0.949 |
Broadcasting | -0.925 |
scope | -0.893 |
hra | -0.876 |
aviour | -0.854 |
Token | the |
Token ID | 262 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
Surgery | +1.330 |
Broadcasting | +1.310 |
scope | +1.260 |
hra | +1.236 |
aviour | +1.202 |
Neg logit contributions | |
ishes | -1.440 |
ery | -1.420 |
adoes | -1.385 |
WI | -1.369 |
suscept | -1.349 |
Token | Tories |
Token ID | 25098 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
ishes | +0.259 |
ery | +0.257 |
adoes | +0.248 |
WI | +0.247 |
suscept | +0.244 |
Neg logit contributions | |
Surgery | -0.199 |
Broadcasting | -0.193 |
scope | -0.187 |
hra | -0.185 |
aviour | -0.178 |
Token | 1990 |
Token ID | 6303 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ishes | +1.320 |
ery | +1.294 |
adoes | +1.262 |
WI | +1.236 |
suscept | +1.228 |
Neg logit contributions | |
Surgery | -1.335 |
Broadcasting | -1.311 |
scope | -1.277 |
hra | -1.249 |
aviour | -1.227 |
Token | and |
Token ID | 290 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
ishes | +0.933 |
ery | +0.920 |
adoes | +0.892 |
WI | +0.880 |
suscept | +0.868 |
Neg logit contributions | |
Surgery | -0.980 |
Broadcasting | -0.964 |
scope | -0.931 |
hra | -0.915 |
aviour | -0.892 |
Token | 2000 |
Token ID | 4751 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
ishes | +0.057 |
ery | +0.056 |
adoes | +0.055 |
WI | +0.054 |
suscept | +0.053 |
Neg logit contributions | |
Surgery | -0.056 |
Broadcasting | -0.055 |
scope | -0.053 |
hra | -0.052 |
aviour | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | -6.43e-01 |
Pos logit contributions | |
ishes | +0.370 |
ery | +0.366 |
adoes | +0.356 |
WI | +0.352 |
suscept | +0.347 |
Neg logit contributions | |
Surgery | -0.302 |
Broadcasting | -0.297 |
scope | -0.287 |
hra | -0.280 |
aviour | -0.273 |
Token | according |
Token ID | 1864 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
ishes | +0.516 |
ery | +0.510 |
adoes | +0.495 |
WI | +0.489 |
suscept | +0.482 |
Neg logit contributions | |
Surgery | -0.475 |
Broadcasting | -0.466 |
scope | -0.452 |
hra | -0.441 |
aviour | -0.431 |
Token | to |
Token ID | 284 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
ishes | +0.182 |
ery | +0.180 |
adoes | +0.174 |
WI | +0.173 |
iors | +0.167 |
Neg logit contributions | |
Surgery | -0.201 |
Broadcasting | -0.197 |
scope | -0.192 |
hra | -0.187 |
aviour | -0.186 |
Token | the |
Token ID | 262 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
Surgery | +0.372 |
Broadcasting | +0.364 |
scope | +0.350 |
hra | +0.343 |
aviour | +0.332 |
Neg logit contributions | |
ishes | -0.479 |
ery | -0.474 |
adoes | -0.461 |
WI | -0.456 |
suscept | -0.450 |
Token | American |
Token ID | 1605 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
ishes | +0.067 |
ery | +0.066 |
adoes | +0.064 |
WI | +0.063 |
suscept | +0.063 |
Neg logit contributions | |
Surgery | -0.051 |
Broadcasting | -0.050 |
scope | -0.049 |
hra | -0.047 |
aviour | -0.046 |
Token | Religious |
Token ID | 23945 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
Surgery | +0.676 |
Broadcasting | +0.665 |
scope | +0.638 |
hra | +0.624 |
aviour | +0.607 |
Neg logit contributions | |
ishes | -0.787 |
ery | -0.777 |
adoes | -0.758 |
WI | -0.747 |
suscept | -0.737 |
Token | Identity |
Token ID | 27207 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
Broadcasting | +3.258 |
Surgery | +3.252 |
scope | +3.048 |
hra | +2.949 |
aviour | +2.834 |
Neg logit contributions | |
ishes | -4.405 |
ery | -4.341 |
adoes | -4.252 |
WI | -4.163 |
iors | -4.161 |
Token | Survey |
Token ID | 13084 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
ishes | +0.334 |
ery | +0.332 |
adoes | +0.320 |
WI | +0.319 |
suscept | +0.313 |
Neg logit contributions | |
Surgery | -0.283 |
Broadcasting | -0.277 |
scope | -0.268 |
hra | -0.265 |
aviour | -0.258 |
Token | in |
Token ID | 287 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
ishes | +0.193 |
ery | +0.191 |
adoes | +0.184 |
WI | +0.182 |
suscept | +0.180 |
Neg logit contributions | |
Surgery | -0.186 |
Broadcasting | -0.183 |
scope | -0.177 |
hra | -0.173 |
aviour | -0.169 |
Token | January |
Token ID | 3269 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
Surgery | +0.075 |
Broadcasting | +0.074 |
scope | +0.071 |
hra | +0.070 |
aviour | +0.068 |
Neg logit contributions | |
ishes | -0.080 |
ery | -0.079 |
adoes | -0.076 |
WI | -0.075 |
suscept | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Surgery | +0.415 |
Broadcasting | +0.408 |
scope | +0.394 |
hra | +0.387 |
aviour | +0.377 |
Neg logit contributions | |
ishes | -0.404 |
ery | -0.399 |
adoes | -0.387 |
WI | -0.382 |
suscept | -0.376 |
Token |
|
Token ID | 198 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
Surgery | +0.116 |
Broadcasting | +0.114 |
scope | +0.108 |
hra | +0.106 |
aviour | +0.102 |
Neg logit contributions | |
ishes | -0.162 |
ery | -0.160 |
adoes | -0.157 |
WI | -0.155 |
suscept | -0.154 |
Token |
|
Token ID | 198 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
ishes | +0.199 |
ery | +0.198 |
adoes | +0.190 |
WI | +0.190 |
iors | +0.186 |
Neg logit contributions | |
Surgery | -0.151 |
Broadcasting | -0.148 |
scope | -0.141 |
hra | -0.138 |
aviour | -0.136 |
Token | Par |
Token ID | 10044 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
ishes | +0.341 |
ery | +0.337 |
adoes | +0.326 |
WI | +0.323 |
suscept | +0.317 |
Neg logit contributions | |
Surgery | -0.325 |
Broadcasting | -0.320 |
scope | -0.308 |
hra | -0.302 |
aviour | -0.294 |
Token | liament |
Token ID | 5130 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
ishes | +0.053 |
ery | +0.053 |
adoes | +0.051 |
WI | +0.050 |
suscept | +0.050 |
Neg logit contributions | |
Surgery | -0.068 |
Broadcasting | -0.068 |
scope | -0.065 |
hra | -0.064 |
aviour | -0.063 |
Token | Hill |
Token ID | 3327 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
ishes | +2.044 |
ery | +2.039 |
adoes | +1.952 |
WI | +1.933 |
iors | +1.887 |
Neg logit contributions | |
Surgery | -2.152 |
Broadcasting | -2.107 |
scope | -2.057 |
hra | -2.023 |
aviour | -1.963 |
Token | on |
Token ID | 319 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
Surgery | +0.094 |
Broadcasting | +0.092 |
scope | +0.089 |
hra | +0.087 |
aviour | +0.084 |
Neg logit contributions | |
ishes | -0.105 |
ery | -0.104 |
adoes | -0.101 |
WI | -0.100 |
suscept | -0.099 |
Token | Wed |
Token ID | 3300 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ishes | +1.608 |
ery | +1.591 |
adoes | +1.534 |
WI | +1.519 |
suscept | +1.493 |
Neg logit contributions | |
Surgery | -1.692 |
Broadcasting | -1.665 |
scope | -1.613 |
hra | -1.588 |
aviour | -1.550 |
Token | nes |
Token ID | 2516 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
ishes | +0.927 |
ery | +0.911 |
adoes | +0.879 |
WI | +0.866 |
iors | +0.850 |
Neg logit contributions | |
Surgery | -1.106 |
Broadcasting | -1.090 |
scope | -1.055 |
hra | -1.039 |
aviour | -1.016 |
Token | work |
Token ID | 670 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
Surgery | +0.334 |
Broadcasting | +0.328 |
scope | +0.316 |
hra | +0.310 |
aviour | +0.301 |
Neg logit contributions | |
ishes | -0.369 |
ery | -0.364 |
adoes | -0.354 |
WI | -0.350 |
suscept | -0.346 |
Token | as |
Token ID | 355 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
Surgery | +0.390 |
Broadcasting | +0.383 |
scope | +0.370 |
hra | +0.363 |
aviour | +0.353 |
Neg logit contributions | |
ishes | -0.410 |
ery | -0.405 |
adoes | -0.393 |
WI | -0.388 |
suscept | -0.383 |
Token | an |
Token ID | 281 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
Surgery | +1.393 |
Broadcasting | +1.361 |
scope | +1.301 |
hra | +1.265 |
aviour | +1.231 |
Neg logit contributions | |
ishes | -2.053 |
ery | -2.022 |
adoes | -1.986 |
WI | -1.956 |
suscept | -1.940 |
Token | officer |
Token ID | 3818 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
Surgery | +0.399 |
Broadcasting | +0.392 |
scope | +0.378 |
hra | +0.371 |
aviour | +0.361 |
Neg logit contributions | |
ishes | -0.415 |
ery | -0.410 |
adoes | -0.398 |
WI | -0.393 |
suscept | -0.387 |
Token | with |
Token ID | 351 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
ishes | +0.431 |
ery | +0.427 |
adoes | +0.410 |
WI | +0.407 |
suscept | +0.401 |
Neg logit contributions | |
Surgery | -0.447 |
Broadcasting | -0.438 |
scope | -0.423 |
hra | -0.416 |
aviour | -0.405 |
Token | another |
Token ID | 1194 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
Surgery | +0.958 |
Broadcasting | +0.938 |
scope | +0.899 |
hra | +0.876 |
aviour | +0.851 |
Neg logit contributions | |
ishes | -1.196 |
ery | -1.179 |
adoes | -1.155 |
WI | -1.134 |
suscept | -1.124 |
Token | agency |
Token ID | 4086 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
Surgery | +0.206 |
Broadcasting | +0.202 |
scope | +0.195 |
hra | +0.191 |
aviour | +0.185 |
Neg logit contributions | |
ishes | -0.243 |
ery | -0.240 |
adoes | -0.233 |
WI | -0.230 |
suscept | -0.227 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
ishes | +1.505 |
ery | +1.502 |
adoes | +1.430 |
WI | +1.428 |
suscept | +1.396 |
Neg logit contributions | |
Surgery | -1.705 |
Broadcasting | -1.667 |
scope | -1.613 |
hra | -1.589 |
aviour | -1.560 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
ishes | +0.109 |
ery | +0.108 |
WI | +0.105 |
adoes | +0.104 |
iors | +0.102 |
Neg logit contributions | |
Surgery | -0.077 |
Broadcasting | -0.075 |
scope | -0.074 |
hra | -0.072 |
aviour | -0.070 |
Token | An |
Token ID | 2025 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
Surgery | +0.318 |
Broadcasting | +0.312 |
scope | +0.302 |
hra | +0.296 |
aviour | +0.288 |
Neg logit contributions | |
ishes | -0.340 |
ery | -0.335 |
adoes | -0.326 |
WI | -0.321 |
suscept | -0.319 |
Token | All |
Token ID | 1439 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
ishes | +1.199 |
ery | +1.186 |
adoes | +1.151 |
WI | +1.135 |
suscept | +1.112 |
Neg logit contributions | |
Surgery | -1.203 |
Broadcasting | -1.182 |
scope | -1.150 |
hra | -1.120 |
aviour | -1.099 |
Token | and |
Token ID | 290 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
ishes | +0.536 |
ery | +0.530 |
adoes | +0.513 |
WI | +0.508 |
suscept | +0.502 |
Neg logit contributions | |
Surgery | -0.483 |
Broadcasting | -0.473 |
scope | -0.457 |
hra | -0.447 |
aviour | -0.436 |
Token | to |
Token ID | 284 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
ishes | +0.604 |
ery | +0.597 |
adoes | +0.577 |
WI | +0.572 |
suscept | +0.565 |
Neg logit contributions | |
Surgery | -0.539 |
Broadcasting | -0.527 |
scope | -0.507 |
hra | -0.496 |
aviour | -0.486 |
Token | work |
Token ID | 670 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
ishes | +0.150 |
ery | +0.148 |
adoes | +0.144 |
WI | +0.142 |
suscept | +0.140 |
Neg logit contributions | |
Surgery | -0.123 |
Broadcasting | -0.120 |
scope | -0.116 |
hra | -0.114 |
aviour | -0.111 |
Token | with |
Token ID | 351 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
ishes | +0.200 |
ery | +0.197 |
adoes | +0.191 |
WI | +0.189 |
suscept | +0.186 |
Neg logit contributions | |
Surgery | -0.204 |
Broadcasting | -0.201 |
scope | -0.194 |
hra | -0.190 |
aviour | -0.186 |
Token | local |
Token ID | 1957 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
Surgery | +0.407 |
Broadcasting | +0.400 |
scope | +0.385 |
hra | +0.377 |
aviour | +0.366 |
Neg logit contributions | |
ishes | -0.465 |
ery | -0.459 |
adoes | -0.447 |
WI | -0.441 |
suscept | -0.435 |
Token | retailers |
Token ID | 16167 |
Feature activation | -8.40e-01 |
Pos logit contributions | |
Surgery | +0.447 |
Broadcasting | +0.440 |
scope | +0.425 |
hra | +0.417 |
aviour | +0.407 |
Neg logit contributions | |
ishes | -0.416 |
ery | -0.411 |
adoes | -0.399 |
WI | -0.393 |
suscept | -0.387 |
Token | and |
Token ID | 290 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ishes | +0.643 |
ery | +0.637 |
adoes | +0.615 |
WI | +0.609 |
suscept | +0.599 |
Neg logit contributions | |
Surgery | -0.655 |
Broadcasting | -0.641 |
scope | -0.620 |
hra | -0.610 |
aviour | -0.593 |
Token | small |
Token ID | 1402 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
ishes | +0.985 |
ery | +0.969 |
adoes | +0.936 |
WI | +0.929 |
suscept | +0.913 |
Neg logit contributions | |
Surgery | -1.032 |
Broadcasting | -1.010 |
scope | -0.980 |
hra | -0.962 |
aviour | -0.941 |
Token | businesses |
Token ID | 5692 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
ishes | +0.366 |
ery | +0.360 |
adoes | +0.347 |
WI | +0.342 |
suscept | +0.336 |
Neg logit contributions | |
Surgery | -0.512 |
Broadcasting | -0.504 |
scope | -0.489 |
hra | -0.482 |
aviour | -0.471 |
Token | to |
Token ID | 284 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
ishes | +0.502 |
ery | +0.498 |
adoes | +0.480 |
WI | +0.476 |
suscept | +0.468 |
Neg logit contributions | |
Surgery | -0.517 |
Broadcasting | -0.506 |
scope | -0.490 |
hra | -0.481 |
aviour | -0.469 |
Token | accept |
Token ID | 2453 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ishes | +0.738 |
ery | +0.730 |
adoes | +0.707 |
WI | +0.699 |
suscept | +0.687 |
Neg logit contributions | |
Surgery | -0.659 |
Broadcasting | -0.644 |
scope | -0.621 |
hra | -0.610 |
aviour | -0.595 |
Token | . |
Token ID | 13 |
Feature activation | -3.85e-01 |
Pos logit contributions | |
ishes | +0.765 |
ery | +0.759 |
adoes | +0.731 |
WI | +0.724 |
suscept | +0.711 |
Neg logit contributions | |
Surgery | -0.859 |
Broadcasting | -0.844 |
scope | -0.819 |
hra | -0.804 |
aviour | -0.783 |
Token | She |
Token ID | 1375 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
ishes | +0.352 |
ery | +0.351 |
adoes | +0.339 |
WI | +0.339 |
iors | +0.330 |
Neg logit contributions | |
Surgery | -0.236 |
Broadcasting | -0.230 |
scope | -0.226 |
hra | -0.220 |
aviour | -0.215 |
Token | once |
Token ID | 1752 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
Surgery | +0.544 |
Broadcasting | +0.538 |
scope | +0.518 |
hra | +0.510 |
aviour | +0.499 |
Neg logit contributions | |
ishes | -0.494 |
ery | -0.486 |
adoes | -0.472 |
WI | -0.466 |
suscept | -0.461 |
Token | wrote |
Token ID | 2630 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
ishes | +0.334 |
ery | +0.329 |
adoes | +0.317 |
WI | +0.313 |
suscept | +0.307 |
Neg logit contributions | |
Surgery | -0.439 |
Broadcasting | -0.432 |
scope | -0.419 |
hra | -0.412 |
aviour | -0.403 |
Token | in |
Token ID | 287 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
ishes | +1.640 |
ery | +1.627 |
adoes | +1.573 |
WI | +1.554 |
suscept | +1.534 |
Neg logit contributions | |
Surgery | -1.730 |
Broadcasting | -1.700 |
scope | -1.648 |
hra | -1.621 |
aviour | -1.579 |
Token | the |
Token ID | 262 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
Surgery | +0.011 |
Broadcasting | +0.011 |
scope | +0.011 |
hra | +0.010 |
aviour | +0.010 |
Neg logit contributions | |
ishes | -0.015 |
ery | -0.014 |
adoes | -0.014 |
WI | -0.014 |
suscept | -0.014 |
Token | guest |
Token ID | 8319 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
Surgery | +0.316 |
Broadcasting | +0.310 |
scope | +0.299 |
hra | +0.293 |
aviour | +0.284 |
Neg logit contributions | |
ishes | -0.362 |
ery | -0.358 |
adoes | -0.348 |
WI | -0.344 |
suscept | -0.340 |
Token | book |
Token ID | 1492 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Surgery | +2.504 |
Broadcasting | +2.476 |
scope | +2.375 |
hra | +2.314 |
aviour | +2.266 |
Neg logit contributions | |
ishes | -2.476 |
ery | -2.408 |
adoes | -2.404 |
WI | -2.347 |
suscept | -2.315 |
Token | of |
Token ID | 286 |
Feature activation | +7.92e-01 |
Pos logit contributions | |
ishes | +0.888 |
ery | +0.879 |
adoes | +0.844 |
WI | +0.837 |
suscept | +0.825 |
Neg logit contributions | |
Surgery | -0.830 |
Broadcasting | -0.815 |
scope | -0.794 |
hra | -0.781 |
aviour | -0.759 |
Token | a |
Token ID | 257 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
Surgery | +0.559 |
Broadcasting | +0.549 |
scope | +0.527 |
hra | +0.516 |
aviour | +0.502 |
Neg logit contributions | |
ishes | -0.663 |
ery | -0.654 |
adoes | -0.638 |
WI | -0.630 |
suscept | -0.622 |
Token | Sh |
Token ID | 911 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
Surgery | +0.253 |
Broadcasting | +0.249 |
scope | +0.240 |
hra | +0.236 |
aviour | +0.229 |
Neg logit contributions | |
ishes | -0.262 |
ery | -0.259 |
adoes | -0.251 |
WI | -0.248 |
suscept | -0.245 |
Token | iro |
Token ID | 7058 |
Feature activation | +4.98e+00 |
Pos logit contributions | |
Surgery | +0.893 |
Broadcasting | +0.875 |
scope | +0.849 |
hra | +0.834 |
aviour | +0.807 |
Neg logit contributions | |
ishes | -0.901 |
ery | -0.882 |
adoes | -0.863 |
WI | -0.853 |
suscept | -0.843 |
Token | , |
Token ID | 11 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
Surgery | +3.747 |
Broadcasting | +3.696 |
scope | +3.552 |
hra | +3.515 |
aviour | +3.373 |
Neg logit contributions | |
ishes | -3.889 |
ery | -3.789 |
adoes | -3.757 |
suscept | -3.643 |
WI | -3.622 |
Token | the |
Token ID | 262 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Surgery | +0.213 |
Broadcasting | +0.209 |
scope | +0.202 |
hra | +0.197 |
aviour | +0.192 |
Neg logit contributions | |
ishes | -0.246 |
ery | -0.243 |
adoes | -0.236 |
WI | -0.234 |
suscept | -0.231 |
Token | president |
Token ID | 1893 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
Surgery | +0.096 |
Broadcasting | +0.094 |
scope | +0.091 |
hra | +0.089 |
aviour | +0.087 |
Neg logit contributions | |
ishes | -0.101 |
ery | -0.100 |
adoes | -0.097 |
WI | -0.096 |
suscept | -0.094 |
Token | of |
Token ID | 286 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Surgery | +1.297 |
Broadcasting | +1.282 |
scope | +1.223 |
hra | +1.198 |
aviour | +1.160 |
Neg logit contributions | |
ishes | -1.599 |
ery | -1.558 |
adoes | -1.543 |
WI | -1.507 |
suscept | -1.503 |
Token | Northwestern |
Token ID | 30197 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Surgery | +0.716 |
Broadcasting | +0.702 |
scope | +0.674 |
hra | +0.660 |
aviour | +0.641 |
Neg logit contributions | |
ishes | -0.871 |
ery | -0.860 |
adoes | -0.838 |
WI | -0.828 |
suscept | -0.818 |
Token | . |
Token ID | 13 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
ishes | +1.724 |
ery | +1.714 |
adoes | +1.651 |
WI | +1.647 |
suscept | +1.618 |
Neg logit contributions | |
Surgery | -1.450 |
Broadcasting | -1.412 |
scope | -1.373 |
hra | -1.352 |
aviour | -1.314 |
Token | � |
Token ID | 564 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
Surgery | +0.263 |
Broadcasting | +0.257 |
scope | +0.247 |
hra | +0.242 |
aviour | +0.235 |
Neg logit contributions | |
ishes | -0.335 |
ery | -0.331 |
adoes | -0.323 |
WI | -0.319 |
suscept | -0.315 |
Token | � |
Token ID | 250 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
ishes | +0.111 |
ery | +0.110 |
adoes | +0.107 |
WI | +0.106 |
suscept | +0.105 |
Neg logit contributions | |
Surgery | -0.088 |
Broadcasting | -0.087 |
scope | -0.083 |
hra | -0.082 |
aviour | -0.079 |
Token | We |
Token ID | 1135 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ishes | +0.543 |
ery | +0.538 |
adoes | +0.519 |
WI | +0.517 |
suscept | +0.504 |
Neg logit contributions | |
Surgery | -0.519 |
Broadcasting | -0.509 |
scope | -0.491 |
hra | -0.482 |
aviour | -0.471 |
Token | � |
Token ID | 447 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
Surgery | +0.741 |
Broadcasting | +0.728 |
scope | +0.702 |
hra | +0.690 |
aviour | +0.670 |
Neg logit contributions | |
ishes | -0.830 |
ery | -0.818 |
adoes | -0.798 |
WI | -0.786 |
suscept | -0.778 |
Token | man |
Token ID | 805 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
Surgery | +0.141 |
Broadcasting | +0.137 |
scope | +0.128 |
hra | +0.124 |
aviour | +0.117 |
Neg logit contributions | |
ishes | -0.385 |
ery | -0.382 |
adoes | -0.375 |
WI | -0.371 |
suscept | -0.368 |
Token | had |
Token ID | 550 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Surgery | +0.303 |
Broadcasting | +0.298 |
scope | +0.288 |
hra | +0.282 |
aviour | +0.275 |
Neg logit contributions | |
ishes | -0.285 |
ery | -0.279 |
adoes | -0.272 |
WI | -0.267 |
suscept | -0.265 |
Token | written |
Token ID | 3194 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
Surgery | +0.878 |
Broadcasting | +0.862 |
scope | +0.832 |
hra | +0.816 |
aviour | +0.790 |
Neg logit contributions | |
ishes | -0.869 |
ery | -0.851 |
adoes | -0.827 |
WI | -0.815 |
suscept | -0.808 |
Token | a |
Token ID | 257 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
ishes | +1.243 |
ery | +1.232 |
adoes | +1.197 |
WI | +1.176 |
suscept | +1.157 |
Neg logit contributions | |
Surgery | -1.323 |
Broadcasting | -1.307 |
scope | -1.266 |
hra | -1.247 |
aviour | -1.222 |
Token | guide |
Token ID | 5698 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ishes | +0.034 |
ery | +0.034 |
adoes | +0.033 |
WI | +0.032 |
suscept | +0.032 |
Neg logit contributions | |
Surgery | -0.034 |
Broadcasting | -0.033 |
scope | -0.032 |
hra | -0.031 |
aviour | -0.030 |
Token | book |
Token ID | 2070 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
ishes | +1.237 |
ery | +1.232 |
adoes | +1.184 |
WI | +1.170 |
suscept | +1.145 |
Neg logit contributions | |
Surgery | -1.283 |
Broadcasting | -1.264 |
scope | -1.217 |
hra | -1.197 |
aviour | -1.167 |
Token | about |
Token ID | 546 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ishes | +1.388 |
ery | +1.377 |
adoes | +1.323 |
WI | +1.305 |
suscept | +1.283 |
Neg logit contributions | |
Surgery | -1.658 |
Broadcasting | -1.637 |
scope | -1.587 |
hra | -1.565 |
aviour | -1.526 |
Token | public |
Token ID | 1171 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
ishes | +1.288 |
ery | +1.276 |
adoes | +1.237 |
WI | +1.219 |
suscept | +1.203 |
Neg logit contributions | |
Surgery | -1.232 |
Broadcasting | -1.211 |
scope | -1.170 |
hra | -1.148 |
aviour | -1.120 |
Token | stair |
Token ID | 18761 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
Surgery | +2.027 |
Broadcasting | +2.004 |
scope | +1.928 |
hra | +1.901 |
aviour | +1.853 |
Neg logit contributions | |
ishes | -1.675 |
ery | -1.641 |
adoes | -1.606 |
WI | -1.574 |
suscept | -1.558 |
Token | ways |
Token ID | 1322 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
ishes | +1.804 |
ery | +1.786 |
adoes | +1.721 |
WI | +1.702 |
suscept | +1.676 |
Neg logit contributions | |
Surgery | -1.768 |
Broadcasting | -1.741 |
scope | -1.672 |
hra | -1.646 |
aviour | -1.606 |
Token | in |
Token ID | 287 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
ishes | +1.532 |
ery | +1.523 |
adoes | +1.469 |
WI | +1.452 |
suscept | +1.428 |
Neg logit contributions | |
Surgery | -1.640 |
Broadcasting | -1.617 |
scope | -1.567 |
hra | -1.537 |
aviour | -1.502 |
Token | s |
Token ID | 82 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
ishes | +1.153 |
ery | +1.141 |
adoes | +1.092 |
WI | +1.086 |
iors | +1.062 |
Neg logit contributions | |
Surgery | -1.151 |
Broadcasting | -1.130 |
scope | -1.065 |
hra | -1.054 |
aviour | -1.046 |
Token | 256 |
Token ID | 17759 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
Surgery | +0.608 |
Broadcasting | +0.598 |
scope | +0.580 |
hra | +0.570 |
aviour | +0.556 |
Neg logit contributions | |
ishes | -0.525 |
ery | -0.518 |
adoes | -0.502 |
WI | -0.495 |
suscept | -0.487 |
Token | GB |
Token ID | 13124 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ishes | +2.140 |
ery | +2.126 |
WI | +2.049 |
adoes | +2.040 |
iors | +1.965 |
Neg logit contributions | |
Surgery | -1.865 |
Broadcasting | -1.798 |
scope | -1.784 |
hra | -1.743 |
aviour | -1.734 |
Token | / |
Token ID | 14 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
ery | +0.706 |
ishes | +0.705 |
WI | +0.662 |
adoes | +0.661 |
suscept | +0.630 |
Neg logit contributions | |
Surgery | -1.164 |
Broadcasting | -1.143 |
scope | -1.115 |
hra | -1.100 |
aviour | -1.092 |
Token | s |
Token ID | 82 |
Feature activation | +5.40e-01 |
Pos logit contributions | |
ishes | +1.173 |
ery | +1.161 |
adoes | +1.111 |
WI | +1.107 |
iors | +1.081 |
Neg logit contributions | |
Surgery | -1.178 |
Broadcasting | -1.157 |
scope | -1.088 |
hra | -1.078 |
aviour | -1.070 |
Token | 320 |
Token ID | 20959 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
Surgery | +0.419 |
Broadcasting | +0.410 |
scope | +0.399 |
hra | +0.392 |
aviour | +0.383 |
Neg logit contributions | |
ishes | -0.412 |
ery | -0.408 |
adoes | -0.394 |
WI | -0.390 |
suscept | -0.383 |
Token | GB |
Token ID | 13124 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ishes | +2.781 |
ery | +2.771 |
WI | +2.675 |
adoes | +2.667 |
suscept | +2.573 |
Neg logit contributions | |
Surgery | -2.365 |
Broadcasting | -2.293 |
scope | -2.275 |
hra | -2.217 |
aviour | -2.171 |
Token | / |
Token ID | 14 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ery | +0.803 |
ishes | +0.802 |
WI | +0.752 |
adoes | +0.751 |
suscept | +0.724 |
Neg logit contributions | |
Surgery | -1.316 |
Broadcasting | -1.292 |
scope | -1.270 |
hra | -1.250 |
aviour | -1.239 |
Token | s |
Token ID | 82 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
ishes | +1.285 |
ery | +1.269 |
adoes | +1.216 |
WI | +1.212 |
iors | +1.182 |
Neg logit contributions | |
Surgery | -1.282 |
Broadcasting | -1.259 |
scope | -1.189 |
hra | -1.175 |
aviour | -1.165 |
Token | Bus |
Token ID | 5869 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
ishes | +0.012 |
ery | +0.012 |
adoes | +0.011 |
WI | +0.011 |
suscept | +0.011 |
Neg logit contributions | |
Surgery | -0.013 |
Broadcasting | -0.013 |
scope | -0.012 |
hra | -0.012 |
aviour | -0.012 |
Token | Width |
Token ID | 38807 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
Surgery | +1.730 |
Broadcasting | +1.698 |
scope | +1.611 |
hra | +1.555 |
aviour | +1.522 |
Neg logit contributions | |
ishes | -2.802 |
ery | -2.769 |
adoes | -2.718 |
WI | -2.682 |
suscept | -2.633 |
Token | smart |
Token ID | 4451 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ishes | +0.313 |
ery | +0.308 |
adoes | +0.299 |
WI | +0.294 |
suscept | +0.290 |
Neg logit contributions | |
Surgery | -0.362 |
Broadcasting | -0.357 |
scope | -0.345 |
hra | -0.339 |
aviour | -0.331 |
Token | phone |
Token ID | 3072 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
Surgery | +1.162 |
Broadcasting | +1.148 |
scope | +1.109 |
hra | +1.080 |
aviour | +1.038 |
Neg logit contributions | |
ishes | -1.443 |
ery | -1.419 |
adoes | -1.396 |
WI | -1.363 |
suscept | -1.358 |
Token | in |
Token ID | 287 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
Surgery | +0.623 |
Broadcasting | +0.612 |
scope | +0.591 |
hra | +0.580 |
aviour | +0.565 |
Neg logit contributions | |
ishes | -0.627 |
ery | -0.619 |
adoes | -0.601 |
WI | -0.593 |
suscept | -0.585 |
Token | downtown |
Token ID | 9436 |
Feature activation | -4.88e-03 |
Pos logit contributions | |
Surgery | +0.177 |
Broadcasting | +0.174 |
scope | +0.166 |
hra | +0.162 |
aviour | +0.157 |
Neg logit contributions | |
ishes | -0.252 |
ery | -0.249 |
adoes | -0.243 |
WI | -0.240 |
suscept | -0.238 |
Token | Toronto |
Token ID | 6586 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
ishes | +0.004 |
ery | +0.004 |
adoes | +0.004 |
WI | +0.004 |
suscept | +0.004 |
Neg logit contributions | |
Surgery | -0.003 |
Broadcasting | -0.003 |
scope | -0.003 |
hra | -0.003 |
aviour | -0.003 |
Token | on |
Token ID | 319 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
ishes | +0.103 |
ery | +0.102 |
adoes | +0.099 |
WI | +0.098 |
suscept | +0.096 |
Neg logit contributions | |
Surgery | -0.104 |
Broadcasting | -0.102 |
scope | -0.099 |
hra | -0.097 |
aviour | -0.095 |
Token | Monday |
Token ID | 3321 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
ishes | +1.356 |
ery | +1.334 |
adoes | +1.296 |
WI | +1.277 |
suscept | +1.262 |
Neg logit contributions | |
Surgery | -1.361 |
Broadcasting | -1.338 |
scope | -1.298 |
hra | -1.280 |
aviour | -1.250 |
Token | . |
Token ID | 13 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
Surgery | +1.141 |
Broadcasting | +1.123 |
scope | +1.082 |
hra | +1.061 |
aviour | +1.034 |
Neg logit contributions | |
ishes | -1.116 |
ery | -1.098 |
adoes | -1.070 |
WI | -1.051 |
suscept | -1.041 |
Token | Liberal |
Token ID | 12266 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
ishes | +0.118 |
ery | +0.117 |
adoes | +0.113 |
WI | +0.112 |
suscept | +0.110 |
Neg logit contributions | |
Surgery | -0.097 |
Broadcasting | -0.095 |
scope | -0.091 |
hra | -0.090 |
aviour | -0.087 |
Token | MP |
Token ID | 4904 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
Surgery | +1.110 |
Broadcasting | +1.094 |
scope | +1.035 |
hra | +1.011 |
aviour | +0.974 |
Neg logit contributions | |
ishes | -1.684 |
ery | -1.656 |
adoes | -1.626 |
WI | -1.605 |
suscept | -1.591 |
Token | P |
Token ID | 47 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
Surgery | +1.561 |
Broadcasting | +1.531 |
scope | +1.461 |
hra | +1.432 |
aviour | +1.385 |
Neg logit contributions | |
ishes | -2.224 |
ery | -2.195 |
adoes | -2.151 |
WI | -2.121 |
suscept | -2.101 |
Token | that |
Token ID | 326 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
ishes | +1.186 |
ery | +1.175 |
adoes | +1.132 |
WI | +1.121 |
suscept | +1.105 |
Neg logit contributions | |
Surgery | -1.260 |
Broadcasting | -1.237 |
scope | -1.195 |
hra | -1.178 |
aviour | -1.146 |
Token | further |
Token ID | 2252 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
Surgery | +0.654 |
Broadcasting | +0.644 |
scope | +0.622 |
hra | +0.614 |
aviour | +0.593 |
Neg logit contributions | |
ishes | -0.595 |
ery | -0.588 |
adoes | -0.569 |
WI | -0.563 |
suscept | -0.557 |
Token | connects |
Token ID | 20417 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
ishes | +1.557 |
ery | +1.522 |
adoes | +1.473 |
WI | +1.449 |
suscept | +1.417 |
Neg logit contributions | |
Surgery | -2.214 |
Broadcasting | -2.181 |
scope | -2.112 |
hra | -2.085 |
aviour | -2.041 |
Token | the |
Token ID | 262 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
ishes | +0.056 |
ery | +0.055 |
adoes | +0.053 |
WI | +0.053 |
suscept | +0.052 |
Neg logit contributions | |
Surgery | -0.048 |
Broadcasting | -0.048 |
scope | -0.046 |
hra | -0.045 |
aviour | -0.043 |
Token | settlements |
Token ID | 18573 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
ishes | +0.710 |
ery | +0.699 |
adoes | +0.678 |
WI | +0.672 |
suscept | +0.662 |
Neg logit contributions | |
Surgery | -0.716 |
Broadcasting | -0.703 |
scope | -0.681 |
hra | -0.669 |
aviour | -0.653 |
Token | south |
Token ID | 5366 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
ishes | +0.237 |
ery | +0.234 |
adoes | +0.227 |
WI | +0.224 |
suscept | +0.221 |
Neg logit contributions | |
Surgery | -0.243 |
Broadcasting | -0.239 |
scope | -0.231 |
hra | -0.226 |
aviour | -0.221 |
Token | of |
Token ID | 286 |
Feature activation | +5.28e-03 |
Pos logit contributions | |
ishes | +3.338 |
ery | +3.295 |
adoes | +3.173 |
WI | +3.170 |
iors | +3.083 |
Neg logit contributions | |
Surgery | -3.544 |
Broadcasting | -3.479 |
scope | -3.364 |
hra | -3.288 |
aviour | -3.233 |
Token | Jerusalem |
Token ID | 10843 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
Surgery | +0.004 |
Broadcasting | +0.004 |
scope | +0.003 |
hra | +0.003 |
aviour | +0.003 |
Neg logit contributions | |
ishes | -0.005 |
ery | -0.004 |
adoes | -0.004 |
WI | -0.004 |
suscept | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -6.89e-02 |
Pos logit contributions | |
Surgery | +1.165 |
Broadcasting | +1.151 |
scope | +1.103 |
hra | +1.087 |
aviour | +1.054 |
Neg logit contributions | |
ishes | -1.204 |
ery | -1.187 |
adoes | -1.160 |
WI | -1.136 |
suscept | -1.130 |
Token | Heb |
Token ID | 27030 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
ishes | +0.058 |
ery | +0.057 |
adoes | +0.056 |
WI | +0.055 |
suscept | +0.054 |
Neg logit contributions | |
Surgery | -0.048 |
Broadcasting | -0.048 |
scope | -0.046 |
hra | -0.045 |
aviour | -0.043 |
Token | ron |
Token ID | 1313 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
ishes | +1.936 |
ery | +1.891 |
adoes | +1.820 |
iors | +1.797 |
WI | +1.791 |
Neg logit contributions | |
Surgery | -2.041 |
Broadcasting | -2.008 |
scope | -1.947 |
hra | -1.900 |
aviour | -1.835 |
Token | weapons |
Token ID | 3777 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
ishes | +0.947 |
ery | +0.935 |
adoes | +0.898 |
WI | +0.882 |
suscept | +0.868 |
Neg logit contributions | |
Surgery | -1.505 |
Broadcasting | -1.484 |
scope | -1.441 |
hra | -1.425 |
aviour | -1.396 |
Token | and |
Token ID | 290 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
ishes | +1.944 |
ery | +1.933 |
adoes | +1.867 |
WI | +1.827 |
suscept | +1.803 |
Neg logit contributions | |
Surgery | -2.169 |
Broadcasting | -2.141 |
scope | -2.069 |
hra | -2.034 |
aviour | -1.995 |
Token | high |
Token ID | 1029 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
ishes | +1.533 |
ery | +1.526 |
adoes | +1.477 |
WI | +1.458 |
suscept | +1.444 |
Neg logit contributions | |
Surgery | -1.357 |
Broadcasting | -1.328 |
scope | -1.281 |
hra | -1.258 |
aviour | -1.231 |
Token | - |
Token ID | 12 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
ishes | +2.239 |
ery | +2.222 |
adoes | +2.137 |
WI | +2.124 |
suscept | +2.088 |
Neg logit contributions | |
Surgery | -2.650 |
Broadcasting | -2.619 |
scope | -2.518 |
hra | -2.477 |
aviour | -2.439 |
Token | capacity |
Token ID | 42404 |
Feature activation | -6.64e+00 |
Pos logit contributions | |
ishes | +1.471 |
ery | +1.453 |
adoes | +1.401 |
WI | +1.387 |
suscept | +1.357 |
Neg logit contributions | |
Surgery | -1.792 |
Broadcasting | -1.766 |
scope | -1.700 |
hra | -1.675 |
aviour | -1.641 |
Token | magazines |
Token ID | 16695 |
Feature activation | -7.94e+00 |
Pos logit contributions | |
ishes | +2.894 |
ery | +2.884 |
adoes | +2.831 |
WI | +2.690 |
suscept | +2.647 |
Neg logit contributions | |
Surgery | -7.180 |
Broadcasting | -7.017 |
hra | -6.905 |
scope | -6.900 |
aviour | -6.788 |
Token | . |
Token ID | 13 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ery | +5.253 |
ishes | +5.169 |
adoes | +4.994 |
WI | +4.890 |
iors | +4.808 |
Neg logit contributions | |
Surgery | -6.508 |
Broadcasting | -6.493 |
scope | -6.323 |
hra | -6.189 |
aviour | -6.107 |
Token |
|
Token ID | 198 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ishes | +1.312 |
ery | +1.305 |
adoes | +1.262 |
WI | +1.260 |
iors | +1.227 |
Neg logit contributions | |
Surgery | -0.975 |
Broadcasting | -0.958 |
scope | -0.934 |
hra | -0.916 |
aviour | -0.906 |
Token |
|
Token ID | 198 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
ery | +1.084 |
ishes | +1.082 |
WI | +1.039 |
adoes | +1.030 |
iors | +1.011 |
Neg logit contributions | |
Surgery | -0.857 |
Broadcasting | -0.838 |
scope | -0.797 |
hra | -0.784 |
aviour | -0.776 |
Token | But |
Token ID | 1537 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
ishes | +1.197 |
ery | +1.189 |
WI | +1.145 |
adoes | +1.143 |
iors | +1.108 |
Neg logit contributions | |
Surgery | -1.144 |
Broadcasting | -1.123 |
scope | -1.070 |
hra | -1.058 |
aviour | -1.033 |
Token | he |
Token ID | 339 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
Surgery | +1.868 |
Broadcasting | +1.837 |
scope | +1.765 |
hra | +1.742 |
aviour | +1.669 |
Neg logit contributions | |
ishes | -2.087 |
ery | -2.056 |
adoes | -2.003 |
WI | -1.984 |
suscept | -1.954 |
Token | sound |
Token ID | 2128 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ishes | +0.261 |
ery | +0.258 |
adoes | +0.249 |
WI | +0.246 |
suscept | +0.244 |
Neg logit contributions | |
Surgery | -0.264 |
Broadcasting | -0.260 |
scope | -0.250 |
hra | -0.246 |
aviour | -0.240 |
Token | education |
Token ID | 3707 |
Feature activation | +8.67e-02 |
Pos logit contributions | |
Surgery | +1.189 |
Broadcasting | +1.171 |
scope | +1.132 |
hra | +1.110 |
aviour | +1.085 |
Neg logit contributions | |
ishes | -1.112 |
ery | -1.094 |
adoes | -1.066 |
WI | -1.048 |
suscept | -1.033 |
Token | ." |
Token ID | 526 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
Surgery | +0.064 |
Broadcasting | +0.063 |
scope | +0.060 |
hra | +0.059 |
aviour | +0.058 |
Neg logit contributions | |
ishes | -0.070 |
ery | -0.069 |
adoes | -0.067 |
WI | -0.066 |
suscept | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
ishes | +0.316 |
ery | +0.313 |
adoes | +0.304 |
WI | +0.301 |
suscept | +0.297 |
Neg logit contributions | |
Surgery | -0.224 |
Broadcasting | -0.219 |
scope | -0.210 |
hra | -0.205 |
aviour | -0.199 |
Token |
|
Token ID | 198 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
ishes | +0.434 |
ery | +0.433 |
WI | +0.415 |
adoes | +0.415 |
iors | +0.405 |
Neg logit contributions | |
Surgery | -0.335 |
Broadcasting | -0.327 |
scope | -0.311 |
hra | -0.306 |
aviour | -0.302 |
Token | Mis |
Token ID | 31281 |
Feature activation | -8.26e+00 |
Pos logit contributions | |
ishes | +0.545 |
ery | +0.540 |
adoes | +0.521 |
WI | +0.519 |
suscept | +0.506 |
Neg logit contributions | |
Surgery | -0.495 |
Broadcasting | -0.486 |
scope | -0.465 |
hra | -0.457 |
aviour | -0.446 |
Token | guided |
Token ID | 23657 |
Feature activation | +2.61e-03 |
Pos logit contributions | |
ery | +6.519 |
ishes | +6.281 |
adoes | +5.965 |
WI | +5.890 |
iors | +5.836 |
Neg logit contributions | |
Surgery | -6.260 |
Broadcasting | -6.117 |
scope | -5.789 |
hra | -5.648 |
ndum | -5.538 |
Token | political |
Token ID | 1964 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
Surgery | +0.002 |
Broadcasting | +0.002 |
scope | +0.002 |
hra | +0.002 |
aviour | +0.002 |
Neg logit contributions | |
ishes | -0.002 |
ery | -0.002 |
adoes | -0.002 |
WI | -0.002 |
suscept | -0.002 |
Token | effort |
Token ID | 3626 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
Surgery | +1.773 |
Broadcasting | +1.745 |
scope | +1.685 |
hra | +1.644 |
aviour | +1.616 |
Neg logit contributions | |
ishes | -1.550 |
ery | -1.522 |
adoes | -1.494 |
WI | -1.477 |
iors | -1.429 |
Token |
|
Token ID | 198 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ishes | +1.789 |
ery | +1.768 |
adoes | +1.723 |
WI | +1.704 |
suscept | +1.683 |
Neg logit contributions | |
Surgery | -1.291 |
Broadcasting | -1.265 |
scope | -1.212 |
hra | -1.186 |
aviour | -1.149 |
Token |
|
Token ID | 198 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ery | +1.167 |
ishes | +1.165 |
WI | +1.118 |
adoes | +1.109 |
iors | +1.090 |
Neg logit contributions | |
Surgery | -0.959 |
Broadcasting | -0.936 |
scope | -0.889 |
hra | -0.877 |
aviour | -0.875 |
Token | are |
Token ID | 389 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ishes | +6.094 |
ery | +5.855 |
adoes | +5.500 |
WI | +5.385 |
suscept | +5.353 |
Neg logit contributions | |
Surgery | -7.081 |
Broadcasting | -6.974 |
scope | -6.696 |
hra | -6.598 |
ndum | -6.509 |
Token | labeled |
Token ID | 15494 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ishes | +0.797 |
ery | +0.784 |
adoes | +0.763 |
WI | +0.752 |
suscept | +0.738 |
Neg logit contributions | |
Surgery | -0.882 |
Broadcasting | -0.867 |
scope | -0.839 |
hra | -0.824 |
aviour | -0.803 |
Token | as |
Token ID | 355 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
ishes | +0.880 |
ery | +0.868 |
adoes | +0.844 |
WI | +0.833 |
suscept | +0.821 |
Neg logit contributions | |
Surgery | -0.848 |
Broadcasting | -0.833 |
scope | -0.804 |
hra | -0.789 |
aviour | -0.768 |
Token | plant |
Token ID | 4618 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Surgery | +0.951 |
Broadcasting | +0.930 |
scope | +0.899 |
hra | +0.886 |
aviour | +0.856 |
Neg logit contributions | |
ishes | -0.894 |
ery | -0.879 |
adoes | -0.862 |
WI | -0.848 |
suscept | -0.830 |
Token | ain |
Token ID | 391 |
Feature activation | -6.86e-01 |
Pos logit contributions | |
Surgery | +1.172 |
Broadcasting | +1.161 |
scope | +1.120 |
hra | +1.114 |
aviour | +1.094 |
Neg logit contributions | |
ishes | -1.215 |
ery | -1.190 |
adoes | -1.178 |
WI | -1.177 |
suscept | -1.152 |
Token | flour |
Token ID | 10601 |
Feature activation | -7.95e+00 |
Pos logit contributions | |
ishes | +0.529 |
ery | +0.522 |
adoes | +0.507 |
WI | +0.500 |
suscept | +0.493 |
Neg logit contributions | |
Surgery | -0.530 |
Broadcasting | -0.521 |
scope | -0.503 |
hra | -0.494 |
aviour | -0.481 |
Token | but |
Token ID | 475 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
ishes | +5.792 |
ery | +5.640 |
adoes | +5.348 |
WI | +5.326 |
suscept | +5.202 |
Neg logit contributions | |
Surgery | -6.311 |
Broadcasting | -6.238 |
scope | -5.971 |
hra | -5.903 |
aviour | -5.844 |
Token | actually |
Token ID | 1682 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
Surgery | +2.462 |
Broadcasting | +2.411 |
scope | +2.327 |
hra | +2.287 |
aviour | +2.180 |
Neg logit contributions | |
ishes | -2.962 |
ery | -2.921 |
adoes | -2.875 |
WI | -2.825 |
suscept | -2.777 |
Token | contain |
Token ID | 3994 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
Surgery | +0.111 |
Broadcasting | +0.109 |
scope | +0.105 |
hra | +0.103 |
aviour | +0.100 |
Neg logit contributions | |
ishes | -0.118 |
ery | -0.116 |
adoes | -0.113 |
WI | -0.112 |
suscept | -0.110 |
Token | a |
Token ID | 257 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
Surgery | +0.469 |
Broadcasting | +0.461 |
scope | +0.445 |
hra | +0.437 |
aviour | +0.425 |
Neg logit contributions | |
ishes | -0.482 |
ery | -0.474 |
adoes | -0.463 |
WI | -0.456 |
suscept | -0.449 |
Token | mix |
Token ID | 5022 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
ishes | +0.151 |
ery | +0.149 |
adoes | +0.145 |
WI | +0.143 |
suscept | +0.140 |
Neg logit contributions | |
Surgery | -0.169 |
Broadcasting | -0.166 |
scope | -0.160 |
hra | -0.158 |
aviour | -0.154 |
Token | used |
Token ID | 973 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
ishes | +0.715 |
ery | +0.702 |
adoes | +0.675 |
WI | +0.662 |
suscept | +0.649 |
Neg logit contributions | |
Surgery | -1.217 |
Broadcasting | -1.200 |
scope | -1.167 |
hra | -1.150 |
aviour | -1.126 |
Token | the |
Token ID | 262 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ishes | +1.322 |
ery | +1.307 |
adoes | +1.264 |
WI | +1.247 |
suscept | +1.228 |
Neg logit contributions | |
Surgery | -1.273 |
Broadcasting | -1.253 |
scope | -1.209 |
hra | -1.186 |
aviour | -1.158 |
Token | same |
Token ID | 976 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
ishes | +0.142 |
ery | +0.140 |
adoes | +0.136 |
WI | +0.134 |
suscept | +0.132 |
Neg logit contributions | |
Surgery | -0.150 |
Broadcasting | -0.147 |
scope | -0.142 |
hra | -0.140 |
aviour | -0.136 |
Token | as |
Token ID | 355 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
Surgery | +1.183 |
Broadcasting | +1.165 |
scope | +1.129 |
hra | +1.104 |
aviour | +1.071 |
Neg logit contributions | |
ishes | -1.225 |
ery | -1.207 |
adoes | -1.184 |
WI | -1.175 |
suscept | -1.151 |
Token | almond |
Token ID | 26948 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
Surgery | +2.334 |
Broadcasting | +2.291 |
scope | +2.233 |
hra | +2.205 |
aviour | +2.125 |
Neg logit contributions | |
ishes | -1.820 |
ery | -1.762 |
adoes | -1.741 |
WI | -1.712 |
iors | -1.682 |
Token | flour |
Token ID | 10601 |
Feature activation | -7.70e+00 |
Pos logit contributions | |
ishes | +0.106 |
ery | +0.103 |
adoes | +0.101 |
WI | +0.099 |
suscept | +0.097 |
Neg logit contributions | |
Surgery | -0.188 |
Broadcasting | -0.185 |
scope | -0.181 |
hra | -0.178 |
aviour | -0.174 |
Token | , |
Token ID | 11 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
ishes | +5.460 |
ery | +5.385 |
adoes | +5.051 |
WI | +5.013 |
suscept | +4.850 |
Neg logit contributions | |
Surgery | -6.220 |
Broadcasting | -6.104 |
scope | -5.943 |
hra | -5.858 |
aviour | -5.760 |
Token | but |
Token ID | 475 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
ishes | +0.033 |
ery | +0.033 |
adoes | +0.032 |
WI | +0.031 |
suscept | +0.031 |
Neg logit contributions | |
Surgery | -0.029 |
Broadcasting | -0.028 |
scope | -0.027 |
hra | -0.027 |
aviour | -0.026 |
Token | yields |
Token ID | 19299 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
Surgery | +2.811 |
Broadcasting | +2.767 |
scope | +2.668 |
hra | +2.647 |
rop | +2.501 |
Neg logit contributions | |
ishes | -3.412 |
ery | -3.362 |
adoes | -3.329 |
WI | -3.270 |
iors | -3.245 |
Token | a |
Token ID | 257 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
Surgery | +0.025 |
Broadcasting | +0.024 |
scope | +0.023 |
hra | +0.023 |
aviour | +0.022 |
Neg logit contributions | |
ishes | -0.028 |
ery | -0.028 |
adoes | -0.027 |
WI | -0.027 |
suscept | -0.026 |
Token | different |
Token ID | 1180 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
ishes | +0.149 |
ery | +0.146 |
adoes | +0.142 |
WI | +0.140 |
suscept | +0.138 |
Neg logit contributions | |
Surgery | -0.173 |
Broadcasting | -0.170 |
scope | -0.165 |
hra | -0.162 |
aviour | -0.158 |
Token | almonds |
Token ID | 48793 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ishes | +1.279 |
ery | +1.266 |
adoes | +1.228 |
WI | +1.207 |
suscept | +1.201 |
Neg logit contributions | |
Surgery | -1.320 |
Broadcasting | -1.300 |
scope | -1.255 |
hra | -1.230 |
aviour | -1.210 |
Token | , |
Token ID | 11 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
ishes | +0.933 |
ery | +0.921 |
adoes | +0.890 |
WI | +0.879 |
suscept | +0.866 |
Neg logit contributions | |
Surgery | -1.011 |
Broadcasting | -0.995 |
scope | -0.962 |
hra | -0.945 |
aviour | -0.923 |
Token | sun |
Token ID | 4252 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
ishes | +0.038 |
ery | +0.038 |
adoes | +0.037 |
WI | +0.036 |
suscept | +0.036 |
Neg logit contributions | |
Surgery | -0.033 |
Broadcasting | -0.033 |
scope | -0.032 |
hra | -0.031 |
aviour | -0.030 |
Token | flower |
Token ID | 25547 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Surgery | +0.379 |
Broadcasting | +0.374 |
scope | +0.364 |
hra | +0.358 |
aviour | +0.350 |
Neg logit contributions | |
ishes | -0.262 |
ery | -0.258 |
adoes | -0.251 |
WI | -0.246 |
suscept | -0.244 |
Token | seed |
Token ID | 9403 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
ishes | +0.887 |
ery | +0.874 |
adoes | +0.852 |
WI | +0.842 |
suscept | +0.831 |
Neg logit contributions | |
Surgery | -0.855 |
Broadcasting | -0.840 |
scope | -0.811 |
hra | -0.795 |
aviour | -0.775 |
Token | flour |
Token ID | 10601 |
Feature activation | -8.19e+00 |
Pos logit contributions | |
ishes | +1.354 |
ery | +1.328 |
adoes | +1.269 |
WI | +1.244 |
suscept | +1.216 |
Neg logit contributions | |
Surgery | -2.600 |
Broadcasting | -2.567 |
scope | -2.500 |
hra | -2.465 |
aviour | -2.419 |
Token | can |
Token ID | 460 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
ishes | +5.476 |
ery | +5.300 |
adoes | +4.999 |
WI | +4.872 |
suscept | +4.722 |
Neg logit contributions | |
Surgery | -7.108 |
Broadcasting | -7.004 |
scope | -6.770 |
hra | -6.688 |
aviour | -6.589 |
Token | be |
Token ID | 307 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ishes | +0.531 |
ery | +0.524 |
adoes | +0.510 |
WI | +0.504 |
suscept | +0.497 |
Neg logit contributions | |
Surgery | -0.436 |
Broadcasting | -0.427 |
scope | -0.411 |
hra | -0.402 |
aviour | -0.391 |
Token | used |
Token ID | 973 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ishes | +1.053 |
ery | +1.038 |
adoes | +1.006 |
WI | +0.992 |
suscept | +0.978 |
Neg logit contributions | |
Surgery | -1.172 |
Broadcasting | -1.153 |
scope | -1.115 |
hra | -1.096 |
aviour | -1.070 |
Token | the |
Token ID | 262 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
ishes | +1.174 |
ery | +1.163 |
adoes | +1.122 |
WI | +1.107 |
suscept | +1.088 |
Neg logit contributions | |
Surgery | -1.184 |
Broadcasting | -1.166 |
scope | -1.124 |
hra | -1.103 |
aviour | -1.079 |
Token | same |
Token ID | 976 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
ishes | +0.194 |
ery | +0.191 |
adoes | +0.185 |
WI | +0.183 |
suscept | +0.180 |
Neg logit contributions | |
Surgery | -0.202 |
Broadcasting | -0.199 |
scope | -0.192 |
hra | -0.188 |
aviour | -0.184 |