Token | respected |
Token ID | 14462 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
ッド | +1.933 |
eus | +1.832 |
ushima | +1.817 |
ptoms | +1.766 |
lyak | +1.765 |
Neg logit contributions | |
rewritten | -1.591 |
rewriting | -1.382 |
dodged | -1.337 |
acci | -1.320 |
disreg | -1.277 |
Token | her |
Token ID | 607 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
rewritten | +0.979 |
rewriting | +0.859 |
dodged | +0.833 |
acci | +0.811 |
disreg | +0.804 |
Neg logit contributions | |
ッド | -1.124 |
ushima | -1.060 |
eus | -1.049 |
ptoms | -1.030 |
lyak | -1.019 |
Token | . |
Token ID | 13 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
ッド | +0.554 |
eus | +0.520 |
ushima | +0.520 |
ptoms | +0.503 |
lyak | +0.500 |
Neg logit contributions | |
rewritten | -0.513 |
rewriting | -0.449 |
dodged | -0.436 |
acci | -0.426 |
disreg | -0.420 |
Token | We |
Token ID | 775 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
ッド | +0.935 |
eus | +0.875 |
ushima | +0.871 |
ptoms | +0.840 |
lyak | +0.838 |
Neg logit contributions | |
rewritten | -1.025 |
rewriting | -0.906 |
dodged | -0.885 |
acci | -0.869 |
disreg | -0.853 |
Token | would |
Token ID | 561 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
ッド | +0.324 |
eus | +0.306 |
ushima | +0.305 |
ptoms | +0.296 |
lyak | +0.294 |
Neg logit contributions | |
rewritten | -0.237 |
rewriting | -0.202 |
dodged | -0.198 |
acci | -0.190 |
disreg | -0.189 |
Token | fart |
Token ID | 15189 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
ッド | +0.156 |
eus | +0.144 |
ushima | +0.144 |
ptoms | +0.139 |
lyak | +0.137 |
Neg logit contributions | |
rewritten | -0.174 |
rewriting | -0.154 |
dodged | -0.151 |
acci | -0.147 |
disreg | -0.146 |
Token | around |
Token ID | 1088 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
rewritten | +7.577 |
rewriting | +6.629 |
acci | +6.553 |
dodged | +6.529 |
disreg | +6.403 |
Neg logit contributions | |
ッド | -10.452 |
ushima | -9.847 |
ptoms | -9.692 |
lyak | -9.434 |
eus | -9.363 |
Token | a |
Token ID | 257 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
rewritten | +3.456 |
rewriting | +3.058 |
dodged | +2.925 |
disreg | +2.852 |
travers | +2.836 |
Neg logit contributions | |
ッド | -4.318 |
ushima | -4.085 |
eus | -4.016 |
ptoms | -3.971 |
lyak | -3.938 |
Token | lot |
Token ID | 1256 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
rewritten | +3.728 |
rewriting | +3.227 |
dodged | +3.134 |
disreg | +3.012 |
travers | +2.997 |
Neg logit contributions | |
ッド | -4.295 |
ushima | -4.087 |
eus | -4.024 |
ptoms | -3.980 |
lyak | -3.932 |
Token | and |
Token ID | 290 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
rewritten | +3.019 |
rewriting | +2.652 |
dodged | +2.573 |
acci | +2.507 |
disreg | +2.491 |
Neg logit contributions | |
ッド | -3.450 |
ushima | -3.260 |
eus | -3.241 |
lyak | -3.163 |
ptoms | -3.162 |
Token | waste |
Token ID | 7030 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
rewritten | +1.087 |
rewriting | +0.952 |
dodged | +0.930 |
acci | +0.900 |
disreg | +0.892 |
Neg logit contributions | |
ッド | -1.205 |
ushima | -1.132 |
eus | -1.130 |
ptoms | -1.092 |
lyak | -1.085 |
Token | index |
Token ID | 9630 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
rewritten | +1.915 |
rewriting | +1.605 |
dodged | +1.542 |
acci | +1.511 |
disreg | +1.442 |
Neg logit contributions | |
ッド | -3.335 |
eus | -3.190 |
ushima | -3.166 |
lyak | -3.095 |
ptoms | -3.072 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
rewritten | +4.889 |
rewriting | +3.986 |
acci | +3.928 |
dodged | +3.822 |
disreg | +3.697 |
Neg logit contributions | |
ッド | -6.172 |
eus | -5.556 |
ushima | -5.555 |
lyak | -5.479 |
ptoms | -5.453 |
Token | php |
Token ID | 10121 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
rewritten | +3.394 |
rewriting | +2.877 |
acci | +2.875 |
dodged | +2.770 |
postponed | +2.635 |
Neg logit contributions | |
ッド | -5.096 |
eus | -4.691 |
ushima | -4.602 |
ptoms | -4.526 |
lyak | -4.524 |
Token | ? |
Token ID | 30 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
rewritten | +4.691 |
rewriting | +3.953 |
dodged | +3.845 |
disreg | +3.604 |
acci | +3.595 |
Neg logit contributions | |
ッド | -6.481 |
eus | -6.117 |
ushima | -6.062 |
ptoms | -5.999 |
lyak | -5.888 |
Token | topic |
Token ID | 26652 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
rewritten | +1.961 |
rewriting | +1.355 |
acci | +1.285 |
dodged | +1.236 |
travers | +1.044 |
Neg logit contributions | |
ッド | -7.527 |
eus | -7.318 |
ushima | -7.221 |
lyak | -7.042 |
ptoms | -7.039 |
Token | = |
Token ID | 28 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
rewritten | +3.703 |
rewriting | +2.936 |
dodged | +2.887 |
acci | +2.802 |
disreg | +2.694 |
Neg logit contributions | |
ッド | -7.119 |
ushima | -6.796 |
eus | -6.785 |
ptoms | -6.619 |
rises | -6.561 |
Token | 28 |
Token ID | 2078 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
rewritten | +8.259 |
rewriting | +7.052 |
acci | +6.986 |
dodged | +6.916 |
disreg | +6.774 |
Neg logit contributions | |
ッド | -9.179 |
ushima | -8.640 |
ptoms | -8.516 |
eus | -8.451 |
lyak | -8.378 |
Token | 18 |
Token ID | 1507 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
rewritten | +4.762 |
rewriting | +4.170 |
acci | +4.049 |
dodged | +4.035 |
disreg | +3.969 |
Neg logit contributions | |
ッド | -4.515 |
ushima | -4.207 |
eus | -4.132 |
lyak | -4.114 |
ptoms | -4.062 |
Token | 48 |
Token ID | 2780 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
rewritten | +4.816 |
rewriting | +4.235 |
dodged | +4.090 |
acci | +4.064 |
disreg | +4.021 |
Neg logit contributions | |
ッド | -4.551 |
ushima | -4.222 |
eus | -4.174 |
lyak | -4.085 |
ptoms | -4.083 |
Token | . |
Token ID | 13 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
rewritten | +2.308 |
rewriting | +2.039 |
dodged | +1.983 |
acci | +1.938 |
disreg | +1.917 |
Neg logit contributions | |
ッド | -2.187 |
eus | -2.037 |
ushima | -2.026 |
lyak | -1.958 |
ptoms | -1.951 |
Token | 0 |
Token ID | 15 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
rewritten | +3.722 |
rewriting | +3.243 |
dodged | +3.177 |
acci | +3.144 |
disreg | +3.013 |
Neg logit contributions | |
ッド | -4.232 |
ushima | -3.956 |
eus | -3.947 |
ptoms | -3.805 |
lyak | -3.798 |
Token | . |
Token ID | 13 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
ッド | +3.374 |
ushima | +3.190 |
eus | +3.186 |
lyak | +3.035 |
ptoms | +3.024 |
Neg logit contributions | |
rewritten | -3.689 |
rewriting | -3.251 |
dodged | -3.195 |
acci | -3.070 |
disreg | -3.044 |
Token |
|
Token ID | 198 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
ッド | +1.165 |
eus | +1.094 |
ushima | +1.085 |
ptoms | +1.047 |
lyak | +1.043 |
Neg logit contributions | |
rewritten | -1.193 |
rewriting | -1.053 |
dodged | -1.030 |
acci | -1.004 |
disreg | -0.986 |
Token |
|
Token ID | 198 |
Feature activation | +4.91e-04 |
Pos logit contributions | |
rewritten | +1.184 |
rewriting | +1.015 |
dodged | +0.981 |
acci | +0.968 |
disreg | +0.954 |
Neg logit contributions | |
ッド | -1.435 |
ushima | -1.349 |
eus | -1.336 |
ptoms | -1.309 |
lyak | -1.300 |
Token | The |
Token ID | 464 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
rewritten | +0.002 |
rewriting | +0.002 |
dodged | +0.002 |
acci | +0.002 |
disreg | +0.002 |
Neg logit contributions | |
ッド | -0.001 |
ushima | -0.001 |
eus | -0.001 |
lyak | -0.001 |
ptoms | -0.001 |
Token | book |
Token ID | 1492 |
Feature activation | +9.27e-01 |
Pos logit contributions | |
ッド | +1.130 |
eus | +1.067 |
ushima | +1.064 |
ptoms | +1.029 |
lyak | +1.025 |
Neg logit contributions | |
rewritten | -0.957 |
rewriting | -0.832 |
dodged | -0.814 |
acci | -0.795 |
disreg | -0.775 |
Token | Haunted |
Token ID | 43884 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
rewritten | +2.747 |
rewriting | +2.368 |
dodged | +2.285 |
acci | +2.222 |
disreg | +2.193 |
Neg logit contributions | |
ッド | -3.252 |
ushima | -3.065 |
eus | -3.061 |
lyak | -2.973 |
ptoms | -2.964 |
Token | Se |
Token ID | 1001 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
rewritten | +8.257 |
rewriting | +7.222 |
dodged | +7.098 |
acci | +6.838 |
postponed | +6.750 |
Neg logit contributions | |
ッド | -8.807 |
ushima | -8.374 |
eus | -8.290 |
ptoms | -8.145 |
lyak | -8.134 |
Token | gu |
Token ID | 5162 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
rewritten | +1.457 |
rewriting | +1.305 |
dodged | +1.281 |
acci | +1.262 |
disreg | +1.236 |
Neg logit contributions | |
ッド | -1.051 |
eus | -0.973 |
ushima | -0.963 |
lyak | -0.924 |
ptoms | -0.923 |
Token | in |
Token ID | 259 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
rewritten | +0.607 |
rewriting | +0.533 |
dodged | +0.521 |
acci | +0.510 |
disreg | +0.500 |
Neg logit contributions | |
ッド | -0.611 |
eus | -0.574 |
ushima | -0.571 |
ptoms | -0.552 |
lyak | -0.550 |
Token | traces |
Token ID | 20675 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
rewritten | +3.472 |
rewriting | +2.979 |
dodged | +2.920 |
acci | +2.830 |
disreg | +2.799 |
Neg logit contributions | |
ッド | -4.008 |
eus | -3.774 |
ushima | -3.766 |
lyak | -3.658 |
ptoms | -3.649 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
rewritten | +1.440 |
rewriting | +1.268 |
dodged | +1.221 |
acci | +1.201 |
travers | +1.180 |
Neg logit contributions | |
ッド | -1.495 |
eus | -1.408 |
ushima | -1.404 |
ptoms | -1.362 |
lyak | -1.361 |
Token | Q |
Token ID | 48 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
ッド | +0.743 |
eus | +0.682 |
ushima | +0.677 |
ptoms | +0.644 |
lyak | +0.642 |
Neg logit contributions | |
rewritten | -1.283 |
rewriting | -1.161 |
dodged | -1.140 |
acci | -1.122 |
disreg | -1.105 |
Token | ARS |
Token ID | 27415 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
ッド | +1.025 |
eus | +0.900 |
ushima | +0.881 |
ptoms | +0.817 |
lyak | +0.812 |
Neg logit contributions | |
rewritten | -3.175 |
rewriting | -2.922 |
dodged | -2.882 |
acci | -2.842 |
disreg | -2.804 |
Token | ). |
Token ID | 737 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
rewritten | +0.509 |
rewriting | +0.454 |
dodged | +0.443 |
acci | +0.436 |
disreg | +0.430 |
Neg logit contributions | |
ッド | -0.390 |
ushima | -0.361 |
eus | -0.360 |
lyak | -0.346 |
ptoms | -0.345 |
Token | Atlanta |
Token ID | 9371 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
rewritten | +0.226 |
rewriting | +0.197 |
dodged | +0.192 |
acci | +0.187 |
disreg | +0.184 |
Neg logit contributions | |
ッド | -0.259 |
eus | -0.244 |
ushima | -0.243 |
ptoms | -0.236 |
lyak | -0.235 |
Token | , |
Token ID | 11 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
ッド | +0.134 |
eus | +0.126 |
ushima | +0.125 |
ptoms | +0.121 |
lyak | +0.120 |
Neg logit contributions | |
rewritten | -0.134 |
rewriting | -0.118 |
dodged | -0.115 |
acci | -0.113 |
disreg | -0.110 |
Token | GA |
Token ID | 14545 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
rewritten | +0.845 |
rewriting | +0.741 |
dodged | +0.723 |
acci | +0.707 |
disreg | +0.694 |
Neg logit contributions | |
ッド | -0.876 |
eus | -0.823 |
ushima | -0.819 |
ptoms | -0.793 |
lyak | -0.791 |
Token | : |
Token ID | 25 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
rewritten | +7.809 |
rewriting | +6.763 |
dodged | +6.626 |
travers | +6.465 |
disreg | +6.444 |
Neg logit contributions | |
ッド | -8.982 |
ushima | -8.320 |
lyak | -8.290 |
ptoms | -8.256 |
eus | -8.216 |
Token | US |
Token ID | 1294 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
ッド | +0.115 |
eus | +0.109 |
ushima | +0.108 |
ptoms | +0.105 |
lyak | +0.104 |
Neg logit contributions | |
rewritten | -0.100 |
rewriting | -0.087 |
dodged | -0.085 |
acci | -0.083 |
disreg | -0.081 |
Token | Department |
Token ID | 2732 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
rewritten | +1.031 |
rewriting | +0.880 |
dodged | +0.861 |
acci | +0.836 |
disreg | +0.818 |
Neg logit contributions | |
ッド | -1.474 |
eus | -1.393 |
ushima | -1.389 |
ptoms | -1.352 |
lyak | -1.350 |
Token | of |
Token ID | 286 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
rewritten | +1.939 |
rewriting | +1.619 |
dodged | +1.575 |
acci | +1.544 |
disreg | +1.507 |
Neg logit contributions | |
ッド | -3.030 |
ushima | -2.862 |
eus | -2.839 |
lyak | -2.766 |
ptoms | -2.766 |
Token | Health |
Token ID | 3893 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
rewritten | +1.620 |
rewriting | +1.444 |
dodged | +1.417 |
acci | +1.390 |
disreg | +1.366 |
Neg logit contributions | |
ッド | -1.276 |
eus | -1.186 |
ushima | -1.182 |
ptoms | -1.136 |
lyak | -1.134 |
Token | a |
Token ID | 257 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
ッド | +2.643 |
eus | +2.479 |
ushima | +2.473 |
ptoms | +2.369 |
lyak | +2.365 |
Neg logit contributions | |
rewritten | -2.747 |
rewriting | -2.428 |
dodged | -2.346 |
acci | -2.328 |
disreg | -2.260 |
Token | moment |
Token ID | 2589 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
ッド | +0.656 |
eus | +0.617 |
ushima | +0.615 |
ptoms | +0.594 |
lyak | +0.593 |
Neg logit contributions | |
rewritten | -0.616 |
rewriting | -0.539 |
dodged | -0.526 |
acci | -0.515 |
disreg | -0.504 |
Token | to |
Token ID | 284 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
rewritten | +1.468 |
rewriting | +1.281 |
dodged | +1.248 |
acci | +1.225 |
disreg | +1.195 |
Neg logit contributions | |
ッド | -1.573 |
eus | -1.479 |
ushima | -1.472 |
ptoms | -1.428 |
lyak | -1.421 |
Token | discuss |
Token ID | 2112 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
rewritten | +1.009 |
rewriting | +0.891 |
dodged | +0.879 |
acci | +0.846 |
disreg | +0.845 |
Neg logit contributions | |
ッド | -0.957 |
eus | -0.881 |
ushima | -0.880 |
ptoms | -0.849 |
lyak | -0.839 |
Token | how |
Token ID | 703 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
rewritten | +1.263 |
rewriting | +1.116 |
dodged | +1.090 |
acci | +1.057 |
disreg | +1.048 |
Neg logit contributions | |
ッド | -1.280 |
eus | -1.196 |
ushima | -1.194 |
ptoms | -1.160 |
lyak | -1.148 |
Token | long |
Token ID | 890 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
ッド | +1.756 |
eus | +1.662 |
ushima | +1.639 |
lyak | +1.592 |
ptoms | +1.583 |
Neg logit contributions | |
rewritten | -1.664 |
rewriting | -1.449 |
dodged | -1.417 |
acci | -1.399 |
disreg | -1.352 |
Token | EV |
Token ID | 8696 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
rewritten | +7.804 |
rewriting | +7.013 |
dodged | +6.838 |
postponed | +6.599 |
travers | +6.591 |
Neg logit contributions | |
ッド | -8.956 |
ushima | -8.425 |
ptoms | -8.335 |
eus | -8.284 |
lyak | -7.960 |
Token | O |
Token ID | 46 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
rewritten | +3.319 |
rewriting | +2.984 |
dodged | +2.930 |
acci | +2.911 |
disreg | +2.884 |
Neg logit contributions | |
ッド | -1.631 |
ushima | -1.455 |
eus | -1.412 |
ptoms | -1.400 |
lyak | -1.360 |
Token | Japan |
Token ID | 2869 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
rewritten | +2.516 |
rewriting | +2.141 |
dodged | +2.102 |
acci | +1.998 |
disreg | +1.979 |
Neg logit contributions | |
ッド | -3.896 |
eus | -3.696 |
ushima | -3.663 |
ptoms | -3.591 |
lyak | -3.567 |
Token | has |
Token ID | 468 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ッド | +0.320 |
eus | +0.304 |
ushima | +0.303 |
lyak | +0.293 |
ptoms | +0.292 |
Neg logit contributions | |
rewritten | -0.257 |
rewriting | -0.221 |
dodged | -0.214 |
acci | -0.212 |
disreg | -0.204 |
Token | been |
Token ID | 587 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
rewritten | +2.737 |
dodged | +2.183 |
rewriting | +2.168 |
disreg | +1.989 |
postponed | +1.980 |
Neg logit contributions | |
ッド | -6.002 |
eus | -5.666 |
ushima | -5.666 |
ptoms | -5.567 |
lyak | -5.526 |
Token | ( |
Token ID | 357 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
rewritten | +3.322 |
rewriting | +2.945 |
acci | +2.903 |
dodged | +2.875 |
travers | +2.786 |
Neg logit contributions | |
ッド | -3.112 |
ushima | -2.870 |
eus | -2.797 |
ptoms | -2.788 |
lyak | -2.712 |
Token | New |
Token ID | 3791 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ッド | +0.794 |
eus | +0.730 |
ushima | +0.728 |
ptoms | +0.696 |
lyak | +0.693 |
Neg logit contributions | |
rewritten | -1.227 |
rewriting | -1.105 |
dodged | -1.083 |
acci | -1.065 |
disreg | -1.048 |
Token | York |
Token ID | 1971 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
ッド | +4.121 |
eus | +3.909 |
ushima | +3.806 |
lyak | +3.634 |
ptoms | +3.625 |
Neg logit contributions | |
rewritten | -5.694 |
rewriting | -5.095 |
dodged | -5.009 |
acci | -4.918 |
disreg | -4.841 |
Token | : |
Token ID | 25 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
ッド | +2.111 |
eus | +1.980 |
ushima | +1.945 |
ptoms | +1.833 |
lyak | +1.828 |
Neg logit contributions | |
rewritten | -2.983 |
rewriting | -2.687 |
dodged | -2.631 |
acci | -2.591 |
disreg | -2.525 |
Token | Sh |
Token ID | 911 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
rewritten | +1.409 |
rewriting | +1.227 |
dodged | +1.190 |
acci | +1.168 |
disreg | +1.158 |
Neg logit contributions | |
ッド | -1.513 |
ushima | -1.431 |
eus | -1.413 |
ptoms | -1.394 |
lyak | -1.385 |
Token | ilo |
Token ID | 18526 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
rewritten | +2.477 |
rewriting | +2.210 |
dodged | +2.166 |
acci | +2.142 |
travers | +2.078 |
Neg logit contributions | |
ッド | -2.025 |
ushima | -1.848 |
eus | -1.844 |
ptoms | -1.791 |
lyak | -1.764 |
Token | Publishing |
Token ID | 23499 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
rewritten | +8.008 |
rewriting | +6.987 |
acci | +6.803 |
dodged | +6.802 |
Pratt | +6.723 |
Neg logit contributions | |
ッド | -8.313 |
ushima | -8.110 |
ptoms | -7.867 |
lyak | -7.650 |
rises | -7.608 |
Token | House |
Token ID | 2097 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
rewritten | +0.099 |
rewriting | +0.087 |
dodged | +0.085 |
acci | +0.083 |
disreg | +0.082 |
Neg logit contributions | |
ッド | -0.094 |
eus | -0.088 |
ushima | -0.088 |
ptoms | -0.084 |
lyak | -0.084 |
Token | , |
Token ID | 11 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
rewritten | +0.742 |
rewriting | +0.648 |
dodged | +0.630 |
acci | +0.619 |
disreg | +0.605 |
Neg logit contributions | |
ッド | -0.813 |
eus | -0.763 |
ushima | -0.762 |
ptoms | -0.739 |
lyak | -0.736 |
Token | 1971 |
Token ID | 16382 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
rewritten | +0.299 |
rewriting | +0.255 |
dodged | +0.246 |
acci | +0.239 |
disreg | +0.236 |
Neg logit contributions | |
ッド | -0.424 |
ushima | -0.401 |
eus | -0.400 |
ptoms | -0.391 |
lyak | -0.388 |
Token | ), |
Token ID | 828 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
ッド | +1.462 |
eus | +1.354 |
ushima | +1.336 |
ptoms | +1.272 |
lyak | +1.270 |
Neg logit contributions | |
rewritten | -2.240 |
rewriting | -2.020 |
dodged | -1.993 |
acci | -1.955 |
disreg | -1.922 |
Token | ably |
Token ID | 1346 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
rewritten | +4.069 |
rewriting | +3.610 |
acci | +3.556 |
dodged | +3.552 |
disreg | +3.445 |
Neg logit contributions | |
ッド | -3.108 |
ushima | -2.875 |
eus | -2.830 |
lyak | -2.780 |
ptoms | -2.709 |
Token | gives |
Token ID | 3607 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
rewritten | +1.992 |
rewriting | +1.671 |
dodged | +1.639 |
acci | +1.553 |
disreg | +1.548 |
Neg logit contributions | |
ッド | -3.151 |
ushima | -3.001 |
eus | -2.976 |
lyak | -2.915 |
ptoms | -2.914 |
Token | his |
Token ID | 465 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
ッド | +0.856 |
eus | +0.807 |
ushima | +0.797 |
ptoms | +0.771 |
lyak | +0.770 |
Neg logit contributions | |
rewritten | -0.814 |
rewriting | -0.712 |
dodged | -0.696 |
acci | -0.682 |
disreg | -0.666 |
Token | penis |
Token ID | 16360 |
Feature activation | +5.18e-03 |
Pos logit contributions | |
rewritten | +1.231 |
rewriting | +1.079 |
dodged | +1.047 |
acci | +1.023 |
disreg | +1.006 |
Neg logit contributions | |
ッド | -1.313 |
ushima | -1.234 |
eus | -1.232 |
ptoms | -1.196 |
lyak | -1.186 |
Token | in |
Token ID | 287 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
rewritten | +0.016 |
rewriting | +0.014 |
dodged | +0.014 |
acci | +0.013 |
disreg | +0.013 |
Neg logit contributions | |
ッド | -0.017 |
ushima | -0.016 |
eus | -0.016 |
ptoms | -0.016 |
lyak | -0.016 |
Token | Forever |
Token ID | 28285 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
ッド | +0.818 |
eus | +0.770 |
ushima | +0.765 |
ptoms | +0.740 |
lyak | +0.738 |
Neg logit contributions | |
rewritten | -0.772 |
rewriting | -0.676 |
dodged | -0.660 |
acci | -0.646 |
disreg | -0.632 |
Token | . |
Token ID | 13 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
rewritten | +8.731 |
rewriting | +7.698 |
dodged | +7.249 |
postponed | +7.121 |
travers | +7.113 |
Neg logit contributions | |
ッド | -7.979 |
ushima | -7.823 |
eus | -7.696 |
ptoms | -7.507 |
rises | -7.176 |
Token | But |
Token ID | 887 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
rewritten | +1.185 |
rewriting | +1.037 |
dodged | +1.001 |
acci | +0.989 |
disreg | +0.975 |
Neg logit contributions | |
ッド | -1.225 |
ushima | -1.159 |
eus | -1.153 |
ptoms | -1.126 |
lyak | -1.122 |
Token | in |
Token ID | 287 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
rewritten | +2.204 |
rewriting | +1.925 |
dodged | +1.847 |
acci | +1.794 |
disreg | +1.790 |
Neg logit contributions | |
ッド | -2.654 |
ushima | -2.499 |
eus | -2.493 |
ptoms | -2.427 |
lyak | -2.412 |
Token | the |
Token ID | 262 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
rewritten | +1.877 |
rewriting | +1.657 |
dodged | +1.589 |
acci | +1.560 |
disreg | +1.539 |
Neg logit contributions | |
ッド | -2.042 |
ushima | -1.922 |
eus | -1.919 |
ptoms | -1.867 |
lyak | -1.859 |
Token | main |
Token ID | 1388 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
rewritten | +1.578 |
rewriting | +1.385 |
dodged | +1.332 |
acci | +1.304 |
disreg | +1.285 |
Neg logit contributions | |
ッド | -1.686 |
ushima | -1.588 |
eus | -1.586 |
ptoms | -1.533 |
lyak | -1.531 |
Token | ? |
Token ID | 30 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
ッド | +0.164 |
eus | +0.155 |
ushima | +0.154 |
ptoms | +0.149 |
lyak | +0.149 |
Neg logit contributions | |
rewritten | -0.149 |
rewriting | -0.130 |
dodged | -0.127 |
acci | -0.124 |
disreg | -0.121 |
Token | Want |
Token ID | 16168 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ッド | +0.554 |
eus | +0.519 |
ushima | +0.516 |
ptoms | +0.499 |
lyak | +0.496 |
Neg logit contributions | |
rewritten | -0.608 |
rewriting | -0.538 |
dodged | -0.526 |
acci | -0.516 |
disreg | -0.507 |
Token | to |
Token ID | 284 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
rewritten | +3.095 |
rewriting | +2.685 |
dodged | +2.608 |
acci | +2.565 |
disreg | +2.505 |
Neg logit contributions | |
ッド | -3.657 |
ushima | -3.475 |
eus | -3.473 |
ptoms | -3.356 |
lyak | -3.340 |
Token | see |
Token ID | 766 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
rewritten | +1.007 |
rewriting | +0.905 |
dodged | +0.888 |
acci | +0.867 |
disreg | +0.864 |
Neg logit contributions | |
ッド | -0.685 |
eus | -0.630 |
ushima | -0.624 |
ptoms | -0.599 |
lyak | -0.591 |
Token | how |
Token ID | 703 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
rewritten | +0.283 |
rewriting | +0.241 |
dodged | +0.235 |
acci | +0.228 |
disreg | +0.224 |
Neg logit contributions | |
ッド | -0.410 |
eus | -0.388 |
ushima | -0.387 |
ptoms | -0.376 |
lyak | -0.375 |
Token | long |
Token ID | 890 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
ッド | +0.807 |
eus | +0.759 |
ushima | +0.754 |
ptoms | +0.729 |
lyak | +0.728 |
Neg logit contributions | |
rewritten | -0.786 |
rewriting | -0.690 |
dodged | -0.674 |
acci | -0.659 |
disreg | -0.646 |
Token | the |
Token ID | 262 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
rewritten | +7.764 |
rewriting | +6.816 |
dodged | +6.689 |
acci | +6.608 |
travers | +6.567 |
Neg logit contributions | |
ッド | -8.618 |
ushima | -8.016 |
eus | -7.880 |
ptoms | -7.862 |
lyak | -7.734 |
Token | line |
Token ID | 1627 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
rewritten | +3.660 |
rewriting | +3.190 |
dodged | +3.131 |
acci | +3.055 |
travers | +3.010 |
Neg logit contributions | |
ッド | -4.008 |
ushima | -3.765 |
eus | -3.730 |
ptoms | -3.627 |
lyak | -3.606 |
Token | is |
Token ID | 318 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
rewritten | +3.927 |
rewriting | +3.354 |
dodged | +3.285 |
acci | +3.179 |
travers | +3.143 |
Neg logit contributions | |
ッド | -5.652 |
ushima | -5.324 |
eus | -5.314 |
lyak | -5.205 |
ptoms | -5.190 |
Token | at |
Token ID | 379 |
Feature activation | -6.22e-04 |
Pos logit contributions | |
rewritten | +3.764 |
rewriting | +3.267 |
dodged | +3.152 |
acci | +3.063 |
travers | +3.028 |
Neg logit contributions | |
ッド | -4.854 |
ushima | -4.571 |
eus | -4.545 |
ptoms | -4.466 |
lyak | -4.415 |
Token | your |
Token ID | 534 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
ッド | +0.002 |
eus | +0.002 |
ushima | +0.002 |
ptoms | +0.002 |
lyak | +0.002 |
Neg logit contributions | |
rewritten | -0.002 |
rewriting | -0.002 |
dodged | -0.002 |
acci | -0.002 |
disreg | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -3.66e-03 |
Pos logit contributions | |
ッド | +3.431 |
eus | +3.257 |
ushima | +3.220 |
ptoms | +3.086 |
lyak | +3.080 |
Neg logit contributions | |
rewritten | -4.225 |
rewriting | -3.778 |
acci | -3.693 |
dodged | -3.690 |
disreg | -3.542 |
Token |
|
Token ID | 198 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ッド | +0.012 |
eus | +0.011 |
ushima | +0.011 |
lyak | +0.011 |
ptoms | +0.011 |
Neg logit contributions | |
rewritten | -0.012 |
rewriting | -0.010 |
dodged | -0.010 |
acci | -0.010 |
disreg | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
rewritten | +0.509 |
rewriting | +0.439 |
dodged | +0.427 |
acci | +0.419 |
disreg | +0.412 |
Neg logit contributions | |
ッド | -0.591 |
ushima | -0.554 |
eus | -0.553 |
ptoms | -0.537 |
lyak | -0.535 |
Token | Shares |
Token ID | 43566 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
ッド | +0.435 |
ushima | +0.398 |
eus | +0.397 |
lyak | +0.380 |
ptoms | +0.377 |
Neg logit contributions | |
rewritten | -0.719 |
rewriting | -0.648 |
dodged | -0.635 |
acci | -0.627 |
disreg | -0.617 |
Token | of |
Token ID | 286 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
ッド | +0.230 |
eus | +0.218 |
ushima | +0.216 |
ptoms | +0.210 |
lyak | +0.210 |
Neg logit contributions | |
rewritten | -0.169 |
rewriting | -0.145 |
dodged | -0.141 |
acci | -0.137 |
disreg | -0.134 |
Token | Alt |
Token ID | 12344 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
ッド | +1.006 |
eus | +0.917 |
ushima | +0.907 |
ptoms | +0.857 |
lyak | +0.854 |
Neg logit contributions | |
rewritten | -2.103 |
rewriting | -1.912 |
dodged | -1.884 |
acci | -1.862 |
disreg | -1.826 |
Token | ria |
Token ID | 7496 |
Feature activation | +5.79e-01 |
Pos logit contributions | |
rewritten | +5.022 |
acci | +4.121 |
dodged | +4.104 |
rewriting | +3.924 |
disreg | +3.873 |
Neg logit contributions | |
ッド | -11.116 |
ptoms | -10.194 |
ushima | -10.050 |
GOODMAN | -10.022 |
lyak | -9.892 |
Token | Group |
Token ID | 4912 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
rewritten | +1.753 |
dodged | +1.521 |
rewriting | +1.519 |
acci | +1.452 |
disreg | +1.441 |
Neg logit contributions | |
ッド | -1.981 |
eus | -1.860 |
ushima | -1.851 |
ptoms | -1.790 |
lyak | -1.788 |
Token | , |
Token ID | 11 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
rewritten | +3.383 |
dodged | +2.989 |
rewriting | +2.979 |
disreg | +2.851 |
acci | +2.842 |
Neg logit contributions | |
ッド | -3.150 |
eus | -2.914 |
ushima | -2.901 |
lyak | -2.812 |
ptoms | -2.785 |
Token | maker |
Token ID | 16009 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
ッド | +0.162 |
eus | +0.151 |
ushima | +0.151 |
ptoms | +0.146 |
lyak | +0.145 |
Neg logit contributions | |
rewritten | -0.178 |
rewriting | -0.158 |
dodged | -0.154 |
acci | -0.151 |
disreg | -0.149 |
Token | of |
Token ID | 286 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
rewritten | +0.069 |
rewriting | +0.059 |
dodged | +0.057 |
acci | +0.055 |
disreg | +0.054 |
Neg logit contributions | |
ッド | -0.096 |
eus | -0.091 |
ushima | -0.091 |
ptoms | -0.088 |
lyak | -0.088 |
Token | is |
Token ID | 318 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
ッド | +0.949 |
eus | +0.901 |
ushima | +0.894 |
ptoms | +0.867 |
lyak | +0.866 |
Neg logit contributions | |
rewritten | -0.742 |
rewriting | -0.639 |
dodged | -0.622 |
acci | -0.608 |
disreg | -0.590 |
Token | about |
Token ID | 546 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
rewritten | +1.217 |
rewriting | +1.048 |
dodged | +1.017 |
acci | +0.986 |
disreg | +0.977 |
Neg logit contributions | |
ッド | -1.555 |
ushima | -1.468 |
eus | -1.461 |
ptoms | -1.424 |
lyak | -1.411 |
Token | 67 |
Token ID | 8275 |
Feature activation | +8.26e-01 |
Pos logit contributions | |
rewritten | +0.181 |
rewriting | +0.158 |
dodged | +0.154 |
acci | +0.150 |
disreg | +0.147 |
Neg logit contributions | |
ッド | -0.199 |
eus | -0.186 |
ushima | -0.186 |
ptoms | -0.180 |
lyak | -0.179 |
Token | , |
Token ID | 11 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
rewritten | +2.428 |
rewriting | +2.116 |
dodged | +2.043 |
travers | +1.978 |
acci | +1.975 |
Neg logit contributions | |
ッド | -2.896 |
ushima | -2.759 |
eus | -2.701 |
ptoms | -2.646 |
lyak | -2.610 |
Token | 000 |
Token ID | 830 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
rewritten | +1.296 |
rewriting | +1.142 |
dodged | +1.116 |
acci | +1.089 |
disreg | +1.076 |
Neg logit contributions | |
ッド | -1.247 |
ushima | -1.163 |
eus | -1.163 |
lyak | -1.125 |
ptoms | -1.125 |
Token | mph |
Token ID | 16462 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
rewritten | +2.331 |
rewriting | +1.902 |
dodged | +1.823 |
disreg | +1.727 |
travers | +1.716 |
Neg logit contributions | |
ッド | -4.886 |
ushima | -4.724 |
eus | -4.635 |
ptoms | -4.568 |
lyak | -4.522 |
Token | ( |
Token ID | 357 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
rewritten | +7.166 |
rewriting | +6.236 |
dodged | +6.095 |
travers | +5.982 |
disreg | +5.957 |
Neg logit contributions | |
ッド | -9.161 |
ushima | -8.576 |
eus | -8.311 |
ptoms | -8.295 |
lyak | -8.216 |
Token | 107 |
Token ID | 15982 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
rewritten | +2.883 |
rewriting | +2.536 |
dodged | +2.464 |
acci | +2.437 |
disreg | +2.393 |
Neg logit contributions | |
ッド | -2.689 |
ushima | -2.494 |
ptoms | -2.440 |
eus | -2.435 |
lyak | -2.401 |
Token | , |
Token ID | 11 |
Feature activation | +5.21e-02 |
Pos logit contributions | |
rewritten | +0.513 |
rewriting | +0.450 |
dodged | +0.439 |
acci | +0.430 |
disreg | +0.420 |
Neg logit contributions | |
ッド | -0.537 |
eus | -0.506 |
ushima | -0.502 |
ptoms | -0.486 |
lyak | -0.485 |
Token | 000 |
Token ID | 830 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
rewritten | +0.177 |
rewriting | +0.157 |
dodged | +0.154 |
acci | +0.150 |
disreg | +0.148 |
Neg logit contributions | |
ッド | -0.159 |
eus | -0.149 |
ushima | -0.148 |
ptoms | -0.142 |
lyak | -0.142 |
Token | km |
Token ID | 10571 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
rewritten | +1.215 |
rewriting | +0.998 |
dodged | +0.975 |
acci | +0.923 |
disreg | +0.912 |
Neg logit contributions | |
ッド | -2.391 |
ushima | -2.277 |
eus | -2.277 |
ptoms | -2.226 |
lyak | -2.202 |
Token |
|
Token ID | 198 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
rewritten | +1.129 |
rewriting | +0.989 |
dodged | +0.961 |
acci | +0.956 |
disreg | +0.938 |
Neg logit contributions | |
ッド | -1.160 |
ushima | -1.095 |
eus | -1.087 |
ptoms | -1.063 |
lyak | -1.048 |
Token |
|
Token ID | 198 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
rewritten | +0.914 |
rewriting | +0.784 |
dodged | +0.759 |
acci | +0.747 |
disreg | +0.738 |
Neg logit contributions | |
ッド | -1.107 |
ushima | -1.041 |
eus | -1.034 |
ptoms | -1.012 |
lyak | -1.004 |
Token | Craig |
Token ID | 40441 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
ッド | +0.437 |
eus | +0.404 |
ushima | +0.403 |
ptoms | +0.386 |
lyak | +0.384 |
Neg logit contributions | |
rewritten | -0.644 |
rewriting | -0.579 |
dodged | -0.566 |
acci | -0.560 |
disreg | -0.549 |
Token | : |
Token ID | 1058 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
ッド | +0.797 |
ushima | +0.743 |
eus | +0.743 |
ptoms | +0.708 |
lyak | +0.707 |
Neg logit contributions | |
rewritten | -0.940 |
rewriting | -0.835 |
dodged | -0.812 |
acci | -0.798 |
disreg | -0.783 |
Token | How |
Token ID | 1374 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
ッド | +1.952 |
eus | +1.822 |
ushima | +1.808 |
ptoms | +1.739 |
lyak | +1.736 |
Neg logit contributions | |
rewritten | -2.401 |
rewriting | -2.139 |
dodged | -2.094 |
acci | -2.054 |
disreg | -2.018 |
Token | long |
Token ID | 890 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ッド | +2.540 |
eus | +2.392 |
ushima | +2.370 |
lyak | +2.287 |
ptoms | +2.283 |
Neg logit contributions | |
rewritten | -2.699 |
rewriting | -2.382 |
dodged | -2.329 |
acci | -2.277 |
disreg | -2.236 |
Token | were |
Token ID | 547 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
rewritten | +7.728 |
rewriting | +6.817 |
dodged | +6.808 |
travers | +6.481 |
postponed | +6.475 |
Neg logit contributions | |
ッド | -8.761 |
ptoms | -7.842 |
eus | -7.805 |
ushima | -7.803 |
lyak | -7.534 |
Token | you |
Token ID | 345 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
rewritten | +4.987 |
rewriting | +4.408 |
dodged | +4.288 |
acci | +4.275 |
travers | +4.155 |
Neg logit contributions | |
ッド | -5.596 |
ushima | -5.234 |
ptoms | -5.202 |
eus | -5.079 |
lyak | -5.015 |
Token | given |
Token ID | 1813 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
rewritten | +1.072 |
rewriting | +0.931 |
dodged | +0.899 |
acci | +0.872 |
disreg | +0.859 |
Neg logit contributions | |
ッド | -1.397 |
eus | -1.317 |
ushima | -1.315 |
ptoms | -1.279 |
lyak | -1.264 |
Token | to |
Token ID | 284 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
rewritten | +0.336 |
rewriting | +0.296 |
dodged | +0.289 |
acci | +0.282 |
disreg | +0.278 |
Neg logit contributions | |
ッド | -0.346 |
ushima | -0.325 |
eus | -0.325 |
ptoms | -0.314 |
lyak | -0.311 |
Token | solve |
Token ID | 8494 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
ッド | +0.567 |
eus | +0.523 |
ushima | +0.521 |
ptoms | +0.500 |
lyak | +0.495 |
Neg logit contributions | |
rewritten | -0.789 |
rewriting | -0.709 |
dodged | -0.693 |
acci | -0.679 |
disreg | -0.671 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
ッド | +2.704 |
eus | +2.579 |
ushima | +2.538 |
ptoms | +2.459 |
lyak | +2.421 |
Neg logit contributions | |
rewritten | -3.055 |
rewriting | -2.694 |
dodged | -2.648 |
acci | -2.621 |
disreg | -2.518 |
Token | A |
Token ID | 317 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
ッド | +0.468 |
eus | +0.439 |
ushima | +0.437 |
ptoms | +0.422 |
lyak | +0.421 |
Neg logit contributions | |
rewritten | -0.489 |
rewriting | -0.431 |
dodged | -0.421 |
acci | -0.412 |
disreg | -0.405 |
Token | movie |
Token ID | 3807 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
rewritten | +0.900 |
rewriting | +0.780 |
dodged | +0.754 |
acci | +0.730 |
disreg | +0.723 |
Neg logit contributions | |
ッド | -1.095 |
eus | -1.034 |
ushima | -1.029 |
ptoms | -0.999 |
lyak | -0.995 |
Token | adaptation |
Token ID | 16711 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
rewritten | +1.927 |
rewriting | +1.660 |
dodged | +1.595 |
disreg | +1.523 |
acci | +1.519 |
Neg logit contributions | |
ッド | -2.577 |
ushima | -2.421 |
eus | -2.420 |
ptoms | -2.350 |
lyak | -2.327 |
Token | of |
Token ID | 286 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
rewritten | +2.778 |
rewriting | +2.361 |
dodged | +2.289 |
disreg | +2.190 |
acci | +2.152 |
Neg logit contributions | |
ッド | -4.042 |
ushima | -3.811 |
eus | -3.784 |
ptoms | -3.693 |
lyak | -3.678 |
Token | Charlie |
Token ID | 11526 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
rewritten | +0.682 |
rewriting | +0.597 |
dodged | +0.581 |
acci | +0.567 |
disreg | +0.559 |
Neg logit contributions | |
ッド | -0.716 |
eus | -0.673 |
ushima | -0.671 |
ptoms | -0.649 |
lyak | -0.646 |
Token | and |
Token ID | 290 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
rewritten | +7.251 |
rewriting | +6.199 |
dodged | +6.087 |
disreg | +5.791 |
postponed | +5.743 |
Neg logit contributions | |
ッド | -9.173 |
ushima | -8.729 |
eus | -8.702 |
lyak | -8.463 |
ptoms | -8.453 |
Token | the |
Token ID | 262 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
rewritten | +1.051 |
rewriting | +0.928 |
dodged | +0.907 |
acci | +0.887 |
disreg | +0.874 |
Neg logit contributions | |
ッド | -0.967 |
eus | -0.900 |
ushima | -0.898 |
ptoms | -0.866 |
lyak | -0.860 |
Token | Chocolate |
Token ID | 24777 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
ッド | +0.464 |
eus | +0.437 |
ushima | +0.435 |
ptoms | +0.421 |
lyak | +0.420 |
Neg logit contributions | |
rewritten | -0.423 |
rewriting | -0.370 |
dodged | -0.361 |
acci | -0.353 |
disreg | -0.345 |
Token | Factory |
Token ID | 19239 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
rewritten | +0.948 |
rewriting | +0.826 |
dodged | +0.809 |
acci | +0.792 |
disreg | +0.778 |
Neg logit contributions | |
ッド | -0.989 |
ushima | -0.924 |
eus | -0.924 |
ptoms | -0.887 |
lyak | -0.885 |
Token | initially |
Token ID | 7317 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
rewritten | +3.124 |
rewriting | +2.653 |
dodged | +2.610 |
disreg | +2.495 |
travers | +2.442 |
Neg logit contributions | |
ッド | -4.378 |
ushima | -4.116 |
eus | -4.092 |
ptoms | -4.011 |
lyak | -4.005 |
Token | , |
Token ID | 11 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
rewritten | +2.093 |
rewriting | +1.844 |
dodged | +1.795 |
acci | +1.738 |
disreg | +1.735 |
Neg logit contributions | |
ッド | -2.163 |
ushima | -2.025 |
eus | -2.014 |
ptoms | -1.962 |
lyak | -1.947 |
Token | or |
Token ID | 393 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
rewritten | +1.608 |
rewriting | +1.389 |
dodged | +1.347 |
disreg | +1.299 |
acci | +1.291 |
Neg logit contributions | |
ッド | -2.113 |
ushima | -1.992 |
eus | -1.984 |
ptoms | -1.936 |
lyak | -1.933 |
Token | 66 |
Token ID | 7930 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
rewritten | +4.063 |
rewriting | +3.526 |
dodged | +3.430 |
travers | +3.310 |
disreg | +3.295 |
Neg logit contributions | |
ッド | -5.131 |
ushima | -4.840 |
ptoms | -4.731 |
eus | -4.723 |
lyak | -4.642 |
Token | , |
Token ID | 11 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
rewritten | +2.929 |
rewriting | +2.543 |
dodged | +2.479 |
acci | +2.404 |
travers | +2.391 |
Neg logit contributions | |
ッド | -3.378 |
ushima | -3.204 |
eus | -3.171 |
ptoms | -3.070 |
lyak | -3.058 |
Token | 627 |
Token ID | 49856 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
rewritten | +0.679 |
rewriting | +0.597 |
dodged | +0.582 |
acci | +0.571 |
disreg | +0.559 |
Neg logit contributions | |
ッド | -0.680 |
eus | -0.639 |
ushima | -0.635 |
ptoms | -0.614 |
lyak | -0.613 |
Token | mph |
Token ID | 16462 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
rewritten | +0.925 |
rewriting | +0.795 |
dodged | +0.774 |
acci | +0.751 |
disreg | +0.737 |
Neg logit contributions | |
ッド | -1.223 |
eus | -1.155 |
ushima | -1.153 |
ptoms | -1.118 |
lyak | -1.114 |
Token | ( |
Token ID | 357 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
rewritten | +7.205 |
rewriting | +6.352 |
dodged | +6.166 |
travers | +6.050 |
disreg | +5.945 |
Neg logit contributions | |
ッド | -9.089 |
ushima | -8.324 |
eus | -8.266 |
lyak | -8.134 |
ptoms | -8.108 |
Token | 107 |
Token ID | 15982 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
rewritten | +2.670 |
rewriting | +2.337 |
dodged | +2.262 |
acci | +2.239 |
disreg | +2.198 |
Neg logit contributions | |
ッド | -2.731 |
ushima | -2.559 |
eus | -2.508 |
ptoms | -2.500 |
lyak | -2.481 |
Token | , |
Token ID | 11 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
rewritten | +0.687 |
rewriting | +0.598 |
dodged | +0.582 |
acci | +0.568 |
disreg | +0.560 |
Neg logit contributions | |
ッド | -0.773 |
eus | -0.725 |
ushima | -0.725 |
ptoms | -0.702 |
lyak | -0.698 |
Token | 226 |
Token ID | 24909 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
rewritten | +0.902 |
rewriting | +0.793 |
dodged | +0.774 |
acci | +0.760 |
disreg | +0.745 |
Neg logit contributions | |
ッド | -0.894 |
eus | -0.839 |
ushima | -0.835 |
ptoms | -0.807 |
lyak | -0.805 |
Token | km |
Token ID | 10571 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
rewritten | +0.954 |
rewriting | +0.766 |
dodged | +0.735 |
acci | +0.703 |
disreg | +0.680 |
Neg logit contributions | |
ッド | -2.186 |
eus | -2.091 |
ushima | -2.082 |
ptoms | -2.032 |
lyak | -2.025 |
Token | 21 |
Token ID | 2481 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
rewritten | +0.577 |
rewriting | +0.518 |
dodged | +0.507 |
acci | +0.499 |
disreg | +0.490 |
Neg logit contributions | |
ッド | -0.414 |
ushima | -0.383 |
eus | -0.381 |
lyak | -0.365 |
ptoms | -0.365 |
Token | , |
Token ID | 11 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
ッド | +2.295 |
eus | +2.175 |
ushima | +2.138 |
lyak | +2.064 |
ptoms | +2.055 |
Neg logit contributions | |
rewritten | -2.377 |
rewriting | -2.116 |
dodged | -2.065 |
acci | -2.007 |
travers | -1.959 |
Token | 21 |
Token ID | 2310 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
rewritten | +0.700 |
rewriting | +0.608 |
dodged | +0.591 |
acci | +0.579 |
disreg | +0.569 |
Neg logit contributions | |
ッド | -0.798 |
ushima | -0.753 |
eus | -0.745 |
lyak | -0.725 |
ptoms | -0.725 |
Token | )) |
Token ID | 4008 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
rewritten | +2.632 |
rewriting | +2.265 |
dodged | +2.223 |
disreg | +2.161 |
acci | +2.160 |
Neg logit contributions | |
ッド | -2.826 |
ushima | -2.679 |
eus | -2.628 |
ptoms | -2.571 |
lyak | -2.558 |
Token | # |
Token ID | 1303 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
rewritten | +4.150 |
rewriting | +3.580 |
dodged | +3.452 |
acci | +3.404 |
travers | +3.342 |
Neg logit contributions | |
ッド | -5.695 |
eus | -5.268 |
ushima | -5.260 |
lyak | -5.200 |
ptoms | -5.101 |
Token | loop |
Token ID | 9052 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
rewritten | +1.157 |
rewriting | +0.910 |
dodged | +0.847 |
acci | +0.822 |
travers | +0.806 |
Neg logit contributions | |
ッド | -3.006 |
ushima | -2.852 |
eus | -2.843 |
ptoms | -2.804 |
lyak | -2.777 |
Token | over |
Token ID | 625 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
rewritten | +5.758 |
rewriting | +4.671 |
travers | +4.514 |
disreg | +4.487 |
dodged | +4.284 |
Neg logit contributions | |
ッド | -10.956 |
ushima | -10.224 |
ptoms | -10.206 |
lyak | -9.987 |
eus | -9.898 |
Token | the |
Token ID | 262 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
rewritten | +4.399 |
rewriting | +3.736 |
disreg | +3.592 |
dodged | +3.565 |
acci | +3.503 |
Neg logit contributions | |
ッド | -5.489 |
ushima | -5.056 |
ptoms | -4.995 |
eus | -4.982 |
rises | -4.831 |
Token | input |
Token ID | 5128 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
rewritten | +1.872 |
rewriting | +1.642 |
dodged | +1.608 |
acci | +1.564 |
disreg | +1.552 |
Neg logit contributions | |
ッド | -1.782 |
ushima | -1.656 |
eus | -1.643 |
ptoms | -1.606 |
lyak | -1.579 |
Token | image |
Token ID | 2939 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
rewritten | +1.796 |
rewriting | +1.505 |
dodged | +1.451 |
disreg | +1.405 |
travers | +1.392 |
Neg logit contributions | |
ッド | -2.810 |
ushima | -2.644 |
eus | -2.619 |
ptoms | -2.611 |
rises | -2.564 |
Token | paths |
Token ID | 13532 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
ッド | +1.026 |
eus | +0.958 |
ushima | +0.946 |
ptoms | +0.919 |
lyak | +0.907 |
Neg logit contributions | |
rewritten | -1.261 |
rewriting | -1.124 |
dodged | -1.104 |
acci | -1.086 |
disreg | -1.059 |
Token | ( |
Token ID | 357 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
rewritten | +3.309 |
rewriting | +2.918 |
dodged | +2.840 |
acci | +2.774 |
disreg | +2.725 |
Neg logit contributions | |
ッド | -3.802 |
eus | -3.546 |
ushima | -3.546 |
ptoms | -3.482 |
lyak | -3.438 |
Token | for |
Token ID | 1640 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
rewritten | +0.498 |
rewriting | +0.442 |
dodged | +0.432 |
acci | +0.423 |
disreg | +0.415 |
Neg logit contributions | |
ッド | -0.447 |
eus | -0.417 |
ushima | -0.416 |
ptoms | -0.402 |
lyak | -0.401 |
Token | their |
Token ID | 511 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
rewritten | +0.265 |
rewriting | +0.230 |
dodged | +0.225 |
acci | +0.221 |
disreg | +0.214 |
Neg logit contributions | |
ッド | -0.306 |
eus | -0.289 |
ushima | -0.287 |
ptoms | -0.278 |
lyak | -0.278 |
Token | Scientology |
Token ID | 24990 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
rewritten | +0.335 |
rewriting | +0.288 |
dodged | +0.280 |
acci | +0.275 |
disreg | +0.267 |
Neg logit contributions | |
ッド | -0.439 |
eus | -0.416 |
ushima | -0.414 |
ptoms | -0.401 |
lyak | -0.401 |
Token | ); |
Token ID | 1776 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
rewritten | +1.405 |
rewriting | +1.230 |
dodged | +1.197 |
acci | +1.159 |
disreg | +1.153 |
Neg logit contributions | |
ッド | -1.553 |
eus | -1.464 |
ushima | -1.456 |
ptoms | -1.420 |
lyak | -1.410 |
Token | Charlie |
Token ID | 11526 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
rewritten | +1.651 |
rewriting | +1.427 |
dodged | +1.394 |
acci | +1.339 |
disreg | +1.329 |
Neg logit contributions | |
ッド | -2.025 |
eus | -1.915 |
ushima | -1.914 |
ptoms | -1.854 |
lyak | -1.850 |
Token | Sheen |
Token ID | 50189 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
rewritten | +6.807 |
rewriting | +5.688 |
dodged | +5.667 |
disreg | +5.414 |
postponed | +5.264 |
Neg logit contributions | |
ッド | -9.278 |
ushima | -9.234 |
lyak | -8.990 |
eus | -8.962 |
ptoms | -8.821 |
Token | for |
Token ID | 329 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
rewritten | +5.222 |
rewriting | +4.568 |
dodged | +4.534 |
acci | +4.320 |
disreg | +4.312 |
Neg logit contributions | |
ッド | -5.863 |
ushima | -5.520 |
eus | -5.492 |
ptoms | -5.352 |
lyak | -5.319 |
Token | his |
Token ID | 465 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
rewritten | +1.089 |
rewriting | +0.946 |
dodged | +0.918 |
acci | +0.893 |
disreg | +0.878 |
Neg logit contributions | |
ッド | -1.343 |
eus | -1.270 |
ushima | -1.264 |
ptoms | -1.228 |
lyak | -1.225 |
Token | antics |
Token ID | 39343 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
rewritten | +2.385 |
rewriting | +2.078 |
dodged | +2.010 |
acci | +1.924 |
disreg | +1.923 |
Neg logit contributions | |
ッド | -3.045 |
eus | -2.865 |
ushima | -2.860 |
ptoms | -2.784 |
lyak | -2.770 |
Token | at |
Token ID | 379 |
Feature activation | +7.63e-02 |
Pos logit contributions | |
rewritten | +2.365 |
rewriting | +2.078 |
dodged | +2.019 |
acci | +1.951 |
disreg | +1.943 |
Neg logit contributions | |
ッド | -2.722 |
eus | -2.551 |
ushima | -2.542 |
ptoms | -2.475 |
lyak | -2.462 |
Token | visible |
Token ID | 7424 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
rewritten | +2.002 |
rewriting | +1.715 |
dodged | +1.711 |
disreg | +1.675 |
travers | +1.626 |
Neg logit contributions | |
ッド | -2.160 |
ushima | -2.049 |
eus | -2.028 |
lyak | -1.980 |
ptoms | -1.975 |
Token | since |
Token ID | 1201 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
rewritten | +0.850 |
rewriting | +0.590 |
dodged | +0.539 |
acci | +0.490 |
disreg | +0.490 |
Neg logit contributions | |
ッド | -3.319 |
eus | -3.180 |
ushima | -3.177 |
lyak | -3.122 |
ptoms | -3.085 |
Token | at |
Token ID | 379 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
ッド | +0.171 |
eus | +0.161 |
ushima | +0.160 |
ptoms | +0.155 |
lyak | +0.155 |
Neg logit contributions | |
rewritten | -0.151 |
rewriting | -0.131 |
dodged | -0.128 |
acci | -0.125 |
disreg | -0.123 |
Token | a |
Token ID | 257 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
ッド | +1.253 |
eus | +1.177 |
ushima | +1.168 |
ptoms | +1.125 |
lyak | +1.122 |
Neg logit contributions | |
rewritten | -1.379 |
rewriting | -1.222 |
dodged | -1.197 |
acci | -1.172 |
disreg | -1.146 |
Token | 90 |
Token ID | 4101 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
rewritten | +1.739 |
rewriting | +1.519 |
dodged | +1.458 |
acci | +1.428 |
disreg | +1.424 |
Neg logit contributions | |
ッド | -1.912 |
ushima | -1.787 |
eus | -1.763 |
lyak | -1.716 |
ptoms | -1.713 |
Token | F |
Token ID | 376 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
rewritten | +3.051 |
rewriting | +2.649 |
dodged | +2.564 |
disreg | +2.498 |
acci | +2.488 |
Neg logit contributions | |
ッド | -3.365 |
ushima | -3.175 |
eus | -3.097 |
ptoms | -3.026 |
lyak | -3.020 |
Token | OV |
Token ID | 8874 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
rewritten | +8.354 |
rewriting | +7.059 |
acci | +7.059 |
disreg | +6.973 |
dodged | +6.855 |
Neg logit contributions | |
ッド | -7.579 |
ushima | -6.873 |
eus | -6.854 |
ptoms | -6.764 |
lyak | -6.589 |
Token | even |
Token ID | 772 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
rewritten | +1.857 |
rewriting | +1.513 |
dodged | +1.436 |
disreg | +1.420 |
travers | +1.375 |
Neg logit contributions | |
ッド | -3.595 |
ushima | -3.471 |
eus | -3.411 |
ptoms | -3.353 |
lyak | -3.340 |
Token | double |
Token ID | 4274 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
rewritten | +0.912 |
rewriting | +0.825 |
dodged | +0.809 |
acci | +0.791 |
disreg | +0.789 |
Neg logit contributions | |
ッド | -0.510 |
ushima | -0.465 |
eus | -0.464 |
ptoms | -0.444 |
lyak | -0.438 |
Token | or |
Token ID | 393 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
rewritten | +2.303 |
rewriting | +1.920 |
dodged | +1.870 |
disreg | +1.793 |
acci | +1.777 |
Neg logit contributions | |
ッド | -3.567 |
ushima | -3.379 |
eus | -3.322 |
ptoms | -3.281 |
lyak | -3.249 |
Token | 4 |
Token ID | 604 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
rewritten | +2.241 |
rewriting | +1.919 |
dodged | +1.857 |
disreg | +1.813 |
acci | +1.794 |
Neg logit contributions | |
ッド | -2.626 |
eus | -2.450 |
ushima | -2.438 |
ptoms | -2.397 |
lyak | -2.379 |
Token | administration |
Token ID | 3662 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
ッド | +0.307 |
eus | +0.290 |
ushima | +0.289 |
ptoms | +0.279 |
lyak | +0.279 |
Neg logit contributions | |
rewritten | -0.268 |
rewriting | -0.234 |
dodged | -0.228 |
acci | -0.223 |
disreg | -0.218 |
Token | . |
Token ID | 13 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
rewritten | +0.832 |
rewriting | +0.735 |
dodged | +0.716 |
acci | +0.703 |
disreg | +0.691 |
Neg logit contributions | |
ッド | -0.787 |
eus | -0.736 |
ushima | -0.733 |
ptoms | -0.708 |
lyak | -0.705 |
Token | Sub |
Token ID | 3834 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
ッド | +0.245 |
eus | +0.229 |
ushima | +0.228 |
ptoms | +0.221 |
lyak | +0.220 |
Neg logit contributions | |
rewritten | -0.249 |
rewriting | -0.219 |
dodged | -0.214 |
acci | -0.210 |
disreg | -0.206 |
Token | sequently |
Token ID | 20415 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
rewritten | +2.023 |
rewriting | +1.796 |
dodged | +1.750 |
acci | +1.721 |
disreg | +1.682 |
Neg logit contributions | |
ッド | -1.787 |
eus | -1.654 |
ushima | -1.634 |
ptoms | -1.575 |
lyak | -1.561 |
Token | , |
Token ID | 11 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
ッド | +1.426 |
eus | +1.340 |
ushima | +1.328 |
lyak | +1.293 |
ptoms | +1.272 |
Neg logit contributions | |
rewritten | -1.468 |
rewriting | -1.291 |
dodged | -1.263 |
acci | -1.257 |
disreg | -1.209 |
Token | Elijah |
Token ID | 38646 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
rewritten | +0.543 |
rewriting | +0.473 |
dodged | +0.461 |
acci | +0.449 |
disreg | +0.442 |
Neg logit contributions | |
ッド | -0.618 |
eus | -0.581 |
ushima | -0.579 |
ptoms | -0.562 |
lyak | -0.558 |
Token | Cummings |
Token ID | 38154 |
Feature activation | +9.70e-01 |
Pos logit contributions | |
rewritten | +7.338 |
rewriting | +6.259 |
dodged | +6.032 |
acci | +5.962 |
disreg | +5.929 |
Neg logit contributions | |
ッド | -8.991 |
ushima | -8.416 |
ptoms | -8.289 |
lyak | -8.241 |
eus | -8.221 |
Token | , |
Token ID | 11 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
rewritten | +2.823 |
rewriting | +2.438 |
dodged | +2.410 |
acci | +2.321 |
disreg | +2.287 |
Neg logit contributions | |
ッド | -3.435 |
ushima | -3.227 |
eus | -3.212 |
ptoms | -3.139 |
lyak | -3.095 |
Token | the |
Token ID | 262 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
rewritten | +0.586 |
rewriting | +0.512 |
dodged | +0.500 |
acci | +0.488 |
disreg | +0.479 |
Neg logit contributions | |
ッド | -0.640 |
eus | -0.601 |
ushima | -0.599 |
ptoms | -0.581 |
lyak | -0.578 |
Token | ranking |
Token ID | 12759 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
rewritten | +0.917 |
rewriting | +0.776 |
dodged | +0.752 |
acci | +0.730 |
disreg | +0.712 |
Neg logit contributions | |
ッド | -1.420 |
eus | -1.348 |
ushima | -1.344 |
ptoms | -1.307 |
lyak | -1.304 |
Token | member |
Token ID | 2888 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
rewritten | +0.605 |
rewriting | +0.509 |
dodged | +0.492 |
acci | +0.479 |
disreg | +0.465 |
Neg logit contributions | |
ッド | -0.984 |
eus | -0.936 |
ushima | -0.931 |
ptoms | -0.906 |
lyak | -0.905 |
Token | , |
Token ID | 11 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
rewritten | +0.929 |
rewriting | +0.810 |
dodged | +0.786 |
acci | +0.767 |
disreg | +0.754 |
Neg logit contributions | |
ッド | -1.049 |
ushima | -0.986 |
eus | -0.983 |
lyak | -0.955 |
ptoms | -0.954 |
Token | and |
Token ID | 290 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
ッド | +1.868 |
eus | +1.768 |
ushima | +1.758 |
ptoms | +1.703 |
lyak | +1.700 |
Neg logit contributions | |
rewritten | -1.535 |
rewriting | -1.330 |
dodged | -1.296 |
acci | -1.267 |
disreg | -1.236 |
Token | without |
Token ID | 1231 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
rewritten | +1.368 |
rewriting | +1.187 |
dodged | +1.156 |
disreg | +1.115 |
acci | +1.109 |
Neg logit contributions | |
ッド | -1.462 |
eus | -1.355 |
ushima | -1.353 |
ptoms | -1.323 |
lyak | -1.313 |
Token | ads |
Token ID | 9011 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
rewritten | +0.364 |
rewriting | +0.320 |
dodged | +0.309 |
acci | +0.300 |
disreg | +0.298 |
Neg logit contributions | |
ッド | -0.384 |
eus | -0.360 |
ushima | -0.360 |
ptoms | -0.349 |
lyak | -0.348 |
Token | . |
Token ID | 13 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
ッド | +2.665 |
eus | +2.520 |
ushima | +2.506 |
lyak | +2.398 |
ptoms | +2.398 |
Neg logit contributions | |
rewritten | -2.833 |
rewriting | -2.499 |
dodged | -2.458 |
acci | -2.417 |
disreg | -2.356 |
Token | Forever |
Token ID | 28285 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
ッド | +1.585 |
eus | +1.490 |
ushima | +1.482 |
ptoms | +1.433 |
lyak | +1.430 |
Neg logit contributions | |
rewritten | -1.550 |
rewriting | -1.361 |
dodged | -1.329 |
acci | -1.300 |
disreg | -1.274 |
Token | . |
Token ID | 13 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
rewritten | +7.457 |
rewriting | +6.474 |
dodged | +5.995 |
disreg | +5.897 |
postponed | +5.825 |
Neg logit contributions | |
ッド | -8.605 |
ushima | -8.168 |
eus | -8.109 |
ptoms | -7.972 |
lyak | -7.889 |
Token |
|
Token ID | 198 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
rewritten | +1.519 |
rewriting | +1.320 |
dodged | +1.276 |
acci | +1.263 |
disreg | +1.226 |
Neg logit contributions | |
ッド | -1.803 |
ushima | -1.685 |
eus | -1.679 |
ptoms | -1.642 |
lyak | -1.630 |
Token |
|
Token ID | 198 |
Feature activation | -6.17e-03 |
Pos logit contributions | |
rewritten | +1.751 |
rewriting | +1.486 |
dodged | +1.435 |
acci | +1.420 |
disreg | +1.399 |
Neg logit contributions | |
ッド | -2.226 |
ushima | -2.086 |
eus | -2.065 |
ptoms | -2.030 |
lyak | -2.011 |
Token | With |
Token ID | 3152 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
ッド | +0.016 |
ushima | +0.015 |
eus | +0.015 |
ptoms | +0.014 |
lyak | +0.014 |
Neg logit contributions | |
rewritten | -0.024 |
rewriting | -0.021 |
dodged | -0.021 |
acci | -0.021 |
disreg | -0.020 |
Token | love |
Token ID | 1842 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
ッド | +1.694 |
eus | +1.587 |
ushima | +1.578 |
ptoms | +1.522 |
lyak | +1.519 |
Neg logit contributions | |
rewritten | -1.821 |
rewriting | -1.609 |
dodged | -1.573 |
acci | -1.541 |
disreg | -1.512 |
Token | on |
Token ID | 319 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
ッド | +0.067 |
eus | +0.064 |
ushima | +0.063 |
lyak | +0.061 |
ptoms | +0.061 |
Neg logit contributions | |
rewritten | -0.052 |
rewriting | -0.044 |
dodged | -0.043 |
acci | -0.042 |
disreg | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
rewritten | +0.924 |
rewriting | +0.809 |
dodged | +0.782 |
acci | +0.763 |
disreg | +0.754 |
Neg logit contributions | |
ッド | -1.028 |
ushima | -0.968 |
eus | -0.965 |
ptoms | -0.937 |
lyak | -0.930 |
Token | film |
Token ID | 2646 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
rewritten | +1.798 |
rewriting | +1.558 |
dodged | +1.498 |
disreg | +1.444 |
acci | +1.442 |
Neg logit contributions | |
ッド | -2.184 |
ushima | -2.064 |
eus | -2.048 |
ptoms | -1.999 |
lyak | -1.979 |
Token | with |
Token ID | 351 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
rewritten | +0.916 |
rewriting | +0.794 |
dodged | +0.765 |
acci | +0.744 |
disreg | +0.733 |
Neg logit contributions | |
ッド | -1.141 |
ushima | -1.073 |
eus | -1.071 |
ptoms | -1.043 |
lyak | -1.032 |
Token | Robert |
Token ID | 5199 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
rewritten | +0.676 |
rewriting | +0.590 |
dodged | +0.573 |
acci | +0.558 |
disreg | +0.551 |
Neg logit contributions | |
ッド | -0.763 |
ushima | -0.717 |
eus | -0.717 |
ptoms | -0.695 |
lyak | -0.692 |
Token | Alt |
Token ID | 12344 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
rewritten | +1.607 |
rewriting | +1.444 |
dodged | +1.408 |
acci | +1.391 |
disreg | +1.359 |
Neg logit contributions | |
ッド | -1.392 |
ushima | -1.292 |
eus | -1.284 |
ptoms | -1.249 |
lyak | -1.245 |
Token | man |
Token ID | 805 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
rewritten | +6.131 |
rewriting | +5.236 |
acci | +5.011 |
travers | +4.965 |
dodged | +4.811 |
Neg logit contributions | |
ッド | -10.025 |
ptoms | -9.347 |
ushima | -9.250 |
lyak | -9.138 |
eus | -9.134 |
Token | to |
Token ID | 284 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
ッド | +1.016 |
eus | +0.966 |
ushima | +0.950 |
lyak | +0.917 |
ptoms | +0.916 |
Neg logit contributions | |
rewritten | -0.974 |
rewriting | -0.852 |
dodged | -0.838 |
acci | -0.820 |
disreg | -0.798 |
Token | be |
Token ID | 307 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ッド | +1.330 |
eus | +1.242 |
ushima | +1.236 |
ptoms | +1.191 |
lyak | +1.188 |
Neg logit contributions | |
rewritten | -1.515 |
rewriting | -1.345 |
dodged | -1.314 |
acci | -1.287 |
disreg | -1.266 |
Token | titled |
Token ID | 11946 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
rewritten | +0.337 |
rewriting | +0.284 |
dodged | +0.275 |
acci | +0.267 |
disreg | +0.260 |
Neg logit contributions | |
ッド | -0.531 |
eus | -0.505 |
ushima | -0.503 |
ptoms | -0.489 |
lyak | -0.488 |
Token | Oh |
Token ID | 3966 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
rewritten | +4.443 |
rewriting | +3.906 |
dodged | +3.767 |
acci | +3.654 |
disreg | +3.640 |
Neg logit contributions | |
ッド | -4.767 |
ushima | -4.476 |
eus | -4.378 |
ptoms | -4.344 |
lyak | -4.234 |
Token | . |
Token ID | 764 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
rewritten | +0.253 |
rewriting | +0.221 |
dodged | +0.215 |
acci | +0.212 |
disreg | +0.208 |
Neg logit contributions | |
ッド | -0.255 |
eus | -0.238 |
ushima | -0.238 |
ptoms | -0.230 |
lyak | -0.230 |
Token |
|
Token ID | 198 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
rewritten | +1.996 |
rewriting | +1.719 |
dodged | +1.661 |
acci | +1.653 |
disreg | +1.604 |
Neg logit contributions | |
ッド | -2.386 |
eus | -2.234 |
ushima | -2.228 |
ptoms | -2.176 |
lyak | -2.159 |
Token |
|
Token ID | 198 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
rewritten | +1.139 |
rewriting | +0.979 |
dodged | +0.949 |
acci | +0.936 |
disreg | +0.920 |
Neg logit contributions | |
ッド | -1.364 |
ushima | -1.277 |
eus | -1.270 |
ptoms | -1.242 |
lyak | -1.231 |
Token | Google |
Token ID | 11708 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
ッド | +0.637 |
eus | +0.590 |
ushima | +0.587 |
ptoms | +0.563 |
lyak | +0.561 |
Neg logit contributions | |
rewritten | -0.895 |
rewriting | -0.802 |
dodged | -0.786 |
acci | -0.773 |
disreg | -0.760 |
Token | Scholar |
Token ID | 11713 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
rewritten | +0.788 |
rewriting | +0.664 |
dodged | +0.645 |
acci | +0.633 |
disreg | +0.623 |
Neg logit contributions | |
ッド | -1.060 |
ushima | -1.005 |
eus | -0.999 |
ptoms | -0.983 |
lyak | -0.977 |
Token | Crossref |
Token ID | 23613 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
rewritten | +2.750 |
rewriting | +2.377 |
dodged | +2.308 |
acci | +2.264 |
disreg | +2.216 |
Neg logit contributions | |
ッド | -3.105 |
eus | -2.919 |
ushima | -2.882 |
ptoms | -2.829 |
lyak | -2.822 |
Token |
|
Token ID | 198 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
rewritten | +6.755 |
rewriting | +5.819 |
acci | +5.719 |
dodged | +5.566 |
disreg | +5.352 |
Neg logit contributions | |
ッド | -8.872 |
eus | -8.447 |
ushima | -8.420 |
ptoms | -8.338 |
lyak | -8.189 |
Token |
|
Token ID | 198 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
rewritten | +0.177 |
rewriting | +0.154 |
dodged | +0.150 |
acci | +0.147 |
disreg | +0.144 |
Neg logit contributions | |
ッド | -0.197 |
eus | -0.185 |
ushima | -0.185 |
ptoms | -0.179 |
lyak | -0.178 |
Token | Pe |
Token ID | 6435 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
ッド | +1.193 |
eus | +1.100 |
ushima | +1.093 |
ptoms | +1.044 |
lyak | +1.039 |
Neg logit contributions | |
rewritten | -1.881 |
rewriting | -1.698 |
dodged | -1.668 |
acci | -1.637 |
disreg | -1.613 |
Token | eters |
Token ID | 7307 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
ッド | +4.833 |
ushima | +4.708 |
eus | +4.706 |
lyak | +4.569 |
ptoms | +4.504 |
Neg logit contributions | |
rewritten | -5.479 |
rewriting | -4.827 |
dodged | -4.687 |
disreg | -4.568 |
travers | -4.476 |
Token | , |
Token ID | 11 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
ッド | +1.162 |
eus | +1.102 |
ushima | +1.097 |
lyak | +1.057 |
ptoms | +1.050 |
Neg logit contributions | |
rewritten | -1.200 |
rewriting | -1.056 |
dodged | -1.028 |
acci | -0.994 |
disreg | -0.990 |
Token | , |
Token ID | 11 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
ッド | +0.355 |
eus | +0.334 |
ushima | +0.333 |
ptoms | +0.322 |
lyak | +0.321 |
Neg logit contributions | |
rewritten | -0.337 |
rewriting | -0.295 |
dodged | -0.288 |
acci | -0.281 |
disreg | -0.276 |
Token | Xbox |
Token ID | 9445 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ッド | +1.346 |
eus | +1.278 |
ushima | +1.259 |
ptoms | +1.212 |
lyak | +1.209 |
Neg logit contributions | |
rewritten | -1.394 |
rewriting | -1.228 |
dodged | -1.202 |
acci | -1.190 |
disreg | -1.147 |
Token | 720 |
Token ID | 26250 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ッド | +1.623 |
eus | +1.514 |
ushima | +1.485 |
ptoms | +1.436 |
lyak | +1.433 |
Neg logit contributions | |
rewritten | -1.964 |
rewriting | -1.746 |
dodged | -1.716 |
acci | -1.693 |
disreg | -1.643 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e-03 |
Pos logit contributions | |
ッド | +4.076 |
eus | +3.921 |
ushima | +3.835 |
lyak | +3.688 |
ptoms | +3.653 |
Neg logit contributions | |
rewritten | -4.351 |
rewriting | -3.836 |
dodged | -3.800 |
acci | -3.797 |
disreg | -3.603 |
Token | the |
Token ID | 262 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
rewritten | +0.007 |
rewriting | +0.007 |
dodged | +0.006 |
acci | +0.006 |
disreg | +0.006 |
Neg logit contributions | |
ッド | -0.007 |
eus | -0.007 |
ushima | -0.007 |
ptoms | -0.006 |
lyak | -0.006 |
Token | Sony |
Token ID | 10184 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
ッド | +1.772 |
eus | +1.687 |
ushima | +1.657 |
ptoms | +1.606 |
lyak | +1.599 |
Neg logit contributions | |
rewritten | -1.806 |
rewriting | -1.592 |
dodged | -1.561 |
acci | -1.536 |
disreg | -1.489 |
Token | PlayStation |
Token ID | 14047 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
ッド | +7.680 |
eus | +7.613 |
ushima | +7.239 |
loo | +6.894 |
lar | +6.835 |
Neg logit contributions | |
rewritten | -12.145 |
acci | -11.084 |
rewriting | -10.944 |
dodged | -10.708 |
grop | -10.268 |
Token | 4 |
Token ID | 604 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
ッド | +5.533 |
eus | +5.501 |
ushima | +5.359 |
lyak | +5.091 |
rises | +5.046 |
Neg logit contributions | |
rewritten | -6.671 |
rewriting | -5.957 |
acci | -5.890 |
dodged | -5.874 |
disreg | -5.521 |
Token | ," |
Token ID | 553 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
ッド | +2.486 |
eus | +2.339 |
ushima | +2.311 |
ptoms | +2.215 |
lyak | +2.215 |
Neg logit contributions | |
rewritten | -3.247 |
rewriting | -2.899 |
dodged | -2.852 |
acci | -2.817 |
disreg | -2.738 |
Token | as |
Token ID | 355 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
rewritten | +0.422 |
rewriting | +0.365 |
dodged | +0.355 |
acci | +0.347 |
disreg | +0.338 |
Neg logit contributions | |
ッド | -0.532 |
eus | -0.503 |
ushima | -0.500 |
ptoms | -0.485 |
lyak | -0.484 |
Token | well |
Token ID | 880 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
rewritten | +0.819 |
rewriting | +0.712 |
dodged | +0.690 |
acci | +0.669 |
disreg | +0.666 |
Neg logit contributions | |
ッド | -0.914 |
eus | -0.854 |
ushima | -0.852 |
ptoms | -0.825 |
lyak | -0.821 |
Token | on |
Token ID | 319 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
rewritten | +1.827 |
rewriting | +1.574 |
dodged | +1.525 |
disreg | +1.467 |
acci | +1.458 |
Neg logit contributions | |
ッド | -2.410 |
ushima | -2.262 |
eus | -2.258 |
ptoms | -2.197 |
lyak | -2.187 |
Token | the |
Token ID | 262 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
rewritten | +1.222 |
rewriting | +1.073 |
dodged | +1.045 |
acci | +1.014 |
disreg | +1.005 |
Neg logit contributions | |
ッド | -1.286 |
ushima | -1.204 |
eus | -1.202 |
ptoms | -1.163 |
lyak | -1.160 |
Token | PlayStation |
Token ID | 14047 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
rewritten | +1.138 |
rewriting | +0.976 |
dodged | +0.946 |
acci | +0.908 |
disreg | +0.906 |
Neg logit contributions | |
ッド | -1.510 |
ushima | -1.425 |
eus | -1.420 |
lyak | -1.378 |
ptoms | -1.377 |
Token | 3 |
Token ID | 513 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
ッド | +3.708 |
eus | +3.558 |
ushima | +3.543 |
lyak | +3.366 |
ptoms | +3.352 |
Neg logit contributions | |
rewritten | -4.703 |
acci | -4.228 |
rewriting | -4.211 |
dodged | -4.082 |
disreg | -3.862 |
Token | and |
Token ID | 290 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
rewritten | +0.643 |
rewriting | +0.557 |
dodged | +0.543 |
acci | +0.529 |
disreg | +0.519 |
Neg logit contributions | |
ッド | -0.774 |
eus | -0.730 |
ushima | -0.727 |
ptoms | -0.705 |
lyak | -0.703 |
Token | Sony |
Token ID | 10184 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
rewritten | +0.700 |
rewriting | +0.614 |
dodged | +0.600 |
acci | +0.586 |
disreg | +0.576 |
Neg logit contributions | |
ッド | -0.716 |
eus | -0.672 |
ushima | -0.670 |
ptoms | -0.647 |
lyak | -0.646 |
Token | 's |
Token ID | 338 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
eus | +7.021 |
ushima | +6.889 |
ッド | +6.867 |
Cs | +6.340 |
loo | +6.214 |
Neg logit contributions | |
rewritten | -10.802 |
acci | -9.840 |
rewriting | -9.617 |
dodged | -9.304 |
disreg | -8.908 |
Token | next |
Token ID | 1306 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
ッド | +0.432 |
eus | +0.409 |
ushima | +0.406 |
ptoms | +0.392 |
lyak | +0.392 |
Neg logit contributions | |
rewritten | -0.402 |
rewriting | -0.352 |
dodged | -0.343 |
acci | -0.337 |
disreg | -0.328 |
Token | - |
Token ID | 12 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ッド | +2.007 |
eus | +1.874 |
ushima | +1.833 |
lyak | +1.752 |
ptoms | +1.725 |
Neg logit contributions | |
rewritten | -3.527 |
rewriting | -3.187 |
dodged | -3.143 |
acci | -3.136 |
disreg | -3.038 |
Token | generation |
Token ID | 20158 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
ッド | +0.480 |
eus | +0.453 |
ushima | +0.451 |
ptoms | +0.437 |
lyak | +0.436 |
Neg logit contributions | |
rewritten | -0.407 |
rewriting | -0.354 |
dodged | -0.345 |
acci | -0.337 |
disreg | -0.329 |
Token | PlayStation |
Token ID | 14047 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
ッド | +1.710 |
eus | +1.620 |
ushima | +1.597 |
ptoms | +1.541 |
lyak | +1.537 |
Neg logit contributions | |
rewritten | -1.818 |
rewriting | -1.602 |
dodged | -1.568 |
acci | -1.554 |
disreg | -1.501 |
Token | 30 |
Token ID | 1270 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
ッド | +2.053 |
eus | +1.925 |
ushima | +1.894 |
ptoms | +1.825 |
lyak | +1.818 |
Neg logit contributions | |
rewritten | -2.674 |
rewriting | -2.393 |
dodged | -2.352 |
acci | -2.303 |
disreg | -2.255 |
Token | ] |
Token ID | 60 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
ッド | +0.287 |
eus | +0.267 |
ushima | +0.265 |
ptoms | +0.255 |
lyak | +0.254 |
Neg logit contributions | |
rewritten | -0.361 |
rewriting | -0.322 |
dodged | -0.316 |
acci | -0.309 |
disreg | -0.304 |
Token | By |
Token ID | 2750 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
rewritten | +1.907 |
rewriting | +1.682 |
dodged | +1.643 |
acci | +1.621 |
disreg | +1.604 |
Neg logit contributions | |
ッド | -1.782 |
ushima | -1.669 |
eus | -1.651 |
ptoms | -1.604 |
lyak | -1.602 |
Token | mid |
Token ID | 3095 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ッド | +2.100 |
eus | +1.975 |
ushima | +1.965 |
ptoms | +1.894 |
lyak | +1.893 |
Neg logit contributions | |
rewritten | -2.151 |
rewriting | -1.891 |
dodged | -1.850 |
acci | -1.814 |
disreg | -1.771 |
Token | - |
Token ID | 12 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ッド | +2.904 |
eus | +2.683 |
ushima | +2.668 |
ptoms | +2.546 |
lyak | +2.507 |
Neg logit contributions | |
rewritten | -4.789 |
rewriting | -4.298 |
dodged | -4.258 |
acci | -4.179 |
disreg | -4.089 |
Token | April |
Token ID | 16784 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
ッド | +2.586 |
eus | +2.395 |
ushima | +2.368 |
ptoms | +2.239 |
lyak | +2.237 |
Neg logit contributions | |
rewritten | -4.580 |
rewriting | -4.135 |
dodged | -4.084 |
acci | -4.018 |
disreg | -3.933 |
Token | , |
Token ID | 11 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
ッド | +8.414 |
eus | +7.959 |
ushima | +7.857 |
lyak | +7.698 |
ptoms | +7.530 |
Neg logit contributions | |
rewritten | -8.995 |
acci | -8.002 |
rewriting | -7.856 |
dodged | -7.594 |
disreg | -7.149 |
Token | the |
Token ID | 262 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
ッド | +2.731 |
eus | +2.577 |
ushima | +2.550 |
lyak | +2.479 |
ptoms | +2.464 |
Neg logit contributions | |
rewritten | -2.789 |
rewriting | -2.448 |
dodged | -2.394 |
acci | -2.363 |
disreg | -2.267 |
Token | final |
Token ID | 2457 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
ッド | +0.123 |
eus | +0.115 |
ushima | +0.115 |
ptoms | +0.111 |
lyak | +0.110 |
Neg logit contributions | |
rewritten | -0.123 |
rewriting | -0.109 |
dodged | -0.106 |
acci | -0.104 |
disreg | -0.102 |
Token | Allied |
Token ID | 28234 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
ッド | +1.185 |
eus | +1.111 |
ushima | +1.104 |
ptoms | +1.066 |
lyak | +1.065 |
Neg logit contributions | |
rewritten | -1.232 |
rewriting | -1.086 |
dodged | -1.063 |
acci | -1.043 |
disreg | -1.020 |
Token | offensive |
Token ID | 5859 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
ッド | +2.278 |
eus | +2.132 |
ushima | +2.113 |
lyak | +2.038 |
ptoms | +2.032 |
Neg logit contributions | |
rewritten | -2.557 |
rewriting | -2.258 |
dodged | -2.221 |
acci | -2.185 |
disreg | -2.125 |
Token | between |
Token ID | 1022 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
ッド | +2.796 |
eus | +2.678 |
ushima | +2.643 |
ptoms | +2.551 |
lyak | +2.546 |
Neg logit contributions | |
rewritten | -2.452 |
rewriting | -2.130 |
dodged | -2.088 |
acci | -2.074 |
disreg | -1.974 |
Token | academic |
Token ID | 8233 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
ッド | +1.173 |
eus | +1.102 |
ushima | +1.096 |
ptoms | +1.055 |
lyak | +1.053 |
Neg logit contributions | |
rewritten | -1.250 |
rewriting | -1.102 |
dodged | -1.081 |
acci | -1.063 |
disreg | -1.036 |
Token | researchers |
Token ID | 4837 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
rewritten | +1.114 |
rewriting | +0.981 |
dodged | +0.947 |
disreg | +0.917 |
acci | +0.914 |
Neg logit contributions | |
ッド | -1.248 |
eus | -1.170 |
ushima | -1.163 |
ptoms | -1.134 |
lyak | -1.133 |
Token | and |
Token ID | 290 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
ッド | +0.371 |
eus | +0.349 |
ushima | +0.349 |
lyak | +0.336 |
ptoms | +0.336 |
Neg logit contributions | |
rewritten | -0.345 |
rewriting | -0.300 |
dodged | -0.294 |
acci | -0.289 |
disreg | -0.280 |
Token | the |
Token ID | 262 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ッド | +0.982 |
eus | +0.920 |
ushima | +0.914 |
ptoms | +0.881 |
lyak | +0.879 |
Neg logit contributions | |
rewritten | -1.091 |
rewriting | -0.966 |
dodged | -0.945 |
acci | -0.927 |
disreg | -0.908 |
Token | pharmaceutical |
Token ID | 18423 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
ッド | +0.469 |
eus | +0.441 |
ushima | +0.438 |
ptoms | +0.424 |
lyak | +0.423 |
Neg logit contributions | |
rewritten | -0.457 |
rewriting | -0.401 |
dodged | -0.391 |
acci | -0.382 |
disreg | -0.375 |
Token | and |
Token ID | 290 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
ッド | +7.215 |
eus | +6.852 |
ushima | +6.767 |
lyak | +6.474 |
rises | +6.440 |
Neg logit contributions | |
rewritten | -9.117 |
acci | -8.004 |
rewriting | -7.986 |
dodged | -7.963 |
postponed | -7.544 |
Token | bi |
Token ID | 3182 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ッド | +2.047 |
eus | +1.925 |
ushima | +1.914 |
ptoms | +1.819 |
lyak | +1.818 |
Neg logit contributions | |
rewritten | -2.397 |
rewriting | -2.124 |
dodged | -2.091 |
acci | -2.075 |
disreg | -1.995 |
Token | otechnology |
Token ID | 31201 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ッド | +3.429 |
eus | +3.301 |
ushima | +3.277 |
ptoms | +3.150 |
lyak | +3.133 |
Neg logit contributions | |
rewritten | -3.069 |
rewriting | -2.669 |
dodged | -2.602 |
acci | -2.508 |
disreg | -2.471 |
Token | industries |
Token ID | 11798 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
ッド | +4.269 |
eus | +4.003 |
ushima | +3.956 |
rises | +3.768 |
lyak | +3.721 |
Neg logit contributions | |
rewritten | -6.068 |
rewriting | -5.366 |
dodged | -5.327 |
acci | -5.314 |
disreg | -5.078 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ッド | +1.703 |
eus | +1.604 |
ushima | +1.593 |
ptoms | +1.522 |
lyak | +1.519 |
Neg logit contributions | |
rewritten | -2.127 |
rewriting | -1.887 |
dodged | -1.853 |
acci | -1.826 |
disreg | -1.777 |
Token | that |
Token ID | 326 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ッド | +1.635 |
eus | +1.548 |
ushima | +1.540 |
ptoms | +1.485 |
lyak | +1.483 |
Neg logit contributions | |
rewritten | -1.449 |
rewriting | -1.256 |
dodged | -1.231 |
acci | -1.207 |
disreg | -1.166 |
Token | state |
Token ID | 1181 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
rewritten | +0.491 |
rewriting | +0.434 |
dodged | +0.424 |
acci | +0.416 |
disreg | +0.408 |
Neg logit contributions | |
ッド | -0.452 |
eus | -0.423 |
ushima | -0.421 |
ptoms | -0.406 |
lyak | -0.405 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
rewritten | +2.087 |
rewriting | +1.847 |
dodged | +1.803 |
acci | +1.763 |
disreg | +1.742 |
Neg logit contributions | |
ッド | -1.929 |
eus | -1.802 |
ushima | -1.795 |
ptoms | -1.740 |
lyak | -1.733 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
rewritten | +0.320 |
rewriting | +0.281 |
dodged | +0.274 |
acci | +0.269 |
disreg | +0.264 |
Neg logit contributions | |
ッド | -0.331 |
eus | -0.310 |
ushima | -0.309 |
ptoms | -0.299 |
lyak | -0.299 |
Token |
|
Token ID | 198 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
rewritten | +0.717 |
rewriting | +0.620 |
dodged | +0.602 |
acci | +0.591 |
disreg | +0.581 |
Neg logit contributions | |
ッド | -0.839 |
ushima | -0.786 |
eus | -0.783 |
ptoms | -0.762 |
lyak | -0.759 |
Token | Medic |
Token ID | 39112 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
ッド | +0.485 |
eus | +0.447 |
ushima | +0.445 |
ptoms | +0.425 |
lyak | +0.425 |
Neg logit contributions | |
rewritten | -0.732 |
rewriting | -0.659 |
dodged | -0.646 |
acci | -0.635 |
disreg | -0.625 |
Token | are |
Token ID | 533 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
eus | +4.791 |
ushima | +4.739 |
ptoms | +4.445 |
ッド | +4.378 |
arate | +4.039 |
Neg logit contributions | |
rewritten | -11.529 |
rewriting | -10.820 |
dodged | -10.221 |
acci | -10.027 |
travers | -9.975 |
Token | is |
Token ID | 318 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
ッド | +2.608 |
eus | +2.479 |
ushima | +2.456 |
ptoms | +2.397 |
lyak | +2.371 |
Neg logit contributions | |
rewritten | -2.274 |
rewriting | -1.978 |
acci | -1.926 |
dodged | -1.917 |
disreg | -1.840 |
Token | thought |
Token ID | 1807 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
ッド | +0.924 |
eus | +0.875 |
ushima | +0.864 |
ptoms | +0.833 |
lyak | +0.831 |
Neg logit contributions | |
rewritten | -0.768 |
rewriting | -0.663 |
dodged | -0.651 |
acci | -0.648 |
disreg | -0.619 |
Token | to |
Token ID | 284 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
ッド | +0.998 |
eus | +0.946 |
ushima | +0.938 |
ptoms | +0.905 |
lyak | +0.901 |
Neg logit contributions | |
rewritten | -0.972 |
rewriting | -0.854 |
dodged | -0.834 |
acci | -0.823 |
disreg | -0.800 |
Token | be |
Token ID | 307 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
ッド | +1.292 |
eus | +1.205 |
ushima | +1.199 |
ptoms | +1.152 |
lyak | +1.149 |
Neg logit contributions | |
rewritten | -1.595 |
rewriting | -1.421 |
dodged | -1.391 |
acci | -1.366 |
disreg | -1.341 |
Token | 000 |
Token ID | 830 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
ッド | +0.268 |
eus | +0.250 |
ushima | +0.249 |
ptoms | +0.240 |
lyak | +0.239 |
Neg logit contributions | |
rewritten | -0.310 |
rewriting | -0.274 |
dodged | -0.268 |
acci | -0.264 |
disreg | -0.259 |
Token | app |
Token ID | 598 |
Feature activation | -1.34e-03 |
Pos logit contributions | |
rewritten | +2.088 |
rewriting | +1.837 |
dodged | +1.778 |
disreg | +1.733 |
acci | +1.728 |
Neg logit contributions | |
ッド | -2.050 |
ushima | -1.932 |
eus | -1.922 |
ptoms | -1.873 |
lyak | -1.827 |
Token | number |
Token ID | 1271 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
ッド | +0.004 |
ushima | +0.004 |
eus | +0.004 |
ptoms | +0.004 |
lyak | +0.004 |
Neg logit contributions | |
rewritten | -0.004 |
rewriting | -0.004 |
dodged | -0.004 |
acci | -0.004 |
disreg | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
rewritten | +2.848 |
rewriting | +2.503 |
dodged | +2.431 |
disreg | +2.380 |
acci | +2.339 |
Neg logit contributions | |
ッド | -2.807 |
ushima | -2.595 |
eus | -2.573 |
ptoms | -2.531 |
lyak | -2.493 |
Token | [ |
Token ID | 685 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
rewritten | +0.937 |
rewriting | +0.827 |
dodged | +0.799 |
acci | +0.775 |
disreg | +0.775 |
Neg logit contributions | |
ッド | -0.970 |
ushima | -0.910 |
eus | -0.908 |
ptoms | -0.883 |
lyak | -0.870 |
Token | Samsung |
Token ID | 32334 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
rewritten | +2.017 |
rewriting | +1.815 |
dodged | +1.769 |
acci | +1.751 |
disreg | +1.719 |
Neg logit contributions | |
ッド | -1.437 |
ushima | -1.316 |
eus | -1.300 |
ptoms | -1.283 |
lyak | -1.259 |
Token | Galaxy |
Token ID | 9252 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
ッド | +7.210 |
eus | +7.004 |
ushima | +6.711 |
lyak | +6.451 |
ptoms | +6.442 |
Neg logit contributions | |
rewritten | -8.942 |
rewriting | -8.017 |
dodged | -7.909 |
acci | -7.798 |
travers | -7.512 |
Token | ] |
Token ID | 60 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
ッド | +1.903 |
eus | +1.777 |
ushima | +1.762 |
ptoms | +1.697 |
lyak | +1.694 |
Neg logit contributions | |
rewritten | -2.345 |
rewriting | -2.090 |
dodged | -2.046 |
acci | -2.007 |
disreg | -1.971 |
Token | Tab |
Token ID | 16904 |
Feature activation | -7.71e-01 |
Pos logit contributions | |
rewritten | +1.457 |
rewriting | +1.280 |
dodged | +1.252 |
acci | +1.223 |
disreg | +1.217 |
Neg logit contributions | |
ッド | -1.380 |
ushima | -1.277 |
eus | -1.268 |
ptoms | -1.243 |
lyak | -1.227 |
Token | 10 |
Token ID | 838 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
ッド | +2.355 |
eus | +2.229 |
ushima | +2.204 |
lyak | +2.122 |
ptoms | +2.121 |
Neg logit contributions | |
rewritten | -2.593 |
rewriting | -2.295 |
dodged | -2.248 |
acci | -2.200 |
disreg | -2.154 |
Token | . |
Token ID | 13 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
rewritten | +1.339 |
rewriting | +1.180 |
dodged | +1.142 |
disreg | +1.119 |
acci | +1.105 |
Neg logit contributions | |
ッド | -1.241 |
ushima | -1.150 |
eus | -1.139 |
ptoms | -1.108 |
lyak | -1.098 |
Token | fields |
Token ID | 7032 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
rewritten | +1.513 |
rewriting | +1.328 |
dodged | +1.287 |
acci | +1.245 |
disreg | +1.245 |
Neg logit contributions | |
ッド | -1.634 |
eus | -1.539 |
ushima | -1.527 |
ptoms | -1.486 |
lyak | -1.474 |
Token | to |
Token ID | 284 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
ッド | +1.223 |
eus | +1.155 |
ushima | +1.148 |
ptoms | +1.111 |
lyak | +1.109 |
Neg logit contributions | |
rewritten | -1.064 |
rewriting | -0.926 |
dodged | -0.903 |
acci | -0.885 |
disreg | -0.861 |
Token | appear |
Token ID | 1656 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
ッド | +1.565 |
eus | +1.465 |
ushima | +1.458 |
ptoms | +1.406 |
lyak | +1.403 |
Neg logit contributions | |
rewritten | -1.657 |
rewriting | -1.465 |
dodged | -1.431 |
acci | -1.398 |
disreg | -1.376 |
Token | � |
Token ID | 564 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
ッド | +0.177 |
eus | +0.166 |
ushima | +0.166 |
ptoms | +0.161 |
lyak | +0.160 |
Neg logit contributions | |
rewritten | -0.149 |
rewriting | -0.129 |
dodged | -0.126 |
acci | -0.122 |
disreg | -0.121 |
Token | � |
Token ID | 250 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
ッド | +0.854 |
eus | +0.806 |
ushima | +0.802 |
ptoms | +0.776 |
lyak | +0.774 |
Neg logit contributions | |
rewritten | -0.753 |
rewriting | -0.656 |
dodged | -0.640 |
acci | -0.625 |
disreg | -0.612 |
Token | fall |
Token ID | 7207 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
rewritten | +1.624 |
rewriting | +1.439 |
dodged | +1.419 |
acci | +1.393 |
disreg | +1.367 |
Neg logit contributions | |
ッド | -1.383 |
ushima | -1.278 |
eus | -1.271 |
ptoms | -1.229 |
lyak | -1.223 |
Token | ow |
Token ID | 322 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
ッド | +6.243 |
eus | +5.818 |
ushima | +5.758 |
ordial | +5.441 |
lyak | +5.389 |
Neg logit contributions | |
rewritten | -10.091 |
rewriting | -8.977 |
dodged | -8.919 |
acci | -8.644 |
disreg | -8.617 |
Token | , |
Token ID | 11 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
ッド | +0.626 |
eus | +0.588 |
ushima | +0.585 |
ptoms | +0.564 |
lyak | +0.563 |
Neg logit contributions | |
rewritten | -0.644 |
rewriting | -0.568 |
dodged | -0.554 |
acci | -0.542 |
disreg | -0.534 |
Token | � |
Token ID | 447 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
ッド | +1.935 |
eus | +1.827 |
ushima | +1.819 |
ptoms | +1.759 |
lyak | +1.758 |
Neg logit contributions | |
rewritten | -1.676 |
rewriting | -1.457 |
dodged | -1.419 |
acci | -1.395 |
disreg | -1.352 |
Token | � |
Token ID | 251 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
rewritten | +1.305 |
rewriting | +1.146 |
dodged | +1.121 |
acci | +1.089 |
disreg | +1.074 |
Neg logit contributions | |
ッド | -1.342 |
eus | -1.270 |
ushima | -1.263 |
ptoms | -1.219 |
lyak | -1.210 |
Token | but |
Token ID | 475 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
ッド | +0.503 |
eus | +0.474 |
ushima | +0.472 |
ptoms | +0.458 |
lyak | +0.456 |
Neg logit contributions | |
rewritten | -0.438 |
rewriting | -0.382 |
dodged | -0.372 |
acci | -0.362 |
disreg | -0.358 |
Token | III |
Token ID | 6711 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
rewritten | +0.234 |
rewriting | +0.206 |
dodged | +0.202 |
acci | +0.198 |
disreg | +0.194 |
Neg logit contributions | |
ッド | -0.231 |
eus | -0.218 |
ushima | -0.216 |
ptoms | -0.208 |
lyak | -0.208 |
Token | / |
Token ID | 14 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
ッド | +2.007 |
eus | +1.870 |
ushima | +1.865 |
ptoms | +1.785 |
lyak | +1.775 |
Neg logit contributions | |
rewritten | -2.196 |
rewriting | -1.946 |
dodged | -1.912 |
acci | -1.885 |
disreg | -1.830 |
Token | Dem |
Token ID | 11522 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ッド | +4.041 |
eus | +3.747 |
ushima | +3.708 |
ptoms | +3.512 |
lyak | +3.452 |
Neg logit contributions | |
rewritten | -7.006 |
rewriting | -6.357 |
dodged | -6.247 |
acci | -6.176 |
disreg | -6.022 |
Token | ocrat |
Token ID | 35128 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
ッド | +4.738 |
eus | +4.607 |
ushima | +4.531 |
ptoms | +4.325 |
lyak | +4.316 |
Neg logit contributions | |
rewritten | -6.016 |
rewriting | -5.353 |
dodged | -5.311 |
acci | -5.068 |
travers | -5.045 |
Token | ) |
Token ID | 8 |
Feature activation | +7.45e-02 |
Pos logit contributions | |
ッド | +1.478 |
eus | +1.358 |
ushima | +1.329 |
ptoms | +1.264 |
lyak | +1.245 |
Neg logit contributions | |
rewritten | -2.411 |
rewriting | -2.191 |
dodged | -2.167 |
acci | -2.126 |
travers | -2.073 |
Token | Buy |
Token ID | 14518 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
rewritten | +0.337 |
rewriting | +0.308 |
dodged | +0.303 |
acci | +0.299 |
disreg | +0.295 |
Neg logit contributions | |
ッド | -0.143 |
eus | -0.129 |
ushima | -0.127 |
ptoms | -0.120 |
lyak | -0.119 |
Token | Photo |
Token ID | 5555 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
ッド | +5.083 |
eus | +4.681 |
ushima | +4.289 |
Cs | +4.094 |
ptoms | +4.066 |
Neg logit contributions | |
rewritten | -10.938 |
rewriting | -10.272 |
acci | -10.167 |
dodged | -9.916 |
fict | -9.781 |
Token |
|
Token ID | 198 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
ッド | +2.218 |
eus | +2.109 |
ushima | +2.044 |
ptoms | +2.012 |
lyak | +1.946 |
Neg logit contributions | |
rewritten | -2.420 |
rewriting | -2.182 |
dodged | -2.132 |
acci | -2.102 |
disreg | -2.025 |
Token |
|
Token ID | 198 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
ッド | +2.282 |
eus | +2.171 |
ushima | +2.124 |
lyak | +2.052 |
ptoms | +2.048 |
Neg logit contributions | |
rewritten | -2.418 |
rewriting | -2.155 |
dodged | -2.120 |
acci | -2.064 |
travers | -2.011 |
Token | Sign |
Token ID | 11712 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ッド | +2.019 |
eus | +1.878 |
ushima | +1.834 |
ptoms | +1.755 |
lyak | +1.749 |
Neg logit contributions | |
rewritten | -3.295 |
rewriting | -2.980 |
dodged | -2.928 |
acci | -2.878 |
disreg | -2.831 |
Token | ing |
Token ID | 278 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ッド | +3.318 |
eus | +3.154 |
ushima | +3.036 |
ptoms | +2.943 |
lyak | +2.878 |
Neg logit contributions | |
rewritten | -4.656 |
rewriting | -4.182 |
dodged | -4.127 |
acci | -4.030 |
disreg | -3.960 |
Token | Cl |
Token ID | 1012 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
rewritten | +0.086 |
rewriting | +0.076 |
dodged | +0.074 |
acci | +0.072 |
disreg | +0.071 |
Neg logit contributions | |
ッド | -0.081 |
eus | -0.076 |
ushima | -0.076 |
ptoms | -0.073 |
lyak | -0.073 |
Token | arity |
Token ID | 6806 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
rewritten | +4.569 |
acci | +4.075 |
rewriting | +3.893 |
dodged | +3.874 |
postponed | +3.758 |
Neg logit contributions | |
ッド | -4.201 |
ushima | -3.654 |
eus | -3.562 |
lyak | -3.546 |
ptoms | -3.490 |
Token | Gaming |
Token ID | 14426 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
rewritten | +1.632 |
rewriting | +1.392 |
dodged | +1.352 |
acci | +1.309 |
disreg | +1.299 |
Neg logit contributions | |
ッド | -2.235 |
ushima | -2.111 |
eus | -2.096 |
ptoms | -2.047 |
lyak | -2.038 |
Token |
|
Token ID | 198 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
rewritten | +2.400 |
rewriting | +2.099 |
dodged | +2.040 |
acci | +1.979 |
disreg | +1.964 |
Neg logit contributions | |
ッド | -2.643 |
ushima | -2.474 |
eus | -2.465 |
ptoms | -2.390 |
lyak | -2.375 |
Token |
|
Token ID | 198 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
ッド | +1.266 |
eus | +1.198 |
ushima | +1.185 |
lyak | +1.145 |
ptoms | +1.143 |
Neg logit contributions | |
rewritten | -1.245 |
rewriting | -1.097 |
dodged | -1.073 |
acci | -1.047 |
disreg | -1.023 |
Token | Good |
Token ID | 10248 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
ッド | +1.507 |
eus | +1.398 |
ushima | +1.385 |
ptoms | +1.327 |
lyak | +1.323 |
Neg logit contributions | |
rewritten | -2.229 |
rewriting | -2.003 |
dodged | -1.964 |
acci | -1.927 |
disreg | -1.899 |
Token | updates |
Token ID | 5992 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
ッド | +8.124 |
eus | +7.604 |
ushima | +7.372 |
ptoms | +7.298 |
rises | +7.245 |
Neg logit contributions | |
rewritten | -8.069 |
rewriting | -7.018 |
dodged | -6.819 |
acci | -6.805 |
disreg | -6.574 |
Token | ? |
Token ID | 30 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
ッド | +0.360 |
eus | +0.338 |
ushima | +0.337 |
ptoms | +0.326 |
lyak | +0.325 |
Neg logit contributions | |
rewritten | -0.327 |
rewriting | -0.286 |
dodged | -0.279 |
acci | -0.272 |
disreg | -0.267 |
Token | No |
Token ID | 1400 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
rewritten | +1.060 |
rewriting | +0.935 |
dodged | +0.914 |
acci | +0.886 |
disreg | +0.880 |
Neg logit contributions | |
ッド | -1.032 |
ushima | -0.962 |
eus | -0.960 |
ptoms | -0.929 |
lyak | -0.920 |
Token | more |
Token ID | 517 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
ッド | +2.701 |
eus | +2.540 |
ushima | +2.513 |
lyak | +2.424 |
ptoms | +2.420 |
Neg logit contributions | |
rewritten | -2.894 |
rewriting | -2.549 |
dodged | -2.498 |
acci | -2.458 |
disreg | -2.399 |
Token | bad |
Token ID | 2089 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
ッド | +1.914 |
eus | +1.814 |
ushima | +1.799 |
ptoms | +1.744 |
lyak | +1.744 |
Neg logit contributions | |
rewritten | -1.675 |
rewriting | -1.450 |
dodged | -1.417 |
acci | -1.401 |
disreg | -1.353 |
Token | managed |
Token ID | 5257 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
rewritten | +1.787 |
rewriting | +1.501 |
dodged | +1.484 |
acci | +1.431 |
disreg | +1.418 |
Neg logit contributions | |
ッド | -2.778 |
eus | -2.624 |
ushima | -2.608 |
ptoms | -2.556 |
lyak | -2.546 |
Token | to |
Token ID | 284 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
rewritten | +2.316 |
rewriting | +2.004 |
dodged | +1.944 |
disreg | +1.882 |
acci | +1.881 |
Neg logit contributions | |
ッド | -3.015 |
eus | -2.836 |
ushima | -2.820 |
ptoms | -2.784 |
lyak | -2.745 |
Token | achieve |
Token ID | 4620 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ッド | +0.549 |
eus | +0.510 |
ushima | +0.503 |
ptoms | +0.487 |
lyak | +0.484 |
Neg logit contributions | |
rewritten | -0.706 |
rewriting | -0.632 |
dodged | -0.619 |
acci | -0.604 |
disreg | -0.598 |
Token | after |
Token ID | 706 |
Feature activation | -6.99e-02 |
Pos logit contributions | |
rewritten | +0.399 |
rewriting | +0.346 |
dodged | +0.336 |
acci | +0.328 |
disreg | +0.323 |
Neg logit contributions | |
ッド | -0.492 |
eus | -0.465 |
ushima | -0.462 |
ptoms | -0.450 |
lyak | -0.447 |
Token | the |
Token ID | 262 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
ッド | +0.235 |
eus | +0.221 |
ushima | +0.219 |
ptoms | +0.213 |
lyak | +0.212 |
Neg logit contributions | |
rewritten | -0.217 |
rewriting | -0.190 |
dodged | -0.185 |
acci | -0.180 |
disreg | -0.178 |
Token | fall |
Token ID | 2121 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
rewritten | +0.923 |
rewriting | +0.828 |
dodged | +0.799 |
acci | +0.783 |
disreg | +0.777 |
Neg logit contributions | |
ッド | -0.787 |
eus | -0.734 |
ushima | -0.727 |
ptoms | -0.709 |
lyak | -0.701 |
Token | of |
Token ID | 286 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ッド | +7.988 |
ushima | +7.735 |
eus | +7.322 |
ptoms | +7.255 |
lyak | +7.167 |
Neg logit contributions | |
rewritten | -8.090 |
rewriting | -6.909 |
dodged | -6.839 |
acci | -6.702 |
disreg | -6.537 |
Token | Ukrainian |
Token ID | 13053 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ッド | +5.912 |
ushima | +5.727 |
eus | +5.483 |
lyak | +5.428 |
rises | +5.266 |
Neg logit contributions | |
rewritten | -7.613 |
dodged | -6.783 |
rewriting | -6.705 |
acci | -6.579 |
disreg | -6.287 |
Token | President |
Token ID | 1992 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
ッド | +5.474 |
ushima | +5.168 |
eus | +5.082 |
lyak | +5.026 |
rises | +4.911 |
Neg logit contributions | |
rewritten | -5.407 |
rewriting | -4.716 |
dodged | -4.711 |
acci | -4.586 |
travers | -4.371 |
Token | Viktor |
Token ID | 31096 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ッド | +0.527 |
eus | +0.493 |
ushima | +0.490 |
ptoms | +0.473 |
lyak | +0.471 |
Neg logit contributions | |
rewritten | -0.574 |
rewriting | -0.508 |
dodged | -0.496 |
acci | -0.487 |
disreg | -0.478 |
Token | Yanukovych |
Token ID | 43202 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
ッド | +3.429 |
ushima | +3.252 |
eus | +3.223 |
lyak | +3.187 |
ptoms | +3.085 |
Neg logit contributions | |
rewritten | -4.039 |
rewriting | -3.567 |
dodged | -3.522 |
acci | -3.459 |
disreg | -3.330 |
Token | is |
Token ID | 318 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
rewritten | +0.884 |
rewriting | +0.760 |
dodged | +0.738 |
acci | +0.711 |
disreg | +0.706 |
Neg logit contributions | |
ッド | -1.206 |
ushima | -1.127 |
eus | -1.127 |
ptoms | -1.101 |
lyak | -1.094 |
Token | subtly |
Token ID | 37847 |
Feature activation | +7.85e-01 |
Pos logit contributions | |
rewritten | +0.474 |
rewriting | +0.410 |
dodged | +0.398 |
acci | +0.385 |
disreg | +0.381 |
Neg logit contributions | |
ッド | -0.587 |
ushima | -0.554 |
eus | -0.551 |
ptoms | -0.538 |
lyak | -0.535 |
Token | — |
Token ID | 960 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
rewritten | +2.225 |
rewriting | +1.919 |
dodged | +1.864 |
disreg | +1.779 |
travers | +1.748 |
Neg logit contributions | |
ッド | -2.894 |
ushima | -2.757 |
eus | -2.722 |
ptoms | -2.689 |
lyak | -2.645 |
Token | and |
Token ID | 392 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
ッド | +1.020 |
eus | +0.952 |
ushima | +0.948 |
ptoms | +0.913 |
lyak | +0.909 |
Neg logit contributions | |
rewritten | -1.205 |
rewriting | -1.071 |
dodged | -1.048 |
acci | -1.028 |
disreg | -1.010 |
Token | not |
Token ID | 407 |
Feature activation | -9.10e-01 |
Pos logit contributions | |
ッド | +0.714 |
eus | +0.673 |
ushima | +0.670 |
ptoms | +0.649 |
lyak | +0.647 |
Neg logit contributions | |
rewritten | -0.635 |
rewriting | -0.554 |
dodged | -0.540 |
acci | -0.527 |
disreg | -0.517 |
Token | so |
Token ID | 523 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
ッド | +3.156 |
eus | +3.001 |
ushima | +2.958 |
lyak | +2.864 |
ptoms | +2.855 |
Neg logit contributions | |
rewritten | -2.648 |
rewriting | -2.301 |
dodged | -2.238 |
acci | -2.221 |
disreg | -2.121 |
Token | subtly |
Token ID | 37847 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ッド | +7.255 |
eus | +7.073 |
ushima | +6.696 |
lyak | +6.595 |
ptoms | +6.395 |
Neg logit contributions | |
rewritten | -8.239 |
rewriting | -7.379 |
acci | -7.308 |
dodged | -7.162 |
postponed | -6.860 |
Token | — |
Token ID | 960 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
rewritten | +0.505 |
rewriting | +0.443 |
dodged | +0.433 |
acci | +0.418 |
disreg | +0.417 |
Neg logit contributions | |
ッド | -0.513 |
ushima | -0.481 |
eus | -0.480 |
ptoms | -0.465 |
lyak | -0.462 |
Token | enc |
Token ID | 12685 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
ッド | +0.136 |
eus | +0.126 |
ushima | +0.126 |
ptoms | +0.122 |
lyak | +0.121 |
Neg logit contributions | |
rewritten | -0.160 |
rewriting | -0.143 |
dodged | -0.139 |
acci | -0.136 |
disreg | -0.134 |
Token | our |
Token ID | 454 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
ッド | +0.167 |
eus | +0.155 |
ushima | +0.153 |
ptoms | +0.147 |
lyak | +0.147 |
Neg logit contributions | |
rewritten | -0.244 |
rewriting | -0.219 |
dodged | -0.215 |
acci | -0.212 |
disreg | -0.208 |
Token | aging |
Token ID | 3039 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
ッド | +1.182 |
ushima | +1.103 |
eus | +1.101 |
ptoms | +1.053 |
lyak | +1.049 |
Neg logit contributions | |
rewritten | -1.726 |
rewriting | -1.557 |
dodged | -1.521 |
acci | -1.490 |
disreg | -1.480 |
Token | finances |
Token ID | 20903 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
rewritten | +2.233 |
rewriting | +1.904 |
dodged | +1.816 |
disreg | +1.750 |
acci | +1.741 |
Neg logit contributions | |
ッド | -3.228 |
eus | -3.077 |
ushima | -3.036 |
lyak | -2.969 |
ptoms | -2.959 |
Token | . |
Token ID | 13 |
Feature activation | -6.80e-02 |
Pos logit contributions | |
rewritten | +0.559 |
rewriting | +0.494 |
dodged | +0.482 |
acci | +0.471 |
disreg | +0.464 |
Neg logit contributions | |
ッド | -0.529 |
eus | -0.495 |
ushima | -0.493 |
ptoms | -0.476 |
lyak | -0.475 |
Token |
|
Token ID | 198 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
ッド | +0.220 |
eus | +0.207 |
ushima | +0.206 |
ptoms | +0.199 |
lyak | +0.198 |
Neg logit contributions | |
rewritten | -0.218 |
rewriting | -0.192 |
dodged | -0.187 |
acci | -0.183 |
disreg | -0.180 |
Token |
|
Token ID | 198 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
ッド | +0.701 |
eus | +0.662 |
ushima | +0.657 |
ptoms | +0.635 |
lyak | +0.634 |
Neg logit contributions | |
rewritten | -0.669 |
rewriting | -0.587 |
dodged | -0.573 |
acci | -0.560 |
disreg | -0.548 |
Token | R |
Token ID | 49 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
ッド | +1.177 |
eus | +1.077 |
ushima | +1.068 |
ptoms | +1.017 |
lyak | +1.013 |
Neg logit contributions | |
rewritten | -2.110 |
rewriting | -1.912 |
dodged | -1.879 |
acci | -1.847 |
disreg | -1.821 |
Token | ising |
Token ID | 1710 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
ッド | +0.208 |
eus | +0.195 |
ushima | +0.193 |
ptoms | +0.188 |
lyak | +0.187 |
Neg logit contributions | |
rewritten | -0.196 |
rewriting | -0.172 |
dodged | -0.168 |
acci | -0.165 |
disreg | -0.161 |
Token | costs |
Token ID | 3484 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
ッド | +7.602 |
ushima | +7.313 |
eus | +7.185 |
rises | +7.044 |
loo | +6.944 |
Neg logit contributions | |
rewritten | -7.834 |
acci | -6.957 |
rewriting | -6.822 |
dodged | -6.794 |
travers | -6.375 |
Token | for |
Token ID | 329 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
ッド | +1.816 |
eus | +1.728 |
ushima | +1.707 |
ptoms | +1.665 |
lyak | +1.653 |
Neg logit contributions | |
rewritten | -1.720 |
rewriting | -1.509 |
dodged | -1.463 |
acci | -1.457 |
disreg | -1.401 |
Token | the |
Token ID | 262 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
rewritten | +1.418 |
rewriting | +1.264 |
dodged | +1.233 |
acci | +1.196 |
disreg | +1.193 |
Neg logit contributions | |
ッド | -1.178 |
eus | -1.101 |
ushima | -1.085 |
lyak | -1.052 |
ptoms | -1.042 |
Token | B |
Token ID | 347 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
rewritten | +1.679 |
rewriting | +1.482 |
dodged | +1.444 |
acci | +1.404 |
disreg | +1.394 |
Neg logit contributions | |
ッド | -1.618 |
eus | -1.518 |
ushima | -1.501 |
lyak | -1.454 |
ptoms | -1.448 |
Token | ip |
Token ID | 541 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
rewritten | +1.820 |
rewriting | +1.626 |
dodged | +1.586 |
acci | +1.569 |
disreg | +1.539 |
Neg logit contributions | |
ッド | -1.509 |
eus | -1.409 |
ushima | -1.406 |
lyak | -1.353 |
ptoms | -1.347 |
Token | a |
Token ID | 257 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
ッド | +2.117 |
eus | +1.986 |
ushima | +1.982 |
ptoms | +1.913 |
lyak | +1.891 |
Neg logit contributions | |
rewritten | -2.509 |
rewriting | -2.211 |
dodged | -2.186 |
acci | -2.141 |
disreg | -2.084 |
Token | time |
Token ID | 640 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
rewritten | +0.776 |
rewriting | +0.633 |
dodged | +0.604 |
acci | +0.578 |
disreg | +0.565 |
Neg logit contributions | |
ッド | -1.635 |
eus | -1.559 |
ushima | -1.555 |
ptoms | -1.518 |
lyak | -1.513 |
Token | . |
Token ID | 13 |
Feature activation | -5.62e-02 |
Pos logit contributions | |
rewritten | +2.358 |
rewriting | +2.099 |
dodged | +2.018 |
acci | +1.983 |
disreg | +1.971 |
Neg logit contributions | |
ッド | -2.459 |
eus | -2.302 |
ushima | -2.283 |
ptoms | -2.215 |
lyak | -2.208 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
ッド | +0.181 |
eus | +0.170 |
ushima | +0.169 |
ptoms | +0.164 |
lyak | +0.163 |
Neg logit contributions | |
rewritten | -0.182 |
rewriting | -0.160 |
dodged | -0.156 |
acci | -0.153 |
disreg | -0.150 |
Token |
|
Token ID | 198 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
ッド | +0.079 |
eus | +0.074 |
ushima | +0.074 |
ptoms | +0.071 |
lyak | +0.071 |
Neg logit contributions | |
rewritten | -0.070 |
rewriting | -0.061 |
dodged | -0.060 |
acci | -0.058 |
disreg | -0.057 |
Token | Perfect |
Token ID | 36635 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
ッド | +1.346 |
eus | +1.241 |
ushima | +1.231 |
ptoms | +1.176 |
lyak | +1.173 |
Neg logit contributions | |
rewritten | -2.140 |
rewriting | -1.930 |
dodged | -1.894 |
acci | -1.862 |
disreg | -1.833 |
Token | Trade |
Token ID | 9601 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ッド | +7.642 |
eus | +7.447 |
ushima | +7.324 |
ptoms | +7.169 |
rises | +6.990 |
Neg logit contributions | |
rewritten | -7.955 |
rewriting | -6.954 |
dodged | -6.826 |
acci | -6.786 |
grop | -6.441 |
Token | School |
Token ID | 3961 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
ッド | +3.053 |
eus | +2.848 |
ushima | +2.833 |
ptoms | +2.730 |
lyak | +2.716 |
Neg logit contributions | |
rewritten | -3.790 |
rewriting | -3.377 |
dodged | -3.301 |
acci | -3.251 |
disreg | -3.186 |
Token | For |
Token ID | 1114 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
rewritten | +0.978 |
rewriting | +0.859 |
dodged | +0.834 |
acci | +0.818 |
disreg | +0.805 |
Neg logit contributions | |
ッド | -0.999 |
eus | -0.923 |
ushima | -0.921 |
lyak | -0.886 |
ptoms | -0.884 |
Token | Veterans |
Token ID | 18950 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ッド | +0.044 |
eus | +0.041 |
ushima | +0.041 |
ptoms | +0.040 |
lyak | +0.040 |
Neg logit contributions | |
rewritten | -0.041 |
rewriting | -0.036 |
dodged | -0.035 |
acci | -0.034 |
disreg | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
rewritten | +3.220 |
rewriting | +2.829 |
dodged | +2.731 |
acci | +2.660 |
disreg | +2.643 |
Neg logit contributions | |
ッド | -3.795 |
eus | -3.523 |
ushima | -3.507 |
ptoms | -3.388 |
lyak | -3.374 |
Token | and |
Token ID | 290 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
ッド | +1.143 |
eus | +1.074 |
ushima | +1.070 |
ptoms | +1.036 |
lyak | +1.035 |
Neg logit contributions | |
rewritten | -1.041 |
rewriting | -0.908 |
dodged | -0.885 |
acci | -0.867 |
disreg | -0.848 |
Token | they |
Token ID | 484 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ッド | +1.900 |
eus | +1.789 |
ushima | +1.768 |
ptoms | +1.714 |
lyak | +1.712 |
Neg logit contributions | |
rewritten | -1.855 |
rewriting | -1.627 |
dodged | -1.591 |
acci | -1.566 |
disreg | -1.519 |
Token | � |
Token ID | 447 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
ッド | +3.475 |
eus | +3.283 |
ushima | +3.242 |
ptoms | +3.133 |
lyak | +3.117 |
Neg logit contributions | |
rewritten | -3.880 |
rewriting | -3.449 |
dodged | -3.349 |
acci | -3.297 |
disreg | -3.214 |
Token | � |
Token ID | 247 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
ッド | +0.375 |
eus | +0.353 |
ushima | +0.351 |
ptoms | +0.339 |
lyak | +0.337 |
Neg logit contributions | |
rewritten | -0.373 |
rewriting | -0.328 |
dodged | -0.321 |
acci | -0.313 |
disreg | -0.307 |
Token | re |
Token ID | 260 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
ッド | +0.829 |
eus | +0.766 |
ushima | +0.763 |
ptoms | +0.734 |
lyak | +0.733 |
Neg logit contributions | |
rewritten | -1.066 |
rewriting | -0.951 |
dodged | -0.933 |
acci | -0.918 |
disreg | -0.900 |
Token | pouring |
Token ID | 23147 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
rewritten | +0.831 |
rewriting | +0.724 |
dodged | +0.699 |
acci | +0.677 |
disreg | +0.671 |
Neg logit contributions | |
ッド | -1.024 |
ushima | -0.970 |
eus | -0.968 |
ptoms | -0.943 |
lyak | -0.938 |
Token | ink |
Token ID | 16882 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
ッド | +7.597 |
eus | +7.111 |
ushima | +6.856 |
lyak | +6.784 |
ptoms | +6.723 |
Neg logit contributions | |
rewritten | -8.235 |
rewriting | -7.292 |
dodged | -7.224 |
acci | -7.075 |
disreg | -6.814 |
Token | into |
Token ID | 656 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
rewritten | +0.609 |
rewriting | +0.533 |
dodged | +0.517 |
acci | +0.509 |
disreg | +0.500 |
Neg logit contributions | |
ッド | -0.668 |
ushima | -0.627 |
eus | -0.624 |
lyak | -0.605 |
ptoms | -0.603 |
Token | them |
Token ID | 606 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
ッド | +2.871 |
eus | +2.689 |
ushima | +2.645 |
ptoms | +2.554 |
lyak | +2.549 |
Neg logit contributions | |
rewritten | -3.401 |
rewriting | -3.016 |
dodged | -2.956 |
acci | -2.900 |
disreg | -2.837 |
Token | , |
Token ID | 11 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
ッド | +2.354 |
eus | +2.219 |
ushima | +2.202 |
ptoms | +2.129 |
lyak | +2.119 |
Neg logit contributions | |
rewritten | -2.315 |
rewriting | -2.029 |
dodged | -1.987 |
acci | -1.935 |
disreg | -1.894 |
Token | and |
Token ID | 290 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
ッド | +1.789 |
eus | +1.690 |
ushima | +1.681 |
ptoms | +1.627 |
lyak | +1.624 |
Neg logit contributions | |
rewritten | -1.503 |
rewriting | -1.301 |
dodged | -1.269 |
acci | -1.241 |
disreg | -1.207 |
Token | the |
Token ID | 262 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ッド | +1.139 |
eus | +1.069 |
ushima | +1.062 |
ptoms | +1.022 |
lyak | +1.022 |
Neg logit contributions | |
rewritten | -1.227 |
rewriting | -1.083 |
dodged | -1.060 |
acci | -1.040 |
disreg | -1.016 |
Token | Brune |
Token ID | 46070 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ッド | +0.557 |
eus | +0.525 |
ushima | +0.521 |
ptoms | +0.504 |
lyak | +0.503 |
Neg logit contributions | |
rewritten | -0.520 |
rewriting | -0.455 |
dodged | -0.445 |
acci | -0.435 |
disreg | -0.425 |
Token | i |
Token ID | 72 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
ッド | +0.139 |
eus | +0.110 |
ushima | +0.108 |
ptoms | +0.097 |
lyak | +0.096 |
Neg logit contributions | |
rewritten | -0.692 |
rewriting | -0.641 |
dodged | -0.633 |
acci | -0.627 |
disreg | -0.619 |
Token | Currency |
Token ID | 20113 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
ッド | +1.766 |
eus | +1.663 |
ushima | +1.652 |
ptoms | +1.596 |
lyak | +1.593 |
Neg logit contributions | |
rewritten | -1.666 |
rewriting | -1.458 |
dodged | -1.427 |
acci | -1.393 |
disreg | -1.360 |
Token | and |
Token ID | 290 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
ッド | +2.298 |
eus | +2.171 |
ushima | +2.148 |
lyak | +2.069 |
ptoms | +2.064 |
Neg logit contributions | |
rewritten | -2.176 |
rewriting | -1.903 |
dodged | -1.864 |
acci | -1.819 |
disreg | -1.768 |
Token | Monetary |
Token ID | 34112 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
ッド | +1.407 |
eus | +1.326 |
ushima | +1.311 |
lyak | +1.273 |
ptoms | +1.268 |
Neg logit contributions | |
rewritten | -1.400 |
rewriting | -1.231 |
dodged | -1.202 |
acci | -1.180 |
disreg | -1.148 |
Token | Board |
Token ID | 5926 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
ッド | +7.065 |
eus | +6.676 |
ushima | +6.435 |
ptoms | +6.305 |
arov | +6.209 |
Neg logit contributions | |
rewritten | -8.596 |
rewriting | -7.619 |
dodged | -7.585 |
acci | -7.321 |
travers | -7.146 |
Token | .[ |
Token ID | 3693 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
ッド | +0.183 |
eus | +0.172 |
ushima | +0.172 |
ptoms | +0.166 |
lyak | +0.166 |
Neg logit contributions | |
rewritten | -0.168 |
rewriting | -0.147 |
dodged | -0.143 |
acci | -0.140 |
disreg | -0.138 |
Token | 18 |
Token ID | 1507 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
ッド | +1.600 |
eus | +1.505 |
ushima | +1.487 |
ptoms | +1.430 |
lyak | +1.426 |
Neg logit contributions | |
rewritten | -1.799 |
rewriting | -1.591 |
dodged | -1.560 |
acci | -1.522 |
disreg | -1.500 |
Token | ] |
Token ID | 60 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
ッド | +2.482 |
eus | +2.302 |
ushima | +2.266 |
ptoms | +2.186 |
lyak | +2.153 |
Neg logit contributions | |
rewritten | -3.376 |
rewriting | -3.039 |
dodged | -2.978 |
acci | -2.906 |
disreg | -2.853 |
Token |
|
Token ID | 198 |
Feature activation | +7.02e-03 |
Pos logit contributions | |
rewritten | +1.719 |
rewriting | +1.493 |
dodged | +1.454 |
acci | +1.438 |
disreg | +1.417 |
Neg logit contributions | |
ッド | -1.910 |
ushima | -1.784 |
eus | -1.769 |
ptoms | -1.733 |
lyak | -1.732 |
Token | to |
Token ID | 284 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
rewritten | +1.152 |
rewriting | +1.004 |
dodged | +0.973 |
acci | +0.945 |
travers | +0.934 |
Neg logit contributions | |
ッド | -1.341 |
ushima | -1.273 |
eus | -1.262 |
ptoms | -1.235 |
lyak | -1.227 |
Token | the |
Token ID | 262 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
ッド | +2.442 |
eus | +2.299 |
ushima | +2.283 |
ptoms | +2.205 |
lyak | +2.203 |
Neg logit contributions | |
rewritten | -2.349 |
rewriting | -2.052 |
dodged | -2.011 |
acci | -1.975 |
disreg | -1.923 |
Token | NFL |
Token ID | 5134 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ッド | +0.603 |
eus | +0.569 |
ushima | +0.566 |
ptoms | +0.548 |
lyak | +0.547 |
Neg logit contributions | |
rewritten | -0.534 |
rewriting | -0.465 |
dodged | -0.453 |
acci | -0.443 |
disreg | -0.434 |
Token | Championship |
Token ID | 10749 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
ッド | +3.433 |
eus | +3.209 |
ushima | +3.181 |
ptoms | +3.089 |
lyak | +3.087 |
Neg logit contributions | |
rewritten | -3.466 |
rewriting | -3.036 |
dodged | -2.997 |
acci | -2.921 |
disreg | -2.848 |
Token | in |
Token ID | 287 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
rewritten | +0.133 |
rewriting | +0.117 |
dodged | +0.114 |
acci | +0.111 |
disreg | +0.110 |
Neg logit contributions | |
ッド | -0.136 |
eus | -0.128 |
ushima | -0.128 |
lyak | -0.123 |
ptoms | -0.123 |
Token | 1949 |
Token ID | 24977 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
ッド | +0.293 |
eus | +0.277 |
ushima | +0.276 |
ptoms | +0.268 |
lyak | +0.268 |
Neg logit contributions | |
rewritten | -0.203 |
rewriting | -0.174 |
dodged | -0.168 |
acci | -0.162 |
disreg | -0.160 |
Token | — |
Token ID | 960 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
ッド | +6.825 |
ushima | +6.245 |
eus | +6.221 |
lyak | +6.001 |
rises | +5.904 |
Neg logit contributions | |
rewritten | -8.854 |
dodged | -7.887 |
rewriting | -7.864 |
acci | -7.779 |
redirected | -7.345 |
Token | a |
Token ID | 64 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
ッド | +1.907 |
eus | +1.782 |
ushima | +1.770 |
ptoms | +1.703 |
lyak | +1.700 |
Neg logit contributions | |
rewritten | -2.211 |
rewriting | -1.959 |
dodged | -1.919 |
acci | -1.883 |
disreg | -1.846 |
Token | game |
Token ID | 983 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
ッド | +0.824 |
eus | +0.777 |
ushima | +0.774 |
ptoms | +0.749 |
lyak | +0.747 |
Neg logit contributions | |
rewritten | -0.725 |
rewriting | -0.631 |
dodged | -0.615 |
acci | -0.601 |
disreg | -0.588 |
Token | in |
Token ID | 287 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
rewritten | +1.738 |
rewriting | +1.534 |
dodged | +1.495 |
disreg | +1.445 |
acci | +1.441 |
Neg logit contributions | |
ッド | -1.660 |
eus | -1.564 |
ushima | -1.552 |
ptoms | -1.498 |
lyak | -1.485 |
Token | which |
Token ID | 543 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
rewritten | +2.531 |
rewriting | +2.250 |
dodged | +2.182 |
disreg | +2.124 |
acci | +2.110 |
Neg logit contributions | |
ッド | -2.339 |
eus | -2.189 |
ushima | -2.166 |
ptoms | -2.102 |
lyak | -2.080 |
Token | was |
Token ID | 373 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
ッド | +0.534 |
eus | +0.508 |
ushima | +0.505 |
ptoms | +0.490 |
lyak | +0.490 |
Neg logit contributions | |
rewritten | -0.371 |
rewriting | -0.316 |
dodged | -0.307 |
acci | -0.300 |
disreg | -0.291 |
Token | done |
Token ID | 1760 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
rewritten | +1.388 |
rewriting | +1.114 |
dodged | +1.081 |
disreg | +1.016 |
acci | +1.009 |
Neg logit contributions | |
ッド | -2.899 |
ushima | -2.776 |
eus | -2.738 |
ptoms | -2.713 |
lyak | -2.687 |
Token | in |
Token ID | 287 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
rewritten | +2.392 |
rewriting | +2.064 |
dodged | +1.980 |
disreg | +1.923 |
acci | +1.921 |
Neg logit contributions | |
ッド | -3.262 |
ushima | -3.053 |
eus | -3.043 |
ptoms | -2.989 |
lyak | -2.957 |
Token | April |
Token ID | 3035 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
rewritten | +2.089 |
rewriting | +1.831 |
dodged | +1.763 |
disreg | +1.709 |
acci | +1.700 |
Neg logit contributions | |
ッド | -2.420 |
eus | -2.275 |
ushima | -2.264 |
ptoms | -2.201 |
lyak | -2.178 |
Token | and |
Token ID | 290 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
ッド | +3.740 |
eus | +3.547 |
ushima | +3.533 |
lyak | +3.419 |
ptoms | +3.380 |
Neg logit contributions | |
rewritten | -3.510 |
rewriting | -3.061 |
dodged | -2.990 |
acci | -2.947 |
disreg | -2.828 |
Token | May |
Token ID | 1737 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
ッド | +0.454 |
eus | +0.428 |
ushima | +0.426 |
ptoms | +0.413 |
lyak | +0.412 |
Neg logit contributions | |
rewritten | -0.384 |
rewriting | -0.334 |
dodged | -0.325 |
acci | -0.317 |
disreg | -0.311 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
ッド | +7.458 |
eus | +7.283 |
ushima | +7.184 |
lyak | +6.998 |
rises | +6.908 |
Neg logit contributions | |
rewritten | -7.668 |
rewriting | -6.654 |
acci | -6.606 |
dodged | -6.555 |
disreg | -6.100 |
Token | , |
Token ID | 11 |
Feature activation | -7.36e-02 |
Pos logit contributions | |
ッド | +5.027 |
eus | +4.791 |
ushima | +4.705 |
lyak | +4.598 |
ptoms | +4.566 |
Neg logit contributions | |
rewritten | -4.945 |
rewriting | -4.310 |
dodged | -4.271 |
acci | -4.244 |
disreg | -3.985 |
Token | involving |
Token ID | 7411 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
ッド | +0.255 |
eus | +0.240 |
ushima | +0.239 |
ptoms | +0.232 |
lyak | +0.231 |
Neg logit contributions | |
rewritten | -0.221 |
rewriting | -0.193 |
dodged | -0.187 |
acci | -0.182 |
disreg | -0.180 |
Token | responses |
Token ID | 9109 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
rewritten | +2.613 |
rewriting | +2.275 |
dodged | +2.197 |
disreg | +2.135 |
travers | +2.116 |
Neg logit contributions | |
ッド | -3.160 |
eus | -2.968 |
ushima | -2.962 |
ptoms | -2.901 |
lyak | -2.873 |
Token | from |
Token ID | 422 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
ッド | +1.941 |
eus | +1.854 |
ushima | +1.835 |
lyak | +1.782 |
ptoms | +1.777 |
Neg logit contributions | |
rewritten | -1.592 |
rewriting | -1.388 |
dodged | -1.338 |
acci | -1.335 |
disreg | -1.276 |
Token | station |
Token ID | 4429 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
ッド | +2.747 |
eus | +2.556 |
ushima | +2.525 |
lyak | +2.386 |
ptoms | +2.374 |
Neg logit contributions | |
rewritten | -4.919 |
rewriting | -4.438 |
dodged | -4.366 |
acci | -4.344 |
disreg | -4.206 |
Token | on |
Token ID | 319 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
ッド | +0.845 |
eus | +0.799 |
ushima | +0.794 |
ptoms | +0.769 |
lyak | +0.768 |
Neg logit contributions | |
rewritten | -0.726 |
rewriting | -0.631 |
dodged | -0.615 |
acci | -0.601 |
disreg | -0.587 |
Token | the |
Token ID | 262 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
rewritten | +0.109 |
rewriting | +0.096 |
dodged | +0.093 |
acci | +0.091 |
disreg | +0.090 |
Neg logit contributions | |
ッド | -0.115 |
eus | -0.108 |
ushima | -0.107 |
ptoms | -0.104 |
lyak | -0.104 |
Token | Russian |
Token ID | 3394 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
rewritten | +1.090 |
rewriting | +0.967 |
dodged | +0.932 |
disreg | +0.909 |
travers | +0.906 |
Neg logit contributions | |
ッド | -1.050 |
eus | -0.976 |
ushima | -0.968 |
ptoms | -0.941 |
lyak | -0.924 |
Token | - |
Token ID | 12 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ッド | +0.782 |
eus | +0.733 |
ushima | +0.726 |
ptoms | +0.701 |
lyak | +0.693 |
Neg logit contributions | |
rewritten | -0.921 |
rewriting | -0.819 |
dodged | -0.799 |
acci | -0.781 |
disreg | -0.771 |
Token | Uk |
Token ID | 28425 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
ッド | +3.337 |
eus | +3.140 |
ushima | +3.097 |
lyak | +2.987 |
ptoms | +2.962 |
Neg logit contributions | |
rewritten | -4.622 |
rewriting | -4.135 |
dodged | -4.057 |
acci | -4.011 |
disreg | -3.908 |
Token | rain |
Token ID | 3201 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
ッド | +2.496 |
ushima | +2.375 |
eus | +2.162 |
lyak | +2.087 |
rises | +1.984 |
Neg logit contributions | |
rewritten | -13.137 |
dodged | -12.222 |
rewriting | -12.219 |
acci | -12.050 |
travers | -11.771 |
Token | ian |
Token ID | 666 |
Feature activation | -9.44e-01 |
Pos logit contributions | |
ッド | +2.037 |
eus | +1.913 |
ushima | +1.905 |
lyak | +1.797 |
ptoms | +1.791 |
Neg logit contributions | |
rewritten | -3.331 |
rewriting | -3.013 |
dodged | -2.956 |
acci | -2.901 |
disreg | -2.860 |
Token | border |
Token ID | 4865 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
ッド | +3.079 |
eus | +2.898 |
ushima | +2.897 |
lyak | +2.805 |
ptoms | +2.780 |
Neg logit contributions | |
rewritten | -2.968 |
rewriting | -2.601 |
dodged | -2.545 |
acci | -2.499 |
disreg | -2.430 |
Token | , |
Token ID | 11 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
ッド | +0.982 |
eus | +0.927 |
ushima | +0.926 |
ptoms | +0.891 |
lyak | +0.891 |
Neg logit contributions | |
rewritten | -0.942 |
rewriting | -0.826 |
dodged | -0.804 |
acci | -0.791 |
disreg | -0.771 |
Token | says |
Token ID | 1139 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
ッド | +0.278 |
eus | +0.265 |
ushima | +0.264 |
ptoms | +0.257 |
lyak | +0.256 |
Neg logit contributions | |
rewritten | -0.169 |
rewriting | -0.142 |
dodged | -0.137 |
acci | -0.133 |
disreg | -0.129 |
Token | 12 |
Token ID | 1105 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
rewritten | +2.173 |
rewriting | +1.898 |
dodged | +1.862 |
acci | +1.812 |
disreg | +1.788 |
Neg logit contributions | |
ッド | -2.278 |
eus | -2.114 |
ushima | -2.110 |
lyak | -2.064 |
ptoms | -2.043 |
Token | ms |
Token ID | 907 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
ッド | +1.435 |
eus | +1.329 |
ushima | +1.320 |
ptoms | +1.268 |
lyak | +1.260 |
Neg logit contributions | |
rewritten | -2.011 |
rewriting | -1.804 |
dodged | -1.769 |
acci | -1.738 |
disreg | -1.709 |
Token | G |
Token ID | 402 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
rewritten | +1.938 |
rewriting | +1.709 |
dodged | +1.668 |
acci | +1.626 |
disreg | +1.610 |
Neg logit contributions | |
ッド | -1.864 |
ushima | -1.739 |
eus | -1.721 |
lyak | -1.676 |
ptoms | -1.673 |
Token | 2 |
Token ID | 17 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
rewritten | +3.257 |
rewriting | +2.788 |
dodged | +2.668 |
acci | +2.604 |
disreg | +2.567 |
Neg logit contributions | |
ッド | -4.627 |
eus | -4.370 |
ushima | -4.341 |
ptoms | -4.223 |
lyak | -4.221 |
Token | G |
Token ID | 38 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
ッド | +0.575 |
ushima | +0.528 |
eus | +0.525 |
ptoms | +0.506 |
lyak | +0.502 |
Neg logit contributions | |
rewritten | -0.896 |
rewriting | -0.804 |
dodged | -0.789 |
acci | -0.775 |
disreg | -0.766 |
Token | Samsung |
Token ID | 10397 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
rewritten | +1.412 |
rewriting | +1.111 |
dodged | +1.059 |
acci | +1.007 |
disreg | +0.977 |
Neg logit contributions | |
ッド | -3.587 |
ushima | -3.419 |
eus | -3.407 |
lyak | -3.344 |
ptoms | -3.344 |
Token | P |
Token ID | 350 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
ッド | +7.216 |
eus | +6.991 |
ushima | +6.598 |
ptoms | +6.496 |
rises | +6.407 |
Neg logit contributions | |
rewritten | -8.206 |
rewriting | -7.384 |
dodged | -7.188 |
acci | -7.155 |
travers | -6.893 |
Token | LS |
Token ID | 6561 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
rewritten | +1.426 |
rewriting | +1.255 |
dodged | +1.228 |
acci | +1.206 |
disreg | +1.188 |
Neg logit contributions | |
ッド | -1.327 |
eus | -1.236 |
ushima | -1.226 |
ptoms | -1.187 |
lyak | -1.176 |
Token | ( |
Token ID | 357 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
rewritten | +1.902 |
rewriting | +1.684 |
dodged | +1.639 |
acci | +1.606 |
disreg | +1.592 |
Neg logit contributions | |
ッド | -1.689 |
ushima | -1.569 |
eus | -1.552 |
lyak | -1.524 |
ptoms | -1.519 |
Token | Response |
Token ID | 31077 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
ッド | +0.735 |
eus | +0.677 |
ushima | +0.674 |
ptoms | +0.645 |
lyak | +0.644 |
Neg logit contributions | |
rewritten | -1.112 |
rewriting | -1.000 |
dodged | -0.981 |
acci | -0.964 |
disreg | -0.949 |
Token | Time |
Token ID | 3862 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
ッド | +2.224 |
eus | +2.112 |
ushima | +2.101 |
ptoms | +2.043 |
lyak | +2.033 |
Neg logit contributions | |
rewritten | -1.447 |
rewriting | -1.222 |
dodged | -1.179 |
acci | -1.155 |
disreg | -1.122 |
Token | - |
Token ID | 12 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ッド | +2.578 |
eus | +2.415 |
ushima | +2.402 |
ptoms | +2.317 |
lyak | +2.311 |
Neg logit contributions | |
rewritten | -2.763 |
rewriting | -2.441 |
dodged | -2.386 |
acci | -2.338 |
disreg | -2.293 |
Token | based |
Token ID | 3106 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
ッド | +3.366 |
eus | +3.147 |
ushima | +3.116 |
ptoms | +3.017 |
lyak | +2.986 |
Neg logit contributions | |
rewritten | -4.211 |
rewriting | -3.750 |
dodged | -3.659 |
acci | -3.598 |
disreg | -3.540 |
Token | remedies |
Token ID | 29604 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
ッド | +2.751 |
eus | +2.568 |
ushima | +2.546 |
ptoms | +2.454 |
lyak | +2.444 |
Neg logit contributions | |
rewritten | -3.293 |
rewriting | -2.927 |
dodged | -2.865 |
acci | -2.819 |
disreg | -2.757 |
Token | instead |
Token ID | 2427 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
ッド | +2.858 |
eus | +2.704 |
ushima | +2.683 |
lyak | +2.598 |
ptoms | +2.592 |
Neg logit contributions | |
rewritten | -2.756 |
rewriting | -2.424 |
dodged | -2.363 |
acci | -2.326 |
disreg | -2.246 |
Token | of |
Token ID | 286 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
rewritten | +2.001 |
rewriting | +1.724 |
dodged | +1.668 |
acci | +1.629 |
disreg | +1.599 |
Neg logit contributions | |
ッド | -2.688 |
eus | -2.535 |
ushima | -2.530 |
ptoms | -2.453 |
lyak | -2.451 |
Token | pharmaceutical |
Token ID | 18423 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
ッド | +0.198 |
eus | +0.186 |
ushima | +0.185 |
lyak | +0.179 |
ptoms | +0.179 |
Neg logit contributions | |
rewritten | -0.198 |
rewriting | -0.175 |
dodged | -0.170 |
acci | -0.165 |
disreg | -0.164 |
Token | s |
Token ID | 82 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
ッド | +6.298 |
eus | +5.906 |
ushima | +5.768 |
ptoms | +5.590 |
rises | +5.473 |
Neg logit contributions | |
rewritten | -9.172 |
rewriting | -8.180 |
acci | -8.134 |
dodged | -8.099 |
postponed | -7.726 |
Token | . |
Token ID | 13 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
ッド | +2.368 |
eus | +2.222 |
ushima | +2.210 |
ptoms | +2.133 |
lyak | +2.127 |
Neg logit contributions | |
rewritten | -2.573 |
rewriting | -2.281 |
dodged | -2.233 |
acci | -2.195 |
disreg | -2.134 |
Token | Home |
Token ID | 5995 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
rewritten | +0.759 |
rewriting | +0.666 |
dodged | +0.648 |
acci | +0.638 |
disreg | +0.632 |
Neg logit contributions | |
ッド | -0.723 |
ushima | -0.682 |
eus | -0.680 |
ptoms | -0.659 |
lyak | -0.659 |
Token | opathy |
Token ID | 27189 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
rewritten | +1.136 |
rewriting | +1.027 |
dodged | +0.998 |
acci | +0.980 |
disreg | +0.972 |
Neg logit contributions | |
ッド | -0.716 |
eus | -0.657 |
ushima | -0.653 |
ptoms | -0.623 |
lyak | -0.621 |
Token | advocated |
Token ID | 25828 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
ッド | +1.288 |
eus | +1.223 |
ushima | +1.213 |
ptoms | +1.178 |
lyak | +1.178 |
Neg logit contributions | |
rewritten | -0.971 |
rewriting | -0.837 |
dodged | -0.808 |
acci | -0.800 |
disreg | -0.766 |
Token | not |
Token ID | 407 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
rewritten | +2.093 |
rewriting | +1.816 |
dodged | +1.777 |
acci | +1.740 |
disreg | +1.725 |
Neg logit contributions | |
ッド | -2.349 |
ushima | -2.193 |
eus | -2.187 |
lyak | -2.118 |
ptoms | -2.117 |
Token | only |
Token ID | 691 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
rewritten | +0.397 |
rewriting | +0.342 |
dodged | +0.334 |
acci | +0.321 |
disreg | +0.320 |
Neg logit contributions | |
ッド | -0.511 |
ushima | -0.482 |
eus | -0.482 |
lyak | -0.467 |
ptoms | -0.466 |
Token | on |
Token ID | 319 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
rewritten | +0.878 |
rewriting | +0.759 |
dodged | +0.739 |
acci | +0.722 |
disreg | +0.710 |
Neg logit contributions | |
ッド | -1.091 |
eus | -1.028 |
ushima | -1.025 |
lyak | -0.993 |
ptoms | -0.992 |
Token | her |
Token ID | 607 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
rewritten | +1.507 |
rewriting | +1.324 |
dodged | +1.293 |
acci | +1.253 |
disreg | +1.250 |
Neg logit contributions | |
ッド | -1.636 |
ushima | -1.542 |
eus | -1.528 |
ptoms | -1.488 |
lyak | -1.480 |
Token | ce |
Token ID | 2906 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
ッド | +0.209 |
eus | +0.197 |
ushima | +0.196 |
ptoms | +0.190 |
lyak | +0.189 |
Neg logit contributions | |
rewritten | -0.188 |
rewriting | -0.164 |
dodged | -0.160 |
acci | -0.156 |
disreg | -0.153 |
Token | int |
Token ID | 600 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
ッド | +2.280 |
eus | +2.124 |
ushima | +2.099 |
ptoms | +2.027 |
lyak | +2.012 |
Neg logit contributions | |
rewritten | -3.621 |
rewriting | -3.276 |
dodged | -3.194 |
acci | -3.134 |
disreg | -3.098 |
Token | ure |
Token ID | 495 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
rewritten | +5.350 |
rewriting | +4.716 |
acci | +4.673 |
dodged | +4.670 |
disreg | +4.593 |
Neg logit contributions | |
ッド | -4.207 |
ushima | -3.857 |
eus | -3.768 |
lyak | -3.732 |
ptoms | -3.728 |
Token | , |
Token ID | 11 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
ッド | +1.074 |
eus | +1.011 |
ushima | +1.006 |
ptoms | +0.973 |
lyak | +0.971 |
Neg logit contributions | |
rewritten | -0.996 |
rewriting | -0.871 |
dodged | -0.850 |
acci | -0.831 |
disreg | -0.815 |
Token | but |
Token ID | 475 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ッド | +0.920 |
eus | +0.869 |
ushima | +0.865 |
ptoms | +0.836 |
lyak | +0.834 |
Neg logit contributions | |
rewritten | -0.808 |
rewriting | -0.704 |
dodged | -0.685 |
acci | -0.671 |
disreg | -0.652 |
Token | on |
Token ID | 319 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
rewritten | +3.489 |
rewriting | +2.988 |
dodged | +2.875 |
disreg | +2.863 |
travers | +2.810 |
Neg logit contributions | |
ッド | -5.125 |
ushima | -4.799 |
eus | -4.778 |
ptoms | -4.685 |
lyak | -4.667 |
Token | her |
Token ID | 607 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
rewritten | +2.195 |
rewriting | +1.937 |
dodged | +1.889 |
acci | +1.846 |
disreg | +1.834 |
Neg logit contributions | |
ッド | -2.255 |
ushima | -2.130 |
eus | -2.104 |
ptoms | -2.051 |
lyak | -2.046 |
Token | of |
Token ID | 286 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
rewritten | +0.799 |
rewriting | +0.684 |
dodged | +0.668 |
acci | +0.637 |
disreg | +0.635 |
Neg logit contributions | |
ッド | -1.123 |
ushima | -1.058 |
eus | -1.056 |
ptoms | -1.025 |
lyak | -1.022 |
Token | the |
Token ID | 262 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
ッド | +0.952 |
eus | +0.896 |
ushima | +0.889 |
ptoms | +0.859 |
lyak | +0.858 |
Neg logit contributions | |
rewritten | -0.954 |
rewriting | -0.839 |
dodged | -0.820 |
acci | -0.802 |
disreg | -0.786 |
Token | asteroid |
Token ID | 27460 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
ッド | +0.569 |
eus | +0.534 |
ushima | +0.533 |
ptoms | +0.515 |
lyak | +0.513 |
Neg logit contributions | |
rewritten | -0.544 |
rewriting | -0.476 |
dodged | -0.465 |
acci | -0.454 |
disreg | -0.446 |
Token | 99 |
Token ID | 7388 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
rewritten | +1.439 |
rewriting | +1.244 |
dodged | +1.219 |
acci | +1.178 |
disreg | +1.167 |
Neg logit contributions | |
ッド | -1.768 |
ushima | -1.665 |
eus | -1.653 |
ptoms | -1.616 |
lyak | -1.602 |
Token | 69 |
Token ID | 3388 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
rewritten | +2.896 |
rewriting | +2.550 |
dodged | +2.493 |
acci | +2.483 |
disreg | +2.414 |
Neg logit contributions | |
ッド | -2.459 |
eus | -2.264 |
ushima | -2.258 |
ptoms | -2.218 |
lyak | -2.191 |
Token | Bra |
Token ID | 9718 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
rewritten | +1.461 |
rewriting | +1.255 |
dodged | +1.233 |
acci | +1.187 |
disreg | +1.172 |
Neg logit contributions | |
ッド | -1.874 |
ushima | -1.762 |
eus | -1.755 |
ptoms | -1.707 |
lyak | -1.702 |
Token | ille |
Token ID | 8270 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
rewritten | +3.989 |
rewriting | +3.326 |
dodged | +3.324 |
acci | +3.267 |
disreg | +3.090 |
Neg logit contributions | |
ッド | -5.465 |
ushima | -4.934 |
eus | -4.863 |
ptoms | -4.817 |
rises | -4.751 |
Token | was |
Token ID | 373 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
rewritten | +3.369 |
rewriting | +2.858 |
dodged | +2.801 |
disreg | +2.724 |
travers | +2.706 |
Neg logit contributions | |
ッド | -4.877 |
ushima | -4.572 |
eus | -4.516 |
ptoms | -4.446 |
lyak | -4.381 |
Token | only |
Token ID | 691 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
ッド | +0.956 |
eus | +0.910 |
ushima | +0.901 |
lyak | +0.876 |
ptoms | +0.875 |
Neg logit contributions | |
rewritten | -0.679 |
rewriting | -0.582 |
dodged | -0.564 |
acci | -0.555 |
disreg | -0.535 |
Token | a |
Token ID | 257 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
rewritten | +2.005 |
rewriting | +1.721 |
dodged | +1.692 |
postponed | +1.627 |
disreg | +1.616 |
Neg logit contributions | |
ッド | -2.397 |
ushima | -2.264 |
eus | -2.222 |
ptoms | -2.193 |
lyak | -2.168 |
Token | partial |
Token ID | 13027 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
ッド | +1.034 |
eus | +0.976 |
ushima | +0.969 |
lyak | +0.937 |
ptoms | +0.935 |
Neg logit contributions | |
rewritten | -0.932 |
rewriting | -0.812 |
dodged | -0.794 |
acci | -0.780 |
disreg | -0.758 |
Token | :// |
Token ID | 1378 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ッド | +0.409 |
eus | +0.382 |
ushima | +0.382 |
ptoms | +0.369 |
lyak | +0.367 |
Neg logit contributions | |
rewritten | -0.474 |
rewriting | -0.417 |
dodged | -0.408 |
acci | -0.403 |
disreg | -0.396 |
Token | www |
Token ID | 2503 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
ッド | +0.360 |
eus | +0.332 |
ushima | +0.330 |
ptoms | +0.315 |
lyak | +0.315 |
Neg logit contributions | |
rewritten | -0.563 |
rewriting | -0.508 |
dodged | -0.498 |
acci | -0.490 |
disreg | -0.482 |
Token | . |
Token ID | 13 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
rewritten | +0.778 |
rewriting | +0.678 |
dodged | +0.663 |
acci | +0.656 |
disreg | +0.644 |
Neg logit contributions | |
ッド | -0.743 |
ushima | -0.692 |
eus | -0.687 |
ptoms | -0.670 |
lyak | -0.665 |
Token | board |
Token ID | 3526 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
rewritten | +1.491 |
rewriting | +1.331 |
dodged | +1.314 |
acci | +1.291 |
disreg | +1.264 |
Neg logit contributions | |
ッド | -1.150 |
eus | -1.062 |
ushima | -1.060 |
lyak | -1.017 |
ptoms | -1.015 |
Token | game |
Token ID | 6057 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
rewritten | +2.868 |
rewriting | +2.511 |
dodged | +2.466 |
acci | +2.412 |
disreg | +2.379 |
Neg logit contributions | |
ッド | -2.669 |
eus | -2.458 |
ushima | -2.454 |
lyak | -2.369 |
ptoms | -2.369 |
Token | ge |
Token ID | 469 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
rewritten | +2.151 |
rewriting | +1.932 |
dodged | +1.887 |
acci | +1.863 |
disreg | +1.834 |
Neg logit contributions | |
ッド | -1.443 |
eus | -1.321 |
ushima | -1.318 |
ptoms | -1.253 |
lyak | -1.252 |
Token | ek |
Token ID | 988 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
rewritten | +4.983 |
acci | +4.337 |
rewriting | +4.295 |
dodged | +4.266 |
disreg | +4.060 |
Neg logit contributions | |
ッド | -4.778 |
ushima | -4.334 |
rises | -4.126 |
lyak | -4.102 |
ptoms | -4.069 |
Token | . |
Token ID | 13 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
rewritten | +0.214 |
rewriting | +0.188 |
dodged | +0.184 |
acci | +0.182 |
disreg | +0.178 |
Neg logit contributions | |
ッド | -0.198 |
ushima | -0.184 |
eus | -0.183 |
ptoms | -0.177 |
lyak | -0.176 |
Token | com |
Token ID | 785 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
rewritten | +1.898 |
rewriting | +1.694 |
dodged | +1.663 |
acci | +1.659 |
disreg | +1.629 |
Neg logit contributions | |
ッド | -1.286 |
ushima | -1.179 |
eus | -1.157 |
ptoms | -1.125 |
lyak | -1.115 |
Token | / |
Token ID | 14 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
rewritten | +2.870 |
rewriting | +2.440 |
dodged | +2.370 |
acci | +2.342 |
disreg | +2.314 |
Neg logit contributions | |
ッド | -3.412 |
ushima | -3.214 |
eus | -3.170 |
ptoms | -3.122 |
lyak | -3.102 |
Token | thread |
Token ID | 16663 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
rewritten | +2.098 |
rewriting | +1.711 |
dodged | +1.644 |
acci | +1.614 |
disreg | +1.542 |
Neg logit contributions | |
ッド | -4.263 |
eus | -4.083 |
ushima | -4.055 |
lyak | -3.956 |
ptoms | -3.952 |
Token | , |
Token ID | 11 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
rewritten | +1.730 |
rewriting | +1.511 |
dodged | +1.479 |
acci | +1.425 |
disreg | +1.413 |
Neg logit contributions | |
ッド | -1.881 |
ushima | -1.763 |
eus | -1.755 |
lyak | -1.707 |
ptoms | -1.697 |
Token | it |
Token ID | 340 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
rewritten | +1.114 |
rewriting | +0.977 |
dodged | +0.949 |
acci | +0.921 |
disreg | +0.915 |
Neg logit contributions | |
ッド | -1.210 |
eus | -1.138 |
ushima | -1.135 |
ptoms | -1.097 |
lyak | -1.097 |
Token | 's |
Token ID | 338 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
rewritten | +0.248 |
rewriting | +0.214 |
dodged | +0.208 |
acci | +0.202 |
disreg | +0.198 |
Neg logit contributions | |
ッド | -0.315 |
eus | -0.297 |
ushima | -0.296 |
ptoms | -0.287 |
lyak | -0.287 |
Token | the |
Token ID | 262 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
rewritten | +1.010 |
rewriting | +0.878 |
dodged | +0.853 |
acci | +0.827 |
disreg | +0.817 |
Neg logit contributions | |
ッド | -1.230 |
eus | -1.155 |
ushima | -1.155 |
ptoms | -1.120 |
lyak | -1.117 |
Token | moment |
Token ID | 2589 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ッド | +0.744 |
eus | +0.704 |
ushima | +0.701 |
ptoms | +0.680 |
lyak | +0.677 |
Neg logit contributions | |
rewritten | -0.584 |
rewriting | -0.501 |
dodged | -0.490 |
acci | -0.478 |
disreg | -0.465 |
Token | 30 |
Token ID | 1542 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ッド | +0.427 |
eus | +0.400 |
ushima | +0.398 |
ptoms | +0.385 |
lyak | +0.384 |
Neg logit contributions | |
rewritten | -0.428 |
rewriting | -0.377 |
dodged | -0.368 |
acci | -0.360 |
disreg | -0.353 |
Token | seconds |
Token ID | 4201 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
rewritten | +4.194 |
rewriting | +3.618 |
dodged | +3.537 |
disreg | +3.401 |
acci | +3.393 |
Neg logit contributions | |
ッド | -5.191 |
eus | -5.017 |
ushima | -4.969 |
lyak | -4.835 |
ptoms | -4.733 |
Token | in |
Token ID | 287 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
rewritten | +3.370 |
rewriting | +2.889 |
dodged | +2.822 |
disreg | +2.696 |
acci | +2.694 |
Neg logit contributions | |
ッド | -4.571 |
eus | -4.282 |
ushima | -4.267 |
lyak | -4.134 |
ptoms | -4.126 |
Token | when |
Token ID | 618 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
rewritten | +4.285 |
rewriting | +3.790 |
dodged | +3.669 |
disreg | +3.573 |
acci | +3.544 |
Neg logit contributions | |
ッド | -4.463 |
ushima | -4.199 |
eus | -4.198 |
ptoms | -4.064 |
lyak | -4.040 |
Token | we |
Token ID | 356 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
rewritten | +0.861 |
rewriting | +0.760 |
dodged | +0.739 |
acci | +0.723 |
disreg | +0.713 |
Neg logit contributions | |
ッド | -0.860 |
eus | -0.807 |
ushima | -0.805 |
ptoms | -0.777 |
lyak | -0.774 |
Token | realize |
Token ID | 6537 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
rewritten | +0.269 |
rewriting | +0.237 |
dodged | +0.234 |
acci | +0.227 |
disreg | +0.225 |
Neg logit contributions | |
ッド | -0.253 |
ushima | -0.235 |
eus | -0.234 |
ptoms | -0.225 |
lyak | -0.224 |
Token | 10 |
Token ID | 940 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
rewritten | +1.450 |
rewriting | +1.242 |
dodged | +1.213 |
acci | +1.206 |
disreg | +1.175 |
Neg logit contributions | |
ッド | -1.667 |
ushima | -1.581 |
eus | -1.554 |
ptoms | -1.535 |
lyak | -1.517 |
Token | . |
Token ID | 13 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
ッド | +0.989 |
eus | +0.925 |
ushima | +0.918 |
lyak | +0.886 |
ptoms | +0.885 |
Neg logit contributions | |
rewritten | -1.163 |
rewriting | -1.035 |
dodged | -1.010 |
acci | -0.996 |
disreg | -0.974 |
Token | 5 |
Token ID | 20 |
Feature activation | -6.19e-02 |
Pos logit contributions | |
ッド | +0.198 |
eus | +0.186 |
ushima | +0.186 |
ptoms | +0.180 |
lyak | +0.179 |
Neg logit contributions | |
rewritten | -0.170 |
rewriting | -0.147 |
dodged | -0.144 |
acci | -0.140 |
disreg | -0.137 |
Token | million |
Token ID | 1510 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
ッド | +0.206 |
ushima | +0.193 |
eus | +0.193 |
ptoms | +0.186 |
lyak | +0.185 |
Neg logit contributions | |
rewritten | -0.195 |
rewriting | -0.170 |
dodged | -0.167 |
acci | -0.162 |
disreg | -0.160 |
Token | ), |
Token ID | 828 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
rewritten | +2.320 |
rewriting | +2.012 |
dodged | +1.972 |
disreg | +1.888 |
acci | +1.884 |
Neg logit contributions | |
ッド | -2.722 |
ushima | -2.545 |
eus | -2.531 |
ptoms | -2.494 |
lyak | -2.460 |
Token | Byron |
Token ID | 36719 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
rewritten | +2.223 |
rewriting | +1.927 |
dodged | +1.871 |
disreg | +1.794 |
acci | +1.787 |
Neg logit contributions | |
ッド | -2.594 |
ushima | -2.464 |
eus | -2.428 |
ptoms | -2.398 |
lyak | -2.361 |
Token | Maxwell |
Token ID | 28276 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
rewritten | +5.612 |
rewriting | +4.605 |
dodged | +4.497 |
Pratt | +4.416 |
deliber | +4.339 |
Neg logit contributions | |
ッド | -9.724 |
ushima | -8.538 |
eus | -8.356 |
loo | -8.198 |
lyak | -8.180 |
Token | ($ |
Token ID | 7198 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
rewritten | +1.651 |
rewriting | +1.451 |
dodged | +1.417 |
acci | +1.384 |
disreg | +1.360 |
Neg logit contributions | |
ッド | -1.680 |
eus | -1.578 |
ushima | -1.572 |
ptoms | -1.522 |
lyak | -1.514 |
Token | 8 |
Token ID | 23 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
rewritten | +0.710 |
rewriting | +0.609 |
dodged | +0.594 |
acci | +0.588 |
disreg | +0.572 |
Neg logit contributions | |
ッド | -0.859 |
ushima | -0.813 |
eus | -0.806 |
ptoms | -0.791 |
lyak | -0.784 |
Token | . |
Token ID | 13 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
ッド | +2.966 |
eus | +2.773 |
ushima | +2.741 |
lyak | +2.657 |
ptoms | +2.633 |
Neg logit contributions | |
rewritten | -3.918 |
rewriting | -3.507 |
dodged | -3.422 |
acci | -3.401 |
disreg | -3.307 |
Token | 5 |
Token ID | 20 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
rewritten | +0.562 |
rewriting | +0.489 |
dodged | +0.476 |
acci | +0.465 |
disreg | +0.455 |
Neg logit contributions | |
ッド | -0.656 |
eus | -0.620 |
ushima | -0.616 |
ptoms | -0.597 |
lyak | -0.596 |
Token | , |
Token ID | 11 |
Feature activation | -8.26e-01 |
Pos logit contributions | |
rewritten | +0.300 |
rewriting | +0.262 |
dodged | +0.255 |
acci | +0.249 |
disreg | +0.244 |
Neg logit contributions | |
ッド | -0.331 |
eus | -0.312 |
ushima | -0.311 |
ptoms | -0.300 |
lyak | -0.300 |
Token | runs |
Token ID | 4539 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
ッド | +2.996 |
eus | +2.872 |
ushima | +2.831 |
ptoms | +2.786 |
lyak | +2.755 |
Neg logit contributions | |
rewritten | -2.225 |
rewriting | -1.918 |
dodged | -1.829 |
acci | -1.822 |
disreg | -1.729 |
Token | in |
Token ID | 287 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
rewritten | +1.407 |
rewriting | +1.212 |
dodged | +1.175 |
acci | +1.135 |
disreg | +1.120 |
Neg logit contributions | |
ッド | -1.908 |
ushima | -1.800 |
eus | -1.799 |
lyak | -1.737 |
ptoms | -1.730 |
Token | 33 |
Token ID | 4747 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
rewritten | +1.903 |
rewriting | +1.654 |
dodged | +1.595 |
acci | +1.546 |
disreg | +1.536 |
Neg logit contributions | |
ッド | -2.351 |
eus | -2.224 |
ushima | -2.209 |
lyak | -2.149 |
ptoms | -2.129 |
Token | cities |
Token ID | 4736 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
rewritten | +2.257 |
rewriting | +1.971 |
dodged | +1.924 |
acci | +1.884 |
disreg | +1.854 |
Neg logit contributions | |
ッド | -2.503 |
eus | -2.375 |
ushima | -2.335 |
lyak | -2.289 |
ptoms | -2.259 |
Token | around |
Token ID | 1088 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
rewritten | +1.313 |
rewriting | +1.159 |
dodged | +1.133 |
acci | +1.103 |
disreg | +1.087 |
Neg logit contributions | |
ッド | -1.274 |
eus | -1.193 |
ushima | -1.185 |
ptoms | -1.149 |
lyak | -1.144 |
Token | the |
Token ID | 262 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
rewritten | +0.959 |
rewriting | +0.846 |
dodged | +0.824 |
acci | +0.801 |
disreg | +0.796 |
Neg logit contributions | |
ッド | -0.930 |
eus | -0.866 |
ushima | -0.864 |
ptoms | -0.836 |
lyak | -0.829 |
Token | world |
Token ID | 995 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
rewritten | +1.757 |
rewriting | +1.555 |
dodged | +1.509 |
acci | +1.471 |
disreg | +1.461 |
Neg logit contributions | |
ッド | -1.588 |
eus | -1.486 |
ushima | -1.476 |
ptoms | -1.424 |
lyak | -1.419 |
Token | . |
Token ID | 13 |
Feature activation | -9.29e-02 |
Pos logit contributions | |
ッド | +0.160 |
eus | +0.150 |
ushima | +0.149 |
ptoms | +0.144 |
lyak | +0.143 |
Neg logit contributions | |
rewritten | -0.174 |
rewriting | -0.154 |
dodged | -0.150 |
acci | -0.147 |
disreg | -0.144 |
Token | ( |
Token ID | 357 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
ッド | +0.289 |
eus | +0.271 |
ushima | +0.269 |
ptoms | +0.260 |
lyak | +0.259 |
Neg logit contributions | |
rewritten | -0.310 |
rewriting | -0.274 |
dodged | -0.268 |
acci | -0.262 |
disreg | -0.258 |
Token | Wa |
Token ID | 33484 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
ッド | +0.413 |
eus | +0.383 |
ushima | +0.381 |
ptoms | +0.365 |
lyak | +0.364 |
Neg logit contributions | |
rewritten | -0.569 |
rewriting | -0.510 |
dodged | -0.499 |
acci | -0.491 |
disreg | -0.482 |
Token | , |
Token ID | 11 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ッド | +1.496 |
eus | +1.416 |
ushima | +1.412 |
lyak | +1.353 |
ptoms | +1.348 |
Neg logit contributions | |
rewritten | -1.332 |
rewriting | -1.157 |
dodged | -1.124 |
acci | -1.106 |
disreg | -1.071 |
Token | Chicago |
Token ID | 4842 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
rewritten | +0.469 |
rewriting | +0.408 |
dodged | +0.397 |
acci | +0.388 |
disreg | +0.380 |
Neg logit contributions | |
ッド | -0.546 |
eus | -0.515 |
ushima | -0.513 |
ptoms | -0.497 |
lyak | -0.496 |
Token | , |
Token ID | 11 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
rewritten | +1.374 |
rewriting | +1.205 |
dodged | +1.175 |
acci | +1.151 |
disreg | +1.126 |
Neg logit contributions | |
ッド | -1.429 |
eus | -1.346 |
ushima | -1.338 |
ptoms | -1.290 |
lyak | -1.288 |
Token | San |
Token ID | 2986 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
rewritten | +1.919 |
rewriting | +1.662 |
dodged | +1.610 |
acci | +1.559 |
disreg | +1.537 |
Neg logit contributions | |
ッド | -2.428 |
ushima | -2.297 |
eus | -2.285 |
ptoms | -2.222 |
lyak | -2.219 |
Token | Diego |
Token ID | 9500 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
ッド | +1.996 |
eus | +1.870 |
ushima | +1.854 |
lyak | +1.780 |
ptoms | +1.776 |
Neg logit contributions | |
rewritten | -2.455 |
rewriting | -2.188 |
dodged | -2.130 |
acci | -2.083 |
disreg | -2.065 |
Token | , |
Token ID | 11 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ッド | +2.414 |
eus | +2.286 |
ushima | +2.276 |
lyak | +2.144 |
ptoms | +2.121 |
Neg logit contributions | |
rewritten | -2.628 |
rewriting | -2.323 |
dodged | -2.268 |
acci | -2.214 |
disreg | -2.166 |
Token | Dallas |
Token ID | 8533 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
rewritten | +0.404 |
rewriting | +0.350 |
dodged | +0.341 |
acci | +0.333 |
disreg | +0.325 |
Neg logit contributions | |
ッド | -0.488 |
eus | -0.461 |
ushima | -0.459 |
ptoms | -0.445 |
lyak | -0.444 |
Token | , |
Token ID | 11 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
ッド | +0.737 |
eus | +0.693 |
ushima | +0.689 |
lyak | +0.661 |
ptoms | +0.660 |
Neg logit contributions | |
rewritten | -0.745 |
rewriting | -0.655 |
dodged | -0.641 |
acci | -0.629 |
disreg | -0.616 |
Token | Palm |
Token ID | 18358 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
rewritten | +0.093 |
rewriting | +0.081 |
dodged | +0.078 |
acci | +0.077 |
disreg | +0.075 |
Neg logit contributions | |
ッド | -0.114 |
eus | -0.107 |
ushima | -0.107 |
ptoms | -0.104 |
lyak | -0.103 |
Token | Beach |
Token ID | 8511 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ッド | +0.277 |
eus | +0.259 |
ushima | +0.257 |
ptoms | +0.248 |
lyak | +0.247 |
Neg logit contributions | |
rewritten | -0.327 |
rewriting | -0.291 |
dodged | -0.285 |
acci | -0.279 |
disreg | -0.274 |
Token | , |
Token ID | 11 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
rewritten | +0.531 |
rewriting | +0.465 |
dodged | +0.453 |
acci | +0.444 |
disreg | +0.434 |
Neg logit contributions | |
ッド | -0.566 |
eus | -0.533 |
ushima | -0.531 |
ptoms | -0.512 |
lyak | -0.512 |
Token | to |
Token ID | 284 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
rewritten | +1.544 |
rewriting | +1.357 |
dodged | +1.305 |
acci | +1.284 |
travers | +1.284 |
Neg logit contributions | |
ッド | -1.769 |
ushima | -1.673 |
eus | -1.669 |
ptoms | -1.608 |
lyak | -1.594 |
Token | place |
Token ID | 1295 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
rewritten | +2.381 |
rewriting | +2.118 |
dodged | +2.048 |
travers | +2.039 |
disreg | +2.015 |
Neg logit contributions | |
ッド | -2.220 |
eus | -2.031 |
ushima | -2.024 |
ptoms | -1.965 |
lyak | -1.910 |
Token | buildings |
Token ID | 6832 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
rewritten | +0.037 |
rewriting | +0.033 |
dodged | +0.032 |
acci | +0.031 |
travers | +0.031 |
Neg logit contributions | |
ッド | -0.034 |
ushima | -0.031 |
eus | -0.031 |
ptoms | -0.030 |
lyak | -0.030 |
Token | in |
Token ID | 287 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
ッド | +0.037 |
ushima | +0.035 |
eus | +0.035 |
ptoms | +0.034 |
lyak | +0.034 |
Neg logit contributions | |
rewritten | -0.034 |
rewriting | -0.030 |
dodged | -0.029 |
acci | -0.028 |
disreg | -0.028 |
Token | this |
Token ID | 428 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
rewritten | +0.529 |
rewriting | +0.471 |
dodged | +0.457 |
acci | +0.446 |
travers | +0.445 |
Neg logit contributions | |
ッド | -0.506 |
ushima | -0.471 |
eus | -0.470 |
ptoms | -0.452 |
lyak | -0.448 |
Token | area |
Token ID | 1989 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
ッド | +0.265 |
ushima | +0.246 |
eus | +0.246 |
ptoms | +0.237 |
lyak | +0.236 |
Neg logit contributions | |
rewritten | -0.314 |
rewriting | -0.280 |
dodged | -0.273 |
acci | -0.267 |
disreg | -0.264 |
Token | . |
Token ID | 13 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
ッド | +0.246 |
eus | +0.229 |
ushima | +0.228 |
ptoms | +0.220 |
lyak | +0.220 |
Neg logit contributions | |
rewritten | -0.270 |
rewriting | -0.240 |
dodged | -0.234 |
acci | -0.229 |
disreg | -0.226 |
Token | The |
Token ID | 383 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
ッド | +1.111 |
eus | +1.037 |
ushima | +1.035 |
ptoms | +0.998 |
lyak | +0.995 |
Neg logit contributions | |
rewritten | -1.197 |
rewriting | -1.058 |
dodged | -1.032 |
acci | -1.012 |
disreg | -0.996 |
Token | spaces |
Token ID | 9029 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
rewritten | +0.096 |
rewriting | +0.085 |
dodged | +0.082 |
acci | +0.080 |
disreg | +0.079 |
Neg logit contributions | |
ッド | -0.101 |
ushima | -0.095 |
eus | -0.095 |
lyak | -0.092 |
ptoms | -0.092 |
Token | buildings |
Token ID | 6832 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
ッド | +1.517 |
eus | +1.427 |
ushima | +1.423 |
ptoms | +1.380 |
lyak | +1.375 |
Neg logit contributions | |
rewritten | -1.264 |
rewriting | -1.097 |
dodged | -1.068 |
acci | -1.040 |
disreg | -1.021 |
Token | are |
Token ID | 389 |
Feature activation | +8.99e-02 |
Pos logit contributions | |
rewritten | +0.478 |
rewriting | +0.419 |
dodged | +0.412 |
travers | +0.399 |
acci | +0.398 |
Neg logit contributions | |
ッド | -0.523 |
ushima | -0.485 |
eus | -0.482 |
ptoms | -0.468 |
lyak | -0.467 |
Token | , |
Token ID | 11 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
rewritten | +2.377 |
rewriting | +2.109 |
dodged | +2.031 |
acci | +1.991 |
disreg | +1.975 |
Neg logit contributions | |
ッド | -2.492 |
eus | -2.350 |
ushima | -2.331 |
ptoms | -2.274 |
lyak | -2.250 |
Token | eh |
Token ID | 32622 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
ッド | +0.104 |
eus | +0.098 |
ushima | +0.097 |
ptoms | +0.094 |
lyak | +0.094 |
Neg logit contributions | |
rewritten | -0.096 |
rewriting | -0.084 |
dodged | -0.082 |
acci | -0.080 |
disreg | -0.078 |
Token | ? |
Token ID | 30 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
rewritten | +2.386 |
rewriting | +2.094 |
dodged | +2.032 |
acci | +1.988 |
disreg | +1.963 |
Neg logit contributions | |
ッド | -2.630 |
eus | -2.492 |
ushima | -2.471 |
ptoms | -2.416 |
lyak | -2.384 |
Token | Can |
Token ID | 1680 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
rewritten | +0.493 |
rewriting | +0.438 |
dodged | +0.427 |
acci | +0.419 |
disreg | +0.415 |
Neg logit contributions | |
ッド | -0.428 |
eus | -0.399 |
ushima | -0.398 |
ptoms | -0.386 |
lyak | -0.381 |
Token | you |
Token ID | 345 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
ッド | +0.139 |
eus | +0.130 |
ushima | +0.129 |
ptoms | +0.125 |
lyak | +0.124 |
Neg logit contributions | |
rewritten | -0.161 |
rewriting | -0.143 |
dodged | -0.140 |
acci | -0.137 |
disreg | -0.135 |
Token | prove |
Token ID | 5879 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
rewritten | +0.082 |
rewriting | +0.073 |
dodged | +0.071 |
acci | +0.070 |
disreg | +0.069 |
Neg logit contributions | |
ッド | -0.069 |
eus | -0.064 |
ushima | -0.064 |
ptoms | -0.062 |
lyak | -0.061 |
Token | it |
Token ID | 340 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
rewritten | +0.968 |
rewriting | +0.861 |
dodged | +0.839 |
acci | +0.816 |
disreg | +0.810 |
Neg logit contributions | |
ッド | -0.868 |
eus | -0.810 |
ushima | -0.806 |
ptoms | -0.784 |
lyak | -0.777 |
Token | ? |
Token ID | 30 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
ッド | +1.832 |
eus | +1.736 |
ushima | +1.727 |
lyak | +1.662 |
ptoms | +1.651 |
Neg logit contributions | |
rewritten | -1.830 |
rewriting | -1.607 |
dodged | -1.567 |
acci | -1.538 |
disreg | -1.495 |
Token | Can |
Token ID | 1680 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
rewritten | +0.566 |
rewriting | +0.501 |
dodged | +0.488 |
acci | +0.478 |
disreg | +0.473 |
Neg logit contributions | |
ッド | -0.524 |
eus | -0.489 |
ushima | -0.487 |
ptoms | -0.473 |
lyak | -0.468 |
Token | you |
Token ID | 345 |
Feature activation | -6.35e-03 |
Pos logit contributions | |
rewritten | +0.132 |
rewriting | +0.117 |
dodged | +0.114 |
acci | +0.112 |
disreg | +0.110 |
Neg logit contributions | |
ッド | -0.115 |
eus | -0.107 |
ushima | -0.107 |
ptoms | -0.103 |
lyak | -0.103 |
Token | stop |
Token ID | 2245 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
ッド | +0.018 |
eus | +0.016 |
ushima | +0.016 |
ptoms | +0.016 |
lyak | +0.015 |
Neg logit contributions | |
rewritten | -0.024 |
rewriting | -0.021 |
dodged | -0.021 |
acci | -0.020 |
disreg | -0.020 |
Token | Nest |
Token ID | 21420 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
rewritten | +0.661 |
rewriting | +0.576 |
dodged | +0.561 |
acci | +0.546 |
disreg | +0.537 |
Neg logit contributions | |
ッド | -0.759 |
eus | -0.714 |
ushima | -0.712 |
ptoms | -0.690 |
lyak | -0.687 |
Token | or |
Token ID | 273 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
rewritten | +0.719 |
rewriting | +0.642 |
dodged | +0.632 |
acci | +0.621 |
disreg | +0.610 |
Neg logit contributions | |
ッド | -0.543 |
ushima | -0.497 |
eus | -0.493 |
ptoms | -0.480 |
lyak | -0.471 |
Token | Kir |
Token ID | 7385 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
rewritten | +2.175 |
rewriting | +1.897 |
dodged | +1.883 |
acci | +1.839 |
disreg | +1.797 |
Neg logit contributions | |
ッド | -2.391 |
ushima | -2.240 |
eus | -2.230 |
ptoms | -2.181 |
lyak | -2.153 |
Token | ch |
Token ID | 354 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
ッド | +3.773 |
eus | +3.656 |
ushima | +3.618 |
lyak | +3.530 |
ptoms | +3.529 |
Neg logit contributions | |
rewritten | -2.460 |
rewriting | -2.096 |
dodged | -2.025 |
acci | -1.932 |
disreg | -1.911 |
Token | ner |
Token ID | 1008 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
rewritten | +2.902 |
rewriting | +2.611 |
dodged | +2.567 |
acci | +2.497 |
travers | +2.472 |
Neg logit contributions | |
ッド | -2.313 |
ushima | -2.076 |
eus | -2.068 |
ptoms | -2.028 |
rises | -1.958 |
Token | assumed |
Token ID | 9672 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
rewritten | +0.958 |
rewriting | +0.821 |
dodged | +0.804 |
acci | +0.777 |
disreg | +0.765 |
Neg logit contributions | |
ッド | -1.298 |
ushima | -1.218 |
eus | -1.217 |
ptoms | -1.186 |
lyak | -1.178 |
Token | office |
Token ID | 2607 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
ッド | +0.101 |
eus | +0.094 |
ushima | +0.094 |
ptoms | +0.091 |
lyak | +0.090 |
Neg logit contributions | |
rewritten | -0.122 |
rewriting | -0.109 |
dodged | -0.106 |
acci | -0.104 |
disreg | -0.103 |
Token | in |
Token ID | 287 |
Feature activation | -2.44e-03 |
Pos logit contributions | |
rewritten | +0.770 |
rewriting | +0.668 |
dodged | +0.646 |
acci | +0.627 |
disreg | +0.620 |
Neg logit contributions | |
ッド | -0.981 |
ushima | -0.923 |
eus | -0.920 |
ptoms | -0.897 |
lyak | -0.890 |
Token | 2003 |
Token ID | 5816 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ッド | +0.009 |
ushima | +0.008 |
eus | +0.008 |
ptoms | +0.008 |
lyak | +0.008 |
Neg logit contributions | |
rewritten | -0.007 |
rewriting | -0.006 |
dodged | -0.006 |
acci | -0.006 |
disreg | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
ッド | +4.064 |
eus | +3.859 |
ushima | +3.746 |
lyak | +3.665 |
ptoms | +3.570 |
Neg logit contributions | |
rewritten | -4.305 |
rewriting | -3.753 |
dodged | -3.698 |
acci | -3.653 |
disreg | -3.509 |
Token | neoliberal |
Token ID | 31972 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ッド | +1.326 |
eus | +1.258 |
ushima | +1.245 |
lyak | +1.208 |
ptoms | +1.203 |
Neg logit contributions | |
rewritten | -1.134 |
rewriting | -0.982 |
dodged | -0.959 |
acci | -0.942 |
disreg | -0.912 |
Token | then |
Token ID | 788 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
ッド | +2.163 |
eus | +2.050 |
ushima | +2.040 |
ptoms | +1.982 |
lyak | +1.979 |
Neg logit contributions | |
rewritten | -1.581 |
rewriting | -1.352 |
dodged | -1.317 |
acci | -1.285 |
disreg | -1.248 |
Token | it |
Token ID | 340 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
ッド | +1.906 |
eus | +1.788 |
ushima | +1.777 |
ptoms | +1.717 |
lyak | +1.717 |
Neg logit contributions | |
rewritten | -2.001 |
rewriting | -1.759 |
dodged | -1.725 |
acci | -1.693 |
disreg | -1.651 |
Token | offers |
Token ID | 4394 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
ッド | +1.619 |
eus | +1.543 |
ushima | +1.523 |
lyak | +1.477 |
ptoms | +1.477 |
Neg logit contributions | |
rewritten | -1.424 |
rewriting | -1.239 |
dodged | -1.206 |
acci | -1.187 |
disreg | -1.147 |
Token | you |
Token ID | 345 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ッド | +2.765 |
eus | +2.611 |
ushima | +2.580 |
ptoms | +2.493 |
lyak | +2.485 |
Neg logit contributions | |
rewritten | -3.047 |
rewriting | -2.690 |
dodged | -2.633 |
acci | -2.602 |
disreg | -2.526 |
Token | the |
Token ID | 262 |
Feature activation | -9.91e-01 |
Pos logit contributions | |
ッド | +4.035 |
eus | +3.805 |
ushima | +3.762 |
ptoms | +3.646 |
lyak | +3.616 |
Neg logit contributions | |
rewritten | -4.699 |
rewriting | -4.151 |
dodged | -4.060 |
acci | -4.033 |
disreg | -3.885 |
Token | way |
Token ID | 835 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
ッド | +2.853 |
eus | +2.687 |
ushima | +2.655 |
ptoms | +2.574 |
lyak | +2.536 |
Neg logit contributions | |
rewritten | -3.445 |
rewriting | -3.054 |
dodged | -3.002 |
acci | -2.954 |
disreg | -2.874 |
Token | to |
Token ID | 284 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
ッド | +3.419 |
eus | +3.244 |
ushima | +3.199 |
ptoms | +3.116 |
lyak | +3.101 |
Neg logit contributions | |
rewritten | -3.418 |
rewriting | -2.989 |
dodged | -2.932 |
acci | -2.879 |
disreg | -2.793 |
Token | modify |
Token ID | 13096 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
ッド | +0.734 |
eus | +0.678 |
ushima | +0.675 |
ptoms | +0.648 |
lyak | +0.645 |
Neg logit contributions | |
rewritten | -0.941 |
rewriting | -0.844 |
dodged | -0.824 |
acci | -0.805 |
disreg | -0.800 |
Token | it |
Token ID | 340 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
rewritten | +2.424 |
rewriting | +2.160 |
dodged | +2.083 |
disreg | +2.033 |
travers | +2.016 |
Neg logit contributions | |
ッド | -2.645 |
ushima | -2.469 |
eus | -2.421 |
ptoms | -2.374 |
lyak | -2.344 |
Token | ," |
Token ID | 553 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
rewritten | +0.293 |
rewriting | +0.259 |
dodged | +0.251 |
acci | +0.245 |
disreg | +0.243 |
Neg logit contributions | |
ッド | -0.297 |
ushima | -0.277 |
eus | -0.275 |
ptoms | -0.267 |
lyak | -0.266 |
Token | said |
Token ID | 531 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
rewritten | +0.259 |
rewriting | +0.212 |
dodged | +0.204 |
acci | +0.197 |
disreg | +0.190 |
Neg logit contributions | |
ッド | -0.525 |
eus | -0.502 |
ushima | -0.499 |
ptoms | -0.487 |
lyak | -0.486 |
Token | , |
Token ID | 11 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
ッド | +1.061 |
eus | +0.995 |
ushima | +0.990 |
ptoms | +0.955 |
lyak | +0.953 |
Neg logit contributions | |
rewritten | -1.133 |
rewriting | -1.000 |
dodged | -0.978 |
acci | -0.958 |
disreg | -0.940 |
Token | let |
Token ID | 1309 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
ッド | +0.417 |
eus | +0.394 |
ushima | +0.392 |
ptoms | +0.380 |
lyak | +0.380 |
Neg logit contributions | |
rewritten | -0.350 |
rewriting | -0.304 |
dodged | -0.296 |
acci | -0.288 |
disreg | -0.284 |
Token | alone |
Token ID | 3436 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
rewritten | +1.373 |
rewriting | +1.225 |
dodged | +1.191 |
disreg | +1.160 |
acci | +1.147 |
Neg logit contributions | |
ッド | -1.187 |
eus | -1.135 |
ushima | -1.113 |
ptoms | -1.084 |
lyak | -1.075 |
Token | graduating |
Token ID | 30024 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
rewritten | +1.894 |
rewriting | +1.651 |
dodged | +1.598 |
disreg | +1.541 |
travers | +1.539 |
Neg logit contributions | |
ッド | -2.232 |
ushima | -2.101 |
eus | -2.095 |
ptoms | -2.036 |
lyak | -2.029 |
Token | with |
Token ID | 351 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
ッド | +2.494 |
eus | +2.360 |
ushima | +2.344 |
ptoms | +2.268 |
lyak | +2.260 |
Neg logit contributions | |
rewritten | -2.224 |
rewriting | -1.937 |
dodged | -1.891 |
acci | -1.845 |
disreg | -1.800 |
Token | a |
Token ID | 257 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
ッド | +1.852 |
eus | +1.736 |
ushima | +1.727 |
ptoms | +1.666 |
lyak | +1.663 |
Neg logit contributions | |
rewritten | -1.964 |
rewriting | -1.735 |
dodged | -1.694 |
acci | -1.658 |
disreg | -1.630 |
Token | medical |
Token ID | 3315 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ッド | +1.204 |
eus | +1.122 |
ushima | +1.118 |
ptoms | +1.075 |
lyak | +1.071 |
Neg logit contributions | |
rewritten | -1.443 |
rewriting | -1.283 |
dodged | -1.254 |
acci | -1.228 |
disreg | -1.211 |
Token | degree |
Token ID | 4922 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
ッド | +2.667 |
eus | +2.474 |
ushima | +2.437 |
lyak | +2.336 |
ptoms | +2.331 |
Neg logit contributions | |
rewritten | -4.025 |
rewriting | -3.616 |
dodged | -3.555 |
acci | -3.500 |
disreg | -3.431 |
Token | . |
Token ID | 13 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
ッド | +1.447 |
eus | +1.356 |
ushima | +1.348 |
ptoms | +1.299 |
lyak | +1.295 |
Neg logit contributions | |
rewritten | -1.647 |
rewriting | -1.460 |
dodged | -1.429 |
acci | -1.400 |
disreg | -1.373 |
Token | The |
Token ID | 383 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
rewritten | +0.332 |
rewriting | +0.293 |
dodged | +0.286 |
acci | +0.282 |
disreg | +0.278 |
Neg logit contributions | |
ッド | -0.303 |
ushima | -0.285 |
eus | -0.285 |
ptoms | -0.275 |
lyak | -0.275 |
Token | rising |
Token ID | 7396 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
rewritten | +0.352 |
rewriting | +0.309 |
dodged | +0.298 |
acci | +0.291 |
disreg | +0.288 |
Neg logit contributions | |
ッド | -0.384 |
eus | -0.361 |
ushima | -0.360 |
lyak | -0.348 |
ptoms | -0.348 |
Token | this |
Token ID | 428 |
Feature activation | -6.23e-03 |
Pos logit contributions | |
rewritten | +1.376 |
rewriting | +1.198 |
dodged | +1.166 |
acci | +1.133 |
disreg | +1.123 |
Neg logit contributions | |
ッド | -1.665 |
eus | -1.581 |
ushima | -1.558 |
lyak | -1.506 |
ptoms | -1.504 |
Token | month |
Token ID | 1227 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
ッド | +0.024 |
eus | +0.023 |
ushima | +0.023 |
ptoms | +0.022 |
lyak | +0.022 |
Neg logit contributions | |
rewritten | -0.016 |
rewriting | -0.014 |
dodged | -0.013 |
acci | -0.013 |
disreg | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +8.69e-02 |
Pos logit contributions | |
ッド | +2.564 |
eus | +2.406 |
ushima | +2.404 |
ptoms | +2.332 |
lyak | +2.324 |
Neg logit contributions | |
rewritten | -2.598 |
rewriting | -2.278 |
dodged | -2.226 |
acci | -2.190 |
disreg | -2.123 |
Token | O |
Token ID | 440 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
rewritten | +0.274 |
rewriting | +0.240 |
dodged | +0.234 |
acci | +0.228 |
disreg | +0.227 |
Neg logit contributions | |
ッド | -0.289 |
eus | -0.272 |
ushima | -0.269 |
lyak | -0.259 |
ptoms | -0.258 |
Token | la |
Token ID | 5031 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
rewritten | +3.235 |
rewriting | +2.899 |
dodged | +2.862 |
acci | +2.820 |
disreg | +2.760 |
Neg logit contributions | |
ッド | -2.455 |
eus | -2.256 |
ushima | -2.191 |
ptoms | -2.113 |
lyak | -2.081 |
Token | , |
Token ID | 11 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
rewritten | +3.338 |
dodged | +2.922 |
rewriting | +2.922 |
acci | +2.811 |
disreg | +2.796 |
Neg logit contributions | |
ッド | -3.284 |
ushima | -3.056 |
eus | -3.034 |
ptoms | -2.954 |
lyak | -2.934 |
Token | a |
Token ID | 257 |
Feature activation | -9.51e-02 |
Pos logit contributions | |
ッド | +0.930 |
eus | +0.871 |
ushima | +0.866 |
ptoms | +0.836 |
lyak | +0.833 |
Neg logit contributions | |
rewritten | -0.984 |
rewriting | -0.869 |
dodged | -0.849 |
acci | -0.831 |
disreg | -0.818 |
Token | local |
Token ID | 1957 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
ッド | +0.319 |
eus | +0.300 |
ushima | +0.298 |
ptoms | +0.288 |
lyak | +0.287 |
Neg logit contributions | |
rewritten | -0.297 |
rewriting | -0.260 |
dodged | -0.253 |
acci | -0.246 |
disreg | -0.244 |
Token | taxi |
Token ID | 17536 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
rewritten | +0.948 |
rewriting | +0.831 |
dodged | +0.808 |
acci | +0.785 |
disreg | +0.781 |
Neg logit contributions | |
ッド | -1.024 |
eus | -0.967 |
ushima | -0.967 |
ptoms | -0.931 |
lyak | -0.930 |
Token | aggreg |
Token ID | 13262 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
ッド | +2.252 |
eus | +2.121 |
ushima | +2.105 |
lyak | +2.037 |
ptoms | +2.027 |
Neg logit contributions | |
rewritten | -2.350 |
rewriting | -2.071 |
dodged | -2.014 |
acci | -1.983 |
disreg | -1.932 |
Token | ator |
Token ID | 1352 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
rewritten | +1.312 |
rewriting | +1.160 |
acci | +1.156 |
dodged | +1.151 |
disreg | +1.106 |
Neg logit contributions | |
ッド | -1.065 |
ushima | -0.955 |
eus | -0.954 |
ptoms | -0.933 |
lyak | -0.925 |
Token | - |
Token ID | 12 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
ッド | +0.311 |
eus | +0.286 |
ushima | +0.284 |
ptoms | +0.271 |
lyak | +0.271 |
Neg logit contributions | |
rewritten | -0.503 |
rewriting | -0.453 |
dodged | -0.445 |
acci | -0.437 |
disreg | -0.431 |
Token | founder |
Token ID | 15454 |
Feature activation | -7.63e-01 |
Pos logit contributions | |
ッド | +0.930 |
eus | +0.860 |
ushima | +0.854 |
ptoms | +0.817 |
lyak | +0.815 |
Neg logit contributions | |
rewritten | -1.386 |
rewriting | -1.246 |
dodged | -1.222 |
acci | -1.201 |
disreg | -1.182 |
Token | and |
Token ID | 290 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
ッド | +2.616 |
eus | +2.478 |
ushima | +2.438 |
lyak | +2.380 |
ptoms | +2.361 |
Neg logit contributions | |
rewritten | -2.263 |
rewriting | -1.955 |
dodged | -1.928 |
acci | -1.859 |
disreg | -1.818 |
Token | former |
Token ID | 1966 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
ッド | +0.923 |
eus | +0.872 |
ushima | +0.865 |
lyak | +0.837 |
ptoms | +0.833 |
Neg logit contributions | |
rewritten | -0.836 |
rewriting | -0.729 |
dodged | -0.712 |
acci | -0.699 |
disreg | -0.680 |
Token | chief |
Token ID | 4039 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
ッド | +0.800 |
eus | +0.755 |
ushima | +0.749 |
lyak | +0.725 |
ptoms | +0.724 |
Neg logit contributions | |
rewritten | -0.731 |
rewriting | -0.639 |
dodged | -0.623 |
acci | -0.610 |
disreg | -0.596 |
Token | executive |
Token ID | 4640 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
ッド | +1.418 |
eus | +1.337 |
ushima | +1.325 |
lyak | +1.281 |
ptoms | +1.279 |
Neg logit contributions | |
rewritten | -1.312 |
rewriting | -1.145 |
dodged | -1.120 |
acci | -1.104 |
disreg | -1.068 |
Token | of |
Token ID | 286 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
ッド | +0.881 |
eus | +0.833 |
ushima | +0.824 |
lyak | +0.799 |
ptoms | +0.795 |
Neg logit contributions | |
rewritten | -0.818 |
rewriting | -0.714 |
dodged | -0.698 |
acci | -0.686 |
disreg | -0.663 |
Token | Hy |
Token ID | 6707 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
ッド | +2.117 |
eus | +1.996 |
ushima | +1.977 |
lyak | +1.907 |
ptoms | +1.899 |
Neg logit contributions | |
rewritten | -2.089 |
rewriting | -1.828 |
dodged | -1.797 |
acci | -1.765 |
disreg | -1.707 |
Token | att |
Token ID | 1078 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
rewritten | +2.354 |
rewriting | +2.046 |
dodged | +2.006 |
acci | +1.974 |
disreg | +1.904 |
Neg logit contributions | |
ッド | -3.013 |
ushima | -2.792 |
eus | -2.786 |
lyak | -2.660 |
ptoms | -2.657 |
Token | Hot |
Token ID | 6964 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
rewritten | +2.844 |
rewriting | +2.473 |
dodged | +2.407 |
acci | +2.344 |
disreg | +2.337 |
Neg logit contributions | |
ッド | -3.217 |
ushima | -3.038 |
eus | -2.998 |
ptoms | -2.952 |
lyak | -2.930 |
Token | els |
Token ID | 1424 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
rewritten | +2.539 |
rewriting | +2.167 |
dodged | +2.151 |
acci | +2.067 |
disreg | +2.045 |
Neg logit contributions | |
ッド | -3.715 |
ushima | -3.494 |
eus | -3.444 |
ptoms | -3.375 |
lyak | -3.338 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
ッド | +2.629 |
eus | +2.455 |
ushima | +2.452 |
ptoms | +2.347 |
lyak | +2.346 |
Neg logit contributions | |
rewritten | -2.893 |
rewriting | -2.561 |
dodged | -2.510 |
acci | -2.465 |
disreg | -2.386 |
Token | A |
Token ID | 317 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
ッド | +0.769 |
eus | +0.722 |
ushima | +0.716 |
ptoms | +0.691 |
lyak | +0.690 |
Neg logit contributions | |
rewritten | -0.798 |
rewriting | -0.705 |
dodged | -0.688 |
acci | -0.673 |
disreg | -0.660 |
Token | concrete |
Token ID | 10017 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
rewritten | +1.337 |
rewriting | +1.159 |
dodged | +1.119 |
acci | +1.086 |
disreg | +1.074 |
Neg logit contributions | |
ッド | -1.589 |
eus | -1.501 |
ushima | -1.497 |
ptoms | -1.454 |
lyak | -1.443 |
Token | footing |
Token ID | 32687 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
ッド | +4.449 |
eus | +4.173 |
ushima | +4.058 |
lyak | +3.974 |
ptoms | +3.907 |
Neg logit contributions | |
rewritten | -4.455 |
rewriting | -3.888 |
acci | -3.854 |
dodged | -3.839 |
disreg | -3.663 |
Token | remains |
Token ID | 3793 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
ッド | +2.228 |
eus | +2.117 |
ushima | +2.102 |
ptoms | +2.037 |
lyak | +2.034 |
Neg logit contributions | |
rewritten | -1.749 |
rewriting | -1.505 |
dodged | -1.467 |
acci | -1.447 |
disreg | -1.391 |
Token | near |
Token ID | 1474 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
rewritten | +0.066 |
rewriting | +0.057 |
dodged | +0.056 |
acci | +0.054 |
disreg | +0.054 |
Neg logit contributions | |
ッド | -0.079 |
ushima | -0.074 |
eus | -0.074 |
ptoms | -0.072 |
lyak | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
ッド | +2.159 |
eus | +2.023 |
ushima | +1.998 |
lyak | +1.932 |
ptoms | +1.931 |
Neg logit contributions | |
rewritten | -2.457 |
rewriting | -2.171 |
dodged | -2.130 |
acci | -2.093 |
disreg | -2.045 |
Token | pitcher |
Token ID | 18086 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
rewritten | +0.079 |
rewriting | +0.069 |
dodged | +0.067 |
acci | +0.065 |
disreg | +0.064 |
Neg logit contributions | |
ッド | -0.087 |
eus | -0.082 |
ushima | -0.082 |
ptoms | -0.079 |
lyak | -0.079 |
Token | � |
Token ID | 447 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
ッド | +0.877 |
eus | +0.824 |
ushima | +0.820 |
ptoms | +0.792 |
lyak | +0.790 |
Neg logit contributions | |
rewritten | -0.854 |
rewriting | -0.750 |
dodged | -0.732 |
acci | -0.716 |
disreg | -0.702 |
Token | � |
Token ID | 247 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
ッド | +0.810 |
eus | +0.755 |
ushima | +0.753 |
ptoms | +0.727 |
lyak | +0.726 |
Neg logit contributions | |
rewritten | -0.872 |
rewriting | -0.771 |
dodged | -0.752 |
acci | -0.739 |
disreg | -0.724 |
Token | s |
Token ID | 82 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
ッド | +1.656 |
eus | +1.535 |
ushima | +1.525 |
ptoms | +1.459 |
lyak | +1.452 |
Neg logit contributions | |
rewritten | -2.438 |
rewriting | -2.188 |
dodged | -2.146 |
acci | -2.109 |
disreg | -2.074 |
Token | Mary |
Token ID | 5335 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
ッド | +5.591 |
eus | +5.369 |
ushima | +5.252 |
lyak | +5.109 |
ptoms | +5.009 |
Neg logit contributions | |
rewritten | -4.820 |
rewriting | -4.257 |
dodged | -4.191 |
acci | -4.159 |
travers | -3.930 |
Token | Sch |
Token ID | 3059 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
ッド | +6.061 |
eus | +5.693 |
ushima | +5.639 |
lyak | +5.486 |
ptoms | +5.394 |
Neg logit contributions | |
rewritten | -7.070 |
rewriting | -6.271 |
dodged | -6.146 |
disreg | -5.964 |
acci | -5.899 |
Token | wal |
Token ID | 16783 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
ッド | +1.246 |
eus | +1.158 |
ushima | +1.152 |
ptoms | +1.104 |
lyak | +1.101 |
Neg logit contributions | |
rewritten | -1.708 |
rewriting | -1.530 |
dodged | -1.499 |
acci | -1.471 |
disreg | -1.448 |
Token | m |
Token ID | 76 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
ッド | +0.106 |
eus | +0.097 |
ushima | +0.097 |
ptoms | +0.092 |
lyak | +0.091 |
Neg logit contributions | |
rewritten | -0.171 |
rewriting | -0.154 |
dodged | -0.151 |
acci | -0.149 |
disreg | -0.146 |
Token | / |
Token ID | 14 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
rewritten | +0.777 |
rewriting | +0.685 |
dodged | +0.669 |
acci | +0.656 |
disreg | +0.643 |
Neg logit contributions | |
ッド | -0.753 |
eus | -0.706 |
ushima | -0.702 |
ptoms | -0.678 |
lyak | -0.677 |
Token | AP |
Token ID | 2969 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
ッド | +3.821 |
eus | +3.578 |
ushima | +3.534 |
lyak | +3.319 |
ptoms | +3.313 |
Neg logit contributions | |
rewritten | -5.952 |
rewriting | -5.338 |
dodged | -5.245 |
acci | -5.153 |
disreg | -5.061 |
Token | ) |
Token ID | 8 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
ッド | +3.975 |
eus | +3.621 |
ushima | +3.605 |
lyak | +3.476 |
ptoms | +3.416 |
Neg logit contributions | |
rewritten | -8.255 |
rewriting | -7.589 |
acci | -7.441 |
dodged | -7.421 |
disreg | -7.156 |
Token |
|
Token ID | 198 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
rewritten | +1.386 |
rewriting | +1.212 |
dodged | +1.183 |
acci | +1.156 |
disreg | +1.133 |
Neg logit contributions | |
ッド | -1.495 |
eus | -1.408 |
ushima | -1.400 |
ptoms | -1.355 |
lyak | -1.352 |
Token |
|
Token ID | 198 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
ッド | +1.083 |
eus | +1.023 |
ushima | +1.014 |
ptoms | +0.980 |
lyak | +0.979 |
Neg logit contributions | |
rewritten | -1.048 |
rewriting | -0.923 |
dodged | -0.903 |
acci | -0.881 |
disreg | -0.861 |
Token | Texas |
Token ID | 21607 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
ッド | +0.863 |
eus | +0.799 |
ushima | +0.792 |
ptoms | +0.758 |
lyak | +0.756 |
Neg logit contributions | |
rewritten | -1.287 |
rewriting | -1.157 |
dodged | -1.135 |
acci | -1.115 |
disreg | -1.098 |
Token | Sen |
Token ID | 2311 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ッド | +1.140 |
eus | +1.077 |
ushima | +1.071 |
ptoms | +1.035 |
lyak | +1.034 |
Neg logit contributions | |
rewritten | -0.961 |
rewriting | -0.833 |
dodged | -0.808 |
acci | -0.798 |
disreg | -0.774 |
Token | ( |
Token ID | 357 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
rewritten | +0.693 |
rewriting | +0.605 |
dodged | +0.592 |
acci | +0.581 |
disreg | +0.571 |
Neg logit contributions | |
ッド | -0.694 |
ushima | -0.649 |
eus | -0.646 |
ptoms | -0.628 |
lyak | -0.622 |
Token | 25 |
Token ID | 1495 |
Feature activation | -8.78e-01 |
Pos logit contributions | |
ッド | +0.150 |
ushima | +0.140 |
eus | +0.139 |
ptoms | +0.135 |
lyak | +0.134 |
Neg logit contributions | |
rewritten | -0.178 |
rewriting | -0.157 |
dodged | -0.154 |
acci | -0.152 |
disreg | -0.149 |
Token | ) |
Token ID | 8 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
ッド | +2.074 |
eus | +1.914 |
ushima | +1.881 |
ptoms | +1.784 |
lyak | +1.784 |
Neg logit contributions | |
rewritten | -3.566 |
rewriting | -3.245 |
dodged | -3.187 |
acci | -3.120 |
disreg | -3.063 |
Token |
|
Token ID | 198 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
rewritten | +0.415 |
rewriting | +0.365 |
dodged | +0.356 |
acci | +0.347 |
disreg | +0.341 |
Neg logit contributions | |
ッド | -0.425 |
eus | -0.401 |
ushima | -0.399 |
lyak | -0.384 |
ptoms | -0.384 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
rewritten | +0.818 |
rewriting | +0.703 |
dodged | +0.682 |
acci | +0.671 |
disreg | +0.661 |
Neg logit contributions | |
ッド | -0.973 |
ushima | -0.912 |
eus | -0.907 |
ptoms | -0.887 |
lyak | -0.880 |
Token | 13 |
Token ID | 1485 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
ッド | +0.381 |
eus | +0.354 |
ushima | +0.351 |
ptoms | +0.337 |
lyak | +0.336 |
Neg logit contributions | |
rewritten | -0.523 |
rewriting | -0.469 |
dodged | -0.459 |
acci | -0.451 |
disreg | -0.444 |
Token | . |
Token ID | 13 |
Feature activation | +2.67e-03 |
Pos logit contributions | |
ッド | +4.425 |
eus | +4.252 |
ushima | +4.024 |
Cs | +3.924 |
lyak | +3.845 |
Neg logit contributions | |
rewritten | -7.443 |
rewriting | -6.875 |
dodged | -6.776 |
acci | -6.600 |
travers | -6.500 |
Token | Mark |
Token ID | 2940 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
rewritten | +0.008 |
rewriting | +0.007 |
dodged | +0.007 |
acci | +0.007 |
disreg | +0.006 |
Neg logit contributions | |
ッド | -0.009 |
eus | -0.009 |
ushima | -0.009 |
ptoms | -0.008 |
lyak | -0.008 |
Token | Bennett |
Token ID | 17468 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ッド | +2.879 |
eus | +2.749 |
ushima | +2.714 |
ptoms | +2.617 |
lyak | +2.605 |
Neg logit contributions | |
rewritten | -2.168 |
rewriting | -1.876 |
dodged | -1.782 |
acci | -1.761 |
disreg | -1.696 |
Token | ( |
Token ID | 357 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
ッド | +3.672 |
eus | +3.504 |
ushima | +3.449 |
lyak | +3.320 |
ptoms | +3.225 |
Neg logit contributions | |
rewritten | -5.036 |
rewriting | -4.495 |
dodged | -4.396 |
acci | -4.214 |
disreg | -4.205 |
Token | 63 |
Token ID | 5066 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
ッド | +0.904 |
eus | +0.838 |
ushima | +0.837 |
ptoms | +0.805 |
lyak | +0.803 |
Neg logit contributions | |
rewritten | -1.170 |
rewriting | -1.043 |
dodged | -1.021 |
acci | -1.005 |
disreg | -0.989 |
Token | Russia |
Token ID | 3284 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
rewritten | +2.226 |
rewriting | +1.963 |
dodged | +1.919 |
disreg | +1.857 |
acci | +1.837 |
Neg logit contributions | |
ッド | -2.274 |
ushima | -2.110 |
eus | -2.108 |
ptoms | -2.036 |
lyak | -2.014 |
Token | is |
Token ID | 318 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ッド | +1.159 |
eus | +1.102 |
ushima | +1.095 |
lyak | +1.062 |
ptoms | +1.061 |
Neg logit contributions | |
rewritten | -0.877 |
rewriting | -0.754 |
dodged | -0.728 |
acci | -0.719 |
disreg | -0.692 |
Token | still |
Token ID | 991 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
rewritten | +0.395 |
rewriting | +0.338 |
dodged | +0.327 |
acci | +0.316 |
disreg | +0.311 |
Neg logit contributions | |
ッド | -0.575 |
eus | -0.545 |
ushima | -0.543 |
ptoms | -0.528 |
lyak | -0.526 |
Token | not |
Token ID | 407 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
ッド | +0.773 |
eus | +0.734 |
ushima | +0.730 |
lyak | +0.708 |
ptoms | +0.707 |
Neg logit contributions | |
rewritten | -0.567 |
rewriting | -0.483 |
dodged | -0.471 |
acci | -0.462 |
disreg | -0.446 |
Token | providing |
Token ID | 4955 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
ッド | +1.149 |
eus | +1.085 |
ushima | +1.083 |
ptoms | +1.051 |
lyak | +1.046 |
Neg logit contributions | |
rewritten | -0.897 |
rewriting | -0.776 |
dodged | -0.753 |
acci | -0.731 |
disreg | -0.719 |
Token | Kiev |
Token ID | 22761 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ッド | +2.115 |
eus | +1.999 |
ushima | +1.976 |
ptoms | +1.911 |
lyak | +1.909 |
Neg logit contributions | |
rewritten | -2.217 |
rewriting | -1.956 |
dodged | -1.915 |
acci | -1.887 |
disreg | -1.837 |
Token | with |
Token ID | 351 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ッド | +7.352 |
eus | +7.235 |
ushima | +7.050 |
lyak | +6.864 |
rises | +6.852 |
Neg logit contributions | |
rewritten | -6.995 |
rewriting | -6.117 |
dodged | -6.055 |
acci | -6.041 |
disreg | -5.728 |
Token | any |
Token ID | 597 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
ッド | +4.021 |
eus | +3.848 |
ushima | +3.777 |
lyak | +3.676 |
ptoms | +3.656 |
Neg logit contributions | |
rewritten | -4.196 |
rewriting | -3.695 |
dodged | -3.642 |
acci | -3.624 |
disreg | -3.475 |
Token | gas |
Token ID | 3623 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ッド | +1.065 |
eus | +0.996 |
ushima | +0.995 |
ptoms | +0.959 |
lyak | +0.956 |
Neg logit contributions | |
rewritten | -1.039 |
rewriting | -0.913 |
dodged | -0.890 |
acci | -0.867 |
disreg | -0.854 |
Token | for |
Token ID | 329 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
ッド | +3.613 |
eus | +3.432 |
ushima | +3.386 |
lyak | +3.283 |
ptoms | +3.274 |
Neg logit contributions | |
rewritten | -4.058 |
rewriting | -3.599 |
dodged | -3.529 |
acci | -3.490 |
disreg | -3.381 |
Token | its |
Token ID | 663 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
ッド | +1.975 |
eus | +1.861 |
ushima | +1.852 |
ptoms | +1.791 |
lyak | +1.788 |
Neg logit contributions | |
rewritten | -1.843 |
rewriting | -1.611 |
dodged | -1.574 |
acci | -1.541 |
disreg | -1.506 |
Token | cent |
Token ID | 1247 |
Feature activation | -8.86e-02 |
Pos logit contributions | |
rewritten | +1.549 |
rewriting | +1.348 |
dodged | +1.301 |
acci | +1.271 |
disreg | +1.255 |
Neg logit contributions | |
ッド | -1.922 |
ushima | -1.806 |
eus | -1.805 |
ptoms | -1.752 |
lyak | -1.742 |
Token | . |
Token ID | 13 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
ッド | +0.291 |
eus | +0.273 |
ushima | +0.272 |
ptoms | +0.263 |
lyak | +0.263 |
Neg logit contributions | |
rewritten | -0.281 |
rewriting | -0.247 |
dodged | -0.242 |
acci | -0.235 |
disreg | -0.232 |
Token |
|
Token ID | 198 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
ッド | +0.830 |
eus | +0.779 |
ushima | +0.775 |
ptoms | +0.748 |
lyak | +0.746 |
Neg logit contributions | |
rewritten | -0.859 |
rewriting | -0.757 |
dodged | -0.739 |
acci | -0.724 |
disreg | -0.710 |
Token |
|
Token ID | 198 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
rewritten | +0.282 |
rewriting | +0.244 |
dodged | +0.237 |
acci | +0.232 |
disreg | +0.229 |
Neg logit contributions | |
ッド | -0.325 |
ushima | -0.305 |
eus | -0.304 |
ptoms | -0.296 |
lyak | -0.294 |
Token | Australian |
Token ID | 38036 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
ッド | +0.691 |
eus | +0.636 |
ushima | +0.633 |
ptoms | +0.604 |
lyak | +0.603 |
Neg logit contributions | |
rewritten | -1.106 |
rewriting | -0.997 |
dodged | -0.979 |
acci | -0.963 |
disreg | -0.948 |
Token | clinical |
Token ID | 8668 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ッド | +2.273 |
eus | +2.137 |
ushima | +2.117 |
ptoms | +2.059 |
lyak | +2.041 |
Neg logit contributions | |
rewritten | -2.170 |
rewriting | -1.903 |
dodged | -1.858 |
acci | -1.831 |
disreg | -1.778 |
Token | psychologist |
Token ID | 23540 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
ッド | +5.893 |
eus | +5.565 |
ptoms | +5.406 |
ushima | +5.390 |
rises | +5.269 |
Neg logit contributions | |
rewritten | -6.515 |
acci | -5.745 |
rewriting | -5.730 |
dodged | -5.700 |
travers | -5.334 |
Token | Dr |
Token ID | 1583 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
ッド | +0.992 |
eus | +0.937 |
ushima | +0.928 |
ptoms | +0.899 |
lyak | +0.898 |
Neg logit contributions | |
rewritten | -0.989 |
rewriting | -0.868 |
dodged | -0.848 |
acci | -0.833 |
disreg | -0.812 |
Token | Deborah |
Token ID | 36976 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
rewritten | +0.928 |
rewriting | +0.805 |
dodged | +0.785 |
acci | +0.768 |
disreg | +0.749 |
Neg logit contributions | |
ッド | -1.112 |
eus | -1.045 |
ushima | -1.043 |
ptoms | -1.011 |
lyak | -1.006 |
Token | Thomas |
Token ID | 5658 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
ッド | +2.882 |
eus | +2.747 |
ushima | +2.700 |
lyak | +2.619 |
ptoms | +2.605 |
Neg logit contributions | |
rewritten | -2.658 |
rewriting | -2.324 |
dodged | -2.253 |
acci | -2.187 |
disreg | -2.159 |
Token | said |
Token ID | 531 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
ッド | +2.989 |
eus | +2.860 |
ushima | +2.838 |
ptoms | +2.767 |
lyak | +2.764 |
Neg logit contributions | |
rewritten | -1.874 |
rewriting | -1.591 |
dodged | -1.506 |
acci | -1.472 |
disreg | -1.429 |
Token | wave |
Token ID | 6769 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
ッド | +1.329 |
eus | +1.239 |
ushima | +1.233 |
ptoms | +1.188 |
lyak | +1.184 |
Neg logit contributions | |
rewritten | -1.578 |
rewriting | -1.403 |
dodged | -1.373 |
acci | -1.349 |
disreg | -1.322 |
Token | possible |
Token ID | 1744 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
ッド | +1.109 |
eus | +1.047 |
ushima | +1.042 |
ptoms | +1.009 |
lyak | +1.007 |
Neg logit contributions | |
rewritten | -0.950 |
rewriting | -0.826 |
dodged | -0.804 |
acci | -0.786 |
disreg | -0.769 |
Token | . |
Token ID | 13 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
rewritten | +0.335 |
rewriting | +0.295 |
dodged | +0.289 |
acci | +0.281 |
disreg | +0.278 |
Neg logit contributions | |
ッド | -0.334 |
eus | -0.314 |
ushima | -0.312 |
lyak | -0.300 |
ptoms | -0.299 |
Token |
|
Token ID | 198 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
ッド | +0.159 |
eus | +0.149 |
ushima | +0.149 |
lyak | +0.143 |
ptoms | +0.143 |
Neg logit contributions | |
rewritten | -0.164 |
rewriting | -0.144 |
dodged | -0.141 |
acci | -0.138 |
disreg | -0.136 |
Token |
|
Token ID | 198 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
rewritten | +0.782 |
rewriting | +0.674 |
dodged | +0.654 |
acci | +0.643 |
disreg | +0.633 |
Neg logit contributions | |
ッド | -0.912 |
ushima | -0.856 |
eus | -0.855 |
ptoms | -0.831 |
lyak | -0.827 |
Token | Post |
Token ID | 6307 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
ッド | +1.154 |
eus | +1.066 |
ushima | +1.059 |
ptoms | +1.013 |
lyak | +1.010 |
Neg logit contributions | |
rewritten | -1.745 |
rewriting | -1.570 |
dodged | -1.540 |
acci | -1.514 |
disreg | -1.490 |
Token | - |
Token ID | 12 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ッド | +5.241 |
eus | +4.938 |
ushima | +4.896 |
ptoms | +4.741 |
lyak | +4.502 |
Neg logit contributions | |
rewritten | -8.206 |
rewriting | -7.433 |
dodged | -7.249 |
disreg | -7.181 |
acci | -7.138 |
Token | p |
Token ID | 79 |
Feature activation | -9.35e-01 |
Pos logit contributions | |
ッド | +3.571 |
eus | +3.351 |
ushima | +3.304 |
ptoms | +3.167 |
rises | +3.088 |
Neg logit contributions | |
rewritten | -7.694 |
rewriting | -7.000 |
dodged | -6.854 |
disreg | -6.739 |
acci | -6.709 |
Token | and |
Token ID | 392 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
ッド | +3.003 |
ushima | +2.875 |
eus | +2.859 |
lyak | +2.778 |
ptoms | +2.777 |
Neg logit contributions | |
rewritten | -2.963 |
rewriting | -2.615 |
dodged | -2.552 |
acci | -2.457 |
disreg | -2.430 |
Token | emic |
Token ID | 5314 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
ッド | +0.538 |
eus | +0.500 |
ushima | +0.497 |
ptoms | +0.478 |
lyak | +0.478 |
Neg logit contributions | |
rewritten | -0.660 |
rewriting | -0.587 |
dodged | -0.576 |
acci | -0.566 |
disreg | -0.554 |
Token | : |
Token ID | 25 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
ッド | +5.186 |
eus | +4.924 |
ushima | +4.859 |
ptoms | +4.788 |
lyak | +4.734 |
Neg logit contributions | |
rewritten | -5.294 |
rewriting | -4.664 |
acci | -4.564 |
dodged | -4.557 |
disreg | -4.374 |