Token | } |
Token ID | 92 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Soup | +3.316 |
esville | +2.926 |
Uniform | +2.906 |
efe | +2.886 |
eers | +2.881 |
Neg logit contributions | |
Fact | -5.333 |
itability | -5.212 |
amus | -5.058 |
weights | -4.998 |
campaigner | -4.808 |
Token | at |
Token ID | 379 |
Feature activation | +9.07e+00 |
Pos logit contributions | |
Soup | +1.757 |
esville | +1.562 |
Uniform | +1.519 |
eers | +1.491 |
icho | +1.481 |
Neg logit contributions | |
Fact | -2.567 |
itability | -2.500 |
amus | -2.430 |
weights | -2.426 |
quished | -2.315 |
Token | least |
Token ID | 1551 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
Soup | +4.160 |
esville | +3.666 |
Uniform | +3.536 |
eers | +3.449 |
Albion | +3.364 |
Neg logit contributions | |
Fact | -8.442 |
itability | -8.249 |
ccording | -8.107 |
weights | -8.020 |
amus | -7.945 |
Token | one |
Token ID | 530 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
Soup | +1.867 |
esville | +1.688 |
Uniform | +1.652 |
eers | +1.625 |
icho | +1.622 |
Neg logit contributions | |
Fact | -1.989 |
itability | -1.909 |
amus | -1.859 |
weights | -1.858 |
quished | -1.761 |
Token | blank |
Token ID | 9178 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
Soup | +0.480 |
esville | +0.439 |
Uniform | +0.422 |
eers | +0.420 |
icho | +0.416 |
Neg logit contributions | |
Fact | -0.528 |
itability | -0.509 |
weights | -0.495 |
amus | -0.494 |
quished | -0.469 |
Token | character |
Token ID | 2095 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
Soup | +3.128 |
esville | +2.781 |
Uniform | +2.755 |
eers | +2.658 |
icho | +2.657 |
Neg logit contributions | |
Fact | -5.013 |
itability | -4.916 |
weights | -4.845 |
amus | -4.781 |
kson | -4.514 |
Token | (< |
Token ID | 38155 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
Soup | +6.222 |
Uniform | +5.185 |
Arche | +5.093 |
esville | +4.982 |
icho | +4.935 |
Neg logit contributions | |
itability | -11.966 |
Fact | -11.959 |
amus | -11.781 |
weights | -11.490 |
kson | -10.878 |
Token | Space |
Token ID | 14106 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
Soup | +0.521 |
esville | +0.460 |
Uniform | +0.440 |
eers | +0.436 |
icho | +0.431 |
Neg logit contributions | |
Fact | -0.903 |
itability | -0.877 |
weights | -0.858 |
amus | -0.857 |
quished | -0.821 |
Token | > |
Token ID | 29 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
Soup | +2.680 |
esville | +2.363 |
Uniform | +2.287 |
eers | +2.249 |
icho | +2.217 |
Neg logit contributions | |
Fact | -4.219 |
itability | -4.152 |
weights | -4.032 |
amus | -4.017 |
ccording | -3.901 |
Token | or |
Token ID | 393 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
Fact | +1.838 |
itability | +1.765 |
weights | +1.721 |
amus | +1.715 |
quished | +1.629 |
Neg logit contributions | |
Soup | -1.766 |
esville | -1.624 |
Uniform | -1.551 |
eers | -1.549 |
icho | -1.538 |
Token | < |
Token ID | 1279 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Soup | +1.968 |
esville | +1.645 |
Uniform | +1.583 |
eers | +1.515 |
Albion | +1.502 |
Neg logit contributions | |
Fact | -4.936 |
itability | -4.822 |
weights | -4.714 |
amus | -4.710 |
quished | -4.547 |
Token | font |
Token ID | 10331 |
Feature activation | +8.64e+00 |
Pos logit contributions | |
Soup | +1.994 |
esville | +1.802 |
eers | +1.736 |
Uniform | +1.729 |
icho | +1.695 |
Neg logit contributions | |
Fact | -2.746 |
itability | -2.687 |
amus | -2.593 |
weights | -2.582 |
quished | -2.491 |
Token | 2 |
Token ID | 17 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Soup | +4.786 |
esville | +4.155 |
Uniform | +4.095 |
ingu | +3.909 |
efe | +3.894 |
Neg logit contributions | |
itability | -6.971 |
Fact | -6.959 |
amus | -6.744 |
campaigner | -6.533 |
weights | -6.529 |
Token | = |
Token ID | 796 |
Feature activation | +5.53e+00 |
Pos logit contributions | |
Fact | +0.919 |
itability | +0.871 |
weights | +0.864 |
amus | +0.858 |
anger | +0.813 |
Neg logit contributions | |
Soup | -0.665 |
esville | -0.605 |
Uniform | -0.577 |
icho | -0.576 |
eers | -0.574 |
Token | New |
Token ID | 968 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
Soup | +2.802 |
Uniform | +2.366 |
esville | +2.303 |
Arche | +2.199 |
Albion | +2.194 |
Neg logit contributions | |
Fact | -5.100 |
itability | -5.069 |
weights | -4.857 |
amus | -4.824 |
campaigner | -4.646 |
Token | - |
Token ID | 12 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
Soup | +2.365 |
esville | +2.127 |
Uniform | +2.057 |
eers | +2.040 |
Albion | +2.013 |
Neg logit contributions | |
Fact | -3.056 |
itability | -2.994 |
amus | -2.901 |
weights | -2.894 |
ccording | -2.792 |
Token | Object |
Token ID | 10267 |
Feature activation | +1.38e+01 |
Pos logit contributions | |
Soup | +2.403 |
esville | +2.162 |
Uniform | +2.092 |
eers | +2.074 |
Albion | +2.038 |
Neg logit contributions | |
itability | -3.223 |
Fact | -3.219 |
amus | -3.134 |
weights | -3.108 |
ccording | -3.040 |
Token | System |
Token ID | 4482 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
Soup | +5.647 |
Uniform | +5.110 |
Arche | +4.259 |
esville | +4.192 |
Albion | +4.102 |
Neg logit contributions | |
Fact | -12.981 |
itability | -12.938 |
amus | -12.463 |
weights | -12.436 |
ennes | -11.913 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Soup | +1.353 |
esville | +1.195 |
Uniform | +1.139 |
eers | +1.128 |
icho | +1.118 |
Neg logit contributions | |
Fact | -2.398 |
itability | -2.322 |
weights | -2.276 |
amus | -2.270 |
quished | -2.171 |
Token | Draw |
Token ID | 25302 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
Soup | +0.733 |
esville | +0.665 |
icho | +0.638 |
Uniform | +0.637 |
eers | +0.634 |
Neg logit contributions | |
Fact | -0.914 |
itability | -0.863 |
weights | -0.854 |
amus | -0.847 |
ysc | -0.805 |
Token | ing |
Token ID | 278 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
Soup | +1.566 |
esville | +1.354 |
Uniform | +1.250 |
eers | +1.245 |
Albion | +1.211 |
Neg logit contributions | |
Fact | -4.003 |
itability | -3.918 |
weights | -3.841 |
amus | -3.832 |
quished | -3.727 |
Token | . |
Token ID | 13 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
Soup | +2.441 |
esville | +2.133 |
Uniform | +2.022 |
eers | +2.005 |
Albion | +1.964 |
Neg logit contributions | |
Fact | -4.805 |
itability | -4.764 |
weights | -4.624 |
amus | -4.618 |
ccording | -4.550 |
Token | Super |
Token ID | 3115 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
Soup | +1.526 |
esville | +1.327 |
Uniform | +1.276 |
eers | +1.257 |
Albion | +1.254 |
Neg logit contributions | |
Fact | -2.635 |
itability | -2.566 |
weights | -2.514 |
amus | -2.509 |
quished | -2.410 |
Token | NES |
Token ID | 31925 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
Soup | +0.967 |
esville | +0.897 |
efe | +0.866 |
Uniform | +0.860 |
icho | +0.857 |
Neg logit contributions | |
Fact | -1.580 |
itability | -1.493 |
amus | -1.480 |
weights | -1.477 |
anger | -1.417 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
Soup | +0.008 |
esville | +0.008 |
icho | +0.007 |
eers | +0.007 |
Uniform | +0.007 |
Neg logit contributions | |
Fact | -0.013 |
itability | -0.012 |
weights | -0.012 |
amus | -0.012 |
quished | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
Fact | +1.272 |
itability | +1.228 |
weights | +1.197 |
amus | +1.195 |
anger | +1.147 |
Neg logit contributions | |
Soup | -1.123 |
esville | -1.042 |
icho | -0.999 |
efe | -0.988 |
Uniform | -0.983 |
Token | Game |
Token ID | 3776 |
Feature activation | +5.82e+00 |
Pos logit contributions | |
Soup | +1.096 |
esville | +1.043 |
eers | +0.980 |
Uniform | +0.980 |
icho | +0.977 |
Neg logit contributions | |
Fact | -1.143 |
itability | -1.092 |
amus | -1.075 |
weights | -1.067 |
anger | -1.018 |
Token | Boy |
Token ID | 6387 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
Soup | +3.425 |
esville | +3.040 |
Uniform | +2.963 |
eers | +2.895 |
Albion | +2.883 |
Neg logit contributions | |
Fact | -4.845 |
itability | -4.744 |
weights | -4.595 |
amus | -4.588 |
quished | -4.443 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
Soup | +4.879 |
Uniform | +4.105 |
Arche | +3.815 |
Enhance | +3.587 |
XIV | +3.585 |
Neg logit contributions | |
Fact | -13.256 |
itability | -12.846 |
amus | -12.663 |
weights | -12.505 |
oiler | -12.441 |
Token | Advance |
Token ID | 25170 |
Feature activation | +8.02e+00 |
Pos logit contributions | |
Soup | +1.474 |
esville | +1.195 |
Uniform | +1.104 |
eers | +1.088 |
icho | +1.062 |
Neg logit contributions | |
Fact | -5.013 |
itability | -4.887 |
weights | -4.804 |
amus | -4.801 |
quished | -4.632 |
Token | . |
Token ID | 13 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
Soup | +3.605 |
esville | +2.984 |
eers | +2.915 |
Uniform | +2.882 |
Albion | +2.820 |
Neg logit contributions | |
Fact | -7.797 |
itability | -7.521 |
amus | -7.496 |
weights | -7.370 |
ysc | -7.181 |
Token | We |
Token ID | 775 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
Soup | +1.831 |
esville | +1.707 |
Uniform | +1.654 |
eers | +1.652 |
icho | +1.645 |
Neg logit contributions | |
Fact | -1.280 |
itability | -1.208 |
weights | -1.178 |
amus | -1.169 |
quished | -1.090 |
Token | hope |
Token ID | 2911 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
Soup | +0.578 |
esville | +0.522 |
Uniform | +0.500 |
eers | +0.497 |
Albion | +0.494 |
Neg logit contributions | |
Fact | -0.769 |
itability | -0.744 |
weights | -0.727 |
amus | -0.724 |
quished | -0.692 |
Token | was |
Token ID | 373 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Fact | +0.206 |
itability | +0.198 |
weights | +0.193 |
amus | +0.193 |
quished | +0.183 |
Neg logit contributions | |
Soup | -0.181 |
esville | -0.165 |
Uniform | -0.159 |
eers | -0.158 |
icho | -0.157 |
Token | to |
Token ID | 284 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Fact | +1.232 |
itability | +1.182 |
amus | +1.155 |
weights | +1.148 |
quished | +1.082 |
Neg logit contributions | |
Soup | -1.263 |
esville | -1.154 |
eers | -1.114 |
Uniform | -1.113 |
icho | -1.100 |
Token | explain |
Token ID | 4727 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
Soup | +1.061 |
esville | +0.974 |
Uniform | +0.944 |
eers | +0.938 |
icho | +0.930 |
Neg logit contributions | |
Fact | -1.019 |
itability | -0.977 |
weights | -0.950 |
amus | -0.948 |
quished | -0.899 |
Token | away |
Token ID | 1497 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
Soup | +1.182 |
esville | +1.081 |
Uniform | +1.052 |
eers | +1.041 |
Albion | +1.038 |
Neg logit contributions | |
Fact | -1.190 |
itability | -1.142 |
weights | -1.118 |
amus | -1.110 |
quished | -1.058 |
Token | this |
Token ID | 428 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Soup | +2.258 |
esville | +2.068 |
Uniform | +2.020 |
Albion | +1.997 |
Arche | +1.984 |
Neg logit contributions | |
Fact | -2.244 |
itability | -2.173 |
weights | -2.110 |
amus | -2.109 |
quished | -2.010 |
Token | stain |
Token ID | 18746 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
Soup | +0.741 |
esville | +0.679 |
Uniform | +0.655 |
eers | +0.653 |
icho | +0.648 |
Neg logit contributions | |
Fact | -0.754 |
itability | -0.725 |
weights | -0.704 |
amus | -0.703 |
quished | -0.663 |
Token | of |
Token ID | 286 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
Soup | +6.745 |
esville | +6.255 |
Albion | +6.064 |
Arche | +6.047 |
ingu | +5.952 |
Neg logit contributions | |
Fact | -10.339 |
itability | -10.162 |
weights | -9.891 |
amus | -9.625 |
quished | -9.574 |
Token | irrational |
Token ID | 25086 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
Soup | +2.702 |
esville | +2.435 |
Uniform | +2.420 |
Albion | +2.386 |
Arche | +2.372 |
Neg logit contributions | |
Fact | -2.792 |
itability | -2.692 |
weights | -2.637 |
amus | -2.620 |
quished | -2.512 |
Token | ity |
Token ID | 414 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
Fact | +1.758 |
itability | +1.710 |
amus | +1.653 |
weights | +1.649 |
anger | +1.560 |
Neg logit contributions | |
Soup | -1.763 |
esville | -1.615 |
Uniform | -1.555 |
eers | -1.552 |
icho | -1.545 |
Token | ( |
Token ID | 357 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
Soup | +1.443 |
esville | +1.306 |
Uniform | +1.251 |
eers | +1.241 |
Albion | +1.239 |
Neg logit contributions | |
Fact | -1.890 |
itability | -1.826 |
weights | -1.789 |
amus | -1.776 |
quished | -1.703 |
Token | Det |
Token ID | 11242 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
Soup | +0.654 |
esville | +0.596 |
Uniform | +0.576 |
eers | +0.571 |
icho | +0.567 |
Neg logit contributions | |
Fact | -0.738 |
itability | -0.715 |
weights | -0.695 |
amus | -0.694 |
quished | -0.664 |
Token | - |
Token ID | 12 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
Fact | +0.311 |
itability | +0.297 |
weights | +0.293 |
amus | +0.289 |
anger | +0.274 |
Neg logit contributions | |
Soup | -0.275 |
esville | -0.251 |
Uniform | -0.241 |
eers | -0.240 |
icho | -0.239 |
Token | sl |
Token ID | 6649 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
Soup | +1.485 |
esville | +1.379 |
eers | +1.331 |
Uniform | +1.323 |
icho | +1.316 |
Neg logit contributions | |
Fact | -1.267 |
itability | -1.216 |
amus | -1.180 |
weights | -1.164 |
quished | -1.111 |
Token | it |
Token ID | 270 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Soup | +1.875 |
esville | +1.721 |
icho | +1.665 |
Uniform | +1.656 |
eers | +1.651 |
Neg logit contributions | |
Fact | -1.708 |
itability | -1.626 |
weights | -1.600 |
amus | -1.586 |
quished | -1.514 |
Token | experiment |
Token ID | 6306 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
Soup | +1.400 |
esville | +1.301 |
Uniform | +1.266 |
eers | +1.260 |
icho | +1.252 |
Neg logit contributions | |
Fact | -0.946 |
itability | -0.901 |
weights | -0.871 |
amus | -0.867 |
quished | -0.809 |
Token | with |
Token ID | 351 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
Soup | +2.523 |
esville | +2.245 |
Uniform | +2.152 |
eers | +2.146 |
Albion | +2.111 |
Neg logit contributions | |
Fact | -3.965 |
itability | -3.851 |
weights | -3.753 |
amus | -3.746 |
ccording | -3.653 |
Token | Cloud |
Token ID | 10130 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
Soup | +1.049 |
esville | +0.906 |
Uniform | +0.855 |
eers | +0.846 |
icho | +0.834 |
Neg logit contributions | |
Fact | -2.341 |
itability | -2.276 |
weights | -2.232 |
amus | -2.227 |
quished | -2.143 |
Token | � |
Token ID | 447 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
Soup | +6.220 |
Uniform | +5.293 |
esville | +5.291 |
Albion | +5.200 |
Arche | +5.198 |
Neg logit contributions | |
itability | -11.398 |
Fact | -11.301 |
ccording | -11.240 |
weights | -10.916 |
amus | -10.902 |
Token | � |
Token ID | 247 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Soup | +1.446 |
esville | +1.299 |
Uniform | +1.246 |
eers | +1.235 |
icho | +1.224 |
Neg logit contributions | |
Fact | -2.071 |
itability | -2.002 |
weights | -1.958 |
amus | -1.952 |
quished | -1.865 |
Token | s |
Token ID | 82 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
Fact | +1.305 |
itability | +1.245 |
weights | +1.242 |
amus | +1.225 |
anger | +1.178 |
Neg logit contributions | |
Soup | -0.944 |
esville | -0.841 |
Uniform | -0.823 |
icho | -0.810 |
Albion | -0.805 |
Token | old |
Token ID | 1468 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
Soup | +0.593 |
esville | +0.533 |
Uniform | +0.510 |
eers | +0.508 |
icho | +0.508 |
Neg logit contributions | |
Fact | -0.803 |
itability | -0.774 |
weights | -0.761 |
amus | -0.757 |
campaigner | -0.720 |
Token | Match |
Token ID | 13225 |
Feature activation | +5.82e+00 |
Pos logit contributions | |
Soup | +2.158 |
esville | +1.941 |
Uniform | +1.896 |
eers | +1.864 |
Albion | +1.856 |
Neg logit contributions | |
Fact | -2.496 |
itability | -2.415 |
weights | -2.346 |
amus | -2.339 |
quished | -2.238 |
Token | issue |
Token ID | 2071 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
Soup | +0.884 |
esville | +0.803 |
Uniform | +0.780 |
eers | +0.773 |
Albion | +0.765 |
Neg logit contributions | |
Fact | -0.963 |
itability | -0.927 |
weights | -0.903 |
amus | -0.900 |
quished | -0.855 |
Token | of |
Token ID | 286 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
Soup | +2.341 |
esville | +2.067 |
Uniform | +2.046 |
eers | +2.026 |
Albion | +1.999 |
Neg logit contributions | |
Fact | -2.627 |
itability | -2.537 |
weights | -2.466 |
amus | -2.446 |
quished | -2.330 |
Token | Frank |
Token ID | 5278 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
Soup | +1.564 |
esville | +1.392 |
Uniform | +1.373 |
eers | +1.341 |
Albion | +1.340 |
Neg logit contributions | |
Fact | -1.900 |
itability | -1.836 |
amus | -1.785 |
weights | -1.784 |
quished | -1.701 |
Token | Miller |
Token ID | 7920 |
Feature activation | +4.21e+00 |
Pos logit contributions | |
Soup | +2.185 |
esville | +1.949 |
Uniform | +1.936 |
icho | +1.882 |
eers | +1.878 |
Neg logit contributions | |
Fact | -2.317 |
itability | -2.229 |
amus | -2.140 |
weights | -2.130 |
quished | -2.058 |
Token | 's |
Token ID | 338 |
Feature activation | +8.73e+00 |
Pos logit contributions | |
Soup | +2.309 |
esville | +1.992 |
Uniform | +1.967 |
eers | +1.925 |
Albion | +1.897 |
Neg logit contributions | |
Fact | -3.676 |
itability | -3.594 |
weights | -3.507 |
amus | -3.478 |
campaigner | -3.341 |
Token | comic |
Token ID | 9048 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
Soup | +5.734 |
Uniform | +5.097 |
Albion | +4.941 |
esville | +4.938 |
Legion | +4.882 |
Neg logit contributions | |
Fact | -6.499 |
itability | -6.133 |
weights | -6.082 |
amus | -6.058 |
quished | -5.764 |
Token | mini |
Token ID | 9927 |
Feature activation | +5.70e+00 |
Pos logit contributions | |
Soup | +8.463 |
Uniform | +7.450 |
eers | +7.317 |
Legion | +7.212 |
Albion | +7.193 |
Neg logit contributions | |
Fact | -9.106 |
itability | -8.912 |
amus | -8.600 |
weights | -8.494 |
ccording | -8.231 |
Token | - |
Token ID | 12 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
Soup | +3.441 |
esville | +3.031 |
eers | +2.985 |
Uniform | +2.975 |
icho | +2.947 |
Neg logit contributions | |
Fact | -4.577 |
itability | -4.466 |
amus | -4.354 |
weights | -4.303 |
ysc | -4.150 |
Token | series |
Token ID | 25076 |
Feature activation | +6.79e+00 |
Pos logit contributions | |
Soup | +3.885 |
esville | +3.606 |
Uniform | +3.519 |
eers | +3.514 |
icho | +3.496 |
Neg logit contributions | |
Fact | -2.243 |
itability | -2.142 |
amus | -2.060 |
weights | -2.024 |
campaigner | -1.924 |
Token | Ron |
Token ID | 6575 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Soup | +4.113 |
Uniform | +3.610 |
esville | +3.568 |
eers | +3.499 |
Albion | +3.482 |
Neg logit contributions | |
Fact | -5.491 |
itability | -5.349 |
amus | -5.185 |
weights | -5.177 |
campaigner | -4.975 |
Token | in |
Token ID | 259 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
Fact | +1.419 |
itability | +1.376 |
amus | +1.351 |
weights | +1.347 |
anger | +1.294 |
Neg logit contributions | |
Soup | -0.717 |
esville | -0.630 |
eers | -0.602 |
Uniform | -0.594 |
Albion | -0.584 |
Token | Grove |
Token ID | 24144 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
Soup | +2.286 |
esville | +2.059 |
Uniform | +1.887 |
eers | +1.870 |
Albion | +1.860 |
Neg logit contributions | |
Fact | -4.654 |
itability | -4.509 |
weights | -4.404 |
amus | -4.369 |
quished | -4.256 |
Token | , |
Token ID | 11 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Soup | +2.275 |
esville | +2.036 |
Uniform | +1.910 |
icho | +1.875 |
Albion | +1.867 |
Neg logit contributions | |
Fact | -3.894 |
itability | -3.798 |
weights | -3.707 |
amus | -3.674 |
ysc | -3.536 |
Token | California |
Token ID | 3442 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
Soup | +0.862 |
esville | +0.770 |
Uniform | +0.744 |
eers | +0.735 |
Albion | +0.732 |
Neg logit contributions | |
Fact | -1.238 |
itability | -1.205 |
weights | -1.173 |
amus | -1.172 |
quished | -1.125 |
Token | . |
Token ID | 13 |
Feature activation | +9.40e-01 |
Pos logit contributions | |
Fact | +1.575 |
itability | +1.524 |
amus | +1.493 |
weights | +1.491 |
quished | +1.429 |
Neg logit contributions | |
Soup | -0.945 |
esville | -0.838 |
Uniform | -0.804 |
eers | -0.796 |
icho | -0.789 |
Token | Hide |
Token ID | 10415 |
Feature activation | +7.89e+00 |
Pos logit contributions | |
Soup | +0.787 |
esville | +0.724 |
Uniform | +0.713 |
eers | +0.701 |
Albion | +0.697 |
Neg logit contributions | |
Fact | -0.556 |
itability | -0.539 |
weights | -0.519 |
amus | -0.518 |
quished | -0.484 |
Token | Caption |
Token ID | 11260 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
Soup | +2.024 |
esville | +1.416 |
Uniform | +1.414 |
Albion | +1.239 |
Arche | +1.222 |
Neg logit contributions | |
Fact | -9.075 |
itability | -9.057 |
ccording | -8.990 |
weights | -8.871 |
amus | -8.759 |
Token | 15 |
Token ID | 1315 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
Soup | +6.003 |
esville | +5.141 |
Arche | +5.093 |
Uniform | +4.970 |
XIV | +4.833 |
Neg logit contributions | |
ccording | -12.621 |
itability | -11.848 |
ategory | -11.527 |
Fact | -11.465 |
quished | -11.252 |
Token | of |
Token ID | 286 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Soup | +0.030 |
esville | +0.028 |
Uniform | +0.027 |
eers | +0.027 |
Albion | +0.027 |
Neg logit contributions | |
Fact | -0.027 |
itability | -0.026 |
weights | -0.025 |
amus | -0.025 |
anger | -0.024 |
Token | 15 |
Token ID | 1315 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
Soup | +0.937 |
esville | +0.852 |
Uniform | +0.817 |
eers | +0.815 |
icho | +0.809 |
Neg logit contributions | |
Fact | -1.141 |
itability | -1.097 |
weights | -1.074 |
amus | -1.069 |
ysc | -1.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
Soup | +1.907 |
esville | +1.739 |
Uniform | +1.683 |
eers | +1.669 |
icho | +1.655 |
Neg logit contributions | |
Fact | -2.025 |
itability | -1.955 |
weights | -1.901 |
amus | -1.896 |
quished | -1.800 |
Token |
|
Token ID | 198 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
Soup | +1.128 |
esville | +1.032 |
Uniform | +0.997 |
eers | +0.992 |
icho | +0.986 |
Neg logit contributions | |
Fact | -1.148 |
itability | -1.096 |
weights | -1.070 |
amus | -1.067 |
quished | -1.005 |
Token | / |
Token ID | 14 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
Fact | +1.156 |
itability | +1.098 |
weights | +1.078 |
amus | +1.076 |
ysc | +1.018 |
Neg logit contributions | |
Soup | -1.022 |
esville | -0.933 |
Uniform | -0.897 |
icho | -0.896 |
eers | -0.896 |
Token | 3 |
Token ID | 18 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
Fact | +1.150 |
itability | +1.077 |
weights | +1.058 |
amus | +1.046 |
anger | +0.989 |
Neg logit contributions | |
Soup | -1.307 |
esville | -1.209 |
Uniform | -1.181 |
Albion | -1.166 |
eers | -1.165 |
Token | minutes |
Token ID | 2431 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
Soup | +2.231 |
esville | +2.039 |
Uniform | +1.984 |
eers | +1.966 |
icho | +1.955 |
Neg logit contributions | |
Fact | -2.019 |
itability | -1.938 |
weights | -1.880 |
amus | -1.874 |
quished | -1.778 |
Token | until |
Token ID | 1566 |
Feature activation | +6.02e+00 |
Pos logit contributions | |
Soup | +1.070 |
esville | +0.944 |
Uniform | +0.889 |
eers | +0.884 |
icho | +0.877 |
Neg logit contributions | |
Fact | -2.102 |
itability | -2.044 |
weights | -2.001 |
amus | -2.001 |
ysc | -1.917 |
Token | golden |
Token ID | 10861 |
Feature activation | +6.93e+00 |
Pos logit contributions | |
Soup | +3.018 |
esville | +2.514 |
Uniform | +2.465 |
icho | +2.425 |
eers | +2.418 |
Neg logit contributions | |
Fact | -5.613 |
itability | -5.414 |
amus | -5.310 |
ccording | -5.224 |
weights | -5.221 |
Token | brown |
Token ID | 7586 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
Soup | +4.035 |
esville | +3.545 |
Uniform | +3.507 |
Albion | +3.420 |
eers | +3.410 |
Neg logit contributions | |
Fact | -5.779 |
itability | -5.571 |
ccording | -5.398 |
amus | -5.372 |
weights | -5.335 |
Token | . |
Token ID | 13 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
Soup | +4.295 |
esville | +3.598 |
eers | +3.458 |
Uniform | +3.169 |
Arche | +3.142 |
Neg logit contributions | |
ccording | -13.842 |
Fact | -12.788 |
itability | -12.604 |
quished | -12.418 |
amus | -12.253 |
Token | These |
Token ID | 2312 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
Soup | +2.531 |
esville | +2.267 |
Uniform | +2.255 |
eers | +2.197 |
Albion | +2.194 |
Neg logit contributions | |
Fact | -2.340 |
itability | -2.328 |
amus | -2.221 |
weights | -2.207 |
quished | -2.091 |
Token | are |
Token ID | 389 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
Soup | +0.247 |
esville | +0.227 |
Uniform | +0.220 |
eers | +0.219 |
icho | +0.217 |
Neg logit contributions | |
Fact | -0.229 |
itability | -0.219 |
weights | -0.214 |
amus | -0.213 |
ysc | -0.200 |
Token | lightly |
Token ID | 15376 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
Soup | +2.832 |
esville | +2.550 |
Uniform | +2.453 |
icho | +2.453 |
Albion | +2.429 |
Neg logit contributions | |
Fact | -3.231 |
itability | -3.141 |
weights | -3.019 |
amus | -3.012 |
ccording | -2.877 |
Token | fried |
Token ID | 23018 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
Fact | +2.141 |
itability | +2.043 |
weights | +2.040 |
amus | +1.960 |
quished | +1.915 |
Neg logit contributions | |
Soup | -2.195 |
esville | -2.055 |
Uniform | -1.993 |
eers | -1.960 |
Albion | -1.945 |
Token | figure |
Token ID | 3785 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Soup | +3.586 |
esville | +3.267 |
Uniform | +3.223 |
eers | +3.160 |
Arche | +3.131 |
Neg logit contributions | |
Fact | -3.813 |
itability | -3.591 |
weights | -3.584 |
amus | -3.526 |
ysc | -3.421 |
Token | in |
Token ID | 287 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
Soup | +1.211 |
esville | +1.094 |
Uniform | +1.049 |
eers | +1.045 |
icho | +1.032 |
Neg logit contributions | |
Fact | -1.619 |
itability | -1.564 |
weights | -1.525 |
amus | -1.522 |
quished | -1.455 |
Token | DC |
Token ID | 6257 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
Fact | +0.832 |
itability | +0.806 |
weights | +0.786 |
amus | +0.781 |
anger | +0.742 |
Neg logit contributions | |
Soup | -0.683 |
esville | -0.622 |
Uniform | -0.594 |
eers | -0.591 |
icho | -0.588 |
Token | Entertainment |
Token ID | 11058 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
Soup | +1.217 |
esville | +1.051 |
Uniform | +1.018 |
eers | +0.994 |
icho | +0.986 |
Neg logit contributions | |
Fact | -2.346 |
itability | -2.278 |
weights | -2.231 |
amus | -2.227 |
quished | -2.142 |
Token | 's |
Token ID | 338 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
Soup | +2.077 |
esville | +1.846 |
Uniform | +1.787 |
eers | +1.768 |
Albion | +1.734 |
Neg logit contributions | |
Fact | -3.166 |
itability | -3.074 |
weights | -3.000 |
amus | -2.991 |
quished | -2.883 |
Token | comic |
Token ID | 9048 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
Soup | +2.615 |
esville | +2.347 |
Uniform | +2.328 |
icho | +2.268 |
Albion | +2.261 |
Neg logit contributions | |
Fact | -2.818 |
itability | -2.673 |
weights | -2.629 |
amus | -2.625 |
quished | -2.510 |
Token | book |
Token ID | 1492 |
Feature activation | +8.96e+00 |
Pos logit contributions | |
Soup | +7.266 |
eers | +6.454 |
Uniform | +6.209 |
esville | +6.197 |
icho | +6.115 |
Neg logit contributions | |
Fact | -9.780 |
itability | -9.300 |
amus | -9.116 |
ccording | -8.816 |
weights | -8.754 |
Token | mythology |
Token ID | 25871 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Soup | +5.932 |
eers | +5.262 |
Uniform | +5.246 |
esville | +5.210 |
Legion | +5.156 |
Neg logit contributions | |
Fact | -6.669 |
itability | -6.252 |
amus | -6.108 |
weights | -6.056 |
ccording | -5.909 |
Token | . |
Token ID | 13 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
Soup | +0.587 |
esville | +0.521 |
Uniform | +0.498 |
eers | +0.494 |
icho | +0.489 |
Neg logit contributions | |
Fact | -0.964 |
itability | -0.934 |
weights | -0.914 |
amus | -0.912 |
quished | -0.873 |
Token | Part |
Token ID | 2142 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
Fact | +0.573 |
itability | +0.542 |
weights | +0.530 |
amus | +0.528 |
quished | +0.495 |
Neg logit contributions | |
Soup | -0.664 |
esville | -0.618 |
eers | -0.595 |
icho | -0.592 |
Uniform | -0.592 |
Token | of |
Token ID | 286 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Soup | +0.293 |
esville | +0.268 |
Uniform | +0.259 |
eers | +0.258 |
icho | +0.256 |
Neg logit contributions | |
Fact | -0.298 |
itability | -0.286 |
weights | -0.279 |
amus | -0.278 |
quished | -0.263 |
Token | Pasadena |
Token ID | 41010 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
Soup | +1.201 |
esville | +1.077 |
Uniform | +1.044 |
Albion | +1.032 |
eers | +1.022 |
Neg logit contributions | |
Fact | -1.615 |
itability | -1.568 |
weights | -1.526 |
amus | -1.520 |
quished | -1.465 |
Token | , |
Token ID | 11 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
Soup | +0.293 |
esville | +0.262 |
Uniform | +0.251 |
eers | +0.249 |
icho | +0.246 |
Neg logit contributions | |
Fact | -0.450 |
itability | -0.435 |
weights | -0.426 |
amus | -0.425 |
quished | -0.406 |
Token | California |
Token ID | 3442 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Soup | +0.259 |
esville | +0.233 |
Uniform | +0.224 |
eers | +0.223 |
icho | +0.220 |
Neg logit contributions | |
Fact | -0.354 |
itability | -0.342 |
weights | -0.334 |
amus | -0.333 |
quished | -0.318 |
Token | . |
Token ID | 13 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
Fact | +1.224 |
itability | +1.183 |
weights | +1.159 |
amus | +1.159 |
quished | +1.111 |
Neg logit contributions | |
Soup | -0.728 |
esville | -0.646 |
Uniform | -0.617 |
eers | -0.611 |
icho | -0.606 |
Token | Hide |
Token ID | 10415 |
Feature activation | +8.32e+00 |
Pos logit contributions | |
Soup | +0.430 |
esville | +0.396 |
Uniform | +0.388 |
eers | +0.383 |
icho | +0.381 |
Neg logit contributions | |
Fact | -0.308 |
itability | -0.297 |
weights | -0.286 |
amus | -0.286 |
quished | -0.267 |
Token | Caption |
Token ID | 11260 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
Soup | +2.049 |
Uniform | +1.407 |
esville | +1.403 |
Albion | +1.217 |
Arche | +1.199 |
Neg logit contributions | |
Fact | -9.629 |
itability | -9.612 |
ccording | -9.593 |
weights | -9.420 |
amus | -9.295 |
Token | 11 |
Token ID | 1367 |
Feature activation | -3.76e+00 |
Pos logit contributions | |
Soup | +5.656 |
Arche | +4.809 |
Uniform | +4.690 |
XIV | +4.617 |
esville | +4.469 |
Neg logit contributions | |
ccording | -12.448 |
itability | -11.362 |
amus | -10.981 |
ategory | -10.956 |
quished | -10.932 |
Token | of |
Token ID | 286 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Fact | +2.283 |
weights | +2.087 |
amus | +2.072 |
itability | +2.055 |
anger | +1.940 |
Neg logit contributions | |
Soup | -3.073 |
esville | -2.823 |
icho | -2.764 |
Uniform | -2.750 |
ingu | -2.746 |
Token | 15 |
Token ID | 1315 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
Soup | +0.724 |
esville | +0.660 |
Uniform | +0.633 |
eers | +0.630 |
icho | +0.627 |
Neg logit contributions | |
Fact | -0.873 |
itability | -0.838 |
weights | -0.820 |
amus | -0.818 |
anger | -0.773 |
Token | Photos |
Token ID | 9434 |
Feature activation | +7.75e+00 |
Pos logit contributions | |
Soup | +1.238 |
esville | +1.099 |
Uniform | +1.056 |
eers | +1.043 |
icho | +1.029 |
Neg logit contributions | |
Fact | -1.952 |
itability | -1.903 |
weights | -1.853 |
amus | -1.851 |
quished | -1.778 |
Token | : |
Token ID | 25 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
Soup | +3.644 |
esville | +3.208 |
eers | +3.071 |
Uniform | +3.064 |
Arche | +2.989 |
Neg logit contributions | |
ccording | -7.365 |
itability | -7.104 |
Fact | -7.046 |
quished | -6.818 |
amus | -6.795 |
Token | Irvine |
Token ID | 37277 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
Soup | +0.400 |
esville | +0.357 |
Uniform | +0.344 |
eers | +0.340 |
Albion | +0.337 |
Neg logit contributions | |
Fact | -0.613 |
itability | -0.594 |
weights | -0.582 |
amus | -0.578 |
quished | -0.554 |
Token | , |
Token ID | 11 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
Fact | +0.450 |
itability | +0.434 |
amus | +0.426 |
weights | +0.424 |
quished | +0.407 |
Neg logit contributions | |
Soup | -0.305 |
esville | -0.273 |
Uniform | -0.262 |
eers | -0.261 |
icho | -0.260 |
Token | California |
Token ID | 3442 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Soup | +1.293 |
esville | +1.159 |
Uniform | +1.122 |
eers | +1.114 |
Albion | +1.104 |
Neg logit contributions | |
Fact | -1.918 |
itability | -1.867 |
weights | -1.819 |
amus | -1.818 |
ysc | -1.733 |
Token | . |
Token ID | 13 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
Fact | +1.139 |
itability | +1.103 |
weights | +1.080 |
amus | +1.080 |
quished | +1.037 |
Neg logit contributions | |
Soup | -0.667 |
esville | -0.593 |
Uniform | -0.564 |
eers | -0.560 |
icho | -0.553 |
Token | Hide |
Token ID | 10415 |
Feature activation | +8.66e+00 |
Pos logit contributions | |
Soup | +0.208 |
esville | +0.191 |
Uniform | +0.187 |
eers | +0.185 |
icho | +0.184 |
Neg logit contributions | |
Fact | -0.152 |
itability | -0.147 |
weights | -0.142 |
amus | -0.141 |
quished | -0.132 |
Token | Caption |
Token ID | 11260 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
Soup | +2.137 |
Uniform | +1.465 |
esville | +1.463 |
Albion | +1.267 |
Arche | +1.249 |
Neg logit contributions | |
ccording | -10.015 |
Fact | -9.976 |
itability | -9.976 |
weights | -9.768 |
amus | -9.631 |
Token | 10 |
Token ID | 838 |
Feature activation | -6.37e+00 |
Pos logit contributions | |
Soup | +5.458 |
Uniform | +4.329 |
Arche | +4.323 |
esville | +4.134 |
XIV | +4.106 |
Neg logit contributions | |
ccording | -12.676 |
itability | -11.522 |
quished | -11.398 |
ategory | -11.385 |
amus | -11.282 |
Token | of |
Token ID | 286 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
Fact | +3.603 |
weights | +3.229 |
amus | +3.218 |
itability | +3.135 |
anger | +3.002 |
Neg logit contributions | |
Soup | -5.370 |
esville | -4.934 |
ingu | -4.898 |
icho | -4.863 |
Uniform | -4.826 |
Token | 15 |
Token ID | 1315 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
Soup | +0.537 |
esville | +0.489 |
Uniform | +0.470 |
eers | +0.467 |
icho | +0.465 |
Neg logit contributions | |
Fact | -0.643 |
itability | -0.618 |
weights | -0.604 |
amus | -0.603 |
anger | -0.570 |
Token | Photos |
Token ID | 9434 |
Feature activation | +7.70e+00 |
Pos logit contributions | |
Soup | +1.284 |
esville | +1.140 |
Uniform | +1.096 |
eers | +1.081 |
icho | +1.067 |
Neg logit contributions | |
Fact | -2.023 |
itability | -1.972 |
weights | -1.921 |
amus | -1.918 |
quished | -1.842 |
Token | : |
Token ID | 25 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
Soup | +3.622 |
esville | +3.187 |
eers | +3.050 |
Uniform | +3.043 |
Arche | +2.969 |
Neg logit contributions | |
ccording | -7.327 |
itability | -7.059 |
Fact | -6.999 |
quished | -6.779 |
amus | -6.757 |
Token | 2 |
Token ID | 17 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
Soup | +2.298 |
esville | +2.083 |
Uniform | +2.019 |
eers | +1.989 |
Albion | +1.983 |
Neg logit contributions | |
Fact | -2.602 |
itability | -2.547 |
weights | -2.464 |
amus | -2.461 |
ccording | -2.396 |
Token | . |
Token ID | 764 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
Soup | +1.696 |
esville | +1.507 |
Uniform | +1.441 |
eers | +1.425 |
Albion | +1.408 |
Neg logit contributions | |
Fact | -2.749 |
itability | -2.676 |
weights | -2.607 |
amus | -2.605 |
quished | -2.520 |
Token | im |
Token ID | 545 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
Soup | +0.743 |
esville | +0.684 |
Uniform | +0.661 |
eers | +0.659 |
icho | +0.653 |
Neg logit contributions | |
Fact | -0.682 |
itability | -0.655 |
weights | -0.637 |
amus | -0.635 |
quished | -0.598 |
Token | read |
Token ID | 961 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
Fact | +0.163 |
itability | +0.155 |
weights | +0.149 |
amus | +0.147 |
ysc | +0.136 |
Neg logit contributions | |
Soup | -0.340 |
esville | -0.318 |
Uniform | -0.311 |
eers | -0.310 |
Albion | -0.309 |
Token | ( |
Token ID | 357 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
Soup | +2.623 |
esville | +2.353 |
Uniform | +2.285 |
Albion | +2.226 |
eers | +2.226 |
Neg logit contributions | |
Fact | -3.569 |
itability | -3.493 |
amus | -3.433 |
ccording | -3.402 |
weights | -3.377 |
Token | image |
Token ID | 2939 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
Soup | +1.237 |
esville | +1.130 |
Uniform | +1.092 |
eers | +1.086 |
icho | +1.077 |
Neg logit contributions | |
Fact | -1.308 |
itability | -1.259 |
weights | -1.226 |
amus | -1.222 |
quished | -1.157 |
Token | Path |
Token ID | 15235 |
Feature activation | +9.50e+00 |
Pos logit contributions | |
Soup | +3.951 |
esville | +2.904 |
Uniform | +2.854 |
Arche | +2.636 |
Albion | +2.612 |
Neg logit contributions | |
ccording | -13.292 |
itability | -12.711 |
quished | -12.590 |
weights | -12.450 |
amus | -12.348 |
Token | ) |
Token ID | 1267 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
Soup | +4.091 |
esville | +3.440 |
Uniform | +3.320 |
Albion | +3.256 |
eers | +3.130 |
Neg logit contributions | |
ccording | -9.549 |
itability | -9.081 |
Fact | -8.948 |
quished | -8.825 |
amus | -8.708 |
Token | image |
Token ID | 2939 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
Soup | +0.526 |
esville | +0.483 |
eers | +0.466 |
Uniform | +0.464 |
icho | +0.460 |
Neg logit contributions | |
Fact | -0.534 |
itability | -0.513 |
weights | -0.500 |
amus | -0.499 |
ysc | -0.471 |
Token | = |
Token ID | 796 |
Feature activation | +6.63e+00 |
Pos logit contributions | |
Soup | +4.797 |
esville | +3.796 |
Uniform | +3.753 |
Arche | +3.553 |
Albion | +3.542 |
Neg logit contributions | |
ccording | -12.053 |
itability | -11.229 |
quished | -11.006 |
Fact | -10.992 |
amus | -10.948 |
Token | im |
Token ID | 545 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
Soup | +4.420 |
Uniform | +3.931 |
esville | +3.915 |
icho | +3.822 |
Albion | +3.793 |
Neg logit contributions | |
Fact | -4.952 |
itability | -4.772 |
weights | -4.656 |
amus | -4.635 |
ccording | -4.629 |
Token | aga |
Token ID | 8126 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Soup | +0.095 |
esville | +0.088 |
Uniform | +0.086 |
eers | +0.086 |
icho | +0.085 |
Neg logit contributions | |
Fact | -0.059 |
itability | -0.056 |
amus | -0.054 |
weights | -0.054 |
quished | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -4.44e-03 |
Pos logit contributions | |
Fact | +0.981 |
itability | +0.938 |
amus | +0.928 |
weights | +0.921 |
quished | +0.882 |
Neg logit contributions | |
Soup | -0.702 |
esville | -0.625 |
Uniform | -0.609 |
eers | -0.608 |
Albion | -0.597 |
Token | California |
Token ID | 3442 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
Fact | +0.004 |
itability | +0.004 |
weights | +0.004 |
amus | +0.004 |
quished | +0.003 |
Neg logit contributions | |
Soup | -0.003 |
esville | -0.002 |
Uniform | -0.002 |
eers | -0.002 |
icho | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
Fact | +1.785 |
itability | +1.724 |
weights | +1.688 |
amus | +1.688 |
quished | +1.621 |
Neg logit contributions | |
Soup | -1.094 |
esville | -0.975 |
Uniform | -0.931 |
eers | -0.925 |
icho | -0.914 |
Token | Hide |
Token ID | 10415 |
Feature activation | +8.30e+00 |
Pos logit contributions | |
Soup | +0.067 |
esville | +0.062 |
Uniform | +0.061 |
eers | +0.060 |
Albion | +0.060 |
Neg logit contributions | |
Fact | -0.048 |
itability | -0.047 |
amus | -0.045 |
weights | -0.045 |
quished | -0.042 |
Token | Caption |
Token ID | 11260 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
Soup | +2.062 |
esville | +1.441 |
Uniform | +1.436 |
Albion | +1.251 |
eers | +1.240 |
Neg logit contributions | |
Fact | -9.569 |
itability | -9.566 |
ccording | -9.542 |
weights | -9.356 |
amus | -9.242 |
Token | 13 |
Token ID | 1511 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
Soup | +4.891 |
eers | +4.174 |
Arche | +4.112 |
esville | +4.048 |
Uniform | +3.907 |
Neg logit contributions | |
ccording | -12.508 |
itability | -12.006 |
weights | -11.547 |
Fact | -11.502 |
amus | -11.467 |
Token | of |
Token ID | 286 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Fact | +0.930 |
itability | +0.868 |
weights | +0.859 |
amus | +0.855 |
anger | +0.800 |
Neg logit contributions | |
Soup | -1.102 |
esville | -1.010 |
Uniform | -0.981 |
icho | -0.976 |
eers | -0.974 |
Token | 15 |
Token ID | 1315 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
Soup | +0.868 |
esville | +0.789 |
Uniform | +0.758 |
eers | +0.755 |
icho | +0.749 |
Neg logit contributions | |
Fact | -1.058 |
itability | -1.018 |
weights | -0.995 |
amus | -0.993 |
anger | -0.938 |
Token | Photos |
Token ID | 9434 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
Soup | +1.044 |
esville | +0.931 |
Uniform | +0.893 |
eers | +0.884 |
icho | +0.874 |
Neg logit contributions | |
Fact | -1.616 |
itability | -1.569 |
weights | -1.531 |
amus | -1.528 |
quished | -1.464 |
Token | : |
Token ID | 25 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
Soup | +3.480 |
esville | +3.070 |
Uniform | +2.934 |
eers | +2.932 |
Arche | +2.858 |
Neg logit contributions | |
ccording | -6.725 |
itability | -6.565 |
Fact | -6.532 |
amus | -6.291 |
quished | -6.286 |
Token | Pasadena |
Token ID | 41010 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
Soup | +0.358 |
esville | +0.316 |
Uniform | +0.302 |
eers | +0.298 |
Albion | +0.296 |
Neg logit contributions | |
Fact | -0.627 |
itability | -0.608 |
weights | -0.595 |
amus | -0.594 |
quished | -0.570 |
Token | , |
Token ID | 11 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
Fact | +0.013 |
itability | +0.013 |
weights | +0.012 |
amus | +0.012 |
quished | +0.012 |
Neg logit contributions | |
Soup | -0.008 |
esville | -0.007 |
Uniform | -0.007 |
eers | -0.007 |
icho | -0.007 |
Token | California |
Token ID | 3442 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
Soup | +1.411 |
esville | +1.265 |
Uniform | +1.215 |
eers | +1.210 |
Albion | +1.207 |
Neg logit contributions | |
Fact | -2.136 |
itability | -2.084 |
amus | -2.033 |
weights | -2.018 |
quished | -1.934 |
Token | . |
Token ID | 13 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
Fact | +0.819 |
itability | +0.794 |
weights | +0.777 |
amus | +0.776 |
quished | +0.744 |
Neg logit contributions | |
Soup | -0.477 |
esville | -0.423 |
Uniform | -0.403 |
eers | -0.399 |
icho | -0.395 |
Token | Hide |
Token ID | 10415 |
Feature activation | +8.25e+00 |
Pos logit contributions | |
Soup | +0.041 |
esville | +0.038 |
Uniform | +0.037 |
eers | +0.037 |
icho | +0.037 |
Neg logit contributions | |
Fact | -0.029 |
itability | -0.028 |
weights | -0.027 |
amus | -0.027 |
quished | -0.025 |
Token | Caption |
Token ID | 11260 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
Soup | +2.062 |
Uniform | +1.422 |
esville | +1.416 |
Albion | +1.242 |
Arche | +1.213 |
Neg logit contributions | |
Fact | -9.546 |
itability | -9.530 |
ccording | -9.490 |
weights | -9.327 |
amus | -9.213 |
Token | 12 |
Token ID | 1105 |
Feature activation | -5.19e+00 |
Pos logit contributions | |
Soup | +5.197 |
Uniform | +4.286 |
Arche | +4.251 |
esville | +4.224 |
Albion | +3.981 |
Neg logit contributions | |
ccording | -12.643 |
itability | -11.838 |
ategory | -11.417 |
amus | -11.304 |
quished | -11.292 |
Token | of |
Token ID | 286 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
Fact | +3.029 |
weights | +2.753 |
amus | +2.728 |
itability | +2.697 |
anger | +2.552 |
Neg logit contributions | |
Soup | -4.322 |
esville | -3.976 |
ingu | -3.911 |
icho | -3.902 |
Uniform | -3.889 |
Token | 15 |
Token ID | 1315 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
Soup | +1.115 |
esville | +1.012 |
Uniform | +0.973 |
eers | +0.967 |
icho | +0.959 |
Neg logit contributions | |
Fact | -1.387 |
itability | -1.336 |
weights | -1.305 |
amus | -1.302 |
quished | -1.234 |
Token | Photos |
Token ID | 9434 |
Feature activation | +7.56e+00 |
Pos logit contributions | |
Soup | +1.233 |
esville | +1.095 |
Uniform | +1.052 |
eers | +1.039 |
icho | +1.026 |
Neg logit contributions | |
Fact | -1.941 |
itability | -1.892 |
weights | -1.842 |
amus | -1.840 |
quished | -1.767 |
Token | : |
Token ID | 25 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
Soup | +3.591 |
esville | +3.161 |
eers | +3.026 |
Uniform | +3.019 |
Arche | +2.948 |
Neg logit contributions | |
ccording | -7.136 |
itability | -6.913 |
Fact | -6.863 |
quished | -6.632 |
amus | -6.618 |
Token | aggressive |
Token ID | 8361 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
Fact | +1.854 |
itability | +1.791 |
weights | +1.744 |
amus | +1.731 |
anger | +1.629 |
Neg logit contributions | |
Soup | -1.897 |
esville | -1.748 |
Uniform | -1.680 |
eers | -1.670 |
icho | -1.667 |
Token | victims |
Token ID | 4970 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Fact | +3.422 |
itability | +3.303 |
weights | +3.256 |
amus | +3.209 |
campaigner | +3.043 |
Neg logit contributions | |
Soup | -3.675 |
esville | -3.410 |
icho | -3.272 |
Uniform | -3.258 |
efe | -3.246 |
Token | of |
Token ID | 286 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Soup | +0.650 |
esville | +0.583 |
Uniform | +0.563 |
eers | +0.559 |
icho | +0.553 |
Neg logit contributions | |
Fact | -0.874 |
itability | -0.846 |
amus | -0.824 |
weights | -0.824 |
quished | -0.789 |
Token | bullying |
Token ID | 20714 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Fact | +0.970 |
itability | +0.930 |
weights | +0.906 |
amus | +0.899 |
quished | +0.847 |
Neg logit contributions | |
Soup | -1.083 |
esville | -0.997 |
Uniform | -0.961 |
eers | -0.958 |
icho | -0.956 |
Token | . |
Token ID | 764 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
Fact | +0.122 |
itability | +0.118 |
weights | +0.115 |
amus | +0.115 |
quished | +0.110 |
Neg logit contributions | |
Soup | -0.085 |
esville | -0.076 |
Uniform | -0.073 |
eers | -0.073 |
icho | -0.072 |
Token | Child |
Token ID | 5932 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
Soup | +1.935 |
esville | +1.701 |
Uniform | +1.674 |
Albion | +1.631 |
eers | +1.614 |
Neg logit contributions | |
Fact | -2.988 |
itability | -2.911 |
weights | -2.864 |
amus | -2.843 |
quished | -2.754 |
Token | Development |
Token ID | 7712 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
Soup | +6.291 |
esville | +5.626 |
Arche | +5.543 |
Uniform | +5.432 |
eers | +5.364 |
Neg logit contributions | |
Fact | -9.796 |
itability | -9.326 |
quished | -9.255 |
weights | -9.226 |
amus | -9.202 |
Token | , |
Token ID | 11 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
Soup | +1.875 |
esville | +1.684 |
Uniform | +1.609 |
eers | +1.590 |
icho | +1.564 |
Neg logit contributions | |
Fact | -2.799 |
itability | -2.738 |
weights | -2.667 |
amus | -2.645 |
quished | -2.555 |
Token | 68 |
Token ID | 8257 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
Soup | +0.226 |
esville | +0.200 |
Uniform | +0.189 |
eers | +0.189 |
icho | +0.187 |
Neg logit contributions | |
Fact | -0.425 |
itability | -0.411 |
weights | -0.404 |
amus | -0.403 |
quished | -0.385 |
Token | , |
Token ID | 11 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
Soup | +1.756 |
esville | +1.551 |
Uniform | +1.492 |
eers | +1.468 |
Albion | +1.446 |
Neg logit contributions | |
Fact | -2.860 |
itability | -2.805 |
weights | -2.728 |
amus | -2.722 |
quished | -2.626 |
Token | 6 |
Token ID | 718 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Soup | +1.562 |
esville | +1.360 |
Uniform | +1.329 |
eers | +1.290 |
Albion | +1.290 |
Neg logit contributions | |
Fact | -2.675 |
itability | -2.609 |
weights | -2.541 |
amus | -2.539 |
quished | -2.447 |
Token | extended |
Token ID | 7083 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
Soup | +0.665 |
esville | +0.599 |
Uniform | +0.582 |
eers | +0.576 |
Albion | +0.571 |
Neg logit contributions | |
Fact | -0.805 |
itability | -0.775 |
weights | -0.758 |
amus | -0.755 |
quished | -0.721 |
Token | universe |
Token ID | 6881 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
Fact | +1.382 |
itability | +1.333 |
weights | +1.298 |
amus | +1.283 |
anger | +1.209 |
Neg logit contributions | |
Soup | -1.523 |
esville | -1.411 |
icho | -1.353 |
eers | -1.352 |
Uniform | -1.347 |
Token | , |
Token ID | 11 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Soup | +0.408 |
esville | +0.367 |
Uniform | +0.353 |
eers | +0.351 |
icho | +0.347 |
Neg logit contributions | |
Fact | -0.553 |
itability | -0.534 |
weights | -0.522 |
amus | -0.521 |
quished | -0.497 |
Token | like |
Token ID | 588 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
Soup | +0.920 |
esville | +0.836 |
Uniform | +0.816 |
eers | +0.807 |
Albion | +0.803 |
Neg logit contributions | |
Fact | -0.946 |
itability | -0.908 |
weights | -0.885 |
amus | -0.882 |
quished | -0.839 |
Token | the |
Token ID | 262 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
Soup | +0.589 |
esville | +0.538 |
Uniform | +0.517 |
eers | +0.515 |
icho | +0.512 |
Neg logit contributions | |
Fact | -0.658 |
itability | -0.635 |
weights | -0.618 |
amus | -0.616 |
quished | -0.584 |
Token | comic |
Token ID | 9048 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
Soup | +0.596 |
esville | +0.542 |
Uniform | +0.524 |
eers | +0.520 |
icho | +0.515 |
Neg logit contributions | |
Fact | -0.673 |
itability | -0.648 |
weights | -0.633 |
amus | -0.631 |
quished | -0.599 |
Token | books |
Token ID | 3835 |
Feature activation | +6.61e+00 |
Pos logit contributions | |
Soup | +6.425 |
eers | +5.611 |
Uniform | +5.449 |
Arche | +5.353 |
Albion | +5.324 |
Neg logit contributions | |
Fact | -9.886 |
itability | -9.593 |
amus | -9.451 |
weights | -9.077 |
campaigner | -8.923 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
Soup | +3.729 |
esville | +3.214 |
Uniform | +3.214 |
eers | +3.167 |
Albion | +3.140 |
Neg logit contributions | |
Fact | -5.590 |
itability | -5.383 |
weights | -5.264 |
amus | -5.216 |
quished | -5.066 |
Token | the |
Token ID | 262 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
Soup | +1.625 |
esville | +1.465 |
Uniform | +1.441 |
Albion | +1.416 |
eers | +1.416 |
Neg logit contributions | |
Fact | -1.743 |
itability | -1.671 |
weights | -1.631 |
amus | -1.626 |
quished | -1.552 |
Token | games |
Token ID | 1830 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Soup | +1.774 |
esville | +1.574 |
Uniform | +1.549 |
eers | +1.521 |
Albion | +1.516 |
Neg logit contributions | |
Fact | -2.359 |
itability | -2.267 |
weights | -2.222 |
amus | -2.220 |
quished | -2.132 |
Token | . |
Token ID | 13 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
Soup | +0.595 |
esville | +0.534 |
Uniform | +0.514 |
eers | +0.510 |
icho | +0.504 |
Neg logit contributions | |
Fact | -0.841 |
itability | -0.813 |
weights | -0.795 |
amus | -0.793 |
quished | -0.757 |
Token |
|
Token ID | 198 |
Feature activation | -4.61e+00 |
Pos logit contributions | |
Fact | +1.006 |
itability | +0.963 |
weights | +0.943 |
amus | +0.940 |
ysc | +0.886 |
Neg logit contributions | |
Soup | -0.970 |
esville | -0.895 |
eers | -0.858 |
icho | -0.857 |
Uniform | -0.856 |
Token |
|
Token ID | 198 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
Fact | +2.860 |
weights | +2.611 |
amus | +2.556 |
itability | +2.554 |
anger | +2.387 |
Neg logit contributions | |
Soup | -3.706 |
esville | -3.447 |
icho | -3.354 |
ingu | -3.324 |
Uniform | -3.304 |
Token | OB |
Token ID | 9864 |
Feature activation | +9.11e-01 |
Pos logit contributions | |
Fact | +3.902 |
weights | +3.541 |
itability | +3.535 |
amus | +3.491 |
anger | +3.317 |
Neg logit contributions | |
Soup | -3.891 |
esville | -3.536 |
Uniform | -3.471 |
Albion | -3.411 |
eers | -3.399 |
Token | B |
Token ID | 33 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
Soup | +0.615 |
esville | +0.560 |
Uniform | +0.542 |
eers | +0.536 |
Albion | +0.533 |
Neg logit contributions | |
Fact | -0.684 |
itability | -0.659 |
weights | -0.644 |
amus | -0.643 |
quished | -0.611 |
Token | ( |
Token ID | 357 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Soup | +1.130 |
esville | +0.927 |
Uniform | +0.890 |
eers | +0.841 |
icho | +0.839 |
Neg logit contributions | |
Fact | -3.113 |
itability | -3.050 |
weights | -2.987 |
amus | -2.980 |
quished | -2.894 |
Token | Object |
Token ID | 10267 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
Soup | +1.088 |
esville | +0.992 |
Uniform | +0.961 |
eers | +0.949 |
icho | +0.945 |
Neg logit contributions | |
Fact | -1.192 |
itability | -1.153 |
amus | -1.122 |
weights | -1.120 |
quished | -1.067 |
Token | Al |
Token ID | 978 |
Feature activation | +1.15e-03 |
Pos logit contributions | |
Soup | +7.903 |
esville | +6.997 |
Uniform | +6.989 |
Arche | +6.888 |
Albion | +6.630 |
Neg logit contributions | |
Fact | -8.307 |
itability | -8.198 |
amus | -7.920 |
weights | -7.842 |
quished | -7.711 |
Token | igned |
Token ID | 3916 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
Soup | +0.001 |
esville | +0.001 |
Uniform | +0.001 |
eers | +0.001 |
icho | +0.001 |
Neg logit contributions | |
Fact | -0.001 |
itability | -0.001 |
weights | -0.001 |
amus | -0.001 |
quished | -0.001 |
Token | B |
Token ID | 347 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
Soup | +2.954 |
Uniform | +2.626 |
esville | +2.607 |
Albion | +2.509 |
Arche | +2.498 |
Neg logit contributions | |
Fact | -3.535 |
itability | -3.433 |
amus | -3.350 |
weights | -3.342 |
quished | -3.258 |
Token | ounding |
Token ID | 9969 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
Soup | +0.457 |
esville | +0.423 |
Uniform | +0.412 |
eers | +0.409 |
icho | +0.408 |
Neg logit contributions | |
Fact | -0.339 |
itability | -0.323 |
amus | -0.314 |
weights | -0.313 |
quished | -0.293 |
Token | Box |
Token ID | 8315 |
Feature activation | +9.01e+00 |
Pos logit contributions | |
Soup | +1.976 |
esville | +1.746 |
Uniform | +1.731 |
eers | +1.686 |
Albion | +1.674 |
Neg logit contributions | |
Fact | -2.650 |
itability | -2.583 |
amus | -2.509 |
weights | -2.507 |
quished | -2.415 |
Token | once |
Token ID | 1752 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
Fact | +2.891 |
itability | +2.845 |
weights | +2.730 |
amus | +2.683 |
ysc | +2.583 |
Neg logit contributions | |
Soup | -2.697 |
esville | -2.494 |
eers | -2.390 |
Uniform | -2.381 |
icho | -2.371 |
Token | again |
Token ID | 757 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
Soup | +2.648 |
esville | +2.461 |
Uniform | +2.389 |
Albion | +2.369 |
eers | +2.360 |
Neg logit contributions | |
Fact | -1.897 |
itability | -1.820 |
amus | -1.769 |
weights | -1.763 |
ccording | -1.691 |
Token | that |
Token ID | 326 |
Feature activation | +6.29e+00 |
Pos logit contributions | |
Soup | +1.828 |
esville | +1.682 |
Uniform | +1.637 |
eers | +1.627 |
Albion | +1.616 |
Neg logit contributions | |
Fact | -1.529 |
itability | -1.465 |
weights | -1.417 |
amus | -1.413 |
quished | -1.338 |
Token | the |
Token ID | 262 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
Soup | +3.890 |
esville | +3.476 |
Uniform | +3.404 |
Albion | +3.339 |
eers | +3.316 |
Neg logit contributions | |
Fact | -5.035 |
itability | -4.881 |
amus | -4.730 |
weights | -4.725 |
ccording | -4.654 |
Token | old |
Token ID | 1468 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
Soup | +2.280 |
esville | +2.027 |
Uniform | +1.949 |
eers | +1.917 |
Albion | +1.900 |
Neg logit contributions | |
Fact | -3.925 |
itability | -3.810 |
amus | -3.721 |
weights | -3.712 |
quished | -3.592 |
Token | address |
Token ID | 2209 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
Soup | +3.679 |
esville | +3.242 |
Uniform | +3.193 |
Albion | +3.089 |
efe | +3.086 |
Neg logit contributions | |
Fact | -6.539 |
itability | -6.282 |
weights | -6.226 |
amus | -6.187 |
quished | -6.129 |
Token | https |
Token ID | 3740 |
Feature activation | +6.89e+00 |
Pos logit contributions | |
Soup | +6.578 |
Uniform | +5.758 |
Arche | +5.535 |
esville | +5.504 |
efe | +5.462 |
Neg logit contributions | |
Fact | -9.227 |
itability | -9.096 |
weights | -8.798 |
amus | -8.793 |
quished | -8.558 |
Token | :// |
Token ID | 1378 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
Soup | +1.879 |
esville | +1.488 |
Uniform | +1.361 |
eers | +1.307 |
icho | +1.260 |
Neg logit contributions | |
Fact | -7.710 |
itability | -7.590 |
weights | -7.470 |
ccording | -7.450 |
amus | -7.409 |
Token | . |
Token ID | 764 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
Soup | +5.761 |
esville | +5.023 |
Uniform | +4.924 |
eers | +4.920 |
efe | +4.768 |
Neg logit contributions | |
Fact | -8.648 |
itability | -8.439 |
amus | -8.085 |
weights | -8.071 |
quished | -7.894 |
Token | nice |
Token ID | 44460 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Soup | +1.273 |
esville | +1.118 |
eers | +1.074 |
icho | +1.071 |
Uniform | +1.066 |
Neg logit contributions | |
Fact | -2.288 |
itability | -2.204 |
weights | -2.160 |
amus | -2.147 |
quished | -2.081 |
Token | hash |
Token ID | 17831 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
Soup | +0.985 |
esville | +0.917 |
Uniform | +0.894 |
eers | +0.889 |
icho | +0.882 |
Neg logit contributions | |
Fact | -0.624 |
itability | -0.593 |
weights | -0.573 |
amus | -0.571 |
ysc | -0.528 |
Token | sent |
Token ID | 1908 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
Fact | +1.043 |
itability | +1.000 |
weights | +0.985 |
amus | +0.976 |
quished | +0.924 |
Neg logit contributions | |
Soup | -0.950 |
esville | -0.870 |
Uniform | -0.836 |
eers | -0.829 |
icho | -0.829 |
Token | in |
Token ID | 259 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Soup | +0.371 |
esville | +0.335 |
Uniform | +0.322 |
eers | +0.320 |
icho | +0.317 |
Neg logit contributions | |
Fact | -0.484 |
itability | -0.468 |
weights | -0.457 |
amus | -0.455 |
quished | -0.434 |
Token | els |
Token ID | 1424 |
Feature activation | +6.23e+00 |
Pos logit contributions | |
Fact | +0.568 |
itability | +0.538 |
weights | +0.523 |
amus | +0.522 |
ysc | +0.480 |
Neg logit contributions | |
Soup | -0.983 |
esville | -0.914 |
Uniform | -0.891 |
eers | -0.886 |
icho | -0.882 |
Token | who |
Token ID | 508 |
Feature activation | +6.72e+00 |
Pos logit contributions | |
Soup | +3.864 |
esville | +3.435 |
Uniform | +3.363 |
Albion | +3.334 |
eers | +3.310 |
Neg logit contributions | |
Fact | -4.968 |
itability | -4.798 |
amus | -4.640 |
weights | -4.620 |
quished | -4.461 |
Token | stand |
Token ID | 1302 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
Soup | +4.702 |
esville | +4.310 |
Uniform | +4.220 |
Albion | +4.175 |
eers | +4.150 |
Neg logit contributions | |
Fact | -4.776 |
itability | -4.533 |
weights | -4.393 |
amus | -4.348 |
quished | -4.145 |
Token | guard |
Token ID | 4860 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
Soup | +5.770 |
esville | +5.169 |
Uniform | +5.139 |
Albion | +5.114 |
Arche | +5.013 |
Neg logit contributions | |
Fact | -8.195 |
itability | -8.062 |
amus | -7.656 |
weights | -7.510 |
ysc | -7.440 |
Token | today |
Token ID | 1909 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
Soup | +6.174 |
esville | +5.878 |
Albion | +5.612 |
Uniform | +5.576 |
Arche | +5.512 |
Neg logit contributions | |
Fact | -9.714 |
itability | -9.413 |
amus | -9.012 |
weights | -8.974 |
ccording | -8.862 |
Token | along |
Token ID | 1863 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
Soup | +2.871 |
esville | +2.596 |
Uniform | +2.471 |
Albion | +2.434 |
eers | +2.426 |
Neg logit contributions | |
Fact | -4.607 |
itability | -4.475 |
amus | -4.354 |
weights | -4.332 |
quished | -4.160 |
Token | the |
Token ID | 262 |
Feature activation | -7.41e-02 |
Pos logit contributions | |
Soup | +2.577 |
esville | +2.336 |
Uniform | +2.264 |
Albion | +2.236 |
Arche | +2.220 |
Neg logit contributions | |
Fact | -3.165 |
itability | -3.065 |
amus | -2.982 |
weights | -2.956 |
quished | -2.800 |
Token | front |
Token ID | 2166 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
Fact | +0.053 |
itability | +0.051 |
weights | +0.050 |
amus | +0.050 |
quished | +0.047 |
Neg logit contributions | |
Soup | -0.053 |
esville | -0.048 |
Uniform | -0.047 |
eers | -0.046 |
icho | -0.046 |
Token | iers |
Token ID | 3183 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
Soup | +1.861 |
esville | +1.640 |
Uniform | +1.566 |
eers | +1.552 |
icho | +1.533 |
Neg logit contributions | |
Fact | -3.590 |
itability | -3.486 |
amus | -3.392 |
weights | -3.374 |
quished | -3.242 |
Token | b |
Token ID | 65 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Fact | +0.808 |
itability | +0.775 |
weights | +0.761 |
amus | +0.756 |
ysc | +0.718 |
Neg logit contributions | |
Soup | -0.693 |
esville | -0.626 |
Uniform | -0.606 |
eers | -0.602 |
icho | -0.597 |
Token | mp |
Token ID | 3149 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
Fact | +0.540 |
itability | +0.511 |
weights | +0.496 |
amus | +0.493 |
ysc | +0.456 |
Neg logit contributions | |
Soup | -0.905 |
esville | -0.842 |
Uniform | -0.822 |
eers | -0.820 |
Albion | -0.814 |
Token | = |
Token ID | 796 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Soup | +1.438 |
esville | +1.269 |
Uniform | +1.219 |
eers | +1.195 |
icho | +1.184 |
Neg logit contributions | |
Fact | -2.349 |
itability | -2.309 |
weights | -2.249 |
amus | -2.240 |
quished | -2.150 |
Token | new |
Token ID | 649 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
Soup | +0.941 |
esville | +0.838 |
Uniform | +0.816 |
Albion | +0.796 |
eers | +0.794 |
Neg logit contributions | |
Fact | -1.283 |
itability | -1.249 |
weights | -1.213 |
amus | -1.212 |
quished | -1.159 |
Token | - |
Token ID | 12 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
Soup | +2.386 |
esville | +2.145 |
Uniform | +2.085 |
eers | +2.046 |
Albion | +2.035 |
Neg logit contributions | |
Fact | -3.086 |
itability | -3.012 |
amus | -2.923 |
weights | -2.912 |
quished | -2.800 |
Token | object |
Token ID | 15252 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
Soup | +2.992 |
esville | +2.811 |
Uniform | +2.710 |
eers | +2.698 |
efe | +2.690 |
Neg logit contributions | |
Fact | -2.225 |
itability | -2.177 |
amus | -2.108 |
weights | -2.068 |
ccording | -2.012 |
Token | system |
Token ID | 1080 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
Soup | +7.037 |
Uniform | +6.630 |
esville | +6.241 |
Albion | +6.139 |
Arche | +6.120 |
Neg logit contributions | |
Fact | -8.820 |
itability | -8.749 |
amus | -8.464 |
weights | -8.314 |
quished | -8.039 |
Token | . |
Token ID | 13 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
Soup | +0.746 |
esville | +0.658 |
Uniform | +0.634 |
icho | +0.625 |
eers | +0.622 |
Neg logit contributions | |
Fact | -1.239 |
itability | -1.189 |
weights | -1.173 |
amus | -1.167 |
ysc | -1.111 |
Token | draw |
Token ID | 19334 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
Soup | +0.291 |
esville | +0.266 |
Uniform | +0.256 |
eers | +0.254 |
icho | +0.252 |
Neg logit contributions | |
Fact | -0.335 |
itability | -0.324 |
amus | -0.315 |
weights | -0.314 |
quished | -0.299 |
Token | ing |
Token ID | 278 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
Soup | +0.981 |
esville | +0.840 |
Uniform | +0.795 |
eers | +0.785 |
icho | +0.775 |
Neg logit contributions | |
Fact | -2.275 |
itability | -2.210 |
weights | -2.168 |
amus | -2.164 |
ysc | -2.078 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Soup | +1.702 |
esville | +1.502 |
Uniform | +1.425 |
eers | +1.416 |
icho | +1.393 |
Neg logit contributions | |
Fact | -3.129 |
itability | -3.045 |
weights | -2.975 |
amus | -2.970 |
quished | -2.863 |
Token | pass |
Token ID | 1208 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
Fact | +0.110 |
itability | +0.106 |
weights | +0.104 |
amus | +0.104 |
quished | +0.099 |
Neg logit contributions | |
Soup | -0.072 |
esville | -0.064 |
Uniform | -0.061 |
eers | -0.061 |
icho | -0.060 |
Token | over |
Token ID | 625 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
Soup | +1.318 |
esville | +1.185 |
Uniform | +1.154 |
Albion | +1.147 |
eers | +1.137 |
Neg logit contributions | |
Fact | -1.646 |
itability | -1.578 |
amus | -1.548 |
weights | -1.538 |
quished | -1.477 |
Token | the |
Token ID | 262 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
Fact | +0.854 |
itability | +0.822 |
weights | +0.800 |
amus | +0.798 |
quished | +0.756 |
Neg logit contributions | |
Soup | -0.806 |
esville | -0.736 |
Uniform | -0.711 |
eers | -0.707 |
icho | -0.701 |
Token | middle |
Token ID | 3504 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
Fact | +1.465 |
itability | +1.382 |
weights | +1.363 |
amus | +1.348 |
campaigner | +1.263 |
Neg logit contributions | |
Soup | -1.891 |
esville | -1.764 |
icho | -1.716 |
eers | -1.704 |
Uniform | -1.692 |
Token | by |
Token ID | 416 |
Feature activation | -4.89e+00 |
Pos logit contributions | |
Fact | +2.284 |
itability | +2.204 |
weights | +2.175 |
amus | +2.149 |
quished | +2.028 |
Neg logit contributions | |
Soup | -2.144 |
esville | -1.972 |
eers | -1.882 |
Uniform | -1.878 |
icho | -1.863 |
Token | Siem |
Token ID | 45196 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
Fact | +3.297 |
itability | +3.140 |
weights | +3.108 |
amus | +3.054 |
campaigner | +2.904 |
Neg logit contributions | |
Soup | -3.571 |
esville | -3.311 |
eers | -3.209 |
icho | -3.196 |
Uniform | -3.173 |
Token | ian |
Token ID | 666 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
Fact | +6.189 |
itability | +6.122 |
amus | +6.076 |
weights | +5.477 |
kson | +5.389 |
Neg logit contributions | |
Soup | -10.559 |
esville | -9.681 |
eers | -9.572 |
WARE | -9.383 |
Uniform | -9.286 |
Token | . |
Token ID | 13 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Fact | +1.708 |
itability | +1.641 |
weights | +1.610 |
amus | +1.599 |
quished | +1.507 |
Neg logit contributions | |
Soup | -1.640 |
esville | -1.498 |
Uniform | -1.438 |
eers | -1.437 |
icho | -1.420 |
Token | Several |
Token ID | 12168 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
Fact | +0.870 |
itability | +0.830 |
weights | +0.813 |
amus | +0.808 |
quished | +0.765 |
Neg logit contributions | |
Soup | -0.884 |
esville | -0.816 |
eers | -0.784 |
Uniform | -0.782 |
icho | -0.781 |
Token | plays |
Token ID | 5341 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
Fact | +2.901 |
itability | +2.762 |
weights | +2.757 |
amus | +2.674 |
campaigner | +2.541 |
Neg logit contributions | |
Soup | -3.343 |
esville | -3.102 |
icho | -3.028 |
efe | -2.986 |
Uniform | -2.963 |
Token | later |
Token ID | 1568 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
Soup | +0.040 |
esville | +0.036 |
Uniform | +0.035 |
eers | +0.035 |
icho | +0.035 |
Neg logit contributions | |
Fact | -0.040 |
itability | -0.038 |
weights | -0.037 |
amus | -0.037 |
quished | -0.035 |
Token | co |
Token ID | 1073 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Fact | +1.544 |
itability | +1.423 |
weights | +1.420 |
amus | +1.378 |
anger | +1.250 |
Neg logit contributions | |
Soup | -4.198 |
esville | -3.927 |
Uniform | -3.919 |
eers | -3.862 |
XIV | -3.816 |
Token | / |
Token ID | 14 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
Fact | +1.406 |
itability | +1.312 |
weights | +1.310 |
amus | +1.297 |
anger | +1.225 |
Neg logit contributions | |
Soup | -1.298 |
esville | -1.152 |
Uniform | -1.132 |
eers | -1.118 |
icho | -1.111 |
Token | ni |
Token ID | 8461 |
Feature activation | -7.66e+00 |
Pos logit contributions | |
Fact | +3.163 |
itability | +2.989 |
amus | +2.928 |
weights | +2.920 |
ysc | +2.778 |
Neg logit contributions | |
Soup | -3.275 |
esville | -2.981 |
Uniform | -2.923 |
Albion | -2.872 |
eers | -2.861 |
Token | WB |
Token ID | 45607 |
Feature activation | -6.83e+00 |
Pos logit contributions | |
Fact | +5.156 |
itability | +4.772 |
amus | +4.720 |
weights | +4.667 |
ysc | +4.481 |
Neg logit contributions | |
Soup | -5.665 |
esville | -5.218 |
Uniform | -5.104 |
Albion | -5.008 |
eers | -4.983 |
Token | qv |
Token ID | 44179 |
Feature activation | -7.91e+00 |
Pos logit contributions | |
Fact | +4.620 |
itability | +4.310 |
amus | +4.249 |
weights | +4.233 |
anger | +4.037 |
Neg logit contributions | |
Soup | -5.097 |
esville | -4.606 |
Uniform | -4.570 |
Albion | -4.496 |
eers | -4.444 |
Token | fi |
Token ID | 12463 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
Fact | +5.321 |
itability | +4.949 |
amus | +4.886 |
weights | +4.838 |
ysc | +4.651 |
Neg logit contributions | |
Soup | -5.827 |
esville | -5.328 |
Uniform | -5.256 |
Albion | -5.160 |
Arche | -5.104 |
Token | W |
Token ID | 54 |
Feature activation | -9.12e+00 |
Pos logit contributions | |
Fact | +8.170 |
itability | +7.455 |
amus | +7.411 |
weights | +7.319 |
anger | +7.151 |
Neg logit contributions | |
Soup | -8.824 |
esville | -8.051 |
Uniform | -8.002 |
Albion | -7.890 |
Arche | -7.770 |
Token | u |
Token ID | 84 |
Feature activation | -9.56e+00 |
Pos logit contributions | |
Fact | +6.114 |
itability | +5.656 |
amus | +5.621 |
weights | +5.581 |
anger | +5.356 |
Neg logit contributions | |
Soup | -6.622 |
esville | -5.976 |
Uniform | -5.946 |
Albion | -5.852 |
Arche | -5.775 |
Token | — |
Token ID | 851 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Fact | +6.137 |
itability | +5.653 |
amus | +5.494 |
weights | +5.471 |
anger | +5.229 |
Neg logit contributions | |
Soup | -7.057 |
esville | -6.572 |
eers | -6.421 |
icho | -6.337 |
Uniform | -6.296 |
Token | Dan |
Token ID | 6035 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
Fact | +1.461 |
itability | +1.393 |
amus | +1.353 |
weights | +1.352 |
quished | +1.270 |
Neg logit contributions | |
Soup | -1.635 |
esville | -1.511 |
eers | -1.462 |
Uniform | -1.457 |
icho | -1.448 |
Token | Mer |
Token ID | 4638 |
Feature activation | -4.79e+00 |
Pos logit contributions | |
Fact | +2.060 |
itability | +1.969 |
amus | +1.943 |
weights | +1.920 |
anger | +1.819 |
Neg logit contributions | |
Soup | -1.887 |
esville | -1.716 |
eers | -1.684 |
icho | -1.676 |
Uniform | -1.665 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
Soup | +0.735 |
esville | +0.665 |
Uniform | +0.647 |
eers | +0.637 |
icho | +0.634 |
Neg logit contributions | |
Fact | -0.785 |
itability | -0.760 |
weights | -0.740 |
amus | -0.737 |
quished | -0.698 |
Token | Haw |
Token ID | 33055 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Soup | +3.612 |
esville | +3.340 |
eers | +3.234 |
icho | +3.170 |
Uniform | +3.158 |
Neg logit contributions | |
itability | -3.747 |
Fact | -3.696 |
amus | -3.614 |
weights | -3.576 |
campaigner | -3.467 |
Token | aii |
Token ID | 42648 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
Fact | +0.560 |
itability | +0.530 |
amus | +0.506 |
weights | +0.503 |
quished | +0.460 |
Neg logit contributions | |
Soup | -1.180 |
esville | -1.101 |
Uniform | -1.083 |
eers | -1.075 |
Albion | -1.068 |
Token | Congress |
Token ID | 3162 |
Feature activation | -5.56e+00 |
Pos logit contributions | |
Fact | +2.341 |
itability | +2.254 |
amus | +2.202 |
weights | +2.194 |
campaigner | +2.080 |
Neg logit contributions | |
Soup | -2.073 |
esville | -1.915 |
eers | -1.829 |
icho | -1.814 |
Uniform | -1.811 |
Token | woman |
Token ID | 8580 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
Fact | +3.430 |
itability | +3.231 |
weights | +3.198 |
amus | +3.165 |
campaigner | +2.949 |
Neg logit contributions | |
Soup | -4.413 |
esville | -4.093 |
Albion | -3.962 |
icho | -3.932 |
Uniform | -3.923 |
Token | Tuls |
Token ID | 33219 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
Fact | +1.829 |
itability | +1.732 |
weights | +1.710 |
amus | +1.689 |
ysc | +1.573 |
Neg logit contributions | |
Soup | -2.373 |
esville | -2.220 |
eers | -2.136 |
Uniform | -2.123 |
icho | -2.121 |
Token | i |
Token ID | 72 |
Feature activation | -7.18e+00 |
Pos logit contributions | |
Fact | +8.560 |
amus | +8.246 |
itability | +8.160 |
weights | +7.842 |
anger | +7.666 |
Neg logit contributions | |
eers | -7.852 |
Soup | -7.766 |
esville | -7.656 |
WARE | -7.543 |
Uniform | -7.264 |
Token | G |
Token ID | 402 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
Fact | +5.083 |
amus | +4.745 |
itability | +4.708 |
weights | +4.686 |
campaigner | +4.486 |
Neg logit contributions | |
Soup | -4.896 |
esville | -4.688 |
eers | -4.640 |
efe | -4.485 |
icho | -4.475 |
Token | abb |
Token ID | 6485 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
Fact | +1.664 |
itability | +1.565 |
weights | +1.543 |
amus | +1.505 |
anger | +1.464 |
Neg logit contributions | |
Soup | -2.011 |
esville | -1.861 |
Albion | -1.825 |
eers | -1.779 |
Uniform | -1.772 |
Token | ard |
Token ID | 446 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
Fact | +3.531 |
amus | +3.355 |
itability | +3.307 |
anger | +3.257 |
weights | +3.244 |
Neg logit contributions | |
Soup | -2.663 |
esville | -2.392 |
eers | -2.381 |
Albion | -2.362 |
Uniform | -2.296 |
Token | is |
Token ID | 318 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
Fact | +0.558 |
itability | +0.537 |
weights | +0.524 |
amus | +0.523 |
quished | +0.496 |
Neg logit contributions | |
Soup | -0.503 |
esville | -0.458 |
Uniform | -0.442 |
eers | -0.439 |
icho | -0.436 |
Token | The |
Token ID | 383 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
Fact | +0.259 |
itability | +0.248 |
weights | +0.241 |
amus | +0.240 |
quished | +0.227 |
Neg logit contributions | |
Soup | -0.270 |
esville | -0.248 |
Uniform | -0.239 |
eers | -0.239 |
icho | -0.237 |
Token | two |
Token ID | 734 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Fact | +0.581 |
itability | +0.559 |
weights | +0.544 |
amus | +0.544 |
ysc | +0.512 |
Neg logit contributions | |
Soup | -0.547 |
esville | -0.501 |
eers | -0.481 |
icho | -0.480 |
Uniform | -0.478 |
Token | form |
Token ID | 1296 |
Feature activation | -8.94e+00 |
Pos logit contributions | |
Fact | +0.918 |
itability | +0.881 |
weights | +0.863 |
amus | +0.860 |
ysc | +0.812 |
Neg logit contributions | |
Soup | -0.830 |
esville | -0.758 |
eers | -0.727 |
Uniform | -0.726 |
icho | -0.722 |
Token | - |
Token ID | 12 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
Fact | +6.098 |
weights | +5.792 |
itability | +5.675 |
amus | +5.597 |
anger | +5.362 |
Neg logit contributions | |
Soup | -6.384 |
esville | -5.755 |
Uniform | -5.596 |
efe | -5.579 |
Albion | -5.518 |
Token | in |
Token ID | 259 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
Fact | +4.927 |
weights | +4.729 |
itability | +4.692 |
amus | +4.614 |
anger | +4.493 |
Neg logit contributions | |
Soup | -2.521 |
esville | -2.172 |
Uniform | -2.088 |
Albion | -2.040 |
icho | -2.013 |
Token | - |
Token ID | 12 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
Fact | +3.853 |
itability | +3.685 |
weights | +3.657 |
amus | +3.590 |
anger | +3.420 |
Neg logit contributions | |
Soup | -3.613 |
esville | -3.335 |
Uniform | -3.171 |
efe | -3.149 |
Albion | -3.146 |
Token | one |
Token ID | 505 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
Fact | +8.673 |
weights | +8.656 |
itability | +8.307 |
anger | +7.970 |
amus | +7.947 |
Neg logit contributions | |
Soup | -7.906 |
esville | -7.244 |
Uniform | -7.095 |
Albion | -7.067 |
XIV | -6.944 |
Token | heat |
Token ID | 4894 |
Feature activation | -5.41e+00 |
Pos logit contributions | |
Fact | +0.115 |
itability | +0.108 |
weights | +0.104 |
amus | +0.103 |
quished | +0.094 |
Neg logit contributions | |
Soup | -0.234 |
esville | -0.219 |
Uniform | -0.214 |
eers | -0.213 |
icho | -0.212 |
Token | sinks |
Token ID | 38614 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Fact | +3.677 |
weights | +3.554 |
itability | +3.529 |
amus | +3.441 |
anger | +3.283 |
Neg logit contributions | |
Soup | -3.905 |
esville | -3.591 |
icho | -3.463 |
Albion | -3.452 |
Uniform | -3.448 |
Token | cool |
Token ID | 3608 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
Fact | +1.097 |
itability | +1.055 |
weights | +1.034 |
amus | +1.029 |
anger | +0.967 |
Neg logit contributions | |
Soup | -1.087 |
esville | -0.996 |
eers | -0.955 |
Uniform | -0.955 |
icho | -0.951 |
Token | the |
Token ID | 262 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
Fact | +2.104 |
itability | +2.024 |
amus | +1.984 |
weights | +1.982 |
quished | +1.864 |
Neg logit contributions | |
Soup | -2.095 |
esville | -1.915 |
Uniform | -1.844 |
icho | -1.843 |
eers | -1.837 |
Token | drops |
Token ID | 49253 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
Fact | +4.314 |
itability | +4.078 |
weights | +4.072 |
amus | +4.064 |
anger | +3.853 |
Neg logit contributions | |
Soup | -4.171 |
esville | -3.784 |
Uniform | -3.689 |
Albion | -3.668 |
XIV | -3.633 |
Token | - |
Token ID | 12 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
Fact | +6.038 |
itability | +5.759 |
weights | +5.642 |
amus | +5.598 |
anger | +5.340 |
Neg logit contributions | |
Soup | -6.137 |
esville | -5.642 |
eers | -5.564 |
Uniform | -5.482 |
Albion | -5.472 |
Token | re |
Token ID | 260 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
Fact | +3.921 |
itability | +3.713 |
weights | +3.684 |
amus | +3.640 |
anger | +3.469 |
Neg logit contributions | |
Soup | -3.807 |
esville | -3.448 |
Uniform | -3.397 |
Albion | -3.345 |
eers | -3.335 |
Token | za |
Token ID | 4496 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
Fact | +3.752 |
itability | +3.633 |
amus | +3.601 |
weights | +3.566 |
anger | +3.414 |
Neg logit contributions | |
Soup | -2.707 |
Uniform | -2.364 |
Albion | -2.305 |
esville | -2.292 |
eers | -2.252 |
Token | - |
Token ID | 12 |
Feature activation | -7.03e+00 |
Pos logit contributions | |
Fact | +3.417 |
itability | +3.270 |
weights | +3.226 |
amus | +3.216 |
quished | +3.044 |
Neg logit contributions | |
Soup | -3.189 |
esville | -2.902 |
Uniform | -2.803 |
Albion | -2.769 |
eers | -2.760 |
Token | as |
Token ID | 292 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
Fact | +4.336 |
amus | +4.226 |
itability | +4.196 |
weights | +4.144 |
anger | +3.951 |
Neg logit contributions | |
Soup | -5.574 |
Uniform | -5.049 |
esville | -5.029 |
Albion | -4.985 |
Arche | -4.871 |
Token | lan |
Token ID | 9620 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
itability | +4.154 |
amus | +3.729 |
Fact | +3.558 |
weights | +3.448 |
anger | +3.196 |
Neg logit contributions | |
Soup | -13.044 |
Uniform | -12.179 |
XIV | -12.101 |
Gleaming | -11.870 |
��極 | -11.816 |
Token | - |
Token ID | 12 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
Fact | +0.603 |
itability | +0.582 |
weights | +0.568 |
amus | +0.566 |
quished | +0.538 |
Neg logit contributions | |
Soup | -0.501 |
esville | -0.455 |
Uniform | -0.438 |
eers | -0.435 |
icho | -0.431 |
Token | after |
Token ID | 8499 |
Feature activation | -7.43e+00 |
Pos logit contributions | |
Fact | +3.477 |
itability | +3.309 |
weights | +3.282 |
amus | +3.251 |
anger | +3.084 |
Neg logit contributions | |
Soup | -3.440 |
esville | -3.142 |
Uniform | -3.064 |
Albion | -3.023 |
eers | -3.018 |
Token | - |
Token ID | 12 |
Feature activation | -7.07e+00 |
Pos logit contributions | |
Fact | +5.171 |
itability | +4.971 |
weights | +4.907 |
amus | +4.834 |
quished | +4.577 |
Neg logit contributions | |
Soup | -5.226 |
esville | -4.718 |
Uniform | -4.648 |
Albion | -4.604 |
eers | -4.586 |
Token | prof |
Token ID | 5577 |
Feature activation | -8.49e-01 |
Pos logit contributions | |
Fact | +5.178 |
itability | +4.886 |
weights | +4.881 |
amus | +4.849 |
anger | +4.709 |
Neg logit contributions | |
Soup | -4.847 |
esville | -4.335 |
Uniform | -4.320 |
Albion | -4.278 |
Arche | -4.231 |
Token | The |
Token ID | 464 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
Fact | +2.970 |
amus | +2.676 |
itability | +2.676 |
weights | +2.674 |
anger | +2.544 |
Neg logit contributions | |
Soup | -2.929 |
esville | -2.726 |
Uniform | -2.653 |
icho | -2.607 |
eers | -2.572 |
Token | Two |
Token ID | 4930 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
Fact | +0.701 |
itability | +0.672 |
weights | +0.656 |
amus | +0.655 |
quished | +0.619 |
Neg logit contributions | |
Soup | -0.694 |
esville | -0.640 |
Uniform | -0.616 |
eers | -0.612 |
icho | -0.610 |
Token | Towers |
Token ID | 29031 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
Soup | +0.202 |
esville | +0.186 |
Uniform | +0.180 |
eers | +0.179 |
icho | +0.178 |
Neg logit contributions | |
Fact | -0.187 |
itability | -0.179 |
weights | -0.174 |
amus | -0.173 |
quished | -0.164 |
Token | [ |
Token ID | 685 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
Fact | +2.430 |
itability | +2.318 |
weights | +2.300 |
amus | +2.283 |
anger | +2.170 |
Neg logit contributions | |
Soup | -2.594 |
esville | -2.438 |
Uniform | -2.329 |
icho | -2.311 |
eers | -2.295 |
Token | edit |
Token ID | 4370 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
Fact | +2.749 |
itability | +2.477 |
weights | +2.463 |
amus | +2.446 |
ysc | +2.261 |
Neg logit contributions | |
Soup | -4.497 |
esville | -4.291 |
icho | -4.128 |
eers | -4.124 |
Uniform | -4.109 |
Token | ] |
Token ID | 2361 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
Fact | +0.236 |
itability | +0.222 |
amus | +0.220 |
weights | +0.220 |
anger | +0.205 |
Neg logit contributions | |
Soup | -0.213 |
esville | -0.196 |
icho | -0.191 |
Uniform | -0.187 |
eers | -0.186 |
Token |
|
Token ID | 198 |
Feature activation | -8.25e+00 |
Pos logit contributions | |
Fact | +5.877 |
amus | +4.953 |
weights | +4.856 |
itability | +4.557 |
anger | +4.480 |
Neg logit contributions | |
Soup | -10.217 |
ingu | -10.128 |
esville | -10.038 |
icho | -9.902 |
��極 | -9.701 |
Token |
|
Token ID | 198 |
Feature activation | -7.35e+00 |
Pos logit contributions | |
Fact | +4.404 |
weights | +3.833 |
amus | +3.782 |
itability | +3.679 |
anger | +3.511 |
Neg logit contributions | |
Soup | -7.071 |
esville | -6.643 |
ingu | -6.578 |
icho | -6.510 |
��極 | -6.411 |
Token | Or |
Token ID | 5574 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
Fact | +5.200 |
weights | +4.572 |
amus | +4.551 |
itability | +4.521 |
anger | +4.404 |
Neg logit contributions | |
Soup | -5.349 |
esville | -4.981 |
Uniform | -4.916 |
icho | -4.794 |
efe | -4.673 |
Token | cs |
Token ID | 6359 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
Fact | +1.981 |
itability | +1.898 |
amus | +1.889 |
weights | +1.855 |
anger | +1.770 |
Neg logit contributions | |
Soup | -1.949 |
esville | -1.797 |
eers | -1.735 |
Uniform | -1.728 |
efe | -1.693 |
Token | sent |
Token ID | 1908 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
Soup | +0.350 |
esville | +0.320 |
Uniform | +0.311 |
eers | +0.309 |
Albion | +0.306 |
Neg logit contributions | |
Fact | -0.347 |
itability | -0.333 |
weights | -0.324 |
amus | -0.323 |
quished | -0.306 |
Token | war |
Token ID | 1175 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
Soup | +0.640 |
esville | +0.583 |
Uniform | +0.570 |
Albion | +0.561 |
eers | +0.560 |
Neg logit contributions | |
Fact | -0.637 |
itability | -0.614 |
amus | -0.595 |
weights | -0.594 |
quished | -0.563 |
Token | with |
Token ID | 351 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
Fact | +2.092 |
itability | +2.012 |
weights | +1.967 |
amus | +1.953 |
quished | +1.857 |
Neg logit contributions | |
Soup | -1.900 |
esville | -1.745 |
Uniform | -1.664 |
eers | -1.661 |
icho | -1.651 |
Token | Germany |
Token ID | 4486 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
Fact | +2.399 |
itability | +2.305 |
weights | +2.273 |
amus | +2.234 |
campaigner | +2.113 |
Neg logit contributions | |
Soup | -2.438 |
esville | -2.269 |
eers | -2.169 |
icho | -2.157 |
Uniform | -2.145 |
Token | [ |
Token ID | 685 |
Feature activation | -5.96e+00 |
Pos logit contributions | |
Fact | +2.460 |
itability | +2.342 |
weights | +2.300 |
amus | +2.299 |
ysc | +2.171 |
Neg logit contributions | |
Soup | -2.371 |
esville | -2.184 |
icho | -2.093 |
eers | -2.089 |
Uniform | -2.084 |
Token | edit |
Token ID | 4370 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
Fact | +3.164 |
itability | +2.800 |
weights | +2.798 |
amus | +2.767 |
anger | +2.578 |
Neg logit contributions | |
Soup | -5.330 |
esville | -5.005 |
eers | -4.847 |
icho | -4.846 |
Uniform | -4.826 |
Token | ] |
Token ID | 2361 |
Feature activation | -1.22e+01 |
Pos logit contributions | |
Fact | +0.304 |
itability | +0.288 |
weights | +0.284 |
amus | +0.283 |
anger | +0.266 |
Neg logit contributions | |
Soup | -0.268 |
esville | -0.245 |
icho | -0.237 |
Uniform | -0.234 |
eers | -0.233 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e+01 |
Pos logit contributions | |
Fact | +6.085 |
amus | +5.074 |
weights | +5.068 |
itability | +4.815 |
anger | +4.601 |
Neg logit contributions | |
Soup | -10.062 |
ingu | -9.781 |
esville | -9.733 |
icho | -9.686 |
��極 | -9.409 |
Token |
|
Token ID | 198 |
Feature activation | -9.96e+00 |
Pos logit contributions | |
Fact | +5.309 |
weights | +4.475 |
amus | +4.374 |
itability | +4.210 |
anger | +4.101 |
Neg logit contributions | |
Soup | -9.475 |
��極 | -9.035 |
esville | -8.929 |
ingu | -8.916 |
icho | -8.900 |
Token | From |
Token ID | 4863 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
Fact | +7.052 |
weights | +6.022 |
itability | +5.877 |
amus | +5.823 |
anger | +5.791 |
Neg logit contributions | |
Soup | -7.304 |
esville | -6.830 |
Uniform | -6.571 |
icho | -6.568 |
eers | -6.386 |
Token | 2 |
Token ID | 362 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
Fact | +2.519 |
itability | +2.406 |
weights | +2.366 |
amus | +2.352 |
anger | +2.230 |
Neg logit contributions | |
Soup | -2.224 |
esville | -2.033 |
eers | -1.963 |
icho | -1.949 |
Uniform | -1.927 |
Token | February |
Token ID | 3945 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
Fact | +2.830 |
itability | +2.669 |
weights | +2.637 |
amus | +2.619 |
ysc | +2.446 |
Neg logit contributions | |
Soup | -3.564 |
esville | -3.289 |
icho | -3.202 |
eers | -3.193 |
Uniform | -3.179 |
Token | Stone |
Token ID | 8026 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
Fact | +2.674 |
itability | +2.625 |
weights | +2.535 |
amus | +2.472 |
campaigner | +2.388 |
Neg logit contributions | |
Soup | -2.626 |
esville | -2.479 |
icho | -2.362 |
Uniform | -2.345 |
eers | -2.342 |
Token | talked |
Token ID | 6619 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
Soup | +3.400 |
esville | +3.022 |
efe | +2.903 |
Uniform | +2.897 |
Albion | +2.894 |
Neg logit contributions | |
Fact | -4.063 |
itability | -3.927 |
amus | -3.904 |
weights | -3.829 |
quished | -3.704 |
Token | with |
Token ID | 351 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
Fact | +2.977 |
itability | +2.902 |
weights | +2.831 |
amus | +2.803 |
campaigner | +2.688 |
Neg logit contributions | |
Soup | -1.718 |
esville | -1.568 |
eers | -1.480 |
Uniform | -1.466 |
icho | -1.445 |
Token | WikiLeaks |
Token ID | 19766 |
Feature activation | -8.53e+00 |
Pos logit contributions | |
Fact | +2.297 |
itability | +2.205 |
weights | +2.158 |
amus | +2.154 |
quished | +2.047 |
Neg logit contributions | |
Soup | -2.086 |
esville | -1.890 |
Uniform | -1.831 |
eers | -1.814 |
icho | -1.800 |
Token | about |
Token ID | 546 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
itability | +6.643 |
Fact | +6.541 |
campaigner | +6.409 |
weights | +6.333 |
kson | +6.097 |
Neg logit contributions | |
Soup | -5.128 |
esville | -4.815 |
icho | -4.554 |
ingu | -4.426 |
Arche | -4.417 |
Token | DNC |
Token ID | 17824 |
Feature activation | -1.22e+01 |
Pos logit contributions | |
Fact | +2.199 |
itability | +2.091 |
weights | +2.032 |
amus | +2.024 |
anger | +1.893 |
Neg logit contributions | |
Soup | -2.871 |
esville | -2.658 |
Uniform | -2.592 |
eers | -2.563 |
Albion | -2.550 |
Token | emails |
Token ID | 7237 |
Feature activation | -9.19e+00 |
Pos logit contributions | |
Fact | +6.741 |
itability | +6.567 |
weights | +6.548 |
campaigner | +6.298 |
hinges | +5.875 |
Neg logit contributions | |
esville | -9.621 |
icho | -9.457 |
Soup | -9.394 |
Arche | -9.054 |
Albion | -9.002 |
Token | Trump |
Token ID | 1301 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
Fact | +6.232 |
itability | +6.195 |
weights | +5.809 |
campaigner | +5.660 |
kson | +5.641 |
Neg logit contributions | |
Soup | -6.426 |
esville | -6.253 |
icho | -5.912 |
Uniform | -5.776 |
Arche | -5.763 |
Token | urges |
Token ID | 28804 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
Fact | +1.372 |
itability | +1.323 |
weights | +1.300 |
amus | +1.277 |
ysc | +1.223 |
Neg logit contributions | |
Soup | -1.256 |
esville | -1.144 |
icho | -1.102 |
Uniform | -1.092 |
eers | -1.091 |
Token | North |
Token ID | 2258 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
Fact | +2.401 |
itability | +2.317 |
weights | +2.247 |
amus | +2.244 |
campaigner | +2.163 |
Neg logit contributions | |
Soup | -2.130 |
esville | -1.958 |
eers | -1.884 |
Albion | -1.862 |
Uniform | -1.858 |
Token | Korea |
Token ID | 4969 |
Feature activation | -6.87e+00 |
Pos logit contributions | |
Fact | +0.381 |
itability | +0.373 |
weights | +0.368 |
amus | +0.362 |
anger | +0.359 |
Neg logit contributions | |
Soup | -0.298 |
esville | -0.280 |
eers | -0.269 |
Uniform | -0.264 |
Albion | -0.255 |
Token | the |
Token ID | 262 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
Fact | +2.121 |
itability | +2.045 |
weights | +2.008 |
amus | +1.968 |
ysc | +1.856 |
Neg logit contributions | |
Soup | -2.279 |
esville | -2.109 |
Uniform | -2.030 |
eers | -2.029 |
efe | -1.999 |
Token | feeling |
Token ID | 4203 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
Fact | +1.877 |
itability | +1.792 |
weights | +1.756 |
amus | +1.739 |
ysc | +1.638 |
Neg logit contributions | |
Soup | -1.898 |
esville | -1.768 |
eers | -1.703 |
Uniform | -1.691 |
icho | -1.676 |
Token | of |
Token ID | 286 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
Fact | +2.257 |
itability | +2.158 |
weights | +2.125 |
amus | +2.090 |
anger | +1.975 |
Neg logit contributions | |
Soup | -2.525 |
esville | -2.316 |
Uniform | -2.262 |
eers | -2.255 |
efe | -2.225 |
Token | a |
Token ID | 257 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
Fact | +2.341 |
itability | +2.257 |
weights | +2.216 |
amus | +2.168 |
anger | +2.042 |
Neg logit contributions | |
Soup | -2.643 |
esville | -2.441 |
eers | -2.365 |
Uniform | -2.355 |
efe | -2.337 |
Token | big |
Token ID | 1263 |
Feature activation | -8.36e+00 |
Pos logit contributions | |
Fact | +1.638 |
itability | +1.569 |
weights | +1.533 |
amus | +1.519 |
anger | +1.426 |
Neg logit contributions | |
Soup | -1.872 |
esville | -1.744 |
eers | -1.679 |
Uniform | -1.676 |
Albion | -1.654 |
Token | yo |
Token ID | 27406 |
Feature activation | -1.21e+01 |
Pos logit contributions | |
Fact | +5.332 |
weights | +5.115 |
itability | +5.061 |
amus | +4.936 |
anger | +4.717 |
Neg logit contributions | |
Soup | -6.061 |
esville | -5.805 |
eers | -5.590 |
Uniform | -5.442 |
efe | -5.419 |
Token | - |
Token ID | 12 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
Fact | +7.485 |
itability | +6.929 |
amus | +6.884 |
weights | +6.828 |
ysc | +6.558 |
Neg logit contributions | |
Soup | -8.746 |
esville | -8.251 |
eers | -8.024 |
Uniform | -7.768 |
Albion | -7.765 |
Token | yo |
Token ID | 8226 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
Fact | +1.722 |
itability | +1.641 |
weights | +1.607 |
amus | +1.595 |
ysc | +1.485 |
Neg logit contributions | |
Soup | -2.560 |
esville | -2.387 |
eers | -2.318 |
Uniform | -2.312 |
Albion | -2.295 |
Token | then |
Token ID | 788 |
Feature activation | -5.90e+00 |
Pos logit contributions | |
Fact | +3.035 |
weights | +2.912 |
itability | +2.889 |
amus | +2.831 |
campaigner | +2.686 |
Neg logit contributions | |
Soup | -3.124 |
esville | -2.845 |
Uniform | -2.745 |
eers | -2.739 |
efe | -2.736 |
Token | this |
Token ID | 428 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
Fact | +3.881 |
itability | +3.711 |
weights | +3.708 |
amus | +3.580 |
campaigner | +3.409 |
Neg logit contributions | |
Soup | -4.294 |
esville | -4.027 |
efe | -3.895 |
eers | -3.865 |
Uniform | -3.847 |
Token | isn |
Token ID | 2125 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
Fact | +1.704 |
itability | +1.632 |
weights | +1.603 |
amus | +1.589 |
anger | +1.489 |
Neg logit contributions | |
Soup | -1.905 |
esville | -1.780 |
eers | -1.713 |
Uniform | -1.712 |
efe | -1.686 |
Token | prime |
Token ID | 35505 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
Fact | +3.385 |
itability | +3.341 |
weights | +3.257 |
amus | +3.214 |
anger | +3.120 |
Neg logit contributions | |
Soup | -4.436 |
Uniform | -4.010 |
esville | -3.904 |
Albion | -3.892 |
Arche | -3.817 |
Token | - |
Token ID | 12 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
Fact | +0.057 |
itability | +0.055 |
weights | +0.054 |
amus | +0.053 |
anger | +0.050 |
Neg logit contributions | |
Soup | -0.050 |
esville | -0.045 |
Uniform | -0.043 |
eers | -0.043 |
Albion | -0.043 |
Token | chall |
Token ID | 36747 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
Fact | +1.136 |
itability | +1.076 |
weights | +1.051 |
amus | +1.036 |
anger | +0.978 |
Neg logit contributions | |
Soup | -1.889 |
esville | -1.738 |
Uniform | -1.727 |
Albion | -1.707 |
eers | -1.687 |
Token | enge |
Token ID | 3540 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Fact | +0.970 |
itability | +0.925 |
amus | +0.900 |
weights | +0.891 |
anger | +0.855 |
Neg logit contributions | |
Soup | -1.200 |
esville | -1.108 |
Uniform | -1.078 |
Albion | -1.068 |
eers | -1.062 |
Token | - |
Token ID | 12 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
Fact | +0.950 |
itability | +0.917 |
weights | +0.893 |
amus | +0.893 |
quished | +0.846 |
Neg logit contributions | |
Soup | -0.809 |
esville | -0.736 |
Uniform | -0.710 |
eers | -0.701 |
icho | -0.697 |
Token | 35 |
Token ID | 2327 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
Fact | +1.726 |
itability | +1.638 |
amus | +1.588 |
weights | +1.585 |
ysc | +1.473 |
Neg logit contributions | |
Soup | -2.602 |
esville | -2.424 |
Uniform | -2.382 |
eers | -2.345 |
icho | -2.327 |
Token | 790 |
Token ID | 37750 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
Fact | +6.887 |
amus | +6.456 |
itability | +6.395 |
weights | +6.268 |
anger | +6.053 |
Neg logit contributions | |
Soup | -9.311 |
Uniform | -8.378 |
esville | -8.355 |
eers | -8.194 |
Albion | -8.175 |
Token | 347 |
Token ID | 30995 |
Feature activation | -9.88e+00 |
Pos logit contributions | |
Fact | +5.919 |
itability | +5.513 |
amus | +5.498 |
weights | +5.381 |
ysc | +5.079 |
Neg logit contributions | |
Soup | -7.971 |
esville | -7.223 |
Uniform | -7.162 |
XIV | -7.018 |
Albion | -6.963 |
Token | . |
Token ID | 13 |
Feature activation | -9.13e+00 |
Pos logit contributions | |
Fact | +7.002 |
itability | +6.657 |
amus | +6.644 |
weights | +6.441 |
ysc | +6.330 |
Neg logit contributions | |
Soup | -6.413 |
esville | -5.850 |
Uniform | -5.623 |
icho | -5.592 |
Albion | -5.585 |
Token | html |
Token ID | 6494 |
Feature activation | -9.95e+00 |
Pos logit contributions | |
Fact | +4.254 |
itability | +3.993 |
weights | +3.885 |
amus | +3.876 |
anger | +3.665 |
Neg logit contributions | |
Soup | -8.434 |
Uniform | -7.891 |
esville | -7.785 |
Arche | -7.774 |
icho | -7.732 |
Token | https |
Token ID | 3740 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
Fact | +5.291 |
itability | +4.568 |
weights | +4.489 |
amus | +4.343 |
anger | +4.172 |
Neg logit contributions | |
Soup | -8.106 |
esville | -7.904 |
icho | -7.871 |
ingu | -7.653 |
rations | -7.519 |
Token | . |
Token ID | 13 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Fact | +0.910 |
itability | +0.877 |
weights | +0.865 |
amus | +0.853 |
anger | +0.806 |
Neg logit contributions | |
Soup | -0.825 |
esville | -0.754 |
Uniform | -0.726 |
eers | -0.720 |
icho | -0.716 |
Token | And |
Token ID | 843 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
Soup | +0.867 |
esville | +0.799 |
Uniform | +0.778 |
eers | +0.772 |
icho | +0.766 |
Neg logit contributions | |
Fact | -0.685 |
itability | -0.657 |
weights | -0.636 |
amus | -0.635 |
quished | -0.597 |
Token | if |
Token ID | 611 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
Fact | +2.508 |
itability | +2.392 |
weights | +2.360 |
amus | +2.318 |
campaigner | +2.173 |
Neg logit contributions | |
Soup | -2.954 |
esville | -2.761 |
Uniform | -2.682 |
icho | -2.656 |
eers | -2.650 |
Token | it |
Token ID | 340 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
Fact | +0.113 |
itability | +0.109 |
weights | +0.106 |
amus | +0.105 |
quished | +0.099 |
Neg logit contributions | |
Soup | -0.119 |
esville | -0.109 |
Uniform | -0.106 |
eers | -0.105 |
icho | -0.104 |
Token | had |
Token ID | 550 |
Feature activation | -8.27e+00 |
Pos logit contributions | |
Fact | +3.775 |
amus | +3.652 |
weights | +3.638 |
itability | +3.629 |
ysc | +3.404 |
Neg logit contributions | |
Soup | -3.739 |
esville | -3.440 |
Uniform | -3.346 |
Albion | -3.261 |
eers | -3.235 |
Token | all |
Token ID | 477 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
Fact | +5.102 |
weights | +5.006 |
amus | +4.963 |
itability | +4.911 |
hinges | +4.504 |
Neg logit contributions | |
Soup | -6.164 |
esville | -5.690 |
Uniform | -5.561 |
efe | -5.481 |
Albion | -5.456 |
Token | four |
Token ID | 1440 |
Feature activation | -6.22e+00 |
Pos logit contributions | |
amus | +6.913 |
Fact | +6.839 |
weights | +6.779 |
itability | +6.727 |
hinges | +6.140 |
Neg logit contributions | |
Soup | -8.638 |
esville | -7.851 |
Uniform | -7.829 |
efe | -7.657 |
Albion | -7.541 |
Token | it |
Token ID | 340 |
Feature activation | -5.12e+00 |
Pos logit contributions | |
Fact | +3.093 |
itability | +2.996 |
weights | +2.953 |
amus | +2.912 |
anger | +2.609 |
Neg logit contributions | |
Soup | -5.612 |
esville | -5.220 |
Uniform | -5.099 |
efe | -5.051 |
eers | -5.015 |
Token | was |
Token ID | 373 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Fact | +3.457 |
itability | +3.334 |
amus | +3.314 |
weights | +3.305 |
anger | +3.076 |
Neg logit contributions | |
Soup | -3.756 |
esville | -3.468 |
Uniform | -3.388 |
Albion | -3.301 |
eers | -3.293 |
Token | a |
Token ID | 257 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
Fact | +0.931 |
itability | +0.901 |
weights | +0.883 |
amus | +0.878 |
campaigner | +0.822 |
Neg logit contributions | |
Soup | -0.910 |
esville | -0.841 |
Uniform | -0.809 |
icho | -0.803 |
eers | -0.801 |
Token | nickel |
Token ID | 30666 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
Fact | +1.671 |
itability | +1.612 |
weights | +1.569 |
amus | +1.564 |
campaigner | +1.481 |
Neg logit contributions | |
Soup | -1.635 |
esville | -1.511 |
Uniform | -1.442 |
icho | -1.433 |
eers | -1.432 |
Token | # |
Token ID | 1303 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
Fact | +2.972 |
weights | +2.827 |
itability | +2.783 |
amus | +2.680 |
campaigner | +2.627 |
Neg logit contributions | |
Soup | -4.249 |
esville | -3.942 |
icho | -3.859 |
Uniform | -3.810 |
efe | -3.777 |
Token | Never |
Token ID | 12295 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
Soup | +0.232 |
esville | +0.197 |
Uniform | +0.184 |
eers | +0.182 |
icho | +0.179 |
Neg logit contributions | |
Fact | -0.605 |
itability | -0.591 |
weights | -0.580 |
amus | -0.579 |
quished | -0.558 |
Token | Trump |
Token ID | 6170 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
Fact | +2.471 |
itability | +2.233 |
weights | +2.228 |
amus | +2.158 |
anger | +2.058 |
Neg logit contributions | |
Soup | -2.877 |
esville | -2.600 |
Uniform | -2.596 |
Albion | -2.557 |
eers | -2.547 |
Token | ers |
Token ID | 364 |
Feature activation | -6.76e+00 |
Pos logit contributions | |
Fact | +1.550 |
itability | +1.470 |
weights | +1.446 |
amus | +1.426 |
campaigner | +1.341 |
Neg logit contributions | |
Soup | -2.251 |
esville | -2.083 |
Uniform | -2.033 |
icho | -2.020 |
Albion | -1.999 |
Token | of |
Token ID | 286 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
Fact | +4.885 |
weights | +4.589 |
itability | +4.567 |
amus | +4.565 |
campaigner | +4.370 |
Neg logit contributions | |
Soup | -4.589 |
esville | -4.172 |
icho | -4.082 |
eers | -4.067 |
Uniform | -4.042 |
Token | all |
Token ID | 477 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
Fact | +1.938 |
itability | +1.864 |
weights | +1.818 |
amus | +1.799 |
anger | +1.703 |
Neg logit contributions | |
Soup | -1.975 |
esville | -1.817 |
Uniform | -1.755 |
eers | -1.749 |
icho | -1.740 |
Token | , |
Token ID | 11 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
Fact | +6.097 |
weights | +5.774 |
itability | +5.667 |
amus | +5.442 |
campaigner | +5.258 |
Neg logit contributions | |
Soup | -9.695 |
esville | -8.875 |
Uniform | -8.862 |
eers | -8.816 |
icho | -8.757 |
Token | Wisconsin |
Token ID | 9279 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
Fact | +2.005 |
itability | +1.920 |
weights | +1.863 |
amus | +1.857 |
campaigner | +1.764 |
Neg logit contributions | |
Soup | -2.077 |
esville | -1.919 |
eers | -1.846 |
Uniform | -1.839 |
icho | -1.838 |
Token | conservative |
Token ID | 5940 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
Fact | +2.962 |
itability | +2.838 |
amus | +2.771 |
weights | +2.750 |
campaigner | +2.648 |
Neg logit contributions | |
Soup | -2.913 |
esville | -2.632 |
eers | -2.553 |
Uniform | -2.543 |
Albion | -2.529 |
Token | radio |
Token ID | 5243 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
Fact | +3.039 |
itability | +2.865 |
weights | +2.791 |
amus | +2.779 |
campaigner | +2.711 |
Neg logit contributions | |
Soup | -4.428 |
esville | -4.139 |
icho | -4.031 |
eers | -3.968 |
Uniform | -3.955 |
Token | host |
Token ID | 2583 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Fact | +1.788 |
itability | +1.696 |
weights | +1.637 |
amus | +1.635 |
campaigner | +1.540 |
Neg logit contributions | |
Soup | -2.424 |
esville | -2.262 |
Uniform | -2.192 |
eers | -2.181 |
icho | -2.176 |
Token | tubing |
Token ID | 43128 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
Soup | +0.958 |
esville | +0.865 |
Uniform | +0.851 |
eers | +0.843 |
Albion | +0.842 |
Neg logit contributions | |
Fact | -1.043 |
itability | -0.998 |
amus | -0.970 |
weights | -0.957 |
quished | -0.930 |
Token | , |
Token ID | 11 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Fact | +1.880 |
itability | +1.807 |
weights | +1.791 |
amus | +1.770 |
campaigner | +1.669 |
Neg logit contributions | |
Soup | -1.618 |
esville | -1.485 |
icho | -1.412 |
Uniform | -1.410 |
eers | -1.406 |
Token | features |
Token ID | 3033 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
Fact | +0.754 |
itability | +0.726 |
amus | +0.703 |
weights | +0.702 |
quished | +0.670 |
Neg logit contributions | |
Soup | -0.715 |
esville | -0.652 |
Uniform | -0.633 |
eers | -0.627 |
Albion | -0.622 |
Token | a |
Token ID | 257 |
Feature activation | -8.99e-01 |
Pos logit contributions | |
Fact | +1.373 |
itability | +1.320 |
weights | +1.288 |
amus | +1.276 |
quished | +1.194 |
Neg logit contributions | |
Soup | -1.616 |
esville | -1.491 |
Uniform | -1.441 |
eers | -1.436 |
icho | -1.430 |
Token | full |
Token ID | 1336 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
Fact | +0.633 |
itability | +0.610 |
weights | +0.594 |
amus | +0.591 |
quished | +0.558 |
Neg logit contributions | |
Soup | -0.648 |
esville | -0.595 |
Uniform | -0.574 |
eers | -0.572 |
icho | -0.568 |
Token | Shim |
Token ID | 31698 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
Soup | +0.359 |
esville | +0.329 |
Uniform | +0.319 |
eers | +0.316 |
icho | +0.314 |
Neg logit contributions | |
Fact | -0.357 |
itability | -0.342 |
amus | -0.333 |
weights | -0.332 |
quished | -0.315 |
Token | ano |
Token ID | 5733 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
Fact | +6.802 |
weights | +6.504 |
itability | +6.378 |
amus | +6.242 |
anger | +6.136 |
Neg logit contributions | |
Soup | -8.795 |
esville | -8.510 |
eers | -8.040 |
XIV | -7.820 |
Uniform | -7.759 |
Token | 600 |
Token ID | 10053 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
Fact | +2.577 |
itability | +2.478 |
weights | +2.421 |
amus | +2.395 |
anger | +2.284 |
Neg logit contributions | |
Soup | -2.503 |
esville | -2.321 |
eers | -2.203 |
icho | -2.187 |
Uniform | -2.185 |
Token | tri |
Token ID | 1333 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
Fact | +0.875 |
itability | +0.841 |
weights | +0.819 |
amus | +0.816 |
quished | +0.771 |
Neg logit contributions | |
Soup | -0.894 |
esville | -0.821 |
Uniform | -0.793 |
eers | -0.789 |
icho | -0.783 |
Token | - |
Token ID | 12 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
Fact | +2.924 |
itability | +2.819 |
weights | +2.791 |
amus | +2.761 |
anger | +2.661 |
Neg logit contributions | |
Soup | -2.726 |
esville | -2.501 |
eers | -2.405 |
Uniform | -2.384 |
Albion | -2.357 |
Token | color |
Token ID | 8043 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
Soup | +0.098 |
esville | +0.090 |
Uniform | +0.087 |
eers | +0.086 |
Albion | +0.086 |
Neg logit contributions | |
Fact | -0.101 |
itability | -0.098 |
amus | -0.095 |
weights | -0.094 |
quished | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
Fact | +0.078 |
itability | +0.073 |
weights | +0.072 |
amus | +0.072 |
anger | +0.067 |
Neg logit contributions | |
Soup | -0.082 |
esville | -0.075 |
Uniform | -0.072 |
icho | -0.072 |
eers | -0.072 |
Token | Follow |
Token ID | 7155 |
Feature activation | -9.23e-01 |
Pos logit contributions | |
Fact | +0.680 |
itability | +0.648 |
weights | +0.635 |
amus | +0.633 |
quished | +0.599 |
Neg logit contributions | |
Soup | -0.617 |
esville | -0.562 |
Uniform | -0.544 |
eers | -0.537 |
icho | -0.534 |
Token | @ |
Token ID | 2488 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Fact | +0.665 |
itability | +0.638 |
weights | +0.624 |
amus | +0.623 |
ysc | +0.588 |
Neg logit contributions | |
Soup | -0.648 |
esville | -0.599 |
Uniform | -0.573 |
eers | -0.572 |
icho | -0.566 |
Token | Pro |
Token ID | 2964 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
Fact | +0.934 |
itability | +0.893 |
weights | +0.872 |
amus | +0.872 |
quished | +0.823 |
Neg logit contributions | |
Soup | -0.927 |
esville | -0.849 |
Uniform | -0.824 |
eers | -0.814 |
icho | -0.809 |
Token | _ |
Token ID | 62 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
Fact | +0.393 |
itability | +0.377 |
weights | +0.368 |
amus | +0.368 |
quished | +0.349 |
Neg logit contributions | |
Soup | -0.336 |
esville | -0.304 |
Uniform | -0.294 |
eers | -0.292 |
icho | -0.289 |
Token | Cy |
Token ID | 20418 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
Fact | +2.594 |
itability | +2.390 |
weights | +2.358 |
amus | +2.336 |
ysc | +2.266 |
Neg logit contributions | |
Soup | -2.768 |
esville | -2.578 |
Uniform | -2.507 |
icho | -2.458 |
XIV | -2.441 |
Token | cling |
Token ID | 8493 |
Feature activation | -8.47e+00 |
Pos logit contributions | |
Fact | +2.279 |
amus | +1.750 |
weights | +1.595 |
itability | +1.425 |
ysc | +1.365 |
Neg logit contributions | |
Soup | -13.642 |
esville | -12.891 |
XIV | -12.841 |
Albion | -12.731 |
ariat | -12.719 |
Token |
|
Token ID | 198 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
Fact | +5.608 |
amus | +5.118 |
weights | +5.094 |
ysc | +4.962 |
itability | +4.923 |
Neg logit contributions | |
Soup | -6.242 |
esville | -5.763 |
icho | -5.500 |
efe | -5.436 |
eers | -5.396 |
Token |
|
Token ID | 198 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
Fact | +2.333 |
weights | +2.101 |
itability | +2.092 |
amus | +2.085 |
anger | +1.943 |
Neg logit contributions | |
Soup | -2.933 |
esville | -2.724 |
icho | -2.656 |
ingu | -2.634 |
Uniform | -2.628 |
Token | Tweet |
Token ID | 47845 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
Fact | +2.018 |
itability | +1.856 |
amus | +1.839 |
weights | +1.834 |
anger | +1.724 |
Neg logit contributions | |
Soup | -2.040 |
esville | -1.863 |
Uniform | -1.823 |
eers | -1.783 |
icho | -1.779 |
Token |
|
Token ID | 198 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
Soup | +0.481 |
esville | +0.436 |
eers | +0.420 |
Uniform | +0.420 |
Albion | +0.414 |
Neg logit contributions | |
Fact | -0.555 |
itability | -0.536 |
weights | -0.525 |
amus | -0.520 |
quished | -0.495 |
Token | , |
Token ID | 11 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
Fact | +2.897 |
itability | +2.790 |
amus | +2.734 |
weights | +2.699 |
anger | +2.580 |
Neg logit contributions | |
Soup | -2.496 |
esville | -2.322 |
eers | -2.212 |
Uniform | -2.205 |
icho | -2.194 |
Token | Death |
Token ID | 5830 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Fact | +1.825 |
itability | +1.754 |
amus | +1.706 |
weights | +1.702 |
quished | +1.609 |
Neg logit contributions | |
Soup | -1.772 |
esville | -1.646 |
Uniform | -1.577 |
eers | -1.576 |
icho | -1.572 |
Token | , |
Token ID | 11 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
Fact | +1.166 |
itability | +1.122 |
weights | +1.103 |
amus | +1.093 |
quished | +1.042 |
Neg logit contributions | |
Soup | -1.005 |
esville | -0.911 |
Uniform | -0.886 |
icho | -0.879 |
eers | -0.878 |
Token | Where |
Token ID | 6350 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
Fact | +0.537 |
itability | +0.517 |
weights | +0.503 |
amus | +0.502 |
quished | +0.475 |
Neg logit contributions | |
Soup | -0.515 |
esville | -0.470 |
Uniform | -0.454 |
eers | -0.451 |
icho | -0.448 |
Token | Is |
Token ID | 1148 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
Fact | +2.025 |
itability | +1.942 |
amus | +1.915 |
weights | +1.900 |
anger | +1.790 |
Neg logit contributions | |
Soup | -1.943 |
esville | -1.831 |
eers | -1.746 |
icho | -1.732 |
Uniform | -1.728 |
Token | Thy |
Token ID | 31468 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
Fact | +2.190 |
itability | +2.146 |
amus | +2.068 |
weights | +2.044 |
anger | +1.935 |
Neg logit contributions | |
Soup | -2.189 |
esville | -2.049 |
eers | -1.967 |
Uniform | -1.937 |
icho | -1.937 |
Token | Sting |
Token ID | 35785 |
Feature activation | +9.68e-01 |
Pos logit contributions | |
itability | +7.994 |
Fact | +7.918 |
amus | +7.735 |
weights | +7.353 |
anger | +7.210 |
Neg logit contributions | |
esville | -6.905 |
eers | -6.718 |
icho | -6.512 |
Soup | -6.488 |
ingu | -6.375 |
Token | - |
Token ID | 12 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
Soup | +0.749 |
esville | +0.691 |
Uniform | +0.671 |
eers | +0.666 |
icho | +0.661 |
Neg logit contributions | |
Fact | -0.636 |
itability | -0.608 |
weights | -0.590 |
amus | -0.587 |
quished | -0.553 |
Token | a |
Token ID | 64 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
Soup | +2.112 |
esville | +1.917 |
Uniform | +1.850 |
eers | +1.843 |
icho | +1.816 |
Neg logit contributions | |
Fact | -2.214 |
itability | -2.166 |
weights | -2.100 |
amus | -2.089 |
quished | -1.982 |
Token | - |
Token ID | 12 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
Fact | +0.300 |
itability | +0.289 |
amus | +0.281 |
weights | +0.281 |
quished | +0.267 |
Neg logit contributions | |
Soup | -0.253 |
esville | -0.230 |
Uniform | -0.221 |
eers | -0.220 |
icho | -0.218 |
Token | ling |
Token ID | 1359 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Fact | +3.170 |
itability | +2.975 |
weights | +2.896 |
amus | +2.873 |
anger | +2.749 |
Neg logit contributions | |
Soup | -2.888 |
esville | -2.699 |
icho | -2.614 |
Uniform | -2.599 |
eers | -2.584 |
Token |
|
Token ID | 198 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
Soup | +1.129 |
esville | +1.028 |
Uniform | +0.999 |
eers | +0.993 |
Albion | +0.981 |
Neg logit contributions | |
Fact | -1.261 |
itability | -1.209 |
weights | -1.176 |
amus | -1.169 |
quished | -1.121 |
Token |
|
Token ID | 198 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Soup | +2.040 |
esville | +1.856 |
Uniform | +1.795 |
eers | +1.784 |
Albion | +1.763 |
Neg logit contributions | |
Fact | -2.279 |
itability | -2.218 |
weights | -2.145 |
amus | -2.145 |
ccording | -2.066 |
Token | With |
Token ID | 3152 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
Soup | +0.969 |
esville | +0.885 |
Uniform | +0.860 |
eers | +0.855 |
Albion | +0.845 |
Neg logit contributions | |
Fact | -0.945 |
itability | -0.920 |
weights | -0.891 |
amus | -0.887 |
ysc | -0.846 |
Token | a |
Token ID | 257 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
Soup | +0.501 |
esville | +0.460 |
Uniform | +0.446 |
eers | +0.444 |
Albion | +0.439 |
Neg logit contributions | |
Fact | -0.495 |
itability | -0.473 |
weights | -0.460 |
amus | -0.459 |
quished | -0.436 |
Token | bare |
Token ID | 6247 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
Soup | +0.365 |
esville | +0.334 |
Uniform | +0.325 |
eers | +0.323 |
icho | +0.320 |
Neg logit contributions | |
Fact | -0.360 |
itability | -0.345 |
weights | -0.336 |
amus | -0.334 |
quished | -0.318 |
Token | bod |
Token ID | 16429 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
Soup | +0.980 |
esville | +0.902 |
Uniform | +0.877 |
eers | +0.871 |
Albion | +0.863 |
Neg logit contributions | |
Fact | -0.928 |
itability | -0.889 |
weights | -0.862 |
amus | -0.857 |
quished | -0.818 |
Token | kin |
Token ID | 5116 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
itability | +7.595 |
Fact | +7.502 |
amus | +7.386 |
weights | +7.152 |
ysc | +7.136 |
Neg logit contributions | |
Soup | -8.038 |
esville | -7.367 |
eers | -7.131 |
Albion | -7.098 |
Uniform | -6.966 |
Token | ? |
Token ID | 30 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
Soup | +0.312 |
esville | +0.284 |
Uniform | +0.275 |
eers | +0.273 |
icho | +0.270 |
Neg logit contributions | |
Fact | -0.349 |
itability | -0.336 |
weights | -0.327 |
amus | -0.326 |
quished | -0.310 |
Token | who |
Token ID | 508 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
Soup | +1.470 |
esville | +1.331 |
Uniform | +1.300 |
eers | +1.279 |
Albion | +1.276 |
Neg logit contributions | |
Fact | -1.543 |
itability | -1.494 |
weights | -1.447 |
amus | -1.442 |
quished | -1.378 |
Token | would |
Token ID | 561 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
Soup | +1.499 |
esville | +1.379 |
Uniform | +1.331 |
eers | +1.326 |
Albion | +1.313 |
Neg logit contributions | |
Fact | -1.472 |
itability | -1.410 |
weights | -1.367 |
amus | -1.362 |
quished | -1.301 |
Token | f |
Token ID | 277 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
Fact | +1.901 |
itability | +1.844 |
amus | +1.791 |
weights | +1.789 |
campaigner | +1.683 |
Neg logit contributions | |
Soup | -2.140 |
esville | -1.961 |
Uniform | -1.900 |
icho | -1.882 |
eers | -1.876 |
Token | between |
Token ID | 1022 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
Fact | +0.901 |
itability | +0.871 |
weights | +0.853 |
amus | +0.848 |
quished | +0.810 |
Neg logit contributions | |
Soup | -0.632 |
esville | -0.568 |
Uniform | -0.544 |
eers | -0.540 |
icho | -0.536 |
Token | Russian |
Token ID | 3394 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
Fact | +2.671 |
itability | +2.564 |
weights | +2.505 |
amus | +2.473 |
campaigner | +2.340 |
Neg logit contributions | |
Soup | -3.051 |
esville | -2.818 |
Uniform | -2.725 |
icho | -2.709 |
eers | -2.705 |
Token | hackers |
Token ID | 17110 |
Feature activation | -7.34e+00 |
Pos logit contributions | |
Fact | +1.322 |
itability | +1.254 |
weights | +1.221 |
amus | +1.204 |
campaigner | +1.121 |
Neg logit contributions | |
Soup | -1.933 |
esville | -1.803 |
Uniform | -1.749 |
eers | -1.733 |
icho | -1.729 |
Token | and |
Token ID | 290 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
Fact | +4.520 |
itability | +4.339 |
weights | +4.251 |
amus | +4.165 |
campaigner | +3.985 |
Neg logit contributions | |
Soup | -5.725 |
esville | -5.376 |
icho | -5.159 |
Uniform | -5.142 |
eers | -5.081 |
Token | the |
Token ID | 262 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
Fact | +2.333 |
itability | +2.246 |
weights | +2.194 |
amus | +2.179 |
quished | +2.071 |
Neg logit contributions | |
Soup | -2.214 |
esville | -2.023 |
Uniform | -1.953 |
eers | -1.942 |
icho | -1.932 |
Token | DNC |
Token ID | 17824 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
Fact | +2.315 |
itability | +2.223 |
weights | +2.165 |
amus | +2.138 |
quished | +2.024 |
Neg logit contributions | |
Soup | -2.792 |
esville | -2.594 |
icho | -2.500 |
eers | -2.496 |
Uniform | -2.495 |
Token | leak |
Token ID | 13044 |
Feature activation | -4.23e+00 |
Pos logit contributions | |
Fact | +6.491 |
itability | +6.258 |
weights | +6.044 |
campaigner | +5.816 |
hinges | +5.805 |
Neg logit contributions | |
Soup | -8.837 |
esville | -8.521 |
icho | -8.354 |
Albion | -8.180 |
Uniform | -7.963 |
Token | isn |
Token ID | 2125 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
Fact | +2.966 |
itability | +2.833 |
weights | +2.797 |
amus | +2.756 |
campaigner | +2.614 |
Neg logit contributions | |
Soup | -3.010 |
esville | -2.787 |
icho | -2.678 |
Uniform | -2.676 |
eers | -2.659 |
Token | 't |
Token ID | 470 |
Feature activation | -7.55e+00 |
Pos logit contributions | |
Fact | +2.079 |
itability | +1.991 |
amus | +1.970 |
weights | +1.958 |
anger | +1.875 |
Neg logit contributions | |
Soup | -0.949 |
esville | -0.809 |
Uniform | -0.780 |
icho | -0.769 |
eers | -0.758 |
Token | a |
Token ID | 257 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
Fact | +5.316 |
itability | +4.945 |
weights | +4.841 |
amus | +4.789 |
campaigner | +4.516 |
Neg logit contributions | |
Soup | -5.256 |
esville | -4.958 |
eers | -4.685 |
icho | -4.670 |
Uniform | -4.660 |
Token | ' |
Token ID | 705 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
Fact | +3.172 |
itability | +2.983 |
amus | +2.813 |
weights | +2.810 |
campaigner | +2.658 |
Neg logit contributions | |
Soup | -4.453 |
esville | -4.207 |
eers | -4.023 |
icho | -4.006 |
Uniform | -4.000 |
Token | , |
Token ID | 11 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
Fact | +0.274 |
itability | +0.263 |
weights | +0.257 |
amus | +0.257 |
quished | +0.243 |
Neg logit contributions | |
Soup | -0.232 |
esville | -0.211 |
Uniform | -0.203 |
eers | -0.201 |
icho | -0.200 |
Token | from |
Token ID | 422 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Fact | +0.322 |
itability | +0.309 |
weights | +0.302 |
amus | +0.301 |
quished | +0.285 |
Neg logit contributions | |
Soup | -0.304 |
esville | -0.278 |
Uniform | -0.268 |
eers | -0.267 |
icho | -0.265 |
Token | £ |
Token ID | 4248 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
Fact | +0.990 |
itability | +0.952 |
weights | +0.930 |
amus | +0.926 |
quished | +0.873 |
Neg logit contributions | |
Soup | -0.972 |
esville | -0.892 |
Uniform | -0.859 |
eers | -0.854 |
icho | -0.850 |
Token | 9 |
Token ID | 24 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
Fact | +0.315 |
itability | +0.303 |
weights | +0.294 |
amus | +0.294 |
quished | +0.278 |
Neg logit contributions | |
Soup | -0.320 |
esville | -0.293 |
Uniform | -0.284 |
eers | -0.282 |
icho | -0.280 |
Token | 79 |
Token ID | 3720 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
Fact | +0.230 |
itability | +0.220 |
weights | +0.214 |
amus | +0.214 |
quished | +0.202 |
Neg logit contributions | |
Soup | -0.244 |
esville | -0.224 |
Uniform | -0.217 |
eers | -0.216 |
icho | -0.214 |
Token | all |
Token ID | 477 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
Fact | +1.721 |
itability | +1.657 |
weights | +1.622 |
amus | +1.617 |
quished | +1.526 |
Neg logit contributions | |
Soup | -1.647 |
esville | -1.505 |
Uniform | -1.441 |
eers | -1.435 |
icho | -1.430 |
Token | - |
Token ID | 12 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
Fact | +7.037 |
weights | +6.806 |
itability | +6.728 |
amus | +6.554 |
campaigner | +6.377 |
Neg logit contributions | |
Soup | -8.361 |
esville | -7.776 |
icho | -7.511 |
Uniform | -7.351 |
efe | -7.346 |
Token | in |
Token ID | 259 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
Fact | +3.784 |
itability | +3.665 |
weights | +3.627 |
amus | +3.589 |
anger | +3.473 |
Neg logit contributions | |
Soup | -1.619 |
esville | -1.380 |
Uniform | -1.298 |
icho | -1.262 |
eers | -1.258 |
Token | clusive |
Token ID | 5731 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Fact | +2.099 |
itability | +2.049 |
amus | +1.987 |
weights | +1.986 |
quished | +1.898 |
Neg logit contributions | |
Soup | -1.514 |
esville | -1.359 |
Uniform | -1.298 |
eers | -1.290 |
icho | -1.290 |
Token | , |
Token ID | 11 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
Fact | +1.126 |
itability | +1.078 |
weights | +1.054 |
amus | +1.049 |
quished | +0.986 |
Neg logit contributions | |
Soup | -1.339 |
esville | -1.235 |
Uniform | -1.191 |
icho | -1.186 |
eers | -1.181 |
Token | departing |
Token ID | 37717 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
Fact | +1.547 |
itability | +1.483 |
weights | +1.453 |
amus | +1.442 |
quished | +1.363 |
Neg logit contributions | |
Soup | -1.524 |
esville | -1.403 |
Uniform | -1.343 |
eers | -1.339 |
icho | -1.336 |
Token | Media |
Token ID | 13152 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
Fact | +1.440 |
itability | +1.353 |
weights | +1.322 |
amus | +1.312 |
anger | +1.242 |
Neg logit contributions | |
Soup | -1.449 |
esville | -1.332 |
Uniform | -1.297 |
icho | -1.276 |
eers | -1.272 |
Token | playback |
Token ID | 16388 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
Fact | +1.328 |
itability | +1.270 |
amus | +1.236 |
weights | +1.235 |
campaigner | +1.171 |
Neg logit contributions | |
Soup | -1.232 |
esville | -1.131 |
Uniform | -1.085 |
eers | -1.075 |
icho | -1.072 |
Token | is |
Token ID | 318 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
Fact | +1.143 |
itability | +1.091 |
amus | +1.076 |
weights | +1.069 |
ysc | +1.000 |
Neg logit contributions | |
Soup | -1.236 |
esville | -1.135 |
icho | -1.098 |
Uniform | -1.094 |
eers | -1.093 |
Token | unsupported |
Token ID | 24222 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
Soup | +0.239 |
esville | +0.218 |
Uniform | +0.209 |
eers | +0.209 |
icho | +0.207 |
Neg logit contributions | |
Fact | -0.274 |
itability | -0.264 |
weights | -0.257 |
amus | -0.256 |
ysc | -0.243 |
Token | on |
Token ID | 319 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
Fact | +3.227 |
itability | +2.988 |
weights | +2.960 |
amus | +2.942 |
ysc | +2.781 |
Neg logit contributions | |
Soup | -3.413 |
esville | -3.225 |
icho | -3.148 |
ingu | -3.078 |
Uniform | -3.033 |
Token | your |
Token ID | 534 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
Fact | +3.047 |
itability | +2.857 |
weights | +2.838 |
amus | +2.769 |
anger | +2.662 |
Neg logit contributions | |
Soup | -3.385 |
icho | -3.098 |
esville | -3.095 |
Uniform | -3.054 |
efe | -3.009 |
Token | device |
Token ID | 3335 |
Feature activation | -6.56e+00 |
Pos logit contributions | |
Fact | +4.224 |
itability | +3.786 |
weights | +3.778 |
amus | +3.658 |
campaigner | +3.636 |
Neg logit contributions | |
Soup | -10.822 |
icho | -10.618 |
esville | -10.329 |
efe | -10.137 |
Arche | -10.007 |
Token | Media |
Token ID | 6343 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
Fact | +3.433 |
weights | +3.070 |
amus | +3.069 |
itability | +3.039 |
kson | +2.774 |
Neg logit contributions | |
Soup | -5.579 |
esville | -5.466 |
icho | -5.379 |
ingu | -5.245 |
efe | -5.167 |
Token | caption |
Token ID | 8305 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
Fact | +0.525 |
itability | +0.506 |
weights | +0.490 |
amus | +0.487 |
ysc | +0.458 |
Neg logit contributions | |
Soup | -0.551 |
esville | -0.508 |
Uniform | -0.491 |
icho | -0.488 |
efe | -0.483 |
Token | Prof |
Token ID | 4415 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Fact | +1.732 |
itability | +1.643 |
weights | +1.601 |
amus | +1.594 |
quished | +1.505 |
Neg logit contributions | |
Soup | -1.819 |
esville | -1.696 |
eers | -1.631 |
icho | -1.626 |
Uniform | -1.617 |
Token | Re |
Token ID | 797 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
Fact | +1.317 |
itability | +1.271 |
weights | +1.224 |
amus | +1.218 |
quished | +1.146 |
Neg logit contributions | |
Soup | -1.624 |
esville | -1.505 |
Uniform | -1.461 |
eers | -1.456 |
icho | -1.446 |
Token | two |
Token ID | 734 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
Fact | +2.192 |
itability | +2.089 |
weights | +2.062 |
amus | +2.023 |
quished | +1.910 |
Neg logit contributions | |
Soup | -2.540 |
esville | -2.361 |
eers | -2.268 |
Uniform | -2.258 |
icho | -2.253 |
Token | basketball |
Token ID | 9669 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
Fact | +2.114 |
itability | +2.035 |
weights | +2.031 |
amus | +1.965 |
campaigner | +1.869 |
Neg logit contributions | |
Soup | -2.242 |
esville | -2.085 |
icho | -1.985 |
Uniform | -1.980 |
efe | -1.975 |
Token | goals |
Token ID | 4661 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
Soup | +0.452 |
esville | +0.413 |
Uniform | +0.400 |
eers | +0.397 |
icho | +0.394 |
Neg logit contributions | |
Fact | -0.470 |
itability | -0.453 |
weights | -0.441 |
amus | -0.439 |
quished | -0.416 |
Token | on |
Token ID | 319 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
Fact | +1.891 |
itability | +1.820 |
weights | +1.793 |
amus | +1.781 |
ysc | +1.689 |
Neg logit contributions | |
Soup | -1.761 |
esville | -1.614 |
Uniform | -1.542 |
eers | -1.535 |
icho | -1.524 |
Token | opposite |
Token ID | 6697 |
Feature activation | -4.90e+00 |
Pos logit contributions | |
Fact | +1.076 |
itability | +1.032 |
weights | +1.008 |
amus | +0.999 |
quished | +0.943 |
Neg logit contributions | |
Soup | -1.208 |
esville | -1.114 |
Uniform | -1.076 |
eers | -1.072 |
icho | -1.065 |
Token | ends |
Token ID | 5645 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
Fact | +2.327 |
weights | +2.213 |
itability | +2.162 |
amus | +2.070 |
hinges | +1.959 |
Neg logit contributions | |
Soup | -4.541 |
esville | -4.277 |
efe | -4.126 |
Uniform | -4.105 |
icho | -4.095 |
Token | stand |
Token ID | 1302 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Fact | +7.439 |
weights | +7.254 |
amus | +7.229 |
itability | +7.113 |
quished | +6.966 |
Neg logit contributions | |
Soup | -7.694 |
esville | -7.028 |
eers | -6.840 |
Uniform | -6.748 |
icho | -6.733 |
Token | over |
Token ID | 625 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Soup | +0.937 |
esville | +0.846 |
Uniform | +0.819 |
eers | +0.813 |
icho | +0.806 |
Neg logit contributions | |
Fact | -1.162 |
itability | -1.124 |
amus | -1.089 |
weights | -1.089 |
quished | -1.036 |
Token | an |
Token ID | 281 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
Fact | +1.016 |
itability | +0.973 |
weights | +0.949 |
amus | +0.943 |
quished | +0.887 |
Neg logit contributions | |
Soup | -1.173 |
esville | -1.082 |
Uniform | -1.047 |
eers | -1.043 |
icho | -1.037 |
Token | empty |
Token ID | 6565 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
Fact | +1.728 |
itability | +1.661 |
weights | +1.629 |
amus | +1.604 |
quished | +1.527 |
Neg logit contributions | |
Soup | -1.734 |
esville | -1.587 |
eers | -1.534 |
Uniform | -1.526 |
icho | -1.525 |
Token | court |
Token ID | 2184 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
Fact | +0.601 |
itability | +0.577 |
weights | +0.563 |
amus | +0.560 |
quished | +0.526 |
Neg logit contributions | |
Soup | -0.663 |
esville | -0.612 |
Uniform | -0.590 |
eers | -0.588 |
icho | -0.586 |
Token | see |
Token ID | 766 |
Feature activation | +5.66e+00 |
Pos logit contributions | |
Soup | +3.551 |
esville | +3.154 |
Uniform | +3.133 |
Albion | +3.078 |
eers | +2.996 |
Neg logit contributions | |
Fact | -4.096 |
itability | -3.880 |
weights | -3.841 |
amus | -3.818 |
quished | -3.705 |
Token | her |
Token ID | 607 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
Soup | +4.229 |
Uniform | +3.727 |
esville | +3.722 |
Albion | +3.705 |
Arche | +3.682 |
Neg logit contributions | |
Fact | -3.875 |
itability | -3.640 |
amus | -3.588 |
weights | -3.577 |
quished | -3.468 |
Token | GP |
Token ID | 14714 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
Soup | +1.995 |
esville | +1.825 |
Uniform | +1.771 |
eers | +1.748 |
Albion | +1.743 |
Neg logit contributions | |
Fact | -1.835 |
itability | -1.736 |
weights | -1.694 |
amus | -1.690 |
quished | -1.617 |
Token | , |
Token ID | 11 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
Fact | +0.435 |
itability | +0.421 |
weights | +0.411 |
amus | +0.410 |
quished | +0.390 |
Neg logit contributions | |
Soup | -0.341 |
esville | -0.308 |
Uniform | -0.297 |
eers | -0.295 |
icho | -0.292 |
Token | the |
Token ID | 262 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
Soup | +1.768 |
esville | +1.602 |
Uniform | +1.574 |
Albion | +1.561 |
eers | +1.546 |
Neg logit contributions | |
Fact | -1.563 |
itability | -1.500 |
amus | -1.466 |
weights | -1.459 |
quished | -1.389 |
Token | child |
Token ID | 1200 |
Feature activation | +8.35e+00 |
Pos logit contributions | |
Soup | +2.183 |
esville | +1.990 |
Uniform | +1.942 |
eers | +1.919 |
Albion | +1.918 |
Neg logit contributions | |
Fact | -2.006 |
itability | -1.914 |
amus | -1.875 |
weights | -1.863 |
quished | -1.790 |
Token | stated |
Token ID | 5081 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
Soup | +4.947 |
esville | +4.580 |
eers | +4.411 |
Uniform | +4.312 |
Arche | +4.267 |
Neg logit contributions | |
Fact | -6.618 |
itability | -6.345 |
amus | -6.116 |
weights | -6.061 |
quished | -6.002 |
Token | clearly |
Token ID | 4084 |
Feature activation | +4.90e+00 |
Pos logit contributions | |
Fact | +0.071 |
itability | +0.068 |
weights | +0.066 |
amus | +0.066 |
quished | +0.063 |
Neg logit contributions | |
Soup | -0.076 |
esville | -0.070 |
Uniform | -0.068 |
eers | -0.067 |
icho | -0.067 |
Token | that |
Token ID | 326 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
Soup | +3.485 |
esville | +3.136 |
Albion | +3.090 |
Uniform | +3.080 |
icho | +3.048 |
Neg logit contributions | |
Fact | -3.464 |
itability | -3.311 |
amus | -3.233 |
weights | -3.221 |
quished | -3.100 |
Token | she |
Token ID | 673 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
Soup | +0.993 |
esville | +0.908 |
Uniform | +0.886 |
eers | +0.879 |
Albion | +0.875 |
Neg logit contributions | |
Fact | -0.843 |
itability | -0.807 |
weights | -0.784 |
amus | -0.783 |
quished | -0.742 |
Token | had |
Token ID | 550 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
Soup | +1.337 |
esville | +1.210 |
Uniform | +1.164 |
eers | +1.163 |
Albion | +1.148 |
Neg logit contributions | |
Fact | -1.704 |
itability | -1.642 |
weights | -1.593 |
amus | -1.592 |
quished | -1.540 |
Token | - |
Token ID | 12 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
Soup | +2.466 |
esville | +2.237 |
Uniform | +2.160 |
Albion | +2.094 |
icho | +2.093 |
Neg logit contributions | |
Fact | -3.534 |
itability | -3.413 |
weights | -3.320 |
amus | -3.292 |
quished | -3.198 |
Token | hour |
Token ID | 9769 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
Soup | +1.640 |
esville | +1.511 |
Uniform | +1.451 |
icho | +1.439 |
Albion | +1.435 |
Neg logit contributions | |
Fact | -1.529 |
itability | -1.477 |
amus | -1.425 |
weights | -1.422 |
quished | -1.346 |
Token | Turkey |
Token ID | 7137 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
Fact | +1.073 |
itability | +1.032 |
weights | +1.007 |
amus | +1.000 |
quished | +0.942 |
Neg logit contributions | |
Soup | -1.243 |
esville | -1.151 |
eers | -1.108 |
Uniform | -1.105 |
icho | -1.096 |
Token | - |
Token ID | 12 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
Fact | +1.263 |
itability | +1.212 |
weights | +1.185 |
amus | +1.180 |
quished | +1.122 |
Neg logit contributions | |
Soup | -1.154 |
esville | -1.063 |
Uniform | -1.020 |
eers | -1.019 |
icho | -1.004 |
Token | backed |
Token ID | 17078 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
Soup | +2.858 |
esville | +2.577 |
icho | +2.484 |
eers | +2.473 |
Uniform | +2.453 |
Neg logit contributions | |
Fact | -3.508 |
itability | -3.477 |
weights | -3.339 |
amus | -3.328 |
quished | -3.196 |
Token | offensive |
Token ID | 5859 |
Feature activation | +8.04e+00 |
Pos logit contributions | |
Soup | +2.850 |
esville | +2.556 |
Uniform | +2.554 |
Albion | +2.512 |
icho | +2.505 |
Neg logit contributions | |
Fact | -2.816 |
itability | -2.707 |
amus | -2.621 |
weights | -2.610 |
quished | -2.489 |
Token | . |
Token ID | 13 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
Soup | +4.910 |
esville | +4.358 |
Albion | +4.312 |
icho | +4.290 |
eers | +4.253 |
Neg logit contributions | |
Fact | -6.261 |
itability | -6.048 |
amus | -5.867 |
weights | -5.842 |
quished | -5.631 |
Token | AFP |
Token ID | 8916 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
Soup | +1.650 |
esville | +1.497 |
Uniform | +1.465 |
Albion | +1.439 |
eers | +1.438 |
Neg logit contributions | |
Fact | -1.723 |
itability | -1.682 |
weights | -1.632 |
amus | -1.627 |
ysc | -1.537 |
Token | has |
Token ID | 468 |
Feature activation | -5.94e+00 |
Pos logit contributions | |
Fact | +0.964 |
itability | +0.918 |
weights | +0.896 |
amus | +0.893 |
quished | +0.846 |
Neg logit contributions | |
Soup | -0.955 |
esville | -0.875 |
Uniform | -0.845 |
eers | -0.840 |
icho | -0.836 |
Token | more |
Token ID | 517 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
Fact | +4.075 |
itability | +3.839 |
weights | +3.718 |
amus | +3.709 |
campaigner | +3.490 |
Neg logit contributions | |
Soup | -4.338 |
esville | -3.948 |
icho | -3.855 |
eers | -3.834 |
Uniform | -3.803 |
Token | from |
Token ID | 422 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
Fact | +1.552 |
itability | +1.489 |
weights | +1.467 |
amus | +1.446 |
quished | +1.382 |
Neg logit contributions | |
Soup | -1.182 |
esville | -1.071 |
icho | -1.019 |
eers | -1.017 |
Uniform | -1.017 |
Token | | |
Token ID | 930 |
Feature activation | +4.40e+00 |
Pos logit contributions | |
Soup | +0.410 |
esville | +0.375 |
eers | +0.358 |
Uniform | +0.356 |
icho | +0.355 |
Neg logit contributions | |
Fact | -0.514 |
itability | -0.490 |
amus | -0.481 |
weights | -0.481 |
anger | -0.455 |
Token | int |
Token ID | 493 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
Soup | +2.227 |
Uniform | +1.914 |
esville | +1.901 |
icho | +1.818 |
eers | +1.802 |
Neg logit contributions | |
Fact | -4.017 |
itability | -3.959 |
amus | -3.840 |
weights | -3.832 |
quished | -3.656 |
Token | f |
Token ID | 277 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
Soup | +1.329 |
esville | +1.231 |
Uniform | +1.199 |
eers | +1.193 |
icho | +1.183 |
Neg logit contributions | |
Fact | -1.033 |
itability | -0.990 |
amus | -0.961 |
weights | -0.956 |
quished | -0.897 |
Token | d |
Token ID | 67 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
Fact | +1.295 |
itability | +1.217 |
weights | +1.206 |
amus | +1.180 |
anger | +1.097 |
Neg logit contributions | |
Soup | -2.174 |
esville | -1.980 |
Uniform | -1.962 |
eers | -1.953 |
Albion | -1.952 |
Token | | |
Token ID | 930 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
Soup | +1.169 |
esville | +1.057 |
Uniform | +1.014 |
eers | +1.010 |
icho | +1.001 |
Neg logit contributions | |
Fact | -1.529 |
itability | -1.470 |
weights | -1.439 |
amus | -1.437 |
quished | -1.361 |
Token | + |
Token ID | 1343 |
Feature activation | +8.68e+00 |
Pos logit contributions | |
Soup | +2.238 |
esville | +1.826 |
Uniform | +1.808 |
eers | +1.725 |
icho | +1.705 |
Neg logit contributions | |
Fact | -5.253 |
itability | -5.207 |
amus | -5.049 |
weights | -5.035 |
quished | -4.811 |
Token | -------------------- |
Token ID | 19351 |
Feature activation | +7.04e+00 |
Pos logit contributions | |
Soup | +5.975 |
esville | +5.441 |
eers | +5.254 |
efe | +5.232 |
icho | +5.226 |
Neg logit contributions | |
Fact | -5.941 |
itability | -5.904 |
ccording | -5.833 |
amus | -5.735 |
weights | -5.664 |
Token | --+ |
Token ID | 44785 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
Soup | +1.993 |
esville | +1.647 |
eers | +1.462 |
Uniform | +1.447 |
icho | +1.428 |
Neg logit contributions | |
Fact | -7.871 |
itability | -7.710 |
weights | -7.596 |
amus | -7.547 |
ccording | -7.451 |
Token | | |
Token ID | 930 |
Feature activation | +4.79e+00 |
Pos logit contributions | |
Soup | +2.140 |
esville | +1.911 |
Uniform | +1.849 |
eers | +1.822 |
icho | +1.803 |
Neg logit contributions | |
Fact | -2.987 |
itability | -2.905 |
weights | -2.832 |
amus | -2.821 |
quished | -2.707 |
Token | [ |
Token ID | 685 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
Soup | +3.246 |
Uniform | +2.870 |
esville | +2.865 |
eers | +2.770 |
icho | +2.757 |
Neg logit contributions | |
Fact | -3.564 |
itability | -3.517 |
weights | -3.386 |
amus | -3.368 |
quished | -3.163 |
Token | address |
Token ID | 21975 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
Soup | +1.790 |
esville | +1.631 |
Uniform | +1.585 |
eers | +1.561 |
icho | +1.544 |
Neg logit contributions | |
Fact | -2.175 |
itability | -2.109 |
weights | -2.038 |
amus | -2.034 |
quished | -1.962 |
Token | fram |
Token ID | 5346 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Soup | +1.256 |
esville | +1.129 |
Uniform | +1.091 |
eers | +1.085 |
icho | +1.072 |
Neg logit contributions | |
Fact | -1.610 |
itability | -1.556 |
amus | -1.516 |
weights | -1.513 |
quished | -1.454 |
Token | work |
Token ID | 1818 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
Fact | +1.201 |
itability | +1.157 |
weights | +1.129 |
amus | +1.128 |
ysc | +1.068 |
Neg logit contributions | |
Soup | -1.037 |
esville | -0.940 |
Uniform | -0.909 |
eers | -0.899 |
Albion | -0.895 |
Token | using |
Token ID | 1262 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
Fact | +0.064 |
itability | +0.061 |
weights | +0.060 |
amus | +0.060 |
quished | +0.057 |
Neg logit contributions | |
Soup | -0.058 |
esville | -0.053 |
Uniform | -0.051 |
eers | -0.051 |
icho | -0.050 |
Token | the |
Token ID | 262 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
Fact | +1.414 |
itability | +1.355 |
weights | +1.332 |
amus | +1.329 |
quished | +1.256 |
Neg logit contributions | |
Soup | -1.336 |
esville | -1.227 |
Uniform | -1.175 |
eers | -1.171 |
icho | -1.169 |
Token | command |
Token ID | 3141 |
Feature activation | +6.37e+00 |
Pos logit contributions | |
Soup | +0.173 |
esville | +0.157 |
Uniform | +0.151 |
eers | +0.150 |
icho | +0.149 |
Neg logit contributions | |
Fact | -0.206 |
itability | -0.199 |
weights | -0.194 |
amus | -0.193 |
quished | -0.184 |
Token | server |
Token ID | 4382 |
Feature activation | +7.25e+00 |
Pos logit contributions | |
Soup | +4.155 |
esville | +3.697 |
eers | +3.672 |
Uniform | +3.630 |
XIV | +3.516 |
Neg logit contributions | |
Fact | -4.888 |
itability | -4.756 |
amus | -4.528 |
weights | -4.525 |
quished | -4.312 |
Token | less |
Token ID | 1203 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
Soup | +2.258 |
esville | +1.896 |
eers | +1.722 |
Uniform | +1.713 |
efe | +1.604 |
Neg logit contributions | |
Fact | -7.722 |
itability | -7.571 |
amus | -7.431 |
quished | -7.417 |
weights | -7.192 |
Token | or |
Token ID | 393 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
Soup | +1.668 |
esville | +1.517 |
Uniform | +1.470 |
eers | +1.456 |
icho | +1.440 |
Neg logit contributions | |
Fact | -1.852 |
itability | -1.799 |
amus | -1.748 |
weights | -1.736 |
quished | -1.673 |
Token | sl |
Token ID | 1017 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
Soup | +0.966 |
esville | +0.881 |
Uniform | +0.854 |
eers | +0.847 |
icho | +0.839 |
Neg logit contributions | |
Fact | -0.995 |
itability | -0.957 |
weights | -0.929 |
amus | -0.926 |
quished | -0.878 |
Token | s |
Token ID | 82 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Soup | +1.551 |
esville | +1.338 |
eers | +1.276 |
icho | +1.260 |
Uniform | +1.252 |
Neg logit contributions | |
Fact | -2.862 |
itability | -2.785 |
amus | -2.712 |
weights | -2.699 |
quished | -2.607 |
Token | . |
Token ID | 764 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
Soup | +0.200 |
esville | +0.183 |
Uniform | +0.177 |
eers | +0.176 |
icho | +0.175 |
Neg logit contributions | |
Fact | -0.203 |
itability | -0.195 |
weights | -0.190 |
amus | -0.189 |
quished | -0.180 |
Token | 15 |
Token ID | 1315 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
Soup | +1.845 |
esville | +1.664 |
Uniform | +1.601 |
eers | +1.586 |
icho | +1.573 |
Neg logit contributions | |
Fact | -2.403 |
itability | -2.330 |
weights | -2.267 |
amus | -2.264 |
quished | -2.159 |
Token | | |
Token ID | 930 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
Soup | +1.218 |
esville | +1.097 |
Uniform | +1.054 |
eers | +1.047 |
icho | +1.037 |
Neg logit contributions | |
Fact | -1.643 |
itability | -1.586 |
weights | -1.550 |
amus | -1.546 |
quished | -1.474 |
Token | org |
Token ID | 8745 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
Soup | +1.358 |
esville | +1.234 |
Uniform | +1.211 |
eers | +1.194 |
icho | +1.182 |
Neg logit contributions | |
Fact | -1.348 |
itability | -1.301 |
weights | -1.259 |
amus | -1.256 |
quished | -1.184 |
Token | . |
Token ID | 764 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Soup | +2.508 |
esville | +2.226 |
Uniform | +2.114 |
eers | +2.108 |
Albion | +2.070 |
Neg logit contributions | |
Fact | -4.485 |
itability | -4.399 |
ccording | -4.300 |
weights | -4.272 |
amus | -4.252 |
Token | cl |
Token ID | 537 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
Fact | +0.741 |
itability | +0.706 |
weights | +0.685 |
amus | +0.677 |
campaigner | +0.632 |
Neg logit contributions | |
Soup | -1.105 |
esville | -1.035 |
Uniform | -0.999 |
icho | -0.994 |
eers | -0.991 |
Token | ojure |
Token ID | 32511 |
Feature activation | +7.91e+00 |
Pos logit contributions | |
Soup | +1.652 |
esville | +1.403 |
icho | +1.318 |
eers | +1.312 |
Uniform | +1.270 |
Neg logit contributions | |
Fact | -4.468 |
itability | -4.310 |
weights | -4.250 |
amus | -4.239 |
ccording | -4.150 |
Token | / |
Token ID | 1220 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
Soup | +3.557 |
esville | +3.088 |
Uniform | +2.977 |
eers | +2.890 |
Albion | +2.858 |
Neg logit contributions | |
Fact | -7.483 |
ccording | -7.421 |
itability | -7.335 |
weights | -7.149 |
amus | -7.110 |
Token | tools |
Token ID | 4899 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
Soup | +0.192 |
esville | +0.176 |
Uniform | +0.170 |
eers | +0.169 |
icho | +0.168 |
Neg logit contributions | |
Fact | -0.195 |
itability | -0.187 |
weights | -0.182 |
amus | -0.181 |
quished | -0.172 |
Token | . |
Token ID | 764 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
Soup | +2.416 |
esville | +2.113 |
Uniform | +2.041 |
eers | +2.020 |
icho | +1.985 |
Neg logit contributions | |
Fact | -4.158 |
itability | -4.049 |
weights | -3.945 |
amus | -3.940 |
quished | -3.824 |
Token | reader |
Token ID | 9173 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
Fact | +0.111 |
itability | +0.107 |
weights | +0.104 |
amus | +0.103 |
quished | +0.098 |
Neg logit contributions | |
Soup | -0.112 |
esville | -0.103 |
Uniform | -0.100 |
eers | -0.099 |
icho | -0.098 |
Token | | |
Token ID | 930 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Soup | +0.375 |
esville | +0.336 |
Uniform | +0.325 |
eers | +0.321 |
icho | +0.317 |
Neg logit contributions | |
Fact | -0.509 |
itability | -0.494 |
amus | -0.482 |
weights | -0.479 |
quished | -0.460 |
Token | reported |
Token ID | 2098 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
Soup | +4.035 |
esville | +3.649 |
Uniform | +3.588 |
Albion | +3.532 |
eers | +3.477 |
Neg logit contributions | |
Fact | -4.189 |
itability | -4.066 |
amus | -3.959 |
weights | -3.921 |
quished | -3.798 |
Token | as |
Token ID | 355 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
Soup | +0.230 |
esville | +0.209 |
Uniform | +0.201 |
eers | +0.200 |
icho | +0.198 |
Neg logit contributions | |
Fact | -0.270 |
itability | -0.260 |
weights | -0.253 |
amus | -0.253 |
quished | -0.240 |
Token | abstract |
Token ID | 12531 |
Feature activation | +6.07e+00 |
Pos logit contributions | |
Fact | +1.821 |
itability | +1.752 |
weights | +1.708 |
amus | +1.685 |
quished | +1.590 |
Neg logit contributions | |
Soup | -1.762 |
esville | -1.635 |
eers | -1.572 |
efe | -1.554 |
Uniform | -1.554 |
Token | s |
Token ID | 82 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
Soup | +3.525 |
esville | +3.144 |
Uniform | +3.049 |
eers | +3.009 |
Albion | +2.957 |
Neg logit contributions | |
Fact | -5.038 |
itability | -4.892 |
amus | -4.786 |
weights | -4.720 |
ysc | -4.573 |
Token | . |
Token ID | 13 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
Soup | +1.397 |
esville | +1.262 |
Uniform | +1.219 |
eers | +1.203 |
icho | +1.195 |
Neg logit contributions | |
Fact | -1.745 |
itability | -1.684 |
weights | -1.644 |
amus | -1.639 |
quished | -1.569 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
Fact | +0.159 |
itability | +0.152 |
weights | +0.148 |
amus | +0.147 |
quished | +0.138 |
Neg logit contributions | |
Soup | -0.173 |
esville | -0.160 |
eers | -0.154 |
Uniform | -0.154 |
icho | -0.153 |
Token | A |
Token ID | 32 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
Soup | +3.491 |
esville | +3.227 |
eers | +3.125 |
icho | +3.064 |
Uniform | +3.055 |
Neg logit contributions | |
itability | -3.630 |
Fact | -3.583 |
amus | -3.506 |
weights | -3.462 |
campaigner | -3.362 |
Token | group |
Token ID | 1448 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
Soup | +2.799 |
esville | +2.572 |
Uniform | +2.505 |
eers | +2.458 |
icho | +2.449 |
Neg logit contributions | |
Fact | -2.685 |
itability | -2.557 |
weights | -2.518 |
amus | -2.479 |
quished | -2.367 |
Token | of |
Token ID | 286 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
Soup | +2.379 |
esville | +2.165 |
Uniform | +2.101 |
eers | +2.089 |
Albion | +2.051 |
Neg logit contributions | |
Fact | -2.605 |
itability | -2.535 |
amus | -2.444 |
weights | -2.443 |
quished | -2.339 |
Token | Ottawa |
Token ID | 14074 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
Fact | +1.391 |
itability | +1.346 |
weights | +1.319 |
amus | +1.302 |
campaigner | +1.236 |
Neg logit contributions | |
Soup | -1.417 |
esville | -1.308 |
Uniform | -1.252 |
icho | -1.250 |
eers | -1.249 |
Token | fitness |
Token ID | 13547 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
Fact | +2.921 |
itability | +2.802 |
weights | +2.761 |
amus | +2.739 |
campaigner | +2.601 |
Neg logit contributions | |
Soup | -3.028 |
esville | -2.796 |
Uniform | -2.703 |
eers | -2.684 |
efe | -2.675 |
Token | s |
Token ID | 82 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
Soup | +1.638 |
esville | +1.439 |
Uniform | +1.375 |
eers | +1.367 |
icho | +1.341 |
Neg logit contributions | |
Fact | -2.840 |
itability | -2.777 |
amus | -2.714 |
weights | -2.666 |
quished | -2.611 |
Token | or |
Token ID | 393 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
Fact | +0.556 |
itability | +0.536 |
weights | +0.525 |
amus | +0.523 |
quished | +0.497 |
Neg logit contributions | |
Soup | -0.428 |
esville | -0.387 |
Uniform | -0.372 |
eers | -0.370 |
icho | -0.366 |
Token | adject |
Token ID | 31129 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Fact | +0.428 |
itability | +0.411 |
weights | +0.400 |
amus | +0.399 |
quished | +0.377 |
Neg logit contributions | |
Soup | -0.445 |
esville | -0.408 |
Uniform | -0.395 |
eers | -0.393 |
icho | -0.390 |
Token | ives |
Token ID | 1083 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
Soup | +0.721 |
esville | +0.640 |
Uniform | +0.613 |
eers | +0.603 |
icho | +0.601 |
Neg logit contributions | |
Fact | -1.208 |
itability | -1.168 |
weights | -1.147 |
amus | -1.143 |
ysc | -1.089 |
Token | , |
Token ID | 11 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
Soup | +2.128 |
esville | +1.899 |
Uniform | +1.831 |
eers | +1.820 |
Albion | +1.813 |
Neg logit contributions | |
Fact | -2.992 |
itability | -2.940 |
amus | -2.846 |
weights | -2.826 |
quished | -2.748 |
Token | representing |
Token ID | 10200 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Soup | +1.906 |
esville | +1.728 |
Uniform | +1.699 |
Albion | +1.664 |
icho | +1.660 |
Neg logit contributions | |
Fact | -1.957 |
itability | -1.901 |
amus | -1.850 |
weights | -1.829 |
quished | -1.762 |
Token | qualities |
Token ID | 14482 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Soup | +1.067 |
esville | +0.977 |
Uniform | +0.950 |
eers | +0.943 |
icho | +0.935 |
Neg logit contributions | |
Fact | -1.016 |
itability | -0.978 |
amus | -0.950 |
weights | -0.947 |
quished | -0.901 |
Token | such |
Token ID | 884 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Fact | +0.843 |
itability | +0.808 |
weights | +0.789 |
amus | +0.784 |
anger | +0.742 |
Neg logit contributions | |
Soup | -0.771 |
esville | -0.704 |
Uniform | -0.678 |
eers | -0.676 |
icho | -0.671 |
Token | as |
Token ID | 355 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Soup | +0.863 |
esville | +0.791 |
Uniform | +0.764 |
eers | +0.760 |
icho | +0.755 |
Neg logit contributions | |
Fact | -0.866 |
itability | -0.830 |
weights | -0.810 |
amus | -0.807 |
quished | -0.760 |
Token | solid |
Token ID | 4735 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
Soup | +0.105 |
esville | +0.096 |
Uniform | +0.093 |
eers | +0.092 |
icho | +0.091 |
Neg logit contributions | |
Fact | -0.113 |
itability | -0.109 |
weights | -0.106 |
amus | -0.105 |
quished | -0.100 |
Token | ity |
Token ID | 414 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
Fact | +1.567 |
itability | +1.520 |
weights | +1.476 |
amus | +1.466 |
ysc | +1.384 |
Neg logit contributions | |
Soup | -1.501 |
esville | -1.387 |
eers | -1.333 |
Uniform | -1.330 |
icho | -1.319 |
Token | Co |
Token ID | 1766 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
Fact | +0.160 |
itability | +0.152 |
weights | +0.149 |
amus | +0.148 |
campaigner | +0.138 |
Neg logit contributions | |
Soup | -0.171 |
esville | -0.159 |
eers | -0.154 |
icho | -0.153 |
Uniform | -0.152 |
Token | - |
Token ID | 12 |
Feature activation | +5.94e+00 |
Pos logit contributions | |
Soup | +2.933 |
esville | +2.564 |
Albion | +2.513 |
Uniform | +2.508 |
efe | +2.478 |
Neg logit contributions | |
itability | -4.336 |
Fact | -4.320 |
amus | -4.251 |
weights | -4.199 |
quished | -4.149 |
Token | operation |
Token ID | 27184 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
Soup | +2.971 |
esville | +2.556 |
Uniform | +2.460 |
Albion | +2.449 |
efe | +2.399 |
Neg logit contributions | |
itability | -5.308 |
Fact | -5.274 |
amus | -5.203 |
weights | -5.128 |
quished | -5.076 |
Token | and |
Token ID | 290 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
Soup | +1.446 |
esville | +1.321 |
Uniform | +1.281 |
eers | +1.271 |
Albion | +1.255 |
Neg logit contributions | |
Fact | -1.476 |
itability | -1.441 |
weights | -1.394 |
amus | -1.387 |
quished | -1.331 |
Token | Development |
Token ID | 7712 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
Soup | +2.146 |
esville | +1.849 |
Uniform | +1.840 |
eers | +1.767 |
Albion | +1.766 |
Neg logit contributions | |
Fact | -3.242 |
itability | -3.184 |
weights | -3.080 |
amus | -3.071 |
quished | -2.993 |
Token | ( |
Token ID | 357 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
Soup | +0.939 |
esville | +0.843 |
Uniform | +0.821 |
eers | +0.812 |
Albion | +0.801 |
Neg logit contributions | |
Fact | -1.249 |
itability | -1.215 |
weights | -1.179 |
amus | -1.178 |
quished | -1.131 |
Token | O |
Token ID | 46 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
Soup | +3.246 |
esville | +2.988 |
eers | +2.902 |
Uniform | +2.885 |
Albion | +2.849 |
Neg logit contributions | |
Fact | -3.156 |
itability | -3.069 |
amus | -2.987 |
weights | -2.953 |
quished | -2.855 |
Token | ECD |
Token ID | 27295 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Soup | +2.497 |
esville | +2.344 |
Uniform | +2.281 |
eers | +2.280 |
icho | +2.260 |
Neg logit contributions | |
Fact | -1.298 |
itability | -1.243 |
weights | -1.176 |
amus | -1.170 |
quished | -1.099 |
Token | ). |
Token ID | 737 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
Soup | +0.945 |
esville | +0.857 |
Uniform | +0.839 |
eers | +0.831 |
icho | +0.819 |
Neg logit contributions | |
Fact | -1.003 |
itability | -0.966 |
amus | -0.938 |
weights | -0.935 |
quished | -0.889 |
Token | In |
Token ID | 554 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
Soup | +0.202 |
esville | +0.187 |
Uniform | +0.180 |
eers | +0.180 |
icho | +0.179 |
Neg logit contributions | |
Fact | -0.172 |
itability | -0.164 |
weights | -0.160 |
amus | -0.159 |
quished | -0.149 |
Token | the |
Token ID | 262 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
Fact | +0.008 |
itability | +0.008 |
weights | +0.008 |
amus | +0.008 |
quished | +0.007 |
Neg logit contributions | |
Soup | -0.009 |
esville | -0.008 |
Uniform | -0.008 |
eers | -0.008 |
icho | -0.008 |
Token | by |
Token ID | 416 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
Soup | +2.294 |
esville | +2.101 |
Uniform | +2.014 |
eers | +2.011 |
Albion | +1.990 |
Neg logit contributions | |
Fact | -3.219 |
itability | -3.119 |
weights | -3.047 |
amus | -3.018 |
quished | -2.880 |
Token | Warriors |
Token ID | 12090 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Soup | +0.319 |
esville | +0.292 |
Uniform | +0.283 |
eers | +0.280 |
Albion | +0.278 |
Neg logit contributions | |
Fact | -0.330 |
itability | -0.318 |
weights | -0.309 |
amus | -0.308 |
quished | -0.293 |
Token | TV |
Token ID | 3195 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
Soup | +1.379 |
esville | +1.257 |
Uniform | +1.223 |
eers | +1.199 |
Albion | +1.188 |
Neg logit contributions | |
Fact | -1.458 |
itability | -1.415 |
weights | -1.380 |
amus | -1.370 |
quished | -1.308 |
Token | � |
Token ID | 447 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
Soup | +1.957 |
esville | +1.775 |
Uniform | +1.738 |
eers | +1.712 |
Albion | +1.683 |
Neg logit contributions | |
Fact | -2.135 |
itability | -2.083 |
weights | -2.023 |
amus | -2.014 |
ysc | -1.926 |
Token | � |
Token ID | 247 |
Feature activation | -8.93e-02 |
Pos logit contributions | |
Soup | +0.298 |
esville | +0.269 |
Uniform | +0.259 |
eers | +0.258 |
icho | +0.256 |
Neg logit contributions | |
Fact | -0.375 |
itability | -0.362 |
amus | -0.352 |
weights | -0.351 |
quished | -0.334 |
Token | s |
Token ID | 82 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Fact | +0.078 |
itability | +0.075 |
weights | +0.074 |
amus | +0.074 |
quished | +0.070 |
Neg logit contributions | |
Soup | -0.050 |
esville | -0.044 |
Uniform | -0.042 |
eers | -0.042 |
icho | -0.042 |
Token | Laure |
Token ID | 13826 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
Soup | +0.962 |
esville | +0.886 |
Uniform | +0.862 |
eers | +0.853 |
Albion | +0.848 |
Neg logit contributions | |
Fact | -0.821 |
itability | -0.787 |
weights | -0.765 |
amus | -0.762 |
quished | -0.720 |
Token | nce |
Token ID | 1198 |
Feature activation | +7.10e-01 |
Pos logit contributions | |
Soup | +2.362 |
esville | +2.157 |
Uniform | +2.092 |
eers | +2.066 |
Albion | +2.065 |
Neg logit contributions | |
Fact | -2.421 |
itability | -2.316 |
weights | -2.262 |
amus | -2.198 |
quished | -2.136 |
Token | Scott |
Token ID | 4746 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
Soup | +0.507 |
esville | +0.464 |
Uniform | +0.449 |
eers | +0.446 |
icho | +0.443 |
Neg logit contributions | |
Fact | -0.508 |
itability | -0.488 |
weights | -0.475 |
amus | -0.473 |
quished | -0.449 |
Token | and |
Token ID | 290 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
Soup | +1.971 |
esville | +1.789 |
Uniform | +1.732 |
eers | +1.706 |
Albion | +1.697 |
Neg logit contributions | |
Fact | -2.237 |
itability | -2.175 |
weights | -2.107 |
amus | -2.090 |
quished | -1.997 |
Token | radio |
Token ID | 5243 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Soup | +0.491 |
esville | +0.452 |
Uniform | +0.441 |
Albion | +0.436 |
eers | +0.436 |
Neg logit contributions | |
Fact | -0.392 |
itability | -0.376 |
weights | -0.364 |
amus | -0.362 |
quished | -0.342 |
Token | . |
Token ID | 13 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
Fact | +1.303 |
itability | +1.249 |
amus | +1.232 |
weights | +1.231 |
anger | +1.166 |
Neg logit contributions | |
Soup | -0.974 |
esville | -0.880 |
Uniform | -0.842 |
eers | -0.833 |
Albion | -0.829 |
Token | � |
Token ID | 447 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Fact | +0.530 |
itability | +0.505 |
weights | +0.492 |
amus | +0.490 |
quished | +0.461 |
Neg logit contributions | |
Soup | -0.583 |
esville | -0.539 |
eers | -0.519 |
Uniform | -0.519 |
icho | -0.516 |
Token | � |
Token ID | 251 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Soup | +1.000 |
esville | +0.901 |
Uniform | +0.867 |
eers | +0.860 |
icho | +0.852 |
Neg logit contributions | |
Fact | -1.330 |
itability | -1.285 |
weights | -1.256 |
amus | -1.252 |
quished | -1.193 |
Token | Well |
Token ID | 3894 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Soup | +0.752 |
esville | +0.688 |
Uniform | +0.671 |
eers | +0.664 |
icho | +0.659 |
Neg logit contributions | |
Fact | -0.684 |
itability | -0.663 |
weights | -0.642 |
amus | -0.640 |
quished | -0.606 |
Token | , |
Token ID | 11 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
Soup | +0.609 |
esville | +0.546 |
Uniform | +0.524 |
eers | +0.520 |
icho | +0.515 |
Neg logit contributions | |
Fact | -0.855 |
itability | -0.829 |
weights | -0.810 |
amus | -0.807 |
quished | -0.769 |
Token | after |
Token ID | 706 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Fact | +0.241 |
itability | +0.232 |
weights | +0.226 |
amus | +0.225 |
quished | +0.214 |
Neg logit contributions | |
Soup | -0.234 |
esville | -0.214 |
Uniform | -0.207 |
eers | -0.205 |
icho | -0.204 |
Token | that |
Token ID | 326 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
Soup | +1.068 |
esville | +0.978 |
Uniform | +0.943 |
eers | +0.935 |
Albion | +0.935 |
Neg logit contributions | |
Fact | -1.015 |
itability | -0.974 |
amus | -0.945 |
weights | -0.944 |
quished | -0.901 |
Token | I |
Token ID | 314 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
Soup | +0.554 |
esville | +0.503 |
Uniform | +0.486 |
eers | +0.482 |
icho | +0.478 |
Neg logit contributions | |
Fact | -0.630 |
itability | -0.608 |
weights | -0.593 |
amus | -0.590 |
quished | -0.562 |
Token | did |
Token ID | 750 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
Soup | +1.996 |
esville | +1.819 |
Uniform | +1.756 |
eers | +1.745 |
Albion | +1.733 |
Neg logit contributions | |
Fact | -2.153 |
itability | -2.078 |
weights | -2.019 |
amus | -1.993 |
quished | -1.924 |
Token | give |
Token ID | 1577 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Soup | +0.077 |
esville | +0.071 |
Uniform | +0.069 |
eers | +0.068 |
icho | +0.068 |
Neg logit contributions | |
Fact | -0.079 |
itability | -0.076 |
amus | -0.074 |
weights | -0.074 |
ysc | -0.069 |
Token | it |
Token ID | 340 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
Soup | +1.151 |
esville | +1.047 |
Uniform | +1.011 |
eers | +1.006 |
Albion | +0.997 |
Neg logit contributions | |
Fact | -1.196 |
itability | -1.154 |
weights | -1.125 |
amus | -1.114 |
quished | -1.077 |
Token | Kane |
Token ID | 20055 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Fact | +2.444 |
itability | +2.327 |
weights | +2.306 |
amus | +2.300 |
campaigner | +2.151 |
Neg logit contributions | |
Soup | -2.656 |
esville | -2.480 |
eers | -2.399 |
Uniform | -2.390 |
icho | -2.380 |
Token | had |
Token ID | 550 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
Fact | +1.084 |
itability | +1.042 |
weights | +1.018 |
amus | +1.017 |
quished | +0.962 |
Neg logit contributions | |
Soup | -1.009 |
esville | -0.921 |
Uniform | -0.889 |
eers | -0.883 |
icho | -0.877 |
Token | come |
Token ID | 1282 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Fact | +2.986 |
itability | +2.895 |
amus | +2.846 |
weights | +2.845 |
campaigner | +2.656 |
Neg logit contributions | |
Soup | -3.408 |
esville | -3.135 |
Uniform | -3.049 |
icho | -3.029 |
efe | -3.016 |
Token | inches |
Token ID | 8331 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
Fact | +1.125 |
itability | +1.085 |
weights | +1.075 |
amus | +1.070 |
ysc | +1.010 |
Neg logit contributions | |
Soup | -0.985 |
esville | -0.902 |
Uniform | -0.875 |
eers | -0.867 |
icho | -0.864 |
Token | from |
Token ID | 422 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
Fact | +0.009 |
itability | +0.008 |
weights | +0.008 |
amus | +0.008 |
quished | +0.008 |
Neg logit contributions | |
Soup | -0.008 |
esville | -0.007 |
Uniform | -0.007 |
eers | -0.007 |
icho | -0.007 |
Token | putting |
Token ID | 5137 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
Fact | +1.442 |
itability | +1.396 |
weights | +1.360 |
amus | +1.357 |
campaigner | +1.279 |
Neg logit contributions | |
Soup | -1.513 |
esville | -1.380 |
Uniform | -1.343 |
eers | -1.334 |
icho | -1.330 |
Token | Ireland |
Token ID | 7517 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
Fact | +1.759 |
itability | +1.690 |
weights | +1.664 |
amus | +1.651 |
anger | +1.552 |
Neg logit contributions | |
Soup | -1.808 |
esville | -1.662 |
Uniform | -1.610 |
icho | -1.596 |
eers | -1.590 |
Token | in |
Token ID | 287 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
Fact | +2.549 |
itability | +2.455 |
weights | +2.444 |
amus | +2.418 |
campaigner | +2.258 |
Neg logit contributions | |
Soup | -2.832 |
esville | -2.556 |
Uniform | -2.548 |
icho | -2.537 |
eers | -2.510 |
Token | front |
Token ID | 2166 |
Feature activation | +8.63e-03 |
Pos logit contributions | |
Fact | +0.739 |
itability | +0.699 |
amus | +0.675 |
weights | +0.666 |
anger | +0.607 |
Neg logit contributions | |
Soup | -1.426 |
esville | -1.330 |
Uniform | -1.297 |
eers | -1.297 |
icho | -1.293 |
Token | just |
Token ID | 655 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Soup | +0.006 |
esville | +0.005 |
Uniform | +0.005 |
eers | +0.005 |
icho | +0.005 |
Neg logit contributions | |
Fact | -0.007 |
itability | -0.006 |
weights | -0.006 |
amus | -0.006 |
quished | -0.006 |
Token | two |
Token ID | 734 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
Soup | +0.251 |
esville | +0.231 |
Uniform | +0.223 |
eers | +0.222 |
icho | +0.220 |
Neg logit contributions | |
Fact | -0.250 |
itability | -0.240 |
weights | -0.234 |
amus | -0.233 |
quished | -0.220 |
Token | . |
Token ID | 13 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
Fact | +3.017 |
itability | +2.901 |
amus | +2.848 |
weights | +2.840 |
campaigner | +2.699 |
Neg logit contributions | |
Soup | -2.333 |
esville | -2.060 |
icho | -2.004 |
Uniform | -1.984 |
eers | -1.982 |
Token | The |
Token ID | 383 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
Fact | +2.474 |
itability | +2.290 |
amus | +2.279 |
weights | +2.268 |
anger | +2.107 |
Neg logit contributions | |
Soup | -2.680 |
esville | -2.524 |
eers | -2.425 |
icho | -2.411 |
ingu | -2.392 |
Token | Liberals |
Token ID | 20162 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
Fact | +0.427 |
itability | +0.408 |
weights | +0.398 |
amus | +0.396 |
campaigner | +0.370 |
Neg logit contributions | |
Soup | -0.481 |
esville | -0.445 |
Uniform | -0.428 |
eers | -0.428 |
icho | -0.426 |
Token | are |
Token ID | 389 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
Fact | +1.207 |
itability | +1.149 |
weights | +1.122 |
amus | +1.121 |
campaigner | +1.057 |
Neg logit contributions | |
Soup | -1.253 |
esville | -1.146 |
Uniform | -1.105 |
eers | -1.103 |
icho | -1.100 |
Token | offering |
Token ID | 6011 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
Soup | +1.029 |
esville | +0.944 |
Uniform | +0.915 |
eers | +0.909 |
icho | +0.903 |
Neg logit contributions | |
Fact | -0.967 |
itability | -0.928 |
weights | -0.903 |
amus | -0.900 |
quished | -0.849 |
Token | the |
Token ID | 262 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
Fact | +0.027 |
itability | +0.026 |
weights | +0.026 |
amus | +0.025 |
quished | +0.024 |
Neg logit contributions | |
Soup | -0.026 |
esville | -0.024 |
Uniform | -0.023 |
eers | -0.023 |
icho | -0.023 |
Token | same |
Token ID | 976 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
Fact | +0.275 |
itability | +0.264 |
weights | +0.258 |
amus | +0.257 |
quished | +0.241 |
Neg logit contributions | |
Soup | -0.270 |
esville | -0.248 |
Uniform | -0.238 |
eers | -0.238 |
icho | -0.236 |
Token | raises |
Token ID | 12073 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
Soup | +1.904 |
esville | +1.731 |
Uniform | +1.674 |
eers | +1.660 |
Albion | +1.645 |
Neg logit contributions | |
Fact | -2.102 |
itability | -2.025 |
amus | -1.974 |
weights | -1.973 |
quished | -1.878 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
Fact | +0.322 |
itability | +0.309 |
weights | +0.301 |
amus | +0.300 |
quished | +0.284 |
Neg logit contributions | |
Soup | -0.316 |
esville | -0.289 |
Uniform | -0.279 |
eers | -0.277 |
icho | -0.275 |
Token | Conservatives |
Token ID | 17897 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
Fact | +1.078 |
itability | +1.022 |
weights | +0.993 |
amus | +0.984 |
campaigner | +0.914 |
Neg logit contributions | |
Soup | -1.551 |
esville | -1.448 |
Uniform | -1.398 |
eers | -1.395 |
icho | -1.390 |
Token | gave |
Token ID | 2921 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
Soup | +1.667 |
esville | +1.525 |
Uniform | +1.475 |
eers | +1.469 |
Albion | +1.456 |
Neg logit contributions | |
Fact | -1.757 |
itability | -1.688 |
weights | -1.640 |
amus | -1.637 |
quished | -1.563 |
Token | into |
Token ID | 656 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
Fact | +0.197 |
itability | +0.190 |
weights | +0.186 |
amus | +0.185 |
quished | +0.176 |
Neg logit contributions | |
Soup | -0.166 |
esville | -0.150 |
Uniform | -0.145 |
eers | -0.144 |
icho | -0.143 |
Token | the |
Token ID | 262 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
Fact | +1.674 |
itability | +1.605 |
weights | +1.555 |
amus | +1.552 |
quished | +1.452 |
Neg logit contributions | |
Soup | -1.982 |
esville | -1.838 |
Uniform | -1.777 |
eers | -1.766 |
icho | -1.763 |
Token | meditation |
Token ID | 16901 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
Fact | +1.531 |
itability | +1.489 |
amus | +1.436 |
weights | +1.434 |
anger | +1.354 |
Neg logit contributions | |
Soup | -1.468 |
esville | -1.357 |
Uniform | -1.297 |
icho | -1.296 |
eers | -1.292 |
Token | on |
Token ID | 319 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
Fact | +1.888 |
itability | +1.823 |
weights | +1.783 |
amus | +1.775 |
ysc | +1.685 |
Neg logit contributions | |
Soup | -1.712 |
esville | -1.565 |
Uniform | -1.504 |
eers | -1.496 |
icho | -1.488 |
Token | relationships |
Token ID | 6958 |
Feature activation | -4.57e+00 |
Pos logit contributions | |
Soup | +0.677 |
esville | +0.618 |
Uniform | +0.602 |
Albion | +0.595 |
eers | +0.594 |
Neg logit contributions | |
Fact | -0.675 |
itability | -0.643 |
weights | -0.629 |
amus | -0.629 |
quished | -0.598 |
Token | and |
Token ID | 290 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Fact | +3.469 |
itability | +3.331 |
weights | +3.238 |
amus | +3.212 |
anger | +3.070 |
Neg logit contributions | |
Soup | -3.014 |
esville | -2.760 |
Uniform | -2.649 |
eers | -2.649 |
icho | -2.605 |
Token | life |
Token ID | 1204 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
Fact | +1.528 |
itability | +1.475 |
weights | +1.437 |
amus | +1.425 |
anger | +1.347 |
Neg logit contributions | |
Soup | -1.620 |
esville | -1.489 |
Uniform | -1.434 |
eers | -1.433 |
icho | -1.431 |
Token | that |
Token ID | 326 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
Fact | +2.276 |
itability | +2.183 |
weights | +2.153 |
amus | +2.117 |
anger | +2.033 |
Neg logit contributions | |
Soup | -1.821 |
esville | -1.635 |
Uniform | -1.572 |
eers | -1.565 |
icho | -1.561 |
Token | we |
Token ID | 356 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
Soup | +2.350 |
esville | +2.120 |
Uniform | +2.063 |
eers | +2.060 |
Albion | +2.044 |
Neg logit contributions | |
Fact | -2.392 |
itability | -2.304 |
weights | -2.242 |
amus | -2.233 |
quished | -2.139 |
Token | all |
Token ID | 477 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
Soup | +2.068 |
esville | +1.913 |
eers | +1.850 |
Uniform | +1.841 |
Albion | +1.825 |
Neg logit contributions | |
Fact | -2.006 |
itability | -1.906 |
weights | -1.866 |
amus | -1.849 |
quished | -1.770 |
Token | know |
Token ID | 760 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
Fact | +1.391 |
itability | +1.357 |
amus | +1.309 |
weights | +1.301 |
campaigner | +1.220 |
Neg logit contributions | |
Soup | -1.657 |
esville | -1.519 |
Uniform | -1.479 |
icho | -1.470 |
eers | -1.467 |
Token | a |
Token ID | 257 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
Soup | +0.753 |
esville | +0.691 |
Uniform | +0.669 |
eers | +0.666 |
icho | +0.659 |
Neg logit contributions | |
Fact | -0.728 |
itability | -0.699 |
weights | -0.680 |
amus | -0.678 |
quished | -0.643 |
Token | lot |
Token ID | 1256 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
Fact | +0.237 |
itability | +0.228 |
amus | +0.220 |
weights | +0.220 |
campaigner | +0.206 |
Neg logit contributions | |
Soup | -0.272 |
esville | -0.251 |
Uniform | -0.242 |
eers | -0.241 |
icho | -0.240 |
Token | of |
Token ID | 286 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
Fact | +1.623 |
itability | +1.555 |
weights | +1.515 |
amus | +1.513 |
campaigner | +1.415 |
Neg logit contributions | |
Soup | -1.808 |
esville | -1.655 |
Uniform | -1.605 |
icho | -1.599 |
eers | -1.591 |
Token | other |
Token ID | 584 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
Fact | +1.788 |
itability | +1.729 |
weights | +1.682 |
amus | +1.667 |
campaigner | +1.572 |
Neg logit contributions | |
Soup | -2.134 |
esville | -1.973 |
icho | -1.905 |
Uniform | -1.896 |
efe | -1.881 |
Token | ways |
Token ID | 2842 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
Fact | +1.852 |
itability | +1.752 |
weights | +1.732 |
amus | +1.698 |
campaigner | +1.606 |
Neg logit contributions | |
Soup | -2.114 |
esville | -1.949 |
Uniform | -1.877 |
icho | -1.868 |
eers | -1.852 |
Token | to |
Token ID | 284 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
Fact | +2.648 |
itability | +2.526 |
amus | +2.469 |
weights | +2.460 |
ysc | +2.308 |
Neg logit contributions | |
Soup | -2.773 |
esville | -2.535 |
Uniform | -2.464 |
efe | -2.442 |
eers | -2.424 |
Token | use |
Token ID | 779 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
Fact | +0.537 |
itability | +0.516 |
weights | +0.500 |
amus | +0.500 |
ysc | +0.468 |
Neg logit contributions | |
Soup | -0.573 |
esville | -0.526 |
Uniform | -0.510 |
eers | -0.507 |
icho | -0.504 |
Token | cannabis |
Token ID | 9499 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
Fact | +2.663 |
itability | +2.551 |
amus | +2.487 |
weights | +2.477 |
anger | +2.310 |
Neg logit contributions | |
Soup | -3.242 |
esville | -2.969 |
Uniform | -2.883 |
icho | -2.872 |
eers | -2.862 |
Token | . |
Token ID | 13 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
Fact | +2.159 |
itability | +2.077 |
weights | +2.040 |
amus | +2.036 |
campaigner | +1.930 |
Neg logit contributions | |
Soup | -1.580 |
esville | -1.403 |
Uniform | -1.360 |
eers | -1.346 |
icho | -1.344 |
Token | Ed |
Token ID | 1717 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
Fact | +0.401 |
itability | +0.379 |
weights | +0.370 |
amus | +0.369 |
ysc | +0.344 |
Neg logit contributions | |
Soup | -0.464 |
esville | -0.431 |
eers | -0.415 |
icho | -0.414 |
Uniform | -0.414 |
Token | ibles |
Token ID | 18764 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
Soup | +1.348 |
esville | +1.202 |
Uniform | +1.147 |
eers | +1.136 |
icho | +1.134 |
Neg logit contributions | |
Fact | -2.094 |
itability | -2.016 |
weights | -1.979 |
amus | -1.965 |
quished | -1.898 |
Token | said |
Token ID | 531 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
Fact | +2.940 |
itability | +2.776 |
amus | +2.750 |
weights | +2.731 |
campaigner | +2.617 |
Neg logit contributions | |
Soup | -3.183 |
esville | -2.918 |
Uniform | -2.857 |
icho | -2.843 |
eers | -2.802 |
Token | . |
Token ID | 13 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
Soup | +0.317 |
esville | +0.283 |
Uniform | +0.272 |
eers | +0.270 |
icho | +0.267 |
Neg logit contributions | |
Fact | -0.474 |
itability | -0.459 |
weights | -0.448 |
amus | -0.447 |
quished | -0.427 |
Token | " |
Token ID | 366 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
Soup | +2.324 |
esville | +2.066 |
Uniform | +2.011 |
Albion | +1.974 |
icho | +1.967 |
Neg logit contributions | |
Fact | -3.347 |
itability | -3.254 |
weights | -3.193 |
amus | -3.171 |
quished | -3.013 |
Token | They |
Token ID | 2990 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
Soup | +0.212 |
esville | +0.193 |
Uniform | +0.188 |
eers | +0.187 |
Albion | +0.185 |
Neg logit contributions | |
Fact | -0.205 |
itability | -0.199 |
weights | -0.193 |
amus | -0.192 |
quished | -0.183 |
Token | don |
Token ID | 836 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Fact | +0.302 |
itability | +0.290 |
weights | +0.282 |
amus | +0.281 |
quished | +0.264 |
Neg logit contributions | |
Soup | -0.337 |
esville | -0.310 |
Uniform | -0.301 |
eers | -0.299 |
icho | -0.297 |
Token | 't |
Token ID | 470 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Soup | +0.443 |
esville | +0.374 |
Uniform | +0.353 |
eers | +0.346 |
icho | +0.345 |
Neg logit contributions | |
Fact | -1.146 |
itability | -1.109 |
weights | -1.091 |
amus | -1.091 |
anger | -1.045 |
Token | appreciate |
Token ID | 9144 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
Fact | +0.715 |
itability | +0.686 |
weights | +0.663 |
amus | +0.662 |
campaigner | +0.620 |
Neg logit contributions | |
Soup | -0.870 |
esville | -0.803 |
Uniform | -0.778 |
eers | -0.776 |
icho | -0.772 |
Token | the |
Token ID | 262 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
Soup | +0.570 |
esville | +0.521 |
Uniform | +0.506 |
eers | +0.503 |
Albion | +0.499 |
Neg logit contributions | |
Fact | -0.547 |
itability | -0.525 |
weights | -0.511 |
amus | -0.510 |
quished | -0.486 |
Token | fear |
Token ID | 3252 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Fact | +0.271 |
itability | +0.260 |
weights | +0.252 |
amus | +0.251 |
campaigner | +0.236 |
Neg logit contributions | |
Soup | -0.289 |
esville | -0.266 |
Uniform | -0.257 |
eers | -0.256 |
icho | -0.255 |
Token | m |
Token ID | 285 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
Fact | +1.451 |
itability | +1.392 |
weights | +1.359 |
amus | +1.353 |
campaigner | +1.272 |
Neg logit contributions | |
Soup | -1.505 |
esville | -1.377 |
Uniform | -1.341 |
eers | -1.327 |
icho | -1.322 |
Token | ong |
Token ID | 506 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
Fact | +1.712 |
itability | +1.654 |
amus | +1.607 |
weights | +1.601 |
anger | +1.543 |
Neg logit contributions | |
Soup | -1.915 |
Uniform | -1.725 |
esville | -1.709 |
XIV | -1.678 |
Albion | -1.667 |
Token | I |
Token ID | 314 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
Fact | +2.233 |
itability | +2.122 |
weights | +2.030 |
amus | +2.018 |
campaigner | +1.931 |
Neg logit contributions | |
Soup | -2.820 |
esville | -2.642 |
eers | -2.572 |
Uniform | -2.534 |
icho | -2.527 |
Token | 'll |
Token ID | 1183 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
Fact | +3.572 |
itability | +3.444 |
amus | +3.408 |
weights | +3.369 |
campaigner | +3.252 |
Neg logit contributions | |
Soup | -2.453 |
esville | -2.217 |
eers | -2.164 |
icho | -2.122 |
Uniform | -2.096 |
Token | make |
Token ID | 787 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
Fact | +0.670 |
itability | +0.621 |
weights | +0.587 |
amus | +0.581 |
quished | +0.516 |
Neg logit contributions | |
Soup | -2.038 |
esville | -1.927 |
Uniform | -1.881 |
eers | -1.873 |
icho | -1.864 |
Token | the |
Token ID | 262 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
Fact | +4.367 |
itability | +4.164 |
weights | +4.094 |
amus | +4.032 |
quished | +3.857 |
Neg logit contributions | |
Soup | -3.934 |
esville | -3.692 |
eers | -3.547 |
Albion | -3.446 |
Arche | -3.430 |
Token | large |
Token ID | 1588 |
Feature activation | -4.86e+00 |
Pos logit contributions | |
Fact | +2.084 |
itability | +1.998 |
weights | +1.929 |
campaigner | +1.872 |
amus | +1.868 |
Neg logit contributions | |
Soup | -2.338 |
esville | -2.189 |
eers | -2.143 |
Uniform | -2.108 |
icho | -2.090 |
Token | corporations |
Token ID | 10225 |
Feature activation | -7.52e+00 |
Pos logit contributions | |
Fact | +3.014 |
itability | +2.936 |
weights | +2.874 |
amus | +2.671 |
campaigner | +2.641 |
Neg logit contributions | |
Soup | -3.711 |
esville | -3.622 |
eers | -3.609 |
Uniform | -3.486 |
Albion | -3.479 |
Token | pay |
Token ID | 1414 |
Feature activation | -6.05e+00 |
Pos logit contributions | |
Fact | +5.342 |
itability | +5.215 |
weights | +5.086 |
campaigner | +4.945 |
quished | +4.856 |
Neg logit contributions | |
eers | -4.876 |
Soup | -4.839 |
esville | -4.762 |
icho | -4.492 |
Uniform | -4.407 |
Token | ' |
Token ID | 705 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
Fact | +3.925 |
itability | +3.680 |
weights | +3.612 |
amus | +3.506 |
anger | +3.445 |
Neg logit contributions | |
Soup | -4.649 |
esville | -4.405 |
icho | -4.341 |
eers | -4.182 |
Albion | -4.181 |
Token | fair |
Token ID | 22043 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
Fact | +3.895 |
itability | +3.739 |
weights | +3.712 |
amus | +3.657 |
anger | +3.477 |
Neg logit contributions | |
Soup | -2.727 |
esville | -2.409 |
Albion | -2.302 |
icho | -2.296 |
Uniform | -2.267 |
Token | share |
Token ID | 2648 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
Fact | +1.553 |
itability | +1.471 |
weights | +1.426 |
amus | +1.400 |
anger | +1.315 |
Neg logit contributions | |
Soup | -2.331 |
esville | -2.176 |
icho | -2.125 |
Uniform | -2.123 |
Albion | -2.106 |
Token | of |
Token ID | 286 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
Fact | +3.403 |
itability | +3.217 |
amus | +3.135 |
weights | +3.131 |
campaigner | +2.939 |
Neg logit contributions | |
Soup | -4.250 |
esville | -3.882 |
Uniform | -3.835 |
eers | -3.797 |
icho | -3.796 |
Token | � |
Token ID | 447 |
Feature activation | +7.29e-01 |
Pos logit contributions | |
Soup | +1.383 |
esville | +1.246 |
Uniform | +1.225 |
Albion | +1.195 |
eers | +1.195 |
Neg logit contributions | |
Fact | -1.446 |
itability | -1.412 |
amus | -1.370 |
weights | -1.370 |
quished | -1.293 |
Token | � |
Token ID | 247 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
Soup | +0.492 |
esville | +0.448 |
Uniform | +0.433 |
eers | +0.430 |
icho | +0.427 |
Neg logit contributions | |
Fact | -0.547 |
itability | -0.528 |
weights | -0.516 |
amus | -0.513 |
quished | -0.487 |
Token |
|
Token ID | 198 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
Fact | +0.256 |
itability | +0.245 |
weights | +0.239 |
amus | +0.238 |
quished | +0.225 |
Neg logit contributions | |
Soup | -0.254 |
esville | -0.233 |
Uniform | -0.225 |
eers | -0.224 |
icho | -0.222 |
Token |
|
Token ID | 198 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Soup | +1.484 |
esville | +1.355 |
Uniform | +1.309 |
eers | +1.302 |
icho | +1.289 |
Neg logit contributions | |
Fact | -1.574 |
itability | -1.518 |
weights | -1.477 |
amus | -1.473 |
quished | -1.401 |
Token | Other |
Token ID | 6395 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
Soup | +1.049 |
esville | +0.958 |
eers | +0.924 |
Uniform | +0.923 |
icho | +0.913 |
Neg logit contributions | |
Fact | -1.035 |
itability | -1.021 |
amus | -0.986 |
weights | -0.984 |
quished | -0.934 |
Token | executions |
Token ID | 30632 |
Feature activation | -6.54e+00 |
Pos logit contributions | |
Fact | +2.184 |
itability | +2.081 |
weights | +2.067 |
amus | +2.004 |
campaigner | +1.933 |
Neg logit contributions | |
Soup | -2.419 |
esville | -2.264 |
eers | -2.168 |
efe | -2.157 |
Uniform | -2.148 |
Token | took |
Token ID | 1718 |
Feature activation | -5.99e+00 |
Pos logit contributions | |
Fact | +4.121 |
itability | +3.920 |
weights | +3.867 |
amus | +3.783 |
campaigner | +3.716 |
Neg logit contributions | |
Soup | -5.025 |
esville | -4.574 |
icho | -4.462 |
efe | -4.436 |
Uniform | -4.430 |
Token | place |
Token ID | 1295 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
Fact | +2.109 |
weights | +1.956 |
itability | +1.925 |
amus | +1.836 |
campaigner | +1.641 |
Neg logit contributions | |
Soup | -6.292 |
esville | -5.932 |
icho | -5.827 |
Uniform | -5.798 |
eers | -5.785 |
Token | in |
Token ID | 287 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
Fact | +2.200 |
itability | +2.110 |
weights | +2.069 |
amus | +2.041 |
ysc | +1.959 |
Neg logit contributions | |
Soup | -1.980 |
esville | -1.787 |
Uniform | -1.729 |
eers | -1.729 |
efe | -1.705 |
Token | Chong |
Token ID | 46892 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Fact | +0.082 |
itability | +0.079 |
weights | +0.077 |
amus | +0.077 |
quished | +0.072 |
Neg logit contributions | |
Soup | -0.083 |
esville | -0.076 |
Uniform | -0.074 |
eers | -0.073 |
icho | -0.073 |
Token | jin |
Token ID | 18594 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
Soup | +1.056 |
esville | +0.957 |
Uniform | +0.924 |
eers | +0.916 |
icho | +0.915 |
Neg logit contributions | |
Fact | -1.202 |
itability | -1.156 |
weights | -1.127 |
amus | -1.124 |
quished | -1.072 |
Token | [ |
Token ID | 685 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
Fact | +0.501 |
itability | +0.481 |
weights | +0.468 |
amus | +0.467 |
quished | +0.441 |
Neg logit contributions | |
Soup | -0.519 |
esville | -0.476 |
Uniform | -0.460 |
eers | -0.458 |
icho | -0.454 |
Token | 10 |
Token ID | 940 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
Fact | +0.284 |
itability | +0.273 |
weights | +0.265 |
amus | +0.264 |
quished | +0.250 |
Neg logit contributions | |
Soup | -0.307 |
esville | -0.282 |
Uniform | -0.273 |
eers | -0.271 |
icho | -0.270 |
Token | : |
Token ID | 25 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
Fact | +1.714 |
itability | +1.633 |
weights | +1.612 |
amus | +1.609 |
anger | +1.525 |
Neg logit contributions | |
Soup | -1.277 |
esville | -1.141 |
Uniform | -1.104 |
eers | -1.095 |
icho | -1.089 |
Token | 47 |
Token ID | 2857 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
Fact | +1.068 |
itability | +0.996 |
weights | +0.980 |
amus | +0.977 |
anger | +0.912 |
Neg logit contributions | |
Soup | -1.265 |
esville | -1.172 |
Uniform | -1.134 |
eers | -1.130 |
icho | -1.125 |
Token | ] |
Token ID | 60 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
Fact | +1.571 |
itability | +1.499 |
weights | +1.477 |
amus | +1.476 |
anger | +1.394 |
Neg logit contributions | |
Soup | -1.289 |
esville | -1.178 |
eers | -1.128 |
Uniform | -1.128 |
icho | -1.115 |
Token | < |
Token ID | 1279 |
Feature activation | -7.67e+00 |
Pos logit contributions | |
Fact | +5.041 |
itability | +4.825 |
amus | +4.759 |
weights | +4.746 |
anger | +4.572 |
Neg logit contributions | |
Soup | -2.317 |
esville | -2.045 |
eers | -1.918 |
Uniform | -1.912 |
icho | -1.853 |
Token | met |
Token ID | 4164 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
Fact | +5.670 |
amus | +5.380 |
itability | +5.243 |
weights | +5.184 |
anger | +5.062 |
Neg logit contributions | |
Soup | -5.149 |
Uniform | -4.652 |
esville | -4.647 |
Albion | -4.477 |
eers | -4.449 |
Token | ros |
Token ID | 4951 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
Fact | +1.416 |
itability | +1.366 |
amus | +1.332 |
weights | +1.326 |
ysc | +1.258 |
Neg logit contributions | |
Soup | -1.377 |
esville | -1.250 |
Uniform | -1.219 |
Albion | -1.196 |
eers | -1.196 |
Token | xual |
Token ID | 5541 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
Fact | +3.011 |
itability | +2.887 |
amus | +2.841 |
weights | +2.813 |
ysc | +2.685 |
Neg logit contributions | |
Soup | -2.746 |
esville | -2.498 |
Uniform | -2.421 |
Albion | -2.387 |
eers | -2.376 |
Token | th |
Token ID | 400 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
Fact | +0.575 |
itability | +0.555 |
weights | +0.542 |
amus | +0.541 |
quished | +0.515 |
Neg logit contributions | |
Soup | -0.439 |
esville | -0.396 |
Uniform | -0.381 |
eers | -0.378 |
icho | -0.375 |
Token | ug |
Token ID | 1018 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
Fact | +3.046 |
itability | +2.918 |
amus | +2.898 |
weights | +2.845 |
ysc | +2.744 |
Neg logit contributions | |
Soup | -2.863 |
esville | -2.609 |
Uniform | -2.534 |
Albion | -2.507 |
eers | -2.506 |
Token | J |
Token ID | 41 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
Fact | +5.188 |
itability | +4.870 |
amus | +4.813 |
weights | +4.783 |
ysc | +4.588 |
Neg logit contributions | |
Soup | -5.157 |
esville | -4.666 |
Uniform | -4.612 |
Albion | -4.525 |
eers | -4.456 |
Token | m |
Token ID | 76 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
Fact | +2.333 |
itability | +2.211 |
amus | +2.159 |
weights | +2.157 |
ysc | +2.038 |
Neg logit contributions | |
Soup | -2.561 |
esville | -2.344 |
Uniform | -2.289 |
eers | -2.258 |
Albion | -2.255 |
Token | B |
Token ID | 33 |
Feature activation | -7.56e+00 |
Pos logit contributions | |
Fact | +6.897 |
itability | +6.376 |
weights | +6.270 |
amus | +6.269 |
ysc | +6.060 |
Neg logit contributions | |
Soup | -7.575 |
esville | -6.896 |
Uniform | -6.836 |
Albion | -6.730 |
eers | -6.631 |
Token | z |
Token ID | 89 |
Feature activation | -6.75e+00 |
Pos logit contributions | |
Fact | +5.043 |
itability | +4.699 |
weights | +4.622 |
amus | +4.621 |
anger | +4.426 |
Neg logit contributions | |
Soup | -5.666 |
esville | -5.116 |
Uniform | -5.058 |
Albion | -5.005 |
eers | -4.941 |
Token | 5 |
Token ID | 20 |
Feature activation | -9.67e+00 |
Pos logit contributions | |
Fact | +4.599 |
itability | +4.292 |
amus | +4.232 |
weights | +4.225 |
ysc | +4.024 |
Neg logit contributions | |
Soup | -4.982 |
esville | -4.553 |
Uniform | -4.475 |
Albion | -4.397 |
eers | -4.365 |
Token | s |
Token ID | 82 |
Feature activation | -8.55e+00 |
Pos logit contributions | |
Fact | +6.836 |
itability | +6.364 |
amus | +6.344 |
weights | +6.279 |
ysc | +6.052 |
Neg logit contributions | |
Soup | -6.581 |
Uniform | -5.900 |
esville | -5.898 |
Albion | -5.789 |
Arche | -5.704 |
Token | 8 |
Token ID | 23 |
Feature activation | -7.81e+00 |
Pos logit contributions | |
Fact | +5.707 |
itability | +5.267 |
amus | +5.217 |
weights | +5.180 |
anger | +4.948 |
Neg logit contributions | |
Soup | -6.303 |
esville | -5.739 |
Uniform | -5.686 |
eers | -5.599 |
Albion | -5.597 |
Token | — |
Token ID | 851 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
Fact | +5.271 |
itability | +4.919 |
amus | +4.855 |
weights | +4.812 |
anger | +4.549 |
Neg logit contributions | |
Soup | -5.614 |
esville | -5.177 |
icho | -5.032 |
eers | -4.985 |
Uniform | -4.952 |
Token | General |
Token ID | 3611 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
Soup | +0.393 |
esville | +0.359 |
Uniform | +0.352 |
eers | +0.346 |
Albion | +0.344 |
Neg logit contributions | |
Fact | -0.365 |
itability | -0.352 |
weights | -0.341 |
amus | -0.339 |
quished | -0.322 |
Token | Flynn |
Token ID | 14654 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
Soup | +2.933 |
esville | +2.671 |
Uniform | +2.638 |
efe | +2.565 |
eers | +2.560 |
Neg logit contributions | |
Fact | -2.291 |
itability | -2.250 |
weights | -2.167 |
amus | -2.136 |
ysc | -2.028 |
Token | (@ |
Token ID | 4275 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
Fact | +4.203 |
itability | +3.998 |
amus | +3.929 |
weights | +3.913 |
anger | +3.757 |
Neg logit contributions | |
Soup | -2.855 |
esville | -2.627 |
eers | -2.513 |
icho | -2.489 |
Albion | -2.464 |
Token | was |
Token ID | 373 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Soup | +1.355 |
esville | +1.236 |
Uniform | +1.191 |
eers | +1.186 |
Albion | +1.172 |
Neg logit contributions | |
Fact | -1.570 |
itability | -1.515 |
weights | -1.474 |
amus | -1.461 |
quished | -1.411 |
Token | coerced |
Token ID | 48474 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
Soup | +0.815 |
esville | +0.741 |
Uniform | +0.715 |
eers | +0.710 |
icho | +0.705 |
Neg logit contributions | |
Fact | -0.930 |
itability | -0.896 |
weights | -0.874 |
amus | -0.869 |
quished | -0.828 |
Token | into |
Token ID | 656 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
Fact | +0.193 |
itability | +0.187 |
weights | +0.182 |
amus | +0.182 |
quished | +0.174 |
Neg logit contributions | |
Soup | -0.134 |
esville | -0.119 |
Uniform | -0.115 |
eers | -0.114 |
icho | -0.113 |
Token | filing |
Token ID | 12180 |
Feature activation | -6.23e+00 |
Pos logit contributions | |
Fact | +2.549 |
itability | +2.420 |
amus | +2.352 |
weights | +2.308 |
campaigner | +2.218 |
Neg logit contributions | |
Soup | -3.589 |
esville | -3.333 |
eers | -3.279 |
icho | -3.239 |
efe | -3.238 |
Token | a |
Token ID | 257 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
Fact | +4.115 |
itability | +3.895 |
amus | +3.769 |
weights | +3.719 |
anger | +3.578 |
Neg logit contributions | |
Soup | -4.657 |
esville | -4.258 |
eers | -4.191 |
icho | -4.185 |
Albion | -4.110 |
Token | police |
Token ID | 1644 |
Feature activation | -6.05e+00 |
Pos logit contributions | |
Fact | +1.108 |
itability | +1.060 |
amus | +1.025 |
weights | +1.020 |
anger | +0.960 |
Neg logit contributions | |
Soup | -1.282 |
esville | -1.182 |
eers | -1.146 |
icho | -1.137 |
Uniform | -1.136 |
Token | complaint |
Token ID | 8224 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
Fact | +2.638 |
itability | +2.411 |
amus | +2.258 |
weights | +2.222 |
anger | +2.098 |
Neg logit contributions | |
Soup | -5.989 |
esville | -5.626 |
eers | -5.516 |
icho | -5.471 |
Albion | -5.426 |
Token | . |
Token ID | 13 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
Fact | +2.907 |
itability | +2.794 |
amus | +2.740 |
weights | +2.708 |
campaigner | +2.576 |
Neg logit contributions | |
Soup | -2.642 |
esville | -2.415 |
icho | -2.327 |
eers | -2.318 |
Uniform | -2.309 |
Token | She |
Token ID | 1375 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Fact | +0.516 |
itability | +0.494 |
weights | +0.481 |
amus | +0.480 |
quished | +0.453 |
Neg logit contributions | |
Soup | -0.528 |
esville | -0.485 |
Uniform | -0.468 |
eers | -0.467 |
icho | -0.463 |
Token | stated |
Token ID | 5081 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
Fact | +0.839 |
itability | +0.805 |
amus | +0.786 |
weights | +0.785 |
anger | +0.737 |
Neg logit contributions | |
Soup | -0.863 |
esville | -0.788 |
Uniform | -0.764 |
eers | -0.759 |
icho | -0.757 |
Token | they |
Token ID | 484 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
Fact | +2.069 |
itability | +1.976 |
weights | +1.912 |
amus | +1.894 |
anger | +1.796 |
Neg logit contributions | |
Soup | -2.485 |
esville | -2.292 |
eers | -2.220 |
Uniform | -2.214 |
icho | -2.203 |