Token | � |
Token ID | 447 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
policeman | +0.907 |
viol | +0.907 |
Frontier | +0.890 |
trump | +0.864 |
Warsaw | +0.851 |
Neg logit contributions | |
20439 | -1.224 |
Ire | -1.146 |
=# | -1.105 |
ゴン | -1.058 |
Reply | -1.054 |
Token | � |
Token ID | 247 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
20439 | +0.334 |
Ire | +0.305 |
=# | +0.287 |
ゴン | +0.271 |
Reply | +0.270 |
Neg logit contributions | |
policeman | -0.438 |
viol | -0.438 |
Frontier | -0.432 |
trump | -0.423 |
Warsaw | -0.415 |
Token | s |
Token ID | 82 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
policeman | +0.263 |
viol | +0.262 |
Frontier | +0.258 |
trump | +0.251 |
Warsaw | +0.247 |
Neg logit contributions | |
20439 | -0.286 |
Ire | -0.266 |
=# | -0.256 |
Reply | -0.243 |
ゴン | -0.242 |
Token | presidency |
Token ID | 12112 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
policeman | +0.334 |
viol | +0.332 |
Frontier | +0.326 |
trump | +0.316 |
Warsaw | +0.312 |
Neg logit contributions | |
20439 | -0.572 |
Ire | -0.537 |
=# | -0.520 |
ゴン | -0.502 |
Reply | -0.501 |
Token | . |
Token ID | 13 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
policeman | +0.847 |
viol | +0.844 |
Frontier | +0.828 |
trump | +0.805 |
Warsaw | +0.791 |
Neg logit contributions | |
20439 | -1.101 |
Ire | -1.036 |
=# | -0.998 |
Reply | -0.952 |
ゴン | -0.950 |
Token | Spons |
Token ID | 43522 |
Feature activation | +6.16e+00 |
Pos logit contributions | |
policeman | +0.310 |
viol | +0.309 |
Frontier | +0.305 |
trump | +0.293 |
Warsaw | +0.290 |
Neg logit contributions | |
20439 | -0.394 |
Ire | -0.368 |
=# | -0.356 |
Reply | -0.339 |
ゴン | -0.338 |
Token | ored |
Token ID | 1850 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
policeman | +1.942 |
viol | +1.937 |
Frontier | +1.929 |
trump | +1.854 |
Warsaw | +1.719 |
Neg logit contributions | |
20439 | -7.462 |
Ire | -7.095 |
ゴン | -6.743 |
=# | -6.722 |
Reply | -6.659 |
Token | by |
Token ID | 416 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
policeman | +0.777 |
viol | +0.773 |
Frontier | +0.770 |
trump | +0.741 |
Warsaw | +0.730 |
Neg logit contributions | |
20439 | -1.143 |
Ire | -1.076 |
=# | -1.032 |
Reply | -0.991 |
ゴン | -0.988 |
Token | Democrats |
Token ID | 4956 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
policeman | +0.705 |
viol | +0.703 |
Frontier | +0.698 |
trump | +0.663 |
Warsaw | +0.659 |
Neg logit contributions | |
20439 | -1.105 |
Ire | -1.039 |
=# | -1.002 |
Reply | -0.963 |
ゴン | -0.962 |
Token | Ab |
Token ID | 2275 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
20439 | +0.195 |
Ire | +0.182 |
=# | +0.175 |
Reply | +0.166 |
ゴン | +0.165 |
Neg logit contributions | |
policeman | -0.191 |
viol | -0.190 |
Frontier | -0.188 |
trump | -0.181 |
Warsaw | -0.180 |
Token | road |
Token ID | 6344 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
policeman | +0.278 |
Frontier | +0.273 |
viol | +0.263 |
trump | +0.251 |
Warsaw | +0.243 |
Neg logit contributions | |
20439 | -0.818 |
Ire | -0.779 |
=# | -0.760 |
ゴン | -0.738 |
Reply | -0.731 |
Token | ( |
Token ID | 7 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
Frontier | +0.971 |
policeman | +0.966 |
viol | +0.961 |
trump | +0.918 |
Warsaw | +0.913 |
Neg logit contributions | |
20439 | -1.312 |
Ire | -1.221 |
=# | -1.191 |
ゴン | -1.136 |
actionDate | -1.107 |
Token | r |
Token ID | 374 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
viol | +1.990 |
policeman | +1.979 |
Frontier | +1.942 |
trump | +1.880 |
Warsaw | +1.806 |
Neg logit contributions | |
20439 | -3.557 |
Ire | -3.345 |
=# | -3.243 |
ゴン | -3.136 |
Reply | -3.121 |
Token | Position |
Token ID | 26545 |
Feature activation | +9.56e-01 |
Pos logit contributions | |
policeman | +1.470 |
viol | +1.461 |
Frontier | +1.454 |
trump | +1.400 |
Warsaw | +1.385 |
Neg logit contributions | |
20439 | -1.380 |
Ire | -1.270 |
=# | -1.226 |
ゴン | -1.157 |
Reply | -1.148 |
Token | ) |
Token ID | 1267 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
policeman | +0.595 |
viol | +0.595 |
Frontier | +0.592 |
trump | +0.563 |
Warsaw | +0.560 |
Neg logit contributions | |
20439 | -0.889 |
Ire | -0.835 |
=# | -0.806 |
ゴン | -0.769 |
Reply | -0.765 |
Token | ; |
Token ID | 2162 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
policeman | +0.083 |
viol | +0.083 |
Frontier | +0.082 |
trump | +0.078 |
Warsaw | +0.077 |
Neg logit contributions | |
20439 | -0.113 |
Ire | -0.106 |
=# | -0.103 |
Reply | -0.098 |
ゴン | -0.097 |
Token | // |
Token ID | 3373 |
Feature activation | +5.87e+00 |
Pos logit contributions | |
policeman | +0.216 |
viol | +0.216 |
Frontier | +0.212 |
trump | +0.204 |
Warsaw | +0.201 |
Neg logit contributions | |
20439 | -0.308 |
Ire | -0.289 |
=# | -0.279 |
ゴン | -0.268 |
Reply | -0.267 |
Token | Note |
Token ID | 5740 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
Frontier | +3.690 |
viol | +3.662 |
policeman | +3.658 |
trump | +3.483 |
Warsaw | +3.469 |
Neg logit contributions | |
20439 | -5.376 |
Ire | -5.012 |
=# | -4.725 |
ゴン | -4.713 |
Reply | -4.513 |
Token | that |
Token ID | 326 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
policeman | +0.285 |
viol | +0.284 |
Frontier | +0.279 |
trump | +0.267 |
Warsaw | +0.263 |
Neg logit contributions | |
20439 | -0.450 |
Ire | -0.424 |
=# | -0.410 |
Reply | -0.393 |
ゴン | -0.391 |
Token | the |
Token ID | 262 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
viol | +1.268 |
policeman | +1.254 |
Frontier | +1.243 |
trump | +1.184 |
Warsaw | +1.162 |
Neg logit contributions | |
20439 | -2.065 |
Ire | -1.932 |
=# | -1.854 |
ゴン | -1.814 |
Reply | -1.793 |
Token | formula |
Token ID | 10451 |
Feature activation | -9.71e-02 |
Pos logit contributions | |
policeman | +1.093 |
viol | +1.091 |
Frontier | +1.073 |
trump | +1.028 |
Warsaw | +1.010 |
Neg logit contributions | |
20439 | -1.749 |
Ire | -1.639 |
=# | -1.583 |
ゴン | -1.533 |
Reply | -1.517 |
Token | below |
Token ID | 2174 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
20439 | +0.092 |
Ire | +0.087 |
=# | +0.084 |
ゴン | +0.081 |
Reply | +0.081 |
Neg logit contributions | |
viol | -0.059 |
policeman | -0.059 |
Frontier | -0.058 |
trump | -0.056 |
Warsaw | -0.055 |
Token | � |
Token ID | 250 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
policeman | +0.386 |
viol | +0.386 |
Frontier | +0.379 |
trump | +0.367 |
Warsaw | +0.360 |
Neg logit contributions | |
20439 | -0.447 |
Ire | -0.415 |
=# | -0.400 |
Reply | -0.382 |
ゴン | -0.378 |
Token | If |
Token ID | 1532 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
viol | +0.471 |
policeman | +0.468 |
Frontier | +0.462 |
trump | +0.443 |
Warsaw | +0.438 |
Neg logit contributions | |
20439 | -0.537 |
Ire | -0.489 |
=# | -0.477 |
ゴン | -0.462 |
iatus | -0.450 |
Token | a |
Token ID | 257 |
Feature activation | +9.96e-02 |
Pos logit contributions | |
viol | +0.484 |
policeman | +0.481 |
Frontier | +0.468 |
trump | +0.450 |
Warsaw | +0.438 |
Neg logit contributions | |
20439 | -0.839 |
Ire | -0.794 |
=# | -0.765 |
ゴン | -0.742 |
Reply | -0.739 |
Token | lady |
Token ID | 10846 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
policeman | +0.040 |
viol | +0.039 |
Frontier | +0.038 |
trump | +0.036 |
Warsaw | +0.035 |
Neg logit contributions | |
20439 | -0.116 |
Ire | -0.111 |
=# | -0.108 |
Reply | -0.104 |
ゴン | -0.104 |
Token | w |
Token ID | 266 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
policeman | +0.370 |
viol | +0.370 |
Frontier | +0.357 |
trump | +0.348 |
Warsaw | +0.337 |
Neg logit contributions | |
20439 | -0.557 |
Ire | -0.528 |
=# | -0.507 |
ゴン | -0.491 |
Reply | -0.488 |
Token | alt |
Token ID | 2501 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
policeman | +0.367 |
viol | +0.365 |
Frontier | +0.360 |
trump | +0.345 |
Warsaw | +0.340 |
Neg logit contributions | |
20439 | -0.543 |
Ire | -0.512 |
=# | -0.493 |
Reply | -0.473 |
ゴン | -0.473 |
Token | z |
Token ID | 89 |
Feature activation | +5.96e-01 |
Pos logit contributions | |
policeman | +2.796 |
viol | +2.758 |
Frontier | +2.726 |
Warsaw | +2.667 |
trump | +2.646 |
Neg logit contributions | |
20439 | -6.162 |
Ire | -5.909 |
=# | -5.594 |
ゴン | -5.587 |
Reply | -5.480 |
Token | with |
Token ID | 351 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
policeman | +0.392 |
viol | +0.392 |
Frontier | +0.381 |
trump | +0.370 |
Warsaw | +0.361 |
Neg logit contributions | |
20439 | -0.546 |
Ire | -0.511 |
=# | -0.492 |
ゴン | -0.474 |
Reply | -0.471 |
Token | you |
Token ID | 345 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
policeman | +0.237 |
viol | +0.237 |
Frontier | +0.231 |
trump | +0.224 |
Warsaw | +0.219 |
Neg logit contributions | |
20439 | -0.339 |
Ire | -0.318 |
=# | -0.308 |
Reply | -0.295 |
ゴン | -0.295 |
Token | , |
Token ID | 11 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
viol | +0.379 |
policeman | +0.378 |
Frontier | +0.367 |
trump | +0.357 |
Warsaw | +0.347 |
Neg logit contributions | |
20439 | -0.538 |
Ire | -0.506 |
=# | -0.487 |
ゴン | -0.470 |
Reply | -0.466 |
Token | beware |
Token ID | 40600 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
viol | +0.369 |
policeman | +0.368 |
Frontier | +0.357 |
trump | +0.349 |
Warsaw | +0.339 |
Neg logit contributions | |
20439 | -0.519 |
Ire | -0.485 |
=# | -0.468 |
ゴン | -0.456 |
Reply | -0.448 |
Token | , |
Token ID | 837 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
policeman | +0.111 |
Frontier | +0.111 |
viol | +0.110 |
trump | +0.105 |
Warsaw | +0.103 |
Neg logit contributions | |
20439 | -0.198 |
Ire | -0.188 |
=# | -0.181 |
ゴン | -0.177 |
Reply | -0.174 |
Token | Float |
Token ID | 48436 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Frontier | +0.801 |
viol | +0.777 |
policeman | +0.773 |
trump | +0.745 |
Warsaw | +0.733 |
Neg logit contributions | |
20439 | -1.051 |
Ire | -0.988 |
=# | -0.935 |
ゴン | -0.927 |
Reply | -0.901 |
Token | ) |
Token ID | 1267 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
Frontier | +0.891 |
policeman | +0.887 |
viol | +0.884 |
trump | +0.844 |
Warsaw | +0.832 |
Neg logit contributions | |
20439 | -1.602 |
Ire | -1.529 |
ゴン | -1.471 |
=# | -1.471 |
Instr | -1.418 |
Token | = |
Token ID | 796 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
Frontier | +1.682 |
policeman | +1.669 |
viol | +1.656 |
trump | +1.616 |
Warsaw | +1.602 |
Neg logit contributions | |
20439 | -1.964 |
Ire | -1.848 |
ゴン | -1.777 |
=# | -1.695 |
Instr | -1.679 |
Token | 1 |
Token ID | 352 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
policeman | +1.267 |
Frontier | +1.258 |
viol | +1.242 |
trump | +1.200 |
Warsaw | +1.179 |
Neg logit contributions | |
20439 | -2.207 |
Ire | -2.050 |
ゴン | -1.985 |
=# | -1.946 |
iatus | -1.884 |
Token | // |
Token ID | 3373 |
Feature activation | +5.84e+00 |
Pos logit contributions | |
Frontier | +0.818 |
policeman | +0.814 |
viol | +0.807 |
trump | +0.775 |
Warsaw | +0.768 |
Neg logit contributions | |
20439 | -1.190 |
Ire | -1.128 |
=# | -1.085 |
ゴン | -1.062 |
Reply | -1.036 |
Token | seconds |
Token ID | 4201 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
viol | +2.713 |
policeman | +2.711 |
Frontier | +2.708 |
trump | +2.555 |
ted | +2.437 |
Neg logit contributions | |
20439 | -6.289 |
Ire | -5.853 |
=# | -5.777 |
ゴン | -5.701 |
Reply | -5.572 |
Token |
|
Token ID | 198 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
policeman | +0.364 |
viol | +0.362 |
Frontier | +0.355 |
trump | +0.339 |
Warsaw | +0.335 |
Neg logit contributions | |
20439 | -0.578 |
Ire | -0.547 |
=# | -0.530 |
Reply | -0.508 |
ゴン | -0.502 |
Token |
|
Token ID | 198 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
policeman | +0.296 |
viol | +0.293 |
Frontier | +0.292 |
trump | +0.276 |
Warsaw | +0.272 |
Neg logit contributions | |
20439 | -0.626 |
Ire | -0.592 |
=# | -0.572 |
ゴン | -0.560 |
Reply | -0.550 |
Token | We |
Token ID | 1135 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
20439 | +0.037 |
Ire | +0.034 |
=# | +0.033 |
ゴン | +0.032 |
Reply | +0.032 |
Neg logit contributions | |
policeman | -0.030 |
viol | -0.030 |
Frontier | -0.029 |
trump | -0.028 |
Warsaw | -0.028 |
Token | can |
Token ID | 460 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
20439 | +1.066 |
Ire | +1.004 |
=# | +0.971 |
Reply | +0.931 |
ゴン | +0.923 |
Neg logit contributions | |
policeman | -0.719 |
viol | -0.715 |
Frontier | -0.703 |
trump | -0.674 |
Warsaw | -0.665 |
Token | { |
Token ID | 1391 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
Frontier | +1.418 |
policeman | +1.407 |
viol | +1.400 |
trump | +1.361 |
Warsaw | +1.354 |
Neg logit contributions | |
20439 | -1.156 |
Ire | -1.065 |
=# | -0.998 |
ゴン | -0.992 |
Reply | -0.952 |
Token | type |
Token ID | 2099 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
20439 | +2.296 |
Ire | +2.142 |
=# | +2.067 |
Reply | +1.955 |
ゴン | +1.907 |
Neg logit contributions | |
policeman | -2.380 |
viol | -2.369 |
Frontier | -2.344 |
trump | -2.255 |
Warsaw | -2.235 |
Token | Target |
Token ID | 12744 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
20439 | +0.758 |
Ire | +0.720 |
=# | +0.701 |
Reply | +0.674 |
ゴン | +0.669 |
Neg logit contributions | |
policeman | -0.356 |
viol | -0.353 |
Frontier | -0.347 |
trump | -0.328 |
Warsaw | -0.322 |
Token | = |
Token ID | 796 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
20439 | +0.048 |
Ire | +0.045 |
=# | +0.043 |
ゴン | +0.042 |
Reply | +0.040 |
Neg logit contributions | |
policeman | -0.047 |
Frontier | -0.047 |
viol | -0.047 |
trump | -0.045 |
Warsaw | -0.045 |
Token | & |
Token ID | 1222 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
20439 | +1.684 |
Ire | +1.578 |
=# | +1.540 |
Reply | +1.465 |
ゴン | +1.442 |
Neg logit contributions | |
policeman | -1.223 |
viol | -1.215 |
Frontier | -1.189 |
trump | -1.142 |
Warsaw | -1.127 |
Token | ' |
Token ID | 705 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
policeman | +1.313 |
viol | +1.303 |
Frontier | +1.284 |
trump | +1.226 |
Warsaw | +1.209 |
Neg logit contributions | |
20439 | -2.096 |
Ire | -1.978 |
=# | -1.920 |
Reply | -1.839 |
ゴン | -1.822 |
Token | a |
Token ID | 64 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
policeman | +3.506 |
Frontier | +3.459 |
trump | +3.362 |
viol | +3.357 |
tab | +3.243 |
Neg logit contributions | |
20439 | -5.297 |
Ire | -5.049 |
ゴン | -4.785 |
=# | -4.768 |
Instr | -4.646 |
Token | T |
Token ID | 309 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
policeman | +0.584 |
viol | +0.580 |
Frontier | +0.575 |
trump | +0.550 |
Warsaw | +0.541 |
Neg logit contributions | |
20439 | -0.849 |
Ire | -0.801 |
=# | -0.773 |
Reply | -0.739 |
ゴン | -0.735 |
Token | ; |
Token ID | 2162 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
policeman | +0.684 |
viol | +0.679 |
Frontier | +0.672 |
trump | +0.644 |
Warsaw | +0.633 |
Neg logit contributions | |
20439 | -0.999 |
Ire | -0.944 |
=# | -0.911 |
Reply | -0.870 |
ゴン | -0.866 |
Token | ... |
Token ID | 2644 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
policeman | +0.867 |
viol | +0.862 |
Frontier | +0.852 |
trump | +0.821 |
Warsaw | +0.813 |
Neg logit contributions | |
20439 | -0.887 |
Ire | -0.827 |
=# | -0.798 |
Reply | -0.757 |
ゴン | -0.746 |
Token | } |
Token ID | 1782 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
policeman | +0.228 |
viol | +0.224 |
Frontier | +0.221 |
trump | +0.212 |
Warsaw | +0.206 |
Neg logit contributions | |
20439 | -0.445 |
Ire | -0.420 |
=# | -0.406 |
ゴン | -0.395 |
Reply | -0.391 |
Token | _ |
Token ID | 4808 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
20439 | +0.162 |
Ire | +0.153 |
=# | +0.149 |
ゴン | +0.146 |
Reply | +0.143 |
Neg logit contributions | |
policeman | -0.074 |
viol | -0.073 |
Frontier | -0.072 |
trump | -0.068 |
Warsaw | -0.067 |
Token | full |
Token ID | 1336 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
policeman | +1.685 |
viol | +1.646 |
Frontier | +1.598 |
trump | +1.530 |
garrison | +1.494 |
Neg logit contributions | |
20439 | -4.220 |
Ire | -3.998 |
=# | -3.859 |
ゴン | -3.788 |
Reply | -3.723 |
Token | v |
Token ID | 410 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
viol | +1.520 |
policeman | +1.481 |
Frontier | +1.429 |
trump | +1.413 |
garrison | +1.313 |
Neg logit contributions | |
20439 | -3.614 |
Ire | -3.446 |
=# | -3.334 |
ゴン | -3.305 |
Reply | -3.218 |
Token | ) |
Token ID | 1267 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
viol | +1.243 |
policeman | +1.227 |
Frontier | +1.196 |
trump | +1.171 |
Warsaw | +1.128 |
Neg logit contributions | |
20439 | -2.431 |
Ire | -2.296 |
=# | -2.241 |
ゴン | -2.239 |
Reply | -2.133 |
Token | { |
Token ID | 1391 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
Frontier | +1.483 |
policeman | +1.473 |
viol | +1.471 |
trump | +1.429 |
Warsaw | +1.399 |
Neg logit contributions | |
20439 | -1.320 |
Ire | -1.214 |
ゴン | -1.165 |
=# | -1.140 |
Reply | -1.086 |
Token | // |
Token ID | 3373 |
Feature activation | +5.80e+00 |
Pos logit contributions | |
20439 | +2.263 |
Ire | +2.136 |
=# | +2.066 |
Reply | +1.943 |
iatus | +1.913 |
Neg logit contributions | |
policeman | -2.137 |
viol | -2.107 |
Frontier | -2.082 |
trump | -2.017 |
Warsaw | -2.014 |
Token | Time |
Token ID | 3862 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
Frontier | +3.199 |
policeman | +3.195 |
viol | +3.179 |
trump | +2.977 |
Warsaw | +2.960 |
Neg logit contributions | |
20439 | -5.816 |
Ire | -5.461 |
=# | -5.285 |
Reply | -5.079 |
ゴン | -5.055 |
Token | component |
Token ID | 7515 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
policeman | +0.616 |
viol | +0.599 |
Frontier | +0.588 |
trump | +0.561 |
garrison | +0.535 |
Neg logit contributions | |
20439 | -1.408 |
Ire | -1.327 |
=# | -1.307 |
ゴン | -1.255 |
Reply | -1.240 |
Token | const |
Token ID | 1500 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
policeman | +0.871 |
viol | +0.864 |
Frontier | +0.848 |
trump | +0.813 |
garrison | +0.785 |
Neg logit contributions | |
20439 | -1.496 |
Ire | -1.407 |
=# | -1.365 |
ゴン | -1.335 |
Reply | -1.298 |
Token | float |
Token ID | 12178 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
20439 | +0.764 |
Ire | +0.716 |
=# | +0.682 |
ゴン | +0.659 |
Reply | +0.658 |
Neg logit contributions | |
policeman | -0.606 |
viol | -0.603 |
Frontier | -0.591 |
trump | -0.573 |
Warsaw | -0.559 |
Token | TA |
Token ID | 21664 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
policeman | +0.543 |
viol | +0.533 |
Frontier | +0.530 |
trump | +0.495 |
Warsaw | +0.488 |
Neg logit contributions | |
20439 | -1.586 |
Ire | -1.505 |
=# | -1.457 |
ゴン | -1.430 |
Reply | -1.430 |
Token | . |
Token ID | 13 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
policeman | +0.729 |
viol | +0.728 |
Frontier | +0.717 |
trump | +0.687 |
Warsaw | +0.680 |
Neg logit contributions | |
20439 | -1.063 |
Ire | -1.007 |
=# | -0.971 |
Reply | -0.929 |
ゴン | -0.918 |
Token | 0 |
Token ID | 15 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
20439 | +0.064 |
Ire | +0.060 |
=# | +0.058 |
ゴン | +0.055 |
Reply | +0.055 |
Neg logit contributions | |
Frontier | -0.050 |
policeman | -0.050 |
viol | -0.050 |
trump | -0.048 |
Warsaw | -0.048 |
Token | f |
Token ID | 69 |
Feature activation | +6.54e-01 |
Pos logit contributions | |
viol | +0.842 |
policeman | +0.839 |
Frontier | +0.837 |
trump | +0.795 |
Warsaw | +0.791 |
Neg logit contributions | |
20439 | -1.248 |
Ire | -1.187 |
=# | -1.152 |
Reply | -1.095 |
ゴン | -1.084 |
Token | ) |
Token ID | 1267 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
policeman | +0.419 |
viol | +0.418 |
Frontier | +0.415 |
trump | +0.395 |
Warsaw | +0.393 |
Neg logit contributions | |
20439 | -0.594 |
Ire | -0.560 |
=# | -0.544 |
Reply | -0.517 |
ゴン | -0.514 |
Token | ; |
Token ID | 2162 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
policeman | +0.008 |
viol | +0.008 |
Frontier | +0.008 |
trump | +0.008 |
Warsaw | +0.008 |
Neg logit contributions | |
20439 | -0.011 |
Ire | -0.010 |
=# | -0.010 |
Reply | -0.009 |
ゴン | -0.009 |
Token | // |
Token ID | 3373 |
Feature activation | +5.64e+00 |
Pos logit contributions | |
policeman | +0.330 |
viol | +0.330 |
Frontier | +0.324 |
trump | +0.308 |
Warsaw | +0.303 |
Neg logit contributions | |
20439 | -0.573 |
Ire | -0.539 |
=# | -0.522 |
Reply | -0.503 |
ゴン | -0.501 |
Token | The |
Token ID | 383 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
viol | +3.681 |
policeman | +3.659 |
Frontier | +3.653 |
trump | +3.464 |
Warsaw | +3.414 |
Neg logit contributions | |
20439 | -5.008 |
Ire | -4.659 |
=# | -4.413 |
ゴン | -4.307 |
iatus | -4.208 |
Token | grid |
Token ID | 10706 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
policeman | +0.578 |
viol | +0.576 |
Frontier | +0.569 |
trump | +0.540 |
Warsaw | +0.532 |
Neg logit contributions | |
20439 | -0.963 |
Ire | -0.906 |
=# | -0.876 |
ゴン | -0.842 |
Reply | -0.842 |
Token | resulting |
Token ID | 7186 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
policeman | +0.484 |
viol | +0.478 |
Frontier | +0.477 |
trump | +0.450 |
Warsaw | +0.443 |
Neg logit contributions | |
20439 | -0.801 |
Ire | -0.751 |
=# | -0.721 |
ゴン | -0.699 |
Reply | -0.696 |
Token | from |
Token ID | 422 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
20439 | +0.773 |
Ire | +0.722 |
=# | +0.698 |
Reply | +0.662 |
ゴン | +0.655 |
Neg logit contributions | |
policeman | -0.699 |
viol | -0.694 |
Frontier | -0.686 |
trump | -0.661 |
Warsaw | -0.654 |
Token | this |
Token ID | 428 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
20439 | +0.170 |
Ire | +0.160 |
=# | +0.155 |
Reply | +0.149 |
ゴン | +0.148 |
Neg logit contributions | |
policeman | -0.106 |
viol | -0.105 |
Frontier | -0.104 |
trump | -0.099 |
Warsaw | -0.098 |
Token | Get |
Token ID | 3855 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
policeman | +0.402 |
viol | +0.399 |
Frontier | +0.399 |
trump | +0.382 |
Warsaw | +0.380 |
Neg logit contributions | |
20439 | -0.446 |
Ire | -0.414 |
=# | -0.403 |
Reply | -0.380 |
ゴン | -0.378 |
Token | Cell |
Token ID | 28780 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Frontier | +0.978 |
policeman | +0.975 |
viol | +0.967 |
trump | +0.913 |
Warsaw | +0.908 |
Neg logit contributions | |
20439 | -1.771 |
Ire | -1.671 |
=# | -1.627 |
ゴン | -1.561 |
Reply | -1.553 |
Token | Volume |
Token ID | 31715 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
policeman | +1.002 |
Frontier | +1.000 |
viol | +0.997 |
trump | +0.950 |
Warsaw | +0.949 |
Neg logit contributions | |
20439 | -1.256 |
Ire | -1.179 |
=# | -1.139 |
ゴン | -1.085 |
Reply | -1.076 |
Token | () |
Token ID | 3419 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
policeman | +0.941 |
Frontier | +0.935 |
viol | +0.934 |
trump | +0.887 |
Warsaw | +0.880 |
Neg logit contributions | |
20439 | -1.252 |
Ire | -1.168 |
=# | -1.134 |
ゴン | -1.082 |
Reply | -1.073 |
Token | ; |
Token ID | 2162 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
policeman | +1.201 |
viol | +1.188 |
Frontier | +1.181 |
trump | +1.125 |
Warsaw | +1.111 |
Neg logit contributions | |
20439 | -1.829 |
Ire | -1.719 |
=# | -1.668 |
Reply | -1.590 |
ゴン | -1.581 |
Token | // |
Token ID | 3373 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
20439 | +0.538 |
Ire | +0.505 |
=# | +0.489 |
Reply | +0.467 |
ゴン | +0.466 |
Neg logit contributions | |
policeman | -0.385 |
viol | -0.383 |
Frontier | -0.377 |
trump | -0.361 |
Warsaw | -0.356 |
Token | Pop |
Token ID | 8099 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
viol | +3.418 |
Frontier | +3.409 |
policeman | +3.406 |
trump | +3.226 |
Warsaw | +3.198 |
Neg logit contributions | |
20439 | -5.066 |
Ire | -4.752 |
=# | -4.485 |
ゴン | -4.444 |
iatus | -4.294 |
Token | ulate |
Token ID | 5039 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
20439 | +0.249 |
Ire | +0.237 |
=# | +0.231 |
Reply | +0.222 |
ゴン | +0.220 |
Neg logit contributions | |
policeman | -0.112 |
viol | -0.111 |
Frontier | -0.109 |
trump | -0.103 |
Warsaw | -0.101 |
Token | density |
Token ID | 12109 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
policeman | +0.191 |
viol | +0.191 |
Frontier | +0.188 |
trump | +0.179 |
Warsaw | +0.176 |
Neg logit contributions | |
20439 | -0.324 |
Ire | -0.306 |
=# | -0.296 |
Reply | -0.285 |
ゴン | -0.284 |
Token | grid |
Token ID | 10706 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
policeman | +0.517 |
viol | +0.516 |
Frontier | +0.511 |
trump | +0.487 |
Warsaw | +0.478 |
Neg logit contributions | |
20439 | -0.728 |
Ire | -0.681 |
=# | -0.654 |
ゴン | -0.628 |
Reply | -0.627 |
Token | . |
Token ID | 13 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
policeman | +0.130 |
viol | +0.130 |
Frontier | +0.128 |
trump | +0.122 |
Warsaw | +0.120 |
Neg logit contributions | |
20439 | -0.182 |
Ire | -0.170 |
=# | -0.163 |
ゴン | -0.157 |
Reply | -0.156 |
Token | . |
Token ID | 764 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
policeman | +0.592 |
viol | +0.587 |
Frontier | +0.578 |
trump | +0.552 |
Warsaw | +0.544 |
Neg logit contributions | |
20439 | -0.978 |
Ire | -0.923 |
=# | -0.897 |
Reply | -0.859 |
ゴン | -0.852 |
Token | copy |
Token ID | 30073 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
policeman | +0.850 |
viol | +0.845 |
Frontier | +0.831 |
trump | +0.810 |
Warsaw | +0.793 |
Neg logit contributions | |
20439 | -1.121 |
Ire | -1.045 |
=# | -1.015 |
ゴン | -0.967 |
Reply | -0.957 |
Token | _ |
Token ID | 62 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
policeman | +1.474 |
viol | +1.455 |
Frontier | +1.444 |
trump | +1.395 |
Warsaw | +1.373 |
Neg logit contributions | |
20439 | -2.076 |
Ire | -1.948 |
=# | -1.890 |
ゴン | -1.806 |
Reply | -1.767 |
Token | board |
Token ID | 3526 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
policeman | +0.662 |
viol | +0.655 |
Frontier | +0.644 |
trump | +0.618 |
Warsaw | +0.607 |
Neg logit contributions | |
20439 | -1.123 |
Ire | -1.059 |
=# | -1.028 |
Reply | -0.984 |
ゴン | -0.979 |
Token | (); |
Token ID | 13979 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
policeman | +0.011 |
viol | +0.010 |
Frontier | +0.010 |
trump | +0.010 |
Warsaw | +0.010 |
Neg logit contributions | |
20439 | -0.015 |
Ire | -0.014 |
=# | -0.014 |
ゴン | -0.013 |
Reply | -0.013 |
Token | // |
Token ID | 3373 |
Feature activation | +5.45e+00 |
Pos logit contributions | |
viol | +1.340 |
policeman | +1.334 |
Frontier | +1.316 |
trump | +1.285 |
Warsaw | +1.247 |
Neg logit contributions | |
20439 | -1.903 |
Ire | -1.761 |
=# | -1.696 |
ゴン | -1.668 |
Reply | -1.600 |
Token | !! |
Token ID | 37867 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
viol | +3.454 |
policeman | +3.404 |
Frontier | +3.359 |
trump | +3.245 |
Warsaw | +3.162 |
Neg logit contributions | |
20439 | -5.082 |
Ire | -4.745 |
ゴン | -4.461 |
=# | -4.449 |
actionDate | -4.280 |
Token | !!! |
Token ID | 10185 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
20439 | +0.147 |
Ire | +0.138 |
=# | +0.133 |
ゴン | +0.127 |
Reply | +0.127 |
Neg logit contributions | |
policeman | -0.105 |
viol | -0.105 |
Frontier | -0.103 |
trump | -0.100 |
Warsaw | -0.098 |
Token | self |
Token ID | 2116 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
viol | +0.211 |
policeman | +0.211 |
Frontier | +0.207 |
trump | +0.199 |
Warsaw | +0.196 |
Neg logit contributions | |
20439 | -0.302 |
Ire | -0.283 |
=# | -0.272 |
ゴン | -0.261 |
Reply | -0.260 |
Token | . |
Token ID | 764 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
20439 | +0.135 |
Ire | +0.128 |
=# | +0.124 |
Reply | +0.119 |
ゴン | +0.118 |
Neg logit contributions | |
policeman | -0.082 |
viol | -0.082 |
Frontier | -0.080 |
trump | -0.077 |
Warsaw | -0.076 |
Token | prev |
Token ID | 47050 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
policeman | +0.115 |
viol | +0.114 |
Frontier | +0.112 |
trump | +0.108 |
Warsaw | +0.106 |
Neg logit contributions | |
20439 | -0.161 |
Ire | -0.151 |
=# | -0.146 |
Reply | -0.139 |
ゴン | -0.139 |
Token | u |
Token ID | 334 |
Feature activation | +8.84e-01 |
Pos logit contributions | |
policeman | +0.534 |
viol | +0.531 |
Frontier | +0.514 |
trump | +0.496 |
Warsaw | +0.483 |
Neg logit contributions | |
20439 | -0.949 |
Ire | -0.899 |
=# | -0.870 |
ゴン | -0.837 |
Reply | -0.832 |
Token | Part |
Token ID | 7841 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
policeman | +0.486 |
Frontier | +0.477 |
viol | +0.477 |
trump | +0.447 |
Warsaw | +0.440 |
Neg logit contributions | |
20439 | -0.891 |
Ire | -0.837 |
=# | -0.810 |
ゴン | -0.791 |
Reply | -0.772 |
Token | icle |
Token ID | 1548 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
policeman | +0.965 |
viol | +0.922 |
Frontier | +0.920 |
trump | +0.890 |
garrison | +0.865 |
Neg logit contributions | |
20439 | -1.516 |
Ire | -1.423 |
ゴン | -1.362 |
=# | -1.334 |
Instr | -1.307 |
Token | ) |
Token ID | 1267 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
policeman | +0.139 |
viol | +0.137 |
Frontier | +0.136 |
trump | +0.130 |
Warsaw | +0.128 |
Neg logit contributions | |
20439 | -0.197 |
Ire | -0.185 |
=# | -0.180 |
Reply | -0.172 |
ゴン | -0.170 |
Token | { |
Token ID | 1391 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
20439 | +1.038 |
Ire | +0.954 |
=# | +0.914 |
Reply | +0.853 |
ゴン | +0.838 |
Neg logit contributions | |
policeman | -1.449 |
viol | -1.442 |
Frontier | -1.428 |
trump | -1.386 |
Warsaw | -1.375 |
Token | // |
Token ID | 3373 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
20439 | +4.471 |
Ire | +4.220 |
=# | +4.137 |
Reply | +3.921 |
Instr | +3.825 |
Neg logit contributions | |
policeman | -3.293 |
viol | -3.240 |
Frontier | -3.220 |
trump | -3.090 |
Warsaw | -3.061 |
Token | For |
Token ID | 1114 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
Frontier | +3.405 |
viol | +3.397 |
policeman | +3.374 |
trump | +3.210 |
Warsaw | +3.185 |
Neg logit contributions | |
20439 | -4.994 |
Ire | -4.675 |
=# | -4.406 |
ゴン | -4.359 |
iatus | -4.223 |
Token | each |
Token ID | 1123 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
policeman | +0.355 |
viol | +0.355 |
Frontier | +0.348 |
trump | +0.331 |
Warsaw | +0.325 |
Neg logit contributions | |
20439 | -0.627 |
Ire | -0.595 |
=# | -0.574 |
ゴン | -0.552 |
Reply | -0.550 |
Token | particle |
Token ID | 18758 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
policeman | +1.037 |
viol | +1.032 |
Frontier | +1.011 |
trump | +0.963 |
Warsaw | +0.936 |
Neg logit contributions | |
20439 | -1.834 |
Ire | -1.735 |
=# | -1.668 |
ゴン | -1.614 |
Reply | -1.612 |
Token | in |
Token ID | 287 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
viol | +1.808 |
policeman | +1.803 |
Frontier | +1.770 |
trump | +1.700 |
Warsaw | +1.651 |
Neg logit contributions | |
20439 | -2.783 |
Ire | -2.669 |
=# | -2.506 |
ゴン | -2.442 |
Reply | -2.430 |
Token | the |
Token ID | 262 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
policeman | +0.021 |
viol | +0.020 |
Frontier | +0.020 |
trump | +0.019 |
Warsaw | +0.019 |
Neg logit contributions | |
20439 | -0.034 |
Ire | -0.033 |
=# | -0.032 |
Reply | -0.030 |
ゴン | -0.030 |
Token | ide |
Token ID | 485 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
20439 | +0.981 |
Ire | +0.923 |
=# | +0.888 |
Reply | +0.847 |
ゴン | +0.847 |
Neg logit contributions | |
policeman | -0.767 |
viol | -0.763 |
Frontier | -0.755 |
trump | -0.725 |
Warsaw | -0.717 |
Token | ( |
Token ID | 7 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
20439 | +0.198 |
Ire | +0.186 |
=# | +0.180 |
Reply | +0.171 |
ゴン | +0.170 |
Neg logit contributions | |
policeman | -0.161 |
viol | -0.160 |
Frontier | -0.158 |
trump | -0.152 |
Warsaw | -0.150 |
Token | space |
Token ID | 2272 |
Feature activation | -8.49e-01 |
Pos logit contributions | |
viol | +1.003 |
policeman | +1.000 |
Frontier | +0.981 |
trump | +0.942 |
Warsaw | +0.933 |
Neg logit contributions | |
20439 | -1.378 |
Ire | -1.306 |
=# | -1.254 |
Reply | -1.197 |
ゴン | -1.192 |
Token | ) |
Token ID | 1267 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
20439 | +0.769 |
Ire | +0.725 |
=# | +0.699 |
Reply | +0.669 |
ゴン | +0.663 |
Neg logit contributions | |
policeman | -0.552 |
viol | -0.549 |
Frontier | -0.541 |
trump | -0.519 |
Warsaw | -0.513 |
Token | { |
Token ID | 1391 |
Feature activation | -5.14e+00 |
Pos logit contributions | |
20439 | +1.154 |
Ire | +1.071 |
=# | +1.034 |
Reply | +0.973 |
ゴン | +0.960 |
Neg logit contributions | |
policeman | -1.281 |
viol | -1.273 |
Frontier | -1.262 |
trump | -1.220 |
Warsaw | -1.208 |
Token | // |
Token ID | 3373 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
20439 | +4.369 |
Ire | +4.100 |
=# | +4.027 |
Reply | +3.798 |
ゴン | +3.719 |
Neg logit contributions | |
policeman | -3.432 |
Frontier | -3.369 |
viol | -3.368 |
trump | -3.216 |
Warsaw | -3.172 |
Token | 1 |
Token ID | 352 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
viol | +3.475 |
policeman | +3.418 |
Frontier | +3.413 |
Warsaw | +3.231 |
trump | +3.229 |
Neg logit contributions | |
20439 | -4.807 |
Ire | -4.496 |
=# | -4.238 |
ゴン | -4.129 |
iatus | -4.048 |
Token | . |
Token ID | 13 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
20439 | +0.320 |
Ire | +0.303 |
=# | +0.290 |
Reply | +0.278 |
ゴン | +0.277 |
Neg logit contributions | |
viol | -0.213 |
policeman | -0.212 |
Frontier | -0.209 |
trump | -0.200 |
Warsaw | -0.198 |
Token | Fit |
Token ID | 25048 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
viol | +0.604 |
policeman | +0.602 |
Frontier | +0.598 |
trump | +0.569 |
Warsaw | +0.563 |
Neg logit contributions | |
20439 | -0.879 |
Ire | -0.832 |
=# | -0.798 |
ゴン | -0.765 |
Reply | -0.763 |
Token | an |
Token ID | 281 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
policeman | +0.310 |
viol | +0.310 |
Frontier | +0.304 |
trump | +0.288 |
Warsaw | +0.286 |
Neg logit contributions | |
20439 | -0.513 |
Ire | -0.486 |
=# | -0.468 |
Reply | -0.449 |
ゴン | -0.448 |
Token | Object |
Token ID | 9515 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
20439 | +0.282 |
Ire | +0.267 |
=# | +0.255 |
Reply | +0.246 |
ゴン | +0.245 |
Neg logit contributions | |
viol | -0.188 |
policeman | -0.188 |
Frontier | -0.184 |
trump | -0.177 |
Warsaw | -0.174 |
Token | _ |
Token ID | 62 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
20439 | +0.920 |
Ire | +0.865 |
=# | +0.834 |
ゴン | +0.799 |
Reply | +0.796 |
Neg logit contributions | |
policeman | -0.684 |
viol | -0.679 |
Frontier | -0.671 |
trump | -0.645 |
Warsaw | -0.637 |
Token | full |
Token ID | 12853 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
20439 | +0.724 |
Ire | +0.678 |
=# | +0.663 |
Reply | +0.623 |
ゴン | +0.618 |
Neg logit contributions | |
viol | -0.639 |
policeman | -0.638 |
Frontier | -0.630 |
trump | -0.605 |
Warsaw | -0.600 |
Token | v |
Token ID | 410 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
policeman | +0.086 |
viol | +0.085 |
Frontier | +0.084 |
trump | +0.080 |
Warsaw | +0.078 |
Neg logit contributions | |
20439 | -0.161 |
Ire | -0.152 |
=# | -0.148 |
Reply | -0.142 |
ゴン | -0.141 |
Token | ) |
Token ID | 8 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
policeman | +0.024 |
viol | +0.024 |
Frontier | +0.023 |
trump | +0.023 |
Warsaw | +0.022 |
Neg logit contributions | |
20439 | -0.026 |
Ire | -0.024 |
=# | -0.023 |
Reply | -0.022 |
ゴン | -0.022 |
Token | { |
Token ID | 1391 |
Feature activation | -5.44e+00 |
Pos logit contributions | |
20439 | +0.604 |
Ire | +0.553 |
=# | +0.530 |
Reply | +0.489 |
ゴン | +0.482 |
Neg logit contributions | |
policeman | -0.955 |
viol | -0.948 |
Frontier | -0.940 |
trump | -0.915 |
Warsaw | -0.908 |
Token | // |
Token ID | 3373 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
20439 | +4.998 |
Ire | +4.751 |
=# | +4.646 |
Reply | +4.421 |
ゴン | +4.333 |
Neg logit contributions | |
policeman | -3.248 |
Frontier | -3.160 |
viol | -3.157 |
Warsaw | -3.017 |
trump | -2.995 |
Token | Time |
Token ID | 3862 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
viol | +3.380 |
policeman | +3.313 |
Frontier | +3.311 |
trump | +3.150 |
Warsaw | +3.105 |
Neg logit contributions | |
20439 | -4.751 |
Ire | -4.427 |
=# | -4.191 |
ゴン | -4.037 |
Reply | -4.036 |
Token | component |
Token ID | 7515 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
viol | +0.276 |
policeman | +0.275 |
Frontier | +0.273 |
trump | +0.260 |
Warsaw | +0.257 |
Neg logit contributions | |
20439 | -0.402 |
Ire | -0.378 |
=# | -0.362 |
Reply | -0.347 |
ゴン | -0.346 |
Token | const |
Token ID | 1500 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
viol | +0.197 |
policeman | +0.197 |
Frontier | +0.194 |
trump | +0.186 |
Warsaw | +0.182 |
Neg logit contributions | |
20439 | -0.270 |
Ire | -0.254 |
=# | -0.242 |
Reply | -0.234 |
ゴン | -0.232 |
Token | float |
Token ID | 12178 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
20439 | +2.229 |
Ire | +2.091 |
=# | +2.029 |
Reply | +1.930 |
ゴン | +1.911 |
Neg logit contributions | |
policeman | -1.767 |
viol | -1.754 |
Frontier | -1.730 |
trump | -1.662 |
Warsaw | -1.647 |
Token | TA |
Token ID | 21664 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
20439 | +0.325 |
Ire | +0.306 |
=# | +0.297 |
Reply | +0.285 |
ゴン | +0.282 |
Neg logit contributions | |
policeman | -0.203 |
viol | -0.202 |
Frontier | -0.200 |
trump | -0.191 |
Warsaw | -0.188 |
Token | appalled |
Token ID | 41586 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
20439 | +0.481 |
Ire | +0.451 |
=# | +0.440 |
ゴン | +0.422 |
Reply | +0.414 |
Neg logit contributions | |
policeman | -0.343 |
viol | -0.342 |
Frontier | -0.336 |
trump | -0.327 |
Warsaw | -0.317 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
20439 | +1.624 |
Ire | +1.529 |
=# | +1.482 |
Reply | +1.416 |
ゴン | +1.402 |
Neg logit contributions | |
policeman | -1.140 |
viol | -1.131 |
Frontier | -1.116 |
trump | -1.069 |
Warsaw | -1.055 |
Token | The |
Token ID | 383 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
20439 | +0.256 |
Ire | +0.239 |
=# | +0.232 |
ゴン | +0.223 |
Reply | +0.222 |
Neg logit contributions | |
Frontier | -0.181 |
policeman | -0.178 |
viol | -0.178 |
Warsaw | -0.171 |
trump | -0.168 |
Token | fact |
Token ID | 1109 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
20439 | +0.305 |
Ire | +0.284 |
=# | +0.274 |
ゴン | +0.262 |
Reply | +0.260 |
Neg logit contributions | |
policeman | -0.245 |
viol | -0.244 |
Frontier | -0.240 |
trump | -0.231 |
Warsaw | -0.228 |
Token | that |
Token ID | 326 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
policeman | +0.932 |
viol | +0.926 |
Frontier | +0.917 |
trump | +0.876 |
Warsaw | +0.860 |
Neg logit contributions | |
20439 | -1.505 |
Ire | -1.419 |
=# | -1.371 |
Reply | -1.325 |
ゴン | -1.315 |
Token | Psych |
Token ID | 31923 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
Frontier | +1.402 |
viol | +1.400 |
policeman | +1.398 |
trump | +1.331 |
Warsaw | +1.329 |
Neg logit contributions | |
20439 | -1.576 |
Ire | -1.462 |
=# | -1.394 |
ゴン | -1.353 |
Reply | -1.345 |
Token | o |
Token ID | 78 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
policeman | +3.017 |
viol | +2.931 |
Frontier | +2.816 |
trump | +2.714 |
om | +2.698 |
Neg logit contributions | |
20439 | -5.298 |
Ire | -4.892 |
ゴン | -4.778 |
DAQ | -4.571 |
=# | -4.540 |
Token | was |
Token ID | 373 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
20439 | +0.833 |
Ire | +0.784 |
=# | +0.760 |
Reply | +0.728 |
ゴン | +0.721 |
Neg logit contributions | |
policeman | -0.535 |
viol | -0.531 |
Frontier | -0.524 |
trump | -0.501 |
Warsaw | -0.493 |
Token | the |
Token ID | 262 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
viol | +0.403 |
policeman | +0.403 |
Frontier | +0.394 |
trump | +0.379 |
Warsaw | +0.369 |
Neg logit contributions | |
20439 | -0.751 |
Ire | -0.708 |
=# | -0.686 |
Reply | -0.662 |
ゴン | -0.656 |
Token | first |
Token ID | 717 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
20439 | +0.503 |
Ire | +0.474 |
=# | +0.460 |
Reply | +0.442 |
ゴン | +0.441 |
Neg logit contributions | |
policeman | -0.268 |
viol | -0.267 |
Frontier | -0.261 |
trump | -0.250 |
Warsaw | -0.245 |
Token | studio |
Token ID | 8034 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
policeman | +0.116 |
viol | +0.116 |
Frontier | +0.115 |
trump | +0.110 |
Warsaw | +0.109 |
Neg logit contributions | |
20439 | -0.139 |
Ire | -0.130 |
=# | -0.124 |
ゴン | -0.120 |
Reply | -0.119 |
Token | ( |
Token ID | 7 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
policeman | +0.342 |
viol | +0.341 |
Frontier | +0.339 |
trump | +0.321 |
Warsaw | +0.318 |
Neg logit contributions | |
20439 | -0.463 |
Ire | -0.435 |
=# | -0.417 |
Reply | -0.398 |
ゴン | -0.398 |
Token | ambient |
Token ID | 25237 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
viol | +1.599 |
policeman | +1.595 |
Frontier | +1.578 |
trump | +1.512 |
Warsaw | +1.473 |
Neg logit contributions | |
20439 | -2.275 |
Ire | -2.129 |
=# | -2.054 |
ゴン | -1.971 |
Reply | -1.965 |
Token | D |
Token ID | 35 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
20439 | +0.224 |
Ire | +0.205 |
=# | +0.196 |
Reply | +0.184 |
ゴン | +0.182 |
Neg logit contributions | |
policeman | -0.299 |
viol | -0.298 |
Frontier | -0.295 |
trump | -0.285 |
Warsaw | -0.283 |
Token | ensity |
Token ID | 6377 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
policeman | +0.406 |
viol | +0.397 |
Frontier | +0.394 |
trump | +0.379 |
Warsaw | +0.372 |
Neg logit contributions | |
20439 | -0.631 |
Ire | -0.593 |
=# | -0.567 |
ゴン | -0.554 |
Reply | -0.546 |
Token | ). |
Token ID | 6739 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
policeman | +0.037 |
viol | +0.037 |
Frontier | +0.036 |
trump | +0.035 |
Warsaw | +0.034 |
Neg logit contributions | |
20439 | -0.049 |
Ire | -0.046 |
=# | -0.044 |
Reply | -0.042 |
ゴン | -0.042 |
Token | // |
Token ID | 3373 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
policeman | +1.018 |
viol | +1.015 |
Frontier | +1.003 |
trump | +0.959 |
Warsaw | +0.948 |
Neg logit contributions | |
20439 | -1.418 |
Ire | -1.334 |
=# | -1.285 |
ゴン | -1.231 |
Reply | -1.228 |
Token | This |
Token ID | 770 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
viol | +3.127 |
Frontier | +3.080 |
policeman | +3.072 |
trump | +2.900 |
Warsaw | +2.879 |
Neg logit contributions | |
20439 | -4.771 |
Ire | -4.474 |
=# | -4.232 |
ゴン | -4.218 |
Reply | -4.076 |
Token | is |
Token ID | 318 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
policeman | +0.009 |
viol | +0.009 |
Frontier | +0.009 |
trump | +0.008 |
Warsaw | +0.008 |
Neg logit contributions | |
20439 | -0.013 |
Ire | -0.012 |
=# | -0.011 |
ゴン | -0.011 |
Reply | -0.011 |
Token | because |
Token ID | 780 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
policeman | +0.543 |
viol | +0.542 |
Frontier | +0.532 |
trump | +0.507 |
Warsaw | +0.499 |
Neg logit contributions | |
20439 | -0.931 |
Ire | -0.878 |
=# | -0.846 |
ゴン | -0.821 |
Reply | -0.816 |
Token | we |
Token ID | 356 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
viol | +0.193 |
policeman | +0.192 |
Frontier | +0.190 |
trump | +0.181 |
Warsaw | +0.177 |
Neg logit contributions | |
20439 | -0.317 |
Ire | -0.299 |
=# | -0.288 |
ゴン | -0.278 |
Reply | -0.278 |
Token | first |
Token ID | 717 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
policeman | +0.208 |
viol | +0.207 |
Frontier | +0.203 |
trump | +0.197 |
Warsaw | +0.192 |
Neg logit contributions | |
20439 | -0.342 |
Ire | -0.322 |
=# | -0.308 |
ゴン | -0.299 |
Reply | -0.298 |
Token | ) |
Token ID | 1267 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
20439 | +1.435 |
Ire | +1.347 |
=# | +1.303 |
Reply | +1.237 |
ゴン | +1.233 |
Neg logit contributions | |
policeman | -1.099 |
viol | -1.094 |
Frontier | -1.080 |
trump | -1.034 |
Warsaw | -1.024 |
Token | ) |
Token ID | 1267 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
20439 | +0.736 |
Ire | +0.687 |
=# | +0.663 |
Reply | +0.630 |
ゴン | +0.628 |
Neg logit contributions | |
policeman | -0.609 |
viol | -0.603 |
Frontier | -0.596 |
trump | -0.575 |
Warsaw | -0.564 |
Token | return |
Token ID | 1441 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
20439 | +0.902 |
Ire | +0.833 |
=# | +0.798 |
Reply | +0.751 |
ゴン | +0.744 |
Neg logit contributions | |
policeman | -1.085 |
viol | -1.080 |
Frontier | -1.067 |
trump | -1.035 |
Warsaw | -1.024 |
Token | space |
Token ID | 2272 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
20439 | +1.243 |
Ire | +1.169 |
=# | +1.128 |
Reply | +1.081 |
ゴン | +1.080 |
Neg logit contributions | |
policeman | -0.830 |
viol | -0.829 |
Frontier | -0.815 |
trump | -0.781 |
Warsaw | -0.767 |
Token | ; |
Token ID | 26 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
20439 | +0.328 |
Ire | +0.308 |
=# | +0.296 |
ゴン | +0.281 |
Reply | +0.280 |
Neg logit contributions | |
policeman | -0.245 |
viol | -0.244 |
Frontier | -0.242 |
trump | -0.231 |
Warsaw | -0.228 |
Token | // |
Token ID | 3373 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
20439 | +0.326 |
Ire | +0.306 |
=# | +0.296 |
Reply | +0.283 |
ゴン | +0.282 |
Neg logit contributions | |
policeman | -0.222 |
viol | -0.222 |
Frontier | -0.217 |
trump | -0.209 |
Warsaw | -0.205 |
Token | 4 |
Token ID | 604 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
viol | +3.010 |
policeman | +2.973 |
Frontier | +2.936 |
trump | +2.806 |
Warsaw | +2.780 |
Neg logit contributions | |
20439 | -4.763 |
Ire | -4.500 |
=# | -4.225 |
ゴン | -4.186 |
Reply | -4.057 |
Token | . |
Token ID | 13 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
20439 | +0.023 |
Ire | +0.022 |
=# | +0.021 |
ゴン | +0.020 |
Reply | +0.020 |
Neg logit contributions | |
policeman | -0.016 |
viol | -0.016 |
Frontier | -0.016 |
trump | -0.015 |
Warsaw | -0.015 |
Token | Rec |
Token ID | 3311 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
viol | +0.594 |
policeman | +0.592 |
Frontier | +0.591 |
trump | +0.560 |
Warsaw | +0.556 |
Neg logit contributions | |
20439 | -0.804 |
Ire | -0.755 |
=# | -0.722 |
ゴン | -0.699 |
Reply | -0.684 |
Token | urs |
Token ID | 1834 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
policeman | +0.461 |
viol | +0.459 |
Frontier | +0.454 |
trump | +0.439 |
Warsaw | +0.436 |
Neg logit contributions | |
20439 | -0.413 |
Ire | -0.385 |
=# | -0.367 |
ゴン | -0.350 |
Reply | -0.342 |
Token | ively |
Token ID | 2280 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
20439 | +1.362 |
Ire | +1.298 |
=# | +1.260 |
Reply | +1.213 |
ゴン | +1.211 |
Neg logit contributions | |
policeman | -0.562 |
viol | -0.555 |
Frontier | -0.546 |
trump | -0.512 |
Warsaw | -0.503 |
Token | ing |
Token ID | 278 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
policeman | +0.050 |
viol | +0.050 |
Frontier | +0.049 |
trump | +0.047 |
Warsaw | +0.047 |
Neg logit contributions | |
20439 | -0.058 |
Ire | -0.055 |
=# | -0.052 |
ゴン | -0.050 |
Reply | -0.049 |
Token | , |
Token ID | 837 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
20439 | +1.030 |
Ire | +0.972 |
=# | +0.933 |
ゴン | +0.897 |
Reply | +0.884 |
Neg logit contributions | |
policeman | -0.789 |
viol | -0.784 |
Frontier | -0.779 |
trump | -0.737 |
Warsaw | -0.736 |
Token | _ |
Token ID | 4808 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
20439 | +0.558 |
Ire | +0.526 |
=# | +0.509 |
Reply | +0.487 |
ゴン | +0.485 |
Neg logit contributions | |
policeman | -0.358 |
viol | -0.355 |
Frontier | -0.349 |
trump | -0.335 |
Warsaw | -0.330 |
Token | : |
Token ID | 1058 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
20439 | +0.325 |
Ire | +0.306 |
=# | +0.296 |
Reply | +0.282 |
ゴン | +0.281 |
Neg logit contributions | |
policeman | -0.231 |
viol | -0.230 |
Frontier | -0.226 |
trump | -0.217 |
Warsaw | -0.214 |
Token | & |
Token ID | 1222 |
Feature activation | +9.55e-01 |
Pos logit contributions | |
20439 | +2.528 |
Ire | +2.360 |
=# | +2.292 |
Reply | +2.176 |
ゴン | +2.139 |
Neg logit contributions | |
policeman | -2.240 |
viol | -2.228 |
Frontier | -2.195 |
trump | -2.115 |
Warsaw | -2.090 |
Token | ' |
Token ID | 705 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
policeman | +0.609 |
viol | +0.604 |
Frontier | +0.603 |
trump | +0.575 |
Warsaw | +0.569 |
Neg logit contributions | |
20439 | -0.879 |
Ire | -0.838 |
=# | -0.797 |
ゴン | -0.767 |
Reply | -0.764 |
Token | a |
Token ID | 64 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
policeman | +3.300 |
Frontier | +3.256 |
viol | +3.236 |
trump | +3.111 |
Warsaw | +3.080 |
Neg logit contributions | |
20439 | -4.343 |
Ire | -4.133 |
=# | -3.939 |
ゴン | -3.775 |
Reply | -3.750 |
Token | Guard |
Token ID | 4932 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
20439 | +0.166 |
Ire | +0.157 |
=# | +0.150 |
ゴン | +0.144 |
Reply | +0.143 |
Neg logit contributions | |
policeman | -0.121 |
viol | -0.120 |
Frontier | -0.120 |
trump | -0.114 |
Warsaw | -0.113 |
Token | ) |
Token ID | 1267 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
20439 | +1.669 |
Ire | +1.571 |
=# | +1.519 |
Reply | +1.449 |
ゴン | +1.439 |
Neg logit contributions | |
policeman | -1.223 |
viol | -1.213 |
Frontier | -1.198 |
trump | -1.149 |
Warsaw | -1.135 |
Token | -> |
Token ID | 4613 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
20439 | +0.563 |
Ire | +0.528 |
=# | +0.512 |
Reply | +0.485 |
ゴン | +0.478 |
Neg logit contributions | |
policeman | -0.487 |
viol | -0.485 |
Frontier | -0.479 |
trump | -0.460 |
Warsaw | -0.456 |
Token | Option |
Token ID | 16018 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
20439 | +2.383 |
Ire | +2.226 |
=# | +2.171 |
Reply | +2.062 |
ゴン | +2.018 |
Neg logit contributions | |
policeman | -1.995 |
viol | -1.991 |
Frontier | -1.954 |
trump | -1.882 |
Warsaw | -1.858 |
Token | , |
Token ID | 11 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
viol | +0.498 |
policeman | +0.497 |
Frontier | +0.489 |
trump | +0.474 |
Warsaw | +0.464 |
Neg logit contributions | |
20439 | -0.659 |
Ire | -0.618 |
=# | -0.595 |
ゴン | -0.567 |
Reply | -0.566 |
Token | this |
Token ID | 428 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
20439 | +0.037 |
Ire | +0.035 |
=# | +0.034 |
ゴン | +0.033 |
Reply | +0.032 |
Neg logit contributions | |
viol | -0.023 |
policeman | -0.022 |
Frontier | -0.022 |
trump | -0.021 |
Warsaw | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
20439 | +0.625 |
Ire | +0.586 |
=# | +0.566 |
Reply | +0.541 |
ゴン | +0.540 |
Neg logit contributions | |
policeman | -0.436 |
viol | -0.433 |
Frontier | -0.426 |
trump | -0.412 |
Warsaw | -0.404 |
Token | completely |
Token ID | 3190 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
viol | +0.422 |
policeman | +0.422 |
Frontier | +0.414 |
trump | +0.404 |
Warsaw | +0.389 |
Neg logit contributions | |
20439 | -0.710 |
Ire | -0.668 |
=# | -0.646 |
ゴン | -0.627 |
Reply | -0.618 |
Token | made |
Token ID | 925 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
viol | +0.124 |
policeman | +0.124 |
Frontier | +0.122 |
trump | +0.119 |
Warsaw | +0.116 |
Neg logit contributions | |
20439 | -0.142 |
Ire | -0.131 |
=# | -0.126 |
ゴン | -0.121 |
Reply | -0.120 |
Token | up |
Token ID | 510 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
policeman | +0.658 |
viol | +0.657 |
Frontier | +0.636 |
trump | +0.622 |
Warsaw | +0.604 |
Neg logit contributions | |
20439 | -1.121 |
Ire | -1.054 |
=# | -1.019 |
ゴン | -0.981 |
Reply | -0.974 |
Token | . |
Token ID | 13 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
viol | +2.832 |
policeman | +2.804 |
trump | +2.788 |
Frontier | +2.746 |
Warsaw | +2.582 |
Neg logit contributions | |
20439 | -4.950 |
Ire | -4.586 |
=# | -4.392 |
ゴン | -4.275 |
Reply | -4.173 |
Token | There |
Token ID | 1318 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
Frontier | +0.223 |
policeman | +0.218 |
viol | +0.217 |
trump | +0.209 |
Warsaw | +0.209 |
Neg logit contributions | |
20439 | -0.358 |
Ire | -0.336 |
=# | -0.324 |
ゴン | -0.318 |
Reply | -0.309 |
Token | was |
Token ID | 373 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
20439 | +0.011 |
Ire | +0.010 |
=# | +0.010 |
Reply | +0.009 |
ゴン | +0.009 |
Neg logit contributions | |
policeman | -0.007 |
viol | -0.007 |
Frontier | -0.007 |
trump | -0.007 |
Warsaw | -0.007 |
Token | no |
Token ID | 645 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
policeman | +0.555 |
viol | +0.555 |
Frontier | +0.543 |
trump | +0.527 |
Warsaw | +0.514 |
Neg logit contributions | |
20439 | -0.911 |
Ire | -0.860 |
=# | -0.832 |
ゴン | -0.802 |
Reply | -0.796 |
Token | actual |
Token ID | 4036 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
policeman | +0.018 |
viol | +0.018 |
Frontier | +0.018 |
trump | +0.017 |
Warsaw | +0.017 |
Neg logit contributions | |
20439 | -0.025 |
Ire | -0.023 |
=# | -0.023 |
ゴン | -0.022 |
Reply | -0.022 |
Token | is |
Token ID | 318 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
viol | +0.696 |
policeman | +0.694 |
Frontier | +0.681 |
trump | +0.657 |
Warsaw | +0.640 |
Neg logit contributions | |
20439 | -1.044 |
Ire | -0.983 |
=# | -0.939 |
Reply | -0.906 |
ゴン | -0.901 |
Token | used |
Token ID | 973 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
policeman | +0.400 |
viol | +0.398 |
Frontier | +0.389 |
trump | +0.375 |
Warsaw | +0.366 |
Neg logit contributions | |
20439 | -0.683 |
Ire | -0.649 |
=# | -0.625 |
Reply | -0.601 |
ゴン | -0.599 |
Token | to |
Token ID | 284 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
policeman | +0.033 |
viol | +0.033 |
Frontier | +0.032 |
trump | +0.031 |
Warsaw | +0.030 |
Neg logit contributions | |
20439 | -0.050 |
Ire | -0.047 |
=# | -0.045 |
Reply | -0.044 |
ゴン | -0.043 |
Token | treat |
Token ID | 2190 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
policeman | +0.412 |
viol | +0.408 |
Frontier | +0.397 |
trump | +0.387 |
Warsaw | +0.372 |
Neg logit contributions | |
20439 | -0.815 |
Ire | -0.771 |
=# | -0.737 |
ゴン | -0.717 |
Reply | -0.715 |
Token | th |
Token ID | 294 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
viol | +0.639 |
policeman | +0.638 |
Frontier | +0.619 |
trump | +0.605 |
Warsaw | +0.594 |
Neg logit contributions | |
20439 | -0.943 |
Ire | -0.884 |
=# | -0.843 |
ゴン | -0.823 |
Reply | -0.812 |
Token | al |
Token ID | 282 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
policeman | +0.459 |
viol | +0.457 |
Frontier | +0.447 |
trump | +0.426 |
Warsaw | +0.419 |
Neg logit contributions | |
20439 | -0.837 |
Ire | -0.797 |
=# | -0.770 |
ゴン | -0.745 |
Reply | -0.739 |
Token | asse |
Token ID | 21612 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
policeman | +3.608 |
viol | +3.602 |
Frontier | +3.530 |
trump | +3.453 |
om | +3.442 |
Neg logit contributions | |
20439 | -3.885 |
Ire | -3.784 |
=# | -3.621 |
ゴン | -3.483 |
Reply | -3.459 |
Token | mia |
Token ID | 20730 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
viol | +0.360 |
policeman | +0.357 |
Frontier | +0.352 |
trump | +0.335 |
Warsaw | +0.319 |
Neg logit contributions | |
20439 | -0.866 |
Ire | -0.831 |
=# | -0.820 |
Reply | -0.784 |
ゴン | -0.784 |
Token | . |
Token ID | 13 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
viol | +0.722 |
policeman | +0.712 |
Frontier | +0.697 |
trump | +0.684 |
Warsaw | +0.661 |
Neg logit contributions | |
20439 | -1.095 |
Ire | -1.032 |
=# | -1.004 |
ゴン | -0.956 |
Reply | -0.954 |
Token | The |
Token ID | 383 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
Frontier | +0.472 |
viol | +0.470 |
policeman | +0.469 |
Warsaw | +0.444 |
trump | +0.439 |
Neg logit contributions | |
20439 | -0.744 |
Ire | -0.699 |
=# | -0.683 |
ゴン | -0.653 |
Reply | -0.652 |
Token | company |
Token ID | 1664 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
policeman | +0.033 |
viol | +0.032 |
Frontier | +0.032 |
trump | +0.030 |
Warsaw | +0.030 |
Neg logit contributions | |
20439 | -0.060 |
Ire | -0.056 |
=# | -0.055 |
ゴン | -0.052 |
Reply | -0.052 |
Token | -> |
Token ID | 4613 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
20439 | +0.563 |
Ire | +0.528 |
=# | +0.512 |
Reply | +0.485 |
ゴン | +0.478 |
Neg logit contributions | |
policeman | -0.487 |
viol | -0.485 |
Frontier | -0.479 |
trump | -0.460 |
Warsaw | -0.456 |
Token | Option |
Token ID | 16018 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
20439 | +2.383 |
Ire | +2.226 |
=# | +2.171 |
Reply | +2.062 |
ゴン | +2.018 |
Neg logit contributions | |
policeman | -1.995 |
viol | -1.991 |
Frontier | -1.954 |
trump | -1.882 |
Warsaw | -1.858 |
Token | < |
Token ID | 1279 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
20439 | +1.249 |
Ire | +1.166 |
=# | +1.121 |
ゴン | +1.056 |
Reply | +1.056 |
Neg logit contributions | |
policeman | -1.276 |
viol | -1.268 |
Frontier | -1.256 |
trump | -1.211 |
Warsaw | -1.197 |
Token | Shared |
Token ID | 39403 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
20439 | +1.008 |
Ire | +0.952 |
=# | +0.919 |
Reply | +0.877 |
ゴン | +0.874 |
Neg logit contributions | |
policeman | -0.689 |
viol | -0.683 |
Frontier | -0.675 |
trump | -0.645 |
Warsaw | -0.638 |
Token | < |
Token ID | 1279 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
policeman | +0.338 |
viol | +0.335 |
Frontier | +0.333 |
trump | +0.320 |
Warsaw | +0.318 |
Neg logit contributions | |
20439 | -0.357 |
Ire | -0.336 |
=# | -0.324 |
ゴン | -0.305 |
Reply | -0.303 |
Token | ' |
Token ID | 705 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
20439 | +0.573 |
Ire | +0.542 |
=# | +0.522 |
ゴン | +0.501 |
Reply | +0.500 |
Neg logit contributions | |
policeman | -0.352 |
viol | -0.346 |
Frontier | -0.345 |
trump | -0.328 |
Warsaw | -0.323 |
Token | a |
Token ID | 64 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
policeman | +3.024 |
Frontier | +3.014 |
viol | +2.907 |
trump | +2.893 |
Warsaw | +2.828 |
Neg logit contributions | |
20439 | -4.450 |
Ire | -4.268 |
=# | -4.060 |
ゴン | -4.059 |
iatus | -3.886 |
Token | , |
Token ID | 837 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
policeman | +0.343 |
viol | +0.341 |
Frontier | +0.335 |
trump | +0.321 |
Warsaw | +0.318 |
Neg logit contributions | |
20439 | -0.464 |
Ire | -0.435 |
=# | -0.422 |
Reply | -0.403 |
ゴン | -0.399 |
Token | T |
Token ID | 309 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
20439 | +0.742 |
Ire | +0.699 |
=# | +0.679 |
Reply | +0.649 |
ゴン | +0.642 |
Neg logit contributions | |
policeman | -0.488 |
viol | -0.485 |
Frontier | -0.477 |
trump | -0.457 |
Warsaw | -0.451 |
Token | >> |
Token ID | 9609 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
policeman | +0.498 |
viol | +0.495 |
Frontier | +0.491 |
trump | +0.474 |
Warsaw | +0.468 |
Neg logit contributions | |
20439 | -0.512 |
Ire | -0.479 |
=# | -0.458 |
ゴン | -0.433 |
Reply | -0.433 |
Token | ; |
Token ID | 2162 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
20439 | +0.383 |
Ire | +0.352 |
=# | +0.336 |
Reply | +0.315 |
ゴン | +0.311 |
Neg logit contributions | |
policeman | -0.521 |
viol | -0.518 |
Frontier | -0.513 |
trump | -0.498 |
Warsaw | -0.493 |
Token | = |
Token ID | 796 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
20439 | +1.754 |
Ire | +1.634 |
=# | +1.584 |
Reply | +1.503 |
ゴン | +1.479 |
Neg logit contributions | |
policeman | -1.656 |
viol | -1.646 |
Frontier | -1.625 |
trump | -1.564 |
Warsaw | -1.549 |
Token | fit |
Token ID | 4197 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
20439 | +2.796 |
Ire | +2.642 |
=# | +2.567 |
Reply | +2.461 |
ゴン | +2.439 |
Neg logit contributions | |
policeman | -1.649 |
viol | -1.635 |
Frontier | -1.610 |
trump | -1.535 |
Warsaw | -1.513 |
Token | ( |
Token ID | 7 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
policeman | +0.261 |
viol | +0.259 |
Frontier | +0.257 |
trump | +0.245 |
Warsaw | +0.243 |
Neg logit contributions | |
20439 | -0.323 |
Ire | -0.304 |
=# | -0.292 |
ゴン | -0.277 |
Reply | -0.275 |
Token | space |
Token ID | 13200 |
Feature activation | -9.12e-01 |
Pos logit contributions | |
viol | +0.943 |
policeman | +0.943 |
Frontier | +0.924 |
trump | +0.876 |
Warsaw | +0.861 |
Neg logit contributions | |
20439 | -2.048 |
Ire | -1.953 |
=# | -1.885 |
Reply | -1.818 |
ゴン | -1.811 |
Token | ); |
Token ID | 1776 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
20439 | +0.754 |
Ire | +0.706 |
=# | +0.673 |
Reply | +0.639 |
ゴン | +0.635 |
Neg logit contributions | |
policeman | -0.671 |
viol | -0.668 |
Frontier | -0.662 |
trump | -0.635 |
Warsaw | -0.631 |
Token | // |
Token ID | 3373 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
20439 | +0.284 |
Ire | +0.267 |
=# | +0.258 |
Reply | +0.247 |
ゴン | +0.247 |
Neg logit contributions | |
policeman | -0.189 |
viol | -0.188 |
Frontier | -0.185 |
trump | -0.177 |
Warsaw | -0.174 |
Token | 2 |
Token ID | 362 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
viol | +3.248 |
Frontier | +3.206 |
policeman | +3.189 |
trump | +3.071 |
Warsaw | +3.012 |
Neg logit contributions | |
20439 | -4.310 |
Ire | -4.066 |
=# | -3.806 |
ゴン | -3.730 |
iatus | -3.591 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
policeman | +0.114 |
viol | +0.112 |
Frontier | +0.112 |
trump | +0.106 |
Warsaw | +0.106 |
Neg logit contributions | |
20439 | -0.145 |
Ire | -0.136 |
=# | -0.133 |
Reply | -0.126 |
ゴン | -0.125 |
Token | Sl |
Token ID | 3454 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
viol | +0.687 |
policeman | +0.686 |
Frontier | +0.685 |
trump | +0.648 |
Warsaw | +0.645 |
Neg logit contributions | |
20439 | -0.963 |
Ire | -0.905 |
=# | -0.867 |
ゴン | -0.839 |
Reply | -0.823 |
Token | ice |
Token ID | 501 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
20439 | +1.513 |
Ire | +1.441 |
=# | +1.406 |
Reply | +1.355 |
ゴン | +1.343 |
Neg logit contributions | |
policeman | -0.598 |
viol | -0.593 |
Frontier | -0.580 |
trump | -0.544 |
Warsaw | -0.534 |
Token | the |
Token ID | 262 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
viol | +0.267 |
policeman | +0.267 |
Frontier | +0.262 |
trump | +0.249 |
Warsaw | +0.247 |
Neg logit contributions | |
20439 | -0.445 |
Ire | -0.421 |
=# | -0.407 |
ゴン | -0.390 |
Reply | -0.390 |
Token | Own |
Token ID | 11744 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
20439 | +2.158 |
Ire | +2.031 |
=# | +1.978 |
Reply | +1.887 |
ゴン | +1.855 |
Neg logit contributions | |
policeman | -1.539 |
viol | -1.532 |
Frontier | -1.497 |
trump | -1.444 |
Warsaw | -1.415 |
Token | ed |
Token ID | 276 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
policeman | +0.304 |
viol | +0.301 |
Frontier | +0.300 |
trump | +0.289 |
Warsaw | +0.287 |
Neg logit contributions | |
20439 | -0.299 |
Ire | -0.279 |
=# | -0.267 |
ゴン | -0.253 |
Reply | -0.251 |
Token | < |
Token ID | 1279 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
20439 | +0.131 |
Ire | +0.122 |
=# | +0.117 |
Reply | +0.110 |
ゴン | +0.109 |
Neg logit contributions | |
policeman | -0.142 |
viol | -0.141 |
Frontier | -0.140 |
trump | -0.135 |
Warsaw | -0.134 |
Token | T |
Token ID | 309 |
Feature activation | -8.19e-03 |
Pos logit contributions | |
20439 | +1.549 |
Ire | +1.463 |
=# | +1.415 |
Reply | +1.354 |
ゴン | +1.347 |
Neg logit contributions | |
policeman | -0.989 |
viol | -0.981 |
Frontier | -0.968 |
trump | -0.924 |
Warsaw | -0.913 |
Token | > |
Token ID | 1875 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
20439 | +0.007 |
Ire | +0.006 |
=# | +0.006 |
Reply | +0.006 |
ゴン | +0.006 |
Neg logit contributions | |
policeman | -0.006 |
viol | -0.006 |
Frontier | -0.006 |
trump | -0.006 |
Warsaw | -0.005 |
Token | { |
Token ID | 1391 |
Feature activation | -6.39e+00 |
Pos logit contributions | |
20439 | +0.759 |
Ire | +0.701 |
=# | +0.670 |
Reply | +0.630 |
ゴン | +0.623 |
Neg logit contributions | |
policeman | -0.944 |
viol | -0.938 |
Frontier | -0.930 |
trump | -0.901 |
Warsaw | -0.893 |
Token | ... |
Token ID | 2644 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
20439 | +5.416 |
Ire | +5.086 |
=# | +4.995 |
Reply | +4.787 |
iatus | +4.659 |
Neg logit contributions | |
viol | -4.137 |
policeman | -4.127 |
Frontier | -4.034 |
trump | -3.867 |
Warsaw | -3.769 |
Token | } |
Token ID | 1782 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
20439 | +0.529 |
Ire | +0.500 |
=# | +0.485 |
Reply | +0.465 |
ゴン | +0.461 |
Neg logit contributions | |
policeman | -0.312 |
viol | -0.310 |
Frontier | -0.305 |
trump | -0.291 |
Warsaw | -0.286 |
Token | impl |
Token ID | 4114 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
20439 | +1.471 |
Ire | +1.390 |
=# | +1.353 |
Reply | +1.296 |
ゴン | +1.280 |
Neg logit contributions | |
policeman | -0.900 |
viol | -0.895 |
Frontier | -0.879 |
trump | -0.839 |
Warsaw | -0.826 |
Token | < |
Token ID | 1279 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
20439 | +1.842 |
Ire | +1.725 |
=# | +1.669 |
Reply | +1.588 |
ゴン | +1.569 |
Neg logit contributions | |
policeman | -1.531 |
viol | -1.521 |
Frontier | -1.501 |
trump | -1.444 |
Warsaw | -1.427 |
Token | T |
Token ID | 309 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
20439 | +0.135 |
Ire | +0.128 |
=# | +0.123 |
ゴン | +0.118 |
Reply | +0.118 |
Neg logit contributions | |
policeman | -0.084 |
viol | -0.083 |
Frontier | -0.083 |
trump | -0.078 |
Warsaw | -0.077 |
Token | s |
Token ID | 82 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
policeman | +1.153 |
viol | +1.150 |
Frontier | +1.139 |
trump | +1.097 |
Warsaw | +1.079 |
Neg logit contributions | |
20439 | -1.426 |
Ire | -1.322 |
=# | -1.274 |
ゴン | -1.234 |
Reply | -1.212 |
Token | Format |
Token ID | 26227 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
policeman | +0.520 |
viol | +0.515 |
Frontier | +0.510 |
trump | +0.490 |
Warsaw | +0.484 |
Neg logit contributions | |
20439 | -0.664 |
Ire | -0.622 |
=# | -0.600 |
Reply | -0.573 |
ゴン | -0.569 |
Token | ) |
Token ID | 8 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
20439 | +0.637 |
Ire | +0.596 |
=# | +0.573 |
Reply | +0.546 |
ゴン | +0.546 |
Neg logit contributions | |
policeman | -0.504 |
viol | -0.502 |
Frontier | -0.494 |
trump | -0.475 |
Warsaw | -0.471 |
Token | } |
Token ID | 1782 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
20439 | +0.971 |
Ire | +0.914 |
=# | +0.883 |
Reply | +0.846 |
ゴン | +0.842 |
Neg logit contributions | |
policeman | -0.638 |
viol | -0.634 |
Frontier | -0.625 |
trump | -0.598 |
Warsaw | -0.589 |
Token | else |
Token ID | 2073 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
20439 | +2.576 |
Ire | +2.434 |
=# | +2.369 |
Reply | +2.267 |
ゴン | +2.233 |
Neg logit contributions | |
policeman | -1.624 |
viol | -1.609 |
Frontier | -1.584 |
trump | -1.513 |
Warsaw | -1.496 |
Token | { |
Token ID | 1391 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
20439 | +1.902 |
Ire | +1.759 |
=# | +1.686 |
Reply | +1.554 |
iatus | +1.496 |
Neg logit contributions | |
policeman | -3.043 |
viol | -3.009 |
Frontier | -2.975 |
trump | -2.893 |
Warsaw | -2.885 |
Token | $ |
Token ID | 720 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
20439 | +5.194 |
Ire | +4.915 |
=# | +4.882 |
Reply | +4.556 |
Instr | +4.408 |
Neg logit contributions | |
policeman | -4.420 |
viol | -4.335 |
Frontier | -4.286 |
Warsaw | -4.105 |
trump | -4.098 |
Token | font |
Token ID | 10331 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
20439 | +0.912 |
Ire | +0.853 |
=# | +0.826 |
Reply | +0.786 |
ゴン | +0.782 |
Neg logit contributions | |
policeman | -0.725 |
viol | -0.721 |
Frontier | -0.712 |
trump | -0.684 |
Warsaw | -0.676 |
Token | 2 |
Token ID | 17 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
20439 | +0.761 |
Ire | +0.716 |
=# | +0.693 |
Reply | +0.661 |
ゴン | +0.654 |
Neg logit contributions | |
policeman | -0.570 |
viol | -0.566 |
Frontier | -0.558 |
trump | -0.535 |
Warsaw | -0.529 |
Token | = |
Token ID | 796 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
20439 | +0.309 |
Ire | +0.289 |
ゴン | +0.270 |
=# | +0.269 |
Reply | +0.258 |
Neg logit contributions | |
policeman | -0.294 |
viol | -0.293 |
Frontier | -0.292 |
trump | -0.284 |
Warsaw | -0.280 |
Token | New |
Token ID | 968 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
20439 | +2.924 |
Ire | +2.774 |
=# | +2.717 |
Reply | +2.596 |
ゴン | +2.559 |
Neg logit contributions | |
policeman | -1.612 |
viol | -1.597 |
Frontier | -1.546 |
trump | -1.480 |
Warsaw | -1.449 |
Token | : |
Token ID | 25 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
20439 | +0.091 |
Ire | +0.086 |
=# | +0.083 |
Reply | +0.079 |
ゴン | +0.079 |
Neg logit contributions | |
policeman | -0.057 |
viol | -0.057 |
Frontier | -0.056 |
trump | -0.053 |
Warsaw | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
20439 | +0.847 |
Ire | +0.803 |
=# | +0.783 |
Reply | +0.750 |
ゴン | +0.741 |
Neg logit contributions | |
policeman | -0.502 |
viol | -0.494 |
Frontier | -0.489 |
trump | -0.466 |
Warsaw | -0.458 |
Token |
|
Token ID | 198 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
policeman | +0.202 |
viol | +0.200 |
Frontier | +0.199 |
trump | +0.188 |
Warsaw | +0.186 |
Neg logit contributions | |
20439 | -0.421 |
Ire | -0.398 |
=# | -0.385 |
ゴン | -0.376 |
Reply | -0.370 |
Token | struct |
Token ID | 7249 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
20439 | +0.034 |
Ire | +0.031 |
=# | +0.030 |
ゴン | +0.029 |
Reply | +0.029 |
Neg logit contributions | |
policeman | -0.026 |
viol | -0.025 |
Frontier | -0.025 |
trump | -0.024 |
Warsaw | -0.024 |
Token | Board |
Token ID | 5926 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
20439 | +2.545 |
Ire | +2.387 |
=# | +2.321 |
Reply | +2.210 |
ゴン | +2.172 |
Neg logit contributions | |
policeman | -2.046 |
viol | -2.027 |
Frontier | -1.996 |
trump | -1.919 |
Warsaw | -1.895 |
Token | { |
Token ID | 1391 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
20439 | +1.213 |
Ire | +1.115 |
=# | +1.067 |
Reply | +1.000 |
ゴン | +0.985 |
Neg logit contributions | |
policeman | -1.623 |
viol | -1.615 |
Frontier | -1.599 |
trump | -1.550 |
Warsaw | -1.537 |
Token | ... |
Token ID | 2644 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
20439 | +5.539 |
Ire | +5.187 |
=# | +5.094 |
Reply | +4.934 |
iatus | +4.716 |
Neg logit contributions | |
policeman | -4.050 |
viol | -3.969 |
Frontier | -3.963 |
trump | -3.719 |
Warsaw | -3.713 |
Token | prev |
Token ID | 8654 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
20439 | +0.247 |
Ire | +0.233 |
=# | +0.225 |
ゴン | +0.215 |
Reply | +0.214 |
Neg logit contributions | |
policeman | -0.166 |
viol | -0.165 |
Frontier | -0.162 |
trump | -0.157 |
Warsaw | -0.153 |
Token | _ |
Token ID | 62 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
policeman | +0.297 |
viol | +0.295 |
Frontier | +0.292 |
trump | +0.280 |
Warsaw | +0.278 |
Neg logit contributions | |
20439 | -0.333 |
Ire | -0.312 |
=# | -0.302 |
Reply | -0.288 |
ゴン | -0.282 |
Token | state |
Token ID | 5219 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
20439 | +0.587 |
Ire | +0.542 |
=# | +0.524 |
Reply | +0.493 |
ゴン | +0.481 |
Neg logit contributions | |
policeman | -0.716 |
viol | -0.711 |
Frontier | -0.709 |
trump | -0.679 |
Warsaw | -0.678 |
Token | : |
Token ID | 1058 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
20439 | +1.425 |
Ire | +1.334 |
=# | +1.304 |
Reply | +1.242 |
ゴン | +1.223 |
Neg logit contributions | |
policeman | -1.102 |
viol | -1.089 |
Frontier | -1.080 |
trump | -1.030 |
Warsaw | -1.022 |
Token | : |
Token ID | 25 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
viol | +0.211 |
policeman | +0.210 |
Frontier | +0.205 |
trump | +0.199 |
Warsaw | +0.194 |
Neg logit contributions | |
20439 | -0.308 |
Ire | -0.291 |
=# | -0.280 |
ゴン | -0.270 |
Reply | -0.268 |
Token |
|
Token ID | 198 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
20439 | +0.438 |
Ire | +0.413 |
=# | +0.400 |
Reply | +0.385 |
ゴン | +0.384 |
Neg logit contributions | |
viol | -0.237 |
policeman | -0.236 |
Frontier | -0.232 |
trump | -0.221 |
Warsaw | -0.218 |
Token |
|
Token ID | 198 |
Feature activation | -6.81e-03 |
Pos logit contributions | |
policeman | +0.259 |
viol | +0.256 |
Frontier | +0.254 |
trump | +0.241 |
Warsaw | +0.237 |
Neg logit contributions | |
20439 | -0.543 |
Ire | -0.514 |
=# | -0.496 |
ゴン | -0.486 |
Reply | -0.478 |
Token | impl |
Token ID | 23928 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
20439 | +0.006 |
Ire | +0.006 |
=# | +0.006 |
ゴン | +0.005 |
Reply | +0.005 |
Neg logit contributions | |
policeman | -0.004 |
viol | -0.004 |
Frontier | -0.004 |
trump | -0.004 |
Warsaw | -0.004 |
Token | Board |
Token ID | 5926 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
20439 | +2.359 |
Ire | +2.210 |
=# | +2.126 |
Reply | +2.028 |
ゴン | +2.016 |
Neg logit contributions | |
policeman | -1.988 |
viol | -1.977 |
Frontier | -1.955 |
trump | -1.881 |
Warsaw | -1.863 |
Token | { |
Token ID | 1391 |
Feature activation | -6.28e+00 |
Pos logit contributions | |
20439 | +1.162 |
Ire | +1.082 |
=# | +1.035 |
ゴン | +0.982 |
Reply | +0.970 |
Neg logit contributions | |
policeman | -1.226 |
viol | -1.220 |
Frontier | -1.212 |
trump | -1.169 |
Warsaw | -1.160 |
Token | pub |
Token ID | 2240 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
20439 | +5.366 |
Ire | +5.033 |
=# | +4.964 |
Reply | +4.748 |
iatus | +4.608 |
Neg logit contributions | |
policeman | -4.134 |
viol | -4.103 |
Frontier | -4.045 |
trump | -3.853 |
Warsaw | -3.815 |
Token | fn |
Token ID | 24714 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
20439 | +0.281 |
Ire | +0.266 |
=# | +0.255 |
ゴン | +0.246 |
Reply | +0.245 |
Neg logit contributions | |
policeman | -0.177 |
viol | -0.176 |
Frontier | -0.175 |
trump | -0.168 |
Warsaw | -0.165 |
Token | copy |
Token ID | 4866 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
20439 | +1.889 |
Ire | +1.785 |
=# | +1.734 |
Reply | +1.662 |
ゴン | +1.647 |
Neg logit contributions | |
policeman | -1.124 |
viol | -1.115 |
Frontier | -1.098 |
trump | -1.046 |
Warsaw | -1.032 |
Token | _ |
Token ID | 62 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
policeman | +0.791 |
viol | +0.785 |
Frontier | +0.775 |
trump | +0.749 |
Warsaw | +0.738 |
Neg logit contributions | |
20439 | -1.003 |
Ire | -0.947 |
=# | -0.913 |
ゴン | -0.870 |
Reply | -0.860 |
Token | board |
Token ID | 3526 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
20439 | +0.120 |
Ire | +0.114 |
=# | +0.109 |
ゴン | +0.104 |
Reply | +0.104 |
Neg logit contributions | |
policeman | -0.089 |
viol | -0.089 |
Frontier | -0.087 |
trump | -0.084 |
Warsaw | -0.083 |
Token | : |
Token ID | 25 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
20439 | +0.557 |
Ire | +0.526 |
=# | +0.508 |
ゴン | +0.488 |
Reply | +0.486 |
Neg logit contributions | |
policeman | -0.373 |
viol | -0.372 |
Frontier | -0.367 |
trump | -0.352 |
Warsaw | -0.346 |
Token |
|
Token ID | 198 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
20439 | +0.548 |
Ire | +0.518 |
=# | +0.503 |
Reply | +0.483 |
ゴン | +0.480 |
Neg logit contributions | |
policeman | -0.302 |
viol | -0.300 |
Frontier | -0.295 |
trump | -0.281 |
Warsaw | -0.277 |
Token |
|
Token ID | 198 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
policeman | +0.193 |
viol | +0.190 |
Frontier | +0.189 |
trump | +0.179 |
Warsaw | +0.177 |
Neg logit contributions | |
20439 | -0.402 |
Ire | -0.380 |
=# | -0.367 |
ゴン | -0.359 |
Reply | -0.353 |
Token | struct |
Token ID | 7249 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
policeman | +0.140 |
viol | +0.139 |
Frontier | +0.137 |
trump | +0.132 |
Warsaw | +0.130 |
Neg logit contributions | |
20439 | -0.168 |
Ire | -0.155 |
=# | -0.149 |
ゴン | -0.143 |
Reply | -0.141 |
Token | Board |
Token ID | 5926 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
20439 | +1.923 |
Ire | +1.806 |
=# | +1.745 |
Reply | +1.664 |
ゴン | +1.654 |
Neg logit contributions | |
policeman | -1.464 |
viol | -1.456 |
Frontier | -1.438 |
trump | -1.381 |
Warsaw | -1.364 |
Token | { |
Token ID | 1391 |
Feature activation | -6.20e+00 |
Pos logit contributions | |
20439 | +1.560 |
Ire | +1.443 |
=# | +1.379 |
Reply | +1.286 |
ゴン | +1.283 |
Neg logit contributions | |
policeman | -1.978 |
viol | -1.970 |
Frontier | -1.952 |
trump | -1.889 |
Warsaw | -1.873 |
Token | turn |
Token ID | 1210 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
20439 | +5.359 |
Ire | +5.039 |
=# | +4.936 |
Reply | +4.715 |
iatus | +4.578 |
Neg logit contributions | |
policeman | -4.059 |
viol | -3.998 |
Frontier | -3.946 |
trump | -3.751 |
Warsaw | -3.737 |
Token | : |
Token ID | 1058 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
20439 | +1.398 |
Ire | +1.312 |
=# | +1.269 |
Reply | +1.210 |
ゴン | +1.198 |
Neg logit contributions | |
policeman | -1.088 |
viol | -1.081 |
Frontier | -1.067 |
trump | -1.025 |
Warsaw | -1.012 |
Token | Player |
Token ID | 7853 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
20439 | +3.330 |
Ire | +3.123 |
=# | +3.024 |
Reply | +2.882 |
ゴン | +2.848 |
Neg logit contributions | |
policeman | -2.660 |
viol | -2.641 |
Frontier | -2.607 |
trump | -2.504 |
Warsaw | -2.477 |
Token | , |
Token ID | 837 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
20439 | +1.163 |
Ire | +1.092 |
=# | +1.056 |
Reply | +1.001 |
ゴン | +0.997 |
Neg logit contributions | |
policeman | -0.916 |
viol | -0.909 |
Frontier | -0.900 |
trump | -0.864 |
Warsaw | -0.854 |
Token | // |
Token ID | 3373 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
20439 | +1.153 |
Ire | +1.091 |
=# | +1.060 |
Reply | +1.017 |
ゴン | +1.010 |
Neg logit contributions | |
policeman | -0.644 |
viol | -0.639 |
Frontier | -0.629 |
trump | -0.599 |
Warsaw | -0.589 |
Token | Wall |
Token ID | 22401 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
policeman | +0.813 |
viol | +0.808 |
Frontier | +0.807 |
trump | +0.773 |
Warsaw | +0.767 |
Neg logit contributions | |
20439 | -0.837 |
Ire | -0.767 |
=# | -0.740 |
ゴン | -0.709 |
Reply | -0.697 |
Token | paper |
Token ID | 20189 |
Feature activation | -9.53e-02 |
Pos logit contributions | |
policeman | +0.512 |
viol | +0.506 |
Frontier | +0.499 |
trump | +0.470 |
Warsaw | +0.467 |
Neg logit contributions | |
20439 | -1.304 |
Ire | -1.232 |
=# | -1.191 |
ゴン | -1.163 |
Reply | -1.147 |
Token | As |
Token ID | 1722 |
Feature activation | -9.42e-01 |
Pos logit contributions | |
20439 | +0.080 |
Ire | +0.074 |
=# | +0.071 |
ゴン | +0.069 |
Reply | +0.067 |
Neg logit contributions | |
policeman | -0.068 |
viol | -0.068 |
Frontier | -0.067 |
trump | -0.065 |
Warsaw | -0.064 |
Token | Source |
Token ID | 7416 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
20439 | +0.753 |
Ire | +0.700 |
=# | +0.675 |
Reply | +0.639 |
ゴン | +0.636 |
Neg logit contributions | |
policeman | -0.714 |
viol | -0.709 |
Frontier | -0.703 |
trump | -0.676 |
Warsaw | -0.670 |
Token | ) |
Token ID | 8 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
policeman | +1.447 |
viol | +1.434 |
Frontier | +1.428 |
trump | +1.374 |
Warsaw | +1.357 |
Neg logit contributions | |
20439 | -1.795 |
Ire | -1.644 |
=# | -1.556 |
ゴン | -1.528 |
Reply | -1.501 |
Token | { |
Token ID | 1391 |
Feature activation | -6.12e+00 |
Pos logit contributions | |
20439 | +1.076 |
Ire | +0.983 |
=# | +0.932 |
Reply | +0.873 |
ゴン | +0.869 |
Neg logit contributions | |
policeman | -1.520 |
viol | -1.513 |
Frontier | -1.500 |
trump | -1.457 |
Warsaw | -1.441 |
Token | $ |
Token ID | 720 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
20439 | +4.723 |
Ire | +4.466 |
=# | +4.456 |
Reply | +4.162 |
Instr | +4.006 |
Neg logit contributions | |
policeman | -4.495 |
viol | -4.439 |
Frontier | -4.357 |
trump | -4.203 |
Warsaw | -4.186 |
Token | w |
Token ID | 86 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
20439 | +0.566 |
Ire | +0.530 |
=# | +0.512 |
Reply | +0.488 |
ゴン | +0.486 |
Neg logit contributions | |
policeman | -0.448 |
viol | -0.445 |
Frontier | -0.439 |
trump | -0.422 |
Warsaw | -0.417 |
Token | path |
Token ID | 6978 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
20439 | +0.547 |
Ire | +0.510 |
=# | +0.491 |
ゴン | +0.467 |
Reply | +0.466 |
Neg logit contributions | |
policeman | -0.471 |
viol | -0.466 |
Frontier | -0.463 |
trump | -0.445 |
Warsaw | -0.440 |
Token | = |
Token ID | 796 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
policeman | +0.268 |
viol | +0.265 |
Frontier | +0.265 |
trump | +0.256 |
Warsaw | +0.252 |
Neg logit contributions | |
20439 | -0.274 |
Ire | -0.254 |
=# | -0.240 |
ゴン | -0.232 |
Reply | -0.229 |
Token | ( |
Token ID | 357 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
20439 | +1.894 |
Ire | +1.776 |
=# | +1.720 |
Reply | +1.636 |
ゴン | +1.616 |
Neg logit contributions | |
policeman | -1.558 |
viol | -1.548 |
Frontier | -1.527 |
trump | -1.468 |
Warsaw | -1.452 |
Token |
|
Token ID | 198 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
policeman | +0.494 |
Frontier | +0.490 |
viol | +0.488 |
trump | +0.463 |
Warsaw | +0.456 |
Neg logit contributions | |
20439 | -1.092 |
Ire | -1.035 |
=# | -0.995 |
ゴン | -0.986 |
Instr | -0.956 |
Token | 55 |
Token ID | 2816 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
policeman | +0.714 |
Frontier | +0.713 |
viol | +0.710 |
trump | +0.682 |
Warsaw | +0.675 |
Neg logit contributions | |
20439 | -0.794 |
Ire | -0.741 |
=# | -0.714 |
ゴン | -0.700 |
iatus | -0.681 |
Token | Recommend |
Token ID | 30361 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
20439 | +0.357 |
Ire | +0.333 |
=# | +0.319 |
ゴン | +0.313 |
iatus | +0.301 |
Neg logit contributions | |
policeman | -0.308 |
viol | -0.307 |
Frontier | -0.306 |
trump | -0.295 |
Warsaw | -0.293 |
Token |
|
Token ID | 198 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
20439 | +3.986 |
Ire | +3.782 |
=# | +3.706 |
Reply | +3.579 |
iatus | +3.425 |
Neg logit contributions | |
policeman | -2.459 |
viol | -2.443 |
Frontier | -2.367 |
trump | -2.254 |
Warsaw | -2.234 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
policeman | +0.648 |
Frontier | +0.646 |
viol | +0.638 |
trump | +0.609 |
Warsaw | +0.599 |
Neg logit contributions | |
20439 | -1.536 |
Ire | -1.452 |
ゴン | -1.399 |
=# | -1.394 |
Instr | -1.352 |
Token | Recommend |
Token ID | 41248 |
Feature activation | -6.09e+00 |
Pos logit contributions | |
Frontier | +0.844 |
policeman | +0.844 |
viol | +0.833 |
trump | +0.806 |
Warsaw | +0.791 |
Neg logit contributions | |
20439 | -0.882 |
Ire | -0.806 |
=# | -0.792 |
ゴン | -0.784 |
iatus | -0.757 |
Token | Share |
Token ID | 8734 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
20439 | +4.182 |
Ire | +3.931 |
Reply | +3.756 |
=# | +3.725 |
DAQ | +3.391 |
Neg logit contributions | |
viol | -5.112 |
policeman | -5.074 |
Frontier | -4.970 |
Warsaw | -4.800 |
trump | -4.795 |
Token | this |
Token ID | 428 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
20439 | +0.260 |
Ire | +0.246 |
=# | +0.237 |
ゴン | +0.232 |
iatus | +0.226 |
Neg logit contributions | |
policeman | -0.144 |
Frontier | -0.142 |
viol | -0.142 |
trump | -0.136 |
Warsaw | -0.131 |
Token | comment |
Token ID | 2912 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
policeman | +0.163 |
viol | +0.160 |
Frontier | +0.157 |
trump | +0.152 |
Warsaw | +0.148 |
Neg logit contributions | |
20439 | -0.252 |
Ire | -0.238 |
=# | -0.229 |
ゴン | -0.222 |
Reply | -0.218 |
Token | on |
Token ID | 319 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
20439 | +1.633 |
Ire | +1.528 |
=# | +1.471 |
Reply | +1.399 |
ゴン | +1.389 |
Neg logit contributions | |
policeman | -1.460 |
viol | -1.451 |
Frontier | -1.434 |
trump | -1.382 |
Warsaw | -1.367 |
Token | Facebook |
Token ID | 3203 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
20439 | +1.209 |
Ire | +1.151 |
=# | +1.109 |
ゴン | +1.087 |
Reply | +1.060 |
Neg logit contributions | |
policeman | -0.614 |
viol | -0.612 |
Frontier | -0.611 |
trump | -0.575 |
Warsaw | -0.567 |
Token |
|
Token ID | 198 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
policeman | +0.299 |
Frontier | +0.297 |
viol | +0.295 |
trump | +0.280 |
Warsaw | +0.277 |
Neg logit contributions | |
20439 | -0.647 |
Ire | -0.613 |
=# | -0.590 |
ゴン | -0.581 |
Instr | -0.566 |
Token | 14 |
Token ID | 1415 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
policeman | +0.437 |
Frontier | +0.435 |
viol | +0.434 |
trump | +0.417 |
Warsaw | +0.412 |
Neg logit contributions | |
20439 | -0.482 |
Ire | -0.451 |
=# | -0.434 |
ゴン | -0.419 |
iatus | -0.410 |
Token | Recommend |
Token ID | 30361 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
20439 | +0.169 |
Ire | +0.159 |
=# | +0.153 |
ゴン | +0.149 |
iatus | +0.144 |
Neg logit contributions | |
viol | -0.144 |
policeman | -0.144 |
Frontier | -0.144 |
trump | -0.138 |
Warsaw | -0.136 |
Token |
|
Token ID | 198 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
20439 | +3.696 |
Ire | +3.501 |
=# | +3.412 |
Reply | +3.294 |
ゴン | +3.188 |
Neg logit contributions | |
policeman | -2.162 |
viol | -2.153 |
Frontier | -2.089 |
trump | -1.995 |
Warsaw | -1.967 |
Token |
|
Token ID | 198 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
policeman | +0.638 |
Frontier | +0.636 |
viol | +0.628 |
trump | +0.600 |
Warsaw | +0.592 |
Neg logit contributions | |
20439 | -1.531 |
Ire | -1.451 |
ゴン | -1.399 |
=# | -1.392 |
Instr | -1.359 |
Token | Recommend |
Token ID | 41248 |
Feature activation | -6.09e+00 |
Pos logit contributions | |
Frontier | +0.775 |
policeman | +0.769 |
viol | +0.761 |
trump | +0.733 |
Warsaw | +0.712 |
Neg logit contributions | |
20439 | -0.914 |
ゴン | -0.843 |
Ire | -0.838 |
=# | -0.832 |
iatus | -0.803 |
Token | Share |
Token ID | 8734 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
20439 | +4.092 |
Ire | +3.865 |
Reply | +3.685 |
=# | +3.626 |
iatus | +3.296 |
Neg logit contributions | |
viol | -5.162 |
policeman | -5.132 |
Frontier | -5.033 |
Warsaw | -4.902 |
trump | -4.850 |
Token | this |
Token ID | 428 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
Frontier | +0.049 |
policeman | +0.048 |
viol | +0.047 |
trump | +0.046 |
Warsaw | +0.044 |
Neg logit contributions | |
20439 | -0.094 |
Ire | -0.088 |
=# | -0.086 |
ゴン | -0.085 |
iatus | -0.082 |
Token | comment |
Token ID | 2912 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
policeman | +0.182 |
Frontier | +0.178 |
viol | +0.176 |
trump | +0.169 |
Warsaw | +0.165 |
Neg logit contributions | |
20439 | -0.286 |
Ire | -0.265 |
=# | -0.257 |
ゴン | -0.251 |
Reply | -0.244 |
Token | on |
Token ID | 319 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
20439 | +1.402 |
Ire | +1.314 |
=# | +1.259 |
ゴン | +1.205 |
Reply | +1.198 |
Neg logit contributions | |
policeman | -1.203 |
viol | -1.193 |
Frontier | -1.181 |
trump | -1.142 |
Warsaw | -1.124 |
Token | Facebook |
Token ID | 3203 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
20439 | +1.273 |
Ire | +1.205 |
=# | +1.170 |
ゴン | +1.145 |
iatus | +1.116 |
Neg logit contributions | |
policeman | -0.626 |
viol | -0.614 |
Frontier | -0.609 |
trump | -0.584 |
Warsaw | -0.566 |
Token | Count |
Token ID | 12332 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
20439 | +0.110 |
Ire | +0.102 |
=# | +0.098 |
ゴン | +0.092 |
Reply | +0.092 |
Neg logit contributions | |
policeman | -0.111 |
viol | -0.110 |
Frontier | -0.109 |
trump | -0.105 |
Warsaw | -0.104 |
Token | ; |
Token ID | 2162 |
Feature activation | -9.06e-01 |
Pos logit contributions | |
20439 | +0.024 |
Ire | +0.023 |
=# | +0.022 |
ゴン | +0.021 |
Reply | +0.021 |
Neg logit contributions | |
policeman | -0.021 |
viol | -0.020 |
Frontier | -0.020 |
trump | -0.019 |
Warsaw | -0.019 |
Token | $ |
Token ID | 720 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
20439 | +0.679 |
Ire | +0.626 |
=# | +0.599 |
ゴン | +0.570 |
Reply | +0.565 |
Neg logit contributions | |
policeman | -0.737 |
viol | -0.735 |
Frontier | -0.727 |
trump | -0.703 |
Warsaw | -0.694 |
Token | i |
Token ID | 72 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
policeman | +0.183 |
viol | +0.183 |
Frontier | +0.179 |
trump | +0.175 |
Warsaw | +0.171 |
Neg logit contributions | |
20439 | -0.174 |
Ire | -0.161 |
=# | -0.154 |
ゴン | -0.149 |
Reply | -0.145 |
Token | ++) |
Token ID | 29577 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
policeman | +0.316 |
viol | +0.314 |
Frontier | +0.310 |
trump | +0.299 |
Warsaw | +0.295 |
Neg logit contributions | |
20439 | -0.380 |
Ire | -0.356 |
=# | -0.340 |
ゴン | -0.326 |
Reply | -0.324 |
Token | { |
Token ID | 1391 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
20439 | +0.091 |
Ire | +0.083 |
=# | +0.079 |
ゴン | +0.074 |
Reply | +0.074 |
Neg logit contributions | |
policeman | -0.125 |
viol | -0.124 |
Frontier | -0.123 |
trump | -0.120 |
Warsaw | -0.118 |
Token | if |
Token ID | 611 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
20439 | +4.754 |
Ire | +4.468 |
=# | +4.432 |
Reply | +4.160 |
Instr | +4.003 |
Neg logit contributions | |
policeman | -4.339 |
viol | -4.289 |
Frontier | -4.217 |
trump | -4.048 |
Warsaw | -4.042 |
Token | ($ |
Token ID | 7198 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
20439 | +1.844 |
Ire | +1.700 |
=# | +1.631 |
Reply | +1.530 |
ゴン | +1.506 |
Neg logit contributions | |
policeman | -2.334 |
viol | -2.322 |
Frontier | -2.297 |
trump | -2.226 |
Warsaw | -2.206 |
Token | i |
Token ID | 72 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
policeman | +0.107 |
viol | +0.106 |
Frontier | +0.105 |
trump | +0.102 |
Warsaw | +0.100 |
Neg logit contributions | |
20439 | -0.104 |
Ire | -0.095 |
=# | -0.092 |
ゴン | -0.088 |
Reply | -0.086 |
Token | - |
Token ID | 532 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
20439 | +0.629 |
Ire | +0.589 |
=# | +0.561 |
ゴン | +0.535 |
Reply | +0.532 |
Neg logit contributions | |
policeman | -0.584 |
viol | -0.580 |
Frontier | -0.575 |
trump | -0.556 |
Warsaw | -0.548 |
Token | eq |
Token ID | 27363 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
20439 | +1.122 |
Ire | +1.061 |
=# | +1.028 |
Reply | +0.987 |
ゴン | +0.982 |
Neg logit contributions | |
policeman | -0.650 |
viol | -0.643 |
Frontier | -0.634 |
trump | -0.605 |
Warsaw | -0.595 |
Token |
|
Token ID | 198 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
policeman | +0.410 |
Frontier | +0.406 |
viol | +0.406 |
trump | +0.384 |
Warsaw | +0.378 |
Neg logit contributions | |
20439 | -0.893 |
Ire | -0.845 |
=# | -0.814 |
ゴン | -0.803 |
Reply | -0.780 |
Token | 19 |
Token ID | 1129 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Frontier | +0.530 |
policeman | +0.530 |
viol | +0.528 |
trump | +0.507 |
Warsaw | +0.502 |
Neg logit contributions | |
20439 | -0.594 |
Ire | -0.554 |
=# | -0.533 |
ゴン | -0.522 |
iatus | -0.510 |
Token | Recommend |
Token ID | 30361 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
policeman | +0.075 |
viol | +0.075 |
Frontier | +0.075 |
trump | +0.072 |
Warsaw | +0.071 |
Neg logit contributions | |
20439 | -0.087 |
Ire | -0.080 |
=# | -0.077 |
ゴン | -0.075 |
iatus | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
20439 | +3.421 |
Ire | +3.246 |
=# | +3.174 |
Reply | +3.061 |
iatus | +2.945 |
Neg logit contributions | |
policeman | -2.044 |
viol | -2.027 |
Frontier | -1.972 |
trump | -1.877 |
Warsaw | -1.856 |
Token |
|
Token ID | 198 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
policeman | +0.765 |
Frontier | +0.764 |
viol | +0.754 |
trump | +0.720 |
Warsaw | +0.708 |
Neg logit contributions | |
20439 | -1.832 |
Ire | -1.732 |
ゴン | -1.669 |
=# | -1.662 |
Instr | -1.613 |
Token | Recommend |
Token ID | 41248 |
Feature activation | -5.85e+00 |
Pos logit contributions | |
Frontier | +1.003 |
policeman | +1.002 |
viol | +0.991 |
trump | +0.956 |
Warsaw | +0.947 |
Neg logit contributions | |
20439 | -0.999 |
Ire | -0.918 |
=# | -0.891 |
ゴン | -0.880 |
iatus | -0.847 |
Token | Share |
Token ID | 8734 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
20439 | +4.017 |
Ire | +3.783 |
Reply | +3.614 |
=# | +3.591 |
DAQ | +3.256 |
Neg logit contributions | |
viol | -4.894 |
policeman | -4.869 |
Frontier | -4.766 |
Warsaw | -4.602 |
trump | -4.586 |
Token | this |
Token ID | 428 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
20439 | +0.164 |
Ire | +0.155 |
=# | +0.149 |
ゴン | +0.146 |
Reply | +0.142 |
Neg logit contributions | |
policeman | -0.092 |
viol | -0.091 |
Frontier | -0.090 |
trump | -0.087 |
Warsaw | -0.084 |
Token | comment |
Token ID | 2912 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
policeman | +0.144 |
viol | +0.142 |
Frontier | +0.139 |
trump | +0.135 |
Warsaw | +0.131 |
Neg logit contributions | |
20439 | -0.219 |
Ire | -0.207 |
=# | -0.199 |
ゴン | -0.192 |
Reply | -0.190 |
Token | on |
Token ID | 319 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
20439 | +1.631 |
Ire | +1.526 |
=# | +1.469 |
Reply | +1.397 |
ゴン | +1.386 |
Neg logit contributions | |
policeman | -1.460 |
viol | -1.451 |
Frontier | -1.434 |
trump | -1.382 |
Warsaw | -1.367 |
Token | Facebook |
Token ID | 3203 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
20439 | +1.181 |
Ire | +1.124 |
=# | +1.083 |
ゴン | +1.059 |
Reply | +1.035 |
Neg logit contributions | |
policeman | -0.607 |
viol | -0.605 |
Frontier | -0.604 |
trump | -0.569 |
Warsaw | -0.562 |
Token |
|
Token ID | 198 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
20439 | +0.827 |
Ire | +0.783 |
=# | +0.761 |
Reply | +0.731 |
ゴン | +0.724 |
Neg logit contributions | |
policeman | -0.452 |
viol | -0.448 |
Frontier | -0.441 |
trump | -0.419 |
Warsaw | -0.413 |
Token |
|
Token ID | 198 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
policeman | +0.222 |
viol | +0.220 |
Frontier | +0.219 |
trump | +0.207 |
Warsaw | +0.204 |
Neg logit contributions | |
20439 | -0.470 |
Ire | -0.444 |
=# | -0.428 |
ゴン | -0.420 |
Reply | -0.412 |
Token | pub |
Token ID | 12984 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
policeman | +0.026 |
viol | +0.025 |
Frontier | +0.025 |
trump | +0.024 |
Warsaw | +0.024 |
Neg logit contributions | |
20439 | -0.034 |
Ire | -0.032 |
=# | -0.030 |
ゴン | -0.029 |
Reply | -0.029 |
Token | struct |
Token ID | 2878 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
20439 | +0.775 |
Ire | +0.730 |
=# | +0.698 |
Reply | +0.669 |
ゴン | +0.667 |
Neg logit contributions | |
policeman | -0.636 |
viol | -0.630 |
Frontier | -0.623 |
trump | -0.601 |
Warsaw | -0.593 |
Token | Guard |
Token ID | 4932 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
20439 | +1.539 |
Ire | +1.446 |
=# | +1.400 |
Reply | +1.335 |
ゴン | +1.321 |
Neg logit contributions | |
policeman | -1.171 |
viol | -1.163 |
Frontier | -1.148 |
trump | -1.102 |
Warsaw | -1.089 |
Token | { |
Token ID | 1391 |
Feature activation | -5.84e+00 |
Pos logit contributions | |
20439 | +1.767 |
Ire | +1.631 |
=# | +1.565 |
Reply | +1.472 |
ゴン | +1.451 |
Neg logit contributions | |
policeman | -2.163 |
viol | -2.151 |
Frontier | -2.129 |
trump | -2.063 |
Warsaw | -2.043 |
Token | ... |
Token ID | 2644 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
20439 | +5.216 |
Ire | +4.919 |
=# | +4.876 |
Reply | +4.643 |
iatus | +4.550 |
Neg logit contributions | |
policeman | -3.552 |
viol | -3.521 |
Frontier | -3.459 |
trump | -3.308 |
Warsaw | -3.263 |
Token | } |
Token ID | 1782 |
Feature activation | -7.12e-01 |
Pos logit contributions | |
policeman | +0.156 |
viol | +0.155 |
Frontier | +0.152 |
trump | +0.146 |
Warsaw | +0.144 |
Neg logit contributions | |
20439 | -0.245 |
Ire | -0.231 |
=# | -0.222 |
ゴン | -0.213 |
Reply | -0.212 |
Token | pub |
Token ID | 2240 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
20439 | +0.686 |
Ire | +0.648 |
=# | +0.629 |
Reply | +0.603 |
ゴン | +0.597 |
Neg logit contributions | |
policeman | -0.421 |
viol | -0.418 |
Frontier | -0.412 |
trump | -0.393 |
Warsaw | -0.388 |
Token | fn |
Token ID | 24714 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
20439 | +0.725 |
Ire | +0.676 |
=# | +0.654 |
Reply | +0.619 |
ゴン | +0.612 |
Neg logit contributions | |
policeman | -0.680 |
viol | -0.676 |
Frontier | -0.668 |
trump | -0.644 |
Warsaw | -0.637 |
Token | pin |
Token ID | 6757 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
20439 | +2.175 |
Ire | +2.053 |
=# | +2.017 |
Reply | +1.925 |
ゴン | +1.891 |
Neg logit contributions | |
policeman | -1.228 |
viol | -1.222 |
Frontier | -1.203 |
trump | -1.139 |
Warsaw | -1.125 |
Token | struct |
Token ID | 2878 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
20439 | +1.542 |
Ire | +1.425 |
=# | +1.392 |
Reply | +1.293 |
ゴン | +1.274 |
Neg logit contributions | |
policeman | -1.730 |
viol | -1.726 |
Frontier | -1.704 |
trump | -1.647 |
Warsaw | -1.626 |
Token | Atomic |
Token ID | 28976 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
20439 | +2.545 |
Ire | +2.378 |
=# | +2.317 |
Reply | +2.202 |
ゴン | +2.164 |
Neg logit contributions | |
policeman | -2.057 |
viol | -2.048 |
Frontier | -2.005 |
trump | -1.938 |
Warsaw | -1.902 |
Token | < |
Token ID | 1279 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
20439 | +0.813 |
Ire | +0.759 |
=# | +0.728 |
Reply | +0.686 |
ゴン | +0.683 |
Neg logit contributions | |
policeman | -0.845 |
viol | -0.838 |
Frontier | -0.831 |
trump | -0.802 |
Warsaw | -0.796 |
Token | T |
Token ID | 309 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
20439 | +2.240 |
Ire | +2.113 |
=# | +2.048 |
Reply | +1.960 |
ゴン | +1.944 |
Neg logit contributions | |
policeman | -1.466 |
viol | -1.455 |
Frontier | -1.435 |
trump | -1.371 |
Warsaw | -1.354 |
Token | > |
Token ID | 1875 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
policeman | +0.052 |
viol | +0.052 |
Frontier | +0.051 |
trump | +0.049 |
Warsaw | +0.049 |
Neg logit contributions | |
20439 | -0.057 |
Ire | -0.053 |
=# | -0.051 |
Reply | -0.049 |
ゴン | -0.048 |
Token | { |
Token ID | 1391 |
Feature activation | -5.79e+00 |
Pos logit contributions | |
20439 | +0.675 |
Ire | +0.622 |
=# | +0.592 |
Reply | +0.554 |
ゴン | +0.548 |
Neg logit contributions | |
policeman | -0.935 |
viol | -0.930 |
Frontier | -0.923 |
trump | -0.895 |
Warsaw | -0.888 |
Token | ... |
Token ID | 2644 |
Feature activation | -8.69e-01 |
Pos logit contributions | |
20439 | +4.443 |
Ire | +4.104 |
=# | +4.041 |
Reply | +3.823 |
iatus | +3.755 |
Neg logit contributions | |
viol | -4.226 |
policeman | -4.215 |
Frontier | -4.146 |
trump | -3.984 |
Warsaw | -3.883 |
Token | } |
Token ID | 1782 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
20439 | +0.828 |
Ire | +0.780 |
=# | +0.759 |
Reply | +0.731 |
ゴン | +0.720 |
Neg logit contributions | |
policeman | -0.517 |
viol | -0.514 |
Frontier | -0.505 |
trump | -0.480 |
Warsaw | -0.473 |
Token | impl |
Token ID | 4114 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
20439 | +1.386 |
Ire | +1.309 |
=# | +1.274 |
Reply | +1.218 |
ゴン | +1.206 |
Neg logit contributions | |
policeman | -0.843 |
viol | -0.838 |
Frontier | -0.825 |
trump | -0.786 |
Warsaw | -0.774 |
Token | < |
Token ID | 1279 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
20439 | +1.847 |
Ire | +1.701 |
=# | +1.647 |
Reply | +1.548 |
ゴン | +1.512 |
Neg logit contributions | |
policeman | -2.051 |
viol | -2.048 |
Frontier | -2.011 |
trump | -1.951 |
Warsaw | -1.919 |
Token | T |
Token ID | 309 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
20439 | +0.788 |
Ire | +0.743 |
=# | +0.721 |
Reply | +0.690 |
ゴン | +0.684 |
Neg logit contributions | |
policeman | -0.506 |
viol | -0.502 |
Frontier | -0.495 |
trump | -0.473 |
Warsaw | -0.466 |
Token | Width |
Token ID | 38807 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
policeman | +0.033 |
viol | +0.033 |
Frontier | +0.033 |
trump | +0.031 |
Warsaw | +0.031 |
Neg logit contributions | |
20439 | -0.045 |
Ire | -0.042 |
=# | -0.040 |
ゴン | -0.039 |
Reply | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
20439 | +0.108 |
Ire | +0.101 |
=# | +0.096 |
ゴン | +0.093 |
Reply | +0.092 |
Neg logit contributions | |
policeman | -0.088 |
viol | -0.087 |
Frontier | -0.087 |
trump | -0.083 |
Warsaw | -0.083 |
Token | Height |
Token ID | 23106 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
20439 | +0.592 |
Ire | +0.561 |
=# | +0.541 |
ゴン | +0.519 |
Reply | +0.519 |
Neg logit contributions | |
policeman | -0.345 |
viol | -0.344 |
Frontier | -0.342 |
trump | -0.322 |
Warsaw | -0.321 |
Token | } |
Token ID | 1782 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
policeman | +1.085 |
Frontier | +1.084 |
viol | +1.080 |
trump | +1.032 |
Warsaw | +1.027 |
Neg logit contributions | |
20439 | -1.400 |
Ire | -1.319 |
=# | -1.234 |
ゴン | -1.203 |
Reply | -1.185 |
Token | else |
Token ID | 2073 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
20439 | +0.994 |
Ire | +0.933 |
=# | +0.899 |
Reply | +0.859 |
ゴン | +0.857 |
Neg logit contributions | |
policeman | -0.751 |
viol | -0.746 |
Frontier | -0.738 |
trump | -0.708 |
Warsaw | -0.699 |
Token | { |
Token ID | 1391 |
Feature activation | -5.66e+00 |
Pos logit contributions | |
20439 | +1.300 |
Ire | +1.192 |
=# | +1.137 |
Reply | +1.065 |
ゴン | +1.054 |
Neg logit contributions | |
policeman | -1.795 |
viol | -1.787 |
Frontier | -1.772 |
trump | -1.718 |
Warsaw | -1.702 |
Token | Write |
Token ID | 19430 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
20439 | +4.010 |
Ire | +3.760 |
=# | +3.714 |
Reply | +3.445 |
Instr | +3.312 |
Neg logit contributions | |
policeman | -4.594 |
viol | -4.548 |
Frontier | -4.473 |
trump | -4.323 |
Warsaw | -4.309 |
Token | - |
Token ID | 12 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
20439 | +0.347 |
Ire | +0.326 |
=# | +0.316 |
Reply | +0.301 |
ゴン | +0.299 |
Neg logit contributions | |
policeman | -0.250 |
viol | -0.248 |
Frontier | -0.245 |
trump | -0.234 |
Warsaw | -0.231 |
Token | Warning |
Token ID | 20361 |
Feature activation | +8.26e-01 |
Pos logit contributions | |
20439 | +0.517 |
Ire | +0.485 |
=# | +0.469 |
ゴン | +0.444 |
Reply | +0.443 |
Neg logit contributions | |
policeman | -0.411 |
viol | -0.409 |
Frontier | -0.403 |
trump | -0.386 |
Warsaw | -0.382 |
Token | - |
Token ID | 532 |
Feature activation | -7.64e-02 |
Pos logit contributions | |
policeman | +0.645 |
viol | +0.645 |
Frontier | +0.639 |
trump | +0.616 |
Warsaw | +0.608 |
Neg logit contributions | |
20439 | -0.642 |
Ire | -0.597 |
=# | -0.564 |
ゴン | -0.543 |
Reply | -0.535 |
Token | Message |
Token ID | 12837 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
20439 | +0.059 |
Ire | +0.054 |
=# | +0.052 |
ゴン | +0.049 |
iatus | +0.048 |
Neg logit contributions | |
policeman | -0.060 |
viol | -0.060 |
Frontier | -0.060 |
trump | -0.057 |
Warsaw | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
policeman | +0.387 |
Frontier | +0.383 |
viol | +0.383 |
trump | +0.362 |
Warsaw | +0.357 |
Neg logit contributions | |
20439 | -0.841 |
Ire | -0.796 |
=# | -0.767 |
ゴン | -0.757 |
Reply | -0.735 |
Token | 27 |
Token ID | 1983 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
Frontier | +0.568 |
policeman | +0.567 |
viol | +0.566 |
trump | +0.543 |
Warsaw | +0.539 |
Neg logit contributions | |
20439 | -0.609 |
Ire | -0.568 |
=# | -0.546 |
ゴン | -0.535 |
iatus | -0.521 |
Token | Recommend |
Token ID | 30361 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
policeman | +0.068 |
viol | +0.068 |
Frontier | +0.068 |
trump | +0.065 |
Warsaw | +0.065 |
Neg logit contributions | |
20439 | -0.077 |
Ire | -0.072 |
=# | -0.069 |
ゴン | -0.067 |
iatus | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
20439 | +3.506 |
Ire | +3.325 |
=# | +3.254 |
Reply | +3.136 |
iatus | +3.019 |
Neg logit contributions | |
policeman | -2.101 |
viol | -2.083 |
Frontier | -2.028 |
trump | -1.931 |
Warsaw | -1.907 |
Token |
|
Token ID | 198 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
policeman | +0.716 |
Frontier | +0.713 |
viol | +0.706 |
trump | +0.673 |
Warsaw | +0.661 |
Neg logit contributions | |
20439 | -1.698 |
Ire | -1.604 |
ゴン | -1.544 |
=# | -1.540 |
Instr | -1.491 |
Token | Recommend |
Token ID | 41248 |
Feature activation | -5.66e+00 |
Pos logit contributions | |
policeman | +0.930 |
Frontier | +0.926 |
viol | +0.920 |
trump | +0.886 |
Warsaw | +0.877 |
Neg logit contributions | |
20439 | -0.932 |
Ire | -0.852 |
=# | -0.830 |
ゴン | -0.817 |
iatus | -0.785 |
Token | Share |
Token ID | 8734 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
20439 | +3.948 |
Ire | +3.700 |
=# | +3.522 |
Reply | +3.515 |
DAQ | +3.204 |
Neg logit contributions | |
viol | -4.690 |
policeman | -4.677 |
Frontier | -4.576 |
trump | -4.412 |
Warsaw | -4.408 |
Token | this |
Token ID | 428 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
20439 | +0.292 |
Ire | +0.275 |
=# | +0.265 |
ゴン | +0.260 |
Reply | +0.254 |
Neg logit contributions | |
policeman | -0.166 |
viol | -0.164 |
Frontier | -0.162 |
trump | -0.156 |
Warsaw | -0.152 |
Token | comment |
Token ID | 2912 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
policeman | +0.217 |
viol | +0.214 |
Frontier | +0.210 |
trump | +0.203 |
Warsaw | +0.198 |
Neg logit contributions | |
20439 | -0.329 |
Ire | -0.311 |
=# | -0.299 |
ゴン | -0.289 |
Reply | -0.285 |
Token | on |
Token ID | 319 |
Feature activation | -9.36e-01 |
Pos logit contributions | |
20439 | +1.702 |
Ire | +1.591 |
=# | +1.535 |
Reply | +1.458 |
ゴン | +1.442 |
Neg logit contributions | |
policeman | -1.539 |
viol | -1.529 |
Frontier | -1.511 |
trump | -1.456 |
Warsaw | -1.440 |
Token | Facebook |
Token ID | 3203 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
20439 | +0.967 |
Ire | +0.920 |
=# | +0.886 |
ゴン | +0.867 |
Reply | +0.846 |
Neg logit contributions | |
policeman | -0.495 |
viol | -0.493 |
Frontier | -0.492 |
trump | -0.464 |
Warsaw | -0.458 |
Token | > |
Token ID | 1875 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
20439 | +0.360 |
Ire | +0.337 |
=# | +0.323 |
Reply | +0.307 |
ゴン | +0.306 |
Neg logit contributions | |
policeman | -0.331 |
viol | -0.328 |
Frontier | -0.325 |
trump | -0.313 |
Warsaw | -0.310 |
Token | Atomic |
Token ID | 28976 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
20439 | +0.519 |
Ire | +0.493 |
=# | +0.473 |
Reply | +0.455 |
ゴン | +0.455 |
Neg logit contributions | |
policeman | -0.331 |
viol | -0.326 |
Frontier | -0.325 |
trump | -0.308 |
Warsaw | -0.307 |
Token | < |
Token ID | 1279 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
20439 | +1.422 |
Ire | +1.326 |
=# | +1.275 |
Reply | +1.206 |
ゴン | +1.195 |
Neg logit contributions | |
policeman | -1.442 |
viol | -1.431 |
Frontier | -1.418 |
trump | -1.367 |
Warsaw | -1.355 |
Token | T |
Token ID | 309 |
Feature activation | -8.67e-03 |
Pos logit contributions | |
20439 | +2.069 |
Ire | +1.944 |
=# | +1.896 |
Reply | +1.810 |
ゴン | +1.786 |
Neg logit contributions | |
policeman | -1.384 |
viol | -1.379 |
Frontier | -1.353 |
trump | -1.298 |
Warsaw | -1.280 |
Token | > |
Token ID | 1875 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
20439 | +0.007 |
Ire | +0.006 |
=# | +0.006 |
Reply | +0.006 |
ゴン | +0.006 |
Neg logit contributions | |
policeman | -0.007 |
viol | -0.007 |
Frontier | -0.007 |
trump | -0.006 |
Warsaw | -0.006 |
Token | { |
Token ID | 1391 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
20439 | +0.067 |
Ire | +0.062 |
=# | +0.059 |
Reply | +0.056 |
ゴン | +0.056 |
Neg logit contributions | |
policeman | -0.079 |
viol | -0.078 |
Frontier | -0.078 |
trump | -0.075 |
Warsaw | -0.075 |
Token | pub |
Token ID | 2240 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
20439 | +4.844 |
Ire | +4.565 |
=# | +4.479 |
Reply | +4.245 |
iatus | +4.190 |
Neg logit contributions | |
viol | -3.660 |
policeman | -3.623 |
Frontier | -3.576 |
trump | -3.449 |
Warsaw | -3.353 |
Token | fn |
Token ID | 24714 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
20439 | +0.722 |
Ire | +0.681 |
=# | +0.661 |
Reply | +0.633 |
ゴン | +0.627 |
Neg logit contributions | |
policeman | -0.460 |
viol | -0.456 |
Frontier | -0.449 |
trump | -0.429 |
Warsaw | -0.424 |
Token | load |
Token ID | 3440 |
Feature activation | +2.46e-03 |
Pos logit contributions | |
20439 | +2.344 |
Ire | +2.205 |
=# | +2.177 |
Reply | +2.086 |
ゴン | +2.045 |
Neg logit contributions | |
policeman | -1.066 |
viol | -1.063 |
Frontier | -1.046 |
trump | -0.974 |
Warsaw | -0.961 |
Token | < |
Token ID | 1279 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
policeman | +0.002 |
viol | +0.002 |
Frontier | +0.002 |
trump | +0.002 |
Warsaw | +0.002 |
Neg logit contributions | |
20439 | -0.002 |
Ire | -0.002 |
=# | -0.002 |
Reply | -0.002 |
ゴン | -0.002 |
Token | ' |
Token ID | 705 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
20439 | +1.074 |
Ire | +1.016 |
=# | +0.973 |
ゴン | +0.933 |
Reply | +0.931 |
Neg logit contributions | |
policeman | -0.749 |
viol | -0.741 |
Frontier | -0.733 |
trump | -0.703 |
Warsaw | -0.694 |
Token | data |
Token ID | 7890 |
Feature activation | -9.33e-01 |
Pos logit contributions | |
20439 | +0.029 |
Ire | +0.027 |
=# | +0.026 |
Reply | +0.025 |
ゴン | +0.025 |
Neg logit contributions | |
policeman | -0.027 |
viol | -0.027 |
Frontier | -0.027 |
trump | -0.026 |
Warsaw | -0.026 |
Token | _ |
Token ID | 62 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
20439 | +0.772 |
Ire | +0.722 |
=# | +0.701 |
Reply | +0.664 |
ゴン | +0.656 |
Neg logit contributions | |
policeman | -0.679 |
viol | -0.675 |
Frontier | -0.667 |
trump | -0.642 |
Warsaw | -0.635 |
Token | full |
Token ID | 12853 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
20439 | +1.379 |
Ire | +1.292 |
=# | +1.253 |
Reply | +1.192 |
ゴン | +1.179 |
Neg logit contributions | |
policeman | -1.098 |
viol | -1.095 |
Frontier | -1.079 |
trump | -1.037 |
Warsaw | -1.027 |
Token | v |
Token ID | 410 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
20439 | +0.170 |
Ire | +0.160 |
=# | +0.153 |
Reply | +0.147 |
ゴン | +0.146 |
Neg logit contributions | |
policeman | -0.125 |
viol | -0.124 |
Frontier | -0.122 |
trump | -0.117 |
Warsaw | -0.115 |
Token | ) |
Token ID | 8 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
policeman | +0.008 |
viol | +0.008 |
Frontier | +0.008 |
trump | +0.008 |
Warsaw | +0.008 |
Neg logit contributions | |
20439 | -0.011 |
Ire | -0.011 |
=# | -0.010 |
Reply | -0.010 |
ゴン | -0.010 |
Token | { |
Token ID | 1391 |
Feature activation | -5.60e+00 |
Pos logit contributions | |
20439 | +0.763 |
Ire | +0.703 |
=# | +0.667 |
Reply | +0.629 |
ゴン | +0.623 |
Neg logit contributions | |
policeman | -0.979 |
viol | -0.974 |
Frontier | -0.965 |
trump | -0.934 |
Warsaw | -0.925 |
Token | v |
Token ID | 410 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
20439 | +4.852 |
Ire | +4.592 |
=# | +4.508 |
Reply | +4.266 |
Instr | +4.184 |
Neg logit contributions | |
policeman | -3.612 |
viol | -3.597 |
Frontier | -3.547 |
trump | -3.404 |
Warsaw | -3.399 |
Token | . |
Token ID | 13 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
policeman | +0.694 |
viol | +0.690 |
Frontier | +0.682 |
trump | +0.651 |
Warsaw | +0.643 |
Neg logit contributions | |
20439 | -0.946 |
Ire | -0.883 |
=# | -0.846 |
ゴン | -0.811 |
Reply | -0.807 |
Token | ver |
Token ID | 332 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
20439 | +0.306 |
Ire | +0.289 |
=# | +0.279 |
Reply | +0.267 |
ゴン | +0.266 |
Neg logit contributions | |
policeman | -0.196 |
viol | -0.194 |
Frontier | -0.192 |
trump | -0.183 |
Warsaw | -0.180 |
Token | tex |
Token ID | 16886 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
policeman | +0.529 |
viol | +0.519 |
Frontier | +0.516 |
trump | +0.479 |
Warsaw | +0.466 |
Neg logit contributions | |
20439 | -1.528 |
Ire | -1.457 |
=# | -1.416 |
ゴン | -1.365 |
Reply | -1.363 |
Token | . |
Token ID | 13 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
20439 | +0.158 |
Ire | +0.147 |
=# | +0.141 |
Reply | +0.133 |
ゴン | +0.133 |
Neg logit contributions | |
policeman | -0.153 |
viol | -0.152 |
Frontier | -0.151 |
trump | -0.145 |
Warsaw | -0.143 |
Token | i |
Token ID | 72 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
policeman | +0.107 |
viol | +0.106 |
Frontier | +0.105 |
trump | +0.102 |
Warsaw | +0.100 |
Neg logit contributions | |
20439 | -0.104 |
Ire | -0.095 |
=# | -0.092 |
ゴン | -0.088 |
Reply | -0.086 |
Token | - |
Token ID | 532 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
20439 | +0.629 |
Ire | +0.589 |
=# | +0.561 |
ゴン | +0.535 |
Reply | +0.532 |
Neg logit contributions | |
policeman | -0.584 |
viol | -0.580 |
Frontier | -0.575 |
trump | -0.556 |
Warsaw | -0.548 |
Token | eq |
Token ID | 27363 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
20439 | +1.122 |
Ire | +1.061 |
=# | +1.028 |
Reply | +0.987 |
ゴン | +0.982 |
Neg logit contributions | |
policeman | -0.650 |
viol | -0.643 |
Frontier | -0.634 |
trump | -0.605 |
Warsaw | -0.595 |
Token | 1 |
Token ID | 352 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
20439 | +1.625 |
Ire | +1.521 |
=# | +1.450 |
ゴン | +1.398 |
Reply | +1.392 |
Neg logit contributions | |
policeman | -1.395 |
viol | -1.387 |
Frontier | -1.377 |
trump | -1.325 |
Warsaw | -1.302 |
Token | ) |
Token ID | 8 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
20439 | +0.907 |
Ire | +0.850 |
=# | +0.821 |
Reply | +0.781 |
ゴン | +0.773 |
Neg logit contributions | |
policeman | -0.762 |
viol | -0.757 |
Frontier | -0.748 |
trump | -0.720 |
Warsaw | -0.711 |
Token | { |
Token ID | 1391 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
20439 | +1.250 |
Ire | +1.144 |
=# | +1.092 |
Reply | +1.020 |
ゴン | +1.005 |
Neg logit contributions | |
policeman | -1.798 |
viol | -1.789 |
Frontier | -1.772 |
trump | -1.720 |
Warsaw | -1.705 |
Token | $ |
Token ID | 720 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
20439 | +4.528 |
Ire | +4.260 |
=# | +4.217 |
Reply | +3.971 |
Instr | +3.824 |
Neg logit contributions | |
policeman | -3.979 |
viol | -3.919 |
Frontier | -3.863 |
Warsaw | -3.706 |
trump | -3.703 |
Token | font |
Token ID | 10331 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
20439 | +0.951 |
Ire | +0.882 |
=# | +0.849 |
ゴン | +0.802 |
Reply | +0.802 |
Neg logit contributions | |
policeman | -0.944 |
viol | -0.939 |
Frontier | -0.928 |
trump | -0.897 |
Warsaw | -0.885 |
Token | 1 |
Token ID | 16 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
20439 | +1.169 |
Ire | +1.082 |
=# | +1.034 |
ゴン | +0.981 |
Reply | +0.980 |
Neg logit contributions | |
policeman | -1.153 |
viol | -1.143 |
Frontier | -1.133 |
trump | -1.096 |
Warsaw | -1.082 |
Token | = |
Token ID | 796 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
20439 | +0.554 |
Ire | +0.514 |
=# | +0.488 |
ゴン | +0.465 |
Reply | +0.464 |
Neg logit contributions | |
policeman | -0.568 |
viol | -0.564 |
Frontier | -0.560 |
trump | -0.542 |
Warsaw | -0.535 |
Token | New |
Token ID | 968 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
20439 | +1.978 |
Ire | +1.847 |
=# | +1.783 |
Reply | +1.691 |
ゴン | +1.671 |
Neg logit contributions | |
policeman | -1.847 |
viol | -1.836 |
Frontier | -1.814 |
trump | -1.748 |
Warsaw | -1.730 |
Token | Own |
Token ID | 11744 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
20439 | +0.572 |
Ire | +0.539 |
=# | +0.522 |
Reply | +0.499 |
ゴン | +0.496 |
Neg logit contributions | |
policeman | -0.381 |
viol | -0.377 |
Frontier | -0.373 |
trump | -0.356 |
Warsaw | -0.352 |
Token | ed |
Token ID | 276 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
20439 | +0.246 |
Ire | +0.229 |
=# | +0.220 |
ゴン | +0.207 |
Reply | +0.207 |
Neg logit contributions | |
policeman | -0.252 |
viol | -0.250 |
Frontier | -0.249 |
trump | -0.239 |
Warsaw | -0.238 |
Token | < |
Token ID | 1279 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
20439 | +0.767 |
Ire | +0.716 |
=# | +0.690 |
Reply | +0.655 |
ゴン | +0.647 |
Neg logit contributions | |
policeman | -0.717 |
viol | -0.713 |
Frontier | -0.704 |
trump | -0.680 |
Warsaw | -0.672 |
Token | T |
Token ID | 309 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
20439 | +1.636 |
Ire | +1.542 |
=# | +1.500 |
Reply | +1.433 |
ゴン | +1.417 |
Neg logit contributions | |
policeman | -1.085 |
viol | -1.079 |
Frontier | -1.060 |
trump | -1.016 |
Warsaw | -1.002 |
Token | > |
Token ID | 1875 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
policeman | +0.139 |
viol | +0.138 |
Frontier | +0.137 |
trump | +0.131 |
Warsaw | +0.130 |
Neg logit contributions | |
20439 | -0.152 |
Ire | -0.142 |
=# | -0.137 |
Reply | -0.130 |
ゴン | -0.129 |
Token | { |
Token ID | 1391 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
20439 | +0.010 |
Ire | +0.009 |
=# | +0.009 |
Reply | +0.008 |
ゴン | +0.008 |
Neg logit contributions | |
policeman | -0.008 |
viol | -0.008 |
Frontier | -0.008 |
trump | -0.008 |
Warsaw | -0.008 |
Token | pub |
Token ID | 2240 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
20439 | +4.727 |
Ire | +4.465 |
=# | +4.350 |
Reply | +4.159 |
iatus | +4.056 |
Neg logit contributions | |
viol | -3.592 |
policeman | -3.583 |
Frontier | -3.483 |
trump | -3.368 |
Warsaw | -3.320 |
Token | fn |
Token ID | 24714 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
20439 | +0.264 |
Ire | +0.249 |
=# | +0.240 |
Reply | +0.230 |
ゴン | +0.229 |
Neg logit contributions | |
policeman | -0.183 |
viol | -0.182 |
Frontier | -0.180 |
trump | -0.172 |
Warsaw | -0.170 |
Token | new |
Token ID | 649 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
20439 | +2.031 |
Ire | +1.901 |
=# | +1.873 |
Reply | +1.783 |
ゴン | +1.740 |
Neg logit contributions | |
policeman | -1.325 |
viol | -1.323 |
Frontier | -1.301 |
trump | -1.230 |
Warsaw | -1.216 |
Token | ( |
Token ID | 357 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
policeman | +0.784 |
viol | +0.776 |
Frontier | +0.772 |
trump | +0.738 |
Warsaw | +0.733 |
Neg logit contributions | |
20439 | -1.051 |
Ire | -0.991 |
=# | -0.959 |
ゴン | -0.911 |
Reply | -0.910 |
Token | t |
Token ID | 256 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
20439 | +0.022 |
Ire | +0.021 |
=# | +0.020 |
Reply | +0.019 |
ゴン | +0.019 |
Neg logit contributions | |
policeman | -0.021 |
viol | -0.021 |
Frontier | -0.021 |
trump | -0.020 |
Warsaw | -0.020 |
Token | )" |
Token ID | 16725 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
20439 | +0.672 |
Ire | +0.629 |
=# | +0.602 |
ゴン | +0.576 |
Reply | +0.574 |
Neg logit contributions | |
policeman | -0.540 |
viol | -0.535 |
Frontier | -0.529 |
trump | -0.509 |
Warsaw | -0.504 |
Token | break |
Token ID | 2270 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
20439 | +0.330 |
Ire | +0.309 |
=# | +0.296 |
ゴン | +0.285 |
Reply | +0.283 |
Neg logit contributions | |
policeman | -0.245 |
viol | -0.243 |
Frontier | -0.242 |
trump | -0.232 |
Warsaw | -0.229 |
Token | } |
Token ID | 1782 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
20439 | +2.171 |
Ire | +2.049 |
=# | +1.989 |
Reply | +1.905 |
ゴン | +1.886 |
Neg logit contributions | |
policeman | -1.368 |
viol | -1.357 |
Frontier | -1.337 |
trump | -1.277 |
Warsaw | -1.260 |
Token | } |
Token ID | 1782 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
20439 | +1.345 |
Ire | +1.266 |
=# | +1.226 |
Reply | +1.171 |
ゴン | +1.160 |
Neg logit contributions | |
policeman | -0.963 |
viol | -0.957 |
Frontier | -0.943 |
trump | -0.904 |
Warsaw | -0.893 |
Token | else |
Token ID | 2073 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
20439 | +1.294 |
Ire | +1.216 |
=# | +1.178 |
Reply | +1.123 |
ゴン | +1.112 |
Neg logit contributions | |
policeman | -0.978 |
viol | -0.971 |
Frontier | -0.958 |
trump | -0.919 |
Warsaw | -0.908 |
Token | { |
Token ID | 1391 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
20439 | +1.826 |
Ire | +1.677 |
=# | +1.603 |
Reply | +1.495 |
ゴン | +1.466 |
Neg logit contributions | |
policeman | -2.620 |
viol | -2.606 |
Frontier | -2.578 |
trump | -2.504 |
Warsaw | -2.485 |
Token | $ |
Token ID | 720 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
20439 | +4.717 |
Ire | +4.476 |
=# | +4.424 |
Reply | +4.160 |
Instr | +4.006 |
Neg logit contributions | |
policeman | -3.633 |
viol | -3.589 |
Frontier | -3.521 |
trump | -3.387 |
Warsaw | -3.373 |
Token | SR |
Token ID | 12562 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
20439 | +0.428 |
Ire | +0.401 |
=# | +0.388 |
Reply | +0.370 |
ゴン | +0.370 |
Neg logit contributions | |
policeman | -0.308 |
viol | -0.306 |
Frontier | -0.302 |
trump | -0.290 |
Warsaw | -0.286 |
Token | = |
Token ID | 796 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
policeman | +0.152 |
viol | +0.151 |
Frontier | +0.149 |
trump | +0.144 |
Warsaw | +0.142 |
Neg logit contributions | |
20439 | -0.176 |
Ire | -0.163 |
=# | -0.155 |
ゴン | -0.149 |
Reply | -0.148 |
Token | [ |
Token ID | 685 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
20439 | +1.413 |
Ire | +1.319 |
=# | +1.268 |
Reply | +1.207 |
ゴン | +1.199 |
Neg logit contributions | |
policeman | -1.281 |
viol | -1.274 |
Frontier | -1.260 |
trump | -1.215 |
Warsaw | -1.201 |
Token | System |
Token ID | 11964 |
Feature activation | -9.64e-01 |
Pos logit contributions | |
20439 | +0.667 |
Ire | +0.626 |
=# | +0.605 |
Reply | +0.578 |
ゴン | +0.577 |
Neg logit contributions | |
policeman | -0.468 |
viol | -0.463 |
Frontier | -0.458 |
trump | -0.438 |
Warsaw | -0.432 |
Token | data |
Token ID | 7890 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
20439 | +0.016 |
Ire | +0.015 |
=# | +0.014 |
Reply | +0.013 |
ゴン | +0.013 |
Neg logit contributions | |
policeman | -0.014 |
viol | -0.014 |
Frontier | -0.013 |
trump | -0.013 |
Warsaw | -0.013 |
Token | _ |
Token ID | 62 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
20439 | +0.920 |
Ire | +0.865 |
=# | +0.834 |
ゴン | +0.799 |
Reply | +0.796 |
Neg logit contributions | |
policeman | -0.684 |
viol | -0.679 |
Frontier | -0.671 |
trump | -0.645 |
Warsaw | -0.637 |
Token | full |
Token ID | 12853 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
20439 | +0.724 |
Ire | +0.678 |
=# | +0.663 |
Reply | +0.623 |
ゴン | +0.618 |
Neg logit contributions | |
viol | -0.639 |
policeman | -0.638 |
Frontier | -0.630 |
trump | -0.605 |
Warsaw | -0.600 |
Token | v |
Token ID | 410 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
policeman | +0.086 |
viol | +0.085 |
Frontier | +0.084 |
trump | +0.080 |
Warsaw | +0.078 |
Neg logit contributions | |
20439 | -0.161 |
Ire | -0.152 |
=# | -0.148 |
Reply | -0.142 |
ゴン | -0.141 |
Token | ) |
Token ID | 8 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
policeman | +0.024 |
viol | +0.024 |
Frontier | +0.023 |
trump | +0.023 |
Warsaw | +0.022 |
Neg logit contributions | |
20439 | -0.026 |
Ire | -0.024 |
=# | -0.023 |
Reply | -0.022 |
ゴン | -0.022 |
Token | { |
Token ID | 1391 |
Feature activation | -5.44e+00 |
Pos logit contributions | |
20439 | +0.604 |
Ire | +0.553 |
=# | +0.530 |
Reply | +0.489 |
ゴン | +0.482 |
Neg logit contributions | |
policeman | -0.955 |
viol | -0.948 |
Frontier | -0.940 |
trump | -0.915 |
Warsaw | -0.908 |
Token | // |
Token ID | 3373 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
20439 | +4.998 |
Ire | +4.751 |
=# | +4.646 |
Reply | +4.421 |
ゴン | +4.333 |
Neg logit contributions | |
policeman | -3.248 |
Frontier | -3.160 |
viol | -3.157 |
Warsaw | -3.017 |
trump | -2.995 |
Token | Time |
Token ID | 3862 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
viol | +3.380 |
policeman | +3.313 |
Frontier | +3.311 |
trump | +3.150 |
Warsaw | +3.105 |
Neg logit contributions | |
20439 | -4.751 |
Ire | -4.427 |
=# | -4.191 |
ゴン | -4.037 |
Reply | -4.036 |
Token | component |
Token ID | 7515 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
viol | +0.276 |
policeman | +0.275 |
Frontier | +0.273 |
trump | +0.260 |
Warsaw | +0.257 |
Neg logit contributions | |
20439 | -0.402 |
Ire | -0.378 |
=# | -0.362 |
Reply | -0.347 |
ゴン | -0.346 |
Token | const |
Token ID | 1500 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
viol | +0.197 |
policeman | +0.197 |
Frontier | +0.194 |
trump | +0.186 |
Warsaw | +0.182 |
Neg logit contributions | |
20439 | -0.270 |
Ire | -0.254 |
=# | -0.242 |
Reply | -0.234 |
ゴン | -0.232 |
Token | float |
Token ID | 12178 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
20439 | +2.229 |
Ire | +2.091 |
=# | +2.029 |
Reply | +1.930 |
ゴン | +1.911 |
Neg logit contributions | |
policeman | -1.767 |
viol | -1.754 |
Frontier | -1.730 |
trump | -1.662 |
Warsaw | -1.647 |
Token | Im |
Token ID | 1846 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
20439 | +0.848 |
Ire | +0.794 |
=# | +0.771 |
Reply | +0.739 |
ゴン | +0.736 |
Neg logit contributions | |
policeman | -0.547 |
viol | -0.545 |
Frontier | -0.540 |
trump | -0.514 |
Warsaw | -0.508 |
Token | more |
Token ID | 3549 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
20439 | +0.139 |
Ire | +0.130 |
=# | +0.127 |
Reply | +0.121 |
ゴン | +0.120 |
Neg logit contributions | |
policeman | -0.102 |
viol | -0.101 |
Frontier | -0.101 |
trump | -0.096 |
Warsaw | -0.095 |
Token | n |
Token ID | 77 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
20439 | +2.271 |
Ire | +2.124 |
=# | +2.053 |
Reply | +1.957 |
ゴン | +1.937 |
Neg logit contributions | |
policeman | -1.879 |
viol | -1.868 |
Frontier | -1.847 |
trump | -1.775 |
Warsaw | -1.755 |
Token | are |
Token ID | 389 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
20439 | +1.415 |
Ire | +1.326 |
=# | +1.288 |
Reply | +1.233 |
ゴン | +1.225 |
Neg logit contributions | |
policeman | -0.978 |
viol | -0.975 |
Frontier | -0.966 |
trump | -0.925 |
Warsaw | -0.910 |
Token | made |
Token ID | 925 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
20439 | +0.217 |
Ire | +0.203 |
=# | +0.198 |
Reply | +0.190 |
ゴン | +0.189 |
Neg logit contributions | |
policeman | -0.127 |
viol | -0.127 |
Frontier | -0.126 |
trump | -0.119 |
Warsaw | -0.118 |
Token | up |
Token ID | 510 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
20439 | +0.586 |
Ire | +0.551 |
=# | +0.534 |
Reply | +0.511 |
ゴン | +0.507 |
Neg logit contributions | |
policeman | -0.403 |
viol | -0.400 |
Frontier | -0.395 |
trump | -0.378 |
Warsaw | -0.373 |
Token | of |
Token ID | 286 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
viol | +3.007 |
policeman | +3.002 |
Frontier | +2.968 |
trump | +2.852 |
Warsaw | +2.804 |
Neg logit contributions | |
20439 | -3.692 |
Ire | -3.434 |
=# | -3.317 |
Reply | -3.195 |
ゴン | -3.128 |
Token | a |
Token ID | 257 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
policeman | +0.405 |
viol | +0.405 |
Frontier | +0.400 |
trump | +0.380 |
Warsaw | +0.374 |
Neg logit contributions | |
20439 | -0.661 |
Ire | -0.618 |
=# | -0.601 |
Reply | -0.579 |
ゴン | -0.577 |
Token | mot |
Token ID | 2369 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
policeman | +0.436 |
viol | +0.433 |
Frontier | +0.428 |
trump | +0.406 |
Warsaw | +0.401 |
Neg logit contributions | |
20439 | -0.716 |
Ire | -0.671 |
=# | -0.650 |
Reply | -0.628 |
ゴン | -0.620 |
Token | ley |
Token ID | 1636 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
policeman | +1.047 |
viol | +1.043 |
Frontier | +1.030 |
trump | +1.003 |
Warsaw | +0.989 |
Neg logit contributions | |
20439 | -0.808 |
Ire | -0.746 |
=# | -0.693 |
ゴン | -0.664 |
Reply | -0.663 |
Token | crew |
Token ID | 5462 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
policeman | +0.318 |
viol | +0.316 |
Frontier | +0.313 |
trump | +0.298 |
Warsaw | +0.295 |
Neg logit contributions | |
20439 | -0.442 |
Ire | -0.412 |
=# | -0.397 |
Reply | -0.382 |
ゴン | -0.377 |
Token | a |
Token ID | 257 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
policeman | +0.710 |
viol | +0.701 |
Frontier | +0.691 |
trump | +0.657 |
Warsaw | +0.646 |
Neg logit contributions | |
20439 | -1.279 |
Ire | -1.204 |
=# | -1.161 |
ゴン | -1.123 |
Reply | -1.117 |
Token | Viking |
Token ID | 27921 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
policeman | +0.163 |
viol | +0.161 |
Frontier | +0.158 |
trump | +0.152 |
Warsaw | +0.149 |
Neg logit contributions | |
20439 | -0.269 |
Ire | -0.253 |
=# | -0.244 |
ゴン | -0.234 |
Reply | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
20439 | +0.129 |
Ire | +0.121 |
=# | +0.117 |
Reply | +0.111 |
ゴン | +0.111 |
Neg logit contributions | |
policeman | -0.089 |
viol | -0.089 |
Frontier | -0.087 |
trump | -0.084 |
Warsaw | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
20439 | +0.224 |
Ire | +0.211 |
=# | +0.205 |
ゴン | +0.197 |
Reply | +0.197 |
Neg logit contributions | |
policeman | -0.136 |
Frontier | -0.136 |
viol | -0.136 |
Warsaw | -0.129 |
trump | -0.128 |
Token |
|
Token ID | 198 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
policeman | +0.064 |
Frontier | +0.063 |
viol | +0.063 |
trump | +0.060 |
Warsaw | +0.059 |
Neg logit contributions | |
20439 | -0.140 |
Ire | -0.133 |
=# | -0.129 |
ゴン | -0.127 |
Instr | -0.124 |
Token | Wall |
Token ID | 22401 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
20439 | +0.142 |
Ire | +0.132 |
=# | +0.128 |
ゴン | +0.122 |
Reply | +0.121 |
Neg logit contributions | |
policeman | -0.120 |
viol | -0.119 |
Frontier | -0.118 |
trump | -0.114 |
Warsaw | -0.113 |
Token | ace |
Token ID | 558 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
viol | +1.553 |
policeman | +1.537 |
Frontier | +1.476 |
trump | +1.413 |
Warsaw | +1.405 |
Neg logit contributions | |
20439 | -4.152 |
Ire | -3.943 |
=# | -3.813 |
ゴン | -3.758 |
Reply | -3.693 |
Token | � |
Token ID | 564 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
viol | +0.922 |
policeman | +0.919 |
Frontier | +0.898 |
trump | +0.881 |
Warsaw | +0.845 |
Neg logit contributions | |
20439 | -1.562 |
Ire | -1.462 |
=# | -1.405 |
ゴン | -1.360 |
Reply | -1.360 |
Token | � |
Token ID | 250 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
20439 | +0.108 |
Ire | +0.098 |
=# | +0.094 |
Reply | +0.088 |
ゴン | +0.088 |
Neg logit contributions | |
viol | -0.140 |
policeman | -0.140 |
Frontier | -0.138 |
trump | -0.134 |
Warsaw | -0.132 |
Token | Out |
Token ID | 7975 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
20439 | +0.430 |
Ire | +0.403 |
=# | +0.390 |
Reply | +0.374 |
ゴン | +0.372 |
Neg logit contributions | |
viol | -0.308 |
policeman | -0.308 |
Frontier | -0.304 |
trump | -0.291 |
Warsaw | -0.288 |
Token | � |
Token ID | 447 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
policeman | +1.237 |
viol | +1.230 |
Frontier | +1.226 |
trump | +1.173 |
Warsaw | +1.157 |
Neg logit contributions | |
20439 | -1.654 |
Ire | -1.546 |
=# | -1.478 |
ゴン | -1.428 |
Reply | -1.411 |
Token | involving |
Token ID | 7411 |
Feature activation | +8.97e-01 |
Pos logit contributions | |
policeman | +0.383 |
viol | +0.381 |
Frontier | +0.379 |
trump | +0.363 |
Warsaw | +0.357 |
Neg logit contributions | |
20439 | -0.531 |
Ire | -0.499 |
=# | -0.484 |
Reply | -0.462 |
ゴン | -0.462 |
Token | the |
Token ID | 262 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
policeman | +0.535 |
viol | +0.534 |
Frontier | +0.528 |
trump | +0.506 |
Warsaw | +0.495 |
Neg logit contributions | |
20439 | -0.869 |
Ire | -0.818 |
=# | -0.796 |
Reply | -0.766 |
ゴン | -0.765 |
Token | biggest |
Token ID | 4094 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
20439 | +0.229 |
Ire | +0.214 |
=# | +0.208 |
Reply | +0.200 |
ゴン | +0.199 |
Neg logit contributions | |
policeman | -0.151 |
viol | -0.150 |
Frontier | -0.148 |
trump | -0.142 |
Warsaw | -0.140 |
Token | alleged |
Token ID | 4260 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
policeman | +0.330 |
viol | +0.330 |
Frontier | +0.321 |
trump | +0.312 |
Warsaw | +0.304 |
Neg logit contributions | |
20439 | -0.475 |
Ire | -0.441 |
=# | -0.427 |
Reply | -0.414 |
ゴン | -0.412 |
Token | bribe |
Token ID | 40879 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
viol | +0.015 |
policeman | +0.015 |
Frontier | +0.015 |
trump | +0.014 |
Warsaw | +0.014 |
Neg logit contributions | |
20439 | -0.022 |
Ire | -0.020 |
=# | -0.020 |
ゴン | -0.019 |
Reply | -0.019 |
Token | yet |
Token ID | 1865 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
policeman | +0.320 |
viol | +0.320 |
Frontier | +0.313 |
trump | +0.304 |
Warsaw | +0.296 |
Neg logit contributions | |
20439 | -0.445 |
Ire | -0.421 |
=# | -0.407 |
ゴン | -0.387 |
Reply | -0.386 |
Token | in |
Token ID | 287 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
viol | +2.718 |
policeman | +2.706 |
Frontier | +2.642 |
trump | +2.623 |
Warsaw | +2.455 |
Neg logit contributions | |
20439 | -4.602 |
Ire | -4.447 |
=# | -4.362 |
ゴン | -4.198 |
Reply | -4.117 |
Token | Australia |
Token ID | 4505 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
policeman | +0.753 |
viol | +0.748 |
Frontier | +0.743 |
trump | +0.712 |
Warsaw | +0.701 |
Neg logit contributions | |
20439 | -1.259 |
Ire | -1.187 |
=# | -1.161 |
Reply | -1.115 |
ゴン | -1.111 |
Token | � |
Token ID | 447 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
20439 | +0.191 |
Ire | +0.180 |
=# | +0.175 |
Reply | +0.167 |
ゴン | +0.165 |
Neg logit contributions | |
policeman | -0.130 |
viol | -0.129 |
Frontier | -0.127 |
trump | -0.122 |
Warsaw | -0.120 |
Token | � |
Token ID | 247 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
20439 | +0.163 |
Ire | +0.150 |
=# | +0.144 |
ゴン | +0.136 |
Reply | +0.135 |
Neg logit contributions | |
policeman | -0.186 |
viol | -0.185 |
Frontier | -0.184 |
trump | -0.179 |
Warsaw | -0.176 |
Token | s |
Token ID | 82 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
policeman | +0.212 |
viol | +0.210 |
Frontier | +0.208 |
trump | +0.199 |
Warsaw | +0.195 |
Neg logit contributions | |
20439 | -0.318 |
Ire | -0.300 |
=# | -0.292 |
Reply | -0.279 |
ゴン | -0.278 |
Token | CEO |
Token ID | 6123 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
policeman | +0.236 |
viol | +0.235 |
Frontier | +0.231 |
trump | +0.222 |
Warsaw | +0.218 |
Neg logit contributions | |
20439 | -0.359 |
Ire | -0.337 |
=# | -0.324 |
ゴン | -0.314 |
Reply | -0.312 |
Token | Richard |
Token ID | 6219 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
policeman | +0.347 |
viol | +0.345 |
Frontier | +0.340 |
trump | +0.326 |
Warsaw | +0.322 |
Neg logit contributions | |
20439 | -0.480 |
Ire | -0.451 |
=# | -0.437 |
Reply | -0.417 |
ゴン | -0.413 |
Token | M |
Token ID | 337 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
20439 | +0.641 |
Ire | +0.601 |
=# | +0.579 |
ゴン | +0.553 |
Reply | +0.553 |
Neg logit contributions | |
policeman | -0.479 |
viol | -0.475 |
Frontier | -0.474 |
trump | -0.453 |
Warsaw | -0.449 |
Token | oller |
Token ID | 49252 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
policeman | +0.160 |
viol | +0.158 |
Frontier | +0.157 |
trump | +0.150 |
Warsaw | +0.148 |
Neg logit contributions | |
20439 | -0.238 |
Ire | -0.224 |
=# | -0.217 |
Reply | -0.208 |
ゴン | -0.207 |
Token | would |
Token ID | 561 |
Feature activation | -9.88e-01 |
Pos logit contributions | |
20439 | +0.226 |
Ire | +0.212 |
=# | +0.206 |
Reply | +0.197 |
ゴン | +0.196 |
Neg logit contributions | |
policeman | -0.153 |
viol | -0.152 |
Frontier | -0.150 |
trump | -0.145 |
Warsaw | -0.142 |
Token | like |
Token ID | 588 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
20439 | +0.952 |
Ire | +0.900 |
=# | +0.871 |
Reply | +0.835 |
ゴン | +0.831 |
Neg logit contributions | |
policeman | -0.589 |
viol | -0.585 |
Frontier | -0.575 |
trump | -0.552 |
Warsaw | -0.542 |
Token | to |
Token ID | 284 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
policeman | +1.869 |
viol | +1.859 |
Frontier | +1.838 |
trump | +1.760 |
Warsaw | +1.717 |
Neg logit contributions | |
20439 | -3.173 |
Ire | -3.008 |
=# | -2.895 |
ゴン | -2.804 |
Reply | -2.781 |
Token | see |
Token ID | 766 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
policeman | +0.150 |
viol | +0.149 |
Frontier | +0.147 |
trump | +0.142 |
Warsaw | +0.138 |
Neg logit contributions | |
20439 | -0.230 |
Ire | -0.215 |
=# | -0.207 |
ゴン | -0.200 |
Reply | -0.199 |
Token | one |
Token ID | 530 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
20439 | +0.767 |
Ire | +0.724 |
=# | +0.702 |
Reply | +0.674 |
ゴン | +0.670 |
Neg logit contributions | |
policeman | -0.466 |
viol | -0.463 |
Frontier | -0.457 |
trump | -0.436 |
Warsaw | -0.429 |
Token | or |
Token ID | 393 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
20439 | +0.054 |
Ire | +0.051 |
=# | +0.049 |
Reply | +0.047 |
ゴン | +0.046 |
Neg logit contributions | |
policeman | -0.040 |
viol | -0.039 |
Frontier | -0.039 |
trump | -0.037 |
Warsaw | -0.037 |
Token | two |
Token ID | 734 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
20439 | +0.234 |
Ire | +0.223 |
=# | +0.216 |
Reply | +0.208 |
ゴン | +0.207 |
Neg logit contributions | |
policeman | -0.111 |
viol | -0.110 |
Frontier | -0.108 |
trump | -0.103 |
Warsaw | -0.101 |
Token | and |
Token ID | 290 |
Feature activation | +6.78e-01 |
Pos logit contributions | |
viol | +0.424 |
Frontier | +0.423 |
policeman | +0.423 |
trump | +0.400 |
Warsaw | +0.394 |
Neg logit contributions | |
20439 | -0.690 |
Ire | -0.649 |
=# | -0.629 |
ゴン | -0.605 |
Reply | -0.603 |
Token | Koch |
Token ID | 17009 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
viol | +0.402 |
policeman | +0.401 |
Frontier | +0.400 |
trump | +0.379 |
Warsaw | +0.374 |
Neg logit contributions | |
20439 | -0.659 |
Ire | -0.620 |
=# | -0.600 |
ゴン | -0.579 |
Reply | -0.576 |
Token | Industries |
Token ID | 20171 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
viol | +0.231 |
Frontier | +0.231 |
policeman | +0.230 |
trump | +0.223 |
Warsaw | +0.220 |
Neg logit contributions | |
20439 | -0.163 |
Ire | -0.147 |
=# | -0.140 |
ゴン | -0.133 |
Reply | -0.130 |
Token | , |
Token ID | 11 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
20439 | +0.345 |
Ire | +0.326 |
=# | +0.316 |
ゴン | +0.303 |
Reply | +0.302 |
Neg logit contributions | |
viol | -0.204 |
policeman | -0.203 |
Frontier | -0.200 |
trump | -0.191 |
Warsaw | -0.187 |
Token | among |
Token ID | 1871 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
viol | +0.240 |
policeman | +0.238 |
Frontier | +0.235 |
trump | +0.226 |
Warsaw | +0.220 |
Neg logit contributions | |
20439 | -0.402 |
Ire | -0.380 |
=# | -0.369 |
Reply | -0.353 |
ゴン | -0.353 |
Token | other |
Token ID | 584 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
viol | +0.339 |
policeman | +0.337 |
Frontier | +0.329 |
trump | +0.307 |
Warsaw | +0.294 |
Neg logit contributions | |
20439 | -1.125 |
Ire | -1.075 |
=# | -1.044 |
Reply | -1.012 |
ゴン | -1.011 |
Token | companies |
Token ID | 2706 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
viol | +1.982 |
Frontier | +1.918 |
policeman | +1.890 |
trump | +1.842 |
Warsaw | +1.776 |
Neg logit contributions | |
20439 | -3.008 |
Ire | -2.818 |
ゴン | -2.702 |
=# | -2.699 |
Reply | -2.613 |
Token | that |
Token ID | 326 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
viol | +0.771 |
policeman | +0.760 |
Frontier | +0.755 |
trump | +0.731 |
Warsaw | +0.707 |
Neg logit contributions | |
20439 | -1.207 |
Ire | -1.142 |
=# | -1.104 |
ゴン | -1.067 |
Reply | -1.053 |
Token | he |
Token ID | 339 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
viol | +0.373 |
policeman | +0.370 |
Frontier | +0.365 |
trump | +0.354 |
Warsaw | +0.343 |
Neg logit contributions | |
20439 | -0.641 |
Ire | -0.605 |
=# | -0.585 |
ゴン | -0.566 |
Reply | -0.560 |
Token | says |
Token ID | 1139 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
viol | +0.575 |
policeman | +0.572 |
Frontier | +0.560 |
trump | +0.544 |
Warsaw | +0.521 |
Neg logit contributions | |
20439 | -0.998 |
Ire | -0.946 |
=# | -0.911 |
ゴン | -0.877 |
Reply | -0.876 |
Token | have |
Token ID | 423 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
viol | +1.017 |
policeman | +0.999 |
Frontier | +0.984 |
trump | +0.959 |
Warsaw | +0.928 |
Neg logit contributions | |
20439 | -1.619 |
Ire | -1.535 |
=# | -1.473 |
ゴン | -1.429 |
Reply | -1.421 |
Token | , |
Token ID | 11 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
viol | +0.632 |
policeman | +0.632 |
Frontier | +0.620 |
trump | +0.599 |
Warsaw | +0.590 |
Neg logit contributions | |
20439 | -0.828 |
Ire | -0.777 |
=# | -0.746 |
Reply | -0.715 |
ゴン | -0.710 |
Token | Mont |
Token ID | 5575 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
20439 | +0.179 |
Ire | +0.168 |
=# | +0.163 |
ゴン | +0.156 |
Reply | +0.156 |
Neg logit contributions | |
viol | -0.113 |
policeman | -0.113 |
Frontier | -0.111 |
trump | -0.107 |
Warsaw | -0.105 |
Token | pel |
Token ID | 30242 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
20439 | +0.636 |
Ire | +0.592 |
=# | +0.570 |
ゴン | +0.547 |
Reply | +0.543 |
Neg logit contributions | |
policeman | -0.550 |
viol | -0.546 |
Frontier | -0.539 |
trump | -0.521 |
Warsaw | -0.515 |
Token | ier |
Token ID | 959 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
policeman | +0.096 |
viol | +0.094 |
Frontier | +0.093 |
trump | +0.085 |
Warsaw | +0.082 |
Neg logit contributions | |
20439 | -0.359 |
Ire | -0.340 |
=# | -0.332 |
ゴン | -0.324 |
Reply | -0.322 |
Token | is |
Token ID | 318 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
20439 | +0.496 |
Ire | +0.467 |
=# | +0.452 |
Reply | +0.434 |
ゴン | +0.431 |
Neg logit contributions | |
policeman | -0.314 |
viol | -0.312 |
Frontier | -0.307 |
trump | -0.294 |
Warsaw | -0.289 |
Token | supposedly |
Token ID | 13519 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
viol | +0.095 |
policeman | +0.095 |
Frontier | +0.093 |
trump | +0.090 |
Warsaw | +0.088 |
Neg logit contributions | |
20439 | -0.164 |
Ire | -0.155 |
=# | -0.150 |
ゴン | -0.145 |
Reply | -0.144 |
Token | the |
Token ID | 262 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
viol | +0.309 |
policeman | +0.309 |
Frontier | +0.302 |
trump | +0.293 |
Warsaw | +0.285 |
Neg logit contributions | |
20439 | -0.528 |
Ire | -0.497 |
=# | -0.481 |
ゴン | -0.465 |
Reply | -0.463 |
Token | � |
Token ID | 564 |
Feature activation | +7.20e-02 |
Pos logit contributions | |
20439 | +0.963 |
Ire | +0.908 |
=# | +0.881 |
Reply | +0.847 |
ゴン | +0.845 |
Neg logit contributions | |
policeman | -0.529 |
viol | -0.525 |
Frontier | -0.512 |
trump | -0.492 |
Warsaw | -0.482 |
Token | � |
Token ID | 250 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
policeman | +0.059 |
viol | +0.059 |
Frontier | +0.058 |
trump | +0.056 |
Warsaw | +0.056 |
Neg logit contributions | |
20439 | -0.053 |
Ire | -0.049 |
=# | -0.047 |
Reply | -0.045 |
ゴン | -0.044 |
Token | heart |
Token ID | 11499 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
policeman | +0.111 |
viol | +0.110 |
Frontier | +0.108 |
trump | +0.104 |
Warsaw | +0.103 |
Neg logit contributions | |
20439 | -0.155 |
Ire | -0.145 |
=# | -0.141 |
ゴン | -0.134 |
Reply | -0.134 |
Token | � |
Token ID | 447 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
policeman | +0.007 |
viol | +0.007 |
Frontier | +0.007 |
trump | +0.007 |
Warsaw | +0.007 |
Neg logit contributions | |
20439 | -0.009 |
Ire | -0.008 |
=# | -0.008 |
Reply | -0.007 |
ゴン | -0.007 |
Token | ' |
Token ID | 6 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
20439 | +0.613 |
Ire | +0.570 |
=# | +0.554 |
ゴン | +0.525 |
Reply | +0.525 |
Neg logit contributions | |
policeman | -0.504 |
viol | -0.500 |
Frontier | -0.495 |
trump | -0.476 |
Warsaw | -0.471 |
Token | We |
Token ID | 1135 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
Frontier | +1.050 |
policeman | +1.046 |
viol | +1.035 |
trump | +0.996 |
Warsaw | +0.990 |
Neg logit contributions | |
20439 | -1.373 |
Ire | -1.248 |
=# | -1.233 |
ゴン | -1.207 |
Instr | -1.176 |
Token | will |
Token ID | 481 |
Feature activation | -5.95e-03 |
Pos logit contributions | |
20439 | +0.757 |
Ire | +0.712 |
=# | +0.690 |
ゴン | +0.665 |
Reply | +0.663 |
Neg logit contributions | |
policeman | -0.465 |
viol | -0.460 |
Frontier | -0.454 |
trump | -0.437 |
Warsaw | -0.428 |
Token | close |
Token ID | 1969 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
20439 | +0.006 |
Ire | +0.005 |
=# | +0.005 |
ゴン | +0.005 |
Reply | +0.005 |
Neg logit contributions | |
policeman | -0.004 |
viol | -0.004 |
Frontier | -0.003 |
trump | -0.003 |
Warsaw | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
policeman | +1.191 |
Frontier | +1.181 |
viol | +1.172 |
trump | +1.123 |
Warsaw | +1.106 |
Neg logit contributions | |
20439 | -2.049 |
Ire | -1.901 |
=# | -1.854 |
ゴン | -1.798 |
Reply | -1.798 |
Token | wall |
Token ID | 3355 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
20439 | +0.825 |
Ire | +0.775 |
=# | +0.754 |
Reply | +0.720 |
ゴン | +0.719 |
Neg logit contributions | |
policeman | -0.511 |
viol | -0.506 |
Frontier | -0.504 |
trump | -0.480 |
Warsaw | -0.472 |
Token | up |
Token ID | 510 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
policeman | +0.216 |
viol | +0.214 |
Frontier | +0.212 |
trump | +0.203 |
Warsaw | +0.200 |
Neg logit contributions | |
20439 | -0.315 |
Ire | -0.298 |
=# | -0.287 |
ゴン | -0.275 |
Reply | -0.274 |
Token | with |
Token ID | 351 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
policeman | +0.732 |
viol | +0.722 |
Frontier | +0.719 |
trump | +0.684 |
Warsaw | +0.678 |
Neg logit contributions | |
20439 | -1.133 |
Ire | -1.072 |
=# | -1.034 |
ゴン | -0.992 |
Reply | -0.986 |
Token | the |
Token ID | 262 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
20439 | +1.012 |
Ire | +0.952 |
=# | +0.925 |
Reply | +0.887 |
ゴン | +0.881 |
Neg logit contributions | |
policeman | -0.627 |
viol | -0.622 |
Frontier | -0.614 |
trump | -0.587 |
Warsaw | -0.578 |
Token | English |
Token ID | 3594 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
policeman | +0.297 |
viol | +0.293 |
Frontier | +0.290 |
trump | +0.281 |
Warsaw | +0.274 |
Neg logit contributions | |
20439 | -0.429 |
Ire | -0.399 |
=# | -0.391 |
ゴン | -0.373 |
Reply | -0.372 |
Token | dead |
Token ID | 2636 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
policeman | +0.682 |
viol | +0.671 |
Frontier | +0.670 |
trump | +0.639 |
Warsaw | +0.623 |
Neg logit contributions | |
20439 | -1.120 |
Ire | -1.052 |
=# | -1.025 |
Reply | -0.981 |
ゴン | -0.981 |
Token | for |
Token ID | 329 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
20439 | +0.108 |
Ire | +0.102 |
=# | +0.098 |
ゴン | +0.094 |
Reply | +0.093 |
Neg logit contributions | |
policeman | -0.076 |
viol | -0.076 |
Frontier | -0.075 |
trump | -0.072 |
Warsaw | -0.071 |
Token | free |
Token ID | 1479 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
20439 | +0.489 |
Ire | +0.463 |
=# | +0.449 |
Reply | +0.432 |
ゴン | +0.431 |
Neg logit contributions | |
policeman | -0.259 |
viol | -0.257 |
Frontier | -0.253 |
trump | -0.241 |
Warsaw | -0.236 |
Token | . |
Token ID | 13 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
20439 | +0.239 |
Ire | +0.225 |
=# | +0.219 |
Reply | +0.208 |
ゴン | +0.207 |
Neg logit contributions | |
policeman | -0.170 |
viol | -0.169 |
Frontier | -0.167 |
trump | -0.161 |
Warsaw | -0.158 |
Token | " |
Token ID | 366 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
20439 | +0.315 |
Ire | +0.295 |
=# | +0.285 |
ゴン | +0.274 |
Reply | +0.273 |
Neg logit contributions | |
policeman | -0.211 |
viol | -0.210 |
Frontier | -0.210 |
trump | -0.199 |
Warsaw | -0.198 |
Token | However |
Token ID | 4864 |
Feature activation | -8.78e-01 |
Pos logit contributions | |
20439 | +0.274 |
Ire | +0.254 |
=# | +0.246 |
ゴン | +0.234 |
Reply | +0.233 |
Neg logit contributions | |
policeman | -0.220 |
viol | -0.219 |
Frontier | -0.217 |
trump | -0.207 |
Warsaw | -0.206 |
Token | at |
Token ID | 379 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
20439 | +0.783 |
Ire | +0.735 |
=# | +0.712 |
Reply | +0.679 |
ゴン | +0.673 |
Neg logit contributions | |
policeman | -0.585 |
viol | -0.581 |
Frontier | -0.573 |
trump | -0.550 |
Warsaw | -0.543 |
Token | the |
Token ID | 262 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
policeman | +0.079 |
viol | +0.079 |
Frontier | +0.078 |
trump | +0.074 |
Warsaw | +0.073 |
Neg logit contributions | |
20439 | -0.139 |
Ire | -0.132 |
=# | -0.128 |
Reply | -0.123 |
ゴン | -0.122 |
Token | same |
Token ID | 976 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
policeman | +0.122 |
viol | +0.121 |
Frontier | +0.119 |
trump | +0.114 |
Warsaw | +0.112 |
Neg logit contributions | |
20439 | -0.190 |
Ire | -0.179 |
=# | -0.173 |
Reply | -0.166 |
ゴン | -0.165 |
Token | time |
Token ID | 640 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
policeman | +0.719 |
viol | +0.701 |
Frontier | +0.695 |
trump | +0.658 |
Warsaw | +0.645 |
Neg logit contributions | |
20439 | -1.527 |
Ire | -1.440 |
=# | -1.392 |
ゴン | -1.356 |
Reply | -1.351 |
Token | you |
Token ID | 345 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
policeman | +0.314 |
viol | +0.313 |
Frontier | +0.309 |
trump | +0.294 |
Warsaw | +0.291 |
Neg logit contributions | |
20439 | -0.481 |
Ire | -0.452 |
=# | -0.435 |
Reply | -0.420 |
ゴン | -0.419 |
Token | don |
Token ID | 836 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
policeman | +0.088 |
viol | +0.087 |
Frontier | +0.085 |
trump | +0.083 |
Warsaw | +0.081 |
Neg logit contributions | |
20439 | -0.131 |
Ire | -0.124 |
=# | -0.119 |
ゴン | -0.115 |
Reply | -0.114 |
Token |
|
Token ID | 198 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
policeman | +0.337 |
viol | +0.334 |
Frontier | +0.333 |
trump | +0.314 |
Warsaw | +0.310 |
Neg logit contributions | |
20439 | -0.719 |
Ire | -0.682 |
=# | -0.657 |
ゴン | -0.645 |
Reply | -0.632 |
Token | The |
Token ID | 464 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
viol | +0.489 |
policeman | +0.488 |
Frontier | +0.486 |
trump | +0.462 |
Warsaw | +0.457 |
Neg logit contributions | |
20439 | -0.599 |
Ire | -0.556 |
=# | -0.536 |
ゴン | -0.518 |
iatus | -0.506 |
Token | two |
Token ID | 734 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
policeman | +0.298 |
viol | +0.297 |
Frontier | +0.296 |
trump | +0.280 |
Warsaw | +0.274 |
Neg logit contributions | |
20439 | -0.496 |
Ire | -0.469 |
=# | -0.450 |
ゴン | -0.436 |
Reply | -0.431 |
Token | adults |
Token ID | 6490 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
policeman | +0.239 |
viol | +0.239 |
Frontier | +0.234 |
trump | +0.223 |
Warsaw | +0.218 |
Neg logit contributions | |
20439 | -0.388 |
Ire | -0.370 |
=# | -0.356 |
Reply | -0.343 |
ゴン | -0.343 |
Token | will |
Token ID | 481 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
viol | +0.607 |
policeman | +0.597 |
Frontier | +0.588 |
trump | +0.575 |
Warsaw | +0.558 |
Neg logit contributions | |
20439 | -0.939 |
Ire | -0.889 |
=# | -0.853 |
ゴン | -0.842 |
Reply | -0.826 |
Token | also |
Token ID | 635 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
viol | +0.658 |
policeman | +0.654 |
Frontier | +0.641 |
trump | +0.623 |
Warsaw | +0.601 |
Neg logit contributions | |
20439 | -1.206 |
Ire | -1.148 |
=# | -1.101 |
ゴン | -1.076 |
Reply | -1.065 |
Token | need |
Token ID | 761 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
policeman | +0.358 |
viol | +0.357 |
Frontier | +0.350 |
trump | +0.339 |
Warsaw | +0.330 |
Neg logit contributions | |
20439 | -0.613 |
Ire | -0.580 |
=# | -0.560 |
ゴン | -0.540 |
Reply | -0.539 |
Token | to |
Token ID | 284 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
policeman | +0.642 |
viol | +0.638 |
Frontier | +0.629 |
trump | +0.601 |
Warsaw | +0.593 |
Neg logit contributions | |
20439 | -0.964 |
Ire | -0.909 |
=# | -0.881 |
Reply | -0.843 |
ゴン | -0.836 |
Token | authorize |
Token ID | 29145 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
policeman | +0.264 |
viol | +0.263 |
Frontier | +0.257 |
trump | +0.249 |
Warsaw | +0.242 |
Neg logit contributions | |
20439 | -0.455 |
Ire | -0.431 |
=# | -0.414 |
ゴン | -0.401 |
Reply | -0.399 |
Token | each |
Token ID | 1123 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
policeman | +0.187 |
viol | +0.186 |
Frontier | +0.185 |
trump | +0.174 |
Warsaw | +0.171 |
Neg logit contributions | |
20439 | -0.327 |
Ire | -0.312 |
=# | -0.301 |
ゴン | -0.290 |
Reply | -0.290 |
Token | other |
Token ID | 584 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
policeman | +0.440 |
viol | +0.435 |
Frontier | +0.428 |
trump | +0.395 |
Warsaw | +0.386 |
Neg logit contributions | |
20439 | -1.126 |
Ire | -1.085 |
=# | -1.052 |
Reply | -1.015 |
ゴン | -1.012 |
Token | , |
Token ID | 11 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
20439 | +0.331 |
Ire | +0.309 |
=# | +0.299 |
Reply | +0.284 |
ゴン | +0.281 |
Neg logit contributions | |
policeman | -0.280 |
viol | -0.277 |
Frontier | -0.274 |
trump | -0.263 |
Warsaw | -0.260 |
Token | could |
Token ID | 714 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
20439 | +0.499 |
Ire | +0.469 |
=# | +0.453 |
Reply | +0.435 |
ゴン | +0.433 |
Neg logit contributions | |
policeman | -0.345 |
viol | -0.344 |
Frontier | -0.338 |
trump | -0.327 |
Warsaw | -0.320 |
Token | even |
Token ID | 772 |
Feature activation | +8.25e-01 |
Pos logit contributions | |
20439 | +0.419 |
Ire | +0.397 |
=# | +0.385 |
Reply | +0.371 |
ゴン | +0.368 |
Neg logit contributions | |
policeman | -0.218 |
viol | -0.217 |
Frontier | -0.213 |
trump | -0.203 |
Warsaw | -0.199 |
Token | be |
Token ID | 307 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
policeman | +0.456 |
viol | +0.452 |
Frontier | +0.442 |
trump | +0.433 |
Warsaw | +0.415 |
Neg logit contributions | |
20439 | -0.842 |
Ire | -0.792 |
=# | -0.764 |
Reply | -0.741 |
ゴン | -0.740 |
Token | used |
Token ID | 973 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
policeman | +0.011 |
viol | +0.011 |
Frontier | +0.011 |
trump | +0.010 |
Warsaw | +0.010 |
Neg logit contributions | |
20439 | -0.019 |
Ire | -0.018 |
=# | -0.017 |
Reply | -0.016 |
ゴン | -0.016 |
Token | as |
Token ID | 355 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
20439 | +0.218 |
Ire | +0.206 |
=# | +0.199 |
Reply | +0.191 |
ゴン | +0.190 |
Neg logit contributions | |
policeman | -0.138 |
viol | -0.137 |
Frontier | -0.135 |
trump | -0.129 |
Warsaw | -0.127 |
Token | part |
Token ID | 636 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
policeman | +0.280 |
viol | +0.278 |
Frontier | +0.272 |
trump | +0.262 |
Warsaw | +0.256 |
Neg logit contributions | |
20439 | -0.469 |
Ire | -0.440 |
=# | -0.426 |
ゴン | -0.412 |
Reply | -0.411 |
Token | of |
Token ID | 286 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
20439 | +0.101 |
Ire | +0.094 |
=# | +0.090 |
ゴン | +0.087 |
Reply | +0.086 |
Neg logit contributions | |
policeman | -0.088 |
viol | -0.087 |
Frontier | -0.086 |
trump | -0.084 |
Warsaw | -0.081 |
Token | a |
Token ID | 257 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
20439 | +0.958 |
Ire | +0.905 |
=# | +0.877 |
Reply | +0.842 |
ゴン | +0.836 |
Neg logit contributions | |
policeman | -0.565 |
viol | -0.560 |
Frontier | -0.553 |
trump | -0.527 |
Warsaw | -0.520 |
Token | universal |
Token ID | 10112 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
policeman | +0.030 |
viol | +0.029 |
Frontier | +0.029 |
trump | +0.028 |
Warsaw | +0.028 |
Neg logit contributions | |
20439 | -0.051 |
Ire | -0.048 |
=# | -0.046 |
ゴン | -0.045 |
Reply | -0.045 |
Token | basic |
Token ID | 4096 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
policeman | +0.047 |
viol | +0.046 |
Frontier | +0.045 |
trump | +0.043 |
Warsaw | +0.043 |
Neg logit contributions | |
20439 | -0.073 |
Ire | -0.068 |
=# | -0.066 |
ゴン | -0.064 |
Reply | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
20439 | +1.499 |
Ire | +1.412 |
=# | +1.371 |
Reply | +1.309 |
ゴン | +1.295 |
Neg logit contributions | |
policeman | -1.061 |
viol | -1.052 |
Frontier | -1.038 |
trump | -0.994 |
Warsaw | -0.983 |
Token | max |
Token ID | 3509 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
policeman | +0.092 |
viol | +0.092 |
Frontier | +0.091 |
trump | +0.086 |
Warsaw | +0.085 |
Neg logit contributions | |
20439 | -0.140 |
Ire | -0.131 |
=# | -0.127 |
ゴン | -0.121 |
Reply | -0.121 |
Token | gain |
Token ID | 4461 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
20439 | +0.475 |
Ire | +0.447 |
=# | +0.432 |
Reply | +0.413 |
ゴン | +0.410 |
Neg logit contributions | |
policeman | -0.327 |
viol | -0.326 |
Frontier | -0.321 |
trump | -0.308 |
Warsaw | -0.303 |
Token | ] |
Token ID | 60 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
20439 | +0.497 |
Ire | +0.470 |
=# | +0.449 |
Reply | +0.432 |
ゴン | +0.430 |
Neg logit contributions | |
policeman | -0.341 |
viol | -0.340 |
Frontier | -0.335 |
trump | -0.322 |
Warsaw | -0.315 |
Token | to |
Token ID | 284 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
viol | +0.777 |
policeman | +0.771 |
Frontier | +0.758 |
trump | +0.734 |
Warsaw | +0.714 |
Neg logit contributions | |
20439 | -1.144 |
Ire | -1.089 |
=# | -1.038 |
Reply | -1.000 |
ゴン | -0.998 |
Token | be |
Token ID | 307 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
20439 | +0.229 |
Ire | +0.216 |
=# | +0.209 |
Reply | +0.201 |
ゴン | +0.200 |
Neg logit contributions | |
policeman | -0.135 |
viol | -0.135 |
Frontier | -0.133 |
trump | -0.127 |
Warsaw | -0.125 |
Token | independent |
Token ID | 4795 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
20439 | +1.027 |
Ire | +0.972 |
=# | +0.943 |
Reply | +0.906 |
ゴン | +0.899 |
Neg logit contributions | |
policeman | -0.571 |
viol | -0.567 |
Frontier | -0.558 |
trump | -0.531 |
Warsaw | -0.522 |
Token | of |
Token ID | 286 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
20439 | +0.382 |
Ire | +0.358 |
=# | +0.344 |
Reply | +0.327 |
ゴン | +0.325 |
Neg logit contributions | |
viol | -0.339 |
policeman | -0.338 |
Frontier | -0.333 |
trump | -0.322 |
Warsaw | -0.316 |
Token | the |
Token ID | 262 |
Feature activation | +7.75e-01 |
Pos logit contributions | |
policeman | +0.329 |
viol | +0.329 |
Frontier | +0.324 |
trump | +0.309 |
Warsaw | +0.303 |
Neg logit contributions | |
20439 | -0.565 |
Ire | -0.532 |
=# | -0.515 |
Reply | -0.492 |
ゴン | -0.492 |
Token | queue |
Token ID | 16834 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
viol | +0.450 |
policeman | +0.450 |
Frontier | +0.442 |
trump | +0.423 |
Warsaw | +0.415 |
Neg logit contributions | |
20439 | -0.770 |
Ire | -0.723 |
=# | -0.699 |
ゴン | -0.669 |
Reply | -0.667 |
Token | growth |
Token ID | 3349 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
20439 | +0.851 |
Ire | +0.805 |
=# | +0.778 |
Reply | +0.746 |
ゴン | +0.742 |
Neg logit contributions | |
policeman | -0.515 |
viol | -0.511 |
Frontier | -0.504 |
trump | -0.481 |
Warsaw | -0.472 |
Token | of |
Token ID | 286 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
20439 | +0.112 |
Ire | +0.106 |
=# | +0.102 |
ゴン | +0.098 |
Reply | +0.098 |
Neg logit contributions | |
policeman | -0.075 |
viol | -0.074 |
Frontier | -0.073 |
trump | -0.070 |
Warsaw | -0.069 |
Token | our |
Token ID | 674 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
policeman | +0.449 |
viol | +0.447 |
Frontier | +0.436 |
trump | +0.422 |
Warsaw | +0.412 |
Neg logit contributions | |
20439 | -0.672 |
Ire | -0.632 |
=# | -0.607 |
Reply | -0.584 |
ゴン | -0.583 |
Token | own |
Token ID | 898 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
policeman | +0.126 |
viol | +0.124 |
Frontier | +0.121 |
trump | +0.116 |
Warsaw | +0.114 |
Neg logit contributions | |
20439 | -0.265 |
Ire | -0.252 |
=# | -0.244 |
Reply | -0.236 |
ゴン | -0.235 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
20439 | +0.981 |
Ire | +0.922 |
=# | +0.892 |
Reply | +0.851 |
ゴン | +0.848 |
Neg logit contributions | |
policeman | -0.732 |
viol | -0.730 |
Frontier | -0.719 |
trump | -0.692 |
Warsaw | -0.682 |
Token |
|
Token ID | 198 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
20439 | +1.358 |
Ire | +1.274 |
=# | +1.236 |
ゴン | +1.183 |
Reply | +1.179 |
Neg logit contributions | |
policeman | -0.940 |
viol | -0.937 |
Frontier | -0.934 |
trump | -0.886 |
Warsaw | -0.885 |
Token |
|
Token ID | 198 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
20439 | +0.509 |
Ire | +0.482 |
=# | +0.468 |
ゴン | +0.457 |
iatus | +0.448 |
Neg logit contributions | |
policeman | -0.239 |
viol | -0.237 |
Frontier | -0.237 |
trump | -0.222 |
Warsaw | -0.221 |
Token | The |
Token ID | 464 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
20439 | +0.496 |
Ire | +0.460 |
=# | +0.446 |
ゴン | +0.425 |
Reply | +0.419 |
Neg logit contributions | |
policeman | -0.424 |
viol | -0.423 |
Frontier | -0.417 |
trump | -0.402 |
Warsaw | -0.397 |
Token | Ford |
Token ID | 8092 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
policeman | +0.018 |
viol | +0.018 |
Frontier | +0.017 |
trump | +0.017 |
Warsaw | +0.016 |
Neg logit contributions | |
20439 | -0.027 |
Ire | -0.025 |
=# | -0.024 |
ゴン | -0.023 |
Reply | -0.023 |
Token | saga |
Token ID | 26784 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
policeman | +0.021 |
viol | +0.020 |
Frontier | +0.020 |
trump | +0.019 |
Warsaw | +0.019 |
Neg logit contributions | |
20439 | -0.028 |
Ire | -0.027 |
=# | -0.026 |
ゴン | -0.025 |
Reply | -0.024 |
Token | began |
Token ID | 2540 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
policeman | +0.442 |
viol | +0.440 |
Frontier | +0.434 |
trump | +0.419 |
Warsaw | +0.407 |
Neg logit contributions | |
20439 | -0.717 |
Ire | -0.675 |
=# | -0.652 |
ゴン | -0.626 |
Reply | -0.625 |
Token | in |
Token ID | 287 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
20439 | +0.287 |
Ire | +0.269 |
=# | +0.260 |
ゴン | +0.249 |
Reply | +0.248 |
Neg logit contributions | |
policeman | -0.204 |
viol | -0.203 |
Frontier | -0.200 |
trump | -0.193 |
Warsaw | -0.191 |
Token |
|
Token ID | 198 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
policeman | +0.735 |
Frontier | +0.732 |
viol | +0.725 |
trump | +0.689 |
Warsaw | +0.681 |
Neg logit contributions | |
20439 | -1.680 |
Ire | -1.592 |
=# | -1.529 |
ゴン | -1.520 |
Instr | -1.478 |
Token | 1995 |
Token ID | 21908 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
policeman | +1.005 |
viol | +0.996 |
Frontier | +0.986 |
trump | +0.959 |
Warsaw | +0.945 |
Neg logit contributions | |
20439 | -0.680 |
Ire | -0.620 |
=# | -0.592 |
ゴン | -0.545 |
Reply | -0.545 |
Token | Garry |
Token ID | 45354 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
Frontier | +1.319 |
policeman | +1.317 |
viol | +1.311 |
trump | +1.248 |
Warsaw | +1.235 |
Neg logit contributions | |
20439 | -1.809 |
Ire | -1.688 |
=# | -1.627 |
ゴン | -1.561 |
Reply | -1.546 |
Token | Sh |
Token ID | 911 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
Frontier | +0.697 |
policeman | +0.689 |
viol | +0.677 |
trump | +0.663 |
Warsaw | +0.662 |
Neg logit contributions | |
20439 | -0.863 |
Ire | -0.804 |
=# | -0.764 |
ゴン | -0.761 |
iatus | -0.728 |
Token | and |
Token ID | 392 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
policeman | +0.381 |
viol | +0.376 |
Frontier | +0.374 |
trump | +0.360 |
Warsaw | +0.351 |
Neg logit contributions | |
20439 | -0.551 |
Ire | -0.521 |
=# | -0.500 |
ゴン | -0.486 |
Reply | -0.478 |
Token | ling |
Token ID | 1359 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
policeman | +0.279 |
viol | +0.276 |
Frontier | +0.270 |
trump | +0.260 |
Warsaw | +0.254 |
Neg logit contributions | |
20439 | -0.480 |
Ire | -0.459 |
=# | -0.442 |
ゴン | -0.433 |
Reply | -0.424 |
Token | 1996 |
Token ID | 8235 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
20439 | +0.639 |
Ire | +0.600 |
=# | +0.581 |
Reply | +0.553 |
ゴン | +0.548 |
Neg logit contributions | |
policeman | -0.507 |
viol | -0.504 |
Frontier | -0.498 |
trump | -0.478 |
Warsaw | -0.473 |
Token | John |
Token ID | 1757 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
policeman | +1.058 |
Frontier | +1.055 |
viol | +1.051 |
trump | +1.003 |
Warsaw | +0.992 |
Neg logit contributions | |
20439 | -1.343 |
Ire | -1.259 |
=# | -1.211 |
Reply | -1.153 |
ゴン | -1.151 |
Token | Lith |
Token ID | 19730 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
policeman | +1.153 |
Frontier | +1.145 |
viol | +1.137 |
Warsaw | +1.084 |
trump | +1.082 |
Neg logit contributions | |
20439 | -1.781 |
Ire | -1.664 |
=# | -1.619 |
ゴン | -1.548 |
Reply | -1.546 |
Token | gow |
Token ID | 21175 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
policeman | +0.183 |
Frontier | +0.179 |
viol | +0.178 |
trump | +0.156 |
Warsaw | +0.153 |
Neg logit contributions | |
20439 | -1.053 |
Ire | -1.012 |
=# | -0.980 |
Reply | -0.961 |
ゴン | -0.957 |
Token |
|
Token ID | 198 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
policeman | +0.064 |
viol | +0.063 |
Frontier | +0.062 |
trump | +0.059 |
Warsaw | +0.058 |
Neg logit contributions | |
20439 | -0.114 |
Ire | -0.108 |
=# | -0.105 |
Reply | -0.101 |
ゴン | -0.100 |
Token | is |
Token ID | 318 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
viol | +0.236 |
policeman | +0.235 |
Frontier | +0.233 |
trump | +0.222 |
Warsaw | +0.219 |
Neg logit contributions | |
20439 | -0.362 |
Ire | -0.343 |
=# | -0.329 |
Reply | -0.318 |
ゴン | -0.317 |
Token | right |
Token ID | 826 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
20439 | +0.593 |
Ire | +0.560 |
=# | +0.542 |
Reply | +0.520 |
ゴン | +0.518 |
Neg logit contributions | |
policeman | -0.378 |
viol | -0.376 |
Frontier | -0.372 |
trump | -0.354 |
Warsaw | -0.350 |
Token | on |
Token ID | 319 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
20439 | +2.624 |
Ire | +2.478 |
=# | +2.407 |
Reply | +2.305 |
ゴン | +2.282 |
Neg logit contributions | |
policeman | -1.653 |
viol | -1.638 |
Frontier | -1.614 |
trump | -1.542 |
Warsaw | -1.521 |
Token | the |
Token ID | 262 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
viol | +0.293 |
policeman | +0.291 |
Frontier | +0.290 |
trump | +0.275 |
Warsaw | +0.271 |
Neg logit contributions | |
20439 | -0.529 |
Ire | -0.498 |
=# | -0.481 |
Reply | -0.460 |
ゴン | -0.460 |
Token | B |
Token ID | 347 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
viol | +0.793 |
policeman | +0.792 |
Frontier | +0.783 |
trump | +0.743 |
Warsaw | +0.733 |
Neg logit contributions | |
20439 | -1.345 |
Ire | -1.261 |
=# | -1.216 |
Reply | -1.170 |
ゴン | -1.168 |
Token | DP |
Token ID | 6322 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
policeman | +1.153 |
viol | +1.145 |
Frontier | +1.136 |
trump | +1.096 |
Warsaw | +1.082 |
Neg logit contributions | |
20439 | -1.225 |
Ire | -1.145 |
=# | -1.096 |
Reply | -1.041 |
ゴン | -1.028 |
Token | threshold |
Token ID | 11387 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
20439 | +0.303 |
Ire | +0.286 |
=# | +0.276 |
Reply | +0.265 |
ゴン | +0.264 |
Neg logit contributions | |
viol | -0.185 |
policeman | -0.184 |
Frontier | -0.182 |
trump | -0.173 |
Warsaw | -0.170 |
Token | ( |
Token ID | 357 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
viol | +0.862 |
policeman | +0.856 |
Frontier | +0.846 |
trump | +0.814 |
Warsaw | +0.794 |
Neg logit contributions | |
20439 | -1.238 |
Ire | -1.178 |
=# | -1.117 |
Reply | -1.079 |
ゴン | -1.079 |
Token | blue |
Token ID | 17585 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
20439 | +0.482 |
Ire | +0.451 |
=# | +0.436 |
Reply | +0.415 |
ゴン | +0.413 |
Neg logit contributions | |
policeman | -0.379 |
viol | -0.377 |
Frontier | -0.373 |
trump | -0.357 |
Warsaw | -0.354 |
Token | dot |
Token ID | 16605 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
policeman | +0.084 |
viol | +0.084 |
Frontier | +0.084 |
trump | +0.079 |
Warsaw | +0.078 |
Neg logit contributions | |
20439 | -0.147 |
Ire | -0.138 |
=# | -0.132 |
Reply | -0.129 |
ゴン | -0.128 |
Token | above |
Token ID | 2029 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
20439 | +0.555 |
Ire | +0.524 |
=# | +0.496 |
Reply | +0.481 |
ゴン | +0.479 |
Neg logit contributions | |
viol | -0.429 |
policeman | -0.428 |
Frontier | -0.425 |
trump | -0.404 |
Warsaw | -0.400 |
Token | remove |
Token ID | 4781 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
20439 | +0.125 |
Ire | +0.119 |
=# | +0.115 |
Reply | +0.111 |
ゴン | +0.110 |
Neg logit contributions | |
policeman | -0.067 |
viol | -0.066 |
Frontier | -0.065 |
trump | -0.062 |
Warsaw | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
policeman | +0.116 |
viol | +0.115 |
Frontier | +0.114 |
trump | +0.108 |
Warsaw | +0.107 |
Neg logit contributions | |
20439 | -0.195 |
Ire | -0.184 |
=# | -0.178 |
Reply | -0.171 |
ゴン | -0.170 |
Token | MS |
Token ID | 5653 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
20439 | +0.618 |
Ire | +0.581 |
=# | +0.564 |
Reply | +0.539 |
ゴン | +0.534 |
Neg logit contributions | |
policeman | -0.422 |
viol | -0.419 |
Frontier | -0.416 |
trump | -0.396 |
Warsaw | -0.392 |
Token | I |
Token ID | 40 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
policeman | +2.153 |
viol | +2.125 |
trump | +2.085 |
Frontier | +2.085 |
Warsaw | +1.937 |
Neg logit contributions | |
20439 | -3.230 |
=# | -2.897 |
Ire | -2.883 |
ゴン | -2.846 |
actionDate | -2.655 |
Token | exclusive |
Token ID | 8568 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
policeman | +0.652 |
viol | +0.648 |
Frontier | +0.646 |
trump | +0.619 |
Warsaw | +0.609 |
Neg logit contributions | |
20439 | -0.824 |
Ire | -0.769 |
=# | -0.742 |
Reply | -0.711 |
ゴン | -0.702 |
Token | Un |
Token ID | 791 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
20439 | +0.162 |
Ire | +0.152 |
=# | +0.147 |
Reply | +0.141 |
ゴン | +0.140 |
Neg logit contributions | |
policeman | -0.116 |
viol | -0.115 |
Frontier | -0.114 |
trump | -0.109 |
Warsaw | -0.107 |
Token | locked |
Token ID | 24162 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
20439 | +1.127 |
Ire | +1.043 |
=# | +1.004 |
Reply | +0.947 |
ゴン | +0.936 |
Neg logit contributions | |
policeman | -1.247 |
viol | -1.238 |
Frontier | -1.227 |
trump | -1.185 |
Warsaw | -1.174 |
Token | Digital |
Token ID | 10231 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
20439 | +1.057 |
Ire | +0.963 |
=# | +0.916 |
Reply | +0.852 |
ゴン | +0.838 |
Neg logit contributions | |
policeman | -1.673 |
viol | -1.666 |
Frontier | -1.648 |
trump | -1.603 |
Warsaw | -1.589 |
Token | Power |
Token ID | 13434 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
20439 | +1.040 |
Ire | +0.974 |
=# | +0.937 |
Reply | +0.897 |
ゴン | +0.883 |
Neg logit contributions | |
policeman | -0.923 |
viol | -0.921 |
Frontier | -0.902 |
trump | -0.874 |
Warsaw | -0.861 |
Token | is |
Token ID | 318 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
Frontier | +1.118 |
policeman | +1.112 |
viol | +1.094 |
trump | +1.060 |
Warsaw | +1.042 |
Neg logit contributions | |
20439 | -1.512 |
Ire | -1.416 |
=# | -1.359 |
Reply | -1.303 |
ゴン | -1.300 |
Token | composed |
Token ID | 13160 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
20439 | +0.930 |
Ire | +0.880 |
=# | +0.856 |
Reply | +0.821 |
ゴン | +0.814 |
Neg logit contributions | |
policeman | -0.507 |
viol | -0.503 |
Frontier | -0.494 |
trump | -0.470 |
Warsaw | -0.463 |
Token | / |
Token ID | 14 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
20439 | +1.372 |
Ire | +1.285 |
=# | +1.242 |
Reply | +1.182 |
ゴン | +1.175 |
Neg logit contributions | |
policeman | -1.101 |
viol | -1.094 |
Frontier | -1.081 |
trump | -1.039 |
Warsaw | -1.028 |
Token | Getty |
Token ID | 6633 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
20439 | +0.572 |
Ire | +0.538 |
=# | +0.521 |
ゴン | +0.503 |
Reply | +0.497 |
Neg logit contributions | |
policeman | -0.355 |
viol | -0.353 |
Frontier | -0.352 |
trump | -0.335 |
Warsaw | -0.331 |
Token | Images |
Token ID | 5382 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
20439 | +1.012 |
Ire | +0.914 |
=# | +0.862 |
ゴン | +0.830 |
Reply | +0.805 |
Neg logit contributions | |
policeman | -1.498 |
Frontier | -1.492 |
viol | -1.490 |
trump | -1.443 |
Warsaw | -1.439 |
Token | ) |
Token ID | 8 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
20439 | +2.642 |
Ire | +2.508 |
=# | +2.403 |
Reply | +2.284 |
ゴン | +2.228 |
Neg logit contributions | |
policeman | -2.597 |
viol | -2.571 |
Frontier | -2.534 |
trump | -2.461 |
Warsaw | -2.417 |
Token | Story |
Token ID | 8362 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
20439 | +0.805 |
Ire | +0.757 |
=# | +0.734 |
Reply | +0.700 |
ゴン | +0.698 |
Neg logit contributions | |
policeman | -0.586 |
viol | -0.580 |
Frontier | -0.575 |
trump | -0.550 |
Warsaw | -0.545 |
Token | Highlights |
Token ID | 30776 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
20439 | +0.213 |
Ire | +0.202 |
=# | +0.196 |
Reply | +0.188 |
ゴン | +0.188 |
Neg logit contributions | |
policeman | -0.106 |
viol | -0.104 |
Frontier | -0.103 |
trump | -0.097 |
Warsaw | -0.096 |
Token | The |
Token ID | 383 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
20439 | +2.947 |
Ire | +2.775 |
=# | +2.692 |
Reply | +2.566 |
ゴン | +2.538 |
Neg logit contributions | |
policeman | -2.115 |
viol | -2.101 |
Frontier | -2.068 |
trump | -1.984 |
Warsaw | -1.956 |
Token | United |
Token ID | 1578 |
Feature activation | +7.95e-03 |
Pos logit contributions | |
policeman | +0.218 |
viol | +0.215 |
Frontier | +0.215 |
trump | +0.206 |
Warsaw | +0.204 |
Neg logit contributions | |
20439 | -0.319 |
Ire | -0.298 |
=# | -0.285 |
ゴン | -0.278 |
Reply | -0.274 |
Token | States |
Token ID | 1829 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
policeman | +0.007 |
viol | +0.007 |
Frontier | +0.007 |
trump | +0.006 |
Warsaw | +0.006 |
Neg logit contributions | |
20439 | -0.006 |
Ire | -0.005 |
=# | -0.005 |
Reply | -0.005 |
ゴン | -0.005 |
Token | will |
Token ID | 481 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
policeman | +0.361 |
viol | +0.354 |
Frontier | +0.352 |
trump | +0.336 |
Warsaw | +0.329 |
Neg logit contributions | |
20439 | -0.612 |
Ire | -0.575 |
=# | -0.551 |
ゴン | -0.538 |
Reply | -0.536 |
Token | run |
Token ID | 1057 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
policeman | +0.261 |
viol | +0.258 |
Frontier | +0.254 |
trump | +0.247 |
Warsaw | +0.239 |
Neg logit contributions | |
20439 | -0.422 |
Ire | -0.398 |
=# | -0.380 |
ゴン | -0.371 |
Reply | -0.369 |
Token | 21 |
Token ID | 2481 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
20439 | +0.715 |
Ire | +0.658 |
=# | +0.631 |
Reply | +0.592 |
ゴン | +0.583 |
Neg logit contributions | |
policeman | -0.925 |
viol | -0.919 |
Frontier | -0.910 |
trump | -0.882 |
Warsaw | -0.874 |
Token | ] |
Token ID | 60 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
viol | +0.230 |
policeman | +0.230 |
Frontier | +0.229 |
trump | +0.219 |
Warsaw | +0.214 |
Neg logit contributions | |
20439 | -0.399 |
Ire | -0.379 |
=# | -0.363 |
ゴン | -0.355 |
Reply | -0.349 |
Token | In |
Token ID | 554 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
Frontier | +0.623 |
policeman | +0.609 |
viol | +0.608 |
Warsaw | +0.592 |
trump | +0.579 |
Neg logit contributions | |
20439 | -0.949 |
Ire | -0.888 |
=# | -0.856 |
ゴン | -0.835 |
Reply | -0.819 |
Token | the |
Token ID | 262 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
20439 | +0.524 |
Ire | +0.493 |
=# | +0.478 |
ゴン | +0.458 |
Reply | +0.451 |
Neg logit contributions | |
policeman | -0.323 |
viol | -0.320 |
Frontier | -0.316 |
trump | -0.303 |
Warsaw | -0.300 |
Token | same |
Token ID | 976 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
20439 | +0.358 |
Ire | +0.335 |
=# | +0.325 |
ゴン | +0.311 |
Reply | +0.307 |
Neg logit contributions | |
policeman | -0.248 |
viol | -0.247 |
Frontier | -0.243 |
trump | -0.234 |
Warsaw | -0.233 |
Token | way |
Token ID | 835 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
policeman | +0.013 |
viol | +0.013 |
Frontier | +0.012 |
trump | +0.012 |
Warsaw | +0.012 |
Neg logit contributions | |
20439 | -0.021 |
Ire | -0.019 |
=# | -0.019 |
ゴン | -0.018 |
Reply | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
20439 | +3.191 |
Ire | +3.001 |
=# | +2.916 |
Reply | +2.785 |
ゴン | +2.738 |
Neg logit contributions | |
policeman | -2.502 |
viol | -2.476 |
Frontier | -2.455 |
trump | -2.340 |
Warsaw | -2.309 |
Token | Christianity |
Token ID | 13624 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
20439 | +1.170 |
Ire | +1.099 |
=# | +1.061 |
Reply | +1.013 |
ゴン | +1.011 |
Neg logit contributions | |
policeman | -0.822 |
viol | -0.820 |
Frontier | -0.804 |
trump | -0.776 |
Warsaw | -0.766 |
Token | would |
Token ID | 561 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
policeman | +0.147 |
viol | +0.147 |
Frontier | +0.143 |
trump | +0.139 |
Warsaw | +0.135 |
Neg logit contributions | |
20439 | -0.235 |
Ire | -0.222 |
=# | -0.216 |
Reply | -0.205 |
ゴン | -0.205 |
Token | later |
Token ID | 1568 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
20439 | +1.251 |
Ire | +1.180 |
=# | +1.145 |
Reply | +1.096 |
ゴン | +1.087 |
Neg logit contributions | |
policeman | -0.800 |
viol | -0.794 |
Frontier | -0.782 |
trump | -0.748 |
Warsaw | -0.737 |
Token | on |
Token ID | 319 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
20439 | +0.418 |
Ire | +0.393 |
=# | +0.380 |
Reply | +0.363 |
ゴン | +0.363 |
Neg logit contributions | |
policeman | -0.283 |
viol | -0.280 |
Frontier | -0.275 |
trump | -0.267 |
Warsaw | -0.260 |
Token | . |
Token ID | 13 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
viol | +0.585 |
policeman | +0.583 |
Frontier | +0.574 |
trump | +0.553 |
Warsaw | +0.542 |
Neg logit contributions | |
20439 | -0.843 |
Ire | -0.801 |
=# | -0.763 |
ゴン | -0.733 |
Reply | -0.731 |
Token | Thus |
Token ID | 6660 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
viol | +0.186 |
policeman | +0.185 |
Frontier | +0.183 |
trump | +0.175 |
Warsaw | +0.173 |
Neg logit contributions | |
20439 | -0.275 |
Ire | -0.261 |
=# | -0.249 |
ゴン | -0.241 |
Reply | -0.239 |
Token | to |
Token ID | 284 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
20439 | +1.224 |
Ire | +1.158 |
=# | +1.116 |
Reply | +1.069 |
ゴン | +1.063 |
Neg logit contributions | |
policeman | -0.847 |
viol | -0.844 |
Frontier | -0.830 |
trump | -0.795 |
Warsaw | -0.784 |
Token | include |
Token ID | 2291 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
policeman | +0.280 |
viol | +0.279 |
Frontier | +0.273 |
trump | +0.264 |
Warsaw | +0.257 |
Neg logit contributions | |
20439 | -0.477 |
Ire | -0.452 |
=# | -0.432 |
ゴン | -0.419 |
Reply | -0.416 |
Token | "\ |
Token ID | 37082 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
policeman | +0.230 |
viol | +0.229 |
Frontier | +0.225 |
trump | +0.215 |
Warsaw | +0.212 |
Neg logit contributions | |
20439 | -0.394 |
Ire | -0.375 |
=# | -0.361 |
Reply | -0.348 |
ゴン | -0.347 |
Token | :" |
Token ID | 11097 |
Feature activation | -5.11e+00 |
Pos logit contributions | |
policeman | +0.542 |
viol | +0.536 |
Frontier | +0.525 |
trump | +0.494 |
Warsaw | +0.484 |
Neg logit contributions | |
20439 | -1.333 |
Ire | -1.272 |
=# | -1.234 |
Reply | -1.192 |
ゴン | -1.185 |
Token | you |
Token ID | 345 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
20439 | +4.491 |
=# | +4.149 |
Ire | +4.119 |
Reply | +3.883 |
iatus | +3.778 |
Neg logit contributions | |
policeman | -3.384 |
viol | -3.319 |
Frontier | -3.303 |
Warsaw | -3.143 |
trump | -3.133 |
Token | have |
Token ID | 423 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
20439 | +1.481 |
Ire | +1.395 |
=# | +1.362 |
Reply | +1.299 |
ゴン | +1.280 |
Neg logit contributions | |
policeman | -0.959 |
viol | -0.952 |
Frontier | -0.939 |
trump | -0.895 |
Warsaw | -0.887 |
Token | to |
Token ID | 284 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
20439 | +0.241 |
Ire | +0.226 |
=# | +0.220 |
Reply | +0.210 |
ゴン | +0.208 |
Neg logit contributions | |
policeman | -0.166 |
viol | -0.165 |
Frontier | -0.163 |
trump | -0.156 |
Warsaw | -0.154 |
Token | specify |
Token ID | 11986 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
policeman | +0.231 |
viol | +0.230 |
Frontier | +0.225 |
trump | +0.219 |
Warsaw | +0.212 |
Neg logit contributions | |
20439 | -0.377 |
Ire | -0.355 |
=# | -0.338 |
ゴン | -0.329 |
Reply | -0.327 |
Token | " |
Token ID | 366 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
policeman | +0.422 |
viol | +0.422 |
Frontier | +0.415 |
trump | +0.399 |
Warsaw | +0.391 |
Neg logit contributions | |
20439 | -0.629 |
Ire | -0.596 |
=# | -0.569 |
ゴン | -0.551 |
Reply | -0.548 |
Token | and |
Token ID | 290 |
Feature activation | +9.24e-02 |
Pos logit contributions | |
20439 | +1.527 |
Ire | +1.440 |
=# | +1.397 |
Reply | +1.337 |
ゴン | +1.324 |
Neg logit contributions | |
policeman | -0.997 |
viol | -0.989 |
Frontier | -0.975 |
trump | -0.932 |
Warsaw | -0.920 |
Token | Sony |
Token ID | 10184 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
policeman | +0.059 |
viol | +0.059 |
Frontier | +0.058 |
trump | +0.055 |
Warsaw | +0.055 |
Neg logit contributions | |
20439 | -0.085 |
Ire | -0.081 |
=# | -0.078 |
Reply | -0.074 |
ゴン | -0.074 |
Token | 's |
Token ID | 338 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
policeman | +0.269 |
viol | +0.267 |
Frontier | +0.266 |
trump | +0.254 |
Warsaw | +0.249 |
Neg logit contributions | |
20439 | -0.408 |
Ire | -0.382 |
=# | -0.371 |
Reply | -0.355 |
ゴン | -0.354 |
Token | next |
Token ID | 1306 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
20439 | +1.613 |
Ire | +1.519 |
=# | +1.475 |
Reply | +1.411 |
ゴン | +1.395 |
Neg logit contributions | |
policeman | -1.085 |
viol | -1.077 |
Frontier | -1.060 |
trump | -1.014 |
Warsaw | -1.002 |
Token | - |
Token ID | 12 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
20439 | +0.253 |
Ire | +0.237 |
=# | +0.230 |
Reply | +0.219 |
ゴン | +0.217 |
Neg logit contributions | |
policeman | -0.189 |
viol | -0.187 |
Frontier | -0.184 |
trump | -0.177 |
Warsaw | -0.175 |
Token | generation |
Token ID | 20158 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
20439 | +0.686 |
Ire | +0.636 |
=# | +0.608 |
Reply | +0.569 |
ゴン | +0.563 |
Neg logit contributions | |
policeman | -0.883 |
viol | -0.878 |
Frontier | -0.870 |
trump | -0.843 |
Warsaw | -0.835 |
Token | PlayStation |
Token ID | 14047 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
20439 | +3.057 |
Ire | +2.870 |
=# | +2.791 |
Reply | +2.657 |
ゴン | +2.624 |
Neg logit contributions | |
policeman | -2.298 |
viol | -2.277 |
Frontier | -2.239 |
trump | -2.150 |
Warsaw | -2.131 |
Token | is |
Token ID | 318 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
policeman | +0.146 |
viol | +0.146 |
Frontier | +0.146 |
trump | +0.139 |
Warsaw | +0.136 |
Neg logit contributions | |
20439 | -0.213 |
Ire | -0.199 |
=# | -0.192 |
Reply | -0.185 |
ゴン | -0.184 |
Token | dated |
Token ID | 14567 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
20439 | +0.769 |
Ire | +0.726 |
=# | +0.704 |
Reply | +0.677 |
ゴン | +0.674 |
Neg logit contributions | |
policeman | -0.448 |
viol | -0.447 |
Frontier | -0.441 |
trump | -0.424 |
Warsaw | -0.413 |
Token | for |
Token ID | 329 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
policeman | +0.100 |
viol | +0.100 |
Frontier | +0.099 |
trump | +0.095 |
Warsaw | +0.095 |
Neg logit contributions | |
20439 | -0.129 |
Ire | -0.121 |
=# | -0.116 |
Reply | -0.111 |
ゴン | -0.111 |
Token | fall |
Token ID | 2121 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
policeman | +0.041 |
viol | +0.041 |
Frontier | +0.040 |
trump | +0.039 |
Warsaw | +0.038 |
Neg logit contributions | |
20439 | -0.053 |
Ire | -0.049 |
=# | -0.047 |
ゴン | -0.045 |
Reply | -0.045 |
Token | since |
Token ID | 1201 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
viol | +2.145 |
policeman | +2.133 |
Frontier | +2.098 |
trump | +2.064 |
Warsaw | +1.971 |
Neg logit contributions | |
20439 | -2.924 |
Ire | -2.746 |
=# | -2.572 |
ゴン | -2.540 |
Reply | -2.485 |
Token | we |
Token ID | 356 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
viol | +0.445 |
policeman | +0.445 |
Frontier | +0.439 |
trump | +0.422 |
Warsaw | +0.411 |
Neg logit contributions | |
20439 | -0.773 |
Ire | -0.729 |
=# | -0.706 |
ゴン | -0.677 |
Reply | -0.675 |
Token | started |
Token ID | 2067 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
policeman | +0.453 |
viol | +0.451 |
Frontier | +0.441 |
trump | +0.428 |
Warsaw | +0.416 |
Neg logit contributions | |
20439 | -0.796 |
Ire | -0.750 |
=# | -0.721 |
ゴン | -0.698 |
Reply | -0.696 |
Token | searching |
Token ID | 10342 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
viol | +0.523 |
policeman | +0.518 |
Frontier | +0.508 |
trump | +0.494 |
Warsaw | +0.480 |
Neg logit contributions | |
20439 | -0.981 |
Ire | -0.925 |
=# | -0.886 |
ゴン | -0.862 |
Reply | -0.857 |
Token | check |
Token ID | 2198 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
20439 | +0.627 |
Ire | +0.590 |
=# | +0.567 |
ゴン | +0.545 |
Reply | +0.544 |
Neg logit contributions | |
viol | -0.410 |
policeman | -0.410 |
Frontier | -0.402 |
trump | -0.388 |
Warsaw | -0.379 |
Token | : |
Token ID | 1058 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
20439 | +0.485 |
Ire | +0.456 |
=# | +0.438 |
ゴン | +0.420 |
Reply | +0.416 |
Neg logit contributions | |
viol | -0.378 |
policeman | -0.377 |
Frontier | -0.371 |
trump | -0.359 |
Warsaw | -0.351 |
Token | bool |
Token ID | 20512 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
20439 | +3.769 |
Ire | +3.546 |
=# | +3.451 |
Reply | +3.295 |
ゴン | +3.249 |
Neg logit contributions | |
policeman | -2.623 |
viol | -2.601 |
Frontier | -2.566 |
trump | -2.453 |
Warsaw | -2.430 |
Token | , |
Token ID | 837 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
20439 | +0.096 |
Ire | +0.091 |
=# | +0.088 |
ゴン | +0.084 |
Reply | +0.084 |
Neg logit contributions | |
policeman | -0.063 |
viol | -0.063 |
Frontier | -0.062 |
trump | -0.060 |
Warsaw | -0.058 |
Token | // |
Token ID | 3373 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
20439 | +0.961 |
Ire | +0.902 |
=# | +0.872 |
Reply | +0.832 |
ゴン | +0.825 |
Neg logit contributions | |
policeman | -0.737 |
viol | -0.732 |
Frontier | -0.722 |
trump | -0.694 |
Warsaw | -0.685 |
Token | Are |
Token ID | 4231 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
viol | +2.416 |
Frontier | +2.396 |
policeman | +2.381 |
trump | +2.308 |
Warsaw | +2.235 |
Neg logit contributions | |
20439 | -3.268 |
Ire | -3.072 |
=# | -2.869 |
ゴン | -2.859 |
Reply | -2.747 |
Token | we |
Token ID | 356 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
20439 | +0.764 |
Ire | +0.721 |
=# | +0.695 |
Reply | +0.668 |
ゴン | +0.666 |
Neg logit contributions | |
policeman | -0.482 |
viol | -0.482 |
Frontier | -0.473 |
trump | -0.455 |
Warsaw | -0.444 |