Token | . |
Token ID | 13 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
uracy | +0.089 |
opin | +0.085 |
ription | +0.084 |
romy | +0.077 |
psons | +0.075 |
Neg logit contributions | |
Pixel | -0.093 |
oise | -0.088 |
archy | -0.082 |
Maher | -0.081 |
ould | -0.080 |
Token |
|
Token ID | 198 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
uracy | +0.420 |
opin | +0.403 |
ription | +0.401 |
romy | +0.368 |
psons | +0.358 |
Neg logit contributions | |
Pixel | -0.416 |
oise | -0.392 |
archy | -0.366 |
ould | -0.359 |
Maher | -0.355 |
Token |
|
Token ID | 198 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
uracy | +0.761 |
opin | +0.721 |
ription | +0.714 |
romy | +0.658 |
psons | +0.636 |
Neg logit contributions | |
Pixel | -0.806 |
oise | -0.760 |
archy | -0.715 |
ould | -0.702 |
Maher | -0.693 |
Token | " |
Token ID | 1 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
uracy | +0.511 |
opin | +0.492 |
ription | +0.488 |
romy | +0.447 |
psons | +0.437 |
Neg logit contributions | |
Pixel | -0.498 |
oise | -0.467 |
archy | -0.436 |
ould | -0.427 |
Maher | -0.423 |
Token | En |
Token ID | 4834 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
Pixel | +0.269 |
oise | +0.250 |
archy | +0.231 |
ould | +0.227 |
Maher | +0.224 |
Neg logit contributions | |
uracy | -0.293 |
opin | -0.284 |
ription | -0.282 |
romy | -0.258 |
psons | -0.254 |
Token | vision |
Token ID | 10178 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
uracy | +0.789 |
opin | +0.758 |
ription | +0.750 |
romy | +0.693 |
psons | +0.677 |
Neg logit contributions | |
Pixel | -0.703 |
oise | -0.663 |
archy | -0.617 |
ould | -0.604 |
Maher | -0.597 |
Token | ing |
Token ID | 278 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
uracy | +4.630 |
opin | +4.461 |
ription | +4.440 |
romy | +4.049 |
inval | +3.996 |
Neg logit contributions | |
Pixel | -4.744 |
oise | -4.482 |
archy | -4.183 |
ould | -4.078 |
Maher | -4.077 |
Token | a |
Token ID | 257 |
Feature activation | -9.09e-02 |
Pos logit contributions | |
uracy | +0.088 |
opin | +0.084 |
ription | +0.084 |
romy | +0.078 |
psons | +0.076 |
Neg logit contributions | |
Pixel | -0.075 |
oise | -0.071 |
archy | -0.065 |
ould | -0.064 |
Maher | -0.064 |
Token | world |
Token ID | 995 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
Pixel | +0.192 |
oise | +0.180 |
archy | +0.166 |
ould | +0.162 |
Maher | +0.161 |
Neg logit contributions | |
uracy | -0.245 |
opin | -0.236 |
ription | -0.234 |
romy | -0.217 |
psons | -0.212 |
Token | where |
Token ID | 810 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
Pixel | +0.084 |
oise | +0.079 |
archy | +0.073 |
ould | +0.072 |
Maher | +0.071 |
Neg logit contributions | |
uracy | -0.092 |
ription | -0.088 |
opin | -0.088 |
romy | -0.081 |
psons | -0.079 |
Token | cigarettes |
Token ID | 17626 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
Pixel | +0.401 |
oise | +0.377 |
archy | +0.351 |
ould | +0.344 |
Maher | +0.343 |
Neg logit contributions | |
uracy | -0.425 |
opin | -0.407 |
ription | -0.405 |
romy | -0.373 |
psons | -0.364 |
Token | in |
Token ID | 287 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
uracy | +0.602 |
opin | +0.578 |
ription | +0.574 |
romy | +0.529 |
psons | +0.516 |
Neg logit contributions | |
Pixel | -0.558 |
oise | -0.524 |
archy | -0.488 |
ould | -0.479 |
Maher | -0.476 |
Token | either |
Token ID | 2035 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
uracy | +0.456 |
opin | +0.439 |
ription | +0.436 |
romy | +0.403 |
psons | +0.393 |
Neg logit contributions | |
Pixel | -0.391 |
oise | -0.366 |
archy | -0.340 |
ould | -0.332 |
Maher | -0.329 |
Token | category |
Token ID | 6536 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
Pixel | +0.257 |
oise | +0.241 |
archy | +0.223 |
Maher | +0.217 |
ould | +0.216 |
Neg logit contributions | |
uracy | -0.316 |
opin | -0.303 |
ription | -0.302 |
romy | -0.281 |
psons | -0.273 |
Token | - |
Token ID | 532 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
uracy | +1.886 |
opin | +1.812 |
ription | +1.798 |
romy | +1.645 |
inval | +1.605 |
Neg logit contributions | |
Pixel | -1.859 |
oise | -1.750 |
archy | -1.630 |
ould | -1.602 |
Maher | -1.586 |
Token | D |
Token ID | 360 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
uracy | +1.649 |
opin | +1.585 |
ription | +1.573 |
romy | +1.449 |
inval | +1.414 |
Neg logit contributions | |
Pixel | -1.498 |
oise | -1.407 |
archy | -1.306 |
ould | -1.280 |
Maher | -1.270 |
Token | essert |
Token ID | 20335 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
uracy | +0.773 |
opin | +0.731 |
ription | +0.722 |
romy | +0.640 |
psons | +0.618 |
Neg logit contributions | |
Pixel | -1.316 |
oise | -1.256 |
archy | -1.192 |
ould | -1.174 |
Maher | -1.166 |
Token | s |
Token ID | 82 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
uracy | +4.588 |
ription | +4.422 |
opin | +4.409 |
romy | +3.997 |
inval | +3.953 |
Neg logit contributions | |
Pixel | -4.395 |
oise | -4.158 |
archy | -3.844 |
ould | -3.808 |
Maher | -3.730 |
Token | or |
Token ID | 393 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
uracy | +0.834 |
opin | +0.801 |
ription | +0.796 |
romy | +0.732 |
inval | +0.713 |
Neg logit contributions | |
Pixel | -0.766 |
oise | -0.721 |
archy | -0.670 |
ould | -0.656 |
Maher | -0.651 |
Token | Veget |
Token ID | 23892 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
uracy | +0.997 |
opin | +0.960 |
ription | +0.953 |
romy | +0.885 |
psons | +0.866 |
Neg logit contributions | |
Pixel | -0.762 |
oise | -0.711 |
archy | -0.657 |
ould | -0.642 |
Maher | -0.637 |
Token | ables |
Token ID | 2977 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
Pixel | +0.384 |
oise | +0.363 |
archy | +0.339 |
ould | +0.329 |
Maher | +0.319 |
Neg logit contributions | |
uracy | -0.533 |
opin | -0.518 |
ription | -0.511 |
romy | -0.474 |
inval | -0.466 |
Token | . |
Token ID | 13 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
uracy | +1.257 |
opin | +1.207 |
ription | +1.197 |
romy | +1.097 |
inval | +1.068 |
Neg logit contributions | |
Pixel | -1.279 |
oise | -1.206 |
archy | -1.126 |
ould | -1.104 |
Maher | -1.095 |
Token | spirit |
Token ID | 4437 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Pixel | +0.603 |
oise | +0.574 |
archy | +0.531 |
ould | +0.519 |
Maher | +0.509 |
Neg logit contributions | |
uracy | -0.669 |
opin | -0.642 |
ription | -0.636 |
romy | -0.584 |
psons | -0.571 |
Token | as |
Token ID | 355 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
uracy | +1.151 |
opin | +1.105 |
ription | +1.097 |
romy | +1.010 |
psons | +0.985 |
Neg logit contributions | |
Pixel | -1.057 |
oise | -0.994 |
archy | -0.925 |
ould | -0.906 |
Maher | -0.898 |
Token | a |
Token ID | 257 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
Pixel | +0.159 |
oise | +0.149 |
archy | +0.139 |
ould | +0.137 |
Maher | +0.135 |
Neg logit contributions | |
uracy | -0.174 |
opin | -0.167 |
ription | -0.166 |
romy | -0.153 |
psons | -0.149 |
Token | strong |
Token ID | 1913 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
Pixel | +0.139 |
oise | +0.131 |
archy | +0.121 |
ould | +0.119 |
Maher | +0.118 |
Neg logit contributions | |
uracy | -0.166 |
opin | -0.159 |
ription | -0.159 |
romy | -0.146 |
psons | -0.143 |
Token | pub |
Token ID | 2240 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
uracy | +1.644 |
opin | +1.589 |
ription | +1.568 |
romy | +1.448 |
inval | +1.414 |
Neg logit contributions | |
Pixel | -1.474 |
oise | -1.380 |
archy | -1.286 |
ould | -1.256 |
Maher | -1.248 |
Token | hero |
Token ID | 4293 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
uracy | +3.438 |
opin | +3.341 |
ription | +3.275 |
romy | +3.038 |
inval | +2.970 |
Neg logit contributions | |
Pixel | -3.112 |
oise | -2.881 |
archy | -2.701 |
ould | -2.638 |
Maher | -2.637 |
Token | to |
Token ID | 284 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
uracy | +4.268 |
opin | +4.127 |
ription | +4.100 |
romy | +3.724 |
inval | +3.630 |
Neg logit contributions | |
Pixel | -4.622 |
oise | -4.338 |
archy | -4.058 |
ould | -3.981 |
Maher | -3.970 |
Token | showcase |
Token ID | 21742 |
Feature activation | -7.50e-02 |
Pos logit contributions | |
uracy | +0.342 |
opin | +0.327 |
ription | +0.326 |
romy | +0.301 |
psons | +0.296 |
Neg logit contributions | |
Pixel | -0.279 |
oise | -0.262 |
archy | -0.243 |
ould | -0.238 |
Maher | -0.235 |
Token | . |
Token ID | 13 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
Pixel | +0.172 |
oise | +0.162 |
archy | +0.150 |
ould | +0.147 |
Maher | +0.146 |
Neg logit contributions | |
uracy | -0.189 |
opin | -0.181 |
ription | -0.180 |
romy | -0.165 |
psons | -0.162 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
uracy | +0.122 |
opin | +0.117 |
ription | +0.116 |
romy | +0.107 |
psons | +0.104 |
Neg logit contributions | |
Pixel | -0.126 |
oise | -0.118 |
archy | -0.111 |
ould | -0.109 |
Maher | -0.108 |
Token |
|
Token ID | 198 |
Feature activation | +6.94e-02 |
Pos logit contributions | |
uracy | +0.305 |
opin | +0.292 |
ription | +0.289 |
romy | +0.266 |
psons | +0.258 |
Neg logit contributions | |
Pixel | -0.307 |
oise | -0.290 |
archy | -0.271 |
ould | -0.265 |
Maher | -0.263 |
Token | s |
Token ID | 82 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
uracy | +0.576 |
opin | +0.554 |
ription | +0.550 |
romy | +0.508 |
psons | +0.495 |
Neg logit contributions | |
Pixel | -0.509 |
oise | -0.478 |
archy | -0.445 |
ould | -0.435 |
Maher | -0.430 |
Token | traditions |
Token ID | 15421 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
Pixel | +0.091 |
oise | +0.084 |
archy | +0.078 |
ould | +0.076 |
Maher | +0.075 |
Neg logit contributions | |
uracy | -0.124 |
opin | -0.119 |
ription | -0.119 |
romy | -0.110 |
psons | -0.108 |
Token | of |
Token ID | 286 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
uracy | +1.123 |
opin | +1.079 |
ription | +1.070 |
romy | +0.986 |
psons | +0.961 |
Neg logit contributions | |
Pixel | -1.040 |
oise | -0.977 |
archy | -0.909 |
ould | -0.891 |
Maher | -0.884 |
Token | being |
Token ID | 852 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
uracy | +0.741 |
opin | +0.713 |
ription | +0.708 |
romy | +0.655 |
psons | +0.639 |
Neg logit contributions | |
Pixel | -0.622 |
oise | -0.582 |
archy | -0.540 |
ould | -0.528 |
Maher | -0.524 |
Token | a |
Token ID | 257 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
uracy | +0.551 |
opin | +0.526 |
ription | +0.525 |
romy | +0.483 |
psons | +0.471 |
Neg logit contributions | |
Pixel | -0.522 |
oise | -0.490 |
archy | -0.458 |
ould | -0.446 |
Maher | -0.446 |
Token | helpless |
Token ID | 21144 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
Pixel | +0.154 |
oise | +0.144 |
archy | +0.133 |
ould | +0.130 |
Maher | +0.129 |
Neg logit contributions | |
uracy | -0.186 |
opin | -0.179 |
ription | -0.178 |
romy | -0.165 |
psons | -0.161 |
Token | princess |
Token ID | 21752 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
uracy | +4.957 |
opin | +4.869 |
ription | +4.683 |
romy | +4.443 |
inval | +4.440 |
Neg logit contributions | |
Pixel | -3.832 |
oise | -3.517 |
archy | -3.276 |
ould | -3.249 |
Maher | -3.191 |
Token | in |
Token ID | 287 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
uracy | +1.839 |
opin | +1.765 |
ription | +1.751 |
romy | +1.607 |
inval | +1.572 |
Neg logit contributions | |
Pixel | -1.808 |
oise | -1.697 |
archy | -1.585 |
ould | -1.560 |
Maher | -1.549 |
Token | a |
Token ID | 257 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
uracy | +0.617 |
opin | +0.595 |
ription | +0.589 |
romy | +0.544 |
inval | +0.532 |
Neg logit contributions | |
Pixel | -0.526 |
oise | -0.493 |
archy | -0.457 |
ould | -0.448 |
Maher | -0.443 |
Token | tower |
Token ID | 10580 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
uracy | +0.423 |
opin | +0.407 |
ription | +0.403 |
romy | +0.372 |
psons | +0.362 |
Neg logit contributions | |
Pixel | -0.396 |
oise | -0.372 |
archy | -0.346 |
ould | -0.339 |
Maher | -0.336 |
Token | that |
Token ID | 326 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
uracy | +1.396 |
opin | +1.344 |
ription | +1.331 |
romy | +1.225 |
inval | +1.196 |
Neg logit contributions | |
Pixel | -1.338 |
oise | -1.258 |
archy | -1.167 |
ould | -1.150 |
Maher | -1.140 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
uracy | +2.308 |
opin | +2.208 |
ription | +2.187 |
romy | +1.997 |
inval | +1.955 |
Neg logit contributions | |
Pixel | -2.449 |
oise | -2.317 |
archy | -2.169 |
ould | -2.118 |
Maher | -2.106 |
Token |
|
Token ID | 198 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
uracy | +0.392 |
opin | +0.375 |
ription | +0.373 |
romy | +0.343 |
psons | +0.333 |
Neg logit contributions | |
Pixel | -0.394 |
oise | -0.371 |
archy | -0.346 |
ould | -0.339 |
Maher | -0.338 |
Token |
|
Token ID | 198 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
uracy | +0.191 |
opin | +0.184 |
ription | +0.183 |
romy | +0.168 |
psons | +0.164 |
Neg logit contributions | |
Pixel | -0.188 |
oise | -0.177 |
archy | -0.165 |
ould | -0.162 |
Maher | -0.161 |
Token | sudo |
Token ID | 24032 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
Pixel | +0.087 |
oise | +0.080 |
archy | +0.072 |
ould | +0.070 |
Maher | +0.069 |
Neg logit contributions | |
uracy | -0.150 |
opin | -0.146 |
ription | -0.145 |
romy | -0.135 |
psons | -0.133 |
Token | ch |
Token ID | 442 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
uracy | +2.044 |
opin | +1.977 |
ription | +1.953 |
romy | +1.813 |
psons | +1.776 |
Neg logit contributions | |
Pixel | -1.601 |
oise | -1.494 |
archy | -1.375 |
ould | -1.346 |
Maher | -1.332 |
Token | mod |
Token ID | 4666 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
Pixel | +0.542 |
oise | +0.507 |
archy | +0.473 |
ould | +0.461 |
Maher | +0.456 |
Neg logit contributions | |
uracy | -0.642 |
opin | -0.619 |
ription | -0.614 |
romy | -0.567 |
inval | -0.554 |
Token | + |
Token ID | 1343 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
uracy | +4.436 |
opin | +4.316 |
ription | +4.215 |
romy | +3.898 |
inval | +3.855 |
Neg logit contributions | |
Pixel | -4.175 |
oise | -3.898 |
archy | -3.632 |
ould | -3.556 |
Maher | -3.551 |
Token | x |
Token ID | 87 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
uracy | +0.085 |
opin | +0.081 |
ription | +0.081 |
romy | +0.074 |
psons | +0.072 |
Neg logit contributions | |
Pixel | -0.078 |
oise | -0.073 |
archy | -0.068 |
ould | -0.067 |
Maher | -0.066 |
Token | / |
Token ID | 1220 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
uracy | +1.257 |
opin | +1.217 |
ription | +1.208 |
romy | +1.129 |
psons | +1.106 |
Neg logit contributions | |
Pixel | -0.775 |
oise | -0.716 |
archy | -0.651 |
ould | -0.634 |
Maher | -0.628 |
Token | etc |
Token ID | 14784 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
uracy | +0.656 |
opin | +0.632 |
ription | +0.626 |
romy | +0.584 |
psons | +0.569 |
Neg logit contributions | |
Pixel | -0.542 |
oise | -0.505 |
archy | -0.467 |
ould | -0.458 |
Maher | -0.457 |
Token | / |
Token ID | 14 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
Pixel | +0.605 |
oise | +0.563 |
archy | +0.519 |
ould | +0.507 |
Maher | +0.502 |
Neg logit contributions | |
uracy | -0.815 |
opin | -0.791 |
ription | -0.785 |
romy | -0.727 |
inval | -0.714 |
Token | the |
Token ID | 262 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
Pixel | +0.940 |
oise | +0.880 |
archy | +0.809 |
ould | +0.797 |
Maher | +0.795 |
Neg logit contributions | |
uracy | -1.175 |
ription | -1.122 |
opin | -1.118 |
romy | -1.043 |
psons | -1.016 |
Token | name |
Token ID | 1438 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
uracy | +0.199 |
opin | +0.191 |
ription | +0.190 |
romy | +0.176 |
psons | +0.171 |
Neg logit contributions | |
Pixel | -0.175 |
oise | -0.164 |
archy | -0.152 |
ould | -0.149 |
Maher | -0.148 |
Token | of |
Token ID | 286 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
Pixel | +0.581 |
oise | +0.546 |
archy | +0.504 |
Maher | +0.492 |
ould | +0.492 |
Neg logit contributions | |
uracy | -0.710 |
opin | -0.680 |
ription | -0.677 |
romy | -0.629 |
psons | -0.611 |
Token | his |
Token ID | 465 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
uracy | +0.399 |
opin | +0.384 |
ription | +0.380 |
romy | +0.350 |
psons | +0.342 |
Neg logit contributions | |
Pixel | -0.367 |
oise | -0.345 |
archy | -0.322 |
ould | -0.315 |
Maher | -0.312 |
Token | favorite |
Token ID | 4004 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
uracy | +0.693 |
opin | +0.664 |
ription | +0.662 |
romy | +0.612 |
psons | +0.597 |
Neg logit contributions | |
Pixel | -0.579 |
oise | -0.543 |
archy | -0.502 |
ould | -0.492 |
Maher | -0.488 |
Token | hero |
Token ID | 4293 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
uracy | +0.726 |
opin | +0.697 |
ription | +0.693 |
romy | +0.640 |
psons | +0.624 |
Neg logit contributions | |
Pixel | -0.628 |
oise | -0.589 |
archy | -0.546 |
ould | -0.534 |
Maher | -0.530 |
Token | , |
Token ID | 11 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
uracy | +4.250 |
opin | +4.105 |
ription | +4.064 |
romy | +3.704 |
psons | +3.645 |
Neg logit contributions | |
Pixel | -4.304 |
oise | -4.045 |
archy | -3.769 |
ould | -3.716 |
Maher | -3.695 |
Token | Anti |
Token ID | 12297 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
uracy | +0.552 |
opin | +0.531 |
ription | +0.528 |
romy | +0.488 |
psons | +0.476 |
Neg logit contributions | |
Pixel | -0.464 |
oise | -0.434 |
archy | -0.402 |
ould | -0.394 |
Maher | -0.391 |
Token | - |
Token ID | 12 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
uracy | +0.091 |
opin | +0.087 |
ription | +0.086 |
romy | +0.080 |
inval | +0.078 |
Neg logit contributions | |
Pixel | -0.080 |
oise | -0.075 |
archy | -0.070 |
ould | -0.068 |
Maher | -0.068 |
Token | mage |
Token ID | 25561 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
Pixel | +0.535 |
oise | +0.502 |
archy | +0.466 |
ould | +0.457 |
Maher | +0.453 |
Neg logit contributions | |
uracy | -0.605 |
opin | -0.581 |
ription | -0.577 |
romy | -0.533 |
psons | -0.519 |
Token | , |
Token ID | 11 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
Pixel | +0.107 |
oise | +0.100 |
archy | +0.094 |
ould | +0.092 |
Maher | +0.091 |
Neg logit contributions | |
uracy | -0.110 |
opin | -0.106 |
ription | -0.105 |
romy | -0.096 |
psons | -0.094 |
Token | help |
Token ID | 1037 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
Pixel | +0.119 |
oise | +0.111 |
archy | +0.104 |
ould | +0.101 |
Maher | +0.100 |
Neg logit contributions | |
uracy | -0.143 |
ription | -0.138 |
opin | -0.138 |
romy | -0.127 |
psons | -0.124 |
Token | =" |
Token ID | 2625 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
uracy | +0.115 |
opin | +0.112 |
ription | +0.111 |
romy | +0.104 |
inval | +0.101 |
Neg logit contributions | |
Pixel | -0.070 |
oise | -0.065 |
archy | -0.059 |
ould | -0.058 |
Maher | -0.057 |
Token | path |
Token ID | 6978 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
Pixel | +1.159 |
oise | +1.070 |
archy | +1.001 |
ould | +0.979 |
Maher | +0.961 |
Neg logit contributions | |
uracy | -1.220 |
opin | -1.183 |
ription | -1.179 |
romy | -1.080 |
inval | -1.054 |
Token | to |
Token ID | 284 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
Pixel | +0.640 |
oise | +0.599 |
archy | +0.560 |
ould | +0.548 |
Maher | +0.544 |
Neg logit contributions | |
uracy | -0.698 |
opin | -0.674 |
ription | -0.670 |
romy | -0.619 |
inval | -0.605 |
Token | images |
Token ID | 4263 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
uracy | +1.278 |
opin | +1.234 |
ription | +1.225 |
romy | +1.139 |
psons | +1.114 |
Neg logit contributions | |
Pixel | -0.913 |
oise | -0.849 |
archy | -0.781 |
ould | -0.762 |
Maher | -0.754 |
Token | directory |
Token ID | 8619 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
uracy | +0.539 |
opin | +0.517 |
ription | +0.512 |
romy | +0.469 |
psons | +0.455 |
Neg logit contributions | |
Pixel | -0.604 |
oise | -0.569 |
archy | -0.536 |
ould | -0.525 |
Maher | -0.522 |
Token | ") |
Token ID | 4943 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
uracy | +3.735 |
opin | +3.546 |
ription | +3.518 |
romy | +3.128 |
inval | +3.068 |
Neg logit contributions | |
Pixel | -4.800 |
oise | -4.561 |
archy | -4.265 |
ould | -4.201 |
Maher | -4.152 |
Token | args |
Token ID | 26498 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
uracy | +0.133 |
ription | +0.127 |
opin | +0.126 |
romy | +0.116 |
psons | +0.112 |
Neg logit contributions | |
Pixel | -0.146 |
oise | -0.138 |
archy | -0.129 |
ould | -0.127 |
Maher | -0.127 |
Token | = |
Token ID | 796 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
Pixel | +0.083 |
oise | +0.076 |
archy | +0.070 |
ould | +0.068 |
Maher | +0.067 |
Neg logit contributions | |
uracy | -0.124 |
opin | -0.120 |
ription | -0.119 |
romy | -0.111 |
inval | -0.109 |
Token | v |
Token ID | 410 |
Feature activation | -8.55e-02 |
Pos logit contributions | |
uracy | +0.574 |
opin | +0.551 |
ription | +0.550 |
romy | +0.506 |
psons | +0.493 |
Neg logit contributions | |
Pixel | -0.499 |
oise | -0.467 |
archy | -0.434 |
ould | -0.425 |
Maher | -0.423 |
Token | ars |
Token ID | 945 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
Pixel | +0.217 |
oise | +0.206 |
archy | +0.193 |
ould | +0.190 |
Maher | +0.187 |
Neg logit contributions | |
uracy | -0.193 |
opin | -0.185 |
ription | -0.183 |
romy | -0.167 |
inval | -0.163 |
Token |
|
Token ID | 198 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
uracy | +0.374 |
opin | +0.359 |
ription | +0.356 |
romy | +0.327 |
psons | +0.318 |
Neg logit contributions | |
Pixel | -0.378 |
oise | -0.357 |
archy | -0.332 |
ould | -0.326 |
Maher | -0.325 |
Token |
|
Token ID | 198 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
uracy | +0.140 |
opin | +0.134 |
ription | +0.133 |
romy | +0.122 |
psons | +0.119 |
Neg logit contributions | |
Pixel | -0.138 |
oise | -0.130 |
archy | -0.121 |
ould | -0.119 |
Maher | -0.118 |
Token | It |
Token ID | 1026 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
Pixel | +0.049 |
oise | +0.045 |
archy | +0.042 |
ould | +0.041 |
Maher | +0.041 |
Neg logit contributions | |
uracy | -0.053 |
opin | -0.051 |
ription | -0.051 |
romy | -0.047 |
psons | -0.046 |
Token | is |
Token ID | 318 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
uracy | +0.071 |
opin | +0.069 |
ription | +0.069 |
romy | +0.064 |
psons | +0.062 |
Neg logit contributions | |
Pixel | -0.050 |
oise | -0.047 |
archy | -0.043 |
ould | -0.042 |
Maher | -0.042 |
Token | very |
Token ID | 845 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
uracy | +0.982 |
opin | +0.941 |
ription | +0.938 |
romy | +0.865 |
psons | +0.843 |
Neg logit contributions | |
Pixel | -0.872 |
oise | -0.819 |
archy | -0.761 |
ould | -0.745 |
Maher | -0.742 |
Token | unlikely |
Token ID | 7485 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
uracy | +0.136 |
ription | +0.131 |
opin | +0.129 |
romy | +0.120 |
psons | +0.117 |
Neg logit contributions | |
Pixel | -0.117 |
oise | -0.110 |
archy | -0.102 |
Maher | -0.100 |
ould | -0.100 |
Token | the |
Token ID | 262 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
uracy | +4.120 |
opin | +4.042 |
ription | +3.965 |
romy | +3.646 |
inval | +3.596 |
Neg logit contributions | |
Pixel | -4.225 |
oise | -3.969 |
archy | -3.732 |
ould | -3.598 |
Maher | -3.500 |
Token | North |
Token ID | 2258 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
Pixel | +0.058 |
oise | +0.055 |
archy | +0.051 |
ould | +0.049 |
Maher | +0.049 |
Neg logit contributions | |
uracy | -0.072 |
opin | -0.069 |
ription | -0.069 |
romy | -0.064 |
psons | -0.063 |
Token | Korean |
Token ID | 6983 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
uracy | +0.765 |
opin | +0.717 |
ription | +0.706 |
romy | +0.610 |
psons | +0.584 |
Neg logit contributions | |
Pixel | -1.675 |
oise | -1.604 |
archy | -1.527 |
ould | -1.506 |
Maher | -1.498 |
Token | authorities |
Token ID | 4773 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
Pixel | +0.076 |
oise | +0.071 |
archy | +0.065 |
ould | +0.064 |
Maher | +0.063 |
Neg logit contributions | |
uracy | -0.095 |
opin | -0.092 |
ription | -0.091 |
romy | -0.084 |
psons | -0.082 |
Token | would |
Token ID | 561 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
uracy | +0.798 |
opin | +0.770 |
ription | +0.770 |
romy | +0.721 |
psons | +0.707 |
Neg logit contributions | |
Pixel | -0.420 |
oise | -0.385 |
archy | -0.348 |
ould | -0.338 |
Maher | -0.335 |
Token | professionally |
Token ID | 28049 |
Feature activation | +6.80e-01 |
Pos logit contributions | |
uracy | +0.408 |
opin | +0.392 |
ription | +0.390 |
romy | +0.362 |
psons | +0.354 |
Neg logit contributions | |
Pixel | -0.323 |
oise | -0.302 |
archy | -0.280 |
ould | -0.273 |
Maher | -0.271 |
Token | packaged |
Token ID | 25555 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
uracy | +1.620 |
opin | +1.557 |
ription | +1.541 |
romy | +1.414 |
inval | +1.379 |
Neg logit contributions | |
Pixel | -1.640 |
oise | -1.545 |
archy | -1.443 |
ould | -1.413 |
Maher | -1.403 |
Token | — |
Token ID | 851 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
Pixel | +0.644 |
oise | +0.605 |
archy | +0.563 |
ould | +0.553 |
Maher | +0.549 |
Neg logit contributions | |
uracy | -0.714 |
opin | -0.682 |
ription | -0.680 |
romy | -0.628 |
psons | -0.616 |
Token | little |
Token ID | 1310 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
uracy | +0.491 |
opin | +0.472 |
ription | +0.468 |
romy | +0.433 |
psons | +0.424 |
Neg logit contributions | |
Pixel | -0.417 |
oise | -0.390 |
archy | -0.363 |
ould | -0.355 |
Maher | -0.352 |
Token | ch |
Token ID | 442 |
Feature activation | -7.03e-01 |
Pos logit contributions | |
uracy | +1.194 |
opin | +1.156 |
ription | +1.144 |
romy | +1.059 |
inval | +1.036 |
Neg logit contributions | |
Pixel | -0.992 |
oise | -0.928 |
archy | -0.860 |
ould | -0.842 |
Maher | -0.832 |
Token | ocol |
Token ID | 4668 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
Pixel | +1.365 |
oise | +1.288 |
archy | +1.199 |
ould | +1.176 |
Maher | +1.111 |
Neg logit contributions | |
uracy | -1.978 |
opin | -1.942 |
ription | -1.890 |
romy | -1.755 |
inval | -1.753 |
Token | ates |
Token ID | 689 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
uracy | +3.915 |
ription | +3.779 |
opin | +3.727 |
romy | +3.408 |
psons | +3.320 |
Neg logit contributions | |
Pixel | -4.529 |
oise | -4.197 |
Maher | -3.890 |
archy | -3.879 |
ould | -3.828 |
Token | , |
Token ID | 11 |
Feature activation | -5.36e-03 |
Pos logit contributions | |
uracy | +0.413 |
opin | +0.397 |
ription | +0.394 |
romy | +0.363 |
psons | +0.354 |
Neg logit contributions | |
Pixel | -0.379 |
oise | -0.356 |
archy | -0.332 |
ould | -0.325 |
Maher | -0.322 |
Token | little |
Token ID | 1310 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
Pixel | +0.011 |
oise | +0.011 |
archy | +0.010 |
ould | +0.010 |
Maher | +0.010 |
Neg logit contributions | |
uracy | -0.014 |
opin | -0.014 |
ription | -0.014 |
romy | -0.013 |
psons | -0.012 |
Token | mint |
Token ID | 21043 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
uracy | +1.789 |
opin | +1.734 |
ription | +1.713 |
romy | +1.586 |
inval | +1.551 |
Neg logit contributions | |
Pixel | -1.535 |
oise | -1.437 |
archy | -1.335 |
ould | -1.307 |
Maher | -1.292 |
Token | s |
Token ID | 82 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
Pixel | +0.568 |
oise | +0.539 |
archy | +0.497 |
ould | +0.485 |
Maher | +0.477 |
Neg logit contributions | |
uracy | -0.692 |
opin | -0.664 |
ription | -0.661 |
romy | -0.609 |
psons | -0.597 |
Token | help |
Token ID | 1037 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
uracy | +0.300 |
opin | +0.291 |
ription | +0.290 |
romy | +0.273 |
psons | +0.268 |
Neg logit contributions | |
Pixel | -0.129 |
oise | -0.117 |
archy | -0.104 |
ould | -0.100 |
Maher | -0.099 |
Token | =" |
Token ID | 2625 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
uracy | +0.886 |
opin | +0.853 |
ription | +0.849 |
romy | +0.799 |
psons | +0.783 |
Neg logit contributions | |
Pixel | -0.418 |
oise | -0.382 |
archy | -0.343 |
ould | -0.332 |
Maher | -0.327 |
Token | path |
Token ID | 6978 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Pixel | +0.636 |
oise | +0.608 |
archy | +0.582 |
ould | +0.572 |
Maher | +0.561 |
Neg logit contributions | |
uracy | -0.302 |
opin | -0.291 |
ription | -0.286 |
romy | -0.241 |
inval | -0.238 |
Token | to |
Token ID | 284 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
Pixel | +0.244 |
oise | +0.230 |
archy | +0.214 |
ould | +0.208 |
Maher | +0.208 |
Neg logit contributions | |
uracy | -0.264 |
opin | -0.255 |
ription | -0.254 |
romy | -0.233 |
inval | -0.228 |
Token | images |
Token ID | 4263 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
uracy | +2.027 |
opin | +1.974 |
ription | +1.954 |
romy | +1.857 |
psons | +1.828 |
Neg logit contributions | |
Pixel | -0.685 |
oise | -0.606 |
archy | -0.522 |
ould | -0.499 |
Maher | -0.482 |
Token | directory |
Token ID | 8619 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
uracy | +0.827 |
opin | +0.780 |
ription | +0.771 |
romy | +0.680 |
psons | +0.653 |
Neg logit contributions | |
Pixel | -1.495 |
oise | -1.427 |
archy | -1.355 |
ould | -1.335 |
Maher | -1.327 |
Token | ") |
Token ID | 4943 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
uracy | +3.356 |
opin | +3.068 |
ription | +3.060 |
romy | +2.696 |
psons | +2.668 |
Neg logit contributions | |
Pixel | -5.098 |
oise | -4.865 |
archy | -4.599 |
ould | -4.529 |
Maher | -4.492 |
Token | args |
Token ID | 26498 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
uracy | +0.711 |
opin | +0.679 |
ription | +0.673 |
romy | +0.613 |
psons | +0.595 |
Neg logit contributions | |
Pixel | -0.840 |
oise | -0.795 |
archy | -0.746 |
ould | -0.733 |
Maher | -0.728 |
Token | = |
Token ID | 796 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
uracy | +0.769 |
opin | +0.738 |
ription | +0.733 |
romy | +0.686 |
psons | +0.671 |
Neg logit contributions | |
Pixel | -0.512 |
oise | -0.476 |
archy | -0.437 |
ould | -0.426 |
Maher | -0.422 |
Token | v |
Token ID | 410 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
uracy | +1.429 |
opin | +1.378 |
ription | +1.356 |
romy | +1.271 |
inval | +1.242 |
Neg logit contributions | |
Pixel | -1.071 |
oise | -1.002 |
archy | -0.923 |
ould | -0.902 |
Maher | -0.892 |
Token | ars |
Token ID | 945 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
uracy | +0.327 |
opin | +0.312 |
ription | +0.309 |
romy | +0.280 |
psons | +0.272 |
Neg logit contributions | |
Pixel | -0.409 |
oise | -0.387 |
archy | -0.364 |
ould | -0.358 |
Maher | -0.356 |
Token | should |
Token ID | 815 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
Pixel | +0.350 |
oise | +0.325 |
archy | +0.299 |
ould | +0.294 |
Maher | +0.290 |
Neg logit contributions | |
uracy | -0.510 |
ription | -0.491 |
opin | -0.490 |
romy | -0.455 |
psons | -0.446 |
Token | applaud |
Token ID | 22517 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Pixel | +0.851 |
oise | +0.802 |
archy | +0.744 |
ould | +0.737 |
Maher | +0.726 |
Neg logit contributions | |
uracy | -0.984 |
ription | -0.946 |
opin | -0.940 |
romy | -0.870 |
psons | -0.854 |
Token | them |
Token ID | 606 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
Pixel | +1.209 |
oise | +1.129 |
archy | +1.032 |
Maher | +1.029 |
ould | +1.018 |
Neg logit contributions | |
uracy | -1.635 |
ription | -1.580 |
opin | -1.561 |
romy | -1.455 |
psons | -1.426 |
Token | for |
Token ID | 329 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
Pixel | +0.560 |
oise | +0.525 |
archy | +0.487 |
ould | +0.481 |
Maher | +0.478 |
Neg logit contributions | |
uracy | -0.632 |
ription | -0.608 |
opin | -0.606 |
romy | -0.561 |
psons | -0.543 |
Token | their |
Token ID | 511 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
Pixel | +0.067 |
oise | +0.063 |
archy | +0.059 |
ould | +0.058 |
Maher | +0.057 |
Neg logit contributions | |
uracy | -0.079 |
ription | -0.075 |
opin | -0.075 |
romy | -0.069 |
psons | -0.068 |
Token | realism |
Token ID | 30787 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
uracy | +0.725 |
opin | +0.695 |
ription | +0.691 |
romy | +0.635 |
psons | +0.618 |
Neg logit contributions | |
Pixel | -0.707 |
oise | -0.665 |
archy | -0.621 |
ould | -0.608 |
Maher | -0.605 |
Token | and |
Token ID | 290 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
uracy | +3.994 |
opin | +3.797 |
ription | +3.698 |
romy | +3.377 |
inval | +3.347 |
Neg logit contributions | |
Pixel | -4.409 |
oise | -4.178 |
archy | -3.970 |
ould | -3.811 |
Maher | -3.786 |
Token | bold |
Token ID | 10758 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
uracy | +0.064 |
ription | +0.062 |
opin | +0.061 |
romy | +0.057 |
psons | +0.055 |
Neg logit contributions | |
Pixel | -0.057 |
oise | -0.053 |
archy | -0.049 |
ould | -0.049 |
Maher | -0.048 |
Token | ness |
Token ID | 1108 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
uracy | +0.368 |
opin | +0.352 |
ription | +0.351 |
romy | +0.322 |
inval | +0.314 |
Neg logit contributions | |
Pixel | -0.365 |
oise | -0.344 |
archy | -0.321 |
ould | -0.315 |
Maher | -0.312 |
Token | , |
Token ID | 11 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
uracy | +1.228 |
opin | +1.178 |
ription | +1.165 |
romy | +1.069 |
psons | +1.043 |
Neg logit contributions | |
Pixel | -1.246 |
oise | -1.175 |
archy | -1.099 |
ould | -1.075 |
Maher | -1.066 |
Token | and |
Token ID | 290 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
uracy | +0.539 |
opin | +0.518 |
ription | +0.515 |
romy | +0.475 |
psons | +0.462 |
Neg logit contributions | |
Pixel | -0.481 |
oise | -0.451 |
archy | -0.419 |
ould | -0.411 |
Maher | -0.409 |
Token | facts |
Token ID | 6419 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
Pixel | +0.574 |
oise | +0.544 |
archy | +0.504 |
ould | +0.497 |
Maher | +0.495 |
Neg logit contributions | |
uracy | -0.584 |
ription | -0.556 |
opin | -0.554 |
romy | -0.509 |
psons | -0.497 |
Token | right |
Token ID | 826 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
uracy | +1.055 |
opin | +1.012 |
ription | +1.007 |
romy | +0.924 |
psons | +0.903 |
Neg logit contributions | |
Pixel | -0.990 |
oise | -0.932 |
archy | -0.867 |
ould | -0.852 |
Maher | -0.844 |
Token | , |
Token ID | 11 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
uracy | +1.457 |
opin | +1.396 |
ription | +1.382 |
romy | +1.271 |
psons | +1.239 |
Neg logit contributions | |
Pixel | -1.459 |
oise | -1.367 |
archy | -1.282 |
ould | -1.242 |
Maher | -1.236 |
Token | are |
Token ID | 389 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
uracy | +0.251 |
opin | +0.241 |
ription | +0.240 |
romy | +0.222 |
psons | +0.217 |
Neg logit contributions | |
Pixel | -0.195 |
oise | -0.183 |
archy | -0.169 |
ould | -0.166 |
Maher | -0.164 |
Token | more |
Token ID | 517 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
Pixel | +0.584 |
oise | +0.547 |
archy | +0.512 |
Maher | +0.505 |
ould | +0.499 |
Neg logit contributions | |
uracy | -0.701 |
ription | -0.671 |
opin | -0.665 |
romy | -0.619 |
psons | -0.602 |
Token | likely |
Token ID | 1884 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
uracy | +0.423 |
ription | +0.404 |
opin | +0.403 |
romy | +0.371 |
psons | +0.363 |
Neg logit contributions | |
Pixel | -0.398 |
oise | -0.374 |
archy | -0.351 |
Maher | -0.343 |
ould | -0.341 |
Token | to |
Token ID | 284 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
uracy | +4.083 |
opin | +4.000 |
ription | +3.903 |
romy | +3.551 |
inval | +3.536 |
Neg logit contributions | |
Pixel | -4.241 |
oise | -4.001 |
archy | -3.711 |
ould | -3.662 |
Maher | -3.577 |
Token | self |
Token ID | 2116 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
uracy | +1.080 |
opin | +1.042 |
ription | +1.034 |
romy | +0.958 |
inval | +0.937 |
Neg logit contributions | |
Pixel | -0.840 |
oise | -0.783 |
archy | -0.724 |
ould | -0.707 |
Maher | -0.701 |
Token | - |
Token ID | 12 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
Pixel | +0.491 |
oise | +0.462 |
archy | +0.429 |
ould | +0.418 |
Maher | +0.416 |
Neg logit contributions | |
uracy | -0.588 |
ription | -0.561 |
opin | -0.560 |
romy | -0.518 |
psons | -0.506 |
Token | ident |
Token ID | 738 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
Pixel | +0.561 |
oise | +0.530 |
archy | +0.498 |
ould | +0.484 |
Maher | +0.479 |
Neg logit contributions | |
uracy | -0.543 |
opin | -0.518 |
ription | -0.516 |
romy | -0.471 |
psons | -0.461 |
Token | ify |
Token ID | 1958 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
Pixel | +0.839 |
oise | +0.804 |
archy | +0.757 |
ould | +0.735 |
Maher | +0.731 |
Neg logit contributions | |
uracy | -0.629 |
opin | -0.600 |
ription | -0.593 |
romy | -0.533 |
inval | -0.519 |
Token | likely |
Token ID | 1884 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
uracy | +0.950 |
opin | +0.914 |
ription | +0.908 |
romy | +0.838 |
psons | +0.818 |
Neg logit contributions | |
Pixel | -0.817 |
oise | -0.765 |
archy | -0.711 |
ould | -0.695 |
Maher | -0.690 |
Token | just |
Token ID | 655 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
uracy | +3.012 |
opin | +2.939 |
ription | +2.852 |
romy | +2.601 |
inval | +2.589 |
Neg logit contributions | |
Pixel | -3.307 |
oise | -3.144 |
archy | -2.910 |
ould | -2.887 |
Maher | -2.809 |
Token | because |
Token ID | 780 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
uracy | +0.106 |
opin | +0.101 |
ription | +0.101 |
romy | +0.093 |
psons | +0.091 |
Neg logit contributions | |
Pixel | -0.101 |
oise | -0.095 |
archy | -0.089 |
Maher | -0.087 |
ould | -0.087 |
Token | they |
Token ID | 484 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
uracy | +1.020 |
opin | +0.981 |
ription | +0.972 |
romy | +0.895 |
inval | +0.872 |
Neg logit contributions | |
Pixel | -0.967 |
oise | -0.908 |
archy | -0.846 |
ould | -0.829 |
Maher | -0.821 |
Token | are |
Token ID | 389 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
uracy | +0.652 |
opin | +0.628 |
ription | +0.624 |
romy | +0.578 |
psons | +0.565 |
Neg logit contributions | |
Pixel | -0.509 |
oise | -0.476 |
archy | -0.440 |
ould | -0.430 |
Maher | -0.426 |
Token | unwilling |
Token ID | 19084 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
uracy | +1.012 |
ription | +0.968 |
opin | +0.966 |
romy | +0.890 |
psons | +0.874 |
Neg logit contributions | |
Pixel | -0.845 |
oise | -0.793 |
archy | -0.737 |
Maher | -0.719 |
ould | -0.716 |
Token | to |
Token ID | 284 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
uracy | +3.719 |
opin | +3.667 |
ription | +3.530 |
romy | +3.255 |
inval | +3.215 |
Neg logit contributions | |
Pixel | -4.421 |
oise | -4.175 |
archy | -3.929 |
ould | -3.885 |
Maher | -3.800 |
Token | report |
Token ID | 989 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
uracy | +0.351 |
ription | +0.335 |
opin | +0.335 |
romy | +0.310 |
psons | +0.306 |
Neg logit contributions | |
Pixel | -0.274 |
oise | -0.259 |
archy | -0.239 |
ould | -0.234 |
Maher | -0.232 |
Token | to |
Token ID | 284 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
uracy | +1.137 |
opin | +1.094 |
ription | +1.086 |
romy | +1.004 |
psons | +0.980 |
Neg logit contributions | |
Pixel | -0.971 |
oise | -0.909 |
archy | -0.844 |
ould | -0.826 |
Maher | -0.819 |
Token | the |
Token ID | 262 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
uracy | +0.408 |
opin | +0.390 |
ription | +0.389 |
romy | +0.360 |
psons | +0.353 |
Neg logit contributions | |
Pixel | -0.328 |
oise | -0.308 |
archy | -0.285 |
ould | -0.278 |
Maher | -0.278 |
Token | IMF |
Token ID | 25375 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
uracy | +0.332 |
opin | +0.317 |
ription | +0.316 |
romy | +0.290 |
psons | +0.284 |
Neg logit contributions | |
Pixel | -0.314 |
oise | -0.297 |
archy | -0.277 |
ould | -0.270 |
Maher | -0.269 |
Token | Cullen |
Token ID | 47681 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
Pixel | +0.005 |
oise | +0.004 |
archy | +0.004 |
Maher | +0.004 |
ould | +0.004 |
Neg logit contributions | |
uracy | -0.006 |
opin | -0.005 |
ription | -0.005 |
romy | -0.005 |
psons | -0.005 |
Token | Schwarz |
Token ID | 34835 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
Pixel | +0.089 |
oise | +0.083 |
archy | +0.077 |
ould | +0.075 |
Maher | +0.075 |
Neg logit contributions | |
uracy | -0.111 |
opin | -0.107 |
ription | -0.106 |
romy | -0.098 |
psons | -0.096 |
Token | , |
Token ID | 11 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
Pixel | +0.885 |
oise | +0.831 |
archy | +0.774 |
ould | +0.759 |
Maher | +0.753 |
Neg logit contributions | |
uracy | -0.945 |
opin | -0.908 |
ription | -0.901 |
romy | -0.829 |
psons | -0.808 |
Token | was |
Token ID | 373 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
Pixel | +0.313 |
oise | +0.294 |
archy | +0.273 |
ould | +0.267 |
Maher | +0.265 |
Neg logit contributions | |
uracy | -0.373 |
opin | -0.358 |
ription | -0.355 |
romy | -0.329 |
psons | -0.321 |
Token | more |
Token ID | 517 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
uracy | +0.066 |
opin | +0.063 |
ription | +0.062 |
romy | +0.058 |
psons | +0.056 |
Neg logit contributions | |
Pixel | -0.065 |
oise | -0.061 |
archy | -0.057 |
ould | -0.056 |
Maher | -0.056 |
Token | blunt |
Token ID | 19861 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Pixel | +0.211 |
oise | +0.200 |
archy | +0.187 |
Maher | +0.184 |
ould | +0.184 |
Neg logit contributions | |
uracy | -0.182 |
ription | -0.174 |
opin | -0.172 |
romy | -0.158 |
psons | -0.154 |
Token | , |
Token ID | 11 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
uracy | +3.779 |
opin | +3.605 |
ription | +3.529 |
romy | +3.205 |
inval | +3.171 |
Neg logit contributions | |
Pixel | -4.391 |
oise | -4.140 |
archy | -3.953 |
ould | -3.823 |
Maher | -3.747 |
Token | sounding |
Token ID | 22655 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
uracy | +1.162 |
opin | +1.119 |
ription | +1.108 |
romy | +1.017 |
inval | +0.993 |
Neg logit contributions | |
Pixel | -1.124 |
oise | -1.058 |
archy | -0.987 |
ould | -0.965 |
Maher | -0.954 |
Token | off |
Token ID | 572 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
Pixel | +0.068 |
oise | +0.064 |
archy | +0.059 |
ould | +0.058 |
Maher | +0.058 |
Neg logit contributions | |
uracy | -0.084 |
ription | -0.080 |
opin | -0.080 |
romy | -0.074 |
psons | -0.073 |
Token | after |
Token ID | 706 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
Pixel | +0.485 |
oise | +0.457 |
archy | +0.424 |
Maher | +0.420 |
ould | +0.417 |
Neg logit contributions | |
uracy | -0.508 |
ription | -0.486 |
opin | -0.485 |
romy | -0.447 |
inval | -0.436 |
Token | failing |
Token ID | 9894 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
uracy | +1.469 |
opin | +1.417 |
ription | +1.402 |
romy | +1.293 |
psons | +1.266 |
Neg logit contributions | |
Pixel | -1.304 |
oise | -1.223 |
archy | -1.138 |
ould | -1.114 |
Maher | -1.096 |
Token | opposed |
Token ID | 6886 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
uracy | +0.354 |
ription | +0.338 |
opin | +0.337 |
romy | +0.310 |
psons | +0.301 |
Neg logit contributions | |
Pixel | -0.337 |
oise | -0.318 |
archy | -0.295 |
ould | -0.290 |
Maher | -0.290 |
Token | by |
Token ID | 416 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
uracy | +0.588 |
opin | +0.568 |
ription | +0.564 |
romy | +0.523 |
inval | +0.512 |
Neg logit contributions | |
Pixel | -0.438 |
oise | -0.408 |
archy | -0.376 |
ould | -0.367 |
Maher | -0.363 |
Token | Pennsylvania |
Token ID | 9589 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
uracy | +0.522 |
opin | +0.502 |
ription | +0.500 |
romy | +0.465 |
psons | +0.454 |
Neg logit contributions | |
Pixel | -0.378 |
oise | -0.352 |
archy | -0.324 |
ould | -0.316 |
Maher | -0.315 |
Token | 's |
Token ID | 338 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
Pixel | +0.359 |
oise | +0.324 |
archy | +0.284 |
Maher | +0.274 |
ould | +0.271 |
Neg logit contributions | |
uracy | -0.910 |
ription | -0.879 |
opin | -0.876 |
romy | -0.829 |
psons | -0.807 |
Token | Th |
Token ID | 536 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
uracy | +0.602 |
opin | +0.580 |
ription | +0.578 |
romy | +0.539 |
psons | +0.527 |
Neg logit contributions | |
Pixel | -0.387 |
oise | -0.359 |
archy | -0.328 |
ould | -0.319 |
Maher | -0.318 |
Token | add |
Token ID | 2860 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Pixel | +0.552 |
oise | +0.527 |
archy | +0.492 |
ould | +0.484 |
Maher | +0.474 |
Neg logit contributions | |
uracy | -0.542 |
opin | -0.524 |
ription | -0.517 |
romy | -0.474 |
inval | -0.466 |
Token | eus |
Token ID | 27650 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
uracy | +3.471 |
opin | +3.243 |
ription | +3.229 |
romy | +2.923 |
psons | +2.851 |
Neg logit contributions | |
Pixel | -4.777 |
oise | -4.458 |
archy | -4.225 |
Maher | -4.137 |
ould | -4.126 |
Token | Stevens |
Token ID | 20019 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
Pixel | +0.121 |
oise | +0.113 |
archy | +0.104 |
ould | +0.102 |
Maher | +0.102 |
Neg logit contributions | |
uracy | -0.154 |
opin | -0.148 |
ription | -0.147 |
romy | -0.136 |
psons | -0.133 |
Token | , |
Token ID | 11 |
Feature activation | -9.06e-02 |
Pos logit contributions | |
uracy | +0.361 |
opin | +0.346 |
ription | +0.344 |
romy | +0.317 |
psons | +0.309 |
Neg logit contributions | |
Pixel | -0.326 |
oise | -0.306 |
archy | -0.285 |
ould | -0.278 |
Maher | -0.277 |
Token | who |
Token ID | 508 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
Pixel | +0.191 |
oise | +0.178 |
archy | +0.166 |
Maher | +0.162 |
ould | +0.160 |
Neg logit contributions | |
uracy | -0.245 |
opin | -0.235 |
ription | -0.234 |
romy | -0.217 |
psons | -0.212 |
Token | represented |
Token ID | 7997 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
uracy | +0.210 |
ription | +0.201 |
opin | +0.200 |
romy | +0.184 |
psons | +0.180 |
Neg logit contributions | |
Pixel | -0.187 |
oise | -0.175 |
archy | -0.163 |
ould | -0.160 |
Maher | -0.160 |
Token | to |
Token ID | 284 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Pixel | +1.536 |
oise | +1.446 |
archy | +1.349 |
ould | +1.322 |
Maher | +1.313 |
Neg logit contributions | |
uracy | -1.560 |
opin | -1.495 |
ription | -1.484 |
romy | -1.365 |
psons | -1.328 |
Token | help |
Token ID | 1037 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
Pixel | +0.215 |
oise | +0.201 |
archy | +0.185 |
ould | +0.181 |
Maher | +0.179 |
Neg logit contributions | |
uracy | -0.289 |
opin | -0.279 |
ription | -0.277 |
romy | -0.257 |
psons | -0.252 |
Token | them |
Token ID | 606 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
Pixel | +1.139 |
oise | +1.060 |
archy | +0.971 |
ould | +0.955 |
Maher | +0.948 |
Neg logit contributions | |
uracy | -1.566 |
opin | -1.512 |
ription | -1.510 |
romy | -1.385 |
psons | -1.357 |
Token | realise |
Token ID | 18996 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
uracy | +0.267 |
opin | +0.258 |
ription | +0.256 |
romy | +0.236 |
psons | +0.232 |
Neg logit contributions | |
Pixel | -0.212 |
oise | -0.199 |
archy | -0.184 |
ould | -0.179 |
Maher | -0.178 |
Token | their |
Token ID | 511 |
Feature activation | +7.87e-02 |
Pos logit contributions | |
uracy | +1.115 |
opin | +1.072 |
ription | +1.060 |
romy | +0.979 |
inval | +0.959 |
Neg logit contributions | |
Pixel | -1.043 |
oise | -0.980 |
archy | -0.913 |
ould | -0.893 |
Maher | -0.885 |
Token | dream |
Token ID | 4320 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
uracy | +0.207 |
opin | +0.199 |
ription | +0.197 |
romy | +0.183 |
psons | +0.178 |
Neg logit contributions | |
Pixel | -0.171 |
oise | -0.160 |
archy | -0.149 |
ould | -0.145 |
Maher | -0.144 |
Token | . |
Token ID | 13 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
uracy | +3.883 |
ription | +3.690 |
opin | +3.681 |
romy | +3.355 |
psons | +3.330 |
Neg logit contributions | |
Pixel | -4.183 |
oise | -3.996 |
archy | -3.709 |
ould | -3.660 |
Maher | -3.626 |
Token | Since |
Token ID | 4619 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
uracy | +0.302 |
opin | +0.289 |
ription | +0.287 |
romy | +0.262 |
psons | +0.255 |
Neg logit contributions | |
Pixel | -0.322 |
oise | -0.303 |
archy | -0.284 |
ould | -0.278 |
Maher | -0.277 |
Token | we |
Token ID | 356 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
uracy | +0.998 |
opin | +0.959 |
ription | +0.951 |
romy | +0.876 |
psons | +0.854 |
Neg logit contributions | |
Pixel | -0.941 |
oise | -0.884 |
archy | -0.824 |
ould | -0.807 |
Maher | -0.801 |
Token | heard |
Token ID | 2982 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
Pixel | +0.298 |
oise | +0.279 |
archy | +0.257 |
ould | +0.252 |
Maher | +0.250 |
Neg logit contributions | |
uracy | -0.386 |
ription | -0.372 |
opin | -0.371 |
romy | -0.341 |
psons | -0.337 |
Token | the |
Token ID | 262 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
uracy | +1.362 |
opin | +1.310 |
ription | +1.294 |
romy | +1.192 |
inval | +1.162 |
Neg logit contributions | |
Pixel | -1.355 |
oise | -1.278 |
archy | -1.195 |
ould | -1.169 |
Maher | -1.157 |
Token | ? |
Token ID | 30 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
uracy | +0.009 |
opin | +0.009 |
ription | +0.009 |
romy | +0.008 |
psons | +0.008 |
Neg logit contributions | |
Pixel | -0.008 |
oise | -0.008 |
archy | -0.007 |
ould | -0.007 |
Maher | -0.007 |
Token | Or |
Token ID | 1471 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
uracy | +0.080 |
opin | +0.076 |
ription | +0.076 |
romy | +0.070 |
psons | +0.068 |
Neg logit contributions | |
Pixel | -0.079 |
oise | -0.074 |
archy | -0.069 |
Maher | -0.068 |
ould | -0.068 |
Token | is |
Token ID | 318 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
Pixel | +0.623 |
oise | +0.583 |
archy | +0.535 |
Maher | +0.526 |
ould | +0.524 |
Neg logit contributions | |
uracy | -0.913 |
ription | -0.882 |
opin | -0.872 |
romy | -0.810 |
psons | -0.795 |
Token | it |
Token ID | 340 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
uracy | +0.630 |
opin | +0.606 |
ription | +0.600 |
romy | +0.553 |
inval | +0.540 |
Neg logit contributions | |
Pixel | -0.573 |
oise | -0.538 |
archy | -0.500 |
ould | -0.490 |
Maher | -0.484 |
Token | more |
Token ID | 517 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
uracy | +0.028 |
ription | +0.027 |
opin | +0.027 |
romy | +0.025 |
psons | +0.024 |
Neg logit contributions | |
Pixel | -0.022 |
oise | -0.021 |
archy | -0.020 |
Maher | -0.019 |
ould | -0.019 |
Token | likely |
Token ID | 1884 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
uracy | +0.609 |
opin | +0.584 |
ription | +0.582 |
romy | +0.537 |
psons | +0.524 |
Neg logit contributions | |
Pixel | -0.550 |
oise | -0.516 |
archy | -0.482 |
ould | -0.470 |
Maher | -0.470 |
Token | to |
Token ID | 284 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
uracy | +3.968 |
opin | +3.861 |
ription | +3.781 |
inval | +3.424 |
romy | +3.413 |
Neg logit contributions | |
Pixel | -4.109 |
oise | -3.894 |
archy | -3.584 |
ould | -3.559 |
Maher | -3.363 |
Token | be |
Token ID | 307 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
uracy | +0.639 |
opin | +0.616 |
ription | +0.611 |
romy | +0.567 |
psons | +0.555 |
Neg logit contributions | |
Pixel | -0.485 |
oise | -0.452 |
archy | -0.417 |
ould | -0.408 |
Maher | -0.404 |
Token | a |
Token ID | 257 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
uracy | +0.802 |
opin | +0.770 |
ription | +0.767 |
romy | +0.708 |
psons | +0.691 |
Neg logit contributions | |
Pixel | -0.706 |
oise | -0.662 |
archy | -0.616 |
ould | -0.602 |
Maher | -0.601 |
Token | team |
Token ID | 1074 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
uracy | +0.060 |
opin | +0.057 |
ription | +0.057 |
romy | +0.053 |
psons | +0.051 |
Neg logit contributions | |
Pixel | -0.051 |
oise | -0.048 |
archy | -0.045 |
ould | -0.044 |
Maher | -0.044 |
Token | in |
Token ID | 287 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
uracy | +0.134 |
ription | +0.129 |
opin | +0.129 |
romy | +0.119 |
psons | +0.116 |
Neg logit contributions | |
Pixel | -0.114 |
oise | -0.107 |
archy | -0.099 |
Maher | -0.097 |
ould | -0.097 |
Token | in |
Token ID | 287 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
Pixel | +1.101 |
oise | +1.020 |
archy | +0.954 |
Maher | +0.940 |
ould | +0.935 |
Neg logit contributions | |
uracy | -1.301 |
ription | -1.245 |
opin | -1.237 |
romy | -1.139 |
psons | -1.119 |
Token | Ohio |
Token ID | 6835 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
Pixel | +0.407 |
oise | +0.382 |
archy | +0.352 |
Maher | +0.349 |
ould | +0.343 |
Neg logit contributions | |
uracy | -0.523 |
ription | -0.499 |
opin | -0.499 |
romy | -0.462 |
psons | -0.454 |
Token | objected |
Token ID | 29811 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
Pixel | +0.586 |
oise | +0.547 |
archy | +0.509 |
ould | +0.500 |
Maher | +0.498 |
Neg logit contributions | |
uracy | -0.676 |
ription | -0.647 |
opin | -0.646 |
romy | -0.593 |
psons | -0.582 |
Token | to |
Token ID | 284 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
Pixel | +0.147 |
oise | +0.139 |
archy | +0.129 |
ould | +0.127 |
Maher | +0.126 |
Neg logit contributions | |
uracy | -0.147 |
opin | -0.141 |
ription | -0.140 |
romy | -0.129 |
psons | -0.125 |
Token | their |
Token ID | 511 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
Pixel | +0.615 |
oise | +0.579 |
archy | +0.534 |
Maher | +0.532 |
ould | +0.523 |
Neg logit contributions | |
uracy | -0.713 |
ription | -0.681 |
opin | -0.677 |
romy | -0.628 |
psons | -0.613 |
Token | cant |
Token ID | 18548 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Pixel | +0.128 |
oise | +0.120 |
archy | +0.110 |
Maher | +0.108 |
ould | +0.108 |
Neg logit contributions | |
uracy | -0.174 |
ription | -0.167 |
opin | -0.166 |
romy | -0.154 |
psons | -0.151 |
Token | een |
Token ID | 6429 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
uracy | +3.969 |
ription | +3.758 |
opin | +3.735 |
romy | +3.446 |
pent | +3.303 |
Neg logit contributions | |
Pixel | -4.112 |
oise | -3.743 |
Maher | -3.503 |
ould | -3.483 |
archy | -3.474 |
Token | serving |
Token ID | 7351 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
uracy | +3.154 |
opin | +3.080 |
ription | +3.018 |
romy | +2.760 |
inval | +2.739 |
Neg logit contributions | |
Pixel | -2.843 |
oise | -2.660 |
archy | -2.478 |
ould | -2.426 |
Maher | -2.367 |
Token | sushi |
Token ID | 36324 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
Pixel | +0.694 |
oise | +0.656 |
archy | +0.608 |
Maher | +0.599 |
ould | +0.592 |
Neg logit contributions | |
uracy | -0.783 |
ription | -0.750 |
opin | -0.740 |
romy | -0.692 |
psons | -0.673 |
Token | , |
Token ID | 11 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
uracy | +1.285 |
opin | +1.234 |
ription | +1.229 |
romy | +1.133 |
psons | +1.106 |
Neg logit contributions | |
Pixel | -1.136 |
oise | -1.066 |
archy | -0.992 |
ould | -0.971 |
Maher | -0.964 |
Token | describing |
Token ID | 12059 |
Feature activation | +7.43e-02 |
Pos logit contributions | |
uracy | +0.031 |
ription | +0.029 |
opin | +0.029 |
romy | +0.027 |
psons | +0.026 |
Neg logit contributions | |
Pixel | -0.027 |
oise | -0.025 |
archy | -0.024 |
Maher | -0.023 |
ould | -0.023 |
Token | alling |
Token ID | 9221 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
uracy | +1.216 |
opin | +1.156 |
ription | +1.144 |
romy | +1.058 |
inval | +1.019 |
Neg logit contributions | |
Pixel | -1.485 |
oise | -1.387 |
archy | -1.289 |
Maher | -1.276 |
ould | -1.258 |
Token | down |
Token ID | 866 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
uracy | +0.286 |
opin | +0.273 |
ription | +0.273 |
romy | +0.251 |
psons | +0.244 |
Neg logit contributions | |
Pixel | -0.278 |
oise | -0.262 |
archy | -0.245 |
ould | -0.240 |
Maher | -0.240 |
Token | into |
Token ID | 656 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
uracy | +0.428 |
opin | +0.410 |
ription | +0.407 |
romy | +0.379 |
psons | +0.368 |
Neg logit contributions | |
Pixel | -0.372 |
oise | -0.349 |
archy | -0.327 |
Maher | -0.319 |
ould | -0.319 |
Token | a |
Token ID | 257 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Pixel | +0.313 |
oise | +0.293 |
archy | +0.273 |
ould | +0.265 |
Maher | +0.264 |
Neg logit contributions | |
uracy | -0.386 |
opin | -0.371 |
ription | -0.369 |
romy | -0.342 |
psons | -0.333 |
Token | basic |
Token ID | 4096 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
Pixel | +0.258 |
oise | +0.241 |
archy | +0.224 |
ould | +0.219 |
Maher | +0.217 |
Neg logit contributions | |
uracy | -0.314 |
opin | -0.303 |
ription | -0.300 |
romy | -0.278 |
psons | -0.272 |
Token | indec |
Token ID | 29077 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Pixel | +0.478 |
oise | +0.445 |
archy | +0.410 |
ould | +0.400 |
Maher | +0.396 |
Neg logit contributions | |
uracy | -0.641 |
opin | -0.619 |
ription | -0.614 |
romy | -0.570 |
psons | -0.557 |
Token | ency |
Token ID | 1387 |
Feature activation | +7.14e-01 |
Pos logit contributions | |
uracy | +3.669 |
ription | +3.479 |
opin | +3.383 |
romy | +3.069 |
psons | +2.995 |
Neg logit contributions | |
Pixel | -4.512 |
oise | -4.205 |
Maher | -3.969 |
archy | -3.958 |
ould | -3.920 |
Token | and |
Token ID | 290 |
Feature activation | +8.04e-02 |
Pos logit contributions | |
uracy | +1.847 |
opin | +1.777 |
ription | +1.761 |
romy | +1.627 |
inval | +1.590 |
Neg logit contributions | |
Pixel | -1.585 |
oise | -1.483 |
archy | -1.374 |
ould | -1.349 |
Maher | -1.335 |
Token | call |
Token ID | 869 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
uracy | +0.209 |
opin | +0.201 |
ription | +0.200 |
romy | +0.185 |
psons | +0.180 |
Neg logit contributions | |
Pixel | -0.177 |
oise | -0.166 |
archy | -0.154 |
ould | -0.151 |
Maher | -0.150 |
Token | ous |
Token ID | 516 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
uracy | +1.146 |
opin | +1.080 |
ription | +1.060 |
romy | +0.942 |
inval | +0.912 |
Neg logit contributions | |
Pixel | -1.981 |
oise | -1.865 |
archy | -1.771 |
ould | -1.751 |
Maher | -1.744 |
Token | ness |
Token ID | 1108 |
Feature activation | +6.39e-01 |
Pos logit contributions | |
uracy | +0.078 |
opin | +0.075 |
ription | +0.075 |
romy | +0.068 |
psons | +0.067 |
Neg logit contributions | |
Pixel | -0.076 |
oise | -0.072 |
archy | -0.067 |
ould | -0.066 |
Maher | -0.065 |
Token | players |
Token ID | 1938 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
Pixel | +0.831 |
oise | +0.778 |
archy | +0.732 |
ould | +0.712 |
Maher | +0.708 |
Neg logit contributions | |
uracy | -0.900 |
ription | -0.862 |
opin | -0.862 |
romy | -0.789 |
psons | -0.775 |
Token | ? |
Token ID | 30 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
Pixel | +0.019 |
oise | +0.018 |
archy | +0.017 |
ould | +0.016 |
Maher | +0.016 |
Neg logit contributions | |
uracy | -0.020 |
opin | -0.020 |
ription | -0.019 |
romy | -0.018 |
psons | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
Pixel | +0.515 |
oise | +0.485 |
archy | +0.453 |
ould | +0.444 |
Maher | +0.441 |
Neg logit contributions | |
uracy | -0.507 |
opin | -0.486 |
ription | -0.482 |
romy | -0.441 |
psons | -0.430 |
Token |
|
Token ID | 198 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
Pixel | +1.047 |
oise | +0.983 |
archy | +0.907 |
Maher | +0.888 |
ould | +0.887 |
Neg logit contributions | |
uracy | -1.155 |
opin | -1.126 |
ription | -1.123 |
romy | -1.022 |
psons | -1.008 |
Token | D |
Token ID | 35 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Pixel | +1.008 |
oise | +0.927 |
archy | +0.866 |
ould | +0.847 |
Maher | +0.827 |
Neg logit contributions | |
uracy | -1.112 |
opin | -1.082 |
ription | -1.076 |
romy | -0.977 |
psons | -0.965 |
Token | read |
Token ID | 961 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
uracy | +0.472 |
opin | +0.455 |
ription | +0.452 |
romy | +0.418 |
psons | +0.409 |
Neg logit contributions | |
Pixel | -0.374 |
oise | -0.349 |
archy | -0.323 |
ould | -0.316 |
Maher | -0.312 |
Token | will |
Token ID | 481 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
uracy | +4.435 |
opin | +4.264 |
ription | +4.215 |
romy | +3.942 |
inval | +3.856 |
Neg logit contributions | |
Pixel | -3.564 |
oise | -3.333 |
archy | -3.022 |
Maher | -2.976 |
ould | -2.939 |
Token | only |
Token ID | 691 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
Pixel | +0.105 |
oise | +0.099 |
archy | +0.092 |
ould | +0.090 |
Maher | +0.089 |
Neg logit contributions | |
uracy | -0.123 |
opin | -0.118 |
ription | -0.118 |
romy | -0.108 |
psons | -0.106 |
Token | really |
Token ID | 1107 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
Pixel | +0.764 |
oise | +0.722 |
archy | +0.667 |
ould | +0.655 |
Maher | +0.643 |
Neg logit contributions | |
uracy | -0.980 |
ription | -0.943 |
opin | -0.943 |
romy | -0.872 |
psons | -0.859 |
Token | increase |
Token ID | 2620 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
uracy | +1.447 |
opin | +1.393 |
ription | +1.381 |
romy | +1.278 |
inval | +1.249 |
Neg logit contributions | |
Pixel | -1.209 |
oise | -1.130 |
archy | -1.047 |
ould | -1.024 |
Maher | -1.016 |
Token | if |
Token ID | 611 |
Feature activation | +7.60e-02 |
Pos logit contributions | |
Pixel | +0.131 |
oise | +0.121 |
archy | +0.113 |
ould | +0.110 |
Maher | +0.110 |
Neg logit contributions | |
uracy | -0.163 |
ription | -0.157 |
opin | -0.156 |
romy | -0.145 |
psons | -0.142 |
Token | residential |
Token ID | 12420 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
Pixel | +0.614 |
oise | +0.574 |
archy | +0.533 |
ould | +0.522 |
Maher | +0.519 |
Neg logit contributions | |
uracy | -0.747 |
opin | -0.715 |
ription | -0.714 |
romy | -0.660 |
psons | -0.646 |
Token | entrance |
Token ID | 10384 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
Pixel | +0.286 |
oise | +0.266 |
archy | +0.246 |
ould | +0.240 |
Maher | +0.238 |
Neg logit contributions | |
uracy | -0.384 |
opin | -0.369 |
ription | -0.368 |
romy | -0.341 |
psons | -0.334 |
Token | along |
Token ID | 1863 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
Pixel | +1.058 |
oise | +0.996 |
archy | +0.924 |
ould | +0.899 |
Maher | +0.893 |
Neg logit contributions | |
uracy | -1.316 |
opin | -1.257 |
ription | -1.250 |
romy | -1.163 |
psons | -1.140 |
Token | Cow |
Token ID | 10417 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
uracy | +0.914 |
opin | +0.881 |
ription | +0.873 |
romy | +0.803 |
psons | +0.783 |
Neg logit contributions | |
Pixel | -0.852 |
oise | -0.798 |
archy | -0.745 |
ould | -0.728 |
Maher | -0.717 |
Token | bell |
Token ID | 7923 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
Pixel | +0.822 |
oise | +0.793 |
archy | +0.758 |
ould | +0.752 |
Maher | +0.744 |
Neg logit contributions | |
uracy | -0.295 |
opin | -0.279 |
ription | -0.269 |
romy | -0.222 |
inval | -0.220 |
Token | Lane |
Token ID | 15016 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
Pixel | +0.933 |
oise | +0.850 |
archy | +0.760 |
ould | +0.736 |
Maher | +0.727 |
Neg logit contributions | |
uracy | -1.931 |
opin | -1.873 |
ription | -1.861 |
romy | -1.750 |
psons | -1.717 |
Token | , |
Token ID | 11 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
Pixel | +4.290 |
oise | +3.989 |
archy | +3.670 |
Maher | +3.591 |
ould | +3.574 |
Neg logit contributions | |
uracy | -5.766 |
opin | -5.465 |
ription | -5.435 |
romy | -5.091 |
psons | -4.962 |
Token | which |
Token ID | 543 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
Pixel | +1.171 |
oise | +1.095 |
archy | +1.026 |
Maher | +1.004 |
ould | +0.999 |
Neg logit contributions | |
uracy | -1.357 |
opin | -1.295 |
ription | -1.294 |
romy | -1.194 |
psons | -1.168 |
Token | they |
Token ID | 484 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Pixel | +1.092 |
oise | +1.018 |
archy | +0.941 |
Maher | +0.922 |
ould | +0.919 |
Neg logit contributions | |
uracy | -1.434 |
ription | -1.375 |
opin | -1.370 |
romy | -1.272 |
psons | -1.242 |
Token | claimed |
Token ID | 4752 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
uracy | +0.415 |
opin | +0.398 |
ription | +0.396 |
romy | +0.365 |
psons | +0.357 |
Neg logit contributions | |
Pixel | -0.371 |
oise | -0.347 |
archy | -0.324 |
ould | -0.317 |
Maher | -0.315 |
Token | was |
Token ID | 373 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
uracy | +0.715 |
opin | +0.691 |
ription | +0.685 |
romy | +0.635 |
inval | +0.623 |
Neg logit contributions | |
Pixel | -0.562 |
oise | -0.524 |
archy | -0.484 |
ould | -0.473 |
Maher | -0.467 |
Token | Hide |
Token ID | 10415 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
Pixel | +0.602 |
oise | +0.567 |
archy | +0.528 |
Maher | +0.518 |
ould | +0.518 |
Neg logit contributions | |
uracy | -0.602 |
opin | -0.578 |
ription | -0.574 |
romy | -0.527 |
inval | -0.513 |
Token | Caption |
Token ID | 11260 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
Pixel | +0.537 |
oise | +0.465 |
archy | +0.394 |
ould | +0.376 |
Maher | +0.372 |
Neg logit contributions | |
uracy | -1.875 |
opin | -1.824 |
ription | -1.814 |
romy | -1.736 |
inval | -1.694 |
Token | 4 |
Token ID | 604 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
Pixel | +2.386 |
oise | +2.221 |
Maher | +2.036 |
archy | +1.996 |
Pixel | +1.978 |
Neg logit contributions | |
uracy | -3.154 |
ription | -3.062 |
opin | -3.061 |
romy | -2.850 |
inval | -2.770 |
Token | of |
Token ID | 286 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
Pixel | +0.816 |
oise | +0.762 |
archy | +0.706 |
ould | +0.688 |
Maher | +0.686 |
Neg logit contributions | |
uracy | -0.991 |
opin | -0.959 |
ription | -0.950 |
romy | -0.880 |
inval | -0.862 |
Token | 15 |
Token ID | 1315 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
Pixel | +1.620 |
oise | +1.508 |
Maher | +1.388 |
archy | +1.372 |
ould | +1.351 |
Neg logit contributions | |
uracy | -2.099 |
ription | -2.040 |
opin | -2.027 |
romy | -1.880 |
inval | -1.830 |
Token | Photos |
Token ID | 9434 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
Pixel | +0.889 |
oise | +0.800 |
archy | +0.704 |
Maher | +0.679 |
ould | +0.658 |
Neg logit contributions | |
uracy | -2.374 |
opin | -2.317 |
ription | -2.294 |
romy | -2.172 |
inval | -2.146 |
Token | : |
Token ID | 25 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
Pixel | +3.882 |
oise | +3.536 |
archy | +3.285 |
Maher | +3.263 |
Pixel | +3.178 |
Neg logit contributions | |
opin | -5.784 |
ription | -5.737 |
uracy | -5.685 |
romy | -5.298 |
obser | -5.287 |
Token | Carly |
Token ID | 35982 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
Pixel | +1.741 |
oise | +1.612 |
Maher | +1.498 |
archy | +1.489 |
ould | +1.429 |
Neg logit contributions | |
uracy | -2.550 |
opin | -2.473 |
ription | -2.458 |
romy | -2.298 |
inval | -2.256 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
Pixel | +1.937 |
oise | +1.668 |
archy | +1.482 |
Maher | +1.461 |
ould | +1.388 |
Neg logit contributions | |
uracy | -5.347 |
opin | -5.268 |
ription | -5.211 |
romy | -4.967 |
psons | -4.847 |
Token | 's |
Token ID | 338 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
Pixel | +3.457 |
oise | +3.228 |
archy | +3.048 |
Maher | +3.040 |
ould | +2.936 |
Neg logit contributions | |
uracy | -4.157 |
ription | -3.986 |
opin | -3.864 |
romy | -3.663 |
psons | -3.586 |
Token | political |
Token ID | 1964 |
Feature activation | -9.36e-01 |
Pos logit contributions | |
Pixel | +1.156 |
oise | +1.093 |
archy | +1.016 |
Maher | +0.993 |
ould | +0.989 |
Neg logit contributions | |
uracy | -1.370 |
ription | -1.309 |
opin | -1.306 |
romy | -1.212 |
psons | -1.184 |
Token | 80 |
Token ID | 4019 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
Pixel | +0.428 |
oise | +0.406 |
archy | +0.370 |
Maher | +0.363 |
ould | +0.363 |
Neg logit contributions | |
uracy | -0.534 |
opin | -0.516 |
ription | -0.515 |
romy | -0.476 |
psons | -0.467 |
Token | minutes |
Token ID | 2431 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
uracy | +0.144 |
opin | +0.139 |
ription | +0.138 |
romy | +0.128 |
psons | +0.125 |
Neg logit contributions | |
Pixel | -0.117 |
oise | -0.110 |
archy | -0.101 |
ould | -0.099 |
Maher | -0.098 |
Token | at |
Token ID | 379 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
uracy | +1.673 |
opin | +1.607 |
ription | +1.596 |
romy | +1.467 |
psons | +1.430 |
Neg logit contributions | |
Pixel | -1.602 |
oise | -1.507 |
archy | -1.405 |
ould | -1.377 |
Maher | -1.367 |
Token | Wel |
Token ID | 22268 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
uracy | +0.305 |
opin | +0.294 |
ription | +0.292 |
romy | +0.269 |
psons | +0.262 |
Neg logit contributions | |
Pixel | -0.268 |
oise | -0.251 |
archy | -0.232 |
ould | -0.228 |
Maher | -0.228 |
Token | ford |
Token ID | 3841 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
Pixel | +0.759 |
oise | +0.718 |
archy | +0.675 |
ould | +0.664 |
Maher | +0.645 |
Neg logit contributions | |
uracy | -0.704 |
opin | -0.685 |
ription | -0.675 |
romy | -0.612 |
inval | -0.603 |
Token | Road |
Token ID | 5567 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
Pixel | +0.936 |
oise | +0.868 |
archy | +0.768 |
ould | +0.763 |
Maher | +0.741 |
Neg logit contributions | |
uracy | -1.867 |
opin | -1.819 |
ription | -1.793 |
romy | -1.683 |
inval | -1.655 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
Pixel | +4.134 |
oise | +3.946 |
ould | +3.671 |
archy | +3.605 |
Maher | +3.549 |
Neg logit contributions | |
uracy | -4.867 |
ription | -4.798 |
opin | -4.794 |
romy | -4.396 |
psons | -4.336 |
Token |
|
Token ID | 198 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
Pixel | +0.039 |
oise | +0.036 |
archy | +0.034 |
ould | +0.033 |
Maher | +0.033 |
Neg logit contributions | |
uracy | -0.039 |
opin | -0.038 |
ription | -0.038 |
romy | -0.035 |
psons | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Pixel | +0.064 |
oise | +0.061 |
archy | +0.056 |
ould | +0.055 |
Maher | +0.055 |
Neg logit contributions | |
uracy | -0.066 |
opin | -0.063 |
ription | -0.063 |
romy | -0.058 |
psons | -0.056 |
Token | � |
Token ID | 447 |
Feature activation | -9.62e-02 |
Pos logit contributions | |
Pixel | +0.314 |
oise | +0.297 |
archy | +0.280 |
ould | +0.276 |
Maher | +0.273 |
Neg logit contributions | |
uracy | -0.228 |
opin | -0.219 |
ription | -0.217 |
romy | -0.195 |
psons | -0.189 |
Token | � |
Token ID | 250 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
Pixel | +0.177 |
oise | +0.165 |
archy | +0.150 |
ould | +0.146 |
Maher | +0.145 |
Neg logit contributions | |
uracy | -0.284 |
opin | -0.275 |
ription | -0.272 |
romy | -0.254 |
inval | -0.250 |
Token | days |
Token ID | 1528 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
uracy | +0.721 |
opin | +0.698 |
ription | +0.691 |
romy | +0.646 |
psons | +0.634 |
Neg logit contributions | |
Pixel | -0.497 |
oise | -0.459 |
archy | -0.422 |
ould | -0.413 |
Maher | -0.408 |
Token | . |
Token ID | 13 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
uracy | +0.548 |
opin | +0.525 |
ription | +0.524 |
romy | +0.482 |
psons | +0.469 |
Neg logit contributions | |
Pixel | -0.490 |
oise | -0.460 |
archy | -0.429 |
ould | -0.420 |
Maher | -0.418 |
Token |
|
Token ID | 198 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
Pixel | +0.201 |
oise | +0.189 |
archy | +0.175 |
Maher | +0.174 |
ould | +0.172 |
Neg logit contributions | |
uracy | -0.198 |
opin | -0.191 |
ription | -0.190 |
romy | -0.174 |
inval | -0.169 |
Token |
|
Token ID | 198 |
Feature activation | -3.59e-03 |
Pos logit contributions | |
uracy | +0.145 |
opin | +0.140 |
ription | +0.139 |
romy | +0.127 |
psons | +0.125 |
Neg logit contributions | |
Pixel | -0.141 |
oise | -0.133 |
archy | -0.124 |
ould | -0.121 |
Maher | -0.121 |
Token | If |
Token ID | 1532 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
Pixel | +0.009 |
oise | +0.008 |
archy | +0.007 |
ould | +0.007 |
Maher | +0.007 |
Neg logit contributions | |
uracy | -0.009 |
opin | -0.009 |
ription | -0.009 |
romy | -0.008 |
inval | -0.008 |
Token | bail |
Token ID | 12274 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
Pixel | +0.046 |
oise | +0.044 |
archy | +0.040 |
Maher | +0.040 |
ould | +0.040 |
Neg logit contributions | |
uracy | -0.050 |
ription | -0.048 |
opin | -0.048 |
romy | -0.044 |
psons | -0.043 |
Token | is |
Token ID | 318 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
Pixel | +3.759 |
oise | +3.548 |
ould | +3.248 |
archy | +3.239 |
Maher | +3.226 |
Neg logit contributions | |
uracy | -5.412 |
ription | -5.218 |
opin | -5.175 |
romy | -4.799 |
psons | -4.669 |
Token | granted |
Token ID | 7520 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
Pixel | +0.558 |
oise | +0.528 |
archy | +0.492 |
ould | +0.485 |
Maher | +0.484 |
Neg logit contributions | |
uracy | -0.573 |
ription | -0.546 |
opin | -0.545 |
romy | -0.498 |
psons | -0.485 |
Token | for |
Token ID | 329 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
Pixel | +0.443 |
oise | +0.415 |
archy | +0.385 |
Maher | +0.384 |
ould | +0.381 |
Neg logit contributions | |
uracy | -0.508 |
opin | -0.488 |
ription | -0.487 |
romy | -0.447 |
inval | -0.434 |
Token | the |
Token ID | 262 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
Pixel | +0.418 |
oise | +0.393 |
archy | +0.366 |
Maher | +0.364 |
ould | +0.361 |
Neg logit contributions | |
uracy | -0.418 |
ription | -0.401 |
opin | -0.400 |
romy | -0.365 |
psons | -0.355 |
Token | n |
Token ID | 299 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
uracy | +0.675 |
opin | +0.650 |
ription | +0.647 |
romy | +0.601 |
psons | +0.587 |
Neg logit contributions | |
Pixel | -0.497 |
oise | -0.463 |
archy | -0.427 |
ould | -0.417 |
Maher | -0.414 |
Token | le |
Token ID | 293 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
Pixel | +0.621 |
oise | +0.568 |
archy | +0.511 |
ould | +0.496 |
Maher | +0.490 |
Neg logit contributions | |
uracy | -1.198 |
opin | -1.161 |
ription | -1.153 |
romy | -1.083 |
psons | -1.062 |
Token | k |
Token ID | 74 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Pixel | +0.649 |
oise | +0.595 |
archy | +0.537 |
ould | +0.523 |
Maher | +0.517 |
Neg logit contributions | |
uracy | -1.192 |
opin | -1.156 |
ription | -1.149 |
romy | -1.077 |
inval | -1.055 |
Token | ( |
Token ID | 357 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
uracy | +0.389 |
opin | +0.375 |
ription | +0.373 |
romy | +0.347 |
psons | +0.339 |
Neg logit contributions | |
Pixel | -0.263 |
oise | -0.244 |
archy | -0.224 |
ould | -0.217 |
Maher | -0.215 |
Token | O |
Token ID | 46 |
Feature activation | -7.88e-03 |
Pos logit contributions | |
uracy | +0.678 |
opin | +0.656 |
ription | +0.651 |
romy | +0.602 |
psons | +0.591 |
Neg logit contributions | |
Pixel | -0.565 |
oise | -0.525 |
archy | -0.484 |
ould | -0.476 |
Maher | -0.470 |
Token | mega |
Token ID | 13731 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
Pixel | +0.014 |
oise | +0.013 |
archy | +0.012 |
ould | +0.012 |
Maher | +0.012 |
Neg logit contributions | |
uracy | -0.023 |
opin | -0.023 |
ription | -0.023 |
romy | -0.021 |
inval | -0.021 |
Token | Pharma |
Token ID | 41845 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
Pixel | +1.563 |
oise | +1.429 |
archy | +1.295 |
ould | +1.294 |
Maher | +1.272 |
Neg logit contributions | |
uracy | -2.807 |
opin | -2.750 |
ription | -2.684 |
romy | -2.513 |
psons | -2.499 |
Token | - |
Token ID | 532 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
Pixel | +3.975 |
oise | +3.746 |
archy | +3.460 |
ould | +3.460 |
Maher | +3.410 |
Neg logit contributions | |
uracy | -4.708 |
opin | -4.551 |
ription | -4.492 |
romy | -4.248 |
psons | -4.214 |
Token | Quick |
Token ID | 12029 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
uracy | +0.666 |
opin | +0.638 |
ription | +0.633 |
romy | +0.578 |
psons | +0.562 |
Neg logit contributions | |
Pixel | -0.725 |
oise | -0.684 |
archy | -0.642 |
ould | -0.629 |
Maher | -0.624 |
Token | step |
Token ID | 9662 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
Pixel | +0.674 |
oise | +0.638 |
archy | +0.595 |
ould | +0.584 |
Maher | +0.577 |
Neg logit contributions | |
uracy | -0.632 |
opin | -0.608 |
ription | -0.602 |
romy | -0.549 |
inval | -0.538 |
Token | ) |
Token ID | 8 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Pixel | +1.866 |
oise | +1.744 |
archy | +1.557 |
Maher | +1.530 |
ould | +1.520 |
Neg logit contributions | |
uracy | -2.904 |
opin | -2.889 |
ription | -2.880 |
romy | -2.639 |
obser | -2.628 |
Token | s |
Token ID | 264 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
uracy | +0.434 |
opin | +0.417 |
ription | +0.412 |
romy | +0.378 |
inval | +0.369 |
Neg logit contributions | |
Pixel | -0.430 |
oise | -0.404 |
archy | -0.379 |
ould | -0.369 |
Maher | -0.365 |
Token | let |
Token ID | 1309 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
Pixel | +0.099 |
oise | +0.093 |
archy | +0.086 |
ould | +0.084 |
Maher | +0.083 |
Neg logit contributions | |
uracy | -0.127 |
opin | -0.122 |
ription | -0.122 |
romy | -0.113 |
psons | -0.111 |
Token | the |
Token ID | 262 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
Pixel | +0.516 |
oise | +0.484 |
archy | +0.451 |
ould | +0.440 |
Maher | +0.436 |
Neg logit contributions | |
uracy | -0.633 |
ription | -0.606 |
opin | -0.605 |
romy | -0.558 |
psons | -0.546 |
Token | future |
Token ID | 2003 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Pixel | +0.220 |
oise | +0.205 |
archy | +0.189 |
ould | +0.185 |
Maher | +0.184 |
Neg logit contributions | |
uracy | -0.276 |
opin | -0.266 |
ription | -0.264 |
romy | -0.245 |
psons | -0.239 |
Token | be |
Token ID | 307 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
uracy | +0.477 |
opin | +0.459 |
ription | +0.456 |
romy | +0.422 |
psons | +0.412 |
Neg logit contributions | |
Pixel | -0.386 |
oise | -0.361 |
archy | -0.335 |
ould | -0.327 |
Maher | -0.324 |
Token | held |
Token ID | 2714 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
Pixel | +0.666 |
oise | +0.627 |
archy | +0.587 |
ould | +0.573 |
Maher | +0.571 |
Neg logit contributions | |
uracy | -0.706 |
ription | -0.676 |
opin | -0.674 |
romy | -0.622 |
psons | -0.607 |
Token | hostage |
Token ID | 23229 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
Pixel | +2.836 |
oise | +2.632 |
archy | +2.455 |
Maher | +2.382 |
ould | +2.374 |
Neg logit contributions | |
uracy | -3.977 |
ription | -3.830 |
opin | -3.763 |
romy | -3.564 |
psons | -3.478 |
Token | by |
Token ID | 416 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
Pixel | +3.781 |
oise | +3.563 |
archy | +3.324 |
ould | +3.237 |
Maher | +3.188 |
Neg logit contributions | |
uracy | -4.879 |
opin | -4.704 |
ription | -4.702 |
romy | -4.408 |
psons | -4.278 |
Token | the |
Token ID | 262 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
Pixel | +0.592 |
oise | +0.555 |
archy | +0.515 |
ould | +0.503 |
Maher | +0.501 |
Neg logit contributions | |
uracy | -0.714 |
opin | -0.685 |
ription | -0.683 |
romy | -0.632 |
psons | -0.617 |
Token | failures |
Token ID | 15536 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
Pixel | +0.558 |
oise | +0.521 |
archy | +0.482 |
ould | +0.471 |
Maher | +0.467 |
Neg logit contributions | |
uracy | -0.721 |
opin | -0.693 |
ription | -0.690 |
romy | -0.640 |
psons | -0.625 |
Token | of |
Token ID | 286 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
uracy | +1.088 |
opin | +1.048 |
ription | +1.038 |
romy | +0.954 |
psons | +0.934 |
Neg logit contributions | |
Pixel | -0.975 |
oise | -0.914 |
archy | -0.848 |
ould | -0.830 |
Maher | -0.823 |
Token | the |
Token ID | 262 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
Pixel | +0.222 |
oise | +0.206 |
archy | +0.190 |
ould | +0.186 |
Maher | +0.184 |
Neg logit contributions | |
uracy | -0.304 |
opin | -0.293 |
ription | -0.291 |
romy | -0.271 |
psons | -0.265 |
Token | at |
Token ID | 379 |
Feature activation | -9.53e-02 |
Pos logit contributions | |
Pixel | +0.831 |
oise | +0.765 |
archy | +0.703 |
ould | +0.679 |
Maher | +0.678 |
Neg logit contributions | |
uracy | -1.413 |
opin | -1.362 |
ription | -1.357 |
romy | -1.262 |
psons | -1.241 |
Token | 3 |
Token ID | 513 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
Pixel | +0.208 |
oise | +0.197 |
archy | +0.181 |
ould | +0.179 |
Maher | +0.178 |
Neg logit contributions | |
uracy | -0.248 |
opin | -0.239 |
ription | -0.236 |
romy | -0.218 |
inval | -0.217 |
Token | 184 |
Token ID | 22883 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
Pixel | +0.145 |
oise | +0.136 |
archy | +0.126 |
ould | +0.124 |
Maher | +0.123 |
Neg logit contributions | |
uracy | -0.158 |
opin | -0.153 |
ription | -0.151 |
romy | -0.140 |
inval | -0.136 |
Token | Ocean |
Token ID | 10692 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
Pixel | +0.042 |
oise | +0.039 |
archy | +0.036 |
Maher | +0.036 |
ould | +0.036 |
Neg logit contributions | |
uracy | -0.049 |
opin | -0.048 |
ription | -0.047 |
romy | -0.044 |
inval | -0.043 |
Token | Beach |
Token ID | 8511 |
Feature activation | -8.99e-01 |
Pos logit contributions | |
Pixel | +0.225 |
oise | +0.209 |
archy | +0.189 |
ould | +0.185 |
Maher | +0.184 |
Neg logit contributions | |
uracy | -0.366 |
opin | -0.354 |
ription | -0.351 |
romy | -0.327 |
inval | -0.322 |
Token | Highway |
Token ID | 14914 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
Pixel | +1.482 |
oise | +1.372 |
archy | +1.222 |
Maher | +1.204 |
ould | +1.199 |
Neg logit contributions | |
uracy | -2.818 |
opin | -2.721 |
ription | -2.688 |
romy | -2.518 |
psons | -2.483 |
Token | with |
Token ID | 351 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
Pixel | +3.666 |
oise | +3.468 |
Maher | +3.253 |
Campus | +3.182 |
archy | +3.164 |
Neg logit contributions | |
uracy | -4.923 |
opin | -4.716 |
ription | -4.637 |
romy | -4.342 |
psons | -4.200 |
Token | the |
Token ID | 262 |
Feature activation | +5.74e-03 |
Pos logit contributions | |
uracy | +0.148 |
opin | +0.142 |
ription | +0.141 |
romy | +0.130 |
psons | +0.126 |
Neg logit contributions | |
Pixel | -0.141 |
oise | -0.133 |
archy | -0.124 |
ould | -0.121 |
Maher | -0.120 |
Token | stolen |
Token ID | 9909 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
uracy | +0.012 |
opin | +0.011 |
ription | +0.011 |
romy | +0.010 |
psons | +0.010 |
Neg logit contributions | |
Pixel | -0.016 |
oise | -0.015 |
archy | -0.014 |
Maher | -0.014 |
ould | -0.014 |
Token | items |
Token ID | 3709 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
Pixel | +0.363 |
oise | +0.343 |
archy | +0.316 |
Maher | +0.311 |
ould | +0.308 |
Neg logit contributions | |
uracy | -0.449 |
opin | -0.431 |
ription | -0.430 |
romy | -0.396 |
inval | -0.386 |
Token | . |
Token ID | 13 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
uracy | +0.942 |
opin | +0.904 |
ription | +0.897 |
romy | +0.824 |
psons | +0.802 |
Neg logit contributions | |
Pixel | -0.934 |
oise | -0.879 |
archy | -0.821 |
ould | -0.805 |
Maher | -0.800 |
Token | Trent |
Token ID | 24269 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
Pixel | +1.019 |
oise | +0.958 |
archy | +0.898 |
ould | +0.886 |
Maher | +0.886 |
Neg logit contributions | |
uracy | -0.926 |
opin | -0.891 |
ription | -0.882 |
romy | -0.802 |
psons | -0.782 |
Token | in |
Token ID | 259 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
Pixel | +1.721 |
oise | +1.650 |
archy | +1.524 |
ould | +1.522 |
Maher | +1.488 |
Neg logit contributions | |
uracy | -2.222 |
ription | -2.137 |
opin | -2.135 |
romy | -2.014 |
inval | -1.962 |
Token | ( |
Token ID | 357 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
uracy | +0.039 |
opin | +0.038 |
ription | +0.038 |
romy | +0.035 |
psons | +0.034 |
Neg logit contributions | |
Pixel | -0.026 |
oise | -0.024 |
archy | -0.022 |
ould | -0.021 |
Maher | -0.021 |
Token | O |
Token ID | 46 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
uracy | +0.752 |
opin | +0.722 |
ription | +0.717 |
romy | +0.659 |
psons | +0.643 |
Neg logit contributions | |
Pixel | -0.716 |
oise | -0.673 |
archy | -0.628 |
ould | -0.615 |
Maher | -0.610 |
Token | mega |
Token ID | 13731 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
uracy | +0.146 |
opin | +0.142 |
ription | +0.141 |
romy | +0.129 |
inval | +0.127 |
Neg logit contributions | |
Pixel | -0.124 |
oise | -0.117 |
archy | -0.108 |
ould | -0.107 |
Maher | -0.105 |
Token | Pharma |
Token ID | 41845 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
Pixel | +1.597 |
oise | +1.474 |
archy | +1.346 |
ould | +1.324 |
Maher | +1.305 |
Neg logit contributions | |
uracy | -2.494 |
opin | -2.419 |
ription | -2.394 |
romy | -2.233 |
psons | -2.201 |
Token | - |
Token ID | 532 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
Pixel | +3.700 |
oise | +3.444 |
ould | +3.316 |
Maher | +3.236 |
archy | +3.218 |
Neg logit contributions | |
uracy | -4.902 |
opin | -4.694 |
ription | -4.619 |
romy | -4.435 |
psons | -4.404 |
Token | Quick |
Token ID | 12029 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
uracy | +0.703 |
opin | +0.674 |
ription | +0.668 |
romy | +0.612 |
psons | +0.595 |
Neg logit contributions | |
Pixel | -0.755 |
oise | -0.712 |
archy | -0.666 |
ould | -0.655 |
Maher | -0.650 |
Token | step |
Token ID | 9662 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
Pixel | +0.611 |
oise | +0.575 |
archy | +0.539 |
ould | +0.528 |
Maher | +0.523 |
Neg logit contributions | |
uracy | -0.646 |
opin | -0.620 |
ription | -0.614 |
romy | -0.566 |
psons | -0.548 |
Token | ) |
Token ID | 8 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
Pixel | +1.888 |
oise | +1.766 |
archy | +1.605 |
ould | +1.580 |
Maher | +1.572 |
Neg logit contributions | |
uracy | -2.739 |
opin | -2.676 |
ription | -2.638 |
romy | -2.462 |
psons | -2.446 |
Token | s |
Token ID | 264 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Pixel | +0.048 |
oise | +0.045 |
archy | +0.042 |
ould | +0.041 |
Maher | +0.041 |
Neg logit contributions | |
uracy | -0.050 |
opin | -0.048 |
ription | -0.047 |
romy | -0.044 |
psons | -0.042 |
Token | Commission |
Token ID | 4513 |
Feature activation | +6.90e-02 |
Pos logit contributions | |
Pixel | +0.027 |
oise | +0.025 |
archy | +0.024 |
Maher | +0.023 |
ould | +0.023 |
Neg logit contributions | |
uracy | -0.024 |
opin | -0.023 |
ription | -0.023 |
romy | -0.021 |
inval | -0.020 |
Token | has |
Token ID | 468 |
Feature activation | +9.27e-02 |
Pos logit contributions | |
uracy | +0.187 |
ription | +0.180 |
opin | +0.180 |
romy | +0.166 |
psons | +0.163 |
Neg logit contributions | |
Pixel | -0.144 |
oise | -0.135 |
archy | -0.124 |
ould | -0.122 |
Maher | -0.121 |
Token | announced |
Token ID | 3414 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
uracy | +0.219 |
ription | +0.209 |
opin | +0.208 |
romy | +0.190 |
psons | +0.186 |
Neg logit contributions | |
Pixel | -0.226 |
oise | -0.214 |
archy | -0.200 |
Maher | -0.197 |
ould | -0.196 |
Token | plans |
Token ID | 3352 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
uracy | +0.417 |
opin | +0.401 |
ription | +0.398 |
romy | +0.368 |
psons | +0.359 |
Neg logit contributions | |
Pixel | -0.351 |
oise | -0.328 |
archy | -0.305 |
ould | -0.298 |
Maher | -0.296 |
Token | to |
Token ID | 284 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
uracy | +0.904 |
opin | +0.867 |
ription | +0.861 |
romy | +0.791 |
inval | +0.770 |
Neg logit contributions | |
Pixel | -0.938 |
oise | -0.887 |
archy | -0.828 |
ould | -0.809 |
Maher | -0.802 |
Token | fine |
Token ID | 3734 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
Pixel | +0.265 |
oise | +0.250 |
archy | +0.232 |
ould | +0.228 |
Maher | +0.225 |
Neg logit contributions | |
uracy | -0.325 |
opin | -0.311 |
ription | -0.310 |
psons | -0.284 |
romy | -0.284 |
Token | a |
Token ID | 257 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Pixel | +3.776 |
oise | +3.446 |
Maher | +3.302 |
Pixel | +3.241 |
archy | +3.144 |
Neg logit contributions | |
uracy | -4.750 |
ription | -4.574 |
opin | -4.497 |
romy | -4.214 |
psons | -4.176 |
Token | mobile |
Token ID | 5175 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
Pixel | +0.225 |
oise | +0.209 |
archy | +0.193 |
ould | +0.188 |
Maher | +0.188 |
Neg logit contributions | |
uracy | -0.303 |
opin | -0.291 |
ription | -0.290 |
romy | -0.269 |
inval | -0.263 |
Token | provider |
Token ID | 10131 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
Pixel | +1.593 |
oise | +1.488 |
archy | +1.376 |
Maher | +1.363 |
Pixel | +1.341 |
Neg logit contributions | |
uracy | -1.773 |
ription | -1.709 |
opin | -1.685 |
romy | -1.554 |
psons | -1.515 |
Token | $ |
Token ID | 720 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
uracy | +1.340 |
opin | +1.286 |
ription | +1.281 |
romy | +1.186 |
psons | +1.157 |
Neg logit contributions | |
Pixel | -1.117 |
oise | -1.043 |
archy | -0.968 |
ould | -0.945 |
Maher | -0.944 |
Token | 51 |
Token ID | 4349 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
Pixel | +0.322 |
oise | +0.297 |
archy | +0.273 |
ould | +0.267 |
Maher | +0.265 |
Neg logit contributions | |
uracy | -0.443 |
opin | -0.431 |
ription | -0.427 |
romy | -0.395 |
inval | -0.390 |
Token | directed |
Token ID | 7924 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
Pixel | +0.329 |
oise | +0.310 |
archy | +0.287 |
ould | +0.282 |
Maher | +0.280 |
Neg logit contributions | |
uracy | -0.383 |
ription | -0.367 |
opin | -0.367 |
romy | -0.336 |
psons | -0.330 |
Token | that |
Token ID | 326 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
Pixel | +0.594 |
oise | +0.555 |
archy | +0.509 |
ould | +0.505 |
Maher | +0.505 |
Neg logit contributions | |
uracy | -0.764 |
ription | -0.733 |
opin | -0.729 |
romy | -0.676 |
psons | -0.665 |
Token | more |
Token ID | 517 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
uracy | +0.902 |
opin | +0.867 |
ription | +0.862 |
romy | +0.797 |
psons | +0.779 |
Neg logit contributions | |
Pixel | -0.746 |
oise | -0.698 |
archy | -0.646 |
ould | -0.634 |
Maher | -0.629 |
Token | work |
Token ID | 670 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Pixel | +0.596 |
oise | +0.554 |
archy | +0.512 |
ould | +0.503 |
Maher | +0.498 |
Neg logit contributions | |
uracy | -0.790 |
ription | -0.761 |
opin | -0.756 |
romy | -0.700 |
psons | -0.689 |
Token | be |
Token ID | 307 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
Pixel | +0.215 |
oise | +0.201 |
archy | +0.184 |
ould | +0.182 |
Maher | +0.177 |
Neg logit contributions | |
uracy | -0.295 |
opin | -0.285 |
ription | -0.284 |
romy | -0.262 |
psons | -0.257 |
Token | put |
Token ID | 1234 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
Pixel | +0.795 |
oise | +0.755 |
archy | +0.707 |
ould | +0.699 |
Maher | +0.686 |
Neg logit contributions | |
uracy | -0.639 |
opin | -0.604 |
ription | -0.603 |
romy | -0.544 |
psons | -0.532 |
Token | into |
Token ID | 656 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
Pixel | +3.948 |
oise | +3.731 |
archy | +3.462 |
ould | +3.398 |
Maher | +3.325 |
Neg logit contributions | |
uracy | -4.574 |
ription | -4.256 |
opin | -4.244 |
romy | -3.873 |
inval | -3.855 |
Token | trying |
Token ID | 2111 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
Pixel | +1.923 |
oise | +1.789 |
archy | +1.641 |
ould | +1.633 |
Maher | +1.631 |
Neg logit contributions | |
uracy | -2.570 |
ription | -2.459 |
opin | -2.456 |
romy | -2.275 |
psons | -2.247 |
Token | to |
Token ID | 284 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
Pixel | +2.012 |
oise | +1.893 |
archy | +1.769 |
ould | +1.735 |
Maher | +1.717 |
Neg logit contributions | |
uracy | -2.127 |
opin | -2.041 |
ription | -2.026 |
romy | -1.858 |
psons | -1.817 |
Token | route |
Token ID | 6339 |
Feature activation | -8.56e-01 |
Pos logit contributions | |
Pixel | +0.445 |
oise | +0.418 |
archy | +0.385 |
ould | +0.377 |
Maher | +0.372 |
Neg logit contributions | |
uracy | -0.595 |
opin | -0.574 |
ription | -0.571 |
romy | -0.528 |
psons | -0.520 |
Token | wider |
Token ID | 10595 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Pixel | +1.783 |
oise | +1.653 |
ould | +1.545 |
Maher | +1.533 |
archy | +1.530 |
Neg logit contributions | |
uracy | -2.308 |
ription | -2.218 |
opin | -2.174 |
romy | -2.006 |
psons | -1.994 |
Token |
|
Token ID | 198 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
uracy | +0.123 |
opin | +0.118 |
ription | +0.117 |
romy | +0.108 |
psons | +0.105 |
Neg logit contributions | |
Pixel | -0.119 |
oise | -0.112 |
archy | -0.104 |
ould | -0.102 |
Maher | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
Pixel | +0.692 |
oise | +0.649 |
archy | +0.601 |
ould | +0.588 |
Maher | +0.586 |
Neg logit contributions | |
uracy | -0.747 |
opin | -0.728 |
ription | -0.724 |
romy | -0.661 |
inval | -0.651 |
Token | Section |
Token ID | 16375 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
Pixel | +0.662 |
oise | +0.617 |
archy | +0.571 |
ould | +0.559 |
Maher | +0.548 |
Neg logit contributions | |
uracy | -0.753 |
opin | -0.729 |
ription | -0.722 |
romy | -0.665 |
psons | -0.651 |
Token | III |
Token ID | 6711 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
uracy | +2.272 |
opin | +2.184 |
ription | +2.173 |
romy | +1.998 |
inval | +1.944 |
Neg logit contributions | |
Pixel | -2.062 |
oise | -1.934 |
archy | -1.797 |
ould | -1.759 |
Maher | -1.746 |
Token | : |
Token ID | 25 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
Pixel | +0.807 |
oise | +0.759 |
archy | +0.708 |
ould | +0.693 |
Maher | +0.685 |
Neg logit contributions | |
uracy | -0.879 |
opin | -0.845 |
ription | -0.835 |
romy | -0.774 |
psons | -0.757 |
Token | Lean |
Token ID | 45661 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
Pixel | +0.502 |
oise | +0.471 |
archy | +0.437 |
ould | +0.427 |
Maher | +0.426 |
Neg logit contributions | |
uracy | -0.574 |
opin | -0.552 |
ription | -0.548 |
romy | -0.507 |
psons | -0.496 |
Token | Six |
Token ID | 9699 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
Pixel | +3.667 |
oise | +3.424 |
archy | +3.208 |
Maher | +3.024 |
ould | +3.022 |
Neg logit contributions | |
uracy | -4.803 |
opin | -4.637 |
ription | -4.602 |
romy | -4.266 |
psons | -4.260 |
Token | Sigma |
Token ID | 31669 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
Pixel | +0.528 |
oise | +0.498 |
archy | +0.465 |
ould | +0.456 |
Maher | +0.453 |
Neg logit contributions | |
uracy | -0.500 |
opin | -0.480 |
ription | -0.475 |
romy | -0.436 |
inval | -0.424 |
Token |
|
Token ID | 198 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Pixel | +1.390 |
oise | +1.299 |
archy | +1.213 |
ould | +1.179 |
Maher | +1.175 |
Neg logit contributions | |
uracy | -1.728 |
opin | -1.657 |
ription | -1.653 |
romy | -1.533 |
psons | -1.509 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Pixel | +0.333 |
oise | +0.314 |
archy | +0.292 |
ould | +0.286 |
Maher | +0.284 |
Neg logit contributions | |
uracy | -0.344 |
opin | -0.332 |
ription | -0.330 |
romy | -0.302 |
psons | -0.295 |
Token | What |
Token ID | 2061 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
Pixel | +0.243 |
oise | +0.228 |
archy | +0.212 |
ould | +0.207 |
Maher | +0.206 |
Neg logit contributions | |
uracy | -0.283 |
opin | -0.272 |
ription | -0.270 |
romy | -0.250 |
psons | -0.244 |
Token | Hayes |
Token ID | 25109 |
Feature activation | -8.99e-01 |
Pos logit contributions | |
Pixel | +0.828 |
oise | +0.777 |
archy | +0.717 |
Maher | +0.704 |
ould | +0.701 |
Neg logit contributions | |
uracy | -1.017 |
opin | -0.981 |
ription | -0.979 |
romy | -0.903 |
psons | -0.884 |
Token | at |
Token ID | 379 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
Pixel | +1.856 |
oise | +1.746 |
archy | +1.604 |
Maher | +1.572 |
ould | +1.571 |
Neg logit contributions | |
uracy | -2.430 |
ription | -2.356 |
opin | -2.353 |
romy | -2.168 |
psons | -2.139 |
Token | nickel |
Token ID | 30666 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
Pixel | +0.323 |
oise | +0.301 |
archy | +0.275 |
ould | +0.268 |
Maher | +0.267 |
Neg logit contributions | |
uracy | -0.473 |
opin | -0.457 |
ription | -0.455 |
romy | -0.423 |
psons | -0.416 |
Token | ( |
Token ID | 357 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
Pixel | +2.338 |
oise | +2.219 |
archy | +2.012 |
ould | +1.971 |
Maher | +1.949 |
Neg logit contributions | |
uracy | -2.895 |
opin | -2.825 |
ription | -2.800 |
romy | -2.602 |
psons | -2.564 |
Token | or |
Token ID | 273 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
Pixel | +0.671 |
oise | +0.627 |
archy | +0.583 |
ould | +0.570 |
Maher | +0.563 |
Neg logit contributions | |
uracy | -0.754 |
opin | -0.730 |
ription | -0.723 |
romy | -0.664 |
psons | -0.650 |
Token | vice |
Token ID | 7927 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Pixel | +1.136 |
oise | +1.078 |
archy | +0.989 |
Maher | +0.976 |
ould | +0.969 |
Neg logit contributions | |
uracy | -1.386 |
ription | -1.340 |
opin | -1.335 |
romy | -1.221 |
psons | -1.213 |
Token | versa |
Token ID | 25470 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Pixel | +3.437 |
oise | +3.309 |
archy | +3.012 |
ould | +2.885 |
Maher | +2.867 |
Neg logit contributions | |
uracy | -4.822 |
ription | -4.707 |
opin | -4.572 |
romy | -4.403 |
psons | -4.295 |
Token | ), |
Token ID | 828 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
uracy | +0.956 |
opin | +0.920 |
ription | +0.913 |
romy | +0.847 |
psons | +0.827 |
Neg logit contributions | |
Pixel | -0.730 |
oise | -0.679 |
archy | -0.627 |
ould | -0.613 |
Maher | -0.607 |
Token | then |
Token ID | 788 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
uracy | +0.107 |
opin | +0.103 |
ription | +0.102 |
romy | +0.094 |
psons | +0.092 |
Neg logit contributions | |
Pixel | -0.093 |
oise | -0.087 |
archy | -0.081 |
ould | -0.079 |
Maher | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
Pixel | +0.094 |
oise | +0.088 |
archy | +0.081 |
ould | +0.080 |
Maher | +0.079 |
Neg logit contributions | |
uracy | -0.111 |
opin | -0.107 |
ription | -0.106 |
romy | -0.098 |
psons | -0.096 |
Token | following |
Token ID | 1708 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Pixel | +0.566 |
oise | +0.532 |
archy | +0.491 |
Maher | +0.479 |
ould | +0.478 |
Neg logit contributions | |
uracy | -0.685 |
ription | -0.659 |
opin | -0.658 |
romy | -0.608 |
psons | -0.595 |
Token | As |
Token ID | 1081 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Pixel | +0.106 |
oise | +0.100 |
archy | +0.093 |
ould | +0.091 |
Maher | +0.091 |
Neg logit contributions | |
uracy | -0.106 |
opin | -0.102 |
ription | -0.101 |
romy | -0.093 |
psons | -0.091 |
Token | well |
Token ID | 880 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
Pixel | +0.563 |
oise | +0.528 |
archy | +0.492 |
ould | +0.481 |
Maher | +0.477 |
Neg logit contributions | |
uracy | -0.651 |
opin | -0.624 |
ription | -0.620 |
romy | -0.574 |
psons | -0.560 |
Token | , |
Token ID | 11 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
Pixel | +0.196 |
oise | +0.183 |
archy | +0.169 |
ould | +0.165 |
Maher | +0.164 |
Neg logit contributions | |
uracy | -0.252 |
opin | -0.243 |
ription | -0.241 |
romy | -0.224 |
inval | -0.218 |
Token | Cow |
Token ID | 10417 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
Pixel | +0.552 |
oise | +0.514 |
archy | +0.477 |
ould | +0.470 |
Maher | +0.467 |
Neg logit contributions | |
uracy | -0.680 |
opin | -0.650 |
ription | -0.647 |
romy | -0.600 |
psons | -0.589 |
Token | bell |
Token ID | 7923 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
Pixel | +0.557 |
oise | +0.525 |
archy | +0.487 |
ould | +0.481 |
Maher | +0.477 |
Neg logit contributions | |
uracy | -0.577 |
opin | -0.552 |
ription | -0.548 |
romy | -0.504 |
psons | -0.493 |
Token | Lane |
Token ID | 15016 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Pixel | +1.034 |
oise | +0.961 |
archy | +0.879 |
ould | +0.864 |
Maher | +0.852 |
Neg logit contributions | |
uracy | -1.467 |
opin | -1.411 |
ription | -1.404 |
romy | -1.307 |
psons | -1.280 |
Token | will |
Token ID | 481 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
Pixel | +3.408 |
oise | +3.147 |
archy | +2.867 |
ould | +2.851 |
Maher | +2.823 |
Neg logit contributions | |
uracy | -5.059 |
ription | -4.833 |
opin | -4.781 |
romy | -4.479 |
psons | -4.404 |
Token | be |
Token ID | 307 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
Pixel | +1.181 |
oise | +1.120 |
archy | +1.025 |
ould | +1.005 |
Maher | +0.984 |
Neg logit contributions | |
uracy | -1.671 |
opin | -1.607 |
ription | -1.604 |
romy | -1.483 |
psons | -1.471 |
Token | completely |
Token ID | 3190 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
Pixel | +0.613 |
oise | +0.578 |
archy | +0.537 |
ould | +0.525 |
Maher | +0.525 |
Neg logit contributions | |
uracy | -0.687 |
ription | -0.656 |
opin | -0.655 |
romy | -0.605 |
psons | -0.592 |
Token | redesigned |
Token ID | 36240 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Pixel | +1.617 |
oise | +1.526 |
archy | +1.414 |
ould | +1.388 |
Maher | +1.381 |
Neg logit contributions | |
uracy | -1.971 |
ription | -1.893 |
opin | -1.858 |
romy | -1.747 |
psons | -1.737 |
Token | , |
Token ID | 11 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
Pixel | +2.777 |
oise | +2.590 |
archy | +2.414 |
Maher | +2.356 |
ould | +2.335 |
Neg logit contributions | |
uracy | -3.401 |
ription | -3.266 |
opin | -3.244 |
romy | -2.996 |
psons | -2.981 |
Token | There |
Token ID | 1858 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
Pixel | +0.867 |
oise | +0.815 |
archy | +0.764 |
ould | +0.748 |
Maher | +0.741 |
Neg logit contributions | |
uracy | -0.900 |
opin | -0.864 |
ription | -0.857 |
romy | -0.790 |
psons | -0.768 |
Token | is |
Token ID | 318 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
uracy | +0.745 |
opin | +0.720 |
ription | +0.714 |
romy | +0.666 |
psons | +0.651 |
Neg logit contributions | |
Pixel | -0.522 |
oise | -0.484 |
archy | -0.444 |
Maher | -0.430 |
ould | -0.429 |
Token | heavy |
Token ID | 4334 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
uracy | +0.411 |
opin | +0.393 |
ription | +0.390 |
romy | +0.358 |
psons | +0.348 |
Neg logit contributions | |
Pixel | -0.431 |
oise | -0.407 |
archy | -0.381 |
ould | -0.374 |
Maher | -0.371 |
Token | police |
Token ID | 1644 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Pixel | +0.941 |
oise | +0.880 |
archy | +0.817 |
ould | +0.800 |
Maher | +0.793 |
Neg logit contributions | |
uracy | -1.161 |
opin | -1.112 |
ription | -1.107 |
romy | -1.022 |
psons | -1.002 |
Token | and |
Token ID | 290 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
Pixel | +0.554 |
oise | +0.521 |
archy | +0.484 |
ould | +0.474 |
Maher | +0.470 |
Neg logit contributions | |
uracy | -0.641 |
opin | -0.614 |
ription | -0.612 |
romy | -0.564 |
psons | -0.554 |
Token | SWAT |
Token ID | 36272 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
uracy | +0.777 |
opin | +0.750 |
ription | +0.745 |
romy | +0.691 |
inval | +0.676 |
Neg logit contributions | |
Pixel | -0.589 |
oise | -0.549 |
archy | -0.506 |
ould | -0.494 |
Maher | -0.489 |
Token | presence |
Token ID | 4931 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
Pixel | +3.634 |
oise | +3.424 |
archy | +3.153 |
ould | +3.123 |
Maher | +3.037 |
Neg logit contributions | |
uracy | -4.650 |
opin | -4.477 |
ription | -4.460 |
romy | -4.098 |
psons | -4.092 |
Token | at |
Token ID | 379 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
Pixel | +0.023 |
oise | +0.021 |
archy | +0.020 |
ould | +0.019 |
Maher | +0.019 |
Neg logit contributions | |
uracy | -0.025 |
opin | -0.024 |
ription | -0.024 |
romy | -0.022 |
psons | -0.022 |
Token | Boston |
Token ID | 6182 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
uracy | +0.256 |
opin | +0.247 |
ription | +0.245 |
romy | +0.227 |
psons | +0.222 |
Neg logit contributions | |
Pixel | -0.200 |
oise | -0.186 |
archy | -0.172 |
ould | -0.168 |
Maher | -0.166 |
Token | Ave |
Token ID | 12761 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
Pixel | +0.690 |
oise | +0.634 |
archy | +0.579 |
ould | +0.564 |
Maher | +0.561 |
Neg logit contributions | |
uracy | -1.129 |
opin | -1.089 |
ription | -1.083 |
romy | -1.012 |
inval | -0.995 |
Token | and |
Token ID | 290 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
Pixel | +1.334 |
oise | +1.248 |
archy | +1.168 |
ould | +1.152 |
Maher | +1.139 |
Neg logit contributions | |
uracy | -1.448 |
opin | -1.389 |
ription | -1.380 |
romy | -1.271 |
psons | -1.242 |
Token | m |
Token ID | 76 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
uracy | +0.173 |
opin | +0.167 |
ription | +0.166 |
romy | +0.155 |
psons | +0.151 |
Neg logit contributions | |
Pixel | -0.120 |
oise | -0.112 |
archy | -0.102 |
ould | -0.100 |
Maher | -0.099 |
Token | ote |
Token ID | 1258 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
uracy | +1.630 |
opin | +1.571 |
ription | +1.569 |
romy | +1.470 |
psons | +1.438 |
Neg logit contributions | |
Pixel | -0.824 |
oise | -0.733 |
archy | -0.671 |
Maher | -0.645 |
ould | -0.638 |
Token | ( |
Token ID | 357 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
Pixel | +1.165 |
oise | +1.080 |
archy | +0.989 |
ould | +0.964 |
Maher | +0.954 |
Neg logit contributions | |
uracy | -1.761 |
opin | -1.702 |
ription | -1.691 |
romy | -1.576 |
psons | -1.543 |
Token | O |
Token ID | 46 |
Feature activation | -4.04e-03 |
Pos logit contributions | |
uracy | +0.701 |
opin | +0.676 |
ription | +0.671 |
romy | +0.621 |
psons | +0.607 |
Neg logit contributions | |
Pixel | -0.568 |
oise | -0.531 |
archy | -0.491 |
ould | -0.480 |
Maher | -0.476 |
Token | mega |
Token ID | 13731 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
Pixel | +0.006 |
oise | +0.006 |
archy | +0.005 |
ould | +0.005 |
Maher | +0.005 |
Neg logit contributions | |
uracy | -0.013 |
opin | -0.013 |
ription | -0.012 |
romy | -0.012 |
inval | -0.011 |
Token | Pharma |
Token ID | 41845 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Pixel | +1.278 |
oise | +1.159 |
archy | +1.044 |
Maher | +1.035 |
ould | +1.025 |
Neg logit contributions | |
uracy | -2.310 |
opin | -2.242 |
ription | -2.220 |
romy | -2.059 |
inval | -2.052 |
Token | - |
Token ID | 532 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
Pixel | +3.710 |
oise | +3.448 |
Maher | +3.171 |
archy | +3.143 |
Pixel | +3.129 |
Neg logit contributions | |
uracy | -4.618 |
ription | -4.550 |
opin | -4.490 |
romy | -4.200 |
psons | -4.166 |
Token | Quick |
Token ID | 12029 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
uracy | +0.699 |
opin | +0.672 |
ription | +0.665 |
romy | +0.612 |
psons | +0.594 |
Neg logit contributions | |
Pixel | -0.728 |
oise | -0.685 |
archy | -0.638 |
Maher | -0.628 |
ould | -0.627 |
Token | step |
Token ID | 9662 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
Pixel | +0.429 |
oise | +0.406 |
archy | +0.379 |
ould | +0.372 |
Maher | +0.367 |
Neg logit contributions | |
uracy | -0.403 |
opin | -0.388 |
ription | -0.384 |
romy | -0.350 |
inval | -0.343 |
Token | ) |
Token ID | 8 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
Pixel | +1.690 |
oise | +1.579 |
archy | +1.413 |
Maher | +1.388 |
ould | +1.381 |
Neg logit contributions | |
uracy | -2.609 |
opin | -2.593 |
ription | -2.589 |
romy | -2.372 |
inval | -2.360 |
Token | s |
Token ID | 264 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
uracy | +0.484 |
opin | +0.465 |
ription | +0.460 |
romy | +0.422 |
inval | +0.413 |
Neg logit contributions | |
Pixel | -0.479 |
oise | -0.451 |
archy | -0.422 |
ould | -0.412 |
Maher | -0.407 |
Token | s |
Token ID | 82 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
Pixel | +0.088 |
oise | +0.082 |
archy | +0.076 |
ould | +0.075 |
Maher | +0.074 |
Neg logit contributions | |
uracy | -0.107 |
opin | -0.103 |
ription | -0.103 |
romy | -0.095 |
psons | -0.093 |
Token | approach |
Token ID | 3164 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
Pixel | +0.035 |
oise | +0.033 |
archy | +0.031 |
ould | +0.030 |
Maher | +0.030 |
Neg logit contributions | |
uracy | -0.045 |
opin | -0.043 |
ription | -0.043 |
romy | -0.040 |
psons | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | +8.70e-02 |
Pos logit contributions | |
Pixel | +1.170 |
oise | +1.097 |
archy | +1.025 |
ould | +1.006 |
Maher | +0.994 |
Neg logit contributions | |
uracy | -1.227 |
opin | -1.179 |
ription | -1.172 |
romy | -1.075 |
psons | -1.049 |
Token | storytelling |
Token ID | 23689 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
uracy | +0.228 |
opin | +0.219 |
ription | +0.218 |
romy | +0.201 |
psons | +0.197 |
Neg logit contributions | |
Pixel | -0.190 |
oise | -0.178 |
archy | -0.165 |
ould | -0.161 |
Maher | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
uracy | +0.085 |
opin | +0.081 |
ription | +0.081 |
romy | +0.074 |
psons | +0.072 |
Neg logit contributions | |
Pixel | -0.086 |
oise | -0.080 |
archy | -0.075 |
ould | -0.074 |
Maher | -0.073 |
Token | Bio |
Token ID | 16024 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Pixel | +0.505 |
oise | +0.473 |
archy | +0.441 |
Maher | +0.436 |
ould | +0.433 |
Neg logit contributions | |
uracy | -0.489 |
opin | -0.469 |
ription | -0.468 |
romy | -0.429 |
inval | -0.419 |
Token | Shock |
Token ID | 31646 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
Pixel | +3.413 |
oise | +3.094 |
archy | +2.847 |
ould | +2.801 |
Maher | +2.727 |
Neg logit contributions | |
uracy | -5.037 |
ription | -4.870 |
opin | -4.838 |
romy | -4.492 |
psons | -4.378 |
Token | is |
Token ID | 318 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
Pixel | +1.304 |
oise | +1.211 |
archy | +1.117 |
ould | +1.106 |
Maher | +1.089 |
Neg logit contributions | |
uracy | -1.566 |
opin | -1.504 |
ription | -1.498 |
romy | -1.385 |
psons | -1.350 |
Token | famous |
Token ID | 5863 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
Pixel | +0.538 |
oise | +0.501 |
archy | +0.470 |
ould | +0.460 |
Maher | +0.459 |
Neg logit contributions | |
uracy | -0.556 |
ription | -0.531 |
opin | -0.529 |
romy | -0.487 |
psons | -0.475 |
Token | for |
Token ID | 329 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
Pixel | +1.226 |
oise | +1.146 |
archy | +1.068 |
ould | +1.048 |
Maher | +1.034 |
Neg logit contributions | |
uracy | -1.407 |
opin | -1.351 |
ription | -1.346 |
romy | -1.239 |
psons | -1.208 |
Token | its |
Token ID | 663 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
Pixel | +0.775 |
oise | +0.724 |
archy | +0.673 |
ould | +0.660 |
Maher | +0.658 |
Neg logit contributions | |
uracy | -0.875 |
opin | -0.840 |
ription | -0.839 |
romy | -0.773 |
psons | -0.754 |
Token | will |
Token ID | 481 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
Pixel | +1.035 |
oise | +0.969 |
archy | +0.898 |
ould | +0.879 |
Maher | +0.879 |
Neg logit contributions | |
uracy | -1.214 |
opin | -1.159 |
ription | -1.158 |
romy | -1.072 |
psons | -1.043 |
Token | potential |
Token ID | 2785 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
Pixel | +0.215 |
oise | +0.202 |
archy | +0.187 |
ould | +0.184 |
Maher | +0.183 |
Neg logit contributions | |
uracy | -0.238 |
opin | -0.229 |
ription | -0.227 |
romy | -0.210 |
psons | -0.204 |
Token | new |
Token ID | 649 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
uracy | +0.701 |
opin | +0.675 |
ription | +0.670 |
romy | +0.621 |
psons | +0.607 |
Neg logit contributions | |
Pixel | -0.556 |
oise | -0.520 |
archy | -0.481 |
ould | -0.470 |
Maher | -0.465 |
Token | entrants |
Token ID | 49117 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
Pixel | +1.590 |
oise | +1.495 |
archy | +1.355 |
ould | +1.339 |
Maher | +1.336 |
Neg logit contributions | |
uracy | -2.223 |
ription | -2.141 |
opin | -2.125 |
romy | -1.993 |
psons | -1.931 |
Token | be |
Token ID | 307 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
Pixel | +1.379 |
oise | +1.301 |
archy | +1.207 |
Maher | +1.184 |
ould | +1.182 |
Neg logit contributions | |
uracy | -1.557 |
opin | -1.479 |
ription | -1.476 |
romy | -1.380 |
psons | -1.334 |
Token | put |
Token ID | 1234 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
Pixel | +0.481 |
oise | +0.453 |
archy | +0.421 |
Maher | +0.415 |
ould | +0.411 |
Neg logit contributions | |
uracy | -0.502 |
opin | -0.478 |
ription | -0.477 |
romy | -0.444 |
psons | -0.429 |
Token | off |
Token ID | 572 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Pixel | +3.518 |
oise | +3.399 |
archy | +3.061 |
Maher | +3.031 |
ould | +2.950 |
Neg logit contributions | |
uracy | -4.783 |
ription | -4.483 |
opin | -4.436 |
romy | -4.162 |
psons | -4.069 |
Token | ? |
Token ID | 30 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
Pixel | +2.532 |
oise | +2.380 |
Maher | +2.153 |
archy | +2.151 |
ould | +2.140 |
Neg logit contributions | |
uracy | -3.132 |
opin | -3.002 |
ription | -2.975 |
romy | -2.836 |
psons | -2.733 |
Token |
|
Token ID | 198 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
Pixel | +0.290 |
oise | +0.269 |
archy | +0.246 |
ould | +0.240 |
Maher | +0.239 |
Neg logit contributions | |
uracy | -0.430 |
opin | -0.415 |
ription | -0.413 |
romy | -0.385 |
psons | -0.377 |
Token |
|
Token ID | 198 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
Pixel | +0.913 |
oise | +0.858 |
archy | +0.793 |
ould | +0.777 |
Maher | +0.775 |
Neg logit contributions | |
uracy | -0.991 |
opin | -0.966 |
ription | -0.960 |
romy | -0.878 |
inval | -0.864 |
Token | Some |
Token ID | 4366 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
Pixel | +0.939 |
oise | +0.869 |
archy | +0.804 |
ould | +0.789 |
Maher | +0.777 |
Neg logit contributions | |
uracy | -1.043 |
opin | -1.015 |
ription | -1.007 |
romy | -0.921 |
psons | -0.905 |
Token | experiment |
Token ID | 6306 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
uracy | +0.156 |
opin | +0.150 |
ription | +0.149 |
romy | +0.138 |
psons | +0.134 |
Neg logit contributions | |
Pixel | -0.141 |
oise | -0.133 |
archy | -0.124 |
ould | -0.122 |
Maher | -0.120 |
Token | . |
Token ID | 13 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
uracy | +0.275 |
opin | +0.263 |
ription | +0.262 |
romy | +0.241 |
psons | +0.234 |
Neg logit contributions | |
Pixel | -0.267 |
oise | -0.251 |
archy | -0.235 |
ould | -0.231 |
Maher | -0.228 |
Token |
|
Token ID | 198 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Pixel | +0.577 |
oise | +0.543 |
archy | +0.508 |
ould | +0.500 |
Maher | +0.499 |
Neg logit contributions | |
uracy | -0.565 |
opin | -0.541 |
ription | -0.539 |
romy | -0.494 |
psons | -0.482 |
Token |
|
Token ID | 198 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
Pixel | +0.355 |
oise | +0.333 |
archy | +0.309 |
ould | +0.303 |
Maher | +0.302 |
Neg logit contributions | |
uracy | -0.372 |
opin | -0.360 |
ription | -0.357 |
romy | -0.327 |
inval | -0.321 |
Token | An |
Token ID | 2025 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
Pixel | +0.431 |
oise | +0.402 |
archy | +0.374 |
ould | +0.367 |
Maher | +0.361 |
Neg logit contributions | |
uracy | -0.488 |
opin | -0.470 |
ription | -0.468 |
romy | -0.430 |
inval | -0.420 |
Token | arch |
Token ID | 998 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Pixel | +0.104 |
oise | +0.086 |
archy | +0.067 |
ould | +0.057 |
Maher | +0.054 |
Neg logit contributions | |
uracy | -0.607 |
opin | -0.592 |
ription | -0.590 |
romy | -0.563 |
psons | -0.554 |
Token | ism |
Token ID | 1042 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
Pixel | +3.456 |
oise | +3.323 |
archy | +3.128 |
ould | +2.987 |
Maher | +2.883 |
Neg logit contributions | |
uracy | -4.695 |
opin | -4.509 |
ription | -4.483 |
romy | -4.155 |
inval | -4.115 |
Token | has |
Token ID | 468 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
Pixel | +0.445 |
oise | +0.416 |
archy | +0.385 |
ould | +0.377 |
Maher | +0.373 |
Neg logit contributions | |
uracy | -0.564 |
opin | -0.542 |
ription | -0.539 |
romy | -0.500 |
psons | -0.488 |
Token | strong |
Token ID | 1913 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
uracy | +0.314 |
opin | +0.300 |
ription | +0.298 |
romy | +0.273 |
psons | +0.266 |
Neg logit contributions | |
Pixel | -0.328 |
oise | -0.309 |
archy | -0.289 |
ould | -0.284 |
Maher | -0.282 |
Token | roots |
Token ID | 11135 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
Pixel | +0.102 |
oise | +0.095 |
archy | +0.088 |
ould | +0.086 |
Maher | +0.085 |
Neg logit contributions | |
uracy | -0.131 |
opin | -0.127 |
ription | -0.126 |
romy | -0.117 |
psons | -0.114 |
Token | in |
Token ID | 287 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
Pixel | +0.555 |
oise | +0.519 |
archy | +0.484 |
ould | +0.475 |
Maher | +0.469 |
Neg logit contributions | |
uracy | -0.657 |
opin | -0.630 |
ription | -0.627 |
romy | -0.580 |
psons | -0.565 |
Token | Champions |
Token ID | 6662 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
Pixel | +0.433 |
oise | +0.401 |
archy | +0.373 |
Maher | +0.364 |
ould | +0.364 |
Neg logit contributions | |
uracy | -0.545 |
opin | -0.525 |
ription | -0.522 |
romy | -0.487 |
psons | -0.471 |
Token | Cup |
Token ID | 5454 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
Pixel | +0.068 |
oise | +0.064 |
archy | +0.061 |
Maher | +0.060 |
ould | +0.060 |
Neg logit contributions | |
uracy | -0.042 |
opin | -0.040 |
ription | -0.039 |
romy | -0.035 |
inval | -0.034 |
Token | at |
Token ID | 379 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
Pixel | +1.422 |
oise | +1.328 |
archy | +1.216 |
ould | +1.203 |
Maher | +1.198 |
Neg logit contributions | |
uracy | -1.715 |
opin | -1.649 |
ription | -1.636 |
romy | -1.523 |
psons | -1.487 |
Token | Wel |
Token ID | 22268 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
uracy | +0.684 |
opin | +0.652 |
ription | +0.646 |
romy | +0.585 |
psons | +0.566 |
Neg logit contributions | |
Pixel | -0.891 |
oise | -0.845 |
archy | -0.795 |
ould | -0.783 |
Maher | -0.778 |
Token | ford |
Token ID | 3841 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
Pixel | +0.476 |
oise | +0.451 |
archy | +0.423 |
ould | +0.416 |
Maher | +0.408 |
Neg logit contributions | |
uracy | -0.431 |
opin | -0.418 |
ription | -0.413 |
romy | -0.374 |
inval | -0.368 |
Token | Road |
Token ID | 5567 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
Pixel | +0.738 |
oise | +0.677 |
archy | +0.591 |
Maher | +0.585 |
ould | +0.579 |
Neg logit contributions | |
uracy | -1.624 |
opin | -1.592 |
ription | -1.569 |
romy | -1.470 |
inval | -1.462 |
Token | , |
Token ID | 11 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
Pixel | +3.572 |
oise | +3.359 |
ould | +3.109 |
Maher | +3.082 |
archy | +3.045 |
Neg logit contributions | |
uracy | -4.490 |
ription | -4.354 |
opin | -4.336 |
romy | -4.036 |
psons | -3.965 |
Token | Saturday |
Token ID | 3909 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
Pixel | +0.438 |
oise | +0.412 |
archy | +0.382 |
Maher | +0.374 |
ould | +0.372 |
Neg logit contributions | |
uracy | -0.526 |
opin | -0.506 |
ription | -0.503 |
romy | -0.467 |
psons | -0.456 |
Token | 21 |
Token ID | 2310 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
Pixel | +1.132 |
oise | +1.065 |
archy | +0.990 |
Maher | +0.968 |
ould | +0.966 |
Neg logit contributions | |
uracy | -1.302 |
ription | -1.255 |
opin | -1.249 |
romy | -1.159 |
psons | -1.125 |
Token | January |
Token ID | 3269 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
Pixel | +0.105 |
oise | +0.100 |
archy | +0.094 |
ould | +0.092 |
Maher | +0.092 |
Neg logit contributions | |
uracy | -0.089 |
opin | -0.085 |
ription | -0.085 |
romy | -0.077 |
psons | -0.075 |
Token | live |
Token ID | 2107 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
Pixel | +0.511 |
oise | +0.479 |
archy | +0.444 |
ould | +0.436 |
Maher | +0.433 |
Neg logit contributions | |
uracy | -0.604 |
opin | -0.582 |
ription | -0.579 |
romy | -0.535 |
psons | -0.522 |
Token |
|
Token ID | 198 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
uracy | +0.093 |
opin | +0.090 |
ription | +0.090 |
romy | +0.082 |
inval | +0.080 |
Neg logit contributions | |
Pixel | -0.090 |
oise | -0.085 |
archy | -0.079 |
ould | -0.077 |
Maher | -0.077 |
Token | She |
Token ID | 3347 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
Pixel | +0.135 |
oise | +0.126 |
archy | +0.117 |
ould | +0.115 |
Maher | +0.114 |
Neg logit contributions | |
uracy | -0.139 |
ription | -0.134 |
opin | -0.134 |
romy | -0.122 |
inval | -0.119 |
Token | confirmed |
Token ID | 4999 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Pixel | +0.204 |
oise | +0.192 |
archy | +0.178 |
ould | +0.174 |
Maher | +0.173 |
Neg logit contributions | |
uracy | -0.246 |
opin | -0.236 |
ription | -0.235 |
romy | -0.218 |
psons | -0.212 |
Token | that |
Token ID | 326 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
uracy | +0.452 |
opin | +0.434 |
ription | +0.433 |
romy | +0.398 |
psons | +0.388 |
Neg logit contributions | |
Pixel | -0.393 |
oise | -0.368 |
archy | -0.342 |
ould | -0.335 |
Maher | -0.334 |
Token | when |
Token ID | 618 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
uracy | +0.892 |
opin | +0.855 |
ription | +0.851 |
romy | +0.782 |
psons | +0.761 |
Neg logit contributions | |
Pixel | -0.829 |
oise | -0.779 |
archy | -0.725 |
ould | -0.711 |
Maher | -0.709 |
Token | bail |
Token ID | 12274 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
Pixel | +0.135 |
oise | +0.127 |
archy | +0.119 |
Maher | +0.118 |
ould | +0.117 |
Neg logit contributions | |
uracy | -0.138 |
ription | -0.132 |
opin | -0.132 |
romy | -0.121 |
psons | -0.117 |
Token | was |
Token ID | 373 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
Pixel | +3.006 |
oise | +2.861 |
archy | +2.565 |
Maher | +2.537 |
ould | +2.529 |
Neg logit contributions | |
uracy | -5.067 |
ription | -4.858 |
opin | -4.843 |
romy | -4.532 |
psons | -4.440 |
Token | originally |
Token ID | 6198 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
uracy | +0.198 |
opin | +0.188 |
ription | +0.187 |
romy | +0.171 |
psons | +0.166 |
Neg logit contributions | |
Pixel | -0.212 |
oise | -0.201 |
archy | -0.188 |
ould | -0.185 |
Maher | -0.185 |
Token | set |
Token ID | 900 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
Pixel | +0.663 |
oise | +0.628 |
archy | +0.587 |
Maher | +0.579 |
ould | +0.576 |
Neg logit contributions | |
uracy | -0.723 |
ription | -0.684 |
opin | -0.683 |
romy | -0.630 |
psons | -0.612 |
Token | , |
Token ID | 11 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
uracy | +0.131 |
opin | +0.126 |
ription | +0.125 |
romy | +0.116 |
psons | +0.112 |
Neg logit contributions | |
Pixel | -0.114 |
oise | -0.106 |
archy | -0.099 |
ould | -0.097 |
Maher | -0.097 |
Token | a |
Token ID | 257 |
Feature activation | -6.05e-02 |
Pos logit contributions | |
uracy | +0.383 |
opin | +0.367 |
ription | +0.365 |
romy | +0.336 |
psons | +0.327 |
Neg logit contributions | |
Pixel | -0.351 |
oise | -0.330 |
archy | -0.307 |
ould | -0.301 |
Maher | -0.301 |
Token | mana |
Token ID | 13149 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
uracy | +0.643 |
opin | +0.618 |
ription | +0.616 |
romy | +0.567 |
psons | +0.554 |
Neg logit contributions | |
Pixel | -0.551 |
oise | -0.517 |
archy | -0.479 |
ould | -0.470 |
Maher | -0.465 |
Token | to |
Token ID | 284 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
uracy | +3.562 |
opin | +3.438 |
ription | +3.392 |
romy | +3.128 |
inval | +3.063 |
Neg logit contributions | |
Pixel | -3.226 |
oise | -3.014 |
archy | -2.820 |
ould | -2.739 |
Maher | -2.723 |
Token | get |
Token ID | 651 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
uracy | +0.114 |
opin | +0.109 |
ription | +0.109 |
romy | +0.100 |
psons | +0.098 |
Neg logit contributions | |
Pixel | -0.094 |
oise | -0.089 |
archy | -0.082 |
ould | -0.081 |
Maher | -0.080 |
Token | a |
Token ID | 257 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
uracy | +1.282 |
opin | +1.230 |
ription | +1.224 |
romy | +1.124 |
psons | +1.098 |
Neg logit contributions | |
Pixel | -1.159 |
oise | -1.089 |
archy | -1.012 |
ould | -0.991 |
Maher | -0.985 |
Token | g |
Token ID | 308 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
uracy | +0.737 |
opin | +0.707 |
ription | +0.705 |
romy | +0.648 |
inval | +0.633 |
Neg logit contributions | |
Pixel | -0.646 |
oise | -0.607 |
archy | -0.562 |
ould | -0.552 |
Maher | -0.547 |
Token | ank |
Token ID | 962 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
uracy | +2.057 |
opin | +1.986 |
ription | +1.971 |
romy | +1.830 |
inval | +1.792 |
Neg logit contributions | |
Pixel | -1.517 |
oise | -1.416 |
archy | -1.306 |
ould | -1.276 |
Maher | -1.258 |
Token | off |
Token ID | 572 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
uracy | +3.471 |
opin | +3.332 |
ription | +3.303 |
romy | +3.043 |
inval | +2.961 |
Neg logit contributions | |
Pixel | -3.255 |
oise | -3.034 |
archy | -2.844 |
ould | -2.779 |
Maher | -2.769 |
Token | , |
Token ID | 11 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
Pixel | +0.628 |
oise | +0.592 |
archy | +0.544 |
ould | +0.539 |
Maher | +0.534 |
Neg logit contributions | |
uracy | -0.716 |
opin | -0.687 |
ription | -0.684 |
romy | -0.628 |
psons | -0.615 |
Token | it |
Token ID | 340 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
uracy | +0.107 |
ription | +0.102 |
opin | +0.102 |
romy | +0.093 |
psons | +0.091 |
Neg logit contributions | |
Pixel | -0.105 |
oise | -0.099 |
archy | -0.092 |
ould | -0.091 |
Maher | -0.091 |
Token | gets |
Token ID | 3011 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
Pixel | +0.322 |
oise | +0.303 |
archy | +0.280 |
ould | +0.274 |
Maher | +0.271 |
Neg logit contributions | |
uracy | -0.405 |
opin | -0.390 |
ription | -0.389 |
romy | -0.361 |
psons | -0.350 |
Token | my |
Token ID | 616 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
uracy | +0.961 |
opin | +0.923 |
ription | +0.919 |
romy | +0.846 |
psons | +0.827 |
Neg logit contributions | |
Pixel | -0.829 |
oise | -0.778 |
archy | -0.722 |
ould | -0.706 |
Maher | -0.703 |
Token | io |
Token ID | 952 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
uracy | +0.477 |
opin | +0.456 |
ription | +0.452 |
romy | +0.411 |
psons | +0.399 |
Neg logit contributions | |
Pixel | -0.571 |
oise | -0.540 |
archy | -0.508 |
ould | -0.499 |
Maher | -0.495 |
Token | beach |
Token ID | 10481 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
Pixel | +0.048 |
oise | +0.044 |
archy | +0.041 |
ould | +0.040 |
Maher | +0.040 |
Neg logit contributions | |
uracy | -0.063 |
opin | -0.061 |
ription | -0.061 |
romy | -0.056 |
psons | -0.055 |
Token | head |
Token ID | 2256 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
uracy | +0.548 |
opin | +0.526 |
ription | +0.523 |
romy | +0.482 |
psons | +0.469 |
Neg logit contributions | |
Pixel | -0.504 |
oise | -0.473 |
archy | -0.440 |
ould | -0.432 |
Maher | -0.428 |
Token | where |
Token ID | 810 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
Pixel | +1.219 |
oise | +1.149 |
archy | +1.073 |
Maher | +1.047 |
ould | +1.032 |
Neg logit contributions | |
uracy | -1.410 |
opin | -1.350 |
ription | -1.346 |
romy | -1.243 |
psons | -1.211 |
Token | they |
Token ID | 484 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
uracy | +0.696 |
opin | +0.667 |
ription | +0.661 |
romy | +0.606 |
inval | +0.590 |
Neg logit contributions | |
Pixel | -0.728 |
oise | -0.686 |
archy | -0.642 |
ould | -0.630 |
Maher | -0.624 |
Token | fought |
Token ID | 8350 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
uracy | +0.448 |
opin | +0.431 |
ription | +0.429 |
romy | +0.394 |
psons | +0.387 |
Neg logit contributions | |
Pixel | -0.383 |
oise | -0.360 |
archy | -0.333 |
ould | -0.327 |
Maher | -0.324 |
Token | alongside |
Token ID | 7848 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
uracy | +2.555 |
opin | +2.470 |
ription | +2.429 |
romy | +2.245 |
inval | +2.219 |
Neg logit contributions | |
Pixel | -2.492 |
oise | -2.344 |
archy | -2.191 |
ould | -2.151 |
Maher | -2.102 |
Token | their |
Token ID | 511 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
uracy | +0.915 |
opin | +0.884 |
ription | +0.874 |
romy | +0.810 |
inval | +0.794 |
Neg logit contributions | |
Pixel | -0.771 |
oise | -0.721 |
archy | -0.671 |
ould | -0.656 |
Maher | -0.647 |
Token | German |
Token ID | 2679 |
Feature activation | -8.75e-02 |
Pos logit contributions | |
uracy | +0.822 |
opin | +0.790 |
ription | +0.786 |
romy | +0.726 |
psons | +0.709 |
Neg logit contributions | |
Pixel | -0.699 |
oise | -0.655 |
archy | -0.607 |
ould | -0.594 |
Maher | -0.590 |
Token | allies |
Token ID | 7681 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Pixel | +0.200 |
oise | +0.189 |
archy | +0.174 |
ould | +0.171 |
Maher | +0.170 |
Neg logit contributions | |
uracy | -0.221 |
opin | -0.211 |
ription | -0.210 |
romy | -0.194 |
psons | -0.189 |
Token | , |
Token ID | 11 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
uracy | +0.469 |
opin | +0.450 |
ription | +0.447 |
romy | +0.411 |
psons | +0.400 |
Neg logit contributions | |
Pixel | -0.443 |
oise | -0.416 |
archy | -0.388 |
ould | -0.379 |
Maher | -0.378 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
Pixel | +0.576 |
oise | +0.535 |
archy | +0.500 |
ould | +0.489 |
Maher | +0.484 |
Neg logit contributions | |
uracy | -0.676 |
ription | -0.643 |
opin | -0.643 |
romy | -0.594 |
psons | -0.583 |
Token | null |
Token ID | 9242 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
Pixel | +0.269 |
oise | +0.250 |
archy | +0.233 |
ould | +0.228 |
Maher | +0.225 |
Neg logit contributions | |
uracy | -0.329 |
ription | -0.315 |
opin | -0.314 |
romy | -0.291 |
psons | -0.285 |
Token | atomic |
Token ID | 17226 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
Pixel | +0.693 |
oise | +0.650 |
archy | +0.610 |
ould | +0.597 |
Maher | +0.590 |
Neg logit contributions | |
uracy | -0.701 |
opin | -0.668 |
ription | -0.665 |
romy | -0.613 |
psons | -0.598 |
Token | pointer |
Token ID | 17562 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
uracy | +0.410 |
opin | +0.394 |
ription | +0.392 |
romy | +0.362 |
psons | +0.354 |
Neg logit contributions | |
Pixel | -0.354 |
oise | -0.331 |
archy | -0.308 |
ould | -0.301 |
Maher | -0.298 |
Token | . |
Token ID | 13 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
uracy | +0.908 |
opin | +0.870 |
ription | +0.865 |
romy | +0.797 |
psons | +0.779 |
Neg logit contributions | |
Pixel | -0.835 |
oise | -0.783 |
archy | -0.731 |
ould | -0.714 |
Maher | -0.710 |
Token | pub |
Token ID | 2240 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Pixel | +0.237 |
oise | +0.221 |
archy | +0.208 |
ould | +0.204 |
Maher | +0.202 |
Neg logit contributions | |
uracy | -0.245 |
opin | -0.234 |
ription | -0.234 |
romy | -0.214 |
psons | -0.210 |
Token | fn |
Token ID | 24714 |
Feature activation | -1.29e-04 |
Pos logit contributions | |
uracy | +2.593 |
opin | +2.513 |
ription | +2.482 |
romy | +2.262 |
inval | +2.244 |
Neg logit contributions | |
Pixel | -2.677 |
oise | -2.550 |
archy | -2.342 |
ould | -2.290 |
Maher | -2.288 |
Token | null |
Token ID | 9242 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
Pixel | +0.000 |
oise | +0.000 |
archy | +0.000 |
ould | +0.000 |
Maher | +0.000 |
Neg logit contributions | |
uracy | -0.000 |
ription | -0.000 |
opin | -0.000 |
romy | -0.000 |
psons | -0.000 |
Token | () |
Token ID | 7499 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
Pixel | +0.557 |
oise | +0.521 |
archy | +0.481 |
ould | +0.470 |
Maher | +0.466 |
Neg logit contributions | |
uracy | -0.707 |
opin | -0.682 |
ription | -0.676 |
romy | -0.627 |
psons | -0.613 |
Token | -> |
Token ID | 4613 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
Pixel | +1.456 |
oise | +1.372 |
archy | +1.283 |
ould | +1.257 |
Maher | +1.242 |
Neg logit contributions | |
uracy | -1.464 |
opin | -1.410 |
ription | -1.402 |
romy | -1.294 |
psons | -1.255 |
Token | Atomic |
Token ID | 28976 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
Pixel | +0.954 |
oise | +0.887 |
archy | +0.823 |
ould | +0.808 |
Maher | +0.806 |
Neg logit contributions | |
uracy | -1.235 |
ription | -1.183 |
opin | -1.179 |
romy | -1.100 |
psons | -1.093 |
Token | the |
Token ID | 262 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
uracy | +0.948 |
opin | +0.911 |
ription | +0.908 |
romy | +0.841 |
psons | +0.822 |
Neg logit contributions | |
Pixel | -0.763 |
oise | -0.714 |
archy | -0.662 |
ould | -0.646 |
Maher | -0.645 |
Token | magazine |
Token ID | 7093 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Pixel | +0.328 |
oise | +0.308 |
archy | +0.285 |
Maher | +0.279 |
ould | +0.277 |
Neg logit contributions | |
uracy | -0.411 |
ription | -0.395 |
opin | -0.394 |
romy | -0.366 |
psons | -0.357 |
Token | responded |
Token ID | 7082 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
uracy | +0.299 |
opin | +0.287 |
ription | +0.286 |
romy | +0.264 |
psons | +0.258 |
Neg logit contributions | |
Pixel | -0.253 |
oise | -0.237 |
archy | -0.220 |
ould | -0.215 |
Maher | -0.214 |
Token | , |
Token ID | 11 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Pixel | +0.511 |
oise | +0.480 |
archy | +0.448 |
Maher | +0.439 |
ould | +0.439 |
Neg logit contributions | |
uracy | -0.530 |
opin | -0.509 |
ription | -0.506 |
romy | -0.466 |
psons | -0.453 |
Token | and |
Token ID | 290 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
uracy | +0.391 |
opin | +0.375 |
ription | +0.372 |
romy | +0.341 |
psons | +0.331 |
Neg logit contributions | |
Pixel | -0.410 |
oise | -0.387 |
archy | -0.362 |
ould | -0.355 |
Maher | -0.352 |
Token | declined |
Token ID | 7392 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
uracy | +0.128 |
opin | +0.123 |
ription | +0.122 |
romy | +0.113 |
psons | +0.110 |
Neg logit contributions | |
Pixel | -0.108 |
oise | -0.101 |
archy | -0.094 |
ould | -0.092 |
Maher | -0.092 |
Token | to |
Token ID | 284 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
uracy | +2.636 |
opin | +2.534 |
ription | +2.512 |
romy | +2.256 |
inval | +2.212 |
Neg logit contributions | |
Pixel | -2.906 |
oise | -2.778 |
archy | -2.582 |
ould | -2.549 |
Maher | -2.438 |
Token | comment |
Token ID | 2912 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
uracy | +1.247 |
opin | +1.194 |
ription | +1.184 |
romy | +1.088 |
psons | +1.061 |
Neg logit contributions | |
Pixel | -1.294 |
oise | -1.222 |
archy | -1.144 |
ould | -1.122 |
Maher | -1.114 |
Token | at |
Token ID | 379 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
uracy | +1.087 |
opin | +1.045 |
ription | +1.035 |
romy | +0.952 |
inval | +0.931 |
Neg logit contributions | |
Pixel | -1.013 |
oise | -0.950 |
archy | -0.884 |
ould | -0.868 |
Maher | -0.855 |
Token | all |
Token ID | 477 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
uracy | +1.086 |
opin | +1.043 |
ription | +1.035 |
romy | +0.954 |
psons | +0.930 |
Neg logit contributions | |
Pixel | -1.004 |
oise | -0.942 |
archy | -0.877 |
ould | -0.859 |
Maher | -0.854 |
Token | except |
Token ID | 2845 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
uracy | +0.772 |
opin | +0.740 |
ription | +0.735 |
romy | +0.673 |
psons | +0.655 |
Neg logit contributions | |
Pixel | -0.811 |
oise | -0.766 |
archy | -0.716 |
ould | -0.702 |
Maher | -0.699 |
Token | economic |
Token ID | 3034 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Pixel | +0.294 |
oise | +0.274 |
archy | +0.253 |
ould | +0.247 |
Maher | +0.245 |
Neg logit contributions | |
uracy | -0.391 |
opin | -0.376 |
ription | -0.374 |
romy | -0.347 |
psons | -0.340 |
Token | interests |
Token ID | 5353 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
uracy | +0.572 |
opin | +0.552 |
ription | +0.548 |
romy | +0.510 |
psons | +0.498 |
Neg logit contributions | |
Pixel | -0.410 |
oise | -0.381 |
archy | -0.350 |
ould | -0.342 |
Maher | -0.339 |
Token | . |
Token ID | 13 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
Pixel | +0.115 |
oise | +0.108 |
archy | +0.101 |
ould | +0.099 |
Maher | +0.099 |
Neg logit contributions | |
uracy | -0.119 |
opin | -0.114 |
ription | -0.114 |
romy | -0.104 |
psons | -0.102 |
Token |
|
Token ID | 198 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
uracy | +0.099 |
opin | +0.094 |
ription | +0.094 |
romy | +0.086 |
inval | +0.084 |
Neg logit contributions | |
Pixel | -0.102 |
oise | -0.096 |
archy | -0.090 |
ould | -0.088 |
Maher | -0.088 |
Token |
|
Token ID | 198 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
Pixel | +0.031 |
oise | +0.029 |
archy | +0.027 |
ould | +0.027 |
Maher | +0.027 |
Neg logit contributions | |
uracy | -0.032 |
opin | -0.031 |
ription | -0.030 |
romy | -0.028 |
inval | -0.027 |
Token | Pet |
Token ID | 25803 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Pixel | +0.147 |
oise | +0.137 |
archy | +0.127 |
ould | +0.125 |
Maher | +0.124 |
Neg logit contributions | |
uracy | -0.158 |
opin | -0.153 |
ription | -0.152 |
romy | -0.139 |
inval | -0.136 |
Token | ro |
Token ID | 305 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
uracy | +2.437 |
opin | +2.340 |
ription | +2.319 |
romy | +2.148 |
psons | +2.060 |
Neg logit contributions | |
Pixel | -2.561 |
oise | -2.407 |
archy | -2.235 |
ould | -2.202 |
Maher | -2.185 |
Token | car |
Token ID | 7718 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
Pixel | +0.234 |
oise | +0.218 |
archy | +0.203 |
ould | +0.198 |
Maher | +0.196 |
Neg logit contributions | |
uracy | -0.289 |
opin | -0.278 |
ription | -0.276 |
romy | -0.255 |
inval | -0.251 |
Token | ibe |
Token ID | 32438 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
Pixel | +0.599 |
oise | +0.567 |
archy | +0.511 |
ould | +0.505 |
Maher | +0.500 |
Neg logit contributions | |
uracy | -0.807 |
opin | -0.779 |
ription | -0.779 |
romy | -0.712 |
inval | -0.702 |
Token | : |
Token ID | 25 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
Pixel | +0.756 |
oise | +0.711 |
archy | +0.661 |
ould | +0.647 |
Maher | +0.643 |
Neg logit contributions | |
uracy | -0.850 |
opin | -0.813 |
ription | -0.812 |
romy | -0.750 |
psons | -0.731 |
Token | Haiti |
Token ID | 25051 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
uracy | +0.061 |
opin | +0.059 |
ription | +0.058 |
romy | +0.054 |
psons | +0.052 |
Neg logit contributions | |
Pixel | -0.058 |
oise | -0.055 |
archy | -0.051 |
ould | -0.050 |
Maher | -0.050 |
Token | pal |
Token ID | 18596 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
Pixel | +0.307 |
oise | +0.288 |
archy | +0.267 |
ould | +0.262 |
Maher | +0.259 |
Neg logit contributions | |
uracy | -0.356 |
opin | -0.343 |
ription | -0.340 |
romy | -0.314 |
inval | -0.307 |
Token | ek |
Token ID | 988 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
Pixel | +1.791 |
oise | +1.696 |
archy | +1.579 |
ould | +1.546 |
Maher | +1.527 |
Neg logit contributions | |
uracy | -2.139 |
opin | -2.066 |
ription | -2.048 |
romy | -1.886 |
inval | -1.851 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Pixel | +0.678 |
oise | +0.636 |
archy | +0.589 |
ould | +0.577 |
Maher | +0.572 |
Neg logit contributions | |
uracy | -0.811 |
opin | -0.780 |
ription | -0.775 |
romy | -0.715 |
psons | -0.700 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
Pixel | +0.608 |
oise | +0.573 |
archy | +0.536 |
ould | +0.525 |
Maher | +0.521 |
Neg logit contributions | |
uracy | -0.587 |
opin | -0.563 |
ription | -0.558 |
romy | -0.511 |
psons | -0.498 |
Token | WASHINGTON |
Token ID | 21793 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
Pixel | +0.057 |
oise | +0.053 |
archy | +0.049 |
ould | +0.048 |
Maher | +0.048 |
Neg logit contributions | |
uracy | -0.061 |
opin | -0.059 |
ription | -0.058 |
romy | -0.053 |
psons | -0.052 |
Token | — |
Token ID | 851 |
Feature activation | +7.25e-02 |
Pos logit contributions | |
Pixel | +0.081 |
oise | +0.075 |
archy | +0.070 |
Maher | +0.068 |
ould | +0.068 |
Neg logit contributions | |
uracy | -0.101 |
ription | -0.097 |
opin | -0.097 |
romy | -0.089 |
psons | -0.087 |
Token | A |
Token ID | 317 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
uracy | +0.180 |
opin | +0.173 |
ription | +0.172 |
romy | +0.158 |
psons | +0.154 |
Neg logit contributions | |
Pixel | -0.169 |
oise | -0.158 |
archy | -0.147 |
Maher | -0.144 |
ould | -0.144 |
Token | Cy |
Token ID | 5934 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
Pixel | +0.629 |
oise | +0.589 |
archy | +0.544 |
ould | +0.531 |
Maher | +0.530 |
Neg logit contributions | |
uracy | -0.798 |
opin | -0.766 |
ription | -0.765 |
romy | -0.705 |
psons | -0.690 |
Token | gn |
Token ID | 4593 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
uracy | +0.173 |
opin | +0.167 |
ription | +0.166 |
romy | +0.154 |
psons | +0.150 |
Neg logit contributions | |
Pixel | -0.152 |
oise | -0.142 |
archy | -0.132 |
ould | -0.129 |
Maher | -0.128 |
Token | us |
Token ID | 385 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
uracy | +0.742 |
opin | +0.701 |
ription | +0.694 |
romy | +0.617 |
psons | +0.593 |
Neg logit contributions | |
Pixel | -1.267 |
oise | -1.203 |
archy | -1.141 |
ould | -1.125 |
Maher | -1.120 |
Token | cargo |
Token ID | 15892 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
Pixel | +0.218 |
oise | +0.204 |
archy | +0.188 |
ould | +0.184 |
Maher | +0.184 |
Neg logit contributions | |
uracy | -0.276 |
ription | -0.265 |
opin | -0.264 |
romy | -0.243 |
psons | -0.239 |
Token | but |
Token ID | 475 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
uracy | +0.257 |
opin | +0.246 |
ription | +0.244 |
romy | +0.225 |
psons | +0.219 |
Neg logit contributions | |
Pixel | -0.249 |
oise | -0.235 |
archy | -0.219 |
Maher | -0.215 |
ould | -0.215 |
Token | said |
Token ID | 531 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
uracy | +0.697 |
opin | +0.670 |
ription | +0.666 |
romy | +0.617 |
psons | +0.601 |
Neg logit contributions | |
Pixel | -0.580 |
oise | -0.543 |
archy | -0.504 |
ould | -0.493 |
Maher | -0.491 |
Token | it |
Token ID | 340 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
uracy | +1.493 |
opin | +1.440 |
ription | +1.426 |
romy | +1.314 |
inval | +1.290 |
Neg logit contributions | |
Pixel | -1.308 |
oise | -1.227 |
archy | -1.136 |
ould | -1.112 |
Maher | -1.094 |
Token | would |
Token ID | 561 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
uracy | +0.410 |
opin | +0.395 |
ription | +0.393 |
romy | +0.364 |
psons | +0.356 |
Neg logit contributions | |
Pixel | -0.318 |
oise | -0.297 |
archy | -0.275 |
ould | -0.269 |
Maher | -0.267 |
Token | try |
Token ID | 1949 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
uracy | +1.697 |
opin | +1.631 |
ription | +1.620 |
romy | +1.496 |
psons | +1.463 |
Neg logit contributions | |
Pixel | -1.462 |
oise | -1.372 |
archy | -1.273 |
ould | -1.248 |
Maher | -1.236 |
Token | to |
Token ID | 284 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
Pixel | +0.596 |
oise | +0.561 |
archy | +0.524 |
ould | +0.514 |
Maher | +0.511 |
Neg logit contributions | |
uracy | -0.626 |
opin | -0.601 |
ription | -0.597 |
romy | -0.548 |
psons | -0.535 |
Token | mitigate |
Token ID | 24237 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
uracy | +0.493 |
opin | +0.474 |
ription | +0.471 |
romy | +0.434 |
psons | +0.425 |
Neg logit contributions | |
Pixel | -0.438 |
oise | -0.413 |
archy | -0.384 |
ould | -0.376 |
Maher | -0.373 |
Token | the |
Token ID | 262 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
Pixel | +0.347 |
oise | +0.325 |
archy | +0.301 |
Maher | +0.297 |
ould | +0.295 |
Neg logit contributions | |
uracy | -0.432 |
opin | -0.414 |
ription | -0.414 |
romy | -0.385 |
psons | -0.375 |
Token | impact |
Token ID | 2928 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
uracy | +0.501 |
opin | +0.481 |
ription | +0.479 |
romy | +0.444 |
psons | +0.433 |
Neg logit contributions | |
Pixel | -0.399 |
oise | -0.375 |
archy | -0.346 |
ould | -0.338 |
Maher | -0.337 |
Token | on |
Token ID | 319 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
Pixel | +0.081 |
oise | +0.076 |
archy | +0.071 |
Maher | +0.069 |
ould | +0.069 |
Neg logit contributions | |
uracy | -0.100 |
opin | -0.097 |
ription | -0.096 |
romy | -0.089 |
psons | -0.087 |
Token | employees |
Token ID | 4409 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
Pixel | +0.669 |
oise | +0.629 |
archy | +0.581 |
Maher | +0.573 |
ould | +0.571 |
Neg logit contributions | |
uracy | -0.794 |
opin | -0.761 |
ription | -0.760 |
romy | -0.704 |
psons | -0.686 |
Token | on |
Token ID | 319 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
Pixel | +0.003 |
oise | +0.003 |
archy | +0.003 |
ould | +0.003 |
Maher | +0.003 |
Neg logit contributions | |
uracy | -0.003 |
opin | -0.003 |
ription | -0.003 |
romy | -0.003 |
psons | -0.003 |
Token | partnering |
Token ID | 41759 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Pixel | +0.630 |
oise | +0.585 |
archy | +0.541 |
Maher | +0.527 |
ould | +0.526 |
Neg logit contributions | |
uracy | -0.793 |
opin | -0.768 |
ription | -0.764 |
romy | -0.709 |
psons | -0.692 |
Token | with |
Token ID | 351 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Pixel | +2.684 |
oise | +2.487 |
archy | +2.300 |
ould | +2.246 |
Maher | +2.246 |
Neg logit contributions | |
uracy | -3.115 |
ription | -3.021 |
opin | -3.019 |
romy | -2.799 |
psons | -2.709 |
Token | television |
Token ID | 5581 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Pixel | +0.291 |
oise | +0.270 |
archy | +0.248 |
ould | +0.243 |
Maher | +0.243 |
Neg logit contributions | |
uracy | -0.366 |
opin | -0.352 |
ription | -0.352 |
romy | -0.325 |
psons | -0.317 |
Token | providers |
Token ID | 9549 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
Pixel | +2.149 |
oise | +1.955 |
archy | +1.800 |
Maher | +1.754 |
Pixel | +1.730 |
Neg logit contributions | |
uracy | -2.810 |
ription | -2.742 |
opin | -2.667 |
romy | -2.494 |
yip | -2.433 |
Token | and |
Token ID | 290 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
uracy | +1.396 |
opin | +1.339 |
ription | +1.331 |
romy | +1.224 |
psons | +1.193 |
Neg logit contributions | |
Pixel | -1.318 |
oise | -1.236 |
archy | -1.153 |
ould | -1.126 |
Maher | -1.121 |
Token | content |
Token ID | 2695 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
uracy | +0.150 |
opin | +0.144 |
ription | +0.144 |
romy | +0.133 |
psons | +0.130 |
Neg logit contributions | |
Pixel | -0.121 |
oise | -0.112 |
archy | -0.104 |
ould | -0.101 |
Maher | -0.101 |
Token | distributors |
Token ID | 32612 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
Pixel | +0.087 |
oise | +0.081 |
archy | +0.076 |
Maher | +0.075 |
ould | +0.074 |
Neg logit contributions | |
uracy | -0.082 |
ription | -0.078 |
opin | -0.078 |
romy | -0.072 |
psons | -0.069 |
Token | to |
Token ID | 284 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
Pixel | +0.330 |
oise | +0.309 |
archy | +0.289 |
Maher | +0.280 |
ould | +0.279 |
Neg logit contributions | |
uracy | -0.360 |
opin | -0.346 |
ription | -0.344 |
romy | -0.318 |
psons | -0.310 |
Token | deliver |
Token ID | 5203 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
uracy | +0.214 |
opin | +0.207 |
ription | +0.206 |
romy | +0.191 |
psons | +0.187 |
Neg logit contributions | |
Pixel | -0.155 |
oise | -0.145 |
archy | -0.133 |
ould | -0.130 |
Maher | -0.128 |
Token | rich |
Token ID | 5527 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
Pixel | +0.283 |
oise | +0.263 |
archy | +0.244 |
Maher | +0.238 |
ould | +0.236 |
Neg logit contributions | |
uracy | -0.361 |
ription | -0.349 |
opin | -0.347 |
romy | -0.322 |
psons | -0.314 |
Token | play |
Token ID | 711 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
uracy | +0.713 |
opin | +0.683 |
ription | +0.676 |
romy | +0.621 |
psons | +0.604 |
Neg logit contributions | |
Pixel | -0.742 |
oise | -0.699 |
archy | -0.654 |
ould | -0.641 |
Maher | -0.637 |
Token | it |
Token ID | 340 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
uracy | +0.955 |
opin | +0.919 |
ription | +0.911 |
romy | +0.842 |
psons | +0.822 |
Neg logit contributions | |
Pixel | -0.830 |
oise | -0.778 |
archy | -0.723 |
ould | -0.707 |
Maher | -0.701 |
Token | with |
Token ID | 351 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
uracy | +0.611 |
opin | +0.586 |
ription | +0.581 |
romy | +0.534 |
psons | +0.520 |
Neg logit contributions | |
Pixel | -0.598 |
oise | -0.562 |
archy | -0.525 |
ould | -0.514 |
Maher | -0.510 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
uracy | +0.445 |
opin | +0.428 |
ription | +0.424 |
romy | +0.392 |
psons | +0.383 |
Neg logit contributions | |
Pixel | -0.380 |
oise | -0.357 |
archy | -0.331 |
ould | -0.324 |
Maher | -0.320 |
Token | j |
Token ID | 474 |
Feature activation | -9.63e-02 |
Pos logit contributions | |
uracy | +0.397 |
ription | +0.382 |
opin | +0.381 |
romy | +0.349 |
inval | +0.340 |
Neg logit contributions | |
Pixel | -0.366 |
oise | -0.344 |
archy | -0.320 |
Maher | -0.314 |
ould | -0.313 |
Token | agg |
Token ID | 9460 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
Pixel | +0.219 |
oise | +0.205 |
archy | +0.191 |
ould | +0.187 |
Maher | +0.186 |
Neg logit contributions | |
uracy | -0.244 |
opin | -0.234 |
ription | -0.231 |
romy | -0.214 |
psons | -0.209 |
Token | ies |
Token ID | 444 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
uracy | +0.120 |
opin | +0.116 |
ription | +0.114 |
romy | +0.104 |
inval | +0.101 |
Neg logit contributions | |
Pixel | -0.144 |
oise | -0.137 |
archy | -0.129 |
ould | -0.127 |
Maher | -0.125 |
Token | since |
Token ID | 1201 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
uracy | +0.930 |
opin | +0.884 |
ription | +0.879 |
romy | +0.794 |
inval | +0.766 |
Neg logit contributions | |
Pixel | -1.287 |
oise | -1.222 |
archy | -1.155 |
ould | -1.136 |
Maher | -1.131 |
Token | even |
Token ID | 772 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
uracy | +1.635 |
opin | +1.572 |
ription | +1.547 |
romy | +1.421 |
psons | +1.396 |
Neg logit contributions | |
Pixel | -1.513 |
oise | -1.426 |
archy | -1.320 |
ould | -1.287 |
Maher | -1.281 |
Token | though |
Token ID | 996 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
uracy | +0.060 |
opin | +0.058 |
ription | +0.057 |
romy | +0.052 |
psons | +0.050 |
Neg logit contributions | |
Pixel | -0.074 |
oise | -0.070 |
archy | -0.066 |
ould | -0.065 |
Maher | -0.064 |
Token | they |
Token ID | 484 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
uracy | +0.472 |
opin | +0.451 |
ription | +0.446 |
romy | +0.408 |
psons | +0.395 |
Neg logit contributions | |
Pixel | -0.547 |
oise | -0.518 |
archy | -0.485 |
ould | -0.477 |
Maher | -0.473 |
Token | with |
Token ID | 351 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
uracy | +1.086 |
opin | +1.044 |
ription | +1.035 |
romy | +0.955 |
psons | +0.931 |
Neg logit contributions | |
Pixel | -0.980 |
oise | -0.920 |
archy | -0.856 |
ould | -0.838 |
Maher | -0.831 |
Token | soft |
Token ID | 2705 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
uracy | +0.051 |
opin | +0.049 |
ription | +0.049 |
romy | +0.045 |
psons | +0.044 |
Neg logit contributions | |
Pixel | -0.045 |
oise | -0.042 |
archy | -0.039 |
ould | -0.038 |
Maher | -0.038 |
Token | or |
Token ID | 393 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
Pixel | +0.407 |
oise | +0.381 |
archy | +0.353 |
ould | +0.346 |
Maher | +0.342 |
Neg logit contributions | |
uracy | -0.487 |
opin | -0.467 |
ription | -0.465 |
romy | -0.429 |
inval | -0.420 |
Token | tough |
Token ID | 5802 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
Pixel | +0.186 |
oise | +0.175 |
archy | +0.163 |
ould | +0.159 |
Maher | +0.159 |
Neg logit contributions | |
uracy | -0.217 |
opin | -0.207 |
ription | -0.207 |
romy | -0.192 |
psons | -0.186 |
Token | conditions |
Token ID | 3403 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
Pixel | +0.121 |
oise | +0.112 |
archy | +0.102 |
ould | +0.099 |
Maher | +0.098 |
Neg logit contributions | |
uracy | -0.196 |
opin | -0.190 |
ription | -0.189 |
romy | -0.176 |
psons | -0.172 |
Token | , |
Token ID | 11 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
Pixel | +0.856 |
oise | +0.802 |
archy | +0.743 |
ould | +0.734 |
Maher | +0.730 |
Neg logit contributions | |
uracy | -0.990 |
opin | -0.947 |
ription | -0.942 |
romy | -0.871 |
psons | -0.850 |
Token | appears |
Token ID | 3568 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
uracy | +0.371 |
opin | +0.357 |
ription | +0.356 |
romy | +0.331 |
psons | +0.324 |
Neg logit contributions | |
Pixel | -0.257 |
oise | -0.239 |
archy | -0.220 |
ould | -0.215 |
Maher | -0.213 |
Token | to |
Token ID | 284 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
Pixel | +1.477 |
oise | +1.397 |
archy | +1.306 |
ould | +1.282 |
Maher | +1.274 |
Neg logit contributions | |
uracy | -1.573 |
ription | -1.503 |
opin | -1.495 |
romy | -1.376 |
psons | -1.339 |
Token | be |
Token ID | 307 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
uracy | +1.045 |
opin | +1.007 |
ription | +0.997 |
romy | +0.927 |
inval | +0.907 |
Neg logit contributions | |
Pixel | -0.815 |
oise | -0.757 |
archy | -0.700 |
ould | -0.682 |
Maher | -0.677 |
Token | inevitable |
Token ID | 13203 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
uracy | +1.196 |
opin | +1.155 |
ription | +1.137 |
romy | +1.050 |
inval | +1.032 |
Neg logit contributions | |
Pixel | -1.100 |
oise | -1.031 |
archy | -0.958 |
ould | -0.939 |
Maher | -0.928 |
Token | now |
Token ID | 783 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
uracy | +1.045 |
opin | +1.003 |
ription | +0.998 |
romy | +0.917 |
psons | +0.891 |
Neg logit contributions | |
Pixel | -1.030 |
oise | -0.968 |
archy | -0.904 |
ould | -0.889 |
Maher | -0.881 |
Token | budget |
Token ID | 4466 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
Pixel | +0.017 |
oise | +0.016 |
archy | +0.015 |
ould | +0.015 |
Maher | +0.014 |
Neg logit contributions | |
uracy | -0.020 |
opin | -0.019 |
ription | -0.019 |
romy | -0.018 |
psons | -0.017 |
Token | sequ |
Token ID | 4726 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
uracy | +0.535 |
opin | +0.514 |
ription | +0.512 |
romy | +0.471 |
psons | +0.460 |
Neg logit contributions | |
Pixel | -0.466 |
oise | -0.438 |
archy | -0.406 |
ould | -0.397 |
Maher | -0.396 |
Token | ester |
Token ID | 7834 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
Pixel | +2.634 |
oise | +2.606 |
ould | +2.387 |
archy | +2.367 |
Maher | +2.270 |
Neg logit contributions | |
opin | -2.512 |
uracy | -2.507 |
ription | -2.416 |
romy | -2.183 |
inval | -2.161 |
Token | that |
Token ID | 326 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
Pixel | +1.058 |
oise | +0.998 |
archy | +0.915 |
Maher | +0.908 |
ould | +0.895 |
Neg logit contributions | |
uracy | -1.309 |
opin | -1.257 |
ription | -1.256 |
romy | -1.160 |
psons | -1.127 |
Token | Obama |
Token ID | 2486 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
uracy | +0.198 |
opin | +0.190 |
ription | +0.190 |
romy | +0.175 |
psons | +0.170 |
Neg logit contributions | |
Pixel | -0.177 |
oise | -0.166 |
archy | -0.154 |
Maher | -0.151 |
ould | -0.151 |
Token | liked |
Token ID | 8288 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
uracy | +0.074 |
opin | +0.071 |
ription | +0.071 |
romy | +0.065 |
psons | +0.064 |
Neg logit contributions | |
Pixel | -0.063 |
oise | -0.059 |
archy | -0.054 |
ould | -0.053 |
Maher | -0.053 |
Token | this |
Token ID | 428 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
Pixel | +1.051 |
oise | +0.992 |
archy | +0.925 |
Maher | +0.906 |
ould | +0.898 |
Neg logit contributions | |
uracy | -1.149 |
opin | -1.106 |
ription | -1.104 |
romy | -1.016 |
psons | -0.987 |
Token | approach |
Token ID | 3164 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
Pixel | +0.101 |
oise | +0.095 |
archy | +0.088 |
Maher | +0.085 |
ould | +0.085 |
Neg logit contributions | |
uracy | -0.128 |
opin | -0.123 |
ription | -0.123 |
romy | -0.114 |
psons | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
Pixel | +0.987 |
oise | +0.926 |
archy | +0.863 |
Maher | +0.847 |
ould | +0.846 |
Neg logit contributions | |
uracy | -1.065 |
opin | -1.025 |
ription | -1.021 |
romy | -0.941 |
psons | -0.913 |
Token | He |
Token ID | 679 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
Pixel | +0.132 |
oise | +0.125 |
archy | +0.116 |
Maher | +0.115 |
ould | +0.114 |
Neg logit contributions | |
uracy | -0.121 |
opin | -0.116 |
ription | -0.115 |
romy | -0.105 |
inval | -0.103 |
Token | ordered |
Token ID | 6149 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
Pixel | +0.206 |
oise | +0.194 |
archy | +0.181 |
ould | +0.177 |
Maher | +0.176 |
Neg logit contributions | |
uracy | -0.234 |
ription | -0.225 |
opin | -0.224 |
romy | -0.206 |
psons | -0.201 |
Token | produce |
Token ID | 4439 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Pixel | +0.398 |
oise | +0.375 |
archy | +0.347 |
ould | +0.343 |
Maher | +0.341 |
Neg logit contributions | |
uracy | -0.460 |
ription | -0.440 |
opin | -0.437 |
romy | -0.403 |
psons | -0.392 |
Token | useful |
Token ID | 4465 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Pixel | +0.244 |
oise | +0.229 |
archy | +0.213 |
ould | +0.208 |
Maher | +0.208 |
Neg logit contributions | |
uracy | -0.286 |
opin | -0.273 |
ription | -0.273 |
romy | -0.252 |
psons | -0.245 |
Token | tricks |
Token ID | 15910 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
uracy | +0.277 |
opin | +0.266 |
ription | +0.265 |
romy | +0.245 |
psons | +0.239 |
Neg logit contributions | |
Pixel | -0.221 |
oise | -0.206 |
archy | -0.191 |
ould | -0.187 |
Maher | -0.186 |
Token | without |
Token ID | 1231 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
Pixel | +1.030 |
oise | +0.972 |
archy | +0.893 |
ould | +0.871 |
Maher | +0.862 |
Neg logit contributions | |
uracy | -1.357 |
opin | -1.307 |
ription | -1.297 |
romy | -1.201 |
psons | -1.172 |
Token | resort |
Token ID | 12600 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
uracy | +0.117 |
opin | +0.112 |
ription | +0.112 |
romy | +0.103 |
psons | +0.100 |
Neg logit contributions | |
Pixel | -0.106 |
oise | -0.100 |
archy | -0.093 |
ould | -0.091 |
Maher | -0.091 |
Token | ing |
Token ID | 278 |
Feature activation | -7.25e-01 |
Pos logit contributions | |
Pixel | +0.444 |
oise | +0.416 |
archy | +0.385 |
ould | +0.374 |
Maher | +0.371 |
Neg logit contributions | |
uracy | -0.560 |
opin | -0.540 |
ription | -0.535 |
romy | -0.495 |
psons | -0.485 |
Token | to |
Token ID | 284 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
Pixel | +1.702 |
oise | +1.603 |
archy | +1.497 |
ould | +1.465 |
Maher | +1.453 |
Neg logit contributions | |
uracy | -1.779 |
opin | -1.705 |
ription | -1.692 |
romy | -1.556 |
psons | -1.520 |
Token | any |
Token ID | 597 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
Pixel | +0.833 |
oise | +0.783 |
archy | +0.726 |
Maher | +0.710 |
ould | +0.709 |
Neg logit contributions | |
uracy | -0.974 |
ription | -0.932 |
opin | -0.930 |
romy | -0.862 |
psons | -0.840 |
Token | mechanical |
Token ID | 12370 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
Pixel | +1.217 |
oise | +1.145 |
archy | +1.052 |
ould | +1.034 |
Maher | +1.030 |
Neg logit contributions | |
uracy | -1.549 |
ription | -1.484 |
opin | -1.482 |
romy | -1.375 |
psons | -1.340 |
Token | complexity |
Token ID | 13357 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
uracy | +0.042 |
opin | +0.041 |
ription | +0.041 |
romy | +0.038 |
psons | +0.037 |
Neg logit contributions | |
Pixel | -0.032 |
oise | -0.030 |
archy | -0.028 |
ould | -0.027 |
Maher | -0.027 |
Token | is |
Token ID | 318 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
Pixel | +0.055 |
oise | +0.051 |
archy | +0.047 |
ould | +0.046 |
Maher | +0.046 |
Neg logit contributions | |
uracy | -0.076 |
opin | -0.073 |
ription | -0.073 |
romy | -0.068 |
psons | -0.066 |
Token | morning |
Token ID | 3329 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Pixel | +0.657 |
oise | +0.616 |
archy | +0.568 |
ould | +0.557 |
Maher | +0.556 |
Neg logit contributions | |
uracy | -0.821 |
opin | -0.785 |
ription | -0.784 |
romy | -0.726 |
psons | -0.712 |
Token | and |
Token ID | 290 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
Pixel | +0.659 |
oise | +0.622 |
archy | +0.574 |
ould | +0.564 |
Maher | +0.563 |
Neg logit contributions | |
uracy | -0.773 |
opin | -0.743 |
ription | -0.739 |
romy | -0.684 |
psons | -0.671 |
Token | cleared |
Token ID | 12539 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
Pixel | +0.494 |
oise | +0.465 |
archy | +0.431 |
Maher | +0.424 |
ould | +0.424 |
Neg logit contributions | |
uracy | -0.585 |
opin | -0.560 |
ription | -0.560 |
romy | -0.517 |
psons | -0.506 |
Token | up |
Token ID | 510 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
Pixel | +0.671 |
oise | +0.632 |
archy | +0.586 |
ould | +0.574 |
Maher | +0.573 |
Neg logit contributions | |
uracy | -0.769 |
opin | -0.738 |
ription | -0.736 |
romy | -0.679 |
psons | -0.664 |
Token | by |
Token ID | 416 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Pixel | +0.047 |
oise | +0.045 |
archy | +0.041 |
ould | +0.041 |
Maher | +0.041 |
Neg logit contributions | |
uracy | -0.053 |
opin | -0.050 |
ription | -0.050 |
romy | -0.046 |
psons | -0.045 |
Token | 2 |
Token ID | 362 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
uracy | +0.299 |
opin | +0.287 |
ription | +0.287 |
romy | +0.264 |
psons | +0.258 |
Neg logit contributions | |
Pixel | -0.260 |
oise | -0.245 |
archy | -0.226 |
ould | -0.223 |
Maher | -0.222 |
Token | pm |
Token ID | 4426 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
Pixel | +0.791 |
oise | +0.749 |
archy | +0.697 |
ould | +0.682 |
Maher | +0.678 |
Neg logit contributions | |
uracy | -0.840 |
opin | -0.811 |
ription | -0.804 |
romy | -0.741 |
psons | -0.724 |
Token | , |
Token ID | 11 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
uracy | +0.060 |
opin | +0.058 |
ription | +0.057 |
romy | +0.053 |
psons | +0.052 |
Neg logit contributions | |
Pixel | -0.056 |
oise | -0.053 |
archy | -0.049 |
ould | -0.048 |
Maher | -0.048 |
Token | so |
Token ID | 523 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Pixel | +0.525 |
oise | +0.494 |
archy | +0.459 |
ould | +0.452 |
Maher | +0.450 |
Neg logit contributions | |
uracy | -0.590 |
opin | -0.567 |
ription | -0.565 |
romy | -0.520 |
psons | -0.508 |
Token | our |
Token ID | 674 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
uracy | +0.714 |
opin | +0.686 |
ription | +0.682 |
romy | +0.628 |
psons | +0.614 |
Neg logit contributions | |
Pixel | -0.640 |
oise | -0.600 |
archy | -0.558 |
ould | -0.548 |
Maher | -0.544 |
Token | wet |
Token ID | 9583 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
Pixel | +0.654 |
oise | +0.616 |
archy | +0.567 |
ould | +0.560 |
Maher | +0.558 |
Neg logit contributions | |
uracy | -0.784 |
ription | -0.753 |
opin | -0.751 |
romy | -0.692 |
psons | -0.680 |
Token | identified |
Token ID | 5174 |
Feature activation | -5.98e-02 |
Pos logit contributions | |
Pixel | +0.427 |
oise | +0.399 |
archy | +0.369 |
ould | +0.363 |
Maher | +0.359 |
Neg logit contributions | |
uracy | -0.550 |
ription | -0.528 |
opin | -0.526 |
romy | -0.488 |
psons | -0.480 |
Token | with |
Token ID | 351 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
Pixel | +0.146 |
oise | +0.138 |
archy | +0.129 |
ould | +0.127 |
Maher | +0.126 |
Neg logit contributions | |
uracy | -0.141 |
opin | -0.135 |
ription | -0.134 |
romy | -0.123 |
psons | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
uracy | +0.144 |
opin | +0.138 |
ription | +0.137 |
romy | +0.127 |
psons | +0.124 |
Neg logit contributions | |
Pixel | -0.125 |
oise | -0.118 |
archy | -0.109 |
ould | -0.107 |
Maher | -0.106 |
Token | middle |
Token ID | 3504 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
Pixel | +0.255 |
oise | +0.240 |
archy | +0.223 |
ould | +0.218 |
Maher | +0.217 |
Neg logit contributions | |
uracy | -0.298 |
opin | -0.284 |
ription | -0.284 |
romy | -0.262 |
psons | -0.257 |
Token | class |
Token ID | 1398 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
Pixel | +0.283 |
oise | +0.262 |
archy | +0.240 |
ould | +0.228 |
Maher | +0.227 |
Neg logit contributions | |
uracy | -0.537 |
opin | -0.517 |
ription | -0.516 |
romy | -0.485 |
psons | -0.475 |
Token | , |
Token ID | 11 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
uracy | +1.202 |
opin | +1.153 |
ription | +1.145 |
romy | +1.054 |
psons | +1.028 |
Neg logit contributions | |
Pixel | -1.132 |
oise | -1.064 |
archy | -0.993 |
ould | -0.972 |
Maher | -0.964 |
Token | despite |
Token ID | 3805 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
Pixel | +0.299 |
oise | +0.279 |
archy | +0.261 |
Maher | +0.254 |
ould | +0.254 |
Neg logit contributions | |
uracy | -0.347 |
ription | -0.332 |
opin | -0.332 |
romy | -0.306 |
psons | -0.299 |
Token | not |
Token ID | 407 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
uracy | +0.241 |
ription | +0.230 |
opin | +0.230 |
romy | +0.213 |
psons | +0.208 |
Neg logit contributions | |
Pixel | -0.201 |
oise | -0.189 |
archy | -0.175 |
ould | -0.171 |
Maher | -0.171 |
Token | necessarily |
Token ID | 6646 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
Pixel | +0.738 |
oise | +0.687 |
archy | +0.641 |
Maher | +0.630 |
ould | +0.624 |
Neg logit contributions | |
uracy | -0.917 |
ription | -0.881 |
opin | -0.870 |
romy | -0.810 |
psons | -0.791 |
Token | enjoying |
Token ID | 13226 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
uracy | +0.069 |
ription | +0.066 |
opin | +0.066 |
romy | +0.061 |
psons | +0.060 |
Neg logit contributions | |
Pixel | -0.056 |
oise | -0.052 |
archy | -0.048 |
Maher | -0.047 |
ould | -0.047 |
Token | all |
Token ID | 477 |
Feature activation | +8.64e-02 |
Pos logit contributions | |
uracy | +0.052 |
opin | +0.050 |
ription | +0.049 |
romy | +0.046 |
psons | +0.045 |
Neg logit contributions | |
Pixel | -0.042 |
oise | -0.039 |
archy | -0.036 |
ould | -0.036 |
Maher | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
uracy | +0.134 |
opin | +0.129 |
ription | +0.128 |
romy | +0.118 |
psons | +0.115 |
Neg logit contributions | |
Pixel | -0.132 |
oise | -0.124 |
archy | -0.116 |
ould | -0.114 |
Maher | -0.113 |
Token | The |
Token ID | 464 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
uracy | +0.041 |
opin | +0.039 |
ription | +0.039 |
romy | +0.036 |
inval | +0.035 |
Neg logit contributions | |
Pixel | -0.040 |
oise | -0.038 |
archy | -0.035 |
ould | -0.034 |
Maher | -0.034 |
Token | high |
Token ID | 1029 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
uracy | +0.471 |
opin | +0.453 |
ription | +0.451 |
romy | +0.416 |
psons | +0.405 |
Neg logit contributions | |
Pixel | -0.408 |
oise | -0.382 |
archy | -0.355 |
ould | -0.347 |
Maher | -0.345 |
Token | court |
Token ID | 2184 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
Pixel | +0.318 |
oise | +0.297 |
archy | +0.275 |
ould | +0.268 |
Maher | +0.267 |
Neg logit contributions | |
uracy | -0.412 |
ription | -0.396 |
opin | -0.396 |
romy | -0.366 |
psons | -0.358 |
Token | 's |
Token ID | 338 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
Pixel | +1.162 |
oise | +1.096 |
archy | +1.006 |
Maher | +0.999 |
ould | +0.984 |
Neg logit contributions | |
uracy | -1.416 |
ription | -1.366 |
opin | -1.340 |
romy | -1.256 |
psons | -1.219 |
Token | order |
Token ID | 1502 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
Pixel | +0.090 |
oise | +0.085 |
archy | +0.078 |
Maher | +0.076 |
ould | +0.076 |
Neg logit contributions | |
uracy | -0.121 |
ription | -0.117 |
opin | -0.116 |
romy | -0.108 |
psons | -0.105 |
Token | had |
Token ID | 550 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Pixel | +1.067 |
oise | +1.001 |
archy | +0.920 |
Maher | +0.911 |
ould | +0.900 |
Neg logit contributions | |
uracy | -1.351 |
ription | -1.296 |
opin | -1.289 |
romy | -1.202 |
psons | -1.168 |
Token | come |
Token ID | 1282 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
uracy | +0.409 |
opin | +0.392 |
ription | +0.390 |
romy | +0.359 |
psons | +0.349 |
Neg logit contributions | |
Pixel | -0.390 |
oise | -0.367 |
archy | -0.342 |
ould | -0.335 |
Maher | -0.335 |
Token | on |
Token ID | 319 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
uracy | +1.359 |
opin | +1.303 |
ription | +1.299 |
romy | +1.194 |
psons | +1.163 |
Neg logit contributions | |
Pixel | -1.247 |
oise | -1.172 |
archy | -1.093 |
ould | -1.068 |
Maher | -1.063 |
Token | a |
Token ID | 257 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
uracy | +0.322 |
opin | +0.310 |
ription | +0.308 |
romy | +0.284 |
psons | +0.278 |
Neg logit contributions | |
Pixel | -0.275 |
oise | -0.258 |
archy | -0.239 |
ould | -0.234 |
Maher | -0.233 |
Token | P |
Token ID | 350 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
Pixel | +0.071 |
oise | +0.066 |
archy | +0.062 |
ould | +0.061 |
Maher | +0.060 |
Neg logit contributions | |
uracy | -0.075 |
opin | -0.072 |
ription | -0.072 |
romy | -0.066 |
psons | -0.064 |
Token | g |
Token ID | 70 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
Pixel | +1.312 |
oise | +1.227 |
archy | +1.142 |
ould | +1.116 |
Maher | +1.089 |
Neg logit contributions | |
uracy | -1.513 |
opin | -1.467 |
ription | -1.456 |
romy | -1.327 |
inval | -1.310 |
Token | 8 |
Token ID | 23 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
Pixel | +0.262 |
oise | +0.245 |
archy | +0.228 |
ould | +0.223 |
Maher | +0.220 |
Neg logit contributions | |
uracy | -0.299 |
opin | -0.288 |
ription | -0.286 |
romy | -0.263 |
inval | -0.258 |
Token | sm |
Token ID | 5796 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
Pixel | +0.158 |
oise | +0.148 |
archy | +0.137 |
ould | +0.134 |
Maher | +0.133 |
Neg logit contributions | |
uracy | -0.181 |
opin | -0.174 |
ription | -0.173 |
romy | -0.159 |
inval | -0.155 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
uracy | +0.056 |
opin | +0.054 |
ription | +0.054 |
romy | +0.049 |
inval | +0.048 |
Neg logit contributions | |
Pixel | -0.066 |
oise | -0.062 |
archy | -0.059 |
ould | -0.057 |
Maher | -0.057 |
Token | Whatever |
Token ID | 21875 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
uracy | +0.029 |
opin | +0.028 |
ription | +0.028 |
romy | +0.025 |
psons | +0.025 |
Neg logit contributions | |
Pixel | -0.027 |
oise | -0.025 |
archy | -0.023 |
ould | -0.023 |
Maher | -0.023 |
Token | reason |
Token ID | 1738 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Pixel | +0.523 |
oise | +0.488 |
archy | +0.452 |
ould | +0.442 |
Maher | +0.441 |
Neg logit contributions | |
uracy | -0.642 |
opin | -0.617 |
ription | -0.615 |
romy | -0.569 |
psons | -0.555 |
Token | Cal |
Token ID | 2199 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
Pixel | +2.456 |
oise | +2.294 |
archy | +2.138 |
Maher | +2.126 |
ould | +2.094 |
Neg logit contributions | |
uracy | -2.987 |
ription | -2.874 |
opin | -2.865 |
romy | -2.645 |
psons | -2.586 |
Token | Quant |
Token ID | 16972 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Pixel | +0.376 |
oise | +0.351 |
archy | +0.324 |
ould | +0.317 |
Maher | +0.314 |
Neg logit contributions | |
uracy | -0.479 |
opin | -0.462 |
ription | -0.459 |
romy | -0.425 |
inval | -0.415 |
Token | rill |
Token ID | 20190 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
uracy | +0.324 |
opin | +0.313 |
ription | +0.310 |
romy | +0.289 |
psons | +0.282 |
Neg logit contributions | |
Pixel | -0.229 |
oise | -0.212 |
archy | -0.195 |
ould | -0.191 |
Maher | -0.189 |
Token | had |
Token ID | 550 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
Pixel | +0.829 |
oise | +0.777 |
archy | +0.719 |
ould | +0.702 |
Maher | +0.701 |
Neg logit contributions | |
uracy | -1.058 |
ription | -1.016 |
opin | -1.015 |
romy | -0.936 |
psons | -0.915 |
Token | to |
Token ID | 284 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
Pixel | +0.056 |
oise | +0.053 |
archy | +0.049 |
ould | +0.048 |
Maher | +0.048 |
Neg logit contributions | |
uracy | -0.060 |
opin | -0.058 |
ription | -0.058 |
romy | -0.053 |
psons | -0.052 |
Token | h |
Token ID | 289 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
Pixel | +0.367 |
oise | +0.340 |
archy | +0.312 |
ould | +0.305 |
Maher | +0.302 |
Neg logit contributions | |
uracy | -0.555 |
opin | -0.535 |
ription | -0.532 |
romy | -0.495 |
psons | -0.486 |
Token | odge |
Token ID | 9728 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
uracy | +0.052 |
opin | +0.050 |
ription | +0.049 |
romy | +0.045 |
psons | +0.044 |
Neg logit contributions | |
Pixel | -0.050 |
oise | -0.047 |
archy | -0.044 |
ould | -0.043 |
Maher | -0.043 |
Token | p |
Token ID | 79 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
uracy | +0.083 |
opin | +0.079 |
ription | +0.078 |
romy | +0.071 |
inval | +0.069 |
Neg logit contributions | |
Pixel | -0.105 |
oise | -0.100 |
archy | -0.093 |
ould | -0.092 |
Maher | -0.091 |
Token | odge |
Token ID | 9728 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
Pixel | +1.083 |
oise | +1.002 |
archy | +0.917 |
ould | +0.901 |
Maher | +0.900 |
Neg logit contributions | |
uracy | -1.528 |
opin | -1.465 |
ription | -1.461 |
romy | -1.360 |
psons | -1.328 |
Token | of |
Token ID | 286 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
Pixel | +0.969 |
oise | +0.907 |
archy | +0.840 |
ould | +0.824 |
Maher | +0.818 |
Neg logit contributions | |
uracy | -1.116 |
opin | -1.076 |
ription | -1.069 |
romy | -0.985 |
inval | -0.962 |
Token | new |
Token ID | 649 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Pixel | +0.728 |
oise | +0.673 |
archy | +0.619 |
ould | +0.604 |
Maher | +0.601 |
Neg logit contributions | |
uracy | -1.091 |
opin | -1.047 |
ription | -1.045 |
romy | -0.971 |
psons | -0.954 |
Token | housing |
Token ID | 5627 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
Pixel | +1.906 |
oise | +1.742 |
archy | +1.569 |
ould | +1.544 |
Maher | +1.537 |
Neg logit contributions | |
uracy | -3.501 |
ription | -3.370 |
opin | -3.344 |
romy | -3.135 |
psons | -3.074 |
Token | programs |
Token ID | 4056 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
Pixel | +1.080 |
oise | +0.992 |
archy | +0.914 |
ould | +0.895 |
Maher | +0.889 |
Neg logit contributions | |
uracy | -1.593 |
ription | -1.533 |
opin | -1.529 |
romy | -1.429 |
psons | -1.397 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
Pixel | +1.411 |
oise | +1.311 |
archy | +1.232 |
Maher | +1.204 |
ould | +1.201 |
Neg logit contributions | |
uracy | -1.573 |
opin | -1.504 |
ription | -1.500 |
romy | -1.378 |
psons | -1.347 |
Token | The |
Token ID | 383 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
Pixel | +0.441 |
oise | +0.413 |
archy | +0.386 |
Maher | +0.383 |
ould | +0.379 |
Neg logit contributions | |
uracy | -0.425 |
opin | -0.405 |
ription | -0.403 |
romy | -0.369 |
inval | -0.360 |
Token | new |
Token ID | 649 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
Pixel | +0.507 |
oise | +0.473 |
archy | +0.435 |
ould | +0.424 |
Maher | +0.421 |
Neg logit contributions | |
uracy | -0.694 |
opin | -0.669 |
ription | -0.665 |
romy | -0.618 |
psons | -0.604 |
Token | v |
Token ID | 85 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
uracy | +0.642 |
opin | +0.618 |
ription | +0.616 |
romy | +0.569 |
psons | +0.557 |
Neg logit contributions | |
Pixel | -0.512 |
oise | -0.479 |
archy | -0.443 |
ould | -0.434 |
Maher | -0.428 |
Token | 4 |
Token ID | 19 |
Feature activation | +6.74e-02 |
Pos logit contributions | |
uracy | +0.340 |
opin | +0.328 |
ription | +0.325 |
romy | +0.301 |
psons | +0.293 |
Neg logit contributions | |
Pixel | -0.289 |
oise | -0.271 |
archy | -0.252 |
ould | -0.246 |
Maher | -0.244 |
Token | against |
Token ID | 1028 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
uracy | +0.176 |
opin | +0.169 |
ription | +0.169 |
romy | +0.156 |
psons | +0.152 |
Neg logit contributions | |
Pixel | -0.148 |
oise | -0.139 |
archy | -0.128 |
ould | -0.126 |
Maher | -0.124 |
Token | pre |
Token ID | 662 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
Pixel | +0.532 |
oise | +0.501 |
archy | +0.462 |
ould | +0.455 |
Maher | +0.453 |
Neg logit contributions | |
uracy | -0.603 |
ription | -0.577 |
opin | -0.577 |
romy | -0.528 |
psons | -0.518 |
Token | - |
Token ID | 12 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
Pixel | +0.905 |
oise | +0.855 |
archy | +0.793 |
ould | +0.782 |
Maher | +0.763 |
Neg logit contributions | |
uracy | -1.054 |
ription | -1.015 |
opin | -1.012 |
romy | -0.928 |
psons | -0.908 |
Token | agh |
Token ID | 10471 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
uracy | +0.619 |
opin | +0.595 |
ription | +0.590 |
romy | +0.546 |
inval | +0.532 |
Neg logit contributions | |
Pixel | -0.534 |
oise | -0.500 |
archy | -0.465 |
Maher | -0.454 |
ould | -0.453 |
Token | s |
Token ID | 82 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
Pixel | +2.779 |
oise | +2.634 |
archy | +2.469 |
ould | +2.446 |
Maher | +2.355 |
Neg logit contributions | |
uracy | -3.171 |
opin | -3.068 |
ription | -3.060 |
romy | -2.775 |
psons | -2.752 |
Token | inv |
Token ID | 800 |
Feature activation | +6.84e-02 |
Pos logit contributions | |
Pixel | +0.274 |
oise | +0.258 |
archy | +0.238 |
ould | +0.235 |
Maher | +0.231 |
Neg logit contributions | |
uracy | -0.330 |
opin | -0.317 |
ription | -0.315 |
romy | -0.291 |
psons | -0.284 |
Token | oker |
Token ID | 11020 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
uracy | +0.151 |
opin | +0.144 |
ription | +0.143 |
romy | +0.130 |
psons | +0.127 |
Neg logit contributions | |
Pixel | -0.178 |
oise | -0.168 |
archy | -0.158 |
Maher | -0.154 |
ould | -0.154 |
Token | is |
Token ID | 318 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
uracy | +1.167 |
opin | +1.126 |
ription | +1.116 |
romy | +1.035 |
inval | +1.011 |
Neg logit contributions | |
Pixel | -0.925 |
oise | -0.863 |
archy | -0.798 |
ould | -0.780 |
Maher | -0.773 |
Token | crucial |
Token ID | 8780 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
Pixel | +0.429 |
oise | +0.404 |
archy | +0.378 |
ould | +0.370 |
Maher | +0.365 |
Neg logit contributions | |
uracy | -0.493 |
ription | -0.471 |
opin | -0.469 |
romy | -0.435 |
psons | -0.424 |
Token | 3 |
Token ID | 513 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
uracy | +0.296 |
opin | +0.285 |
ription | +0.283 |
romy | +0.261 |
psons | +0.255 |
Neg logit contributions | |
Pixel | -0.254 |
oise | -0.238 |
archy | -0.220 |
ould | -0.216 |
Maher | -0.215 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Pixel | +0.092 |
oise | +0.087 |
archy | +0.081 |
ould | +0.079 |
Maher | +0.079 |
Neg logit contributions | |
uracy | -0.088 |
opin | -0.085 |
ription | -0.085 |
romy | -0.077 |
inval | -0.076 |
Token | 5 |
Token ID | 20 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
Pixel | +0.235 |
oise | +0.219 |
archy | +0.200 |
ould | +0.197 |
Maher | +0.195 |
Neg logit contributions | |
uracy | -0.294 |
opin | -0.286 |
ription | -0.284 |
romy | -0.262 |
inval | -0.258 |
Token | - |
Token ID | 12 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
uracy | +0.997 |
opin | +0.959 |
ription | +0.949 |
romy | +0.876 |
psons | +0.856 |
Neg logit contributions | |
Pixel | -0.905 |
oise | -0.850 |
archy | -0.796 |
ould | -0.776 |
Maher | -0.768 |
Token | liter |
Token ID | 17201 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
uracy | +0.486 |
opin | +0.462 |
ription | +0.457 |
romy | +0.399 |
inval | +0.394 |
Neg logit contributions | |
Pixel | -0.848 |
oise | -0.807 |
archy | -0.763 |
ould | -0.751 |
Maher | -0.742 |
Token | Eco |
Token ID | 38719 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Pixel | +0.593 |
oise | +0.558 |
archy | +0.510 |
ould | +0.502 |
Maher | +0.499 |
Neg logit contributions | |
uracy | -0.733 |
opin | -0.704 |
ription | -0.701 |
romy | -0.651 |
inval | -0.634 |
Token | Boost |
Token ID | 45686 |
Feature activation | -7.41e-01 |
Pos logit contributions | |
Pixel | +2.167 |
oise | +1.984 |
archy | +1.772 |
ould | +1.727 |
Maher | +1.679 |
Neg logit contributions | |
uracy | -3.708 |
opin | -3.656 |
ription | -3.641 |
romy | -3.355 |
inval | -3.328 |
Token | proved |
Token ID | 8302 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
Pixel | +1.671 |
oise | +1.584 |
archy | +1.440 |
ould | +1.425 |
Maher | +1.410 |
Neg logit contributions | |
uracy | -1.884 |
opin | -1.813 |
ription | -1.801 |
romy | -1.664 |
inval | -1.617 |
Token | its |
Token ID | 663 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
Pixel | +0.561 |
oise | +0.530 |
archy | +0.489 |
Maher | +0.477 |
ould | +0.476 |
Neg logit contributions | |
uracy | -0.663 |
ription | -0.639 |
opin | -0.639 |
romy | -0.587 |
psons | -0.576 |
Token | met |
Token ID | 1138 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
uracy | +1.015 |
opin | +0.976 |
ription | +0.967 |
romy | +0.895 |
inval | +0.874 |
Neg logit contributions | |
Pixel | -0.857 |
oise | -0.801 |
archy | -0.743 |
ould | -0.728 |
Maher | -0.721 |
Token | tle |
Token ID | 7100 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
uracy | +0.454 |
opin | +0.432 |
ription | +0.428 |
romy | +0.392 |
psons | +0.379 |
Neg logit contributions | |
Pixel | -0.529 |
oise | -0.500 |
archy | -0.469 |
ould | -0.460 |
Maher | -0.458 |
Token | $ |
Token ID | 720 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
Pixel | +0.791 |
oise | +0.729 |
archy | +0.672 |
Maher | +0.663 |
ould | +0.659 |
Neg logit contributions | |
uracy | -1.132 |
opin | -1.082 |
ription | -1.081 |
romy | -1.006 |
psons | -0.980 |
Token | 300 |
Token ID | 6200 |
Feature activation | -6.80e-02 |
Pos logit contributions | |
Pixel | +0.947 |
oise | +0.874 |
archy | +0.803 |
ould | +0.785 |
Maher | +0.775 |
Neg logit contributions | |
uracy | -1.326 |
opin | -1.293 |
ription | -1.278 |
romy | -1.181 |
inval | -1.167 |
Token | , |
Token ID | 11 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Pixel | +0.154 |
oise | +0.145 |
archy | +0.135 |
ould | +0.132 |
Maher | +0.131 |
Neg logit contributions | |
uracy | -0.173 |
opin | -0.166 |
ription | -0.165 |
romy | -0.152 |
psons | -0.148 |
Token | while |
Token ID | 981 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
Pixel | +0.626 |
oise | +0.584 |
archy | +0.541 |
ould | +0.528 |
Maher | +0.524 |
Neg logit contributions | |
uracy | -0.809 |
opin | -0.783 |
ription | -0.775 |
romy | -0.718 |
psons | -0.703 |
Token | five |
Token ID | 1936 |
Feature activation | -9.32e-02 |
Pos logit contributions | |
uracy | +1.046 |
opin | +1.013 |
ription | +1.001 |
romy | +0.931 |
inval | +0.914 |
Neg logit contributions | |
Pixel | -0.816 |
oise | -0.766 |
archy | -0.706 |
ould | -0.691 |
Maher | -0.677 |
Token | backers |
Token ID | 18917 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Pixel | +0.198 |
oise | +0.185 |
archy | +0.171 |
ould | +0.168 |
Maher | +0.166 |
Neg logit contributions | |
uracy | -0.249 |
opin | -0.240 |
ription | -0.238 |
romy | -0.221 |
psons | -0.216 |
Token | opted |
Token ID | 18524 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
Pixel | +2.473 |
oise | +2.281 |
archy | +2.138 |
Maher | +2.117 |
ould | +2.107 |
Neg logit contributions | |
uracy | -2.977 |
ription | -2.823 |
opin | -2.819 |
romy | -2.596 |
psons | -2.543 |
Token | for |
Token ID | 329 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
Pixel | +0.453 |
oise | +0.427 |
archy | +0.398 |
ould | +0.390 |
Maher | +0.387 |
Neg logit contributions | |
uracy | -0.462 |
opin | -0.443 |
ription | -0.440 |
romy | -0.404 |
inval | -0.394 |
Token | the |
Token ID | 262 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
Pixel | +0.129 |
oise | +0.121 |
archy | +0.113 |
ould | +0.111 |
Maher | +0.110 |
Neg logit contributions | |
uracy | -0.139 |
opin | -0.134 |
ription | -0.133 |
romy | -0.122 |
psons | -0.119 |
Token | most |
Token ID | 749 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
uracy | +0.192 |
opin | +0.185 |
ription | +0.184 |
romy | +0.171 |
inval | +0.167 |
Neg logit contributions | |
Pixel | -0.147 |
oise | -0.137 |
archy | -0.127 |
ould | -0.124 |
Maher | -0.122 |
Token | lavish |
Token ID | 38957 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
uracy | +0.644 |
opin | +0.621 |
ription | +0.614 |
romy | +0.568 |
inval | +0.555 |
Neg logit contributions | |
Pixel | -0.569 |
oise | -0.534 |
archy | -0.496 |
ould | -0.486 |
Maher | -0.481 |