Token |
|
Token ID | 198 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
@@ | +0.148 |
Lie | +0.141 |
� | +0.137 |
Krug | +0.136 |
Lt | +0.135 |
Neg logit contributions | |
anson | -0.206 |
rem | -0.173 |
� | -0.166 |
roman | -0.153 |
lement | -0.153 |
Token | pur |
Token ID | 14225 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
@@ | +0.065 |
� | +0.062 |
Lie | +0.062 |
Krug | +0.061 |
� | +0.060 |
Neg logit contributions | |
anson | -0.097 |
rem | -0.083 |
� | -0.080 |
lement | -0.073 |
roman | -0.073 |
Token | ak |
Token ID | 461 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
@@ | +0.083 |
� | +0.080 |
Krug | +0.079 |
Lie | +0.079 |
crossings | +0.078 |
Neg logit contributions | |
anson | -0.096 |
rem | -0.082 |
� | -0.077 |
roman | -0.070 |
lement | -0.070 |
Token | ushi |
Token ID | 17731 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
anson | +0.132 |
rem | +0.113 |
� | +0.107 |
lement | +0.098 |
roman | +0.098 |
Neg logit contributions | |
@@ | -0.096 |
� | -0.091 |
Krug | -0.091 |
Lie | -0.090 |
crossings | -0.089 |
Token | Profile |
Token ID | 13118 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
@@ | +0.421 |
� | +0.397 |
Krug | +0.395 |
Lie | +0.391 |
Nemesis | +0.385 |
Neg logit contributions | |
anson | -0.670 |
rem | -0.581 |
� | -0.552 |
lement | -0.510 |
roman | -0.508 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
@@ | +0.447 |
� | +0.402 |
Krug | +0.400 |
Lie | +0.393 |
Nemesis | +0.390 |
Neg logit contributions | |
anson | -1.251 |
rem | -1.117 |
� | -1.070 |
roman | -1.008 |
lement | -1.007 |
Token | August |
Token ID | 2932 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
@@ | +2.381 |
Krug | +2.223 |
Nemesis | +2.210 |
Claim | +2.182 |
Lie | +2.151 |
Neg logit contributions | |
anson | -3.342 |
rem | -2.808 |
� | -2.699 |
uana | -2.506 |
roman | -2.478 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
@@ | +0.923 |
Krug | +0.895 |
Lie | +0.882 |
Nemesis | +0.881 |
Scand | +0.877 |
Neg logit contributions | |
anson | -0.810 |
rem | -0.666 |
� | -0.615 |
lement | -0.552 |
roman | -0.551 |
Token | United |
Token ID | 1578 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
@@ | +0.369 |
Krug | +0.354 |
Lie | +0.349 |
� | +0.347 |
Nemesis | +0.346 |
Neg logit contributions | |
anson | -0.443 |
rem | -0.375 |
� | -0.353 |
roman | -0.323 |
lement | -0.322 |
Token | States |
Token ID | 1829 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
@@ | +0.555 |
� | +0.516 |
Krug | +0.515 |
Lie | +0.509 |
crossings | +0.505 |
Neg logit contributions | |
anson | -1.217 |
rem | -1.080 |
� | -1.030 |
lement | -0.961 |
roman | -0.957 |
Token | 3 |
Token ID | 513 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
@@ | +0.986 |
Krug | +0.926 |
Nemesis | +0.912 |
Lie | +0.905 |
COUN | +0.902 |
Neg logit contributions | |
anson | -1.915 |
rem | -1.665 |
� | -1.605 |
roman | -1.487 |
lement | -1.479 |
Token |
|
Token ID | 198 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
@@ | +0.320 |
Lie | +0.299 |
Krug | +0.294 |
Nemesis | +0.288 |
Lt | +0.286 |
Neg logit contributions | |
anson | -0.477 |
rem | -0.400 |
� | -0.385 |
roman | -0.357 |
lement | -0.355 |
Token | Dark |
Token ID | 17367 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
@@ | +0.243 |
� | +0.231 |
Lie | +0.231 |
Krug | +0.230 |
� | +0.225 |
Neg logit contributions | |
anson | -0.365 |
rem | -0.315 |
� | -0.300 |
lement | -0.276 |
roman | -0.274 |
Token | horse |
Token ID | 30527 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
@@ | +0.608 |
� | +0.580 |
Lie | +0.577 |
Krug | +0.577 |
crossings | +0.564 |
Neg logit contributions | |
anson | -0.882 |
rem | -0.762 |
� | -0.724 |
lement | -0.663 |
roman | -0.660 |
Token | Profile |
Token ID | 13118 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
anson | +0.579 |
rem | +0.505 |
� | +0.477 |
lement | +0.438 |
roman | +0.437 |
Neg logit contributions | |
@@ | -0.389 |
Krug | -0.367 |
� | -0.367 |
Lie | -0.362 |
crossings | -0.362 |
Token | Blog |
Token ID | 14001 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
@@ | +0.601 |
� | +0.556 |
Krug | +0.556 |
Lie | +0.549 |
crossings | +0.541 |
Neg logit contributions | |
anson | -1.384 |
rem | -1.230 |
� | -1.173 |
lement | -1.095 |
roman | -1.093 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
@@ | +0.861 |
Krug | +0.740 |
Lie | +0.715 |
Claim | +0.713 |
Nemesis | +0.706 |
Neg logit contributions | |
anson | -3.536 |
rem | -3.147 |
� | -3.076 |
lement | -2.902 |
heng | -2.888 |
Token | December |
Token ID | 3426 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
@@ | +2.333 |
Krug | +2.229 |
Nemesis | +2.208 |
Scand | +2.174 |
Claim | +2.169 |
Neg logit contributions | |
anson | -3.305 |
rem | -2.771 |
� | -2.651 |
uana | -2.466 |
roman | -2.464 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
@@ | +1.010 |
Krug | +0.988 |
Scand | +0.969 |
Nemesis | +0.969 |
Lie | +0.966 |
Neg logit contributions | |
anson | -0.955 |
rem | -0.780 |
� | -0.731 |
roman | -0.662 |
uana | -0.661 |
Token | United |
Token ID | 1578 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
@@ | +0.607 |
Krug | +0.585 |
Lie | +0.578 |
� | +0.575 |
Nemesis | +0.574 |
Neg logit contributions | |
anson | -0.709 |
rem | -0.597 |
� | -0.561 |
roman | -0.514 |
lement | -0.512 |
Token | States |
Token ID | 1829 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
@@ | +0.569 |
� | +0.531 |
Krug | +0.527 |
crossings | +0.523 |
Lie | +0.522 |
Neg logit contributions | |
anson | -1.163 |
rem | -1.032 |
� | -0.981 |
lement | -0.917 |
roman | -0.908 |
Token | 228 |
Token ID | 29041 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
@@ | +0.942 |
Krug | +0.889 |
Nemesis | +0.878 |
Lie | +0.863 |
COUN | +0.863 |
Neg logit contributions | |
anson | -1.859 |
rem | -1.612 |
� | -1.564 |
roman | -1.446 |
uana | -1.437 |
Token | J |
Token ID | 41 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
@@ | +0.261 |
Lie | +0.249 |
� | +0.249 |
Krug | +0.247 |
� | +0.242 |
Neg logit contributions | |
anson | -0.385 |
rem | -0.331 |
� | -0.315 |
lement | -0.290 |
roman | -0.288 |
Token | aaaa |
Token ID | 24794 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
anson | +0.074 |
rem | +0.065 |
� | +0.061 |
lement | +0.056 |
roman | +0.056 |
Neg logit contributions | |
@@ | -0.053 |
� | -0.050 |
Lie | -0.050 |
Krug | -0.050 |
� | -0.049 |
Token | asper |
Token ID | 32981 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
anson | +0.858 |
rem | +0.738 |
� | +0.695 |
lement | +0.633 |
roman | +0.633 |
Neg logit contributions | |
@@ | -0.677 |
� | -0.643 |
Krug | -0.642 |
Lie | -0.638 |
crossings | -0.632 |
Token | Profile |
Token ID | 13118 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
@@ | +0.483 |
� | +0.457 |
Krug | +0.455 |
Lie | +0.454 |
crossings | +0.445 |
Neg logit contributions | |
anson | -0.730 |
rem | -0.634 |
� | -0.603 |
lement | -0.553 |
roman | -0.553 |
Token | Blog |
Token ID | 14001 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
@@ | +0.429 |
� | +0.394 |
Krug | +0.393 |
Lie | +0.388 |
Nemesis | +0.380 |
Neg logit contributions | |
anson | -1.070 |
rem | -0.953 |
� | -0.911 |
roman | -0.853 |
lement | -0.853 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
@@ | +0.801 |
Krug | +0.715 |
Lie | +0.668 |
Nemesis | +0.667 |
Claim | +0.663 |
Neg logit contributions | |
anson | -3.261 |
rem | -2.911 |
� | -2.830 |
lement | -2.672 |
uana | -2.667 |
Token | April |
Token ID | 3035 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
@@ | +2.325 |
Krug | +2.210 |
Nemesis | +2.187 |
Claim | +2.156 |
Scand | +2.154 |
Neg logit contributions | |
anson | -3.309 |
rem | -2.775 |
� | -2.650 |
uana | -2.467 |
roman | -2.461 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
@@ | +0.792 |
Krug | +0.773 |
Scand | +0.762 |
Nemesis | +0.760 |
Lie | +0.756 |
Neg logit contributions | |
anson | -0.779 |
rem | -0.637 |
� | -0.599 |
roman | -0.547 |
lement | -0.539 |
Token | United |
Token ID | 1578 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
@@ | +0.283 |
Krug | +0.272 |
Lie | +0.269 |
� | +0.266 |
Nemesis | +0.266 |
Neg logit contributions | |
anson | -0.340 |
rem | -0.287 |
� | -0.270 |
roman | -0.248 |
lement | -0.246 |
Token | States |
Token ID | 1829 |
Feature activation | +6.78e-01 |
Pos logit contributions | |
@@ | +0.593 |
� | +0.555 |
Krug | +0.549 |
crossings | +0.545 |
Lie | +0.544 |
Neg logit contributions | |
anson | -1.189 |
rem | -1.053 |
� | -0.999 |
lement | -0.936 |
roman | -0.927 |
Token | 5 |
Token ID | 642 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
@@ | +0.925 |
Krug | +0.871 |
Nemesis | +0.857 |
Lie | +0.846 |
COUN | +0.845 |
Neg logit contributions | |
anson | -1.820 |
rem | -1.579 |
� | -1.529 |
roman | -1.418 |
lement | -1.406 |
Token |
|
Token ID | 198 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
@@ | +0.663 |
Lie | +0.620 |
Krug | +0.601 |
Lt | +0.593 |
territory | +0.590 |
Neg logit contributions | |
anson | -1.047 |
rem | -0.871 |
� | -0.844 |
roman | -0.788 |
uana | -0.786 |
Token | re |
Token ID | 260 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
@@ | +0.474 |
Lie | +0.454 |
� | +0.453 |
Krug | +0.449 |
� | +0.440 |
Neg logit contributions | |
anson | -0.716 |
rem | -0.617 |
� | -0.588 |
lement | -0.541 |
roman | -0.536 |
Token | i |
Token ID | 72 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
@@ | +0.658 |
� | +0.629 |
Krug | +0.627 |
Lie | +0.626 |
crossings | +0.613 |
Neg logit contributions | |
anson | -0.856 |
rem | -0.733 |
� | -0.692 |
roman | -0.633 |
lement | -0.632 |
Token | Profile |
Token ID | 13118 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
anson | +0.258 |
rem | +0.224 |
� | +0.212 |
lement | +0.195 |
roman | +0.194 |
Neg logit contributions | |
@@ | -0.175 |
� | -0.166 |
Krug | -0.165 |
Lie | -0.164 |
crossings | -0.162 |
Token | Blog |
Token ID | 14001 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
@@ | +0.711 |
� | +0.649 |
Krug | +0.648 |
Lie | +0.640 |
Nemesis | +0.628 |
Neg logit contributions | |
anson | -1.809 |
rem | -1.613 |
� | -1.541 |
roman | -1.448 |
lement | -1.443 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
@@ | +0.788 |
Krug | +0.705 |
Lie | +0.659 |
Nemesis | +0.648 |
Claim | +0.647 |
Neg logit contributions | |
anson | -3.259 |
rem | -2.907 |
� | -2.837 |
roman | -2.679 |
lement | -2.668 |
Token | October |
Token ID | 3267 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
@@ | +2.362 |
Krug | +2.233 |
Nemesis | +2.220 |
Claim | +2.188 |
Lie | +2.175 |
Neg logit contributions | |
anson | -3.262 |
rem | -2.738 |
� | -2.617 |
roman | -2.421 |
uana | -2.418 |
Token | 2002 |
Token ID | 6244 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
@@ | +1.064 |
Krug | +1.032 |
Lie | +1.018 |
Scand | +1.017 |
� | +1.012 |
Neg logit contributions | |
anson | -0.945 |
rem | -0.782 |
� | -0.721 |
lement | -0.645 |
roman | -0.645 |
Token | United |
Token ID | 1578 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
@@ | +0.270 |
Krug | +0.259 |
Lie | +0.256 |
� | +0.254 |
Nemesis | +0.253 |
Neg logit contributions | |
anson | -0.324 |
rem | -0.274 |
� | -0.258 |
roman | -0.235 |
lement | -0.235 |
Token | States |
Token ID | 1829 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
@@ | +0.480 |
� | +0.448 |
Krug | +0.445 |
crossings | +0.441 |
Lie | +0.440 |
Neg logit contributions | |
anson | -0.957 |
rem | -0.849 |
� | -0.806 |
lement | -0.754 |
roman | -0.747 |
Token | 3 |
Token ID | 513 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
@@ | +1.010 |
Krug | +0.946 |
Nemesis | +0.932 |
Lie | +0.923 |
COUN | +0.921 |
Neg logit contributions | |
anson | -1.982 |
rem | -1.723 |
� | -1.668 |
roman | -1.543 |
lement | -1.533 |
Token | TB |
Token ID | 22737 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
@@ | +0.395 |
Lie | +0.379 |
� | +0.378 |
Krug | +0.375 |
� | +0.367 |
Neg logit contributions | |
anson | -0.591 |
rem | -0.508 |
� | -0.485 |
lement | -0.445 |
roman | -0.442 |
Token | one |
Token ID | 505 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
@@ | +0.335 |
� | +0.318 |
Krug | +0.316 |
Lie | +0.315 |
� | +0.308 |
Neg logit contributions | |
anson | -0.502 |
rem | -0.434 |
� | -0.413 |
lement | -0.379 |
roman | -0.378 |
Token | - |
Token ID | 12 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
anson | +0.100 |
rem | +0.087 |
� | +0.082 |
lement | +0.075 |
roman | +0.075 |
Neg logit contributions | |
@@ | -0.067 |
� | -0.063 |
Krug | -0.063 |
Lie | -0.063 |
crossings | -0.062 |
Token | Profile |
Token ID | 13118 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
@@ | +0.137 |
� | +0.130 |
Lie | +0.129 |
Krug | +0.128 |
� | +0.126 |
Neg logit contributions | |
anson | -0.212 |
rem | -0.184 |
� | -0.175 |
lement | -0.161 |
roman | -0.160 |
Token | Blog |
Token ID | 14001 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
@@ | +0.565 |
� | +0.521 |
Krug | +0.520 |
Lie | +0.514 |
crossings | +0.504 |
Neg logit contributions | |
anson | -1.327 |
rem | -1.180 |
� | -1.126 |
lement | -1.052 |
roman | -1.052 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
@@ | +0.781 |
Krug | +0.679 |
Nemesis | +0.649 |
Lie | +0.648 |
Claim | +0.642 |
Neg logit contributions | |
anson | -3.230 |
rem | -2.887 |
� | -2.824 |
lement | -2.670 |
roman | -2.652 |
Token | November |
Token ID | 3389 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
@@ | +2.234 |
Krug | +2.137 |
Nemesis | +2.108 |
Scand | +2.087 |
Claim | +2.077 |
Neg logit contributions | |
anson | -3.183 |
rem | -2.665 |
� | -2.541 |
roman | -2.369 |
uana | -2.365 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
@@ | +1.126 |
Krug | +1.099 |
Nemesis | +1.076 |
Lie | +1.076 |
Scand | +1.076 |
Neg logit contributions | |
anson | -1.014 |
rem | -0.828 |
� | -0.770 |
roman | -0.697 |
uana | -0.696 |
Token | United |
Token ID | 1578 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
@@ | +0.218 |
Krug | +0.209 |
� | +0.207 |
Lie | +0.207 |
Nemesis | +0.205 |
Neg logit contributions | |
anson | -0.249 |
rem | -0.211 |
� | -0.198 |
roman | -0.180 |
lement | -0.180 |
Token | States |
Token ID | 1829 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
@@ | +0.580 |
� | +0.541 |
Krug | +0.537 |
Lie | +0.532 |
crossings | +0.531 |
Neg logit contributions | |
anson | -1.186 |
rem | -1.050 |
� | -0.998 |
lement | -0.933 |
roman | -0.925 |
Token | 21 |
Token ID | 2310 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
@@ | +0.940 |
Krug | +0.888 |
Nemesis | +0.875 |
Lie | +0.861 |
Scand | +0.860 |
Neg logit contributions | |
anson | -1.845 |
rem | -1.601 |
� | -1.552 |
roman | -1.437 |
uana | -1.428 |
Token | ada |
Token ID | 4763 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
@@ | +0.124 |
� | +0.117 |
Lie | +0.117 |
Krug | +0.117 |
� | +0.114 |
Neg logit contributions | |
anson | -0.176 |
rem | -0.152 |
� | -0.145 |
lement | -0.132 |
roman | -0.132 |
Token | ver |
Token ID | 332 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
@@ | +0.409 |
� | +0.386 |
Krug | +0.386 |
Lie | +0.385 |
crossings | +0.379 |
Neg logit contributions | |
anson | -0.594 |
rem | -0.513 |
� | -0.485 |
lement | -0.446 |
roman | -0.446 |
Token | BB |
Token ID | 15199 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
anson | +0.380 |
rem | +0.330 |
� | +0.312 |
lement | +0.287 |
roman | +0.286 |
Neg logit contributions | |
@@ | -0.258 |
� | -0.244 |
Krug | -0.243 |
Lie | -0.241 |
crossings | -0.239 |
Token | Profile |
Token ID | 13118 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
@@ | +0.720 |
� | +0.687 |
Krug | +0.678 |
Lie | +0.675 |
� | +0.669 |
Neg logit contributions | |
anson | -1.131 |
rem | -0.979 |
� | -0.934 |
lement | -0.855 |
roman | -0.851 |
Token | Blog |
Token ID | 14001 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
@@ | +0.643 |
� | +0.593 |
Krug | +0.592 |
Lie | +0.584 |
Nemesis | +0.573 |
Neg logit contributions | |
anson | -1.459 |
rem | -1.294 |
� | -1.236 |
lement | -1.155 |
roman | -1.154 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
@@ | +0.774 |
Krug | +0.658 |
Lie | +0.641 |
Claim | +0.638 |
Nemesis | +0.633 |
Neg logit contributions | |
anson | -3.114 |
rem | -2.779 |
� | -2.714 |
lement | -2.563 |
heng | -2.547 |
Token | June |
Token ID | 2795 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
@@ | +2.229 |
Krug | +2.131 |
Nemesis | +2.098 |
Scand | +2.081 |
Claim | +2.068 |
Neg logit contributions | |
anson | -3.145 |
rem | -2.638 |
� | -2.517 |
roman | -2.340 |
uana | -2.327 |
Token | 2009 |
Token ID | 3717 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
@@ | +1.169 |
Krug | +1.141 |
Nemesis | +1.125 |
Scand | +1.122 |
Lie | +1.117 |
Neg logit contributions | |
anson | -1.150 |
rem | -0.938 |
� | -0.882 |
roman | -0.802 |
uana | -0.800 |
Token | Germany |
Token ID | 4486 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
@@ | +0.278 |
Krug | +0.266 |
Lie | +0.264 |
� | +0.263 |
COUN | +0.261 |
Neg logit contributions | |
anson | -0.319 |
rem | -0.269 |
� | -0.253 |
roman | -0.231 |
lement | -0.229 |
Token | 247 |
Token ID | 30179 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
@@ | +0.652 |
Krug | +0.613 |
Nemesis | +0.599 |
Lie | +0.599 |
Scand | +0.596 |
Neg logit contributions | |
anson | -1.269 |
rem | -1.103 |
� | -1.063 |
roman | -0.989 |
lement | -0.982 |
Token | 34 |
Token ID | 2682 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
@@ | +0.431 |
Krug | +0.402 |
� | +0.399 |
Lie | +0.398 |
COUN | +0.394 |
Neg logit contributions | |
anson | -0.672 |
rem | -0.579 |
� | -0.554 |
roman | -0.510 |
lement | -0.507 |
Token | of |
Token ID | 286 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
@@ | +0.344 |
� | +0.325 |
Krug | +0.325 |
Lie | +0.324 |
crossings | +0.317 |
Neg logit contributions | |
anson | -0.567 |
rem | -0.494 |
� | -0.469 |
lement | -0.433 |
roman | -0.432 |
Token | Dal |
Token ID | 12348 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
anson | +0.046 |
rem | +0.040 |
� | +0.037 |
lement | +0.034 |
roman | +0.034 |
Neg logit contributions | |
@@ | -0.036 |
Krug | -0.034 |
� | -0.034 |
Lie | -0.034 |
crossings | -0.033 |
Token | K |
Token ID | 42 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
anson | +0.991 |
rem | +0.864 |
� | +0.820 |
lement | +0.757 |
roman | +0.756 |
Neg logit contributions | |
@@ | -0.609 |
� | -0.574 |
Krug | -0.573 |
Lie | -0.569 |
crossings | -0.565 |
Token | Profile |
Token ID | 13118 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
@@ | +0.075 |
� | +0.071 |
Krug | +0.071 |
Lie | +0.071 |
crossings | +0.070 |
Neg logit contributions | |
anson | -0.100 |
rem | -0.086 |
� | -0.081 |
lement | -0.074 |
roman | -0.074 |
Token | Blog |
Token ID | 14001 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
@@ | +0.655 |
� | +0.603 |
Krug | +0.601 |
Lie | +0.595 |
Nemesis | +0.586 |
Neg logit contributions | |
anson | -1.490 |
rem | -1.321 |
� | -1.261 |
roman | -1.180 |
lement | -1.178 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
@@ | +0.788 |
Krug | +0.696 |
Nemesis | +0.666 |
Lie | +0.662 |
Claim | +0.661 |
Neg logit contributions | |
anson | -3.224 |
rem | -2.866 |
� | -2.802 |
lement | -2.637 |
roman | -2.630 |
Token | June |
Token ID | 2795 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
@@ | +2.198 |
Krug | +2.092 |
Nemesis | +2.072 |
Claim | +2.039 |
Scand | +2.039 |
Neg logit contributions | |
anson | -3.086 |
rem | -2.582 |
� | -2.478 |
uana | -2.298 |
roman | -2.292 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
@@ | +1.080 |
Krug | +1.050 |
Nemesis | +1.037 |
Scand | +1.035 |
Lie | +1.027 |
Neg logit contributions | |
anson | -1.062 |
rem | -0.872 |
� | -0.816 |
roman | -0.744 |
uana | -0.737 |
Token | Canada |
Token ID | 3340 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
@@ | +0.146 |
Krug | +0.140 |
Lie | +0.138 |
� | +0.138 |
Nemesis | +0.137 |
Neg logit contributions | |
anson | -0.169 |
rem | -0.143 |
� | -0.134 |
roman | -0.123 |
lement | -0.122 |
Token | 16 |
Token ID | 1467 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
@@ | +0.298 |
Krug | +0.282 |
Nemesis | +0.275 |
Lie | +0.274 |
COUN | +0.272 |
Neg logit contributions | |
anson | -0.579 |
rem | -0.505 |
� | -0.486 |
roman | -0.451 |
lement | -0.447 |
Token | 10 |
Token ID | 940 |
Feature activation | -8.19e-04 |
Pos logit contributions | |
@@ | +0.196 |
Krug | +0.184 |
Lie | +0.182 |
� | +0.182 |
COUN | +0.180 |
Neg logit contributions | |
anson | -0.286 |
rem | -0.246 |
� | -0.235 |
roman | -0.215 |
lement | -0.213 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
@@ | +0.366 |
Lie | +0.342 |
Krug | +0.334 |
Nemesis | +0.328 |
Lt | +0.327 |
Neg logit contributions | |
anson | -0.555 |
rem | -0.464 |
� | -0.447 |
roman | -0.416 |
uana | -0.414 |
Token | heart |
Token ID | 11499 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
@@ | +0.226 |
Lie | +0.216 |
� | +0.216 |
Krug | +0.214 |
� | +0.209 |
Neg logit contributions | |
anson | -0.347 |
rem | -0.299 |
� | -0.285 |
lement | -0.262 |
roman | -0.261 |
Token | l |
Token ID | 75 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
@@ | +0.258 |
� | +0.245 |
Krug | +0.243 |
Lie | +0.242 |
crossings | +0.240 |
Neg logit contributions | |
anson | -0.339 |
rem | -0.292 |
� | -0.277 |
lement | -0.252 |
roman | -0.252 |
Token | xp |
Token ID | 42372 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
@@ | +0.356 |
� | +0.336 |
Lie | +0.334 |
Krug | +0.334 |
crossings | +0.329 |
Neg logit contributions | |
anson | -0.457 |
rem | -0.394 |
� | -0.372 |
roman | -0.340 |
lement | -0.338 |
Token | Profile |
Token ID | 13118 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
@@ | +0.041 |
� | +0.038 |
Krug | +0.038 |
Lie | +0.038 |
crossings | +0.038 |
Neg logit contributions | |
anson | -0.058 |
rem | -0.050 |
� | -0.047 |
lement | -0.043 |
roman | -0.043 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
@@ | +0.440 |
� | +0.402 |
Krug | +0.401 |
Lie | +0.395 |
Nemesis | +0.388 |
Neg logit contributions | |
anson | -1.132 |
rem | -1.009 |
� | -0.965 |
roman | -0.905 |
lement | -0.904 |
Token | September |
Token ID | 2693 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
@@ | +2.150 |
Krug | +2.013 |
Nemesis | +1.997 |
Claim | +1.978 |
Lie | +1.951 |
Neg logit contributions | |
anson | -3.082 |
rem | -2.580 |
� | -2.479 |
uana | -2.310 |
roman | -2.285 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
@@ | +0.674 |
Krug | +0.653 |
Scand | +0.645 |
Lie | +0.643 |
Nemesis | +0.641 |
Neg logit contributions | |
anson | -0.614 |
rem | -0.508 |
� | -0.471 |
roman | -0.425 |
lement | -0.425 |
Token | United |
Token ID | 1578 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
@@ | +0.077 |
Krug | +0.074 |
Lie | +0.073 |
� | +0.073 |
Nemesis | +0.073 |
Neg logit contributions | |
anson | -0.091 |
rem | -0.077 |
� | -0.072 |
roman | -0.066 |
lement | -0.066 |
Token | States |
Token ID | 1829 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
@@ | +0.423 |
� | +0.395 |
Krug | +0.392 |
Lie | +0.388 |
crossings | +0.388 |
Neg logit contributions | |
anson | -0.852 |
rem | -0.755 |
� | -0.718 |
lement | -0.671 |
roman | -0.664 |
Token | 10 |
Token ID | 838 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
@@ | +0.879 |
Krug | +0.828 |
Nemesis | +0.815 |
Lie | +0.806 |
COUN | +0.802 |
Neg logit contributions | |
anson | -1.744 |
rem | -1.515 |
� | -1.466 |
roman | -1.360 |
lement | -1.351 |
Token |
|
Token ID | 198 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
@@ | +0.268 |
Lie | +0.250 |
Krug | +0.244 |
Nemesis | +0.238 |
territory | +0.238 |
Neg logit contributions | |
anson | -0.410 |
rem | -0.342 |
� | -0.331 |
roman | -0.309 |
uana | -0.306 |
Token | Big |
Token ID | 12804 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
@@ | +0.189 |
Lie | +0.180 |
� | +0.180 |
Krug | +0.178 |
� | +0.175 |
Neg logit contributions | |
anson | -0.285 |
rem | -0.245 |
� | -0.234 |
lement | -0.216 |
roman | -0.214 |
Token | Fan |
Token ID | 22480 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
@@ | +0.842 |
Lie | +0.803 |
Krug | +0.802 |
� | +0.795 |
Nemesis | +0.784 |
Neg logit contributions | |
anson | -1.068 |
rem | -0.911 |
� | -0.870 |
lement | -0.792 |
roman | -0.790 |
Token | Profile |
Token ID | 13118 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
@@ | +0.532 |
� | +0.504 |
Lie | +0.501 |
Krug | +0.497 |
� | +0.487 |
Neg logit contributions | |
anson | -0.813 |
rem | -0.704 |
� | -0.669 |
lement | -0.617 |
roman | -0.614 |
Token | Blog |
Token ID | 14001 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
@@ | +0.465 |
� | +0.430 |
Krug | +0.429 |
Lie | +0.424 |
crossings | +0.416 |
Neg logit contributions | |
anson | -1.073 |
rem | -0.953 |
� | -0.910 |
lement | -0.849 |
roman | -0.848 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
@@ | +0.732 |
Krug | +0.626 |
Lie | +0.607 |
Claim | +0.606 |
Nemesis | +0.600 |
Neg logit contributions | |
anson | -2.850 |
rem | -2.541 |
� | -2.481 |
lement | -2.347 |
heng | -2.323 |
Token | December |
Token ID | 3426 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
@@ | +2.109 |
Krug | +2.021 |
Nemesis | +1.988 |
Scand | +1.966 |
Claim | +1.963 |
Neg logit contributions | |
anson | -2.997 |
rem | -2.519 |
� | -2.400 |
uana | -2.237 |
roman | -2.233 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +9.44e-02 |
Pos logit contributions | |
@@ | +1.080 |
Krug | +1.058 |
Scand | +1.038 |
Nemesis | +1.038 |
Lie | +1.033 |
Neg logit contributions | |
anson | -1.021 |
rem | -0.831 |
� | -0.780 |
uana | -0.713 |
roman | -0.706 |
Token | TL |
Token ID | 24811 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
@@ | +0.180 |
Krug | +0.173 |
� | +0.172 |
Lie | +0.171 |
Nemesis | +0.169 |
Neg logit contributions | |
anson | -0.201 |
rem | -0.170 |
� | -0.159 |
roman | -0.144 |
lement | -0.144 |
Token | AD |
Token ID | 2885 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
@@ | +1.030 |
Krug | +0.982 |
COUN | +0.976 |
Nemesis | +0.964 |
Lie | +0.960 |
Neg logit contributions | |
anson | -1.749 |
rem | -1.523 |
� | -1.464 |
roman | -1.346 |
lement | -1.338 |
Token | T |
Token ID | 51 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
anson | +0.055 |
rem | +0.048 |
� | +0.046 |
roman | +0.042 |
lement | +0.042 |
Neg logit contributions | |
@@ | -0.030 |
� | -0.028 |
Krug | -0.028 |
Lie | -0.028 |
COUN | -0.027 |
Token | K |
Token ID | 42 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
@@ | +0.558 |
� | +0.527 |
Krug | +0.526 |
Lie | +0.524 |
crossings | +0.513 |
Neg logit contributions | |
anson | -0.885 |
rem | -0.771 |
� | -0.731 |
lement | -0.673 |
roman | -0.671 |
Token | ier |
Token ID | 959 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
anson | +0.238 |
rem | +0.205 |
� | +0.195 |
lement | +0.178 |
roman | +0.178 |
Neg logit contributions | |
@@ | -0.173 |
� | -0.164 |
Lie | -0.163 |
Krug | -0.163 |
� | -0.159 |
Token | an |
Token ID | 272 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
@@ | +0.843 |
Lie | +0.808 |
� | +0.808 |
Krug | +0.803 |
� | +0.790 |
Neg logit contributions | |
anson | -1.113 |
rem | -0.963 |
� | -0.915 |
roman | -0.829 |
lement | -0.824 |
Token | Profile |
Token ID | 13118 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
anson | +0.308 |
rem | +0.268 |
� | +0.254 |
lement | +0.233 |
roman | +0.233 |
Neg logit contributions | |
@@ | -0.207 |
� | -0.196 |
Krug | -0.195 |
Lie | -0.194 |
crossings | -0.192 |
Token | Blog |
Token ID | 14001 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
@@ | +0.550 |
� | +0.511 |
Krug | +0.510 |
Lie | +0.504 |
crossings | +0.496 |
Neg logit contributions | |
anson | -1.195 |
rem | -1.059 |
� | -1.009 |
lement | -0.941 |
roman | -0.940 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
@@ | +0.721 |
Krug | +0.625 |
Lie | +0.607 |
Nemesis | +0.599 |
Claim | +0.597 |
Neg logit contributions | |
anson | -2.884 |
rem | -2.570 |
� | -2.508 |
lement | -2.370 |
uana | -2.352 |
Token | August |
Token ID | 2932 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
@@ | +2.074 |
Krug | +1.992 |
Nemesis | +1.959 |
Scand | +1.940 |
Claim | +1.924 |
Neg logit contributions | |
anson | -2.937 |
rem | -2.465 |
� | -2.353 |
roman | -2.188 |
uana | -2.181 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
@@ | +0.928 |
Krug | +0.906 |
Lie | +0.892 |
Nemesis | +0.891 |
Scand | +0.890 |
Neg logit contributions | |
anson | -0.847 |
rem | -0.690 |
� | -0.645 |
lement | -0.578 |
roman | -0.578 |
Token | United |
Token ID | 1578 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
@@ | +0.264 |
Krug | +0.254 |
Lie | +0.251 |
� | +0.251 |
Nemesis | +0.249 |
Neg logit contributions | |
anson | -0.302 |
rem | -0.255 |
� | -0.239 |
roman | -0.218 |
lement | -0.218 |
Token | States |
Token ID | 1829 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
@@ | +0.469 |
� | +0.437 |
Krug | +0.432 |
crossings | +0.431 |
Lie | +0.429 |
Neg logit contributions | |
anson | -0.953 |
rem | -0.846 |
� | -0.804 |
lement | -0.752 |
roman | -0.745 |
Token | 6 |
Token ID | 718 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
@@ | +0.778 |
Krug | +0.735 |
Nemesis | +0.725 |
Lie | +0.714 |
COUN | +0.711 |
Neg logit contributions | |
anson | -1.514 |
rem | -1.313 |
� | -1.273 |
roman | -1.178 |
lement | -1.169 |
Token | screen |
Token ID | 3159 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
@@ | +0.289 |
Krug | +0.273 |
� | +0.272 |
crossings | +0.271 |
Lie | +0.268 |
Neg logit contributions | |
anson | -0.420 |
rem | -0.360 |
� | -0.343 |
roman | -0.312 |
lement | -0.312 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
@@ | +0.557 |
Krug | +0.523 |
crossings | +0.521 |
� | +0.519 |
Lie | +0.513 |
Neg logit contributions | |
anson | -0.749 |
rem | -0.636 |
� | -0.606 |
lement | -0.551 |
roman | -0.550 |
Token | Source |
Token ID | 8090 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
anson | +0.284 |
rem | +0.240 |
� | +0.226 |
lement | +0.206 |
roman | +0.206 |
Neg logit contributions | |
@@ | -0.249 |
Krug | -0.237 |
� | -0.235 |
Lie | -0.233 |
Nemesis | -0.233 |
Token | : |
Token ID | 25 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
anson | +0.583 |
rem | +0.505 |
� | +0.477 |
roman | +0.437 |
lement | +0.437 |
Neg logit contributions | |
@@ | -0.410 |
Krug | -0.387 |
� | -0.387 |
Lie | -0.383 |
crossings | -0.380 |
Token | Game |
Token ID | 3776 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
anson | +0.239 |
rem | +0.207 |
� | +0.196 |
lement | +0.180 |
roman | +0.180 |
Neg logit contributions | |
@@ | -0.164 |
Krug | -0.155 |
� | -0.154 |
Lie | -0.152 |
Nemesis | -0.151 |
Token | FAQ |
Token ID | 42680 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
anson | +0.542 |
rem | +0.474 |
� | +0.449 |
lement | +0.415 |
roman | +0.414 |
Neg logit contributions | |
@@ | -0.328 |
� | -0.308 |
Krug | -0.308 |
Lie | -0.305 |
crossings | -0.303 |
Token | s |
Token ID | 82 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
@@ | +1.837 |
Lie | +1.682 |
territory | +1.676 |
Krug | +1.659 |
Nemesis | +1.659 |
Neg logit contributions | |
anson | -3.070 |
rem | -2.671 |
� | -2.554 |
lement | -2.354 |
roman | -2.343 |
Token |
|
Token ID | 198 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
anson | +0.306 |
rem | +0.263 |
� | +0.249 |
roman | +0.226 |
lement | +0.225 |
Neg logit contributions | |
@@ | -0.234 |
� | -0.221 |
Krug | -0.221 |
Nemesis | -0.217 |
crossings | -0.217 |
Token |
|
Token ID | 198 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
anson | +0.142 |
rem | +0.118 |
� | +0.114 |
roman | +0.106 |
uana | +0.105 |
Neg logit contributions | |
@@ | -0.097 |
Lie | -0.091 |
Krug | -0.089 |
� | -0.087 |
territory | -0.087 |
Token | The |
Token ID | 464 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
anson | +0.137 |
rem | +0.117 |
� | +0.112 |
lement | +0.103 |
roman | +0.103 |
Neg logit contributions | |
@@ | -0.087 |
Lie | -0.084 |
� | -0.082 |
Krug | -0.082 |
� | -0.080 |
Token | focus |
Token ID | 2962 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
anson | +0.032 |
rem | +0.028 |
� | +0.027 |
lement | +0.025 |
roman | +0.024 |
Neg logit contributions | |
@@ | -0.020 |
Krug | -0.019 |
� | -0.019 |
crossings | -0.019 |
Nemesis | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +8.64e-02 |
Pos logit contributions | |
@@ | +0.725 |
� | +0.696 |
Krug | +0.694 |
Lie | +0.691 |
crossings | +0.686 |
Neg logit contributions | |
anson | -0.644 |
rem | -0.540 |
� | -0.501 |
lement | -0.446 |
roman | -0.444 |
Token | Several |
Token ID | 12168 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
@@ | +0.138 |
Krug | +0.129 |
� | +0.128 |
Lie | +0.128 |
COUN | +0.125 |
Neg logit contributions | |
anson | -0.214 |
rem | -0.184 |
� | -0.175 |
roman | -0.162 |
lement | -0.161 |
Token | trees |
Token ID | 7150 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
@@ | +0.427 |
Krug | +0.400 |
� | +0.398 |
crossings | +0.398 |
Lie | +0.393 |
Neg logit contributions | |
anson | -0.737 |
rem | -0.643 |
� | -0.613 |
lement | -0.566 |
roman | -0.565 |
Token | have |
Token ID | 423 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
@@ | +0.507 |
� | +0.477 |
Krug | +0.475 |
crossings | +0.474 |
Lie | +0.471 |
Neg logit contributions | |
anson | -0.681 |
rem | -0.582 |
� | -0.553 |
roman | -0.506 |
lement | -0.502 |
Token | been |
Token ID | 587 |
Feature activation | -9.11e-04 |
Pos logit contributions | |
@@ | +0.137 |
� | +0.130 |
Krug | +0.130 |
Lie | +0.129 |
crossings | +0.129 |
Neg logit contributions | |
anson | -0.170 |
rem | -0.145 |
� | -0.137 |
lement | -0.125 |
roman | -0.125 |
Token | cut |
Token ID | 2005 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
anson | +0.002 |
rem | +0.002 |
� | +0.002 |
lement | +0.002 |
roman | +0.002 |
Neg logit contributions | |
@@ | -0.001 |
� | -0.001 |
Krug | -0.001 |
crossings | -0.001 |
Lie | -0.001 |
Token | in |
Token ID | 287 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
@@ | +2.001 |
crossings | +1.888 |
territory | +1.846 |
Krug | +1.832 |
� | +1.822 |
Neg logit contributions | |
anson | -2.916 |
rem | -2.487 |
� | -2.368 |
roman | -2.181 |
lement | -2.157 |
Token | the |
Token ID | 262 |
Feature activation | +6.28e-03 |
Pos logit contributions | |
@@ | +0.177 |
Krug | +0.167 |
� | +0.166 |
crossings | +0.166 |
Lie | +0.164 |
Neg logit contributions | |
anson | -0.248 |
rem | -0.214 |
� | -0.202 |
lement | -0.186 |
roman | -0.185 |
Token | area |
Token ID | 1989 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
@@ | +0.009 |
Krug | +0.008 |
� | +0.008 |
crossings | +0.008 |
Lie | +0.008 |
Neg logit contributions | |
anson | -0.016 |
rem | -0.014 |
� | -0.014 |
lement | -0.013 |
roman | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
anson | +0.334 |
rem | +0.286 |
� | +0.270 |
lement | +0.246 |
roman | +0.246 |
Neg logit contributions | |
@@ | -0.255 |
Krug | -0.241 |
� | -0.240 |
crossings | -0.239 |
Lie | -0.238 |
Token | make |
Token ID | 787 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
@@ | +0.168 |
� | +0.156 |
Krug | +0.156 |
Lie | +0.155 |
crossings | +0.155 |
Neg logit contributions | |
anson | -0.258 |
rem | -0.221 |
� | -0.212 |
roman | -0.195 |
lement | -0.194 |
Token |
|
Token ID | 198 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
@@ | +0.065 |
Lie | +0.061 |
Krug | +0.059 |
Lt | +0.058 |
Claim | +0.058 |
Neg logit contributions | |
anson | -0.098 |
rem | -0.081 |
� | -0.079 |
roman | -0.073 |
uana | -0.073 |
Token | Australian |
Token ID | 38036 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
anson | +0.109 |
rem | +0.093 |
� | +0.089 |
lement | +0.082 |
roman | +0.081 |
Neg logit contributions | |
@@ | -0.073 |
Lie | -0.071 |
� | -0.069 |
Krug | -0.068 |
� | -0.067 |
Token | clinical |
Token ID | 8668 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
anson | +0.468 |
rem | +0.408 |
� | +0.387 |
lement | +0.356 |
roman | +0.356 |
Neg logit contributions | |
@@ | -0.302 |
� | -0.285 |
Krug | -0.284 |
Lie | -0.282 |
crossings | -0.281 |
Token | psychologist |
Token ID | 23540 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
@@ | +0.282 |
Krug | +0.265 |
� | +0.264 |
Lie | +0.262 |
crossings | +0.261 |
Neg logit contributions | |
anson | -0.462 |
rem | -0.404 |
� | -0.384 |
lement | -0.353 |
roman | -0.353 |
Token | Dr |
Token ID | 1583 |
Feature activation | -9.12e-02 |
Pos logit contributions | |
@@ | +0.798 |
Krug | +0.759 |
� | +0.748 |
Lie | +0.746 |
crossings | +0.744 |
Neg logit contributions | |
anson | -1.102 |
rem | -0.957 |
� | -0.908 |
roman | -0.834 |
lement | -0.830 |
Token | Deborah |
Token ID | 36976 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
anson | +0.204 |
rem | +0.175 |
� | +0.165 |
roman | +0.150 |
lement | +0.150 |
Neg logit contributions | |
@@ | -0.163 |
Krug | -0.156 |
� | -0.155 |
Lie | -0.154 |
crossings | -0.152 |
Token | Thomas |
Token ID | 5658 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
@@ | +1.855 |
Krug | +1.814 |
� | +1.744 |
Lie | +1.740 |
Nemesis | +1.732 |
Neg logit contributions | |
anson | -2.979 |
rem | -2.598 |
� | -2.474 |
roman | -2.293 |
lement | -2.282 |
Token | said |
Token ID | 531 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
@@ | +0.212 |
Krug | +0.199 |
� | +0.198 |
Lie | +0.195 |
crossings | +0.194 |
Neg logit contributions | |
anson | -0.425 |
rem | -0.376 |
� | -0.359 |
lement | -0.334 |
roman | -0.333 |
Token | while |
Token ID | 981 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
@@ | +0.020 |
Krug | +0.019 |
crossings | +0.019 |
� | +0.019 |
Lie | +0.018 |
Neg logit contributions | |
anson | -0.028 |
rem | -0.024 |
� | -0.023 |
roman | -0.021 |
lement | -0.021 |
Token | she |
Token ID | 673 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
@@ | +0.245 |
Krug | +0.235 |
crossings | +0.232 |
� | +0.230 |
Lie | +0.228 |
Neg logit contributions | |
anson | -0.354 |
rem | -0.306 |
� | -0.293 |
roman | -0.270 |
lement | -0.268 |
Token | did |
Token ID | 750 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
@@ | +0.408 |
Krug | +0.387 |
crossings | +0.386 |
� | +0.378 |
Lie | +0.376 |
Neg logit contributions | |
anson | -0.587 |
rem | -0.500 |
� | -0.483 |
roman | -0.446 |
lement | -0.432 |
Token | une |
Token ID | 1726 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
anson | +0.019 |
rem | +0.016 |
� | +0.015 |
lement | +0.013 |
roman | +0.013 |
Neg logit contributions | |
@@ | -0.022 |
� | -0.021 |
Krug | -0.021 |
Lie | -0.021 |
crossings | -0.020 |
Token | limit |
Token ID | 4179 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
anson | +0.807 |
rem | +0.700 |
� | +0.662 |
roman | +0.608 |
lement | +0.608 |
Neg logit contributions | |
@@ | -0.550 |
� | -0.519 |
Krug | -0.518 |
Lie | -0.514 |
crossings | -0.510 |
Token | ( |
Token ID | 357 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
@@ | +0.226 |
� | +0.211 |
Krug | +0.210 |
crossings | +0.208 |
Lie | +0.208 |
Neg logit contributions | |
anson | -0.344 |
rem | -0.297 |
� | -0.283 |
roman | -0.261 |
lement | -0.259 |
Token | up |
Token ID | 929 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
@@ | +0.024 |
� | +0.023 |
Lie | +0.023 |
Krug | +0.023 |
crossings | +0.022 |
Neg logit contributions | |
anson | -0.031 |
rem | -0.026 |
� | -0.025 |
lement | -0.023 |
roman | -0.022 |
Token | 57 |
Token ID | 7632 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
@@ | +0.522 |
Krug | +0.500 |
� | +0.499 |
Lie | +0.493 |
crossings | +0.490 |
Neg logit contributions | |
anson | -0.612 |
rem | -0.529 |
� | -0.489 |
lement | -0.444 |
roman | -0.442 |
Token | %) |
Token ID | 4407 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
@@ | +0.687 |
� | +0.646 |
Krug | +0.646 |
crossings | +0.638 |
Lie | +0.638 |
Neg logit contributions | |
anson | -1.025 |
rem | -0.888 |
� | -0.843 |
lement | -0.781 |
roman | -0.775 |
Token | as |
Token ID | 355 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
@@ | +2.136 |
� | +1.962 |
Krug | +1.962 |
Lie | +1.958 |
territory | +1.955 |
Neg logit contributions | |
anson | -2.705 |
rem | -2.262 |
� | -2.165 |
roman | -1.978 |
lement | -1.974 |
Token | well |
Token ID | 880 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
anson | +0.542 |
rem | +0.477 |
� | +0.456 |
roman | +0.423 |
lement | +0.422 |
Neg logit contributions | |
@@ | -0.282 |
� | -0.262 |
Krug | -0.262 |
Lie | -0.258 |
crossings | -0.257 |
Token | as |
Token ID | 355 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
@@ | +0.519 |
Krug | +0.498 |
� | +0.494 |
Lie | +0.490 |
crossings | +0.485 |
Neg logit contributions | |
anson | -0.743 |
rem | -0.649 |
� | -0.613 |
lement | -0.563 |
roman | -0.559 |
Token | more |
Token ID | 517 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
anson | +0.720 |
rem | +0.622 |
� | +0.589 |
lement | +0.539 |
roman | +0.539 |
Neg logit contributions | |
@@ | -0.512 |
Krug | -0.483 |
� | -0.483 |
Lie | -0.478 |
crossings | -0.476 |
Token | relaxed |
Token ID | 18397 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
anson | +0.658 |
rem | +0.571 |
� | +0.541 |
lement | +0.496 |
roman | +0.496 |
Neg logit contributions | |
@@ | -0.450 |
� | -0.424 |
Krug | -0.422 |
Lie | -0.418 |
crossings | -0.418 |
Token | court |
Token ID | 2184 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
anson | +0.471 |
rem | +0.414 |
� | +0.394 |
lement | +0.365 |
roman | +0.364 |
Neg logit contributions | |
@@ | -0.263 |
Krug | -0.247 |
� | -0.246 |
Lie | -0.244 |
crossings | -0.243 |
Token | on |
Token ID | 319 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
@@ | +0.281 |
Krug | +0.263 |
crossings | +0.261 |
� | +0.261 |
Lie | +0.258 |
Neg logit contributions | |
anson | -0.466 |
rem | -0.406 |
� | -0.388 |
roman | -0.358 |
lement | -0.357 |
Token | Wednesday |
Token ID | 3583 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
@@ | +0.534 |
Krug | +0.506 |
� | +0.504 |
crossings | +0.501 |
Lie | +0.500 |
Neg logit contributions | |
anson | -0.626 |
rem | -0.534 |
� | -0.504 |
roman | -0.458 |
lement | -0.457 |
Token | handed |
Token ID | 10158 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
@@ | +0.398 |
Krug | +0.373 |
crossings | +0.370 |
� | +0.366 |
territory | +0.359 |
Neg logit contributions | |
anson | -0.861 |
rem | -0.757 |
� | -0.726 |
roman | -0.675 |
lement | -0.675 |
Token | ab |
Token ID | 450 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
anson | +0.310 |
rem | +0.267 |
� | +0.253 |
lement | +0.231 |
roman | +0.230 |
Neg logit contributions | |
@@ | -0.226 |
Krug | -0.214 |
� | -0.213 |
Lie | -0.211 |
crossings | -0.211 |
Token | orig |
Token ID | 11612 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
anson | +0.325 |
rem | +0.286 |
� | +0.272 |
lement | +0.252 |
roman | +0.252 |
Neg logit contributions | |
@@ | -0.172 |
Krug | -0.161 |
� | -0.161 |
Lie | -0.159 |
crossings | -0.158 |
Token | inals |
Token ID | 6897 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
@@ | +1.700 |
� | +1.586 |
Krug | +1.585 |
crossings | +1.575 |
Lie | +1.574 |
Neg logit contributions | |
anson | -3.077 |
rem | -2.695 |
� | -2.581 |
roman | -2.393 |
lement | -2.378 |
Token | a |
Token ID | 257 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
@@ | +0.105 |
Krug | +0.101 |
� | +0.100 |
crossings | +0.100 |
Lie | +0.099 |
Neg logit contributions | |
anson | -0.139 |
rem | -0.118 |
� | -0.112 |
roman | -0.103 |
lement | -0.102 |
Token | partial |
Token ID | 13027 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
@@ | +0.352 |
Krug | +0.331 |
� | +0.329 |
Lie | +0.325 |
crossings | +0.325 |
Neg logit contributions | |
anson | -0.528 |
rem | -0.454 |
� | -0.433 |
roman | -0.399 |
lement | -0.397 |
Token | victory |
Token ID | 5373 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
anson | +0.893 |
rem | +0.767 |
� | +0.728 |
roman | +0.671 |
lement | +0.671 |
Neg logit contributions | |
@@ | -0.600 |
Krug | -0.566 |
� | -0.563 |
crossings | -0.559 |
Lie | -0.557 |
Token | on |
Token ID | 319 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
anson | +0.321 |
rem | +0.277 |
� | +0.264 |
roman | +0.241 |
lement | +0.240 |
Neg logit contributions | |
@@ | -0.218 |
Krug | -0.204 |
� | -0.203 |
crossings | -0.202 |
Lie | -0.201 |
Token | YouTube |
Token ID | 7444 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
@@ | +0.089 |
Krug | +0.084 |
� | +0.083 |
Lie | +0.083 |
crossings | +0.082 |
Neg logit contributions | |
anson | -0.150 |
rem | -0.131 |
� | -0.125 |
roman | -0.115 |
lement | -0.115 |
Token | spokeswoman |
Token ID | 12262 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
@@ | +0.077 |
Krug | +0.073 |
� | +0.072 |
Lie | +0.071 |
Nemesis | +0.070 |
Neg logit contributions | |
anson | -0.114 |
rem | -0.098 |
� | -0.094 |
roman | -0.086 |
lement | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
@@ | +0.680 |
Krug | +0.642 |
� | +0.640 |
Lie | +0.633 |
crossings | +0.631 |
Neg logit contributions | |
anson | -0.850 |
rem | -0.735 |
� | -0.695 |
roman | -0.631 |
lement | -0.630 |
Token |
|
Token ID | 198 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
@@ | +0.435 |
Krug | +0.412 |
Lie | +0.405 |
� | +0.401 |
COUN | +0.398 |
Neg logit contributions | |
anson | -0.607 |
rem | -0.521 |
� | -0.496 |
roman | -0.458 |
lement | -0.453 |
Token |
|
Token ID | 198 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
@@ | +0.302 |
Lie | +0.284 |
Krug | +0.274 |
Lt | +0.272 |
Claim | +0.268 |
Neg logit contributions | |
anson | -0.461 |
rem | -0.383 |
� | -0.372 |
roman | -0.347 |
uana | -0.344 |
Token | Newsletter |
Token ID | 33031 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
@@ | +0.184 |
Lie | +0.178 |
� | +0.173 |
Krug | +0.172 |
� | +0.170 |
Neg logit contributions | |
anson | -0.274 |
rem | -0.233 |
� | -0.224 |
lement | -0.206 |
roman | -0.205 |
Token | Sign |
Token ID | 5865 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
@@ | +1.039 |
Krug | +0.914 |
crossings | +0.893 |
Lie | +0.883 |
Nemesis | +0.881 |
Neg logit contributions | |
anson | -3.712 |
rem | -3.361 |
� | -3.253 |
roman | -3.045 |
lement | -3.026 |
Token | Up |
Token ID | 3205 |
Feature activation | +7.91e-02 |
Pos logit contributions | |
anson | +0.040 |
rem | +0.035 |
� | +0.034 |
roman | +0.031 |
lement | +0.031 |
Neg logit contributions | |
@@ | -0.020 |
Krug | -0.018 |
� | -0.018 |
Lie | -0.018 |
Nemesis | -0.018 |
Token | Continue |
Token ID | 10054 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
@@ | +0.117 |
Lie | +0.108 |
Krug | +0.106 |
Nemesis | +0.106 |
� | +0.105 |
Neg logit contributions | |
anson | -0.205 |
rem | -0.178 |
� | -0.170 |
lement | -0.158 |
aston | -0.157 |
Token | reading |
Token ID | 3555 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
@@ | +0.237 |
Krug | +0.224 |
crossings | +0.223 |
Lie | +0.222 |
territory | +0.220 |
Neg logit contributions | |
anson | -0.316 |
rem | -0.272 |
� | -0.258 |
roman | -0.237 |
lement | -0.233 |
Token | the |
Token ID | 262 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
anson | +0.511 |
rem | +0.441 |
� | +0.418 |
roman | +0.385 |
lement | +0.380 |
Neg logit contributions | |
@@ | -0.351 |
Lie | -0.325 |
Krug | -0.325 |
� | -0.324 |
crossings | -0.321 |
Token | these |
Token ID | 777 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
anson | +0.236 |
rem | +0.203 |
� | +0.191 |
lement | +0.173 |
roman | +0.173 |
Neg logit contributions | |
@@ | -0.194 |
� | -0.185 |
Krug | -0.184 |
Lie | -0.183 |
crossings | -0.182 |
Token | new |
Token ID | 649 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
@@ | +0.182 |
crossings | +0.172 |
Krug | +0.172 |
� | +0.171 |
Lie | +0.171 |
Neg logit contributions | |
anson | -0.253 |
rem | -0.218 |
� | -0.207 |
roman | -0.190 |
lement | -0.189 |
Token | changes |
Token ID | 2458 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
@@ | +0.350 |
crossings | +0.334 |
Krug | +0.333 |
Lie | +0.332 |
� | +0.331 |
Neg logit contributions | |
anson | -0.461 |
rem | -0.393 |
� | -0.374 |
roman | -0.341 |
lement | -0.339 |
Token | . |
Token ID | 13 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
anson | +0.743 |
rem | +0.641 |
� | +0.610 |
roman | +0.556 |
lement | +0.554 |
Neg logit contributions | |
@@ | -0.506 |
Krug | -0.476 |
� | -0.476 |
crossings | -0.473 |
Lie | -0.471 |
Token | The |
Token ID | 383 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
@@ | +0.151 |
Krug | +0.144 |
Lie | +0.142 |
Claim | +0.141 |
� | +0.140 |
Neg logit contributions | |
anson | -0.233 |
rem | -0.201 |
� | -0.192 |
roman | -0.177 |
lement | -0.174 |
Token | taxp |
Token ID | 9065 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
anson | +0.136 |
rem | +0.119 |
� | +0.114 |
lement | +0.105 |
roman | +0.105 |
Neg logit contributions | |
@@ | -0.075 |
Krug | -0.070 |
� | -0.070 |
Lie | -0.069 |
crossings | -0.069 |
Token | aying |
Token ID | 8369 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
@@ | +1.846 |
� | +1.716 |
Lie | +1.708 |
crossings | +1.674 |
Krug | +1.669 |
Neg logit contributions | |
anson | -2.803 |
rem | -2.427 |
� | -2.409 |
finger | -2.132 |
roman | -2.129 |
Token | rail |
Token ID | 6787 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
@@ | +0.169 |
Krug | +0.159 |
crossings | +0.158 |
� | +0.157 |
Lie | +0.157 |
Neg logit contributions | |
anson | -0.310 |
rem | -0.273 |
� | -0.261 |
lement | -0.240 |
roman | -0.239 |
Token | roads |
Token ID | 21372 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
anson | +0.344 |
rem | +0.306 |
� | +0.296 |
lement | +0.276 |
roman | +0.275 |
Neg logit contributions | |
@@ | -0.123 |
� | -0.114 |
crossings | -0.112 |
Krug | -0.112 |
Lie | -0.111 |
Token | challenged |
Token ID | 12827 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
anson | +0.850 |
rem | +0.740 |
� | +0.706 |
roman | +0.650 |
lement | +0.647 |
Neg logit contributions | |
@@ | -0.522 |
Krug | -0.490 |
� | -0.489 |
crossings | -0.485 |
Lie | -0.483 |
Token | this |
Token ID | 428 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
anson | +0.613 |
rem | +0.527 |
� | +0.496 |
lement | +0.450 |
roman | +0.450 |
Neg logit contributions | |
@@ | -0.497 |
Krug | -0.473 |
� | -0.472 |
Lie | -0.469 |
crossings | -0.467 |
Token | the |
Token ID | 262 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
anson | +0.076 |
rem | +0.066 |
� | +0.062 |
lement | +0.057 |
roman | +0.057 |
Neg logit contributions | |
@@ | -0.052 |
Krug | -0.049 |
� | -0.049 |
Lie | -0.048 |
crossings | -0.048 |
Token | two |
Token ID | 734 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
@@ | +0.197 |
Krug | +0.183 |
� | +0.183 |
Lie | +0.180 |
crossings | +0.180 |
Neg logit contributions | |
anson | -0.389 |
rem | -0.344 |
� | -0.328 |
lement | -0.304 |
roman | -0.303 |
Token | cities |
Token ID | 4736 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
@@ | +0.084 |
Krug | +0.079 |
� | +0.079 |
crossings | +0.078 |
Lie | +0.077 |
Neg logit contributions | |
anson | -0.151 |
rem | -0.132 |
� | -0.126 |
lement | -0.117 |
roman | -0.117 |
Token | — |
Token ID | 960 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
anson | +0.355 |
rem | +0.304 |
� | +0.286 |
lement | +0.260 |
roman | +0.259 |
Neg logit contributions | |
@@ | -0.294 |
� | -0.280 |
Krug | -0.280 |
Lie | -0.277 |
crossings | -0.276 |
Token | let |
Token ID | 1616 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
@@ | +0.430 |
Lie | +0.413 |
Krug | +0.411 |
� | +0.410 |
crossings | +0.404 |
Neg logit contributions | |
anson | -0.504 |
rem | -0.423 |
� | -0.401 |
lement | -0.367 |
roman | -0.364 |
Token | alone |
Token ID | 3436 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
@@ | +0.062 |
Krug | +0.058 |
� | +0.058 |
Lie | +0.058 |
crossings | +0.057 |
Neg logit contributions | |
anson | -0.115 |
rem | -0.101 |
� | -0.096 |
roman | -0.089 |
lement | -0.089 |
Token | 59 |
Token ID | 7863 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
@@ | +1.864 |
Krug | +1.777 |
crossings | +1.749 |
� | +1.740 |
Lie | +1.738 |
Neg logit contributions | |
anson | -2.775 |
rem | -2.398 |
� | -2.256 |
lement | -2.093 |
aston | -2.080 |
Token | community |
Token ID | 2055 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
@@ | +0.211 |
� | +0.199 |
Krug | +0.199 |
Lie | +0.197 |
crossings | +0.197 |
Neg logit contributions | |
anson | -0.316 |
rem | -0.274 |
� | -0.259 |
lement | -0.239 |
roman | -0.238 |
Token | districts |
Token ID | 12815 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
@@ | +0.260 |
crossings | +0.221 |
Krug | +0.220 |
� | +0.219 |
Lie | +0.211 |
Neg logit contributions | |
anson | -1.432 |
rem | -1.295 |
� | -1.249 |
roman | -1.182 |
lement | -1.180 |
Token | — |
Token ID | 960 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
@@ | +0.375 |
� | +0.355 |
Krug | +0.355 |
crossings | +0.354 |
Lie | +0.353 |
Neg logit contributions | |
anson | -0.394 |
rem | -0.332 |
� | -0.310 |
roman | -0.279 |
lement | -0.279 |
Token | that |
Token ID | 5562 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
@@ | +0.614 |
� | +0.586 |
Krug | +0.584 |
Lie | +0.582 |
crossings | +0.575 |
Neg logit contributions | |
anson | -0.665 |
rem | -0.559 |
� | -0.527 |
lement | -0.477 |
roman | -0.474 |
Token |
|
Token ID | 198 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
anson | +0.041 |
rem | +0.035 |
� | +0.033 |
lement | +0.030 |
roman | +0.030 |
Neg logit contributions | |
@@ | -0.031 |
� | -0.029 |
Krug | -0.029 |
Lie | -0.029 |
crossings | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
@@ | +0.445 |
Lie | +0.414 |
Krug | +0.404 |
Nemesis | +0.397 |
territory | +0.397 |
Neg logit contributions | |
anson | -0.682 |
rem | -0.569 |
� | -0.550 |
roman | -0.512 |
uana | -0.508 |
Token | Ko |
Token ID | 48735 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
@@ | +0.325 |
Lie | +0.311 |
� | +0.310 |
Krug | +0.307 |
� | +0.302 |
Neg logit contributions | |
anson | -0.497 |
rem | -0.429 |
� | -0.409 |
lement | -0.376 |
roman | -0.373 |
Token | Do |
Token ID | 5211 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
@@ | +0.066 |
� | +0.062 |
Lie | +0.062 |
Krug | +0.062 |
crossings | +0.061 |
Neg logit contributions | |
anson | -0.094 |
rem | -0.081 |
� | -0.077 |
lement | -0.070 |
roman | -0.070 |
Token | Profile |
Token ID | 13118 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
anson | +0.040 |
rem | +0.035 |
� | +0.033 |
lement | +0.031 |
roman | +0.030 |
Neg logit contributions | |
@@ | -0.025 |
� | -0.024 |
Krug | -0.024 |
Lie | -0.024 |
crossings | -0.024 |
Token | Joined |
Token ID | 16709 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
@@ | +0.479 |
� | +0.443 |
Krug | +0.442 |
Lie | +0.437 |
crossings | +0.428 |
Neg logit contributions | |
anson | -1.084 |
rem | -0.962 |
� | -0.918 |
lement | -0.857 |
roman | -0.857 |
Token | December |
Token ID | 3426 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
@@ | +1.913 |
Krug | +1.805 |
Nemesis | +1.781 |
Claim | +1.760 |
Scand | +1.756 |
Neg logit contributions | |
anson | -2.719 |
rem | -2.282 |
� | -2.192 |
uana | -2.034 |
roman | -2.016 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
@@ | +1.149 |
Krug | +1.125 |
Scand | +1.107 |
Nemesis | +1.105 |
Lie | +1.097 |
Neg logit contributions | |
anson | -1.071 |
rem | -0.875 |
� | -0.814 |
uana | -0.745 |
lement | -0.739 |
Token | United |
Token ID | 1578 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
@@ | +0.338 |
Krug | +0.325 |
Lie | +0.321 |
� | +0.320 |
Nemesis | +0.318 |
Neg logit contributions | |
anson | -0.400 |
rem | -0.339 |
� | -0.318 |
roman | -0.291 |
lement | -0.290 |
Token | States |
Token ID | 1829 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
@@ | +0.481 |
Krug | +0.450 |
COUN | +0.440 |
Lie | +0.438 |
Scand | +0.436 |
Neg logit contributions | |
anson | -1.298 |
rem | -1.151 |
� | -1.102 |
roman | -1.037 |
lement | -1.024 |
Token | 6 |
Token ID | 718 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
@@ | +0.860 |
Krug | +0.808 |
Nemesis | +0.796 |
Lie | +0.787 |
COUN | +0.784 |
Neg logit contributions | |
anson | -1.711 |
rem | -1.487 |
� | -1.440 |
roman | -1.334 |
lement | -1.325 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
@@ | +0.674 |
Krug | +0.653 |
Scand | +0.645 |
Lie | +0.643 |
Nemesis | +0.641 |
Neg logit contributions | |
anson | -0.614 |
rem | -0.508 |
� | -0.471 |
roman | -0.425 |
lement | -0.425 |
Token | United |
Token ID | 1578 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
@@ | +0.077 |
Krug | +0.074 |
Lie | +0.073 |
� | +0.073 |
Nemesis | +0.073 |
Neg logit contributions | |
anson | -0.091 |
rem | -0.077 |
� | -0.072 |
roman | -0.066 |
lement | -0.066 |
Token | States |
Token ID | 1829 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
@@ | +0.423 |
� | +0.395 |
Krug | +0.392 |
Lie | +0.388 |
crossings | +0.388 |
Neg logit contributions | |
anson | -0.852 |
rem | -0.755 |
� | -0.718 |
lement | -0.671 |
roman | -0.664 |
Token | 10 |
Token ID | 838 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
@@ | +0.879 |
Krug | +0.828 |
Nemesis | +0.815 |
Lie | +0.806 |
COUN | +0.802 |
Neg logit contributions | |
anson | -1.744 |
rem | -1.515 |
� | -1.466 |
roman | -1.360 |
lement | -1.351 |
Token | 53 |
Token ID | 4310 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
@@ | +0.270 |
Krug | +0.253 |
� | +0.251 |
Lie | +0.250 |
COUN | +0.247 |
Neg logit contributions | |
anson | -0.423 |
rem | -0.367 |
� | -0.350 |
roman | -0.321 |
lement | -0.320 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
@@ | +0.136 |
Krug | +0.126 |
Lie | +0.124 |
� | +0.124 |
Nemesis | +0.123 |
Neg logit contributions | |
anson | -0.220 |
rem | -0.190 |
� | -0.182 |
roman | -0.167 |
lement | -0.167 |
Token | # |
Token ID | 1303 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
@@ | +1.737 |
� | +1.644 |
Krug | +1.643 |
Lie | +1.627 |
crossings | +1.617 |
Neg logit contributions | |
anson | -2.750 |
rem | -2.403 |
� | -2.272 |
lement | -2.096 |
roman | -2.089 |
Token | 15 |
Token ID | 1314 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
@@ | +1.314 |
Lie | +1.268 |
Lt | +1.212 |
flat | +1.192 |
� | +1.162 |
Neg logit contributions | |
anson | -2.260 |
rem | -1.827 |
� | -1.822 |
roman | -1.744 |
lement | -1.715 |
Token | honestly |
Token ID | 12698 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
@@ | +0.102 |
Krug | +0.095 |
� | +0.095 |
Nemesis | +0.093 |
Lie | +0.093 |
Neg logit contributions | |
anson | -0.181 |
rem | -0.158 |
� | -0.152 |
roman | -0.141 |
lement | -0.140 |
Token | if |
Token ID | 611 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
anson | +0.431 |
rem | +0.374 |
� | +0.356 |
lement | +0.329 |
roman | +0.327 |
Neg logit contributions | |
@@ | -0.277 |
Krug | -0.260 |
� | -0.260 |
Lie | -0.257 |
crossings | -0.256 |
Token | we |
Token ID | 356 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
anson | +0.300 |
rem | +0.259 |
� | +0.247 |
roman | +0.226 |
lement | +0.225 |
Neg logit contributions | |
@@ | -0.216 |
Krug | -0.205 |
� | -0.205 |
Lie | -0.201 |
crossings | -0.201 |
Token | the |
Token ID | 262 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
@@ | +0.443 |
Krug | +0.427 |
� | +0.420 |
crossings | +0.416 |
Lie | +0.416 |
Neg logit contributions | |
anson | -0.624 |
rem | -0.539 |
� | -0.509 |
lement | -0.471 |
roman | -0.468 |
Token | quarterback |
Token ID | 7880 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
anson | +0.147 |
rem | +0.125 |
� | +0.118 |
lement | +0.109 |
roman | +0.109 |
Neg logit contributions | |
@@ | -0.109 |
Krug | -0.104 |
� | -0.103 |
crossings | -0.103 |
Lie | -0.102 |
Token | competition |
Token ID | 5449 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
@@ | +0.283 |
Krug | +0.268 |
� | +0.268 |
Lie | +0.266 |
crossings | +0.264 |
Neg logit contributions | |
anson | -0.410 |
rem | -0.353 |
� | -0.331 |
roman | -0.307 |
lement | -0.306 |
Token | between |
Token ID | 1022 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
anson | +0.091 |
rem | +0.078 |
� | +0.074 |
roman | +0.068 |
lement | +0.068 |
Neg logit contributions | |
@@ | -0.065 |
Krug | -0.062 |
crossings | -0.061 |
Lie | -0.061 |
� | -0.061 |
Token | Trevor |
Token ID | 25389 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
@@ | +0.034 |
Krug | +0.033 |
� | +0.032 |
Lie | +0.032 |
crossings | +0.032 |
Neg logit contributions | |
anson | -0.049 |
rem | -0.042 |
� | -0.040 |
lement | -0.037 |
roman | -0.036 |
Token | Siem |
Token ID | 45196 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
@@ | +0.291 |
� | +0.284 |
territory | +0.282 |
Krug | +0.279 |
Lie | +0.277 |
Neg logit contributions | |
anson | -0.342 |
rem | -0.294 |
� | -0.271 |
roman | -0.259 |
heng | -0.255 |
Token | ian |
Token ID | 666 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
anson | +4.119 |
rem | +3.643 |
� | +3.423 |
lement | +3.205 |
roman | +3.200 |
Neg logit contributions | |
@@ | -2.256 |
Krug | -2.137 |
� | -2.102 |
crossings | -2.089 |
Lie | -2.084 |
Token | and |
Token ID | 290 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
anson | +0.058 |
rem | +0.050 |
� | +0.046 |
lement | +0.042 |
roman | +0.042 |
Neg logit contributions | |
@@ | -0.050 |
Krug | -0.047 |
� | -0.047 |
Lie | -0.047 |
crossings | -0.047 |
Token | Pa |
Token ID | 11243 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
anson | +0.587 |
rem | +0.507 |
� | +0.478 |
lement | +0.440 |
roman | +0.439 |
Neg logit contributions | |
@@ | -0.425 |
Krug | -0.406 |
� | -0.405 |
Lie | -0.400 |
crossings | -0.397 |
Token | xton |
Token ID | 22874 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
anson | +0.881 |
rem | +0.684 |
� | +0.608 |
lement | +0.517 |
roman | +0.513 |
Neg logit contributions | |
@@ | -1.593 |
� | -1.543 |
Krug | -1.540 |
Lie | -1.536 |
crossings | -1.523 |
Token | Lynch |
Token ID | 16093 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
anson | +0.840 |
rem | +0.727 |
� | +0.681 |
roman | +0.626 |
lement | +0.623 |
Neg logit contributions | |
@@ | -0.652 |
� | -0.621 |
Krug | -0.619 |
Lie | -0.615 |
crossings | -0.606 |
Token | ow |
Token ID | 322 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
anson | +0.870 |
rem | +0.776 |
� | +0.766 |
lement | +0.712 |
uana | +0.701 |
Neg logit contributions | |
@@ | -0.256 |
� | -0.239 |
Lie | -0.239 |
Krug | -0.228 |
� | -0.221 |
Token | support |
Token ID | 1104 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
anson | +0.623 |
rem | +0.535 |
� | +0.504 |
lement | +0.458 |
roman | +0.458 |
Neg logit contributions | |
@@ | -0.519 |
Krug | -0.496 |
� | -0.495 |
Lie | -0.491 |
crossings | -0.488 |
Token | from |
Token ID | 422 |
Feature activation | -8.00e-02 |
Pos logit contributions | |
@@ | +0.407 |
� | +0.386 |
Krug | +0.384 |
Lie | +0.380 |
crossings | +0.378 |
Neg logit contributions | |
anson | -0.604 |
rem | -0.525 |
� | -0.499 |
lement | -0.459 |
roman | -0.456 |
Token | former |
Token ID | 1966 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
anson | +0.188 |
rem | +0.163 |
� | +0.154 |
lement | +0.142 |
roman | +0.141 |
Neg logit contributions | |
@@ | -0.134 |
� | -0.128 |
Krug | -0.127 |
Lie | -0.126 |
crossings | -0.125 |
Token | Ku |
Token ID | 12554 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
@@ | +0.331 |
Krug | +0.312 |
� | +0.310 |
crossings | +0.307 |
Lie | +0.307 |
Neg logit contributions | |
anson | -0.516 |
rem | -0.453 |
� | -0.436 |
roman | -0.402 |
lement | -0.402 |
Token | Klux |
Token ID | 45092 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
anson | +0.287 |
rem | +0.247 |
� | +0.239 |
roman | +0.221 |
uana | +0.218 |
Neg logit contributions | |
@@ | -0.178 |
� | -0.169 |
Krug | -0.169 |
Lie | -0.166 |
� | -0.165 |
Token | Klan |
Token ID | 31151 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
anson | +3.163 |
rem | +2.683 |
� | +2.520 |
lement | +2.264 |
roman | +2.262 |
Neg logit contributions | |
@@ | -3.015 |
� | -2.882 |
Krug | -2.879 |
Lie | -2.853 |
crossings | -2.830 |
Token | leader |
Token ID | 3554 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
anson | +1.200 |
rem | +1.045 |
� | +0.992 |
lement | +0.910 |
roman | +0.909 |
Neg logit contributions | |
@@ | -0.793 |
Krug | -0.751 |
� | -0.749 |
Lie | -0.742 |
crossings | -0.737 |
Token | David |
Token ID | 3271 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
anson | +0.339 |
rem | +0.286 |
� | +0.270 |
roman | +0.243 |
lement | +0.242 |
Neg logit contributions | |
@@ | -0.324 |
Krug | -0.310 |
� | -0.309 |
Lie | -0.306 |
crossings | -0.305 |
Token | Duke |
Token ID | 11083 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
anson | +1.050 |
rem | +0.934 |
� | +0.887 |
lement | +0.830 |
roman | +0.826 |
Neg logit contributions | |
@@ | -0.476 |
� | -0.441 |
Krug | -0.439 |
Lie | -0.435 |
crossings | -0.432 |
Token | . |
Token ID | 13 |
Feature activation | -8.00e-02 |
Pos logit contributions | |
anson | +1.973 |
rem | +1.694 |
� | +1.593 |
roman | +1.450 |
lement | +1.449 |
Neg logit contributions | |
@@ | -1.615 |
� | -1.537 |
Krug | -1.536 |
Lie | -1.522 |
crossings | -1.512 |
Token | Jets |
Token ID | 14728 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
anson | +0.127 |
rem | +0.110 |
� | +0.104 |
lement | +0.096 |
roman | +0.096 |
Neg logit contributions | |
@@ | -0.089 |
Krug | -0.085 |
� | -0.084 |
Lie | -0.083 |
Nemesis | -0.082 |
Token | � |
Token ID | 447 |
Feature activation | -7.97e-02 |
Pos logit contributions | |
anson | +0.627 |
rem | +0.537 |
� | +0.506 |
lement | +0.462 |
roman | +0.457 |
Neg logit contributions | |
@@ | -0.521 |
Krug | -0.496 |
� | -0.494 |
Lie | -0.490 |
� | -0.484 |
Token | � |
Token ID | 247 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
anson | +0.257 |
rem | +0.230 |
� | +0.222 |
lement | +0.210 |
roman | +0.209 |
Neg logit contributions | |
@@ | -0.068 |
� | -0.064 |
� | -0.062 |
Lie | -0.058 |
Krug | -0.057 |
Token | un |
Token ID | 555 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
anson | +0.152 |
rem | +0.131 |
� | +0.124 |
lement | +0.114 |
roman | +0.113 |
Neg logit contributions | |
@@ | -0.109 |
� | -0.104 |
Krug | -0.104 |
Lie | -0.103 |
� | -0.102 |
Token | quest |
Token ID | 6138 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
@@ | +1.111 |
Lie | +1.059 |
Krug | +1.054 |
� | +1.052 |
crossings | +1.036 |
Neg logit contributions | |
anson | -1.119 |
rem | -0.921 |
� | -0.878 |
lement | -0.783 |
roman | -0.779 |
Token | ioned |
Token ID | 14994 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
anson | +0.614 |
rem | +0.539 |
� | +0.515 |
lement | +0.477 |
roman | +0.474 |
Neg logit contributions | |
@@ | -0.320 |
� | -0.298 |
Krug | -0.298 |
Lie | -0.296 |
crossings | -0.291 |
Token | best |
Token ID | 1266 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
anson | +3.100 |
rem | +2.708 |
� | +2.555 |
roman | +2.339 |
lement | +2.338 |
Neg logit contributions | |
@@ | -2.104 |
� | -1.999 |
Krug | -1.976 |
crossings | -1.972 |
Lie | -1.969 |
Token | sniper |
Token ID | 26160 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
anson | +0.248 |
rem | +0.216 |
� | +0.205 |
lement | +0.190 |
roman | +0.188 |
Neg logit contributions | |
@@ | -0.157 |
Krug | -0.148 |
� | -0.148 |
Lie | -0.146 |
crossings | -0.145 |
Token | , |
Token ID | 11 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
anson | +1.109 |
rem | +0.965 |
� | +0.913 |
lement | +0.839 |
roman | +0.837 |
Neg logit contributions | |
@@ | -0.742 |
Krug | -0.701 |
� | -0.701 |
Lie | -0.694 |
crossings | -0.689 |
Token | and |
Token ID | 290 |
Feature activation | -3.77e-03 |
Pos logit contributions | |
anson | +0.070 |
rem | +0.060 |
� | +0.056 |
lement | +0.052 |
roman | +0.052 |
Neg logit contributions | |
@@ | -0.054 |
Krug | -0.051 |
� | -0.051 |
Lie | -0.050 |
crossings | -0.050 |
Token | he |
Token ID | 339 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
anson | +0.009 |
rem | +0.007 |
� | +0.007 |
lement | +0.006 |
roman | +0.006 |
Neg logit contributions | |
@@ | -0.007 |
Krug | -0.006 |
� | -0.006 |
Lie | -0.006 |
crossings | -0.006 |
Token | ."[ |
Token ID | 18161 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
anson | +1.582 |
rem | +1.352 |
� | +1.270 |
roman | +1.158 |
lement | +1.154 |
Neg logit contributions | |
@@ | -1.328 |
Krug | -1.257 |
� | -1.255 |
Lie | -1.248 |
crossings | -1.245 |
Token | 45 |
Token ID | 2231 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
@@ | +0.282 |
Lie | +0.266 |
� | +0.264 |
Krug | +0.261 |
� | +0.256 |
Neg logit contributions | |
anson | -0.500 |
rem | -0.426 |
� | -0.410 |
roman | -0.383 |
uana | -0.379 |
Token | ] |
Token ID | 60 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
@@ | +0.348 |
Lie | +0.324 |
Krug | +0.315 |
� | +0.306 |
COUN | +0.303 |
Neg logit contributions | |
anson | -0.858 |
rem | -0.735 |
� | -0.718 |
roman | -0.682 |
uana | -0.674 |
Token | Damascus |
Token ID | 21028 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
@@ | +0.533 |
Krug | +0.503 |
Lie | +0.495 |
Claim | +0.491 |
Nemesis | +0.490 |
Neg logit contributions | |
anson | -0.857 |
rem | -0.732 |
� | -0.696 |
roman | -0.650 |
uana | -0.641 |
Token | rel |
Token ID | 823 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
@@ | +0.984 |
Krug | +0.932 |
crossings | +0.923 |
Lie | +0.912 |
territory | +0.909 |
Neg logit contributions | |
anson | -1.678 |
rem | -1.452 |
� | -1.392 |
roman | -1.285 |
lement | -1.276 |
Token | ented |
Token ID | 4714 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
@@ | +0.932 |
Krug | +0.879 |
� | +0.866 |
Lie | +0.862 |
crossings | +0.859 |
Neg logit contributions | |
anson | -1.125 |
rem | -0.971 |
� | -0.911 |
roman | -0.832 |
lement | -0.821 |
Token | . |
Token ID | 13 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
anson | +2.975 |
rem | +2.585 |
� | +2.438 |
lement | +2.240 |
roman | +2.227 |
Neg logit contributions | |
@@ | -2.136 |
� | -2.027 |
Krug | -2.019 |
Lie | -2.004 |
crossings | -1.982 |
Token | Cos |
Token ID | 10437 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
anson | +0.194 |
rem | +0.168 |
� | +0.160 |
roman | +0.148 |
lement | +0.147 |
Neg logit contributions | |
@@ | -0.115 |
Krug | -0.108 |
Lie | -0.106 |
� | -0.106 |
Nemesis | -0.105 |
Token | kun |
Token ID | 28374 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
@@ | +0.381 |
Krug | +0.364 |
� | +0.356 |
Lie | +0.356 |
Nemesis | +0.350 |
Neg logit contributions | |
anson | -0.507 |
rem | -0.435 |
� | -0.416 |
lement | -0.382 |
roman | -0.377 |
Token | K |
Token ID | 509 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
@@ | +0.247 |
� | +0.227 |
Krug | +0.227 |
Lie | +0.223 |
crossings | +0.221 |
Neg logit contributions | |
anson | -0.644 |
rem | -0.575 |
� | -0.550 |
lement | -0.515 |
roman | -0.514 |
Token | irc |
Token ID | 1980 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
@@ | +0.083 |
Lie | +0.079 |
Krug | +0.078 |
crossings | +0.078 |
� | +0.078 |
Neg logit contributions | |
anson | -0.062 |
rem | -0.050 |
� | -0.047 |
roman | -0.041 |
lement | -0.041 |
Token | that |
Token ID | 326 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
anson | +0.298 |
rem | +0.253 |
� | +0.237 |
lement | +0.217 |
roman | +0.216 |
Neg logit contributions | |
@@ | -0.256 |
Krug | -0.245 |
� | -0.243 |
Lie | -0.241 |
crossings | -0.240 |
Token | Republican |
Token ID | 3415 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
@@ | +0.163 |
Krug | +0.154 |
� | +0.154 |
Lie | +0.153 |
crossings | +0.152 |
Neg logit contributions | |
anson | -0.237 |
rem | -0.205 |
� | -0.193 |
lement | -0.179 |
roman | -0.178 |
Token | Rep |
Token ID | 1432 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
@@ | +0.611 |
Krug | +0.578 |
� | +0.576 |
territory | +0.572 |
crossings | +0.571 |
Neg logit contributions | |
anson | -0.921 |
rem | -0.803 |
� | -0.761 |
lement | -0.704 |
roman | -0.700 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
anson | +0.787 |
rem | +0.667 |
� | +0.632 |
roman | +0.583 |
lement | +0.581 |
Neg logit contributions | |
@@ | -0.597 |
Lie | -0.564 |
Krug | -0.563 |
� | -0.555 |
� | -0.550 |
Token | Ryan |
Token ID | 6047 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
@@ | +0.191 |
Krug | +0.184 |
� | +0.180 |
Lie | +0.179 |
Nemesis | +0.178 |
Neg logit contributions | |
anson | -0.246 |
rem | -0.211 |
� | -0.198 |
roman | -0.183 |
lement | -0.182 |
Token | Zin |
Token ID | 36971 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
@@ | +0.418 |
� | +0.395 |
Krug | +0.393 |
crossings | +0.385 |
Lie | +0.384 |
Neg logit contributions | |
anson | -0.662 |
rem | -0.577 |
� | -0.544 |
lement | -0.508 |
roman | -0.506 |
Token | ke |
Token ID | 365 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
anson | +2.513 |
rem | +2.098 |
� | +1.974 |
roman | +1.775 |
lement | +1.773 |
Neg logit contributions | |
@@ | -2.547 |
� | -2.457 |
Krug | -2.421 |
Lie | -2.415 |
� | -2.398 |
Token | won |
Token ID | 1839 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
anson | +0.913 |
rem | +0.802 |
� | +0.760 |
lement | +0.705 |
roman | +0.704 |
Neg logit contributions | |
@@ | -0.506 |
� | -0.476 |
Krug | -0.474 |
Lie | -0.469 |
crossings | -0.465 |
Token | by |
Token ID | 416 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
anson | +0.385 |
rem | +0.337 |
� | +0.319 |
roman | +0.298 |
lement | +0.295 |
Neg logit contributions | |
@@ | -0.225 |
Krug | -0.212 |
� | -0.211 |
Lie | -0.210 |
� | -0.209 |
Token | 15 |
Token ID | 1315 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
anson | +0.439 |
rem | +0.383 |
� | +0.367 |
lement | +0.336 |
roman | +0.336 |
Neg logit contributions | |
@@ | -0.263 |
Krug | -0.246 |
crossings | -0.243 |
� | -0.243 |
territory | -0.241 |
Token | points |
Token ID | 2173 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
anson | +0.616 |
rem | +0.535 |
� | +0.511 |
roman | +0.473 |
lement | +0.471 |
Neg logit contributions | |
@@ | -0.376 |
Krug | -0.353 |
� | -0.352 |
crossings | -0.350 |
Lie | -0.348 |
Token | re |
Token ID | 260 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
@@ | +0.048 |
� | +0.045 |
Lie | +0.045 |
� | +0.044 |
Krug | +0.044 |
Neg logit contributions | |
anson | -0.080 |
rem | -0.069 |
� | -0.066 |
roman | -0.061 |
lement | -0.061 |
Token | crying |
Token ID | 13774 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
@@ | +0.372 |
� | +0.353 |
Lie | +0.349 |
Krug | +0.347 |
crossings | +0.345 |
Neg logit contributions | |
anson | -0.558 |
rem | -0.481 |
� | -0.457 |
roman | -0.419 |
lement | -0.419 |
Token | alone |
Token ID | 3436 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
anson | +0.403 |
rem | +0.348 |
� | +0.326 |
lement | +0.298 |
roman | +0.298 |
Neg logit contributions | |
@@ | -0.312 |
� | -0.294 |
Krug | -0.294 |
Lie | -0.291 |
crossings | -0.291 |
Token | into |
Token ID | 656 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
@@ | +1.681 |
� | +1.564 |
Lie | +1.561 |
crossings | +1.558 |
Krug | +1.541 |
Neg logit contributions | |
anson | -2.337 |
rem | -1.992 |
� | -1.877 |
roman | -1.722 |
aston | -1.721 |
Token | a |
Token ID | 257 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
@@ | +0.096 |
� | +0.090 |
Krug | +0.090 |
Lie | +0.089 |
crossings | +0.089 |
Neg logit contributions | |
anson | -0.132 |
rem | -0.114 |
� | -0.107 |
roman | -0.099 |
lement | -0.098 |
Token | pint |
Token ID | 35245 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
@@ | +0.210 |
� | +0.196 |
Krug | +0.195 |
Lie | +0.194 |
crossings | +0.191 |
Neg logit contributions | |
anson | -0.366 |
rem | -0.320 |
� | -0.304 |
roman | -0.282 |
lement | -0.281 |
Token | of |
Token ID | 286 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
anson | +3.039 |
rem | +2.676 |
� | +2.537 |
lement | +2.354 |
roman | +2.318 |
Neg logit contributions | |
@@ | -1.939 |
Krug | -1.836 |
Lie | -1.836 |
� | -1.832 |
crossings | -1.826 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
anson | +0.045 |
rem | +0.039 |
� | +0.037 |
roman | +0.034 |
lement | +0.034 |
Neg logit contributions | |
@@ | -0.029 |
Krug | -0.028 |
� | -0.027 |
Lie | -0.027 |
crossings | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
@@ | +0.419 |
Krug | +0.397 |
� | +0.393 |
Lie | +0.391 |
crossings | +0.383 |
Neg logit contributions | |
anson | -0.685 |
rem | -0.598 |
� | -0.566 |
roman | -0.525 |
lement | -0.524 |
Token | Jerry |
Token ID | 13075 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
@@ | +0.058 |
� | +0.055 |
Krug | +0.055 |
Lie | +0.055 |
� | +0.054 |
Neg logit contributions | |
anson | -0.072 |
rem | -0.061 |
� | -0.057 |
roman | -0.053 |
lement | -0.052 |
Token | � |
Token ID | 447 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
@@ | +0.262 |
Krug | +0.247 |
Lie | +0.246 |
� | +0.245 |
crossings | +0.242 |
Neg logit contributions | |
anson | -0.401 |
rem | -0.347 |
� | -0.327 |
roman | -0.304 |
lement | -0.302 |
Token |
|
Token ID | 198 |
Feature activation | -9.92e-02 |
Pos logit contributions | |
anson | +0.134 |
rem | +0.112 |
� | +0.108 |
roman | +0.101 |
uana | +0.100 |
Neg logit contributions | |
@@ | -0.091 |
Lie | -0.085 |
Krug | -0.083 |
Lt | -0.081 |
� | -0.081 |
Token | The |
Token ID | 464 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
anson | +0.241 |
rem | +0.207 |
� | +0.197 |
lement | +0.182 |
roman | +0.181 |
Neg logit contributions | |
@@ | -0.159 |
Lie | -0.152 |
� | -0.151 |
Krug | -0.150 |
� | -0.147 |
Token | trade |
Token ID | 3292 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
anson | +0.242 |
rem | +0.209 |
� | +0.198 |
lement | +0.182 |
roman | +0.182 |
Neg logit contributions | |
@@ | -0.171 |
Krug | -0.162 |
� | -0.161 |
Lie | -0.159 |
crossings | -0.159 |
Token | also |
Token ID | 635 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
anson | +0.336 |
rem | +0.290 |
� | +0.276 |
lement | +0.253 |
roman | +0.252 |
Neg logit contributions | |
@@ | -0.228 |
Krug | -0.218 |
� | -0.214 |
crossings | -0.214 |
Lie | -0.212 |
Token | reun |
Token ID | 15398 |
Feature activation | +7.37e-01 |
Pos logit contributions | |
anson | +0.290 |
rem | +0.251 |
� | +0.242 |
roman | +0.223 |
lement | +0.221 |
Neg logit contributions | |
@@ | -0.174 |
Krug | -0.165 |
crossings | -0.164 |
� | -0.160 |
territory | -0.158 |
Token | ites |
Token ID | 2737 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
@@ | +1.104 |
Krug | +1.042 |
� | +1.027 |
crossings | +1.016 |
Lie | +1.009 |
Neg logit contributions | |
anson | -1.858 |
rem | -1.606 |
� | -1.535 |
roman | -1.430 |
lement | -1.399 |
Token | Fel |
Token ID | 13937 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
anson | +2.894 |
rem | +2.519 |
� | +2.381 |
lement | +2.185 |
roman | +2.180 |
Neg logit contributions | |
@@ | -1.949 |
� | -1.848 |
Lie | -1.830 |
Krug | -1.824 |
crossings | -1.810 |
Token | ton |
Token ID | 1122 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
@@ | +0.357 |
� | +0.340 |
Krug | +0.337 |
Lie | +0.337 |
crossings | +0.336 |
Neg logit contributions | |
anson | -0.347 |
rem | -0.292 |
� | -0.273 |
roman | -0.245 |
lement | -0.244 |
Token | and |
Token ID | 290 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
anson | +1.137 |
rem | +0.984 |
� | +0.929 |
lement | +0.850 |
roman | +0.847 |
Neg logit contributions | |
@@ | -0.834 |
� | -0.790 |
Krug | -0.789 |
Lie | -0.783 |
crossings | -0.777 |
Token | Thomas |
Token ID | 5658 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
anson | +0.719 |
rem | +0.627 |
� | +0.593 |
lement | +0.545 |
roman | +0.544 |
Neg logit contributions | |
@@ | -0.476 |
� | -0.449 |
Krug | -0.449 |
Lie | -0.445 |
crossings | -0.441 |
Token | with |
Token ID | 351 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
anson | +0.860 |
rem | +0.751 |
� | +0.712 |
lement | +0.656 |
roman | +0.654 |
Neg logit contributions | |
@@ | -0.544 |
Krug | -0.514 |
� | -0.513 |
Lie | -0.507 |
crossings | -0.503 |
Token | — |
Token ID | 851 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
@@ | +0.273 |
� | +0.258 |
Krug | +0.258 |
crossings | +0.257 |
Lie | +0.256 |
Neg logit contributions | |
anson | -0.311 |
rem | -0.263 |
� | -0.247 |
roman | -0.224 |
lement | -0.222 |
Token | are |
Token ID | 389 |
Feature activation | +5.13e-03 |
Pos logit contributions | |
anson | +0.129 |
rem | +0.111 |
� | +0.107 |
roman | +0.099 |
lement | +0.098 |
Neg logit contributions | |
@@ | -0.083 |
� | -0.078 |
Krug | -0.078 |
crossings | -0.077 |
Lie | -0.077 |
Token | no |
Token ID | 645 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
@@ | +0.009 |
Krug | +0.008 |
� | +0.008 |
crossings | +0.008 |
Lie | +0.008 |
Neg logit contributions | |
anson | -0.012 |
rem | -0.010 |
� | -0.010 |
roman | -0.009 |
lement | -0.009 |
Token | longer |
Token ID | 2392 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
anson | +0.425 |
rem | +0.363 |
� | +0.340 |
lement | +0.312 |
roman | +0.311 |
Neg logit contributions | |
@@ | -0.350 |
� | -0.336 |
Krug | -0.334 |
Lie | -0.330 |
crossings | -0.330 |
Token | unt |
Token ID | 1418 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
anson | +0.291 |
rem | +0.251 |
� | +0.238 |
roman | +0.220 |
lement | +0.218 |
Neg logit contributions | |
@@ | -0.193 |
Krug | -0.182 |
� | -0.180 |
crossings | -0.180 |
Lie | -0.178 |
Token | ouch |
Token ID | 7673 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
@@ | +0.370 |
Krug | +0.348 |
� | +0.348 |
Lie | +0.347 |
crossings | +0.343 |
Neg logit contributions | |
anson | -0.284 |
rem | -0.225 |
� | -0.216 |
aston | -0.186 |
lement | -0.185 |
Token | able |
Token ID | 540 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
anson | +2.236 |
rem | +1.872 |
� | +1.735 |
lement | +1.544 |
roman | +1.536 |
Neg logit contributions | |
@@ | -2.551 |
Krug | -2.448 |
� | -2.446 |
Lie | -2.429 |
crossings | -2.414 |
Token | . |
Token ID | 13 |
Feature activation | +6.10e-03 |
Pos logit contributions | |
anson | +0.301 |
rem | +0.259 |
� | +0.244 |
roman | +0.224 |
lement | +0.222 |
Neg logit contributions | |
@@ | -0.227 |
Krug | -0.215 |
� | -0.215 |
crossings | -0.214 |
Lie | -0.213 |
Token | Further |
Token ID | 7735 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
@@ | +0.010 |
Krug | +0.010 |
Lie | +0.010 |
� | +0.010 |
COUN | +0.009 |
Neg logit contributions | |
anson | -0.015 |
rem | -0.012 |
� | -0.012 |
roman | -0.011 |
lement | -0.011 |
Token | tax |
Token ID | 1687 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
anson | +0.857 |
rem | +0.748 |
� | +0.708 |
lement | +0.652 |
roman | +0.650 |
Neg logit contributions | |
@@ | -0.549 |
� | -0.518 |
Krug | -0.518 |
Lie | -0.513 |
crossings | -0.509 |
Token | rises |
Token ID | 16736 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
@@ | +0.231 |
� | +0.221 |
Krug | +0.221 |
crossings | +0.220 |
Lie | +0.218 |
Neg logit contributions | |
anson | -0.250 |
rem | -0.209 |
� | -0.197 |
roman | -0.178 |
lement | -0.177 |
Token | south |
Token ID | 5366 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
@@ | +0.349 |
� | +0.332 |
Krug | +0.332 |
Lie | +0.329 |
crossings | +0.327 |
Neg logit contributions | |
anson | -0.416 |
rem | -0.356 |
� | -0.335 |
lement | -0.304 |
roman | -0.303 |
Token | as |
Token ID | 355 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
anson | +0.936 |
rem | +0.812 |
� | +0.763 |
lement | +0.698 |
roman | +0.697 |
Neg logit contributions | |
@@ | -0.677 |
� | -0.645 |
Krug | -0.639 |
Lie | -0.639 |
crossings | -0.627 |
Token | the |
Token ID | 262 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
anson | +0.741 |
rem | +0.648 |
� | +0.616 |
lement | +0.569 |
roman | +0.568 |
Neg logit contributions | |
@@ | -0.438 |
Krug | -0.415 |
� | -0.410 |
Lie | -0.406 |
crossings | -0.404 |
Token | city |
Token ID | 1748 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
anson | +0.328 |
rem | +0.288 |
� | +0.274 |
lement | +0.253 |
roman | +0.253 |
Neg logit contributions | |
@@ | -0.187 |
Krug | -0.177 |
� | -0.176 |
crossings | -0.174 |
Lie | -0.174 |
Token | of |
Token ID | 286 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
@@ | +0.031 |
Krug | +0.029 |
� | +0.029 |
crossings | +0.029 |
Lie | +0.029 |
Neg logit contributions | |
anson | -0.061 |
rem | -0.053 |
� | -0.051 |
lement | -0.047 |
roman | -0.047 |
Token | Zin |
Token ID | 36971 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
anson | +0.558 |
rem | +0.490 |
� | +0.467 |
lement | +0.434 |
roman | +0.432 |
Neg logit contributions | |
@@ | -0.294 |
Krug | -0.278 |
� | -0.274 |
Lie | -0.271 |
crossings | -0.270 |
Token | , |
Token ID | 11 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
anson | +2.715 |
rem | +2.355 |
� | +2.211 |
lement | +2.020 |
roman | +2.016 |
Neg logit contributions | |
@@ | -2.066 |
� | -1.960 |
Krug | -1.953 |
Lie | -1.940 |
crossings | -1.926 |
Token | and |
Token ID | 290 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
anson | +0.112 |
rem | +0.097 |
� | +0.091 |
lement | +0.084 |
roman | +0.083 |
Neg logit contributions | |
@@ | -0.087 |
Krug | -0.083 |
crossings | -0.082 |
Nemesis | -0.081 |
Lie | -0.081 |
Token | has |
Token ID | 468 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
anson | +0.335 |
rem | +0.290 |
� | +0.274 |
lement | +0.254 |
roman | +0.253 |
Neg logit contributions | |
@@ | -0.227 |
Krug | -0.218 |
crossings | -0.216 |
� | -0.213 |
Lie | -0.211 |
Token | even |
Token ID | 772 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
anson | +0.605 |
rem | +0.522 |
� | +0.493 |
roman | +0.455 |
lement | +0.452 |
Neg logit contributions | |
@@ | -0.415 |
Krug | -0.393 |
crossings | -0.391 |
� | -0.390 |
territory | -0.384 |
Token | landed |
Token ID | 11406 |
Feature activation | +6.69e-02 |
Pos logit contributions | |
anson | +0.100 |
rem | +0.086 |
� | +0.082 |
roman | +0.075 |
lement | +0.075 |
Neg logit contributions | |
@@ | -0.065 |
Krug | -0.062 |
� | -0.061 |
crossings | -0.061 |
Lie | -0.060 |
Token | them |
Token ID | 606 |
Feature activation | -6.02e-03 |
Pos logit contributions | |
anson | +0.209 |
rem | +0.181 |
� | +0.172 |
lement | +0.158 |
roman | +0.157 |
Neg logit contributions | |
@@ | -0.149 |
Krug | -0.143 |
crossings | -0.140 |
� | -0.140 |
Lie | -0.138 |
Token | , |
Token ID | 11 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
anson | +0.015 |
rem | +0.013 |
� | +0.012 |
roman | +0.011 |
lement | +0.011 |
Neg logit contributions | |
@@ | -0.010 |
Krug | -0.009 |
� | -0.009 |
crossings | -0.009 |
Lie | -0.009 |
Token | they |
Token ID | 484 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
@@ | +0.105 |
Krug | +0.100 |
� | +0.099 |
crossings | +0.098 |
Lie | +0.097 |
Neg logit contributions | |
anson | -0.140 |
rem | -0.121 |
� | -0.115 |
lement | -0.105 |
roman | -0.104 |
Token | were |
Token ID | 547 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
anson | +0.050 |
rem | +0.043 |
� | +0.041 |
roman | +0.038 |
lement | +0.038 |
Neg logit contributions | |
@@ | -0.033 |
� | -0.031 |
Krug | -0.031 |
Lie | -0.031 |
crossings | -0.031 |
Token | unt |
Token ID | 1418 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
anson | +0.222 |
rem | +0.191 |
� | +0.183 |
lement | +0.169 |
roman | +0.168 |
Neg logit contributions | |
@@ | -0.138 |
Krug | -0.130 |
� | -0.130 |
Lie | -0.128 |
crossings | -0.128 |
Token | ouch |
Token ID | 7673 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
@@ | +0.533 |
Lie | +0.515 |
Krug | +0.510 |
� | +0.509 |
Nemesis | +0.504 |
Neg logit contributions | |
anson | -0.280 |
rem | -0.205 |
� | -0.196 |
roman | -0.155 |
lement | -0.154 |
Token | able |
Token ID | 540 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
anson | +2.265 |
rem | +1.896 |
� | +1.761 |
lement | +1.571 |
roman | +1.566 |
Neg logit contributions | |
@@ | -2.494 |
� | -2.389 |
Krug | -2.388 |
Lie | -2.371 |
crossings | -2.357 |
Token | . |
Token ID | 13 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
anson | +0.399 |
rem | +0.342 |
� | +0.324 |
lement | +0.295 |
roman | +0.294 |
Neg logit contributions | |
@@ | -0.313 |
Krug | -0.297 |
� | -0.296 |
Lie | -0.294 |
crossings | -0.293 |
Token |
|
Token ID | 198 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
@@ | +0.065 |
Krug | +0.062 |
� | +0.061 |
Lie | +0.061 |
COUN | +0.060 |
Neg logit contributions | |
anson | -0.094 |
rem | -0.080 |
� | -0.076 |
roman | -0.070 |
lement | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
anson | +0.103 |
rem | +0.086 |
� | +0.083 |
roman | +0.077 |
uana | +0.077 |
Neg logit contributions | |
@@ | -0.069 |
Lie | -0.065 |
Krug | -0.063 |
Lt | -0.062 |
territory | -0.062 |
Token | Still |
Token ID | 9590 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
anson | +0.157 |
rem | +0.133 |
� | +0.127 |
lement | +0.118 |
roman | +0.117 |
Neg logit contributions | |
@@ | -0.101 |
Lie | -0.097 |
� | -0.095 |
Krug | -0.095 |
� | -0.093 |
Token | He |
Token ID | 679 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
@@ | +0.024 |
Krug | +0.023 |
Lie | +0.022 |
� | +0.022 |
Nemesis | +0.022 |
Neg logit contributions | |
anson | -0.037 |
rem | -0.032 |
� | -0.030 |
roman | -0.028 |
lement | -0.028 |
Token | was |
Token ID | 373 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
anson | +0.041 |
rem | +0.035 |
� | +0.033 |
lement | +0.030 |
roman | +0.030 |
Neg logit contributions | |
@@ | -0.030 |
� | -0.029 |
Krug | -0.029 |
Lie | -0.028 |
crossings | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
anson | +0.182 |
rem | +0.158 |
� | +0.151 |
lement | +0.139 |
roman | +0.139 |
Neg logit contributions | |
@@ | -0.118 |
Krug | -0.112 |
� | -0.111 |
Lie | -0.110 |
crossings | -0.109 |
Token | major |
Token ID | 1688 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
@@ | +0.215 |
Krug | +0.203 |
� | +0.201 |
Lie | +0.199 |
Nemesis | +0.197 |
Neg logit contributions | |
anson | -0.330 |
rem | -0.285 |
� | -0.272 |
roman | -0.250 |
lement | -0.249 |
Token | trade |
Token ID | 3292 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
@@ | +0.505 |
Krug | +0.482 |
� | +0.477 |
Lie | +0.472 |
Nemesis | +0.471 |
Neg logit contributions | |
anson | -0.728 |
rem | -0.626 |
� | -0.601 |
roman | -0.548 |
lement | -0.545 |
Token | asset |
Token ID | 11171 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
anson | +0.422 |
rem | +0.364 |
� | +0.346 |
lement | +0.314 |
roman | +0.314 |
Neg logit contributions | |
@@ | -0.305 |
Krug | -0.289 |
� | -0.288 |
Lie | -0.287 |
crossings | -0.282 |
Token | . |
Token ID | 13 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
anson | +2.660 |
rem | +2.302 |
� | +2.171 |
lement | +1.988 |
roman | +1.983 |
Neg logit contributions | |
@@ | -1.948 |
� | -1.847 |
Krug | -1.846 |
Lie | -1.830 |
crossings | -1.816 |
Token | And |
Token ID | 843 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
anson | +0.029 |
rem | +0.025 |
� | +0.024 |
roman | +0.022 |
lement | +0.022 |
Neg logit contributions | |
@@ | -0.019 |
Krug | -0.018 |
� | -0.018 |
Lie | -0.018 |
Nemesis | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
@@ | +0.075 |
Krug | +0.071 |
� | +0.070 |
Lie | +0.069 |
crossings | +0.069 |
Neg logit contributions | |
anson | -0.103 |
rem | -0.089 |
� | -0.084 |
lement | -0.078 |
roman | -0.077 |
Token | Padres |
Token ID | 40188 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
anson | +0.183 |
rem | +0.158 |
� | +0.151 |
lement | +0.139 |
roman | +0.138 |
Neg logit contributions | |
@@ | -0.127 |
Krug | -0.120 |
� | -0.119 |
crossings | -0.118 |
Lie | -0.118 |
Token | should |
Token ID | 815 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
@@ | +0.555 |
Krug | +0.523 |
� | +0.519 |
Lie | +0.514 |
crossings | +0.510 |
Neg logit contributions | |
anson | -0.828 |
rem | -0.717 |
� | -0.682 |
lement | -0.626 |
roman | -0.625 |
Token | did |
Token ID | 750 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
anson | +0.154 |
rem | +0.133 |
� | +0.126 |
roman | +0.116 |
lement | +0.116 |
Neg logit contributions | |
@@ | -0.103 |
� | -0.097 |
Krug | -0.097 |
Lie | -0.096 |
crossings | -0.095 |
Token | just |
Token ID | 655 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
anson | +0.362 |
rem | +0.313 |
� | +0.296 |
roman | +0.271 |
lement | +0.271 |
Neg logit contributions | |
@@ | -0.247 |
� | -0.233 |
Krug | -0.232 |
Lie | -0.231 |
crossings | -0.229 |
Token | that |
Token ID | 326 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
anson | +0.605 |
rem | +0.504 |
� | +0.468 |
roman | +0.415 |
lement | +0.413 |
Neg logit contributions | |
@@ | -0.697 |
� | -0.667 |
Krug | -0.666 |
Lie | -0.662 |
crossings | -0.657 |
Token | – |
Token ID | 784 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
anson | +0.409 |
rem | +0.352 |
� | +0.332 |
roman | +0.304 |
lement | +0.304 |
Neg logit contributions | |
@@ | -0.301 |
Krug | -0.286 |
� | -0.285 |
Lie | -0.282 |
crossings | -0.281 |
Token | via |
Token ID | 2884 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
anson | +0.391 |
rem | +0.337 |
� | +0.319 |
roman | +0.292 |
lement | +0.292 |
Neg logit contributions | |
@@ | -0.288 |
Krug | -0.276 |
� | -0.275 |
Lie | -0.272 |
crossings | -0.270 |
Token | Bray |
Token ID | 43050 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
@@ | +0.295 |
Krug | +0.282 |
� | +0.280 |
Lie | +0.277 |
crossings | +0.276 |
Neg logit contributions | |
anson | -0.393 |
rem | -0.340 |
� | -0.321 |
lement | -0.293 |
roman | -0.293 |
Token | den |
Token ID | 6559 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
anson | +2.168 |
rem | +1.831 |
� | +1.688 |
lement | +1.513 |
roman | +1.497 |
Neg logit contributions | |
@@ | -2.399 |
� | -2.281 |
Lie | -2.262 |
Krug | -2.259 |
OHN | -2.249 |
Token | Sc |
Token ID | 1446 |
Feature activation | +9.80e-02 |
Pos logit contributions | |
anson | +0.434 |
rem | +0.381 |
� | +0.367 |
roman | +0.340 |
lement | +0.338 |
Neg logit contributions | |
@@ | -0.223 |
Krug | -0.215 |
� | -0.209 |
Scand | -0.206 |
Nemesis | -0.205 |
Token | hen |
Token ID | 831 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
@@ | +0.264 |
Krug | +0.256 |
Lie | +0.254 |
� | +0.254 |
crossings | +0.252 |
Neg logit contributions | |
anson | -0.130 |
rem | -0.097 |
� | -0.089 |
lement | -0.072 |
roman | -0.071 |
Token | n |
Token ID | 77 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
@@ | +0.582 |
Krug | +0.535 |
� | +0.532 |
Lie | +0.529 |
crossings | +0.517 |
Neg logit contributions | |
anson | -1.234 |
rem | -1.097 |
� | -1.045 |
roman | -0.964 |
lement | -0.961 |
Token | � |
Token ID | 447 |
Feature activation | -9.74e-03 |
Pos logit contributions | |
@@ | +0.336 |
Krug | +0.322 |
� | +0.322 |
Lie | +0.316 |
crossings | +0.316 |
Neg logit contributions | |
anson | -0.508 |
rem | -0.443 |
� | -0.418 |
roman | -0.385 |
lement | -0.384 |
Token | ask |
Token ID | 2093 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
anson | +1.195 |
rem | +1.033 |
� | +0.984 |
lement | +0.901 |
roman | +0.897 |
Neg logit contributions | |
@@ | -0.849 |
Krug | -0.812 |
� | -0.800 |
Lie | -0.799 |
crossings | -0.793 |
Token | y |
Token ID | 88 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
@@ | +0.134 |
� | +0.127 |
Krug | +0.127 |
Lie | +0.125 |
crossings | +0.125 |
Neg logit contributions | |
anson | -0.181 |
rem | -0.158 |
� | -0.148 |
lement | -0.136 |
roman | -0.135 |
Token | from |
Token ID | 422 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
anson | +0.153 |
rem | +0.133 |
� | +0.126 |
lement | +0.116 |
roman | +0.115 |
Neg logit contributions | |
@@ | -0.105 |
Krug | -0.099 |
� | -0.099 |
Lie | -0.098 |
crossings | -0.097 |
Token | Steel |
Token ID | 7851 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
anson | +0.335 |
rem | +0.289 |
� | +0.274 |
lement | +0.254 |
roman | +0.251 |
Neg logit contributions | |
@@ | -0.237 |
Krug | -0.225 |
� | -0.224 |
Lie | -0.220 |
crossings | -0.220 |
Token | Pole |
Token ID | 29385 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
anson | +0.118 |
rem | +0.102 |
� | +0.096 |
lement | +0.089 |
roman | +0.088 |
Neg logit contributions | |
@@ | -0.081 |
� | -0.077 |
Krug | -0.077 |
Lie | -0.076 |
crossings | -0.076 |
Token | Bath |
Token ID | 24967 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
anson | +0.123 |
rem | +0.106 |
� | +0.100 |
lement | +0.093 |
roman | +0.093 |
Neg logit contributions | |
@@ | -0.087 |
Krug | -0.083 |
� | -0.083 |
Lie | -0.082 |
crossings | -0.082 |
Token | tub |
Token ID | 37995 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
anson | +2.623 |
rem | +2.274 |
� | +2.143 |
lement | +1.953 |
roman | +1.951 |
Neg logit contributions | |
@@ | -1.949 |
Krug | -1.848 |
� | -1.848 |
Lie | -1.830 |
crossings | -1.816 |
Token | , |
Token ID | 11 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
anson | +0.298 |
rem | +0.257 |
� | +0.244 |
lement | +0.225 |
roman | +0.223 |
Neg logit contributions | |
@@ | -0.209 |
Krug | -0.198 |
� | -0.198 |
Lie | -0.196 |
crossings | -0.195 |
Token | but |
Token ID | 475 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
anson | +0.149 |
rem | +0.127 |
� | +0.118 |
lement | +0.109 |
roman | +0.108 |
Neg logit contributions | |
@@ | -0.132 |
Krug | -0.125 |
� | -0.124 |
Lie | -0.124 |
crossings | -0.122 |
Token | then |
Token ID | 788 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
anson | +0.266 |
rem | +0.231 |
� | +0.217 |
lement | +0.201 |
roman | +0.200 |
Neg logit contributions | |
@@ | -0.184 |
Krug | -0.175 |
� | -0.173 |
Lie | -0.172 |
crossings | -0.171 |
Token | Klaus |
Token ID | 49739 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
anson | +0.634 |
rem | +0.555 |
� | +0.525 |
lement | +0.486 |
roman | +0.485 |
Neg logit contributions | |
@@ | -0.383 |
Krug | -0.362 |
� | -0.360 |
Lie | -0.357 |
crossings | -0.354 |
Token | said |
Token ID | 531 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
@@ | +0.384 |
Krug | +0.362 |
� | +0.358 |
crossings | +0.356 |
Lie | +0.356 |
Neg logit contributions | |
anson | -0.707 |
rem | -0.618 |
� | -0.589 |
roman | -0.546 |
lement | -0.543 |
Token | Trump |
Token ID | 1301 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
anson | +0.473 |
rem | +0.408 |
� | +0.387 |
lement | +0.355 |
roman | +0.353 |
Neg logit contributions | |
@@ | -0.337 |
Krug | -0.321 |
� | -0.318 |
crossings | -0.316 |
Lie | -0.315 |
Token | had |
Token ID | 550 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
@@ | +0.507 |
� | +0.479 |
crossings | +0.478 |
Krug | +0.478 |
Lie | +0.472 |
Neg logit contributions | |
anson | -0.775 |
rem | -0.662 |
� | -0.638 |
roman | -0.582 |
lement | -0.580 |
Token | � |
Token ID | 564 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
anson | +0.577 |
rem | +0.502 |
� | +0.478 |
lement | +0.440 |
roman | +0.439 |
Neg logit contributions | |
@@ | -0.351 |
Krug | -0.330 |
� | -0.330 |
Lie | -0.326 |
crossings | -0.326 |
Token | � |
Token ID | 250 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
@@ | +0.194 |
� | +0.188 |
� | +0.177 |
Lie | +0.169 |
Krug | +0.167 |
Neg logit contributions | |
anson | -0.642 |
rem | -0.572 |
� | -0.552 |
lement | -0.522 |
roman | -0.521 |
Token | ex |
Token ID | 1069 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
anson | +0.025 |
rem | +0.022 |
� | +0.021 |
lement | +0.019 |
roman | +0.019 |
Neg logit contributions | |
@@ | -0.018 |
� | -0.017 |
Lie | -0.017 |
Krug | -0.017 |
� | -0.017 |
Token | cessive |
Token ID | 45428 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
anson | +2.431 |
rem | +2.108 |
� | +1.955 |
lement | +1.778 |
roman | +1.769 |
Neg logit contributions | |
@@ | -2.124 |
� | -2.017 |
Krug | -2.016 |
Lie | -2.000 |
crossings | -1.997 |
Token | expectations |
Token ID | 9027 |
Feature activation | +5.51e-03 |
Pos logit contributions | |
@@ | +0.544 |
� | +0.520 |
crossings | +0.519 |
Krug | +0.515 |
Lie | +0.514 |
Neg logit contributions | |
anson | -0.686 |
rem | -0.577 |
� | -0.550 |
roman | -0.502 |
lement | -0.497 |
Token | � |
Token ID | 447 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
@@ | +0.009 |
� | +0.009 |
Krug | +0.009 |
crossings | +0.009 |
Lie | +0.009 |
Neg logit contributions | |
anson | -0.013 |
rem | -0.011 |
� | -0.010 |
lement | -0.010 |
roman | -0.010 |
Token | � |
Token ID | 251 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
� | +0.069 |
@@ | +0.061 |
� | +0.060 |
Lie | +0.043 |
Krug | +0.042 |
Neg logit contributions | |
anson | -0.439 |
rem | -0.392 |
� | -0.381 |
roman | -0.364 |
lement | -0.363 |
Token | of |
Token ID | 286 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
anson | +0.266 |
rem | +0.228 |
� | +0.216 |
roman | +0.198 |
lement | +0.197 |
Neg logit contributions | |
@@ | -0.204 |
� | -0.193 |
Krug | -0.192 |
Lie | -0.189 |
crossings | -0.189 |
Token | When |
Token ID | 2215 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
anson | +0.138 |
rem | +0.116 |
� | +0.111 |
lement | +0.102 |
roman | +0.101 |
Neg logit contributions | |
@@ | -0.103 |
Lie | -0.100 |
Krug | -0.098 |
� | -0.098 |
� | -0.095 |
Token | we |
Token ID | 356 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
@@ | +0.040 |
Krug | +0.038 |
� | +0.038 |
Lie | +0.037 |
crossings | +0.037 |
Neg logit contributions | |
anson | -0.054 |
rem | -0.047 |
� | -0.045 |
lement | -0.041 |
roman | -0.041 |
Token | watched |
Token ID | 7342 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
anson | +0.010 |
rem | +0.009 |
� | +0.008 |
roman | +0.008 |
lement | +0.008 |
Neg logit contributions | |
@@ | -0.007 |
Krug | -0.006 |
crossings | -0.006 |
Lie | -0.006 |
� | -0.006 |
Token | this |
Token ID | 428 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
anson | +0.505 |
rem | +0.433 |
� | +0.411 |
lement | +0.377 |
roman | +0.375 |
Neg logit contributions | |
@@ | -0.379 |
Krug | -0.362 |
� | -0.360 |
Lie | -0.356 |
crossings | -0.354 |
Token | little |
Token ID | 1310 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
anson | +0.186 |
rem | +0.161 |
� | +0.152 |
lement | +0.140 |
roman | +0.140 |
Neg logit contributions | |
@@ | -0.130 |
Krug | -0.123 |
� | -0.122 |
Lie | -0.122 |
crossings | -0.121 |
Token | gag |
Token ID | 32261 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
anson | +0.446 |
rem | +0.386 |
� | +0.366 |
lement | +0.336 |
roman | +0.335 |
Neg logit contributions | |
@@ | -0.307 |
Krug | -0.292 |
� | -0.290 |
Lie | -0.288 |
crossings | -0.286 |
Token | with |
Token ID | 351 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
anson | +2.684 |
rem | +2.343 |
� | +2.209 |
lement | +2.032 |
roman | +2.030 |
Neg logit contributions | |
@@ | -1.814 |
� | -1.712 |
Krug | -1.705 |
Lie | -1.693 |
crossings | -1.684 |
Token | Erin |
Token ID | 28894 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
anson | +0.466 |
rem | +0.401 |
� | +0.381 |
lement | +0.349 |
roman | +0.348 |
Neg logit contributions | |
@@ | -0.337 |
Krug | -0.323 |
� | -0.320 |
Lie | -0.317 |
crossings | -0.314 |
Token | and |
Token ID | 290 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
@@ | +0.031 |
Krug | +0.029 |
� | +0.029 |
Lie | +0.029 |
crossings | +0.029 |
Neg logit contributions | |
anson | -0.048 |
rem | -0.042 |
� | -0.040 |
lement | -0.037 |
roman | -0.037 |
Token | Jill |
Token ID | 22772 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
anson | +0.395 |
rem | +0.342 |
� | +0.323 |
lement | +0.296 |
roman | +0.295 |
Neg logit contributions | |
@@ | -0.288 |
Krug | -0.274 |
� | -0.273 |
Lie | -0.271 |
crossings | -0.269 |
Token | ian |
Token ID | 666 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
@@ | +0.328 |
Krug | +0.312 |
� | +0.311 |
Lie | +0.310 |
crossings | +0.305 |
Neg logit contributions | |
anson | -0.527 |
rem | -0.461 |
� | -0.438 |
lement | -0.405 |
roman | -0.403 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
anson | +1.240 |
rem | +1.116 |
� | +1.069 |
lement | +1.005 |
roman | +1.002 |
Neg logit contributions | |
@@ | -0.376 |
� | -0.343 |
Krug | -0.340 |
Lie | -0.336 |
crossings | -0.332 |
Token | arrives |
Token ID | 14443 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
@@ | +0.993 |
Krug | +0.946 |
crossings | +0.944 |
� | +0.918 |
Lie | +0.918 |
Neg logit contributions | |
anson | -1.256 |
rem | -1.062 |
� | -1.012 |
roman | -0.936 |
lement | -0.921 |
Token | for |
Token ID | 329 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
@@ | +0.328 |
Krug | +0.314 |
� | +0.311 |
crossings | +0.310 |
Lie | +0.309 |
Neg logit contributions | |
anson | -0.480 |
rem | -0.413 |
� | -0.393 |
roman | -0.363 |
lement | -0.357 |
Token | a |
Token ID | 257 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
anson | +0.532 |
rem | +0.457 |
� | +0.435 |
roman | +0.399 |
lement | +0.397 |
Neg logit contributions | |
@@ | -0.408 |
Krug | -0.389 |
� | -0.385 |
Lie | -0.383 |
crossings | -0.382 |
Token | news |
Token ID | 1705 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
anson | +0.418 |
rem | +0.359 |
� | +0.344 |
roman | +0.317 |
lement | +0.315 |
Neg logit contributions | |
@@ | -0.296 |
Krug | -0.281 |
� | -0.276 |
Lie | -0.275 |
crossings | -0.275 |
Token | conference |
Token ID | 4495 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
anson | +0.638 |
rem | +0.515 |
� | +0.490 |
roman | +0.433 |
lement | +0.423 |
Neg logit contributions | |
@@ | -0.788 |
� | -0.761 |
Krug | -0.759 |
Lie | -0.746 |
crossings | -0.745 |
Token | after |
Token ID | 706 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
anson | +2.568 |
rem | +2.212 |
� | +2.092 |
roman | +1.912 |
lement | +1.906 |
Neg logit contributions | |
@@ | -1.929 |
Krug | -1.831 |
� | -1.817 |
crossings | -1.805 |
Lie | -1.804 |
Token | participating |
Token ID | 11983 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
anson | +0.649 |
rem | +0.564 |
� | +0.541 |
lement | +0.498 |
roman | +0.497 |
Neg logit contributions | |
@@ | -0.418 |
Krug | -0.399 |
Lie | -0.392 |
crossings | -0.392 |
� | -0.391 |
Token | in |
Token ID | 287 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
anson | +0.313 |
rem | +0.272 |
� | +0.256 |
lement | +0.235 |
roman | +0.233 |
Neg logit contributions | |
@@ | -0.220 |
� | -0.209 |
Krug | -0.207 |
Lie | -0.206 |
crossings | -0.205 |
Token | a |
Token ID | 257 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
@@ | +0.062 |
Krug | +0.059 |
� | +0.059 |
Lie | +0.058 |
crossings | +0.058 |
Neg logit contributions | |
anson | -0.082 |
rem | -0.070 |
� | -0.067 |
roman | -0.061 |
lement | -0.061 |
Token | debate |
Token ID | 4384 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
anson | +0.085 |
rem | +0.073 |
� | +0.070 |
roman | +0.064 |
lement | +0.064 |
Neg logit contributions | |
@@ | -0.058 |
Krug | -0.055 |
� | -0.054 |
Lie | -0.054 |
crossings | -0.054 |
Token | Ste |
Token ID | 7447 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
anson | +2.020 |
rem | +1.766 |
� | +1.673 |
lement | +1.543 |
roman | +1.540 |
Neg logit contributions | |
@@ | -1.253 |
� | -1.180 |
Krug | -1.179 |
Lie | -1.169 |
crossings | -1.157 |
Token | lla |
Token ID | 8466 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
anson | +0.253 |
rem | +0.214 |
� | +0.208 |
roman | +0.193 |
uana | +0.190 |
Neg logit contributions | |
@@ | -0.164 |
� | -0.157 |
Lie | -0.155 |
� | -0.152 |
Krug | -0.150 |
Token | - |
Token ID | 12 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
anson | +0.781 |
rem | +0.685 |
� | +0.649 |
lement | +0.600 |
roman | +0.599 |
Neg logit contributions | |
@@ | -0.463 |
� | -0.436 |
Krug | -0.436 |
Lie | -0.431 |
crossings | -0.427 |
Token | san |
Token ID | 12807 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
anson | +0.395 |
rem | +0.329 |
� | +0.308 |
roman | +0.276 |
lement | +0.276 |
Neg logit contributions | |
@@ | -0.425 |
� | -0.408 |
Lie | -0.407 |
Krug | -0.403 |
� | -0.400 |
Token | collapsed |
Token ID | 14707 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
anson | +0.736 |
rem | +0.643 |
� | +0.611 |
roman | +0.565 |
lement | +0.564 |
Neg logit contributions | |
@@ | -0.434 |
� | -0.408 |
Krug | -0.407 |
Lie | -0.404 |
crossings | -0.400 |
Token | ?」 |
Token ID | 42943 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
anson | +0.290 |
rem | +0.250 |
� | +0.237 |
roman | +0.221 |
lement | +0.216 |
Neg logit contributions | |
@@ | -0.202 |
� | -0.192 |
Krug | -0.192 |
Lie | -0.191 |
crossings | -0.189 |
Token |
|
Token ID | 198 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
anson | +2.307 |
rem | +1.960 |
� | +1.845 |
lement | +1.674 |
roman | +1.623 |
Neg logit contributions | |
@@ | -2.111 |
� | -2.030 |
Krug | -2.020 |
crossings | -2.014 |
Lie | -1.994 |
Token |
|
Token ID | 198 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
anson | +0.532 |
rem | +0.444 |
� | +0.426 |
roman | +0.399 |
uana | +0.395 |
Neg logit contributions | |
@@ | -0.361 |
Lie | -0.338 |
Krug | -0.330 |
territory | -0.324 |
� | -0.324 |
Token | " |
Token ID | 1 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
anson | +0.471 |
rem | +0.391 |
� | +0.365 |
roman | +0.327 |
lement | +0.326 |
Neg logit contributions | |
@@ | -0.506 |
Lie | -0.487 |
� | -0.484 |
Krug | -0.482 |
� | -0.476 |
Token | Yes |
Token ID | 5297 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
anson | +1.352 |
rem | +1.178 |
� | +1.120 |
roman | +1.041 |
lement | +1.038 |
Neg logit contributions | |
@@ | -0.759 |
Lie | -0.718 |
� | -0.712 |
Krug | -0.708 |
� | -0.692 |
Token | . |
Token ID | 13 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
anson | +1.294 |
rem | +1.125 |
� | +1.063 |
lement | +0.976 |
roman | +0.974 |
Neg logit contributions | |
@@ | -0.888 |
� | -0.840 |
Krug | -0.839 |
Lie | -0.831 |
crossings | -0.825 |
Token | to |
Token ID | 284 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
@@ | +0.543 |
� | +0.513 |
Krug | +0.512 |
crossings | +0.508 |
Lie | +0.506 |
Neg logit contributions | |
anson | -0.860 |
rem | -0.742 |
� | -0.707 |
roman | -0.648 |
lement | -0.644 |
Token | NATO |
Token ID | 11332 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
@@ | +0.049 |
� | +0.046 |
Krug | +0.046 |
Lie | +0.046 |
crossings | +0.045 |
Neg logit contributions | |
anson | -0.077 |
rem | -0.067 |
� | -0.063 |
roman | -0.058 |
lement | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
@@ | +0.379 |
crossings | +0.361 |
Krug | +0.360 |
territory | +0.359 |
� | +0.358 |
Neg logit contributions | |
anson | -0.576 |
rem | -0.495 |
� | -0.474 |
roman | -0.437 |
aston | -0.430 |
Token | A |
Token ID | 317 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
anson | +0.204 |
rem | +0.174 |
� | +0.166 |
roman | +0.154 |
uana | +0.153 |
Neg logit contributions | |
@@ | -0.132 |
Krug | -0.126 |
Lie | -0.124 |
� | -0.123 |
COUN | -0.122 |
Token | l |
Token ID | 300 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
anson | +0.118 |
rem | +0.101 |
� | +0.097 |
lement | +0.089 |
roman | +0.089 |
Neg logit contributions | |
@@ | -0.077 |
� | -0.074 |
Krug | -0.074 |
Lie | -0.073 |
crossings | -0.072 |
Token | uke |
Token ID | 4649 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
anson | +0.075 |
rem | +0.066 |
� | +0.064 |
lement | +0.059 |
roman | +0.059 |
Neg logit contributions | |
@@ | -0.032 |
Krug | -0.029 |
crossings | -0.029 |
� | -0.029 |
Lie | -0.028 |
Token | warm |
Token ID | 31975 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
anson | +1.718 |
rem | +1.383 |
� | +1.257 |
lement | +1.083 |
roman | +1.077 |
Neg logit contributions | |
@@ | -2.687 |
� | -2.593 |
Krug | -2.585 |
Lie | -2.575 |
crossings | -2.564 |
Token | response |
Token ID | 2882 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
anson | +1.037 |
rem | +0.895 |
� | +0.856 |
lement | +0.788 |
roman | +0.787 |
Neg logit contributions | |
@@ | -0.678 |
Krug | -0.649 |
� | -0.635 |
Lie | -0.633 |
crossings | -0.628 |
Token | from |
Token ID | 422 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
anson | +0.485 |
rem | +0.419 |
� | +0.400 |
lement | +0.367 |
roman | +0.366 |
Neg logit contributions | |
@@ | -0.316 |
Krug | -0.297 |
� | -0.295 |
crossings | -0.294 |
Lie | -0.293 |
Token | the |
Token ID | 262 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
anson | +0.545 |
rem | +0.472 |
� | +0.445 |
lement | +0.409 |
roman | +0.408 |
Neg logit contributions | |
@@ | -0.380 |
Krug | -0.362 |
� | -0.360 |
Lie | -0.356 |
crossings | -0.355 |
Token | United |
Token ID | 1578 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
anson | +0.419 |
rem | +0.364 |
� | +0.345 |
lement | +0.319 |
roman | +0.317 |
Neg logit contributions | |
@@ | -0.260 |
Krug | -0.248 |
� | -0.246 |
crossings | -0.244 |
Lie | -0.243 |
Token | , |
Token ID | 11 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
anson | +0.437 |
rem | +0.380 |
� | +0.359 |
lement | +0.331 |
roman | +0.330 |
Neg logit contributions | |
@@ | -0.293 |
� | -0.277 |
Krug | -0.277 |
Lie | -0.274 |
crossings | -0.271 |
Token | House |
Token ID | 2097 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
anson | +0.372 |
rem | +0.323 |
� | +0.306 |
roman | +0.281 |
lement | +0.281 |
Neg logit contributions | |
@@ | -0.257 |
Krug | -0.245 |
� | -0.242 |
Lie | -0.241 |
crossings | -0.239 |
Token | Speaker |
Token ID | 14931 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
anson | +0.045 |
rem | +0.039 |
� | +0.037 |
roman | +0.034 |
lement | +0.034 |
Neg logit contributions | |
@@ | -0.030 |
Krug | -0.029 |
� | -0.028 |
Lie | -0.028 |
� | -0.028 |
Token | Tip |
Token ID | 23095 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
anson | +0.315 |
rem | +0.272 |
� | +0.258 |
lement | +0.235 |
roman | +0.234 |
Neg logit contributions | |
@@ | -0.238 |
Krug | -0.230 |
� | -0.225 |
Lie | -0.224 |
Nemesis | -0.223 |
Token | O |
Token ID | 440 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
@@ | +0.056 |
Krug | +0.051 |
Lie | +0.046 |
� | +0.045 |
Nemesis | +0.042 |
Neg logit contributions | |
anson | -0.406 |
rem | -0.369 |
� | -0.357 |
lement | -0.342 |
roman | -0.340 |
Token | ' |
Token ID | 6 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
@@ | +0.740 |
Krug | +0.724 |
� | +0.702 |
Lie | +0.701 |
Scand | +0.699 |
Neg logit contributions | |
anson | -0.820 |
rem | -0.649 |
� | -0.635 |
roman | -0.578 |
lement | -0.573 |
Token | Neill |
Token ID | 26538 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
anson | +2.587 |
rem | +2.287 |
� | +2.139 |
roman | +1.943 |
lement | +1.937 |
Neg logit contributions | |
@@ | -1.815 |
� | -1.727 |
Krug | -1.687 |
crossings | -1.686 |
� | -1.682 |
Token | , |
Token ID | 11 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
anson | +0.553 |
rem | +0.486 |
� | +0.462 |
lement | +0.427 |
roman | +0.426 |
Neg logit contributions | |
@@ | -0.309 |
Krug | -0.289 |
� | -0.289 |
Lie | -0.286 |
crossings | -0.284 |
Token | could |
Token ID | 714 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
anson | +0.082 |
rem | +0.071 |
� | +0.067 |
roman | +0.063 |
lement | +0.062 |
Neg logit contributions | |
@@ | -0.053 |
Krug | -0.050 |
� | -0.050 |
crossings | -0.049 |
Lie | -0.049 |
Token | hardly |
Token ID | 8941 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
@@ | +0.226 |
Krug | +0.211 |
Lie | +0.211 |
� | +0.211 |
crossings | +0.209 |
Neg logit contributions | |
anson | -0.346 |
rem | -0.299 |
� | -0.285 |
roman | -0.263 |
lement | -0.260 |
Token | have |
Token ID | 423 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
@@ | +0.729 |
Lie | +0.673 |
crossings | +0.666 |
Krug | +0.665 |
COUN | +0.657 |
Neg logit contributions | |
anson | -1.164 |
rem | -0.991 |
� | -0.953 |
roman | -0.873 |
lement | -0.862 |
Token | of |
Token ID | 286 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
@@ | +0.417 |
Krug | +0.389 |
� | +0.389 |
Lie | +0.386 |
crossings | +0.381 |
Neg logit contributions | |
anson | -0.838 |
rem | -0.743 |
� | -0.706 |
lement | -0.656 |
roman | -0.654 |
Token | the |
Token ID | 262 |
Feature activation | -1.90e-03 |
Pos logit contributions | |
anson | +0.354 |
rem | +0.307 |
� | +0.291 |
lement | +0.268 |
roman | +0.268 |
Neg logit contributions | |
@@ | -0.234 |
Krug | -0.221 |
� | -0.220 |
Lie | -0.217 |
crossings | -0.216 |
Token | ruling |
Token ID | 6922 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
anson | +0.005 |
rem | +0.004 |
� | +0.004 |
lement | +0.004 |
roman | +0.004 |
Neg logit contributions | |
@@ | -0.003 |
� | -0.003 |
Krug | -0.003 |
Lie | -0.003 |
crossings | -0.003 |
Token | class |
Token ID | 1398 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
anson | +0.167 |
rem | +0.143 |
� | +0.134 |
roman | +0.122 |
lement | +0.122 |
Neg logit contributions | |
@@ | -0.133 |
Krug | -0.126 |
� | -0.126 |
Lie | -0.124 |
crossings | -0.124 |
Token | to |
Token ID | 284 |
Feature activation | -1.64e-03 |
Pos logit contributions | |
@@ | +0.400 |
Krug | +0.379 |
� | +0.378 |
Lie | +0.375 |
crossings | +0.374 |
Neg logit contributions | |
anson | -0.526 |
rem | -0.451 |
� | -0.428 |
roman | -0.388 |
lement | -0.387 |
Token | demon |
Token ID | 3222 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
anson | +0.004 |
rem | +0.003 |
� | +0.003 |
roman | +0.003 |
lement | +0.003 |
Neg logit contributions | |
@@ | -0.003 |
Krug | -0.002 |
Lie | -0.002 |
� | -0.002 |
crossings | -0.002 |
Token | ize |
Token ID | 1096 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
anson | +2.550 |
rem | +2.229 |
� | +2.098 |
roman | +1.929 |
lement | +1.920 |
Neg logit contributions | |
@@ | -1.821 |
Krug | -1.725 |
� | -1.723 |
Lie | -1.700 |
crossings | -1.699 |
Token | or |
Token ID | 393 |
Feature activation | +6.95e-02 |
Pos logit contributions | |
anson | +0.121 |
rem | +0.104 |
� | +0.099 |
roman | +0.090 |
lement | +0.090 |
Neg logit contributions | |
@@ | -0.089 |
Krug | -0.084 |
� | -0.083 |
Lie | -0.083 |
crossings | -0.082 |
Token | ridicule |
Token ID | 38364 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
@@ | +0.107 |
Krug | +0.100 |
� | +0.100 |
Lie | +0.100 |
crossings | +0.099 |
Neg logit contributions | |
anson | -0.174 |
rem | -0.152 |
� | -0.144 |
lement | -0.133 |
roman | -0.133 |
Token | anger |
Token ID | 8993 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
anson | +0.982 |
rem | +0.848 |
� | +0.801 |
lement | +0.732 |
roman | +0.731 |
Neg logit contributions | |
@@ | -0.733 |
Krug | -0.695 |
� | -0.692 |
Lie | -0.687 |
crossings | -0.683 |
Token | in |
Token ID | 287 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
anson | +1.117 |
rem | +0.967 |
� | +0.912 |
lement | +0.834 |
roman | +0.832 |
Neg logit contributions | |
@@ | -0.822 |
� | -0.779 |
Krug | -0.779 |
Lie | -0.772 |
crossings | -0.766 |
Token | a |
Token ID | 257 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
anson | +0.573 |
rem | +0.495 |
� | +0.470 |
lement | +0.431 |
roman | +0.429 |
Neg logit contributions | |
@@ | -0.409 |
Krug | -0.388 |
� | -0.388 |
Lie | -0.382 |
crossings | -0.381 |
Token | sc |
Token ID | 629 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
@@ | +0.125 |
Krug | +0.117 |
� | +0.117 |
Lie | +0.116 |
Nemesis | +0.115 |
Neg logit contributions | |
anson | -0.191 |
rem | -0.165 |
� | -0.158 |
lement | -0.145 |
roman | -0.144 |
Token | 2 |
Token ID | 17 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
@@ | +0.577 |
� | +0.548 |
Krug | +0.542 |
Lie | +0.537 |
� | +0.535 |
Neg logit contributions | |
anson | -0.799 |
rem | -0.690 |
� | -0.661 |
lement | -0.599 |
roman | -0.596 |
Token | drama |
Token ID | 10512 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
@@ | +1.130 |
� | +1.067 |
Krug | +1.057 |
� | +1.044 |
Lie | +1.044 |
Neg logit contributions | |
anson | -1.532 |
rem | -1.310 |
� | -1.261 |
lement | -1.156 |
roman | -1.146 |
Token | sub |
Token ID | 850 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
@@ | +0.771 |
Krug | +0.727 |
� | +0.725 |
Lie | +0.714 |
Nemesis | +0.712 |
Neg logit contributions | |
anson | -1.092 |
rem | -0.943 |
� | -0.902 |
lement | -0.821 |
roman | -0.818 |
Token | forum |
Token ID | 27302 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
anson | +0.616 |
rem | +0.533 |
� | +0.507 |
lement | +0.463 |
roman | +0.462 |
Neg logit contributions | |
@@ | -0.426 |
� | -0.402 |
Krug | -0.401 |
Lie | -0.398 |
crossings | -0.393 |
Token | everything |
Token ID | 2279 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
@@ | +1.389 |
Krug | +1.306 |
� | +1.301 |
Lie | +1.280 |
Nemesis | +1.274 |
Neg logit contributions | |
anson | -2.072 |
rem | -1.790 |
� | -1.710 |
lement | -1.568 |
roman | -1.557 |
Token | 'd |
Token ID | 1549 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
@@ | +0.288 |
Krug | +0.270 |
� | +0.268 |
crossings | +0.268 |
Lie | +0.266 |
Neg logit contributions | |
anson | -0.426 |
rem | -0.369 |
� | -0.351 |
lement | -0.323 |
roman | -0.322 |
Token | be |
Token ID | 307 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
@@ | +0.245 |
� | +0.229 |
Krug | +0.229 |
Lie | +0.228 |
crossings | +0.226 |
Neg logit contributions | |
anson | -0.414 |
rem | -0.361 |
� | -0.344 |
lement | -0.318 |
roman | -0.317 |
Token | okay |
Token ID | 8788 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
@@ | +0.375 |
Krug | +0.353 |
� | +0.351 |
Lie | +0.348 |
crossings | +0.346 |
Neg logit contributions | |
anson | -0.551 |
rem | -0.475 |
� | -0.452 |
lement | -0.417 |
roman | -0.416 |
Token |
|
Token ID | 198 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
anson | +0.579 |
rem | +0.499 |
� | +0.471 |
lement | +0.431 |
roman | +0.430 |
Neg logit contributions | |
@@ | -0.435 |
� | -0.411 |
Krug | -0.411 |
Lie | -0.408 |
crossings | -0.405 |
Token | s |
Token ID | 82 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
@@ | +0.252 |
� | +0.236 |
Krug | +0.235 |
Lie | +0.232 |
crossings | +0.230 |
Neg logit contributions | |
anson | -0.427 |
rem | -0.371 |
� | -0.354 |
roman | -0.328 |
lement | -0.327 |
Token | press |
Token ID | 1803 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
@@ | +0.229 |
� | +0.218 |
Krug | +0.216 |
Lie | +0.214 |
crossings | +0.214 |
Neg logit contributions | |
anson | -0.328 |
rem | -0.281 |
� | -0.269 |
lement | -0.246 |
roman | -0.245 |
Token | secretary |
Token ID | 7705 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
anson | +1.086 |
rem | +0.919 |
� | +0.866 |
roman | +0.783 |
lement | +0.778 |
Neg logit contributions | |
@@ | -0.990 |
� | -0.944 |
Krug | -0.939 |
Lie | -0.933 |
crossings | -0.929 |
Token | , |
Token ID | 11 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
anson | +0.404 |
rem | +0.355 |
� | +0.337 |
roman | +0.312 |
lement | +0.312 |
Neg logit contributions | |
@@ | -0.228 |
Krug | -0.214 |
� | -0.214 |
Lie | -0.211 |
crossings | -0.209 |
Token | Cullen |
Token ID | 47681 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
anson | +0.101 |
rem | +0.087 |
� | +0.082 |
roman | +0.075 |
lement | +0.075 |
Neg logit contributions | |
@@ | -0.078 |
Krug | -0.074 |
� | -0.073 |
Lie | -0.073 |
crossings | -0.072 |
Token | Schwarz |
Token ID | 34835 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
@@ | +0.320 |
Krug | +0.303 |
� | +0.300 |
Lie | +0.298 |
crossings | +0.294 |
Neg logit contributions | |
anson | -0.525 |
rem | -0.460 |
� | -0.436 |
lement | -0.403 |
roman | -0.402 |
Token | , |
Token ID | 11 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
@@ | +1.308 |
� | +1.230 |
Krug | +1.230 |
Lie | +1.216 |
crossings | +1.206 |
Neg logit contributions | |
anson | -2.220 |
rem | -1.946 |
� | -1.846 |
lement | -1.705 |
roman | -1.702 |
Token | was |
Token ID | 373 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
anson | +0.241 |
rem | +0.212 |
� | +0.203 |
roman | +0.189 |
lement | +0.188 |
Neg logit contributions | |
@@ | -0.115 |
Krug | -0.106 |
� | -0.106 |
Lie | -0.104 |
crossings | -0.103 |
Token | more |
Token ID | 517 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
@@ | +0.079 |
Krug | +0.075 |
� | +0.075 |
crossings | +0.074 |
Lie | +0.074 |
Neg logit contributions | |
anson | -0.117 |
rem | -0.101 |
� | -0.096 |
roman | -0.088 |
lement | -0.088 |
Token | blunt |
Token ID | 19861 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
@@ | +0.026 |
Krug | +0.024 |
� | +0.024 |
crossings | +0.024 |
Lie | +0.024 |
Neg logit contributions | |
anson | -0.040 |
rem | -0.034 |
� | -0.033 |
roman | -0.030 |
lement | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
anson | +0.689 |
rem | +0.592 |
� | +0.561 |
roman | +0.512 |
lement | +0.509 |
Neg logit contributions | |
@@ | -0.511 |
Krug | -0.484 |
� | -0.483 |
Lie | -0.476 |
crossings | -0.476 |
Token | and |
Token ID | 290 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
anson | +0.056 |
rem | +0.048 |
� | +0.045 |
roman | +0.041 |
lement | +0.041 |
Neg logit contributions | |
@@ | -0.041 |
crossings | -0.039 |
Krug | -0.039 |
� | -0.039 |
territory | -0.038 |
Token | violated |
Token ID | 13998 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
@@ | +0.047 |
Krug | +0.045 |
crossings | +0.044 |
� | +0.044 |
Lie | +0.044 |
Neg logit contributions | |
anson | -0.068 |
rem | -0.059 |
� | -0.056 |
roman | -0.052 |
lement | -0.051 |
Token | Turkish |
Token ID | 9663 |
Feature activation | +9.04e-02 |
Pos logit contributions | |
@@ | +0.306 |
Krug | +0.289 |
� | +0.287 |
crossings | +0.287 |
Lie | +0.286 |
Neg logit contributions | |
anson | -0.491 |
rem | -0.426 |
� | -0.403 |
roman | -0.376 |
lement | -0.371 |
Token | laws |
Token ID | 3657 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
@@ | +0.125 |
� | +0.117 |
Krug | +0.117 |
crossings | +0.116 |
Lie | +0.116 |
Neg logit contributions | |
anson | -0.240 |
rem | -0.210 |
� | -0.201 |
roman | -0.187 |
lement | -0.186 |
Token | and |
Token ID | 290 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
@@ | +0.087 |
Krug | +0.082 |
crossings | +0.082 |
� | +0.082 |
Lie | +0.081 |
Neg logit contributions | |
anson | -0.114 |
rem | -0.097 |
� | -0.092 |
roman | -0.083 |
lement | -0.083 |
Token | sensibilities |
Token ID | 50222 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
@@ | +0.053 |
Krug | +0.050 |
� | +0.050 |
crossings | +0.050 |
Lie | +0.049 |
Neg logit contributions | |
anson | -0.077 |
rem | -0.067 |
� | -0.063 |
roman | -0.059 |
lement | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
@@ | +1.454 |
crossings | +1.383 |
Krug | +1.376 |
territory | +1.363 |
� | +1.362 |
Neg logit contributions | |
anson | -1.852 |
rem | -1.554 |
� | -1.482 |
aston | -1.338 |
roman | -1.335 |
Token | That |
Token ID | 1320 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
anson | +0.026 |
rem | +0.022 |
� | +0.021 |
roman | +0.019 |
lement | +0.019 |
Neg logit contributions | |
@@ | -0.018 |
Krug | -0.017 |
Lie | -0.016 |
� | -0.016 |
Nemesis | -0.016 |
Token | is |
Token ID | 318 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
@@ | +0.090 |
Krug | +0.085 |
� | +0.084 |
Lie | +0.084 |
crossings | +0.084 |
Neg logit contributions | |
anson | -0.129 |
rem | -0.111 |
� | -0.105 |
lement | -0.097 |
roman | -0.096 |
Token | quite |
Token ID | 2407 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
@@ | +0.072 |
Krug | +0.068 |
� | +0.068 |
crossings | +0.067 |
Lie | +0.067 |
Neg logit contributions | |
anson | -0.119 |
rem | -0.103 |
� | -0.098 |
roman | -0.090 |
lement | -0.090 |
Token | enough |
Token ID | 1576 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
@@ | +0.536 |
� | +0.505 |
Krug | +0.503 |
crossings | +0.499 |
Lie | +0.497 |
Neg logit contributions | |
anson | -0.773 |
rem | -0.656 |
� | -0.633 |
roman | -0.581 |
lement | -0.576 |
Token | p |
Token ID | 79 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
@@ | +0.450 |
Krug | +0.424 |
� | +0.423 |
Lie | +0.421 |
crossings | +0.415 |
Neg logit contributions | |
anson | -0.630 |
rem | -0.542 |
� | -0.516 |
roman | -0.472 |
lement | -0.470 |
Token | to |
Token ID | 284 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
@@ | +0.047 |
Krug | +0.045 |
� | +0.044 |
Lie | +0.044 |
crossings | +0.043 |
Neg logit contributions | |
anson | -0.070 |
rem | -0.061 |
� | -0.058 |
roman | -0.053 |
lement | -0.053 |
Token | a |
Token ID | 257 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
anson | +0.126 |
rem | +0.109 |
� | +0.104 |
lement | +0.096 |
roman | +0.095 |
Neg logit contributions | |
@@ | -0.080 |
Krug | -0.076 |
� | -0.075 |
Lie | -0.074 |
crossings | -0.073 |
Token | side |
Token ID | 1735 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
@@ | +0.186 |
Krug | +0.177 |
� | +0.177 |
Lie | +0.174 |
� | +0.172 |
Neg logit contributions | |
anson | -0.285 |
rem | -0.246 |
� | -0.235 |
lement | -0.216 |
roman | -0.215 |
Token | lane |
Token ID | 11193 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
@@ | +0.170 |
Krug | +0.164 |
� | +0.162 |
Lie | +0.160 |
crossings | +0.159 |
Neg logit contributions | |
anson | -0.246 |
rem | -0.211 |
� | -0.202 |
lement | -0.184 |
roman | -0.183 |
Token | + |
Token ID | 1343 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
@@ | +1.300 |
Krug | +1.255 |
� | +1.231 |
Lie | +1.210 |
Nemesis | +1.204 |
Neg logit contributions | |
anson | -2.044 |
rem | -1.759 |
� | -1.669 |
lement | -1.546 |
roman | -1.539 |
Token | all |
Token ID | 477 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
@@ | +1.188 |
Krug | +1.153 |
� | +1.132 |
Lie | +1.113 |
Nemesis | +1.101 |
Neg logit contributions | |
anson | -1.871 |
rem | -1.616 |
� | -1.538 |
lement | -1.423 |
roman | -1.416 |
Token | my |
Token ID | 616 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
@@ | +0.096 |
Krug | +0.091 |
� | +0.090 |
Lie | +0.089 |
crossings | +0.088 |
Neg logit contributions | |
anson | -0.140 |
rem | -0.122 |
� | -0.116 |
lement | -0.106 |
roman | -0.106 |
Token | mana |
Token ID | 13149 |
Feature activation | +6.84e-02 |
Pos logit contributions | |
anson | +0.290 |
rem | +0.252 |
� | +0.240 |
lement | +0.220 |
roman | +0.220 |
Neg logit contributions | |
@@ | -0.189 |
Krug | -0.180 |
� | -0.178 |
Lie | -0.176 |
crossings | -0.175 |
Token | to |
Token ID | 284 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
@@ | +0.110 |
Krug | +0.104 |
� | +0.103 |
Lie | +0.102 |
crossings | +0.101 |
Neg logit contributions | |
anson | -0.166 |
rem | -0.145 |
� | -0.138 |
lement | -0.126 |
roman | -0.126 |
Token | get |
Token ID | 651 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
anson | +0.272 |
rem | +0.235 |
� | +0.224 |
lement | +0.206 |
roman | +0.206 |
Neg logit contributions | |
@@ | -0.169 |
Krug | -0.161 |
� | -0.158 |
Lie | -0.157 |
crossings | -0.156 |
Token | and |
Token ID | 290 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
@@ | +0.835 |
� | +0.796 |
Krug | +0.796 |
Lie | +0.790 |
crossings | +0.785 |
Neg logit contributions | |
anson | -0.915 |
rem | -0.778 |
� | -0.729 |
lement | -0.658 |
roman | -0.658 |
Token | the |
Token ID | 262 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
@@ | +0.288 |
Krug | +0.273 |
Lie | +0.272 |
� | +0.271 |
crossings | +0.270 |
Neg logit contributions | |
anson | -0.380 |
rem | -0.326 |
� | -0.309 |
roman | -0.282 |
lement | -0.281 |
Token | Northern |
Token ID | 8342 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
@@ | +0.150 |
Krug | +0.143 |
� | +0.142 |
Lie | +0.142 |
crossings | +0.142 |
Neg logit contributions | |
anson | -0.185 |
rem | -0.159 |
� | -0.149 |
roman | -0.136 |
lement | -0.136 |
Token | Territory |
Token ID | 25219 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
anson | +0.442 |
rem | +0.394 |
� | +0.381 |
roman | +0.360 |
aston | +0.356 |
Neg logit contributions | |
@@ | -0.134 |
territory | -0.127 |
� | -0.123 |
Lie | -0.121 |
Krug | -0.120 |
Token | ( |
Token ID | 357 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
anson | +0.479 |
rem | +0.405 |
� | +0.379 |
lement | +0.341 |
roman | +0.340 |
Neg logit contributions | |
@@ | -0.467 |
� | -0.446 |
Krug | -0.446 |
Lie | -0.443 |
crossings | -0.440 |
Token | NT |
Token ID | 11251 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
@@ | +0.277 |
Lie | +0.262 |
� | +0.262 |
Krug | +0.259 |
� | +0.257 |
Neg logit contributions | |
anson | -0.457 |
rem | -0.396 |
� | -0.379 |
roman | -0.349 |
lement | -0.348 |
Token | ). |
Token ID | 737 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
@@ | +1.632 |
� | +1.563 |
Lie | +1.557 |
territory | +1.547 |
crossings | +1.542 |
Neg logit contributions | |
anson | -2.807 |
rem | -2.440 |
� | -2.325 |
roman | -2.164 |
lement | -2.138 |
Token | In |
Token ID | 554 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
@@ | +0.259 |
Lie | +0.246 |
Krug | +0.245 |
� | +0.241 |
Nemesis | +0.241 |
Neg logit contributions | |
anson | -0.385 |
rem | -0.328 |
� | -0.311 |
roman | -0.291 |
lement | -0.287 |
Token | most |
Token ID | 749 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
@@ | +0.097 |
crossings | +0.092 |
Krug | +0.092 |
� | +0.092 |
Lie | +0.092 |
Neg logit contributions | |
anson | -0.135 |
rem | -0.116 |
� | -0.110 |
lement | -0.102 |
roman | -0.102 |
Token | respects |
Token ID | 19410 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
@@ | +0.228 |
� | +0.213 |
Krug | +0.213 |
Lie | +0.211 |
crossings | +0.209 |
Neg logit contributions | |
anson | -0.433 |
rem | -0.382 |
� | -0.363 |
lement | -0.337 |
roman | -0.336 |
Token | these |
Token ID | 777 |
Feature activation | -8.74e-02 |
Pos logit contributions | |
anson | +0.932 |
rem | +0.818 |
� | +0.775 |
lement | +0.715 |
roman | +0.713 |
Neg logit contributions | |
@@ | -0.556 |
� | -0.524 |
Krug | -0.524 |
Lie | -0.518 |
crossings | -0.512 |
Token | easy |
Token ID | 2562 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
anson | +0.272 |
rem | +0.235 |
� | +0.224 |
lement | +0.205 |
roman | +0.204 |
Neg logit contributions | |
@@ | -0.182 |
Krug | -0.170 |
crossings | -0.169 |
� | -0.169 |
Lie | -0.168 |
Token | to |
Token ID | 284 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
@@ | +0.498 |
� | +0.472 |
crossings | +0.469 |
Krug | +0.468 |
Lie | +0.468 |
Neg logit contributions | |
anson | -0.653 |
rem | -0.561 |
� | -0.527 |
roman | -0.483 |
lement | -0.481 |
Token | obtain |
Token ID | 7330 |
Feature activation | -6.86e-02 |
Pos logit contributions | |
@@ | +0.091 |
Lie | +0.084 |
� | +0.083 |
Krug | +0.083 |
crossings | +0.083 |
Neg logit contributions | |
anson | -0.156 |
rem | -0.135 |
� | -0.130 |
roman | -0.119 |
lement | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
anson | +0.160 |
rem | +0.138 |
� | +0.130 |
roman | +0.119 |
lement | +0.119 |
Neg logit contributions | |
@@ | -0.117 |
� | -0.110 |
crossings | -0.109 |
Krug | -0.109 |
Lie | -0.109 |
Token | Ms |
Token ID | 6997 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
@@ | +0.093 |
Krug | +0.088 |
Lie | +0.086 |
Nemesis | +0.086 |
� | +0.086 |
Neg logit contributions | |
anson | -0.153 |
rem | -0.132 |
� | -0.125 |
roman | -0.117 |
aston | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
anson | +0.929 |
rem | +0.802 |
� | +0.758 |
lement | +0.688 |
roman | +0.685 |
Neg logit contributions | |
@@ | -0.748 |
� | -0.714 |
Krug | -0.709 |
crossings | -0.707 |
Lie | -0.705 |
Token | Gab |
Token ID | 12300 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
@@ | +0.968 |
Krug | +0.917 |
� | +0.908 |
Lie | +0.903 |
Nemesis | +0.887 |
Neg logit contributions | |
anson | -1.509 |
rem | -1.312 |
� | -1.241 |
lement | -1.150 |
roman | -1.146 |
Token | ald |
Token ID | 1940 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
@@ | +0.915 |
� | +0.875 |
Krug | +0.860 |
Lie | +0.859 |
crossings | +0.853 |
Neg logit contributions | |
anson | -1.096 |
rem | -0.929 |
� | -0.883 |
lement | -0.807 |
roman | -0.799 |
Token | on |
Token ID | 261 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
@@ | +0.510 |
� | +0.488 |
Krug | +0.481 |
crossings | +0.476 |
Lie | +0.474 |
Neg logit contributions | |
anson | -0.816 |
rem | -0.710 |
� | -0.670 |
roman | -0.628 |
lement | -0.622 |
Token | � |
Token ID | 447 |
Feature activation | +8.48e-02 |
Pos logit contributions | |
@@ | +0.132 |
� | +0.126 |
Krug | +0.126 |
Lie | +0.124 |
crossings | +0.123 |
Neg logit contributions | |
anson | -0.169 |
rem | -0.146 |
� | -0.137 |
lement | -0.125 |
roman | -0.125 |
Token | � |
Token ID | 247 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
@@ | +0.070 |
� | +0.067 |
� | +0.066 |
Lie | +0.060 |
Krug | +0.059 |
Neg logit contributions | |
anson | -0.274 |
rem | -0.246 |
� | -0.236 |
lement | -0.224 |
roman | -0.223 |
Token | 5 |
Token ID | 642 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
@@ | +0.126 |
Krug | +0.117 |
crossings | +0.116 |
� | +0.116 |
Lie | +0.116 |
Neg logit contributions | |
anson | -0.211 |
rem | -0.184 |
� | -0.176 |
roman | -0.162 |
lement | -0.162 |
Token | years |
Token ID | 812 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
anson | +0.365 |
rem | +0.316 |
� | +0.298 |
lement | +0.275 |
roman | +0.274 |
Neg logit contributions | |
@@ | -0.256 |
� | -0.242 |
Krug | -0.241 |
Lie | -0.239 |
crossings | -0.239 |
Token | instead |
Token ID | 2427 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
anson | +0.265 |
rem | +0.228 |
� | +0.216 |
lement | +0.198 |
roman | +0.197 |
Neg logit contributions | |
@@ | -0.189 |
� | -0.178 |
Krug | -0.177 |
crossings | -0.177 |
Lie | -0.176 |
Token | of |
Token ID | 286 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
anson | +0.280 |
rem | +0.248 |
� | +0.234 |
lement | +0.219 |
roman | +0.216 |
Neg logit contributions | |
@@ | -0.154 |
Krug | -0.145 |
Lie | -0.143 |
� | -0.142 |
crossings | -0.141 |
Token | the |
Token ID | 262 |
Feature activation | -8.70e-03 |
Pos logit contributions | |
anson | +0.105 |
rem | +0.091 |
� | +0.087 |
lement | +0.079 |
roman | +0.079 |
Neg logit contributions | |
@@ | -0.068 |
Krug | -0.064 |
� | -0.064 |
crossings | -0.064 |
Lie | -0.064 |
Token | 100 |
Token ID | 1802 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
anson | +0.021 |
rem | +0.018 |
� | +0.018 |
lement | +0.016 |
roman | +0.016 |
Neg logit contributions | |
@@ | -0.014 |
Krug | -0.013 |
� | -0.013 |
crossings | -0.013 |
Lie | -0.013 |
Token | years |
Token ID | 812 |
Feature activation | -8.17e-02 |
Pos logit contributions | |
@@ | +0.178 |
� | +0.168 |
Krug | +0.168 |
Lie | +0.167 |
crossings | +0.167 |
Neg logit contributions | |
anson | -0.227 |
rem | -0.194 |
� | -0.183 |
roman | -0.168 |
lement | -0.168 |
Token | said |
Token ID | 531 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
anson | +0.194 |
rem | +0.168 |
� | +0.159 |
lement | +0.146 |
roman | +0.146 |
Neg logit contributions | |
@@ | -0.135 |
Krug | -0.128 |
� | -0.127 |
crossings | -0.126 |
Lie | -0.126 |
Token | by |
Token ID | 416 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
@@ | +0.061 |
Krug | +0.057 |
� | +0.057 |
Lie | +0.057 |
crossings | +0.056 |
Neg logit contributions | |
anson | -0.103 |
rem | -0.090 |
� | -0.086 |
lement | -0.079 |
roman | -0.079 |
Token | IPCC |
Token ID | 35814 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
@@ | +0.229 |
Krug | +0.217 |
� | +0.214 |
Lie | +0.212 |
crossings | +0.210 |
Neg logit contributions | |
anson | -0.400 |
rem | -0.351 |
� | -0.334 |
lement | -0.310 |
roman | -0.309 |
Token |
|
Token ID | 198 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
@@ | +1.360 |
� | +1.282 |
Krug | +1.278 |
COUN | +1.273 |
crossings | +1.262 |
Neg logit contributions | |
anson | -2.003 |
rem | -1.729 |
� | -1.666 |
lement | -1.513 |
roman | -1.495 |
Token | allowed |
Token ID | 3142 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
anson | +0.501 |
rem | +0.433 |
� | +0.415 |
roman | +0.384 |
lement | +0.383 |
Neg logit contributions | |
@@ | -0.284 |
� | -0.265 |
Krug | -0.264 |
Lie | -0.262 |
crossings | -0.262 |
Token | a |
Token ID | 257 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
anson | +0.329 |
rem | +0.283 |
� | +0.266 |
lement | +0.243 |
roman | +0.243 |
Neg logit contributions | |
@@ | -0.255 |
Krug | -0.241 |
� | -0.241 |
Lie | -0.240 |
crossings | -0.238 |
Token | parliamentary |
Token ID | 17521 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
@@ | +0.114 |
� | +0.107 |
Krug | +0.107 |
Lie | +0.106 |
crossings | +0.106 |
Neg logit contributions | |
anson | -0.176 |
rem | -0.152 |
� | -0.144 |
roman | -0.133 |
lement | -0.133 |
Token | debate |
Token ID | 4384 |
Feature activation | +3.72e-03 |
Pos logit contributions | |
anson | +0.058 |
rem | +0.049 |
� | +0.047 |
lement | +0.043 |
roman | +0.043 |
Neg logit contributions | |
@@ | -0.047 |
Krug | -0.044 |
� | -0.044 |
crossings | -0.044 |
Lie | -0.044 |
Token | in |
Token ID | 287 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
@@ | +0.006 |
� | +0.006 |
Krug | +0.006 |
Lie | +0.006 |
crossings | +0.006 |
Neg logit contributions | |
anson | -0.009 |
rem | -0.008 |
� | -0.007 |
lement | -0.007 |
roman | -0.007 |
Token | July |
Token ID | 2901 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
@@ | +0.144 |
Krug | +0.136 |
� | +0.135 |
Lie | +0.135 |
crossings | +0.135 |
Neg logit contributions | |
anson | -0.183 |
rem | -0.156 |
� | -0.147 |
lement | -0.135 |
roman | -0.134 |
Token | on |
Token ID | 319 |
Feature activation | +9.64e-02 |
Pos logit contributions | |
anson | +0.070 |
rem | +0.060 |
� | +0.057 |
lement | +0.052 |
roman | +0.052 |
Neg logit contributions | |
@@ | -0.051 |
� | -0.048 |
Krug | -0.048 |
crossings | -0.047 |
Lie | -0.047 |
Token | continuing |
Token ID | 8282 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
@@ | +0.169 |
� | +0.159 |
Krug | +0.158 |
crossings | +0.158 |
Lie | +0.158 |
Neg logit contributions | |
anson | -0.222 |
rem | -0.189 |
� | -0.178 |
roman | -0.163 |
lement | -0.163 |
Token | its |
Token ID | 663 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
anson | +0.215 |
rem | +0.184 |
� | +0.173 |
roman | +0.159 |
lement | +0.158 |
Neg logit contributions | |
@@ | -0.156 |
crossings | -0.147 |
� | -0.147 |
Lie | -0.146 |
Krug | -0.146 |
Token | 18 |
Token ID | 1248 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
anson | +0.505 |
rem | +0.436 |
� | +0.413 |
lement | +0.381 |
roman | +0.380 |
Neg logit contributions | |
@@ | -0.327 |
Krug | -0.307 |
� | -0.307 |
crossings | -0.305 |
Lie | -0.304 |
Token | - |
Token ID | 12 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
@@ | +0.166 |
� | +0.156 |
Krug | +0.156 |
Lie | +0.155 |
crossings | +0.154 |
Neg logit contributions | |
anson | -0.228 |
rem | -0.197 |
� | -0.186 |
roman | -0.171 |
lement | -0.170 |
Token | for |
Token ID | 329 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
@@ | +0.386 |
Krug | +0.365 |
crossings | +0.361 |
Lie | +0.361 |
� | +0.359 |
Neg logit contributions | |
anson | -0.588 |
rem | -0.507 |
� | -0.480 |
lement | -0.443 |
roman | -0.442 |
Token | so |
Token ID | 523 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
@@ | +0.113 |
Krug | +0.107 |
crossings | +0.106 |
� | +0.105 |
Lie | +0.105 |
Neg logit contributions | |
anson | -0.172 |
rem | -0.150 |
� | -0.143 |
roman | -0.131 |
lement | -0.131 |
Token | many |
Token ID | 867 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
@@ | +0.097 |
� | +0.092 |
Krug | +0.092 |
Lie | +0.091 |
crossings | +0.090 |
Neg logit contributions | |
anson | -0.121 |
rem | -0.104 |
� | -0.098 |
roman | -0.090 |
lement | -0.089 |
Token | months |
Token ID | 1933 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
@@ | +0.251 |
Krug | +0.239 |
crossings | +0.237 |
� | +0.235 |
Lie | +0.233 |
Neg logit contributions | |
anson | -0.373 |
rem | -0.323 |
� | -0.307 |
roman | -0.283 |
lement | -0.282 |
Token | that |
Token ID | 326 |
Feature activation | +8.07e-02 |
Pos logit contributions | |
anson | +0.328 |
rem | +0.279 |
� | +0.261 |
roman | +0.238 |
lement | +0.237 |
Neg logit contributions | |
@@ | -0.271 |
Krug | -0.257 |
� | -0.256 |
crossings | -0.254 |
Lie | -0.253 |
Token | we |
Token ID | 356 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
@@ | +0.142 |
Krug | +0.135 |
� | +0.135 |
Lie | +0.134 |
crossings | +0.133 |
Neg logit contributions | |
anson | -0.183 |
rem | -0.158 |
� | -0.148 |
lement | -0.136 |
roman | -0.136 |
Token | all |
Token ID | 477 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
@@ | +0.019 |
� | +0.018 |
Krug | +0.017 |
Lie | +0.017 |
� | +0.017 |
Neg logit contributions | |
anson | -0.027 |
rem | -0.024 |
� | -0.022 |
roman | -0.021 |
lement | -0.020 |
Token | knew |
Token ID | 2993 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
@@ | +0.029 |
Krug | +0.027 |
Lie | +0.027 |
� | +0.026 |
crossings | +0.026 |
Neg logit contributions | |
anson | -0.043 |
rem | -0.037 |
� | -0.035 |
roman | -0.033 |
lement | -0.032 |
Token | about |
Token ID | 546 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
@@ | +0.034 |
Krug | +0.032 |
� | +0.032 |
crossings | +0.031 |
Lie | +0.031 |
Neg logit contributions | |
anson | -0.044 |
rem | -0.037 |
� | -0.035 |
roman | -0.032 |
lement | -0.032 |
Token | his |
Token ID | 465 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
@@ | +0.230 |
Krug | +0.219 |
� | +0.217 |
Lie | +0.215 |
crossings | +0.215 |
Neg logit contributions | |
anson | -0.290 |
rem | -0.248 |
� | -0.234 |
roman | -0.213 |
lement | -0.213 |
Token | fond |
Token ID | 16245 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
@@ | +0.061 |
Krug | +0.058 |
� | +0.058 |
crossings | +0.057 |
Lie | +0.057 |
Neg logit contributions | |
anson | -0.095 |
rem | -0.082 |
� | -0.078 |
roman | -0.072 |
lement | -0.072 |
Token | to |
Token ID | 284 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
anson | +0.134 |
rem | +0.116 |
� | +0.109 |
lement | +0.100 |
roman | +0.100 |
Neg logit contributions | |
@@ | -0.102 |
Krug | -0.098 |
� | -0.097 |
Lie | -0.096 |
crossings | -0.096 |
Token | argue |
Token ID | 7267 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
anson | +0.687 |
rem | +0.598 |
� | +0.567 |
roman | +0.522 |
lement | +0.522 |
Neg logit contributions | |
@@ | -0.434 |
Krug | -0.407 |
� | -0.406 |
Lie | -0.403 |
crossings | -0.399 |
Token | that |
Token ID | 326 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
anson | +0.225 |
rem | +0.192 |
� | +0.179 |
roman | +0.162 |
lement | +0.162 |
Neg logit contributions | |
@@ | -0.206 |
Krug | -0.199 |
� | -0.196 |
Lie | -0.194 |
crossings | -0.193 |
Token | there |
Token ID | 612 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
anson | +0.461 |
rem | +0.403 |
� | +0.382 |
roman | +0.353 |
lement | +0.353 |
Neg logit contributions | |
@@ | -0.275 |
Krug | -0.260 |
� | -0.258 |
Lie | -0.255 |
crossings | -0.254 |
Token | � |
Token ID | 447 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
anson | +0.353 |
rem | +0.306 |
� | +0.289 |
lement | +0.265 |
roman | +0.264 |
Neg logit contributions | |
@@ | -0.253 |
� | -0.240 |
Krug | -0.239 |
Lie | -0.237 |
crossings | -0.236 |
Token | � |
Token ID | 247 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
anson | +0.214 |
rem | +0.191 |
� | +0.184 |
lement | +0.174 |
roman | +0.174 |
Neg logit contributions | |
@@ | -0.056 |
� | -0.055 |
� | -0.054 |
Lie | -0.049 |
Krug | -0.048 |
Token | s |
Token ID | 82 |
Feature activation | +4.94e-03 |
Pos logit contributions | |
anson | +0.152 |
rem | +0.130 |
� | +0.124 |
roman | +0.113 |
lement | +0.113 |
Neg logit contributions | |
@@ | -0.111 |
� | -0.106 |
Krug | -0.104 |
Lie | -0.104 |
� | -0.103 |
Token | only |
Token ID | 691 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
@@ | +0.009 |
Krug | +0.008 |
� | +0.008 |
Lie | +0.008 |
crossings | +0.008 |
Neg logit contributions | |
anson | -0.011 |
rem | -0.010 |
� | -0.009 |
roman | -0.008 |
lement | -0.008 |
Token | room |
Token ID | 2119 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
anson | +1.107 |
rem | +0.983 |
� | +0.937 |
lement | +0.873 |
roman | +0.872 |
Neg logit contributions | |
@@ | -0.498 |
Krug | -0.464 |
� | -0.461 |
Lie | -0.454 |
crossings | -0.450 |
Token | for |
Token ID | 329 |
Feature activation | -8.35e-02 |
Pos logit contributions | |
anson | +0.043 |
rem | +0.037 |
� | +0.035 |
lement | +0.032 |
roman | +0.032 |
Neg logit contributions | |
@@ | -0.032 |
� | -0.030 |
Krug | -0.030 |
Lie | -0.030 |
crossings | -0.030 |
Token | one |
Token ID | 530 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
anson | +0.207 |
rem | +0.181 |
� | +0.172 |
lement | +0.158 |
roman | +0.158 |
Neg logit contributions | |
@@ | -0.130 |
Krug | -0.124 |
� | -0.122 |
Lie | -0.121 |
crossings | -0.120 |
Token | correct |
Token ID | 3376 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
anson | +0.272 |
rem | +0.231 |
� | +0.219 |
roman | +0.202 |
lement | +0.201 |
Neg logit contributions | |
@@ | -0.189 |
Krug | -0.179 |
Lie | -0.178 |
� | -0.176 |
crossings | -0.175 |
Token | the |
Token ID | 262 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
anson | +0.477 |
rem | +0.412 |
� | +0.389 |
lement | +0.356 |
roman | +0.356 |
Neg logit contributions | |
@@ | -0.346 |
Krug | -0.327 |
� | -0.326 |
Lie | -0.324 |
crossings | -0.322 |
Token | error |
Token ID | 4049 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
anson | +0.041 |
rem | +0.036 |
� | +0.035 |
roman | +0.032 |
lement | +0.032 |
Neg logit contributions | |
@@ | -0.017 |
Krug | -0.015 |
� | -0.015 |
Lie | -0.015 |
crossings | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
@@ | +0.191 |
Krug | +0.180 |
� | +0.179 |
Lie | +0.178 |
crossings | +0.178 |
Neg logit contributions | |
anson | -0.267 |
rem | -0.230 |
� | -0.217 |
roman | -0.199 |
lement | -0.198 |
Token | Ryan |
Token ID | 6047 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
@@ | +0.108 |
Krug | +0.104 |
� | +0.103 |
Lie | +0.102 |
crossings | +0.101 |
Neg logit contributions | |
anson | -0.145 |
rem | -0.125 |
� | -0.118 |
lement | -0.108 |
roman | -0.108 |
Token | said |
Token ID | 531 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
@@ | +0.126 |
Krug | +0.117 |
� | +0.117 |
Lie | +0.116 |
crossings | +0.115 |
Neg logit contributions | |
anson | -0.281 |
rem | -0.249 |
� | -0.237 |
lement | -0.221 |
roman | -0.220 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
anson | +0.285 |
rem | +0.243 |
� | +0.227 |
lement | +0.207 |
roman | +0.203 |
Neg logit contributions | |
@@ | -0.269 |
� | -0.258 |
Krug | -0.257 |
Lie | -0.256 |
crossings | -0.253 |
Token | He |
Token ID | 679 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
@@ | +0.254 |
Krug | +0.241 |
� | +0.238 |
Lie | +0.237 |
COUN | +0.234 |
Neg logit contributions | |
anson | -0.346 |
rem | -0.294 |
� | -0.278 |
roman | -0.256 |
lement | -0.255 |
Token | did |
Token ID | 750 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
@@ | +0.016 |
� | +0.015 |
Krug | +0.015 |
crossings | +0.014 |
Lie | +0.014 |
Neg logit contributions | |
anson | -0.033 |
rem | -0.029 |
� | -0.028 |
roman | -0.026 |
lement | -0.026 |
Token | not |
Token ID | 407 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
anson | +0.131 |
rem | +0.113 |
� | +0.108 |
roman | +0.098 |
lement | +0.098 |
Neg logit contributions | |
@@ | -0.090 |
Krug | -0.086 |
� | -0.085 |
Lie | -0.085 |
crossings | -0.084 |
Token | know |
Token ID | 760 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
anson | +0.365 |
rem | +0.315 |
� | +0.304 |
roman | +0.281 |
aston | +0.278 |
Neg logit contributions | |
@@ | -0.208 |
Krug | -0.189 |
Lie | -0.187 |
crossings | -0.186 |
� | -0.184 |
Token | Previously |
Token ID | 29875 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
anson | +0.100 |
rem | +0.087 |
� | +0.083 |
lement | +0.076 |
roman | +0.076 |
Neg logit contributions | |
@@ | -0.063 |
Krug | -0.060 |
� | -0.060 |
Lie | -0.059 |
crossings | -0.058 |
Token | : |
Token ID | 25 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
anson | +0.063 |
rem | +0.054 |
� | +0.052 |
roman | +0.048 |
lement | +0.048 |
Neg logit contributions | |
@@ | -0.037 |
Krug | -0.035 |
Lie | -0.035 |
Nemesis | -0.034 |
Claim | -0.034 |
Token | K |
Token ID | 509 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
@@ | +0.051 |
Krug | +0.049 |
� | +0.048 |
Lie | +0.048 |
Nemesis | +0.047 |
Neg logit contributions | |
anson | -0.087 |
rem | -0.076 |
� | -0.072 |
roman | -0.067 |
lement | -0.067 |
Token | orm |
Token ID | 579 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
anson | +0.364 |
rem | +0.311 |
� | +0.297 |
roman | +0.270 |
lement | +0.269 |
Neg logit contributions | |
@@ | -0.279 |
� | -0.267 |
Krug | -0.267 |
Lie | -0.264 |
crossings | -0.261 |
Token | a |
Token ID | 64 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
@@ | +1.122 |
Krug | +1.075 |
� | +1.069 |
Lie | +1.055 |
crossings | +1.048 |
Neg logit contributions | |
anson | -1.686 |
rem | -1.462 |
� | -1.399 |
lement | -1.283 |
roman | -1.283 |
Token | S |
Token ID | 311 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
anson | +0.117 |
rem | +0.102 |
� | +0.097 |
lement | +0.089 |
roman | +0.089 |
Neg logit contributions | |
@@ | -0.074 |
Krug | -0.070 |
� | -0.070 |
Lie | -0.069 |
crossings | -0.069 |
Token | utra |
Token ID | 35076 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
anson | +0.840 |
rem | +0.727 |
� | +0.687 |
lement | +0.629 |
roman | +0.628 |
Neg logit contributions | |
@@ | -0.607 |
� | -0.576 |
Krug | -0.575 |
Lie | -0.570 |
crossings | -0.566 |
Token | , |
Token ID | 11 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
anson | +1.246 |
rem | +1.101 |
� | +1.044 |
lement | +0.963 |
roman | +0.958 |
Neg logit contributions | |
@@ | -0.716 |
� | -0.671 |
Lie | -0.666 |
Krug | -0.666 |
crossings | -0.662 |
Token | L |
Token ID | 406 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
anson | +0.045 |
rem | +0.039 |
� | +0.037 |
lement | +0.034 |
roman | +0.034 |
Neg logit contributions | |
@@ | -0.027 |
Krug | -0.025 |
� | -0.025 |
Lie | -0.025 |
crossings | -0.025 |
Token | aid |
Token ID | 1698 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
@@ | +0.439 |
Krug | +0.417 |
� | +0.415 |
Lie | +0.412 |
crossings | +0.408 |
Neg logit contributions | |
anson | -0.567 |
rem | -0.488 |
� | -0.464 |
roman | -0.421 |
lement | -0.420 |
Token | back |
Token ID | 1891 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
anson | +1.628 |
rem | +1.422 |
� | +1.334 |
lement | +1.232 |
roman | +1.224 |
Neg logit contributions | |
@@ | -1.109 |
� | -1.049 |
Krug | -1.046 |
Lie | -1.036 |
crossings | -1.036 |
Token | fixed |
Token ID | 5969 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
@@ | +0.585 |
� | +0.552 |
Krug | +0.549 |
Scand | +0.543 |
� | +0.540 |
Neg logit contributions | |
anson | -0.822 |
rem | -0.704 |
� | -0.665 |
roman | -0.618 |
lement | -0.615 |
Token | the |
Token ID | 262 |
Feature activation | -7.90e-03 |
Pos logit contributions | |
@@ | +0.716 |
Krug | +0.674 |
� | +0.671 |
Lie | +0.665 |
crossings | +0.664 |
Neg logit contributions | |
anson | -1.032 |
rem | -0.892 |
� | -0.842 |
lement | -0.773 |
roman | -0.772 |
Token | problem |
Token ID | 1917 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
anson | +0.020 |
rem | +0.018 |
� | +0.017 |
lement | +0.015 |
roman | +0.015 |
Neg logit contributions | |
@@ | -0.012 |
Krug | -0.011 |
� | -0.011 |
crossings | -0.011 |
Lie | -0.011 |
Token | completely |
Token ID | 3190 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
@@ | +0.299 |
Krug | +0.281 |
� | +0.281 |
Lie | +0.278 |
crossings | +0.277 |
Neg logit contributions | |
anson | -0.476 |
rem | -0.415 |
� | -0.393 |
roman | -0.362 |
lement | -0.362 |
Token | . |
Token ID | 13 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
@@ | +0.042 |
� | +0.040 |
Krug | +0.040 |
Lie | +0.039 |
crossings | +0.039 |
Neg logit contributions | |
anson | -0.066 |
rem | -0.058 |
� | -0.054 |
lement | -0.050 |
roman | -0.050 |
Token | The |
Token ID | 383 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
anson | +0.217 |
rem | +0.186 |
� | +0.177 |
roman | +0.164 |
lement | +0.164 |
Neg logit contributions | |
@@ | -0.132 |
Krug | -0.124 |
Nemesis | -0.121 |
� | -0.121 |
Lie | -0.121 |
Token | terrain |
Token ID | 15510 |
Feature activation | +9.49e-03 |
Pos logit contributions | |
anson | +0.239 |
rem | +0.206 |
� | +0.196 |
lement | +0.182 |
roman | +0.180 |
Neg logit contributions | |
@@ | -0.147 |
Krug | -0.138 |
� | -0.137 |
crossings | -0.136 |
Lie | -0.136 |
Token | would |
Token ID | 561 |
Feature activation | -3.09e-03 |
Pos logit contributions | |
@@ | +0.016 |
crossings | +0.015 |
Krug | +0.015 |
� | +0.015 |
Lie | +0.015 |
Neg logit contributions | |
anson | -0.023 |
rem | -0.019 |
� | -0.018 |
lement | -0.017 |
roman | -0.017 |
Token | either |
Token ID | 2035 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
anson | +0.007 |
rem | +0.006 |
� | +0.006 |
roman | +0.005 |
lement | +0.005 |
Neg logit contributions | |
@@ | -0.005 |
� | -0.005 |
Krug | -0.005 |
Lie | -0.005 |
crossings | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
@@ | +1.055 |
� | +0.980 |
Lie | +0.977 |
Krug | +0.974 |
crossings | +0.970 |
Neg logit contributions | |
anson | -1.588 |
rem | -1.357 |
� | -1.294 |
lement | -1.190 |
roman | -1.188 |
Token | ) |
Token ID | 8 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
anson | +0.086 |
rem | +0.075 |
� | +0.071 |
lement | +0.065 |
roman | +0.065 |
Neg logit contributions | |
@@ | -0.059 |
� | -0.056 |
Krug | -0.055 |
Lie | -0.055 |
crossings | -0.054 |
Token | 1998 |
Token ID | 7795 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
anson | +0.682 |
rem | +0.577 |
� | +0.547 |
lement | +0.497 |
roman | +0.496 |
Neg logit contributions | |
@@ | -0.582 |
Krug | -0.558 |
Lie | -0.551 |
� | -0.550 |
Nemesis | -0.542 |
Token | ). |
Token ID | 6739 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
anson | +0.115 |
rem | +0.097 |
� | +0.089 |
lement | +0.081 |
roman | +0.080 |
Neg logit contributions | |
@@ | -0.122 |
� | -0.118 |
Krug | -0.117 |
crossings | -0.116 |
Lie | -0.116 |
Token | Intelligent |
Token ID | 49452 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
@@ | +0.160 |
Krug | +0.151 |
� | +0.149 |
Lie | +0.149 |
Nemesis | +0.147 |
Neg logit contributions | |
anson | -0.259 |
rem | -0.224 |
� | -0.213 |
roman | -0.197 |
lement | -0.196 |
Token | , |
Token ID | 11 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
@@ | +0.197 |
Krug | +0.185 |
� | +0.185 |
crossings | +0.183 |
Lie | +0.183 |
Neg logit contributions | |
anson | -0.290 |
rem | -0.252 |
� | -0.239 |
roman | -0.218 |
lement | -0.218 |
Token | attractive |
Token ID | 10966 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
anson | +0.787 |
rem | +0.688 |
� | +0.653 |
roman | +0.602 |
lement | +0.602 |
Neg logit contributions | |
@@ | -0.480 |
Krug | -0.450 |
� | -0.449 |
Lie | -0.443 |
crossings | -0.442 |
Token | , |
Token ID | 11 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
@@ | +0.020 |
Krug | +0.019 |
� | +0.019 |
crossings | +0.019 |
Lie | +0.018 |
Neg logit contributions | |
anson | -0.035 |
rem | -0.030 |
� | -0.029 |
roman | -0.026 |
lement | -0.026 |
Token | well |
Token ID | 880 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
anson | +0.613 |
rem | +0.528 |
� | +0.499 |
roman | +0.456 |
lement | +0.456 |
Neg logit contributions | |
@@ | -0.455 |
Krug | -0.430 |
� | -0.428 |
Lie | -0.424 |
crossings | -0.423 |
Token | - |
Token ID | 12 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
@@ | +0.310 |
� | +0.288 |
Krug | +0.288 |
Lie | +0.284 |
crossings | +0.284 |
Neg logit contributions | |
anson | -0.527 |
rem | -0.460 |
� | -0.439 |
lement | -0.404 |
roman | -0.403 |
Token | beh |
Token ID | 20709 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
anson | +0.349 |
rem | +0.290 |
� | +0.279 |
lement | +0.252 |
roman | +0.248 |
Neg logit contributions | |
@@ | -0.305 |
� | -0.289 |
Krug | -0.288 |
Lie | -0.288 |
crossings | -0.283 |
Token | aving |
Token ID | 2703 |
Feature activation | +4.28e-04 |
Pos logit contributions | |
@@ | +0.049 |
� | +0.044 |
� | +0.044 |
crossings | +0.044 |
Lie | +0.044 |
Neg logit contributions | |
anson | -0.064 |
rem | -0.053 |
� | -0.052 |
roman | -0.047 |
lement | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
@@ | +0.001 |
Krug | +0.001 |
� | +0.001 |
crossings | +0.001 |
Lie | +0.001 |
Neg logit contributions | |
anson | -0.001 |
rem | -0.001 |
� | -0.001 |
lement | -0.001 |
roman | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
@@ | +0.651 |
Krug | +0.620 |
� | +0.615 |
crossings | +0.611 |
Lie | +0.608 |
Neg logit contributions | |
anson | -0.741 |
rem | -0.634 |
� | -0.595 |
lement | -0.536 |
roman | -0.533 |
Token | different |
Token ID | 1180 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
@@ | +0.048 |
Krug | +0.045 |
Lie | +0.045 |
� | +0.045 |
crossings | +0.044 |
Neg logit contributions | |
anson | -0.072 |
rem | -0.062 |
� | -0.059 |
roman | -0.054 |
lement | -0.054 |
Token | story |
Token ID | 1621 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
anson | +0.140 |
rem | +0.120 |
� | +0.114 |
lement | +0.104 |
roman | +0.103 |
Neg logit contributions | |
@@ | -0.104 |
Krug | -0.099 |
� | -0.097 |
Lie | -0.097 |
crossings | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
anson | +0.752 |
rem | +0.644 |
� | +0.609 |
lement | +0.553 |
roman | +0.552 |
Neg logit contributions | |
@@ | -0.575 |
Krug | -0.543 |
� | -0.542 |
Lie | -0.538 |
crossings | -0.537 |
Token | A |
Token ID | 317 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
@@ | +0.039 |
Krug | +0.037 |
Lie | +0.036 |
� | +0.036 |
Nemesis | +0.035 |
Neg logit contributions | |
anson | -0.058 |
rem | -0.050 |
� | -0.047 |
roman | -0.043 |
lement | -0.043 |
Token | Nav |
Token ID | 13244 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
@@ | +0.097 |
Krug | +0.092 |
� | +0.091 |
Lie | +0.091 |
crossings | +0.089 |
Neg logit contributions | |
anson | -0.139 |
rem | -0.120 |
� | -0.114 |
roman | -0.105 |
lement | -0.105 |
Token | igator |
Token ID | 23823 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
anson | +0.589 |
rem | +0.467 |
� | +0.424 |
lement | +0.359 |
roman | +0.351 |
Neg logit contributions | |
@@ | -0.999 |
Lie | -0.958 |
Krug | -0.956 |
� | -0.956 |
crossings | -0.945 |
Token | news |
Token ID | 1705 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
anson | +0.881 |
rem | +0.770 |
� | +0.729 |
roman | +0.673 |
lement | +0.672 |
Neg logit contributions | |
@@ | -0.557 |
Krug | -0.523 |
� | -0.521 |
Lie | -0.519 |
crossings | -0.513 |
Token | release |
Token ID | 2650 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
anson | +0.422 |
rem | +0.359 |
� | +0.343 |
roman | +0.309 |
lement | +0.308 |
Neg logit contributions | |
@@ | -0.349 |
� | -0.332 |
Krug | -0.331 |
Lie | -0.327 |
� | -0.325 |
Token | on |
Token ID | 319 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
anson | +1.233 |
rem | +1.075 |
� | +1.023 |
lement | +0.938 |
roman | +0.935 |
Neg logit contributions | |
@@ | -0.784 |
Krug | -0.733 |
� | -0.729 |
crossings | -0.726 |
Lie | -0.723 |
Token | March |
Token ID | 2805 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
@@ | +0.069 |
Krug | +0.066 |
� | +0.065 |
Lie | +0.065 |
crossings | +0.065 |
Neg logit contributions | |
anson | -0.090 |
rem | -0.077 |
� | -0.073 |
roman | -0.066 |
lement | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
@@ | +0.502 |
Krug | +0.471 |
� | +0.468 |
crossings | +0.467 |
Lie | +0.464 |
Neg logit contributions | |
anson | -0.837 |
rem | -0.730 |
� | -0.695 |
roman | -0.639 |
lement | -0.637 |
Token | We |
Token ID | 775 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
anson | +0.591 |
rem | +0.510 |
� | +0.484 |
roman | +0.451 |
lement | +0.446 |
Neg logit contributions | |
@@ | -0.374 |
Krug | -0.356 |
Lie | -0.347 |
Nemesis | -0.345 |
� | -0.345 |
Token | need |
Token ID | 761 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
anson | +0.497 |
rem | +0.432 |
� | +0.411 |
roman | +0.383 |
lement | +0.378 |
Neg logit contributions | |
@@ | -0.314 |
Krug | -0.294 |
� | -0.294 |
Lie | -0.293 |
crossings | -0.292 |
Token | brave |
Token ID | 14802 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
anson | +0.564 |
rem | +0.486 |
� | +0.457 |
lement | +0.417 |
roman | +0.416 |
Neg logit contributions | |
@@ | -0.448 |
� | -0.425 |
Krug | -0.425 |
Lie | -0.421 |
crossings | -0.419 |
Token | explorers |
Token ID | 45531 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
anson | +0.713 |
rem | +0.613 |
� | +0.578 |
lement | +0.531 |
roman | +0.528 |
Neg logit contributions | |
@@ | -0.545 |
Krug | -0.521 |
� | -0.517 |
crossings | -0.515 |
Lie | -0.512 |
Token | willing |
Token ID | 4684 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
anson | +1.349 |
rem | +1.167 |
� | +1.103 |
lement | +1.009 |
roman | +1.009 |
Neg logit contributions | |
@@ | -0.977 |
Krug | -0.926 |
� | -0.922 |
Lie | -0.916 |
crossings | -0.914 |
Token | to |
Token ID | 284 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
anson | +1.752 |
rem | +1.490 |
� | +1.386 |
lement | +1.266 |
roman | +1.239 |
Neg logit contributions | |
@@ | -1.639 |
Krug | -1.560 |
� | -1.559 |
Lie | -1.546 |
COUN | -1.527 |
Token | find |
Token ID | 1064 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
anson | +0.160 |
rem | +0.139 |
� | +0.133 |
roman | +0.123 |
lement | +0.122 |
Neg logit contributions | |
@@ | -0.098 |
Krug | -0.092 |
crossings | -0.091 |
Lie | -0.091 |
� | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | -8.22e-02 |
Pos logit contributions | |
@@ | +0.036 |
Krug | +0.034 |
crossings | +0.034 |
� | +0.034 |
Lie | +0.034 |
Neg logit contributions | |
anson | -0.049 |
rem | -0.042 |
� | -0.040 |
roman | -0.037 |
lement | -0.037 |
Token | source |
Token ID | 2723 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
anson | +0.207 |
rem | +0.181 |
� | +0.172 |
roman | +0.159 |
lement | +0.159 |
Neg logit contributions | |
@@ | -0.124 |
Krug | -0.118 |
crossings | -0.117 |
� | -0.116 |
Lie | -0.115 |
Token | of |
Token ID | 286 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
anson | +0.869 |
rem | +0.766 |
� | +0.729 |
lement | +0.679 |
roman | +0.672 |
Neg logit contributions | |
@@ | -0.455 |
� | -0.427 |
Krug | -0.426 |
Lie | -0.422 |
crossings | -0.415 |
Token | waiter |
Token ID | 46612 |
Feature activation | -9.18e-02 |
Pos logit contributions | |
anson | +0.769 |
rem | +0.664 |
� | +0.629 |
lement | +0.576 |
roman | +0.575 |
Neg logit contributions | |
@@ | -0.541 |
� | -0.510 |
Krug | -0.506 |
Lie | -0.504 |
crossings | -0.503 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
anson | +0.219 |
rem | +0.190 |
� | +0.180 |
roman | +0.165 |
lement | +0.165 |
Neg logit contributions | |
@@ | -0.152 |
� | -0.143 |
Krug | -0.142 |
Lie | -0.141 |
crossings | -0.141 |
Token |
|
Token ID | 198 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
@@ | +0.210 |
Lie | +0.194 |
Krug | +0.190 |
Claim | +0.186 |
territory | +0.186 |
Neg logit contributions | |
anson | -0.313 |
rem | -0.261 |
� | -0.252 |
roman | -0.235 |
uana | -0.234 |
Token | � |
Token ID | 447 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
@@ | +0.132 |
Lie | +0.127 |
� | +0.125 |
Krug | +0.125 |
� | +0.122 |
Neg logit contributions | |
anson | -0.172 |
rem | -0.146 |
� | -0.138 |
lement | -0.127 |
roman | -0.126 |
Token | � |
Token ID | 246 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
anson | +0.092 |
rem | +0.083 |
� | +0.080 |
lement | +0.076 |
roman | +0.076 |
Neg logit contributions | |
@@ | -0.019 |
� | -0.016 |
Krug | -0.016 |
Lie | -0.016 |
crossings | -0.015 |
Token | Ex |
Token ID | 3109 |
Feature activation | -9.10e-01 |
Pos logit contributions | |
anson | +0.254 |
rem | +0.219 |
� | +0.209 |
roman | +0.193 |
lement | +0.193 |
Neg logit contributions | |
@@ | -0.159 |
Lie | -0.152 |
� | -0.150 |
Krug | -0.149 |
� | -0.146 |
Token | - |
Token ID | 12 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
anson | +2.146 |
rem | +1.864 |
� | +1.753 |
lement | +1.608 |
roman | +1.605 |
Neg logit contributions | |
@@ | -1.520 |
� | -1.442 |
Krug | -1.438 |
Lie | -1.426 |
crossings | -1.418 |
Token | gay |
Token ID | 22744 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
@@ | +0.232 |
Lie | +0.221 |
� | +0.219 |
Krug | +0.218 |
� | +0.214 |
Neg logit contributions | |
anson | -0.337 |
rem | -0.289 |
� | -0.278 |
lement | -0.255 |
roman | -0.253 |
Token | � |
Token ID | 447 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
@@ | +0.182 |
� | +0.172 |
Krug | +0.171 |
Lie | +0.170 |
crossings | +0.168 |
Neg logit contributions | |
anson | -0.254 |
rem | -0.220 |
� | -0.208 |
lement | -0.191 |
roman | -0.190 |
Token | � |
Token ID | 247 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
@@ | +0.028 |
Krug | +0.025 |
Lie | +0.025 |
crossings | +0.025 |
� | +0.025 |
Neg logit contributions | |
anson | -0.094 |
rem | -0.085 |
� | -0.082 |
lement | -0.077 |
roman | -0.077 |
Token | Christian |
Token ID | 4302 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
@@ | +0.091 |
� | +0.085 |
Krug | +0.085 |
Lie | +0.085 |
crossings | +0.084 |
Neg logit contributions | |
anson | -0.138 |
rem | -0.120 |
� | -0.113 |
lement | -0.104 |
roman | -0.104 |
Token | be |
Token ID | 307 |
Feature activation | -6.43e-03 |
Pos logit contributions | |
anson | +0.780 |
rem | +0.679 |
� | +0.642 |
lement | +0.590 |
roman | +0.589 |
Neg logit contributions | |
@@ | -0.523 |
� | -0.494 |
Krug | -0.493 |
Lie | -0.489 |
crossings | -0.485 |
Token | empty |
Token ID | 6565 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
anson | +0.016 |
rem | +0.014 |
� | +0.013 |
lement | +0.012 |
roman | +0.012 |
Neg logit contributions | |
@@ | -0.010 |
Krug | -0.010 |
� | -0.010 |
Lie | -0.010 |
crossings | -0.010 |
Token | ) |
Token ID | 8 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
@@ | +0.044 |
Krug | +0.042 |
� | +0.041 |
Lie | +0.041 |
crossings | +0.041 |
Neg logit contributions | |
anson | -0.062 |
rem | -0.053 |
� | -0.050 |
lement | -0.046 |
roman | -0.046 |
Token | Example |
Token ID | 17934 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
@@ | +0.370 |
Krug | +0.338 |
crossings | +0.336 |
COUN | +0.334 |
Nemesis | +0.334 |
Neg logit contributions | |
anson | -0.487 |
rem | -0.414 |
� | -0.396 |
lement | -0.363 |
aston | -0.360 |
Token | : |
Token ID | 25 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
anson | +0.153 |
rem | +0.133 |
� | +0.126 |
lement | +0.115 |
roman | +0.115 |
Neg logit contributions | |
@@ | -0.105 |
Krug | -0.098 |
� | -0.098 |
Lie | -0.097 |
crossings | -0.096 |
Token | vi |
Token ID | 25357 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
@@ | +0.228 |
Krug | +0.212 |
� | +0.211 |
Lie | +0.209 |
crossings | +0.208 |
Neg logit contributions | |
anson | -0.319 |
rem | -0.274 |
� | -0.262 |
lement | -0.241 |
roman | -0.239 |
Token | : |
Token ID | 25 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
anson | +2.308 |
rem | +2.010 |
� | +1.885 |
roman | +1.728 |
lement | +1.724 |
Neg logit contributions | |
@@ | -1.696 |
� | -1.622 |
Krug | -1.622 |
Lie | -1.606 |
crossings | -1.594 |
Token | no |
Token ID | 3919 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
@@ | +0.056 |
� | +0.052 |
Krug | +0.052 |
Lie | +0.052 |
crossings | +0.051 |
Neg logit contributions | |
anson | -0.077 |
rem | -0.066 |
� | -0.063 |
lement | -0.057 |
roman | -0.057 |
Token | ai |
Token ID | 1872 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
anson | +1.387 |
rem | +1.197 |
� | +1.137 |
lement | +1.044 |
roman | +1.037 |
Neg logit contributions | |
@@ | -0.976 |
� | -0.921 |
Krug | -0.919 |
Lie | -0.912 |
crossings | -0.905 |
Token | : |
Token ID | 25 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
anson | +1.118 |
rem | +0.965 |
� | +0.913 |
lement | +0.834 |
roman | +0.832 |
Neg logit contributions | |
@@ | -0.821 |
Krug | -0.774 |
� | -0.774 |
Lie | -0.767 |
crossings | -0.763 |
Token | sw |
Token ID | 2032 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
anson | +0.090 |
rem | +0.076 |
� | +0.073 |
lement | +0.067 |
roman | +0.066 |
Neg logit contributions | |
@@ | -0.072 |
� | -0.068 |
Krug | -0.068 |
Lie | -0.067 |
crossings | -0.067 |
Token | featuring |
Token ID | 9593 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
@@ | +0.381 |
Krug | +0.356 |
� | +0.354 |
crossings | +0.352 |
Lie | +0.351 |
Neg logit contributions | |
anson | -0.554 |
rem | -0.481 |
� | -0.455 |
roman | -0.420 |
lement | -0.419 |
Token | Tone |
Token ID | 45362 |
Feature activation | -9.54e-02 |
Pos logit contributions | |
anson | +0.828 |
rem | +0.721 |
� | +0.684 |
lement | +0.629 |
roman | +0.628 |
Neg logit contributions | |
@@ | -0.545 |
Krug | -0.514 |
� | -0.512 |
Lie | -0.507 |
crossings | -0.505 |
Token | controls |
Token ID | 6973 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
anson | +0.239 |
rem | +0.209 |
� | +0.198 |
lement | +0.183 |
roman | +0.182 |
Neg logit contributions | |
@@ | -0.146 |
� | -0.137 |
Krug | -0.137 |
Lie | -0.136 |
crossings | -0.135 |
Token | , |
Token ID | 11 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
anson | +1.111 |
rem | +0.963 |
� | +0.911 |
lement | +0.835 |
roman | +0.834 |
Neg logit contributions | |
@@ | -0.784 |
Krug | -0.738 |
� | -0.737 |
Lie | -0.732 |
crossings | -0.728 |
Token | Comp |
Token ID | 3082 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
anson | +0.592 |
rem | +0.516 |
� | +0.490 |
lement | +0.451 |
roman | +0.451 |
Neg logit contributions | |
@@ | -0.378 |
Krug | -0.355 |
� | -0.354 |
Lie | -0.350 |
crossings | -0.350 |
Token | ression |
Token ID | 2234 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
@@ | +0.345 |
Krug | +0.328 |
� | +0.328 |
Lie | +0.326 |
crossings | +0.324 |
Neg logit contributions | |
anson | -0.388 |
rem | -0.330 |
� | -0.310 |
roman | -0.281 |
lement | -0.280 |
Token | , |
Token ID | 11 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
anson | +2.315 |
rem | +2.033 |
� | +1.899 |
roman | +1.742 |
lement | +1.740 |
Neg logit contributions | |
@@ | -1.565 |
� | -1.520 |
Krug | -1.506 |
Lie | -1.492 |
� | -1.477 |
Token | and |
Token ID | 290 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
anson | +0.518 |
rem | +0.451 |
� | +0.427 |
roman | +0.393 |
lement | +0.393 |
Neg logit contributions | |
@@ | -0.340 |
Krug | -0.320 |
� | -0.319 |
Lie | -0.316 |
crossings | -0.315 |
Token | Analog |
Token ID | 50088 |
Feature activation | +6.95e-02 |
Pos logit contributions | |
anson | +0.367 |
rem | +0.321 |
� | +0.305 |
roman | +0.281 |
lement | +0.281 |
Neg logit contributions | |
@@ | -0.219 |
Krug | -0.205 |
� | -0.205 |
Lie | -0.203 |
crossings | -0.202 |
Token | Tape |
Token ID | 40849 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
@@ | +0.103 |
� | +0.097 |
Krug | +0.096 |
Lie | +0.095 |
crossings | +0.095 |
Neg logit contributions | |
anson | -0.177 |
rem | -0.155 |
� | -0.148 |
lement | -0.136 |
roman | -0.136 |
Token | S |
Token ID | 311 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
anson | +0.352 |
rem | +0.305 |
� | +0.291 |
lement | +0.267 |
roman | +0.266 |
Neg logit contributions | |
@@ | -0.232 |
Krug | -0.217 |
� | -0.217 |
Lie | -0.216 |
crossings | -0.215 |
Token | had |
Token ID | 550 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
anson | +0.397 |
rem | +0.343 |
� | +0.327 |
roman | +0.302 |
lement | +0.301 |
Neg logit contributions | |
@@ | -0.254 |
Krug | -0.238 |
� | -0.236 |
Lie | -0.234 |
crossings | -0.234 |
Token | you |
Token ID | 345 |
Feature activation | -9.31e-02 |
Pos logit contributions | |
anson | +1.030 |
rem | +0.889 |
� | +0.845 |
lement | +0.776 |
roman | +0.771 |
Neg logit contributions | |
@@ | -0.716 |
Krug | -0.682 |
� | -0.672 |
Lie | -0.669 |
crossings | -0.668 |
Token | can |
Token ID | 460 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
anson | +0.232 |
rem | +0.201 |
� | +0.192 |
lement | +0.176 |
roman | +0.176 |
Neg logit contributions | |
@@ | -0.145 |
Krug | -0.137 |
Lie | -0.135 |
crossings | -0.134 |
� | -0.134 |
Token | summon |
Token ID | 11457 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
anson | +0.086 |
rem | +0.074 |
� | +0.071 |
lement | +0.065 |
roman | +0.065 |
Neg logit contributions | |
@@ | -0.057 |
Krug | -0.054 |
� | -0.053 |
Lie | -0.053 |
crossings | -0.053 |
Token | a |
Token ID | 257 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
anson | +0.457 |
rem | +0.395 |
� | +0.375 |
lement | +0.344 |
roman | +0.344 |
Neg logit contributions | |
@@ | -0.319 |
Krug | -0.303 |
� | -0.299 |
Lie | -0.297 |
crossings | -0.296 |
Token | demon |
Token ID | 3222 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
anson | +0.071 |
rem | +0.061 |
� | +0.058 |
roman | +0.054 |
lement | +0.054 |
Neg logit contributions | |
@@ | -0.045 |
Krug | -0.043 |
� | -0.042 |
Lie | -0.042 |
Nemesis | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
anson | +2.106 |
rem | +1.828 |
� | +1.726 |
lement | +1.582 |
roman | +1.579 |
Neg logit contributions | |
@@ | -1.487 |
� | -1.411 |
Krug | -1.408 |
Lie | -1.396 |
crossings | -1.385 |
Token | you |
Token ID | 345 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
anson | +0.140 |
rem | +0.119 |
� | +0.113 |
lement | +0.103 |
roman | +0.103 |
Neg logit contributions | |
@@ | -0.108 |
Krug | -0.103 |
Nemesis | -0.100 |
Lie | -0.100 |
crossings | -0.100 |
Token | can |
Token ID | 460 |
Feature activation | +6.10e-03 |
Pos logit contributions | |
anson | +0.183 |
rem | +0.155 |
� | +0.147 |
lement | +0.134 |
roman | +0.134 |
Neg logit contributions | |
@@ | -0.143 |
Krug | -0.135 |
Lie | -0.133 |
� | -0.133 |
crossings | -0.132 |
Token | do |
Token ID | 466 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
@@ | +0.011 |
Krug | +0.010 |
Lie | +0.010 |
crossings | +0.010 |
Nemesis | +0.010 |
Neg logit contributions | |
anson | -0.014 |
rem | -0.012 |
� | -0.012 |
roman | -0.011 |
lement | -0.011 |
Token | A |
Token ID | 317 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
anson | +0.035 |
rem | +0.030 |
� | +0.028 |
lement | +0.026 |
roman | +0.026 |
Neg logit contributions | |
@@ | -0.025 |
Krug | -0.023 |
Lie | -0.023 |
� | -0.023 |
crossings | -0.023 |