Token |
|
Token ID | 198 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
ude | +1.772 |
cerning | +1.771 |
asm | +1.639 |
ingu | +1.603 |
clus | +1.565 |
Neg logit contributions | |
deadliest | -2.316 |
losses | -2.289 |
victims | -2.250 |
occupants | -2.240 |
quotas | -2.201 |
Token |
|
Token ID | 198 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
losses | +5.599 |
deadliest | +5.558 |
victims | +5.545 |
quotas | +5.380 |
occupants | +5.351 |
Neg logit contributions | |
cerning | -6.247 |
ude | -6.103 |
ingu | -5.839 |
cius | -5.750 |
asm | -5.628 |
Token | Some |
Token ID | 4366 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
cerning | +2.119 |
ude | +2.111 |
asm | +1.967 |
ingu | +1.940 |
clus | +1.888 |
Neg logit contributions | |
deadliest | -2.310 |
losses | -2.302 |
victims | -2.250 |
occupants | -2.241 |
quotas | -2.194 |
Token | small |
Token ID | 1402 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
ude | +10.821 |
cerning | +10.778 |
asm | +10.147 |
ingu | +9.945 |
clus | +9.834 |
Neg logit contributions | |
deadliest | -6.761 |
losses | -6.427 |
occupants | -6.378 |
victims | -6.316 |
quotas | -6.281 |
Token | quick |
Token ID | 2068 |
Feature activation | +1.59e+01 |
Pos logit contributions | |
ude | +12.648 |
cerning | +12.554 |
asm | +11.987 |
clus | +11.771 |
ingu | +11.586 |
Neg logit contributions | |
deadliest | -6.790 |
occupants | -6.233 |
victims | -6.201 |
losses | -6.076 |
safest | -5.969 |
Token | tips |
Token ID | 9040 |
Feature activation | +2.39e+01 |
Pos logit contributions | |
ude | +15.951 |
cerning | +15.765 |
asm | +15.164 |
ingu | +14.791 |
clus | +14.703 |
Neg logit contributions | |
deadliest | -9.933 |
occupants | -9.510 |
victims | -9.104 |
losses | -8.976 |
safest | -8.851 |
Token | ? |
Token ID | 30 |
Feature activation | +9.36e+00 |
Pos logit contributions | |
cerning | +20.825 |
ude | +20.263 |
asm | +19.195 |
cius | +18.563 |
ás | +18.247 |
Neg logit contributions | |
deadliest | -13.952 |
occupants | -13.504 |
losses | -13.443 |
quotas | -13.252 |
victims | -12.824 |
Token | Wear |
Token ID | 27539 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
ude | +7.050 |
cerning | +7.005 |
asm | +6.480 |
ingu | +6.237 |
clus | +6.193 |
Neg logit contributions | |
deadliest | -9.651 |
losses | -9.510 |
occupants | -9.363 |
victims | -9.339 |
quotas | -9.222 |
Token | comfortable |
Token ID | 6792 |
Feature activation | +4.98e+00 |
Pos logit contributions | |
deadliest | +2.510 |
losses | +2.490 |
victims | +2.435 |
occupants | +2.404 |
quotas | +2.358 |
Neg logit contributions | |
cerning | -3.346 |
ude | -3.317 |
asm | -3.109 |
ingu | -3.106 |
cius | -3.067 |
Token | shoes |
Token ID | 10012 |
Feature activation | +6.26e+00 |
Pos logit contributions | |
ude | +6.157 |
cerning | +6.094 |
asm | +5.771 |
ingu | +5.742 |
clus | +5.664 |
Neg logit contributions | |
deadliest | -2.975 |
losses | -2.831 |
victims | -2.809 |
quotas | -2.764 |
occupants | -2.717 |
Token | and |
Token ID | 290 |
Feature activation | +6.17e+00 |
Pos logit contributions | |
cerning | +5.952 |
ude | +5.886 |
asm | +5.499 |
ingu | +5.439 |
cius | +5.337 |
Neg logit contributions | |
deadliest | -5.471 |
losses | -5.410 |
victims | -5.291 |
occupants | -5.274 |
quotas | -5.243 |
Token | are |
Token ID | 389 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
deadliest | +0.571 |
losses | +0.565 |
victims | +0.555 |
occupants | +0.550 |
quotas | +0.541 |
Neg logit contributions | |
cerning | -0.630 |
ude | -0.625 |
asm | -0.584 |
ingu | -0.582 |
cius | -0.570 |
Token | a |
Token ID | 257 |
Feature activation | +9.20e+00 |
Pos logit contributions | |
deadliest | +2.791 |
losses | +2.782 |
victims | +2.709 |
occupants | +2.660 |
quotas | +2.611 |
Neg logit contributions | |
cerning | -5.194 |
ude | -5.172 |
asm | -4.893 |
ingu | -4.883 |
cius | -4.822 |
Token | few |
Token ID | 1178 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
cerning | +9.203 |
ude | +9.135 |
asm | +8.571 |
ingu | +8.393 |
clus | +8.371 |
Neg logit contributions | |
deadliest | -7.186 |
losses | -6.994 |
occupants | -6.967 |
victims | -6.850 |
quotas | -6.667 |
Token | important |
Token ID | 1593 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
cerning | +12.446 |
ude | +12.325 |
asm | +11.605 |
ingu | +11.478 |
yt | +11.178 |
Neg logit contributions | |
deadliest | -6.756 |
occupants | -6.309 |
victims | -6.228 |
losses | -6.210 |
quotas | -5.940 |
Token | notes |
Token ID | 4710 |
Feature activation | +1.80e+01 |
Pos logit contributions | |
cerning | +11.629 |
ude | +11.567 |
asm | +10.969 |
ingu | +10.803 |
yt | +10.548 |
Neg logit contributions | |
deadliest | -6.440 |
victims | -5.901 |
occupants | -5.831 |
losses | -5.825 |
quotas | -5.573 |
Token | to |
Token ID | 284 |
Feature activation | +2.38e+01 |
Pos logit contributions | |
cerning | +17.642 |
ude | +17.025 |
asm | +16.122 |
ingu | +15.858 |
ás | +15.800 |
Neg logit contributions | |
deadliest | -11.460 |
losses | -10.582 |
occupants | -10.502 |
victims | -10.394 |
quotas | -10.276 |
Token | keep |
Token ID | 1394 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
cerning | +20.949 |
ude | +20.194 |
summarize | +18.867 |
dos | +18.772 |
explain | +18.648 |
Neg logit contributions | |
deadliest | -14.011 |
riots | -13.362 |
losses | -13.341 |
quotas | -13.192 |
occupants | -13.139 |
Token | in |
Token ID | 287 |
Feature activation | +1.65e+01 |
Pos logit contributions | |
cerning | +4.140 |
ude | +4.109 |
asm | +3.844 |
ingu | +3.812 |
cius | +3.777 |
Neg logit contributions | |
deadliest | -2.301 |
losses | -2.232 |
victims | -2.165 |
occupants | -2.145 |
quotas | -2.144 |
Token | mind |
Token ID | 2000 |
Feature activation | +1.77e+01 |
Pos logit contributions | |
cerning | +11.225 |
ude | +10.657 |
asm | +9.731 |
clus | +9.570 |
ingu | +9.260 |
Neg logit contributions | |
deadliest | -16.379 |
losses | -15.908 |
victims | -15.803 |
quotas | -15.802 |
occupants | -15.762 |
Token | : |
Token ID | 25 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
cerning | +17.455 |
ude | +16.337 |
asm | +15.600 |
ingu | +15.328 |
ás | +14.911 |
Neg logit contributions | |
deadliest | -12.083 |
losses | -11.356 |
occupants | -11.296 |
quotas | -11.063 |
victims | -10.984 |
Token |
|
Token ID | 198 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
cerning | +2.905 |
ude | +2.864 |
asm | +2.655 |
ingu | +2.645 |
cius | +2.560 |
Neg logit contributions | |
deadliest | -3.410 |
losses | -3.364 |
victims | -3.312 |
occupants | -3.305 |
quotas | -3.239 |
Token | basic |
Token ID | 4096 |
Feature activation | +1.51e+01 |
Pos logit contributions | |
ude | +12.506 |
cerning | +12.214 |
asm | +11.558 |
ingu | +11.233 |
yt | +11.177 |
Neg logit contributions | |
deadliest | -8.254 |
occupants | -8.042 |
losses | -7.816 |
victims | -7.691 |
quotas | -7.624 |
Token | dos |
Token ID | 23430 |
Feature activation | +9.24e+00 |
Pos logit contributions | |
ude | +16.176 |
cerning | +15.537 |
asm | +15.295 |
clus | +14.716 |
ingu | +14.706 |
Neg logit contributions | |
deadliest | -8.464 |
occupants | -8.158 |
losses | -7.783 |
victims | -7.748 |
invaders | -7.372 |
Token | and |
Token ID | 290 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
cerning | +8.464 |
ude | +8.334 |
asm | +7.786 |
ingu | +7.725 |
cius | +7.642 |
Neg logit contributions | |
deadliest | -8.452 |
losses | -8.101 |
victims | -8.098 |
occupants | -8.053 |
quotas | -7.846 |
Token | don |
Token ID | 836 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
cerning | +3.242 |
ude | +3.235 |
ingu | +2.893 |
dos | +2.877 |
cius | +2.858 |
Neg logit contributions | |
deadliest | -5.432 |
occupants | -5.248 |
losses | -5.178 |
victims | -5.159 |
quotas | -5.096 |
Token | 't |
Token ID | 470 |
Feature activation | +9.68e+00 |
Pos logit contributions | |
cerning | +3.488 |
ude | +3.455 |
asm | +3.242 |
ingu | +3.146 |
dos | +3.141 |
Neg logit contributions | |
deadliest | -3.551 |
occupants | -3.399 |
losses | -3.390 |
victims | -3.369 |
quotas | -3.282 |
Token | s |
Token ID | 82 |
Feature activation | +2.31e+01 |
Pos logit contributions | |
ude | +8.111 |
cerning | +8.021 |
cius | +7.550 |
dos | +7.531 |
asm | +7.408 |
Neg logit contributions | |
deadliest | -8.831 |
occupants | -8.547 |
victims | -8.392 |
losses | -8.360 |
revenues | -8.148 |
Token | , |
Token ID | 11 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
cerning | +20.544 |
ude | +20.503 |
cius | +19.221 |
asm | +18.890 |
wn | +18.506 |
Neg logit contributions | |
deadliest | -13.386 |
occupants | -12.862 |
losses | -12.547 |
quotas | -12.331 |
revenues | -12.105 |
Token | using |
Token ID | 1262 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
ude | +11.598 |
cerning | +11.324 |
asm | +10.820 |
cius | +10.612 |
ingu | +10.488 |
Neg logit contributions | |
deadliest | -7.863 |
occupants | -7.574 |
losses | -7.412 |
victims | -7.241 |
quotas | -7.099 |
Token | some |
Token ID | 617 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
ude | +3.924 |
cerning | +3.913 |
asm | +3.698 |
ingu | +3.671 |
cius | +3.604 |
Neg logit contributions | |
deadliest | -2.712 |
losses | -2.664 |
occupants | -2.619 |
victims | -2.603 |
quotas | -2.515 |
Token | recent |
Token ID | 2274 |
Feature activation | +7.02e+00 |
Pos logit contributions | |
ude | +11.131 |
cerning | +10.840 |
asm | +10.373 |
ingu | +10.245 |
clus | +9.995 |
Neg logit contributions | |
deadliest | -7.425 |
occupants | -7.329 |
losses | -7.192 |
victims | -7.027 |
quotas | -6.931 |
Token | examples |
Token ID | 6096 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
ude | +7.597 |
cerning | +7.541 |
asm | +7.255 |
ingu | +7.129 |
yt | +6.972 |
Neg logit contributions | |
deadliest | -4.909 |
occupants | -4.721 |
losses | -4.676 |
victims | -4.630 |
quotas | -4.595 |
Token | . |
Token ID | 13 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
ude | +8.709 |
cerning | +8.541 |
asm | +7.961 |
ingu | +7.913 |
cius | +7.590 |
Neg logit contributions | |
deadliest | -9.408 |
losses | -9.263 |
occupants | -9.212 |
victims | -9.141 |
quotas | -9.090 |
Token | Just |
Token ID | 2329 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
ude | +6.185 |
cerning | +6.108 |
asm | +5.745 |
ingu | +5.408 |
yt | +5.331 |
Neg logit contributions | |
deadliest | -8.985 |
losses | -8.833 |
occupants | -8.792 |
victims | -8.688 |
quotas | -8.525 |
Token | a |
Token ID | 257 |
Feature activation | +1.87e+01 |
Pos logit contributions | |
ude | +11.060 |
cerning | +10.977 |
asm | +10.482 |
ingu | +10.143 |
yt | +9.983 |
Neg logit contributions | |
deadliest | -7.379 |
losses | -7.196 |
occupants | -7.069 |
victims | -6.951 |
quotas | -6.851 |
Token | slight |
Token ID | 3731 |
Feature activation | +2.01e+01 |
Pos logit contributions | |
ude | +17.088 |
cerning | +16.663 |
asm | +16.132 |
yt | +15.311 |
wn | +15.246 |
Neg logit contributions | |
losses | -12.979 |
occupants | -12.926 |
deadliest | -12.800 |
quotas | -12.670 |
victims | -12.369 |
Token | side |
Token ID | 1735 |
Feature activation | +1.49e+01 |
Pos logit contributions | |
ude | +18.921 |
cerning | +18.427 |
asm | +17.900 |
yt | +16.918 |
ingu | +16.753 |
Neg logit contributions | |
deadliest | -12.458 |
occupants | -12.145 |
losses | -11.835 |
victims | -11.784 |
quotas | -11.590 |
Token | question |
Token ID | 1808 |
Feature activation | +2.28e+01 |
Pos logit contributions | |
ude | +12.208 |
cerning | +12.068 |
asm | +11.565 |
ingu | +10.806 |
hack | +10.711 |
Neg logit contributions | |
deadliest | -13.217 |
occupants | -13.002 |
losses | -12.994 |
quotas | -12.759 |
victims | -12.758 |
Token | , |
Token ID | 11 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
ude | +20.641 |
cerning | +20.531 |
asm | +19.049 |
cius | +18.579 |
ingu | +18.437 |
Neg logit contributions | |
deadliest | -14.350 |
losses | -13.940 |
occupants | -13.438 |
quotas | -13.091 |
victims | -12.968 |
Token | in |
Token ID | 287 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
ude | +13.436 |
cerning | +13.210 |
asm | +12.721 |
cius | +12.118 |
ingu | +12.059 |
Neg logit contributions | |
deadliest | -9.274 |
losses | -8.946 |
occupants | -8.908 |
victims | -8.699 |
quotas | -8.661 |
Token | an |
Token ID | 281 |
Feature activation | +1.71e+01 |
Pos logit contributions | |
cerning | +16.926 |
ude | +16.344 |
asm | +16.292 |
clus | +15.780 |
ingu | +15.356 |
Neg logit contributions | |
occupants | -11.001 |
deadliest | -10.715 |
victims | -10.498 |
losses | -10.434 |
quotas | -10.379 |
Token | interview |
Token ID | 2720 |
Feature activation | +1.38e+01 |
Pos logit contributions | |
ude | +17.123 |
cerning | +17.041 |
asm | +16.052 |
ingu | +15.726 |
utics | +15.091 |
Neg logit contributions | |
quotas | -10.383 |
deadliest | -10.350 |
losses | -10.290 |
occupants | -10.220 |
victims | -10.111 |
Token | with |
Token ID | 351 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
cerning | +13.493 |
ude | +13.402 |
asm | +12.562 |
ingu | +12.482 |
clus | +12.221 |
Neg logit contributions | |
deadliest | -10.056 |
occupants | -9.643 |
losses | -9.567 |
quotas | -9.377 |
victims | -9.298 |
Token | � |
Token ID | 247 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
ude | +2.434 |
cerning | +2.353 |
asm | +2.150 |
ingu | +2.120 |
cius | +2.057 |
Neg logit contributions | |
deadliest | -4.663 |
losses | -4.647 |
victims | -4.622 |
occupants | -4.595 |
quotas | -4.536 |
Token | ve |
Token ID | 303 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
ude | +1.423 |
cerning | +1.416 |
asm | +1.195 |
ingu | +1.161 |
cius | +1.126 |
Neg logit contributions | |
deadliest | -4.308 |
losses | -4.303 |
victims | -4.261 |
occupants | -4.256 |
quotas | -4.203 |
Token | talked |
Token ID | 6619 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
ude | +3.843 |
cerning | +3.802 |
asm | +3.579 |
ingu | +3.511 |
wn | +3.497 |
Neg logit contributions | |
deadliest | -2.347 |
losses | -2.345 |
occupants | -2.293 |
victims | -2.270 |
quotas | -2.247 |
Token | about |
Token ID | 546 |
Feature activation | +9.28e+00 |
Pos logit contributions | |
cerning | +0.597 |
ude | +0.593 |
asm | +0.552 |
ingu | +0.549 |
cius | +0.538 |
Neg logit contributions | |
deadliest | -0.466 |
losses | -0.460 |
victims | -0.451 |
occupants | -0.450 |
quotas | -0.440 |
Token | a |
Token ID | 257 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
ude | +9.940 |
cerning | +9.880 |
asm | +9.222 |
ingu | +9.175 |
clus | +9.109 |
Neg logit contributions | |
deadliest | -6.233 |
occupants | -6.045 |
losses | -6.002 |
quotas | -5.841 |
victims | -5.818 |
Token | lot |
Token ID | 1256 |
Feature activation | +2.27e+01 |
Pos logit contributions | |
ude | +9.647 |
cerning | +9.645 |
asm | +8.788 |
wn | +8.669 |
lex | +8.637 |
Neg logit contributions | |
deadliest | -10.816 |
losses | -10.800 |
occupants | -10.793 |
victims | -10.603 |
quotas | -10.546 |
Token | here |
Token ID | 994 |
Feature activation | +1.74e+01 |
Pos logit contributions | |
cerning | +20.082 |
ude | +19.964 |
ingu | +18.408 |
yt | +18.308 |
asm | +18.304 |
Neg logit contributions | |
deadliest | -13.273 |
occupants | -13.087 |
quotas | -12.524 |
safest | -12.432 |
losses | -12.347 |
Token | because |
Token ID | 780 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
cerning | +16.424 |
ude | +16.060 |
asm | +15.009 |
ingu | +14.889 |
yt | +14.754 |
Neg logit contributions | |
deadliest | -12.274 |
occupants | -11.829 |
losses | -11.479 |
quotas | -11.426 |
safest | -11.303 |
Token | he |
Token ID | 339 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
cerning | +2.943 |
ude | +2.937 |
asm | +2.782 |
ingu | +2.724 |
cius | +2.669 |
Neg logit contributions | |
deadliest | -2.029 |
losses | -2.005 |
victims | -1.959 |
occupants | -1.958 |
quotas | -1.927 |
Token | � |
Token ID | 447 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ude | +3.172 |
cerning | +3.139 |
ingu | +2.968 |
asm | +2.967 |
cius | +2.872 |
Neg logit contributions | |
deadliest | -2.225 |
losses | -2.169 |
occupants | -2.145 |
victims | -2.135 |
quotas | -2.083 |
Token | � |
Token ID | 247 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
deadliest | +2.960 |
losses | +2.941 |
victims | +2.897 |
occupants | +2.872 |
quotas | +2.837 |
Neg logit contributions | |
cerning | -1.950 |
ude | -1.923 |
asm | -1.756 |
ingu | -1.751 |
cius | -1.703 |
Token | � |
Token ID | 447 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
deadliest | +0.414 |
losses | +0.412 |
victims | +0.403 |
occupants | +0.396 |
quotas | +0.390 |
Neg logit contributions | |
cerning | -0.576 |
ude | -0.572 |
asm | -0.538 |
ingu | -0.536 |
cius | -0.526 |
Token | � |
Token ID | 247 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
ude | +1.937 |
cerning | +1.914 |
asm | +1.742 |
ingu | +1.699 |
cius | +1.627 |
Neg logit contributions | |
deadliest | -3.706 |
losses | -3.683 |
victims | -3.650 |
occupants | -3.621 |
quotas | -3.589 |
Token | s |
Token ID | 82 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
losses | +0.718 |
deadliest | +0.718 |
victims | +0.702 |
occupants | +0.690 |
quotas | +0.683 |
Neg logit contributions | |
cerning | -0.840 |
ude | -0.829 |
ingu | -0.780 |
asm | -0.779 |
cius | -0.763 |
Token | a |
Token ID | 257 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
ude | +3.815 |
cerning | +3.807 |
asm | +3.603 |
ingu | +3.576 |
cius | +3.514 |
Neg logit contributions | |
deadliest | -2.211 |
losses | -2.160 |
victims | -2.118 |
occupants | -2.109 |
quotas | -2.058 |
Token | picture |
Token ID | 4286 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ude | +11.372 |
cerning | +11.173 |
asm | +10.589 |
ingu | +10.388 |
lex | +10.113 |
Neg logit contributions | |
deadliest | -8.375 |
occupants | -8.311 |
losses | -8.227 |
victims | -8.121 |
quotas | -8.061 |
Token | to |
Token ID | 284 |
Feature activation | +2.26e+01 |
Pos logit contributions | |
ude | +10.072 |
cerning | +10.060 |
asm | +9.421 |
ingu | +9.265 |
cius | +8.989 |
Neg logit contributions | |
deadliest | -9.435 |
losses | -9.183 |
occupants | -9.118 |
victims | -9.027 |
quotas | -9.012 |
Token | help |
Token ID | 1037 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
ude | +19.810 |
cerning | +18.943 |
summarize | +18.804 |
explain | +18.699 |
yt | +18.303 |
Neg logit contributions | |
quotas | -14.116 |
losses | -13.900 |
deadliest | -13.858 |
occupants | -13.182 |
victims | -12.971 |
Token | make |
Token ID | 787 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
ude | +13.055 |
cerning | +12.574 |
ingu | +12.003 |
asm | +11.936 |
summarize | +11.525 |
Neg logit contributions | |
deadliest | -10.670 |
losses | -10.474 |
quotas | -10.226 |
occupants | -9.853 |
revenues | -9.849 |
Token | this |
Token ID | 428 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
cerning | +3.108 |
ude | +3.089 |
asm | +2.891 |
ingu | +2.871 |
clus | +2.783 |
Neg logit contributions | |
deadliest | -2.775 |
losses | -2.711 |
victims | -2.669 |
occupants | -2.653 |
quotas | -2.614 |
Token | clearer |
Token ID | 22363 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
cerning | +0.934 |
ude | +0.931 |
asm | +0.870 |
ingu | +0.867 |
cius | +0.849 |
Neg logit contributions | |
deadliest | -0.755 |
losses | -0.748 |
victims | -0.733 |
occupants | -0.726 |
quotas | -0.714 |
Token | : |
Token ID | 25 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
ude | +10.065 |
cerning | +10.039 |
ingu | +9.246 |
asm | +9.153 |
cius | +8.729 |
Neg logit contributions | |
deadliest | -11.661 |
losses | -11.243 |
occupants | -11.173 |
victims | -11.138 |
quotas | -11.079 |
Token | working |
Token ID | 1762 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
deadliest | +0.095 |
losses | +0.094 |
occupants | +0.092 |
victims | +0.091 |
quotas | +0.091 |
Neg logit contributions | |
cerning | -0.109 |
ude | -0.109 |
asm | -0.102 |
ingu | -0.101 |
cius | -0.099 |
Token | on |
Token ID | 319 |
Feature activation | +6.51e+00 |
Pos logit contributions | |
deadliest | +2.414 |
losses | +2.390 |
victims | +2.330 |
occupants | +2.281 |
quotas | +2.235 |
Neg logit contributions | |
cerning | -4.806 |
ude | -4.775 |
asm | -4.526 |
ingu | -4.524 |
cius | -4.465 |
Token | a |
Token ID | 257 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
cerning | +7.347 |
ude | +7.296 |
asm | +7.081 |
ingu | +6.886 |
clus | +6.779 |
Neg logit contributions | |
deadliest | -4.368 |
losses | -4.176 |
occupants | -4.113 |
victims | -4.080 |
quotas | -3.988 |
Token | blog |
Token ID | 4130 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
cerning | +10.169 |
ude | +10.030 |
asm | +9.566 |
clus | +9.286 |
ingu | +9.279 |
Neg logit contributions | |
deadliest | -7.571 |
losses | -7.431 |
occupants | -7.378 |
victims | -7.172 |
quotas | -7.109 |
Token | post |
Token ID | 1281 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
cerning | +9.793 |
ude | +9.512 |
asm | +9.105 |
ingu | +8.774 |
cius | +8.593 |
Neg logit contributions | |
deadliest | -8.558 |
losses | -8.287 |
occupants | -8.161 |
victims | -8.135 |
quotas | -8.053 |
Token | to |
Token ID | 284 |
Feature activation | +2.26e+01 |
Pos logit contributions | |
cerning | +12.447 |
ude | +12.261 |
asm | +11.628 |
ingu | +11.107 |
cius | +11.008 |
Neg logit contributions | |
deadliest | -9.574 |
losses | -9.081 |
occupants | -9.056 |
victims | -8.834 |
quotas | -8.745 |
Token | describe |
Token ID | 6901 |
Feature activation | +4.93e+00 |
Pos logit contributions | |
ude | +19.540 |
summarize | +19.285 |
explain | +19.154 |
cerning | +18.651 |
discuss | +17.810 |
Neg logit contributions | |
deadliest | -14.319 |
quotas | -14.282 |
losses | -13.935 |
occupants | -13.879 |
invaders | -13.414 |
Token | these |
Token ID | 777 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
ude | +5.412 |
cerning | +5.409 |
asm | +5.162 |
ingu | +5.062 |
clus | +4.968 |
Neg logit contributions | |
deadliest | -3.562 |
losses | -3.424 |
victims | -3.381 |
occupants | -3.373 |
quotas | -3.296 |
Token | new |
Token ID | 649 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
cerning | +0.095 |
ude | +0.095 |
asm | +0.090 |
ingu | +0.089 |
cius | +0.088 |
Neg logit contributions | |
deadliest | -0.051 |
losses | -0.050 |
victims | -0.049 |
occupants | -0.049 |
quotas | -0.047 |
Token | instance |
Token ID | 4554 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
deadliest | +0.521 |
losses | +0.512 |
victims | +0.502 |
occupants | +0.497 |
quotas | +0.486 |
Neg logit contributions | |
cerning | -0.760 |
ude | -0.756 |
asm | -0.714 |
ingu | -0.709 |
cius | -0.695 |
Token | types |
Token ID | 3858 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
deadliest | +2.624 |
losses | +2.617 |
victims | +2.561 |
occupants | +2.530 |
quotas | +2.496 |
Neg logit contributions | |
cerning | -3.188 |
ude | -3.155 |
ingu | -2.955 |
asm | -2.949 |
cius | -2.905 |
Token | on |
Token ID | 319 |
Feature activation | +8.09e+00 |
Pos logit contributions | |
deadliest | +1.672 |
losses | +1.653 |
victims | +1.611 |
occupants | +1.574 |
quotas | +1.533 |
Neg logit contributions | |
cerning | -3.584 |
ude | -3.561 |
asm | -3.387 |
ingu | -3.367 |
cius | -3.333 |
Token | fundamental |
Token ID | 7531 |
Feature activation | +8.55e+00 |
Pos logit contributions | |
ude | +9.186 |
cerning | +9.093 |
asm | +8.870 |
ingu | +8.633 |
clus | +8.428 |
Neg logit contributions | |
deadliest | -5.326 |
occupants | -4.973 |
victims | -4.950 |
losses | -4.921 |
quotas | -4.828 |
Token | measurement |
Token ID | 15558 |
Feature activation | +7.38e+00 |
Pos logit contributions | |
ude | +9.630 |
cerning | +9.409 |
asm | +9.261 |
ingu | +9.015 |
clus | +8.834 |
Neg logit contributions | |
deadliest | -5.617 |
occupants | -5.266 |
victims | -5.234 |
losses | -5.088 |
quotas | -5.010 |
Token | ambig |
Token ID | 18203 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
ude | +8.815 |
cerning | +8.574 |
asm | +8.313 |
ingu | +8.122 |
clus | +8.031 |
Neg logit contributions | |
deadliest | -4.581 |
occupants | -4.346 |
victims | -4.215 |
losses | -4.171 |
safest | -4.007 |
Token | u |
Token ID | 84 |
Feature activation | +9.70e+00 |
Pos logit contributions | |
cerning | +6.686 |
ude | +6.680 |
asm | +5.998 |
ingu | +5.601 |
clus | +5.455 |
Neg logit contributions | |
deadliest | -17.813 |
occupants | -17.452 |
losses | -17.432 |
victims | -17.268 |
quotas | -17.154 |
Token | ities |
Token ID | 871 |
Feature activation | +2.24e+01 |
Pos logit contributions | |
ude | +6.222 |
cerning | +5.980 |
asm | +5.483 |
ingu | +5.436 |
cius | +5.346 |
Neg logit contributions | |
deadliest | -11.550 |
occupants | -11.192 |
losses | -11.178 |
victims | -11.105 |
quotas | -11.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
ude | +21.119 |
cerning | +20.977 |
asm | +20.472 |
cius | +19.735 |
ingu | +19.699 |
Neg logit contributions | |
deadliest | -12.315 |
occupants | -11.756 |
losses | -11.463 |
quotas | -11.273 |
invaders | -11.114 |
Token |
|
Token ID | 198 |
Feature activation | -7.83e+00 |
Pos logit contributions | |
cerning | +1.609 |
ude | +1.605 |
asm | +1.504 |
ingu | +1.459 |
cius | +1.428 |
Neg logit contributions | |
deadliest | -2.008 |
losses | -1.963 |
occupants | -1.945 |
victims | -1.942 |
quotas | -1.896 |
Token |
|
Token ID | 198 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
losses | +6.627 |
deadliest | +6.571 |
victims | +6.556 |
quotas | +6.380 |
occupants | +6.317 |
Neg logit contributions | |
cerning | -7.588 |
ude | -7.395 |
ingu | -7.094 |
cius | -7.002 |
asm | -6.799 |
Token | Although |
Token ID | 7003 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
cerning | +0.553 |
ude | +0.550 |
asm | +0.515 |
ingu | +0.508 |
cius | +0.495 |
Neg logit contributions | |
deadliest | -0.594 |
losses | -0.586 |
victims | -0.578 |
occupants | -0.575 |
quotas | -0.563 |
Token | it |
Token ID | 340 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
deadliest | +0.864 |
losses | +0.850 |
victims | +0.838 |
occupants | +0.831 |
quotas | +0.815 |
Neg logit contributions | |
cerning | -0.995 |
ude | -0.989 |
asm | -0.928 |
ingu | -0.921 |
cius | -0.902 |
Token | semi |
Token ID | 10663 |
Feature activation | +9.44e+00 |
Pos logit contributions | |
ude | +7.531 |
cerning | +7.458 |
asm | +7.225 |
ingu | +7.006 |
cius | +6.847 |
Neg logit contributions | |
deadliest | -3.852 |
losses | -3.778 |
occupants | -3.734 |
victims | -3.667 |
quotas | -3.662 |
Token | - |
Token ID | 12 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
ude | +8.461 |
cerning | +8.449 |
ingu | +7.885 |
asm | +7.850 |
clus | +7.524 |
Neg logit contributions | |
deadliest | -8.629 |
losses | -8.521 |
occupants | -8.456 |
victims | -8.411 |
quotas | -8.246 |
Token | relevant |
Token ID | 49659 |
Feature activation | +1.61e+01 |
Pos logit contributions | |
cerning | +8.880 |
ude | +8.810 |
asm | +8.217 |
ingu | +7.961 |
hack | +7.735 |
Neg logit contributions | |
deadliest | -13.100 |
losses | -13.006 |
occupants | -12.936 |
victims | -12.648 |
quotas | -12.646 |
Token | survey |
Token ID | 5526 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
ude | +16.579 |
asm | +16.148 |
cerning | +16.038 |
yt | +15.566 |
wn | +15.151 |
Neg logit contributions | |
deadliest | -8.901 |
occupants | -8.410 |
victims | -8.044 |
quotas | -7.958 |
losses | -7.924 |
Token | results |
Token ID | 2482 |
Feature activation | +1.88e+01 |
Pos logit contributions | |
ude | +11.362 |
cerning | +11.279 |
asm | +10.877 |
ingu | +10.680 |
yt | +10.476 |
Neg logit contributions | |
deadliest | -4.367 |
occupants | -4.004 |
losses | -3.914 |
victims | -3.901 |
safest | -3.789 |
Token | to |
Token ID | 284 |
Feature activation | +2.24e+01 |
Pos logit contributions | |
ude | +17.319 |
cerning | +17.272 |
asm | +16.553 |
wn | +15.745 |
hack | +15.562 |
Neg logit contributions | |
deadliest | -12.387 |
occupants | -11.796 |
losses | -11.659 |
quotas | -11.552 |
victims | -11.245 |
Token | help |
Token ID | 1037 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
ude | +19.308 |
cerning | +18.444 |
summarize | +18.097 |
asm | +17.810 |
explain | +17.701 |
Neg logit contributions | |
losses | -14.142 |
deadliest | -13.972 |
quotas | -13.910 |
occupants | -13.519 |
victims | -13.102 |
Token | evaluate |
Token ID | 13446 |
Feature activation | +7.59e+00 |
Pos logit contributions | |
ude | +12.762 |
cerning | +12.390 |
asm | +11.789 |
ingu | +11.515 |
explain | +11.230 |
Neg logit contributions | |
deadliest | -9.871 |
losses | -9.726 |
quotas | -9.470 |
occupants | -9.164 |
revenues | -9.069 |
Token | the |
Token ID | 262 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
cerning | +7.322 |
ude | +7.306 |
asm | +7.013 |
ingu | +6.758 |
clus | +6.659 |
Neg logit contributions | |
deadliest | -6.189 |
losses | -5.942 |
occupants | -5.924 |
victims | -5.872 |
quotas | -5.808 |
Token | significance |
Token ID | 12085 |
Feature activation | +6.57e+00 |
Pos logit contributions | |
cerning | +1.699 |
ude | +1.692 |
asm | +1.603 |
ingu | +1.567 |
cius | +1.552 |
Neg logit contributions | |
deadliest | -1.262 |
losses | -1.240 |
victims | -1.219 |
occupants | -1.212 |
quotas | -1.193 |
Token | of |
Token ID | 286 |
Feature activation | +6.92e+00 |
Pos logit contributions | |
cerning | +5.351 |
ude | +5.279 |
asm | +4.834 |
ingu | +4.805 |
cius | +4.710 |
Neg logit contributions | |
deadliest | -6.755 |
losses | -6.646 |
victims | -6.548 |
occupants | -6.522 |
quotas | -6.420 |
Token | 30 |
Token ID | 1270 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
ude | +3.828 |
cerning | +3.816 |
asm | +3.528 |
ingu | +3.436 |
cius | +3.374 |
Neg logit contributions | |
deadliest | -4.880 |
losses | -4.843 |
victims | -4.763 |
occupants | -4.735 |
quotas | -4.664 |
Token | Amazing |
Token ID | 23181 |
Feature activation | +8.29e+00 |
Pos logit contributions | |
cerning | +3.643 |
ude | +3.634 |
asm | +3.364 |
ingu | +3.337 |
cius | +3.285 |
Neg logit contributions | |
deadliest | -3.806 |
losses | -3.774 |
victims | -3.711 |
occupants | -3.702 |
quotas | -3.639 |
Token | Street |
Token ID | 3530 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
ude | +8.924 |
cerning | +8.790 |
asm | +8.382 |
ingu | +8.306 |
cius | +8.154 |
Neg logit contributions | |
deadliest | -5.971 |
losses | -5.722 |
occupants | -5.643 |
victims | -5.639 |
quotas | -5.507 |
Token | Post |
Token ID | 2947 |
Feature activation | +6.73e+00 |
Pos logit contributions | |
cerning | +3.703 |
ude | +3.680 |
asm | +3.441 |
ingu | +3.422 |
cius | +3.302 |
Neg logit contributions | |
deadliest | -4.051 |
losses | -3.955 |
victims | -3.902 |
occupants | -3.876 |
quotas | -3.833 |
Token | ers |
Token ID | 364 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
ude | +6.627 |
cerning | +6.598 |
asm | +6.190 |
ingu | +6.152 |
clus | +5.973 |
Neg logit contributions | |
deadliest | -5.715 |
losses | -5.501 |
victims | -5.430 |
occupants | -5.393 |
quotas | -5.306 |
Token | To |
Token ID | 1675 |
Feature activation | +2.16e+01 |
Pos logit contributions | |
ude | +12.925 |
cerning | +12.609 |
asm | +11.939 |
cius | +11.754 |
ingu | +11.679 |
Neg logit contributions | |
deadliest | -9.899 |
occupants | -9.373 |
losses | -9.367 |
victims | -9.135 |
quotas | -9.108 |
Token | Bright |
Token ID | 17558 |
Feature activation | +6.76e+00 |
Pos logit contributions | |
ude | +20.005 |
cerning | +19.389 |
asm | +18.553 |
dos | +18.315 |
clus | +18.140 |
Neg logit contributions | |
deadliest | -13.323 |
losses | -13.061 |
quotas | -12.873 |
occupants | -12.378 |
safest | -12.276 |
Token | en |
Token ID | 268 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
ude | +4.200 |
cerning | +4.147 |
asm | +3.700 |
ingu | +3.591 |
cius | +3.455 |
Neg logit contributions | |
deadliest | -8.231 |
losses | -8.042 |
victims | -7.890 |
quotas | -7.889 |
occupants | -7.865 |
Token | Your |
Token ID | 3406 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
ude | +2.582 |
cerning | +2.581 |
asm | +2.347 |
ingu | +2.329 |
cius | +2.275 |
Neg logit contributions | |
deadliest | -3.456 |
losses | -3.409 |
victims | -3.350 |
occupants | -3.319 |
quotas | -3.300 |
Token | Day |
Token ID | 3596 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
ude | +4.376 |
cerning | +4.362 |
asm | +4.032 |
ingu | +4.003 |
cius | +3.919 |
Neg logit contributions | |
deadliest | -4.052 |
losses | -3.995 |
victims | -3.917 |
occupants | -3.872 |
quotas | -3.858 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.88e+00 |
Pos logit contributions | |
ude | +4.816 |
cerning | +4.768 |
asm | +4.403 |
ingu | +4.339 |
cius | +4.260 |
Neg logit contributions | |
deadliest | -5.604 |
losses | -5.498 |
victims | -5.404 |
occupants | -5.396 |
quotas | -5.314 |
Token | or |
Token ID | 393 |
Feature activation | +9.61e+00 |
Pos logit contributions | |
ude | +8.517 |
cerning | +8.500 |
asm | +7.932 |
ingu | +7.728 |
cius | +7.707 |
Neg logit contributions | |
deadliest | -7.262 |
losses | -7.069 |
occupants | -6.945 |
victims | -6.882 |
quotas | -6.768 |
Token | Short |
Token ID | 10073 |
Feature activation | +7.56e+00 |
Pos logit contributions | |
cerning | +9.957 |
ude | +9.913 |
asm | +9.400 |
ingu | +9.167 |
cius | +9.139 |
Neg logit contributions | |
deadliest | -7.334 |
losses | -7.142 |
occupants | -6.990 |
victims | -6.915 |
quotas | -6.861 |
Token | Rules |
Token ID | 14252 |
Feature activation | +1.87e+01 |
Pos logit contributions | |
cerning | +6.601 |
ude | +6.539 |
asm | +6.106 |
ingu | +6.003 |
cius | +5.857 |
Neg logit contributions | |
deadliest | -7.263 |
losses | -7.042 |
occupants | -6.977 |
victims | -6.964 |
quotas | -6.875 |
Token | and |
Token ID | 290 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
cerning | +17.971 |
ude | +17.214 |
asm | +16.278 |
cius | +16.115 |
ingu | +15.837 |
Neg logit contributions | |
deadliest | -12.724 |
losses | -12.560 |
occupants | -12.359 |
victims | -12.061 |
quotas | -11.882 |
Token | Ref |
Token ID | 6524 |
Feature activation | +1.16e+01 |
Pos logit contributions | |
cerning | +10.541 |
ude | +10.528 |
asm | +9.831 |
ingu | +9.663 |
clus | +9.550 |
Neg logit contributions | |
deadliest | -8.900 |
losses | -8.634 |
occupants | -8.589 |
victims | -8.586 |
quotas | -8.178 |
Token | lections |
Token ID | 26448 |
Feature activation | +2.14e+01 |
Pos logit contributions | |
ude | +7.974 |
cerning | +7.956 |
ingu | +7.463 |
asm | +7.285 |
lex | +7.175 |
Neg logit contributions | |
deadliest | -12.706 |
losses | -12.461 |
victims | -12.438 |
occupants | -12.289 |
quotas | -11.989 |
Token | for |
Token ID | 329 |
Feature activation | +1.49e+01 |
Pos logit contributions | |
cerning | +20.594 |
ude | +19.759 |
asm | +18.514 |
ingu | +18.240 |
cius | +18.235 |
Neg logit contributions | |
deadliest | -13.065 |
losses | -12.506 |
occupants | -12.443 |
victims | -12.064 |
quotas | -11.936 |
Token | Conduct |
Token ID | 28579 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
cerning | +12.680 |
ude | +12.376 |
asm | +11.843 |
cius | +11.341 |
clus | +11.040 |
Neg logit contributions | |
deadliest | -12.618 |
losses | -12.381 |
quotas | -12.112 |
occupants | -11.721 |
revenues | -11.716 |
Token | in |
Token ID | 287 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
cerning | +7.849 |
ude | +7.741 |
asm | +7.163 |
ingu | +7.100 |
cius | +6.847 |
Neg logit contributions | |
deadliest | -7.799 |
losses | -7.655 |
quotas | -7.433 |
occupants | -7.400 |
victims | -7.383 |
Token | Society |
Token ID | 7023 |
Feature activation | +7.96e+00 |
Pos logit contributions | |
cerning | +15.722 |
ude | +15.205 |
asm | +14.823 |
clus | +14.127 |
cius | +13.834 |
Neg logit contributions | |
deadliest | -12.271 |
losses | -11.953 |
quotas | -11.771 |
victims | -11.638 |
occupants | -11.634 |
Token | , |
Token ID | 11 |
Feature activation | +8.29e+00 |
Pos logit contributions | |
cerning | +6.641 |
ude | +6.579 |
asm | +6.071 |
ingu | +5.999 |
cius | +5.873 |
Neg logit contributions | |
deadliest | -7.891 |
losses | -7.754 |
victims | -7.574 |
occupants | -7.559 |
quotas | -7.472 |
Token | gate |
Token ID | 10494 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
cerning | +2.212 |
ude | +2.193 |
asm | +2.043 |
ingu | +2.034 |
cius | +1.983 |
Neg logit contributions | |
deadliest | -2.238 |
losses | -2.200 |
victims | -2.164 |
occupants | -2.160 |
quotas | -2.113 |
Token | Lit |
Token ID | 25659 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
cerning | +3.072 |
ude | +3.057 |
asm | +2.838 |
ingu | +2.816 |
cius | +2.736 |
Neg logit contributions | |
deadliest | -3.345 |
losses | -3.267 |
occupants | -3.212 |
victims | -3.193 |
quotas | -3.163 |
Token | igation |
Token ID | 7065 |
Feature activation | +6.33e+00 |
Pos logit contributions | |
cerning | +2.995 |
ude | +2.964 |
asm | +2.669 |
ingu | +2.647 |
cius | +2.570 |
Neg logit contributions | |
deadliest | -5.436 |
losses | -5.369 |
victims | -5.306 |
occupants | -5.302 |
quotas | -5.241 |
Token | Q |
Token ID | 1195 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
cerning | +5.571 |
ude | +5.503 |
asm | +5.083 |
ingu | +5.061 |
cius | +4.936 |
Neg logit contributions | |
deadliest | -6.040 |
losses | -5.848 |
occupants | -5.806 |
victims | -5.723 |
quotas | -5.687 |
Token | & |
Token ID | 5 |
Feature activation | +7.46e+00 |
Pos logit contributions | |
ude | +9.836 |
cerning | +9.666 |
asm | +9.197 |
ingu | +8.915 |
dos | +8.586 |
Neg logit contributions | |
deadliest | -13.985 |
losses | -13.665 |
occupants | -13.593 |
quotas | -13.521 |
victims | -13.511 |
Token | As |
Token ID | 1722 |
Feature activation | +2.12e+01 |
Pos logit contributions | |
ude | +6.863 |
cerning | +6.829 |
asm | +6.590 |
ingu | +6.182 |
ás | +6.057 |
Neg logit contributions | |
deadliest | -6.641 |
losses | -6.497 |
quotas | -6.413 |
victims | -6.392 |
occupants | -6.390 |
Token | . |
Token ID | 13 |
Feature activation | +9.26e+00 |
Pos logit contributions | |
cerning | +18.612 |
ude | +18.584 |
cius | +17.047 |
ingu | +16.661 |
asm | +16.606 |
Neg logit contributions | |
deadliest | -14.639 |
occupants | -13.932 |
losses | -13.404 |
quotas | -13.292 |
safest | -13.267 |
Token |
|
Token ID | 198 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
cerning | +7.540 |
ude | +7.526 |
asm | +6.911 |
ingu | +6.743 |
cius | +6.693 |
Neg logit contributions | |
deadliest | -9.120 |
losses | -8.918 |
occupants | -8.873 |
victims | -8.709 |
quotas | -8.607 |
Token |
|
Token ID | 198 |
Feature activation | +7.14e+00 |
Pos logit contributions | |
ude | +0.379 |
cerning | +0.378 |
asm | +0.353 |
ingu | +0.341 |
clus | +0.335 |
Neg logit contributions | |
deadliest | -0.472 |
losses | -0.465 |
occupants | -0.458 |
victims | -0.457 |
quotas | -0.447 |
Token | Maybe |
Token ID | 13300 |
Feature activation | +6.39e+00 |
Pos logit contributions | |
ude | +5.866 |
cerning | +5.826 |
asm | +5.428 |
ingu | +5.294 |
clus | +5.191 |
Neg logit contributions | |
deadliest | -7.153 |
losses | -7.109 |
occupants | -6.941 |
victims | -6.884 |
safest | -6.760 |
Token | you |
Token ID | 345 |
Feature activation | +7.98e+00 |
Pos logit contributions | |
cerning | +6.642 |
ude | +6.632 |
ingu | +6.197 |
asm | +6.177 |
cius | +6.096 |
Neg logit contributions | |
deadliest | -4.949 |
losses | -4.851 |
occupants | -4.763 |
victims | -4.702 |
quotas | -4.626 |
Token | thought |
Token ID | 1807 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ude | +12.530 |
cerning | +12.111 |
asm | +11.632 |
cius | +11.390 |
ingu | +11.358 |
Neg logit contributions | |
deadliest | -8.962 |
losses | -8.791 |
occupants | -8.623 |
quotas | -8.484 |
victims | -8.385 |
Token | it |
Token ID | 340 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
deadliest | +2.383 |
losses | +2.381 |
victims | +2.343 |
occupants | +2.310 |
quotas | +2.285 |
Neg logit contributions | |
cerning | -2.232 |
ude | -2.215 |
ingu | -2.057 |
asm | -2.053 |
cius | -2.009 |
Token | would |
Token ID | 561 |
Feature activation | +7.43e+00 |
Pos logit contributions | |
deadliest | +0.292 |
losses | +0.289 |
victims | +0.282 |
occupants | +0.282 |
quotas | +0.279 |
Neg logit contributions | |
cerning | -0.414 |
ude | -0.412 |
asm | -0.387 |
ingu | -0.383 |
clus | -0.377 |
Token | be |
Token ID | 307 |
Feature activation | +9.27e-01 |
Pos logit contributions | |
ude | +6.182 |
cerning | +6.053 |
asm | +5.677 |
ingu | +5.567 |
clus | +5.388 |
Neg logit contributions | |
deadliest | -7.304 |
losses | -7.228 |
quotas | -7.065 |
occupants | -7.065 |
victims | -7.042 |
Token | interesting |
Token ID | 3499 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
cerning | +0.862 |
ude | +0.859 |
asm | +0.801 |
ingu | +0.790 |
clus | +0.770 |
Neg logit contributions | |
deadliest | -0.853 |
losses | -0.846 |
victims | -0.830 |
occupants | -0.829 |
quotas | -0.815 |
Token | to |
Token ID | 284 |
Feature activation | +2.12e+01 |
Pos logit contributions | |
cerning | +11.064 |
ude | +10.980 |
ingu | +10.446 |
asm | +10.398 |
cius | +10.096 |
Neg logit contributions | |
deadliest | -6.968 |
losses | -6.797 |
occupants | -6.620 |
victims | -6.609 |
quotas | -6.435 |
Token | test |
Token ID | 1332 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ude | +18.031 |
summarize | +17.438 |
cerning | +17.246 |
explain | +16.975 |
ingu | +16.854 |
Neg logit contributions | |
deadliest | -14.330 |
losses | -14.185 |
quotas | -13.805 |
occupants | -13.749 |
revenues | -13.342 |
Token | drive |
Token ID | 3708 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
cerning | +2.937 |
ude | +2.932 |
asm | +2.791 |
ingu | +2.749 |
clus | +2.693 |
Neg logit contributions | |
deadliest | -1.690 |
losses | -1.651 |
victims | -1.599 |
occupants | -1.585 |
quotas | -1.540 |
Token | both |
Token ID | 1111 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
ude | +3.396 |
cerning | +3.395 |
asm | +3.228 |
ingu | +3.146 |
clus | +3.097 |
Neg logit contributions | |
deadliest | -2.566 |
losses | -2.482 |
victims | -2.431 |
occupants | -2.412 |
quotas | -2.359 |
Token | Flat |
Token ID | 21939 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
ude | +3.368 |
cerning | +3.364 |
asm | +3.167 |
ingu | +3.120 |
cius | +3.054 |
Neg logit contributions | |
deadliest | -2.857 |
losses | -2.794 |
victims | -2.741 |
occupants | -2.728 |
quotas | -2.667 |
Token | pak |
Token ID | 41091 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
deadliest | +5.028 |
losses | +5.022 |
victims | +4.962 |
occupants | +4.919 |
quotas | +4.891 |
Neg logit contributions | |
cerning | -2.324 |
ude | -2.295 |
ingu | -2.045 |
asm | -2.012 |
cius | -1.968 |
Token | unlocked |
Token ID | 14838 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
deadliest | +2.122 |
losses | +2.101 |
victims | +2.075 |
occupants | +2.026 |
quotas | +1.992 |
Neg logit contributions | |
cerning | -3.129 |
ude | -3.105 |
asm | -2.936 |
ingu | -2.925 |
cius | -2.857 |
Token | . |
Token ID | 13 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
deadliest | +1.732 |
losses | +1.723 |
victims | +1.698 |
occupants | +1.681 |
quotas | +1.660 |
Neg logit contributions | |
cerning | -1.281 |
ude | -1.267 |
asm | -1.166 |
ingu | -1.161 |
cius | -1.131 |
Token | Another |
Token ID | 6023 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
cerning | +0.564 |
ude | +0.560 |
asm | +0.518 |
ingu | +0.510 |
cius | +0.500 |
Neg logit contributions | |
deadliest | -0.692 |
losses | -0.686 |
victims | -0.674 |
occupants | -0.670 |
quotas | -0.661 |
Token | interesting |
Token ID | 3499 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
ude | +5.499 |
cerning | +5.464 |
asm | +5.163 |
ingu | +4.959 |
cius | +4.915 |
Neg logit contributions | |
deadliest | -4.582 |
losses | -4.487 |
victims | -4.413 |
occupants | -4.408 |
quotas | -4.387 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.21e+01 |
Pos logit contributions | |
ude | +10.587 |
cerning | +10.352 |
asm | +9.980 |
ingu | +9.407 |
clus | +9.337 |
Neg logit contributions | |
deadliest | -9.676 |
losses | -9.253 |
victims | -9.137 |
quotas | -9.092 |
occupants | -9.051 |
Token | to |
Token ID | 284 |
Feature activation | +2.10e+01 |
Pos logit contributions | |
cerning | +12.259 |
ude | +12.139 |
asm | +11.453 |
wn | +10.945 |
cius | +10.889 |
Neg logit contributions | |
deadliest | -9.162 |
losses | -8.694 |
occupants | -8.569 |
victims | -8.565 |
quotas | -8.498 |
Token | note |
Token ID | 3465 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
ude | +17.399 |
cerning | +17.066 |
asm | +15.839 |
dos | +15.403 |
summarize | +15.373 |
Neg logit contributions | |
deadliest | -14.832 |
quotas | -14.594 |
losses | -14.506 |
occupants | -14.060 |
invaders | -13.821 |
Token | is |
Token ID | 318 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
cerning | +11.245 |
ude | +10.982 |
asm | +10.372 |
ingu | +9.965 |
cius | +9.904 |
Neg logit contributions | |
deadliest | -8.924 |
occupants | -8.440 |
losses | -8.390 |
victims | -8.286 |
quotas | -8.144 |
Token | that |
Token ID | 326 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
cerning | +1.267 |
ude | +1.256 |
asm | +1.174 |
ingu | +1.159 |
cius | +1.136 |
Neg logit contributions | |
deadliest | -1.300 |
losses | -1.285 |
victims | -1.264 |
occupants | -1.255 |
quotas | -1.234 |
Token | it |
Token ID | 340 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
deadliest | +3.133 |
losses | +3.120 |
victims | +3.067 |
occupants | +3.026 |
quotas | +2.976 |
Neg logit contributions | |
cerning | -3.474 |
ude | -3.451 |
ingu | -3.223 |
asm | -3.207 |
cius | -3.145 |
Token | says |
Token ID | 1139 |
Feature activation | -7.15e+00 |
Pos logit contributions | |
deadliest | +3.665 |
losses | +3.664 |
victims | +3.575 |
occupants | +3.509 |
quotas | +3.447 |
Neg logit contributions | |
cerning | -5.694 |
ude | -5.630 |
ingu | -5.317 |
asm | -5.315 |
cius | -5.226 |
Token | is |
Token ID | 318 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ude | +0.450 |
cerning | +0.449 |
asm | +0.425 |
ingu | +0.423 |
cius | +0.414 |
Neg logit contributions | |
deadliest | -0.276 |
losses | -0.270 |
victims | -0.264 |
occupants | -0.263 |
quotas | -0.258 |
Token | easy |
Token ID | 2562 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
deadliest | +2.055 |
losses | +2.035 |
victims | +1.994 |
occupants | +1.969 |
quotas | +1.932 |
Neg logit contributions | |
cerning | -2.916 |
ude | -2.895 |
asm | -2.725 |
ingu | -2.717 |
cius | -2.669 |
Token | , |
Token ID | 11 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
cerning | +5.417 |
ude | +5.395 |
ingu | +5.104 |
asm | +5.096 |
cius | +4.967 |
Neg logit contributions | |
deadliest | -2.994 |
losses | -2.938 |
occupants | -2.851 |
victims | -2.845 |
quotas | -2.765 |
Token | then |
Token ID | 788 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
ude | +3.942 |
cerning | +3.931 |
asm | +3.736 |
ingu | +3.680 |
cius | +3.618 |
Neg logit contributions | |
deadliest | -2.472 |
losses | -2.425 |
occupants | -2.356 |
victims | -2.344 |
quotas | -2.297 |
Token | , |
Token ID | 11 |
Feature activation | +9.80e+00 |
Pos logit contributions | |
ude | +2.868 |
cerning | +2.863 |
asm | +2.630 |
ingu | +2.566 |
clus | +2.464 |
Neg logit contributions | |
deadliest | -4.602 |
losses | -4.525 |
occupants | -4.458 |
victims | -4.430 |
quotas | -4.340 |
Token | to |
Token ID | 284 |
Feature activation | +2.10e+01 |
Pos logit contributions | |
cerning | +10.869 |
ude | +10.800 |
ingu | +10.288 |
asm | +10.206 |
cius | +9.872 |
Neg logit contributions | |
deadliest | -6.760 |
losses | -6.593 |
occupants | -6.481 |
victims | -6.356 |
quotas | -6.241 |
Token | imagine |
Token ID | 5967 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
ude | +19.761 |
cerning | +18.595 |
asm | +17.991 |
summarize | +17.591 |
ingu | +17.556 |
Neg logit contributions | |
deadliest | -13.725 |
occupants | -12.852 |
quotas | -12.832 |
losses | -12.654 |
revenues | -12.060 |
Token | this |
Token ID | 428 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
cerning | +2.545 |
ude | +2.540 |
asm | +2.407 |
ingu | +2.364 |
clus | +2.310 |
Neg logit contributions | |
deadliest | -2.090 |
losses | -2.048 |
victims | -2.003 |
occupants | -1.993 |
quotas | -1.963 |
Token | ge |
Token ID | 4903 |
Feature activation | +6.88e+00 |
Pos logit contributions | |
deadliest | +0.498 |
losses | +0.493 |
victims | +0.484 |
occupants | +0.479 |
quotas | +0.471 |
Neg logit contributions | |
cerning | -0.584 |
ude | -0.581 |
asm | -0.544 |
ingu | -0.541 |
cius | -0.530 |
Token | ophysical |
Token ID | 41789 |
Feature activation | +5.93e+00 |
Pos logit contributions | |
ude | +4.104 |
cerning | +4.078 |
asm | +3.739 |
ingu | +3.610 |
clus | +3.486 |
Neg logit contributions | |
losses | -8.537 |
deadliest | -8.537 |
victims | -8.381 |
occupants | -8.324 |
quotas | -8.313 |
Token | boundary |
Token ID | 18645 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
ude | +6.051 |
cerning | +5.977 |
asm | +5.782 |
ingu | +5.602 |
clus | +5.439 |
Neg logit contributions | |
deadliest | -4.681 |
losses | -4.601 |
victims | -4.521 |
occupants | -4.474 |
quotas | -4.384 |
Token | � |
Token ID | 251 |
Feature activation | +7.77e+00 |
Pos logit contributions | |
ude | +6.487 |
cerning | +6.442 |
asm | +5.960 |
ingu | +5.874 |
cius | +5.735 |
Neg logit contributions | |
deadliest | -7.737 |
losses | -7.676 |
victims | -7.560 |
occupants | -7.502 |
quotas | -7.420 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.37e+00 |
Pos logit contributions | |
ude | +6.743 |
cerning | +6.684 |
asm | +6.223 |
ingu | +6.110 |
cius | +5.990 |
Neg logit contributions | |
deadliest | -7.382 |
losses | -7.275 |
occupants | -7.119 |
victims | -7.091 |
quotas | -7.008 |
Token | What |
Token ID | 2061 |
Feature activation | +5.12e+00 |
Pos logit contributions | |
ude | +7.499 |
cerning | +7.454 |
asm | +7.064 |
ingu | +6.891 |
clus | +6.724 |
Neg logit contributions | |
losses | -7.667 |
deadliest | -7.646 |
occupants | -7.417 |
victims | -7.408 |
quotas | -7.305 |
Token | better |
Token ID | 1365 |
Feature activation | +8.74e+00 |
Pos logit contributions | |
cerning | +5.460 |
ude | +5.437 |
asm | +5.184 |
ingu | +5.090 |
cius | +5.002 |
Neg logit contributions | |
deadliest | -3.835 |
losses | -3.746 |
occupants | -3.675 |
victims | -3.615 |
quotas | -3.598 |
Token | way |
Token ID | 835 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
ude | +9.318 |
cerning | +9.256 |
asm | +8.817 |
ingu | +8.647 |
yt | +8.495 |
Neg logit contributions | |
deadliest | -6.172 |
losses | -6.091 |
quotas | -5.940 |
occupants | -5.901 |
victims | -5.821 |
Token | to |
Token ID | 284 |
Feature activation | +2.09e+01 |
Pos logit contributions | |
ude | +12.261 |
cerning | +12.234 |
asm | +11.497 |
ingu | +11.288 |
cius | +11.069 |
Neg logit contributions | |
deadliest | -8.281 |
losses | -7.914 |
occupants | -7.832 |
quotas | -7.722 |
victims | -7.679 |
Token | raise |
Token ID | 5298 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
ude | +18.778 |
cerning | +17.842 |
asm | +17.421 |
explain | +17.141 |
summarize | +17.080 |
Neg logit contributions | |
deadliest | -13.538 |
quotas | -13.494 |
occupants | -13.236 |
losses | -13.142 |
victims | -12.433 |
Token | awareness |
Token ID | 9359 |
Feature activation | +1.41e+01 |
Pos logit contributions | |
deadliest | +0.062 |
losses | +0.061 |
victims | +0.060 |
occupants | +0.059 |
quotas | +0.057 |
Neg logit contributions | |
cerning | -0.134 |
ude | -0.134 |
asm | -0.127 |
ingu | -0.127 |
cius | -0.125 |
Token | of |
Token ID | 286 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
cerning | +12.933 |
ude | +12.582 |
asm | +11.499 |
ingu | +11.320 |
cius | +11.266 |
Neg logit contributions | |
deadliest | -11.643 |
occupants | -11.122 |
losses | -10.984 |
quotas | -10.842 |
safest | -10.815 |
Token | an |
Token ID | 281 |
Feature activation | +9.40e+00 |
Pos logit contributions | |
cerning | +11.423 |
ude | +11.419 |
asm | +10.882 |
ingu | +10.690 |
clus | +10.634 |
Neg logit contributions | |
deadliest | -6.875 |
losses | -6.772 |
occupants | -6.728 |
quotas | -6.421 |
victims | -6.402 |
Token | issue |
Token ID | 2071 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
cerning | +10.072 |
ude | +10.038 |
ingu | +9.485 |
asm | +9.411 |
cius | +9.247 |
Neg logit contributions | |
losses | -6.231 |
deadliest | -6.202 |
quotas | -6.161 |
occupants | -6.154 |
victims | -5.935 |
Token | name |
Token ID | 1438 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
ude | +10.828 |
cerning | +10.735 |
asm | +10.159 |
ingu | +10.125 |
cius | +9.836 |
Neg logit contributions | |
deadliest | -7.245 |
losses | -7.230 |
occupants | -7.063 |
victims | -6.988 |
quotas | -6.972 |
Token | with |
Token ID | 351 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
cerning | +4.048 |
ude | +4.032 |
asm | +3.807 |
ingu | +3.777 |
cius | +3.692 |
Neg logit contributions | |
deadliest | -3.535 |
losses | -3.487 |
victims | -3.403 |
occupants | -3.390 |
quotas | -3.345 |
Token | ausp |
Token ID | 45062 |
Feature activation | +8.56e+00 |
Pos logit contributions | |
cerning | +3.197 |
ude | +3.190 |
asm | +3.029 |
ingu | +2.994 |
cius | +2.933 |
Neg logit contributions | |
deadliest | -1.965 |
losses | -1.922 |
victims | -1.875 |
occupants | -1.860 |
quotas | -1.832 |
Token | icious |
Token ID | 6243 |
Feature activation | +6.79e+00 |
Pos logit contributions | |
ude | +5.940 |
cerning | +5.939 |
asm | +5.507 |
ingu | +5.387 |
cius | +5.287 |
Neg logit contributions | |
deadliest | -9.622 |
losses | -9.538 |
victims | -9.274 |
quotas | -9.273 |
occupants | -9.230 |
Token | con |
Token ID | 369 |
Feature activation | +9.23e+00 |
Pos logit contributions | |
ude | +7.685 |
cerning | +7.519 |
asm | +7.193 |
ingu | +6.992 |
cius | +6.933 |
Neg logit contributions | |
deadliest | -4.697 |
losses | -4.420 |
victims | -4.283 |
quotas | -4.278 |
safest | -4.209 |
Token | notations |
Token ID | 30078 |
Feature activation | +2.09e+01 |
Pos logit contributions | |
cerning | +5.022 |
ude | +4.811 |
asm | +4.269 |
clus | +4.241 |
ingu | +4.172 |
Neg logit contributions | |
deadliest | -11.965 |
losses | -11.839 |
victims | -11.621 |
quotas | -11.568 |
occupants | -11.518 |
Token | being |
Token ID | 852 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
ude | +19.732 |
cerning | +19.592 |
asm | +18.790 |
ingu | +18.539 |
wn | +18.389 |
Neg logit contributions | |
deadliest | -11.968 |
losses | -11.212 |
occupants | -10.954 |
quotas | -10.844 |
victims | -10.644 |
Token | thought |
Token ID | 1807 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
deadliest | +0.578 |
losses | +0.573 |
victims | +0.559 |
occupants | +0.553 |
quotas | +0.544 |
Neg logit contributions | |
cerning | -0.868 |
ude | -0.863 |
asm | -0.814 |
ingu | -0.812 |
cius | -0.796 |
Token | to |
Token ID | 284 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
deadliest | +1.168 |
losses | +1.156 |
victims | +1.135 |
occupants | +1.118 |
quotas | +1.096 |
Neg logit contributions | |
cerning | -1.620 |
ude | -1.607 |
asm | -1.513 |
ingu | -1.507 |
cius | -1.485 |
Token | bring |
Token ID | 2222 |
Feature activation | -6.81e+00 |
Pos logit contributions | |
ude | +11.266 |
cerning | +10.795 |
ingu | +10.249 |
asm | +10.192 |
clus | +9.831 |
Neg logit contributions | |
deadliest | -9.713 |
losses | -9.543 |
occupants | -9.385 |
quotas | -9.334 |
victims | -9.250 |
Token | good |
Token ID | 922 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
losses | +5.496 |
deadliest | +5.484 |
victims | +5.433 |
occupants | +5.331 |
quotas | +5.207 |
Neg logit contributions | |
cerning | -7.035 |
ude | -6.893 |
ingu | -6.530 |
asm | -6.479 |
cius | -6.449 |
Token | give |
Token ID | 1577 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
ude | +11.208 |
cerning | +10.593 |
asm | +10.103 |
ingu | +10.094 |
apply | +9.854 |
Neg logit contributions | |
losses | -9.398 |
deadliest | -9.365 |
occupants | -9.104 |
victims | -8.909 |
quotas | -8.841 |
Token | you |
Token ID | 345 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
deadliest | +2.839 |
losses | +2.819 |
victims | +2.777 |
occupants | +2.741 |
quotas | +2.702 |
Neg logit contributions | |
cerning | -3.051 |
ude | -3.017 |
asm | -2.820 |
ingu | -2.813 |
cius | -2.758 |
Token | some |
Token ID | 617 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
cerning | +0.138 |
ude | +0.138 |
asm | +0.131 |
ingu | +0.130 |
cius | +0.128 |
Neg logit contributions | |
deadliest | -0.071 |
losses | -0.070 |
victims | -0.068 |
occupants | -0.067 |
quotas | -0.065 |
Token | truthful |
Token ID | 42278 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
ude | +10.673 |
cerning | +10.532 |
asm | +9.865 |
ingu | +9.636 |
yt | +9.591 |
Neg logit contributions | |
deadliest | -6.825 |
losses | -6.663 |
occupants | -6.589 |
victims | -6.433 |
quotas | -6.248 |
Token | euphem |
Token ID | 48732 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
ude | +11.074 |
cerning | +10.687 |
asm | +10.149 |
yt | +10.099 |
ingu | +9.948 |
Neg logit contributions | |
deadliest | -6.513 |
losses | -6.169 |
occupants | -6.076 |
victims | -5.993 |
quotas | -5.871 |
Token | isms |
Token ID | 6583 |
Feature activation | +2.09e+01 |
Pos logit contributions | |
ude | +6.172 |
cerning | +5.935 |
asm | +5.575 |
ingu | +5.221 |
iku | +4.888 |
Neg logit contributions | |
deadliest | -15.097 |
losses | -14.709 |
occupants | -14.669 |
victims | -14.495 |
quotas | -14.391 |
Token | , |
Token ID | 11 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
ude | +18.731 |
cerning | +18.180 |
asm | +17.210 |
ingu | +16.886 |
ás | +16.714 |
Neg logit contributions | |
deadliest | -13.489 |
losses | -12.964 |
occupants | -12.617 |
quotas | -12.447 |
riots | -12.122 |
Token | � |
Token ID | 447 |
Feature activation | +7.99e-01 |
Pos logit contributions | |
ude | +8.472 |
cerning | +8.264 |
asm | +7.814 |
cius | +7.632 |
ingu | +7.627 |
Neg logit contributions | |
deadliest | -9.255 |
losses | -9.159 |
occupants | -8.904 |
victims | -8.825 |
quotas | -8.605 |
Token | � |
Token ID | 251 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
cerning | +0.586 |
ude | +0.586 |
asm | +0.536 |
ingu | +0.527 |
cius | +0.515 |
Neg logit contributions | |
deadliest | -0.898 |
losses | -0.890 |
victims | -0.879 |
occupants | -0.872 |
quotas | -0.859 |
Token | offers |
Token ID | 4394 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
deadliest | +0.234 |
losses | +0.232 |
victims | +0.228 |
occupants | +0.226 |
quotas | +0.222 |
Neg logit contributions | |
cerning | -0.231 |
ude | -0.229 |
asm | -0.213 |
ingu | -0.212 |
cius | -0.208 |
Token | Ra |
Token ID | 7567 |
Feature activation | +6.16e+00 |
Pos logit contributions | |
cerning | +2.985 |
ude | +2.968 |
asm | +2.763 |
ingu | +2.736 |
cius | +2.676 |
Neg logit contributions | |
deadliest | -3.449 |
losses | -3.431 |
victims | -3.348 |
occupants | -3.340 |
quotas | -3.291 |
Token | It |
Token ID | 1026 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
cerning | +0.858 |
ude | +0.850 |
asm | +0.802 |
ingu | +0.787 |
cius | +0.771 |
Neg logit contributions | |
deadliest | -0.839 |
losses | -0.835 |
victims | -0.819 |
occupants | -0.813 |
quotas | -0.800 |
Token | � |
Token ID | 447 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
deadliest | +0.328 |
losses | +0.322 |
victims | +0.316 |
occupants | +0.313 |
quotas | +0.307 |
Neg logit contributions | |
cerning | -0.513 |
ude | -0.511 |
asm | -0.483 |
ingu | -0.481 |
cius | -0.472 |
Token | � |
Token ID | 247 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
ude | +1.387 |
cerning | +1.383 |
asm | +1.265 |
ingu | +1.252 |
cius | +1.213 |
Neg logit contributions | |
deadliest | -2.128 |
losses | -2.115 |
victims | -2.091 |
occupants | -2.074 |
quotas | -2.047 |
Token | s |
Token ID | 82 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
deadliest | +0.282 |
losses | +0.279 |
victims | +0.274 |
occupants | +0.272 |
quotas | +0.268 |
Neg logit contributions | |
cerning | -0.303 |
ude | -0.302 |
asm | -0.282 |
ingu | -0.280 |
cius | -0.274 |
Token | interesting |
Token ID | 3499 |
Feature activation | +9.54e+00 |
Pos logit contributions | |
deadliest | +3.354 |
losses | +3.318 |
victims | +3.244 |
occupants | +3.193 |
quotas | +3.129 |
Neg logit contributions | |
cerning | -4.972 |
ude | -4.917 |
ingu | -4.641 |
asm | -4.626 |
cius | -4.550 |
Token | to |
Token ID | 284 |
Feature activation | +2.08e+01 |
Pos logit contributions | |
cerning | +10.171 |
ude | +10.053 |
asm | +9.759 |
ingu | +9.539 |
cius | +9.258 |
Neg logit contributions | |
deadliest | -6.925 |
losses | -6.697 |
occupants | -6.627 |
victims | -6.617 |
quotas | -6.395 |
Token | compare |
Token ID | 8996 |
Feature activation | +4.40e+00 |
Pos logit contributions | |
ude | +17.876 |
asm | +16.957 |
cerning | +16.939 |
summarize | +16.338 |
ingu | +16.233 |
Neg logit contributions | |
deadliest | -14.320 |
quotas | -13.996 |
losses | -13.796 |
occupants | -13.776 |
victims | -13.313 |
Token | this |
Token ID | 428 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
cerning | +4.368 |
ude | +4.346 |
asm | +4.183 |
ingu | +4.037 |
cius | +3.977 |
Neg logit contributions | |
deadliest | -3.669 |
losses | -3.577 |
victims | -3.546 |
occupants | -3.532 |
quotas | -3.463 |
Token | to |
Token ID | 284 |
Feature activation | +9.57e+00 |
Pos logit contributions | |
deadliest | +0.213 |
losses | +0.211 |
victims | +0.207 |
occupants | +0.205 |
quotas | +0.202 |
Neg logit contributions | |
cerning | -0.310 |
ude | -0.308 |
asm | -0.291 |
ingu | -0.289 |
cius | -0.284 |
Token | the |
Token ID | 262 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
cerning | +9.601 |
ude | +9.592 |
asm | +9.378 |
ingu | +8.935 |
cius | +8.801 |
Neg logit contributions | |
deadliest | -7.051 |
losses | -6.894 |
occupants | -6.871 |
victims | -6.793 |
quotas | -6.636 |
Token | standard |
Token ID | 3210 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ude | +2.413 |
cerning | +2.410 |
asm | +2.280 |
ingu | +2.236 |
cius | +2.211 |
Neg logit contributions | |
deadliest | -1.871 |
losses | -1.852 |
victims | -1.822 |
occupants | -1.810 |
quotas | -1.785 |
Token | Perhaps |
Token ID | 13710 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
cerning | +1.456 |
ude | +1.448 |
asm | +1.345 |
ingu | +1.327 |
cius | +1.289 |
Neg logit contributions | |
deadliest | -1.691 |
losses | -1.676 |
victims | -1.643 |
occupants | -1.634 |
quotas | -1.611 |
Token | with |
Token ID | 351 |
Feature activation | +4.76e+00 |
Pos logit contributions | |
ude | +3.257 |
cerning | +3.256 |
asm | +3.112 |
ingu | +3.039 |
cius | +2.959 |
Neg logit contributions | |
deadliest | -2.322 |
losses | -2.282 |
victims | -2.213 |
occupants | -2.212 |
quotas | -2.187 |
Token | some |
Token ID | 617 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
cerning | +5.094 |
ude | +5.079 |
asm | +4.850 |
ingu | +4.730 |
clus | +4.615 |
Neg logit contributions | |
deadliest | -3.614 |
losses | -3.532 |
occupants | -3.493 |
victims | -3.475 |
quotas | -3.427 |
Token | apolog |
Token ID | 8453 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
ude | +14.061 |
cerning | +13.620 |
asm | +13.275 |
ingu | +12.900 |
lex | +12.448 |
Neg logit contributions | |
deadliest | -7.763 |
occupants | -7.190 |
losses | -7.061 |
quotas | -7.000 |
victims | -6.960 |
Token | etic |
Token ID | 5139 |
Feature activation | +1.62e+01 |
Pos logit contributions | |
ude | +9.425 |
cerning | +9.218 |
asm | +8.497 |
ingu | +8.302 |
clus | +8.080 |
Neg logit contributions | |
deadliest | -13.211 |
losses | -12.803 |
quotas | -12.537 |
victims | -12.524 |
occupants | -12.514 |
Token | intent |
Token ID | 6824 |
Feature activation | +2.07e+01 |
Pos logit contributions | |
ude | +17.165 |
cerning | +16.268 |
asm | +16.137 |
ingu | +15.459 |
clus | +15.062 |
Neg logit contributions | |
deadliest | -9.380 |
occupants | -8.443 |
victims | -8.189 |
quotas | -8.175 |
losses | -8.047 |
Token | , |
Token ID | 11 |
Feature activation | +8.30e+00 |
Pos logit contributions | |
ude | +20.675 |
cerning | +20.131 |
asm | +19.358 |
ingu | +18.998 |
clus | +18.461 |
Neg logit contributions | |
deadliest | -11.305 |
quotas | -10.243 |
losses | -10.214 |
safest | -10.069 |
occupants | -9.951 |
Token | the |
Token ID | 262 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ude | +8.579 |
cerning | +8.422 |
asm | +8.103 |
ingu | +7.880 |
cius | +7.636 |
Neg logit contributions | |
deadliest | -6.313 |
losses | -6.150 |
occupants | -5.901 |
quotas | -5.861 |
victims | -5.848 |
Token | late |
Token ID | 2739 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
deadliest | +1.604 |
losses | +1.590 |
victims | +1.563 |
occupants | +1.545 |
quotas | +1.519 |
Neg logit contributions | |
cerning | -1.802 |
ude | -1.787 |
asm | -1.671 |
ingu | -1.666 |
cius | -1.633 |
Token | English |
Token ID | 3594 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
deadliest | +1.480 |
losses | +1.468 |
victims | +1.448 |
occupants | +1.432 |
quotas | +1.407 |
Neg logit contributions | |
cerning | -1.343 |
ude | -1.327 |
asm | -1.230 |
ingu | -1.230 |
cius | -1.203 |
Token | scholar |
Token ID | 15606 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
deadliest | +3.429 |
losses | +3.399 |
victims | +3.366 |
occupants | +3.316 |
quotas | +3.241 |
Neg logit contributions | |
cerning | -3.047 |
ude | -2.999 |
asm | -2.780 |
ingu | -2.766 |
cius | -2.734 |
Token | large |
Token ID | 1588 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
cerning | +2.051 |
ude | +2.039 |
asm | +1.957 |
ingu | +1.888 |
clus | +1.830 |
Neg logit contributions | |
deadliest | -1.509 |
losses | -1.488 |
occupants | -1.450 |
victims | -1.442 |
quotas | -1.405 |
Token | bowl |
Token ID | 9396 |
Feature activation | -9.48e-01 |
Pos logit contributions | |
deadliest | +0.546 |
losses | +0.540 |
victims | +0.530 |
occupants | +0.528 |
quotas | +0.515 |
Neg logit contributions | |
cerning | -0.699 |
ude | -0.694 |
asm | -0.656 |
ingu | -0.648 |
cius | -0.633 |
Token | , |
Token ID | 11 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
deadliest | +1.000 |
losses | +0.995 |
victims | +0.981 |
occupants | +0.969 |
quotas | +0.959 |
Neg logit contributions | |
cerning | -0.765 |
ude | -0.755 |
asm | -0.695 |
ingu | -0.693 |
cius | -0.678 |
Token | combine |
Token ID | 12082 |
Feature activation | -1.63e+01 |
Pos logit contributions | |
deadliest | +3.652 |
losses | +3.619 |
victims | +3.607 |
occupants | +3.509 |
quotas | +3.480 |
Neg logit contributions | |
cerning | -2.771 |
ude | -2.678 |
asm | -2.472 |
cius | -2.465 |
ingu | -2.459 |
Token | the |
Token ID | 262 |
Feature activation | -1.09e+01 |
Pos logit contributions | |
losses | +8.907 |
victims | +8.717 |
only | +8.705 |
deadliest | +8.454 |
occupants | +8.194 |
Neg logit contributions | |
cerning | -18.564 |
ude | -17.641 |
ingu | -17.545 |
cius | -17.484 |
cipled | -17.316 |
Token | broccoli |
Token ID | 44653 |
Feature activation | -2.54e+01 |
Pos logit contributions | |
losses | +7.642 |
deadliest | +7.565 |
victims | +7.563 |
occupants | +7.373 |
quotas | +7.159 |
Neg logit contributions | |
cerning | -11.795 |
ude | -11.506 |
ingu | -11.309 |
asm | -10.968 |
cius | -10.914 |
Token | pieces |
Token ID | 5207 |
Feature activation | -1.49e+01 |
Pos logit contributions | |
victims | +12.359 |
losses | +11.989 |
only | +11.396 |
remnants | +11.202 |
owners | +11.194 |
Neg logit contributions | |
cius | -28.037 |
ingu | -27.724 |
cerning | -27.436 |
ude | -27.156 |
uggest | -26.776 |
Token | with |
Token ID | 351 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
losses | +6.780 |
only | +6.759 |
victims | +6.661 |
closest | +6.435 |
deadliest | +6.374 |
Neg logit contributions | |
cerning | -19.047 |
ude | -17.848 |
ingu | -17.789 |
cius | -17.755 |
cipled | -17.485 |
Token | all |
Token ID | 477 |
Feature activation | -7.99e+00 |
Pos logit contributions | |
losses | +5.546 |
deadliest | +5.425 |
victims | +5.408 |
only | +5.100 |
occupants | +5.090 |
Neg logit contributions | |
cerning | -12.725 |
ude | -12.412 |
ingu | -12.043 |
cius | -11.954 |
asm | -11.757 |
Token | the |
Token ID | 262 |
Feature activation | -9.40e+00 |
Pos logit contributions | |
losses | +5.523 |
victims | +5.406 |
deadliest | +5.343 |
occupants | +5.210 |
quotas | +5.197 |
Neg logit contributions | |
cerning | -9.028 |
ude | -8.688 |
ingu | -8.487 |
cius | -8.475 |
asm | -8.300 |
Token | other |
Token ID | 584 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
losses | +6.631 |
victims | +6.530 |
deadliest | +6.446 |
occupants | +6.312 |
quotas | +6.113 |
Neg logit contributions | |
cerning | -10.503 |
ude | -10.181 |
ingu | -9.920 |
cius | -9.876 |
asm | -9.758 |
Token | the |
Token ID | 262 |
Feature activation | -7.59e+00 |
Pos logit contributions | |
deadliest | +2.010 |
losses | +2.005 |
victims | +1.960 |
occupants | +1.930 |
quotas | +1.910 |
Neg logit contributions | |
cerning | -2.825 |
ude | -2.799 |
ingu | -2.631 |
asm | -2.627 |
cius | -2.589 |
Token | question |
Token ID | 1808 |
Feature activation | -6.11e+00 |
Pos logit contributions | |
losses | +6.083 |
deadliest | +6.042 |
victims | +5.933 |
quotas | +5.850 |
occupants | +5.805 |
Neg logit contributions | |
cerning | -7.902 |
ude | -7.794 |
ingu | -7.422 |
asm | -7.289 |
cius | -7.273 |
Token | of |
Token ID | 286 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
losses | +6.402 |
deadliest | +6.343 |
victims | +6.321 |
quotas | +6.199 |
occupants | +6.197 |
Neg logit contributions | |
cerning | -4.803 |
ude | -4.734 |
ingu | -4.410 |
asm | -4.342 |
cius | -4.232 |
Token | whether |
Token ID | 1771 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
losses | +3.986 |
deadliest | +3.952 |
victims | +3.890 |
quotas | +3.837 |
occupants | +3.822 |
Neg logit contributions | |
cerning | -4.865 |
ude | -4.805 |
ingu | -4.522 |
asm | -4.475 |
cius | -4.447 |
Token | the |
Token ID | 262 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
losses | +7.808 |
quotas | +7.638 |
victims | +7.591 |
occupants | +7.388 |
deadliest | +7.383 |
Neg logit contributions | |
cerning | -11.572 |
ude | -11.269 |
ingu | -10.865 |
asm | -10.576 |
cius | -10.540 |
Token | UK |
Token ID | 3482 |
Feature activation | -2.40e+01 |
Pos logit contributions | |
losses | +9.494 |
victims | +9.252 |
quotas | +9.156 |
deadliest | +9.154 |
occupants | +8.998 |
Neg logit contributions | |
cerning | -12.032 |
ude | -11.632 |
ingu | -11.278 |
asm | -10.940 |
cius | -10.896 |
Token | remains |
Token ID | 3793 |
Feature activation | -2.04e+01 |
Pos logit contributions | |
losses | +16.640 |
authorities | +15.767 |
victims | +15.428 |
quotas | +15.355 |
only | +15.256 |
Neg logit contributions | |
cerning | -22.457 |
ude | -20.772 |
uggest | -19.959 |
utics | -19.818 |
dos | -19.674 |
Token | in |
Token ID | 287 |
Feature activation | -6.65e+00 |
Pos logit contributions | |
victims | +13.492 |
losses | +13.079 |
only | +12.873 |
occupants | +12.861 |
quotas | +12.829 |
Neg logit contributions | |
cerning | -20.534 |
ude | -19.408 |
ingu | -18.799 |
assi | -18.534 |
cius | -18.338 |
Token | the |
Token ID | 262 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
losses | +4.765 |
victims | +4.697 |
deadliest | +4.684 |
quotas | +4.628 |
occupants | +4.524 |
Neg logit contributions | |
cerning | -7.397 |
ude | -7.062 |
ingu | -6.789 |
cius | -6.777 |
asm | -6.731 |
Token | EU |
Token ID | 4576 |
Feature activation | -1.66e+01 |
Pos logit contributions | |
losses | +7.198 |
victims | +7.184 |
deadliest | +7.177 |
quotas | +7.065 |
closest | +6.857 |
Neg logit contributions | |
cerning | -11.757 |
ude | -11.348 |
ingu | -11.007 |
cius | -10.914 |
asm | -10.782 |
Token | - |
Token ID | 532 |
Feature activation | -6.97e+00 |
Pos logit contributions | |
losses | +14.176 |
quotas | +14.173 |
victims | +14.075 |
only | +13.982 |
deadliest | +13.737 |
Neg logit contributions | |
cerning | -12.983 |
ude | -12.317 |
dos | -11.808 |
ingu | -11.766 |
clus | -11.598 |
Token | until |
Token ID | 1566 |
Feature activation | -6.55e+00 |
Pos logit contributions | |
losses | +6.095 |
deadliest | +6.069 |
victims | +5.947 |
occupants | +5.842 |
quotas | +5.724 |
Neg logit contributions | |
cerning | -9.073 |
ude | -8.909 |
ingu | -8.467 |
asm | -8.385 |
cius | -8.364 |
Token | the |
Token ID | 262 |
Feature activation | -8.34e+00 |
Pos logit contributions | |
losses | +5.133 |
deadliest | +5.133 |
victims | +5.029 |
occupants | +4.954 |
quotas | +4.821 |
Neg logit contributions | |
cerning | -6.993 |
ude | -6.897 |
ingu | -6.509 |
asm | -6.477 |
cius | -6.394 |
Token | spacecraft |
Token ID | 16807 |
Feature activation | -1.55e+01 |
Pos logit contributions | |
deadliest | +7.123 |
losses | +7.121 |
victims | +7.007 |
occupants | +6.929 |
closest | +6.695 |
Neg logit contributions | |
cerning | -8.217 |
ude | -8.090 |
ingu | -7.626 |
asm | -7.561 |
cius | -7.464 |
Token | impacted |
Token ID | 20755 |
Feature activation | -1.67e+01 |
Pos logit contributions | |
losses | +11.775 |
deadliest | +11.508 |
occupants | +11.458 |
victims | +11.414 |
closest | +11.068 |
Neg logit contributions | |
cerning | -15.303 |
ude | -14.797 |
ingu | -14.004 |
cius | -13.918 |
asm | -13.810 |
Token | the |
Token ID | 262 |
Feature activation | -9.98e+00 |
Pos logit contributions | |
losses | +13.649 |
deadliest | +13.632 |
victims | +13.554 |
closest | +13.440 |
only | +13.203 |
Neg logit contributions | |
cerning | -14.855 |
ude | -14.119 |
ingu | -13.536 |
cius | -13.440 |
apply | -13.110 |
Token | lunar |
Token ID | 25572 |
Feature activation | -2.39e+01 |
Pos logit contributions | |
deadliest | +7.788 |
losses | +7.574 |
victims | +7.535 |
closest | +7.360 |
occupants | +7.348 |
Neg logit contributions | |
cerning | -10.418 |
ude | -10.186 |
ingu | -9.726 |
asm | -9.552 |
cius | -9.424 |
Token | surface |
Token ID | 4417 |
Feature activation | -1.18e+01 |
Pos logit contributions | |
deadliest | +17.162 |
victims | +17.077 |
closest | +17.035 |
losses | +16.943 |
authorities | +16.688 |
Neg logit contributions | |
cerning | -18.955 |
ude | -17.802 |
ingu | -17.391 |
clus | -17.314 |
utics | -17.299 |
Token | at |
Token ID | 379 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
losses | +11.257 |
deadliest | +11.144 |
victims | +11.104 |
occupants | +10.952 |
closest | +10.794 |
Neg logit contributions | |
cerning | -9.781 |
ude | -9.401 |
cius | -8.876 |
ingu | -8.837 |
asm | -8.713 |
Token | 2 |
Token ID | 362 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
deadliest | +2.425 |
losses | +2.400 |
victims | +2.345 |
occupants | +2.313 |
quotas | +2.267 |
Neg logit contributions | |
cerning | -3.962 |
ude | -3.934 |
asm | -3.715 |
ingu | -3.707 |
cius | -3.643 |
Token | , |
Token ID | 11 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
losses | +3.510 |
deadliest | +3.500 |
victims | +3.452 |
occupants | +3.406 |
quotas | +3.371 |
Neg logit contributions | |
cerning | -2.698 |
ude | -2.653 |
ingu | -2.451 |
asm | -2.443 |
cius | -2.389 |
Token | 600 |
Token ID | 8054 |
Feature activation | -6.14e+00 |
Pos logit contributions | |
deadliest | +3.007 |
losses | +2.997 |
victims | +2.949 |
occupants | +2.914 |
quotas | +2.876 |
Neg logit contributions | |
cerning | -2.594 |
ude | -2.550 |
asm | -2.378 |
ingu | -2.366 |
cius | -2.309 |
Token | had |
Token ID | 550 |
Feature activation | -8.04e+00 |
Pos logit contributions | |
losses | +11.017 |
victims | +11.005 |
deadliest | +10.921 |
occupants | +10.544 |
authorities | +10.419 |
Neg logit contributions | |
cerning | -9.803 |
ude | -9.498 |
asm | -8.839 |
ingu | -8.747 |
cius | -8.657 |
Token | s |
Token ID | 264 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
losses | +5.448 |
deadliest | +5.411 |
victims | +5.318 |
occupants | +5.159 |
quotas | +5.059 |
Neg logit contributions | |
cerning | -9.317 |
ude | -9.205 |
ingu | -8.713 |
asm | -8.693 |
cius | -8.628 |
Token | oured |
Token ID | 8167 |
Feature activation | -6.62e+00 |
Pos logit contributions | |
cerning | +0.842 |
ude | +0.840 |
asm | +0.765 |
ingu | +0.758 |
cius | +0.730 |
Neg logit contributions | |
deadliest | -1.375 |
losses | -1.367 |
victims | -1.348 |
occupants | -1.341 |
quotas | -1.327 |
Token | . |
Token ID | 13 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
losses | +6.867 |
deadliest | +6.844 |
victims | +6.753 |
occupants | +6.616 |
quotas | +6.552 |
Neg logit contributions | |
cerning | -5.272 |
ude | -5.185 |
asm | -4.781 |
cius | -4.760 |
ingu | -4.741 |
Token | The |
Token ID | 383 |
Feature activation | -7.95e+00 |
Pos logit contributions | |
deadliest | +2.836 |
losses | +2.826 |
victims | +2.783 |
occupants | +2.742 |
quotas | +2.721 |
Neg logit contributions | |
cerning | -2.406 |
ude | -2.370 |
ingu | -2.201 |
asm | -2.190 |
cius | -2.148 |
Token | Libyan |
Token ID | 28661 |
Feature activation | -2.39e+01 |
Pos logit contributions | |
losses | +6.763 |
deadliest | +6.725 |
victims | +6.589 |
occupants | +6.409 |
quotas | +6.345 |
Neg logit contributions | |
cerning | -7.952 |
ude | -7.778 |
ingu | -7.416 |
asm | -7.270 |
cius | -7.256 |
Token | leader |
Token ID | 3554 |
Feature activation | -9.35e+00 |
Pos logit contributions | |
losses | +21.952 |
authorities | +21.823 |
victims | +21.646 |
revenues | +21.054 |
deadliest | +20.932 |
Neg logit contributions | |
cerning | -15.502 |
ingu | -14.021 |
cius | -13.785 |
ude | -13.759 |
utics | -13.645 |
Token | demanded |
Token ID | 12284 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
losses | +8.386 |
deadliest | +8.362 |
victims | +8.277 |
occupants | +8.048 |
quotas | +7.959 |
Neg logit contributions | |
cerning | -8.562 |
ude | -8.338 |
ingu | -7.784 |
asm | -7.704 |
cius | -7.624 |
Token | tough |
Token ID | 5802 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
losses | +9.846 |
deadliest | +9.650 |
victims | +9.616 |
quotas | +9.558 |
revenues | +9.415 |
Neg logit contributions | |
cerning | -10.682 |
ude | -10.257 |
cius | -9.862 |
ingu | -9.723 |
asm | -9.559 |
Token | contract |
Token ID | 2775 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
deadliest | +0.874 |
losses | +0.869 |
victims | +0.844 |
occupants | +0.827 |
quotas | +0.815 |
Neg logit contributions | |
cerning | -1.821 |
ude | -1.807 |
asm | -1.716 |
ingu | -1.711 |
cius | -1.689 |
Token | terms |
Token ID | 2846 |
Feature activation | +6.08e+00 |
Pos logit contributions | |
losses | +3.374 |
deadliest | +3.316 |
victims | +3.262 |
quotas | +3.220 |
occupants | +3.166 |
Neg logit contributions | |
cerning | -4.861 |
ude | -4.780 |
asm | -4.539 |
cius | -4.492 |
ingu | -4.484 |
Token | � |
Token ID | 447 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
losses | +4.538 |
deadliest | +4.525 |
victims | +4.429 |
occupants | +4.344 |
quotas | +4.287 |
Neg logit contributions | |
cerning | -6.152 |
ude | -6.046 |
asm | -5.705 |
ingu | -5.668 |
cius | -5.595 |
Token | � |
Token ID | 247 |
Feature activation | -7.69e+00 |
Pos logit contributions | |
losses | +5.653 |
deadliest | +5.646 |
victims | +5.515 |
occupants | +5.464 |
quotas | +5.403 |
Neg logit contributions | |
cerning | -4.602 |
ude | -4.375 |
ingu | -4.115 |
asm | -4.058 |
clus | -4.051 |
Token | s |
Token ID | 82 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
losses | +5.657 |
deadliest | +5.566 |
victims | +5.505 |
occupants | +5.318 |
quotas | +5.315 |
Neg logit contributions | |
cerning | -8.267 |
ude | -7.954 |
ingu | -7.890 |
asm | -7.599 |
cius | -7.594 |
Token | got |
Token ID | 1392 |
Feature activation | -1.43e+01 |
Pos logit contributions | |
deadliest | +7.874 |
losses | +7.842 |
victims | +7.734 |
occupants | +7.488 |
quotas | +7.289 |
Neg logit contributions | |
cerning | -10.846 |
ude | -10.603 |
ingu | -10.007 |
asm | -9.884 |
cius | -9.818 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
losses | +11.112 |
victims | +11.017 |
deadliest | +10.962 |
occupants | +10.524 |
closest | +10.321 |
Neg logit contributions | |
cerning | -14.320 |
ude | -13.979 |
ingu | -13.213 |
cius | -13.023 |
asm | -12.989 |
Token | Republican |
Token ID | 3415 |
Feature activation | -2.35e+01 |
Pos logit contributions | |
deadliest | +8.987 |
losses | +8.869 |
victims | +8.752 |
occupants | +8.493 |
closest | +8.431 |
Neg logit contributions | |
cerning | -10.746 |
ude | -10.588 |
ingu | -10.009 |
cius | -9.792 |
asm | -9.693 |
Token | nomination |
Token ID | 11872 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
losses | +17.498 |
victims | +17.211 |
deadliest | +17.056 |
occupants | +16.956 |
closest | +16.915 |
Neg logit contributions | |
cerning | -20.061 |
ude | -19.024 |
utics | -18.372 |
ingu | -18.093 |
cius | -17.499 |
Token | locked |
Token ID | 8970 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
losses | +13.286 |
deadliest | +13.076 |
victims | +13.036 |
only | +12.756 |
occupants | +12.711 |
Neg logit contributions | |
cerning | -11.509 |
ude | -11.098 |
ingu | -10.349 |
asm | -10.194 |
cius | -10.126 |
Token | up |
Token ID | 510 |
Feature activation | -6.06e+00 |
Pos logit contributions | |
losses | +8.609 |
deadliest | +8.526 |
victims | +8.429 |
occupants | +8.223 |
only | +8.009 |
Neg logit contributions | |
cerning | -13.473 |
ude | -13.293 |
ingu | -12.812 |
asm | -12.790 |
cius | -12.465 |
Token | and |
Token ID | 290 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
losses | +6.208 |
deadliest | +6.194 |
victims | +6.097 |
occupants | +6.001 |
quotas | +5.924 |
Neg logit contributions | |
cerning | -4.966 |
ude | -4.886 |
asm | -4.520 |
ingu | -4.516 |
cius | -4.414 |
Token | is |
Token ID | 318 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
deadliest | +4.890 |
losses | +4.875 |
victims | +4.771 |
occupants | +4.690 |
quotas | +4.620 |
Neg logit contributions | |
cerning | -6.533 |
ude | -6.457 |
ingu | -6.057 |
asm | -6.038 |
cius | -5.949 |
Token |
|
Token ID | 198 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
losses | +3.162 |
deadliest | +3.151 |
victims | +3.110 |
occupants | +3.049 |
quotas | +3.038 |
Neg logit contributions | |
cerning | -3.123 |
ude | -3.071 |
ingu | -2.889 |
asm | -2.850 |
cius | -2.801 |
Token |
|
Token ID | 198 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
victims | +8.796 |
losses | +8.747 |
only | +8.725 |
deadliest | +8.479 |
quotas | +8.462 |
Neg logit contributions | |
cerning | -13.018 |
ingu | -12.496 |
ude | -12.440 |
cius | -12.382 |
uggest | -12.064 |
Token | He |
Token ID | 1544 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
deadliest | +4.456 |
losses | +4.443 |
victims | +4.382 |
occupants | +4.295 |
quotas | +4.283 |
Neg logit contributions | |
cerning | -5.379 |
ude | -5.286 |
ingu | -4.992 |
asm | -4.914 |
cius | -4.880 |
Token | wants |
Token ID | 3382 |
Feature activation | -1.09e+01 |
Pos logit contributions | |
deadliest | +0.143 |
losses | +0.139 |
victims | +0.136 |
occupants | +0.136 |
quotas | +0.133 |
Neg logit contributions | |
ude | -0.211 |
cerning | -0.210 |
asm | -0.198 |
ingu | -0.198 |
cius | -0.194 |
Token | the |
Token ID | 262 |
Feature activation | -9.66e+00 |
Pos logit contributions | |
losses | +7.898 |
victims | +7.759 |
quotas | +7.593 |
deadliest | +7.579 |
occupants | +7.490 |
Neg logit contributions | |
cerning | -11.861 |
ude | -11.598 |
ingu | -11.097 |
cius | -10.900 |
asm | -10.859 |
Token | UK |
Token ID | 3482 |
Feature activation | -2.34e+01 |
Pos logit contributions | |
losses | +7.906 |
deadliest | +7.814 |
victims | +7.784 |
occupants | +7.600 |
quotas | +7.582 |
Neg logit contributions | |
cerning | -9.631 |
ude | -9.516 |
ingu | -9.033 |
asm | -8.876 |
cius | -8.734 |
Token | to |
Token ID | 284 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
losses | +18.323 |
victims | +18.163 |
authorities | +18.015 |
only | +17.756 |
revenues | +17.533 |
Neg logit contributions | |
cerning | -18.322 |
ude | -17.553 |
asm | -16.645 |
dos | -16.491 |
utics | -16.465 |
Token | remain |
Token ID | 3520 |
Feature activation | -2.16e+01 |
Pos logit contributions | |
ude | +0.661 |
cerning | +0.659 |
asm | +0.608 |
ingu | +0.607 |
cius | +0.590 |
Neg logit contributions | |
deadliest | -0.751 |
losses | -0.742 |
victims | -0.730 |
occupants | -0.727 |
quotas | -0.714 |
Token | within |
Token ID | 1626 |
Feature activation | -5.74e+00 |
Pos logit contributions | |
victims | +14.912 |
only | +14.464 |
losses | +14.347 |
occupants | +14.262 |
closest | +14.082 |
Neg logit contributions | |
cerning | -20.670 |
ude | -19.020 |
ingu | -18.630 |
assi | -18.464 |
cius | -18.281 |
Token | the |
Token ID | 262 |
Feature activation | -5.78e+00 |
Pos logit contributions | |
losses | +4.233 |
deadliest | +4.212 |
victims | +4.154 |
quotas | +4.053 |
occupants | +4.026 |
Neg logit contributions | |
cerning | -6.336 |
ude | -6.170 |
ingu | -5.890 |
asm | -5.838 |
cius | -5.776 |
Token | EU |
Token ID | 4576 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
deadliest | +4.529 |
losses | +4.496 |
victims | +4.460 |
quotas | +4.377 |
occupants | +4.297 |
Neg logit contributions | |
cerning | -6.105 |
ude | -5.988 |
ingu | -5.709 |
cius | -5.619 |
asm | -5.618 |
Token | On |
Token ID | 1550 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
losses | +7.133 |
deadliest | +7.098 |
victims | +6.978 |
occupants | +6.831 |
quotas | +6.799 |
Neg logit contributions | |
cerning | -6.579 |
ude | -6.401 |
ingu | -6.063 |
asm | -5.919 |
cius | -5.914 |
Token | average |
Token ID | 2811 |
Feature activation | -3.76e+00 |
Pos logit contributions | |
cerning | +1.599 |
ude | +1.597 |
asm | +1.500 |
ingu | +1.487 |
cius | +1.458 |
Neg logit contributions | |
deadliest | -1.162 |
losses | -1.150 |
victims | -1.127 |
occupants | -1.124 |
quotas | -1.098 |
Token | , |
Token ID | 11 |
Feature activation | -7.54e+00 |
Pos logit contributions | |
deadliest | +3.804 |
losses | +3.804 |
victims | +3.733 |
occupants | +3.682 |
quotas | +3.631 |
Neg logit contributions | |
cerning | -3.187 |
ude | -3.132 |
ingu | -2.901 |
asm | -2.895 |
cius | -2.838 |
Token | the |
Token ID | 262 |
Feature activation | -9.02e+00 |
Pos logit contributions | |
losses | +6.120 |
deadliest | +6.111 |
victims | +5.986 |
occupants | +5.875 |
quotas | +5.777 |
Neg logit contributions | |
cerning | -7.822 |
ude | -7.685 |
ingu | -7.297 |
asm | -7.154 |
cius | -7.131 |
Token | glacier |
Token ID | 44539 |
Feature activation | -1.93e+01 |
Pos logit contributions | |
deadliest | +8.003 |
losses | +7.952 |
victims | +7.760 |
occupants | +7.645 |
quotas | +7.504 |
Neg logit contributions | |
cerning | -8.645 |
ude | -8.378 |
ingu | -8.004 |
cius | -7.852 |
asm | -7.800 |
Token | moved |
Token ID | 3888 |
Feature activation | -2.34e+01 |
Pos logit contributions | |
losses | +16.269 |
deadliest | +15.322 |
victims | +15.129 |
occupants | +14.976 |
only | +14.956 |
Neg logit contributions | |
cerning | -16.727 |
ude | -15.708 |
cius | -15.075 |
ingu | -15.021 |
uggest | -14.836 |
Token | nearly |
Token ID | 3016 |
Feature activation | -7.75e+00 |
Pos logit contributions | |
losses | +18.229 |
only | +17.789 |
deadliest | +17.399 |
victims | +17.355 |
closest | +16.963 |
Neg logit contributions | |
cerning | -20.399 |
cius | -18.919 |
ude | -18.910 |
ingu | -18.469 |
clus | -18.165 |
Token | three |
Token ID | 1115 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
deadliest | +5.380 |
losses | +5.377 |
victims | +5.242 |
occupants | +5.153 |
quotas | +5.063 |
Neg logit contributions | |
cerning | -8.991 |
ude | -8.892 |
ingu | -8.375 |
asm | -8.332 |
cius | -8.278 |
Token | times |
Token ID | 1661 |
Feature activation | -6.09e+00 |
Pos logit contributions | |
deadliest | +1.761 |
losses | +1.746 |
victims | +1.710 |
occupants | +1.688 |
quotas | +1.659 |
Neg logit contributions | |
cerning | -2.450 |
ude | -2.433 |
asm | -2.287 |
ingu | -2.282 |
cius | -2.241 |
Token | faster |
Token ID | 5443 |
Feature activation | -6.42e+00 |
Pos logit contributions | |
deadliest | +3.372 |
losses | +3.327 |
victims | +3.229 |
occupants | +3.157 |
quotas | +3.079 |
Neg logit contributions | |
cerning | -7.971 |
ude | -7.859 |
ingu | -7.461 |
asm | -7.434 |
cius | -7.421 |
Token | in |
Token ID | 287 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
losses | +5.383 |
deadliest | +5.348 |
victims | +5.254 |
occupants | +5.169 |
quotas | +5.065 |
Neg logit contributions | |
cerning | -6.472 |
ude | -6.377 |
ingu | -5.999 |
asm | -5.962 |
cius | -5.905 |
Token | Although |
Token ID | 7003 |
Feature activation | -5.81e+00 |
Pos logit contributions | |
deadliest | +5.433 |
losses | +5.339 |
victims | +5.315 |
quotas | +5.184 |
occupants | +5.178 |
Neg logit contributions | |
cerning | -5.973 |
ude | -5.862 |
ingu | -5.529 |
cius | -5.505 |
asm | -5.414 |
Token | some |
Token ID | 617 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
deadliest | +4.948 |
losses | +4.939 |
victims | +4.848 |
occupants | +4.742 |
quotas | +4.705 |
Neg logit contributions | |
cerning | -5.822 |
ude | -5.766 |
ingu | -5.433 |
asm | -5.373 |
cius | -5.349 |
Token | believe |
Token ID | 1975 |
Feature activation | -7.54e+00 |
Pos logit contributions | |
ude | +1.094 |
cerning | +1.091 |
asm | +1.021 |
ingu | +1.016 |
cius | +0.993 |
Neg logit contributions | |
deadliest | -0.762 |
losses | -0.744 |
victims | -0.726 |
occupants | -0.724 |
quotas | -0.712 |
Token | that |
Token ID | 326 |
Feature activation | -6.64e+00 |
Pos logit contributions | |
losses | +6.330 |
deadliest | +6.254 |
victims | +6.179 |
occupants | +6.011 |
quotas | +6.006 |
Neg logit contributions | |
cerning | -7.568 |
ude | -7.483 |
ingu | -7.100 |
cius | -7.032 |
asm | -6.966 |
Token | the |
Token ID | 262 |
Feature activation | -8.27e+00 |
Pos logit contributions | |
losses | +5.640 |
deadliest | +5.617 |
victims | +5.532 |
occupants | +5.407 |
quotas | +5.378 |
Neg logit contributions | |
cerning | -6.654 |
ude | -6.566 |
ingu | -6.212 |
cius | -6.154 |
asm | -6.094 |
Token | Syrian |
Token ID | 6318 |
Feature activation | -2.32e+01 |
Pos logit contributions | |
deadliest | +7.619 |
losses | +7.604 |
victims | +7.461 |
occupants | +7.323 |
quotas | +7.242 |
Neg logit contributions | |
cerning | -7.596 |
ude | -7.476 |
ingu | -7.130 |
cius | -7.019 |
asm | -6.899 |
Token | government |
Token ID | 1230 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
authorities | +17.903 |
losses | +17.849 |
victims | +17.615 |
deadliest | +17.175 |
occupants | +17.064 |
Neg logit contributions | |
cerning | -19.027 |
cius | -18.284 |
ude | -18.255 |
utics | -17.790 |
ingu | -17.530 |
Token | helped |
Token ID | 4193 |
Feature activation | -8.21e+00 |
Pos logit contributions | |
losses | +9.892 |
deadliest | +9.688 |
victims | +9.629 |
quotas | +9.309 |
authorities | +9.293 |
Neg logit contributions | |
cerning | -13.588 |
ude | -13.248 |
ingu | -12.598 |
asm | -12.423 |
cius | -12.336 |
Token | facilitate |
Token ID | 15570 |
Feature activation | -6.90e+00 |
Pos logit contributions | |
deadliest | +8.091 |
losses | +8.003 |
victims | +7.952 |
occupants | +7.729 |
quotas | +7.672 |
Neg logit contributions | |
cerning | -6.943 |
ude | -6.732 |
cius | -6.373 |
asm | -6.342 |
ingu | -6.239 |
Token | the |
Token ID | 262 |
Feature activation | -6.61e+00 |
Pos logit contributions | |
losses | +5.398 |
deadliest | +5.390 |
victims | +5.247 |
occupants | +5.092 |
quotas | +5.068 |
Neg logit contributions | |
cerning | -7.380 |
ude | -7.251 |
ingu | -6.896 |
cius | -6.893 |
asm | -6.771 |
Token | P |
Token ID | 350 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
deadliest | +5.976 |
losses | +5.942 |
victims | +5.821 |
occupants | +5.739 |
quotas | +5.653 |
Neg logit contributions | |
cerning | -6.308 |
ude | -6.206 |
ingu | -5.883 |
cius | -5.798 |
asm | -5.759 |
Token |
|
Token ID | 198 |
Feature activation | -7.34e+00 |
Pos logit contributions | |
losses | +8.595 |
victims | +8.516 |
only | +8.418 |
deadliest | +8.294 |
quotas | +8.241 |
Neg logit contributions | |
cerning | -11.923 |
ude | -11.517 |
ingu | -11.351 |
cius | -11.244 |
uggest | -10.905 |
Token | is |
Token ID | 271 |
Feature activation | -1.42e+01 |
Pos logit contributions | |
deadliest | +7.694 |
losses | +7.693 |
victims | +7.669 |
quotas | +7.558 |
occupants | +7.536 |
Neg logit contributions | |
ude | -5.441 |
cerning | -5.434 |
cius | -4.907 |
ingu | -4.888 |
asm | -4.749 |
Token | created |
Token ID | 2727 |
Feature activation | -1.49e+01 |
Pos logit contributions | |
victims | +9.264 |
deadliest | +9.241 |
only | +9.132 |
losses | +9.031 |
quotas | +8.971 |
Neg logit contributions | |
cerning | -15.065 |
ude | -14.766 |
clus | -13.906 |
asm | -13.761 |
cius | -13.743 |
Token | and |
Token ID | 290 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
victims | +8.768 |
only | +8.759 |
losses | +8.716 |
quotas | +8.301 |
occupants | +8.116 |
Neg logit contributions | |
cerning | -16.501 |
ude | -15.852 |
ingu | -15.771 |
cius | -15.477 |
clus | -15.311 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
losses | +8.211 |
victims | +8.202 |
only | +7.935 |
quotas | +7.773 |
occupants | +7.649 |
Neg logit contributions | |
cerning | -16.608 |
ude | -15.816 |
ingu | -15.506 |
cius | -15.463 |
clus | -15.341 |
Token | archived |
Token ID | 33962 |
Feature activation | -2.30e+01 |
Pos logit contributions | |
victims | +14.120 |
deadliest | +14.072 |
losses | +13.904 |
only | +13.856 |
quotas | +13.705 |
Neg logit contributions | |
cerning | -12.213 |
ude | -11.989 |
clus | -11.478 |
cius | -11.115 |
ingu | -11.101 |
Token | file |
Token ID | 2393 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
only | +14.056 |
victims | +13.752 |
losses | +13.244 |
closest | +12.567 |
quotas | +12.493 |
Neg logit contributions | |
cerning | -22.678 |
clus | -21.103 |
cius | -20.624 |
utics | -20.409 |
cipled | -20.288 |
Token | is |
Token ID | 318 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
losses | +7.443 |
victims | +7.258 |
only | +7.086 |
quotas | +6.928 |
occupants | +6.667 |
Neg logit contributions | |
cerning | -19.927 |
ude | -19.164 |
ingu | -19.162 |
cius | -19.153 |
clus | -18.432 |
Token | uploaded |
Token ID | 19144 |
Feature activation | -2.23e+01 |
Pos logit contributions | |
victims | +10.000 |
losses | +9.850 |
only | +9.766 |
deadliest | +9.705 |
quotas | +9.596 |
Neg logit contributions | |
cerning | -14.307 |
ude | -13.820 |
ingu | -13.585 |
cius | -13.457 |
asm | -13.234 |
Token | to |
Token ID | 284 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
only | +10.332 |
victims | +9.588 |
losses | +9.301 |
quotas | +9.216 |
authorities | +8.514 |
Neg logit contributions | |
cerning | -25.169 |
cius | -24.823 |
ingu | -24.710 |
ude | -24.169 |
uggest | -24.126 |
Token | it |
Token ID | 340 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
deadliest | +2.547 |
losses | +2.542 |
victims | +2.479 |
occupants | +2.408 |
quotas | +2.367 |
Neg logit contributions | |
cerning | -6.758 |
ude | -6.694 |
ingu | -6.374 |
asm | -6.355 |
cius | -6.321 |
Token | s |
Token ID | 82 |
Feature activation | -7.62e+00 |
Pos logit contributions | |
deadliest | +5.005 |
losses | +4.978 |
victims | +4.883 |
occupants | +4.734 |
quotas | +4.684 |
Neg logit contributions | |
cerning | -6.475 |
ude | -6.312 |
ingu | -6.053 |
asm | -6.023 |
cius | -5.918 |
Token | efforts |
Token ID | 4040 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
deadliest | +6.402 |
losses | +6.363 |
victims | +6.236 |
occupants | +6.083 |
quotas | +5.907 |
Neg logit contributions | |
cerning | -7.750 |
ude | -7.595 |
ingu | -7.204 |
asm | -7.135 |
cius | -7.038 |
Token | to |
Token ID | 284 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
cerning | +4.013 |
ude | +3.988 |
asm | +3.750 |
ingu | +3.736 |
cius | +3.665 |
Neg logit contributions | |
deadliest | -2.526 |
losses | -2.476 |
occupants | -2.426 |
victims | -2.411 |
quotas | -2.362 |
Token | pull |
Token ID | 2834 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
ude | +9.688 |
cerning | +9.410 |
ingu | +8.991 |
asm | +8.786 |
clus | +8.423 |
Neg logit contributions | |
deadliest | -8.469 |
losses | -8.307 |
occupants | -8.174 |
quotas | -8.028 |
revenues | -7.939 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
losses | +9.777 |
victims | +9.741 |
deadliest | +9.669 |
occupants | +9.480 |
quotas | +9.211 |
Neg logit contributions | |
cerning | -13.183 |
ude | -12.684 |
ingu | -12.091 |
cius | -12.079 |
asm | -11.873 |
Token | Fukushima |
Token ID | 34170 |
Feature activation | -2.28e+01 |
Pos logit contributions | |
deadliest | +10.455 |
victims | +10.331 |
losses | +10.329 |
occupants | +10.071 |
authorities | +9.756 |
Neg logit contributions | |
cerning | -12.972 |
ude | -12.686 |
ingu | -11.961 |
cius | -11.939 |
asm | -11.727 |
Token | plant |
Token ID | 4618 |
Feature activation | -1.95e+01 |
Pos logit contributions | |
victims | +20.253 |
losses | +20.013 |
occupants | +19.421 |
authorities | +19.267 |
deadliest | +18.945 |
Neg logit contributions | |
cerning | -16.351 |
ude | -15.181 |
cipled | -14.921 |
cius | -14.887 |
utics | -14.776 |
Token | back |
Token ID | 736 |
Feature activation | -9.89e+00 |
Pos logit contributions | |
losses | +15.807 |
victims | +15.721 |
occupants | +15.637 |
deadliest | +15.363 |
closest | +15.080 |
Neg logit contributions | |
cerning | -16.811 |
ude | -15.920 |
cius | -15.277 |
ingu | -15.243 |
uggest | -15.078 |
Token | from |
Token ID | 422 |
Feature activation | -4.86e+00 |
Pos logit contributions | |
losses | +9.617 |
victims | +9.470 |
deadliest | +9.456 |
occupants | +9.369 |
quotas | +9.099 |
Neg logit contributions | |
cerning | -8.244 |
ude | -7.934 |
asm | -7.465 |
ingu | -7.409 |
cius | -7.356 |
Token | the |
Token ID | 262 |
Feature activation | -7.44e+00 |
Pos logit contributions | |
deadliest | +3.578 |
losses | +3.567 |
victims | +3.482 |
occupants | +3.425 |
quotas | +3.315 |
Neg logit contributions | |
cerning | -5.458 |
ude | -5.393 |
ingu | -5.130 |
asm | -5.095 |
cius | -5.033 |
Token | brink |
Token ID | 29585 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
deadliest | +4.936 |
losses | +4.858 |
victims | +4.725 |
occupants | +4.601 |
quotas | +4.452 |
Neg logit contributions | |
cerning | -8.877 |
ude | -8.781 |
ingu | -8.416 |
asm | -8.278 |
cius | -8.262 |
Token | from |
Token ID | 422 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
losses | +11.757 |
victims | +11.368 |
deadliest | +11.151 |
quotas | +11.122 |
occupants | +11.119 |
Neg logit contributions | |
cerning | -14.333 |
ude | -13.741 |
ingu | -13.261 |
cius | -13.072 |
asm | -12.784 |
Token | its |
Token ID | 663 |
Feature activation | -9.44e+00 |
Pos logit contributions | |
deadliest | +2.475 |
losses | +2.474 |
victims | +2.417 |
occupants | +2.382 |
quotas | +2.354 |
Neg logit contributions | |
cerning | -3.151 |
ude | -3.116 |
ingu | -2.923 |
asm | -2.916 |
cius | -2.873 |
Token | membership |
Token ID | 9931 |
Feature activation | -1.46e+01 |
Pos logit contributions | |
losses | +7.871 |
deadliest | +7.621 |
victims | +7.548 |
quotas | +7.407 |
occupants | +7.406 |
Neg logit contributions | |
cerning | -9.627 |
ude | -9.284 |
ingu | -8.941 |
cius | -8.729 |
asm | -8.694 |
Token | , |
Token ID | 11 |
Feature activation | -9.91e+00 |
Pos logit contributions | |
losses | +12.410 |
quotas | +12.006 |
victims | +11.870 |
only | +11.566 |
authorities | +11.549 |
Neg logit contributions | |
cerning | -13.270 |
ude | -12.383 |
ingu | -12.342 |
asm | -11.890 |
iku | -11.836 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
losses | +7.692 |
victims | +7.464 |
deadliest | +7.429 |
quotas | +7.301 |
occupants | +7.262 |
Neg logit contributions | |
cerning | -10.537 |
ude | -10.300 |
ingu | -9.841 |
cius | -9.671 |
asm | -9.518 |
Token | EU |
Token ID | 4576 |
Feature activation | -2.28e+01 |
Pos logit contributions | |
losses | +10.484 |
victims | +10.145 |
deadliest | +10.083 |
quotas | +9.908 |
occupants | +9.861 |
Neg logit contributions | |
cerning | -11.771 |
ude | -11.509 |
ingu | -11.149 |
cius | -10.823 |
utics | -10.732 |
Token | has |
Token ID | 468 |
Feature activation | -1.21e+01 |
Pos logit contributions | |
losses | +18.128 |
victims | +17.929 |
authorities | +17.844 |
quotas | +17.678 |
only | +17.541 |
Neg logit contributions | |
cerning | -18.286 |
ude | -17.692 |
dos | -16.577 |
cius | -16.463 |
ingu | -16.441 |
Token | benefited |
Token ID | 25050 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
losses | +8.654 |
victims | +8.477 |
deadliest | +8.411 |
quotas | +8.240 |
occupants | +8.134 |
Neg logit contributions | |
cerning | -13.345 |
ude | -13.042 |
ingu | -12.458 |
cius | -12.341 |
asm | -12.186 |
Token | as |
Token ID | 355 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
losses | +8.535 |
deadliest | +8.377 |
victims | +8.350 |
occupants | +8.127 |
quotas | +7.950 |
Neg logit contributions | |
cerning | -10.977 |
ude | -10.640 |
cius | -10.064 |
ingu | -10.031 |
asm | -9.853 |
Token | well |
Token ID | 880 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
losses | +3.770 |
deadliest | +3.742 |
victims | +3.672 |
occupants | +3.584 |
quotas | +3.507 |
Neg logit contributions | |
cerning | -7.453 |
ude | -7.330 |
ingu | -6.969 |
asm | -6.938 |
cius | -6.905 |
Token | . |
Token ID | 13 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
deadliest | +0.398 |
losses | +0.395 |
victims | +0.388 |
occupants | +0.384 |
quotas | +0.379 |
Neg logit contributions | |
cerning | -0.371 |
ude | -0.368 |
asm | -0.342 |
ingu | -0.340 |
cius | -0.333 |
Token | has |
Token ID | 468 |
Feature activation | -7.16e+00 |
Pos logit contributions | |
deadliest | +0.948 |
losses | +0.932 |
victims | +0.911 |
occupants | +0.905 |
quotas | +0.887 |
Neg logit contributions | |
cerning | -1.407 |
ude | -1.403 |
asm | -1.321 |
ingu | -1.314 |
cius | -1.290 |
Token | also |
Token ID | 635 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
deadliest | +5.102 |
losses | +5.101 |
victims | +5.012 |
occupants | +4.891 |
quotas | +4.833 |
Neg logit contributions | |
cerning | -8.134 |
ude | -8.028 |
ingu | -7.600 |
asm | -7.589 |
cius | -7.471 |
Token | warned |
Token ID | 7728 |
Feature activation | -7.42e+00 |
Pos logit contributions | |
deadliest | +2.006 |
losses | +1.975 |
victims | +1.927 |
occupants | +1.912 |
quotas | +1.870 |
Neg logit contributions | |
cerning | -3.119 |
ude | -3.109 |
asm | -2.931 |
ingu | -2.914 |
cius | -2.863 |
Token | that |
Token ID | 326 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
losses | +6.378 |
victims | +6.235 |
deadliest | +6.210 |
occupants | +6.071 |
quotas | +6.046 |
Neg logit contributions | |
cerning | -7.306 |
ude | -7.256 |
ingu | -6.893 |
asm | -6.744 |
cius | -6.713 |
Token | the |
Token ID | 262 |
Feature activation | -9.76e+00 |
Pos logit contributions | |
losses | +4.969 |
deadliest | +4.875 |
victims | +4.822 |
occupants | +4.722 |
quotas | +4.684 |
Neg logit contributions | |
cerning | -7.117 |
ude | -7.046 |
ingu | -6.658 |
asm | -6.584 |
cius | -6.553 |
Token | UK |
Token ID | 3482 |
Feature activation | -2.28e+01 |
Pos logit contributions | |
losses | +7.808 |
deadliest | +7.523 |
victims | +7.480 |
authorities | +7.250 |
occupants | +7.246 |
Neg logit contributions | |
cerning | -10.099 |
ude | -9.964 |
ingu | -9.542 |
asm | -9.295 |
cius | -9.257 |
Token | will |
Token ID | 481 |
Feature activation | -7.55e+00 |
Pos logit contributions | |
losses | +17.313 |
authorities | +16.500 |
victims | +16.333 |
revenues | +16.098 |
only | +15.971 |
Neg logit contributions | |
cerning | -19.894 |
ude | -19.559 |
clus | -18.216 |
ingu | -18.130 |
cipled | -18.118 |
Token | not |
Token ID | 407 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
losses | +6.967 |
deadliest | +6.839 |
victims | +6.838 |
only | +6.598 |
occupants | +6.590 |
Neg logit contributions | |
cerning | -6.918 |
ude | -6.648 |
cius | -6.314 |
asm | -6.305 |
ingu | -6.270 |
Token | escape |
Token ID | 6654 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
losses | +4.102 |
deadliest | +4.056 |
victims | +4.039 |
occupants | +3.943 |
quotas | +3.907 |
Neg logit contributions | |
cerning | -3.597 |
ude | -3.467 |
asm | -3.246 |
cius | -3.238 |
ingu | -3.234 |
Token | the |
Token ID | 262 |
Feature activation | -6.73e+00 |
Pos logit contributions | |
losses | +7.940 |
victims | +7.415 |
deadliest | +7.335 |
quotas | +7.178 |
occupants | +7.118 |
Neg logit contributions | |
cerning | -12.537 |
ude | -12.176 |
cius | -11.763 |
ingu | -11.679 |
clus | -11.484 |
Token | slowdown |
Token ID | 35621 |
Feature activation | -8.61e+00 |
Pos logit contributions | |
losses | +5.032 |
deadliest | +4.974 |
victims | +4.830 |
occupants | +4.724 |
quotas | +4.668 |
Neg logit contributions | |
cerning | -7.471 |
ude | -7.409 |
ingu | -7.003 |
asm | -6.895 |
cius | -6.866 |
Token | 800 |
Token ID | 7410 |
Feature activation | -5.31e+00 |
Pos logit contributions | |
deadliest | +4.131 |
losses | +4.129 |
victims | +4.080 |
occupants | +4.012 |
quotas | +3.935 |
Neg logit contributions | |
cerning | -3.966 |
ude | -3.859 |
asm | -3.668 |
ingu | -3.614 |
cius | -3.565 |
Token | hectares |
Token ID | 43025 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
losses | +4.286 |
deadliest | +4.260 |
victims | +4.259 |
occupants | +4.217 |
quotas | +4.065 |
Neg logit contributions | |
cerning | -5.487 |
ude | -5.399 |
asm | -5.106 |
ingu | -5.069 |
cius | -5.035 |
Token | of |
Token ID | 286 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
deadliest | +3.472 |
losses | +3.470 |
victims | +3.421 |
occupants | +3.370 |
quotas | +3.313 |
Neg logit contributions | |
cerning | -3.227 |
ude | -3.179 |
asm | -2.962 |
ingu | -2.957 |
cius | -2.907 |
Token | forest |
Token ID | 8222 |
Feature activation | -8.97e+00 |
Pos logit contributions | |
deadliest | +2.232 |
losses | +2.215 |
victims | +2.174 |
occupants | +2.142 |
quotas | +2.102 |
Neg logit contributions | |
cerning | -3.038 |
ude | -3.006 |
asm | -2.827 |
ingu | -2.827 |
cius | -2.775 |
Token | , |
Token ID | 11 |
Feature activation | -7.61e+00 |
Pos logit contributions | |
losses | +8.117 |
deadliest | +8.002 |
victims | +7.986 |
occupants | +7.778 |
quotas | +7.674 |
Neg logit contributions | |
cerning | -8.202 |
ude | -7.987 |
ingu | -7.581 |
cius | -7.549 |
asm | -7.513 |
Token | burning |
Token ID | 9482 |
Feature activation | -2.25e+01 |
Pos logit contributions | |
deadliest | +6.586 |
losses | +6.540 |
victims | +6.446 |
occupants | +6.339 |
quotas | +6.210 |
Neg logit contributions | |
cerning | -7.566 |
ude | -7.452 |
ingu | -6.973 |
asm | -6.958 |
cius | -6.881 |
Token | several |
Token ID | 1811 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
victims | +18.372 |
losses | +18.176 |
deadliest | +17.874 |
occupants | +17.828 |
only | +17.073 |
Neg logit contributions | |
cerning | -18.888 |
cius | -17.896 |
ude | -17.887 |
ingu | -17.490 |
cipled | -17.070 |
Token | vehicles |
Token ID | 5672 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
deadliest | +3.473 |
losses | +3.447 |
victims | +3.407 |
occupants | +3.366 |
quotas | +3.250 |
Neg logit contributions | |
cerning | -5.228 |
ude | -5.161 |
ingu | -4.884 |
asm | -4.880 |
cius | -4.825 |
Token | and |
Token ID | 290 |
Feature activation | -5.04e+00 |
Pos logit contributions | |
deadliest | +3.123 |
losses | +3.113 |
victims | +3.071 |
occupants | +3.039 |
quotas | +2.974 |
Neg logit contributions | |
cerning | -2.955 |
ude | -2.913 |
asm | -2.713 |
ingu | -2.704 |
cius | -2.663 |
Token | threatening |
Token ID | 11123 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
deadliest | +3.493 |
losses | +3.457 |
victims | +3.378 |
occupants | +3.331 |
quotas | +3.261 |
Neg logit contributions | |
cerning | -5.895 |
ude | -5.855 |
asm | -5.535 |
ingu | -5.517 |
cius | -5.428 |
Token | homes |
Token ID | 5682 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
losses | +7.665 |
victims | +7.636 |
deadliest | +7.528 |
occupants | +7.502 |
authorities | +7.112 |
Neg logit contributions | |
cerning | -11.628 |
ude | -11.434 |
ingu | -10.920 |
cius | -10.862 |
asm | -10.723 |
Token | payment |
Token ID | 6074 |
Feature activation | -7.50e+00 |
Pos logit contributions | |
deadliest | +4.012 |
losses | +4.003 |
victims | +3.919 |
occupants | +3.849 |
quotas | +3.811 |
Neg logit contributions | |
cerning | -4.746 |
ude | -4.687 |
ingu | -4.396 |
asm | -4.384 |
cius | -4.323 |
Token | schedule |
Token ID | 7269 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
losses | +6.220 |
deadliest | +6.109 |
victims | +6.066 |
quotas | +5.939 |
occupants | +5.855 |
Neg logit contributions | |
cerning | -7.646 |
ude | -7.490 |
ingu | -7.105 |
asm | -7.060 |
cius | -7.012 |
Token | would |
Token ID | 561 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
ude | +1.671 |
cerning | +1.657 |
asm | +1.572 |
ingu | +1.566 |
clus | +1.522 |
Neg logit contributions | |
deadliest | -1.071 |
losses | -1.038 |
occupants | -1.022 |
victims | -1.015 |
quotas | -0.983 |
Token | save |
Token ID | 3613 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
deadliest | +0.052 |
losses | +0.052 |
victims | +0.051 |
occupants | +0.051 |
quotas | +0.050 |
Neg logit contributions | |
cerning | -0.050 |
ude | -0.050 |
asm | -0.046 |
ingu | -0.046 |
cius | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
losses | +9.398 |
victims | +9.386 |
deadliest | +9.131 |
occupants | +8.956 |
revenues | +8.855 |
Neg logit contributions | |
cerning | -11.295 |
ude | -10.949 |
ingu | -10.583 |
cius | -10.480 |
asm | -10.405 |
Token | Haitian |
Token ID | 48723 |
Feature activation | -2.25e+01 |
Pos logit contributions | |
victims | +7.806 |
losses | +7.786 |
deadliest | +7.673 |
occupants | +7.462 |
authorities | +7.258 |
Neg logit contributions | |
cerning | -11.696 |
ude | -11.450 |
ingu | -10.981 |
asm | -10.885 |
cius | -10.859 |
Token | government |
Token ID | 1230 |
Feature activation | -1.56e+01 |
Pos logit contributions | |
victims | +18.363 |
losses | +18.106 |
authorities | +17.824 |
occupants | +17.320 |
revenues | +17.136 |
Neg logit contributions | |
cerning | -17.732 |
ude | -16.999 |
clus | -16.371 |
ingu | -16.289 |
cius | -16.244 |
Token | USD |
Token ID | 11403 |
Feature activation | -4.96e+00 |
Pos logit contributions | |
losses | +12.328 |
victims | +11.895 |
revenues | +11.720 |
deadliest | +11.676 |
only | +11.466 |
Neg logit contributions | |
cerning | -15.219 |
ude | -14.862 |
ingu | -14.200 |
cius | -14.077 |
clus | -14.068 |
Token | 100 |
Token ID | 1802 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
deadliest | +4.234 |
losses | +4.234 |
victims | +4.150 |
occupants | +4.078 |
quotas | +4.032 |
Neg logit contributions | |
cerning | -4.955 |
ude | -4.906 |
asm | -4.606 |
ingu | -4.593 |
cius | -4.510 |
Token | million |
Token ID | 1510 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
deadliest | +2.698 |
losses | +2.694 |
victims | +2.643 |
occupants | +2.587 |
quotas | +2.557 |
Neg logit contributions | |
cerning | -3.388 |
ude | -3.357 |
asm | -3.157 |
ingu | -3.148 |
cius | -3.091 |
Token | per |
Token ID | 583 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
deadliest | +2.996 |
losses | +2.983 |
victims | +2.923 |
occupants | +2.877 |
quotas | +2.838 |
Neg logit contributions | |
cerning | -3.754 |
ude | -3.726 |
asm | -3.495 |
ingu | -3.486 |
cius | -3.425 |
Token | on |
Token ID | 319 |
Feature activation | +7.08e+00 |
Pos logit contributions | |
deadliest | +0.528 |
losses | +0.521 |
victims | +0.507 |
occupants | +0.498 |
quotas | +0.487 |
Neg logit contributions | |
cerning | -1.118 |
ude | -1.111 |
asm | -1.054 |
ingu | -1.052 |
cius | -1.036 |
Token | the |
Token ID | 262 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
ude | +6.428 |
cerning | +6.419 |
asm | +6.205 |
clus | +5.899 |
ingu | +5.891 |
Neg logit contributions | |
deadliest | -6.177 |
occupants | -5.994 |
losses | -5.918 |
victims | -5.891 |
quotas | -5.776 |
Token | fact |
Token ID | 1109 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
cerning | +0.358 |
ude | +0.356 |
asm | +0.339 |
ingu | +0.334 |
cius | +0.329 |
Neg logit contributions | |
deadliest | -0.205 |
losses | -0.201 |
victims | -0.197 |
occupants | -0.196 |
quotas | -0.191 |
Token | that |
Token ID | 326 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
deadliest | +1.305 |
losses | +1.302 |
victims | +1.276 |
occupants | +1.256 |
quotas | +1.241 |
Neg logit contributions | |
cerning | -1.313 |
ude | -1.300 |
ingu | -1.212 |
asm | -1.208 |
cius | -1.184 |
Token | the |
Token ID | 262 |
Feature activation | -7.27e+00 |
Pos logit contributions | |
deadliest | +2.723 |
losses | +2.703 |
victims | +2.658 |
occupants | +2.631 |
quotas | +2.593 |
Neg logit contributions | |
cerning | -2.563 |
ude | -2.542 |
asm | -2.360 |
ingu | -2.352 |
cius | -2.300 |
Token | Democratic |
Token ID | 4390 |
Feature activation | -2.24e+01 |
Pos logit contributions | |
losses | +5.711 |
deadliest | +5.696 |
victims | +5.570 |
occupants | +5.450 |
quotas | +5.343 |
Neg logit contributions | |
cerning | -7.770 |
ude | -7.707 |
ingu | -7.300 |
asm | -7.163 |
cius | -7.114 |
Token | nominating |
Token ID | 48833 |
Feature activation | -1.22e+01 |
Pos logit contributions | |
losses | +16.632 |
victims | +16.045 |
occupants | +15.855 |
authorities | +15.575 |
deadliest | +15.544 |
Neg logit contributions | |
cerning | -19.383 |
ude | -18.708 |
ingu | -18.209 |
utics | -17.892 |
cius | -17.664 |
Token | process |
Token ID | 1429 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
losses | +9.243 |
victims | +9.144 |
occupants | +8.920 |
deadliest | +8.908 |
quotas | +8.842 |
Neg logit contributions | |
cerning | -12.686 |
ude | -12.384 |
ingu | -11.798 |
cius | -11.554 |
asm | -11.400 |
Token | is |
Token ID | 318 |
Feature activation | -5.99e+00 |
Pos logit contributions | |
deadliest | +1.524 |
losses | +1.493 |
victims | +1.466 |
occupants | +1.455 |
quotas | +1.419 |
Neg logit contributions | |
cerning | -2.285 |
ude | -2.276 |
asm | -2.146 |
ingu | -2.133 |
cius | -2.092 |
Token | totally |
Token ID | 6635 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
deadliest | +4.152 |
losses | +4.105 |
victims | +4.028 |
occupants | +3.925 |
quotas | +3.867 |
Neg logit contributions | |
cerning | -6.986 |
ude | -6.945 |
asm | -6.558 |
ingu | -6.557 |
cius | -6.444 |
Token | rigged |
Token ID | 31043 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
cerning | +1.962 |
ude | +1.952 |
asm | +1.849 |
ingu | +1.847 |
cius | +1.809 |
Neg logit contributions | |
deadliest | -0.993 |
losses | -0.976 |
occupants | -0.955 |
victims | -0.939 |
quotas | -0.923 |
Token | and |
Token ID | 290 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
victims | +8.768 |
only | +8.759 |
losses | +8.716 |
quotas | +8.301 |
occupants | +8.116 |
Neg logit contributions | |
cerning | -16.501 |
ude | -15.852 |
ingu | -15.771 |
cius | -15.477 |
clus | -15.311 |
Token | the |
Token ID | 262 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
losses | +8.211 |
victims | +8.202 |
only | +7.935 |
quotas | +7.773 |
occupants | +7.649 |
Neg logit contributions | |
cerning | -16.608 |
ude | -15.816 |
ingu | -15.506 |
cius | -15.463 |
clus | -15.341 |
Token | archived |
Token ID | 33962 |
Feature activation | -2.30e+01 |
Pos logit contributions | |
victims | +14.120 |
deadliest | +14.072 |
losses | +13.904 |
only | +13.856 |
quotas | +13.705 |
Neg logit contributions | |
cerning | -12.213 |
ude | -11.989 |
clus | -11.478 |
cius | -11.115 |
ingu | -11.101 |
Token | file |
Token ID | 2393 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
only | +14.056 |
victims | +13.752 |
losses | +13.244 |
closest | +12.567 |
quotas | +12.493 |
Neg logit contributions | |
cerning | -22.678 |
clus | -21.103 |
cius | -20.624 |
utics | -20.409 |
cipled | -20.288 |
Token | is |
Token ID | 318 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
losses | +7.443 |
victims | +7.258 |
only | +7.086 |
quotas | +6.928 |
occupants | +6.667 |
Neg logit contributions | |
cerning | -19.927 |
ude | -19.164 |
ingu | -19.162 |
cius | -19.153 |
clus | -18.432 |
Token | uploaded |
Token ID | 19144 |
Feature activation | -2.23e+01 |
Pos logit contributions | |
victims | +10.000 |
losses | +9.850 |
only | +9.766 |
deadliest | +9.705 |
quotas | +9.596 |
Neg logit contributions | |
cerning | -14.307 |
ude | -13.820 |
ingu | -13.585 |
cius | -13.457 |
asm | -13.234 |
Token | to |
Token ID | 284 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
only | +10.332 |
victims | +9.588 |
losses | +9.301 |
quotas | +9.216 |
authorities | +8.514 |
Neg logit contributions | |
cerning | -25.169 |
cius | -24.823 |
ingu | -24.710 |
ude | -24.169 |
uggest | -24.126 |
Token | it |
Token ID | 340 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
deadliest | +2.547 |
losses | +2.542 |
victims | +2.479 |
occupants | +2.408 |
quotas | +2.367 |
Neg logit contributions | |
cerning | -6.758 |
ude | -6.694 |
ingu | -6.374 |
asm | -6.355 |
cius | -6.321 |
Token | . |
Token ID | 13 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
losses | +5.940 |
victims | +5.869 |
deadliest | +5.852 |
quotas | +5.748 |
occupants | +5.724 |
Neg logit contributions | |
cerning | -4.164 |
ude | -4.041 |
ingu | -3.807 |
cius | -3.721 |
asm | -3.647 |
Token | In |
Token ID | 554 |
Feature activation | +5.78e+00 |
Pos logit contributions | |
deadliest | +3.694 |
losses | +3.685 |
victims | +3.633 |
occupants | +3.591 |
quotas | +3.554 |
Neg logit contributions | |
cerning | -2.994 |
ude | -2.975 |
ingu | -2.733 |
asm | -2.721 |
cius | -2.674 |
Token | the |
Token ID | 262 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
cerning | +5.967 |
ude | +5.848 |
asm | +5.621 |
ingu | +5.505 |
clus | +5.415 |
Neg logit contributions | |
deadliest | -4.657 |
losses | -4.547 |
occupants | -4.512 |
victims | -4.461 |
quotas | -4.361 |
Token | UN |
Token ID | 4725 |
Feature activation | -1.72e+01 |
Pos logit contributions | |
deadliest | +7.773 |
losses | +7.738 |
victims | +7.620 |
occupants | +7.426 |
quotas | +7.314 |
Neg logit contributions | |
cerning | -9.104 |
ude | -8.938 |
ingu | -8.560 |
cius | -8.486 |
asm | -8.298 |
Token | commission |
Token ID | 5810 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
losses | +16.180 |
victims | +16.150 |
deadliest | +16.021 |
authorities | +15.587 |
quotas | +15.537 |
Neg logit contributions | |
cerning | -12.822 |
ude | -12.497 |
cius | -11.776 |
asm | -11.473 |
ingu | -11.388 |
Token | of |
Token ID | 286 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
deadliest | +3.344 |
losses | +3.327 |
victims | +3.269 |
occupants | +3.215 |
quotas | +3.173 |
Neg logit contributions | |
cerning | -4.177 |
ude | -4.135 |
asm | -3.880 |
ingu | -3.873 |
cius | -3.819 |
Token | inquiry |
Token ID | 12069 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
deadliest | +0.128 |
losses | +0.127 |
victims | +0.125 |
occupants | +0.124 |
quotas | +0.122 |
Neg logit contributions | |
cerning | -0.098 |
ude | -0.098 |
asm | -0.090 |
ingu | -0.089 |
cius | -0.087 |
Token | held |
Token ID | 2714 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
deadliest | +2.250 |
losses | +2.225 |
victims | +2.176 |
occupants | +2.154 |
quotas | +2.111 |
Neg logit contributions | |
cerning | -3.287 |
ude | -3.267 |
asm | -3.079 |
ingu | -3.064 |
cius | -3.007 |
Token | Hamas |
Token ID | 15767 |
Feature activation | -2.22e+01 |
Pos logit contributions | |
losses | +9.666 |
deadliest | +9.469 |
victims | +9.447 |
occupants | +9.224 |
quotas | +9.197 |
Neg logit contributions | |
cerning | -11.872 |
ude | -11.588 |
cius | -11.177 |
ingu | -11.089 |
asm | -10.823 |
Token | accountable |
Token ID | 16689 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
losses | +19.704 |
victims | +19.702 |
occupants | +18.905 |
authorities | +18.891 |
deadliest | +18.806 |
Neg logit contributions | |
cerning | -16.048 |
ude | -15.333 |
ingu | -14.621 |
cius | -14.500 |
utics | -14.249 |
Token | to |
Token ID | 284 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
deadliest | +2.467 |
losses | +2.457 |
victims | +2.419 |
occupants | +2.386 |
quotas | +2.353 |
Neg logit contributions | |
cerning | -2.342 |
ude | -2.317 |
asm | -2.154 |
ingu | -2.150 |
cius | -2.106 |
Token | international |
Token ID | 3230 |
Feature activation | -9.06e+00 |
Pos logit contributions | |
ude | +2.915 |
cerning | +2.896 |
asm | +2.741 |
ingu | +2.693 |
clus | +2.621 |
Neg logit contributions | |
deadliest | -2.138 |
losses | -2.095 |
occupants | -2.041 |
victims | -2.024 |
quotas | -1.979 |
Token | human |
Token ID | 1692 |
Feature activation | -7.87e+00 |
Pos logit contributions | |
victims | +5.549 |
losses | +5.488 |
deadliest | +5.356 |
quotas | +5.242 |
occupants | +5.183 |
Neg logit contributions | |
cerning | -11.041 |
ude | -10.790 |
cius | -10.422 |
ingu | -10.294 |
asm | -10.180 |
Token | rights |
Token ID | 2489 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
losses | +2.738 |
victims | +2.600 |
deadliest | +2.462 |
quotas | +2.300 |
occupants | +2.273 |
Neg logit contributions | |
cerning | -11.803 |
ude | -11.526 |
cius | -11.151 |
asm | -11.049 |
ingu | -11.039 |
Token | law |
Token ID | 1099 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
losses | +10.121 |
victims | +9.828 |
revenues | +9.489 |
authorities | +9.473 |
deadliest | +9.425 |
Neg logit contributions | |
cerning | -14.599 |
ude | -14.364 |
ingu | -13.849 |
cius | -13.531 |
asm | -13.449 |
Token | . |
Token ID | 13 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
losses | +6.958 |
deadliest | +6.847 |
victims | +6.831 |
occupants | +6.711 |
quotas | +6.644 |
Neg logit contributions | |
cerning | -4.981 |
ude | -4.920 |
ingu | -4.574 |
asm | -4.482 |
cius | -4.407 |
Token | Republicans |
Token ID | 4734 |
Feature activation | -7.34e+00 |
Pos logit contributions | |
deadliest | +2.954 |
losses | +2.942 |
victims | +2.892 |
occupants | +2.857 |
quotas | +2.825 |
Neg logit contributions | |
cerning | -2.731 |
ude | -2.703 |
ingu | -2.508 |
asm | -2.502 |
cius | -2.447 |
Token | argue |
Token ID | 7267 |
Feature activation | -8.85e+00 |
Pos logit contributions | |
losses | +7.474 |
deadliest | +7.411 |
victims | +7.335 |
occupants | +7.210 |
quotas | +7.107 |
Neg logit contributions | |
cerning | -6.095 |
ude | -5.858 |
ingu | -5.472 |
asm | -5.440 |
clus | -5.306 |
Token | the |
Token ID | 262 |
Feature activation | -9.74e+00 |
Pos logit contributions | |
losses | +7.528 |
victims | +7.325 |
deadliest | +7.323 |
occupants | +7.114 |
quotas | +7.098 |
Neg logit contributions | |
cerning | -8.815 |
ude | -8.686 |
ingu | -8.230 |
asm | -8.079 |
cius | -8.033 |
Token | healthcare |
Token ID | 11409 |
Feature activation | -2.20e+01 |
Pos logit contributions | |
losses | +7.681 |
deadliest | +7.497 |
victims | +7.448 |
occupants | +7.190 |
quotas | +7.156 |
Neg logit contributions | |
cerning | -10.342 |
ude | -10.165 |
ingu | -9.730 |
asm | -9.553 |
cius | -9.472 |
Token | law |
Token ID | 1099 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
losses | +16.382 |
victims | +15.595 |
authorities | +15.199 |
revenues | +15.147 |
only | +14.510 |
Neg logit contributions | |
cerning | -20.179 |
ude | -19.616 |
ingu | -19.377 |
cius | -18.739 |
asm | -18.160 |
Token | would |
Token ID | 561 |
Feature activation | -5.14e+00 |
Pos logit contributions | |
losses | +12.329 |
victims | +12.053 |
deadliest | +11.712 |
revenues | +11.556 |
only | +11.520 |
Neg logit contributions | |
cerning | -15.802 |
ude | -15.293 |
cius | -14.499 |
ingu | -14.495 |
asm | -14.301 |
Token | " |
Token ID | 366 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
deadliest | +5.152 |
losses | +5.151 |
victims | +5.086 |
occupants | +4.971 |
quotas | +4.911 |
Neg logit contributions | |
cerning | -4.395 |
ude | -4.291 |
asm | -4.010 |
ingu | -3.990 |
cius | -3.961 |
Token | raid |
Token ID | 7086 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
ude | +1.809 |
cerning | +1.802 |
asm | +1.629 |
ingu | +1.608 |
cius | +1.535 |
Neg logit contributions | |
deadliest | -3.212 |
losses | -3.178 |
occupants | -3.135 |
victims | -3.124 |
quotas | -3.096 |
Token | " |
Token ID | 1 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
losses | +4.128 |
deadliest | +4.125 |
victims | +4.063 |
occupants | +3.992 |
quotas | +3.916 |
Neg logit contributions | |
cerning | -5.413 |
ude | -5.356 |
asm | -5.061 |
ingu | -5.047 |
cius | -4.951 |
Token | with |
Token ID | 351 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
deadliest | +4.456 |
losses | +4.445 |
victims | +4.379 |
occupants | +4.283 |
quotas | +4.228 |
Neg logit contributions | |
cerning | -5.675 |
ude | -5.605 |
ingu | -5.283 |
asm | -5.269 |
cius | -5.186 |
Token | Iran |
Token ID | 4068 |
Feature activation | -9.25e+00 |
Pos logit contributions | |
cerning | +2.824 |
ude | +2.824 |
asm | +2.679 |
ingu | +2.632 |
clus | +2.581 |
Neg logit contributions | |
deadliest | -1.954 |
losses | -1.909 |
occupants | -1.873 |
victims | -1.872 |
quotas | -1.813 |
Token | , |
Token ID | 11 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
losses | +8.699 |
deadliest | +8.580 |
victims | +8.573 |
quotas | +8.369 |
occupants | +8.304 |
Neg logit contributions | |
cerning | -8.109 |
ude | -7.860 |
ingu | -7.375 |
asm | -7.291 |
cius | -7.278 |
Token | encouraging |
Token ID | 12577 |
Feature activation | -6.24e+00 |
Pos logit contributions | |
deadliest | +3.133 |
losses | +3.106 |
victims | +3.048 |
occupants | +3.015 |
quotas | +2.966 |
Neg logit contributions | |
cerning | -3.636 |
ude | -3.609 |
asm | -3.378 |
ingu | -3.364 |
cius | -3.298 |
Token | the |
Token ID | 262 |
Feature activation | -7.55e+00 |
Pos logit contributions | |
deadliest | +4.964 |
losses | +4.953 |
victims | +4.876 |
occupants | +4.765 |
quotas | +4.696 |
Neg logit contributions | |
cerning | -6.610 |
ude | -6.489 |
ingu | -6.176 |
cius | -6.100 |
asm | -6.079 |
Token | Iranian |
Token ID | 10704 |
Feature activation | -2.20e+01 |
Pos logit contributions | |
deadliest | +6.141 |
losses | +6.013 |
victims | +5.945 |
occupants | +5.829 |
quotas | +5.659 |
Neg logit contributions | |
cerning | -7.851 |
ude | -7.741 |
ingu | -7.389 |
cius | -7.262 |
asm | -7.151 |
Token | armed |
Token ID | 6936 |
Feature activation | -1.22e+01 |
Pos logit contributions | |
authorities | +17.391 |
victims | +17.080 |
deadliest | +16.554 |
losses | +16.487 |
occupants | +16.401 |
Neg logit contributions | |
cerning | -17.662 |
ude | -16.993 |
utics | -16.567 |
dos | -16.472 |
ingu | -16.455 |
Token | forces |
Token ID | 3386 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
losses | +7.530 |
victims | +7.387 |
deadliest | +7.350 |
occupants | +7.233 |
authorities | +7.115 |
Neg logit contributions | |
cerning | -14.373 |
ude | -14.026 |
ingu | -13.815 |
cius | -13.418 |
apply | -13.378 |
Token | to |
Token ID | 284 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
losses | +5.281 |
deadliest | +5.266 |
victims | +5.149 |
occupants | +4.985 |
quotas | +4.887 |
Neg logit contributions | |
cerning | -9.225 |
ude | -9.089 |
ingu | -8.612 |
asm | -8.596 |
cius | -8.545 |
Token | stay |
Token ID | 2652 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
ude | +3.068 |
cerning | +3.037 |
asm | +2.843 |
ingu | +2.819 |
clus | +2.721 |
Neg logit contributions | |
deadliest | -3.032 |
losses | -3.009 |
occupants | -2.971 |
victims | -2.955 |
quotas | -2.907 |
Token | in |
Token ID | 287 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
victims | +8.668 |
deadliest | +8.664 |
losses | +8.654 |
occupants | +8.277 |
closest | +8.104 |
Neg logit contributions | |
cerning | -13.826 |
ude | -13.380 |
ingu | -13.178 |
cius | -12.710 |
asm | -12.635 |
Token | arity |
Token ID | 6806 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
losses | +8.427 |
deadliest | +8.370 |
victims | +8.261 |
occupants | +8.160 |
quotas | +8.079 |
Neg logit contributions | |
cerning | -3.578 |
ude | -3.384 |
ingu | -3.076 |
cius | -3.028 |
asm | -2.893 |
Token | Gaming |
Token ID | 14426 |
Feature activation | -1.99e+01 |
Pos logit contributions | |
losses | +9.932 |
victims | +9.627 |
deadliest | +9.457 |
only | +9.332 |
occupants | +9.130 |
Neg logit contributions | |
cerning | -11.914 |
ude | -11.730 |
ingu | -11.130 |
cius | -10.885 |
asm | -10.854 |
Token | will |
Token ID | 481 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
victims | +9.416 |
losses | +9.397 |
only | +9.330 |
revenues | +8.649 |
owners | +8.549 |
Neg logit contributions | |
cerning | -22.722 |
ude | -22.426 |
ingu | -21.676 |
cipled | -21.464 |
cius | -21.412 |
Token | also |
Token ID | 635 |
Feature activation | -7.94e+00 |
Pos logit contributions | |
only | +8.361 |
victims | +8.237 |
losses | +8.185 |
deadliest | +7.776 |
occupants | +7.617 |
Neg logit contributions | |
cerning | -15.045 |
ude | -14.362 |
cius | -14.327 |
ingu | -14.222 |
cipled | -13.890 |
Token | be |
Token ID | 307 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
losses | +7.021 |
victims | +6.967 |
deadliest | +6.872 |
quotas | +6.714 |
only | +6.709 |
Neg logit contributions | |
cerning | -7.462 |
ude | -7.126 |
ingu | -6.884 |
cius | -6.861 |
asm | -6.681 |
Token | releasing |
Token ID | 13011 |
Feature activation | -2.18e+01 |
Pos logit contributions | |
victims | +8.455 |
losses | +8.452 |
only | +8.190 |
deadliest | +8.154 |
authorities | +7.839 |
Neg logit contributions | |
ude | -15.840 |
cerning | -15.806 |
cius | -15.450 |
ingu | -15.345 |
cipled | -15.073 |
Token | an |
Token ID | 281 |
Feature activation | -8.95e+00 |
Pos logit contributions | |
only | +8.444 |
victims | +8.236 |
authorities | +8.060 |
revenues | +7.936 |
losses | +7.842 |
Neg logit contributions | |
cerning | -27.112 |
cius | -26.283 |
ingu | -25.664 |
ude | -25.491 |
cipled | -24.969 |
Token | official |
Token ID | 1743 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
losses | +4.001 |
deadliest | +3.948 |
victims | +3.947 |
occupants | +3.681 |
quotas | +3.619 |
Neg logit contributions | |
cerning | -12.231 |
ude | -12.107 |
ingu | -11.761 |
cius | -11.651 |
asm | -11.511 |
Token | statement |
Token ID | 2643 |
Feature activation | -8.35e+00 |
Pos logit contributions | |
losses | +9.585 |
victims | +9.422 |
deadliest | +9.290 |
quotas | +9.097 |
only | +9.026 |
Neg logit contributions | |
cerning | -11.290 |
ude | -11.236 |
cius | -10.678 |
ingu | -10.544 |
dos | -10.401 |
Token | soon |
Token ID | 2582 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
deadliest | +4.521 |
losses | +4.448 |
victims | +4.437 |
occupants | +4.238 |
quotas | +4.161 |
Neg logit contributions | |
cerning | -10.783 |
ude | -10.683 |
ingu | -10.221 |
cius | -10.159 |
asm | -10.101 |
Token | . |
Token ID | 13 |
Feature activation | -7.70e+00 |
Pos logit contributions | |
losses | +3.944 |
deadliest | +3.901 |
victims | +3.897 |
occupants | +3.818 |
quotas | +3.815 |
Neg logit contributions | |
cerning | -2.619 |
ude | -2.563 |
ingu | -2.381 |
cius | -2.330 |
asm | -2.310 |
Token | some |
Token ID | 617 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
losses | +6.255 |
deadliest | +6.171 |
victims | +6.056 |
occupants | +5.964 |
quotas | +5.879 |
Neg logit contributions | |
cerning | -8.414 |
ude | -8.272 |
ingu | -7.809 |
asm | -7.803 |
cius | -7.744 |
Token | real |
Token ID | 1103 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
ude | +2.041 |
cerning | +2.038 |
asm | +1.907 |
ingu | +1.899 |
cius | +1.855 |
Neg logit contributions | |
deadliest | -1.355 |
losses | -1.314 |
victims | -1.292 |
occupants | -1.281 |
quotas | -1.259 |
Token | field |
Token ID | 2214 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
ude | +3.777 |
cerning | +3.754 |
asm | +3.528 |
ingu | +3.508 |
clus | +3.416 |
Neg logit contributions | |
deadliest | -2.474 |
losses | -2.393 |
victims | -2.353 |
occupants | -2.332 |
quotas | -2.302 |
Token | of |
Token ID | 286 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
deadliest | +0.220 |
losses | +0.220 |
victims | +0.216 |
occupants | +0.214 |
quotas | +0.211 |
Neg logit contributions | |
cerning | -0.198 |
ude | -0.196 |
asm | -0.182 |
ingu | -0.181 |
cius | -0.178 |
Token | view |
Token ID | 1570 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
cerning | +1.756 |
ude | +1.734 |
asm | +1.564 |
ingu | +1.533 |
cius | +1.472 |
Neg logit contributions | |
deadliest | -3.507 |
losses | -3.469 |
victims | -3.414 |
occupants | -3.391 |
quotas | -3.384 |
Token | advantages |
Token ID | 13391 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
ude | +4.254 |
cerning | +4.176 |
asm | +3.931 |
ingu | +3.876 |
cius | +3.751 |
Neg logit contributions | |
deadliest | -3.601 |
losses | -3.414 |
victims | -3.402 |
occupants | -3.384 |
quotas | -3.343 |
Token | , |
Token ID | 11 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
ude | +13.234 |
cerning | +12.925 |
asm | +12.188 |
ingu | +12.037 |
yt | +11.773 |
Neg logit contributions | |
deadliest | -8.661 |
occupants | -8.004 |
losses | -7.959 |
quotas | -7.957 |
victims | -7.949 |
Token | but |
Token ID | 475 |
Feature activation | -4.79e+00 |
Pos logit contributions | |
ude | +0.622 |
cerning | +0.620 |
asm | +0.584 |
ingu | +0.577 |
cius | +0.567 |
Neg logit contributions | |
deadliest | -0.474 |
losses | -0.464 |
victims | -0.457 |
occupants | -0.453 |
quotas | -0.445 |
Token | you |
Token ID | 345 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
losses | +4.145 |
deadliest | +4.138 |
victims | +4.043 |
occupants | +3.993 |
quotas | +3.926 |
Neg logit contributions | |
cerning | -4.766 |
ude | -4.707 |
asm | -4.411 |
ingu | -4.406 |
cius | -4.328 |
Token | also |
Token ID | 635 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
ude | +0.026 |
cerning | +0.026 |
asm | +0.024 |
ingu | +0.024 |
cius | +0.023 |
Neg logit contributions | |
deadliest | -0.022 |
losses | -0.022 |
victims | -0.021 |
occupants | -0.021 |
quotas | -0.021 |
Token | give |
Token ID | 1577 |
Feature activation | -6.67e+00 |
Pos logit contributions | |
ude | +2.909 |
cerning | +2.834 |
asm | +2.669 |
ingu | +2.650 |
cius | +2.559 |
Neg logit contributions | |
deadliest | -2.803 |
losses | -2.713 |
occupants | -2.707 |
victims | -2.703 |
quotas | -2.667 |
Token | a |
Token ID | 257 |
Feature activation | +6.33e+00 |
Pos logit contributions | |
deadliest | +1.186 |
losses | +1.164 |
victims | +1.133 |
occupants | +1.127 |
quotas | +1.105 |
Neg logit contributions | |
cerning | -2.244 |
ude | -2.241 |
asm | -2.115 |
ingu | -2.109 |
cius | -2.080 |
Token | lifetime |
Token ID | 10869 |
Feature activation | +6.37e+00 |
Pos logit contributions | |
ude | +7.210 |
cerning | +7.179 |
ingu | +6.736 |
asm | +6.709 |
clus | +6.577 |
Neg logit contributions | |
losses | -4.118 |
deadliest | -4.053 |
occupants | -3.990 |
quotas | -3.971 |
victims | -3.934 |
Token | . |
Token ID | 13 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
ude | +6.647 |
cerning | +6.563 |
ingu | +6.199 |
asm | +6.199 |
cius | +5.921 |
Neg logit contributions | |
deadliest | -4.956 |
losses | -4.740 |
occupants | -4.693 |
quotas | -4.618 |
victims | -4.573 |
Token |
|
Token ID | 198 |
Feature activation | -8.98e+00 |
Pos logit contributions | |
ude | +0.599 |
cerning | +0.599 |
asm | +0.555 |
ingu | +0.547 |
cius | +0.535 |
Neg logit contributions | |
deadliest | -0.695 |
losses | -0.686 |
occupants | -0.672 |
victims | -0.669 |
quotas | -0.659 |
Token |
|
Token ID | 198 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
losses | +7.255 |
victims | +7.205 |
deadliest | +7.153 |
quotas | +6.995 |
only | +6.901 |
Neg logit contributions | |
cerning | -8.888 |
ude | -8.596 |
ingu | -8.375 |
cius | -8.273 |
uggest | -7.936 |
Token | In |
Token ID | 818 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
deadliest | +0.259 |
losses | +0.257 |
victims | +0.251 |
occupants | +0.250 |
quotas | +0.246 |
Neg logit contributions | |
cerning | -0.261 |
ude | -0.260 |
asm | -0.243 |
ingu | -0.240 |
cius | -0.234 |
Token | Short |
Token ID | 10073 |
Feature activation | +9.06e+00 |
Pos logit contributions | |
cerning | +12.685 |
ude | +12.457 |
asm | +12.379 |
clus | +11.799 |
ingu | +11.597 |
Neg logit contributions | |
deadliest | -7.987 |
occupants | -7.860 |
losses | -7.628 |
quotas | -7.477 |
revenues | -7.202 |
Token | � |
Token ID | 447 |
Feature activation | +7.62e+00 |
Pos logit contributions | |
ude | +10.577 |
cerning | +10.519 |
ingu | +10.049 |
asm | +9.951 |
clus | +9.713 |
Neg logit contributions | |
deadliest | -5.611 |
losses | -5.407 |
occupants | -5.345 |
quotas | -5.225 |
victims | -5.223 |
Token | � |
Token ID | 247 |
Feature activation | +7.92e+00 |
Pos logit contributions | |
ude | +5.276 |
cerning | +5.075 |
asm | +4.742 |
ingu | +4.548 |
cius | +4.465 |
Neg logit contributions | |
deadliest | -8.492 |
losses | -8.360 |
victims | -8.273 |
occupants | -8.237 |
quotas | -8.109 |
Token | s |
Token ID | 82 |
Feature activation | +9.34e+00 |
Pos logit contributions | |
ude | +8.254 |
cerning | +8.137 |
asm | +7.625 |
ingu | +7.540 |
cius | +7.372 |
Neg logit contributions | |
deadliest | -6.059 |
losses | -5.952 |
occupants | -5.903 |
quotas | -5.796 |
victims | -5.692 |
Token | case |
Token ID | 1339 |
Feature activation | +5.78e+00 |
Pos logit contributions | |
ude | +9.440 |
cerning | +9.374 |
asm | +8.750 |
ingu | +8.643 |
clus | +8.499 |
Neg logit contributions | |
deadliest | -6.848 |
occupants | -6.839 |
losses | -6.794 |
quotas | -6.713 |
victims | -6.491 |
Token | The |
Token ID | 464 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ude | +2.048 |
cerning | +2.038 |
asm | +1.926 |
ingu | +1.873 |
clus | +1.838 |
Neg logit contributions | |
deadliest | -2.138 |
losses | -2.098 |
occupants | -2.051 |
victims | -2.042 |
quotas | -1.999 |
Token | younger |
Token ID | 7099 |
Feature activation | -6.80e+00 |
Pos logit contributions | |
deadliest | +0.866 |
losses | +0.854 |
victims | +0.836 |
occupants | +0.831 |
quotas | +0.816 |
Neg logit contributions | |
cerning | -1.190 |
ude | -1.188 |
asm | -1.122 |
ingu | -1.109 |
cius | -1.092 |
Token | you |
Token ID | 345 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
losses | +5.244 |
victims | +5.209 |
deadliest | +5.181 |
occupants | +5.095 |
quotas | +4.893 |
Neg logit contributions | |
cerning | -7.292 |
ude | -7.111 |
ingu | -6.806 |
asm | -6.682 |
cius | -6.604 |
Token | are |
Token ID | 389 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
ude | +4.018 |
cerning | +3.888 |
asm | +3.603 |
ingu | +3.587 |
cius | +3.486 |
Neg logit contributions | |
deadliest | -5.496 |
losses | -5.420 |
occupants | -5.304 |
victims | -5.292 |
quotas | -5.268 |
Token | ," |
Token ID | 553 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
cerning | +2.063 |
ude | +2.053 |
asm | +1.893 |
ingu | +1.883 |
cius | +1.843 |
Neg logit contributions | |
deadliest | -2.550 |
losses | -2.524 |
victims | -2.478 |
occupants | -2.467 |
quotas | -2.436 |
Token | D |
Token ID | 360 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
cerning | +0.934 |
ude | +0.928 |
asm | +0.859 |
ingu | +0.851 |
cius | +0.832 |
Neg logit contributions | |
deadliest | -1.128 |
losses | -1.116 |
victims | -1.098 |
occupants | -1.091 |
quotas | -1.074 |
Token | ella |
Token ID | 12627 |
Feature activation | +5.80e+00 |
Pos logit contributions | |
ude | +3.976 |
cerning | +3.811 |
asm | +3.250 |
ingu | +3.009 |
yt | +2.958 |
Neg logit contributions | |
deadliest | -17.886 |
losses | -17.833 |
occupants | -17.650 |
quotas | -17.584 |
victims | -17.505 |
Token | Vol |
Token ID | 4709 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
ude | +4.769 |
cerning | +4.662 |
asm | +4.344 |
ingu | +4.168 |
cius | +4.151 |
Neg logit contributions | |
deadliest | -5.953 |
losses | -5.781 |
occupants | -5.728 |
victims | -5.676 |
quotas | -5.577 |
Token | pe |
Token ID | 431 |
Feature activation | +1.16e+01 |
Pos logit contributions | |
ude | +3.671 |
cerning | +3.204 |
asm | +2.791 |
ingu | +2.574 |
yt | +2.358 |
Neg logit contributions | |
deadliest | -17.496 |
losses | -17.463 |
occupants | -17.124 |
victims | -17.117 |
invaders | -16.967 |
Token | says |
Token ID | 1139 |
Feature activation | +5.68e+00 |
Pos logit contributions | |
ude | +9.420 |
cerning | +9.036 |
asm | +8.675 |
ingu | +8.474 |
yt | +8.201 |
Neg logit contributions | |
deadliest | -11.109 |
occupants | -10.702 |
losses | -10.605 |
quotas | -10.429 |
victims | -10.378 |
Token | , |
Token ID | 11 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
cerning | +4.630 |
ude | +4.607 |
asm | +4.235 |
ingu | +4.208 |
cius | +4.108 |
Neg logit contributions | |
deadliest | -5.915 |
losses | -5.862 |
victims | -5.760 |
occupants | -5.734 |
quotas | -5.649 |
Token | the |
Token ID | 262 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
ude | +6.270 |
cerning | +6.183 |
asm | +5.986 |
ingu | +5.753 |
clus | +5.719 |
Neg logit contributions | |
deadliest | -4.852 |
losses | -4.726 |
quotas | -4.686 |
occupants | -4.586 |
victims | -4.571 |
Token | flesh |
Token ID | 11222 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
cerning | +0.089 |
ude | +0.089 |
asm | +0.084 |
ingu | +0.083 |
cius | +0.081 |
Neg logit contributions | |
deadliest | -0.061 |
losses | -0.060 |
victims | -0.059 |
occupants | -0.058 |
quotas | -0.058 |
Token | ," |
Token ID | 553 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
losses | +3.033 |
deadliest | +3.027 |
victims | +2.990 |
occupants | +2.930 |
quotas | +2.880 |
Neg logit contributions | |
cerning | -2.988 |
ude | -2.929 |
ingu | -2.753 |
asm | -2.720 |
cius | -2.686 |
Token | said |
Token ID | 531 |
Feature activation | +7.69e-01 |
Pos logit contributions | |
cerning | +1.035 |
ude | +1.031 |
asm | +0.965 |
ingu | +0.958 |
cius | +0.940 |
Neg logit contributions | |
deadliest | -0.875 |
losses | -0.866 |
victims | -0.849 |
occupants | -0.841 |
quotas | -0.830 |
Token | Adrian |
Token ID | 21462 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
cerning | +0.655 |
ude | +0.650 |
asm | +0.601 |
ingu | +0.598 |
cius | +0.584 |
Neg logit contributions | |
deadliest | -0.777 |
losses | -0.771 |
victims | -0.759 |
occupants | -0.752 |
quotas | -0.742 |
Token | L |
Token ID | 406 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
ude | +8.750 |
cerning | +8.536 |
asm | +8.005 |
ingu | +7.902 |
clus | +7.623 |
Neg logit contributions | |
deadliest | -10.648 |
losses | -10.421 |
occupants | -10.283 |
victims | -10.214 |
quotas | -10.111 |
Token | ister |
Token ID | 1694 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
ude | +9.171 |
cerning | +8.728 |
ingu | +8.381 |
asm | +8.302 |
ás | +7.801 |
Neg logit contributions | |
deadliest | -13.861 |
losses | -13.662 |
quotas | -13.502 |
occupants | -13.489 |
victims | -13.461 |
Token | , |
Token ID | 11 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
ude | +10.446 |
cerning | +10.115 |
asm | +9.421 |
ingu | +9.392 |
cius | +9.225 |
Neg logit contributions | |
deadliest | -12.223 |
losses | -11.801 |
occupants | -11.682 |
victims | -11.545 |
quotas | -11.492 |
Token | at |
Token ID | 379 |
Feature activation | +9.29e+00 |
Pos logit contributions | |
ude | +5.140 |
cerning | +5.086 |
asm | +4.838 |
ingu | +4.793 |
cius | +4.665 |
Neg logit contributions | |
deadliest | -3.215 |
losses | -3.127 |
occupants | -3.058 |
victims | -3.055 |
quotas | -3.020 |
Token | the |
Token ID | 262 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
ude | +8.853 |
cerning | +8.808 |
ingu | +8.226 |
asm | +8.202 |
clus | +7.869 |
Neg logit contributions | |
deadliest | -7.860 |
losses | -7.692 |
occupants | -7.555 |
quotas | -7.521 |
victims | -7.517 |
Token | Natural |
Token ID | 12068 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
cerning | +2.105 |
ude | +2.098 |
asm | +1.968 |
ingu | +1.957 |
cius | +1.917 |
Neg logit contributions | |
deadliest | -1.682 |
losses | -1.662 |
victims | -1.624 |
occupants | -1.616 |
quotas | -1.600 |
Token |
|
Token ID | 198 |
Feature activation | -9.60e+00 |
Pos logit contributions | |
deadliest | +0.712 |
losses | +0.707 |
victims | +0.696 |
occupants | +0.688 |
quotas | +0.680 |
Neg logit contributions | |
cerning | -0.647 |
ude | -0.641 |
ingu | -0.594 |
asm | -0.593 |
cius | -0.580 |
Token |
|
Token ID | 198 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
losses | +7.536 |
victims | +7.509 |
deadliest | +7.430 |
quotas | +7.248 |
only | +7.244 |
Neg logit contributions | |
cerning | -9.584 |
ude | -9.266 |
ingu | -9.105 |
cius | -8.965 |
uggest | -8.658 |
Token | Let |
Token ID | 5756 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
deadliest | +2.479 |
losses | +2.453 |
victims | +2.424 |
occupants | +2.386 |
quotas | +2.363 |
Neg logit contributions | |
cerning | -2.701 |
ude | -2.668 |
ingu | -2.504 |
asm | -2.483 |
cius | -2.449 |
Token | � |
Token ID | 447 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
deadliest | +0.798 |
losses | +0.791 |
victims | +0.774 |
occupants | +0.762 |
quotas | +0.747 |
Neg logit contributions | |
cerning | -1.310 |
ude | -1.300 |
asm | -1.228 |
ingu | -1.225 |
cius | -1.205 |
Token | � |
Token ID | 247 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ude | +3.108 |
cerning | +3.106 |
asm | +2.855 |
ingu | +2.848 |
cius | +2.766 |
Neg logit contributions | |
deadliest | -3.678 |
losses | -3.643 |
victims | -3.582 |
occupants | -3.554 |
quotas | -3.509 |
Token | s |
Token ID | 82 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
cerning | +1.492 |
ude | +1.486 |
asm | +1.351 |
ingu | +1.341 |
cius | +1.306 |
Neg logit contributions | |
deadliest | -2.243 |
losses | -2.227 |
victims | -2.192 |
occupants | -2.178 |
quotas | -2.152 |
Token | back |
Token ID | 736 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ude | +15.619 |
cerning | +14.897 |
asm | +14.511 |
ingu | +14.346 |
dos | +13.992 |
Neg logit contributions | |
deadliest | -7.511 |
losses | -7.428 |
occupants | -7.332 |
quotas | -7.109 |
victims | -6.799 |
Token | track |
Token ID | 11659 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
ude | +0.146 |
cerning | +0.146 |
asm | +0.138 |
ingu | +0.138 |
cius | +0.136 |
Neg logit contributions | |
deadliest | -0.047 |
losses | -0.045 |
victims | -0.044 |
occupants | -0.043 |
quotas | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
cerning | +3.494 |
ude | +3.491 |
asm | +3.296 |
ingu | +3.262 |
clus | +3.184 |
Neg logit contributions | |
deadliest | -2.404 |
losses | -2.320 |
occupants | -2.279 |
victims | -2.259 |
quotas | -2.241 |
Token | bit |
Token ID | 1643 |
Feature activation | +1.80e+01 |
Pos logit contributions | |
ude | +10.074 |
cerning | +9.802 |
asm | +9.378 |
clus | +8.962 |
ingu | +8.894 |
Neg logit contributions | |
deadliest | -10.061 |
losses | -9.848 |
occupants | -9.792 |
quotas | -9.770 |
victims | -9.658 |
Token | . |
Token ID | 13 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
ude | +16.516 |
cerning | +16.222 |
asm | +15.394 |
ingu | +15.197 |
wn | +14.944 |
Neg logit contributions | |
deadliest | -12.518 |
occupants | -12.013 |
quotas | -11.769 |
losses | -11.582 |
revenues | -11.301 |
Token | in |
Token ID | 287 |
Feature activation | +9.48e+00 |
Pos logit contributions | |
losses | +3.749 |
deadliest | +3.714 |
victims | +3.673 |
occupants | +3.608 |
quotas | +3.551 |
Neg logit contributions | |
cerning | -3.194 |
ude | -3.145 |
ingu | -2.882 |
asm | -2.880 |
cius | -2.867 |
Token | 2001 |
Token ID | 5878 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ude | +8.279 |
cerning | +8.262 |
asm | +7.935 |
clus | +7.592 |
ingu | +7.417 |
Neg logit contributions | |
deadliest | -8.485 |
losses | -8.224 |
occupants | -8.150 |
quotas | -8.097 |
victims | -8.083 |
Token | and |
Token ID | 290 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
deadliest | +1.366 |
losses | +1.361 |
victims | +1.342 |
occupants | +1.325 |
quotas | +1.305 |
Neg logit contributions | |
cerning | -1.100 |
ude | -1.091 |
ingu | -1.003 |
asm | -1.003 |
cius | -0.986 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
deadliest | +1.749 |
losses | +1.736 |
victims | +1.703 |
occupants | +1.684 |
quotas | +1.656 |
Neg logit contributions | |
cerning | -2.071 |
ude | -2.054 |
asm | -1.923 |
ingu | -1.919 |
cius | -1.882 |
Token | spoke |
Token ID | 5158 |
Feature activation | -8.14e+00 |
Pos logit contributions | |
ude | +1.111 |
cerning | +1.107 |
asm | +1.048 |
ingu | +1.034 |
clus | +1.008 |
Neg logit contributions | |
deadliest | -0.786 |
losses | -0.770 |
victims | -0.749 |
occupants | -0.746 |
quotas | -0.738 |
Token | two |
Token ID | 734 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
losses | +5.926 |
deadliest | +5.918 |
victims | +5.855 |
occupants | +5.677 |
quotas | +5.549 |
Neg logit contributions | |
cerning | -8.930 |
ude | -8.748 |
ingu | -8.354 |
cius | -8.338 |
asm | -8.296 |
Token | lines |
Token ID | 3951 |
Feature activation | +6.29e+00 |
Pos logit contributions | |
cerning | +4.347 |
ude | +4.345 |
asm | +4.098 |
ingu | +4.047 |
clus | +3.990 |
Neg logit contributions | |
deadliest | -2.628 |
losses | -2.585 |
occupants | -2.490 |
victims | -2.489 |
quotas | -2.462 |
Token | of |
Token ID | 286 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
ude | +6.493 |
cerning | +6.483 |
asm | +6.098 |
ingu | +5.958 |
clus | +5.805 |
Neg logit contributions | |
deadliest | -4.947 |
losses | -4.793 |
occupants | -4.702 |
victims | -4.658 |
quotas | -4.648 |
Token | dialogue |
Token ID | 10721 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
ude | +4.985 |
cerning | +4.977 |
asm | +4.696 |
ingu | +4.597 |
clus | +4.495 |
Neg logit contributions | |
deadliest | -4.671 |
losses | -4.583 |
occupants | -4.453 |
victims | -4.447 |
quotas | -4.425 |
Token | (" |
Token ID | 5855 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
ude | +5.175 |
cerning | +5.175 |
asm | +4.891 |
ingu | +4.796 |
clus | +4.695 |
Neg logit contributions | |
deadliest | -3.040 |
losses | -2.912 |
occupants | -2.850 |
victims | -2.822 |
quotas | -2.813 |
Token | Hey |
Token ID | 10814 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
cerning | +3.186 |
ude | +3.182 |
asm | +2.947 |
ingu | +2.889 |
cius | +2.809 |
Neg logit contributions | |
deadliest | -3.942 |
losses | -3.899 |
victims | -3.825 |
occupants | -3.807 |
quotas | -3.761 |
Token | gets |
Token ID | 3011 |
Feature activation | -9.41e+00 |
Pos logit contributions | |
ude | +5.167 |
cerning | +5.108 |
ingu | +4.843 |
asm | +4.839 |
cius | +4.697 |
Neg logit contributions | |
deadliest | -3.970 |
losses | -3.835 |
victims | -3.781 |
occupants | -3.778 |
quotas | -3.730 |
Token | the |
Token ID | 262 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
losses | +7.957 |
deadliest | +7.906 |
victims | +7.853 |
occupants | +7.598 |
quotas | +7.505 |
Neg logit contributions | |
cerning | -9.217 |
ude | -9.040 |
ingu | -8.465 |
asm | -8.445 |
cius | -8.432 |
Token | spec |
Token ID | 1020 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
losses | +5.690 |
deadliest | +5.683 |
victims | +5.581 |
occupants | +5.479 |
quotas | +5.405 |
Neg logit contributions | |
cerning | -6.007 |
ude | -5.898 |
ingu | -5.524 |
asm | -5.514 |
cius | -5.432 |
Token | right |
Token ID | 826 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
cerning | +3.495 |
ude | +3.476 |
asm | +3.262 |
ingu | +3.253 |
cius | +3.189 |
Neg logit contributions | |
deadliest | -3.127 |
losses | -3.070 |
victims | -3.005 |
occupants | -3.000 |
quotas | -2.935 |
Token | at |
Token ID | 379 |
Feature activation | +6.53e+00 |
Pos logit contributions | |
cerning | +2.271 |
ude | +2.258 |
ingu | +2.101 |
asm | +2.097 |
cius | +2.053 |
Neg logit contributions | |
deadliest | -2.236 |
losses | -2.206 |
victims | -2.163 |
occupants | -2.157 |
quotas | -2.120 |
Token | first |
Token ID | 717 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ude | +6.773 |
cerning | +6.744 |
asm | +6.421 |
ingu | +6.384 |
clus | +6.146 |
Neg logit contributions | |
deadliest | -5.007 |
losses | -4.891 |
occupants | -4.819 |
victims | -4.796 |
quotas | -4.623 |
Token | go |
Token ID | 467 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
cerning | +0.889 |
ude | +0.886 |
asm | +0.814 |
ingu | +0.810 |
cius | +0.788 |
Neg logit contributions | |
deadliest | -1.157 |
losses | -1.146 |
victims | -1.129 |
occupants | -1.120 |
quotas | -1.106 |
Token | > |
Token ID | 1875 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
deadliest | +1.052 |
losses | +1.051 |
victims | +1.033 |
occupants | +1.018 |
quotas | +1.005 |
Neg logit contributions | |
cerning | -0.935 |
ude | -0.923 |
asm | -0.855 |
ingu | -0.849 |
cius | -0.835 |
Token | analytics |
Token ID | 23696 |
Feature activation | +6.01e+00 |
Pos logit contributions | |
ude | +3.662 |
cerning | +3.661 |
asm | +3.399 |
ingu | +3.381 |
cius | +3.292 |
Neg logit contributions | |
deadliest | -3.599 |
losses | -3.560 |
victims | -3.491 |
occupants | -3.474 |
quotas | -3.407 |
Token | dash |
Token ID | 14470 |
Feature activation | +5.95e+00 |
Pos logit contributions | |
ude | +6.010 |
cerning | +5.894 |
asm | +5.562 |
ingu | +5.549 |
cius | +5.385 |
Neg logit contributions | |
deadliest | -4.987 |
losses | -4.804 |
occupants | -4.772 |
victims | -4.749 |
quotas | -4.606 |
Token | boards |
Token ID | 12821 |
Feature activation | +1.44e+01 |
Pos logit contributions | |
ude | +5.360 |
cerning | +5.354 |
ingu | +5.043 |
asm | +5.000 |
cius | +4.878 |
Neg logit contributions | |
deadliest | -5.576 |
losses | -5.445 |
victims | -5.359 |
occupants | -5.352 |
quotas | -5.241 |
Token | B |
Token ID | 33 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
deadliest | +1.468 |
losses | +1.459 |
victims | +1.435 |
occupants | +1.420 |
quotas | +1.402 |
Neg logit contributions | |
cerning | -1.306 |
ude | -1.294 |
asm | -1.198 |
ingu | -1.194 |
cius | -1.169 |
Token | DP |
Token ID | 6322 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
ude | +2.985 |
cerning | +2.972 |
asm | +2.739 |
ingu | +2.714 |
cius | +2.608 |
Neg logit contributions | |
deadliest | -4.873 |
losses | -4.818 |
victims | -4.773 |
occupants | -4.734 |
quotas | -4.653 |
Token | is |
Token ID | 318 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
losses | +4.307 |
deadliest | +4.255 |
victims | +4.182 |
occupants | +4.089 |
quotas | +4.076 |
Neg logit contributions | |
cerning | -5.511 |
ude | -5.402 |
asm | -5.088 |
ingu | -5.052 |
cius | -4.996 |
Token | central |
Token ID | 4318 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
deadliest | +3.850 |
losses | +3.834 |
victims | +3.743 |
occupants | +3.655 |
quotas | +3.617 |
Neg logit contributions | |
cerning | -5.520 |
ude | -5.453 |
asm | -5.110 |
ingu | -5.096 |
cius | -5.050 |
Token | to |
Token ID | 284 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
deadliest | +1.368 |
losses | +1.362 |
victims | +1.323 |
occupants | +1.295 |
quotas | +1.278 |
Neg logit contributions | |
cerning | -2.620 |
ude | -2.597 |
asm | -2.464 |
ingu | -2.457 |
cius | -2.426 |
Token | understanding |
Token ID | 4547 |
Feature activation | +7.08e+00 |
Pos logit contributions | |
cerning | +11.144 |
ude | +11.048 |
asm | +10.779 |
ingu | +10.551 |
clus | +10.310 |
Neg logit contributions | |
deadliest | -7.494 |
occupants | -7.222 |
victims | -7.106 |
losses | -7.074 |
quotas | -6.911 |
Token | network |
Token ID | 3127 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
cerning | +7.181 |
ude | +7.174 |
asm | +6.878 |
ingu | +6.716 |
cius | +6.556 |
Neg logit contributions | |
deadliest | -5.463 |
occupants | -5.250 |
victims | -5.223 |
losses | -5.201 |
quotas | -5.071 |
Token | performance |
Token ID | 2854 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
losses | +4.697 |
deadliest | +4.567 |
victims | +4.493 |
occupants | +4.448 |
quotas | +4.375 |
Neg logit contributions | |
cerning | -6.877 |
ude | -6.746 |
ingu | -6.344 |
asm | -6.311 |
cius | -6.288 |
Token | . |
Token ID | 13 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
cerning | +0.434 |
ude | +0.431 |
asm | +0.400 |
ingu | +0.399 |
cius | +0.390 |
Neg logit contributions | |
deadliest | -0.452 |
losses | -0.448 |
victims | -0.441 |
occupants | -0.437 |
quotas | -0.430 |
Token | Consider |
Token ID | 12642 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
cerning | +0.157 |
ude | +0.156 |
asm | +0.146 |
ingu | +0.144 |
cius | +0.140 |
Neg logit contributions | |
deadliest | -0.188 |
losses | -0.185 |
victims | -0.183 |
occupants | -0.182 |
quotas | -0.178 |
Token | what |
Token ID | 644 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
cerning | +0.311 |
ude | +0.308 |
asm | +0.289 |
ingu | +0.289 |
cius | +0.284 |
Neg logit contributions | |
deadliest | -0.233 |
losses | -0.231 |
victims | -0.226 |
occupants | -0.223 |
quotas | -0.220 |
Token | ] |
Token ID | 60 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
cerning | +1.236 |
ude | +1.224 |
ingu | +1.121 |
asm | +1.119 |
cius | +1.098 |
Neg logit contributions | |
deadliest | -1.679 |
losses | -1.671 |
victims | -1.645 |
occupants | -1.628 |
quotas | -1.606 |
Token | While |
Token ID | 2893 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
cerning | +0.144 |
ude | +0.143 |
asm | +0.133 |
ingu | +0.132 |
cius | +0.129 |
Neg logit contributions | |
deadliest | -0.161 |
losses | -0.160 |
victims | -0.157 |
occupants | -0.156 |
quotas | -0.154 |
Token | Cob |
Token ID | 14828 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
deadliest | +1.263 |
losses | +1.251 |
victims | +1.226 |
occupants | +1.216 |
quotas | +1.197 |
Neg logit contributions | |
cerning | -1.462 |
ude | -1.450 |
asm | -1.358 |
ingu | -1.352 |
cius | -1.326 |
Token | ain |
Token ID | 391 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
losses | +4.089 |
deadliest | +4.086 |
victims | +4.030 |
occupants | +3.982 |
quotas | +3.940 |
Neg logit contributions | |
cerning | -1.438 |
ude | -1.388 |
ingu | -1.190 |
cius | -1.182 |
asm | -1.181 |
Token | regarded |
Token ID | 11987 |
Feature activation | -7.46e+00 |
Pos logit contributions | |
deadliest | +3.124 |
losses | +3.100 |
victims | +3.039 |
occupants | +2.993 |
quotas | +2.936 |
Neg logit contributions | |
cerning | -4.445 |
ude | -4.407 |
asm | -4.155 |
ingu | -4.141 |
cius | -4.072 |
Token | In |
Token ID | 554 |
Feature activation | +9.19e+00 |
Pos logit contributions | |
losses | +6.614 |
deadliest | +6.532 |
victims | +6.525 |
occupants | +6.236 |
quotas | +6.111 |
Neg logit contributions | |
cerning | -7.122 |
ude | -6.982 |
ingu | -6.553 |
asm | -6.485 |
cius | -6.458 |
Token | U |
Token ID | 471 |
Feature activation | +9.10e+00 |
Pos logit contributions | |
cerning | +8.667 |
ude | +8.626 |
asm | +8.071 |
clus | +7.897 |
cius | +7.730 |
Neg logit contributions | |
deadliest | -8.000 |
losses | -7.794 |
quotas | -7.707 |
occupants | -7.682 |
victims | -7.597 |
Token | ter |
Token ID | 353 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
ude | +5.583 |
cerning | +5.539 |
ingu | +4.811 |
asm | +4.782 |
cius | +4.708 |
Neg logit contributions | |
deadliest | -10.966 |
losses | -10.803 |
victims | -10.759 |
quotas | -10.675 |
occupants | -10.615 |
Token | o |
Token ID | 78 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
cerning | +1.889 |
ude | +1.867 |
asm | +1.667 |
ingu | +1.655 |
cius | +1.596 |
Neg logit contributions | |
deadliest | -3.955 |
losses | -3.931 |
victims | -3.882 |
occupants | -3.853 |
quotas | -3.812 |
Token | " |
Token ID | 366 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
cerning | +3.787 |
ude | +3.781 |
asm | +3.563 |
ingu | +3.558 |
cius | +3.464 |
Neg logit contributions | |
deadliest | -2.378 |
losses | -2.327 |
occupants | -2.267 |
victims | -2.256 |
quotas | -2.245 |
Token | for |
Token ID | 1640 |
Feature activation | +5.16e+00 |
Pos logit contributions | |
cerning | +1.875 |
ude | +1.865 |
asm | +1.717 |
ingu | +1.676 |
cius | +1.618 |
Neg logit contributions | |
deadliest | -2.871 |
losses | -2.846 |
occupants | -2.798 |
victims | -2.795 |
quotas | -2.770 |
Token | should |
Token ID | 815 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
ude | +7.757 |
cerning | +7.634 |
asm | +7.312 |
ingu | +7.056 |
wn | +7.046 |
Neg logit contributions | |
deadliest | -5.148 |
losses | -4.830 |
occupants | -4.782 |
victims | -4.721 |
quotas | -4.630 |
Token | meet |
Token ID | 1826 |
Feature activation | -7.42e+00 |
Pos logit contributions | |
ude | +2.787 |
cerning | +2.775 |
asm | +2.573 |
ingu | +2.565 |
cius | +2.491 |
Neg logit contributions | |
deadliest | -2.704 |
losses | -2.670 |
victims | -2.615 |
occupants | -2.602 |
quotas | -2.560 |
Token | its |
Token ID | 663 |
Feature activation | -5.36e+00 |
Pos logit contributions | |
losses | +5.901 |
deadliest | +5.797 |
victims | +5.728 |
occupants | +5.588 |
quotas | +5.545 |
Neg logit contributions | |
cerning | -7.779 |
ude | -7.536 |
ingu | -7.237 |
cius | -7.104 |
asm | -7.075 |
Token | match |
Token ID | 2872 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
deadliest | +4.093 |
losses | +4.051 |
victims | +3.977 |
occupants | +3.898 |
quotas | +3.869 |
Neg logit contributions | |
cerning | -5.918 |
ude | -5.857 |
ingu | -5.572 |
asm | -5.461 |
cius | -5.455 |
Token | in |
Token ID | 287 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
cerning | +1.348 |
ude | +1.341 |
asm | +1.265 |
ingu | +1.254 |
cius | +1.235 |
Neg logit contributions | |
deadliest | -0.922 |
losses | -0.908 |
victims | -0.891 |
occupants | -0.884 |
quotas | -0.863 |
Token | the |
Token ID | 262 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ude | +10.074 |
cerning | +9.994 |
asm | +9.934 |
clus | +9.298 |
ingu | +9.035 |
Neg logit contributions | |
deadliest | -7.863 |
occupants | -7.499 |
losses | -7.487 |
victims | -7.450 |
quotas | -7.383 |
Token | Soon |
Token ID | 15894 |
Feature activation | -7.02e+00 |
Pos logit contributions | |
cerning | +1.188 |
ude | +1.182 |
asm | +1.113 |
ingu | +1.099 |
cius | +1.081 |
Neg logit contributions | |
deadliest | -0.911 |
losses | -0.900 |
victims | -0.884 |
occupants | -0.876 |
quotas | -0.864 |
Token | ers |
Token ID | 364 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
losses | +5.042 |
deadliest | +4.942 |
victims | +4.923 |
occupants | +4.798 |
quotas | +4.672 |
Neg logit contributions | |
cerning | -7.825 |
ude | -7.621 |
ingu | -7.298 |
cius | -7.215 |
asm | -7.157 |
Token | ' |
Token ID | 6 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
losses | +4.006 |
deadliest | +4.005 |
victims | +3.912 |
occupants | +3.864 |
quotas | +3.785 |
Neg logit contributions | |
cerning | -4.165 |
ude | -4.101 |
ingu | -3.845 |
asm | -3.815 |
cius | -3.764 |
Token | diamond |
Token ID | 15291 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
deadliest | +1.411 |
losses | +1.398 |
victims | +1.369 |
occupants | +1.353 |
quotas | +1.329 |
Neg logit contributions | |
cerning | -1.924 |
ude | -1.910 |
asm | -1.796 |
ingu | -1.790 |
cius | -1.757 |
Token | formations |
Token ID | 30648 |
Feature activation | +4.75e+00 |
Pos logit contributions | |
losses | +4.715 |
deadliest | +4.690 |
victims | +4.642 |
occupants | +4.587 |
quotas | +4.478 |
Neg logit contributions | |
cerning | -4.983 |
ude | -4.890 |
ingu | -4.648 |
cius | -4.550 |
asm | -4.541 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.80e+00 |
Pos logit contributions | |
ude | +7.489 |
cerning | +7.383 |
asm | +6.808 |
ingu | +6.622 |
cius | +6.593 |
Neg logit contributions | |
deadliest | -9.383 |
losses | -9.208 |
occupants | -9.081 |
victims | -8.884 |
quotas | -8.791 |
Token | on |
Token ID | 319 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
ude | +5.729 |
cerning | +5.694 |
asm | +5.357 |
ingu | +5.276 |
cius | +5.199 |
Neg logit contributions | |
deadliest | -4.845 |
losses | -4.783 |
occupants | -4.626 |
victims | -4.574 |
quotas | -4.527 |
Token | YouTube |
Token ID | 7444 |
Feature activation | +4.80e+00 |
Pos logit contributions | |
ude | +10.244 |
cerning | +10.093 |
asm | +9.904 |
ingu | +9.495 |
clus | +9.366 |
Neg logit contributions | |
deadliest | -8.392 |
losses | -8.273 |
occupants | -8.147 |
victims | -7.848 |
quotas | -7.838 |
Token | : |
Token ID | 25 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
ude | +3.762 |
cerning | +3.759 |
asm | +3.473 |
ingu | +3.415 |
cius | +3.307 |
Neg logit contributions | |
deadliest | -5.092 |
losses | -5.033 |
occupants | -4.912 |
victims | -4.908 |
quotas | -4.840 |
Token | - |
Token ID | 532 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
ude | +2.267 |
cerning | +2.259 |
asm | +2.084 |
ingu | +2.048 |
cius | +1.998 |
Neg logit contributions | |
deadliest | -2.829 |
losses | -2.819 |
occupants | -2.751 |
victims | -2.740 |
quotas | -2.701 |
Token | January |
Token ID | 3269 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ude | +2.681 |
cerning | +2.673 |
asm | +2.478 |
ingu | +2.440 |
cius | +2.378 |
Neg logit contributions | |
deadliest | -3.258 |
losses | -3.252 |
occupants | -3.175 |
victims | -3.164 |
quotas | -3.116 |
Token | 18 |
Token ID | 1248 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
deadliest | +1.027 |
losses | +1.015 |
victims | +0.992 |
occupants | +0.984 |
quotas | +0.964 |
Neg logit contributions | |
cerning | -1.427 |
ude | -1.422 |
asm | -1.337 |
ingu | -1.328 |
cius | -1.304 |
Token | , |
Token ID | 11 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
cerning | +2.202 |
ude | +2.201 |
asm | +2.016 |
ingu | +1.986 |
cius | +1.934 |
Neg logit contributions | |
deadliest | -3.129 |
losses | -3.110 |
victims | -3.055 |
occupants | -3.047 |
quotas | -2.998 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ude | +3.077 |
cerning | +3.053 |
asm | +2.862 |
ingu | +2.790 |
clus | +2.718 |
Neg logit contributions | |
deadliest | -3.268 |
losses | -3.230 |
occupants | -3.150 |
victims | -3.147 |
quotas | -3.098 |
Token | It |
Token ID | 1026 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
cerning | +0.945 |
ude | +0.938 |
asm | +0.869 |
ingu | +0.863 |
cius | +0.841 |
Neg logit contributions | |
deadliest | -1.097 |
losses | -1.091 |
victims | -1.070 |
occupants | -1.065 |
quotas | -1.048 |
Token | was |
Token ID | 373 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
ude | +0.674 |
cerning | +0.671 |
asm | +0.636 |
ingu | +0.630 |
cius | +0.618 |
Neg logit contributions | |
deadliest | -0.380 |
losses | -0.372 |
occupants | -0.362 |
victims | -0.360 |
quotas | -0.353 |
Token | added |
Token ID | 2087 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
ude | +9.397 |
cerning | +9.245 |
ingu | +8.636 |
asm | +8.621 |
yt | +8.367 |
Neg logit contributions | |
deadliest | -7.428 |
losses | -7.111 |
occupants | -7.079 |
victims | -7.020 |
quotas | -6.984 |
Token | : |
Token ID | 25 |
Feature activation | -8.89e-01 |
Pos logit contributions | |
losses | +4.549 |
deadliest | +4.501 |
victims | +4.486 |
occupants | +4.427 |
quotas | +4.384 |
Neg logit contributions | |
cerning | -2.578 |
ude | -2.568 |
ingu | -2.329 |
asm | -2.329 |
cius | -2.321 |
Token | " |
Token ID | 366 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
deadliest | +0.556 |
losses | +0.555 |
victims | +0.544 |
occupants | +0.530 |
quotas | +0.519 |
Neg logit contributions | |
cerning | -1.087 |
ude | -1.083 |
asm | -1.025 |
ingu | -1.023 |
cius | -1.018 |
Token | We |
Token ID | 1135 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
ude | +1.671 |
cerning | +1.671 |
asm | +1.553 |
ingu | +1.503 |
clus | +1.480 |
Neg logit contributions | |
deadliest | -2.037 |
losses | -1.994 |
victims | -1.962 |
occupants | -1.960 |
quotas | -1.916 |
Token | are |
Token ID | 389 |
Feature activation | -4.89e+00 |
Pos logit contributions | |
ude | +2.386 |
cerning | +2.358 |
asm | +2.188 |
ingu | +2.173 |
cius | +2.106 |
Neg logit contributions | |
deadliest | -2.815 |
losses | -2.777 |
victims | -2.720 |
occupants | -2.715 |
quotas | -2.682 |
Token | excited |
Token ID | 6568 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
deadliest | +4.591 |
losses | +4.571 |
victims | +4.508 |
occupants | +4.441 |
quotas | +4.368 |
Neg logit contributions | |
cerning | -4.501 |
ude | -4.453 |
asm | -4.133 |
ingu | -4.133 |
cius | -4.056 |
Token | to |
Token ID | 284 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
deadliest | +0.818 |
losses | +0.811 |
victims | +0.791 |
occupants | +0.780 |
quotas | +0.763 |
Neg logit contributions | |
cerning | -1.468 |
ude | -1.458 |
asm | -1.380 |
ingu | -1.377 |
cius | -1.355 |
Token | be |
Token ID | 307 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
ude | +9.924 |
cerning | +9.497 |
asm | +9.085 |
clus | +8.959 |
ingu | +8.838 |
Neg logit contributions | |
deadliest | -10.761 |
losses | -10.430 |
quotas | -10.233 |
occupants | -10.177 |
victims | -10.040 |
Token | the |
Token ID | 262 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
deadliest | +1.754 |
losses | +1.735 |
victims | +1.698 |
occupants | +1.681 |
quotas | +1.655 |
Neg logit contributions | |
cerning | -2.184 |
ude | -2.173 |
asm | -2.037 |
ingu | -2.028 |
cius | -1.988 |
Token | first |
Token ID | 717 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
deadliest | +2.281 |
losses | +2.247 |
victims | +2.214 |
occupants | +2.168 |
quotas | +2.088 |
Neg logit contributions | |
cerning | -5.212 |
ude | -5.152 |
asm | -4.920 |
ingu | -4.903 |
cius | -4.844 |
Token | ever |
Token ID | 1683 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
deadliest | +3.460 |
losses | +3.454 |
victims | +3.400 |
occupants | +3.348 |
quotas | +3.271 |
Neg logit contributions | |
cerning | -4.469 |
ude | -4.407 |
asm | -4.163 |
ingu | -4.143 |
cius | -4.082 |
Token | Conservatives |
Token ID | 17897 |
Feature activation | -1.14e+01 |
Pos logit contributions | |
losses | +4.554 |
deadliest | +4.518 |
victims | +4.447 |
occupants | +4.347 |
quotas | +4.304 |
Neg logit contributions | |
cerning | -5.993 |
ude | -5.903 |
ingu | -5.575 |
asm | -5.555 |
cius | -5.446 |
Token | made |
Token ID | 925 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
losses | +10.675 |
victims | +10.432 |
deadliest | +10.430 |
occupants | +10.123 |
only | +10.081 |
Neg logit contributions | |
cerning | -10.037 |
ude | -9.405 |
asm | -9.018 |
ingu | -8.875 |
cius | -8.702 |
Token | major |
Token ID | 1688 |
Feature activation | -8.45e-01 |
Pos logit contributions | |
losses | +10.054 |
victims | +9.666 |
deadliest | +9.495 |
quotas | +9.400 |
revenues | +9.272 |
Neg logit contributions | |
cerning | -12.958 |
ude | -12.801 |
ingu | -12.174 |
cius | -12.146 |
asm | -11.841 |
Token | changes |
Token ID | 2458 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
deadliest | +0.423 |
losses | +0.401 |
victims | +0.395 |
occupants | +0.391 |
quotas | +0.375 |
Neg logit contributions | |
cerning | -1.149 |
ude | -1.146 |
asm | -1.090 |
ingu | -1.086 |
cius | -1.067 |
Token | to |
Token ID | 284 |
Feature activation | +7.08e+00 |
Pos logit contributions | |
cerning | +1.007 |
ude | +1.004 |
asm | +0.948 |
ingu | +0.942 |
cius | +0.926 |
Neg logit contributions | |
deadliest | -0.599 |
losses | -0.585 |
victims | -0.570 |
occupants | -0.569 |
quotas | -0.555 |
Token | public |
Token ID | 1171 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
ude | +7.638 |
cerning | +7.521 |
asm | +7.180 |
ingu | +7.110 |
clus | +6.996 |
Neg logit contributions | |
deadliest | -5.200 |
losses | -4.883 |
occupants | -4.856 |
victims | -4.792 |
safest | -4.640 |
Token | servants |
Token ID | 17523 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
losses | +3.468 |
deadliest | +3.413 |
victims | +3.371 |
occupants | +3.293 |
quotas | +3.249 |
Neg logit contributions | |
cerning | -6.145 |
ude | -6.096 |
asm | -5.775 |
ingu | -5.768 |
cius | -5.698 |
Token | � |
Token ID | 447 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
ude | +2.155 |
cerning | +2.144 |
ingu | +2.004 |
asm | +2.001 |
cius | +1.962 |
Neg logit contributions | |
deadliest | -1.939 |
losses | -1.876 |
victims | -1.857 |
occupants | -1.855 |
quotas | -1.779 |
Token | � |
Token ID | 247 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
cerning | +0.534 |
ude | +0.528 |
asm | +0.487 |
ingu | +0.485 |
cius | +0.473 |
Neg logit contributions | |
deadliest | -0.680 |
losses | -0.675 |
victims | -0.665 |
occupants | -0.659 |
quotas | -0.650 |
Token | pay |
Token ID | 1414 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
deadliest | +0.769 |
losses | +0.761 |
victims | +0.748 |
occupants | +0.743 |
quotas | +0.730 |
Neg logit contributions | |
cerning | -0.732 |
ude | -0.728 |
asm | -0.675 |
ingu | -0.671 |
cius | -0.657 |
Token | and |
Token ID | 290 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
losses | +5.006 |
victims | +4.866 |
deadliest | +4.862 |
quotas | +4.819 |
occupants | +4.724 |
Neg logit contributions | |
cerning | -5.239 |
ude | -5.112 |
asm | -4.850 |
ingu | -4.804 |
cius | -4.681 |
Token | : |
Token ID | 1058 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
losses | +6.783 |
victims | +6.663 |
deadliest | +6.648 |
occupants | +6.453 |
quotas | +6.412 |
Neg logit contributions | |
cerning | -6.503 |
ude | -6.270 |
ingu | -5.970 |
cius | -5.872 |
asm | -5.790 |
Token | Board |
Token ID | 5926 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
losses | +3.459 |
deadliest | +3.459 |
victims | +3.404 |
occupants | +3.368 |
quotas | +3.308 |
Neg logit contributions | |
cerning | -2.211 |
ude | -2.175 |
ingu | -1.980 |
asm | -1.963 |
cius | -1.922 |
Token | = |
Token ID | 796 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
deadliest | +2.727 |
losses | +2.723 |
victims | +2.683 |
occupants | +2.655 |
quotas | +2.609 |
Neg logit contributions | |
cerning | -2.110 |
ude | -2.074 |
ingu | -1.898 |
asm | -1.892 |
cius | -1.860 |
Token | self |
Token ID | 2116 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
cerning | +0.160 |
ude | +0.159 |
asm | +0.148 |
ingu | +0.147 |
cius | +0.144 |
Neg logit contributions | |
deadliest | -0.168 |
losses | -0.167 |
victims | -0.164 |
occupants | -0.162 |
quotas | -0.160 |
Token | . |
Token ID | 764 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
deadliest | +0.757 |
losses | +0.755 |
victims | +0.745 |
occupants | +0.737 |
quotas | +0.726 |
Neg logit contributions | |
cerning | -0.505 |
ude | -0.496 |
ingu | -0.448 |
asm | -0.448 |
cius | -0.436 |
Token | prev |
Token ID | 47050 |
Feature activation | -5.40e+00 |
Pos logit contributions | |
cerning | +1.161 |
ude | +1.153 |
asm | +1.058 |
ingu | +1.046 |
cius | +1.018 |
Neg logit contributions | |
deadliest | -1.736 |
losses | -1.718 |
victims | -1.694 |
occupants | -1.683 |
quotas | -1.664 |
Token | _ |
Token ID | 62 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
losses | +5.117 |
deadliest | +5.114 |
victims | +5.051 |
occupants | +4.952 |
quotas | +4.825 |
Neg logit contributions | |
cerning | -4.722 |
ude | -4.589 |
ingu | -4.253 |
clus | -4.238 |
asm | -4.228 |
Token | state |
Token ID | 5219 |
Feature activation | +9.44e-02 |
Pos logit contributions | |
deadliest | +2.649 |
losses | +2.645 |
victims | +2.608 |
occupants | +2.576 |
quotas | +2.542 |
Neg logit contributions | |
cerning | -2.028 |
ude | -1.999 |
ingu | -1.839 |
asm | -1.829 |
cius | -1.792 |
Token | . |
Token ID | 764 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
cerning | +0.069 |
ude | +0.068 |
asm | +0.061 |
ingu | +0.061 |
cius | +0.060 |
Neg logit contributions | |
deadliest | -0.107 |
losses | -0.107 |
victims | -0.105 |
occupants | -0.104 |
quotas | -0.103 |
Token | un |
Token ID | 403 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
cerning | +0.931 |
ude | +0.924 |
asm | +0.830 |
ingu | +0.817 |
cius | +0.789 |
Neg logit contributions | |
deadliest | -1.959 |
losses | -1.943 |
victims | -1.918 |
occupants | -1.907 |
quotas | -1.887 |
Token | wrap |
Token ID | 37150 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
cerning | +2.272 |
ude | +2.256 |
asm | +2.038 |
ingu | +1.967 |
apply | +1.938 |
Neg logit contributions | |
deadliest | -4.966 |
losses | -4.914 |
occupants | -4.846 |
victims | -4.837 |
quotas | -4.809 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
deadliest | +4.152 |
losses | +4.144 |
victims | +4.112 |
occupants | +4.024 |
quotas | +3.972 |
Neg logit contributions | |
cerning | -3.947 |
ude | -3.892 |
asm | -3.642 |
ingu | -3.624 |
cius | -3.569 |
Token | � |
Token ID | 564 |
Feature activation | +4.75e+00 |
Pos logit contributions | |
cerning | +1.270 |
ude | +1.268 |
asm | +1.163 |
ingu | +1.136 |
cius | +1.107 |
Neg logit contributions | |
deadliest | -1.955 |
losses | -1.940 |
victims | -1.904 |
occupants | -1.896 |
quotas | -1.866 |
Token | � |
Token ID | 251 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
ude | +3.399 |
cerning | +3.360 |
asm | +3.097 |
ingu | +3.032 |
cius | +2.975 |
Neg logit contributions | |
deadliest | -5.299 |
losses | -5.238 |
victims | -5.193 |
occupants | -5.152 |
quotas | -5.082 |
Token | Then |
Token ID | 3244 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
deadliest | +0.192 |
losses | +0.190 |
victims | +0.187 |
occupants | +0.186 |
quotas | +0.183 |
Neg logit contributions | |
cerning | -0.147 |
ude | -0.146 |
asm | -0.135 |
ingu | -0.133 |
cius | -0.130 |
Token | , |
Token ID | 11 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
cerning | +0.379 |
ude | +0.378 |
asm | +0.351 |
ingu | +0.347 |
cius | +0.339 |
Neg logit contributions | |
deadliest | -0.428 |
losses | -0.422 |
victims | -0.415 |
occupants | -0.413 |
quotas | -0.407 |
Token | it |
Token ID | 340 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
deadliest | +2.450 |
losses | +2.426 |
victims | +2.382 |
occupants | +2.362 |
quotas | +2.328 |
Neg logit contributions | |
cerning | -2.440 |
ude | -2.425 |
asm | -2.262 |
ingu | -2.241 |
cius | -2.194 |
Token | happened |
Token ID | 3022 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
deadliest | +1.133 |
losses | +1.119 |
victims | +1.093 |
occupants | +1.080 |
quotas | +1.057 |
Neg logit contributions | |
cerning | -1.927 |
ude | -1.918 |
asm | -1.814 |
ingu | -1.806 |
cius | -1.774 |
Token | . |
Token ID | 13 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
deadliest | +6.905 |
losses | +6.893 |
victims | +6.856 |
occupants | +6.699 |
quotas | +6.630 |
Neg logit contributions | |
cerning | -4.894 |
ude | -4.792 |
cius | -4.405 |
ingu | -4.369 |
asm | -4.330 |
Token | He |
Token ID | 679 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
deadliest | +2.724 |
losses | +2.709 |
victims | +2.674 |
occupants | +2.644 |
quotas | +2.613 |
Neg logit contributions | |
cerning | -2.034 |
ude | -2.011 |
ingu | -1.844 |
asm | -1.838 |
cius | -1.799 |
Token | said |
Token ID | 531 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
deadliest | +0.121 |
losses | +0.120 |
victims | +0.117 |
occupants | +0.116 |
quotas | +0.114 |
Neg logit contributions | |
cerning | -0.178 |
ude | -0.177 |
asm | -0.167 |
ingu | -0.166 |
cius | -0.163 |
Token | , |
Token ID | 11 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
deadliest | +5.908 |
losses | +5.888 |
victims | +5.854 |
occupants | +5.732 |
quotas | +5.660 |
Neg logit contributions | |
cerning | -5.432 |
ude | -5.338 |
ingu | -4.985 |
asm | -4.916 |
cius | -4.911 |
Token | of |
Token ID | 286 |
Feature activation | +4.98e+00 |
Pos logit contributions | |
cerning | +3.066 |
ude | +3.036 |
asm | +2.813 |
ingu | +2.778 |
cius | +2.721 |
Neg logit contributions | |
deadliest | -3.402 |
losses | -3.348 |
victims | -3.297 |
occupants | -3.294 |
quotas | -3.236 |
Token | hard |
Token ID | 1327 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
cerning | +6.480 |
ude | +6.440 |
asm | +6.171 |
ingu | +6.008 |
clus | +5.959 |
Neg logit contributions | |
deadliest | -2.538 |
occupants | -2.350 |
losses | -2.308 |
quotas | -2.273 |
victims | -2.267 |
Token | trade |
Token ID | 3292 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
deadliest | +1.041 |
losses | +1.024 |
victims | +1.007 |
occupants | +0.998 |
quotas | +0.979 |
Neg logit contributions | |
cerning | -1.387 |
ude | -1.379 |
asm | -1.294 |
ingu | -1.289 |
cius | -1.266 |
Token | offs |
Token ID | 8210 |
Feature activation | +1.71e+01 |
Pos logit contributions | |
losses | +1.685 |
deadliest | +1.679 |
victims | +1.643 |
occupants | +1.616 |
quotas | +1.600 |
Neg logit contributions | |
cerning | -2.048 |
ude | -2.031 |
asm | -1.901 |
ingu | -1.899 |
cius | -1.857 |
Token | between |
Token ID | 1022 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
cerning | +14.973 |
ude | +14.926 |
asm | +14.272 |
yt | +14.014 |
cius | +13.819 |
Neg logit contributions | |
deadliest | -12.295 |
occupants | -11.557 |
losses | -11.326 |
quotas | -11.280 |
safest | -11.153 |
Token | moving |
Token ID | 3867 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
cerning | +5.181 |
ude | +5.159 |
asm | +4.952 |
ingu | +4.807 |
clus | +4.795 |
Neg logit contributions | |
deadliest | -3.423 |
losses | -3.254 |
occupants | -3.250 |
victims | -3.208 |
quotas | -3.141 |
Token | really |
Token ID | 1107 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
losses | +11.392 |
victims | +11.050 |
deadliest | +10.794 |
revenues | +10.744 |
quotas | +10.329 |
Neg logit contributions | |
cerning | -16.576 |
ude | -15.885 |
ingu | -15.384 |
cius | -15.072 |
asm | -14.846 |
Token | quickly |
Token ID | 2952 |
Feature activation | +8.23e-01 |
Pos logit contributions | |
ude | +1.152 |
cerning | +1.150 |
asm | +1.086 |
ingu | +1.074 |
clus | +1.047 |
Neg logit contributions | |
deadliest | -0.752 |
losses | -0.734 |
occupants | -0.725 |
victims | -0.717 |
quotas | -0.706 |
Token | like |
Token ID | 588 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
cerning | +0.831 |
ude | +0.827 |
asm | +0.774 |
ingu | +0.768 |
cius | +0.755 |
Neg logit contributions | |
deadliest | -0.699 |
losses | -0.687 |
victims | -0.675 |
occupants | -0.674 |
quotas | -0.660 |
Token | a |
Token ID | 257 |
Feature activation | +5.92e+00 |
Pos logit contributions | |
ude | +0.875 |
cerning | +0.875 |
asm | +0.820 |
ingu | +0.808 |
cius | +0.797 |
Neg logit contributions | |
deadliest | -0.728 |
losses | -0.712 |
occupants | -0.700 |
victims | -0.696 |
quotas | -0.683 |
Token | traditional |
Token ID | 4569 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
cerning | +6.259 |
ude | +6.249 |
asm | +5.887 |
ingu | +5.771 |
cius | +5.755 |
Neg logit contributions | |
deadliest | -4.456 |
occupants | -4.330 |
losses | -4.324 |
victims | -4.241 |
quotas | -4.221 |
Token | the |
Token ID | 262 |
Feature activation | -6.28e+00 |
Pos logit contributions | |
losses | +5.407 |
deadliest | +5.401 |
victims | +5.336 |
occupants | +5.225 |
quotas | +5.150 |
Neg logit contributions | |
cerning | -6.883 |
ude | -6.681 |
asm | -6.304 |
cius | -6.284 |
ingu | -6.194 |
Token | salt |
Token ID | 8268 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
deadliest | +4.858 |
losses | +4.835 |
victims | +4.758 |
occupants | +4.675 |
quotas | +4.552 |
Neg logit contributions | |
cerning | -6.789 |
ude | -6.694 |
asm | -6.284 |
ingu | -6.273 |
cius | -6.200 |
Token | trucks |
Token ID | 13960 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
losses | +4.959 |
deadliest | +4.890 |
victims | +4.842 |
occupants | +4.755 |
quotas | +4.726 |
Neg logit contributions | |
cerning | -5.089 |
ude | -4.950 |
asm | -4.618 |
clus | -4.577 |
ingu | -4.576 |
Token | because |
Token ID | 780 |
Feature activation | -7.97e+00 |
Pos logit contributions | |
ude | +1.438 |
cerning | +1.435 |
asm | +1.343 |
ingu | +1.336 |
cius | +1.284 |
Neg logit contributions | |
deadliest | -1.372 |
losses | -1.352 |
victims | -1.322 |
occupants | -1.312 |
quotas | -1.292 |
Token | the |
Token ID | 262 |
Feature activation | -9.81e+00 |
Pos logit contributions | |
losses | +6.688 |
deadliest | +6.665 |
victims | +6.593 |
occupants | +6.470 |
quotas | +6.359 |
Neg logit contributions | |
cerning | -7.979 |
ude | -7.822 |
ingu | -7.363 |
asm | -7.275 |
cius | -7.242 |
Token | grid |
Token ID | 10706 |
Feature activation | -1.40e+01 |
Pos logit contributions | |
losses | +7.500 |
deadliest | +7.458 |
victims | +7.360 |
occupants | +7.279 |
quotas | +7.074 |
Neg logit contributions | |
cerning | -10.447 |
ude | -10.295 |
ingu | -9.730 |
cius | -9.667 |
asm | -9.603 |
Token | lock |
Token ID | 5354 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
losses | +13.025 |
deadliest | +12.518 |
victims | +12.511 |
occupants | +12.375 |
only | +12.326 |
Neg logit contributions | |
cerning | -11.235 |
ude | -11.091 |
ingu | -10.382 |
asm | -10.032 |
cius | -10.027 |
Token | was |
Token ID | 373 |
Feature activation | -9.23e+00 |
Pos logit contributions | |
losses | +7.098 |
deadliest | +7.029 |
victims | +7.019 |
occupants | +6.745 |
quotas | +6.699 |
Neg logit contributions | |
cerning | -10.598 |
ude | -10.453 |
ingu | -9.864 |
asm | -9.839 |
cius | -9.770 |
Token | too |
Token ID | 1165 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
deadliest | +6.930 |
losses | +6.765 |
victims | +6.674 |
occupants | +6.453 |
quotas | +6.383 |
Neg logit contributions | |
cerning | -9.948 |
ude | -9.922 |
ingu | -9.306 |
asm | -9.262 |
cius | -9.127 |
Token | thick |
Token ID | 6546 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
deadliest | +0.560 |
losses | +0.556 |
victims | +0.542 |
occupants | +0.536 |
quotas | +0.525 |
Neg logit contributions | |
cerning | -1.002 |
ude | -0.995 |
asm | -0.942 |
ingu | -0.938 |
cius | -0.920 |
Token | to |
Token ID | 284 |
Feature activation | +6.64e+00 |
Pos logit contributions | |
cerning | +0.783 |
ude | +0.777 |
asm | +0.720 |
ingu | +0.717 |
cius | +0.700 |
Neg logit contributions | |
deadliest | -0.883 |
losses | -0.876 |
victims | -0.862 |
occupants | -0.855 |
quotas | -0.842 |
Token | Ung |
Token ID | 44904 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
cerning | +1.915 |
ude | +1.888 |
asm | +1.652 |
ingu | +1.628 |
cius | +1.564 |
Neg logit contributions | |
deadliest | -5.166 |
losses | -5.137 |
victims | -5.072 |
occupants | -5.043 |
quotas | -4.999 |
Token | ol |
Token ID | 349 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
deadliest | +0.801 |
losses | +0.795 |
victims | +0.786 |
occupants | +0.777 |
quotas | +0.765 |
Neg logit contributions | |
cerning | -0.320 |
ude | -0.310 |
ingu | -0.279 |
asm | -0.272 |
cius | -0.265 |
Token | . |
Token ID | 13 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
cerning | +3.303 |
ude | +3.283 |
asm | +3.047 |
ingu | +3.003 |
cius | +2.951 |
Neg logit contributions | |
deadliest | -3.874 |
losses | -3.850 |
victims | -3.786 |
occupants | -3.757 |
quotas | -3.735 |
Token | Fro |
Token ID | 9734 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
deadliest | +1.027 |
losses | +1.019 |
victims | +1.004 |
occupants | +0.994 |
quotas | +0.981 |
Neg logit contributions | |
cerning | -0.839 |
ude | -0.831 |
asm | -0.767 |
ingu | -0.764 |
cius | -0.746 |
Token | do |
Token ID | 4598 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
cerning | +0.190 |
ude | +0.185 |
asm | +0.140 |
ingu | +0.137 |
cius | +0.124 |
Neg logit contributions | |
deadliest | -1.145 |
losses | -1.139 |
victims | -1.128 |
occupants | -1.122 |
quotas | -1.112 |
Token | falls |
Token ID | 8953 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
deadliest | +0.417 |
losses | +0.410 |
victims | +0.402 |
occupants | +0.400 |
quotas | +0.394 |
Neg logit contributions | |
ude | -0.533 |
cerning | -0.532 |
asm | -0.499 |
ingu | -0.497 |
cius | -0.484 |
Token | to |
Token ID | 284 |
Feature activation | +6.07e+00 |
Pos logit contributions | |
victims | +10.550 |
losses | +10.175 |
deadliest | +9.913 |
occupants | +9.604 |
only | +9.520 |
Neg logit contributions | |
cerning | -14.333 |
ude | -14.131 |
cius | -13.844 |
ingu | -13.408 |
clus | -13.371 |
Token | Shel |
Token ID | 15325 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
ude | +6.038 |
cerning | +5.997 |
asm | +5.661 |
ingu | +5.611 |
cius | +5.394 |
Neg logit contributions | |
deadliest | -4.945 |
losses | -4.887 |
occupants | -4.851 |
quotas | -4.809 |
victims | -4.782 |
Token | ob |
Token ID | 672 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
cerning | +1.470 |
ude | +1.457 |
asm | +1.269 |
ingu | +1.228 |
cius | +1.177 |
Neg logit contributions | |
deadliest | -4.604 |
losses | -4.572 |
victims | -4.523 |
occupants | -4.505 |
quotas | -4.473 |
Token | 's |
Token ID | 338 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
cerning | +1.315 |
ude | +1.312 |
asm | +1.236 |
ingu | +1.227 |
cius | +1.206 |
Neg logit contributions | |
deadliest | -0.906 |
losses | -0.894 |
victims | -0.872 |
occupants | -0.870 |
quotas | -0.856 |
Token | sting |
Token ID | 24276 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
deadliest | +0.383 |
losses | +0.380 |
victims | +0.371 |
occupants | +0.367 |
quotas | +0.362 |
Neg logit contributions | |
cerning | -0.614 |
ude | -0.611 |
asm | -0.577 |
ingu | -0.574 |
cius | -0.564 |
Token | 22 |
Token ID | 2534 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
cerning | +6.307 |
ude | +6.297 |
asm | +6.001 |
ingu | +5.813 |
cius | +5.645 |
Neg logit contributions | |
deadliest | -5.053 |
occupants | -4.964 |
losses | -4.901 |
victims | -4.848 |
quotas | -4.772 |
Token | June |
Token ID | 2795 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
cerning | +1.796 |
ude | +1.793 |
asm | +1.650 |
ingu | +1.619 |
cius | +1.589 |
Neg logit contributions | |
deadliest | -2.123 |
losses | -2.100 |
occupants | -2.067 |
victims | -2.063 |
quotas | -2.021 |
Token | 1941 |
Token ID | 23234 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
deadliest | +2.655 |
losses | +2.648 |
victims | +2.600 |
occupants | +2.557 |
quotas | +2.525 |
Neg logit contributions | |
cerning | -2.874 |
ude | -2.855 |
ingu | -2.656 |
asm | -2.655 |
cius | -2.606 |
Token | , |
Token ID | 11 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
deadliest | +3.956 |
losses | +3.953 |
victims | +3.893 |
occupants | +3.826 |
quotas | +3.790 |
Neg logit contributions | |
cerning | -3.235 |
ude | -3.226 |
ingu | -2.964 |
asm | -2.946 |
cius | -2.890 |
Token | Germany |
Token ID | 4486 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
deadliest | +4.138 |
losses | +4.124 |
victims | +4.050 |
occupants | +3.998 |
quotas | +3.941 |
Neg logit contributions | |
cerning | -4.399 |
ude | -4.367 |
ingu | -4.065 |
asm | -4.060 |
cius | -3.977 |
Token | launched |
Token ID | 5611 |
Feature activation | -1.49e+01 |
Pos logit contributions | |
losses | +10.627 |
deadliest | +10.447 |
victims | +10.288 |
authorities | +9.988 |
occupants | +9.959 |
Neg logit contributions | |
cerning | -13.893 |
ude | -13.398 |
ingu | -12.774 |
cius | -12.688 |
asm | -12.666 |
Token | Operation |
Token ID | 14680 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
losses | +11.248 |
deadliest | +11.047 |
victims | +10.644 |
occupants | +10.427 |
quotas | +10.337 |
Neg logit contributions | |
cerning | -14.918 |
ude | -14.495 |
ingu | -13.980 |
cius | -13.687 |
clus | -13.647 |
Token | Barbar |
Token ID | 45952 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
deadliest | +3.018 |
losses | +3.009 |
victims | +2.956 |
occupants | +2.907 |
quotas | +2.902 |
Neg logit contributions | |
cerning | -2.622 |
ude | -2.599 |
ingu | -2.410 |
asm | -2.387 |
cius | -2.350 |
Token | oss |
Token ID | 793 |
Feature activation | +5.43e-01 |
Pos logit contributions | |
losses | +1.339 |
deadliest | +1.334 |
victims | +1.312 |
occupants | +1.297 |
quotas | +1.286 |
Neg logit contributions | |
cerning | -0.826 |
ude | -0.807 |
ingu | -0.738 |
asm | -0.738 |
cius | -0.721 |
Token | a |
Token ID | 64 |
Feature activation | +5.07e+00 |
Pos logit contributions | |
cerning | +0.399 |
ude | +0.393 |
ingu | +0.358 |
asm | +0.357 |
cius | +0.350 |
Neg logit contributions | |
deadliest | -0.614 |
losses | -0.610 |
victims | -0.601 |
occupants | -0.594 |
quotas | -0.588 |
Token | , |
Token ID | 11 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
cerning | +5.301 |
ude | +5.288 |
asm | +4.982 |
ingu | +4.939 |
cius | +4.791 |
Neg logit contributions | |
deadliest | -3.915 |
losses | -3.834 |
occupants | -3.801 |
victims | -3.768 |
quotas | -3.726 |
Token | ul |
Token ID | 377 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
cerning | +0.492 |
ude | +0.487 |
asm | +0.435 |
ingu | +0.431 |
cius | +0.416 |
Neg logit contributions | |
deadliest | -1.028 |
losses | -1.021 |
victims | -1.010 |
occupants | -1.001 |
quotas | -0.993 |
Token | ics |
Token ID | 873 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
deadliest | +0.622 |
losses | +0.621 |
victims | +0.612 |
occupants | +0.605 |
quotas | +0.596 |
Neg logit contributions | |
cerning | -0.338 |
ude | -0.331 |
ingu | -0.297 |
cius | -0.292 |
asm | -0.291 |
Token | issue |
Token ID | 2071 |
Feature activation | +4.44e+00 |
Pos logit contributions | |
ude | +2.452 |
cerning | +2.435 |
asm | +2.315 |
ingu | +2.271 |
clus | +2.220 |
Neg logit contributions | |
deadliest | -1.644 |
losses | -1.589 |
victims | -1.581 |
occupants | -1.561 |
quotas | -1.545 |
Token | with |
Token ID | 351 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
ude | +4.368 |
cerning | +4.338 |
asm | +4.050 |
ingu | +4.028 |
clus | +3.903 |
Neg logit contributions | |
deadliest | -3.744 |
losses | -3.668 |
occupants | -3.616 |
victims | -3.611 |
quotas | -3.575 |
Token | the |
Token ID | 262 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
deadliest | +1.581 |
losses | +1.565 |
victims | +1.532 |
occupants | +1.513 |
quotas | +1.484 |
Neg logit contributions | |
cerning | -2.344 |
ude | -2.328 |
asm | -2.193 |
ingu | -2.187 |
cius | -2.149 |
Token | rocket |
Token ID | 10701 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
deadliest | +4.332 |
losses | +4.303 |
victims | +4.226 |
occupants | +4.174 |
quotas | +4.072 |
Neg logit contributions | |
cerning | -5.070 |
ude | -5.006 |
ingu | -4.705 |
asm | -4.677 |
cius | -4.594 |
Token | � |
Token ID | 447 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
losses | +11.599 |
deadliest | +11.293 |
victims | +11.199 |
occupants | +11.184 |
closest | +10.760 |
Neg logit contributions | |
cerning | -12.055 |
ude | -11.648 |
ingu | -10.892 |
cius | -10.861 |
asm | -10.846 |
Token | � |
Token ID | 247 |
Feature activation | -5.12e+00 |
Pos logit contributions | |
cerning | +0.122 |
ude | +0.122 |
asm | +0.112 |
ingu | +0.111 |
cius | +0.108 |
Neg logit contributions | |
deadliest | -0.168 |
losses | -0.166 |
victims | -0.164 |
occupants | -0.163 |
quotas | -0.161 |
Token | s |
Token ID | 82 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
deadliest | +4.879 |
losses | +4.848 |
victims | +4.757 |
occupants | +4.701 |
quotas | +4.606 |
Neg logit contributions | |
cerning | -4.620 |
ude | -4.548 |
asm | -4.243 |
ingu | -4.242 |
cius | -4.186 |
Token | first |
Token ID | 717 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
deadliest | +3.389 |
losses | +3.375 |
victims | +3.292 |
occupants | +3.266 |
quotas | +3.168 |
Neg logit contributions | |
cerning | -4.663 |
ude | -4.596 |
asm | -4.328 |
ingu | -4.326 |
cius | -4.255 |
Token | stage |
Token ID | 3800 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
deadliest | +3.571 |
losses | +3.553 |
victims | +3.485 |
occupants | +3.461 |
quotas | +3.350 |
Neg logit contributions | |
cerning | -4.571 |
ude | -4.491 |
ingu | -4.234 |
asm | -4.209 |
cius | -4.147 |