Token | particularly |
Token ID | 3573 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
tested | +1.548 |
expected | +1.496 |
coal | +1.477 |
isot | +1.426 |
cube | +1.402 |
Neg logit contributions | |
citiz | -1.889 |
eleph | -1.802 |
lation | -1.732 |
teasp | -1.671 |
TAIN | -1.668 |
Token | harrowing |
Token ID | 45736 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
citiz | +3.016 |
eleph | +2.836 |
lation | +2.765 |
TAIN | +2.586 |
田 | +2.519 |
Neg logit contributions | |
tested | -3.045 |
coal | -2.965 |
expected | -2.957 |
cube | -2.837 |
cli | -2.792 |
Token | example |
Token ID | 1672 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
tested | +1.000 |
expected | +0.966 |
coal | +0.957 |
isot | +0.919 |
cube | +0.910 |
Neg logit contributions | |
citiz | -1.230 |
eleph | -1.168 |
lation | -1.125 |
teasp | -1.079 |
TAIN | -1.076 |
Token | located |
Token ID | 5140 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
citiz | +2.782 |
eleph | +2.649 |
lation | +2.585 |
TAIN | +2.427 |
田 | +2.364 |
Neg logit contributions | |
tested | -2.592 |
coal | -2.504 |
expected | -2.504 |
cube | -2.384 |
isot | -2.361 |
Token | at |
Token ID | 379 |
Feature activation | +1.72e+01 |
Pos logit contributions | |
citiz | +2.129 |
eleph | +2.033 |
lation | +1.971 |
TAIN | +1.861 |
teasp | +1.829 |
Neg logit contributions | |
tested | -1.910 |
expected | -1.858 |
coal | -1.847 |
cube | -1.759 |
isot | -1.737 |
Token | exactly |
Token ID | 3446 |
Feature activation | +2.38e+01 |
Pos logit contributions | |
citiz | +11.221 |
eleph | +10.772 |
lation | +10.307 |
TAIN | +9.908 |
田 | +9.650 |
Neg logit contributions | |
tested | -11.831 |
expected | -11.470 |
coal | -11.416 |
cube | -10.831 |
cli | -10.798 |
Token | the |
Token ID | 262 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
citiz | +11.993 |
lation | +11.899 |
eleph | +11.433 |
TAIN | +10.951 |
田 | +10.887 |
Neg logit contributions | |
tested | -16.505 |
coal | -16.257 |
expected | -15.950 |
Reviewer | -15.500 |
cube | -15.426 |
Token | midway |
Token ID | 37720 |
Feature activation | +9.52e+00 |
Pos logit contributions | |
citiz | +7.905 |
lation | +7.268 |
eleph | +7.182 |
TAIN | +6.671 |
bottom | +6.617 |
Neg logit contributions | |
tested | -11.043 |
coal | -10.772 |
expected | -10.599 |
cli | -10.206 |
cube | -10.116 |
Token | point |
Token ID | 966 |
Feature activation | +5.20e+00 |
Pos logit contributions | |
citiz | +5.774 |
eleph | +5.580 |
lation | +5.475 |
TAIN | +4.936 |
田 | +4.874 |
Neg logit contributions | |
tested | -8.065 |
coal | -7.924 |
expected | -7.907 |
cube | -7.433 |
cli | -7.385 |
Token | of |
Token ID | 286 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
citiz | +4.034 |
eleph | +3.867 |
lation | +3.801 |
TAIN | +3.560 |
田 | +3.488 |
Neg logit contributions | |
tested | -3.776 |
coal | -3.658 |
expected | -3.657 |
cube | -3.464 |
isot | -3.433 |
Token | the |
Token ID | 262 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
citiz | +3.157 |
eleph | +2.997 |
lation | +2.907 |
TAIN | +2.755 |
田 | +2.657 |
Neg logit contributions | |
tested | -3.141 |
expected | -3.037 |
coal | -3.036 |
cube | -2.878 |
isot | -2.860 |
Token | are |
Token ID | 389 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
citiz | +3.433 |
eleph | +3.233 |
lation | +3.186 |
TAIN | +2.958 |
田 | +2.878 |
Neg logit contributions | |
tested | -3.638 |
expected | -3.518 |
coal | -3.496 |
cube | -3.345 |
isot | -3.329 |
Token | also |
Token ID | 635 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
tested | +0.939 |
expected | +0.910 |
coal | +0.903 |
isot | +0.865 |
cube | +0.859 |
Neg logit contributions | |
citiz | -1.051 |
eleph | -1.001 |
lation | -0.965 |
TAIN | -0.918 |
teasp | -0.916 |
Token | major |
Token ID | 1688 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
tested | +1.491 |
expected | +1.446 |
coal | +1.432 |
isot | +1.377 |
cube | +1.361 |
Neg logit contributions | |
citiz | -1.649 |
eleph | -1.573 |
lation | -1.512 |
teasp | -1.444 |
TAIN | -1.439 |
Token | global |
Token ID | 3298 |
Feature activation | -4.70e+00 |
Pos logit contributions | |
tested | +1.279 |
expected | +1.235 |
coal | +1.227 |
isot | +1.176 |
cube | +1.160 |
Neg logit contributions | |
citiz | -1.633 |
eleph | -1.563 |
lation | -1.507 |
teasp | -1.441 |
TAIN | -1.436 |
Token | factors |
Token ID | 5087 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
tested | +2.963 |
expected | +2.852 |
coal | +2.844 |
isot | +2.744 |
shocks | +2.673 |
Neg logit contributions | |
citiz | -4.135 |
eleph | -3.976 |
lation | -3.842 |
teasp | -3.731 |
TAIN | -3.647 |
Token | at |
Token ID | 379 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
citiz | +2.748 |
eleph | +2.564 |
lation | +2.506 |
TAIN | +2.345 |
田 | +2.249 |
Neg logit contributions | |
tested | -2.998 |
coal | -2.919 |
expected | -2.896 |
cube | -2.773 |
cli | -2.761 |
Token | play |
Token ID | 711 |
Feature activation | +5.25e+00 |
Pos logit contributions | |
citiz | +8.469 |
lation | +8.001 |
eleph | +7.560 |
emphasis | +7.511 |
TAIN | +7.127 |
Neg logit contributions | |
coal | -19.402 |
expected | -19.292 |
tested | -19.255 |
cli | -18.807 |
cube | -18.749 |
Token | when |
Token ID | 618 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
citiz | +4.001 |
eleph | +3.766 |
lation | +3.664 |
TAIN | +3.471 |
teasp | +3.328 |
Neg logit contributions | |
tested | -3.909 |
coal | -3.779 |
expected | -3.757 |
cube | -3.587 |
isot | -3.570 |
Token | it |
Token ID | 340 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
citiz | +0.925 |
eleph | +0.879 |
lation | +0.856 |
TAIN | +0.816 |
teasp | +0.799 |
Neg logit contributions | |
tested | -0.674 |
expected | -0.649 |
coal | -0.645 |
cube | -0.610 |
isot | -0.608 |
Token | comes |
Token ID | 2058 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
citiz | +0.693 |
eleph | +0.659 |
lation | +0.642 |
TAIN | +0.611 |
teasp | +0.601 |
Neg logit contributions | |
tested | -0.523 |
expected | -0.505 |
coal | -0.501 |
cube | -0.474 |
isot | -0.474 |
Token | to |
Token ID | 284 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
citiz | +0.622 |
eleph | +0.588 |
lation | +0.572 |
TAIN | +0.542 |
teasp | +0.532 |
Neg logit contributions | |
tested | -0.588 |
expected | -0.570 |
coal | -0.566 |
isot | -0.540 |
cube | -0.540 |
Token | Texas |
Token ID | 3936 |
Feature activation | +7.86e+00 |
Pos logit contributions | |
citiz | +8.932 |
lation | +7.998 |
eleph | +7.931 |
TAIN | +7.811 |
bottom | +7.321 |
Neg logit contributions | |
tested | -8.708 |
coal | -8.478 |
expected | -8.386 |
cube | -8.012 |
crit | -7.885 |
Token | Tech |
Token ID | 9634 |
Feature activation | +8.19e+00 |
Pos logit contributions | |
citiz | +5.310 |
eleph | +4.828 |
lation | +4.728 |
TAIN | +4.463 |
田 | +4.303 |
Neg logit contributions | |
tested | -6.457 |
expected | -6.187 |
coal | -6.158 |
cube | -5.956 |
cli | -5.831 |
Token | Police |
Token ID | 4287 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
citiz | +6.126 |
eleph | +5.606 |
lation | +5.526 |
TAIN | +5.270 |
田 | +5.101 |
Neg logit contributions | |
tested | -6.051 |
coal | -5.875 |
expected | -5.863 |
cube | -5.545 |
cli | -5.446 |
Token | Department |
Token ID | 2732 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
citiz | +0.523 |
eleph | +0.499 |
lation | +0.486 |
TAIN | +0.464 |
teasp | +0.460 |
Neg logit contributions | |
tested | -0.337 |
expected | -0.324 |
coal | -0.322 |
isot | -0.304 |
cube | -0.303 |
Token | . |
Token ID | 13 |
Feature activation | +8.19e+00 |
Pos logit contributions | |
citiz | +3.851 |
eleph | +3.623 |
lation | +3.528 |
TAIN | +3.366 |
田 | +3.243 |
Neg logit contributions | |
tested | -3.464 |
coal | -3.363 |
expected | -3.356 |
cube | -3.187 |
cli | -3.151 |
Token | At |
Token ID | 1629 |
Feature activation | +2.27e+01 |
Pos logit contributions | |
citiz | +5.797 |
TAIN | +5.328 |
eleph | +5.293 |
lation | +5.286 |
田 | +5.022 |
Neg logit contributions | |
tested | -6.024 |
expected | -5.956 |
coal | -5.949 |
cube | -5.726 |
isot | -5.724 |
Token | this |
Token ID | 428 |
Feature activation | +1.30e+01 |
Pos logit contributions | |
citiz | +12.310 |
lation | +11.112 |
TAIN | +11.099 |
eleph | +10.972 |
bottom | +10.282 |
Neg logit contributions | |
tested | -16.136 |
expected | -15.874 |
coal | -15.534 |
crit | -14.776 |
Reviewer | -14.492 |
Token | time |
Token ID | 640 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
citiz | +6.576 |
eleph | +5.696 |
lation | +5.613 |
TAIN | +5.263 |
emphasis | +4.837 |
Neg logit contributions | |
tested | -11.749 |
coal | -11.525 |
expected | -11.492 |
cube | -11.081 |
cli | -10.769 |
Token | , |
Token ID | 11 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
citiz | +1.820 |
eleph | +1.719 |
lation | +1.666 |
TAIN | +1.591 |
teasp | +1.555 |
Neg logit contributions | |
tested | -1.550 |
expected | -1.503 |
coal | -1.495 |
cube | -1.421 |
isot | -1.411 |
Token | the |
Token ID | 262 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
citiz | +3.028 |
eleph | +2.831 |
lation | +2.728 |
TAIN | +2.618 |
teasp | +2.521 |
Neg logit contributions | |
tested | -2.894 |
expected | -2.814 |
coal | -2.806 |
cube | -2.691 |
isot | -2.635 |
Token | shooter |
Token ID | 14273 |
Feature activation | +8.38e+00 |
Pos logit contributions | |
citiz | +3.593 |
eleph | +3.320 |
lation | +3.248 |
TAIN | +3.092 |
田 | +2.973 |
Neg logit contributions | |
tested | -3.289 |
coal | -3.184 |
expected | -3.170 |
cube | -3.031 |
cli | -2.955 |
Token | health |
Token ID | 1535 |
Feature activation | -6.18e+00 |
Pos logit contributions | |
citiz | +2.282 |
eleph | +2.138 |
lation | +2.094 |
TAIN | +1.986 |
teasp | +1.925 |
Neg logit contributions | |
tested | -2.054 |
expected | -1.991 |
coal | -1.989 |
cube | -1.897 |
isot | -1.864 |
Token | and |
Token ID | 290 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
tested | +3.705 |
expected | +3.518 |
coal | +3.448 |
isot | +3.420 |
crit | +3.280 |
Neg logit contributions | |
citiz | -5.571 |
eleph | -5.466 |
lation | -5.149 |
teasp | -5.097 |
utenant | -4.877 |
Token | recovery |
Token ID | 7628 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
citiz | +2.764 |
eleph | +2.549 |
lation | +2.478 |
TAIN | +2.338 |
teasp | +2.227 |
Neg logit contributions | |
tested | -3.547 |
expected | -3.457 |
coal | -3.444 |
cube | -3.310 |
cli | -3.261 |
Token | purposes |
Token ID | 4959 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
tested | +2.157 |
expected | +2.070 |
coal | +2.047 |
isot | +1.993 |
cube | +1.925 |
Neg logit contributions | |
citiz | -2.896 |
eleph | -2.795 |
lation | -2.656 |
teasp | -2.588 |
TAIN | -2.523 |
Token | . |
Token ID | 13 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
tested | +1.983 |
expected | +1.909 |
coal | +1.884 |
isot | +1.807 |
cube | +1.786 |
Neg logit contributions | |
citiz | -2.738 |
eleph | -2.609 |
lation | -2.508 |
teasp | -2.427 |
TAIN | -2.404 |
Token | At |
Token ID | 1629 |
Feature activation | +2.16e+01 |
Pos logit contributions | |
citiz | +2.642 |
lation | +2.494 |
eleph | +2.461 |
TAIN | +2.407 |
田 | +2.300 |
Neg logit contributions | |
tested | -2.276 |
expected | -2.230 |
coal | -2.198 |
isot | -2.135 |
cube | -2.088 |
Token | most |
Token ID | 749 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
citiz | +12.864 |
lation | +12.281 |
TAIN | +12.058 |
eleph | +11.842 |
bottom | +11.509 |
Neg logit contributions | |
coal | -14.013 |
tested | -14.005 |
expected | -13.653 |
cli | -13.038 |
cube | -12.815 |
Token | , |
Token ID | 11 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
citiz | +9.394 |
lation | +8.798 |
eleph | +8.395 |
TAIN | +8.306 |
emphasis | +7.843 |
Neg logit contributions | |
coal | -9.631 |
expected | -9.346 |
tested | -9.340 |
cube | -9.009 |
cli | -8.825 |
Token | it |
Token ID | 340 |
Feature activation | +5.83e+00 |
Pos logit contributions | |
citiz | +4.081 |
eleph | +3.781 |
lation | +3.776 |
TAIN | +3.560 |
teasp | +3.366 |
Neg logit contributions | |
tested | -3.758 |
coal | -3.681 |
expected | -3.661 |
cube | -3.482 |
cli | -3.442 |
Token | � |
Token ID | 447 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
citiz | +4.372 |
lation | +4.111 |
eleph | +4.085 |
TAIN | +3.862 |
teasp | +3.665 |
Neg logit contributions | |
tested | -4.319 |
expected | -4.210 |
coal | -4.182 |
cube | -3.977 |
cli | -3.890 |
Token | � |
Token ID | 247 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
tested | +0.513 |
expected | +0.497 |
coal | +0.495 |
isot | +0.471 |
cube | +0.470 |
Neg logit contributions | |
citiz | -0.537 |
eleph | -0.510 |
lation | -0.490 |
TAIN | -0.466 |
teasp | -0.460 |
Token | we |
Token ID | 356 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
citiz | +3.763 |
eleph | +3.551 |
lation | +3.486 |
TAIN | +3.270 |
田 | +3.139 |
Neg logit contributions | |
tested | -3.525 |
expected | -3.422 |
coal | -3.419 |
cube | -3.275 |
cli | -3.228 |
Token | are |
Token ID | 389 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
citiz | +3.036 |
eleph | +2.854 |
lation | +2.823 |
TAIN | +2.638 |
田 | +2.546 |
Neg logit contributions | |
tested | -3.161 |
expected | -3.096 |
coal | -3.053 |
cube | -2.936 |
isot | -2.920 |
Token | involved |
Token ID | 2950 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
citiz | +1.422 |
eleph | +1.313 |
lation | +1.257 |
TAIN | +1.159 |
teasp | +1.117 |
Neg logit contributions | |
tested | -2.394 |
expected | -2.335 |
coal | -2.327 |
cube | -2.251 |
isot | -2.247 |
Token | in |
Token ID | 287 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
tested | +1.558 |
expected | +1.491 |
coal | +1.484 |
isot | +1.407 |
cube | +1.405 |
Neg logit contributions | |
citiz | -2.200 |
eleph | -2.094 |
lation | -2.042 |
TAIN | -1.948 |
teasp | -1.933 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
citiz | +0.155 |
eleph | +0.148 |
lation | +0.143 |
TAIN | +0.136 |
teasp | +0.135 |
Neg logit contributions | |
tested | -0.125 |
expected | -0.120 |
coal | -0.120 |
isot | -0.114 |
cube | -0.113 |
Token | At |
Token ID | 1629 |
Feature activation | +2.15e+01 |
Pos logit contributions | |
citiz | +1.070 |
eleph | +1.016 |
lation | +0.993 |
TAIN | +0.947 |
teasp | +0.927 |
Neg logit contributions | |
tested | -0.799 |
expected | -0.772 |
coal | -0.765 |
isot | -0.728 |
cube | -0.725 |
Token | the |
Token ID | 262 |
Feature activation | +1.76e+01 |
Pos logit contributions | |
citiz | +13.800 |
lation | +12.546 |
eleph | +12.338 |
TAIN | +11.729 |
bottom | +11.286 |
Neg logit contributions | |
tested | -13.996 |
expected | -13.479 |
coal | -13.362 |
cli | -12.800 |
csv | -12.693 |
Token | moment |
Token ID | 2589 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
citiz | +11.932 |
lation | +10.548 |
eleph | +10.315 |
bottom | +10.102 |
TAIN | +9.626 |
Neg logit contributions | |
tested | -12.028 |
expected | -11.737 |
coal | -11.347 |
cube | -10.959 |
Reviewer | -10.803 |
Token | our |
Token ID | 674 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
tested | +0.944 |
expected | +0.909 |
coal | +0.906 |
isot | +0.859 |
cube | +0.855 |
Neg logit contributions | |
citiz | -1.254 |
eleph | -1.198 |
lation | -1.158 |
teasp | -1.116 |
TAIN | -1.108 |
Token | main |
Token ID | 1388 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
citiz | +1.041 |
eleph | +0.978 |
lation | +0.946 |
TAIN | +0.890 |
teasp | +0.874 |
Neg logit contributions | |
tested | -1.156 |
expected | -1.122 |
coal | -1.115 |
cube | -1.068 |
isot | -1.067 |
Token | campaigns |
Token ID | 9964 |
Feature activation | +5.68e+00 |
Pos logit contributions | |
citiz | +3.251 |
eleph | +3.015 |
lation | +2.958 |
TAIN | +2.760 |
田 | +2.642 |
Neg logit contributions | |
tested | -3.887 |
expected | -3.797 |
coal | -3.726 |
cube | -3.593 |
isot | -3.574 |
Token | the |
Token ID | 262 |
Feature activation | +6.92e+00 |
Pos logit contributions | |
citiz | +6.715 |
eleph | +6.193 |
lation | +6.122 |
TAIN | +5.571 |
emphasis | +5.499 |
Neg logit contributions | |
tested | -7.708 |
coal | -7.571 |
expected | -7.458 |
cli | -7.181 |
cube | -7.122 |
Token | fact |
Token ID | 1109 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
citiz | +5.012 |
eleph | +4.635 |
lation | +4.607 |
TAIN | +4.234 |
emphasis | +4.102 |
Neg logit contributions | |
tested | -5.330 |
coal | -5.208 |
expected | -5.158 |
cli | -4.902 |
cube | -4.881 |
Token | that |
Token ID | 326 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
citiz | +2.429 |
eleph | +2.266 |
lation | +2.189 |
TAIN | +2.017 |
teasp | +1.990 |
Neg logit contributions | |
tested | -3.539 |
expected | -3.473 |
coal | -3.455 |
isot | -3.307 |
cube | -3.306 |
Token | they |
Token ID | 484 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
tested | +0.281 |
expected | +0.270 |
coal | +0.268 |
isot | +0.258 |
cube | +0.256 |
Neg logit contributions | |
citiz | -0.335 |
eleph | -0.319 |
lation | -0.307 |
TAIN | -0.294 |
teasp | -0.293 |
Token | have |
Token ID | 423 |
Feature activation | +6.60e+00 |
Pos logit contributions | |
citiz | +0.524 |
eleph | +0.496 |
lation | +0.482 |
TAIN | +0.456 |
teasp | +0.448 |
Neg logit contributions | |
tested | -0.486 |
expected | -0.471 |
coal | -0.468 |
cube | -0.445 |
isot | -0.445 |
Token | at |
Token ID | 379 |
Feature activation | +2.14e+01 |
Pos logit contributions | |
citiz | +5.081 |
eleph | +4.730 |
lation | +4.693 |
TAIN | +4.329 |
田 | +4.223 |
Neg logit contributions | |
tested | -4.749 |
coal | -4.663 |
expected | -4.640 |
cube | -4.408 |
cli | -4.388 |
Token | some |
Token ID | 617 |
Feature activation | +1.48e+01 |
Pos logit contributions | |
citiz | +12.228 |
lation | +11.468 |
eleph | +11.299 |
TAIN | +11.123 |
bottom | +11.087 |
Neg logit contributions | |
coal | -14.788 |
expected | -14.406 |
tested | -14.163 |
cli | -14.110 |
cube | -13.765 |
Token | point |
Token ID | 966 |
Feature activation | +8.48e+00 |
Pos logit contributions | |
citiz | +8.434 |
lation | +7.973 |
eleph | +7.515 |
affili | +7.178 |
TAIN | +7.097 |
Neg logit contributions | |
coal | -12.029 |
expected | -11.990 |
tested | -11.895 |
cube | -11.274 |
cli | -11.179 |
Token | left |
Token ID | 1364 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
citiz | +6.075 |
lation | +5.867 |
eleph | +5.766 |
TAIN | +5.275 |
affili | +5.166 |
Neg logit contributions | |
tested | -6.109 |
coal | -6.021 |
expected | -5.953 |
cube | -5.715 |
cli | -5.708 |
Token | this |
Token ID | 428 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
citiz | +4.133 |
eleph | +3.938 |
lation | +3.816 |
TAIN | +3.552 |
teasp | +3.425 |
Neg logit contributions | |
tested | -3.962 |
coal | -3.897 |
expected | -3.894 |
cube | -3.693 |
cli | -3.649 |
Token | path |
Token ID | 3108 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
citiz | +2.592 |
eleph | +2.447 |
lation | +2.377 |
TAIN | +2.232 |
teasp | +2.184 |
Neg logit contributions | |
tested | -2.405 |
expected | -2.343 |
coal | -2.336 |
cube | -2.205 |
cli | -2.190 |
Token | 's |
Token ID | 338 |
Feature activation | -4.82e+00 |
Pos logit contributions | |
tested | +0.287 |
expected | +0.277 |
coal | +0.275 |
isot | +0.263 |
cube | +0.262 |
Neg logit contributions | |
citiz | -0.345 |
eleph | -0.328 |
lation | -0.317 |
TAIN | -0.302 |
teasp | -0.300 |
Token | impact |
Token ID | 2928 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
tested | +3.256 |
expected | +3.153 |
coal | +3.096 |
isot | +3.049 |
cube | +2.920 |
Neg logit contributions | |
citiz | -4.030 |
eleph | -3.948 |
teasp | -3.702 |
lation | -3.699 |
TAIN | -3.580 |
Token | on |
Token ID | 319 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
citiz | +2.008 |
eleph | +1.893 |
lation | +1.816 |
TAIN | +1.713 |
teasp | +1.666 |
Neg logit contributions | |
tested | -2.273 |
expected | -2.211 |
coal | -2.206 |
cube | -2.126 |
isot | -2.099 |
Token | the |
Token ID | 262 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
citiz | +2.153 |
eleph | +2.003 |
lation | +1.961 |
TAIN | +1.850 |
teasp | +1.800 |
Neg logit contributions | |
tested | -2.034 |
expected | -1.964 |
coal | -1.963 |
cube | -1.861 |
isot | -1.851 |
Token | city |
Token ID | 1748 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
citiz | +2.457 |
eleph | +2.247 |
lation | +2.216 |
TAIN | +2.063 |
teasp | +1.999 |
Neg logit contributions | |
tested | -2.767 |
expected | -2.676 |
coal | -2.674 |
cube | -2.545 |
isot | -2.525 |
Token | at |
Token ID | 379 |
Feature activation | +2.10e+01 |
Pos logit contributions | |
citiz | +0.126 |
eleph | +0.120 |
lation | +0.117 |
TAIN | +0.111 |
teasp | +0.110 |
Neg logit contributions | |
tested | -0.102 |
expected | -0.098 |
coal | -0.098 |
isot | -0.093 |
cube | -0.093 |
Token | large |
Token ID | 1588 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
citiz | +7.978 |
lation | +7.314 |
bottom | +7.027 |
eleph | +6.947 |
TAIN | +6.626 |
Neg logit contributions | |
tested | -18.803 |
coal | -18.442 |
expected | -18.233 |
cli | -17.695 |
cube | -17.515 |
Token | . |
Token ID | 13 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
tested | +0.489 |
expected | +0.472 |
coal | +0.468 |
isot | +0.445 |
cube | +0.443 |
Neg logit contributions | |
citiz | -0.651 |
eleph | -0.619 |
lation | -0.602 |
TAIN | -0.573 |
teasp | -0.569 |
Token |
|
Token ID | 198 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
citiz | +0.657 |
eleph | +0.621 |
lation | +0.611 |
TAIN | +0.584 |
teasp | +0.568 |
Neg logit contributions | |
tested | -0.507 |
expected | -0.490 |
coal | -0.486 |
isot | -0.465 |
cube | -0.460 |
Token |
|
Token ID | 198 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
citiz | +1.909 |
lation | +1.905 |
TAIN | +1.817 |
eleph | +1.748 |
田 | +1.725 |
Neg logit contributions | |
tested | -1.742 |
coal | -1.700 |
expected | -1.699 |
isot | -1.646 |
cube | -1.582 |
Token | " |
Token ID | 1 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
citiz | +2.834 |
lation | +2.781 |
TAIN | +2.661 |
eleph | +2.566 |
田 | +2.565 |
Neg logit contributions | |
tested | -2.686 |
expected | -2.601 |
coal | -2.581 |
isot | -2.543 |
cube | -2.428 |
Token | a |
Token ID | 257 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
citiz | +2.362 |
eleph | +2.187 |
lation | +2.169 |
TAIN | +2.021 |
teasp | +1.954 |
Neg logit contributions | |
tested | -2.352 |
expected | -2.280 |
coal | -2.267 |
cube | -2.151 |
isot | -2.134 |
Token | post |
Token ID | 1281 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
citiz | +1.881 |
eleph | +1.752 |
lation | +1.732 |
TAIN | +1.618 |
teasp | +1.583 |
Neg logit contributions | |
tested | -1.805 |
expected | -1.755 |
coal | -1.738 |
cube | -1.655 |
isot | -1.648 |
Token | - |
Token ID | 12 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
citiz | +1.155 |
eleph | +1.103 |
lation | +1.084 |
TAIN | +1.035 |
teasp | +1.022 |
Neg logit contributions | |
tested | -0.592 |
expected | -0.568 |
coal | -0.559 |
isot | -0.524 |
cube | -0.523 |
Token | capitalist |
Token ID | 49970 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
citiz | +1.993 |
eleph | +1.880 |
lation | +1.868 |
TAIN | +1.748 |
teasp | +1.709 |
Neg logit contributions | |
tested | -1.642 |
expected | -1.593 |
coal | -1.566 |
isot | -1.516 |
cube | -1.495 |
Token | economy |
Token ID | 3773 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
citiz | +3.847 |
lation | +3.521 |
eleph | +3.486 |
TAIN | +3.253 |
田 | +3.050 |
Neg logit contributions | |
tested | -4.596 |
expected | -4.467 |
coal | -4.403 |
cube | -4.232 |
cli | -4.173 |
Token | at |
Token ID | 379 |
Feature activation | +2.10e+01 |
Pos logit contributions | |
citiz | +1.547 |
eleph | +1.466 |
lation | +1.429 |
TAIN | +1.353 |
teasp | +1.329 |
Neg logit contributions | |
tested | -1.308 |
expected | -1.265 |
coal | -1.258 |
cube | -1.197 |
isot | -1.194 |
Token | a |
Token ID | 257 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
citiz | +11.662 |
lation | +11.310 |
eleph | +10.218 |
TAIN | +10.001 |
bottom | +9.878 |
Neg logit contributions | |
tested | -15.122 |
coal | -14.909 |
expected | -14.846 |
Reviewer | -14.391 |
cli | -14.294 |
Token | local |
Token ID | 1957 |
Feature activation | +6.74e+00 |
Pos logit contributions | |
citiz | +7.126 |
lation | +6.621 |
eleph | +6.318 |
TAIN | +5.814 |
teasp | +5.611 |
Neg logit contributions | |
tested | -8.750 |
expected | -8.609 |
coal | -8.560 |
cube | -8.209 |
Reviewer | -7.982 |
Token | level |
Token ID | 1241 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
citiz | +3.801 |
eleph | +3.444 |
lation | +3.417 |
TAIN | +3.048 |
teasp | +2.956 |
Neg logit contributions | |
tested | -6.345 |
expected | -6.242 |
coal | -6.163 |
cube | -5.898 |
cli | -5.828 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
citiz | +2.297 |
eleph | +2.179 |
lation | +2.129 |
TAIN | +2.013 |
teasp | +1.975 |
Neg logit contributions | |
tested | -1.905 |
expected | -1.842 |
coal | -1.832 |
cube | -1.740 |
isot | -1.734 |
Token | Alternative |
Token ID | 27182 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
tested | +0.191 |
expected | +0.185 |
coal | +0.183 |
isot | +0.174 |
cube | +0.174 |
Neg logit contributions | |
citiz | -0.244 |
eleph | -0.232 |
lation | -0.225 |
TAIN | -0.215 |
teasp | -0.212 |
Token | .[ |
Token ID | 3693 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
citiz | +2.821 |
eleph | +2.671 |
lation | +2.602 |
TAIN | +2.452 |
teasp | +2.398 |
Neg logit contributions | |
tested | -2.677 |
expected | -2.593 |
coal | -2.580 |
isot | -2.456 |
cube | -2.448 |
Token | 50 |
Token ID | 1120 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
citiz | +2.539 |
eleph | +2.376 |
lation | +2.374 |
TAIN | +2.220 |
teasp | +2.145 |
Neg logit contributions | |
tested | -2.613 |
expected | -2.552 |
coal | -2.518 |
isot | -2.429 |
cube | -2.413 |
Token | ][ |
Token ID | 7131 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
citiz | +2.575 |
eleph | +2.437 |
lation | +2.369 |
TAIN | +2.218 |
teasp | +2.167 |
Neg logit contributions | |
tested | -2.918 |
coal | -2.826 |
expected | -2.822 |
isot | -2.691 |
cube | -2.681 |
Token | 51 |
Token ID | 4349 |
Feature activation | +5.90e+00 |
Pos logit contributions | |
citiz | +3.329 |
lation | +3.166 |
eleph | +3.115 |
TAIN | +2.971 |
田 | +2.818 |
Neg logit contributions | |
tested | -3.521 |
expected | -3.445 |
coal | -3.384 |
isot | -3.305 |
cube | -3.279 |
Token | ] |
Token ID | 60 |
Feature activation | +5.71e+00 |
Pos logit contributions | |
citiz | +3.858 |
lation | +3.622 |
eleph | +3.606 |
TAIN | +3.324 |
teasp | +3.207 |
Neg logit contributions | |
tested | -4.896 |
coal | -4.764 |
expected | -4.716 |
isot | -4.519 |
cli | -4.508 |
Token | At |
Token ID | 1629 |
Feature activation | +2.06e+01 |
Pos logit contributions | |
citiz | +4.286 |
lation | +4.105 |
eleph | +4.047 |
TAIN | +3.923 |
utenant | +3.767 |
Neg logit contributions | |
tested | -4.060 |
expected | -3.984 |
coal | -3.914 |
isot | -3.812 |
cube | -3.747 |
Token | least |
Token ID | 1551 |
Feature activation | +1.65e+01 |
Pos logit contributions | |
citiz | +11.817 |
lation | +11.396 |
eleph | +11.212 |
TAIN | +10.396 |
emphasis | +10.099 |
Neg logit contributions | |
tested | -14.779 |
coal | -13.952 |
expected | -13.821 |
csv | -13.355 |
cli | -13.353 |
Token | 18 |
Token ID | 1248 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
citiz | +8.320 |
lation | +7.684 |
eleph | +7.392 |
emphasis | +7.163 |
TAIN | +7.073 |
Neg logit contributions | |
tested | -13.269 |
coal | -13.240 |
expected | -12.871 |
cube | -12.519 |
csv | -12.514 |
Token | , |
Token ID | 11 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
citiz | +1.711 |
eleph | +1.624 |
lation | +1.575 |
TAIN | +1.496 |
teasp | +1.466 |
Neg logit contributions | |
tested | -1.476 |
expected | -1.428 |
coal | -1.414 |
cube | -1.352 |
isot | -1.344 |
Token | 000 |
Token ID | 830 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
tested | +0.017 |
expected | +0.017 |
coal | +0.016 |
isot | +0.016 |
cube | +0.016 |
Neg logit contributions | |
citiz | -0.026 |
eleph | -0.025 |
lation | -0.024 |
TAIN | -0.023 |
teasp | -0.023 |
Token | people |
Token ID | 661 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
citiz | +2.258 |
eleph | +2.120 |
lation | +2.055 |
TAIN | +1.939 |
teasp | +1.881 |
Neg logit contributions | |
tested | -2.331 |
expected | -2.257 |
coal | -2.238 |
cube | -2.145 |
isot | -2.134 |
Token | legitimate |
Token ID | 9829 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
tested | +2.687 |
coal | +2.555 |
expected | +2.555 |
isot | +2.476 |
cube | +2.426 |
Neg logit contributions | |
citiz | -2.771 |
eleph | -2.674 |
lation | -2.510 |
teasp | -2.426 |
TAIN | -2.398 |
Token | phenomenon |
Token ID | 10733 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
tested | +0.280 |
expected | +0.271 |
coal | +0.269 |
isot | +0.259 |
cube | +0.257 |
Neg logit contributions | |
citiz | -0.285 |
eleph | -0.271 |
lation | -0.260 |
TAIN | -0.247 |
teasp | -0.245 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
tested | +1.484 |
expected | +1.432 |
coal | +1.416 |
isot | +1.354 |
cube | +1.344 |
Neg logit contributions | |
citiz | -1.850 |
eleph | -1.755 |
lation | -1.694 |
TAIN | -1.625 |
teasp | -1.617 |
Token |
|
Token ID | 198 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
citiz | +1.050 |
eleph | +0.995 |
lation | +0.978 |
TAIN | +0.933 |
teasp | +0.908 |
Neg logit contributions | |
tested | -0.819 |
expected | -0.794 |
coal | -0.786 |
isot | -0.753 |
cube | -0.747 |
Token |
|
Token ID | 198 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
citiz | +0.113 |
eleph | +0.107 |
lation | +0.105 |
TAIN | +0.100 |
teasp | +0.098 |
Neg logit contributions | |
tested | -0.084 |
expected | -0.081 |
coal | -0.081 |
isot | -0.077 |
cube | -0.076 |
Token | At |
Token ID | 2953 |
Feature activation | +2.06e+01 |
Pos logit contributions | |
citiz | +1.351 |
lation | +1.279 |
eleph | +1.257 |
TAIN | +1.218 |
田 | +1.176 |
Neg logit contributions | |
tested | -1.239 |
expected | -1.203 |
coal | -1.194 |
isot | -1.159 |
cube | -1.141 |
Token | a |
Token ID | 257 |
Feature activation | +9.32e+00 |
Pos logit contributions | |
citiz | +12.671 |
lation | +11.818 |
eleph | +11.724 |
TAIN | +11.555 |
bottom | +11.391 |
Neg logit contributions | |
tested | -13.257 |
coal | -13.230 |
expected | -12.937 |
cube | -12.711 |
Reviewer | -12.249 |
Token | moment |
Token ID | 2589 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
citiz | +6.640 |
eleph | +6.091 |
lation | +6.082 |
TAIN | +5.709 |
teasp | +5.437 |
Neg logit contributions | |
tested | -6.952 |
expected | -6.708 |
coal | -6.694 |
cube | -6.573 |
isot | -6.330 |
Token | when |
Token ID | 618 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
tested | +2.356 |
expected | +2.264 |
coal | +2.243 |
isot | +2.150 |
cube | +2.134 |
Neg logit contributions | |
citiz | -2.657 |
eleph | -2.510 |
lation | -2.415 |
TAIN | -2.302 |
teasp | -2.283 |
Token | the |
Token ID | 262 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
tested | +4.427 |
expected | +4.267 |
coal | +4.212 |
isot | +4.123 |
cli | +3.948 |
Neg logit contributions | |
citiz | -4.760 |
eleph | -4.593 |
lation | -4.411 |
teasp | -4.330 |
TAIN | -4.185 |
Token | transgender |
Token ID | 10637 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
tested | +1.599 |
expected | +1.549 |
coal | +1.541 |
isot | +1.479 |
cube | +1.463 |
Neg logit contributions | |
citiz | -1.605 |
eleph | -1.522 |
lation | -1.459 |
TAIN | -1.389 |
teasp | -1.382 |
Token | hate |
Token ID | 5465 |
Feature activation | +3.93e+00 |
Pos logit contributions | |
citiz | +2.433 |
eleph | +2.271 |
lation | +2.241 |
TAIN | +2.097 |
teasp | +2.040 |
Neg logit contributions | |
tested | -2.324 |
expected | -2.255 |
coal | -2.226 |
cube | -2.143 |
isot | -2.107 |
Token | consuming |
Token ID | 18587 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
citiz | +3.004 |
eleph | +2.829 |
lation | +2.810 |
TAIN | +2.630 |
teasp | +2.546 |
Neg logit contributions | |
tested | -2.887 |
expected | -2.815 |
coal | -2.766 |
cube | -2.643 |
isot | -2.634 |
Token | his |
Token ID | 465 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
citiz | +1.675 |
eleph | +1.566 |
lation | +1.535 |
TAIN | +1.449 |
teasp | +1.406 |
Neg logit contributions | |
tested | -1.542 |
expected | -1.494 |
coal | -1.475 |
cube | -1.413 |
isot | -1.402 |
Token | supporters |
Token ID | 5941 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
citiz | +0.115 |
eleph | +0.108 |
lation | +0.104 |
TAIN | +0.098 |
teasp | +0.096 |
Neg logit contributions | |
tested | -0.140 |
expected | -0.136 |
coal | -0.136 |
isot | -0.130 |
cube | -0.130 |
Token | . |
Token ID | 13 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
citiz | +1.321 |
eleph | +1.250 |
lation | +1.217 |
TAIN | +1.157 |
teasp | +1.130 |
Neg logit contributions | |
tested | -1.139 |
expected | -1.104 |
coal | -1.096 |
isot | -1.043 |
cube | -1.042 |
Token | At |
Token ID | 1629 |
Feature activation | +2.05e+01 |
Pos logit contributions | |
citiz | +1.763 |
eleph | +1.659 |
lation | +1.646 |
TAIN | +1.568 |
田 | +1.514 |
Neg logit contributions | |
tested | -1.471 |
expected | -1.427 |
coal | -1.413 |
isot | -1.361 |
cube | -1.343 |
Token | Hillary |
Token ID | 5073 |
Feature activation | +4.44e+00 |
Pos logit contributions | |
citiz | +12.314 |
lation | +11.423 |
eleph | +11.326 |
bottom | +11.070 |
TAIN | +11.038 |
Neg logit contributions | |
tested | -13.565 |
coal | -13.018 |
expected | -12.973 |
cube | -12.588 |
isot | -12.475 |
Token | . |
Token ID | 13 |
Feature activation | -4.66e+00 |
Pos logit contributions | |
citiz | +3.687 |
eleph | +3.486 |
lation | +3.435 |
TAIN | +3.261 |
teasp | +3.201 |
Neg logit contributions | |
tested | -2.981 |
coal | -2.894 |
expected | -2.874 |
cube | -2.749 |
isot | -2.746 |
Token | His |
Token ID | 2399 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
tested | +3.104 |
expected | +2.991 |
coal | +2.985 |
cube | +2.778 |
crit | +2.758 |
Neg logit contributions | |
citiz | -4.034 |
eleph | -3.870 |
lation | -3.663 |
teasp | -3.535 |
TAIN | -3.473 |
Token | supporters |
Token ID | 5941 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
tested | +0.837 |
expected | +0.812 |
coal | +0.808 |
isot | +0.774 |
cube | +0.770 |
Neg logit contributions | |
citiz | -0.785 |
eleph | -0.745 |
lation | -0.714 |
TAIN | -0.677 |
teasp | -0.671 |
Token | are |
Token ID | 389 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
tested | +1.933 |
expected | +1.862 |
coal | +1.835 |
isot | +1.763 |
cube | +1.751 |
Neg logit contributions | |
citiz | -2.301 |
eleph | -2.210 |
lation | -2.099 |
teasp | -2.049 |
TAIN | -2.009 |
Token | . |
Token ID | 13 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
citiz | +1.083 |
eleph | +1.029 |
lation | +0.999 |
TAIN | +0.952 |
teasp | +0.939 |
Neg logit contributions | |
tested | -0.859 |
expected | -0.830 |
coal | -0.824 |
isot | -0.782 |
cube | -0.782 |
Token |
|
Token ID | 198 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
citiz | +2.378 |
eleph | +2.239 |
lation | +2.232 |
TAIN | +2.152 |
田 | +2.054 |
Neg logit contributions | |
tested | -1.966 |
expected | -1.911 |
coal | -1.893 |
isot | -1.834 |
cube | -1.801 |
Token |
|
Token ID | 198 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
citiz | +2.143 |
lation | +2.143 |
TAIN | +2.048 |
eleph | +1.968 |
田 | +1.944 |
Neg logit contributions | |
tested | -1.962 |
coal | -1.917 |
expected | -1.914 |
isot | -1.861 |
cube | -1.789 |
Token | Starting |
Token ID | 22851 |
Feature activation | +9.43e+00 |
Pos logit contributions | |
citiz | +2.786 |
lation | +2.707 |
TAIN | +2.611 |
eleph | +2.555 |
田 | +2.490 |
Neg logit contributions | |
tested | -2.702 |
expected | -2.618 |
coal | -2.612 |
isot | -2.567 |
cube | -2.495 |
Token | at |
Token ID | 379 |
Feature activation | +1.89e+01 |
Pos logit contributions | |
citiz | +7.123 |
lation | +6.607 |
eleph | +6.591 |
TAIN | +6.313 |
utenant | +5.940 |
Neg logit contributions | |
tested | -6.615 |
expected | -6.398 |
coal | -6.379 |
cube | -6.242 |
cli | -6.120 |
Token | approximately |
Token ID | 6702 |
Feature activation | +2.04e+01 |
Pos logit contributions | |
lation | +9.410 |
citiz | +9.350 |
Noon | +8.401 |
eleph | +8.278 |
TAIN | +8.054 |
Neg logit contributions | |
tested | -14.720 |
coal | -14.303 |
expected | -14.062 |
cube | -13.961 |
Reviewer | -13.701 |
Token | 11 |
Token ID | 1367 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
lation | +8.811 |
citiz | +8.571 |
Noon | +8.186 |
emphasis | +7.710 |
eleph | +7.556 |
Neg logit contributions | |
tested | -16.447 |
coal | -16.284 |
Reviewer | -15.671 |
cube | -15.637 |
expected | -15.622 |
Token | : |
Token ID | 25 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
citiz | +6.903 |
eleph | +6.446 |
lation | +6.287 |
TAIN | +6.031 |
teasp | +5.764 |
Neg logit contributions | |
tested | -9.463 |
coal | -9.062 |
expected | -8.987 |
cube | -8.898 |
cli | -8.541 |
Token | 30 |
Token ID | 1270 |
Feature activation | +8.29e+00 |
Pos logit contributions | |
citiz | +4.791 |
lation | +4.640 |
eleph | +4.477 |
TAIN | +4.399 |
田 | +4.189 |
Neg logit contributions | |
tested | -5.661 |
expected | -5.499 |
coal | -5.466 |
isot | -5.360 |
cube | -5.227 |
Token | a |
Token ID | 257 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
citiz | +5.603 |
eleph | +5.282 |
lation | +5.076 |
TAIN | +4.848 |
teasp | +4.735 |
Neg logit contributions | |
tested | -6.607 |
coal | -6.369 |
expected | -6.340 |
cube | -6.165 |
cli | -6.031 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
lation | +3.657 |
citiz | +3.624 |
TAIN | +3.433 |
eleph | +3.294 |
utenant | +3.249 |
Neg logit contributions | |
tested | -3.703 |
expected | -3.588 |
coal | -3.576 |
isot | -3.473 |
cube | -3.381 |
Token | io |
Token ID | 952 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
citiz | +0.138 |
eleph | +0.132 |
lation | +0.127 |
TAIN | +0.122 |
teasp | +0.120 |
Neg logit contributions | |
tested | -0.108 |
expected | -0.104 |
coal | -0.103 |
isot | -0.098 |
cube | -0.098 |
Token | � |
Token ID | 447 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
citiz | +2.361 |
eleph | +2.229 |
lation | +2.181 |
TAIN | +2.038 |
teasp | +1.987 |
Neg logit contributions | |
tested | -2.439 |
expected | -2.367 |
coal | -2.332 |
cube | -2.236 |
isot | -2.228 |
Token | � |
Token ID | 251 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
tested | +0.163 |
expected | +0.159 |
coal | +0.158 |
cube | +0.153 |
isot | +0.153 |
Neg logit contributions | |
citiz | -0.089 |
eleph | -0.083 |
lation | -0.078 |
TAIN | -0.072 |
teasp | -0.071 |
Token | ) |
Token ID | 1267 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
citiz | +1.325 |
eleph | +1.254 |
lation | +1.224 |
TAIN | +1.158 |
teasp | +1.139 |
Neg logit contributions | |
tested | -1.115 |
expected | -1.079 |
coal | -1.067 |
isot | -1.016 |
cube | -1.015 |
Token | where |
Token ID | 810 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
citiz | +2.102 |
eleph | +1.993 |
lation | +1.937 |
TAIN | +1.831 |
teasp | +1.801 |
Neg logit contributions | |
tested | -1.856 |
expected | -1.798 |
coal | -1.781 |
cube | -1.703 |
isot | -1.700 |
Token | at |
Token ID | 379 |
Feature activation | +2.02e+01 |
Pos logit contributions | |
citiz | +1.085 |
eleph | +1.029 |
lation | +0.998 |
TAIN | +0.948 |
teasp | +0.934 |
Neg logit contributions | |
tested | -0.918 |
expected | -0.888 |
coal | -0.882 |
cube | -0.839 |
isot | -0.838 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e+01 |
Pos logit contributions | |
citiz | +11.908 |
lation | +10.911 |
eleph | +10.684 |
TAIN | +9.835 |
bottom | +9.381 |
Neg logit contributions | |
tested | -14.658 |
coal | -14.081 |
expected | -14.062 |
cli | -13.612 |
cube | -13.601 |
Token | moment |
Token ID | 2589 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
citiz | +10.024 |
lation | +8.747 |
eleph | +8.657 |
bottom | +8.123 |
TAIN | +7.787 |
Neg logit contributions | |
tested | -11.706 |
expected | -11.449 |
coal | -11.154 |
Reviewer | -10.673 |
csv | -10.633 |
Token | we |
Token ID | 356 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
tested | +1.883 |
coal | +1.807 |
expected | +1.806 |
isot | +1.725 |
cube | +1.700 |
Neg logit contributions | |
citiz | -2.407 |
eleph | -2.314 |
lation | -2.233 |
teasp | -2.180 |
TAIN | -2.143 |
Token | try |
Token ID | 1949 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
citiz | +3.077 |
eleph | +2.876 |
lation | +2.859 |
TAIN | +2.655 |
田 | +2.571 |
Neg logit contributions | |
tested | -3.255 |
expected | -3.189 |
coal | -3.129 |
cube | -3.017 |
isot | -2.983 |
Token | to |
Token ID | 284 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
citiz | +1.049 |
eleph | +0.991 |
lation | +0.955 |
TAIN | +0.904 |
teasp | +0.889 |
Neg logit contributions | |
tested | -1.020 |
expected | -0.996 |
coal | -0.982 |
cube | -0.940 |
isot | -0.936 |
Token | B |
Token ID | 347 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
citiz | +2.006 |
eleph | +1.880 |
lation | +1.835 |
TAIN | +1.729 |
teasp | +1.667 |
Neg logit contributions | |
tested | -2.077 |
expected | -2.008 |
coal | -1.986 |
cube | -1.912 |
isot | -1.897 |
Token | CS |
Token ID | 7902 |
Feature activation | +6.33e+00 |
Pos logit contributions | |
tested | +0.376 |
expected | +0.360 |
coal | +0.359 |
cube | +0.337 |
isot | +0.335 |
Neg logit contributions | |
citiz | -0.639 |
eleph | -0.616 |
lation | -0.592 |
teasp | -0.567 |
TAIN | -0.565 |
Token | . |
Token ID | 13 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
citiz | +5.001 |
eleph | +4.682 |
lation | +4.549 |
TAIN | +4.362 |
teasp | +4.169 |
Neg logit contributions | |
tested | -4.548 |
expected | -4.400 |
coal | -4.302 |
cli | -4.172 |
isot | -4.124 |
Token |
|
Token ID | 198 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
citiz | +1.727 |
eleph | +1.628 |
lation | +1.625 |
TAIN | +1.560 |
田 | +1.498 |
Neg logit contributions | |
tested | -1.397 |
expected | -1.356 |
coal | -1.339 |
isot | -1.298 |
cube | -1.272 |
Token |
|
Token ID | 198 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
citiz | +1.777 |
lation | +1.746 |
TAIN | +1.667 |
eleph | +1.639 |
田 | +1.589 |
Neg logit contributions | |
tested | -1.553 |
coal | -1.513 |
expected | -1.511 |
isot | -1.461 |
cube | -1.413 |
Token | At |
Token ID | 2953 |
Feature activation | +2.02e+01 |
Pos logit contributions | |
citiz | +1.744 |
lation | +1.683 |
eleph | +1.612 |
TAIN | +1.611 |
田 | +1.552 |
Neg logit contributions | |
tested | -1.703 |
expected | -1.651 |
coal | -1.639 |
isot | -1.607 |
cube | -1.562 |
Token | a |
Token ID | 257 |
Feature activation | +8.25e+00 |
Pos logit contributions | |
citiz | +11.356 |
lation | +10.900 |
eleph | +10.864 |
TAIN | +10.411 |
bottom | +10.131 |
Neg logit contributions | |
tested | -14.338 |
coal | -13.718 |
expected | -13.677 |
cli | -13.199 |
crit | -13.059 |
Token | curs |
Token ID | 13882 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
citiz | +5.818 |
lation | +5.339 |
eleph | +5.318 |
TAIN | +4.998 |
teasp | +4.759 |
Neg logit contributions | |
tested | -6.337 |
expected | -6.123 |
coal | -6.111 |
cube | -5.914 |
cli | -5.721 |
Token | ory |
Token ID | 652 |
Feature activation | +5.03e+00 |
Pos logit contributions | |
citiz | +2.473 |
lation | +2.401 |
eleph | +2.344 |
TAIN | +2.265 |
utenant | +2.177 |
Neg logit contributions | |
tested | -1.601 |
expected | -1.545 |
coal | -1.525 |
isot | -1.455 |
cube | -1.452 |
Token | glance |
Token ID | 16086 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
citiz | +3.633 |
lation | +3.379 |
eleph | +3.338 |
TAIN | +3.100 |
teasp | +2.960 |
Neg logit contributions | |
tested | -3.934 |
expected | -3.847 |
coal | -3.804 |
cube | -3.679 |
cli | -3.592 |
Token | , |
Token ID | 11 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
citiz | +3.730 |
eleph | +3.522 |
lation | +3.455 |
TAIN | +3.243 |
teasp | +3.147 |
Neg logit contributions | |
tested | -3.351 |
coal | -3.292 |
expected | -3.280 |
cube | -3.125 |
isot | -3.085 |
Token | total |
Token ID | 2472 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
tested | +3.547 |
expected | +3.408 |
coal | +3.382 |
isot | +3.317 |
cube | +3.219 |
Neg logit contributions | |
citiz | -3.540 |
eleph | -3.364 |
lation | -3.219 |
TAIN | -3.069 |
teasp | -3.055 |
Token | disregard |
Token ID | 25070 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
citiz | +2.814 |
eleph | +2.651 |
lation | +2.591 |
TAIN | +2.447 |
teasp | +2.376 |
Neg logit contributions | |
tested | -2.868 |
expected | -2.787 |
coal | -2.768 |
cube | -2.654 |
isot | -2.614 |
Token | for |
Token ID | 329 |
Feature activation | +9.27e-01 |
Pos logit contributions | |
citiz | +3.596 |
eleph | +3.430 |
lation | +3.293 |
teasp | +3.106 |
TAIN | +3.096 |
Neg logit contributions | |
tested | -3.013 |
coal | -2.961 |
expected | -2.946 |
cube | -2.781 |
isot | -2.768 |
Token | reality |
Token ID | 3950 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
citiz | +0.706 |
eleph | +0.664 |
lation | +0.645 |
TAIN | +0.610 |
teasp | +0.597 |
Neg logit contributions | |
tested | -0.709 |
expected | -0.688 |
coal | -0.686 |
cube | -0.654 |
isot | -0.652 |
Token | . |
Token ID | 13 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
tested | +3.684 |
expected | +3.462 |
coal | +3.416 |
isot | +3.350 |
cube | +3.287 |
Neg logit contributions | |
citiz | -4.864 |
eleph | -4.671 |
lation | -4.497 |
teasp | -4.290 |
TAIN | -4.290 |
Token | At |
Token ID | 1629 |
Feature activation | +2.02e+01 |
Pos logit contributions | |
tested | +0.279 |
expected | +0.269 |
coal | +0.267 |
cube | +0.253 |
isot | +0.253 |
Neg logit contributions | |
citiz | -0.369 |
eleph | -0.351 |
lation | -0.340 |
TAIN | -0.323 |
teasp | -0.322 |
Token | Home |
Token ID | 5995 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
citiz | +12.296 |
TAIN | +11.212 |
lation | +10.936 |
eleph | +10.856 |
bottom | +10.613 |
Neg logit contributions | |
tested | -13.231 |
coal | -13.092 |
expected | -12.814 |
csv | -12.363 |
cube | -12.214 |
Token | With |
Token ID | 2080 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
tested | +2.507 |
coal | +2.323 |
expected | +2.320 |
isot | +2.191 |
cube | +2.152 |
Neg logit contributions | |
citiz | -5.377 |
eleph | -5.190 |
lation | -5.070 |
TAIN | -4.882 |
teasp | -4.831 |
Token | Amy |
Token ID | 14235 |
Feature activation | +9.61e-01 |
Pos logit contributions | |
citiz | +2.695 |
lation | +2.550 |
eleph | +2.478 |
TAIN | +2.406 |
田 | +2.259 |
Neg logit contributions | |
expected | -4.139 |
tested | -4.126 |
coal | -4.070 |
isot | -3.915 |
cube | -3.902 |
Token | Sed |
Token ID | 22710 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
citiz | +0.712 |
eleph | +0.668 |
lation | +0.652 |
TAIN | +0.616 |
teasp | +0.600 |
Neg logit contributions | |
tested | -0.751 |
expected | -0.732 |
coal | -0.728 |
cube | -0.696 |
isot | -0.695 |
Token | aris |
Token ID | 20066 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
citiz | +1.137 |
lation | +1.082 |
eleph | +1.081 |
TAIN | +1.038 |
田 | +1.002 |
Neg logit contributions | |
tested | -0.587 |
expected | -0.564 |
coal | -0.554 |
cube | -0.521 |
isot | -0.521 |
Token | of |
Token ID | 286 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
citiz | +1.424 |
eleph | +1.354 |
lation | +1.317 |
TAIN | +1.252 |
teasp | +1.230 |
Neg logit contributions | |
tested | -1.202 |
expected | -1.161 |
coal | -1.157 |
isot | -1.099 |
cube | -1.097 |
Token | Breitbart |
Token ID | 19923 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
citiz | +1.167 |
eleph | +1.105 |
lation | +1.073 |
TAIN | +1.023 |
teasp | +0.997 |
Neg logit contributions | |
tested | -1.065 |
expected | -1.028 |
coal | -1.023 |
cube | -0.979 |
isot | -0.976 |
Token | . |
Token ID | 13 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
citiz | +8.421 |
eleph | +7.725 |
TAIN | +7.512 |
lation | +7.501 |
teasp | +6.967 |
Neg logit contributions | |
tested | -8.841 |
expected | -8.314 |
coal | -8.275 |
cube | -8.071 |
isot | -7.850 |
Token |
|
Token ID | 198 |
Feature activation | -9.00e-02 |
Pos logit contributions | |
tested | +1.217 |
expected | +1.170 |
coal | +1.166 |
cube | +1.095 |
isot | +1.075 |
Neg logit contributions | |
citiz | -1.699 |
eleph | -1.625 |
lation | -1.535 |
teasp | -1.487 |
TAIN | -1.450 |
Token |
|
Token ID | 198 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
tested | +0.058 |
expected | +0.056 |
coal | +0.055 |
cube | +0.052 |
isot | +0.052 |
Neg logit contributions | |
citiz | -0.080 |
eleph | -0.076 |
lation | -0.074 |
TAIN | -0.071 |
teasp | -0.070 |
Token | At |
Token ID | 2953 |
Feature activation | +2.01e+01 |
Pos logit contributions | |
citiz | +0.249 |
eleph | +0.236 |
lation | +0.228 |
TAIN | +0.217 |
teasp | +0.215 |
Neg logit contributions | |
tested | -0.215 |
expected | -0.208 |
coal | -0.207 |
cube | -0.197 |
isot | -0.197 |
Token | a |
Token ID | 257 |
Feature activation | +6.73e+00 |
Pos logit contributions | |
citiz | +11.688 |
lation | +11.319 |
TAIN | +10.986 |
eleph | +10.984 |
bottom | +10.562 |
Neg logit contributions | |
tested | -13.812 |
coal | -13.209 |
cube | -12.801 |
expected | -12.738 |
Reviewer | -12.640 |
Token | glance |
Token ID | 16086 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
citiz | +5.282 |
lation | +4.889 |
eleph | +4.879 |
TAIN | +4.602 |
田 | +4.381 |
Neg logit contributions | |
tested | -4.710 |
expected | -4.469 |
coal | -4.459 |
cube | -4.361 |
isot | -4.225 |
Token | , |
Token ID | 11 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
citiz | +3.234 |
eleph | +3.060 |
lation | +3.001 |
TAIN | +2.837 |
teasp | +2.743 |
Neg logit contributions | |
tested | -2.861 |
coal | -2.785 |
expected | -2.767 |
cube | -2.654 |
isot | -2.620 |
Token | H |
Token ID | 367 |
Feature activation | +9.55e-01 |
Pos logit contributions | |
tested | +0.691 |
expected | +0.670 |
coal | +0.664 |
isot | +0.634 |
cube | +0.626 |
Neg logit contributions | |
citiz | -0.799 |
eleph | -0.760 |
lation | -0.730 |
teasp | -0.703 |
TAIN | -0.693 |
Token | ahn |
Token ID | 15386 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
citiz | +0.655 |
lation | +0.608 |
eleph | +0.603 |
TAIN | +0.568 |
田 | +0.546 |
Neg logit contributions | |
tested | -0.788 |
expected | -0.766 |
coal | -0.761 |
isot | -0.733 |
cube | -0.731 |
Token | more |
Token ID | 517 |
Feature activation | +1.16e+01 |
Pos logit contributions | |
citiz | +1.815 |
eleph | +1.722 |
lation | +1.670 |
TAIN | +1.578 |
teasp | +1.539 |
Neg logit contributions | |
tested | -1.758 |
expected | -1.700 |
coal | -1.694 |
isot | -1.611 |
cube | -1.608 |
Token | than |
Token ID | 621 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
citiz | +7.870 |
eleph | +7.365 |
lation | +7.361 |
TAIN | +6.811 |
emphasis | +6.650 |
Neg logit contributions | |
tested | -8.792 |
expected | -8.588 |
coal | -8.323 |
cube | -7.923 |
csv | -7.864 |
Token | an |
Token ID | 281 |
Feature activation | -3.76e+00 |
Pos logit contributions | |
citiz | +2.819 |
eleph | +2.639 |
lation | +2.567 |
TAIN | +2.363 |
utenant | +2.267 |
Neg logit contributions | |
tested | -3.409 |
coal | -3.302 |
expected | -3.272 |
cube | -3.135 |
isot | -3.095 |
Token | hour |
Token ID | 1711 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
expected | +2.341 |
tested | +2.339 |
coal | +2.278 |
isot | +2.245 |
cube | +2.198 |
Neg logit contributions | |
citiz | -3.329 |
eleph | -3.129 |
lation | -3.054 |
TAIN | -2.945 |
teasp | -2.924 |
Token | . |
Token ID | 13 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
citiz | +1.139 |
eleph | +1.084 |
lation | +1.052 |
TAIN | +1.001 |
teasp | +0.986 |
Neg logit contributions | |
tested | -0.930 |
expected | -0.897 |
coal | -0.893 |
cube | -0.847 |
isot | -0.847 |
Token | At |
Token ID | 1629 |
Feature activation | +2.01e+01 |
Pos logit contributions | |
citiz | +0.472 |
eleph | +0.449 |
lation | +0.436 |
TAIN | +0.416 |
teasp | +0.410 |
Neg logit contributions | |
tested | -0.366 |
expected | -0.353 |
coal | -0.351 |
isot | -0.333 |
cube | -0.333 |
Token | least |
Token ID | 1551 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
citiz | +10.932 |
lation | +9.885 |
eleph | +9.841 |
TAIN | +9.289 |
bottom | +9.130 |
Neg logit contributions | |
tested | -15.138 |
coal | -14.871 |
expected | -14.579 |
cli | -13.822 |
csv | -13.814 |
Token | 250 |
Token ID | 8646 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
citiz | +6.178 |
lation | +5.481 |
eleph | +5.461 |
TAIN | +5.108 |
emphasis | +4.902 |
Neg logit contributions | |
tested | -9.645 |
coal | -9.630 |
expected | -9.380 |
csv | -9.043 |
cube | -9.017 |
Token | people |
Token ID | 661 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
citiz | +2.673 |
eleph | +2.482 |
lation | +2.417 |
TAIN | +2.292 |
teasp | +2.196 |
Neg logit contributions | |
tested | -3.049 |
expected | -2.962 |
coal | -2.945 |
isot | -2.832 |
cube | -2.810 |
Token | cycle |
Token ID | 6772 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
citiz | +0.951 |
eleph | +0.901 |
lation | +0.878 |
TAIN | +0.833 |
teasp | +0.819 |
Neg logit contributions | |
tested | -0.782 |
expected | -0.757 |
coal | -0.754 |
isot | -0.715 |
cube | -0.714 |
Token | in |
Token ID | 287 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
citiz | +2.261 |
eleph | +2.127 |
lation | +2.085 |
TAIN | +1.971 |
teasp | +1.895 |
Neg logit contributions | |
tested | -2.245 |
expected | -2.184 |
coal | -2.183 |
cube | -2.086 |
isot | -2.056 |
Token | world |
Token ID | 995 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
citiz | +4.115 |
lation | +3.836 |
eleph | +3.796 |
TAIN | +3.543 |
田 | +3.372 |
Neg logit contributions | |
tested | -4.309 |
expected | -4.213 |
coal | -4.133 |
cli | -3.956 |
cube | -3.945 |
Token | than |
Token ID | 621 |
Feature activation | +3.93e+00 |
Pos logit contributions | |
tested | +2.249 |
expected | +2.161 |
coal | +2.149 |
cube | +2.063 |
isot | +2.061 |
Neg logit contributions | |
citiz | -2.448 |
eleph | -2.334 |
lation | -2.208 |
teasp | -2.145 |
TAIN | -2.120 |
Token | the |
Token ID | 262 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
citiz | +2.931 |
eleph | +2.746 |
lation | +2.706 |
TAIN | +2.555 |
田 | +2.425 |
Neg logit contributions | |
tested | -2.981 |
coal | -2.908 |
expected | -2.896 |
cube | -2.772 |
isot | -2.767 |
Token | DC |
Token ID | 6257 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
citiz | +2.393 |
eleph | +2.240 |
lation | +2.209 |
TAIN | +2.083 |
田 | +1.992 |
Neg logit contributions | |
tested | -2.313 |
expected | -2.243 |
coal | -2.239 |
isot | -2.134 |
cube | -2.132 |
Token | universe |
Token ID | 6881 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
citiz | +6.185 |
eleph | +5.853 |
lation | +5.797 |
TAIN | +5.529 |
田 | +5.175 |
Neg logit contributions | |
tested | -6.801 |
coal | -6.577 |
expected | -6.537 |
isot | -6.219 |
cli | -6.121 |
Token | at |
Token ID | 379 |
Feature activation | +2.01e+01 |
Pos logit contributions | |
tested | +0.317 |
expected | +0.306 |
coal | +0.304 |
cube | +0.289 |
isot | +0.289 |
Neg logit contributions | |
citiz | -0.399 |
eleph | -0.379 |
lation | -0.367 |
TAIN | -0.350 |
teasp | -0.348 |
Token | the |
Token ID | 262 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
citiz | +10.268 |
lation | +9.854 |
eleph | +9.452 |
TAIN | +9.238 |
bottom | +8.935 |
Neg logit contributions | |
tested | -15.466 |
coal | -15.126 |
expected | -15.117 |
cli | -14.651 |
csv | -14.463 |
Token | time |
Token ID | 640 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
citiz | +5.865 |
lation | +5.351 |
eleph | +5.259 |
TAIN | +4.875 |
bottom | +4.635 |
Neg logit contributions | |
tested | -11.989 |
expected | -11.856 |
coal | -11.649 |
cli | -11.272 |
cube | -11.135 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
tested | +0.410 |
expected | +0.394 |
coal | +0.392 |
cube | +0.372 |
isot | +0.371 |
Neg logit contributions | |
citiz | -0.538 |
eleph | -0.510 |
lation | -0.494 |
TAIN | -0.471 |
teasp | -0.471 |
Token | It |
Token ID | 632 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
tested | +0.206 |
expected | +0.198 |
coal | +0.197 |
cube | +0.187 |
isot | +0.186 |
Neg logit contributions | |
citiz | -0.278 |
eleph | -0.265 |
lation | -0.256 |
TAIN | -0.243 |
teasp | -0.243 |
Token | was |
Token ID | 373 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
citiz | +3.062 |
lation | +2.898 |
eleph | +2.890 |
TAIN | +2.706 |
utenant | +2.598 |
Neg logit contributions | |
tested | -2.689 |
coal | -2.605 |
expected | -2.600 |
cube | -2.461 |
cli | -2.445 |
Token | . |
Token ID | 13 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
citiz | +2.393 |
eleph | +2.274 |
lation | +2.212 |
TAIN | +2.106 |
teasp | +2.062 |
Neg logit contributions | |
tested | -1.984 |
expected | -1.923 |
coal | -1.912 |
cube | -1.824 |
isot | -1.812 |
Token | � |
Token ID | 447 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
tested | +0.639 |
expected | +0.615 |
coal | +0.612 |
cube | +0.580 |
isot | +0.575 |
Neg logit contributions | |
citiz | -0.855 |
eleph | -0.820 |
lation | -0.785 |
teasp | -0.751 |
TAIN | -0.745 |
Token | � |
Token ID | 251 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
citiz | +1.844 |
eleph | +1.725 |
lation | +1.676 |
TAIN | +1.579 |
teasp | +1.556 |
Neg logit contributions | |
tested | -2.013 |
expected | -1.955 |
coal | -1.935 |
cube | -1.869 |
isot | -1.865 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
citiz | +1.532 |
eleph | +1.451 |
lation | +1.420 |
TAIN | +1.353 |
teasp | +1.325 |
Neg logit contributions | |
tested | -1.208 |
expected | -1.168 |
coal | -1.158 |
isot | -1.104 |
cube | -1.099 |
Token |
|
Token ID | 198 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
citiz | +1.049 |
lation | +1.002 |
eleph | +0.984 |
TAIN | +0.958 |
田 | +0.921 |
Neg logit contributions | |
tested | -0.844 |
expected | -0.818 |
coal | -0.815 |
isot | -0.781 |
cube | -0.766 |
Token | At |
Token ID | 2953 |
Feature activation | +2.01e+01 |
Pos logit contributions | |
citiz | +2.505 |
lation | +2.367 |
eleph | +2.312 |
TAIN | +2.253 |
田 | +2.165 |
Neg logit contributions | |
tested | -2.388 |
expected | -2.314 |
coal | -2.298 |
isot | -2.234 |
cube | -2.198 |
Token | the |
Token ID | 262 |
Feature activation | +1.52e+01 |
Pos logit contributions | |
citiz | +13.694 |
lation | +12.897 |
eleph | +12.581 |
TAIN | +12.495 |
bottom | +11.596 |
Neg logit contributions | |
tested | -12.387 |
expected | -11.714 |
coal | -11.636 |
csv | -11.242 |
Reviewer | -11.141 |
Token | city |
Token ID | 1748 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
citiz | +10.223 |
lation | +9.210 |
bottom | +8.967 |
eleph | +8.744 |
TAIN | +8.707 |
Neg logit contributions | |
tested | -10.715 |
expected | -10.354 |
coal | -10.185 |
Reviewer | -9.859 |
cube | -9.828 |
Token | hall |
Token ID | 6899 |
Feature activation | +7.59e+00 |
Pos logit contributions | |
citiz | +4.057 |
eleph | +3.756 |
lation | +3.751 |
TAIN | +3.491 |
teasp | +3.346 |
Neg logit contributions | |
tested | -4.266 |
expected | -4.148 |
coal | -4.063 |
isot | -3.883 |
cube | -3.856 |
Token | , |
Token ID | 11 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
citiz | +6.004 |
eleph | +5.576 |
lation | +5.553 |
TAIN | +5.204 |
田 | +5.029 |
Neg logit contributions | |
tested | -5.310 |
expected | -5.156 |
coal | -5.150 |
cube | -4.913 |
isot | -4.798 |
Token | player |
Token ID | 2137 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
citiz | +3.241 |
eleph | +3.028 |
lation | +2.962 |
TAIN | +2.813 |
teasp | +2.711 |
Neg logit contributions | |
tested | -2.915 |
expected | -2.826 |
coal | -2.809 |
cube | -2.692 |
isot | -2.667 |
Token | oc |
Token ID | 420 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
tested | +1.141 |
expected | +1.097 |
coal | +1.092 |
isot | +1.033 |
cube | +1.032 |
Neg logit contributions | |
citiz | -1.566 |
eleph | -1.494 |
lation | -1.446 |
TAIN | -1.381 |
teasp | -1.368 |
Token | op |
Token ID | 404 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
tested | +0.663 |
coal | +0.626 |
expected | +0.625 |
cube | +0.568 |
isot | +0.561 |
Neg logit contributions | |
citiz | -1.790 |
eleph | -1.717 |
lation | -1.665 |
TAIN | -1.615 |
teasp | -1.607 |
Token | . |
Token ID | 13 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
citiz | +1.871 |
eleph | +1.774 |
lation | +1.729 |
TAIN | +1.645 |
teasp | +1.619 |
Neg logit contributions | |
tested | -1.525 |
expected | -1.476 |
coal | -1.460 |
isot | -1.388 |
cube | -1.387 |
Token | Continue |
Token ID | 10054 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
tested | +1.918 |
expected | +1.834 |
coal | +1.830 |
cube | +1.735 |
cli | +1.696 |
Neg logit contributions | |
citiz | -2.529 |
eleph | -2.432 |
lation | -2.296 |
teasp | -2.231 |
TAIN | -2.174 |
Token | Reading |
Token ID | 11725 |
Feature activation | +1.37e+01 |
Pos logit contributions | |
lation | +5.222 |
citiz | +4.897 |
TAIN | +4.744 |
田 | +4.203 |
emphasis | +4.187 |
Neg logit contributions | |
expected | -10.822 |
coal | -10.795 |
tested | -10.719 |
cube | -10.294 |
isot | -10.212 |
Token | Below |
Token ID | 10383 |
Feature activation | +1.99e+01 |
Pos logit contributions | |
lation | +3.314 |
citiz | +2.837 |
TAIN | +2.771 |
bottom | +2.460 |
emphasis | +2.362 |
Neg logit contributions | |
tested | -14.843 |
expected | -14.643 |
coal | -14.616 |
isot | -14.356 |
dissatisf | -14.261 |
Token | Advertisement |
Token ID | 39711 |
Feature activation | +7.23e+00 |
Pos logit contributions | |
lation | +5.064 |
TAIN | +4.515 |
citiz | +4.237 |
emphasis | +3.710 |
Racer | +3.479 |
Neg logit contributions | |
expected | -19.481 |
tested | -19.461 |
coal | -19.237 |
isot | -19.066 |
crit | -18.680 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
citiz | +5.416 |
eleph | +5.110 |
lation | +5.058 |
TAIN | +4.868 |
田 | +4.551 |
Neg logit contributions | |
tested | -5.203 |
expected | -5.016 |
coal | -5.005 |
isot | -4.854 |
cube | -4.783 |
Token | UN |
Token ID | 4944 |
Feature activation | -6.32e+00 |
Pos logit contributions | |
tested | +1.765 |
expected | +1.709 |
coal | +1.695 |
cube | +1.612 |
isot | +1.588 |
Neg logit contributions | |
citiz | -2.155 |
eleph | -2.077 |
lation | -1.939 |
teasp | -1.879 |
TAIN | -1.834 |
Token | ITED |
Token ID | 22061 |
Feature activation | -7.52e+00 |
Pos logit contributions | |
tested | +4.758 |
expected | +4.685 |
coal | +4.593 |
isot | +4.424 |
cube | +4.413 |
Neg logit contributions | |
citiz | -4.710 |
eleph | -4.496 |
lation | -4.297 |
teasp | -4.125 |
utenant | -3.950 |
Token | N |
Token ID | 399 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
tested | +4.312 |
coal | +4.092 |
expected | +4.040 |
isot | +3.879 |
cube | +3.866 |
Neg logit contributions | |
citiz | -7.038 |
eleph | -6.682 |
lation | -6.428 |
teasp | -6.375 |
TAIN | -5.955 |
Token | � |
Token ID | 447 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
citiz | +1.066 |
eleph | +1.008 |
lation | +0.984 |
TAIN | +0.936 |
teasp | +0.918 |
Neg logit contributions | |
tested | -0.832 |
expected | -0.804 |
coal | -0.799 |
cube | -0.756 |
isot | -0.755 |
Token | � |
Token ID | 247 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
tested | +1.600 |
expected | +1.555 |
coal | +1.536 |
isot | +1.456 |
cube | +1.450 |
Neg logit contributions | |
citiz | -1.630 |
eleph | -1.562 |
lation | -1.481 |
TAIN | -1.410 |
teasp | -1.399 |
Token | s |
Token ID | 82 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
citiz | +0.276 |
eleph | +0.262 |
lation | +0.255 |
TAIN | +0.242 |
teasp | +0.239 |
Neg logit contributions | |
tested | -0.219 |
expected | -0.212 |
coal | -0.210 |
isot | -0.200 |
cube | -0.200 |
Token | what |
Token ID | 644 |
Feature activation | -6.98e+00 |
Pos logit contributions | |
tested | +3.550 |
expected | +3.481 |
coal | +3.398 |
isot | +3.245 |
cli | +3.178 |
Neg logit contributions | |
citiz | -4.226 |
eleph | -4.076 |
lation | -3.875 |
teasp | -3.809 |
TAIN | -3.752 |
Token | the |
Token ID | 262 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
tested | +4.697 |
expected | +4.521 |
coal | +4.388 |
isot | +4.275 |
shocks | +4.192 |
Neg logit contributions | |
citiz | -5.777 |
eleph | -5.640 |
teasp | -5.340 |
lation | -5.339 |
TAIN | -5.074 |
Token | Consumer |
Token ID | 18110 |
Feature activation | -2.07e+01 |
Pos logit contributions | |
tested | +2.668 |
expected | +2.593 |
coal | +2.576 |
isot | +2.441 |
cube | +2.425 |
Neg logit contributions | |
citiz | -3.383 |
eleph | -3.276 |
lation | -3.118 |
teasp | -3.021 |
TAIN | -2.988 |
Token | Financial |
Token ID | 11302 |
Feature activation | -8.54e+00 |
Pos logit contributions | |
coal | +9.166 |
tested | +9.020 |
Choice | +8.845 |
Prediction | +8.552 |
crit | +8.539 |
Neg logit contributions | |
citiz | -18.369 |
eleph | -18.236 |
teasp | -17.016 |
lation | -16.686 |
田 | -16.583 |
Token | Protection |
Token ID | 9985 |
Feature activation | -1.09e+01 |
Pos logit contributions | |
tested | +2.953 |
coal | +2.819 |
expected | +2.696 |
Choice | +2.601 |
Prediction | +2.550 |
Neg logit contributions | |
eleph | -10.070 |
citiz | -10.018 |
teasp | -9.646 |
田 | -9.203 |
pione | -9.147 |
Token | Bureau |
Token ID | 9840 |
Feature activation | -7.47e+00 |
Pos logit contributions | |
tested | +4.498 |
expected | +4.435 |
coal | +4.414 |
crit | +4.100 |
Prediction | +4.006 |
Neg logit contributions | |
citiz | -11.889 |
eleph | -11.759 |
teasp | -11.315 |
pione | -10.784 |
lation | -10.550 |
Token | — |
Token ID | 960 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
tested | +4.313 |
expected | +4.125 |
coal | +3.990 |
isot | +3.884 |
cube | +3.762 |
Neg logit contributions | |
citiz | -6.975 |
eleph | -6.801 |
lation | -6.440 |
teasp | -6.369 |
TAIN | -6.148 |
Token | which |
Token ID | 4758 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
tested | +0.116 |
expected | +0.111 |
coal | +0.111 |
cube | +0.104 |
isot | +0.104 |
Neg logit contributions | |
citiz | -0.161 |
eleph | -0.154 |
lation | -0.149 |
TAIN | -0.142 |
teasp | -0.141 |
Token | , |
Token ID | 11 |
Feature activation | -9.90e+00 |
Pos logit contributions | |
tested | +2.785 |
expected | +2.602 |
coal | +2.567 |
cube | +2.516 |
cli | +2.400 |
Neg logit contributions | |
citiz | -4.656 |
eleph | -4.590 |
teasp | -4.210 |
pione | -4.050 |
lation | -4.040 |
Token | we |
Token ID | 356 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
tested | +4.704 |
expected | +4.169 |
coal | +4.095 |
cube | +4.031 |
isot | +3.970 |
Neg logit contributions | |
eleph | -10.631 |
citiz | -10.448 |
teasp | -10.009 |
� | -9.768 |
pione | -9.598 |
Token | spent |
Token ID | 3377 |
Feature activation | -8.50e+00 |
Pos logit contributions | |
tested | +0.146 |
expected | +0.141 |
coal | +0.140 |
cube | +0.131 |
isot | +0.131 |
Neg logit contributions | |
citiz | -0.250 |
eleph | -0.239 |
lation | -0.234 |
TAIN | -0.223 |
teasp | -0.221 |
Token | a |
Token ID | 257 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
tested | +4.666 |
expected | +4.458 |
coal | +4.226 |
cube | +4.226 |
isot | +4.198 |
Neg logit contributions | |
eleph | -8.575 |
citiz | -8.564 |
teasp | -7.835 |
� | -7.827 |
pione | -7.727 |
Token | week |
Token ID | 1285 |
Feature activation | -7.28e+00 |
Pos logit contributions | |
tested | +0.719 |
expected | +0.697 |
coal | +0.687 |
cube | +0.659 |
isot | +0.658 |
Neg logit contributions | |
citiz | -0.849 |
eleph | -0.818 |
lation | -0.769 |
TAIN | -0.737 |
teasp | -0.735 |
Token | using |
Token ID | 1262 |
Feature activation | -2.01e+01 |
Pos logit contributions | |
tested | +3.765 |
expected | +3.221 |
coal | +3.149 |
cube | +3.096 |
cli | +2.977 |
Neg logit contributions | |
eleph | -7.797 |
citiz | -7.766 |
teasp | -7.182 |
pione | -7.113 |
� | -7.062 |
Token | the |
Token ID | 262 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
tested | +8.060 |
cube | +6.511 |
expected | +6.451 |
coal | +6.064 |
cli | +5.942 |
Neg logit contributions | |
eleph | -23.005 |
citiz | -22.230 |
teasp | -21.715 |
pione | -21.408 |
� | -21.407 |
Token | tactics |
Token ID | 10815 |
Feature activation | -5.78e+00 |
Pos logit contributions | |
tested | +4.483 |
expected | +4.126 |
coal | +4.094 |
cli | +3.945 |
isot | +3.934 |
Neg logit contributions | |
citiz | -3.762 |
eleph | -3.695 |
teasp | -3.321 |
lation | -3.286 |
pione | -3.205 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
tested | +3.191 |
expected | +2.880 |
coal | +2.783 |
isot | +2.642 |
cli | +2.612 |
Neg logit contributions | |
citiz | -5.954 |
eleph | -5.810 |
teasp | -5.517 |
lation | -5.443 |
� | -5.247 |
Token | in |
Token ID | 287 |
Feature activation | -5.77e+00 |
Pos logit contributions | |
tested | +5.726 |
expected | +5.037 |
cube | +4.951 |
coal | +4.795 |
isot | +4.569 |
Neg logit contributions | |
citiz | -14.584 |
eleph | -14.445 |
teasp | -13.691 |
� | -13.680 |
pione | -13.591 |
Token | this |
Token ID | 428 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
tested | +4.282 |
coal | +4.053 |
expected | +4.048 |
cube | +3.961 |
isot | +3.892 |
Neg logit contributions | |
citiz | -4.787 |
eleph | -4.628 |
teasp | -4.256 |
� | -4.010 |
pione | -4.010 |
Token | � |
Token ID | 247 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
tested | +0.723 |
expected | +0.703 |
coal | +0.700 |
isot | +0.661 |
cli | +0.657 |
Neg logit contributions | |
citiz | -0.749 |
eleph | -0.719 |
lation | -0.674 |
teasp | -0.650 |
TAIN | -0.641 |
Token | s |
Token ID | 82 |
Feature activation | -9.38e+00 |
Pos logit contributions | |
citiz | +1.338 |
eleph | +1.285 |
lation | +1.265 |
TAIN | +1.212 |
teasp | +1.195 |
Neg logit contributions | |
tested | -0.611 |
expected | -0.582 |
coal | -0.580 |
cube | -0.541 |
isot | -0.536 |
Token | an |
Token ID | 281 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
expected | +5.870 |
tested | +5.830 |
coal | +5.535 |
isot | +5.449 |
Prediction | +5.247 |
Neg logit contributions | |
citiz | -8.104 |
eleph | -7.947 |
teasp | -7.516 |
lation | -7.341 |
TAIN | -7.167 |
Token | economic |
Token ID | 3034 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
expected | +7.871 |
isot | +7.692 |
tested | +7.535 |
coal | +7.168 |
shocks | +7.091 |
Neg logit contributions | |
citiz | -11.190 |
eleph | -10.398 |
teasp | -10.314 |
田 | -9.983 |
TAIN | -9.766 |
Token | and |
Token ID | 290 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
tested | +7.847 |
isot | +7.616 |
expected | +7.539 |
coal | +7.504 |
shocks | +7.420 |
Neg logit contributions | |
citiz | -10.458 |
eleph | -10.267 |
teasp | -9.664 |
田 | -9.370 |
TAIN | -9.249 |
Token | political |
Token ID | 1964 |
Feature activation | -1.99e+01 |
Pos logit contributions | |
tested | +2.403 |
expected | +2.353 |
coal | +2.339 |
isot | +2.270 |
cube | +2.236 |
Neg logit contributions | |
citiz | -1.660 |
eleph | -1.589 |
lation | -1.485 |
teasp | -1.419 |
TAIN | -1.403 |
Token | environment |
Token ID | 2858 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
isot | +11.252 |
tested | +11.034 |
shocks | +11.028 |
coal | +10.869 |
expected | +10.669 |
Neg logit contributions | |
citiz | -14.852 |
eleph | -14.657 |
teasp | -14.066 |
TAIN | -13.560 |
田 | -13.515 |
Token | in |
Token ID | 287 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
tested | +9.275 |
expected | +8.986 |
isot | +8.570 |
coal | +8.429 |
shocks | +8.295 |
Neg logit contributions | |
citiz | -10.831 |
eleph | -10.543 |
teasp | -9.836 |
lation | -9.557 |
田 | -9.463 |
Token | which |
Token ID | 543 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
tested | +0.899 |
expected | +0.867 |
coal | +0.858 |
isot | +0.816 |
cube | +0.813 |
Neg logit contributions | |
citiz | -1.229 |
eleph | -1.170 |
lation | -1.134 |
TAIN | -1.082 |
teasp | -1.072 |
Token | you |
Token ID | 345 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
tested | +3.155 |
expected | +3.064 |
coal | +3.032 |
isot | +2.918 |
cube | +2.834 |
Neg logit contributions | |
citiz | -3.626 |
eleph | -3.534 |
lation | -3.302 |
teasp | -3.272 |
TAIN | -3.199 |
Token | � |
Token ID | 447 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
citiz | +1.339 |
eleph | +1.265 |
lation | +1.239 |
TAIN | +1.168 |
teasp | +1.146 |
Neg logit contributions | |
tested | -1.249 |
expected | -1.205 |
coal | -1.203 |
cube | -1.149 |
isot | -1.144 |
Token | ets |
Token ID | 1039 |
Feature activation | -6.65e+00 |
Pos logit contributions | |
tested | +3.795 |
expected | +3.650 |
coal | +3.621 |
cube | +3.475 |
isot | +3.445 |
Neg logit contributions | |
citiz | -4.100 |
eleph | -3.908 |
lation | -3.712 |
TAIN | -3.537 |
teasp | -3.529 |
Token | Per |
Token ID | 2448 |
Feature activation | -6.38e+00 |
Pos logit contributions | |
tested | +4.397 |
expected | +4.214 |
coal | +4.109 |
isot | +3.965 |
cube | +3.958 |
Neg logit contributions | |
citiz | -5.755 |
eleph | -5.518 |
lation | -5.207 |
teasp | -5.131 |
TAIN | -4.975 |
Token | Second |
Token ID | 5498 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
tested | +3.956 |
expected | +3.874 |
coal | +3.855 |
cli | +3.665 |
cube | +3.640 |
Neg logit contributions | |
citiz | -5.646 |
eleph | -5.457 |
lation | -5.150 |
teasp | -4.970 |
田 | -4.913 |
Token | on |
Token ID | 319 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
tested | +6.789 |
expected | +6.466 |
coal | +6.267 |
Prediction | +6.137 |
isot | +6.060 |
Neg logit contributions | |
citiz | -9.489 |
eleph | -9.096 |
teasp | -8.673 |
lation | -8.587 |
田 | -8.176 |
Token | a |
Token ID | 257 |
Feature activation | -1.09e+01 |
Pos logit contributions | |
tested | +7.844 |
expected | +7.469 |
coal | +7.230 |
isot | +7.160 |
Prediction | +7.148 |
Neg logit contributions | |
citiz | -10.459 |
eleph | -10.039 |
teasp | -9.662 |
lation | -9.469 |
田 | -9.006 |
Token | Public |
Token ID | 5094 |
Feature activation | -1.97e+01 |
Pos logit contributions | |
tested | +7.013 |
expected | +6.644 |
coal | +6.549 |
cli | +6.438 |
Prediction | +6.410 |
Neg logit contributions | |
citiz | -9.066 |
eleph | -8.815 |
lation | -8.251 |
teasp | -8.159 |
TAIN | -7.808 |
Token | Cloud |
Token ID | 10130 |
Feature activation | -7.32e+00 |
Pos logit contributions | |
tested | +10.156 |
cube | +10.145 |
coal | +10.051 |
Choice | +9.757 |
Prediction | +9.696 |
Neg logit contributions | |
citiz | -15.980 |
eleph | -15.630 |
teasp | -14.602 |
lation | -14.332 |
田 | -13.580 |
Token | Inst |
Token ID | 2262 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
tested | +4.564 |
expected | +4.347 |
coal | +4.292 |
cube | +4.179 |
isot | +4.081 |
Neg logit contributions | |
citiz | -6.547 |
eleph | -6.271 |
lation | -5.974 |
teasp | -5.870 |
TAIN | -5.660 |
Token | ance |
Token ID | 590 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
tested | +0.178 |
expected | +0.173 |
coal | +0.171 |
isot | +0.163 |
cube | +0.163 |
Neg logit contributions | |
citiz | -0.217 |
eleph | -0.207 |
lation | -0.202 |
TAIN | -0.191 |
teasp | -0.188 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
tested | +4.064 |
expected | +3.822 |
coal | +3.770 |
cube | +3.651 |
isot | +3.589 |
Neg logit contributions | |
citiz | -5.846 |
eleph | -5.638 |
lation | -5.334 |
teasp | -5.269 |
TAIN | -5.119 |
Token | Imp |
Token ID | 9855 |
Feature activation | -4.25e+00 |
Pos logit contributions | |
tested | +1.747 |
expected | +1.679 |
coal | +1.661 |
cube | +1.588 |
isot | +1.562 |
Neg logit contributions | |
citiz | -2.314 |
eleph | -2.223 |
lation | -2.115 |
teasp | -2.044 |
TAIN | -2.001 |
Token | agency |
Token ID | 4086 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
tested | +5.127 |
expected | +4.891 |
coal | +4.863 |
isot | +4.676 |
cube | +4.599 |
Neg logit contributions | |
citiz | -6.629 |
eleph | -6.391 |
lation | -6.012 |
teasp | -5.946 |
TAIN | -5.841 |
Token | Wh |
Token ID | 854 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
tested | +0.397 |
expected | +0.384 |
coal | +0.381 |
isot | +0.362 |
cube | +0.362 |
Neg logit contributions | |
citiz | -0.475 |
eleph | -0.451 |
lation | -0.438 |
TAIN | -0.416 |
teasp | -0.411 |
Token | itt |
Token ID | 715 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
tested | +6.919 |
coal | +6.818 |
expected | +6.791 |
cli | +6.350 |
cube | +6.239 |
Neg logit contributions | |
eleph | -9.825 |
citiz | -9.718 |
teasp | -8.508 |
lation | -8.463 |
� | -8.405 |
Token | les |
Token ID | 829 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
tested | +1.025 |
expected | +0.991 |
coal | +0.989 |
cube | +0.940 |
isot | +0.934 |
Neg logit contributions | |
citiz | -1.130 |
eleph | -1.081 |
lation | -1.024 |
teasp | -0.976 |
TAIN | -0.975 |
Token | ea |
Token ID | 18213 |
Feature activation | -8.03e+00 |
Pos logit contributions | |
citiz | +0.186 |
eleph | +0.178 |
lation | +0.177 |
TAIN | +0.170 |
teasp | +0.166 |
Neg logit contributions | |
tested | -0.073 |
expected | -0.068 |
coal | -0.067 |
isot | -0.063 |
cube | -0.062 |
Token | Community |
Token ID | 8108 |
Feature activation | -1.94e+01 |
Pos logit contributions | |
tested | +4.995 |
expected | +4.830 |
coal | +4.724 |
cube | +4.471 |
isot | +4.470 |
Neg logit contributions | |
citiz | -6.996 |
eleph | -6.829 |
lation | -6.483 |
teasp | -6.396 |
TAIN | -6.232 |
Token | Connect |
Token ID | 8113 |
Feature activation | -6.29e+00 |
Pos logit contributions | |
tested | +10.485 |
coal | +10.060 |
expected | +9.875 |
Choice | +9.798 |
Prediction | +9.791 |
Neg logit contributions | |
eleph | -15.563 |
citiz | -15.301 |
teasp | -14.552 |
lation | -14.065 |
TAIN | -13.682 |
Token | ions |
Token ID | 507 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
tested | +4.635 |
expected | +4.462 |
coal | +4.370 |
isot | +4.154 |
cube | +4.149 |
Neg logit contributions | |
citiz | -4.937 |
eleph | -4.692 |
lation | -4.445 |
teasp | -4.268 |
TAIN | -4.257 |
Token | provides |
Token ID | 3769 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
tested | +7.061 |
expected | +6.786 |
coal | +6.529 |
cube | +6.304 |
Prediction | +6.242 |
Neg logit contributions | |
citiz | -9.885 |
eleph | -9.806 |
teasp | -9.362 |
lation | -9.096 |
TAIN | -8.776 |
Token | emergency |
Token ID | 6334 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
tested | +2.661 |
expected | +2.573 |
coal | +2.554 |
isot | +2.486 |
cube | +2.434 |
Neg logit contributions | |
citiz | -2.671 |
eleph | -2.582 |
lation | -2.441 |
teasp | -2.384 |
TAIN | -2.346 |
Token | relief |
Token ID | 8259 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
tested | +1.580 |
expected | +1.526 |
coal | +1.526 |
isot | +1.466 |
cube | +1.445 |
Neg logit contributions | |
citiz | -1.808 |
eleph | -1.722 |
lation | -1.654 |
TAIN | -1.585 |
teasp | -1.580 |
Token | allies |
Token ID | 7681 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
tested | +1.659 |
expected | +1.608 |
coal | +1.603 |
isot | +1.534 |
cube | +1.518 |
Neg logit contributions | |
citiz | -1.726 |
eleph | -1.642 |
lation | -1.578 |
TAIN | -1.500 |
teasp | -1.492 |
Token | will |
Token ID | 481 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
tested | +3.399 |
expected | +3.326 |
coal | +3.265 |
isot | +3.107 |
cube | +3.068 |
Neg logit contributions | |
citiz | -4.234 |
eleph | -4.099 |
lation | -3.835 |
teasp | -3.817 |
TAIN | -3.687 |
Token | argue |
Token ID | 7267 |
Feature activation | +4.50e+00 |
Pos logit contributions | |
tested | +3.055 |
expected | +2.985 |
coal | +2.925 |
isot | +2.838 |
cube | +2.777 |
Neg logit contributions | |
citiz | -3.797 |
eleph | -3.652 |
lation | -3.417 |
teasp | -3.398 |
TAIN | -3.284 |
Token | that |
Token ID | 326 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
citiz | +2.969 |
eleph | +2.744 |
lation | +2.700 |
TAIN | +2.523 |
teasp | +2.395 |
Neg logit contributions | |
tested | -3.799 |
expected | -3.720 |
coal | -3.694 |
cube | -3.601 |
isot | -3.535 |
Token | the |
Token ID | 262 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
citiz | +0.392 |
eleph | +0.370 |
lation | +0.361 |
TAIN | +0.342 |
teasp | +0.334 |
Neg logit contributions | |
tested | -0.355 |
expected | -0.344 |
coal | -0.342 |
cube | -0.327 |
isot | -0.325 |
Token | Public |
Token ID | 5094 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
tested | +0.562 |
expected | +0.543 |
coal | +0.540 |
cube | +0.514 |
isot | +0.513 |
Neg logit contributions | |
citiz | -0.670 |
eleph | -0.635 |
lation | -0.617 |
TAIN | -0.586 |
teasp | -0.577 |
Token | Policy |
Token ID | 7820 |
Feature activation | -8.87e+00 |
Pos logit contributions | |
tested | +6.434 |
Prediction | +6.154 |
expected | +6.089 |
Choice | +6.067 |
Surve | +5.740 |
Neg logit contributions | |
citiz | -19.318 |
eleph | -19.142 |
teasp | -18.275 |
田 | -18.154 |
TAIN | -17.782 |
Token | Poll |
Token ID | 12868 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
tested | +2.899 |
expected | +2.662 |
coal | +2.647 |
isot | +2.458 |
Prediction | +2.374 |
Neg logit contributions | |
citiz | -10.368 |
eleph | -10.117 |
lation | -9.609 |
TAIN | -9.562 |
teasp | -9.388 |
Token | ing |
Token ID | 278 |
Feature activation | -6.08e+00 |
Pos logit contributions | |
tested | +1.591 |
expected | +1.536 |
coal | +1.522 |
isot | +1.459 |
cube | +1.455 |
Neg logit contributions | |
citiz | -1.640 |
eleph | -1.549 |
lation | -1.492 |
TAIN | -1.421 |
teasp | -1.402 |
Token | survey |
Token ID | 5526 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
tested | +3.388 |
expected | +3.222 |
coal | +3.172 |
isot | +3.042 |
cube | +2.929 |
Neg logit contributions | |
citiz | -5.851 |
eleph | -5.653 |
lation | -5.391 |
teasp | -5.314 |
TAIN | -5.258 |
Token | paints |
Token ID | 32281 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
tested | +1.689 |
expected | +1.616 |
coal | +1.594 |
isot | +1.527 |
cube | +1.515 |
Neg logit contributions | |
citiz | -2.184 |
eleph | -2.083 |
lation | -1.991 |
TAIN | -1.924 |
teasp | -1.917 |
Token | or |
Token ID | 393 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
tested | +4.698 |
expected | +4.513 |
coal | +4.438 |
isot | +4.392 |
cube | +4.388 |
Neg logit contributions | |
citiz | -5.229 |
eleph | -5.164 |
teasp | -4.674 |
pione | -4.590 |
� | -4.552 |
Token | 4 |
Token ID | 604 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
tested | +3.834 |
expected | +3.627 |
cube | +3.536 |
isot | +3.524 |
coal | +3.511 |
Neg logit contributions | |
citiz | -4.527 |
eleph | -4.417 |
teasp | -4.032 |
lation | -3.953 |
pione | -3.858 |
Token | x |
Token ID | 87 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
tested | +2.436 |
expected | +2.305 |
coal | +2.258 |
cube | +2.251 |
cli | +2.130 |
Neg logit contributions | |
citiz | -4.600 |
eleph | -4.529 |
teasp | -4.131 |
pione | -4.001 |
lation | -3.992 |
Token | 1080 |
Token ID | 17729 |
Feature activation | -4.83e+00 |
Pos logit contributions | |
citiz | +2.350 |
lation | +2.196 |
eleph | +2.180 |
TAIN | +2.052 |
田 | +1.963 |
Neg logit contributions | |
tested | -2.797 |
expected | -2.739 |
coal | -2.716 |
isot | -2.591 |
cube | -2.566 |
Token | P |
Token ID | 47 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
tested | +2.388 |
expected | +2.272 |
coal | +2.215 |
cube | +2.160 |
cli | +2.058 |
Neg logit contributions | |
citiz | -5.271 |
eleph | -5.224 |
teasp | -4.736 |
lation | -4.649 |
� | -4.550 |
Token | resolution |
Token ID | 6323 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
citiz | +0.681 |
eleph | +0.643 |
lation | +0.632 |
TAIN | +0.601 |
teasp | +0.586 |
Neg logit contributions | |
tested | -0.558 |
expected | -0.540 |
coal | -0.537 |
isot | -0.509 |
cube | -0.506 |
Token | you |
Token ID | 345 |
Feature activation | -7.28e+00 |
Pos logit contributions | |
tested | +6.171 |
expected | +5.380 |
shocks | +4.962 |
cube | +4.925 |
coal | +4.832 |
Neg logit contributions | |
eleph | -22.211 |
teasp | -21.554 |
pione | -21.470 |
� | -21.269 |
citiz | -21.120 |
Token | will |
Token ID | 481 |
Feature activation | -9.83e+00 |
Pos logit contributions | |
tested | +4.296 |
expected | +4.204 |
coal | +4.019 |
cube | +3.941 |
isot | +3.848 |
Neg logit contributions | |
citiz | -7.031 |
eleph | -7.017 |
teasp | -6.551 |
� | -6.408 |
pione | -6.390 |
Token | still |
Token ID | 991 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
tested | +4.434 |
expected | +4.101 |
cube | +3.887 |
coal | +3.838 |
isot | +3.592 |
Neg logit contributions | |
citiz | -10.822 |
eleph | -10.682 |
teasp | -10.211 |
� | -10.191 |
pione | -10.089 |
Token | be |
Token ID | 307 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
tested | +3.745 |
coal | +3.539 |
expected | +3.532 |
isot | +3.337 |
cube | +3.303 |
Neg logit contributions | |
citiz | -6.409 |
eleph | -6.144 |
teasp | -5.866 |
� | -5.677 |
lation | -5.612 |
Token | able |
Token ID | 1498 |
Feature activation | -1.69e+01 |
Pos logit contributions | |
tested | +0.612 |
expected | +0.593 |
coal | +0.590 |
isot | +0.566 |
cube | +0.563 |
Neg logit contributions | |
citiz | -0.547 |
eleph | -0.514 |
lation | -0.494 |
TAIN | -0.466 |
teasp | -0.465 |
Token | for |
Token ID | 329 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
tested | +1.626 |
expected | +1.567 |
coal | +1.559 |
isot | +1.509 |
cube | +1.472 |
Neg logit contributions | |
citiz | -1.941 |
eleph | -1.856 |
lation | -1.780 |
teasp | -1.716 |
TAIN | -1.699 |
Token | Islamist |
Token ID | 17780 |
Feature activation | +7.95e+00 |
Pos logit contributions | |
tested | +0.230 |
expected | +0.224 |
coal | +0.222 |
isot | +0.212 |
cube | +0.212 |
Neg logit contributions | |
citiz | -0.235 |
eleph | -0.223 |
lation | -0.215 |
TAIN | -0.204 |
teasp | -0.201 |
Token | militants |
Token ID | 13162 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
citiz | +5.768 |
lation | +5.301 |
eleph | +5.293 |
TAIN | +4.931 |
affili | +4.737 |
Neg logit contributions | |
tested | -6.157 |
expected | -5.906 |
coal | -5.748 |
cube | -5.612 |
cli | -5.541 |
Token | , |
Token ID | 11 |
Feature activation | -7.21e+00 |
Pos logit contributions | |
citiz | +1.789 |
eleph | +1.698 |
lation | +1.649 |
TAIN | +1.566 |
teasp | +1.525 |
Neg logit contributions | |
tested | -1.610 |
expected | -1.555 |
coal | -1.551 |
cube | -1.476 |
isot | -1.473 |
Token | a |
Token ID | 257 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
tested | +4.836 |
expected | +4.721 |
coal | +4.701 |
isot | +4.524 |
cube | +4.408 |
Neg logit contributions | |
citiz | -5.921 |
eleph | -5.759 |
lation | -5.493 |
teasp | -5.405 |
TAIN | -5.157 |
Token | Human |
Token ID | 5524 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
tested | +6.163 |
coal | +6.100 |
expected | +6.012 |
isot | +5.746 |
cube | +5.648 |
Neg logit contributions | |
citiz | -6.699 |
eleph | -6.664 |
lation | -6.166 |
teasp | -6.104 |
TAIN | -5.840 |
Token | Rights |
Token ID | 6923 |
Feature activation | -1.40e+01 |
Pos logit contributions | |
coal | +4.219 |
Aid | +3.852 |
tested | +3.777 |
isot | +3.638 |
expected | +3.548 |
Neg logit contributions | |
citiz | -21.102 |
eleph | -20.535 |
田 | -19.622 |
teasp | -19.472 |
lation | -19.303 |
Token | Watch |
Token ID | 6305 |
Feature activation | -7.60e+00 |
Pos logit contributions | |
coal | +6.098 |
tested | +6.036 |
isot | +5.705 |
cube | +5.665 |
Prediction | +5.659 |
Neg logit contributions | |
eleph | -13.934 |
citiz | -13.710 |
teasp | -12.875 |
lation | -12.399 |
� | -12.392 |
Token | ( |
Token ID | 357 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
tested | +3.922 |
coal | +3.844 |
expected | +3.778 |
isot | +3.658 |
cube | +3.589 |
Neg logit contributions | |
citiz | -7.412 |
eleph | -7.306 |
lation | -6.910 |
teasp | -6.781 |
TAIN | -6.636 |
Token | HR |
Token ID | 17184 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
citiz | +1.028 |
eleph | +0.972 |
lation | +0.953 |
TAIN | +0.911 |
田 | +0.886 |
Neg logit contributions | |
tested | -0.950 |
expected | -0.913 |
coal | -0.909 |
isot | -0.877 |
cube | -0.868 |
Token | W |
Token ID | 54 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
tested | +2.314 |
expected | +2.219 |
coal | +2.204 |
cube | +2.101 |
isot | +2.058 |
Neg logit contributions | |
citiz | -3.757 |
eleph | -3.652 |
lation | -3.500 |
teasp | -3.369 |
TAIN | -3.289 |
Token | an |
Token ID | 281 |
Feature activation | -9.79e+00 |
Pos logit contributions | |
citiz | +0.888 |
eleph | +0.840 |
lation | +0.819 |
TAIN | +0.774 |
teasp | +0.761 |
Neg logit contributions | |
tested | -0.844 |
expected | -0.817 |
coal | -0.816 |
isot | -0.777 |
cube | -0.775 |
Token | interview |
Token ID | 2720 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
expected | +5.442 |
isot | +5.296 |
tested | +5.265 |
coal | +5.039 |
MRI | +4.853 |
Neg logit contributions | |
citiz | -9.019 |
eleph | -8.670 |
teasp | -8.327 |
lation | -8.288 |
TAIN | -8.018 |
Token | with |
Token ID | 351 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
tested | +0.349 |
expected | +0.337 |
coal | +0.333 |
isot | +0.316 |
cube | +0.315 |
Neg logit contributions | |
citiz | -0.495 |
eleph | -0.472 |
lation | -0.458 |
TAIN | -0.438 |
teasp | -0.433 |
Token | the |
Token ID | 262 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
citiz | +0.089 |
eleph | +0.085 |
lation | +0.082 |
TAIN | +0.078 |
teasp | +0.078 |
Neg logit contributions | |
tested | -0.075 |
expected | -0.073 |
coal | -0.072 |
isot | -0.069 |
cube | -0.069 |
Token | Colorado |
Token ID | 7492 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
tested | +1.006 |
expected | +0.971 |
coal | +0.968 |
isot | +0.925 |
cube | +0.922 |
Neg logit contributions | |
citiz | -1.044 |
eleph | -1.000 |
lation | -0.959 |
teasp | -0.909 |
TAIN | -0.903 |
Token | Civil |
Token ID | 7511 |
Feature activation | -1.89e+01 |
Pos logit contributions | |
citiz | +2.459 |
eleph | +2.318 |
lation | +2.271 |
TAIN | +2.177 |
teasp | +2.121 |
Neg logit contributions | |
tested | -1.762 |
expected | -1.694 |
coal | -1.667 |
isot | -1.599 |
cube | -1.595 |
Token | Rights |
Token ID | 6923 |
Feature activation | -9.94e+00 |
Pos logit contributions | |
coal | +5.625 |
tested | +5.612 |
Prediction | +5.217 |
felt | +4.930 |
Choice | +4.914 |
Neg logit contributions | |
eleph | -21.526 |
citiz | -20.630 |
teasp | -20.614 |
pione | -19.914 |
� | -19.683 |
Token | Division |
Token ID | 7458 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
tested | +4.939 |
coal | +4.697 |
expected | +4.592 |
isot | +4.377 |
cube | +4.327 |
Neg logit contributions | |
citiz | -9.593 |
eleph | -9.410 |
lation | -8.908 |
TAIN | -8.594 |
田 | -8.547 |
Token | , |
Token ID | 11 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
tested | +0.306 |
expected | +0.295 |
coal | +0.293 |
isot | +0.279 |
cube | +0.278 |
Neg logit contributions | |
citiz | -0.376 |
eleph | -0.358 |
lation | -0.347 |
TAIN | -0.330 |
teasp | -0.326 |
Token | La |
Token ID | 4689 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
citiz | +2.720 |
eleph | +2.565 |
lation | +2.514 |
TAIN | +2.361 |
teasp | +2.299 |
Neg logit contributions | |
tested | -2.721 |
expected | -2.642 |
coal | -2.640 |
cube | -2.529 |
isot | -2.502 |
Token | Feb |
Token ID | 15146 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
citiz | +1.487 |
lation | +1.385 |
eleph | +1.371 |
TAIN | +1.308 |
田 | +1.255 |
Neg logit contributions | |
tested | -1.700 |
expected | -1.652 |
coal | -1.645 |
isot | -1.585 |
cube | -1.580 |
Token | even |
Token ID | 772 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
tested | +4.784 |
cube | +4.415 |
expected | +4.409 |
coal | +4.343 |
isot | +4.055 |
Neg logit contributions | |
eleph | -13.258 |
citiz | -13.165 |
teasp | -12.632 |
pione | -12.626 |
� | -12.552 |
Token | play |
Token ID | 711 |
Feature activation | -6.67e+00 |
Pos logit contributions | |
tested | +4.657 |
coal | +4.466 |
cube | +4.460 |
expected | +4.423 |
isot | +4.281 |
Neg logit contributions | |
citiz | -3.871 |
eleph | -3.844 |
teasp | -3.464 |
� | -3.276 |
pione | -3.236 |
Token | games |
Token ID | 1830 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
tested | +4.823 |
coal | +4.402 |
expected | +4.386 |
cube | +4.325 |
isot | +4.260 |
Neg logit contributions | |
eleph | -5.644 |
citiz | -5.582 |
teasp | -5.158 |
lation | -4.973 |
� | -4.921 |
Token | with |
Token ID | 351 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
tested | +5.395 |
cube | +4.535 |
expected | +4.345 |
coal | +4.254 |
crit | +3.922 |
Neg logit contributions | |
eleph | -17.018 |
citiz | -16.734 |
� | -15.882 |
pione | -15.843 |
teasp | -15.787 |
Token | AA |
Token ID | 15923 |
Feature activation | -8.54e+00 |
Pos logit contributions | |
tested | +5.917 |
isot | +5.432 |
coal | +5.426 |
expected | +5.375 |
cube | +5.182 |
Neg logit contributions | |
citiz | -9.877 |
eleph | -9.784 |
teasp | -9.335 |
pione | -9.021 |
� | -8.904 |
Token | on |
Token ID | 319 |
Feature activation | -1.89e+01 |
Pos logit contributions | |
tested | +5.263 |
coal | +4.829 |
expected | +4.814 |
cube | +4.578 |
cli | +4.503 |
Neg logit contributions | |
citiz | -8.382 |
eleph | -8.189 |
teasp | -7.670 |
� | -7.555 |
pione | -7.485 |
Token | ( |
Token ID | 357 |
Feature activation | -8.79e+00 |
Pos logit contributions | |
tested | +6.328 |
expected | +5.261 |
coal | +5.182 |
cube | +5.058 |
Choice | +4.785 |
Neg logit contributions | |
eleph | -22.306 |
citiz | -21.576 |
pione | -20.650 |
� | -20.566 |
teasp | -20.503 |
Token | Anti |
Token ID | 28795 |
Feature activation | -9.42e+00 |
Pos logit contributions | |
tested | +4.438 |
coal | +3.989 |
expected | +3.909 |
Wide | +3.724 |
Boost | +3.566 |
Neg logit contributions | |
citiz | -10.013 |
eleph | -9.843 |
teasp | -9.089 |
pione | -8.715 |
� | -8.672 |
Token | Ali |
Token ID | 37893 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
tested | +4.673 |
coal | +4.347 |
expected | +4.345 |
cube | +4.072 |
Wide | +4.048 |
Neg logit contributions | |
citiz | -10.417 |
eleph | -10.293 |
teasp | -9.563 |
pione | -9.385 |
� | -9.173 |
Token | asing |
Token ID | 2313 |
Feature activation | -1.22e+01 |
Pos logit contributions | |
citiz | +1.438 |
lation | +1.372 |
eleph | +1.312 |
TAIN | +1.267 |
田 | +1.213 |
Neg logit contributions | |
tested | -1.771 |
expected | -1.724 |
coal | -1.709 |
isot | -1.650 |
cube | -1.645 |
Token | is |
Token ID | 318 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
tested | +5.873 |
coal | +5.287 |
expected | +5.247 |
cube | +5.081 |
isot | +5.048 |
Neg logit contributions | |
citiz | -13.546 |
eleph | -13.432 |
teasp | -12.788 |
pione | -12.709 |
� | -12.684 |
Token | during |
Token ID | 1141 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
tested | +0.789 |
expected | +0.759 |
coal | +0.754 |
isot | +0.723 |
cube | +0.717 |
Neg logit contributions | |
citiz | -1.007 |
eleph | -0.966 |
lation | -0.925 |
teasp | -0.886 |
TAIN | -0.884 |
Token | the |
Token ID | 262 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
citiz | +1.819 |
eleph | +1.698 |
lation | +1.669 |
TAIN | +1.561 |
teasp | +1.512 |
Neg logit contributions | |
tested | -2.118 |
expected | -2.066 |
coal | -2.056 |
cube | -1.956 |
isot | -1.956 |
Token | development |
Token ID | 2478 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
citiz | +2.999 |
lation | +2.799 |
eleph | +2.744 |
TAIN | +2.546 |
田 | +2.456 |
Neg logit contributions | |
tested | -3.624 |
expected | -3.511 |
coal | -3.507 |
cube | -3.326 |
cli | -3.326 |
Token | but |
Token ID | 475 |
Feature activation | -5.83e+00 |
Pos logit contributions | |
tested | +1.321 |
expected | +1.272 |
coal | +1.264 |
isot | +1.211 |
cube | +1.207 |
Neg logit contributions | |
citiz | -1.763 |
eleph | -1.685 |
lation | -1.604 |
teasp | -1.553 |
TAIN | -1.547 |
Token | couldn |
Token ID | 3521 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
tested | +3.596 |
expected | +3.447 |
coal | +3.349 |
isot | +3.233 |
cube | +3.191 |
Neg logit contributions | |
citiz | -5.265 |
eleph | -5.194 |
teasp | -4.851 |
lation | -4.809 |
TAIN | -4.695 |
Token | 't |
Token ID | 470 |
Feature activation | -1.88e+01 |
Pos logit contributions | |
tested | +2.961 |
expected | +2.591 |
cube | +2.365 |
coal | +2.153 |
felt | +2.002 |
Neg logit contributions | |
eleph | -18.273 |
citiz | -17.867 |
pione | -17.312 |
teasp | -17.166 |
� | -17.144 |
Token | fit |
Token ID | 4197 |
Feature activation | -5.16e+00 |
Pos logit contributions | |
tested | +9.728 |
expected | +9.446 |
isot | +9.440 |
cube | +9.324 |
coal | +9.106 |
Neg logit contributions | |
eleph | -15.874 |
teasp | -15.632 |
citiz | -15.613 |
pione | -15.416 |
� | -14.900 |
Token | in |
Token ID | 287 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
tested | +3.521 |
expected | +3.354 |
coal | +3.264 |
isot | +3.252 |
cube | +3.174 |
Neg logit contributions | |
citiz | -4.365 |
eleph | -4.182 |
lation | -3.925 |
teasp | -3.908 |
田 | -3.726 |
Token | - |
Token ID | 532 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
tested | +2.071 |
expected | +1.988 |
coal | +1.951 |
isot | +1.885 |
cube | +1.875 |
Neg logit contributions | |
citiz | -2.626 |
eleph | -2.525 |
lation | -2.414 |
teasp | -2.326 |
TAIN | -2.291 |
Token | those |
Token ID | 883 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
tested | +1.197 |
expected | +1.155 |
coal | +1.145 |
isot | +1.097 |
cube | +1.091 |
Neg logit contributions | |
citiz | -1.451 |
eleph | -1.389 |
lation | -1.333 |
teasp | -1.276 |
TAIN | -1.273 |
Token | were |
Token ID | 547 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
tested | +0.254 |
expected | +0.246 |
coal | +0.244 |
isot | +0.233 |
cube | +0.233 |
Neg logit contributions | |
citiz | -0.300 |
eleph | -0.286 |
lation | -0.275 |
TAIN | -0.262 |
teasp | -0.261 |
Token | � |
Token ID | 447 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
citiz | +1.399 |
eleph | +1.324 |
lation | +1.290 |
TAIN | +1.226 |
teasp | +1.202 |
Neg logit contributions | |
tested | -1.194 |
expected | -1.163 |
coal | -1.148 |
cube | -1.094 |
isot | -1.090 |
Token | � |
Token ID | 247 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
citiz | +0.706 |
eleph | +0.677 |
lation | +0.646 |
teasp | +0.617 |
TAIN | +0.616 |
Neg logit contributions | |
tested | -0.565 |
expected | -0.545 |
coal | -0.544 |
isot | -0.513 |
cube | -0.511 |
Token | s |
Token ID | 82 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
citiz | +0.012 |
eleph | +0.012 |
lation | +0.012 |
TAIN | +0.011 |
teasp | +0.011 |
Neg logit contributions | |
tested | -0.006 |
expected | -0.006 |
coal | -0.006 |
isot | -0.005 |
cube | -0.005 |
Token | Health |
Token ID | 3893 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
citiz | +2.595 |
eleph | +2.424 |
lation | +2.387 |
TAIN | +2.239 |
teasp | +2.155 |
Neg logit contributions | |
tested | -2.701 |
expected | -2.635 |
coal | -2.600 |
cube | -2.496 |
isot | -2.481 |
Token | and |
Token ID | 290 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
tested | +9.499 |
coal | +9.029 |
isot | +8.701 |
expected | +8.670 |
Prediction | +8.581 |
Neg logit contributions | |
citiz | -12.498 |
eleph | -12.167 |
teasp | -11.414 |
lation | -11.387 |
田 | -11.088 |
Token | Social |
Token ID | 5483 |
Feature activation | -1.88e+01 |
Pos logit contributions | |
tested | +2.750 |
expected | +2.603 |
coal | +2.578 |
isot | +2.454 |
cube | +2.387 |
Neg logit contributions | |
citiz | -4.523 |
eleph | -4.457 |
lation | -4.236 |
teasp | -4.188 |
TAIN | -4.055 |
Token | Development |
Token ID | 7712 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
tested | +9.473 |
expected | +8.973 |
coal | +8.949 |
Prediction | +8.830 |
Enh | +8.759 |
Neg logit contributions | |
citiz | -15.659 |
eleph | -15.501 |
teasp | -15.065 |
lation | -14.605 |
田 | -14.160 |
Token | program |
Token ID | 1430 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
tested | +7.376 |
coal | +7.086 |
isot | +6.975 |
expected | +6.799 |
Prediction | +6.705 |
Neg logit contributions | |
eleph | -8.926 |
citiz | -8.897 |
teasp | -8.263 |
lation | -8.121 |
TAIN | -7.967 |
Token | develops |
Token ID | 21126 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
citiz | +1.848 |
eleph | +1.744 |
lation | +1.708 |
TAIN | +1.615 |
teasp | +1.577 |
Neg logit contributions | |
tested | -1.604 |
expected | -1.562 |
coal | -1.544 |
cube | -1.472 |
isot | -1.465 |
Token | knowledge |
Token ID | 3725 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
tested | +1.027 |
expected | +0.989 |
coal | +0.984 |
isot | +0.947 |
cube | +0.939 |
Neg logit contributions | |
citiz | -1.095 |
eleph | -1.045 |
lation | -1.004 |
teasp | -0.953 |
TAIN | -0.952 |
Token | and |
Token ID | 290 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
tested | +1.679 |
expected | +1.615 |
coal | +1.604 |
isot | +1.530 |
cube | +1.525 |
Neg logit contributions | |
citiz | -1.973 |
eleph | -1.884 |
lation | -1.814 |
TAIN | -1.735 |
teasp | -1.728 |
Token | that |
Token ID | 326 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
citiz | +2.231 |
eleph | +2.097 |
lation | +2.060 |
TAIN | +1.940 |
teasp | +1.882 |
Neg logit contributions | |
tested | -2.071 |
expected | -2.016 |
coal | -2.006 |
cube | -1.900 |
isot | -1.890 |
Token | is |
Token ID | 318 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
citiz | +3.834 |
eleph | +3.566 |
lation | +3.510 |
TAIN | +3.324 |
田 | +3.214 |
Neg logit contributions | |
tested | -3.240 |
expected | -3.143 |
coal | -3.103 |
cube | -2.942 |
cli | -2.925 |
Token | both |
Token ID | 1111 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
citiz | +1.256 |
eleph | +1.188 |
lation | +1.158 |
TAIN | +1.097 |
teasp | +1.075 |
Neg logit contributions | |
tested | -1.060 |
expected | -1.022 |
coal | -1.017 |
cube | -0.968 |
isot | -0.963 |
Token | artistic |
Token ID | 17290 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
citiz | +1.570 |
eleph | +1.474 |
lation | +1.433 |
TAIN | +1.350 |
teasp | +1.314 |
Neg logit contributions | |
tested | -1.651 |
expected | -1.600 |
coal | -1.593 |
cube | -1.525 |
isot | -1.513 |
Token | and |
Token ID | 290 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
tested | +0.997 |
expected | +0.957 |
coal | +0.954 |
isot | +0.913 |
cube | +0.908 |
Neg logit contributions | |
citiz | -1.237 |
eleph | -1.180 |
lation | -1.140 |
TAIN | -1.088 |
teasp | -1.082 |
Token | social |
Token ID | 1919 |
Feature activation | -1.87e+01 |
Pos logit contributions | |
citiz | +2.965 |
eleph | +2.688 |
lation | +2.645 |
TAIN | +2.464 |
田 | +2.335 |
Neg logit contributions | |
tested | -4.148 |
expected | -4.041 |
coal | -4.035 |
cube | -3.889 |
cli | -3.843 |
Token | , |
Token ID | 11 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
tested | +10.778 |
crit | +10.219 |
isot | +10.149 |
expected | +9.832 |
shocks | +9.821 |
Neg logit contributions | |
eleph | -13.940 |
citiz | -13.856 |
teasp | -13.737 |
lation | -13.155 |
田 | -12.839 |
Token | public |
Token ID | 1171 |
Feature activation | -9.59e+00 |
Pos logit contributions | |
tested | +0.518 |
expected | +0.499 |
coal | +0.495 |
isot | +0.477 |
cube | +0.471 |
Neg logit contributions | |
citiz | -0.626 |
eleph | -0.599 |
lation | -0.575 |
teasp | -0.552 |
TAIN | -0.551 |
Token | opinions |
Token ID | 9317 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
tested | +6.439 |
crit | +6.309 |
isot | +6.247 |
expected | +6.203 |
shocks | +6.141 |
Neg logit contributions | |
citiz | -7.374 |
eleph | -7.200 |
teasp | -6.738 |
lation | -6.620 |
TAIN | -6.579 |
Token | of |
Token ID | 286 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
tested | +1.813 |
expected | +1.743 |
coal | +1.735 |
isot | +1.678 |
cube | +1.666 |
Neg logit contributions | |
citiz | -2.067 |
eleph | -1.979 |
lation | -1.867 |
teasp | -1.835 |
TAIN | -1.812 |
Token | pun |
Token ID | 4000 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
citiz | +0.939 |
eleph | +0.885 |
lation | +0.853 |
TAIN | +0.805 |
teasp | +0.795 |
Neg logit contributions | |
tested | -1.038 |
expected | -1.008 |
coal | -1.002 |
isot | -0.961 |
cube | -0.959 |
Token | the |
Token ID | 262 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
tested | +3.937 |
expected | +3.707 |
coal | +3.624 |
cube | +3.584 |
isot | +3.545 |
Neg logit contributions | |
eleph | -6.259 |
citiz | -6.255 |
teasp | -5.793 |
� | -5.545 |
pione | -5.535 |
Token | clone |
Token ID | 17271 |
Feature activation | +5.99e+00 |
Pos logit contributions | |
citiz | +2.458 |
lation | +2.306 |
eleph | +2.272 |
TAIN | +2.151 |
utenant | +2.050 |
Neg logit contributions | |
tested | -2.576 |
expected | -2.491 |
coal | -2.466 |
cube | -2.337 |
isot | -2.321 |
Token | � |
Token ID | 447 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
citiz | +3.810 |
lation | +3.613 |
eleph | +3.566 |
TAIN | +3.269 |
teasp | +3.145 |
Neg logit contributions | |
tested | -5.074 |
coal | -4.927 |
expected | -4.925 |
cube | -4.757 |
cli | -4.667 |
Token | � |
Token ID | 247 |
Feature activation | -6.65e+00 |
Pos logit contributions | |
tested | +1.297 |
coal | +1.255 |
expected | +1.253 |
cli | +1.171 |
isot | +1.151 |
Neg logit contributions | |
citiz | -1.850 |
eleph | -1.810 |
lation | -1.666 |
teasp | -1.647 |
TAIN | -1.610 |
Token | s |
Token ID | 82 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
tested | +2.549 |
expected | +2.306 |
coal | +2.263 |
cube | +2.102 |
cli | +2.008 |
Neg logit contributions | |
citiz | -8.087 |
eleph | -7.969 |
teasp | -7.457 |
pione | -7.239 |
� | -7.200 |
Token | state |
Token ID | 1181 |
Feature activation | -1.87e+01 |
Pos logit contributions | |
tested | +2.282 |
expected | +2.202 |
coal | +2.185 |
isot | +2.113 |
cube | +2.080 |
Neg logit contributions | |
citiz | -3.248 |
eleph | -3.130 |
lation | -2.978 |
teasp | -2.882 |
TAIN | -2.864 |
Token | ( |
Token ID | 357 |
Feature activation | -7.35e+00 |
Pos logit contributions | |
tested | +5.099 |
expected | +4.307 |
cube | +4.116 |
felt | +3.966 |
csv | +3.938 |
Neg logit contributions | |
eleph | -22.487 |
citiz | -22.089 |
teasp | -20.919 |
pione | -20.893 |
� | -20.633 |
Token | such |
Token ID | 10508 |
Feature activation | -1.18e+01 |
Pos logit contributions | |
tested | +4.823 |
expected | +4.551 |
coal | +4.503 |
cube | +4.192 |
crit | +4.138 |
Neg logit contributions | |
citiz | -7.277 |
eleph | -7.189 |
teasp | -6.596 |
� | -6.417 |
pione | -6.397 |
Token | as |
Token ID | 355 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
tested | +6.539 |
cube | +5.893 |
expected | +5.841 |
coal | +5.765 |
felt | +5.534 |
Neg logit contributions | |
eleph | -12.642 |
citiz | -12.487 |
� | -11.681 |
teasp | -11.595 |
pione | -11.534 |
Token | through |
Token ID | 832 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
tested | +2.263 |
expected | +2.172 |
coal | +2.146 |
isot | +2.060 |
cube | +2.045 |
Neg logit contributions | |
citiz | -2.662 |
eleph | -2.550 |
lation | -2.442 |
TAIN | -2.322 |
teasp | -2.313 |
Token | a |
Token ID | 257 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
tested | +6.498 |
expected | +6.116 |
cube | +5.943 |
isot | +5.908 |
coal | +5.905 |
Neg logit contributions | |
eleph | -12.467 |
citiz | -12.306 |
teasp | -11.353 |
� | -11.224 |
pione | -11.130 |
Token | Jill |
Token ID | 22772 |
Feature activation | -7.78e+00 |
Pos logit contributions | |
tested | +5.598 |
expected | +5.308 |
coal | +5.275 |
isot | +5.259 |
cube | +5.045 |
Neg logit contributions | |
citiz | -6.487 |
eleph | -6.360 |
lation | -6.019 |
teasp | -5.998 |
TAIN | -5.697 |
Token | ian |
Token ID | 666 |
Feature activation | -8.92e+00 |
Pos logit contributions | |
tested | +5.128 |
expected | +4.906 |
coal | +4.870 |
isot | +4.701 |
cube | +4.692 |
Neg logit contributions | |
citiz | -6.392 |
eleph | -6.195 |
lation | -5.874 |
teasp | -5.649 |
TAIN | -5.618 |
Token | , |
Token ID | 11 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
tested | +5.697 |
expected | +5.528 |
coal | +5.481 |
isot | +5.340 |
cube | +5.261 |
Neg logit contributions | |
citiz | -7.216 |
eleph | -7.029 |
lation | -6.712 |
teasp | -6.467 |
TAIN | -6.346 |
Token | we |
Token ID | 356 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
tested | +2.617 |
expected | +2.498 |
coal | +2.456 |
isot | +2.366 |
cube | +2.325 |
Neg logit contributions | |
citiz | -3.725 |
eleph | -3.606 |
lation | -3.470 |
teasp | -3.388 |
TAIN | -3.293 |
Token | couldn |
Token ID | 3521 |
Feature activation | -1.09e+01 |
Pos logit contributions | |
tested | +3.413 |
expected | +3.327 |
coal | +3.170 |
isot | +3.079 |
cube | +3.020 |
Neg logit contributions | |
citiz | -5.053 |
eleph | -4.853 |
lation | -4.571 |
teasp | -4.549 |
TAIN | -4.445 |
Token | 't |
Token ID | 470 |
Feature activation | -1.85e+01 |
Pos logit contributions | |
tested | +4.616 |
expected | +4.375 |
coal | +4.015 |
cube | +3.974 |
felt | +3.882 |
Neg logit contributions | |
eleph | -13.378 |
citiz | -13.132 |
teasp | -12.248 |
pione | -12.153 |
� | -11.923 |
Token | resist |
Token ID | 4180 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
expected | +8.087 |
isot | +7.745 |
tested | +7.645 |
shocks | +7.210 |
cube | +6.923 |
Neg logit contributions | |
citiz | -18.040 |
teasp | -17.837 |
eleph | -17.760 |
pione | -17.429 |
� | -17.228 |
Token | the |
Token ID | 262 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
tested | +3.242 |
expected | +3.150 |
coal | +3.085 |
isot | +3.046 |
cube | +2.949 |
Neg logit contributions | |
citiz | -4.097 |
eleph | -3.924 |
lation | -3.752 |
teasp | -3.695 |
TAIN | -3.592 |
Token | temptation |
Token ID | 29062 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
tested | +3.082 |
expected | +2.960 |
coal | +2.912 |
isot | +2.845 |
cube | +2.796 |
Neg logit contributions | |
citiz | -3.843 |
eleph | -3.661 |
lation | -3.510 |
TAIN | -3.384 |
teasp | -3.358 |
Token | to |
Token ID | 284 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
tested | +1.370 |
expected | +1.321 |
coal | +1.312 |
isot | +1.255 |
cube | +1.254 |
Neg logit contributions | |
citiz | -1.563 |
eleph | -1.480 |
lation | -1.429 |
TAIN | -1.362 |
teasp | -1.346 |
Token | recall |
Token ID | 10014 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
tested | +3.554 |
expected | +3.447 |
coal | +3.355 |
isot | +3.353 |
cube | +3.237 |
Neg logit contributions | |
citiz | -4.726 |
eleph | -4.577 |
teasp | -4.323 |
lation | -4.274 |
TAIN | -4.115 |
Token |
|
Token ID | 198 |
Feature activation | -9.14e+00 |
Pos logit contributions | |
tested | +2.963 |
expected | +2.849 |
coal | +2.811 |
cube | +2.678 |
csv | +2.553 |
Neg logit contributions | |
citiz | -5.112 |
eleph | -4.989 |
teasp | -4.572 |
lation | -4.550 |
田 | -4.318 |
Token |
|
Token ID | 198 |
Feature activation | -7.75e+00 |
Pos logit contributions | |
tested | +4.405 |
expected | +4.088 |
cube | +3.942 |
coal | +3.942 |
csv | +3.668 |
Neg logit contributions | |
eleph | -10.977 |
citiz | -10.864 |
teasp | -10.143 |
� | -9.941 |
pione | -9.933 |
Token | The |
Token ID | 464 |
Feature activation | -6.14e+00 |
Pos logit contributions | |
tested | +5.056 |
expected | +4.970 |
coal | +4.808 |
cube | +4.597 |
Wide | +4.526 |
Neg logit contributions | |
eleph | -7.299 |
citiz | -7.251 |
teasp | -6.530 |
pione | -6.205 |
� | -6.187 |
Token | Bank |
Token ID | 5018 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
tested | +3.988 |
expected | +3.907 |
coal | +3.801 |
isot | +3.672 |
cube | +3.614 |
Neg logit contributions | |
citiz | -5.264 |
eleph | -5.149 |
lation | -4.750 |
TAIN | -4.721 |
teasp | -4.720 |
Token | 's |
Token ID | 338 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
tested | +3.271 |
expected | +3.153 |
coal | +3.088 |
cube | +2.971 |
isot | +2.959 |
Neg logit contributions | |
citiz | -4.916 |
eleph | -4.750 |
lation | -4.480 |
teasp | -4.382 |
TAIN | -4.361 |
Token | Monetary |
Token ID | 34112 |
Feature activation | -1.85e+01 |
Pos logit contributions | |
tested | +4.284 |
expected | +4.209 |
coal | +4.112 |
isot | +3.994 |
cube | +3.913 |
Neg logit contributions | |
citiz | -4.983 |
eleph | -4.864 |
lation | -4.480 |
teasp | -4.466 |
TAIN | -4.435 |
Token | Policy |
Token ID | 7820 |
Feature activation | -8.33e+00 |
Pos logit contributions | |
tested | +5.454 |
Prediction | +5.212 |
Choice | +4.911 |
Enh | +4.862 |
expected | +4.821 |
Neg logit contributions | |
citiz | -20.205 |
eleph | -20.075 |
teasp | -19.049 |
田 | -18.473 |
lation | -18.373 |
Token | Committee |
Token ID | 4606 |
Feature activation | -5.72e+00 |
Pos logit contributions | |
tested | +2.731 |
coal | +2.515 |
expected | +2.498 |
isot | +2.235 |
Prediction | +2.207 |
Neg logit contributions | |
citiz | -9.844 |
eleph | -9.637 |
TAIN | -8.984 |
lation | -8.960 |
teasp | -8.925 |
Token | has |
Token ID | 468 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
tested | +3.291 |
expected | +3.203 |
coal | +3.139 |
isot | +2.985 |
cube | +2.951 |
Neg logit contributions | |
citiz | -5.364 |
eleph | -5.251 |
lation | -4.906 |
teasp | -4.894 |
TAIN | -4.758 |
Token | held |
Token ID | 2714 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
tested | +0.156 |
expected | +0.150 |
coal | +0.149 |
cube | +0.141 |
isot | +0.141 |
Neg logit contributions | |
citiz | -0.207 |
eleph | -0.197 |
lation | -0.191 |
TAIN | -0.182 |
teasp | -0.180 |
Token | rates |
Token ID | 3965 |
Feature activation | -6.07e+00 |
Pos logit contributions | |
tested | +1.968 |
expected | +1.917 |
coal | +1.892 |
isot | +1.824 |
cube | +1.801 |
Neg logit contributions | |
citiz | -2.107 |
eleph | -2.013 |
lation | -1.919 |
TAIN | -1.842 |
teasp | -1.836 |
Token | rington |
Token ID | 24833 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
tested | +1.851 |
coal | +1.806 |
expected | +1.805 |
cube | +1.719 |
isot | +1.700 |
Neg logit contributions | |
citiz | -1.560 |
eleph | -1.515 |
teasp | -1.348 |
lation | -1.345 |
TAIN | -1.266 |
Token | . |
Token ID | 13 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
tested | +2.628 |
expected | +2.548 |
coal | +2.506 |
isot | +2.443 |
cube | +2.399 |
Neg logit contributions | |
citiz | -3.298 |
eleph | -3.128 |
lation | -2.976 |
teasp | -2.896 |
TAIN | -2.875 |
Token |
|
Token ID | 198 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
tested | +1.302 |
expected | +1.254 |
coal | +1.242 |
cube | +1.188 |
isot | +1.178 |
Neg logit contributions | |
citiz | -1.611 |
eleph | -1.539 |
lation | -1.454 |
teasp | -1.401 |
TAIN | -1.374 |
Token |
|
Token ID | 198 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
tested | +0.171 |
expected | +0.165 |
coal | +0.164 |
isot | +0.155 |
cube | +0.155 |
Neg logit contributions | |
citiz | -0.233 |
eleph | -0.221 |
lation | -0.216 |
TAIN | -0.206 |
teasp | -0.203 |
Token | The |
Token ID | 464 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
tested | +0.922 |
expected | +0.892 |
coal | +0.885 |
cube | +0.842 |
isot | +0.840 |
Neg logit contributions | |
citiz | -1.091 |
eleph | -1.039 |
lation | -0.995 |
TAIN | -0.945 |
teasp | -0.944 |
Token | Aut |
Token ID | 5231 |
Feature activation | -1.84e+01 |
Pos logit contributions | |
tested | +0.661 |
expected | +0.642 |
coal | +0.636 |
isot | +0.611 |
cube | +0.606 |
Neg logit contributions | |
citiz | -0.734 |
eleph | -0.698 |
lation | -0.672 |
TAIN | -0.639 |
teasp | -0.634 |
Token | on |
Token ID | 261 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
tested | +9.855 |
coal | +9.701 |
expected | +9.499 |
opsy | +9.139 |
isot | +9.125 |
Neg logit contributions | |
citiz | -15.079 |
eleph | -14.653 |
teasp | -13.381 |
田 | -13.073 |
� | -12.795 |
Token | av |
Token ID | 615 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
tested | +2.884 |
expected | +2.779 |
coal | +2.750 |
isot | +2.668 |
cube | +2.640 |
Neg logit contributions | |
citiz | -3.471 |
eleph | -3.293 |
lation | -3.108 |
teasp | -3.015 |
TAIN | -3.004 |
Token | system |
Token ID | 1080 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
citiz | +1.219 |
eleph | +1.170 |
lation | +1.147 |
TAIN | +1.099 |
teasp | +1.084 |
Neg logit contributions | |
tested | -0.604 |
expected | -0.577 |
coal | -0.571 |
cube | -0.531 |
isot | -0.529 |
Token | required |
Token ID | 2672 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
tested | +1.419 |
expected | +1.364 |
coal | +1.347 |
isot | +1.300 |
cube | +1.287 |
Neg logit contributions | |
citiz | -1.730 |
eleph | -1.653 |
lation | -1.576 |
teasp | -1.525 |
TAIN | -1.521 |
Token | occasional |
Token ID | 12209 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
tested | +1.412 |
expected | +1.365 |
coal | +1.356 |
isot | +1.320 |
cube | +1.292 |
Neg logit contributions | |
citiz | -1.482 |
eleph | -1.413 |
lation | -1.348 |
teasp | -1.290 |
TAIN | -1.288 |
Token | people |
Token ID | 661 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
citiz | +5.906 |
lation | +5.169 |
eleph | +5.058 |
TAIN | +4.740 |
田 | +4.341 |
Neg logit contributions | |
tested | -8.054 |
expected | -7.835 |
coal | -7.738 |
cube | -7.545 |
cli | -7.390 |
Token | with |
Token ID | 351 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
tested | +1.480 |
expected | +1.432 |
coal | +1.411 |
isot | +1.349 |
cube | +1.341 |
Neg logit contributions | |
citiz | -1.851 |
eleph | -1.768 |
lation | -1.704 |
TAIN | -1.627 |
teasp | -1.625 |
Token | fixed |
Token ID | 5969 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
tested | +0.930 |
expected | +0.901 |
coal | +0.892 |
isot | +0.862 |
cube | +0.851 |
Neg logit contributions | |
citiz | -0.921 |
eleph | -0.878 |
lation | -0.838 |
teasp | -0.802 |
TAIN | -0.799 |
Token | incomes |
Token ID | 18696 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
tested | +0.616 |
expected | +0.592 |
coal | +0.588 |
isot | +0.565 |
cube | +0.554 |
Neg logit contributions | |
citiz | -0.869 |
eleph | -0.833 |
lation | -0.803 |
TAIN | -0.769 |
teasp | -0.767 |
Token | from |
Token ID | 422 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
tested | +3.706 |
expected | +3.585 |
coal | +3.496 |
isot | +3.354 |
cube | +3.331 |
Neg logit contributions | |
citiz | -5.125 |
eleph | -4.893 |
lation | -4.688 |
teasp | -4.557 |
TAIN | -4.529 |
Token | Social |
Token ID | 5483 |
Feature activation | -1.83e+01 |
Pos logit contributions | |
tested | +0.528 |
expected | +0.511 |
coal | +0.508 |
isot | +0.486 |
cube | +0.482 |
Neg logit contributions | |
citiz | -0.535 |
eleph | -0.510 |
lation | -0.488 |
TAIN | -0.464 |
teasp | -0.461 |
Token | Security |
Token ID | 4765 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
tested | +7.600 |
expected | +6.679 |
coal | +6.657 |
Enh | +6.188 |
Prediction | +6.112 |
Neg logit contributions | |
eleph | -17.453 |
citiz | -17.416 |
teasp | -16.996 |
lation | -16.424 |
田 | -16.370 |
Token | ; |
Token ID | 26 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
tested | +0.439 |
expected | +0.423 |
coal | +0.420 |
isot | +0.402 |
cube | +0.400 |
Neg logit contributions | |
citiz | -0.485 |
eleph | -0.463 |
lation | -0.445 |
TAIN | -0.424 |
teasp | -0.421 |
Token | a |
Token ID | 257 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
citiz | +2.150 |
eleph | +2.020 |
lation | +1.960 |
TAIN | +1.848 |
teasp | +1.824 |
Neg logit contributions | |
tested | -1.994 |
expected | -1.936 |
coal | -1.934 |
cube | -1.829 |
isot | -1.821 |
Token | third |
Token ID | 2368 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
citiz | +2.866 |
eleph | +2.633 |
lation | +2.592 |
teasp | +2.392 |
TAIN | +2.384 |
Neg logit contributions | |
tested | -3.428 |
expected | -3.371 |
coal | -3.338 |
isot | -3.191 |
cube | -3.182 |
Token | offers |
Token ID | 4394 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
citiz | +0.332 |
eleph | +0.314 |
lation | +0.305 |
TAIN | +0.290 |
teasp | +0.286 |
Neg logit contributions | |
tested | -0.287 |
expected | -0.278 |
coal | -0.276 |
cube | -0.263 |
isot | -0.263 |
Token | orns |
Token ID | 19942 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
citiz | +1.647 |
eleph | +1.550 |
lation | +1.542 |
TAIN | +1.419 |
teasp | +1.381 |
Neg logit contributions | |
tested | -2.024 |
expected | -1.954 |
coal | -1.941 |
cube | -1.884 |
isot | -1.875 |
Token | ( |
Token ID | 357 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
tested | +2.049 |
expected | +1.971 |
coal | +1.958 |
isot | +1.875 |
cube | +1.854 |
Neg logit contributions | |
citiz | -2.749 |
eleph | -2.596 |
lation | -2.501 |
TAIN | -2.410 |
teasp | -2.393 |
Token | which |
Token ID | 4758 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
tested | +1.894 |
expected | +1.819 |
coal | +1.799 |
cube | +1.698 |
isot | +1.680 |
Neg logit contributions | |
citiz | -2.697 |
eleph | -2.571 |
lation | -2.455 |
teasp | -2.358 |
TAIN | -2.354 |
Token | I |
Token ID | 314 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
tested | +2.420 |
expected | +2.309 |
coal | +2.281 |
isot | +2.220 |
cube | +2.171 |
Neg logit contributions | |
citiz | -3.283 |
eleph | -3.150 |
lation | -3.007 |
teasp | -2.914 |
TAIN | -2.885 |
Token | couldn |
Token ID | 3521 |
Feature activation | -9.59e+00 |
Pos logit contributions | |
tested | +2.931 |
expected | +2.819 |
coal | +2.756 |
isot | +2.676 |
cube | +2.650 |
Neg logit contributions | |
citiz | -3.470 |
eleph | -3.300 |
lation | -3.111 |
teasp | -3.025 |
TAIN | -3.001 |
Token | 't |
Token ID | 470 |
Feature activation | -1.83e+01 |
Pos logit contributions | |
tested | +2.823 |
expected | +2.554 |
cube | +2.353 |
coal | +2.230 |
felt | +2.044 |
Neg logit contributions | |
eleph | -13.231 |
citiz | -13.075 |
teasp | -12.317 |
pione | -12.268 |
� | -12.234 |
Token | really |
Token ID | 1107 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
tested | +9.770 |
isot | +9.611 |
expected | +9.484 |
cube | +9.018 |
crit | +8.556 |
Neg logit contributions | |
citiz | -16.123 |
eleph | -15.628 |
pione | -15.024 |
teasp | -14.955 |
� | -14.940 |
Token | taste |
Token ID | 6938 |
Feature activation | -8.09e+00 |
Pos logit contributions | |
tested | +6.282 |
expected | +6.203 |
isot | +5.977 |
cube | +5.776 |
coal | +5.775 |
Neg logit contributions | |
citiz | -9.138 |
eleph | -8.888 |
teasp | -8.390 |
pione | -8.114 |
� | -8.077 |
Token | ) |
Token ID | 8 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
tested | +5.153 |
expected | +4.959 |
coal | +4.756 |
cube | +4.655 |
isot | +4.646 |
Neg logit contributions | |
citiz | -7.087 |
eleph | -6.792 |
lation | -6.370 |
teasp | -6.353 |
TAIN | -6.181 |
Token | and |
Token ID | 290 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
tested | +2.730 |
expected | +2.624 |
coal | +2.583 |
isot | +2.482 |
cube | +2.456 |
Neg logit contributions | |
citiz | -3.633 |
eleph | -3.438 |
lation | -3.312 |
TAIN | -3.192 |
teasp | -3.157 |
Token | a |
Token ID | 257 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
tested | +3.871 |
expected | +3.714 |
coal | +3.630 |
isot | +3.568 |
cube | +3.469 |
Neg logit contributions | |
citiz | -4.481 |
eleph | -4.332 |
lation | -4.132 |
TAIN | -3.969 |
teasp | -3.960 |
Token | social |
Token ID | 1919 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
tested | +0.727 |
expected | +0.703 |
coal | +0.698 |
isot | +0.664 |
cube | +0.663 |
Neg logit contributions | |
citiz | -0.885 |
eleph | -0.841 |
lation | -0.815 |
TAIN | -0.775 |
teasp | -0.767 |
Token | media |
Token ID | 2056 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
tested | +9.098 |
coal | +8.745 |
expected | +8.614 |
isot | +8.400 |
cube | +8.033 |
Neg logit contributions | |
citiz | -11.405 |
eleph | -11.159 |
lation | -10.822 |
teasp | -10.422 |
田 | -10.379 |
Token | analytics |
Token ID | 23696 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
tested | +0.799 |
expected | +0.772 |
coal | +0.768 |
isot | +0.734 |
cube | +0.731 |
Neg logit contributions | |
citiz | -0.876 |
eleph | -0.833 |
lation | -0.803 |
TAIN | -0.763 |
teasp | -0.757 |
Token | firm |
Token ID | 4081 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
citiz | +1.659 |
eleph | +1.560 |
lation | +1.540 |
TAIN | +1.445 |
teasp | +1.399 |
Neg logit contributions | |
tested | -1.482 |
expected | -1.442 |
coal | -1.432 |
cube | -1.364 |
isot | -1.353 |
Token | , |
Token ID | 11 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
citiz | +1.241 |
eleph | +1.167 |
lation | +1.151 |
TAIN | +1.095 |
teasp | +1.057 |
Neg logit contributions | |
tested | -0.936 |
expected | -0.908 |
coal | -0.908 |
isot | -0.858 |
cube | -0.857 |
Token | Social |
Token ID | 5483 |
Feature activation | -1.82e+01 |
Pos logit contributions | |
tested | +1.132 |
expected | +1.091 |
coal | +1.079 |
isot | +1.024 |
cube | +1.022 |
Neg logit contributions | |
citiz | -1.520 |
eleph | -1.451 |
lation | -1.407 |
TAIN | -1.340 |
teasp | -1.332 |
Token | Sphere |
Token ID | 38882 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
tested | +11.434 |
coal | +11.194 |
expected | +10.943 |
Prediction | +10.634 |
cube | +10.263 |
Neg logit contributions | |
eleph | -13.491 |
citiz | -13.426 |
lation | -12.315 |
teasp | -12.041 |
田 | -11.971 |
Token | , |
Token ID | 11 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
tested | +3.890 |
expected | +3.756 |
coal | +3.661 |
cube | +3.470 |
isot | +3.445 |
Neg logit contributions | |
citiz | -5.745 |
eleph | -5.505 |
lation | -5.257 |
teasp | -5.036 |
TAIN | -5.019 |
Token | says |
Token ID | 1139 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
tested | +2.001 |
expected | +1.928 |
coal | +1.891 |
isot | +1.794 |
cube | +1.769 |
Neg logit contributions | |
citiz | -3.416 |
eleph | -3.279 |
lation | -3.175 |
TAIN | -3.044 |
teasp | -3.037 |
Token | that |
Token ID | 326 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
citiz | +2.533 |
eleph | +2.333 |
lation | +2.291 |
TAIN | +2.144 |
teasp | +2.043 |
Neg logit contributions | |
tested | -3.159 |
expected | -3.075 |
coal | -3.065 |
cube | -2.946 |
isot | -2.925 |
Token | in |
Token ID | 287 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
tested | +0.441 |
expected | +0.428 |
coal | +0.425 |
isot | +0.408 |
cube | +0.407 |
Neg logit contributions | |
citiz | -0.389 |
eleph | -0.368 |
lation | -0.353 |
TAIN | -0.333 |
teasp | -0.331 |
Token | , |
Token ID | 11 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
citiz | +3.993 |
eleph | +3.748 |
lation | +3.686 |
TAIN | +3.514 |
teasp | +3.404 |
Neg logit contributions | |
tested | -3.644 |
expected | -3.502 |
coal | -3.485 |
isot | -3.318 |
cube | -3.312 |
Token | though |
Token ID | 996 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
citiz | +1.753 |
eleph | +1.652 |
lation | +1.615 |
TAIN | +1.530 |
teasp | +1.483 |
Neg logit contributions | |
tested | -1.608 |
expected | -1.555 |
coal | -1.547 |
cube | -1.480 |
isot | -1.473 |
Token | these |
Token ID | 777 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
tested | +3.066 |
expected | +2.968 |
coal | +2.945 |
isot | +2.813 |
cube | +2.770 |
Neg logit contributions | |
citiz | -3.794 |
eleph | -3.646 |
lation | -3.475 |
teasp | -3.389 |
TAIN | -3.321 |
Token | outcomes |
Token ID | 10906 |
Feature activation | -9.85e+00 |
Pos logit contributions | |
tested | +3.146 |
expected | +3.058 |
coal | +3.048 |
isot | +2.958 |
cube | +2.908 |
Neg logit contributions | |
citiz | -3.754 |
eleph | -3.661 |
lation | -3.444 |
teasp | -3.388 |
TAIN | -3.303 |
Token | stray |
Token ID | 28583 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
tested | +6.071 |
expected | +5.904 |
coal | +5.778 |
isot | +5.764 |
cube | +5.503 |
Neg logit contributions | |
citiz | -8.487 |
eleph | -8.358 |
teasp | -7.916 |
lation | -7.474 |
TAIN | -7.403 |
Token | more |
Token ID | 517 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
tested | +0.449 |
expected | +0.435 |
coal | +0.432 |
isot | +0.412 |
cube | +0.411 |
Neg logit contributions | |
citiz | -0.485 |
eleph | -0.460 |
lation | -0.445 |
TAIN | -0.422 |
teasp | -0.417 |
Token | into |
Token ID | 656 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
citiz | +8.216 |
lation | +7.748 |
eleph | +7.518 |
TAIN | +7.139 |
emphasis | +7.125 |
Neg logit contributions | |
tested | -9.861 |
coal | -9.635 |
expected | -9.359 |
cube | -9.160 |
cli | -9.036 |
Token | fairy |
Token ID | 25607 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
citiz | +2.789 |
eleph | +2.615 |
lation | +2.574 |
TAIN | +2.447 |
teasp | +2.340 |
Neg logit contributions | |
tested | -2.684 |
coal | -2.590 |
expected | -2.561 |
cube | -2.470 |
cli | -2.430 |
Token | tale |
Token ID | 12838 |
Feature activation | +9.65e+00 |
Pos logit contributions | |
citiz | +0.271 |
eleph | +0.259 |
lation | +0.252 |
TAIN | +0.241 |
teasp | +0.239 |
Neg logit contributions | |
tested | -0.161 |
expected | -0.155 |
coal | -0.153 |
isot | -0.144 |
cube | -0.144 |
Token | territory |
Token ID | 7674 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
citiz | +6.770 |
lation | +6.337 |
eleph | +6.174 |
TAIN | +5.878 |
田 | +5.572 |
Neg logit contributions | |
tested | -7.426 |
coal | -7.131 |
expected | -7.072 |
cube | -6.745 |
cli | -6.643 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
citiz | +1.312 |
eleph | +1.246 |
lation | +1.209 |
TAIN | +1.149 |
teasp | +1.134 |
Neg logit contributions | |
tested | -1.105 |
expected | -1.068 |
coal | -1.061 |
isot | -1.009 |
cube | -1.009 |
Token | you |
Token ID | 345 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
tested | +1.890 |
expected | +1.821 |
coal | +1.803 |
isot | +1.740 |
cube | +1.703 |
Neg logit contributions | |
citiz | -2.211 |
eleph | -2.120 |
lation | -2.032 |
teasp | -1.959 |
TAIN | -1.943 |
Token | are |
Token ID | 389 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
citiz | +1.255 |
eleph | +1.174 |
lation | +1.161 |
TAIN | +1.092 |
田 | +1.054 |
Neg logit contributions | |
tested | -1.298 |
expected | -1.255 |
coal | -1.253 |
cube | -1.198 |
isot | -1.189 |
Token | paying |
Token ID | 5989 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
tested | +2.266 |
expected | +2.214 |
coal | +2.164 |
isot | +2.083 |
cube | +2.048 |
Neg logit contributions | |
citiz | -2.869 |
eleph | -2.748 |
lation | -2.632 |
teasp | -2.559 |
TAIN | -2.513 |
Token | for |
Token ID | 329 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
tested | +2.185 |
expected | +2.098 |
coal | +2.062 |
isot | +1.987 |
cube | +1.938 |
Neg logit contributions | |
citiz | -3.061 |
eleph | -2.960 |
lation | -2.840 |
teasp | -2.763 |
TAIN | -2.699 |
Token | at |
Token ID | 379 |
Feature activation | +1.30e+01 |
Pos logit contributions | |
tested | +2.348 |
expected | +2.254 |
coal | +2.226 |
isot | +2.160 |
cube | +2.105 |
Neg logit contributions | |
citiz | -2.806 |
eleph | -2.690 |
lation | -2.576 |
teasp | -2.513 |
TAIN | -2.463 |
Token | the |
Token ID | 262 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
citiz | +7.885 |
lation | +7.318 |
eleph | +7.309 |
TAIN | +6.794 |
bottom | +6.633 |
Neg logit contributions | |
tested | -10.249 |
coal | -10.124 |
expected | -9.949 |
cube | -9.653 |
cli | -9.638 |
Token | supermarket |
Token ID | 26454 |
Feature activation | +4.40e+00 |
Pos logit contributions | |
citiz | +6.695 |
eleph | +6.134 |
lation | +6.087 |
bottom | +5.976 |
TAIN | +5.655 |
Neg logit contributions | |
tested | -11.108 |
coal | -10.805 |
expected | -10.783 |
cube | -10.363 |
cli | -10.264 |
Token | ," |
Token ID | 553 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
citiz | +3.657 |
eleph | +3.464 |
lation | +3.395 |
TAIN | +3.218 |
teasp | +3.119 |
Neg logit contributions | |
tested | -2.969 |
expected | -2.874 |
coal | -2.867 |
cube | -2.713 |
isot | -2.676 |
Token | she |
Token ID | 673 |
Feature activation | -8.44e+00 |
Pos logit contributions | |
tested | +1.438 |
expected | +1.393 |
coal | +1.386 |
cube | +1.317 |
isot | +1.317 |
Neg logit contributions | |
citiz | -1.601 |
eleph | -1.518 |
lation | -1.472 |
TAIN | -1.392 |
teasp | -1.382 |
Token | said |
Token ID | 531 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
tested | +3.192 |
expected | +3.147 |
coal | +2.969 |
cube | +2.727 |
isot | +2.708 |
Neg logit contributions | |
citiz | -9.336 |
eleph | -8.917 |
lation | -8.725 |
TAIN | -8.381 |
teasp | -8.368 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
tested | +1.153 |
expected | +1.115 |
coal | +1.104 |
cube | +1.051 |
isot | +1.049 |
Neg logit contributions | |
citiz | -1.538 |
eleph | -1.462 |
lation | -1.422 |
TAIN | -1.354 |
teasp | -1.340 |
Token | and |
Token ID | 290 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
citiz | +8.023 |
lation | +7.335 |
eleph | +7.107 |
TAIN | +6.779 |
emphasis | +6.485 |
Neg logit contributions | |
expected | -9.407 |
tested | -9.394 |
coal | -9.084 |
cube | -8.777 |
cli | -8.663 |
Token | donate |
Token ID | 16565 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
citiz | +2.507 |
eleph | +2.342 |
lation | +2.306 |
TAIN | +2.187 |
teasp | +2.087 |
Neg logit contributions | |
tested | -2.490 |
expected | -2.411 |
coal | -2.399 |
cube | -2.290 |
isot | -2.275 |
Token | please |
Token ID | 3387 |
Feature activation | +5.48e+00 |
Pos logit contributions | |
citiz | +2.102 |
eleph | +1.965 |
lation | +1.919 |
TAIN | +1.818 |
teasp | +1.770 |
Neg logit contributions | |
tested | -1.944 |
expected | -1.893 |
coal | -1.881 |
cube | -1.787 |
isot | -1.765 |
Token | visit |
Token ID | 3187 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
citiz | +4.642 |
eleph | +4.366 |
lation | +4.317 |
TAIN | +4.166 |
田 | +3.944 |
Neg logit contributions | |
tested | -3.540 |
expected | -3.437 |
coal | -3.385 |
cube | -3.253 |
cli | -3.152 |
Token | their |
Token ID | 511 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
citiz | +7.126 |
eleph | +6.618 |
lation | +6.365 |
TAIN | +6.056 |
utenant | +5.608 |
Neg logit contributions | |
tested | -7.712 |
expected | -7.678 |
coal | -7.433 |
cube | -7.186 |
isot | -7.043 |
Token | website |
Token ID | 3052 |
Feature activation | +1.48e+01 |
Pos logit contributions | |
citiz | +2.843 |
eleph | +2.644 |
lation | +2.586 |
TAIN | +2.405 |
teasp | +2.330 |
Neg logit contributions | |
tested | -3.063 |
expected | -2.983 |
coal | -2.930 |
cube | -2.825 |
isot | -2.810 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.27e-01 |
Pos logit contributions | |
citiz | +9.772 |
eleph | +9.073 |
lation | +8.999 |
TAIN | +8.349 |
田 | +7.824 |
Neg logit contributions | |
tested | -10.691 |
expected | -10.674 |
coal | -10.158 |
cube | -9.921 |
cli | -9.771 |
Token | T |
Token ID | 51 |
Feature activation | +7.02e+00 |
Pos logit contributions | |
tested | +0.644 |
expected | +0.623 |
coal | +0.618 |
cube | +0.588 |
isot | +0.586 |
Neg logit contributions | |
citiz | -0.778 |
eleph | -0.742 |
lation | -0.713 |
TAIN | -0.677 |
teasp | -0.675 |
Token | inder |
Token ID | 5540 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
citiz | +5.628 |
lation | +5.329 |
eleph | +5.262 |
TAIN | +5.030 |
田 | +4.837 |
Neg logit contributions | |
tested | -4.816 |
expected | -4.592 |
coal | -4.530 |
isot | -4.376 |
cube | -4.291 |
Token | is |
Token ID | 318 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
tested | +0.293 |
expected | +0.283 |
coal | +0.281 |
cube | +0.267 |
isot | +0.267 |
Neg logit contributions | |
citiz | -0.362 |
eleph | -0.344 |
lation | -0.333 |
TAIN | -0.317 |
teasp | -0.313 |
Token | reportedly |
Token ID | 7478 |
Feature activation | -9.67e-01 |
Pos logit contributions | |
tested | +1.807 |
expected | +1.753 |
coal | +1.721 |
isot | +1.653 |
cube | +1.636 |
Neg logit contributions | |
citiz | -2.030 |
eleph | -1.934 |
lation | -1.854 |
teasp | -1.779 |
TAIN | -1.765 |
Token | and |
Token ID | 290 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
tested | +1.464 |
expected | +1.415 |
coal | +1.402 |
isot | +1.334 |
cube | +1.321 |
Neg logit contributions | |
citiz | -1.903 |
eleph | -1.824 |
lation | -1.759 |
TAIN | -1.682 |
teasp | -1.677 |
Token | Smith |
Token ID | 4176 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
tested | +2.503 |
expected | +2.430 |
coal | +2.369 |
isot | +2.296 |
cube | +2.238 |
Neg logit contributions | |
citiz | -3.034 |
eleph | -2.943 |
lation | -2.798 |
teasp | -2.711 |
TAIN | -2.676 |
Token | did |
Token ID | 750 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
tested | +3.311 |
expected | +3.213 |
coal | +3.114 |
isot | +2.996 |
cube | +2.980 |
Neg logit contributions | |
citiz | -4.643 |
eleph | -4.453 |
lation | -4.251 |
teasp | -4.127 |
TAIN | -4.054 |
Token | just |
Token ID | 655 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
tested | +0.343 |
expected | +0.332 |
coal | +0.329 |
isot | +0.314 |
cube | +0.313 |
Neg logit contributions | |
citiz | -0.400 |
eleph | -0.380 |
lation | -0.368 |
TAIN | -0.349 |
teasp | -0.346 |
Token | score |
Token ID | 4776 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
citiz | +5.812 |
lation | +5.325 |
eleph | +5.134 |
TAIN | +4.864 |
utenant | +4.559 |
Neg logit contributions | |
tested | -8.696 |
coal | -8.509 |
expected | -8.440 |
cube | -8.095 |
cli | -8.081 |
Token | at |
Token ID | 379 |
Feature activation | +1.47e+01 |
Pos logit contributions | |
tested | +2.844 |
expected | +2.780 |
coal | +2.703 |
isot | +2.598 |
cube | +2.585 |
Neg logit contributions | |
citiz | -2.967 |
eleph | -2.881 |
lation | -2.714 |
teasp | -2.646 |
TAIN | -2.572 |
Token | least |
Token ID | 1551 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
citiz | +5.465 |
eleph | +4.615 |
lation | +4.555 |
TAIN | +4.274 |
teasp | +3.709 |
Neg logit contributions | |
tested | -14.921 |
coal | -14.712 |
cube | -14.189 |
isot | -14.079 |
expected | -14.038 |
Token | 25 |
Token ID | 1679 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
citiz | +6.326 |
lation | +5.649 |
TAIN | +5.324 |
eleph | +5.306 |
emphasis | +5.217 |
Neg logit contributions | |
tested | -11.511 |
coal | -11.259 |
expected | -10.845 |
cube | -10.831 |
cli | -10.747 |
Token | Fantasy |
Token ID | 10982 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
citiz | +0.639 |
eleph | +0.602 |
lation | +0.580 |
TAIN | +0.551 |
teasp | +0.539 |
Neg logit contributions | |
tested | -0.721 |
expected | -0.699 |
coal | -0.697 |
isot | -0.669 |
cube | -0.667 |
Token | points |
Token ID | 2173 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
citiz | +2.295 |
eleph | +2.120 |
lation | +2.097 |
TAIN | +1.981 |
teasp | +1.855 |
Neg logit contributions | |
tested | -3.144 |
coal | -3.068 |
expected | -3.062 |
isot | -2.930 |
cube | -2.919 |
Token | in |
Token ID | 287 |
Feature activation | -5.77e+00 |
Pos logit contributions | |
tested | +2.347 |
expected | +2.279 |
coal | +2.210 |
isot | +2.123 |
cube | +2.120 |
Neg logit contributions | |
citiz | -3.239 |
eleph | -3.107 |
lation | -2.988 |
teasp | -2.885 |
TAIN | -2.833 |
Token | on |
Token ID | 319 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
tested | +3.993 |
expected | +3.813 |
coal | +3.787 |
cube | +3.663 |
isot | +3.630 |
Neg logit contributions | |
citiz | -5.000 |
eleph | -4.831 |
lation | -4.530 |
teasp | -4.404 |
TAIN | -4.373 |
Token | the |
Token ID | 262 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
tested | +6.468 |
expected | +6.279 |
isot | +6.106 |
coal | +6.079 |
cube | +6.073 |
Neg logit contributions | |
citiz | -8.357 |
eleph | -8.203 |
lation | -7.618 |
teasp | -7.523 |
田 | -7.450 |
Token | home |
Token ID | 1363 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
tested | +2.378 |
expected | +2.328 |
coal | +2.300 |
isot | +2.231 |
cube | +2.216 |
Neg logit contributions | |
citiz | -2.077 |
eleph | -1.998 |
lation | -1.883 |
teasp | -1.808 |
TAIN | -1.758 |
Token | screen |
Token ID | 3159 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
tested | +6.968 |
coal | +6.783 |
expected | +6.702 |
isot | +6.571 |
cube | +6.546 |
Neg logit contributions | |
eleph | -7.466 |
citiz | -7.452 |
lation | -6.973 |
田 | -6.631 |
teasp | -6.602 |
Token | . |
Token ID | 13 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
tested | +0.386 |
expected | +0.373 |
coal | +0.370 |
cube | +0.353 |
isot | +0.352 |
Neg logit contributions | |
citiz | -0.462 |
eleph | -0.439 |
lation | -0.424 |
TAIN | -0.404 |
teasp | -0.400 |
Token | At |
Token ID | 1629 |
Feature activation | +1.76e+01 |
Pos logit contributions | |
citiz | +2.973 |
eleph | +2.803 |
lation | +2.777 |
TAIN | +2.645 |
teasp | +2.565 |
Neg logit contributions | |
tested | -2.531 |
expected | -2.454 |
coal | -2.440 |
isot | -2.346 |
cube | -2.294 |
Token | the |
Token ID | 262 |
Feature activation | +1.60e+01 |
Pos logit contributions | |
citiz | +11.133 |
lation | +10.485 |
eleph | +10.183 |
TAIN | +9.870 |
bottom | +9.404 |
Neg logit contributions | |
tested | -12.498 |
coal | -12.348 |
expected | -11.956 |
csv | -11.420 |
cli | -11.379 |
Token | moment |
Token ID | 2589 |
Feature activation | -9.36e-01 |
Pos logit contributions | |
citiz | +10.005 |
lation | +9.115 |
bottom | +9.047 |
eleph | +8.841 |
TAIN | +8.524 |
Neg logit contributions | |
tested | -11.742 |
coal | -11.565 |
expected | -11.264 |
cube | -10.770 |
isot | -10.689 |
Token | the |
Token ID | 262 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
tested | +0.640 |
expected | +0.615 |
coal | +0.609 |
cube | +0.584 |
isot | +0.582 |
Neg logit contributions | |
citiz | -0.794 |
eleph | -0.755 |
lation | -0.728 |
teasp | -0.694 |
TAIN | -0.693 |
Token | running |
Token ID | 2491 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
citiz | +2.674 |
eleph | +2.501 |
lation | +2.447 |
TAIN | +2.317 |
teasp | +2.240 |
Neg logit contributions | |
tested | -2.655 |
expected | -2.578 |
coal | -2.569 |
isot | -2.441 |
cube | -2.433 |
Token | Android |
Token ID | 5565 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
citiz | +3.309 |
eleph | +3.120 |
lation | +3.063 |
TAIN | +2.901 |
teasp | +2.793 |
Neg logit contributions | |
tested | -3.192 |
expected | -3.115 |
coal | -3.088 |
cube | -2.921 |
isot | -2.918 |
Token | implementing |
Token ID | 15427 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
citiz | +0.519 |
eleph | +0.492 |
lation | +0.476 |
TAIN | +0.451 |
teasp | +0.446 |
Neg logit contributions | |
tested | -0.478 |
expected | -0.463 |
coal | -0.460 |
isot | -0.439 |
cube | -0.439 |
Token | an |
Token ID | 281 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
citiz | +0.148 |
eleph | +0.141 |
lation | +0.136 |
TAIN | +0.129 |
teasp | +0.128 |
Neg logit contributions | |
tested | -0.134 |
expected | -0.130 |
coal | -0.129 |
isot | -0.123 |
cube | -0.123 |
Token | official |
Token ID | 1743 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
tested | +1.132 |
expected | +1.103 |
coal | +1.083 |
isot | +1.045 |
cube | +1.035 |
Neg logit contributions | |
citiz | -1.407 |
eleph | -1.332 |
lation | -1.283 |
TAIN | -1.233 |
teasp | -1.230 |
Token | version |
Token ID | 2196 |
Feature activation | +6.26e+00 |
Pos logit contributions | |
citiz | +3.048 |
eleph | +2.839 |
lation | +2.830 |
TAIN | +2.645 |
teasp | +2.558 |
Neg logit contributions | |
tested | -2.888 |
expected | -2.801 |
coal | -2.782 |
isot | -2.640 |
cube | -2.639 |
Token | of |
Token ID | 286 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
citiz | +4.808 |
lation | +4.562 |
eleph | +4.542 |
TAIN | +4.172 |
teasp | +4.113 |
Neg logit contributions | |
tested | -4.425 |
coal | -4.411 |
expected | -4.358 |
cube | -4.155 |
isot | -4.138 |
Token | this |
Token ID | 428 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
citiz | +1.027 |
eleph | +0.971 |
lation | +0.942 |
TAIN | +0.894 |
teasp | +0.879 |
Neg logit contributions | |
tested | -0.955 |
expected | -0.925 |
coal | -0.919 |
isot | -0.877 |
cube | -0.876 |
Token | mode |
Token ID | 4235 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
citiz | +1.369 |
eleph | +1.291 |
lation | +1.257 |
TAIN | +1.189 |
teasp | +1.168 |
Neg logit contributions | |
tested | -1.270 |
expected | -1.232 |
coal | -1.226 |
isot | -1.167 |
cube | -1.166 |
Token | , |
Token ID | 11 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
citiz | +0.962 |
eleph | +0.912 |
lation | +0.885 |
TAIN | +0.840 |
teasp | +0.829 |
Neg logit contributions | |
tested | -0.838 |
expected | -0.810 |
coal | -0.805 |
isot | -0.767 |
cube | -0.766 |
Token | although |
Token ID | 3584 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
citiz | +1.328 |
eleph | +1.258 |
lation | +1.226 |
TAIN | +1.163 |
teasp | +1.143 |
Neg logit contributions | |
tested | -1.137 |
expected | -1.099 |
coal | -1.096 |
cube | -1.044 |
isot | -1.040 |
Token | little |
Token ID | 1310 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
tested | +3.539 |
expected | +3.402 |
coal | +3.350 |
isot | +3.229 |
cube | +3.192 |
Neg logit contributions | |
citiz | -4.199 |
eleph | -4.162 |
lation | -3.855 |
teasp | -3.823 |
TAIN | -3.652 |
Token | has |
Token ID | 468 |
Feature activation | +6.28e+00 |
Pos logit contributions | |
citiz | +1.396 |
eleph | +1.322 |
lation | +1.290 |
TAIN | +1.226 |
teasp | +1.204 |
Neg logit contributions | |
tested | -1.122 |
expected | -1.085 |
coal | -1.077 |
cube | -1.023 |
isot | -1.020 |
Token | make |
Token ID | 787 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
citiz | +0.279 |
eleph | +0.265 |
lation | +0.255 |
TAIN | +0.241 |
teasp | +0.241 |
Neg logit contributions | |
tested | -0.258 |
expected | -0.250 |
coal | -0.248 |
isot | -0.238 |
cube | -0.237 |
Token | physical |
Token ID | 3518 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
tested | +0.890 |
expected | +0.860 |
coal | +0.853 |
isot | +0.823 |
cube | +0.813 |
Neg logit contributions | |
citiz | -1.050 |
eleph | -1.005 |
lation | -0.968 |
teasp | -0.925 |
TAIN | -0.916 |
Token | activity |
Token ID | 3842 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
tested | +1.906 |
expected | +1.852 |
coal | +1.842 |
isot | +1.801 |
cube | +1.780 |
Neg logit contributions | |
citiz | -1.246 |
eleph | -1.164 |
lation | -1.109 |
teasp | -1.034 |
TAIN | -1.026 |
Token | more |
Token ID | 517 |
Feature activation | +8.60e+00 |
Pos logit contributions | |
tested | +2.193 |
expected | +2.099 |
coal | +2.070 |
isot | +2.029 |
cube | +1.975 |
Neg logit contributions | |
citiz | -2.854 |
eleph | -2.757 |
lation | -2.618 |
teasp | -2.592 |
TAIN | -2.505 |
Token | accessible |
Token ID | 9857 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
citiz | +5.468 |
lation | +5.054 |
eleph | +4.861 |
TAIN | +4.626 |
emphasis | +4.547 |
Neg logit contributions | |
tested | -7.094 |
expected | -6.961 |
coal | -6.957 |
cube | -6.630 |
cli | -6.629 |
Token | . |
Token ID | 13 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
tested | +2.535 |
expected | +2.427 |
coal | +2.391 |
isot | +2.295 |
cube | +2.287 |
Neg logit contributions | |
citiz | -3.515 |
eleph | -3.374 |
lation | -3.252 |
teasp | -3.168 |
田 | -3.090 |
Token |
|
Token ID | 198 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
citiz | +2.840 |
lation | +2.674 |
eleph | +2.669 |
TAIN | +2.620 |
田 | +2.479 |
Neg logit contributions | |
tested | -2.418 |
expected | -2.363 |
coal | -2.344 |
isot | -2.280 |
cli | -2.215 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
tested | +0.603 |
expected | +0.580 |
coal | +0.575 |
cube | +0.546 |
isot | +0.540 |
Neg logit contributions | |
citiz | -0.878 |
eleph | -0.843 |
lation | -0.800 |
teasp | -0.773 |
TAIN | -0.762 |
Token | Work |
Token ID | 12468 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
citiz | +1.932 |
lation | +1.840 |
eleph | +1.791 |
TAIN | +1.770 |
田 | +1.696 |
Neg logit contributions | |
tested | -1.789 |
expected | -1.736 |
coal | -1.725 |
isot | -1.678 |
cube | -1.640 |
Token | it |
Token ID | 340 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
citiz | +1.978 |
eleph | +1.892 |
lation | +1.852 |
TAIN | +1.771 |
teasp | +1.744 |
Neg logit contributions | |
tested | -1.188 |
expected | -1.145 |
coal | -1.136 |
cube | -1.065 |
isot | -1.065 |
Token | Y |
Token ID | 575 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
tested | +1.765 |
expected | +1.687 |
coal | +1.673 |
cube | +1.603 |
isot | +1.589 |
Neg logit contributions | |
citiz | -2.043 |
eleph | -1.937 |
lation | -1.882 |
teasp | -1.781 |
TAIN | -1.771 |
Token | _ |
Token ID | 62 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
tested | +3.287 |
expected | +3.086 |
coal | +2.960 |
cube | +2.843 |
csv | +2.645 |
Neg logit contributions | |
eleph | -12.250 |
citiz | -12.188 |
teasp | -11.294 |
pione | -11.223 |
� | -11.181 |
Token | args |
Token ID | 22046 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
tested | +1.710 |
expected | +1.615 |
coal | +1.581 |
cube | +1.428 |
cli | +1.417 |
Neg logit contributions | |
citiz | -4.869 |
eleph | -4.781 |
teasp | -4.474 |
lation | -4.434 |
� | -4.311 |
Token | ()) |
Token ID | 28955 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
citiz | +2.263 |
lation | +2.255 |
eleph | +2.059 |
TAIN | +2.058 |
田 | +1.968 |
Neg logit contributions | |
tested | -2.864 |
coal | -2.787 |
expected | -2.773 |
isot | -2.731 |
cube | -2.644 |
Token | # |
Token ID | 1303 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
tested | +1.955 |
expected | +1.900 |
coal | +1.876 |
isot | +1.802 |
cube | +1.793 |
Neg logit contributions | |
citiz | -2.228 |
eleph | -2.117 |
lation | -2.058 |
TAIN | -1.945 |
teasp | -1.927 |
Token | initialize |
Token ID | 41216 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
citiz | +0.889 |
eleph | +0.839 |
lation | +0.807 |
TAIN | +0.757 |
teasp | +0.735 |
Neg logit contributions | |
tested | -1.157 |
expected | -1.124 |
coal | -1.121 |
cube | -1.078 |
isot | -1.071 |
Token | a |
Token ID | 257 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
tested | +4.777 |
expected | +4.640 |
cube | +4.442 |
coal | +4.433 |
isot | +4.359 |
Neg logit contributions | |
citiz | -8.991 |
eleph | -8.851 |
teasp | -8.231 |
� | -8.091 |
pione | -8.026 |
Token | rectangular |
Token ID | 36954 |
Feature activation | -6.43e+00 |
Pos logit contributions | |
citiz | +3.848 |
lation | +3.521 |
eleph | +3.436 |
TAIN | +3.299 |
utenant | +3.133 |
Neg logit contributions | |
tested | -4.376 |
expected | -4.233 |
coal | -4.197 |
crit | -3.957 |
cli | -3.933 |
Token | and |
Token ID | 290 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
tested | +3.552 |
expected | +3.421 |
isot | +3.399 |
coal | +3.361 |
cube | +3.325 |
Neg logit contributions | |
citiz | -6.400 |
eleph | -6.293 |
teasp | -5.893 |
pione | -5.730 |
� | -5.721 |
Token | square |
Token ID | 6616 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
citiz | +1.321 |
eleph | +1.229 |
lation | +1.212 |
TAIN | +1.143 |
teasp | +1.099 |
Neg logit contributions | |
tested | -1.452 |
expected | -1.410 |
coal | -1.406 |
isot | -1.327 |
cube | -1.326 |
Token | struct |
Token ID | 2878 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
citiz | +1.392 |
eleph | +1.276 |
lation | +1.270 |
TAIN | +1.190 |
teasp | +1.122 |
Neg logit contributions | |
tested | -1.826 |
expected | -1.777 |
coal | -1.768 |
cube | -1.687 |
isot | -1.671 |
Token | uring |
Token ID | 870 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
lation | +3.969 |
citiz | +3.514 |
TAIN | +3.467 |
acters | +3.108 |
emphasis | +2.832 |
Neg logit contributions | |
tested | -13.240 |
expected | -13.035 |
coal | -12.948 |
cli | -12.547 |
dissatisf | -12.437 |
Token | to |
Token ID | 284 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
tested | +2.605 |
expected | +2.528 |
coal | +2.485 |
isot | +2.425 |
cube | +2.354 |
Neg logit contributions | |
citiz | -3.208 |
eleph | -3.036 |
lation | -2.923 |
TAIN | -2.816 |
teasp | -2.807 |
Token | the |
Token ID | 262 |
Feature activation | -9.28e-01 |
Pos logit contributions | |
tested | +3.207 |
coal | +3.120 |
expected | +3.110 |
isot | +3.059 |
cube | +2.932 |
Neg logit contributions | |
citiz | -3.585 |
eleph | -3.410 |
lation | -3.250 |
teasp | -3.151 |
TAIN | -3.104 |
Token | deg |
Token ID | 3396 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
tested | +0.675 |
expected | +0.654 |
coal | +0.651 |
isot | +0.626 |
cube | +0.619 |
Neg logit contributions | |
citiz | -0.743 |
eleph | -0.706 |
lation | -0.680 |
TAIN | -0.647 |
teasp | -0.641 |
Token | assing |
Token ID | 19696 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
citiz | +2.006 |
lation | +1.907 |
eleph | +1.906 |
TAIN | +1.777 |
teasp | +1.748 |
Neg logit contributions | |
tested | -1.548 |
expected | -1.487 |
coal | -1.456 |
cube | -1.391 |
isot | -1.387 |
Token | from |
Token ID | 422 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
citiz | +0.475 |
eleph | +0.452 |
lation | +0.440 |
TAIN | +0.418 |
teasp | +0.412 |
Neg logit contributions | |
tested | -0.376 |
expected | -0.364 |
coal | -0.361 |
cube | -0.343 |
isot | -0.341 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
tested | +0.922 |
expected | +0.891 |
coal | +0.890 |
isot | +0.851 |
cube | +0.843 |
Neg logit contributions | |
citiz | -1.052 |
eleph | -0.999 |
lation | -0.964 |
TAIN | -0.917 |
teasp | -0.911 |
Token | ocean |
Token ID | 9151 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
citiz | +1.397 |
eleph | +1.316 |
lation | +1.289 |
TAIN | +1.215 |
teasp | +1.191 |
Neg logit contributions | |
tested | -1.231 |
expected | -1.191 |
coal | -1.176 |
cube | -1.128 |
cli | -1.110 |
Token | ic |
Token ID | 291 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
tested | +0.616 |
expected | +0.594 |
coal | +0.593 |
isot | +0.568 |
cube | +0.564 |
Neg logit contributions | |
citiz | -0.718 |
eleph | -0.680 |
lation | -0.653 |
TAIN | -0.623 |
teasp | -0.620 |
Token | up |
Token ID | 510 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
tested | +1.320 |
coal | +1.276 |
expected | +1.273 |
isot | +1.232 |
cube | +1.210 |
Neg logit contributions | |
citiz | -1.513 |
eleph | -1.444 |
lation | -1.380 |
teasp | -1.320 |
TAIN | -1.314 |
Token | well |
Token ID | 4053 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
tested | +0.732 |
expected | +0.713 |
coal | +0.709 |
isot | +0.683 |
cube | +0.682 |
Neg logit contributions | |
citiz | -0.506 |
eleph | -0.472 |
lation | -0.453 |
TAIN | -0.423 |
teasp | -0.415 |
Token | ings |
Token ID | 654 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
citiz | +2.385 |
lation | +2.249 |
eleph | +2.249 |
TAIN | +2.064 |
teasp | +1.999 |
Neg logit contributions | |
tested | -2.923 |
expected | -2.862 |
coal | -2.835 |
cube | -2.745 |
isot | -2.723 |
Token | similar |
Token ID | 2092 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
tested | +1.309 |
expected | +1.268 |
coal | +1.255 |
isot | +1.206 |
cube | +1.200 |
Neg logit contributions | |
citiz | -1.420 |
eleph | -1.350 |
lation | -1.294 |
TAIN | -1.232 |
teasp | -1.225 |
Token | range |
Token ID | 2837 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
citiz | +0.364 |
eleph | +0.344 |
lation | +0.332 |
TAIN | +0.314 |
teasp | +0.310 |
Neg logit contributions | |
tested | -0.374 |
expected | -0.362 |
coal | -0.360 |
isot | -0.344 |
cube | -0.344 |
Token | , |
Token ID | 11 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
tested | +1.482 |
expected | +1.430 |
coal | +1.400 |
isot | +1.337 |
cube | +1.331 |
Neg logit contributions | |
citiz | -2.004 |
eleph | -1.901 |
lation | -1.830 |
teasp | -1.756 |
TAIN | -1.752 |
Token | such |
Token ID | 884 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
tested | +1.834 |
expected | +1.782 |
coal | +1.752 |
isot | +1.680 |
cube | +1.650 |
Neg logit contributions | |
citiz | -2.284 |
eleph | -2.179 |
lation | -2.096 |
teasp | -2.007 |
TAIN | -2.006 |
Token | as |
Token ID | 355 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
citiz | +2.866 |
lation | +2.677 |
eleph | +2.639 |
TAIN | +2.486 |
田 | +2.388 |
Neg logit contributions | |
tested | -3.368 |
expected | -3.284 |
coal | -3.249 |
cube | -3.102 |
isot | -3.091 |
Token | Alpha |
Token ID | 12995 |
Feature activation | +6.71e+00 |
Pos logit contributions | |
citiz | +3.472 |
eleph | +3.269 |
lation | +3.195 |
TAIN | +3.012 |
田 | +2.937 |
Neg logit contributions | |
tested | -3.104 |
expected | -2.990 |
coal | -2.986 |
cube | -2.860 |
isot | -2.809 |
Token | Delta |
Token ID | 16978 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
citiz | +4.637 |
eleph | +4.299 |
lation | +4.246 |
TAIN | +3.937 |
田 | +3.862 |
Neg logit contributions | |
tested | -5.355 |
expected | -5.200 |
coal | -5.171 |
cli | -4.896 |
cube | -4.880 |
Token | Phi |
Token ID | 47256 |
Feature activation | +8.04e+00 |
Pos logit contributions | |
citiz | +7.383 |
lation | +6.999 |
eleph | +6.764 |
TAIN | +6.298 |
田 | +6.244 |
Neg logit contributions | |
tested | -7.287 |
expected | -7.029 |
coal | -6.919 |
cli | -6.618 |
cube | -6.593 |
Token | ($ |
Token ID | 7198 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
citiz | +6.156 |
eleph | +5.717 |
lation | +5.707 |
TAIN | +5.297 |
teasp | +5.203 |
Neg logit contributions | |
tested | -5.771 |
expected | -5.583 |
coal | -5.523 |
cube | -5.272 |
cli | -5.182 |
Token | 400 |
Token ID | 7029 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
citiz | +1.140 |
eleph | +1.082 |
lation | +1.078 |
TAIN | +1.031 |
田 | +0.998 |
Neg logit contributions | |
tested | -0.904 |
expected | -0.873 |
coal | -0.872 |
isot | -0.840 |
cube | -0.823 |
Token | per |
Token ID | 583 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
citiz | +0.967 |
eleph | +0.916 |
lation | +0.892 |
TAIN | +0.847 |
teasp | +0.834 |
Neg logit contributions | |
tested | -0.820 |
expected | -0.791 |
coal | -0.788 |
cube | -0.748 |
isot | -0.747 |
Token | of |
Token ID | 286 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
citiz | +1.749 |
eleph | +1.641 |
lation | +1.600 |
TAIN | +1.506 |
teasp | +1.461 |
Neg logit contributions | |
tested | -1.835 |
expected | -1.785 |
coal | -1.778 |
cube | -1.705 |
isot | -1.698 |
Token | the |
Token ID | 262 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
citiz | +0.991 |
eleph | +0.938 |
lation | +0.918 |
TAIN | +0.875 |
teasp | +0.847 |
Neg logit contributions | |
tested | -0.776 |
expected | -0.751 |
coal | -0.748 |
cube | -0.711 |
isot | -0.709 |
Token | commenters |
Token ID | 43807 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
citiz | +2.790 |
eleph | +2.599 |
lation | +2.557 |
TAIN | +2.393 |
teasp | +2.300 |
Neg logit contributions | |
tested | -2.948 |
expected | -2.874 |
coal | -2.862 |
cube | -2.746 |
isot | -2.720 |
Token | see |
Token ID | 766 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
tested | +1.480 |
expected | +1.433 |
coal | +1.411 |
isot | +1.356 |
cube | +1.350 |
Neg logit contributions | |
citiz | -1.622 |
eleph | -1.548 |
lation | -1.475 |
teasp | -1.419 |
TAIN | -1.411 |
Token | the |
Token ID | 262 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
tested | +0.432 |
expected | +0.419 |
coal | +0.416 |
isot | +0.398 |
cube | +0.397 |
Neg logit contributions | |
citiz | -0.441 |
eleph | -0.417 |
lation | -0.403 |
TAIN | -0.382 |
teasp | -0.378 |
Token | video |
Token ID | 2008 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
citiz | +0.399 |
eleph | +0.377 |
lation | +0.366 |
TAIN | +0.346 |
teasp | +0.341 |
Neg logit contributions | |
tested | -0.389 |
expected | -0.377 |
coal | -0.375 |
cube | -0.358 |
isot | -0.358 |
Token | as |
Token ID | 355 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
tested | +1.061 |
expected | +1.023 |
coal | +1.013 |
isot | +0.972 |
cube | +0.972 |
Neg logit contributions | |
citiz | -1.146 |
eleph | -1.090 |
lation | -1.045 |
teasp | -0.997 |
TAIN | -0.997 |
Token | a |
Token ID | 257 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
tested | +1.842 |
expected | +1.782 |
coal | +1.761 |
isot | +1.701 |
cube | +1.676 |
Neg logit contributions | |
citiz | -1.999 |
eleph | -1.902 |
lation | -1.819 |
TAIN | -1.753 |
teasp | -1.731 |
Token | cheap |
Token ID | 7026 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
tested | +2.634 |
coal | +2.510 |
expected | +2.510 |
isot | +2.410 |
cube | +2.373 |
Neg logit contributions | |
citiz | -2.836 |
eleph | -2.727 |
lation | -2.561 |
TAIN | -2.470 |
teasp | -2.466 |
Token | attempt |
Token ID | 2230 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
tested | +3.215 |
coal | +3.078 |
expected | +3.056 |
isot | +2.928 |
cube | +2.906 |
Neg logit contributions | |
citiz | -3.601 |
eleph | -3.427 |
lation | -3.231 |
TAIN | -3.147 |
teasp | -3.092 |
Token | to |
Token ID | 284 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
citiz | +1.384 |
eleph | +1.311 |
lation | +1.278 |
TAIN | +1.205 |
teasp | +1.189 |
Neg logit contributions | |
tested | -1.303 |
expected | -1.267 |
coal | -1.259 |
cube | -1.198 |
isot | -1.196 |
Token | ability |
Token ID | 2694 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
citiz | +1.238 |
eleph | +1.154 |
lation | +1.119 |
TAIN | +1.046 |
teasp | +1.016 |
Neg logit contributions | |
tested | -1.504 |
coal | -1.461 |
expected | -1.460 |
cube | -1.398 |
isot | -1.389 |
Token | to |
Token ID | 284 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
tested | +3.668 |
expected | +3.534 |
coal | +3.471 |
isot | +3.345 |
cube | +3.325 |
Neg logit contributions | |
citiz | -4.580 |
eleph | -4.341 |
lation | -4.217 |
TAIN | -4.023 |
teasp | -4.018 |
Token | make |
Token ID | 787 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
tested | +2.564 |
expected | +2.499 |
coal | +2.456 |
isot | +2.400 |
cube | +2.369 |
Neg logit contributions | |
citiz | -2.966 |
eleph | -2.833 |
lation | -2.673 |
teasp | -2.651 |
TAIN | -2.538 |
Token | memorable |
Token ID | 18078 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
tested | +4.081 |
expected | +3.943 |
coal | +3.883 |
isot | +3.853 |
shocks | +3.697 |
Neg logit contributions | |
citiz | -4.776 |
eleph | -4.611 |
lation | -4.391 |
teasp | -4.391 |
田 | -4.145 |
Token | phrases |
Token ID | 20144 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
tested | +3.064 |
expected | +2.954 |
coal | +2.939 |
isot | +2.886 |
cube | +2.771 |
Neg logit contributions | |
citiz | -3.437 |
eleph | -3.297 |
lation | -3.138 |
teasp | -3.096 |
TAIN | -2.993 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
tested | +0.510 |
expected | +0.493 |
coal | +0.489 |
isot | +0.467 |
cube | +0.466 |
Neg logit contributions | |
citiz | -0.621 |
eleph | -0.590 |
lation | -0.570 |
TAIN | -0.544 |
teasp | -0.540 |
Token | She |
Token ID | 1375 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
tested | +0.982 |
expected | +0.941 |
coal | +0.935 |
cube | +0.888 |
isot | +0.879 |
Neg logit contributions | |
citiz | -1.349 |
eleph | -1.293 |
lation | -1.227 |
teasp | -1.189 |
TAIN | -1.170 |
Token | offers |
Token ID | 4394 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
citiz | +1.051 |
eleph | +1.000 |
lation | +0.975 |
TAIN | +0.924 |
teasp | +0.910 |
Neg logit contributions | |
tested | -0.846 |
expected | -0.817 |
coal | -0.812 |
cube | -0.772 |
isot | -0.769 |
Token | the |
Token ID | 262 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
tested | +3.470 |
expected | +3.317 |
coal | +3.302 |
isot | +3.244 |
cube | +3.120 |
Neg logit contributions | |
citiz | -3.988 |
eleph | -3.807 |
lation | -3.644 |
teasp | -3.574 |
TAIN | -3.507 |
Token | grat |
Token ID | 14586 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
tested | +1.543 |
expected | +1.492 |
coal | +1.479 |
isot | +1.424 |
cube | +1.408 |
Neg logit contributions | |
citiz | -1.519 |
eleph | -1.438 |
lation | -1.382 |
TAIN | -1.319 |
teasp | -1.310 |
Token | ifications |
Token ID | 6637 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
tested | +1.370 |
expected | +1.336 |
coal | +1.333 |
isot | +1.288 |
cube | +1.286 |
Neg logit contributions | |
citiz | -0.627 |
eleph | -0.575 |
lation | -0.535 |
TAIN | -0.489 |
teasp | -0.484 |
Token | . |
Token ID | 13 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
tested | +3.847 |
expected | +3.713 |
coal | +3.656 |
isot | +3.458 |
cube | +3.402 |
Neg logit contributions | |
citiz | -5.604 |
eleph | -5.366 |
lation | -5.116 |
teasp | -5.022 |
TAIN | -4.924 |
Token | It |
Token ID | 632 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
tested | +1.299 |
expected | +1.252 |
coal | +1.240 |
cube | +1.177 |
isot | +1.163 |
Neg logit contributions | |
citiz | -1.715 |
eleph | -1.647 |
lation | -1.559 |
teasp | -1.514 |
TAIN | -1.479 |
Token | 's |
Token ID | 338 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
tested | +0.139 |
expected | +0.135 |
coal | +0.134 |
isot | +0.127 |
cube | +0.127 |
Neg logit contributions | |
citiz | -0.179 |
eleph | -0.171 |
lation | -0.166 |
TAIN | -0.158 |
teasp | -0.156 |
Token | investing |
Token ID | 14771 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
tested | +2.879 |
expected | +2.806 |
coal | +2.717 |
isot | +2.632 |
cube | +2.584 |
Neg logit contributions | |
citiz | -3.432 |
eleph | -3.279 |
lation | -3.107 |
teasp | -3.055 |
TAIN | -2.999 |
Token | in |
Token ID | 287 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
tested | +0.958 |
expected | +0.926 |
coal | +0.913 |
isot | +0.876 |
cube | +0.862 |
Neg logit contributions | |
citiz | -1.256 |
eleph | -1.198 |
lation | -1.155 |
TAIN | -1.106 |
teasp | -1.105 |
Token | Firefox |
Token ID | 16802 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
tested | +1.891 |
expected | +1.828 |
coal | +1.801 |
isot | +1.749 |
cube | +1.714 |
Neg logit contributions | |
citiz | -2.123 |
eleph | -2.041 |
lation | -1.946 |
teasp | -1.882 |
TAIN | -1.857 |
Token | for |
Token ID | 329 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
tested | +0.339 |
expected | +0.327 |
coal | +0.324 |
isot | +0.311 |
cube | +0.310 |
Neg logit contributions | |
citiz | -0.401 |
eleph | -0.382 |
lation | -0.370 |
TAIN | -0.350 |
teasp | -0.349 |
Token | phones |
Token ID | 9512 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
tested | +2.497 |
expected | +2.408 |
coal | +2.344 |
isot | +2.285 |
cube | +2.259 |
Neg logit contributions | |
citiz | -3.046 |
eleph | -2.939 |
lation | -2.788 |
teasp | -2.720 |
TAIN | -2.654 |
Token | and |
Token ID | 290 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
tested | +0.424 |
expected | +0.409 |
coal | +0.405 |
isot | +0.386 |
cube | +0.386 |
Neg logit contributions | |
citiz | -0.528 |
eleph | -0.502 |
lation | -0.486 |
TAIN | -0.464 |
teasp | -0.460 |
Token | tablets |
Token ID | 17255 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
tested | +1.149 |
expected | +1.086 |
coal | +1.061 |
isot | +1.009 |
cube | +0.999 |
Neg logit contributions | |
citiz | -2.477 |
eleph | -2.403 |
lation | -2.329 |
teasp | -2.245 |
TAIN | -2.233 |
Token | powered |
Token ID | 13232 |
Feature activation | -6.07e+00 |
Pos logit contributions | |
tested | +1.466 |
expected | +1.418 |
coal | +1.399 |
isot | +1.333 |
cube | +1.332 |
Neg logit contributions | |
citiz | -1.878 |
eleph | -1.790 |
lation | -1.725 |
TAIN | -1.650 |
teasp | -1.648 |
Token | orious |
Token ID | 9982 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
tested | +3.759 |
expected | +3.618 |
coal | +3.555 |
isot | +3.392 |
cube | +3.379 |
Neg logit contributions | |
citiz | -5.053 |
eleph | -4.866 |
lation | -4.615 |
teasp | -4.460 |
TAIN | -4.421 |
Token | � |
Token ID | 447 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
citiz | +2.603 |
eleph | +2.490 |
lation | +2.407 |
TAIN | +2.295 |
teasp | +2.253 |
Neg logit contributions | |
tested | -1.969 |
expected | -1.912 |
coal | -1.897 |
isot | -1.808 |
cube | -1.801 |
Token | � |
Token ID | 251 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
tested | +0.979 |
expected | +0.948 |
coal | +0.943 |
cube | +0.896 |
isot | +0.894 |
Neg logit contributions | |
citiz | -0.959 |
eleph | -0.918 |
lation | -0.861 |
teasp | -0.822 |
TAIN | -0.821 |
Token | Conor |
Token ID | 41708 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
citiz | +0.602 |
eleph | +0.573 |
lation | +0.557 |
TAIN | +0.530 |
teasp | +0.523 |
Neg logit contributions | |
tested | -0.464 |
expected | -0.448 |
coal | -0.445 |
isot | -0.423 |
cube | -0.422 |
Token | McGregor |
Token ID | 32777 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
tested | +0.323 |
expected | +0.315 |
coal | +0.310 |
isot | +0.296 |
cube | +0.295 |
Neg logit contributions | |
citiz | -0.325 |
eleph | -0.307 |
lation | -0.297 |
TAIN | -0.281 |
teasp | -0.278 |
Token | challenged |
Token ID | 12827 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
tested | +1.064 |
expected | +1.029 |
coal | +1.014 |
isot | +0.969 |
cube | +0.964 |
Neg logit contributions | |
citiz | -1.353 |
eleph | -1.291 |
lation | -1.250 |
TAIN | -1.187 |
teasp | -1.184 |
Token | Floyd |
Token ID | 21150 |
Feature activation | +6.71e+00 |
Pos logit contributions | |
tested | +3.178 |
expected | +3.080 |
coal | +3.004 |
isot | +2.916 |
cube | +2.864 |
Neg logit contributions | |
citiz | -3.486 |
eleph | -3.348 |
lation | -3.199 |
teasp | -3.110 |
TAIN | -3.005 |
Token | � |
Token ID | 564 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
citiz | +5.242 |
eleph | +4.872 |
lation | +4.746 |
TAIN | +4.575 |
teasp | +4.427 |
Neg logit contributions | |
tested | -4.785 |
coal | -4.605 |
expected | -4.570 |
cube | -4.415 |
isot | -4.337 |
Token | � |
Token ID | 250 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
tested | +2.458 |
expected | +2.386 |
coal | +2.376 |
cube | +2.238 |
isot | +2.232 |
Neg logit contributions | |
citiz | -2.521 |
eleph | -2.399 |
lation | -2.273 |
teasp | -2.154 |
TAIN | -2.145 |
Token | Money |
Token ID | 26788 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
citiz | +2.813 |
eleph | +2.653 |
lation | +2.640 |
TAIN | +2.514 |
田 | +2.444 |
Neg logit contributions | |
tested | -2.357 |
coal | -2.291 |
expected | -2.270 |
isot | -2.187 |
cube | -2.167 |
Token | � |
Token ID | 447 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
tested | +6.507 |
expected | +6.231 |
coal | +6.085 |
cli | +5.993 |
cube | +5.921 |
Neg logit contributions | |
citiz | -8.864 |
eleph | -8.566 |
lation | -7.969 |
TAIN | -7.672 |
teasp | -7.652 |
Token | s |
Token ID | 82 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
tested | +1.464 |
expected | +1.380 |
coal | +1.362 |
cube | +1.268 |
isot | +1.249 |
Neg logit contributions | |
citiz | -3.373 |
eleph | -3.263 |
lation | -3.127 |
teasp | -3.042 |
TAIN | -3.029 |
Token | method |
Token ID | 2446 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
tested | +1.429 |
expected | +1.389 |
coal | +1.380 |
isot | +1.333 |
cube | +1.322 |
Neg logit contributions | |
citiz | -1.294 |
eleph | -1.239 |
lation | -1.182 |
teasp | -1.123 |
TAIN | -1.120 |
Token | ical |
Token ID | 605 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
citiz | +3.231 |
eleph | +3.057 |
lation | +3.005 |
TAIN | +2.836 |
teasp | +2.759 |
Neg logit contributions | |
tested | -2.829 |
expected | -2.770 |
coal | -2.741 |
isot | -2.602 |
cube | -2.597 |
Token | grapple |
Token ID | 42146 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
citiz | +0.977 |
eleph | +0.918 |
lation | +0.894 |
TAIN | +0.844 |
teasp | +0.826 |
Neg logit contributions | |
tested | -0.967 |
expected | -0.939 |
coal | -0.932 |
cube | -0.890 |
isot | -0.886 |
Token | style |
Token ID | 3918 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
citiz | +2.838 |
eleph | +2.703 |
lation | +2.677 |
TAIN | +2.511 |
teasp | +2.455 |
Neg logit contributions | |
tested | -2.239 |
expected | -2.177 |
coal | -2.155 |
cube | -2.054 |
isot | -2.039 |
Token | and |
Token ID | 290 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
citiz | +1.900 |
eleph | +1.800 |
lation | +1.764 |
TAIN | +1.663 |
teasp | +1.621 |
Neg logit contributions | |
tested | -1.625 |
expected | -1.575 |
coal | -1.560 |
cube | -1.491 |
isot | -1.480 |
Token | quick |
Token ID | 2068 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
citiz | +0.759 |
eleph | +0.717 |
lation | +0.697 |
TAIN | +0.659 |
teasp | +0.647 |
Neg logit contributions | |
tested | -0.719 |
expected | -0.697 |
coal | -0.694 |
cube | -0.662 |
isot | -0.659 |
Token | movement |
Token ID | 3356 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
tested | +2.474 |
expected | +2.397 |
coal | +2.369 |
isot | +2.291 |
cube | +2.271 |
Neg logit contributions | |
citiz | -2.789 |
eleph | -2.666 |
lation | -2.533 |
teasp | -2.428 |
TAIN | -2.410 |
Token | will |
Token ID | 481 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
tested | +1.535 |
expected | +1.481 |
coal | +1.480 |
isot | +1.423 |
cube | +1.405 |
Neg logit contributions | |
citiz | -1.790 |
eleph | -1.711 |
lation | -1.618 |
teasp | -1.576 |
TAIN | -1.558 |
Token | make |
Token ID | 787 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
citiz | +1.034 |
eleph | +0.976 |
lation | +0.953 |
TAIN | +0.902 |
teasp | +0.878 |
Neg logit contributions | |
tested | -1.008 |
expected | -0.977 |
coal | -0.973 |
cube | -0.928 |
isot | -0.922 |
Token | her |
Token ID | 607 |
Feature activation | -7.83e-01 |
Pos logit contributions | |
tested | +1.290 |
expected | +1.242 |
coal | +1.233 |
isot | +1.176 |
cube | +1.168 |
Neg logit contributions | |
citiz | -1.629 |
eleph | -1.552 |
lation | -1.504 |
TAIN | -1.430 |
teasp | -1.427 |
Token | this |
Token ID | 428 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
citiz | +2.870 |
eleph | +2.665 |
lation | +2.622 |
TAIN | +2.458 |
田 | +2.383 |
Neg logit contributions | |
tested | -3.031 |
coal | -2.943 |
expected | -2.924 |
cube | -2.831 |
cli | -2.759 |
Token | is |
Token ID | 318 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
tested | +0.528 |
expected | +0.513 |
coal | +0.507 |
isot | +0.488 |
cube | +0.483 |
Neg logit contributions | |
citiz | -0.551 |
eleph | -0.524 |
lation | -0.506 |
TAIN | -0.480 |
teasp | -0.475 |
Token | now |
Token ID | 783 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
tested | +2.758 |
expected | +2.681 |
coal | +2.622 |
isot | +2.527 |
cube | +2.476 |
Neg logit contributions | |
citiz | -3.179 |
eleph | -3.031 |
lation | -2.895 |
TAIN | -2.774 |
teasp | -2.759 |
Token | three |
Token ID | 1115 |
Feature activation | -8.06e+00 |
Pos logit contributions | |
tested | +1.950 |
expected | +1.896 |
coal | +1.865 |
isot | +1.790 |
cube | +1.761 |
Neg logit contributions | |
citiz | -2.219 |
eleph | -2.109 |
lation | -2.018 |
TAIN | -1.932 |
teasp | -1.927 |
Token | games |
Token ID | 1830 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
tested | +5.669 |
expected | +5.417 |
isot | +5.221 |
coal | +5.169 |
shocks | +4.947 |
Neg logit contributions | |
citiz | -6.520 |
eleph | -6.286 |
lation | -5.956 |
teasp | -5.701 |
田 | -5.676 |
Token | on |
Token ID | 319 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
tested | +3.752 |
expected | +3.577 |
coal | +3.476 |
isot | +3.422 |
cube | +3.358 |
Neg logit contributions | |
citiz | -4.629 |
eleph | -4.394 |
lation | -4.205 |
teasp | -4.096 |
TAIN | -4.008 |
Token | the |
Token ID | 262 |
Feature activation | -5.55e+00 |
Pos logit contributions | |
tested | +6.481 |
expected | +6.284 |
isot | +6.008 |
coal | +5.963 |
cube | +5.759 |
Neg logit contributions | |
citiz | -9.514 |
eleph | -9.187 |
lation | -8.931 |
teasp | -8.737 |
田 | -8.543 |
Token | road |
Token ID | 2975 |
Feature activation | -7.75e+00 |
Pos logit contributions | |
tested | +3.444 |
expected | +3.299 |
coal | +3.192 |
isot | +3.162 |
cube | +3.076 |
Neg logit contributions | |
citiz | -4.985 |
eleph | -4.773 |
lation | -4.600 |
TAIN | -4.490 |
teasp | -4.487 |
Token | with |
Token ID | 351 |
Feature activation | -6.61e+00 |
Pos logit contributions | |
tested | +4.773 |
expected | +4.540 |
coal | +4.354 |
isot | +4.263 |
cube | +4.150 |
Neg logit contributions | |
citiz | -6.955 |
eleph | -6.619 |
lation | -6.423 |
teasp | -6.275 |
TAIN | -6.123 |
Token | less |
Token ID | 1342 |
Feature activation | +5.77e+00 |
Pos logit contributions | |
tested | +4.612 |
expected | +4.463 |
coal | +4.330 |
isot | +4.282 |
cube | +4.114 |
Neg logit contributions | |
citiz | -5.381 |
eleph | -5.229 |
lation | -4.930 |
teasp | -4.861 |
TAIN | -4.705 |
Token | than |
Token ID | 621 |
Feature activation | +4.29e+00 |
Pos logit contributions | |
citiz | +2.934 |
eleph | +2.684 |
lation | +2.574 |
TAIN | +2.451 |
emphasis | +2.246 |
Neg logit contributions | |
tested | -5.618 |
coal | -5.540 |
expected | -5.470 |
cube | -5.316 |
cli | -5.294 |
Token | ( |
Token ID | 357 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
tested | +2.666 |
coal | +2.617 |
expected | +2.575 |
cube | +2.471 |
isot | +2.469 |
Neg logit contributions | |
citiz | -3.784 |
eleph | -3.587 |
lation | -3.454 |
TAIN | -3.344 |
teasp | -3.305 |
Token | k |
Token ID | 74 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
tested | +0.738 |
expected | +0.712 |
coal | +0.710 |
cube | +0.669 |
isot | +0.666 |
Neg logit contributions | |
citiz | -1.058 |
eleph | -1.012 |
lation | -0.973 |
TAIN | -0.932 |
teasp | -0.927 |
Token | ö |
Token ID | 9101 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
tested | +1.141 |
expected | +1.102 |
coal | +1.098 |
cube | +1.038 |
isot | +1.035 |
Neg logit contributions | |
citiz | -1.500 |
eleph | -1.428 |
lation | -1.374 |
TAIN | -1.315 |
teasp | -1.305 |
Token | ke |
Token ID | 365 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
citiz | +3.507 |
lation | +3.380 |
eleph | +3.321 |
TAIN | +3.153 |
utenant | +3.065 |
Neg logit contributions | |
tested | -2.448 |
expected | -2.351 |
coal | -2.297 |
isot | -2.220 |
cube | -2.195 |
Token | ), |
Token ID | 828 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
citiz | +2.110 |
eleph | +2.004 |
lation | +1.968 |
TAIN | +1.852 |
teasp | +1.819 |
Neg logit contributions | |
tested | -1.824 |
expected | -1.761 |
coal | -1.735 |
cube | -1.655 |
isot | -1.654 |
Token | white |
Token ID | 2330 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
citiz | +1.585 |
eleph | +1.497 |
lation | +1.449 |
TAIN | +1.368 |
teasp | +1.350 |
Neg logit contributions | |
tested | -1.518 |
expected | -1.471 |
coal | -1.451 |
cube | -1.390 |
isot | -1.386 |
Token | ( |
Token ID | 357 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
coal | +5.166 |
tested | +5.098 |
expected | +4.901 |
cube | +4.881 |
isot | +4.835 |
Neg logit contributions | |
citiz | -8.953 |
eleph | -8.535 |
lation | -8.255 |
TAIN | -8.158 |
teasp | -7.905 |
Token | ch |
Token ID | 354 |
Feature activation | -6.57e+00 |
Pos logit contributions | |
tested | +1.458 |
coal | +1.407 |
expected | +1.405 |
cube | +1.323 |
isot | +1.308 |
Neg logit contributions | |
citiz | -2.219 |
eleph | -2.127 |
lation | -2.029 |
TAIN | -1.954 |
teasp | -1.948 |
Token | ag |
Token ID | 363 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
tested | +4.176 |
coal | +4.137 |
expected | +4.065 |
isot | +3.809 |
cube | +3.809 |
Neg logit contributions | |
citiz | -5.778 |
eleph | -5.502 |
lation | -5.100 |
teasp | -5.042 |
TAIN | -5.038 |
Token | ha |
Token ID | 3099 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
tested | +2.881 |
coal | +2.818 |
expected | +2.791 |
cube | +2.674 |
isot | +2.640 |
Neg logit contributions | |
citiz | -3.731 |
eleph | -3.574 |
lation | -3.354 |
teasp | -3.280 |
TAIN | -3.257 |
Token | � |
Token ID | 447 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
citiz | +1.012 |
eleph | +0.962 |
lation | +0.936 |
TAIN | +0.885 |
teasp | +0.874 |
Neg logit contributions | |
tested | -0.867 |
expected | -0.837 |
coal | -0.829 |
cube | -0.791 |
isot | -0.789 |
Token | contact |
Token ID | 2800 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
tested | +4.003 |
expected | +3.890 |
coal | +3.792 |
isot | +3.663 |
cube | +3.587 |
Neg logit contributions | |
citiz | -6.124 |
eleph | -5.937 |
teasp | -5.679 |
lation | -5.502 |
� | -5.368 |
Token | Wh |
Token ID | 854 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
citiz | +0.091 |
eleph | +0.086 |
lation | +0.083 |
TAIN | +0.079 |
teasp | +0.078 |
Neg logit contributions | |
tested | -0.081 |
expected | -0.078 |
coal | -0.077 |
isot | -0.074 |
cube | -0.074 |
Token | itt |
Token ID | 715 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
tested | +2.034 |
expected | +1.933 |
coal | +1.928 |
cube | +1.762 |
isot | +1.752 |
Neg logit contributions | |
citiz | -4.893 |
eleph | -4.856 |
lation | -4.480 |
teasp | -4.431 |
TAIN | -4.353 |
Token | les |
Token ID | 829 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
citiz | +2.603 |
lation | +2.511 |
TAIN | +2.339 |
eleph | +2.310 |
田 | +2.220 |
Neg logit contributions | |
tested | -2.968 |
expected | -2.925 |
coal | -2.845 |
isot | -2.822 |
cube | -2.737 |
Token | ea |
Token ID | 18213 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
citiz | +4.167 |
lation | +4.158 |
eleph | +3.937 |
TAIN | +3.883 |
田 | +3.846 |
Neg logit contributions | |
tested | -2.020 |
expected | -1.884 |
coal | -1.845 |
isot | -1.790 |
cube | -1.679 |
Token | Community |
Token ID | 8108 |
Feature activation | -1.21e+01 |
Pos logit contributions | |
citiz | +2.370 |
eleph | +2.202 |
lation | +2.149 |
TAIN | +2.032 |
田 | +1.981 |
Neg logit contributions | |
tested | -2.250 |
expected | -2.194 |
coal | -2.158 |
isot | -2.077 |
cube | -2.066 |
Token | Connect |
Token ID | 8113 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
coal | +6.481 |
tested | +6.403 |
expected | +6.051 |
cube | +5.958 |
cli | +5.701 |
Neg logit contributions | |
citiz | -10.901 |
eleph | -10.691 |
lation | -10.161 |
teasp | -9.824 |
TAIN | -9.724 |
Token | ions |
Token ID | 507 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
citiz | +2.011 |
lation | +1.923 |
eleph | +1.877 |
TAIN | +1.797 |
田 | +1.729 |
Neg logit contributions | |
tested | -1.751 |
expected | -1.700 |
coal | -1.666 |
isot | -1.623 |
cube | -1.601 |
Token | ' |
Token ID | 6 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
citiz | +0.740 |
eleph | +0.699 |
lation | +0.678 |
TAIN | +0.643 |
teasp | +0.631 |
Neg logit contributions | |
tested | -0.690 |
expected | -0.671 |
coal | -0.665 |
isot | -0.635 |
cube | -0.634 |
Token | Emergency |
Token ID | 18154 |
Feature activation | -5.56e+00 |
Pos logit contributions | |
citiz | +0.860 |
eleph | +0.812 |
lation | +0.788 |
TAIN | +0.747 |
teasp | +0.734 |
Neg logit contributions | |
tested | -0.773 |
expected | -0.749 |
coal | -0.743 |
cube | -0.709 |
isot | -0.709 |
Token | Relief |
Token ID | 30323 |
Feature activation | -6.02e+00 |
Pos logit contributions | |
tested | +3.241 |
coal | +3.157 |
expected | +3.121 |
isot | +2.942 |
cube | +2.915 |
Neg logit contributions | |
citiz | -5.160 |
eleph | -4.973 |
lation | -4.740 |
teasp | -4.612 |
TAIN | -4.569 |
Token | . |
Token ID | 13 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
citiz | +0.650 |
eleph | +0.619 |
lation | +0.601 |
TAIN | +0.572 |
teasp | +0.564 |
Neg logit contributions | |
tested | -0.514 |
expected | -0.496 |
coal | -0.492 |
isot | -0.468 |
cube | -0.467 |
Token |
|
Token ID | 198 |
Feature activation | -6.77e+00 |
Pos logit contributions | |
tested | +2.213 |
expected | +2.149 |
coal | +2.128 |
cube | +2.006 |
isot | +1.961 |
Neg logit contributions | |
citiz | -3.218 |
eleph | -3.078 |
lation | -2.924 |
teasp | -2.804 |
TAIN | -2.746 |
Token |
|
Token ID | 198 |
Feature activation | -5.40e+00 |
Pos logit contributions | |
tested | +3.560 |
expected | +3.364 |
coal | +3.251 |
cube | +3.208 |
cli | +2.990 |
Neg logit contributions | |
eleph | -7.554 |
citiz | -7.551 |
teasp | -6.946 |
� | -6.774 |
pione | -6.765 |
Token | The |
Token ID | 464 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
tested | +3.602 |
expected | +3.519 |
coal | +3.479 |
cube | +3.290 |
cli | +3.219 |
Neg logit contributions | |
citiz | -4.887 |
eleph | -4.783 |
teasp | -4.296 |
lation | -4.250 |
� | -4.055 |
Token | move |
Token ID | 1445 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
tested | +3.336 |
expected | +3.280 |
coal | +3.232 |
isot | +3.103 |
cube | +3.061 |
Neg logit contributions | |
citiz | -3.946 |
eleph | -3.809 |
lation | -3.609 |
teasp | -3.477 |
TAIN | -3.466 |
Token | followed |
Token ID | 3940 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
tested | +2.156 |
expected | +2.106 |
coal | +2.072 |
isot | +1.996 |
cube | +1.965 |
Neg logit contributions | |
citiz | -2.513 |
eleph | -2.422 |
lation | -2.297 |
teasp | -2.228 |
TAIN | -2.194 |
Token | criticism |
Token ID | 7734 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
expected | +6.858 |
tested | +6.788 |
coal | +6.578 |
isot | +6.397 |
shocks | +6.345 |
Neg logit contributions | |
citiz | -9.475 |
eleph | -9.195 |
lation | -8.529 |
teasp | -8.514 |
TAIN | -8.211 |
Token | from |
Token ID | 422 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
tested | +4.131 |
expected | +4.040 |
coal | +3.936 |
isot | +3.778 |
cube | +3.741 |
Neg logit contributions | |
citiz | -4.731 |
eleph | -4.591 |
lation | -4.343 |
TAIN | -4.190 |
teasp | -4.189 |
Token | pro |
Token ID | 386 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
tested | +2.279 |
expected | +2.234 |
coal | +2.210 |
isot | +2.109 |
cube | +2.075 |
Neg logit contributions | |
citiz | -2.606 |
eleph | -2.499 |
lation | -2.397 |
teasp | -2.298 |
TAIN | -2.275 |
Token | - |
Token ID | 12 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
tested | +0.698 |
expected | +0.665 |
coal | +0.661 |
isot | +0.610 |
cube | +0.605 |
Neg logit contributions | |
citiz | -1.571 |
eleph | -1.516 |
lation | -1.461 |
teasp | -1.409 |
TAIN | -1.406 |
Token | Brexit |
Token ID | 40730 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
tested | +1.005 |
expected | +0.970 |
coal | +0.970 |
cube | +0.900 |
isot | +0.895 |
Neg logit contributions | |
citiz | -1.660 |
eleph | -1.593 |
lation | -1.539 |
TAIN | -1.475 |
teasp | -1.474 |
Token | glad |
Token ID | 9675 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
citiz | +2.583 |
eleph | +2.423 |
lation | +2.383 |
TAIN | +2.233 |
teasp | +2.157 |
Neg logit contributions | |
tested | -3.031 |
expected | -2.946 |
coal | -2.929 |
isot | -2.819 |
cube | -2.814 |
Token | nan |
Token ID | 15709 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
citiz | +6.357 |
lation | +5.741 |
eleph | +5.435 |
TAIN | +5.148 |
utenant | +5.051 |
Neg logit contributions | |
tested | -8.784 |
expected | -8.684 |
coal | -8.555 |
Reviewer | -7.996 |
forced | -7.948 |
Token | ocl |
Token ID | 38679 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
tested | +3.422 |
coal | +3.329 |
expected | +3.315 |
cli | +3.128 |
cube | +3.107 |
Neg logit contributions | |
citiz | -6.654 |
eleph | -6.541 |
teasp | -5.956 |
pione | -5.917 |
� | -5.888 |
Token | oud |
Token ID | 2778 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
citiz | +1.972 |
lation | +1.916 |
eleph | +1.842 |
TAIN | +1.816 |
utenant | +1.775 |
Neg logit contributions | |
tested | -1.170 |
expected | -1.123 |
coal | -1.104 |
isot | -1.067 |
cube | -1.043 |
Token | wasn |
Token ID | 2492 |
Feature activation | -6.75e+00 |
Pos logit contributions | |
tested | +0.700 |
expected | +0.665 |
coal | +0.651 |
isot | +0.617 |
cube | +0.617 |
Neg logit contributions | |
citiz | -1.111 |
eleph | -1.072 |
lation | -1.012 |
teasp | -0.992 |
TAIN | -0.980 |
Token | 't |
Token ID | 470 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
tested | +2.318 |
expected | +2.098 |
coal | +1.962 |
cube | +1.908 |
felt | +1.742 |
Neg logit contributions | |
citiz | -8.767 |
eleph | -8.757 |
teasp | -8.135 |
� | -8.019 |
pione | -8.008 |
Token | the |
Token ID | 262 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
tested | +6.299 |
expected | +5.771 |
cube | +5.464 |
coal | +5.400 |
forced | +5.193 |
Neg logit contributions | |
eleph | -14.614 |
citiz | -14.324 |
teasp | -13.695 |
� | -13.616 |
pione | -13.539 |
Token | one |
Token ID | 530 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
tested | +3.409 |
coal | +3.296 |
expected | +3.285 |
isot | +3.214 |
cube | +3.204 |
Neg logit contributions | |
citiz | -4.029 |
eleph | -3.971 |
teasp | -3.688 |
� | -3.553 |
pione | -3.512 |
Token | tried |
Token ID | 3088 |
Feature activation | -9.30e+00 |
Pos logit contributions | |
tested | +4.899 |
expected | +4.329 |
coal | +4.055 |
forced | +3.878 |
cube | +3.823 |
Neg logit contributions | |
citiz | -11.207 |
eleph | -11.080 |
� | -10.392 |
teasp | -10.268 |
pione | -10.184 |
Token | to |
Token ID | 284 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
tested | +5.184 |
expected | +4.796 |
cube | +4.712 |
coal | +4.594 |
isot | +4.443 |
Neg logit contributions | |
citiz | -9.889 |
eleph | -9.827 |
� | -9.162 |
pione | -9.066 |
teasp | -9.010 |
Token | steal |
Token ID | 8711 |
Feature activation | +7.63e+00 |
Pos logit contributions | |
tested | +0.168 |
expected | +0.162 |
coal | +0.161 |
isot | +0.154 |
cube | +0.154 |
Neg logit contributions | |
citiz | -0.193 |
eleph | -0.184 |
lation | -0.176 |
teasp | -0.168 |
TAIN | -0.168 |