Token | You |
Token ID | 921 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
buttocks | +0.612 |
ONY | +0.594 |
Bench | +0.594 |
igan | +0.585 |
venth | +0.583 |
Neg logit contributions | |
emort | -0.584 |
urable | -0.573 |
scales | -0.560 |
Cla | -0.559 |
milo | -0.553 |
Token | essentially |
Token ID | 6986 |
Feature activation | +9.33e-01 |
Pos logit contributions | |
buttocks | +0.241 |
ONY | +0.229 |
venth | +0.229 |
Bench | +0.229 |
igan | +0.227 |
Neg logit contributions | |
emort | -0.228 |
urable | -0.224 |
Cla | -0.223 |
scales | -0.219 |
milo | -0.217 |
Token | lay |
Token ID | 3830 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
buttocks | +0.646 |
ONY | +0.613 |
venth | +0.612 |
Bench | +0.611 |
igan | +0.610 |
Neg logit contributions | |
emort | -0.553 |
urable | -0.541 |
Cla | -0.538 |
milo | -0.526 |
scales | -0.525 |
Token | your |
Token ID | 534 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
emort | +0.600 |
urable | +0.596 |
Cla | +0.592 |
scales | +0.582 |
milo | +0.573 |
Neg logit contributions | |
buttocks | -0.670 |
ONY | -0.641 |
Bench | -0.639 |
igan | -0.634 |
venth | -0.633 |
Token | che |
Token ID | 1125 |
Feature activation | -8.80e-01 |
Pos logit contributions | |
buttocks | +0.812 |
Bench | +0.766 |
ONY | +0.765 |
venth | +0.758 |
igan | +0.757 |
Neg logit contributions | |
emort | -0.697 |
urable | -0.687 |
Cla | -0.681 |
milo | -0.662 |
scales | -0.660 |
Token | v |
Token ID | 85 |
Feature activation | +5.91e+00 |
Pos logit contributions | |
emort | +0.466 |
urable | +0.461 |
Cla | +0.461 |
scales | +0.456 |
milo | +0.441 |
Neg logit contributions | |
buttocks | -0.655 |
ONY | -0.629 |
Bench | -0.627 |
venth | -0.626 |
igan | -0.622 |
Token | ron |
Token ID | 1313 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
igan | +2.991 |
buttocks | +2.978 |
ONY | +2.966 |
venth | +2.932 |
Bench | +2.801 |
Neg logit contributions | |
emort | -4.376 |
Cla | -4.304 |
scales | -4.260 |
milo | -4.248 |
urable | -4.248 |
Token | template |
Token ID | 11055 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
buttocks | +1.045 |
ONY | +0.993 |
Bench | +0.992 |
igan | +0.990 |
venth | +0.982 |
Neg logit contributions | |
emort | -0.910 |
urable | -0.903 |
Cla | -0.901 |
scales | -0.877 |
milo | -0.871 |
Token | down |
Token ID | 866 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
buttocks | +0.742 |
ONY | +0.707 |
Bench | +0.703 |
igan | +0.701 |
venth | +0.700 |
Neg logit contributions | |
emort | -0.689 |
urable | -0.681 |
Cla | -0.679 |
scales | -0.666 |
milo | -0.658 |
Token | , |
Token ID | 11 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
buttocks | +0.769 |
ONY | +0.730 |
Bench | +0.727 |
igan | +0.727 |
venth | +0.725 |
Neg logit contributions | |
emort | -0.763 |
urable | -0.754 |
Cla | -0.752 |
scales | -0.739 |
milo | -0.730 |
Token | corners |
Token ID | 14371 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
buttocks | +0.507 |
ONY | +0.481 |
igan | +0.479 |
Bench | +0.479 |
venth | +0.475 |
Neg logit contributions | |
emort | -0.434 |
urable | -0.425 |
Cla | -0.423 |
milo | -0.417 |
scales | -0.409 |
Token | following |
Token ID | 1708 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
buttocks | +1.238 |
ONY | +1.185 |
Bench | +1.180 |
igan | +1.166 |
venth | +1.157 |
Neg logit contributions | |
emort | -1.187 |
urable | -1.163 |
Cla | -1.150 |
scales | -1.141 |
milo | -1.140 |
Token | me |
Token ID | 502 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
emort | +0.544 |
urable | +0.539 |
Cla | +0.538 |
scales | +0.533 |
milo | +0.518 |
Neg logit contributions | |
buttocks | -0.647 |
ONY | -0.619 |
Bench | -0.618 |
venth | -0.615 |
igan | -0.612 |
Token | on |
Token ID | 319 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
buttocks | +0.502 |
ONY | +0.478 |
Bench | +0.478 |
igan | +0.475 |
venth | +0.474 |
Neg logit contributions | |
emort | -0.493 |
urable | -0.490 |
Cla | -0.489 |
scales | -0.483 |
milo | -0.473 |
Token | this |
Token ID | 428 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
buttocks | +0.326 |
ONY | +0.312 |
Bench | +0.311 |
igan | +0.309 |
venth | +0.308 |
Neg logit contributions | |
emort | -0.345 |
urable | -0.341 |
Cla | -0.338 |
scales | -0.333 |
milo | -0.330 |
Token | j |
Token ID | 474 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
buttocks | +0.284 |
ONY | +0.273 |
igan | +0.271 |
Bench | +0.271 |
venth | +0.270 |
Neg logit contributions | |
emort | -0.179 |
urable | -0.177 |
Cla | -0.176 |
scales | -0.173 |
milo | -0.171 |
Token | ourn |
Token ID | 1798 |
Feature activation | +5.62e+00 |
Pos logit contributions | |
buttocks | +0.974 |
venth | +0.955 |
igan | +0.953 |
ONY | +0.944 |
Bench | +0.941 |
Neg logit contributions | |
Cla | -0.905 |
emort | -0.900 |
scales | -0.889 |
urable | -0.888 |
milo | -0.868 |
Token | y |
Token ID | 88 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
igan | +3.920 |
buttocks | +3.797 |
venth | +3.774 |
ONY | +3.755 |
Bench | +3.693 |
Neg logit contributions | |
scales | -3.142 |
Cla | -3.132 |
urable | -3.085 |
Kahn | -3.051 |
emort | -3.033 |
Token | , |
Token ID | 11 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
emort | +0.709 |
Cla | +0.702 |
urable | +0.702 |
scales | +0.695 |
milo | +0.676 |
Neg logit contributions | |
buttocks | -0.788 |
ONY | -0.753 |
Bench | -0.751 |
venth | -0.749 |
igan | -0.745 |
Token | because |
Token ID | 780 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
buttocks | +0.328 |
ONY | +0.314 |
Bench | +0.312 |
igan | +0.310 |
venth | +0.310 |
Neg logit contributions | |
emort | -0.318 |
urable | -0.314 |
Cla | -0.312 |
scales | -0.308 |
milo | -0.302 |
Token | like |
Token ID | 588 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
buttocks | +0.762 |
ONY | +0.726 |
Bench | +0.725 |
igan | +0.720 |
venth | +0.718 |
Neg logit contributions | |
emort | -0.752 |
urable | -0.742 |
Cla | -0.730 |
scales | -0.722 |
milo | -0.714 |
Token | always |
Token ID | 1464 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
emort | +0.297 |
urable | +0.294 |
Cla | +0.290 |
scales | +0.287 |
milo | +0.283 |
Neg logit contributions | |
buttocks | -0.325 |
ONY | -0.310 |
Bench | -0.309 |
venth | -0.308 |
igan | -0.307 |
Token | how |
Token ID | 703 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
buttocks | +0.572 |
ONY | +0.542 |
Bench | +0.539 |
igan | +0.539 |
venth | +0.539 |
Neg logit contributions | |
emort | -0.344 |
urable | -0.339 |
Cla | -0.333 |
scales | -0.328 |
milo | -0.325 |
Token | she |
Token ID | 673 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
buttocks | +0.306 |
Bench | +0.288 |
ONY | +0.288 |
igan | +0.288 |
venth | +0.287 |
Neg logit contributions | |
emort | -0.229 |
urable | -0.224 |
Cla | -0.223 |
scales | -0.220 |
milo | -0.217 |
Token | looked |
Token ID | 3114 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
buttocks | +0.830 |
igan | +0.771 |
Bench | +0.759 |
venth | +0.757 |
ONY | +0.755 |
Neg logit contributions | |
emort | -0.674 |
Cla | -0.655 |
urable | -0.654 |
milo | -0.637 |
scales | -0.629 |
Token | . |
Token ID | 13 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
buttocks | +0.028 |
ONY | +0.027 |
Bench | +0.027 |
venth | +0.027 |
igan | +0.026 |
Neg logit contributions | |
emort | -0.024 |
urable | -0.024 |
Cla | -0.024 |
scales | -0.023 |
milo | -0.023 |
Token | " |
Token ID | 366 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
buttocks | +0.451 |
Bench | +0.436 |
ONY | +0.435 |
igan | +0.433 |
venth | +0.432 |
Neg logit contributions | |
emort | -0.367 |
urable | -0.363 |
scales | -0.356 |
milo | -0.352 |
Cla | -0.350 |
Token | Due |
Token ID | 22229 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
buttocks | +0.326 |
Bench | +0.322 |
ONY | +0.320 |
Newsletter | +0.318 |
venth | +0.316 |
Neg logit contributions | |
emort | -0.213 |
urable | -0.210 |
scales | -0.208 |
Cla | -0.206 |
milo | -0.203 |
Token | to |
Token ID | 284 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
buttocks | +3.405 |
venth | +3.160 |
igan | +3.157 |
secondly | +3.154 |
Bench | +3.130 |
Neg logit contributions | |
emort | -3.610 |
urable | -3.558 |
milo | -3.537 |
Cla | -3.470 |
Kahn | -3.405 |
Token | this |
Token ID | 428 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
emort | +0.155 |
urable | +0.154 |
Cla | +0.150 |
scales | +0.149 |
milo | +0.148 |
Neg logit contributions | |
buttocks | -0.235 |
ONY | -0.223 |
Bench | -0.223 |
venth | -0.223 |
igan | -0.223 |
Token | fl |
Token ID | 781 |
Feature activation | -6.42e-01 |
Pos logit contributions | |
buttocks | +0.177 |
ONY | +0.168 |
venth | +0.167 |
igan | +0.167 |
Bench | +0.167 |
Neg logit contributions | |
emort | -0.143 |
urable | -0.142 |
Cla | -0.140 |
scales | -0.138 |
milo | -0.137 |
Token | attery |
Token ID | 16296 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
Cla | +0.399 |
emort | +0.399 |
scales | +0.394 |
urable | +0.392 |
milo | +0.385 |
Neg logit contributions | |
buttocks | -0.418 |
venth | -0.401 |
ONY | -0.401 |
igan | -0.400 |
Bench | -0.398 |
Token | the |
Token ID | 262 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
buttocks | +0.878 |
igan | +0.830 |
ONY | +0.828 |
Bench | +0.827 |
venth | +0.826 |
Neg logit contributions | |
emort | -0.758 |
urable | -0.753 |
Cla | -0.738 |
scales | -0.737 |
milo | -0.732 |
Token | cable |
Token ID | 7862 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
buttocks | +0.642 |
ONY | +0.609 |
Bench | +0.606 |
venth | +0.606 |
igan | +0.602 |
Neg logit contributions | |
emort | -0.660 |
urable | -0.650 |
scales | -0.645 |
Cla | -0.643 |
milo | -0.633 |
Token | , |
Token ID | 11 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
buttocks | +0.552 |
ONY | +0.527 |
Bench | +0.524 |
venth | +0.523 |
igan | +0.520 |
Neg logit contributions | |
emort | -0.561 |
urable | -0.556 |
Cla | -0.555 |
scales | -0.550 |
milo | -0.536 |
Token | � |
Token ID | 564 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
buttocks | +1.003 |
ONY | +0.956 |
Bench | +0.945 |
igan | +0.942 |
venth | +0.936 |
Neg logit contributions | |
emort | -0.991 |
urable | -0.984 |
scales | -0.966 |
Cla | -0.961 |
milo | -0.944 |
Token | � |
Token ID | 250 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
buttocks | +0.863 |
ONY | +0.826 |
Bench | +0.825 |
venth | +0.823 |
igan | +0.817 |
Neg logit contributions | |
emort | -0.748 |
urable | -0.740 |
Cla | -0.740 |
scales | -0.732 |
milo | -0.711 |
Token | Che |
Token ID | 7376 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
emort | +0.082 |
urable | +0.082 |
Cla | +0.081 |
scales | +0.081 |
Kahn | +0.078 |
Neg logit contributions | |
buttocks | -0.110 |
Bench | -0.106 |
ONY | -0.106 |
venth | -0.105 |
igan | -0.104 |
Token | v |
Token ID | 85 |
Feature activation | +5.26e+00 |
Pos logit contributions | |
buttocks | +0.298 |
ONY | +0.288 |
Bench | +0.286 |
venth | +0.286 |
igan | +0.284 |
Neg logit contributions | |
emort | -0.142 |
urable | -0.140 |
Cla | -0.140 |
scales | -0.138 |
milo | -0.132 |
Token | ron |
Token ID | 1313 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
buttocks | +2.603 |
ONY | +2.588 |
igan | +2.547 |
venth | +2.498 |
Bench | +2.444 |
Neg logit contributions | |
emort | -4.029 |
Cla | -3.942 |
urable | -3.926 |
scales | -3.914 |
milo | -3.877 |
Token | country |
Token ID | 1499 |
Feature activation | +9.14e-01 |
Pos logit contributions | |
buttocks | +0.865 |
ONY | +0.824 |
Bench | +0.823 |
igan | +0.821 |
venth | +0.817 |
Neg logit contributions | |
emort | -0.754 |
urable | -0.754 |
Cla | -0.742 |
scales | -0.733 |
milo | -0.722 |
Token | manager |
Token ID | 4706 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
buttocks | +0.633 |
ONY | +0.603 |
Bench | +0.602 |
igan | +0.599 |
venth | +0.597 |
Neg logit contributions | |
emort | -0.535 |
urable | -0.534 |
Cla | -0.525 |
scales | -0.517 |
milo | -0.508 |
Token | Pat |
Token ID | 3208 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
emort | +0.203 |
urable | +0.201 |
Cla | +0.201 |
scales | +0.199 |
milo | +0.194 |
Neg logit contributions | |
buttocks | -0.232 |
ONY | -0.222 |
Bench | -0.221 |
venth | -0.221 |
igan | -0.220 |
Token | ry |
Token ID | 563 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
emort | +0.411 |
urable | +0.407 |
Cla | +0.406 |
scales | +0.405 |
milo | +0.394 |
Neg logit contributions | |
buttocks | -0.341 |
ONY | -0.322 |
Bench | -0.321 |
venth | -0.321 |
igan | -0.320 |
Token | ) |
Token ID | 1267 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
buttocks | +0.627 |
ONY | +0.601 |
Bench | +0.598 |
venth | +0.595 |
igan | +0.592 |
Neg logit contributions | |
emort | -0.545 |
urable | -0.540 |
Cla | -0.536 |
scales | -0.534 |
milo | -0.522 |
Token | AS |
Token ID | 7054 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
buttocks | +0.899 |
ONY | +0.849 |
Bench | +0.844 |
igan | +0.825 |
venth | +0.824 |
Neg logit contributions | |
emort | -1.461 |
urable | -1.446 |
Cla | -1.431 |
scales | -1.426 |
milo | -1.418 |
Token | version |
Token ID | 2196 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
buttocks | +1.257 |
Bench | +1.202 |
ONY | +1.194 |
venth | +1.173 |
igan | +1.159 |
Neg logit contributions | |
emort | -1.133 |
urable | -1.116 |
Cla | -1.100 |
milo | -1.088 |
scales | -1.080 |
Token | , |
Token ID | 837 |
Feature activation | +9.93e-01 |
Pos logit contributions | |
emort | +0.536 |
urable | +0.532 |
Cla | +0.532 |
scales | +0.527 |
milo | +0.513 |
Neg logit contributions | |
buttocks | -0.527 |
ONY | -0.502 |
Bench | -0.500 |
venth | -0.499 |
igan | -0.495 |
Token | C |
Token ID | 327 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
buttocks | +0.551 |
ONY | +0.523 |
Bench | +0.520 |
venth | +0.510 |
igan | +0.508 |
Neg logit contributions | |
emort | -0.720 |
urable | -0.712 |
Cla | -0.705 |
scales | -0.699 |
milo | -0.697 |
Token | OUNT |
Token ID | 28270 |
Feature activation | +5.26e+00 |
Pos logit contributions | |
buttocks | +0.982 |
Bench | +0.958 |
ONY | +0.955 |
venth | +0.934 |
igan | +0.931 |
Neg logit contributions | |
emort | -0.856 |
Cla | -0.844 |
urable | -0.840 |
scales | -0.838 |
Kahn | -0.826 |
Token | ( |
Token ID | 357 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
buttocks | +3.742 |
ONY | +3.694 |
Bench | +3.683 |
igan | +3.580 |
venth | +3.533 |
Neg logit contributions | |
emort | -2.827 |
urable | -2.824 |
milo | -2.746 |
Kahn | -2.708 |
scales | -2.699 |
Token | * |
Token ID | 1635 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
buttocks | +0.798 |
ONY | +0.766 |
Bench | +0.762 |
venth | +0.753 |
igan | +0.750 |
Neg logit contributions | |
emort | -0.671 |
urable | -0.661 |
Cla | -0.652 |
scales | -0.645 |
milo | -0.642 |
Token | ) |
Token ID | 1267 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
buttocks | +1.380 |
ONY | +1.320 |
Bench | +1.320 |
igan | +1.299 |
venth | +1.295 |
Neg logit contributions | |
emort | -1.344 |
urable | -1.329 |
Cla | -1.309 |
scales | -1.296 |
Kahn | -1.286 |
Token | AS |
Token ID | 7054 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
buttocks | +0.744 |
ONY | +0.708 |
Bench | +0.706 |
igan | +0.694 |
venth | +0.694 |
Neg logit contributions | |
emort | -0.887 |
urable | -0.880 |
Cla | -0.870 |
scales | -0.862 |
milo | -0.855 |
Token | count |
Token ID | 954 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
buttocks | +0.989 |
Bench | +0.950 |
ONY | +0.945 |
venth | +0.932 |
igan | +0.921 |
Neg logit contributions | |
emort | -0.834 |
urable | -0.824 |
Cla | -0.813 |
milo | -0.800 |
scales | -0.799 |
Token | was |
Token ID | 373 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
buttocks | +0.058 |
Bench | +0.055 |
venth | +0.054 |
ONY | +0.054 |
igan | +0.054 |
Neg logit contributions | |
emort | -0.050 |
urable | -0.049 |
Cla | -0.048 |
scales | -0.047 |
milo | -0.047 |
Token | hes |
Token ID | 956 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
buttocks | +0.469 |
venth | +0.452 |
igan | +0.452 |
Bench | +0.451 |
ONY | +0.446 |
Neg logit contributions | |
emort | -0.315 |
scales | -0.303 |
urable | -0.301 |
Cla | -0.296 |
Kahn | -0.293 |
Token | in |
Token ID | 287 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
buttocks | +0.391 |
Bench | +0.372 |
ONY | +0.372 |
venth | +0.371 |
igan | +0.370 |
Neg logit contributions | |
emort | -0.342 |
Cla | -0.337 |
urable | -0.335 |
scales | -0.331 |
milo | -0.323 |
Token | the |
Token ID | 262 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
buttocks | +0.312 |
Bench | +0.297 |
ONY | +0.295 |
igan | +0.294 |
venth | +0.294 |
Neg logit contributions | |
emort | -0.262 |
urable | -0.259 |
Cla | -0.254 |
scales | -0.252 |
milo | -0.247 |
Token | Moj |
Token ID | 40120 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
buttocks | +0.399 |
Bench | +0.375 |
ONY | +0.372 |
venth | +0.371 |
igan | +0.370 |
Neg logit contributions | |
emort | -0.394 |
urable | -0.389 |
Cla | -0.381 |
scales | -0.376 |
milo | -0.376 |
Token | ave |
Token ID | 1015 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
emort | +0.254 |
scales | +0.244 |
Cla | +0.244 |
urable | +0.243 |
milo | +0.237 |
Neg logit contributions | |
buttocks | -0.293 |
igan | -0.282 |
Bench | -0.282 |
ONY | -0.281 |
venth | -0.280 |
Token | Desert |
Token ID | 18692 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
buttocks | +2.321 |
Bench | +2.220 |
igan | +2.058 |
Messenger | +2.048 |
venth | +2.033 |
Neg logit contributions | |
emort | -4.509 |
milo | -4.330 |
urable | -4.284 |
Cla | -4.190 |
scales | -4.084 |
Token | to |
Token ID | 284 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
buttocks | +1.322 |
Bench | +1.262 |
venth | +1.246 |
igan | +1.244 |
ONY | +1.243 |
Neg logit contributions | |
emort | -1.216 |
urable | -1.180 |
Cla | -1.177 |
scales | -1.155 |
milo | -1.146 |
Token | Mo |
Token ID | 4270 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
buttocks | +0.341 |
Bench | +0.323 |
ONY | +0.321 |
venth | +0.320 |
igan | +0.319 |
Neg logit contributions | |
emort | -0.295 |
urable | -0.290 |
Cla | -0.286 |
scales | -0.281 |
milo | -0.278 |
Token | ab |
Token ID | 397 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
emort | +1.223 |
Cla | +1.215 |
urable | +1.213 |
scales | +1.200 |
milo | +1.162 |
Neg logit contributions | |
buttocks | -1.488 |
ONY | -1.423 |
Bench | -1.417 |
venth | -1.415 |
igan | -1.404 |
Token | 's |
Token ID | 338 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
buttocks | +1.224 |
Bench | +1.179 |
igan | +1.158 |
ONY | +1.157 |
venth | +1.144 |
Neg logit contributions | |
emort | -1.209 |
urable | -1.183 |
Cla | -1.174 |
milo | -1.156 |
scales | -1.152 |
Token | record |
Token ID | 1700 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
buttocks | +0.909 |
venth | +0.877 |
Bench | +0.874 |
ONY | +0.868 |
igan | +0.864 |
Neg logit contributions | |
emort | -0.567 |
Cla | -0.563 |
urable | -0.553 |
milo | -0.547 |
scales | -0.545 |
Token | , |
Token ID | 11 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
emort | +0.920 |
urable | +0.918 |
Cla | +0.917 |
scales | +0.911 |
milo | +0.880 |
Neg logit contributions | |
buttocks | -1.020 |
ONY | -0.977 |
Bench | -0.973 |
venth | -0.965 |
igan | -0.961 |
Token | what |
Token ID | 644 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
buttocks | +0.153 |
ONY | +0.145 |
Bench | +0.144 |
venth | +0.143 |
igan | +0.143 |
Neg logit contributions | |
emort | -0.136 |
urable | -0.135 |
Cla | -0.130 |
milo | -0.130 |
scales | -0.129 |
Token | can |
Token ID | 460 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
emort | +0.198 |
urable | +0.196 |
Cla | +0.192 |
scales | +0.189 |
milo | +0.189 |
Neg logit contributions | |
buttocks | -0.228 |
venth | -0.217 |
ONY | -0.217 |
Bench | -0.217 |
igan | -0.215 |
Token | we |
Token ID | 356 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
buttocks | +0.577 |
ONY | +0.560 |
Bench | +0.554 |
igan | +0.554 |
venth | +0.547 |
Neg logit contributions | |
emort | -0.509 |
urable | -0.499 |
scales | -0.489 |
milo | -0.485 |
Cla | -0.485 |
Token | expect |
Token ID | 1607 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
emort | +0.082 |
urable | +0.081 |
Cla | +0.080 |
scales | +0.080 |
milo | +0.079 |
Neg logit contributions | |
buttocks | -0.064 |
Bench | -0.060 |
ONY | -0.060 |
igan | -0.060 |
venth | -0.060 |
Token | from |
Token ID | 422 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
buttocks | +2.966 |
igan | +2.865 |
Bench | +2.802 |
ONY | +2.793 |
venth | +2.728 |
Neg logit contributions | |
emort | -3.588 |
urable | -3.511 |
milo | -3.476 |
scales | -3.420 |
Cla | -3.376 |
Token | the |
Token ID | 262 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
buttocks | +0.061 |
ONY | +0.058 |
Bench | +0.058 |
venth | +0.058 |
igan | +0.058 |
Neg logit contributions | |
emort | -0.044 |
urable | -0.044 |
scales | -0.043 |
Cla | -0.042 |
milo | -0.042 |
Token | team |
Token ID | 1074 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
emort | +0.015 |
urable | +0.015 |
Cla | +0.014 |
scales | +0.014 |
milo | +0.014 |
Neg logit contributions | |
buttocks | -0.017 |
Bench | -0.016 |
venth | -0.016 |
ONY | -0.016 |
igan | -0.016 |
Token | from |
Token ID | 422 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
buttocks | +0.781 |
Bench | +0.745 |
igan | +0.742 |
ONY | +0.741 |
venth | +0.739 |
Neg logit contributions | |
emort | -0.622 |
urable | -0.611 |
Cla | -0.604 |
scales | -0.593 |
milo | -0.590 |
Token | now |
Token ID | 783 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
emort | +0.069 |
urable | +0.068 |
Cla | +0.067 |
scales | +0.067 |
milo | +0.066 |
Neg logit contributions | |
buttocks | -0.082 |
ONY | -0.079 |
Bench | -0.078 |
venth | -0.078 |
igan | -0.078 |
Token | all |
Token ID | 477 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
buttocks | +0.257 |
ONY | +0.247 |
venth | +0.246 |
Bench | +0.245 |
igan | +0.243 |
Neg logit contributions | |
emort | -0.169 |
urable | -0.165 |
Cla | -0.162 |
scales | -0.161 |
milo | -0.160 |
Token | of |
Token ID | 286 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
emort | +1.060 |
Cla | +1.058 |
urable | +1.055 |
scales | +1.048 |
Kahn | +1.016 |
Neg logit contributions | |
buttocks | -1.086 |
ONY | -1.035 |
Bench | -1.034 |
venth | -1.029 |
igan | -1.021 |
Token | them |
Token ID | 606 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
buttocks | +0.135 |
ONY | +0.131 |
venth | +0.130 |
igan | +0.130 |
Bench | +0.130 |
Neg logit contributions | |
emort | -0.085 |
urable | -0.084 |
Cla | -0.082 |
scales | -0.082 |
milo | -0.080 |
Token | are |
Token ID | 389 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
emort | +0.420 |
urable | +0.415 |
Cla | +0.414 |
scales | +0.410 |
milo | +0.400 |
Neg logit contributions | |
buttocks | -0.490 |
ONY | -0.469 |
Bench | -0.467 |
venth | -0.466 |
igan | -0.464 |
Token | C |
Token ID | 327 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
emort | +0.130 |
urable | +0.127 |
Cla | +0.126 |
scales | +0.125 |
milo | +0.123 |
Neg logit contributions | |
buttocks | -0.148 |
ONY | -0.141 |
Bench | -0.140 |
venth | -0.140 |
igan | -0.139 |
Token | eph |
Token ID | 27446 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
buttocks | +0.406 |
ONY | +0.393 |
venth | +0.391 |
igan | +0.391 |
Bench | +0.390 |
Neg logit contributions | |
emort | -0.357 |
Cla | -0.353 |
urable | -0.352 |
scales | -0.350 |
milo | -0.343 |
Token | aly |
Token ID | 3400 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
buttocks | +3.629 |
igan | +3.507 |
ONY | +3.497 |
venth | +3.468 |
Bench | +3.445 |
Neg logit contributions | |
emort | -2.933 |
urable | -2.862 |
Cla | -2.810 |
milo | -2.793 |
scales | -2.769 |
Token | x |
Token ID | 87 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
buttocks | +0.849 |
ONY | +0.812 |
venth | +0.809 |
Bench | +0.808 |
igan | +0.801 |
Neg logit contributions | |
emort | -0.714 |
urable | -0.704 |
Cla | -0.699 |
scales | -0.684 |
milo | -0.675 |
Token | and |
Token ID | 290 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
buttocks | +0.736 |
Bench | +0.700 |
igan | +0.698 |
venth | +0.698 |
ONY | +0.698 |
Neg logit contributions | |
emort | -0.660 |
urable | -0.655 |
Cla | -0.643 |
scales | -0.636 |
milo | -0.634 |
Token | their |
Token ID | 511 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
buttocks | +0.300 |
ONY | +0.287 |
Bench | +0.286 |
venth | +0.285 |
igan | +0.283 |
Neg logit contributions | |
emort | -0.237 |
urable | -0.233 |
Cla | -0.227 |
scales | -0.227 |
milo | -0.225 |
Token | slaves |
Token ID | 13384 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
emort | +0.329 |
urable | +0.326 |
Cla | +0.322 |
scales | +0.318 |
milo | +0.314 |
Neg logit contributions | |
buttocks | -0.357 |
ONY | -0.338 |
Bench | -0.337 |
venth | -0.337 |
igan | -0.334 |
Token | 's |
Token ID | 338 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
emort | +0.179 |
urable | +0.177 |
Cla | +0.176 |
scales | +0.174 |
milo | +0.171 |
Neg logit contributions | |
buttocks | -0.206 |
Bench | -0.196 |
ONY | -0.196 |
venth | -0.195 |
igan | -0.194 |
Token | best |
Token ID | 1266 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
buttocks | +0.885 |
venth | +0.830 |
ONY | +0.824 |
Bench | +0.823 |
igan | +0.815 |
Neg logit contributions | |
emort | -0.815 |
urable | -0.801 |
milo | -0.783 |
Cla | -0.782 |
scales | -0.770 |
Token | - |
Token ID | 12 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
buttocks | +0.601 |
Bench | +0.567 |
ONY | +0.566 |
igan | +0.565 |
venth | +0.563 |
Neg logit contributions | |
emort | -0.524 |
urable | -0.516 |
Cla | -0.508 |
scales | -0.501 |
milo | -0.496 |
Token | known |
Token ID | 4002 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
emort | +0.080 |
Cla | +0.079 |
urable | +0.079 |
scales | +0.078 |
Kahn | +0.076 |
Neg logit contributions | |
buttocks | -0.070 |
Bench | -0.066 |
ONY | -0.066 |
venth | -0.066 |
igan | -0.066 |
Token | mail |
Token ID | 6920 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
buttocks | +1.226 |
Bench | +1.138 |
igan | +1.129 |
venth | +1.125 |
ONY | +1.124 |
Neg logit contributions | |
emort | -0.967 |
urable | -0.954 |
milo | -0.917 |
Cla | -0.915 |
scales | -0.897 |
Token | order |
Token ID | 1502 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
buttocks | +0.149 |
igan | +0.142 |
venth | +0.141 |
Bench | +0.141 |
ONY | +0.141 |
Neg logit contributions | |
emort | -0.149 |
urable | -0.147 |
Cla | -0.147 |
scales | -0.143 |
milo | -0.141 |
Token | catalog |
Token ID | 18388 |
Feature activation | -1.85e-03 |
Pos logit contributions | |
buttocks | +3.497 |
igan | +3.199 |
Protestants | +3.193 |
ONY | +3.161 |
Bench | +3.159 |
Neg logit contributions | |
emort | -3.215 |
urable | -3.119 |
milo | -3.069 |
Cla | -2.998 |
ngth | -2.925 |
Token | ues |
Token ID | 947 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
emort | +0.001 |
Cla | +0.001 |
urable | +0.001 |
scales | +0.001 |
milo | +0.001 |
Neg logit contributions | |
buttocks | -0.002 |
ONY | -0.002 |
igan | -0.002 |
Bench | -0.002 |
venth | -0.002 |
Token | apologised |
Token ID | 46397 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
emort | +0.118 |
urable | +0.117 |
Cla | +0.116 |
scales | +0.114 |
milo | +0.112 |
Neg logit contributions | |
buttocks | -0.149 |
ONY | -0.142 |
Bench | -0.141 |
venth | -0.141 |
igan | -0.141 |
Token | today |
Token ID | 1909 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
emort | +1.113 |
Cla | +1.104 |
urable | +1.103 |
scales | +1.093 |
milo | +1.060 |
Neg logit contributions | |
buttocks | -1.201 |
ONY | -1.149 |
Bench | -1.146 |
venth | -1.143 |
igan | -1.133 |
Token | after |
Token ID | 706 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
emort | +0.050 |
urable | +0.049 |
Cla | +0.049 |
scales | +0.048 |
milo | +0.047 |
Neg logit contributions | |
buttocks | -0.054 |
ONY | -0.052 |
venth | -0.052 |
Bench | -0.051 |
igan | -0.051 |
Token | has |
Token ID | 468 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
buttocks | +0.058 |
ONY | +0.055 |
Bench | +0.055 |
igan | +0.054 |
venth | +0.054 |
Neg logit contributions | |
emort | -0.044 |
urable | -0.043 |
Cla | -0.043 |
scales | -0.042 |
milo | -0.042 |
Token | no |
Token ID | 645 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
emort | +0.372 |
urable | +0.368 |
Cla | +0.367 |
scales | +0.362 |
milo | +0.353 |
Neg logit contributions | |
buttocks | -0.464 |
ONY | -0.442 |
Bench | -0.440 |
venth | -0.439 |
igan | -0.438 |
Token | affinity |
Token ID | 28430 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
buttocks | +0.678 |
ONY | +0.638 |
Bench | +0.635 |
venth | +0.633 |
igan | +0.632 |
Neg logit contributions | |
emort | -0.610 |
urable | -0.603 |
Cla | -0.596 |
scales | -0.585 |
milo | -0.582 |
Token | for |
Token ID | 329 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
emort | +0.653 |
urable | +0.649 |
Cla | +0.649 |
scales | +0.644 |
milo | +0.624 |
Neg logit contributions | |
buttocks | -0.703 |
ONY | -0.673 |
Bench | -0.669 |
venth | -0.669 |
igan | -0.663 |
Token | ge |
Token ID | 4903 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
emort | +0.207 |
urable | +0.206 |
Cla | +0.203 |
scales | +0.201 |
milo | +0.197 |
Neg logit contributions | |
buttocks | -0.238 |
ONY | -0.225 |
Bench | -0.224 |
venth | -0.223 |
igan | -0.223 |
Token | eking |
Token ID | 18754 |
Feature activation | +5.16e+00 |
Pos logit contributions | |
buttocks | +0.463 |
Bench | +0.444 |
igan | +0.444 |
venth | +0.444 |
ONY | +0.443 |
Neg logit contributions | |
emort | -0.312 |
urable | -0.310 |
Cla | -0.308 |
scales | -0.302 |
milo | -0.297 |
Token | out |
Token ID | 503 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
buttocks | +2.884 |
Bench | +2.673 |
igan | +2.668 |
ONY | +2.589 |
venth | +2.535 |
Neg logit contributions | |
milo | -3.698 |
emort | -3.693 |
urable | -3.650 |
Cla | -3.643 |
iott | -3.516 |
Token | on |
Token ID | 319 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
buttocks | +0.342 |
igan | +0.322 |
ONY | +0.320 |
Bench | +0.319 |
venth | +0.318 |
Neg logit contributions | |
emort | -0.312 |
urable | -0.308 |
Cla | -0.305 |
scales | -0.300 |
milo | -0.295 |
Token | the |
Token ID | 262 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
emort | +0.269 |
urable | +0.266 |
Cla | +0.264 |
scales | +0.260 |
milo | +0.255 |
Neg logit contributions | |
buttocks | -0.332 |
ONY | -0.316 |
Bench | -0.315 |
venth | -0.314 |
igan | -0.313 |
Token | gadgets |
Token ID | 35281 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
buttocks | +0.235 |
Bench | +0.218 |
ONY | +0.218 |
venth | +0.217 |
igan | +0.216 |
Neg logit contributions | |
emort | -0.213 |
urable | -0.210 |
Cla | -0.205 |
milo | -0.203 |
scales | -0.200 |
Token | of |
Token ID | 286 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
emort | +0.432 |
urable | +0.426 |
Cla | +0.424 |
scales | +0.420 |
milo | +0.411 |
Neg logit contributions | |
buttocks | -0.484 |
ONY | -0.461 |
Bench | -0.460 |
venth | -0.458 |
igan | -0.456 |
Token | ings |
Token ID | 654 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
buttocks | +0.500 |
ONY | +0.467 |
Bench | +0.463 |
venth | +0.463 |
igan | +0.461 |
Neg logit contributions | |
emort | -0.939 |
urable | -0.931 |
Cla | -0.931 |
scales | -0.926 |
milo | -0.911 |
Token | , |
Token ID | 11 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
buttocks | +0.048 |
ONY | +0.045 |
Bench | +0.045 |
venth | +0.045 |
igan | +0.045 |
Neg logit contributions | |
emort | -0.043 |
urable | -0.043 |
Cla | -0.043 |
scales | -0.042 |
milo | -0.041 |
Token | p |
Token ID | 279 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
buttocks | +0.239 |
ONY | +0.228 |
Bench | +0.227 |
venth | +0.226 |
igan | +0.225 |
Neg logit contributions | |
emort | -0.249 |
urable | -0.247 |
Cla | -0.246 |
scales | -0.244 |
milo | -0.238 |
Token | . |
Token ID | 13 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
buttocks | +1.459 |
ONY | +1.410 |
venth | +1.402 |
igan | +1.394 |
Bench | +1.391 |
Neg logit contributions | |
emort | -1.280 |
Cla | -1.261 |
urable | -1.258 |
scales | -1.248 |
milo | -1.222 |
Token | 232 |
Token ID | 31773 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
buttocks | +0.589 |
ONY | +0.565 |
venth | +0.561 |
Bench | +0.561 |
igan | +0.557 |
Neg logit contributions | |
emort | -0.584 |
urable | -0.574 |
Cla | -0.568 |
scales | -0.563 |
milo | -0.558 |
Token | n |
Token ID | 299 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
buttocks | +1.004 |
venth | +0.957 |
ONY | +0.957 |
Bench | +0.951 |
igan | +0.947 |
Neg logit contributions | |
emort | -0.923 |
urable | -0.913 |
Cla | -0.904 |
scales | -0.895 |
milo | -0.883 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
buttocks | +3.247 |
venth | +3.171 |
ONY | +3.098 |
igan | +3.096 |
Bench | +3.026 |
Neg logit contributions | |
emort | -3.219 |
urable | -3.156 |
Cla | -3.150 |
scales | -3.083 |
milo | -3.083 |
Token | 1 |
Token ID | 16 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
buttocks | +0.738 |
ONY | +0.706 |
venth | +0.704 |
Bench | +0.699 |
igan | +0.696 |
Neg logit contributions | |
emort | -0.757 |
urable | -0.741 |
Cla | -0.731 |
scales | -0.727 |
milo | -0.720 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
buttocks | +0.686 |
venth | +0.653 |
ONY | +0.651 |
Bench | +0.650 |
igan | +0.647 |
Neg logit contributions | |
emort | -0.639 |
urable | -0.631 |
Cla | -0.626 |
scales | -0.621 |
milo | -0.611 |
Token |
|
Token ID | 198 |
Feature activation | +6.90e-01 |
Pos logit contributions | |
buttocks | +0.081 |
ONY | +0.078 |
Bench | +0.077 |
venth | +0.077 |
igan | +0.077 |
Neg logit contributions | |
emort | -0.066 |
urable | -0.065 |
Cla | -0.064 |
scales | -0.064 |
milo | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
buttocks | +0.491 |
Bench | +0.473 |
ONY | +0.472 |
venth | +0.470 |
igan | +0.469 |
Neg logit contributions | |
emort | -0.393 |
urable | -0.383 |
Cla | -0.381 |
scales | -0.377 |
milo | -0.369 |
Token | f |
Token ID | 277 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
buttocks | +0.259 |
ONY | +0.246 |
Bench | +0.246 |
venth | +0.245 |
igan | +0.245 |
Neg logit contributions | |
emort | -0.243 |
urable | -0.240 |
Cla | -0.239 |
scales | -0.233 |
milo | -0.231 |
Token | 2 |
Token ID | 362 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
buttocks | +0.602 |
ONY | +0.576 |
Bench | +0.576 |
igan | +0.575 |
venth | +0.574 |
Neg logit contributions | |
emort | -0.514 |
urable | -0.505 |
Cla | -0.500 |
scales | -0.488 |
milo | -0.480 |
Token | , |
Token ID | 837 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
buttocks | +1.341 |
ONY | +1.270 |
Bench | +1.261 |
igan | +1.259 |
venth | +1.259 |
Neg logit contributions | |
emort | -1.472 |
urable | -1.457 |
Cla | -1.441 |
scales | -1.400 |
milo | -1.394 |
Token | … |
Token ID | 3926 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
buttocks | +0.913 |
Bench | +0.862 |
ONY | +0.862 |
venth | +0.858 |
igan | +0.855 |
Neg logit contributions | |
emort | -0.885 |
urable | -0.875 |
Cla | -0.868 |
scales | -0.850 |
milo | -0.843 |
Token | f |
Token ID | 277 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
emort | +0.345 |
urable | +0.340 |
Cla | +0.338 |
scales | +0.332 |
milo | +0.329 |
Neg logit contributions | |
buttocks | -0.358 |
Bench | -0.342 |
ONY | -0.341 |
venth | -0.340 |
igan | -0.338 |
Token | n |
Token ID | 299 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
buttocks | +0.812 |
ONY | +0.772 |
venth | +0.771 |
Bench | +0.770 |
igan | +0.765 |
Neg logit contributions | |
emort | -0.966 |
urable | -0.953 |
Cla | -0.950 |
scales | -0.930 |
milo | -0.920 |
Token | , |
Token ID | 837 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
buttocks | +3.150 |
ONY | +2.978 |
venth | +2.961 |
Bench | +2.960 |
igan | +2.956 |
Neg logit contributions | |
emort | -3.285 |
Cla | -3.191 |
urable | -3.186 |
milo | -3.117 |
Kahn | -3.061 |
Token | such |
Token ID | 884 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
buttocks | +1.205 |
ONY | +1.136 |
Bench | +1.131 |
igan | +1.124 |
venth | +1.120 |
Neg logit contributions | |
emort | -1.237 |
urable | -1.214 |
Cla | -1.196 |
milo | -1.183 |
scales | -1.165 |
Token | that |
Token ID | 326 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
emort | +0.073 |
urable | +0.072 |
Cla | +0.072 |
scales | +0.071 |
milo | +0.070 |
Neg logit contributions | |
buttocks | -0.078 |
ONY | -0.074 |
Bench | -0.074 |
venth | -0.074 |
igan | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
buttocks | +0.929 |
igan | +0.881 |
Bench | +0.880 |
ONY | +0.878 |
venth | +0.876 |
Neg logit contributions | |
emort | -0.812 |
urable | -0.801 |
Cla | -0.785 |
milo | -0.770 |
scales | -0.767 |
Token | final |
Token ID | 2457 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
buttocks | +0.213 |
venth | +0.201 |
ONY | +0.201 |
Bench | +0.200 |
igan | +0.200 |
Neg logit contributions | |
emort | -0.159 |
urable | -0.156 |
Cla | -0.154 |
milo | -0.151 |
scales | -0.149 |
Token | play |
Token ID | 711 |
Feature activation | -8.26e-01 |
Pos logit contributions | |
buttocks | +1.548 |
Bench | +1.468 |
ONY | +1.457 |
venth | +1.449 |
igan | +1.446 |
Neg logit contributions | |
emort | -1.380 |
urable | -1.351 |
Cla | -1.339 |
milo | -1.329 |
Kahn | -1.307 |
Token | it |
Token ID | 340 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
emort | +0.487 |
urable | +0.482 |
Cla | +0.479 |
scales | +0.473 |
milo | +0.463 |
Neg logit contributions | |
buttocks | -0.569 |
ONY | -0.544 |
Bench | -0.543 |
venth | -0.541 |
igan | -0.538 |
Token | with |
Token ID | 351 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
emort | +0.052 |
urable | +0.052 |
Cla | +0.052 |
scales | +0.051 |
milo | +0.050 |
Neg logit contributions | |
buttocks | -0.056 |
ONY | -0.053 |
Bench | -0.053 |
venth | -0.053 |
igan | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +4.53e-03 |
Pos logit contributions | |
buttocks | +0.088 |
ONY | +0.084 |
Bench | +0.084 |
venth | +0.083 |
igan | +0.083 |
Neg logit contributions | |
emort | -0.071 |
urable | -0.070 |
Cla | -0.070 |
scales | -0.069 |
milo | -0.068 |
Token | j |
Token ID | 474 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
buttocks | +0.003 |
ONY | +0.003 |
Bench | +0.003 |
venth | +0.003 |
igan | +0.003 |
Neg logit contributions | |
emort | -0.003 |
urable | -0.003 |
Cla | -0.003 |
scales | -0.003 |
milo | -0.003 |
Token | agg |
Token ID | 9460 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
buttocks | +0.042 |
Bench | +0.040 |
ONY | +0.040 |
venth | +0.040 |
igan | +0.040 |
Neg logit contributions | |
emort | -0.032 |
Cla | -0.032 |
urable | -0.032 |
scales | -0.031 |
milo | -0.030 |
Token | ies |
Token ID | 444 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
buttocks | +3.596 |
igan | +3.533 |
ONY | +3.460 |
venth | +3.424 |
Bench | +3.409 |
Neg logit contributions | |
Cla | -2.804 |
emort | -2.785 |
urable | -2.748 |
milo | -2.707 |
scales | -2.690 |
Token | since |
Token ID | 1201 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
buttocks | +0.807 |
Bench | +0.772 |
ONY | +0.772 |
igan | +0.770 |
venth | +0.765 |
Neg logit contributions | |
emort | -0.743 |
Cla | -0.734 |
urable | -0.729 |
scales | -0.715 |
milo | -0.708 |
Token | even |
Token ID | 772 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
buttocks | +0.443 |
Bench | +0.424 |
ONY | +0.423 |
venth | +0.421 |
igan | +0.419 |
Neg logit contributions | |
emort | -0.358 |
urable | -0.351 |
Cla | -0.348 |
scales | -0.342 |
milo | -0.337 |
Token | though |
Token ID | 996 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
buttocks | +0.813 |
Bench | +0.776 |
ONY | +0.775 |
venth | +0.767 |
igan | +0.762 |
Neg logit contributions | |
emort | -0.738 |
urable | -0.730 |
Cla | -0.725 |
scales | -0.711 |
milo | -0.702 |
Token | they |
Token ID | 484 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
buttocks | +0.033 |
Bench | +0.032 |
ONY | +0.032 |
venth | +0.031 |
igan | +0.031 |
Neg logit contributions | |
emort | -0.028 |
urable | -0.027 |
Cla | -0.027 |
scales | -0.027 |
milo | -0.026 |
Token | ays |
Token ID | 592 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
buttocks | +0.419 |
ONY | +0.395 |
Bench | +0.394 |
venth | +0.390 |
igan | +0.389 |
Neg logit contributions | |
emort | -0.676 |
Cla | -0.671 |
urable | -0.669 |
scales | -0.662 |
milo | -0.654 |
Token | cale |
Token ID | 38765 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
emort | +0.262 |
Cla | +0.258 |
urable | +0.258 |
scales | +0.254 |
milo | +0.244 |
Neg logit contributions | |
buttocks | -0.542 |
ONY | -0.523 |
Bench | -0.522 |
venth | -0.521 |
igan | -0.518 |
Token | image |
Token ID | 2939 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
buttocks | +0.172 |
Bench | +0.164 |
ONY | +0.164 |
venth | +0.163 |
igan | +0.162 |
Neg logit contributions | |
emort | -0.145 |
urable | -0.144 |
Cla | -0.143 |
milo | -0.140 |
scales | -0.140 |
Token | = |
Token ID | 796 |
Feature activation | -9.89e-02 |
Pos logit contributions | |
buttocks | +0.143 |
Bench | +0.139 |
ONY | +0.138 |
venth | +0.138 |
igan | +0.137 |
Neg logit contributions | |
emort | -0.061 |
Cla | -0.060 |
urable | -0.060 |
scales | -0.058 |
milo | -0.057 |
Token | c |
Token ID | 269 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
emort | +0.056 |
urable | +0.056 |
Cla | +0.055 |
milo | +0.054 |
scales | +0.054 |
Neg logit contributions | |
buttocks | -0.070 |
Bench | -0.067 |
ONY | -0.067 |
igan | -0.067 |
venth | -0.066 |
Token | v |
Token ID | 85 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
buttocks | +1.619 |
igan | +1.586 |
Bench | +1.582 |
venth | +1.575 |
ONY | +1.529 |
Neg logit contributions | |
Cla | -0.995 |
emort | -0.973 |
scales | -0.962 |
Kahn | -0.960 |
urable | -0.945 |
Token | 2 |
Token ID | 17 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
Bench | +3.175 |
buttocks | +3.092 |
igan | +3.070 |
Newsletter | +3.038 |
ONY | +3.027 |
Neg logit contributions | |
emort | -3.313 |
Cla | -3.152 |
Kahn | -3.127 |
urable | -3.114 |
scales | -3.086 |
Token | . |
Token ID | 13 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
buttocks | +0.601 |
Bench | +0.578 |
ONY | +0.577 |
venth | +0.574 |
igan | +0.571 |
Neg logit contributions | |
emort | -0.509 |
Cla | -0.503 |
urable | -0.502 |
scales | -0.497 |
milo | -0.485 |
Token | im |
Token ID | 320 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
emort | +0.269 |
Cla | +0.267 |
urable | +0.265 |
scales | +0.264 |
Kahn | +0.256 |
Neg logit contributions | |
buttocks | -0.331 |
Bench | -0.319 |
ONY | -0.318 |
venth | -0.317 |
igan | -0.315 |
Token | read |
Token ID | 961 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
emort | +0.283 |
Cla | +0.281 |
urable | +0.279 |
scales | +0.277 |
milo | +0.269 |
Neg logit contributions | |
buttocks | -0.369 |
Bench | -0.356 |
ONY | -0.355 |
venth | -0.353 |
igan | -0.352 |
Token | ( |
Token ID | 7 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
emort | +0.407 |
Cla | +0.404 |
urable | +0.402 |
scales | +0.399 |
Kahn | +0.389 |
Neg logit contributions | |
buttocks | -0.496 |
ONY | -0.477 |
Bench | -0.477 |
venth | -0.474 |
igan | -0.474 |
Token | seen |
Token ID | 1775 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
buttocks | +0.178 |
ONY | +0.169 |
venth | +0.168 |
Bench | +0.168 |
igan | +0.168 |
Neg logit contributions | |
emort | -0.144 |
urable | -0.141 |
Cla | -0.140 |
scales | -0.139 |
milo | -0.136 |
Token | as |
Token ID | 355 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
buttocks | +0.849 |
ONY | +0.795 |
igan | +0.792 |
venth | +0.790 |
Bench | +0.790 |
Neg logit contributions | |
emort | -1.115 |
urable | -1.102 |
Cla | -1.097 |
scales | -1.087 |
milo | -1.070 |
Token | a |
Token ID | 257 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
emort | +0.017 |
urable | +0.016 |
Cla | +0.016 |
scales | +0.016 |
milo | +0.016 |
Neg logit contributions | |
buttocks | -0.022 |
ONY | -0.021 |
venth | -0.021 |
igan | -0.021 |
Bench | -0.021 |
Token | racial |
Token ID | 8425 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
emort | +0.507 |
Cla | +0.503 |
urable | +0.503 |
scales | +0.498 |
milo | +0.484 |
Neg logit contributions | |
buttocks | -0.535 |
ONY | -0.511 |
Bench | -0.509 |
venth | -0.508 |
igan | -0.504 |
Token | epit |
Token ID | 21240 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
emort | +0.576 |
urable | +0.571 |
Cla | +0.571 |
scales | +0.565 |
milo | +0.553 |
Neg logit contributions | |
buttocks | -0.499 |
ONY | -0.473 |
Bench | -0.470 |
venth | -0.470 |
igan | -0.466 |
Token | het |
Token ID | 3202 |
Feature activation | +5.07e+00 |
Pos logit contributions | |
emort | +0.022 |
Cla | +0.022 |
urable | +0.022 |
scales | +0.022 |
milo | +0.021 |
Neg logit contributions | |
buttocks | -0.028 |
ONY | -0.027 |
venth | -0.027 |
igan | -0.026 |
Bench | -0.026 |
Token | that |
Token ID | 326 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
buttocks | +3.136 |
igan | +2.942 |
ONY | +2.925 |
venth | +2.883 |
Newsletter | +2.859 |
Neg logit contributions | |
emort | -3.345 |
urable | -3.275 |
Cla | -3.220 |
milo | -3.208 |
scales | -3.188 |
Token | targeted |
Token ID | 7977 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
buttocks | +0.586 |
ONY | +0.548 |
igan | +0.547 |
venth | +0.545 |
Bench | +0.544 |
Neg logit contributions | |
emort | -0.538 |
urable | -0.531 |
Cla | -0.515 |
scales | -0.512 |
milo | -0.508 |
Token | African |
Token ID | 5510 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
buttocks | +0.338 |
ONY | +0.318 |
Bench | +0.316 |
venth | +0.316 |
igan | +0.314 |
Neg logit contributions | |
emort | -0.362 |
urable | -0.359 |
Cla | -0.356 |
scales | -0.353 |
milo | -0.348 |
Token | - |
Token ID | 12 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
emort | +0.024 |
urable | +0.024 |
Cla | +0.024 |
scales | +0.023 |
milo | +0.023 |
Neg logit contributions | |
buttocks | -0.021 |
ONY | -0.020 |
Bench | -0.020 |
venth | -0.020 |
igan | -0.019 |
Token | Americans |
Token ID | 17636 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
buttocks | +0.184 |
Bench | +0.174 |
ONY | +0.174 |
venth | +0.173 |
igan | +0.172 |
Neg logit contributions | |
emort | -0.224 |
Cla | -0.223 |
urable | -0.223 |
scales | -0.222 |
milo | -0.216 |
Token | is |
Token ID | 318 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
emort | +0.448 |
urable | +0.442 |
Cla | +0.439 |
scales | +0.425 |
milo | +0.420 |
Neg logit contributions | |
buttocks | -0.650 |
venth | -0.618 |
igan | -0.618 |
ONY | -0.618 |
Bench | -0.616 |
Token | totally |
Token ID | 6635 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
buttocks | +0.206 |
ONY | +0.195 |
Bench | +0.195 |
igan | +0.194 |
venth | +0.194 |
Neg logit contributions | |
emort | -0.182 |
urable | -0.180 |
Cla | -0.179 |
scales | -0.176 |
milo | -0.174 |
Token | rigged |
Token ID | 31043 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
emort | +0.410 |
urable | +0.405 |
Cla | +0.404 |
scales | +0.398 |
milo | +0.389 |
Neg logit contributions | |
buttocks | -0.574 |
ONY | -0.548 |
Bench | -0.546 |
venth | -0.546 |
igan | -0.543 |
Token | and |
Token ID | 290 |
Feature activation | +8.55e-01 |
Pos logit contributions | |
buttocks | +0.810 |
igan | +0.773 |
ONY | +0.766 |
venth | +0.760 |
Bench | +0.756 |
Neg logit contributions | |
emort | -0.770 |
urable | -0.765 |
Cla | -0.763 |
scales | -0.751 |
milo | -0.734 |
Token | Cro |
Token ID | 9325 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
buttocks | +0.596 |
ONY | +0.568 |
igan | +0.563 |
Bench | +0.563 |
venth | +0.562 |
Neg logit contributions | |
emort | -0.501 |
urable | -0.495 |
Cla | -0.486 |
milo | -0.476 |
scales | -0.476 |
Token | oked |
Token ID | 6545 |
Feature activation | +5.07e+00 |
Pos logit contributions | |
emort | +0.598 |
urable | +0.593 |
Cla | +0.585 |
scales | +0.577 |
milo | +0.570 |
Neg logit contributions | |
buttocks | -0.474 |
ONY | -0.457 |
venth | -0.456 |
Bench | -0.454 |
igan | -0.452 |
Token | Hillary |
Token ID | 5073 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
buttocks | +3.281 |
igan | +3.115 |
ONY | +3.074 |
Bench | +3.033 |
venth | +3.013 |
Neg logit contributions | |
urable | -3.106 |
emort | -3.059 |
scales | -2.973 |
eq | -2.934 |
ngth | -2.924 |
Token | Clinton |
Token ID | 2605 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
emort | +0.062 |
urable | +0.062 |
Cla | +0.061 |
scales | +0.060 |
milo | +0.059 |
Neg logit contributions | |
buttocks | -0.092 |
ONY | -0.088 |
Bench | -0.088 |
venth | -0.088 |
igan | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
emort | +0.142 |
urable | +0.140 |
Cla | +0.138 |
scales | +0.136 |
milo | +0.134 |
Neg logit contributions | |
buttocks | -0.199 |
ONY | -0.190 |
igan | -0.190 |
Bench | -0.189 |
venth | -0.189 |
Token | Deborah |
Token ID | 36976 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
buttocks | +0.350 |
ONY | +0.336 |
Bench | +0.334 |
venth | +0.333 |
igan | +0.332 |
Neg logit contributions | |
emort | -0.244 |
urable | -0.241 |
Cla | -0.238 |
scales | -0.235 |
milo | -0.231 |
Token | Wasserman |
Token ID | 38997 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
buttocks | +0.474 |
ONY | +0.449 |
Bench | +0.447 |
igan | +0.446 |
venth | +0.440 |
Neg logit contributions | |
emort | -0.589 |
urable | -0.581 |
milo | -0.563 |
scales | -0.561 |
Cla | -0.560 |
Token | each |
Token ID | 1123 |
Feature activation | +8.31e-01 |
Pos logit contributions | |
emort | +0.091 |
urable | +0.090 |
Cla | +0.089 |
scales | +0.087 |
milo | +0.086 |
Neg logit contributions | |
buttocks | -0.139 |
ONY | -0.133 |
Bench | -0.132 |
venth | -0.132 |
igan | -0.131 |
Token | pixel |
Token ID | 17465 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
buttocks | +0.622 |
ONY | +0.595 |
venth | +0.589 |
igan | +0.587 |
Bench | +0.587 |
Neg logit contributions | |
emort | -0.446 |
urable | -0.434 |
Cla | -0.434 |
milo | -0.422 |
scales | -0.422 |
Token | and |
Token ID | 290 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
buttocks | +0.599 |
ONY | +0.571 |
Bench | +0.568 |
venth | +0.567 |
igan | +0.566 |
Neg logit contributions | |
emort | -0.529 |
Cla | -0.523 |
urable | -0.523 |
scales | -0.510 |
milo | -0.507 |
Token | have |
Token ID | 423 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
emort | +0.102 |
urable | +0.100 |
Cla | +0.099 |
milo | +0.097 |
scales | +0.097 |
Neg logit contributions | |
buttocks | -0.116 |
ONY | -0.110 |
Bench | -0.110 |
venth | -0.109 |
igan | -0.109 |
Token | j |
Token ID | 474 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
buttocks | +0.053 |
ONY | +0.050 |
Bench | +0.050 |
igan | +0.050 |
venth | +0.050 |
Neg logit contributions | |
emort | -0.038 |
urable | -0.038 |
Cla | -0.037 |
milo | -0.037 |
scales | -0.036 |
Token | agg |
Token ID | 9460 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
buttocks | +0.412 |
ONY | +0.392 |
Bench | +0.391 |
igan | +0.391 |
venth | +0.389 |
Neg logit contributions | |
emort | -0.426 |
Cla | -0.423 |
urable | -0.419 |
scales | -0.417 |
milo | -0.409 |
Token | ies |
Token ID | 444 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
buttocks | +3.405 |
igan | +3.391 |
ONY | +3.246 |
Bench | +3.228 |
venth | +3.207 |
Neg logit contributions | |
emort | -2.890 |
Cla | -2.885 |
milo | -2.764 |
urable | -2.764 |
Kahn | -2.749 |
Token | but |
Token ID | 475 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
buttocks | +0.676 |
ONY | +0.641 |
Bench | +0.640 |
igan | +0.638 |
venth | +0.637 |
Neg logit contributions | |
emort | -0.602 |
Cla | -0.588 |
urable | -0.586 |
milo | -0.575 |
scales | -0.574 |
Token | I |
Token ID | 314 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
buttocks | +0.989 |
Bench | +0.940 |
ONY | +0.940 |
venth | +0.935 |
igan | +0.930 |
Neg logit contributions | |
emort | -0.922 |
urable | -0.905 |
Cla | -0.897 |
scales | -0.877 |
milo | -0.874 |
Token | have |
Token ID | 423 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
emort | +0.030 |
urable | +0.030 |
Cla | +0.030 |
scales | +0.029 |
milo | +0.029 |
Neg logit contributions | |
buttocks | -0.031 |
venth | -0.029 |
ONY | -0.029 |
Bench | -0.029 |
igan | -0.029 |
Token | no |
Token ID | 645 |
Feature activation | +9.38e-01 |
Pos logit contributions | |
emort | +0.448 |
urable | +0.441 |
Cla | +0.436 |
scales | +0.430 |
milo | +0.426 |
Neg logit contributions | |
buttocks | -0.542 |
Bench | -0.518 |
ONY | -0.516 |
venth | -0.513 |
igan | -0.512 |
Token | bank |
Token ID | 3331 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
buttocks | +0.311 |
Bench | +0.299 |
ONY | +0.298 |
igan | +0.297 |
venth | +0.297 |
Neg logit contributions | |
emort | -0.198 |
urable | -0.195 |
Cla | -0.194 |
scales | -0.190 |
milo | -0.187 |
Token | loans |
Token ID | 10021 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
buttocks | +0.865 |
ONY | +0.843 |
Bench | +0.837 |
igan | +0.830 |
Newsletter | +0.824 |
Neg logit contributions | |
emort | -0.733 |
milo | -0.693 |
Cla | -0.691 |
urable | -0.686 |
Kahn | -0.685 |
Token | and |
Token ID | 290 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
buttocks | +1.161 |
igan | +1.106 |
ONY | +1.105 |
Bench | +1.097 |
venth | +1.091 |
Neg logit contributions | |
emort | -1.164 |
Cla | -1.136 |
urable | -1.133 |
scales | -1.121 |
milo | -1.111 |
Token | the |
Token ID | 262 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
buttocks | +0.438 |
Bench | +0.415 |
ONY | +0.415 |
venth | +0.414 |
igan | +0.413 |
Neg logit contributions | |
emort | -0.346 |
urable | -0.341 |
Cla | -0.338 |
milo | -0.333 |
scales | -0.332 |
Token | cac |
Token ID | 40428 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
buttocks | +0.015 |
Bench | +0.015 |
ONY | +0.015 |
venth | +0.015 |
igan | +0.015 |
Neg logit contributions | |
emort | -0.013 |
urable | -0.013 |
Cla | -0.012 |
milo | -0.012 |
scales | -0.012 |
Token | het |
Token ID | 3202 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
buttocks | +0.829 |
ONY | +0.797 |
Bench | +0.793 |
venth | +0.791 |
igan | +0.787 |
Neg logit contributions | |
emort | -0.657 |
Cla | -0.651 |
urable | -0.649 |
scales | -0.644 |
milo | -0.624 |
Token | of |
Token ID | 286 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
buttocks | +2.783 |
igan | +2.631 |
ONY | +2.608 |
Bench | +2.603 |
Newsletter | +2.587 |
Neg logit contributions | |
emort | -3.538 |
urable | -3.425 |
milo | -3.423 |
Cla | -3.414 |
scales | -3.383 |
Token | having |
Token ID | 1719 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
buttocks | +0.924 |
ONY | +0.872 |
Bench | +0.869 |
venth | +0.863 |
igan | +0.862 |
Neg logit contributions | |
emort | -0.853 |
urable | -0.840 |
Cla | -0.829 |
milo | -0.816 |
scales | -0.812 |
Token | planes |
Token ID | 13016 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
buttocks | +0.283 |
ONY | +0.269 |
Bench | +0.268 |
venth | +0.266 |
igan | +0.266 |
Neg logit contributions | |
emort | -0.224 |
urable | -0.221 |
Cla | -0.220 |
scales | -0.217 |
milo | -0.214 |
Token | land |
Token ID | 1956 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
buttocks | +0.716 |
ONY | +0.680 |
Bench | +0.677 |
igan | +0.674 |
venth | +0.674 |
Neg logit contributions | |
emort | -0.682 |
Cla | -0.670 |
urable | -0.669 |
scales | -0.657 |
milo | -0.649 |
Token | in |
Token ID | 287 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
buttocks | +0.326 |
ONY | +0.311 |
Bench | +0.310 |
venth | +0.309 |
igan | +0.308 |
Neg logit contributions | |
emort | -0.300 |
urable | -0.297 |
Cla | -0.297 |
scales | -0.294 |
milo | -0.286 |
Token | precisely |
Token ID | 10582 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
buttocks | +0.054 |
ONY | +0.052 |
Bench | +0.051 |
venth | +0.051 |
igan | +0.051 |
Neg logit contributions | |
emort | -0.045 |
urable | -0.044 |
Cla | -0.044 |
scales | -0.043 |
milo | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
emort | +1.041 |
Cla | +1.041 |
urable | +1.036 |
scales | +1.029 |
Kahn | +0.996 |
Neg logit contributions | |
buttocks | -1.377 |
ONY | -1.332 |
Bench | -1.324 |
venth | -1.322 |
igan | -1.309 |
Token | setting |
Token ID | 4634 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
buttocks | +0.787 |
venth | +0.749 |
ONY | +0.748 |
igan | +0.747 |
Bench | +0.743 |
Neg logit contributions | |
emort | -0.364 |
urable | -0.359 |
Cla | -0.347 |
milo | -0.341 |
scales | -0.335 |
Token | you |
Token ID | 345 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
buttocks | +0.409 |
ONY | +0.390 |
Bench | +0.387 |
venth | +0.386 |
igan | +0.384 |
Neg logit contributions | |
emort | -0.367 |
urable | -0.363 |
Cla | -0.359 |
scales | -0.355 |
milo | -0.350 |
Token | might |
Token ID | 1244 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
buttocks | +0.096 |
ONY | +0.092 |
venth | +0.092 |
igan | +0.091 |
Bench | +0.091 |
Neg logit contributions | |
emort | -0.084 |
Cla | -0.083 |
urable | -0.083 |
scales | -0.082 |
milo | -0.080 |
Token | expect |
Token ID | 1607 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
buttocks | +0.496 |
ONY | +0.470 |
venth | +0.469 |
igan | +0.467 |
Bench | +0.466 |
Neg logit contributions | |
emort | -0.475 |
Cla | -0.470 |
urable | -0.469 |
scales | -0.463 |
milo | -0.455 |
Token | for |
Token ID | 329 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
buttocks | +3.087 |
ONY | +2.900 |
igan | +2.880 |
Bench | +2.829 |
venth | +2.801 |
Neg logit contributions | |
emort | -3.246 |
urable | -3.196 |
milo | -3.074 |
Cla | -3.073 |
scales | -3.053 |
Token | a |
Token ID | 257 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
emort | +0.200 |
urable | +0.198 |
Cla | +0.197 |
scales | +0.195 |
milo | +0.189 |
Neg logit contributions | |
buttocks | -0.290 |
ONY | -0.278 |
Bench | -0.277 |
venth | -0.276 |
igan | -0.275 |
Token | home |
Token ID | 1363 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
emort | +0.010 |
urable | +0.010 |
Cla | +0.010 |
scales | +0.010 |
milo | +0.009 |
Neg logit contributions | |
buttocks | -0.011 |
ONY | -0.011 |
Bench | -0.010 |
venth | -0.010 |
igan | -0.010 |
Token | entertaining |
Token ID | 17774 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
buttocks | +0.806 |
ONY | +0.766 |
venth | +0.766 |
Bench | +0.761 |
igan | +0.759 |
Neg logit contributions | |
emort | -0.793 |
Cla | -0.779 |
urable | -0.773 |
scales | -0.764 |
milo | -0.756 |
Token | show |
Token ID | 905 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
buttocks | +0.091 |
ONY | +0.088 |
venth | +0.087 |
igan | +0.087 |
Bench | +0.087 |
Neg logit contributions | |
emort | -0.068 |
urable | -0.067 |
Cla | -0.067 |
scales | -0.066 |
milo | -0.064 |
Token | find |
Token ID | 1064 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
buttocks | +0.967 |
ONY | +0.925 |
igan | +0.920 |
venth | +0.909 |
Bench | +0.908 |
Neg logit contributions | |
emort | -0.905 |
urable | -0.893 |
Cla | -0.883 |
milo | -0.878 |
scales | -0.857 |
Token | out |
Token ID | 503 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
buttocks | +0.374 |
ONY | +0.357 |
igan | +0.355 |
Bench | +0.355 |
venth | +0.352 |
Neg logit contributions | |
emort | -0.417 |
urable | -0.412 |
Cla | -0.408 |
scales | -0.403 |
milo | -0.401 |
Token | what |
Token ID | 644 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
emort | +0.348 |
urable | +0.345 |
Cla | +0.342 |
scales | +0.338 |
milo | +0.331 |
Neg logit contributions | |
buttocks | -0.447 |
ONY | -0.427 |
Bench | -0.426 |
igan | -0.425 |
venth | -0.424 |
Token | we |
Token ID | 356 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
urable | +0.124 |
emort | +0.123 |
milo | +0.118 |
Cla | +0.118 |
scales | +0.116 |
Neg logit contributions | |
buttocks | -0.185 |
ONY | -0.177 |
igan | -0.177 |
Bench | -0.176 |
venth | -0.176 |
Token | can |
Token ID | 460 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
buttocks | +0.168 |
igan | +0.159 |
venth | +0.159 |
ONY | +0.159 |
Bench | +0.159 |
Neg logit contributions | |
emort | -0.161 |
urable | -0.159 |
Cla | -0.158 |
scales | -0.156 |
milo | -0.153 |
Token | expect |
Token ID | 1607 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
buttocks | +0.842 |
ONY | +0.794 |
Bench | +0.791 |
igan | +0.791 |
venth | +0.787 |
Neg logit contributions | |
emort | -1.161 |
Cla | -1.152 |
urable | -1.149 |
scales | -1.137 |
milo | -1.125 |
Token | from |
Token ID | 422 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
buttocks | +2.739 |
igan | +2.622 |
ONY | +2.567 |
Bench | +2.554 |
secondly | +2.517 |
Neg logit contributions | |
emort | -3.412 |
urable | -3.371 |
milo | -3.327 |
Cla | -3.313 |
scales | -3.305 |
Token | him |
Token ID | 683 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
emort | +0.226 |
urable | +0.224 |
Cla | +0.220 |
scales | +0.218 |
milo | +0.213 |
Neg logit contributions | |
buttocks | -0.384 |
ONY | -0.370 |
Bench | -0.367 |
venth | -0.367 |
igan | -0.366 |
Token | this |
Token ID | 428 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
buttocks | +0.084 |
ONY | +0.080 |
igan | +0.080 |
Bench | +0.080 |
venth | +0.079 |
Neg logit contributions | |
emort | -0.083 |
urable | -0.082 |
Cla | -0.082 |
scales | -0.081 |
milo | -0.079 |
Token | season |
Token ID | 1622 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
emort | +0.068 |
urable | +0.068 |
Cla | +0.067 |
scales | +0.067 |
milo | +0.065 |
Neg logit contributions | |
buttocks | -0.069 |
ONY | -0.066 |
venth | -0.065 |
Bench | -0.065 |
igan | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
buttocks | +1.354 |
igan | +1.294 |
Bench | +1.286 |
ONY | +1.282 |
venth | +1.275 |
Neg logit contributions | |
emort | -1.371 |
urable | -1.350 |
Cla | -1.346 |
scales | -1.319 |
milo | -1.311 |
Token | targeting |
Token ID | 10822 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
emort | +0.141 |
urable | +0.140 |
Cla | +0.139 |
scales | +0.137 |
milo | +0.135 |
Neg logit contributions | |
buttocks | -0.159 |
ONY | -0.151 |
Bench | -0.151 |
venth | -0.150 |
igan | -0.150 |
Token | and |
Token ID | 290 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
buttocks | +0.687 |
ONY | +0.656 |
igan | +0.650 |
Bench | +0.650 |
venth | +0.647 |
Neg logit contributions | |
emort | -0.599 |
urable | -0.588 |
Cla | -0.585 |
scales | -0.577 |
milo | -0.566 |
Token | abuse |
Token ID | 5076 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
buttocks | +0.038 |
ONY | +0.037 |
Bench | +0.036 |
venth | +0.036 |
igan | +0.036 |
Neg logit contributions | |
emort | -0.033 |
urable | -0.032 |
Cla | -0.032 |
scales | -0.032 |
milo | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
buttocks | +1.645 |
ONY | +1.577 |
igan | +1.555 |
Bench | +1.551 |
Newsletter | +1.537 |
Neg logit contributions | |
emort | -1.741 |
urable | -1.703 |
Cla | -1.685 |
scales | -1.663 |
milo | -1.660 |
Token | No |
Token ID | 1400 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
buttocks | +0.142 |
ONY | +0.137 |
Bench | +0.136 |
venth | +0.135 |
igan | +0.135 |
Neg logit contributions | |
emort | -0.128 |
urable | -0.127 |
Cla | -0.125 |
scales | -0.124 |
milo | -0.121 |
Token | wonder |
Token ID | 4240 |
Feature activation | -6.69e+00 |
Pos logit contributions | |
buttocks | +1.227 |
ONY | +1.182 |
venth | +1.161 |
Bench | +1.154 |
igan | +1.154 |
Neg logit contributions | |
emort | -0.968 |
urable | -0.963 |
Cla | -0.955 |
scales | -0.926 |
milo | -0.921 |
Token | the |
Token ID | 262 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
Cla | +3.606 |
scales | +3.553 |
emort | +3.499 |
urable | +3.455 |
Kahn | +3.442 |
Neg logit contributions | |
buttocks | -4.806 |
venth | -4.671 |
Bench | -4.609 |
ONY | -4.573 |
igan | -4.509 |
Token | Obama |
Token ID | 2486 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
emort | +0.338 |
urable | +0.335 |
Cla | +0.329 |
scales | +0.322 |
milo | +0.319 |
Neg logit contributions | |
buttocks | -0.590 |
ONY | -0.565 |
Bench | -0.563 |
venth | -0.560 |
igan | -0.559 |
Token | IRS |
Token ID | 14440 |
Feature activation | -9.53e-02 |
Pos logit contributions | |
buttocks | +0.871 |
ONY | +0.835 |
igan | +0.832 |
Bench | +0.829 |
venth | +0.826 |
Neg logit contributions | |
emort | -0.516 |
urable | -0.516 |
Cla | -0.509 |
scales | -0.497 |
milo | -0.495 |
Token | has |
Token ID | 468 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
emort | +0.054 |
urable | +0.053 |
Cla | +0.052 |
scales | +0.052 |
milo | +0.051 |
Neg logit contributions | |
buttocks | -0.069 |
ONY | -0.066 |
Bench | -0.065 |
igan | -0.065 |
venth | -0.065 |
Token | had |
Token ID | 550 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
buttocks | +0.085 |
ONY | +0.081 |
igan | +0.080 |
Bench | +0.080 |
venth | +0.079 |
Neg logit contributions | |
emort | -0.072 |
urable | -0.071 |
Cla | -0.071 |
scales | -0.069 |
milo | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
buttocks | +0.015 |
venth | +0.014 |
ONY | +0.014 |
Bench | +0.014 |
igan | +0.014 |
Neg logit contributions | |
emort | -0.012 |
urable | -0.012 |
Cla | -0.012 |
scales | -0.012 |
milo | -0.011 |
Token | about |
Token ID | 546 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
emort | +0.873 |
urable | +0.863 |
Cla | +0.862 |
scales | +0.853 |
milo | +0.831 |
Neg logit contributions | |
buttocks | -1.075 |
ONY | -1.026 |
Bench | -1.022 |
venth | -1.020 |
igan | -1.014 |
Token | homosexuality |
Token ID | 19822 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
emort | +0.254 |
urable | +0.250 |
Cla | +0.248 |
scales | +0.245 |
milo | +0.242 |
Neg logit contributions | |
buttocks | -0.304 |
ONY | -0.288 |
Bench | -0.287 |
venth | -0.286 |
igan | -0.285 |
Token |
|
Token ID | 198 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
buttocks | +0.744 |
ONY | +0.709 |
Bench | +0.704 |
venth | +0.701 |
igan | +0.696 |
Neg logit contributions | |
emort | -0.705 |
urable | -0.696 |
Cla | -0.693 |
scales | -0.686 |
milo | -0.674 |
Token |
|
Token ID | 198 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
buttocks | +0.590 |
Bench | +0.571 |
ONY | +0.571 |
venth | +0.564 |
Newsletter | +0.561 |
Neg logit contributions | |
emort | -0.479 |
Cla | -0.467 |
urable | -0.467 |
scales | -0.464 |
Kahn | -0.448 |
Token | Reviewer |
Token ID | 35407 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
buttocks | +0.204 |
Bench | +0.197 |
ONY | +0.196 |
venth | +0.196 |
igan | +0.194 |
Neg logit contributions | |
emort | -0.182 |
Cla | -0.179 |
urable | -0.177 |
scales | -0.177 |
Kahn | -0.171 |
Token | : |
Token ID | 25 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
Cla | +3.138 |
scales | +3.133 |
urable | +3.042 |
eq | +3.016 |
emort | +3.009 |
Neg logit contributions | |
buttocks | -3.947 |
ONY | -3.807 |
Newsletter | -3.753 |
Bench | -3.753 |
venth | -3.719 |
Token | k |
Token ID | 479 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
emort | +0.309 |
urable | +0.306 |
Cla | +0.303 |
scales | +0.300 |
milo | +0.295 |
Neg logit contributions | |
buttocks | -0.328 |
ONY | -0.312 |
Bench | -0.311 |
venth | -0.310 |
igan | -0.308 |
Token | imp |
Token ID | 11011 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
buttocks | +1.067 |
ONY | +1.028 |
venth | +1.027 |
igan | +1.026 |
Bench | +1.019 |
Neg logit contributions | |
Cla | -0.865 |
emort | -0.865 |
urable | -0.854 |
scales | -0.853 |
Kahn | -0.831 |
Token | unk |
Token ID | 2954 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
emort | +0.162 |
Cla | +0.161 |
urable | +0.161 |
scales | +0.159 |
milo | +0.154 |
Neg logit contributions | |
buttocks | -0.190 |
ONY | -0.182 |
venth | -0.181 |
Bench | -0.180 |
igan | -0.180 |
Token | rock |
Token ID | 10823 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
emort | +0.217 |
Cla | +0.216 |
urable | +0.216 |
scales | +0.213 |
Kahn | +0.207 |
Neg logit contributions | |
buttocks | -0.251 |
ONY | -0.240 |
igan | -0.239 |
venth | -0.239 |
Bench | -0.239 |
Token | us |
Token ID | 514 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
buttocks | +1.117 |
igan | +1.051 |
ONY | +1.050 |
Bench | +1.044 |
venth | +1.038 |
Neg logit contributions | |
emort | -1.170 |
urable | -1.167 |
scales | -1.144 |
Cla | -1.140 |
milo | -1.133 |
Token | . |
Token ID | 13 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
emort | +0.263 |
urable | +0.261 |
Cla | +0.261 |
scales | +0.258 |
milo | +0.252 |
Neg logit contributions | |
buttocks | -0.291 |
ONY | -0.277 |
Bench | -0.276 |
venth | -0.276 |
igan | -0.275 |
Token | " |
Token ID | 366 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
buttocks | +0.280 |
ONY | +0.271 |
Bench | +0.270 |
Newsletter | +0.268 |
igan | +0.267 |
Neg logit contributions | |
emort | -0.247 |
urable | -0.244 |
scales | -0.237 |
milo | -0.236 |
Cla | -0.233 |
Token | Don |
Token ID | 3987 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
emort | +0.195 |
urable | +0.192 |
Cla | +0.191 |
scales | +0.190 |
milo | +0.184 |
Neg logit contributions | |
buttocks | -0.242 |
Bench | -0.235 |
ONY | -0.235 |
venth | -0.232 |
Newsletter | -0.232 |
Token | 't |
Token ID | 470 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
buttocks | +0.631 |
igan | +0.589 |
Bench | +0.588 |
venth | +0.588 |
ONY | +0.586 |
Neg logit contributions | |
emort | -1.292 |
urable | -1.269 |
Cla | -1.267 |
scales | -1.263 |
milo | -1.248 |
Token | worry |
Token ID | 5490 |
Feature activation | -5.48e+00 |
Pos logit contributions | |
emort | +0.058 |
urable | +0.057 |
Cla | +0.057 |
milo | +0.056 |
scales | +0.056 |
Neg logit contributions | |
buttocks | -0.071 |
ONY | -0.067 |
venth | -0.067 |
igan | -0.067 |
Bench | -0.066 |
Token | , |
Token ID | 11 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
Cla | +3.018 |
emort | +2.968 |
scales | +2.942 |
urable | +2.938 |
Kahn | +2.872 |
Neg logit contributions | |
buttocks | -3.923 |
ONY | -3.797 |
venth | -3.782 |
Bench | -3.781 |
igan | -3.695 |
Token | I |
Token ID | 314 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
emort | +0.295 |
urable | +0.293 |
Cla | +0.288 |
scales | +0.286 |
milo | +0.281 |
Neg logit contributions | |
buttocks | -0.358 |
ONY | -0.341 |
Bench | -0.338 |
venth | -0.338 |
igan | -0.337 |
Token | 'm |
Token ID | 1101 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
emort | +0.339 |
urable | +0.334 |
Cla | +0.333 |
scales | +0.329 |
milo | +0.323 |
Neg logit contributions | |
buttocks | -0.367 |
ONY | -0.348 |
venth | -0.347 |
igan | -0.345 |
Bench | -0.345 |
Token | not |
Token ID | 407 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
buttocks | +0.544 |
ONY | +0.512 |
Bench | +0.511 |
venth | +0.510 |
igan | +0.508 |
Neg logit contributions | |
emort | -0.585 |
urable | -0.578 |
Cla | -0.568 |
scales | -0.564 |
milo | -0.560 |
Token | talking |
Token ID | 3375 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
emort | +0.827 |
Cla | +0.821 |
urable | +0.821 |
scales | +0.814 |
milo | +0.791 |
Neg logit contributions | |
buttocks | -0.800 |
ONY | -0.762 |
Bench | -0.760 |
venth | -0.758 |
igan | -0.752 |
Token | popular |
Token ID | 2968 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
buttocks | +0.198 |
Bench | +0.187 |
ONY | +0.187 |
venth | +0.186 |
igan | +0.186 |
Neg logit contributions | |
emort | -0.192 |
urable | -0.189 |
Cla | -0.188 |
scales | -0.186 |
milo | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
buttocks | +2.072 |
igan | +1.976 |
Newsletter | +1.934 |
Bench | +1.932 |
venth | +1.927 |
Neg logit contributions | |
emort | -2.202 |
urable | -2.194 |
Cla | -2.156 |
scales | -2.140 |
milo | -2.120 |
Token | Future |
Token ID | 10898 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
emort | +0.043 |
urable | +0.042 |
Cla | +0.042 |
scales | +0.042 |
milo | +0.040 |
Neg logit contributions | |
buttocks | -0.048 |
ONY | -0.046 |
Bench | -0.046 |
venth | -0.046 |
igan | -0.046 |
Token | Hall |
Token ID | 4789 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
emort | +0.686 |
urable | +0.678 |
Cla | +0.676 |
scales | +0.669 |
milo | +0.652 |
Neg logit contributions | |
buttocks | -0.862 |
ONY | -0.823 |
Bench | -0.822 |
venth | -0.818 |
igan | -0.815 |
Token | of |
Token ID | 286 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
emort | +1.387 |
Cla | +1.379 |
urable | +1.376 |
scales | +1.366 |
milo | +1.328 |
Neg logit contributions | |
buttocks | -1.306 |
ONY | -1.243 |
Bench | -1.238 |
venth | -1.235 |
igan | -1.223 |
Token | Fame |
Token ID | 18864 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
buttocks | +0.119 |
Bench | +0.115 |
venth | +0.115 |
ONY | +0.115 |
igan | +0.114 |
Neg logit contributions | |
emort | -0.110 |
urable | -0.108 |
scales | -0.106 |
Cla | -0.105 |
eq | -0.103 |
Token | slug |
Token ID | 31065 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
Cla | +2.743 |
scales | +2.673 |
emort | +2.672 |
urable | +2.658 |
Kahn | +2.585 |
Neg logit contributions | |
buttocks | -4.187 |
ONY | -4.063 |
venth | -4.041 |
Bench | -4.024 |
igan | -4.018 |
Token | ger |
Token ID | 1362 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
emort | +0.408 |
urable | +0.404 |
Cla | +0.403 |
scales | +0.399 |
milo | +0.388 |
Neg logit contributions | |
buttocks | -0.567 |
Bench | -0.543 |
ONY | -0.543 |
venth | -0.541 |
igan | -0.538 |
Token | Stan |
Token ID | 7299 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
emort | +0.088 |
urable | +0.087 |
Cla | +0.087 |
scales | +0.086 |
milo | +0.084 |
Neg logit contributions | |
buttocks | -0.099 |
ONY | -0.094 |
Bench | -0.094 |
venth | -0.094 |
igan | -0.093 |
Token | Mus |
Token ID | 2629 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
buttocks | +0.220 |
igan | +0.211 |
Bench | +0.210 |
venth | +0.210 |
ONY | +0.208 |
Neg logit contributions | |
emort | -0.160 |
urable | -0.157 |
scales | -0.155 |
Cla | -0.153 |
milo | -0.150 |
Token | ial |
Token ID | 498 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
emort | +0.984 |
urable | +0.975 |
Cla | +0.975 |
scales | +0.966 |
milo | +0.940 |
Neg logit contributions | |
buttocks | -0.971 |
ONY | -0.925 |
Bench | -0.922 |
venth | -0.919 |
igan | -0.913 |
Token | National |
Token ID | 2351 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
buttocks | +0.367 |
ONY | +0.348 |
Bench | +0.347 |
venth | +0.345 |
igan | +0.342 |
Neg logit contributions | |
emort | -0.357 |
urable | -0.352 |
scales | -0.346 |
Cla | -0.344 |
milo | -0.339 |
Token | Women |
Token ID | 6926 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
buttocks | +0.299 |
venth | +0.281 |
Bench | +0.281 |
ONY | +0.281 |
igan | +0.280 |
Neg logit contributions | |
emort | -0.307 |
urable | -0.303 |
scales | -0.298 |
milo | -0.294 |
Cla | -0.294 |
Token | 's |
Token ID | 338 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
buttocks | +0.926 |
Bench | +0.883 |
venth | +0.881 |
ONY | +0.880 |
igan | +0.875 |
Neg logit contributions | |
emort | -0.832 |
urable | -0.814 |
scales | -0.809 |
Cla | -0.807 |
milo | -0.790 |
Token | Hall |
Token ID | 4789 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
emort | +0.350 |
urable | +0.344 |
scales | +0.340 |
Cla | +0.340 |
milo | +0.335 |
Neg logit contributions | |
buttocks | -0.373 |
Bench | -0.354 |
venth | -0.353 |
ONY | -0.353 |
igan | -0.350 |
Token | of |
Token ID | 286 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
Cla | +1.345 |
emort | +1.341 |
urable | +1.338 |
scales | +1.330 |
Kahn | +1.292 |
Neg logit contributions | |
buttocks | -1.313 |
ONY | -1.250 |
Bench | -1.241 |
venth | -1.239 |
igan | -1.228 |
Token | Fame |
Token ID | 18864 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
buttocks | +0.287 |
Bench | +0.279 |
igan | +0.279 |
venth | +0.278 |
ONY | +0.278 |
Neg logit contributions | |
emort | -0.126 |
urable | -0.121 |
scales | -0.120 |
Cla | -0.117 |
milo | -0.115 |
Token | .[ |
Token ID | 3693 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
Cla | +3.182 |
urable | +3.123 |
emort | +3.109 |
scales | +3.099 |
Kahn | +3.027 |
Neg logit contributions | |
buttocks | -3.668 |
ONY | -3.495 |
Bench | -3.490 |
venth | -3.484 |
igan | -3.473 |
Token | 49 |
Token ID | 2920 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
emort | +0.222 |
Cla | +0.218 |
urable | +0.218 |
scales | +0.216 |
milo | +0.210 |
Neg logit contributions | |
buttocks | -0.251 |
Bench | -0.241 |
ONY | -0.241 |
venth | -0.240 |
igan | -0.237 |
Token | ] |
Token ID | 60 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
buttocks | +0.568 |
ONY | +0.538 |
Bench | +0.537 |
venth | +0.536 |
igan | +0.531 |
Neg logit contributions | |
emort | -1.015 |
urable | -1.005 |
Cla | -0.996 |
scales | -0.992 |
milo | -0.982 |
Token | On |
Token ID | 1550 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
emort | +0.914 |
urable | +0.904 |
Cla | +0.898 |
scales | +0.892 |
milo | +0.866 |
Neg logit contributions | |
buttocks | -1.246 |
ONY | -1.198 |
Bench | -1.193 |
venth | -1.189 |
igan | -1.183 |
Token | July |
Token ID | 2901 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
buttocks | +0.136 |
ONY | +0.127 |
venth | +0.127 |
Bench | +0.126 |
igan | +0.126 |
Neg logit contributions | |
emort | -0.140 |
urable | -0.136 |
milo | -0.133 |
Cla | -0.133 |
scales | -0.131 |
Token | uck |
Token ID | 1347 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
emort | +0.654 |
Cla | +0.652 |
urable | +0.647 |
scales | +0.645 |
milo | +0.629 |
Neg logit contributions | |
buttocks | -0.739 |
ONY | -0.711 |
Bench | -0.710 |
venth | -0.705 |
igan | -0.705 |
Token | ley |
Token ID | 1636 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
buttocks | +0.605 |
Bench | +0.581 |
igan | +0.578 |
venth | +0.570 |
ONY | +0.566 |
Neg logit contributions | |
urable | -0.759 |
Cla | -0.753 |
emort | -0.753 |
scales | -0.746 |
milo | -0.742 |
Token | NBA |
Token ID | 32470 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
emort | +0.311 |
urable | +0.308 |
Cla | +0.308 |
scales | +0.305 |
milo | +0.297 |
Neg logit contributions | |
buttocks | -0.336 |
Bench | -0.322 |
ONY | -0.322 |
venth | -0.319 |
igan | -0.318 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.66e-01 |
Pos logit contributions | |
emort | +0.767 |
urable | +0.759 |
Cla | +0.758 |
scales | +0.749 |
milo | +0.723 |
Neg logit contributions | |
buttocks | -1.043 |
ONY | -0.998 |
venth | -0.993 |
Bench | -0.989 |
igan | -0.985 |
Token | Ted |
Token ID | 38972 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
emort | +0.520 |
Cla | +0.514 |
urable | +0.512 |
scales | +0.510 |
Kahn | +0.493 |
Neg logit contributions | |
buttocks | -0.585 |
Bench | -0.565 |
ONY | -0.564 |
venth | -0.559 |
igan | -0.555 |
Token | Nug |
Token ID | 45777 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
emort | +0.168 |
urable | +0.165 |
Cla | +0.163 |
scales | +0.162 |
milo | +0.158 |
Neg logit contributions | |
buttocks | -0.307 |
Bench | -0.296 |
ONY | -0.295 |
venth | -0.294 |
igan | -0.294 |
Token | ent |
Token ID | 298 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
urable | +1.797 |
Cla | +1.730 |
scales | +1.727 |
emort | +1.682 |
eq | +1.613 |
Neg logit contributions | |
buttocks | -4.933 |
Bench | -4.804 |
ONY | -4.801 |
Newsletter | -4.757 |
venth | -4.734 |
Token | promised |
Token ID | 8072 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
buttocks | +0.754 |
ONY | +0.715 |
igan | +0.712 |
Bench | +0.710 |
venth | +0.709 |
Neg logit contributions | |
urable | -0.630 |
emort | -0.628 |
Cla | -0.617 |
scales | -0.612 |
milo | -0.608 |
Token | Thursday |
Token ID | 3635 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
buttocks | +0.152 |
ONY | +0.145 |
Bench | +0.143 |
venth | +0.143 |
igan | +0.143 |
Neg logit contributions | |
emort | -0.129 |
urable | -0.128 |
Cla | -0.127 |
scales | -0.125 |
milo | -0.123 |
Token | that |
Token ID | 326 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
emort | +0.312 |
urable | +0.309 |
Cla | +0.308 |
scales | +0.305 |
milo | +0.297 |
Neg logit contributions | |
buttocks | -0.386 |
ONY | -0.369 |
Bench | -0.367 |
venth | -0.367 |
igan | -0.365 |
Token | he |
Token ID | 339 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
buttocks | +0.660 |
ONY | +0.629 |
Bench | +0.626 |
igan | +0.622 |
venth | +0.621 |
Neg logit contributions | |
emort | -0.407 |
urable | -0.402 |
Cla | -0.390 |
scales | -0.387 |
milo | -0.380 |
Token | it |
Token ID | 340 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
emort | +0.211 |
urable | +0.208 |
Cla | +0.205 |
scales | +0.203 |
milo | +0.199 |
Neg logit contributions | |
buttocks | -0.311 |
ONY | -0.299 |
venth | -0.299 |
Bench | -0.298 |
igan | -0.297 |
Token | at |
Token ID | 379 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
buttocks | +0.260 |
ONY | +0.249 |
Bench | +0.248 |
venth | +0.248 |
igan | +0.247 |
Neg logit contributions | |
emort | -0.228 |
urable | -0.225 |
Cla | -0.225 |
scales | -0.222 |
milo | -0.217 |
Token | any |
Token ID | 597 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
emort | +0.284 |
urable | +0.281 |
Cla | +0.278 |
scales | +0.274 |
milo | +0.270 |
Neg logit contributions | |
buttocks | -0.379 |
ONY | -0.364 |
Bench | -0.363 |
venth | -0.362 |
igan | -0.361 |
Token | time |
Token ID | 640 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
buttocks | +1.407 |
venth | +1.363 |
igan | +1.359 |
Bench | +1.359 |
ONY | +1.356 |
Neg logit contributions | |
emort | -0.605 |
urable | -0.586 |
Cla | -0.565 |
milo | -0.562 |
scales | -0.556 |
Token | without |
Token ID | 1231 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
emort | +0.291 |
urable | +0.288 |
Cla | +0.288 |
scales | +0.285 |
milo | +0.277 |
Neg logit contributions | |
buttocks | -0.350 |
ONY | -0.335 |
Bench | -0.334 |
venth | -0.334 |
igan | -0.332 |
Token | worrying |
Token ID | 18916 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
emort | +0.078 |
urable | +0.078 |
Cla | +0.077 |
scales | +0.076 |
milo | +0.075 |
Neg logit contributions | |
buttocks | -0.084 |
ONY | -0.080 |
Bench | -0.080 |
venth | -0.080 |
igan | -0.079 |
Token | about |
Token ID | 546 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
Cla | +2.349 |
scales | +2.320 |
emort | +2.281 |
urable | +2.259 |
Kahn | +2.178 |
Neg logit contributions | |
buttocks | -4.326 |
ONY | -4.146 |
venth | -4.129 |
Bench | -4.116 |
Newsletter | -4.044 |
Token | the |
Token ID | 262 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
emort | +0.529 |
urable | +0.524 |
Cla | +0.521 |
scales | +0.516 |
milo | +0.503 |
Neg logit contributions | |
buttocks | -0.636 |
ONY | -0.610 |
Bench | -0.607 |
venth | -0.607 |
igan | -0.604 |
Token | negative |
Token ID | 4633 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
emort | +0.449 |
urable | +0.444 |
Cla | +0.436 |
scales | +0.433 |
milo | +0.426 |
Neg logit contributions | |
buttocks | -0.569 |
ONY | -0.547 |
venth | -0.545 |
Bench | -0.543 |
igan | -0.542 |
Token | downside |
Token ID | 25320 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
emort | +0.583 |
urable | +0.575 |
Cla | +0.568 |
scales | +0.559 |
milo | +0.540 |
Neg logit contributions | |
buttocks | -1.435 |
ONY | -1.387 |
Bench | -1.384 |
venth | -1.382 |
igan | -1.379 |
Token | . |
Token ID | 13 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
emort | +0.892 |
urable | +0.884 |
Cla | +0.882 |
scales | +0.873 |
milo | +0.851 |
Neg logit contributions | |
buttocks | -0.913 |
ONY | -0.872 |
Bench | -0.868 |
venth | -0.866 |
igan | -0.861 |
Token | slightly |
Token ID | 4622 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
buttocks | +0.485 |
ONY | +0.462 |
Bench | +0.460 |
venth | +0.458 |
igan | +0.457 |
Neg logit contributions | |
emort | -0.423 |
urable | -0.418 |
Cla | -0.417 |
scales | -0.412 |
milo | -0.405 |
Token | incomplete |
Token ID | 17503 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
buttocks | +0.315 |
ONY | +0.299 |
Bench | +0.298 |
venth | +0.297 |
igan | +0.296 |
Neg logit contributions | |
emort | -0.296 |
urable | -0.292 |
Cla | -0.291 |
scales | -0.287 |
milo | -0.282 |
Token | , |
Token ID | 11 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
buttocks | +1.964 |
igan | +1.887 |
ONY | +1.863 |
Newsletter | +1.852 |
venth | +1.837 |
Neg logit contributions | |
emort | -2.110 |
urable | -2.093 |
Cla | -2.062 |
scales | -2.050 |
milo | -2.049 |
Token | don |
Token ID | 836 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
emort | +0.076 |
urable | +0.075 |
Cla | +0.073 |
scales | +0.073 |
milo | +0.072 |
Neg logit contributions | |
buttocks | -0.087 |
ONY | -0.083 |
Bench | -0.082 |
igan | -0.082 |
venth | -0.081 |
Token | 't |
Token ID | 470 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
buttocks | +0.537 |
Bench | +0.503 |
ONY | +0.502 |
venth | +0.499 |
igan | +0.498 |
Neg logit contributions | |
emort | -1.051 |
urable | -1.035 |
Cla | -1.033 |
scales | -1.027 |
milo | -1.010 |
Token | worry |
Token ID | 5490 |
Feature activation | -5.22e+00 |
Pos logit contributions | |
emort | +0.232 |
urable | +0.228 |
Cla | +0.228 |
scales | +0.225 |
milo | +0.223 |
Neg logit contributions | |
buttocks | -0.282 |
ONY | -0.267 |
venth | -0.267 |
igan | -0.267 |
Bench | -0.266 |
Token | . |
Token ID | 13 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
Cla | +2.900 |
emort | +2.878 |
urable | +2.851 |
scales | +2.845 |
Kahn | +2.772 |
Neg logit contributions | |
buttocks | -3.709 |
ONY | -3.568 |
venth | -3.546 |
Bench | -3.535 |
igan | -3.455 |
Token | Full |
Token ID | 6462 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
emort | +0.530 |
urable | +0.525 |
Cla | +0.519 |
scales | +0.518 |
milo | +0.504 |
Neg logit contributions | |
buttocks | -0.600 |
ONY | -0.575 |
Bench | -0.572 |
venth | -0.571 |
igan | -0.568 |
Token | details |
Token ID | 3307 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
emort | +0.534 |
urable | +0.533 |
Cla | +0.526 |
scales | +0.521 |
milo | +0.510 |
Neg logit contributions | |
buttocks | -0.567 |
Bench | -0.538 |
ONY | -0.537 |
venth | -0.534 |
igan | -0.532 |
Token | and |
Token ID | 290 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
emort | +0.311 |
urable | +0.306 |
Cla | +0.305 |
scales | +0.301 |
milo | +0.297 |
Neg logit contributions | |
buttocks | -0.325 |
ONY | -0.308 |
Bench | -0.307 |
igan | -0.306 |
venth | -0.306 |
Token | transcript |
Token ID | 14687 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
emort | +0.348 |
urable | +0.345 |
Cla | +0.340 |
scales | +0.333 |
milo | +0.331 |
Neg logit contributions | |
buttocks | -0.446 |
ONY | -0.422 |
Bench | -0.422 |
igan | -0.420 |
venth | -0.419 |
Token |
|
Token ID | 198 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
buttocks | +0.324 |
ONY | +0.317 |
Bench | +0.314 |
igan | +0.313 |
Newsletter | +0.312 |
Neg logit contributions | |
emort | -0.292 |
urable | -0.287 |
scales | -0.280 |
milo | -0.275 |
Cla | -0.275 |
Token |
|
Token ID | 198 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
buttocks | +0.640 |
Bench | +0.619 |
ONY | +0.617 |
igan | +0.614 |
venth | +0.613 |
Neg logit contributions | |
emort | -0.530 |
urable | -0.514 |
Cla | -0.511 |
scales | -0.506 |
milo | -0.497 |
Token | It |
Token ID | 1026 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
Bench | +0.345 |
buttocks | +0.345 |
ONY | +0.342 |
Newsletter | +0.342 |
venth | +0.337 |
Neg logit contributions | |
emort | -0.306 |
scales | -0.297 |
Cla | -0.297 |
urable | -0.297 |
Kahn | -0.284 |
Token | 's |
Token ID | 338 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
buttocks | +0.260 |
venth | +0.248 |
ONY | +0.248 |
Bench | +0.247 |
igan | +0.247 |
Neg logit contributions | |
emort | -0.226 |
urable | -0.223 |
Cla | -0.221 |
scales | -0.215 |
milo | -0.213 |
Token | not |
Token ID | 407 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
buttocks | +0.305 |
ONY | +0.289 |
Bench | +0.288 |
igan | +0.287 |
venth | +0.286 |
Neg logit contributions | |
emort | -0.256 |
urable | -0.249 |
Cla | -0.247 |
scales | -0.244 |
milo | -0.243 |
Token | hard |
Token ID | 1327 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
emort | +0.224 |
urable | +0.219 |
Cla | +0.219 |
scales | +0.216 |
milo | +0.213 |
Neg logit contributions | |
buttocks | -0.258 |
ONY | -0.245 |
venth | -0.243 |
Bench | -0.243 |
igan | -0.242 |
Token | to |
Token ID | 284 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
urable | +2.587 |
scales | +2.582 |
Cla | +2.562 |
emort | +2.543 |
eq | +2.464 |
Neg logit contributions | |
buttocks | -3.958 |
ONY | -3.823 |
venth | -3.801 |
Bench | -3.776 |
Newsletter | -3.724 |
Token | figure |
Token ID | 3785 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
emort | +0.390 |
Cla | +0.384 |
urable | +0.383 |
scales | +0.377 |
milo | +0.371 |
Neg logit contributions | |
buttocks | -0.528 |
ONY | -0.504 |
venth | -0.503 |
Bench | -0.502 |
igan | -0.502 |
Token | out |
Token ID | 503 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
buttocks | +0.011 |
ONY | +0.010 |
igan | +0.010 |
Bench | +0.010 |
venth | +0.010 |
Neg logit contributions | |
emort | -0.012 |
urable | -0.011 |
Cla | -0.011 |
scales | -0.011 |
milo | -0.011 |
Token | which |
Token ID | 543 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
emort | +0.658 |
urable | +0.652 |
Cla | +0.646 |
scales | +0.640 |
milo | +0.629 |
Neg logit contributions | |
buttocks | -0.648 |
ONY | -0.616 |
Bench | -0.613 |
venth | -0.612 |
igan | -0.611 |
Token | of |
Token ID | 286 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
emort | +0.006 |
urable | +0.005 |
Cla | +0.005 |
milo | +0.005 |
scales | +0.005 |
Neg logit contributions | |
buttocks | -0.008 |
ONY | -0.007 |
igan | -0.007 |
venth | -0.007 |
Bench | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
buttocks | +0.366 |
ONY | +0.352 |
Bench | +0.352 |
venth | +0.350 |
igan | +0.347 |
Neg logit contributions | |
emort | -0.319 |
urable | -0.313 |
Cla | -0.309 |
scales | -0.308 |
milo | -0.302 |
Token |
|
Token ID | 198 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
buttocks | +0.838 |
Bench | +0.811 |
ONY | +0.807 |
igan | +0.803 |
venth | +0.803 |
Neg logit contributions | |
emort | -0.697 |
urable | -0.675 |
Cla | -0.671 |
scales | -0.665 |
milo | -0.652 |
Token | And |
Token ID | 1870 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
buttocks | +0.549 |
Bench | +0.539 |
ONY | +0.533 |
venth | +0.530 |
Newsletter | +0.525 |
Neg logit contributions | |
emort | -0.500 |
Cla | -0.490 |
urable | -0.489 |
scales | -0.486 |
milo | -0.470 |
Token | don |
Token ID | 836 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
emort | +0.097 |
urable | +0.096 |
Cla | +0.096 |
scales | +0.095 |
milo | +0.092 |
Neg logit contributions | |
buttocks | -0.113 |
ONY | -0.108 |
Bench | -0.108 |
venth | -0.108 |
igan | -0.107 |
Token | 't |
Token ID | 470 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
buttocks | +0.409 |
ONY | +0.382 |
Bench | +0.382 |
venth | +0.380 |
igan | +0.378 |
Neg logit contributions | |
emort | -0.788 |
urable | -0.777 |
Cla | -0.777 |
scales | -0.772 |
milo | -0.759 |
Token | worry |
Token ID | 5490 |
Feature activation | -5.14e+00 |
Pos logit contributions | |
emort | +0.184 |
urable | +0.179 |
Cla | +0.176 |
milo | +0.175 |
scales | +0.174 |
Neg logit contributions | |
buttocks | -0.236 |
venth | -0.224 |
ONY | -0.224 |
Bench | -0.224 |
igan | -0.223 |
Token | – |
Token ID | 784 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
Cla | +3.010 |
scales | +2.950 |
urable | +2.948 |
emort | +2.945 |
Kahn | +2.887 |
Neg logit contributions | |
buttocks | -3.497 |
ONY | -3.381 |
venth | -3.363 |
Bench | -3.308 |
igan | -3.283 |
Token | you |
Token ID | 345 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
emort | +0.178 |
urable | +0.176 |
Cla | +0.176 |
scales | +0.174 |
milo | +0.170 |
Neg logit contributions | |
buttocks | -0.202 |
ONY | -0.193 |
Bench | -0.193 |
venth | -0.192 |
igan | -0.191 |
Token | will |
Token ID | 481 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
emort | +0.321 |
urable | +0.318 |
Cla | +0.318 |
scales | +0.315 |
milo | +0.306 |
Neg logit contributions | |
buttocks | -0.354 |
ONY | -0.338 |
Bench | -0.337 |
venth | -0.336 |
igan | -0.334 |
Token | once |
Token ID | 1752 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
emort | +0.217 |
urable | +0.214 |
Cla | +0.214 |
scales | +0.211 |
milo | +0.207 |
Neg logit contributions | |
buttocks | -0.255 |
ONY | -0.244 |
Bench | -0.242 |
venth | -0.242 |
igan | -0.241 |
Token | again |
Token ID | 757 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
buttocks | +0.145 |
ONY | +0.139 |
Bench | +0.138 |
venth | +0.138 |
igan | +0.137 |
Neg logit contributions | |
emort | -0.125 |
urable | -0.124 |
Cla | -0.124 |
scales | -0.122 |
milo | -0.119 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
emort | +0.252 |
Cla | +0.250 |
urable | +0.250 |
scales | +0.248 |
milo | +0.237 |
Neg logit contributions | |
buttocks | -0.433 |
ONY | -0.417 |
Bench | -0.416 |
venth | -0.416 |
igan | -0.413 |
Token | boys |
Token ID | 13202 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
urable | +1.091 |
emort | +1.089 |
Cla | +1.083 |
scales | +1.076 |
milo | +1.066 |
Neg logit contributions | |
buttocks | -0.557 |
ONY | -0.514 |
venth | -0.509 |
Bench | -0.509 |
igan | -0.501 |
Token | beware |
Token ID | 40600 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
emort | +0.023 |
urable | +0.023 |
Cla | +0.023 |
scales | +0.023 |
milo | +0.022 |
Neg logit contributions | |
buttocks | -0.024 |
ONY | -0.023 |
Bench | -0.023 |
venth | -0.023 |
igan | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
emort | +1.205 |
Cla | +1.198 |
urable | +1.192 |
scales | +1.177 |
Kahn | +1.140 |
Neg logit contributions | |
buttocks | -1.744 |
ONY | -1.674 |
Bench | -1.671 |
venth | -1.670 |
igan | -1.654 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Cla | +0.639 |
urable | +0.639 |
emort | +0.639 |
scales | +0.632 |
Kahn | +0.610 |
Neg logit contributions | |
buttocks | -0.925 |
ONY | -0.887 |
venth | -0.882 |
Bench | -0.882 |
igan | -0.874 |
Token | Reviewer |
Token ID | 35407 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
emort | +0.538 |
Cla | +0.527 |
scales | +0.513 |
urable | +0.506 |
Kahn | +0.485 |
Neg logit contributions | |
buttocks | -1.318 |
Bench | -1.296 |
ONY | -1.287 |
venth | -1.284 |
igan | -1.272 |
Token | : |
Token ID | 25 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
urable | +2.772 |
Cla | +2.765 |
scales | +2.757 |
eq | +2.721 |
Kahn | +2.640 |
Neg logit contributions | |
buttocks | -3.748 |
sacrific | -3.604 |
ONY | -3.556 |
venth | -3.489 |
Bench | -3.477 |
Token | j |
Token ID | 474 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
emort | +0.363 |
urable | +0.360 |
Cla | +0.357 |
scales | +0.354 |
milo | +0.347 |
Neg logit contributions | |
buttocks | -0.383 |
ONY | -0.364 |
Bench | -0.362 |
venth | -0.361 |
igan | -0.359 |
Token | wt |
Token ID | 46569 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
buttocks | +0.075 |
ONY | +0.072 |
venth | +0.072 |
Bench | +0.072 |
igan | +0.072 |
Neg logit contributions | |
emort | -0.062 |
Cla | -0.062 |
urable | -0.061 |
scales | -0.061 |
Kahn | -0.059 |
Token | 0001 |
Token ID | 18005 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
buttocks | +0.529 |
igan | +0.507 |
venth | +0.507 |
ONY | +0.507 |
Bench | +0.504 |
Neg logit contributions | |
emort | -0.464 |
Cla | -0.462 |
urable | -0.460 |
scales | -0.456 |
milo | -0.443 |
Token | - |
Token ID | 532 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
emort | +0.542 |
urable | +0.538 |
Cla | +0.537 |
scales | +0.531 |
milo | +0.516 |
Neg logit contributions | |
buttocks | -0.619 |
ONY | -0.593 |
Bench | -0.591 |
venth | -0.591 |
igan | -0.587 |
Token | The |
Token ID | 383 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
emort | +0.326 |
urable | +0.322 |
Cla | +0.317 |
scales | +0.316 |
milo | +0.311 |
Neg logit contributions | |
buttocks | -0.350 |
ONY | -0.337 |
Bench | -0.336 |
venth | -0.334 |
igan | -0.333 |
Token | education |
Token ID | 3707 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
buttocks | +0.095 |
Bench | +0.090 |
ONY | +0.090 |
venth | +0.090 |
igan | +0.089 |
Neg logit contributions | |
emort | -0.077 |
urable | -0.076 |
Cla | -0.075 |
milo | -0.074 |
scales | -0.074 |
Token | sector |
Token ID | 6567 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
emort | +0.019 |
urable | +0.018 |
Cla | +0.018 |
scales | +0.018 |
milo | +0.018 |
Neg logit contributions | |
buttocks | -0.027 |
ONY | -0.025 |
Bench | -0.025 |
venth | -0.025 |
igan | -0.025 |
Token | as |
Token ID | 355 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
buttocks | +0.338 |
ONY | +0.323 |
venth | +0.323 |
Bench | +0.322 |
igan | +0.322 |
Neg logit contributions | |
emort | -0.253 |
urable | -0.250 |
Cla | -0.249 |
scales | -0.243 |
milo | -0.239 |
Token | a |
Token ID | 257 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
emort | +0.212 |
urable | +0.209 |
Cla | +0.208 |
milo | +0.202 |
scales | +0.202 |
Neg logit contributions | |
buttocks | -0.277 |
Bench | -0.263 |
venth | -0.262 |
ONY | -0.262 |
Newsletter | -0.261 |
Token | whole |
Token ID | 2187 |
Feature activation | -5.00e+00 |
Pos logit contributions | |
emort | +0.626 |
urable | +0.621 |
Cla | +0.619 |
scales | +0.611 |
milo | +0.594 |
Neg logit contributions | |
buttocks | -0.890 |
ONY | -0.849 |
venth | -0.848 |
Bench | -0.848 |
igan | -0.843 |
Token | has |
Token ID | 468 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
scales | +2.532 |
Cla | +2.445 |
emort | +2.430 |
urable | +2.404 |
Kahn | +2.342 |
Neg logit contributions | |
buttocks | -3.811 |
ONY | -3.727 |
Bench | -3.699 |
venth | -3.651 |
igan | -3.583 |
Token | a |
Token ID | 257 |
Feature activation | -7.64e-02 |
Pos logit contributions | |
emort | +0.676 |
urable | +0.669 |
Cla | +0.667 |
scales | +0.658 |
milo | +0.646 |
Neg logit contributions | |
buttocks | -0.785 |
ONY | -0.747 |
Bench | -0.746 |
venth | -0.744 |
igan | -0.741 |
Token | low |
Token ID | 1877 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
emort | +0.044 |
urable | +0.044 |
Cla | +0.044 |
scales | +0.043 |
milo | +0.043 |
Neg logit contributions | |
buttocks | -0.053 |
Bench | -0.051 |
venth | -0.051 |
ONY | -0.050 |
igan | -0.050 |
Token | quit |
Token ID | 11238 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
buttocks | +0.186 |
Bench | +0.176 |
igan | +0.176 |
venth | +0.175 |
ONY | +0.175 |
Neg logit contributions | |
emort | -0.164 |
Cla | -0.163 |
urable | -0.162 |
scales | -0.160 |
milo | -0.159 |
Token | rate |
Token ID | 2494 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
emort | +0.327 |
Cla | +0.322 |
urable | +0.317 |
scales | +0.313 |
milo | +0.303 |
Neg logit contributions | |
buttocks | -0.651 |
ONY | -0.629 |
Bench | -0.628 |
igan | -0.627 |
venth | -0.627 |
Token | � |
Token ID | 251 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
emort | +0.346 |
Cla | +0.344 |
urable | +0.343 |
scales | +0.341 |
milo | +0.332 |
Neg logit contributions | |
buttocks | -0.282 |
ONY | -0.267 |
Bench | -0.265 |
venth | -0.265 |
igan | -0.262 |
Token |
|
Token ID | 198 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
emort | +0.828 |
urable | +0.821 |
Cla | +0.817 |
scales | +0.811 |
milo | +0.789 |
Neg logit contributions | |
buttocks | -0.959 |
ONY | -0.918 |
Bench | -0.915 |
venth | -0.912 |
igan | -0.907 |
Token |
|
Token ID | 198 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
emort | +0.065 |
urable | +0.064 |
Cla | +0.064 |
scales | +0.064 |
milo | +0.062 |
Neg logit contributions | |
buttocks | -0.088 |
ONY | -0.085 |
Bench | -0.085 |
venth | -0.084 |
igan | -0.084 |
Token | https |
Token ID | 5450 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
emort | +0.267 |
urable | +0.265 |
Cla | +0.265 |
scales | +0.263 |
Kahn | +0.254 |
Neg logit contributions | |
buttocks | -0.315 |
Bench | -0.305 |
ONY | -0.303 |
venth | -0.302 |
igan | -0.299 |
Token | :// |
Token ID | 1378 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
urable | +0.769 |
emort | +0.766 |
Cla | +0.765 |
scales | +0.763 |
milo | +0.745 |
Neg logit contributions | |
buttocks | -0.532 |
ONY | -0.498 |
Bench | -0.491 |
venth | -0.490 |
igan | -0.485 |
Token | twitter |
Token ID | 6956 |
Feature activation | -4.99e+00 |
Pos logit contributions | |
emort | +0.265 |
Cla | +0.263 |
urable | +0.262 |
scales | +0.260 |
Kahn | +0.253 |
Neg logit contributions | |
buttocks | -0.284 |
ONY | -0.271 |
venth | -0.271 |
Bench | -0.271 |
igan | -0.269 |
Token | . |
Token ID | 13 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
Cla | +2.380 |
urable | +2.363 |
scales | +2.358 |
eq | +2.324 |
track | +2.245 |
Neg logit contributions | |
buttocks | -3.965 |
sacrific | -3.779 |
ONY | -3.774 |
venth | -3.709 |
Bench | -3.697 |
Token | com |
Token ID | 785 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
buttocks | +0.044 |
ONY | +0.041 |
venth | +0.040 |
Bench | +0.040 |
igan | +0.040 |
Neg logit contributions | |
Cla | -0.056 |
urable | -0.056 |
emort | -0.056 |
scales | -0.055 |
eq | -0.054 |
Token | / |
Token ID | 14 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
emort | +0.333 |
urable | +0.330 |
Cla | +0.330 |
scales | +0.327 |
milo | +0.317 |
Neg logit contributions | |
buttocks | -0.377 |
ONY | -0.361 |
Bench | -0.360 |
venth | -0.359 |
igan | -0.356 |
Token | om |
Token ID | 296 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
buttocks | +0.338 |
Bench | +0.327 |
ONY | +0.325 |
venth | +0.323 |
igan | +0.323 |
Neg logit contributions | |
emort | -0.318 |
scales | -0.314 |
Cla | -0.313 |
urable | -0.312 |
Kahn | -0.303 |
Token | g |
Token ID | 70 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
buttocks | +0.134 |
ONY | +0.128 |
venth | +0.128 |
Bench | +0.128 |
igan | +0.127 |
Neg logit contributions | |
emort | -0.119 |
Cla | -0.119 |
urable | -0.118 |
scales | -0.118 |
milo | -0.114 |
Token | Canada |
Token ID | 3340 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
buttocks | +0.798 |
ONY | +0.767 |
Bench | +0.763 |
venth | +0.762 |
igan | +0.760 |
Neg logit contributions | |
emort | -0.533 |
urable | -0.529 |
scales | -0.523 |
Cla | -0.521 |
milo | -0.507 |
Token | 's |
Token ID | 338 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
buttocks | +0.839 |
Bench | +0.807 |
igan | +0.798 |
ONY | +0.798 |
venth | +0.795 |
Neg logit contributions | |
emort | -0.781 |
urable | -0.771 |
scales | -0.758 |
Cla | -0.752 |
milo | -0.751 |
Token | Soccer |
Token ID | 15872 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
buttocks | +1.181 |
Bench | +1.125 |
venth | +1.124 |
ONY | +1.117 |
igan | +1.113 |
Neg logit contributions | |
emort | -0.885 |
urable | -0.880 |
scales | -0.859 |
Cla | -0.852 |
milo | -0.850 |
Token | Hall |
Token ID | 4789 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
buttocks | +0.503 |
venth | +0.489 |
Bench | +0.488 |
igan | +0.485 |
ONY | +0.483 |
Neg logit contributions | |
emort | -0.359 |
scales | -0.353 |
urable | -0.352 |
milo | -0.345 |
Cla | -0.343 |
Token | of |
Token ID | 286 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
emort | +0.754 |
Cla | +0.750 |
urable | +0.749 |
scales | +0.743 |
Kahn | +0.722 |
Neg logit contributions | |
buttocks | -0.719 |
ONY | -0.684 |
Bench | -0.681 |
venth | -0.680 |
igan | -0.672 |
Token | Fame |
Token ID | 18864 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
buttocks | +0.565 |
venth | +0.559 |
Bench | +0.554 |
igan | +0.552 |
ONY | +0.548 |
Neg logit contributions | |
emort | -0.271 |
urable | -0.259 |
scales | -0.259 |
Cla | -0.252 |
milo | -0.244 |
Token | on |
Token ID | 319 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
Cla | +3.035 |
emort | +3.008 |
urable | +2.988 |
scales | +2.974 |
Kahn | +2.899 |
Neg logit contributions | |
buttocks | -3.271 |
ONY | -3.134 |
venth | -3.129 |
Bench | -3.118 |
igan | -3.083 |
Token | Nov |
Token ID | 5267 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
emort | +0.037 |
urable | +0.036 |
Cla | +0.036 |
milo | +0.035 |
scales | +0.035 |
Neg logit contributions | |
buttocks | -0.027 |
ONY | -0.025 |
Bench | -0.025 |
venth | -0.025 |
igan | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
emort | +0.727 |
Cla | +0.724 |
urable | +0.723 |
scales | +0.716 |
Kahn | +0.693 |
Neg logit contributions | |
buttocks | -0.919 |
ONY | -0.880 |
Bench | -0.877 |
venth | -0.874 |
igan | -0.868 |
Token | 8 |
Token ID | 807 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
buttocks | +0.169 |
Bench | +0.161 |
ONY | +0.161 |
venth | +0.160 |
igan | +0.159 |
Neg logit contributions | |
emort | -0.189 |
urable | -0.185 |
Cla | -0.182 |
scales | -0.182 |
milo | -0.180 |
Token | in |
Token ID | 287 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
emort | +0.600 |
urable | +0.594 |
Cla | +0.594 |
scales | +0.588 |
milo | +0.572 |
Neg logit contributions | |
buttocks | -0.679 |
ONY | -0.649 |
Bench | -0.647 |
venth | -0.646 |
igan | -0.641 |
Token | � |
Token ID | 247 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
buttocks | +0.046 |
ONY | +0.043 |
Bench | +0.043 |
venth | +0.042 |
igan | +0.042 |
Neg logit contributions | |
emort | -0.073 |
Cla | -0.072 |
urable | -0.072 |
scales | -0.072 |
milo | -0.070 |
Token | s |
Token ID | 82 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
buttocks | +0.035 |
venth | +0.033 |
Bench | +0.033 |
ONY | +0.033 |
igan | +0.033 |
Neg logit contributions | |
emort | -0.032 |
urable | -0.032 |
Cla | -0.032 |
scales | -0.031 |
Kahn | -0.031 |
Token | politics |
Token ID | 4819 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
buttocks | +0.215 |
venth | +0.205 |
ONY | +0.204 |
igan | +0.204 |
Bench | +0.203 |
Neg logit contributions | |
emort | -0.143 |
urable | -0.140 |
Cla | -0.139 |
scales | -0.136 |
milo | -0.135 |
Token | as |
Token ID | 355 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
emort | +0.298 |
urable | +0.294 |
Cla | +0.293 |
scales | +0.290 |
milo | +0.283 |
Neg logit contributions | |
buttocks | -0.330 |
ONY | -0.314 |
Bench | -0.314 |
venth | -0.313 |
igan | -0.312 |
Token | a |
Token ID | 257 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
emort | +0.186 |
urable | +0.181 |
Cla | +0.177 |
milo | +0.175 |
scales | +0.173 |
Neg logit contributions | |
buttocks | -0.235 |
ONY | -0.224 |
Bench | -0.224 |
igan | -0.222 |
venth | -0.221 |
Token | whole |
Token ID | 2187 |
Feature activation | -4.95e+00 |
Pos logit contributions | |
emort | +0.690 |
urable | +0.680 |
Cla | +0.678 |
scales | +0.669 |
milo | +0.652 |
Neg logit contributions | |
buttocks | -0.974 |
ONY | -0.932 |
venth | -0.928 |
Bench | -0.927 |
igan | -0.922 |
Token | . |
Token ID | 13 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
Cla | +2.662 |
scales | +2.654 |
urable | +2.649 |
emort | +2.646 |
Kahn | +2.551 |
Neg logit contributions | |
buttocks | -3.575 |
ONY | -3.452 |
Bench | -3.436 |
venth | -3.411 |
igan | -3.332 |
Token | But |
Token ID | 887 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
emort | +0.050 |
urable | +0.049 |
scales | +0.048 |
Cla | +0.047 |
milo | +0.047 |
Neg logit contributions | |
buttocks | -0.052 |
ONY | -0.050 |
Bench | -0.050 |
igan | -0.050 |
venth | -0.050 |
Token | it |
Token ID | 340 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
buttocks | +0.635 |
Bench | +0.605 |
ONY | +0.605 |
igan | +0.605 |
venth | +0.601 |
Neg logit contributions | |
emort | -0.494 |
urable | -0.484 |
Cla | -0.469 |
scales | -0.467 |
milo | -0.463 |
Token | � |
Token ID | 447 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
emort | +0.367 |
urable | +0.363 |
Cla | +0.362 |
scales | +0.356 |
milo | +0.348 |
Neg logit contributions | |
buttocks | -0.467 |
ONY | -0.445 |
venth | -0.444 |
Bench | -0.444 |
igan | -0.443 |
Token | � |
Token ID | 247 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
buttocks | +0.120 |
ONY | +0.113 |
Bench | +0.112 |
venth | +0.111 |
igan | +0.110 |
Neg logit contributions | |
emort | -0.193 |
Cla | -0.193 |
urable | -0.193 |
scales | -0.192 |
milo | -0.187 |
Token | Flag |
Token ID | 34227 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
emort | +0.009 |
Cla | +0.009 |
scales | +0.009 |
urable | +0.009 |
eq | +0.009 |
Neg logit contributions | |
buttocks | -0.008 |
Bench | -0.008 |
ONY | -0.008 |
venth | -0.008 |
igan | -0.008 |
Token | Reply |
Token ID | 14883 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
emort | +1.005 |
urable | +0.993 |
Cla | +0.981 |
scales | +0.979 |
milo | +0.958 |
Neg logit contributions | |
buttocks | -1.041 |
ONY | -0.993 |
venth | -0.987 |
Bench | -0.985 |
igan | -0.977 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
emort | +0.213 |
urable | +0.209 |
scales | +0.206 |
Cla | +0.205 |
milo | +0.204 |
Neg logit contributions | |
buttocks | -0.266 |
ONY | -0.256 |
Bench | -0.255 |
venth | -0.254 |
igan | -0.252 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Cla | +0.628 |
urable | +0.625 |
emort | +0.624 |
scales | +0.620 |
Kahn | +0.597 |
Neg logit contributions | |
buttocks | -0.913 |
ONY | -0.875 |
venth | -0.870 |
Bench | -0.868 |
igan | -0.862 |
Token | 14 |
Token ID | 1415 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
emort | +0.922 |
urable | +0.906 |
Cla | +0.904 |
scales | +0.901 |
milo | +0.877 |
Neg logit contributions | |
buttocks | -0.862 |
Bench | -0.829 |
ONY | -0.827 |
venth | -0.822 |
igan | -0.812 |
Token | Recommend |
Token ID | 30361 |
Feature activation | -4.95e+00 |
Pos logit contributions | |
buttocks | +0.285 |
venth | +0.274 |
ONY | +0.273 |
Bench | +0.273 |
igan | +0.270 |
Neg logit contributions | |
emort | -0.172 |
urable | -0.168 |
milo | -0.164 |
scales | -0.164 |
Cla | -0.163 |
Token |
|
Token ID | 198 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
Cla | +2.282 |
urable | +2.180 |
scales | +2.168 |
Kahn | +2.143 |
eq | +2.121 |
Neg logit contributions | |
buttocks | -4.044 |
sacrific | -3.908 |
ONY | -3.867 |
Bench | -3.838 |
venth | -3.833 |
Token |
|
Token ID | 198 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
Cla | +0.983 |
urable | +0.976 |
scales | +0.968 |
emort | +0.958 |
eq | +0.936 |
Neg logit contributions | |
buttocks | -1.525 |
ONY | -1.464 |
venth | -1.455 |
Bench | -1.445 |
igan | -1.433 |
Token | Recommend |
Token ID | 41248 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
emort | +1.055 |
scales | +1.033 |
Cla | +1.025 |
urable | +1.017 |
milo | +1.011 |
Neg logit contributions | |
buttocks | -1.465 |
Bench | -1.458 |
ONY | -1.433 |
Newsletter | -1.432 |
venth | -1.427 |
Token | Share |
Token ID | 8734 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Cla | +2.292 |
scales | +2.243 |
Kahn | +2.228 |
urable | +2.212 |
emort | +2.197 |
Neg logit contributions | |
buttocks | -3.263 |
Bench | -3.160 |
venth | -3.123 |
ONY | -3.103 |
Newsletter | -3.075 |
Token | this |
Token ID | 428 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
emort | +0.603 |
urable | +0.591 |
Cla | +0.589 |
scales | +0.580 |
milo | +0.566 |
Neg logit contributions | |
buttocks | -1.000 |
venth | -0.959 |
Bench | -0.958 |
ONY | -0.955 |
igan | -0.949 |
Token | ( |
Token ID | 357 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
emort | +0.070 |
urable | +0.069 |
Cla | +0.069 |
scales | +0.068 |
milo | +0.067 |
Neg logit contributions | |
buttocks | -0.073 |
ONY | -0.070 |
Bench | -0.070 |
venth | -0.069 |
igan | -0.069 |
Token | Blu |
Token ID | 38676 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
emort | +0.534 |
urable | +0.529 |
Cla | +0.529 |
scales | +0.524 |
milo | +0.510 |
Neg logit contributions | |
buttocks | -0.567 |
ONY | -0.543 |
Bench | -0.541 |
venth | -0.539 |
igan | -0.535 |
Token | - |
Token ID | 12 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
emort | +1.479 |
urable | +1.470 |
Cla | +1.469 |
scales | +1.444 |
Kahn | +1.402 |
Neg logit contributions | |
buttocks | -1.949 |
Bench | -1.852 |
ONY | -1.852 |
venth | -1.846 |
igan | -1.827 |
Token | ray |
Token ID | 2433 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
emort | +0.375 |
Cla | +0.372 |
urable | +0.371 |
scales | +0.369 |
milo | +0.358 |
Neg logit contributions | |
buttocks | -0.437 |
ONY | -0.419 |
Bench | -0.418 |
venth | -0.416 |
igan | -0.414 |
Token | / |
Token ID | 14 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
emort | +0.889 |
urable | +0.880 |
Cla | +0.879 |
scales | +0.869 |
milo | +0.845 |
Neg logit contributions | |
buttocks | -1.105 |
ONY | -1.059 |
Bench | -1.055 |
venth | -1.051 |
igan | -1.046 |
Token | DVD |
Token ID | 12490 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
buttocks | +0.389 |
ONY | +0.375 |
Bench | +0.374 |
venth | +0.369 |
igan | +0.365 |
Neg logit contributions | |
emort | -0.475 |
urable | -0.469 |
Cla | -0.468 |
scales | -0.466 |
milo | -0.456 |
Token | + |
Token ID | 1343 |
Feature activation | -9.15e-01 |
Pos logit contributions | |
Cla | +2.622 |
emort | +2.612 |
urable | +2.594 |
scales | +2.591 |
eq | +2.521 |
Neg logit contributions | |
buttocks | -3.594 |
venth | -3.443 |
Bench | -3.423 |
ONY | -3.415 |
igan | -3.382 |
Token | Digital |
Token ID | 10231 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
emort | +0.609 |
urable | +0.605 |
Cla | +0.601 |
scales | +0.597 |
milo | +0.586 |
Neg logit contributions | |
buttocks | -0.559 |
ONY | -0.533 |
Bench | -0.531 |
venth | -0.527 |
igan | -0.525 |
Token | Copy |
Token ID | 17393 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
buttocks | +0.213 |
ONY | +0.205 |
Bench | +0.204 |
venth | +0.203 |
igan | +0.202 |
Neg logit contributions | |
emort | -0.188 |
urable | -0.186 |
Cla | -0.185 |
scales | -0.183 |
milo | -0.180 |
Token | ) |
Token ID | 8 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
emort | +0.868 |
urable | +0.859 |
Cla | +0.859 |
scales | +0.849 |
milo | +0.824 |
Neg logit contributions | |
buttocks | -1.114 |
ONY | -1.067 |
Bench | -1.064 |
venth | -1.061 |
igan | -1.055 |
Token | --- |
Token ID | 11420 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
emort | +0.213 |
urable | +0.211 |
Cla | +0.210 |
scales | +0.208 |
milo | +0.204 |
Neg logit contributions | |
buttocks | -0.222 |
ONY | -0.213 |
Bench | -0.212 |
venth | -0.211 |
igan | -0.210 |
Token | bad |
Token ID | 2089 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
urable | +0.333 |
emort | +0.331 |
Cla | +0.329 |
scales | +0.329 |
eq | +0.319 |
Neg logit contributions | |
buttocks | -0.433 |
ONY | -0.422 |
venth | -0.417 |
Bench | -0.417 |
igan | -0.411 |
Token | idea |
Token ID | 2126 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
buttocks | +0.016 |
ONY | +0.015 |
venth | +0.015 |
igan | +0.015 |
Bench | +0.015 |
Neg logit contributions | |
emort | -0.015 |
urable | -0.015 |
Cla | -0.015 |
scales | -0.014 |
milo | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
emort | +0.083 |
urable | +0.082 |
Cla | +0.081 |
scales | +0.081 |
milo | +0.079 |
Neg logit contributions | |
buttocks | -0.098 |
ONY | -0.094 |
Bench | -0.093 |
venth | -0.093 |
igan | -0.093 |
Token |
|
Token ID | 198 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
emort | +0.402 |
Cla | +0.399 |
urable | +0.398 |
scales | +0.394 |
Kahn | +0.383 |
Neg logit contributions | |
buttocks | -0.476 |
ONY | -0.455 |
Bench | -0.454 |
venth | -0.453 |
igan | -0.450 |
Token |
|
Token ID | 198 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
emort | +0.218 |
urable | +0.214 |
Cla | +0.214 |
scales | +0.211 |
milo | +0.205 |
Neg logit contributions | |
buttocks | -0.287 |
ONY | -0.275 |
Bench | -0.275 |
venth | -0.274 |
igan | -0.273 |
Token | Reviewer |
Token ID | 35407 |
Feature activation | -4.87e+00 |
Pos logit contributions | |
emort | +0.269 |
Cla | +0.264 |
scales | +0.257 |
Guth | +0.248 |
urable | +0.247 |
Neg logit contributions | |
Bench | -0.564 |
buttocks | -0.558 |
ONY | -0.555 |
venth | -0.552 |
Newsletter | -0.551 |
Token | : |
Token ID | 25 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
Cla | +2.667 |
urable | +2.657 |
scales | +2.653 |
eq | +2.614 |
Kahn | +2.538 |
Neg logit contributions | |
buttocks | -3.514 |
sacrific | -3.354 |
ONY | -3.346 |
venth | -3.290 |
Bench | -3.278 |
Token | Mike |
Token ID | 4995 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
emort | +0.347 |
urable | +0.344 |
Cla | +0.340 |
scales | +0.338 |
milo | +0.331 |
Neg logit contributions | |
buttocks | -0.367 |
ONY | -0.349 |
Bench | -0.347 |
venth | -0.346 |
igan | -0.344 |
Token | M |
Token ID | 44 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
emort | +1.400 |
Cla | +1.391 |
urable | +1.386 |
scales | +1.371 |
Kahn | +1.334 |
Neg logit contributions | |
buttocks | -1.760 |
ONY | -1.683 |
venth | -1.676 |
Bench | -1.675 |
igan | -1.663 |
Token | anners |
Token ID | 15672 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
buttocks | +0.291 |
ONY | +0.280 |
Bench | +0.278 |
igan | +0.277 |
venth | +0.277 |
Neg logit contributions | |
emort | -0.241 |
Cla | -0.240 |
urable | -0.239 |
scales | -0.236 |
milo | -0.233 |
Token | - |
Token ID | 532 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
buttocks | +0.787 |
Bench | +0.758 |
ONY | +0.754 |
igan | +0.751 |
venth | +0.750 |
Neg logit contributions | |
emort | -0.716 |
urable | -0.712 |
Cla | -0.709 |
scales | -0.709 |
milo | -0.683 |
Token | According |
Token ID | 4784 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
emort | +0.078 |
urable | +0.077 |
Cla | +0.076 |
scales | +0.076 |
milo | +0.074 |
Neg logit contributions | |
buttocks | -0.085 |
ONY | -0.082 |
Bench | -0.081 |
venth | -0.081 |
igan | -0.081 |
Token | to |
Token ID | 284 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
buttocks | +0.079 |
ONY | +0.076 |
Bench | +0.075 |
venth | +0.075 |
igan | +0.075 |
Neg logit contributions | |
emort | -0.061 |
Cla | -0.061 |
urable | -0.061 |
scales | -0.060 |
milo | -0.058 |
Token | Mur |
Token ID | 5921 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
buttocks | +0.203 |
ONY | +0.194 |
Bench | +0.193 |
igan | +0.192 |
venth | +0.191 |
Neg logit contributions | |
emort | -0.148 |
urable | -0.147 |
Cla | -0.142 |
scales | -0.141 |
milo | -0.141 |
Token | ad |
Token ID | 324 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
emort | +0.230 |
urable | +0.228 |
Cla | +0.227 |
scales | +0.224 |
milo | +0.217 |
Neg logit contributions | |
buttocks | -0.271 |
ONY | -0.259 |
venth | -0.259 |
igan | -0.258 |
Bench | -0.257 |
Token | Do |
Token ID | 2141 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
emort | +0.531 |
urable | +0.525 |
Cla | +0.523 |
scales | +0.519 |
milo | +0.502 |
Neg logit contributions | |
buttocks | -0.755 |
ONY | -0.725 |
Bench | -0.722 |
venth | -0.722 |
igan | -0.719 |
Token | ğ |
Token ID | 33133 |
Feature activation | -4.83e+00 |
Pos logit contributions | |
emort | +0.109 |
urable | +0.108 |
Cla | +0.108 |
scales | +0.106 |
milo | +0.102 |
Neg logit contributions | |
buttocks | -0.253 |
igan | -0.245 |
venth | -0.245 |
ONY | -0.245 |
Bench | -0.245 |
Token | uk |
Token ID | 2724 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
urable | +2.907 |
emort | +2.901 |
Cla | +2.867 |
scales | +2.859 |
milo | +2.779 |
Neg logit contributions | |
buttocks | -3.203 |
ONY | -3.087 |
Bench | -3.059 |
venth | -3.042 |
Newsletter | -3.008 |
Token | an |
Token ID | 272 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
emort | +1.147 |
urable | +1.144 |
Cla | +1.141 |
scales | +1.130 |
Kahn | +1.094 |
Neg logit contributions | |
buttocks | -1.275 |
ONY | -1.221 |
Bench | -1.218 |
venth | -1.211 |
igan | -1.195 |
Token | li |
Token ID | 4528 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
emort | +0.957 |
urable | +0.949 |
Cla | +0.948 |
scales | +0.939 |
milo | +0.914 |
Neg logit contributions | |
buttocks | -0.994 |
ONY | -0.949 |
Bench | -0.945 |
venth | -0.943 |
igan | -0.938 |
Token | , |
Token ID | 11 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
emort | +0.796 |
urable | +0.788 |
Cla | +0.786 |
scales | +0.778 |
milo | +0.760 |
Neg logit contributions | |
buttocks | -0.802 |
ONY | -0.763 |
Bench | -0.761 |
venth | -0.760 |
igan | -0.758 |
Token | writing |
Token ID | 3597 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
emort | +0.170 |
urable | +0.168 |
Cla | +0.165 |
scales | +0.164 |
milo | +0.161 |
Neg logit contributions | |
buttocks | -0.214 |
igan | -0.204 |
ONY | -0.204 |
Bench | -0.203 |
venth | -0.202 |
Token | , |
Token ID | 11 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
emort | +1.298 |
urable | +1.292 |
Cla | +1.290 |
scales | +1.277 |
Kahn | +1.237 |
Neg logit contributions | |
buttocks | -1.500 |
ONY | -1.434 |
Bench | -1.431 |
venth | -1.428 |
igan | -1.413 |
Token | an |
Token ID | 281 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
emort | +0.122 |
urable | +0.120 |
Cla | +0.120 |
scales | +0.118 |
milo | +0.116 |
Neg logit contributions | |
buttocks | -0.146 |
ONY | -0.139 |
Bench | -0.139 |
igan | -0.138 |
venth | -0.138 |
Token | impoverished |
Token ID | 35014 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
emort | +0.208 |
Cla | +0.205 |
urable | +0.204 |
scales | +0.201 |
milo | +0.196 |
Neg logit contributions | |
buttocks | -0.320 |
ONY | -0.308 |
igan | -0.305 |
venth | -0.305 |
Bench | -0.305 |
Token | and |
Token ID | 290 |
Feature activation | +5.96e-03 |
Pos logit contributions | |
buttocks | +0.711 |
ONY | +0.676 |
igan | +0.674 |
venth | +0.674 |
Bench | +0.674 |
Neg logit contributions | |
emort | -0.528 |
urable | -0.519 |
Cla | -0.515 |
scales | -0.508 |
milo | -0.499 |
Token | sterile |
Token ID | 38697 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
buttocks | +0.004 |
ONY | +0.004 |
Bench | +0.004 |
venth | +0.004 |
igan | +0.004 |
Neg logit contributions | |
emort | -0.004 |
urable | -0.004 |
Cla | -0.004 |
scales | -0.004 |
milo | -0.003 |
Token | amalg |
Token ID | 45541 |
Feature activation | -4.80e+00 |
Pos logit contributions | |
emort | +0.258 |
urable | +0.254 |
Cla | +0.253 |
scales | +0.250 |
milo | +0.243 |
Neg logit contributions | |
buttocks | -0.363 |
ONY | -0.347 |
Bench | -0.346 |
venth | -0.346 |
igan | -0.345 |
Token | am |
Token ID | 321 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
emort | +3.098 |
urable | +3.092 |
Cla | +3.067 |
scales | +3.007 |
milo | +2.967 |
Neg logit contributions | |
buttocks | -2.968 |
Bench | -2.848 |
ONY | -2.820 |
Newsletter | -2.781 |
venth | -2.762 |
Token | of |
Token ID | 286 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
buttocks | +0.350 |
ONY | +0.334 |
venth | +0.331 |
Bench | +0.331 |
igan | +0.329 |
Neg logit contributions | |
emort | -0.384 |
Cla | -0.381 |
urable | -0.379 |
scales | -0.377 |
milo | -0.368 |
Token | social |
Token ID | 1919 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
buttocks | +0.355 |
igan | +0.337 |
ONY | +0.336 |
venth | +0.336 |
Bench | +0.334 |
Neg logit contributions | |
emort | -0.275 |
urable | -0.271 |
Cla | -0.269 |
scales | -0.262 |
milo | -0.262 |
Token | democracy |
Token ID | 7996 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
buttocks | +0.499 |
ONY | +0.475 |
venth | +0.474 |
Bench | +0.473 |
igan | +0.472 |
Neg logit contributions | |
emort | -0.374 |
urable | -0.368 |
Cla | -0.367 |
scales | -0.360 |
milo | -0.357 |
Token | and |
Token ID | 290 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
emort | +0.451 |
urable | +0.446 |
Cla | +0.446 |
scales | +0.441 |
milo | +0.429 |
Neg logit contributions | |
buttocks | -0.493 |
ONY | -0.471 |
Bench | -0.469 |
venth | -0.468 |
igan | -0.466 |
Token | outrage |
Token ID | 11616 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
buttocks | +0.137 |
Bench | +0.131 |
igan | +0.130 |
ONY | +0.130 |
venth | +0.129 |
Neg logit contributions | |
emort | -0.107 |
urable | -0.105 |
Cla | -0.105 |
scales | -0.103 |
milo | -0.101 |
Token | ously |
Token ID | 3481 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
emort | +0.297 |
Cla | +0.294 |
urable | +0.293 |
scales | +0.289 |
milo | +0.282 |
Neg logit contributions | |
buttocks | -0.359 |
ONY | -0.344 |
Bench | -0.343 |
venth | -0.341 |
igan | -0.341 |
Token | high |
Token ID | 1029 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
emort | +0.324 |
Cla | +0.319 |
urable | +0.318 |
scales | +0.314 |
milo | +0.309 |
Neg logit contributions | |
buttocks | -0.342 |
Bench | -0.326 |
ONY | -0.325 |
venth | -0.323 |
igan | -0.323 |
Token | gasoline |
Token ID | 21408 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
emort | +0.024 |
urable | +0.024 |
Cla | +0.023 |
scales | +0.023 |
milo | +0.022 |
Neg logit contributions | |
buttocks | -0.048 |
Bench | -0.046 |
ONY | -0.046 |
igan | -0.046 |
venth | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
emort | +0.016 |
urable | +0.016 |
Cla | +0.016 |
milo | +0.015 |
scales | +0.015 |
Neg logit contributions | |
buttocks | -0.029 |
Bench | -0.028 |
igan | -0.028 |
ONY | -0.028 |
venth | -0.027 |
Token | heating |
Token ID | 16930 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
buttocks | +0.368 |
Bench | +0.352 |
venth | +0.351 |
ONY | +0.350 |
igan | +0.349 |
Neg logit contributions | |
emort | -0.262 |
Cla | -0.256 |
urable | -0.256 |
scales | -0.251 |
milo | -0.248 |
Token | - |
Token ID | 12 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
buttocks | +2.397 |
Bench | +2.276 |
igan | +2.242 |
ONY | +2.225 |
venth | +2.187 |
Neg logit contributions | |
emort | -1.968 |
urable | -1.906 |
Cla | -1.882 |
milo | -1.875 |
Kahn | -1.815 |
Token | oil |
Token ID | 9437 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
emort | +0.171 |
urable | +0.169 |
Cla | +0.169 |
scales | +0.168 |
milo | +0.162 |
Neg logit contributions | |
buttocks | -0.215 |
ONY | -0.206 |
Bench | -0.205 |
venth | -0.205 |
igan | -0.203 |
Token | prices |
Token ID | 4536 |
Feature activation | +9.56e-01 |
Pos logit contributions | |
buttocks | +0.177 |
Bench | +0.171 |
igan | +0.169 |
ONY | +0.169 |
venth | +0.168 |
Neg logit contributions | |
emort | -0.097 |
urable | -0.095 |
Cla | -0.095 |
milo | -0.092 |
scales | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
buttocks | +0.631 |
ONY | +0.599 |
Bench | +0.599 |
igan | +0.597 |
venth | +0.594 |
Neg logit contributions | |
emort | -0.593 |
urable | -0.584 |
Cla | -0.581 |
scales | -0.574 |
milo | -0.565 |
Token | I |
Token ID | 314 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
buttocks | +0.357 |
ONY | +0.348 |
Bench | +0.348 |
Newsletter | +0.344 |
igan | +0.344 |
Neg logit contributions | |
emort | -0.361 |
urable | -0.349 |
scales | -0.343 |
Cla | -0.339 |
milo | -0.337 |
Token | told |
Token ID | 1297 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
buttocks | +0.458 |
venth | +0.435 |
ONY | +0.434 |
igan | +0.434 |
Bench | +0.434 |
Neg logit contributions | |
emort | -0.435 |
urable | -0.434 |
Cla | -0.429 |
scales | -0.424 |
milo | -0.417 |
Token | them |
Token ID | 606 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
emort | +0.176 |
urable | +0.174 |
Cla | +0.171 |
scales | +0.170 |
milo | +0.166 |
Neg logit contributions | |
buttocks | -0.256 |
ONY | -0.246 |
Bench | -0.245 |
venth | -0.244 |
igan | -0.244 |
Token | to |
Token ID | 284 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
buttocks | +0.261 |
ONY | +0.249 |
Bench | +0.247 |
venth | +0.247 |
igan | +0.246 |
Neg logit contributions | |
emort | -0.229 |
urable | -0.227 |
Cla | -0.224 |
scales | -0.222 |
milo | -0.218 |
Token | follow |
Token ID | 1061 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
buttocks | +0.991 |
ONY | +0.947 |
igan | +0.940 |
Bench | +0.939 |
venth | +0.935 |
Neg logit contributions | |
emort | -0.746 |
urable | -0.742 |
Cla | -0.729 |
milo | -0.718 |
scales | -0.717 |
Token | us |
Token ID | 514 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
emort | +0.243 |
urable | +0.242 |
Cla | +0.240 |
scales | +0.238 |
milo | +0.233 |
Neg logit contributions | |
buttocks | -0.270 |
ONY | -0.258 |
Bench | -0.258 |
venth | -0.257 |
igan | -0.255 |
Token | too |
Token ID | 1165 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
buttocks | +0.547 |
ONY | +0.520 |
Bench | +0.520 |
venth | +0.518 |
igan | +0.518 |
Neg logit contributions | |
emort | -0.483 |
urable | -0.481 |
Cla | -0.478 |
scales | -0.473 |
milo | -0.463 |
Token | . |
Token ID | 13 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
buttocks | +2.063 |
igan | +1.964 |
Bench | +1.961 |
ONY | +1.953 |
venth | +1.944 |
Neg logit contributions | |
urable | -2.011 |
emort | -1.986 |
Cla | -1.968 |
scales | -1.954 |
milo | -1.929 |
Token | Was |
Token ID | 8920 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
buttocks | +0.274 |
ONY | +0.263 |
Bench | +0.262 |
venth | +0.261 |
igan | +0.260 |
Neg logit contributions | |
emort | -0.248 |
urable | -0.246 |
scales | -0.241 |
Cla | -0.241 |
milo | -0.236 |
Token | that |
Token ID | 326 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
emort | +0.410 |
urable | +0.406 |
Cla | +0.402 |
scales | +0.400 |
milo | +0.389 |
Neg logit contributions | |
buttocks | -0.509 |
ONY | -0.489 |
Bench | -0.486 |
venth | -0.486 |
igan | -0.485 |
Token | a |
Token ID | 257 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
emort | +0.816 |
urable | +0.809 |
Cla | +0.805 |
scales | +0.797 |
milo | +0.775 |
Neg logit contributions | |
buttocks | -1.089 |
ONY | -1.043 |
Bench | -1.039 |
venth | -1.039 |
igan | -1.032 |
Token | bit |
Token ID | 1643 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
emort | +0.110 |
urable | +0.109 |
Cla | +0.109 |
scales | +0.108 |
milo | +0.105 |
Neg logit contributions | |
buttocks | -0.137 |
ONY | -0.131 |
venth | -0.131 |
Bench | -0.131 |
igan | -0.130 |
Token | f |
Token ID | 277 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
buttocks | +0.259 |
ONY | +0.246 |
Bench | +0.246 |
venth | +0.245 |
igan | +0.245 |
Neg logit contributions | |
emort | -0.243 |
urable | -0.240 |
Cla | -0.239 |
scales | -0.233 |
milo | -0.231 |
Token | 2 |
Token ID | 362 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
buttocks | +0.602 |
ONY | +0.576 |
Bench | +0.576 |
igan | +0.575 |
venth | +0.574 |
Neg logit contributions | |
emort | -0.514 |
urable | -0.505 |
Cla | -0.500 |
scales | -0.488 |
milo | -0.480 |
Token | , |
Token ID | 837 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
buttocks | +1.341 |
ONY | +1.270 |
Bench | +1.261 |
igan | +1.259 |
venth | +1.259 |
Neg logit contributions | |
emort | -1.472 |
urable | -1.457 |
Cla | -1.441 |
scales | -1.400 |
milo | -1.394 |
Token | … |
Token ID | 3926 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
buttocks | +0.913 |
Bench | +0.862 |
ONY | +0.862 |
venth | +0.858 |
igan | +0.855 |
Neg logit contributions | |
emort | -0.885 |
urable | -0.875 |
Cla | -0.868 |
scales | -0.850 |
milo | -0.843 |
Token | f |
Token ID | 277 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
emort | +0.345 |
urable | +0.340 |
Cla | +0.338 |
scales | +0.332 |
milo | +0.329 |
Neg logit contributions | |
buttocks | -0.358 |
Bench | -0.342 |
ONY | -0.341 |
venth | -0.340 |
igan | -0.338 |
Token | n |
Token ID | 299 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
buttocks | +0.812 |
ONY | +0.772 |
venth | +0.771 |
Bench | +0.770 |
igan | +0.765 |
Neg logit contributions | |
emort | -0.966 |
urable | -0.953 |
Cla | -0.950 |
scales | -0.930 |
milo | -0.920 |
Token | , |
Token ID | 837 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
buttocks | +3.150 |
ONY | +2.978 |
venth | +2.961 |
Bench | +2.960 |
igan | +2.956 |
Neg logit contributions | |
emort | -3.285 |
Cla | -3.191 |
urable | -3.186 |
milo | -3.117 |
Kahn | -3.061 |
Token | such |
Token ID | 884 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
buttocks | +1.205 |
ONY | +1.136 |
Bench | +1.131 |
igan | +1.124 |
venth | +1.120 |
Neg logit contributions | |
emort | -1.237 |
urable | -1.214 |
Cla | -1.196 |
milo | -1.183 |
scales | -1.165 |
Token | that |
Token ID | 326 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
emort | +0.073 |
urable | +0.072 |
Cla | +0.072 |
scales | +0.071 |
milo | +0.070 |
Neg logit contributions | |
buttocks | -0.078 |
ONY | -0.074 |
Bench | -0.074 |
venth | -0.074 |
igan | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
buttocks | +0.929 |
igan | +0.881 |
Bench | +0.880 |
ONY | +0.878 |
venth | +0.876 |
Neg logit contributions | |
emort | -0.812 |
urable | -0.801 |
Cla | -0.785 |
milo | -0.770 |
scales | -0.767 |
Token | final |
Token ID | 2457 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
buttocks | +0.213 |
venth | +0.201 |
ONY | +0.201 |
Bench | +0.200 |
igan | +0.200 |
Neg logit contributions | |
emort | -0.159 |
urable | -0.156 |
Cla | -0.154 |
milo | -0.151 |
scales | -0.149 |
Token | at |
Token ID | 379 |
Feature activation | -7.83e-01 |
Pos logit contributions | |
buttocks | +0.204 |
Bench | +0.194 |
ONY | +0.194 |
venth | +0.192 |
igan | +0.191 |
Neg logit contributions | |
emort | -0.185 |
urable | -0.183 |
Cla | -0.182 |
milo | -0.177 |
scales | -0.177 |
Token | a |
Token ID | 257 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
emort | +0.460 |
urable | +0.456 |
Cla | +0.455 |
scales | +0.449 |
milo | +0.439 |
Neg logit contributions | |
buttocks | -0.539 |
ONY | -0.516 |
Bench | -0.515 |
venth | -0.513 |
igan | -0.509 |
Token | refresh |
Token ID | 14976 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
emort | +0.225 |
urable | +0.222 |
Cla | +0.221 |
scales | +0.218 |
milo | +0.217 |
Neg logit contributions | |
buttocks | -0.217 |
Bench | -0.207 |
ONY | -0.206 |
venth | -0.205 |
igan | -0.203 |
Token | rate |
Token ID | 2494 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
buttocks | +0.902 |
Bench | +0.870 |
ONY | +0.870 |
venth | +0.866 |
igan | +0.861 |
Neg logit contributions | |
emort | -0.493 |
Cla | -0.474 |
urable | -0.471 |
scales | -0.461 |
milo | -0.455 |
Token | more |
Token ID | 517 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
emort | +0.149 |
Cla | +0.148 |
urable | +0.148 |
scales | +0.147 |
Kahn | +0.142 |
Neg logit contributions | |
buttocks | -0.153 |
ONY | -0.146 |
Bench | -0.145 |
venth | -0.145 |
igan | -0.144 |
Token | likely |
Token ID | 1884 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
buttocks | +0.011 |
Bench | +0.011 |
ONY | +0.011 |
venth | +0.011 |
igan | +0.011 |
Neg logit contributions | |
emort | -0.009 |
urable | -0.009 |
Cla | -0.009 |
scales | -0.009 |
milo | -0.009 |
Token | achieved |
Token ID | 8793 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
buttocks | +1.921 |
ONY | +1.843 |
Bench | +1.841 |
igan | +1.828 |
venth | +1.811 |
Neg logit contributions | |
emort | -1.753 |
urable | -1.719 |
Cla | -1.718 |
milo | -1.688 |
scales | -1.680 |
Token | during |
Token ID | 1141 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
buttocks | +0.328 |
ONY | +0.311 |
Bench | +0.310 |
igan | +0.308 |
venth | +0.307 |
Neg logit contributions | |
emort | -0.337 |
Cla | -0.333 |
urable | -0.332 |
scales | -0.327 |
milo | -0.325 |
Token | G |
Token ID | 402 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
buttocks | +0.245 |
ONY | +0.234 |
Bench | +0.234 |
venth | +0.232 |
igan | +0.230 |
Neg logit contributions | |
emort | -0.192 |
urable | -0.190 |
Cla | -0.189 |
scales | -0.186 |
milo | -0.184 |
Token | - |
Token ID | 12 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
emort | +0.008 |
urable | +0.008 |
Cla | +0.008 |
scales | +0.008 |
milo | +0.008 |
Neg logit contributions | |
buttocks | -0.010 |
Bench | -0.010 |
ONY | -0.010 |
igan | -0.010 |
venth | -0.010 |
Token | sync |
Token ID | 27261 |
Feature activation | +1.23e-03 |
Pos logit contributions | |
emort | +0.208 |
Cla | +0.206 |
urable | +0.204 |
scales | +0.204 |
milo | +0.199 |
Neg logit contributions | |
buttocks | -0.162 |
Bench | -0.158 |
ONY | -0.156 |
igan | -0.154 |
venth | -0.154 |
Token | when |
Token ID | 618 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
emort | +0.212 |
urable | +0.210 |
Cla | +0.210 |
scales | +0.208 |
milo | +0.202 |
Neg logit contributions | |
buttocks | -0.241 |
ONY | -0.230 |
Bench | -0.229 |
venth | -0.229 |
igan | -0.228 |
Token | the |
Token ID | 262 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
emort | +0.151 |
urable | +0.149 |
Cla | +0.146 |
scales | +0.145 |
milo | +0.143 |
Neg logit contributions | |
buttocks | -0.194 |
ONY | -0.186 |
Bench | -0.185 |
igan | -0.184 |
venth | -0.184 |
Token | Russian |
Token ID | 3394 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
emort | +0.404 |
urable | +0.400 |
Cla | +0.393 |
scales | +0.388 |
milo | +0.386 |
Neg logit contributions | |
buttocks | -0.438 |
Bench | -0.415 |
ONY | -0.415 |
venth | -0.413 |
igan | -0.412 |
Token | was |
Token ID | 373 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
buttocks | +0.348 |
Bench | +0.331 |
venth | +0.330 |
igan | +0.329 |
ONY | +0.328 |
Neg logit contributions | |
emort | -0.305 |
urable | -0.304 |
Cla | -0.298 |
scales | -0.293 |
milo | -0.291 |
Token | withdrawn |
Token ID | 23019 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
buttocks | +0.600 |
venth | +0.567 |
Bench | +0.566 |
igan | +0.565 |
ONY | +0.565 |
Neg logit contributions | |
emort | -0.558 |
urable | -0.552 |
Cla | -0.545 |
scales | -0.540 |
milo | -0.534 |
Token | due |
Token ID | 2233 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
buttocks | +1.358 |
igan | +1.285 |
venth | +1.279 |
ONY | +1.268 |
Bench | +1.266 |
Neg logit contributions | |
emort | -1.417 |
urable | -1.412 |
Cla | -1.391 |
scales | -1.383 |
milo | -1.372 |
Token | to |
Token ID | 284 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
buttocks | +2.513 |
igan | +2.383 |
Bench | +2.359 |
ONY | +2.357 |
venth | +2.343 |
Neg logit contributions | |
emort | -2.421 |
urable | -2.353 |
Cla | -2.314 |
milo | -2.306 |
scales | -2.264 |
Token | weight |
Token ID | 3463 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
emort | +0.159 |
urable | +0.157 |
Cla | +0.156 |
scales | +0.155 |
milo | +0.152 |
Neg logit contributions | |
buttocks | -0.198 |
ONY | -0.188 |
Bench | -0.188 |
venth | -0.187 |
igan | -0.187 |
Token | cutting |
Token ID | 7720 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
buttocks | +1.930 |
Bench | +1.854 |
igan | +1.837 |
ONY | +1.825 |
venth | +1.822 |
Neg logit contributions | |
emort | -1.647 |
urable | -1.620 |
Cla | -1.603 |
milo | -1.570 |
eq | -1.531 |
Token | difficulties |
Token ID | 13156 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
buttocks | +0.824 |
Bench | +0.778 |
ONY | +0.774 |
igan | +0.773 |
venth | +0.772 |
Neg logit contributions | |
emort | -0.712 |
urable | -0.705 |
Cla | -0.698 |
scales | -0.680 |
milo | -0.679 |
Token | . |
Token ID | 13 |
Feature activation | +8.98e-03 |
Pos logit contributions | |
emort | +0.206 |
urable | +0.204 |
Cla | +0.203 |
scales | +0.201 |
milo | +0.196 |
Neg logit contributions | |
buttocks | -0.244 |
ONY | -0.233 |
Bench | -0.233 |
venth | -0.232 |
igan | -0.231 |
Token | early |
Token ID | 1903 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
buttocks | +0.053 |
ONY | +0.051 |
venth | +0.051 |
igan | +0.051 |
Bench | +0.050 |
Neg logit contributions | |
emort | -0.053 |
Cla | -0.053 |
urable | -0.053 |
milo | -0.052 |
scales | -0.052 |
Token | stage |
Token ID | 3800 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
buttocks | +1.420 |
venth | +1.368 |
Bench | +1.362 |
ONY | +1.360 |
igan | +1.350 |
Neg logit contributions | |
emort | -0.776 |
urable | -0.763 |
Cla | -0.750 |
milo | -0.749 |
scales | -0.732 |
Token | � |
Token ID | 447 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
buttocks | +0.665 |
ONY | +0.635 |
Bench | +0.634 |
venth | +0.631 |
igan | +0.630 |
Neg logit contributions | |
emort | -0.711 |
urable | -0.706 |
Cla | -0.699 |
scales | -0.690 |
milo | -0.682 |
Token | � |
Token ID | 251 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
buttocks | +0.352 |
ONY | +0.337 |
Bench | +0.336 |
venth | +0.335 |
igan | +0.333 |
Neg logit contributions | |
emort | -0.297 |
urable | -0.294 |
Cla | -0.294 |
scales | -0.291 |
milo | -0.283 |
Token | by |
Token ID | 416 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
emort | +0.098 |
urable | +0.097 |
Cla | +0.096 |
scales | +0.096 |
milo | +0.093 |
Neg logit contributions | |
buttocks | -0.111 |
ONY | -0.107 |
Bench | -0.107 |
venth | -0.106 |
igan | -0.105 |
Token | the |
Token ID | 262 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
buttocks | +0.334 |
Bench | +0.321 |
ONY | +0.321 |
venth | +0.318 |
igan | +0.317 |
Neg logit contributions | |
emort | -0.285 |
urable | -0.283 |
Cla | -0.277 |
scales | -0.275 |
milo | -0.272 |
Token | Wall |
Token ID | 5007 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
buttocks | +0.375 |
Bench | +0.358 |
ONY | +0.357 |
venth | +0.355 |
igan | +0.354 |
Neg logit contributions | |
emort | -0.296 |
urable | -0.294 |
Cla | -0.285 |
milo | -0.284 |
scales | -0.280 |
Token | Street |
Token ID | 3530 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
buttocks | +0.380 |
Bench | +0.355 |
ONY | +0.353 |
venth | +0.351 |
igan | +0.348 |
Neg logit contributions | |
emort | -0.869 |
urable | -0.863 |
Cla | -0.859 |
scales | -0.854 |
milo | -0.844 |
Token | Journal |
Token ID | 4913 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
buttocks | +0.750 |
Bench | +0.710 |
igan | +0.688 |
ONY | +0.686 |
venth | +0.685 |
Neg logit contributions | |
emort | -1.412 |
urable | -1.406 |
Cla | -1.386 |
milo | -1.369 |
scales | -1.366 |
Token | . |
Token ID | 13 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
buttocks | +1.000 |
Bench | +0.961 |
ONY | +0.958 |
venth | +0.956 |
igan | +0.950 |
Neg logit contributions | |
emort | -0.947 |
urable | -0.944 |
Cla | -0.930 |
scales | -0.924 |
milo | -0.907 |
Token |
|
Token ID | 198 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
buttocks | +0.400 |
Bench | +0.386 |
ONY | +0.386 |
venth | +0.383 |
igan | +0.382 |
Neg logit contributions | |
emort | -0.355 |
urable | -0.350 |
scales | -0.344 |
Cla | -0.343 |
milo | -0.337 |
Token | has |
Token ID | 468 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
buttocks | +0.066 |
Bench | +0.062 |
ONY | +0.062 |
venth | +0.062 |
igan | +0.062 |
Neg logit contributions | |
emort | -0.054 |
urable | -0.053 |
Cla | -0.053 |
scales | -0.052 |
milo | -0.052 |
Token | also |
Token ID | 635 |
Feature activation | +9.56e-01 |
Pos logit contributions | |
emort | +0.612 |
urable | +0.605 |
Cla | +0.604 |
scales | +0.597 |
milo | +0.585 |
Neg logit contributions | |
buttocks | -0.686 |
Bench | -0.649 |
ONY | -0.649 |
venth | -0.648 |
igan | -0.646 |
Token | warned |
Token ID | 7728 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
buttocks | +0.636 |
igan | +0.595 |
Bench | +0.594 |
venth | +0.591 |
ONY | +0.589 |
Neg logit contributions | |
emort | -0.598 |
urable | -0.589 |
Cla | -0.586 |
scales | -0.577 |
milo | -0.574 |
Token | that |
Token ID | 326 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
emort | +1.087 |
Cla | +1.081 |
urable | +1.079 |
scales | +1.070 |
Kahn | +1.034 |
Neg logit contributions | |
buttocks | -1.369 |
ONY | -1.314 |
Bench | -1.308 |
venth | -1.308 |
igan | -1.297 |
Token | the |
Token ID | 262 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
buttocks | +0.538 |
Bench | +0.512 |
igan | +0.509 |
ONY | +0.503 |
venth | +0.499 |
Neg logit contributions | |
emort | -0.422 |
urable | -0.416 |
milo | -0.408 |
Cla | -0.408 |
scales | -0.401 |
Token | UK |
Token ID | 3482 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
buttocks | +0.333 |
Bench | +0.317 |
igan | +0.313 |
venth | +0.310 |
ONY | +0.309 |
Neg logit contributions | |
emort | -0.316 |
urable | -0.310 |
milo | -0.308 |
Cla | -0.306 |
scales | -0.301 |
Token | will |
Token ID | 481 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
emort | +0.082 |
urable | +0.081 |
Cla | +0.081 |
scales | +0.080 |
milo | +0.078 |
Neg logit contributions | |
buttocks | -0.097 |
Bench | -0.093 |
ONY | -0.092 |
venth | -0.092 |
igan | -0.092 |
Token | not |
Token ID | 407 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
emort | +0.057 |
Cla | +0.056 |
urable | +0.056 |
milo | +0.055 |
scales | +0.055 |
Neg logit contributions | |
buttocks | -0.064 |
igan | -0.061 |
ONY | -0.061 |
Bench | -0.061 |
venth | -0.060 |
Token | escape |
Token ID | 6654 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
buttocks | +0.038 |
igan | +0.036 |
ONY | +0.036 |
venth | +0.036 |
Bench | +0.036 |
Neg logit contributions | |
emort | -0.034 |
Cla | -0.034 |
urable | -0.034 |
milo | -0.033 |
scales | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -9.48e-01 |
Pos logit contributions | |
emort | +0.017 |
urable | +0.017 |
Cla | +0.017 |
milo | +0.017 |
scales | +0.016 |
Neg logit contributions | |
buttocks | -0.021 |
Bench | -0.020 |
igan | -0.020 |
ONY | -0.020 |
venth | -0.020 |
Token | slowdown |
Token ID | 35621 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
emort | +0.555 |
urable | +0.546 |
Cla | +0.542 |
milo | +0.535 |
scales | +0.534 |
Neg logit contributions | |
buttocks | -0.663 |
Bench | -0.629 |
venth | -0.623 |
ONY | -0.622 |
igan | -0.621 |
Token | ping |
Token ID | 29400 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
emort | +0.235 |
urable | +0.233 |
Cla | +0.231 |
scales | +0.227 |
milo | +0.224 |
Neg logit contributions | |
buttocks | -0.255 |
ONY | -0.243 |
Bench | -0.243 |
venth | -0.241 |
igan | -0.239 |
Token | tests |
Token ID | 5254 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
buttocks | +0.897 |
Bench | +0.862 |
ONY | +0.855 |
venth | +0.840 |
igan | +0.839 |
Neg logit contributions | |
emort | -0.982 |
Cla | -0.964 |
urable | -0.963 |
scales | -0.941 |
milo | -0.941 |
Token | will |
Token ID | 481 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
buttocks | +1.793 |
Bench | +1.710 |
ONY | +1.695 |
igan | +1.688 |
venth | +1.663 |
Neg logit contributions | |
emort | -1.625 |
Cla | -1.559 |
urable | -1.554 |
milo | -1.542 |
Kahn | -1.513 |
Token | be |
Token ID | 307 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
buttocks | +0.018 |
ONY | +0.017 |
Bench | +0.017 |
igan | +0.017 |
venth | +0.017 |
Neg logit contributions | |
emort | -0.018 |
Cla | -0.018 |
urable | -0.017 |
milo | -0.017 |
Kahn | -0.017 |
Token | stored |
Token ID | 8574 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
emort | +0.102 |
urable | +0.100 |
Cla | +0.100 |
scales | +0.098 |
milo | +0.098 |
Neg logit contributions | |
buttocks | -0.111 |
Bench | -0.106 |
ONY | -0.106 |
venth | -0.105 |
igan | -0.105 |
Token | in |
Token ID | 287 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
buttocks | +0.026 |
ONY | +0.025 |
Bench | +0.025 |
venth | +0.025 |
igan | +0.025 |
Neg logit contributions | |
emort | -0.026 |
Cla | -0.026 |
urable | -0.026 |
scales | -0.026 |
milo | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
buttocks | +0.532 |
Bench | +0.510 |
ONY | +0.509 |
venth | +0.504 |
igan | +0.503 |
Neg logit contributions | |
emort | -0.396 |
urable | -0.391 |
Cla | -0.387 |
scales | -0.381 |
milo | -0.377 |
Token | S |
Token ID | 311 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
buttocks | +0.542 |
Bench | +0.520 |
ONY | +0.519 |
venth | +0.513 |
igan | +0.512 |
Neg logit contributions | |
emort | -0.489 |
urable | -0.479 |
Cla | -0.478 |
scales | -0.472 |
milo | -0.467 |
Token | 3 |
Token ID | 18 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
buttocks | +0.563 |
ONY | +0.536 |
Bench | +0.534 |
venth | +0.532 |
igan | +0.528 |
Neg logit contributions | |
emort | -0.637 |
Cla | -0.631 |
urable | -0.630 |
scales | -0.625 |
milo | -0.611 |
Token | bucket |
Token ID | 19236 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
buttocks | +0.153 |
Bench | +0.146 |
ONY | +0.146 |
venth | +0.144 |
igan | +0.144 |
Neg logit contributions | |
emort | -0.175 |
Cla | -0.173 |
urable | -0.173 |
scales | -0.171 |
milo | -0.168 |
Token | . |
Token ID | 13 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
emort | +0.398 |
urable | +0.395 |
Cla | +0.395 |
scales | +0.391 |
milo | +0.378 |
Neg logit contributions | |
buttocks | -0.461 |
ONY | -0.441 |
Bench | -0.440 |
venth | -0.439 |
igan | -0.435 |
Token |
|
Token ID | 198 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
buttocks | +0.060 |
ONY | +0.057 |
Bench | +0.057 |
venth | +0.057 |
igan | +0.057 |
Neg logit contributions | |
emort | -0.046 |
urable | -0.045 |
Cla | -0.045 |
scales | -0.044 |
milo | -0.043 |
Token | Should |
Token ID | 19926 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
emort | +0.093 |
Cla | +0.092 |
urable | +0.092 |
scales | +0.091 |
milo | +0.088 |
Neg logit contributions | |
buttocks | -0.110 |
Bench | -0.107 |
ONY | -0.107 |
venth | -0.106 |
Newsletter | -0.105 |
Token | Oklahoma |
Token ID | 10433 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
emort | +0.626 |
urable | +0.613 |
Cla | +0.607 |
scales | +0.601 |
milo | +0.593 |
Neg logit contributions | |
buttocks | -0.795 |
ONY | -0.766 |
Bench | -0.761 |
igan | -0.755 |
venth | -0.754 |
Token | use |
Token ID | 779 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
buttocks | +0.027 |
ONY | +0.026 |
Bench | +0.026 |
venth | +0.026 |
igan | +0.026 |
Neg logit contributions | |
emort | -0.022 |
urable | -0.022 |
Cla | -0.021 |
milo | -0.021 |
scales | -0.021 |
Token | its |
Token ID | 663 |
Feature activation | -8.90e-02 |
Pos logit contributions | |
emort | +0.888 |
urable | +0.875 |
Cla | +0.860 |
scales | +0.854 |
milo | +0.837 |
Neg logit contributions | |
buttocks | -1.201 |
Bench | -1.150 |
ONY | -1.149 |
venth | -1.140 |
igan | -1.134 |
Token | personnel |
Token ID | 8213 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
emort | +0.054 |
urable | +0.053 |
Cla | +0.053 |
scales | +0.052 |
milo | +0.051 |
Neg logit contributions | |
buttocks | -0.060 |
Bench | -0.057 |
ONY | -0.057 |
venth | -0.057 |
igan | -0.056 |
Token | optim |
Token ID | 6436 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
emort | +0.013 |
urable | +0.013 |
Cla | +0.013 |
scales | +0.013 |
milo | +0.013 |
Neg logit contributions | |
buttocks | -0.014 |
ONY | -0.013 |
Bench | -0.013 |
venth | -0.013 |
igan | -0.013 |
Token | ally |
Token ID | 453 |
Feature activation | -9.59e-01 |
Pos logit contributions | |
buttocks | +0.850 |
igan | +0.818 |
venth | +0.817 |
ONY | +0.811 |
Bench | +0.809 |
Neg logit contributions | |
emort | -0.611 |
urable | -0.598 |
Cla | -0.597 |
scales | -0.589 |
milo | -0.573 |
Token | , |
Token ID | 11 |
Feature activation | -3.61e-03 |
Pos logit contributions | |
emort | +0.565 |
urable | +0.557 |
Cla | +0.556 |
scales | +0.549 |
milo | +0.536 |
Neg logit contributions | |
buttocks | -0.661 |
ONY | -0.630 |
Bench | -0.629 |
venth | -0.627 |
igan | -0.626 |
Token | it |
Token ID | 340 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
emort | +0.002 |
urable | +0.002 |
Cla | +0.002 |
scales | +0.002 |
milo | +0.002 |
Neg logit contributions | |
buttocks | -0.003 |
ONY | -0.003 |
Bench | -0.003 |
igan | -0.003 |
venth | -0.003 |
Token | 'll |
Token ID | 1183 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
emort | +0.430 |
urable | +0.424 |
Cla | +0.422 |
scales | +0.412 |
milo | +0.410 |
Neg logit contributions | |
buttocks | -0.481 |
ONY | -0.455 |
venth | -0.455 |
Bench | -0.454 |
igan | -0.453 |
Token | FT |
Token ID | 9792 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
emort | +0.633 |
urable | +0.627 |
Cla | +0.627 |
scales | +0.620 |
milo | +0.602 |
Neg logit contributions | |
buttocks | -0.734 |
ONY | -0.702 |
Bench | -0.700 |
venth | -0.699 |
igan | -0.694 |
Token | ( |
Token ID | 357 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
emort | +0.046 |
urable | +0.045 |
Cla | +0.045 |
scales | +0.044 |
milo | +0.044 |
Neg logit contributions | |
buttocks | -0.051 |
ONY | -0.050 |
Bench | -0.049 |
igan | -0.049 |
venth | -0.049 |
Token | total |
Token ID | 23350 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
buttocks | +0.171 |
Bench | +0.165 |
ONY | +0.163 |
venth | +0.162 |
igan | +0.162 |
Neg logit contributions | |
emort | -0.160 |
Cla | -0.157 |
urable | -0.157 |
scales | -0.156 |
Kahn | -0.151 |
Token | loop |
Token ID | 9052 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
emort | +0.207 |
urable | +0.200 |
Cla | +0.199 |
scales | +0.194 |
milo | +0.193 |
Neg logit contributions | |
buttocks | -0.323 |
Bench | -0.310 |
venth | -0.309 |
ONY | -0.309 |
igan | -0.307 |
Token | - |
Token ID | 532 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
emort | +1.437 |
Cla | +1.420 |
urable | +1.420 |
scales | +1.402 |
milo | +1.360 |
Neg logit contributions | |
buttocks | -2.014 |
ONY | -1.932 |
Bench | -1.927 |
venth | -1.922 |
igan | -1.912 |
Token | 42 |
Token ID | 5433 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
emort | +0.110 |
urable | +0.109 |
Cla | +0.108 |
scales | +0.107 |
milo | +0.105 |
Neg logit contributions | |
buttocks | -0.127 |
Bench | -0.120 |
ONY | -0.120 |
venth | -0.120 |
igan | -0.120 |
Token | mile |
Token ID | 10591 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
emort | +0.175 |
Cla | +0.172 |
urable | +0.171 |
scales | +0.168 |
milo | +0.167 |
Neg logit contributions | |
buttocks | -0.198 |
Bench | -0.190 |
venth | -0.189 |
ONY | -0.189 |
igan | -0.187 |
Token | ) |
Token ID | 8 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
emort | +0.961 |
urable | +0.948 |
Cla | +0.946 |
scales | +0.930 |
milo | +0.911 |
Neg logit contributions | |
buttocks | -1.353 |
Bench | -1.295 |
ONY | -1.294 |
venth | -1.288 |
igan | -1.282 |
Token |
|
Token ID | 198 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
emort | +0.026 |
urable | +0.026 |
Cla | +0.026 |
scales | +0.025 |
milo | +0.025 |
Neg logit contributions | |
buttocks | -0.029 |
Bench | -0.028 |
ONY | -0.028 |
igan | -0.028 |
venth | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
buttocks | +0.172 |
Bench | +0.169 |
ONY | +0.167 |
Newsletter | +0.165 |
venth | +0.165 |
Neg logit contributions | |
emort | -0.144 |
urable | -0.141 |
Cla | -0.141 |
scales | -0.139 |
milo | -0.136 |
Token | Date |
Token ID | 10430 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
emort | +0.197 |
Cla | +0.194 |
urable | +0.193 |
scales | +0.191 |
Kahn | +0.186 |
Neg logit contributions | |
buttocks | -0.230 |
Bench | -0.226 |
ONY | -0.222 |
venth | -0.221 |
Newsletter | -0.221 |
Token | hours |
Token ID | 2250 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
emort | +0.006 |
urable | +0.006 |
Cla | +0.006 |
milo | +0.006 |
scales | +0.006 |
Neg logit contributions | |
buttocks | -0.010 |
ONY | -0.009 |
Bench | -0.009 |
venth | -0.009 |
igan | -0.009 |
Token | spent |
Token ID | 3377 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
Cla | +1.633 |
urable | +1.629 |
emort | +1.629 |
scales | +1.628 |
Kahn | +1.565 |
Neg logit contributions | |
buttocks | -1.793 |
ONY | -1.714 |
Bench | -1.714 |
venth | -1.714 |
igan | -1.690 |
Token | studying |
Token ID | 11065 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
buttocks | +0.394 |
ONY | +0.375 |
Bench | +0.374 |
venth | +0.372 |
igan | +0.370 |
Neg logit contributions | |
emort | -0.408 |
urable | -0.403 |
Cla | -0.402 |
scales | -0.397 |
milo | -0.389 |
Token | to |
Token ID | 284 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
emort | +0.128 |
urable | +0.126 |
Cla | +0.126 |
scales | +0.124 |
milo | +0.122 |
Neg logit contributions | |
buttocks | -0.156 |
ONY | -0.149 |
Bench | -0.149 |
venth | -0.148 |
igan | -0.148 |
Token | achieve |
Token ID | 4620 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
buttocks | +0.245 |
ONY | +0.236 |
venth | +0.235 |
Bench | +0.234 |
igan | +0.233 |
Neg logit contributions | |
emort | -0.205 |
urable | -0.201 |
Cla | -0.201 |
scales | -0.196 |
milo | -0.194 |
Token | it |
Token ID | 340 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
buttocks | +0.303 |
ONY | +0.289 |
Bench | +0.287 |
igan | +0.286 |
venth | +0.286 |
Neg logit contributions | |
emort | -0.222 |
urable | -0.214 |
Cla | -0.214 |
milo | -0.210 |
scales | -0.207 |
Token | . |
Token ID | 13 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
emort | +0.301 |
Cla | +0.299 |
urable | +0.297 |
scales | +0.293 |
milo | +0.287 |
Neg logit contributions | |
buttocks | -0.346 |
ONY | -0.331 |
Bench | -0.329 |
venth | -0.329 |
igan | -0.328 |
Token | A |
Token ID | 317 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
buttocks | +0.412 |
ONY | +0.400 |
Bench | +0.399 |
venth | +0.396 |
igan | +0.394 |
Neg logit contributions | |
emort | -0.385 |
urable | -0.376 |
scales | -0.367 |
Cla | -0.367 |
milo | -0.363 |
Token | willingness |
Token ID | 16826 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
buttocks | +0.747 |
Bench | +0.713 |
venth | +0.712 |
ONY | +0.710 |
igan | +0.705 |
Neg logit contributions | |
emort | -0.573 |
Cla | -0.562 |
urable | -0.560 |
scales | -0.550 |
milo | -0.546 |
Token | to |
Token ID | 284 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
buttocks | +0.165 |
ONY | +0.159 |
Bench | +0.158 |
venth | +0.158 |
igan | +0.157 |
Neg logit contributions | |
emort | -0.129 |
urable | -0.128 |
Cla | -0.128 |
scales | -0.127 |
milo | -0.123 |
Token | learn |
Token ID | 2193 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
emort | +0.331 |
urable | +0.327 |
Cla | +0.327 |
scales | +0.322 |
milo | +0.315 |
Neg logit contributions | |
buttocks | -0.397 |
ONY | -0.379 |
Bench | -0.378 |
venth | -0.377 |
igan | -0.375 |
Token | core |
Token ID | 4755 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
buttocks | +0.046 |
venth | +0.044 |
ONY | +0.044 |
Bench | +0.044 |
igan | +0.044 |
Neg logit contributions | |
emort | -0.040 |
urable | -0.039 |
Cla | -0.039 |
scales | -0.038 |
milo | -0.038 |
Token | is |
Token ID | 318 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
emort | +0.467 |
urable | +0.455 |
Cla | +0.452 |
scales | +0.442 |
milo | +0.437 |
Neg logit contributions | |
buttocks | -0.520 |
Bench | -0.491 |
venth | -0.491 |
ONY | -0.490 |
igan | -0.489 |
Token | a |
Token ID | 257 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
emort | +0.406 |
urable | +0.400 |
Cla | +0.395 |
scales | +0.391 |
milo | +0.383 |
Neg logit contributions | |
buttocks | -0.515 |
ONY | -0.492 |
Bench | -0.489 |
venth | -0.489 |
igan | -0.488 |
Token | handful |
Token ID | 10089 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
emort | +0.370 |
urable | +0.363 |
Cla | +0.361 |
scales | +0.357 |
milo | +0.352 |
Neg logit contributions | |
buttocks | -0.429 |
ONY | -0.407 |
venth | -0.406 |
Bench | -0.406 |
igan | -0.404 |
Token | of |
Token ID | 286 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
Cla | +1.230 |
urable | +1.228 |
emort | +1.225 |
scales | +1.224 |
milo | +1.184 |
Neg logit contributions | |
buttocks | -1.187 |
ONY | -1.129 |
Bench | -1.127 |
venth | -1.121 |
igan | -1.108 |
Token | slow |
Token ID | 3105 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
buttocks | +0.056 |
venth | +0.053 |
ONY | +0.053 |
Bench | +0.053 |
igan | +0.053 |
Neg logit contributions | |
emort | -0.050 |
urable | -0.049 |
Cla | -0.048 |
milo | -0.048 |
scales | -0.047 |
Token | - |
Token ID | 12 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
emort | +0.460 |
urable | +0.457 |
Cla | +0.456 |
scales | +0.450 |
milo | +0.440 |
Neg logit contributions | |
buttocks | -0.505 |
ONY | -0.482 |
Bench | -0.481 |
venth | -0.480 |
igan | -0.477 |
Token | cooked |
Token ID | 46591 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
emort | +0.743 |
urable | +0.738 |
Cla | +0.737 |
scales | +0.732 |
milo | +0.716 |
Neg logit contributions | |
buttocks | -0.500 |
ONY | -0.471 |
Bench | -0.468 |
venth | -0.467 |
igan | -0.462 |
Token | meat |
Token ID | 6174 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
emort | +0.316 |
urable | +0.313 |
Cla | +0.310 |
scales | +0.306 |
milo | +0.302 |
Neg logit contributions | |
buttocks | -0.335 |
ONY | -0.317 |
Bench | -0.317 |
venth | -0.316 |
igan | -0.315 |
Token | dishes |
Token ID | 16759 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
emort | +0.586 |
Cla | +0.580 |
urable | +0.580 |
scales | +0.574 |
milo | +0.551 |
Neg logit contributions | |
buttocks | -0.987 |
ONY | -0.951 |
Bench | -0.948 |
venth | -0.947 |
igan | -0.941 |
Token | , |
Token ID | 11 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
emort | +0.577 |
Cla | +0.573 |
urable | +0.572 |
scales | +0.567 |
milo | +0.550 |
Neg logit contributions | |
buttocks | -0.640 |
ONY | -0.612 |
Bench | -0.610 |
venth | -0.609 |
igan | -0.604 |
Token | in |
Token ID | 287 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
emort | +0.386 |
urable | +0.383 |
Cla | +0.375 |
milo | +0.370 |
scales | +0.369 |
Neg logit contributions | |
buttocks | -0.460 |
ONY | -0.437 |
Bench | -0.434 |
venth | -0.432 |
igan | -0.431 |
Token | some |
Token ID | 617 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
emort | +0.572 |
urable | +0.566 |
Cla | +0.564 |
scales | +0.557 |
milo | +0.543 |
Neg logit contributions | |
buttocks | -0.702 |
ONY | -0.672 |
Bench | -0.669 |
venth | -0.669 |
igan | -0.664 |
Token | cases |
Token ID | 2663 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
emort | +0.148 |
urable | +0.146 |
Cla | +0.145 |
scales | +0.141 |
milo | +0.140 |
Neg logit contributions | |
buttocks | -0.191 |
ONY | -0.182 |
Bench | -0.182 |
venth | -0.182 |
igan | -0.181 |
Token | they |
Token ID | 484 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
emort | +0.644 |
urable | +0.641 |
Cla | +0.630 |
scales | +0.620 |
milo | +0.612 |
Neg logit contributions | |
buttocks | -0.780 |
ONY | -0.742 |
Bench | -0.738 |
venth | -0.737 |
igan | -0.733 |
Token | seemed |
Token ID | 3947 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
emort | +0.499 |
urable | +0.494 |
Cla | +0.493 |
scales | +0.487 |
milo | +0.475 |
Neg logit contributions | |
buttocks | -0.566 |
ONY | -0.539 |
venth | -0.537 |
Bench | -0.537 |
igan | -0.533 |
Token | to |
Token ID | 284 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
emort | +0.669 |
urable | +0.662 |
Cla | +0.662 |
scales | +0.655 |
milo | +0.636 |
Neg logit contributions | |
buttocks | -0.845 |
ONY | -0.809 |
Bench | -0.807 |
venth | -0.805 |
igan | -0.800 |
Token | reflect |
Token ID | 4079 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
emort | +0.412 |
urable | +0.404 |
Cla | +0.404 |
scales | +0.397 |
milo | +0.391 |
Neg logit contributions | |
buttocks | -0.546 |
ONY | -0.522 |
venth | -0.519 |
Bench | -0.517 |
igan | -0.516 |
Token | alternate |
Token ID | 13527 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
emort | +0.089 |
urable | +0.088 |
Cla | +0.087 |
scales | +0.085 |
milo | +0.085 |
Neg logit contributions | |
buttocks | -0.100 |
ONY | -0.096 |
venth | -0.095 |
Bench | -0.094 |
igan | -0.094 |
Token | interpretations |
Token ID | 26146 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
buttocks | +0.596 |
igan | +0.567 |
ONY | +0.564 |
venth | +0.564 |
Bench | +0.563 |
Neg logit contributions | |
emort | -0.396 |
urable | -0.394 |
Cla | -0.390 |
milo | -0.376 |
scales | -0.375 |
Token | of |
Token ID | 286 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
emort | +0.643 |
Cla | +0.639 |
urable | +0.638 |
scales | +0.633 |
milo | +0.614 |
Neg logit contributions | |
buttocks | -0.661 |
ONY | -0.630 |
Bench | -0.628 |
venth | -0.626 |
igan | -0.622 |
Token | the |
Token ID | 262 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
urable | +0.090 |
emort | +0.090 |
Cla | +0.089 |
scales | +0.087 |
milo | +0.087 |
Neg logit contributions | |
buttocks | -0.127 |
ONY | -0.121 |
Bench | -0.121 |
venth | -0.120 |
igan | -0.120 |
Token | — |
Token ID | 851 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
emort | +0.124 |
urable | +0.123 |
Cla | +0.122 |
scales | +0.121 |
milo | +0.118 |
Neg logit contributions | |
buttocks | -0.134 |
ONY | -0.128 |
Bench | -0.127 |
venth | -0.127 |
igan | -0.126 |
Token | though |
Token ID | 996 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
emort | +0.163 |
urable | +0.162 |
Cla | +0.160 |
scales | +0.158 |
milo | +0.155 |
Neg logit contributions | |
buttocks | -0.194 |
ONY | -0.184 |
Bench | -0.183 |
venth | -0.182 |
igan | -0.181 |
Token | Major |
Token ID | 8386 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
emort | +0.702 |
urable | +0.696 |
Cla | +0.688 |
scales | +0.681 |
milo | +0.663 |
Neg logit contributions | |
buttocks | -0.954 |
ONY | -0.911 |
Bench | -0.908 |
venth | -0.904 |
igan | -0.902 |
Token | League |
Token ID | 4041 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
buttocks | +1.039 |
igan | +0.988 |
Bench | +0.988 |
ONY | +0.981 |
venth | +0.968 |
Neg logit contributions | |
urable | -1.432 |
emort | -1.422 |
milo | -1.386 |
Cla | -1.379 |
eq | -1.371 |
Token | Eating |
Token ID | 37537 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
emort | +0.501 |
urable | +0.500 |
Cla | +0.491 |
scales | +0.485 |
milo | +0.478 |
Neg logit contributions | |
buttocks | -0.729 |
ONY | -0.699 |
Bench | -0.698 |
venth | -0.692 |
igan | -0.692 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
emort | +0.089 |
urable | +0.088 |
Cla | +0.087 |
scales | +0.086 |
milo | +0.085 |
Neg logit contributions | |
buttocks | -0.102 |
Bench | -0.097 |
ONY | -0.097 |
igan | -0.097 |
venth | -0.097 |
Token | to |
Token ID | 284 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
emort | +0.736 |
Cla | +0.731 |
urable | +0.730 |
scales | +0.723 |
milo | +0.699 |
Neg logit contributions | |
buttocks | -0.939 |
ONY | -0.900 |
Bench | -0.898 |
venth | -0.895 |
igan | -0.889 |
Token | take |
Token ID | 1011 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
emort | +0.919 |
urable | +0.910 |
Cla | +0.910 |
scales | +0.900 |
milo | +0.876 |
Neg logit contributions | |
buttocks | -1.038 |
ONY | -0.990 |
Bench | -0.987 |
venth | -0.985 |
igan | -0.979 |
Token | the |
Token ID | 262 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
emort | +0.433 |
urable | +0.427 |
Cla | +0.419 |
scales | +0.416 |
milo | +0.409 |
Neg logit contributions | |
buttocks | -0.623 |
ONY | -0.593 |
venth | -0.591 |
Bench | -0.590 |
igan | -0.590 |
Token | high |
Token ID | 1029 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
emort | +0.057 |
urable | +0.057 |
milo | +0.055 |
Cla | +0.055 |
scales | +0.054 |
Neg logit contributions | |
buttocks | -0.074 |
ONY | -0.069 |
Bench | -0.069 |
igan | -0.069 |
venth | -0.069 |
Token | road |
Token ID | 2975 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
emort | +0.259 |
urable | +0.255 |
Cla | +0.250 |
scales | +0.245 |
Kahn | +0.243 |
Neg logit contributions | |
buttocks | -0.334 |
Bench | -0.318 |
igan | -0.316 |
ONY | -0.315 |
venth | -0.315 |
Token | into |
Token ID | 656 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
buttocks | +0.546 |
ONY | +0.519 |
Bench | +0.517 |
venth | +0.516 |
igan | +0.516 |
Neg logit contributions | |
emort | -0.495 |
urable | -0.488 |
Cla | -0.488 |
scales | -0.481 |
milo | -0.472 |
Token | the |
Token ID | 262 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
emort | +0.216 |
urable | +0.214 |
Cla | +0.210 |
scales | +0.207 |
milo | +0.204 |
Neg logit contributions | |
buttocks | -0.298 |
ONY | -0.283 |
venth | -0.282 |
Bench | -0.282 |
igan | -0.281 |
Token | trash |
Token ID | 13913 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
buttocks | +0.189 |
ONY | +0.179 |
Bench | +0.179 |
venth | +0.179 |
igan | +0.178 |
Neg logit contributions | |
emort | -0.150 |
urable | -0.148 |
Cla | -0.147 |
scales | -0.144 |
milo | -0.142 |
Token | , |
Token ID | 11 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
emort | +0.704 |
urable | +0.698 |
Cla | +0.698 |
scales | +0.691 |
milo | +0.672 |
Neg logit contributions | |
buttocks | -0.750 |
ONY | -0.716 |
Bench | -0.714 |
venth | -0.712 |
igan | -0.707 |
Token | � |
Token ID | 564 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
buttocks | +0.247 |
ONY | +0.233 |
Bench | +0.233 |
venth | +0.231 |
igan | +0.231 |
Neg logit contributions | |
emort | -0.239 |
urable | -0.238 |
Cla | -0.236 |
scales | -0.232 |
milo | -0.229 |
Token | � |
Token ID | 250 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
buttocks | +0.453 |
ONY | +0.434 |
Bench | +0.432 |
venth | +0.431 |
igan | +0.428 |
Neg logit contributions | |
emort | -0.378 |
urable | -0.374 |
Cla | -0.374 |
scales | -0.370 |
milo | -0.359 |
Token | I |
Token ID | 40 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
emort | +0.384 |
Cla | +0.380 |
urable | +0.380 |
scales | +0.376 |
Kahn | +0.365 |
Neg logit contributions | |
buttocks | -0.447 |
ONY | -0.427 |
Bench | -0.427 |
venth | -0.425 |
igan | -0.423 |
Token | strip |
Token ID | 10283 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
emort | +0.304 |
urable | +0.301 |
Cla | +0.301 |
scales | +0.298 |
milo | +0.290 |
Neg logit contributions | |
buttocks | -0.327 |
ONY | -0.312 |
Bench | -0.311 |
venth | -0.310 |
igan | -0.308 |
Token | off |
Token ID | 572 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
buttocks | +0.353 |
ONY | +0.337 |
Bench | +0.336 |
venth | +0.335 |
igan | +0.334 |
Neg logit contributions | |
emort | -0.309 |
urable | -0.306 |
Cla | -0.306 |
scales | -0.302 |
milo | -0.294 |
Token | the |
Token ID | 262 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
buttocks | +1.551 |
igan | +1.477 |
ONY | +1.464 |
venth | +1.458 |
Bench | +1.449 |
Neg logit contributions | |
emort | -1.335 |
urable | -1.322 |
Cla | -1.302 |
scales | -1.278 |
Kahn | -1.261 |
Token | identity |
Token ID | 5369 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
buttocks | +0.201 |
ONY | +0.191 |
Bench | +0.190 |
venth | +0.190 |
igan | +0.189 |
Neg logit contributions | |
emort | -0.183 |
urable | -0.181 |
Cla | -0.181 |
scales | -0.178 |
milo | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
buttocks | +0.397 |
igan | +0.379 |
Bench | +0.378 |
ONY | +0.378 |
venth | +0.376 |
Neg logit contributions | |
emort | -0.331 |
urable | -0.324 |
Cla | -0.321 |
scales | -0.318 |
milo | -0.313 |
Token | If |
Token ID | 1002 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
emort | +0.042 |
urable | +0.041 |
scales | +0.041 |
Cla | +0.040 |
milo | +0.040 |
Neg logit contributions | |
buttocks | -0.053 |
ONY | -0.052 |
Bench | -0.052 |
igan | -0.051 |
venth | -0.051 |
Token | he |
Token ID | 339 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
buttocks | +0.394 |
ONY | +0.380 |
Bench | +0.379 |
igan | +0.377 |
venth | +0.376 |
Neg logit contributions | |
emort | -0.273 |
urable | -0.272 |
Cla | -0.267 |
scales | -0.264 |
milo | -0.259 |
Token | did |
Token ID | 750 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
emort | +0.351 |
urable | +0.348 |
Cla | +0.348 |
scales | +0.343 |
milo | +0.337 |
Neg logit contributions | |
buttocks | -0.314 |
ONY | -0.296 |
Bench | -0.296 |
venth | -0.296 |
igan | -0.294 |
Token | not |
Token ID | 407 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
emort | +0.446 |
urable | +0.445 |
Cla | +0.442 |
scales | +0.437 |
milo | +0.430 |
Neg logit contributions | |
buttocks | -0.474 |
ONY | -0.452 |
igan | -0.450 |
Bench | -0.449 |
venth | -0.448 |
Token | set |
Token ID | 900 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
emort | +0.445 |
urable | +0.441 |
Cla | +0.440 |
scales | +0.434 |
milo | +0.426 |
Neg logit contributions | |
buttocks | -0.514 |
ONY | -0.487 |
venth | -0.486 |
igan | -0.485 |
Bench | -0.485 |
Token | it |
Token ID | 340 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
emort | +1.985 |
Cla | +1.970 |
urable | +1.964 |
scales | +1.951 |
Kahn | +1.887 |
Neg logit contributions | |
buttocks | -2.524 |
ONY | -2.425 |
Bench | -2.414 |
venth | -2.410 |
igan | -2.392 |
Token | in |
Token ID | 287 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
emort | +0.271 |
urable | +0.268 |
Cla | +0.268 |
scales | +0.266 |
milo | +0.258 |
Neg logit contributions | |
buttocks | -0.301 |
ONY | -0.287 |
Bench | -0.286 |
venth | -0.286 |
igan | -0.284 |
Token | motion |
Token ID | 6268 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
urable | +0.091 |
emort | +0.091 |
Cla | +0.089 |
scales | +0.088 |
milo | +0.085 |
Neg logit contributions | |
buttocks | -0.171 |
ONY | -0.164 |
Bench | -0.164 |
igan | -0.164 |
venth | -0.163 |
Token | , |
Token ID | 11 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
buttocks | +1.267 |
igan | +1.211 |
ONY | +1.185 |
venth | +1.183 |
Bench | +1.180 |
Neg logit contributions | |
urable | -1.285 |
Cla | -1.281 |
emort | -1.279 |
scales | -1.258 |
milo | -1.243 |
Token | he |
Token ID | 339 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
emort | +0.025 |
urable | +0.025 |
Cla | +0.024 |
scales | +0.024 |
milo | +0.024 |
Neg logit contributions | |
buttocks | -0.033 |
ONY | -0.032 |
Bench | -0.032 |
venth | -0.031 |
igan | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
buttocks | +0.324 |
ONY | +0.317 |
Bench | +0.314 |
igan | +0.313 |
Newsletter | +0.312 |
Neg logit contributions | |
emort | -0.292 |
urable | -0.287 |
scales | -0.280 |
milo | -0.275 |
Cla | -0.275 |
Token |
|
Token ID | 198 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
buttocks | +0.640 |
Bench | +0.619 |
ONY | +0.617 |
igan | +0.614 |
venth | +0.613 |
Neg logit contributions | |
emort | -0.530 |
urable | -0.514 |
Cla | -0.511 |
scales | -0.506 |
milo | -0.497 |
Token | It |
Token ID | 1026 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
Bench | +0.345 |
buttocks | +0.345 |
ONY | +0.342 |
Newsletter | +0.342 |
venth | +0.337 |
Neg logit contributions | |
emort | -0.306 |
scales | -0.297 |
Cla | -0.297 |
urable | -0.297 |
Kahn | -0.284 |
Token | 's |
Token ID | 338 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
buttocks | +0.260 |
venth | +0.248 |
ONY | +0.248 |
Bench | +0.247 |
igan | +0.247 |
Neg logit contributions | |
emort | -0.226 |
urable | -0.223 |
Cla | -0.221 |
scales | -0.215 |
milo | -0.213 |
Token | not |
Token ID | 407 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
buttocks | +0.305 |
ONY | +0.289 |
Bench | +0.288 |
igan | +0.287 |
venth | +0.286 |
Neg logit contributions | |
emort | -0.256 |
urable | -0.249 |
Cla | -0.247 |
scales | -0.244 |
milo | -0.243 |
Token | hard |
Token ID | 1327 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
emort | +0.224 |
urable | +0.219 |
Cla | +0.219 |
scales | +0.216 |
milo | +0.213 |
Neg logit contributions | |
buttocks | -0.258 |
ONY | -0.245 |
venth | -0.243 |
Bench | -0.243 |
igan | -0.242 |
Token | to |
Token ID | 284 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
urable | +2.587 |
scales | +2.582 |
Cla | +2.562 |
emort | +2.543 |
eq | +2.464 |
Neg logit contributions | |
buttocks | -3.958 |
ONY | -3.823 |
venth | -3.801 |
Bench | -3.776 |
Newsletter | -3.724 |
Token | figure |
Token ID | 3785 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
emort | +0.390 |
Cla | +0.384 |
urable | +0.383 |
scales | +0.377 |
milo | +0.371 |
Neg logit contributions | |
buttocks | -0.528 |
ONY | -0.504 |
venth | -0.503 |
Bench | -0.502 |
igan | -0.502 |
Token | out |
Token ID | 503 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
buttocks | +0.011 |
ONY | +0.010 |
igan | +0.010 |
Bench | +0.010 |
venth | +0.010 |
Neg logit contributions | |
emort | -0.012 |
urable | -0.011 |
Cla | -0.011 |
scales | -0.011 |
milo | -0.011 |
Token | which |
Token ID | 543 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
emort | +0.658 |
urable | +0.652 |
Cla | +0.646 |
scales | +0.640 |
milo | +0.629 |
Neg logit contributions | |
buttocks | -0.648 |
ONY | -0.616 |
Bench | -0.613 |
venth | -0.612 |
igan | -0.611 |
Token | of |
Token ID | 286 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
emort | +0.006 |
urable | +0.005 |
Cla | +0.005 |
milo | +0.005 |
scales | +0.005 |
Neg logit contributions | |
buttocks | -0.008 |
ONY | -0.007 |
igan | -0.007 |
venth | -0.007 |
Bench | -0.007 |
Token | always |
Token ID | 1464 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
buttocks | +1.359 |
igan | +1.294 |
venth | +1.286 |
ONY | +1.278 |
Bench | +1.270 |
Neg logit contributions | |
emort | -1.352 |
urable | -1.327 |
Cla | -1.305 |
milo | -1.272 |
scales | -1.267 |
Token | done |
Token ID | 1760 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
emort | +0.125 |
urable | +0.123 |
Cla | +0.123 |
scales | +0.121 |
milo | +0.119 |
Neg logit contributions | |
buttocks | -0.120 |
ONY | -0.113 |
venth | -0.113 |
Bench | -0.113 |
igan | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
buttocks | +0.548 |
ONY | +0.521 |
igan | +0.520 |
venth | +0.518 |
Bench | +0.517 |
Neg logit contributions | |
emort | -0.509 |
urable | -0.502 |
Cla | -0.500 |
scales | -0.489 |
milo | -0.485 |
Token | As |
Token ID | 1081 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
buttocks | +0.818 |
ONY | +0.788 |
Bench | +0.784 |
venth | +0.781 |
igan | +0.778 |
Neg logit contributions | |
emort | -0.754 |
urable | -0.743 |
scales | -0.731 |
Cla | -0.730 |
milo | -0.718 |
Token | a |
Token ID | 257 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
buttocks | +0.717 |
ONY | +0.682 |
Bench | +0.680 |
venth | +0.679 |
igan | +0.677 |
Neg logit contributions | |
emort | -0.602 |
urable | -0.591 |
Cla | -0.583 |
scales | -0.576 |
milo | -0.573 |
Token | result |
Token ID | 1255 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
emort | +0.894 |
Cla | +0.893 |
urable | +0.888 |
scales | +0.885 |
Kahn | +0.856 |
Neg logit contributions | |
buttocks | -1.059 |
ONY | -1.018 |
Bench | -1.015 |
venth | -1.010 |
igan | -1.003 |
Token | of |
Token ID | 286 |
Feature activation | -9.10e-01 |
Pos logit contributions | |
Cla | +2.648 |
scales | +2.614 |
urable | +2.585 |
emort | +2.580 |
Kahn | +2.537 |
Neg logit contributions | |
buttocks | -3.087 |
Bench | -2.959 |
venth | -2.944 |
ONY | -2.942 |
igan | -2.872 |
Token | that |
Token ID | 326 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
emort | +0.523 |
Cla | +0.521 |
urable | +0.518 |
scales | +0.516 |
Kahn | +0.499 |
Neg logit contributions | |
buttocks | -0.636 |
ONY | -0.608 |
Bench | -0.607 |
venth | -0.606 |
igan | -0.601 |
Token | , |
Token ID | 11 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
emort | +0.063 |
urable | +0.063 |
Cla | +0.063 |
scales | +0.062 |
milo | +0.060 |
Neg logit contributions | |
buttocks | -0.070 |
ONY | -0.067 |
Bench | -0.067 |
venth | -0.066 |
igan | -0.066 |
Token | Fred |
Token ID | 8559 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
emort | +0.195 |
urable | +0.193 |
Cla | +0.192 |
scales | +0.190 |
milo | +0.186 |
Neg logit contributions | |
buttocks | -0.219 |
ONY | -0.209 |
Bench | -0.208 |
venth | -0.208 |
igan | -0.207 |
Token | 's |
Token ID | 338 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
buttocks | +1.003 |
ONY | +0.965 |
igan | +0.961 |
venth | +0.959 |
Bench | +0.958 |
Neg logit contributions | |
emort | -0.857 |
urable | -0.847 |
Cla | -0.836 |
scales | -0.826 |
milo | -0.816 |
Token | also |
Token ID | 635 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
emort | +0.236 |
urable | +0.231 |
Cla | +0.230 |
scales | +0.227 |
milo | +0.226 |
Neg logit contributions | |
buttocks | -0.269 |
ONY | -0.255 |
Bench | -0.255 |
venth | -0.254 |
igan | -0.252 |
Token | available |
Token ID | 1695 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
buttocks | +0.569 |
ONY | +0.538 |
Bench | +0.537 |
venth | +0.533 |
igan | +0.530 |
Neg logit contributions | |
emort | -0.462 |
urable | -0.447 |
Cla | -0.443 |
milo | -0.441 |
scales | -0.433 |
Token | as |
Token ID | 355 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
buttocks | +1.667 |
igan | +1.573 |
ONY | +1.558 |
venth | +1.545 |
Bench | +1.544 |
Neg logit contributions | |
emort | -1.981 |
urable | -1.937 |
Cla | -1.922 |
milo | -1.900 |
scales | -1.888 |
Token | standalone |
Token ID | 27669 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
buttocks | +0.401 |
ONY | +0.380 |
Bench | +0.379 |
venth | +0.377 |
igan | +0.374 |
Neg logit contributions | |
emort | -0.318 |
urable | -0.312 |
Cla | -0.305 |
milo | -0.302 |
scales | -0.297 |
Token | multimedia |
Token ID | 40162 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
emort | +0.903 |
urable | +0.889 |
Cla | +0.878 |
scales | +0.859 |
milo | +0.850 |
Neg logit contributions | |
buttocks | -1.514 |
ONY | -1.448 |
Bench | -1.444 |
venth | -1.442 |
igan | -1.436 |
Token | CD |
Token ID | 6458 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
emort | +0.608 |
urable | +0.595 |
Cla | +0.588 |
milo | +0.575 |
scales | +0.574 |
Neg logit contributions | |
buttocks | -0.883 |
ONY | -0.841 |
Bench | -0.837 |
venth | -0.836 |
igan | -0.831 |
Token | post |
Token ID | 1281 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
emort | +2.283 |
Cla | +2.281 |
urable | +2.276 |
scales | +2.263 |
Kahn | +2.170 |
Neg logit contributions | |
buttocks | -3.003 |
Bench | -2.871 |
ONY | -2.868 |
venth | -2.863 |
igan | -2.843 |
Token | cards |
Token ID | 27761 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
emort | +0.091 |
Cla | +0.089 |
urable | +0.088 |
scales | +0.087 |
milo | +0.086 |
Neg logit contributions | |
buttocks | -0.165 |
venth | -0.160 |
ONY | -0.160 |
igan | -0.159 |
Bench | -0.159 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
emort | +0.255 |
urable | +0.252 |
Cla | +0.251 |
scales | +0.248 |
milo | +0.243 |
Neg logit contributions | |
buttocks | -0.274 |
ONY | -0.261 |
Bench | -0.260 |
venth | -0.259 |
igan | -0.258 |
Token |
|
Token ID | 198 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
buttocks | +0.114 |
ONY | +0.111 |
Bench | +0.110 |
venth | +0.110 |
igan | +0.109 |
Neg logit contributions | |
emort | -0.100 |
urable | -0.098 |
scales | -0.096 |
Cla | -0.096 |
milo | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
buttocks | +0.764 |
Bench | +0.738 |
ONY | +0.736 |
igan | +0.734 |
venth | +0.731 |
Neg logit contributions | |
emort | -0.644 |
urable | -0.621 |
Cla | -0.618 |
scales | -0.611 |
milo | -0.602 |
Token | frying |
Token ID | 45366 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
emort | +0.072 |
urable | +0.071 |
Cla | +0.070 |
scales | +0.069 |
milo | +0.068 |
Neg logit contributions | |
buttocks | -0.080 |
ONY | -0.076 |
Bench | -0.076 |
venth | -0.076 |
igan | -0.075 |
Token | pan |
Token ID | 3425 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
emort | +1.092 |
Cla | +1.077 |
urable | +1.075 |
scales | +1.063 |
milo | +1.033 |
Neg logit contributions | |
buttocks | -1.500 |
ONY | -1.437 |
Bench | -1.433 |
venth | -1.429 |
igan | -1.422 |
Token | along |
Token ID | 1863 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
emort | +1.045 |
urable | +1.039 |
Cla | +1.037 |
scales | +1.030 |
milo | +0.993 |
Neg logit contributions | |
buttocks | -1.285 |
ONY | -1.231 |
Bench | -1.227 |
venth | -1.225 |
igan | -1.216 |
Token | with |
Token ID | 351 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
emort | +0.316 |
urable | +0.311 |
Cla | +0.310 |
scales | +0.305 |
milo | +0.300 |
Neg logit contributions | |
buttocks | -0.387 |
ONY | -0.371 |
Bench | -0.370 |
venth | -0.369 |
igan | -0.367 |
Token | 1 |
Token ID | 352 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
buttocks | +0.594 |
ONY | +0.563 |
igan | +0.562 |
Bench | +0.561 |
venth | +0.560 |
Neg logit contributions | |
emort | -0.448 |
urable | -0.447 |
Cla | -0.432 |
scales | -0.426 |
milo | -0.421 |
Token | tablespoon |
Token ID | 33107 |
Feature activation | -4.25e+00 |
Pos logit contributions | |
emort | +0.116 |
urable | +0.115 |
Cla | +0.115 |
scales | +0.114 |
milo | +0.111 |
Neg logit contributions | |
buttocks | -0.100 |
ONY | -0.095 |
Bench | -0.095 |
venth | -0.095 |
igan | -0.094 |
Token | coconut |
Token ID | 20132 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
Cla | +2.409 |
scales | +2.387 |
urable | +2.365 |
emort | +2.361 |
milo | +2.299 |
Neg logit contributions | |
buttocks | -2.976 |
Bench | -2.871 |
ONY | -2.865 |
venth | -2.857 |
Newsletter | -2.817 |
Token | oil |
Token ID | 3056 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
emort | +0.006 |
urable | +0.006 |
Cla | +0.006 |
scales | +0.006 |
milo | +0.005 |
Neg logit contributions | |
buttocks | -0.013 |
ONY | -0.013 |
igan | -0.013 |
venth | -0.013 |
Bench | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
emort | +0.090 |
urable | +0.089 |
Cla | +0.089 |
scales | +0.088 |
milo | +0.086 |
Neg logit contributions | |
buttocks | -0.105 |
ONY | -0.100 |
Bench | -0.100 |
venth | -0.100 |
igan | -0.099 |
Token | fry |
Token ID | 32959 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
emort | +0.146 |
urable | +0.144 |
Cla | +0.142 |
scales | +0.140 |
milo | +0.138 |
Neg logit contributions | |
buttocks | -0.170 |
ONY | -0.162 |
Bench | -0.161 |
venth | -0.160 |
igan | -0.160 |
Token | it |
Token ID | 340 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
emort | +0.647 |
urable | +0.641 |
Cla | +0.638 |
scales | +0.627 |
milo | +0.615 |
Neg logit contributions | |
buttocks | -0.867 |
ONY | -0.823 |
Bench | -0.823 |
igan | -0.817 |
venth | -0.817 |