Token | : |
Token ID | 25 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
gements | +0.158 |
attest | +0.154 |
emed | +0.147 |
Blossom | +0.141 |
feature | +0.141 |
Neg logit contributions | |
disrupting | -0.155 |
disruption | -0.148 |
obstruction | -0.145 |
enforcing | -0.144 |
interference | -0.141 |
Token | $ |
Token ID | 720 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
disrupting | +0.637 |
disruption | +0.603 |
obstruction | +0.590 |
enforcing | +0.588 |
interference | +0.574 |
Neg logit contributions | |
gements | -0.778 |
attest | -0.758 |
emed | -0.732 |
Blossom | -0.707 |
creen | -0.705 |
Token | 8 |
Token ID | 23 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
gements | +1.681 |
attest | +1.644 |
emed | +1.565 |
feature | +1.526 |
Blossom | +1.516 |
Neg logit contributions | |
disrupting | -1.378 |
disruption | -1.299 |
obstruction | -1.275 |
enforcing | -1.263 |
HUD | -1.247 |
Token | million |
Token ID | 1510 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
disrupting | +5.590 |
enforcing | +5.311 |
disruption | +5.297 |
obstruction | +5.151 |
punishing | +5.077 |
Neg logit contributions | |
gements | -6.378 |
attest | -6.087 |
Blossom | -5.872 |
emed | -5.869 |
creen | -5.771 |
Token | ( |
Token ID | 357 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
disrupting | +0.725 |
disruption | +0.688 |
enforcing | +0.676 |
obstruction | +0.676 |
interference | +0.657 |
Neg logit contributions | |
gements | -0.795 |
attest | -0.765 |
emed | -0.738 |
Blossom | -0.719 |
feature | -0.710 |
Token | t |
Token ID | 83 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
gements | +1.721 |
attest | +1.686 |
emed | +1.604 |
feature | +1.562 |
Blossom | +1.556 |
Neg logit contributions | |
disrupting | -1.368 |
disruption | -1.290 |
obstruction | -1.272 |
enforcing | -1.254 |
HUD | -1.249 |
Token | ied |
Token ID | 798 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
oxic | +9.311 |
disrupting | +8.900 |
disruption | +8.476 |
AKING | +8.458 |
oglu | +8.457 |
Neg logit contributions | |
gements | -17.382 |
creen | -16.327 |
attest | -16.103 |
soever | -15.750 |
Blossom | -15.381 |
Token | for |
Token ID | 329 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
gements | +1.198 |
attest | +1.164 |
emed | +1.110 |
Blossom | +1.080 |
creen | +1.074 |
Neg logit contributions | |
disrupting | -1.102 |
disruption | -1.048 |
obstruction | -1.038 |
enforcing | -1.028 |
interference | -1.006 |
Token | 25 |
Token ID | 1679 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
gements | +2.027 |
attest | +1.958 |
emed | +1.882 |
creen | +1.799 |
feature | +1.794 |
Neg logit contributions | |
disrupting | -1.949 |
disruption | -1.851 |
obstruction | -1.827 |
enforcing | -1.799 |
interference | -1.772 |
Token | th |
Token ID | 400 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
gements | +3.236 |
attest | +3.124 |
emed | +2.929 |
feature | +2.829 |
Blossom | +2.802 |
Neg logit contributions | |
disrupting | -5.338 |
disruption | -5.120 |
obstruction | -5.053 |
enforcing | -4.999 |
interference | -4.935 |
Token | among |
Token ID | 1871 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
gements | +3.027 |
attest | +2.917 |
emed | +2.765 |
feature | +2.641 |
Blossom | +2.637 |
Neg logit contributions | |
disrupting | -4.418 |
disruption | -4.230 |
obstruction | -4.177 |
enforcing | -4.136 |
interference | -4.074 |
Token |
|
Token ID | 198 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
disrupting | +2.867 |
disruption | +2.685 |
obstruction | +2.644 |
HUD | +2.604 |
enforcing | +2.591 |
Neg logit contributions | |
gements | -5.181 |
attest | -5.060 |
emed | -4.794 |
creen | -4.671 |
Blossom | -4.627 |
Token | Tom |
Token ID | 13787 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
disrupting | +1.186 |
HUD | +1.133 |
disruption | +1.119 |
obstruction | +1.092 |
interference | +1.083 |
Neg logit contributions | |
gements | -1.386 |
attest | -1.369 |
emed | -1.275 |
Blossom | -1.271 |
feature | -1.261 |
Token | Wilson |
Token ID | 8127 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
disrupting | +6.278 |
disruption | +6.144 |
enforcing | +5.715 |
interference | +5.714 |
obstruction | +5.653 |
Neg logit contributions | |
attest | -7.794 |
gements | -7.596 |
emed | -7.490 |
respectively | -7.405 |
feature | -7.394 |
Token |
|
Token ID | 198 |
Feature activation | +9.80e-01 |
Pos logit contributions | |
disrupting | +8.335 |
disruption | +7.832 |
interference | +7.724 |
obstruction | +7.691 |
enforcing | +7.678 |
Neg logit contributions | |
gements | -11.792 |
attest | -11.479 |
Blossom | -11.098 |
emed | -11.029 |
creen | -10.872 |
Token |
|
Token ID | 198 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
disrupting | +2.776 |
disruption | +2.600 |
obstruction | +2.572 |
HUD | +2.520 |
enforcing | +2.517 |
Neg logit contributions | |
gements | -5.244 |
attest | -5.127 |
emed | -4.842 |
creen | -4.707 |
Blossom | -4.672 |
Token | T |
Token ID | 51 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
disrupting | +1.337 |
HUD | +1.296 |
disruption | +1.264 |
obstruction | +1.235 |
interference | +1.229 |
Neg logit contributions | |
gements | -1.565 |
attest | -1.550 |
emed | -1.441 |
Blossom | -1.432 |
feature | -1.424 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
oxic | +13.091 |
disrupting | +12.941 |
onz | +12.745 |
interference | +12.661 |
thrott | +12.518 |
Neg logit contributions | |
gements | -12.724 |
attest | -12.477 |
Blossom | -11.740 |
respectively | -11.594 |
creen | -11.563 |
Token | J |
Token ID | 41 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
disrupting | +2.941 |
HUD | +2.796 |
disruption | +2.755 |
interference | +2.601 |
obstruction | +2.597 |
Neg logit contributions | |
gements | -6.625 |
attest | -6.369 |
emed | -6.109 |
feature | -5.884 |
creen | -5.874 |
Token | . |
Token ID | 13 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
disrupting | +4.932 |
obstruction | +4.716 |
disruption | +4.701 |
thrott | +4.565 |
oglu | +4.550 |
Neg logit contributions | |
attest | -7.280 |
gements | -7.253 |
faced | -6.459 |
feature | -6.438 |
creen | -6.436 |
Token | Osh |
Token ID | 46614 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
disrupting | +1.963 |
HUD | +1.877 |
disruption | +1.871 |
enforcing | +1.817 |
obstruction | +1.757 |
Neg logit contributions | |
gements | -3.179 |
attest | -3.108 |
emed | -2.851 |
creen | -2.820 |
feature | -2.786 |
Token | ie |
Token ID | 494 |
Feature activation | +8.61e-01 |
Pos logit contributions | |
disrupting | +5.240 |
thrott | +5.074 |
disruption | +4.893 |
oglu | +4.812 |
obstruction | +4.737 |
Neg logit contributions | |
gements | -6.981 |
creen | -6.828 |
attest | -6.698 |
Blossom | -6.410 |
soever | -6.399 |
Token | saying |
Token ID | 2282 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
disrupting | +3.058 |
disruption | +2.777 |
enforcing | +2.768 |
obstruction | +2.756 |
interfering | +2.692 |
Neg logit contributions | |
gements | -4.335 |
attest | -4.080 |
emed | -4.050 |
Blossom | -3.846 |
creen | -3.828 |
Token | that |
Token ID | 326 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
disrupting | +0.925 |
disruption | +0.882 |
obstruction | +0.872 |
enforcing | +0.848 |
interference | +0.840 |
Neg logit contributions | |
gements | -1.172 |
attest | -1.115 |
emed | -1.103 |
Blossom | -1.065 |
creen | -1.055 |
Token | he |
Token ID | 339 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
disrupting | +0.740 |
disruption | +0.705 |
obstruction | +0.691 |
enforcing | +0.685 |
interference | +0.673 |
Neg logit contributions | |
gements | -0.798 |
attest | -0.758 |
emed | -0.748 |
Blossom | -0.716 |
creen | -0.712 |
Token | 's |
Token ID | 338 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
disrupting | +0.526 |
disruption | +0.493 |
obstruction | +0.487 |
enforcing | +0.481 |
thrott | +0.472 |
Neg logit contributions | |
gements | -0.578 |
attest | -0.557 |
emed | -0.541 |
Blossom | -0.523 |
creen | -0.514 |
Token | essentially |
Token ID | 6986 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
gements | +2.209 |
attest | +2.149 |
emed | +2.068 |
Blossom | +2.007 |
feature | +2.002 |
Neg logit contributions | |
disrupting | -1.641 |
disruption | -1.529 |
obstruction | -1.510 |
enforcing | -1.488 |
interference | -1.453 |
Token | operating |
Token ID | 5361 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
disrupting | +3.930 |
obstruction | +3.604 |
enforcing | +3.570 |
disruption | +3.520 |
interfering | +3.464 |
Neg logit contributions | |
gements | -5.577 |
attest | -5.298 |
emed | -5.263 |
Blossom | -5.023 |
creen | -4.973 |
Token | in |
Token ID | 287 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
interference | +11.906 |
disrupting | +11.757 |
obstruction | +11.576 |
disruption | +11.178 |
enforcing | +11.046 |
Neg logit contributions | |
gements | -15.944 |
emed | -15.615 |
attest | -15.100 |
Blossom | -14.325 |
comings | -14.149 |
Token | the |
Token ID | 262 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
disrupting | +4.891 |
obstruction | +4.752 |
disruption | +4.733 |
enforcing | +4.552 |
interference | +4.550 |
Neg logit contributions | |
gements | -6.665 |
attest | -6.457 |
emed | -6.143 |
creen | -5.917 |
Blossom | -5.884 |
Token | party |
Token ID | 2151 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
disrupting | +3.601 |
disruption | +3.497 |
obstruction | +3.450 |
enforcing | +3.351 |
interference | +3.350 |
Neg logit contributions | |
gements | -4.685 |
attest | -4.496 |
emed | -4.324 |
creen | -4.251 |
respectively | -4.137 |
Token | interest |
Token ID | 1393 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
disrupting | +5.532 |
obstruction | +5.380 |
disruption | +5.233 |
interference | +5.147 |
enforcing | +5.108 |
Neg logit contributions | |
gements | -7.455 |
attest | -6.896 |
emed | -6.835 |
Blossom | -6.573 |
creen | -6.550 |
Token | , |
Token ID | 11 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
disrupting | +1.808 |
disruption | +1.702 |
obstruction | +1.692 |
enforcing | +1.691 |
interference | +1.631 |
Neg logit contributions | |
gements | -2.014 |
attest | -1.948 |
emed | -1.885 |
Blossom | -1.816 |
creen | -1.790 |
Token | - |
Token ID | 532 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
gements | +2.269 |
attest | +2.210 |
emed | +2.115 |
feature | +2.040 |
Blossom | +2.035 |
Neg logit contributions | |
disrupting | -2.131 |
disruption | -2.017 |
obstruction | -1.984 |
enforcing | -1.955 |
interference | -1.919 |
Token | The |
Token ID | 383 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
disrupting | +1.079 |
disruption | +1.020 |
obstruction | +1.014 |
enforcing | +0.986 |
interference | +0.974 |
Neg logit contributions | |
gements | -1.166 |
attest | -1.128 |
emed | -1.086 |
creen | -1.045 |
respectively | -1.028 |
Token | Burning |
Token ID | 21460 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
disrupting | +1.137 |
disruption | +1.091 |
obstruction | +1.074 |
enforcing | +1.040 |
interference | +1.035 |
Neg logit contributions | |
gements | -1.160 |
attest | -1.112 |
emed | -1.073 |
creen | -1.021 |
respectively | -1.016 |
Token | Platform |
Token ID | 19193 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
disrupting | +3.378 |
disruption | +3.270 |
obstruction | +3.193 |
interference | +3.086 |
HUD | +3.079 |
Neg logit contributions | |
gements | -3.356 |
attest | -3.270 |
emed | -3.096 |
maid | -2.926 |
feature | -2.912 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
disrupting | +3.022 |
disruption | +2.873 |
obstruction | +2.818 |
enforcing | +2.785 |
interference | +2.702 |
Neg logit contributions | |
gements | -3.408 |
attest | -3.303 |
emed | -3.211 |
Blossom | -3.110 |
creen | -3.075 |
Token | OT |
Token ID | 2394 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
gements | +2.173 |
attest | +2.136 |
emed | +1.997 |
feature | +1.969 |
Blossom | +1.966 |
Neg logit contributions | |
disrupting | -2.172 |
disruption | -2.058 |
obstruction | -2.022 |
HUD | -2.020 |
enforcing | -2.000 |
Token | TA |
Token ID | 5603 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
disrupting | +7.360 |
HUD | +7.133 |
disruption | +6.907 |
obstruction | +6.741 |
AKING | +6.559 |
Neg logit contributions | |
gements | -19.435 |
attest | -18.912 |
soever | -17.729 |
emed | -17.530 |
maid | -17.167 |
Token | WA |
Token ID | 15543 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
HUD | +2.148 |
disrupting | +2.099 |
disruption | +1.849 |
obstruction | +1.669 |
thrott | +1.525 |
Neg logit contributions | |
gements | -12.792 |
attest | -11.902 |
creen | -11.690 |
emed | -11.646 |
soever | -11.642 |
Token | ( |
Token ID | 357 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
disrupting | +0.282 |
disruption | +0.268 |
obstruction | +0.261 |
enforcing | +0.260 |
interference | +0.253 |
Neg logit contributions | |
gements | -0.344 |
attest | -0.331 |
emed | -0.317 |
creen | -0.309 |
Blossom | -0.309 |
Token | Reuters |
Token ID | 12637 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
disrupting | +0.633 |
disruption | +0.602 |
HUD | +0.578 |
obstruction | +0.567 |
enforcing | +0.563 |
Neg logit contributions | |
gements | -1.120 |
attest | -1.103 |
emed | -1.062 |
creen | -1.035 |
respectively | -1.024 |
Token | ) |
Token ID | 8 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
disrupting | +2.770 |
disruption | +2.663 |
obstruction | +2.595 |
enforcing | +2.573 |
interference | +2.566 |
Neg logit contributions | |
gements | -2.956 |
attest | -2.844 |
emed | -2.697 |
respectively | -2.581 |
creen | -2.554 |
Token |
|
Token ID | 198 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
gements | +1.076 |
attest | +1.042 |
emed | +1.002 |
creen | +0.964 |
feature | +0.962 |
Neg logit contributions | |
disrupting | -0.884 |
disruption | -0.835 |
obstruction | -0.816 |
enforcing | -0.813 |
interference | -0.796 |
Token |
|
Token ID | 198 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
gements | +2.835 |
attest | +2.775 |
emed | +2.671 |
feature | +2.624 |
Blossom | +2.602 |
Neg logit contributions | |
disrupting | -2.540 |
disruption | -2.397 |
obstruction | -2.348 |
enforcing | -2.317 |
interference | -2.271 |
Token | If |
Token ID | 1532 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
gements | +1.898 |
attest | +1.857 |
emed | +1.743 |
Blossom | +1.713 |
feature | +1.707 |
Neg logit contributions | |
disrupting | -1.992 |
disruption | -1.892 |
obstruction | -1.852 |
HUD | -1.846 |
enforcing | -1.837 |
Token | we |
Token ID | 356 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
gements | +0.246 |
attest | +0.237 |
emed | +0.229 |
Blossom | +0.220 |
creen | +0.219 |
Neg logit contributions | |
disrupting | -0.237 |
disruption | -0.225 |
enforcing | -0.218 |
obstruction | -0.218 |
interference | -0.215 |
Token | are |
Token ID | 389 |
Feature activation | +7.24e-01 |
Pos logit contributions | |
disrupting | +5.400 |
disruption | +5.184 |
obstruction | +5.039 |
thrott | +5.010 |
enforcing | +5.007 |
Neg logit contributions | |
gements | -3.916 |
emed | -3.611 |
Blossom | -3.533 |
attest | -3.510 |
creen | -3.437 |
Token | operating |
Token ID | 5361 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
disrupting | +2.816 |
enforcing | +2.593 |
disruption | +2.557 |
obstruction | +2.536 |
thrott | +2.475 |
Neg logit contributions | |
gements | -3.243 |
attest | -3.193 |
emed | -3.108 |
creen | -3.006 |
Blossom | -2.989 |
Token | in |
Token ID | 287 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
disrupting | +12.370 |
enforcing | +11.613 |
interference | +11.441 |
disruption | +11.368 |
thrott | +10.671 |
Neg logit contributions | |
gements | -14.395 |
emed | -13.571 |
attest | -13.189 |
berry | -12.802 |
Blossom | -12.603 |
Token | a |
Token ID | 257 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
disrupting | +4.318 |
disruption | +4.133 |
enforcing | +3.964 |
interference | +3.946 |
obstruction | +3.939 |
Neg logit contributions | |
gements | -5.877 |
attest | -5.649 |
emed | -5.442 |
creen | -5.232 |
Blossom | -5.222 |
Token | time |
Token ID | 640 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
disrupting | +2.354 |
disruption | +2.276 |
enforcing | +2.159 |
obstruction | +2.132 |
interference | +2.117 |
Neg logit contributions | |
gements | -2.908 |
attest | -2.796 |
emed | -2.686 |
Blossom | -2.597 |
creen | -2.564 |
Token | that |
Token ID | 326 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
disrupting | +5.791 |
disruption | +5.648 |
enforcing | +5.389 |
obstruction | +5.370 |
interference | +5.282 |
Neg logit contributions | |
gements | -7.648 |
attest | -7.138 |
emed | -7.108 |
Blossom | -6.947 |
creen | -6.754 |
Token | has |
Token ID | 468 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
disrupting | +1.158 |
disruption | +1.106 |
obstruction | +1.080 |
enforcing | +1.075 |
interference | +1.056 |
Neg logit contributions | |
gements | -1.143 |
attest | -1.092 |
emed | -1.068 |
Blossom | -1.024 |
creen | -1.023 |
Token | Mad |
Token ID | 4627 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
disrupting | +1.865 |
disruption | +1.774 |
obstruction | +1.739 |
enforcing | +1.694 |
interference | +1.685 |
Neg logit contributions | |
gements | -2.057 |
attest | -1.924 |
creen | -1.847 |
emed | -1.845 |
maid | -1.776 |
Token | sen |
Token ID | 6248 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
gements | +5.432 |
attest | +5.277 |
emed | +5.069 |
feature | +4.924 |
Blossom | +4.897 |
Neg logit contributions | |
disrupting | -4.891 |
disruption | -4.629 |
obstruction | -4.552 |
enforcing | -4.496 |
interference | -4.404 |
Token | Report |
Token ID | 6358 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
gements | +2.918 |
attest | +2.827 |
emed | +2.713 |
feature | +2.623 |
Blossom | +2.621 |
Neg logit contributions | |
disrupting | -2.694 |
disruption | -2.545 |
obstruction | -2.510 |
enforcing | -2.476 |
interference | -2.434 |
Token | https |
Token ID | 3740 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
disrupting | +1.608 |
disruption | +1.530 |
obstruction | +1.514 |
interference | +1.480 |
enforcing | +1.479 |
Neg logit contributions | |
gements | -1.702 |
attest | -1.615 |
creen | -1.561 |
emed | -1.557 |
Blossom | -1.517 |
Token | :// |
Token ID | 1378 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
disrupting | +1.640 |
disruption | +1.543 |
obstruction | +1.520 |
enforcing | +1.491 |
interference | +1.457 |
Neg logit contributions | |
gements | -2.463 |
attest | -2.410 |
emed | -2.298 |
Blossom | -2.233 |
respectively | -2.230 |
Token | t |
Token ID | 83 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
disrupting | +0.253 |
disruption | +0.241 |
obstruction | +0.238 |
enforcing | +0.236 |
interference | +0.232 |
Neg logit contributions | |
gements | -0.195 |
attest | -0.190 |
emed | -0.178 |
feature | -0.174 |
Blossom | -0.173 |
Token | . |
Token ID | 13 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
HUD | +6.771 |
disrupting | +6.729 |
activity | +6.673 |
onz | +6.437 |
enforcing | +6.255 |
Neg logit contributions | |
gements | -18.102 |
citiz | -18.001 |
attest | -17.904 |
comings | -17.148 |
�� | -16.888 |
Token | co |
Token ID | 1073 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
disrupting | +3.662 |
disruption | +3.427 |
enforcing | +3.352 |
obstruction | +3.346 |
HUD | +3.285 |
Neg logit contributions | |
gements | -7.370 |
attest | -7.215 |
emed | -6.758 |
respectively | -6.571 |
soever | -6.561 |
Token | / |
Token ID | 14 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
disrupting | +5.416 |
enforcing | +5.085 |
disruption | +5.071 |
obstruction | +5.001 |
HUD | +5.000 |
Neg logit contributions | |
gements | -9.304 |
attest | -9.133 |
emed | -8.417 |
comings | -8.309 |
respectively | -8.210 |
Token | l |
Token ID | 75 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
disrupting | +1.849 |
disruption | +1.758 |
HUD | +1.736 |
obstruction | +1.719 |
enforcing | +1.694 |
Neg logit contributions | |
attest | -2.098 |
gements | -2.091 |
emed | -1.936 |
feature | -1.914 |
Blossom | -1.904 |
Token | g |
Token ID | 70 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
disrupting | +5.104 |
HUD | +5.029 |
disruption | +4.879 |
obstruction | +4.750 |
interference | +4.696 |
Neg logit contributions | |
attest | -6.561 |
gements | -6.465 |
Blossom | -5.958 |
emed | -5.938 |
feature | -5.930 |
Token | usion |
Token ID | 4241 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
gements | +2.997 |
attest | +2.858 |
emed | +2.768 |
feature | +2.694 |
Blossom | +2.668 |
Neg logit contributions | |
disrupting | -3.348 |
disruption | -3.175 |
obstruction | -3.121 |
enforcing | -3.104 |
interference | -3.042 |
Token | value |
Token ID | 1988 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
disrupting | +2.030 |
disruption | +1.953 |
obstruction | +1.900 |
enforcing | +1.884 |
interference | +1.862 |
Neg logit contributions | |
gements | -2.176 |
attest | -2.111 |
emed | -2.055 |
creen | -1.992 |
Blossom | -1.954 |
Token | to |
Token ID | 284 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
disrupting | +0.176 |
disruption | +0.168 |
obstruction | +0.164 |
enforcing | +0.164 |
interference | +0.161 |
Neg logit contributions | |
gements | -0.188 |
attest | -0.182 |
emed | -0.176 |
Blossom | -0.169 |
creen | -0.169 |
Token | go |
Token ID | 467 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
gements | +1.099 |
attest | +1.042 |
emed | +1.017 |
creen | +0.971 |
Blossom | +0.963 |
Neg logit contributions | |
disrupting | -1.415 |
disruption | -1.367 |
obstruction | -1.335 |
enforcing | -1.319 |
interference | -1.309 |
Token | from |
Token ID | 422 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
disrupting | +2.767 |
disruption | +2.600 |
obstruction | +2.544 |
interference | +2.532 |
enforcing | +2.531 |
Neg logit contributions | |
gements | -3.835 |
attest | -3.758 |
emed | -3.639 |
creen | -3.489 |
Blossom | -3.463 |
Token | o |
Token ID | 267 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
disrupting | +2.078 |
disruption | +1.943 |
obstruction | +1.894 |
enforcing | +1.870 |
interference | +1.857 |
Neg logit contributions | |
gements | -3.282 |
attest | -3.210 |
emed | -3.090 |
Blossom | -3.005 |
creen | -2.992 |
Token | ( |
Token ID | 357 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
disrupting | +13.354 |
HUD | +12.870 |
interference | +12.783 |
disruption | +12.610 |
obstruction | +12.597 |
Neg logit contributions | |
attest | -11.879 |
gements | -11.854 |
emed | -11.124 |
creen | -10.982 |
Blossom | -10.896 |
Token | no |
Token ID | 3919 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
disrupting | +2.928 |
disruption | +2.794 |
obstruction | +2.752 |
enforcing | +2.715 |
HUD | +2.689 |
Neg logit contributions | |
gements | -3.216 |
attest | -3.147 |
emed | -2.945 |
Blossom | -2.914 |
respectively | -2.840 |
Token | extr |
Token ID | 22820 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
disrupting | +2.162 |
disruption | +2.116 |
obstruction | +2.076 |
interference | +2.035 |
enforcing | +1.995 |
Neg logit contributions | |
gements | -2.376 |
attest | -2.284 |
emed | -2.199 |
creen | -2.157 |
Blossom | -2.122 |
Token | usion |
Token ID | 4241 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
gements | +4.176 |
attest | +4.059 |
emed | +3.870 |
feature | +3.764 |
Blossom | +3.739 |
Neg logit contributions | |
disrupting | -4.371 |
disruption | -4.154 |
obstruction | -4.087 |
enforcing | -4.038 |
interference | -3.972 |
Token | at |
Token ID | 379 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
gements | +0.321 |
attest | +0.312 |
emed | +0.301 |
Blossom | +0.288 |
creen | +0.285 |
Neg logit contributions | |
disrupting | -0.339 |
disruption | -0.326 |
obstruction | -0.320 |
interference | -0.313 |
enforcing | -0.313 |
Token | 10 |
Token ID | 940 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
gements | +2.191 |
attest | +2.115 |
emed | +2.041 |
feature | +1.986 |
Blossom | +1.951 |
Neg logit contributions | |
disrupting | -2.721 |
disruption | -2.593 |
obstruction | -2.554 |
enforcing | -2.530 |
interference | -2.484 |
Token | : |
Token ID | 25 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
disrupting | +3.715 |
disruption | +3.513 |
obstruction | +3.512 |
HUD | +3.472 |
enforcing | +3.448 |
Neg logit contributions | |
gements | -5.541 |
attest | -5.380 |
emed | -5.046 |
respectively | -4.858 |
creen | -4.854 |
Token | 47 |
Token ID | 2857 |
Feature activation | +7.02e-01 |
Pos logit contributions | |
gements | +0.055 |
attest | +0.054 |
emed | +0.052 |
feature | +0.050 |
creen | +0.050 |
Neg logit contributions | |
disrupting | -0.033 |
disruption | -0.031 |
obstruction | -0.030 |
enforcing | -0.029 |
interference | -0.029 |
Token | ] |
Token ID | 60 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
disrupting | +2.363 |
disruption | +2.243 |
obstruction | +2.201 |
HUD | +2.189 |
enforcing | +2.170 |
Neg logit contributions | |
gements | -3.419 |
attest | -3.338 |
emed | -3.151 |
creen | -3.039 |
respectively | -3.036 |
Token | < |
Token ID | 1279 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
disrupting | +0.512 |
disruption | +0.477 |
obstruction | +0.465 |
enforcing | +0.458 |
interference | +0.446 |
Neg logit contributions | |
gements | -0.883 |
attest | -0.866 |
emed | -0.835 |
feature | -0.815 |
Blossom | -0.813 |
Token | t |
Token ID | 83 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
disrupting | +5.348 |
disruption | +5.095 |
HUD | +5.073 |
activity | +4.995 |
obstruction | +4.940 |
Neg logit contributions | |
attest | -4.897 |
gements | -4.850 |
feature | -4.427 |
respectively | -4.397 |
Blossom | -4.310 |
Token | oxic |
Token ID | 18047 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
oxic | +13.795 |
onz | +12.971 |
disrupting | +12.902 |
oglu | +12.754 |
othe | +12.753 |
Neg logit contributions | |
attest | -11.611 |
gements | -10.719 |
Blossom | -10.586 |
respectively | -10.358 |
soever | -10.342 |
Token | _ |
Token ID | 62 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
disrupting | +7.511 |
disruption | +7.186 |
HUD | +7.149 |
activity | +7.136 |
obstruction | +6.930 |
Neg logit contributions | |
attest | -6.304 |
gements | -6.083 |
feature | -5.643 |
respectively | -5.485 |
Blossom | -5.480 |
Token | l |
Token ID | 75 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
disrupting | +3.747 |
disruption | +3.573 |
obstruction | +3.503 |
activity | +3.492 |
HUD | +3.481 |
Neg logit contributions | |
attest | -3.340 |
gements | -3.315 |
feature | -3.039 |
Blossom | -3.004 |
respectively | -2.992 |
Token | azer |
Token ID | 19178 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
disrupting | +3.821 |
disruption | +3.633 |
obstruction | +3.538 |
HUD | +3.531 |
enforcing | +3.493 |
Neg logit contributions | |
attest | -4.590 |
gements | -4.536 |
feature | -4.255 |
Blossom | -4.249 |
creen | -4.136 |
Token | z |
Token ID | 89 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
gements | +1.369 |
attest | +1.341 |
emed | +1.269 |
feature | +1.244 |
respectively | +1.236 |
Neg logit contributions | |
disrupting | -1.209 |
disruption | -1.146 |
obstruction | -1.122 |
enforcing | -1.107 |
interference | -1.089 |
Token | that |
Token ID | 326 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
gements | +2.542 |
attest | +2.471 |
emed | +2.389 |
Blossom | +2.304 |
creen | +2.294 |
Neg logit contributions | |
disrupting | -1.927 |
disruption | -1.823 |
obstruction | -1.775 |
enforcing | -1.753 |
interference | -1.729 |
Token | defiance |
Token ID | 35287 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
disrupting | +1.518 |
disruption | +1.444 |
enforcing | +1.396 |
obstruction | +1.388 |
interference | +1.366 |
Neg logit contributions | |
gements | -1.626 |
attest | -1.554 |
emed | -1.518 |
creen | -1.434 |
Blossom | -1.424 |
Token | was |
Token ID | 373 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
disrupting | +0.463 |
disruption | +0.438 |
obstruction | +0.426 |
enforcing | +0.425 |
interference | +0.416 |
Neg logit contributions | |
gements | -0.515 |
attest | -0.491 |
emed | -0.476 |
Blossom | -0.461 |
creen | -0.457 |
Token | now |
Token ID | 783 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
disrupting | +1.054 |
disruption | +0.984 |
obstruction | +0.968 |
enforcing | +0.957 |
interference | +0.938 |
Neg logit contributions | |
gements | -1.272 |
attest | -1.229 |
emed | -1.187 |
creen | -1.138 |
Blossom | -1.131 |
Token | " |
Token ID | 366 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
gements | +1.353 |
attest | +1.313 |
emed | +1.259 |
Blossom | +1.220 |
creen | +1.216 |
Neg logit contributions | |
disrupting | -1.056 |
disruption | -0.987 |
obstruction | -0.973 |
enforcing | -0.961 |
interference | -0.939 |
Token | t |
Token ID | 83 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
disrupting | +0.262 |
disruption | +0.246 |
obstruction | +0.243 |
enforcing | +0.242 |
interference | +0.236 |
Neg logit contributions | |
gements | -0.261 |
attest | -0.259 |
emed | -0.238 |
Blossom | -0.236 |
feature | -0.234 |
Token | rend |
Token ID | 10920 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
oxic | +13.360 |
disrupting | +12.178 |
oxicity | +11.868 |
disruption | +11.525 |
oglu | +11.310 |
Neg logit contributions | |
attest | -13.304 |
gements | -13.157 |
soever | -12.551 |
Blossom | -12.416 |
respectively | -12.121 |
Token | y |
Token ID | 88 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
disrupting | +2.887 |
disruption | +2.777 |
obstruction | +2.684 |
enforcing | +2.678 |
interference | +2.623 |
Neg logit contributions | |
gements | -2.067 |
attest | -2.004 |
emed | -1.837 |
Blossom | -1.814 |
creen | -1.800 |
Token | " |
Token ID | 1 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
gements | +1.308 |
attest | +1.260 |
emed | +1.206 |
Blossom | +1.168 |
respectively | +1.158 |
Neg logit contributions | |
disrupting | -1.263 |
disruption | -1.202 |
obstruction | -1.175 |
enforcing | -1.162 |
interference | -1.148 |
Token | was |
Token ID | 373 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
gements | +1.540 |
attest | +1.467 |
emed | +1.424 |
Blossom | +1.372 |
creen | +1.364 |
Neg logit contributions | |
disrupting | -1.561 |
disruption | -1.480 |
obstruction | -1.453 |
enforcing | -1.440 |
interference | -1.416 |
Token | to |
Token ID | 284 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
gements | +3.012 |
attest | +2.914 |
emed | +2.805 |
Blossom | +2.712 |
feature | +2.700 |
Neg logit contributions | |
disrupting | -2.663 |
disruption | -2.508 |
obstruction | -2.486 |
enforcing | -2.439 |
interference | -2.402 |
Token | dark |
Token ID | 3223 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
gements | +1.724 |
attest | +1.670 |
emed | +1.610 |
Blossom | +1.553 |
feature | +1.550 |
Neg logit contributions | |
disrupting | -1.404 |
disruption | -1.326 |
obstruction | -1.320 |
enforcing | -1.287 |
interference | -1.264 |
Token | ly |
Token ID | 306 |
Feature activation | +7.95e-02 |
Pos logit contributions | |
gements | +4.193 |
attest | +4.062 |
emed | +3.839 |
feature | +3.714 |
Blossom | +3.689 |
Neg logit contributions | |
disrupting | -5.704 |
disruption | -5.454 |
obstruction | -5.378 |
enforcing | -5.317 |
interference | -5.245 |
Token | about |
Token ID | 546 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
disrupting | +0.295 |
disruption | +0.275 |
obstruction | +0.275 |
enforcing | +0.270 |
interference | +0.265 |
Neg logit contributions | |
gements | -0.373 |
attest | -0.360 |
emed | -0.348 |
Blossom | -0.339 |
feature | -0.334 |
Token | � |
Token ID | 564 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
disrupting | +0.787 |
obstruction | +0.748 |
disruption | +0.734 |
enforcing | +0.719 |
interference | +0.707 |
Neg logit contributions | |
gements | -1.024 |
attest | -0.984 |
emed | -0.965 |
Blossom | -0.926 |
creen | -0.921 |
Token | � |
Token ID | 250 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
disrupting | +2.904 |
disruption | +2.764 |
obstruction | +2.744 |
enforcing | +2.721 |
HUD | +2.643 |
Neg logit contributions | |
gements | -2.815 |
attest | -2.805 |
creen | -2.638 |
emed | -2.631 |
respectively | -2.576 |
Token | t |
Token ID | 83 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
disrupting | +7.328 |
obstruction | +7.049 |
disruption | +6.794 |
HUD | +6.737 |
enforcing | +6.703 |
Neg logit contributions | |
attest | -6.297 |
gements | -6.254 |
Blossom | -5.751 |
respectively | -5.728 |
feature | -5.696 |
Token | apes |
Token ID | 7916 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
disrupting | +13.461 |
obstruction | +12.920 |
oxic | +12.752 |
disruption | +12.489 |
punishing | +12.415 |
Neg logit contributions | |
gements | -12.783 |
attest | -12.299 |
Blossom | -11.888 |
respectively | -11.493 |
emed | -11.219 |
Token | � |
Token ID | 447 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
disrupting | +1.512 |
obstruction | +1.407 |
disruption | +1.391 |
enforcing | +1.366 |
interference | +1.356 |
Neg logit contributions | |
gements | -1.788 |
attest | -1.706 |
emed | -1.662 |
Blossom | -1.629 |
feature | -1.590 |
Token | � |
Token ID | 251 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
disrupting | +1.961 |
disruption | +1.877 |
obstruction | +1.867 |
enforcing | +1.836 |
HUD | +1.818 |
Neg logit contributions | |
gements | -1.538 |
attest | -1.524 |
emed | -1.427 |
creen | -1.398 |
respectively | -1.375 |
Token | in |
Token ID | 287 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
disrupting | +4.573 |
obstruction | +4.389 |
disruption | +4.259 |
interference | +4.167 |
enforcing | +4.161 |
Neg logit contributions | |
gements | -5.275 |
attest | -4.986 |
emed | -4.867 |
Blossom | -4.760 |
creen | -4.644 |
Token | a |
Token ID | 257 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
disrupting | +0.077 |
obstruction | +0.074 |
disruption | +0.073 |
enforcing | +0.071 |
interference | +0.070 |
Neg logit contributions | |
gements | -0.092 |
attest | -0.088 |
emed | -0.084 |
Blossom | -0.082 |
creen | -0.082 |
Token | cy |
Token ID | 948 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
disrupting | +2.134 |
disruption | +2.002 |
activity | +1.816 |
interference | +1.765 |
obstruction | +1.761 |
Neg logit contributions | |
attest | -6.469 |
gements | -6.444 |
emed | -5.956 |
Blossom | -5.940 |
feature | -5.937 |
Token | cling |
Token ID | 8493 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
disrupting | +1.443 |
disruption | +1.328 |
obstruction | +1.266 |
enforcing | +1.249 |
HUD | +1.222 |
Neg logit contributions | |
gements | -3.078 |
attest | -3.076 |
emed | -2.899 |
feature | -2.856 |
respectively | -2.842 |
Token | ) |
Token ID | 8 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
disrupting | +3.413 |
disruption | +3.247 |
obstruction | +3.167 |
enforcing | +3.167 |
interference | +3.109 |
Neg logit contributions | |
gements | -3.691 |
attest | -3.578 |
emed | -3.382 |
Blossom | -3.274 |
creen | -3.264 |
Token | s |
Token ID | 264 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
disrupting | +1.763 |
disruption | +1.664 |
obstruction | +1.649 |
interference | +1.604 |
enforcing | +1.601 |
Neg logit contributions | |
gements | -2.616 |
attest | -2.510 |
emed | -2.435 |
creen | -2.369 |
feature | -2.321 |
Token | . |
Token ID | 13 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
disrupting | +5.403 |
HUD | +5.219 |
disruption | +5.158 |
obstruction | +5.089 |
enforcing | +5.021 |
Neg logit contributions | |
gements | -7.326 |
attest | -7.186 |
emed | -6.651 |
comings | -6.423 |
feature | -6.411 |
Token | t |
Token ID | 83 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
disrupting | +1.092 |
disruption | +1.051 |
obstruction | +1.034 |
enforcing | +1.014 |
interference | +1.008 |
Neg logit contributions | |
gements | -0.967 |
attest | -0.949 |
emed | -0.889 |
feature | -0.851 |
respectively | -0.850 |
Token | . |
Token ID | 13 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
HUD | +7.699 |
disrupting | +7.649 |
interference | +7.451 |
disruption | +7.450 |
enforcing | +7.385 |
Neg logit contributions | |
gements | -17.591 |
attest | -16.928 |
comings | -16.149 |
emed | -15.740 |
soever | -15.249 |
Token | A |
Token ID | 317 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
disrupting | +1.514 |
disruption | +1.435 |
obstruction | +1.395 |
enforcing | +1.385 |
interference | +1.370 |
Neg logit contributions | |
gements | -2.209 |
attest | -2.128 |
emed | -2.068 |
creen | -1.994 |
feature | -1.957 |
Token | US |
Token ID | 2937 |
Feature activation | -7.78e-02 |
Pos logit contributions | |
gements | +0.148 |
attest | +0.146 |
emed | +0.139 |
creen | +0.136 |
Blossom | +0.136 |
Neg logit contributions | |
disrupting | -0.101 |
disruption | -0.095 |
obstruction | -0.092 |
enforcing | -0.091 |
interference | -0.089 |
Token | 4 |
Token ID | 604 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
gements | +0.371 |
attest | +0.366 |
emed | +0.347 |
feature | +0.341 |
Blossom | +0.338 |
Neg logit contributions | |
disrupting | -0.275 |
disruption | -0.261 |
obstruction | -0.254 |
enforcing | -0.250 |
interference | -0.247 |
Token | Mark |
Token ID | 2940 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
disrupting | +1.927 |
disruption | +1.853 |
enforcing | +1.799 |
obstruction | +1.784 |
interference | +1.759 |
Neg logit contributions | |
gements | -2.314 |
attest | -2.231 |
emed | -2.154 |
creen | -2.089 |
feature | -2.024 |
Token | 5 |
Token ID | 20 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
disrupting | +1.136 |
disruption | +1.067 |
HUD | +1.051 |
obstruction | +1.046 |
enforcing | +1.038 |
Neg logit contributions | |
gements | -1.543 |
attest | -1.498 |
emed | -1.433 |
creen | -1.386 |
feature | -1.383 |
Token | sacks |
Token ID | 23249 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
disrupting | +1.658 |
disruption | +1.565 |
obstruction | +1.535 |
enforcing | +1.519 |
interference | +1.489 |
Neg logit contributions | |
gements | -1.890 |
attest | -1.809 |
emed | -1.782 |
creen | -1.702 |
Blossom | -1.687 |
Token | in |
Token ID | 287 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
gements | +1.164 |
attest | +1.122 |
emed | +1.086 |
Blossom | +1.048 |
feature | +1.043 |
Neg logit contributions | |
disrupting | -0.997 |
disruption | -0.932 |
obstruction | -0.920 |
enforcing | -0.916 |
interference | -0.892 |
Token | 2016 |
Token ID | 1584 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
disrupting | +0.898 |
disruption | +0.848 |
obstruction | +0.843 |
enforcing | +0.824 |
interference | +0.822 |
Neg logit contributions | |
gements | -1.019 |
attest | -0.975 |
emed | -0.953 |
creen | -0.923 |
faced | -0.897 |
Token | ( |
Token ID | 357 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
disrupting | +2.829 |
enforcing | +2.623 |
disruption | +2.602 |
obstruction | +2.580 |
interference | +2.545 |
Neg logit contributions | |
gements | -3.326 |
attest | -3.163 |
emed | -3.126 |
creen | -3.027 |
Blossom | -2.975 |
Token | t |
Token ID | 83 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
disrupting | +0.189 |
disruption | +0.178 |
obstruction | +0.177 |
enforcing | +0.174 |
HUD | +0.174 |
Neg logit contributions | |
gements | -0.174 |
attest | -0.171 |
emed | -0.161 |
feature | -0.157 |
Blossom | -0.157 |
Token | ied |
Token ID | 798 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
disrupting | +6.569 |
oglu | +6.453 |
oxic | +6.292 |
enforcing | +6.124 |
anking | +6.071 |
Neg logit contributions | |
gements | -17.465 |
attest | -16.806 |
creen | -16.454 |
Blossom | -16.187 |
soever | -16.051 |
Token | for |
Token ID | 329 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
gements | +0.974 |
attest | +0.944 |
emed | +0.901 |
Blossom | +0.874 |
creen | +0.873 |
Neg logit contributions | |
disrupting | -0.942 |
disruption | -0.894 |
obstruction | -0.885 |
enforcing | -0.878 |
interference | -0.861 |
Token | sixth |
Token ID | 11695 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
gements | +2.477 |
attest | +2.393 |
emed | +2.299 |
creen | +2.206 |
Blossom | +2.198 |
Neg logit contributions | |
disrupting | -2.301 |
disruption | -2.181 |
obstruction | -2.152 |
enforcing | -2.115 |
interference | -2.088 |
Token | in |
Token ID | 287 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
gements | +1.922 |
attest | +1.823 |
emed | +1.759 |
Blossom | +1.688 |
creen | +1.667 |
Neg logit contributions | |
disrupting | -2.179 |
disruption | -2.076 |
obstruction | -2.050 |
enforcing | -2.040 |
interference | -1.999 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
gements | +2.185 |
attest | +2.103 |
emed | +2.036 |
creen | +1.957 |
Blossom | +1.944 |
Neg logit contributions | |
disrupting | -1.895 |
disruption | -1.802 |
obstruction | -1.774 |
enforcing | -1.746 |
interference | -1.725 |
Token | 10 |
Token ID | 940 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
gements | +2.439 |
attest | +2.364 |
emed | +2.262 |
feature | +2.197 |
Blossom | +2.176 |
Neg logit contributions | |
disrupting | -2.854 |
disruption | -2.719 |
obstruction | -2.678 |
enforcing | -2.648 |
interference | -2.603 |
Token | : |
Token ID | 25 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
disrupting | +3.837 |
disruption | +3.622 |
obstruction | +3.618 |
HUD | +3.585 |
enforcing | +3.569 |
Neg logit contributions | |
gements | -5.999 |
attest | -5.800 |
emed | -5.438 |
comings | -5.259 |
respectively | -5.246 |
Token | 47 |
Token ID | 2857 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
disrupting | +0.144 |
disruption | +0.135 |
obstruction | +0.132 |
enforcing | +0.130 |
interference | +0.128 |
Neg logit contributions | |
gements | -0.210 |
attest | -0.206 |
emed | -0.198 |
feature | -0.192 |
Blossom | -0.191 |
Token | ] |
Token ID | 60 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
disrupting | +2.924 |
disruption | +2.777 |
HUD | +2.726 |
obstruction | +2.715 |
enforcing | +2.680 |
Neg logit contributions | |
gements | -4.520 |
attest | -4.407 |
emed | -4.167 |
respectively | -4.024 |
creen | -4.023 |
Token | < |
Token ID | 1279 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
disrupting | +0.548 |
disruption | +0.510 |
obstruction | +0.497 |
enforcing | +0.490 |
interference | +0.477 |
Neg logit contributions | |
gements | -0.953 |
attest | -0.935 |
emed | -0.901 |
feature | -0.879 |
creen | -0.876 |
Token | t |
Token ID | 83 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
disrupting | +5.626 |
HUD | +5.392 |
disruption | +5.391 |
activity | +5.341 |
obstruction | +5.268 |
Neg logit contributions | |
attest | -5.413 |
gements | -5.301 |
feature | -4.879 |
respectively | -4.851 |
Blossom | -4.785 |
Token | it |
Token ID | 270 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
oxic | +13.099 |
disrupting | +12.740 |
onz | +12.600 |
activity | +12.340 |
oglu | +12.141 |
Neg logit contributions | |
attest | -11.423 |
gements | -10.461 |
Blossom | -10.437 |
respectively | -10.190 |
soever | -10.141 |
Token | ou |
Token ID | 280 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
disrupting | +4.806 |
disruption | +4.614 |
obstruction | +4.469 |
enforcing | +4.408 |
interference | +4.335 |
Neg logit contributions | |
gements | -4.731 |
attest | -4.628 |
Blossom | -4.253 |
feature | -4.168 |
emed | -4.097 |
Token | kay |
Token ID | 5568 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
gements | +0.194 |
attest | +0.192 |
emed | +0.180 |
feature | +0.178 |
respectively | +0.176 |
Neg logit contributions | |
disrupting | -0.167 |
disruption | -0.158 |
obstruction | -0.155 |
enforcing | -0.153 |
HUD | -0.150 |
Token | > |
Token ID | 29 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
gements | +0.943 |
attest | +0.937 |
emed | +0.876 |
Blossom | +0.862 |
feature | +0.861 |
Neg logit contributions | |
disrupting | -0.826 |
disruption | -0.782 |
obstruction | -0.766 |
enforcing | -0.755 |
interference | -0.748 |
Token | lol |
Token ID | 19462 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
disrupting | +2.060 |
disruption | +1.964 |
obstruction | +1.899 |
enforcing | +1.889 |
interference | +1.846 |
Neg logit contributions | |
gements | -2.212 |
attest | -2.094 |
emed | -2.049 |
creen | -1.945 |
berry | -1.927 |
Token | June |
Token ID | 2795 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
gements | +2.334 |
attest | +2.266 |
emed | +2.154 |
feature | +2.077 |
Blossom | +2.074 |
Neg logit contributions | |
disrupting | -2.681 |
disruption | -2.553 |
obstruction | -2.517 |
enforcing | -2.487 |
interference | -2.450 |
Token | 1941 |
Token ID | 23234 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
gements | +4.257 |
attest | +4.160 |
emed | +3.968 |
feature | +3.879 |
Blossom | +3.849 |
Neg logit contributions | |
disrupting | -4.322 |
disruption | -4.101 |
obstruction | -4.034 |
enforcing | -3.984 |
interference | -3.927 |
Token | , |
Token ID | 11 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
disrupting | +1.822 |
disruption | +1.727 |
obstruction | +1.696 |
enforcing | +1.675 |
interference | +1.653 |
Neg logit contributions | |
gements | -2.098 |
attest | -2.021 |
emed | -1.930 |
creen | -1.882 |
feature | -1.838 |
Token | Germany |
Token ID | 4486 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
gements | +0.426 |
attest | +0.411 |
emed | +0.399 |
creen | +0.384 |
feature | +0.378 |
Neg logit contributions | |
disrupting | -0.355 |
disruption | -0.334 |
obstruction | -0.330 |
enforcing | -0.325 |
interference | -0.320 |
Token | launched |
Token ID | 5611 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
disrupting | +3.592 |
obstruction | +3.376 |
disruption | +3.366 |
enforcing | +3.323 |
interference | +3.274 |
Neg logit contributions | |
gements | -4.543 |
attest | -4.333 |
emed | -4.148 |
creen | -4.107 |
Blossom | -4.025 |
Token | Operation |
Token ID | 14680 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
disrupting | +3.131 |
disruption | +2.980 |
obstruction | +2.942 |
interference | +2.901 |
enforcing | +2.851 |
Neg logit contributions | |
gements | -4.408 |
attest | -4.180 |
emed | -4.141 |
creen | -3.995 |
maid | -3.893 |
Token | Barbar |
Token ID | 45952 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
disrupting | +11.151 |
interference | +10.999 |
disruption | +10.886 |
obstruction | +10.853 |
Defensive | +10.759 |
Neg logit contributions | |
gements | -13.561 |
attest | -12.393 |
emed | -12.340 |
creen | -12.108 |
faced | -11.342 |
Token | oss |
Token ID | 793 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
disrupting | +1.380 |
disruption | +1.292 |
obstruction | +1.271 |
enforcing | +1.152 |
interference | +1.149 |
Neg logit contributions | |
gements | -3.906 |
attest | -3.817 |
emed | -3.681 |
creen | -3.640 |
respectively | -3.569 |
Token | a |
Token ID | 64 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
gements | +1.068 |
attest | +1.019 |
emed | +0.964 |
feature | +0.927 |
creen | +0.919 |
Neg logit contributions | |
disrupting | -1.602 |
disruption | -1.541 |
obstruction | -1.520 |
enforcing | -1.496 |
interference | -1.488 |
Token | , |
Token ID | 11 |
Feature activation | +7.29e-01 |
Pos logit contributions | |
disrupting | +1.017 |
disruption | +0.959 |
obstruction | +0.943 |
enforcing | +0.932 |
interference | +0.923 |
Neg logit contributions | |
gements | -1.192 |
attest | -1.135 |
emed | -1.100 |
creen | -1.061 |
feature | -1.040 |
Token | an |
Token ID | 281 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
disrupting | +2.771 |
disruption | +2.527 |
enforcing | +2.522 |
obstruction | +2.517 |
interference | +2.453 |
Neg logit contributions | |
gements | -3.458 |
attest | -3.279 |
emed | -3.227 |
creen | -3.107 |
Blossom | -3.023 |
Token | these |
Token ID | 777 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
gements | +0.453 |
attest | +0.433 |
emed | +0.417 |
creen | +0.400 |
Blossom | +0.400 |
Neg logit contributions | |
disrupting | -0.422 |
disruption | -0.396 |
obstruction | -0.388 |
enforcing | -0.386 |
interference | -0.376 |
Token | two |
Token ID | 734 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
gements | +1.714 |
attest | +1.633 |
emed | +1.553 |
Blossom | +1.498 |
creen | +1.489 |
Neg logit contributions | |
disrupting | -2.201 |
disruption | -2.100 |
obstruction | -2.071 |
enforcing | -2.041 |
interference | -2.014 |
Token | ? |
Token ID | 30 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
gements | +1.623 |
attest | +1.551 |
emed | +1.482 |
Blossom | +1.441 |
creen | +1.430 |
Neg logit contributions | |
disrupting | -1.831 |
disruption | -1.731 |
obstruction | -1.709 |
enforcing | -1.683 |
interference | -1.656 |
Token |
|
Token ID | 198 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
disrupting | +0.531 |
disruption | +0.502 |
obstruction | +0.487 |
HUD | +0.486 |
enforcing | +0.484 |
Neg logit contributions | |
gements | -0.805 |
attest | -0.791 |
emed | -0.751 |
feature | -0.727 |
respectively | -0.722 |
Token |
|
Token ID | 198 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
gements | +0.858 |
attest | +0.838 |
emed | +0.803 |
feature | +0.784 |
Blossom | +0.780 |
Neg logit contributions | |
disrupting | -0.683 |
disruption | -0.644 |
obstruction | -0.632 |
enforcing | -0.623 |
interference | -0.612 |
Token | T |
Token ID | 51 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
gements | +1.538 |
attest | +1.508 |
emed | +1.388 |
feature | +1.373 |
Blossom | +1.367 |
Neg logit contributions | |
disrupting | -1.787 |
disruption | -1.704 |
HUD | -1.702 |
obstruction | -1.669 |
enforcing | -1.654 |
Token | OM |
Token ID | 2662 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
OB | +7.143 |
AKING | +7.013 |
HUD | +6.901 |
REP | +6.653 |
disrupting | +6.617 |
Neg logit contributions | |
gements | -17.643 |
attest | -16.965 |
soever | -16.552 |
Blossom | -16.438 |
comings | -15.929 |
Token | KING |
Token ID | 32957 |
Feature activation | -9.54e-02 |
Pos logit contributions | |
AKING | +5.935 |
HUD | +5.770 |
disrupting | +5.601 |
disruption | +5.120 |
enforcing | +5.046 |
Neg logit contributions | |
gements | -15.035 |
emed | -14.219 |
attest | -13.843 |
creen | -13.256 |
faced | -13.211 |
Token | : |
Token ID | 25 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
gements | +0.384 |
attest | +0.374 |
emed | +0.356 |
feature | +0.346 |
Blossom | +0.344 |
Neg logit contributions | |
disrupting | -0.410 |
disruption | -0.389 |
obstruction | -0.383 |
enforcing | -0.379 |
interference | -0.373 |
Token | Well |
Token ID | 3894 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
gements | +0.638 |
attest | +0.614 |
emed | +0.592 |
creen | +0.566 |
feature | +0.562 |
Neg logit contributions | |
disrupting | -0.517 |
disruption | -0.489 |
obstruction | -0.479 |
enforcing | -0.473 |
HUD | -0.469 |
Token | , |
Token ID | 11 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
disrupting | +0.853 |
disruption | +0.806 |
obstruction | +0.790 |
enforcing | +0.769 |
interference | +0.762 |
Neg logit contributions | |
gements | -0.997 |
attest | -0.960 |
emed | -0.909 |
Blossom | -0.875 |
creen | -0.870 |
Token | version |
Token ID | 2196 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
gements | +2.155 |
attest | +2.064 |
emed | +1.985 |
creen | +1.866 |
Blossom | +1.863 |
Neg logit contributions | |
disrupting | -3.210 |
disruption | -3.110 |
obstruction | -3.057 |
enforcing | -3.019 |
interference | -2.990 |
Token | of |
Token ID | 286 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
gements | +0.177 |
attest | +0.173 |
emed | +0.166 |
Blossom | +0.162 |
feature | +0.161 |
Neg logit contributions | |
disrupting | -0.138 |
disruption | -0.130 |
obstruction | -0.127 |
enforcing | -0.126 |
interference | -0.123 |
Token | � |
Token ID | 564 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
gements | +1.003 |
attest | +0.982 |
emed | +0.948 |
creen | +0.917 |
respectively | +0.906 |
Neg logit contributions | |
disrupting | -0.876 |
disruption | -0.839 |
obstruction | -0.825 |
enforcing | -0.812 |
interference | -0.800 |
Token | � |
Token ID | 250 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
disrupting | +3.226 |
disruption | +3.112 |
obstruction | +3.083 |
HUD | +3.032 |
enforcing | +3.016 |
Neg logit contributions | |
attest | -3.039 |
gements | -2.949 |
creen | -2.866 |
emed | -2.847 |
Blossom | -2.798 |
Token | The |
Token ID | 464 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
disrupting | +8.456 |
disruption | +8.135 |
HUD | +8.066 |
obstruction | +8.040 |
enforcing | +7.962 |
Neg logit contributions | |
gements | -7.151 |
attest | -7.043 |
creen | -6.613 |
Blossom | -6.519 |
emed | -6.511 |
Token | OS |
Token ID | 7294 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
disrupting | +1.339 |
disruption | +1.291 |
obstruction | +1.266 |
enforcing | +1.246 |
interference | +1.225 |
Neg logit contributions | |
gements | -1.290 |
attest | -1.260 |
emed | -1.210 |
creen | -1.155 |
respectively | -1.150 |
Token | - |
Token ID | 12 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
disrupting | +12.759 |
enforcing | +12.525 |
disruption | +12.328 |
obstruction | +12.164 |
HUD | +12.158 |
Neg logit contributions | |
gements | -10.910 |
attest | -10.820 |
emed | -9.948 |
faced | -9.848 |
creen | -9.523 |
Token | per |
Token ID | 525 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
disrupting | +4.026 |
HUD | +3.870 |
disruption | +3.853 |
enforcing | +3.820 |
activity | +3.756 |
Neg logit contributions | |
attest | -4.127 |
gements | -4.103 |
emed | -3.800 |
Blossom | -3.727 |
creen | -3.717 |
Token | iment |
Token ID | 3681 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
disrupting | +1.734 |
disruption | +1.630 |
obstruction | +1.611 |
enforcing | +1.609 |
activity | +1.547 |
Neg logit contributions | |
attest | -2.036 |
gements | -2.035 |
emed | -1.897 |
respectively | -1.859 |
Blossom | -1.842 |
Token | � |
Token ID | 447 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
disrupting | +2.392 |
disruption | +2.273 |
obstruction | +2.250 |
enforcing | +2.217 |
interference | +2.168 |
Neg logit contributions | |
gements | -2.827 |
attest | -2.757 |
creen | -2.621 |
emed | -2.611 |
Blossom | -2.553 |
Token | � |
Token ID | 247 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
disrupting | +3.085 |
disruption | +2.962 |
obstruction | +2.892 |
enforcing | +2.874 |
HUD | +2.794 |
Neg logit contributions | |
attest | -3.222 |
gements | -3.153 |
emed | -3.044 |
creen | -3.034 |
Blossom | -2.928 |
Token | Cycling |
Token ID | 25550 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
disrupting | +0.494 |
disruption | +0.469 |
obstruction | +0.457 |
enforcing | +0.451 |
interference | +0.446 |
Neg logit contributions | |
gements | -0.611 |
attest | -0.590 |
emed | -0.571 |
Blossom | -0.552 |
creen | -0.544 |
Token | Team |
Token ID | 4816 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
disrupting | +3.976 |
disruption | +3.820 |
enforcing | +3.675 |
interference | +3.639 |
obstruction | +3.558 |
Neg logit contributions | |
attest | -5.841 |
gements | -5.816 |
emed | -5.473 |
Blossom | -5.344 |
faced | -5.308 |
Token | ) |
Token ID | 8 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
disrupting | +7.093 |
disruption | +6.758 |
enforcing | +6.659 |
obstruction | +6.594 |
interference | +6.564 |
Neg logit contributions | |
gements | -7.992 |
attest | -7.791 |
emed | -7.417 |
creen | -7.271 |
faced | -7.210 |
Token | s |
Token ID | 264 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
disrupting | +1.518 |
disruption | +1.447 |
obstruction | +1.426 |
interference | +1.397 |
enforcing | +1.392 |
Neg logit contributions | |
gements | -1.980 |
attest | -1.919 |
emed | -1.865 |
creen | -1.808 |
feature | -1.783 |
Token | . |
Token ID | 13 |
Feature activation | +9.23e-02 |
Pos logit contributions | |
disrupting | +4.861 |
HUD | +4.665 |
disruption | +4.633 |
obstruction | +4.579 |
enforcing | +4.533 |
Neg logit contributions | |
gements | -6.385 |
attest | -6.249 |
emed | -5.794 |
comings | -5.588 |
Blossom | -5.586 |
Token | t |
Token ID | 83 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
disrupting | +0.427 |
disruption | +0.409 |
obstruction | +0.404 |
enforcing | +0.399 |
interference | +0.393 |
Neg logit contributions | |
gements | -0.338 |
attest | -0.330 |
emed | -0.308 |
feature | -0.297 |
Blossom | -0.297 |
Token | . |
Token ID | 13 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
HUD | +7.458 |
disrupting | +7.234 |
interference | +7.016 |
enforcing | +6.988 |
disruption | +6.963 |
Neg logit contributions | |
gements | -16.955 |
attest | -16.410 |
comings | -15.743 |
emed | -15.114 |
citiz | -15.004 |
Token | A |
Token ID | 317 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
disrupting | +2.166 |
disruption | +2.065 |
obstruction | +2.014 |
enforcing | +1.984 |
interference | +1.979 |
Neg logit contributions | |
gements | -3.135 |
attest | -3.010 |
emed | -2.921 |
creen | -2.821 |
faced | -2.757 |
Token | US |
Token ID | 2937 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
gements | +2.850 |
attest | +2.804 |
emed | +2.694 |
Blossom | +2.638 |
feature | +2.638 |
Neg logit contributions | |
disrupting | -1.426 |
disruption | -1.322 |
obstruction | -1.282 |
enforcing | -1.259 |
interference | -1.226 |
Token | 5 |
Token ID | 642 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
gements | +0.099 |
attest | +0.096 |
emed | +0.092 |
Blossom | +0.089 |
feature | +0.089 |
Neg logit contributions | |
disrupting | -0.091 |
disruption | -0.087 |
obstruction | -0.085 |
enforcing | -0.084 |
interference | -0.083 |
Token | Robbie |
Token ID | 34855 |
Feature activation | +7.55e-01 |
Pos logit contributions | |
gements | +0.236 |
attest | +0.229 |
emed | +0.220 |
creen | +0.212 |
feature | +0.210 |
Neg logit contributions | |
disrupting | -0.207 |
disruption | -0.198 |
enforcing | -0.192 |
obstruction | -0.192 |
interference | -0.188 |
Token | in |
Token ID | 287 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
disrupting | +1.769 |
disruption | +1.664 |
obstruction | +1.610 |
enforcing | +1.609 |
interference | +1.590 |
Neg logit contributions | |
gements | -1.984 |
attest | -1.917 |
emed | -1.862 |
Blossom | -1.819 |
creen | -1.791 |
Token | the |
Token ID | 262 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
disrupting | +2.402 |
disruption | +2.238 |
enforcing | +2.212 |
obstruction | +2.156 |
interference | +2.098 |
Neg logit contributions | |
gements | -2.940 |
attest | -2.805 |
emed | -2.771 |
Blossom | -2.690 |
creen | -2.668 |
Token | Russian |
Token ID | 3394 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
disrupting | +3.702 |
disruption | +3.591 |
obstruction | +3.444 |
enforcing | +3.426 |
interference | +3.325 |
Neg logit contributions | |
gements | -4.927 |
emed | -4.699 |
attest | -4.656 |
Blossom | -4.427 |
creen | -4.422 |
Token | Arctic |
Token ID | 15723 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
disrupting | +1.099 |
disruption | +1.053 |
obstruction | +1.025 |
enforcing | +1.012 |
interference | +0.991 |
Neg logit contributions | |
gements | -1.433 |
attest | -1.359 |
emed | -1.333 |
creen | -1.294 |
Blossom | -1.281 |
Token | where |
Token ID | 810 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
gements | +1.358 |
attest | +1.295 |
emed | +1.254 |
Blossom | +1.213 |
creen | +1.211 |
Neg logit contributions | |
disrupting | -1.280 |
disruption | -1.212 |
obstruction | -1.186 |
enforcing | -1.180 |
interference | -1.152 |
Token | operating |
Token ID | 5361 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
disrupting | +1.876 |
disruption | +1.798 |
obstruction | +1.749 |
enforcing | +1.717 |
interference | +1.699 |
Neg logit contributions | |
gements | -2.274 |
attest | -2.170 |
emed | -2.152 |
Blossom | -2.070 |
creen | -2.056 |
Token | conditions |
Token ID | 3403 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
disrupting | +12.010 |
disruption | +11.898 |
interference | +11.485 |
enforcing | +11.307 |
thrott | +11.190 |
Neg logit contributions | |
gements | -13.690 |
emed | -12.392 |
creen | -12.375 |
Blossom | -12.085 |
attest | -11.834 |
Token | are |
Token ID | 389 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
disrupting | +6.429 |
disruption | +6.180 |
obstruction | +6.079 |
enforcing | +6.018 |
interference | +5.962 |
Neg logit contributions | |
gements | -4.511 |
emed | -3.972 |
Blossom | -3.774 |
attest | -3.698 |
creen | -3.679 |
Token | beyond |
Token ID | 3675 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
disrupting | +2.156 |
disruption | +2.019 |
obstruction | +1.988 |
enforcing | +1.960 |
punishing | +1.918 |
Neg logit contributions | |
gements | -2.864 |
attest | -2.667 |
emed | -2.656 |
Blossom | -2.582 |
creen | -2.553 |
Token | the |
Token ID | 262 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
disrupting | +1.438 |
disruption | +1.373 |
obstruction | +1.341 |
interference | +1.303 |
enforcing | +1.295 |
Neg logit contributions | |
gements | -1.807 |
attest | -1.701 |
emed | -1.660 |
Blossom | -1.645 |
creen | -1.639 |
Token | capability |
Token ID | 12971 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
disrupting | +4.029 |
disruption | +3.890 |
obstruction | +3.785 |
interference | +3.745 |
enforcing | +3.677 |
Neg logit contributions | |
gements | -5.310 |
emed | -4.947 |
attest | -4.938 |
creen | -4.816 |
Blossom | -4.790 |
Token | if |
Token ID | 611 |
Feature activation | +7.77e-01 |
Pos logit contributions | |
disrupting | +2.576 |
disruption | +2.426 |
obstruction | +2.384 |
interference | +2.372 |
enforcing | +2.354 |
Neg logit contributions | |
gements | -3.330 |
attest | -3.223 |
emed | -3.153 |
respectively | -2.975 |
creen | -2.955 |
Token | he |
Token ID | 339 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
disrupting | +2.806 |
obstruction | +2.661 |
disruption | +2.633 |
enforcing | +2.613 |
interference | +2.530 |
Neg logit contributions | |
gements | -3.739 |
attest | -3.590 |
emed | -3.518 |
Blossom | -3.336 |
creen | -3.328 |
Token | was |
Token ID | 373 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
disrupting | +4.489 |
obstruction | +4.233 |
enforcing | +4.197 |
thrott | +4.175 |
disruption | +4.162 |
Neg logit contributions | |
gements | -4.272 |
attest | -4.069 |
emed | -3.876 |
Blossom | -3.794 |
creen | -3.676 |
Token | � |
Token ID | 564 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
disrupting | +4.530 |
obstruction | +4.268 |
enforcing | +4.238 |
punishing | +4.084 |
interfering | +4.046 |
Neg logit contributions | |
gements | -5.920 |
attest | -5.682 |
emed | -5.539 |
Blossom | -5.280 |
creen | -5.271 |
Token | � |
Token ID | 250 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
disrupting | +2.388 |
obstruction | +2.268 |
enforcing | +2.265 |
disruption | +2.258 |
thrott | +2.171 |
Neg logit contributions | |
gements | -2.226 |
attest | -2.206 |
creen | -2.037 |
emed | -2.011 |
feature | -2.000 |
Token | t |
Token ID | 83 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
disrupting | +6.481 |
obstruction | +6.208 |
enforcing | +6.118 |
disruption | +5.960 |
activity | +5.958 |
Neg logit contributions | |
attest | -5.705 |
gements | -5.674 |
Blossom | -5.378 |
feature | -5.215 |
creen | -5.077 |
Token | empt |
Token ID | 1791 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
disrupting | +11.436 |
oxic | +10.873 |
punishing | +10.870 |
enforcing | +10.724 |
obstruction | +10.617 |
Neg logit contributions | |
attest | -13.538 |
gements | -13.511 |
Blossom | -12.620 |
soever | -12.375 |
respectively | -12.131 |
Token | ed |
Token ID | 276 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
disrupting | +7.171 |
enforcing | +6.803 |
obstruction | +6.740 |
disruption | +6.717 |
punishing | +6.671 |
Neg logit contributions | |
gements | -7.533 |
attest | -7.222 |
Blossom | -6.991 |
creen | -6.811 |
respectively | -6.620 |
Token | � |
Token ID | 447 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
disrupting | +1.658 |
obstruction | +1.561 |
disruption | +1.555 |
enforcing | +1.511 |
interference | +1.494 |
Neg logit contributions | |
gements | -2.281 |
attest | -2.232 |
emed | -2.160 |
Blossom | -2.118 |
feature | -2.068 |
Token | � |
Token ID | 251 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
disrupting | +2.533 |
disruption | +2.444 |
obstruction | +2.428 |
enforcing | +2.369 |
HUD | +2.327 |
Neg logit contributions | |
attest | -2.398 |
gements | -2.370 |
emed | -2.222 |
creen | -2.205 |
feature | -2.182 |
Token | to |
Token ID | 284 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
disrupting | +5.792 |
obstruction | +5.605 |
disruption | +5.476 |
enforcing | +5.422 |
interference | +5.272 |
Neg logit contributions | |
gements | -6.302 |
attest | -6.067 |
emed | -5.740 |
Blossom | -5.617 |
feature | -5.602 |
Token | lawsuit |
Token ID | 8087 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
gements | +1.293 |
attest | +1.238 |
emed | +1.211 |
respectively | +1.149 |
creen | +1.148 |
Neg logit contributions | |
disrupting | -1.086 |
disruption | -1.036 |
obstruction | -1.021 |
enforcing | -0.984 |
interference | -0.979 |
Token | claiming |
Token ID | 8512 |
Feature activation | -6.86e-01 |
Pos logit contributions | |
gements | +4.241 |
attest | +4.143 |
emed | +3.993 |
feature | +3.898 |
Blossom | +3.883 |
Neg logit contributions | |
disrupting | -2.677 |
disruption | -2.498 |
obstruction | -2.445 |
enforcing | -2.408 |
interference | -2.354 |
Token | that |
Token ID | 326 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
gements | +3.215 |
attest | +3.124 |
emed | +3.006 |
feature | +2.910 |
Blossom | +2.901 |
Neg logit contributions | |
disrupting | -2.545 |
disruption | -2.404 |
obstruction | -2.364 |
enforcing | -2.321 |
interference | -2.289 |
Token | he |
Token ID | 339 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
disrupting | +0.426 |
disruption | +0.400 |
obstruction | +0.397 |
enforcing | +0.393 |
interference | +0.386 |
Neg logit contributions | |
gements | -0.541 |
attest | -0.518 |
emed | -0.504 |
creen | -0.483 |
Blossom | -0.477 |
Token | was |
Token ID | 373 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
disrupting | +2.197 |
disruption | +2.044 |
obstruction | +2.034 |
thrott | +2.029 |
enforcing | +2.004 |
Neg logit contributions | |
gements | -2.716 |
attest | -2.598 |
emed | -2.522 |
Blossom | -2.446 |
creen | -2.411 |
Token | t |
Token ID | 256 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
disrupting | +5.329 |
thrott | +5.034 |
obstruction | +4.916 |
enforcing | +4.883 |
disruption | +4.871 |
Neg logit contributions | |
gements | -7.250 |
attest | -6.889 |
emed | -6.825 |
Blossom | -6.488 |
comings | -6.452 |
Token | aunted |
Token ID | 20227 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
disrupting | +10.426 |
HUD | +10.289 |
oxic | +9.978 |
disruption | +9.771 |
thrott | +9.760 |
Neg logit contributions | |
Blossom | -13.307 |
attest | -13.285 |
gements | -12.822 |
feature | -12.347 |
creen | -12.314 |
Token | and |
Token ID | 290 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
disrupting | +2.906 |
disruption | +2.659 |
enforcing | +2.643 |
obstruction | +2.626 |
interference | +2.557 |
Neg logit contributions | |
gements | -3.260 |
attest | -3.100 |
emed | -3.035 |
Blossom | -2.856 |
creen | -2.833 |
Token | falsely |
Token ID | 24566 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
disrupting | +4.599 |
disruption | +4.246 |
thrott | +4.239 |
obstruction | +4.166 |
enforcing | +3.976 |
Neg logit contributions | |
gements | -7.136 |
attest | -6.643 |
emed | -6.548 |
Blossom | -6.462 |
comings | -6.458 |
Token | arrested |
Token ID | 5169 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
gements | +0.759 |
attest | +0.729 |
emed | +0.713 |
Blossom | +0.695 |
feature | +0.687 |
Neg logit contributions | |
disrupting | -0.461 |
disruption | -0.426 |
obstruction | -0.420 |
enforcing | -0.413 |
thrott | -0.403 |
Token | by |
Token ID | 416 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
disrupting | +3.257 |
enforcing | +2.989 |
disruption | +2.950 |
obstruction | +2.939 |
interference | +2.872 |
Neg logit contributions | |
gements | -3.681 |
attest | -3.479 |
emed | -3.431 |
creen | -3.210 |
Blossom | -3.168 |
Token | public |
Token ID | 1171 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
gements | +4.515 |
attest | +4.412 |
emed | +4.219 |
feature | +4.120 |
Blossom | +4.087 |
Neg logit contributions | |
disrupting | -3.980 |
disruption | -3.757 |
obstruction | -3.696 |
enforcing | -3.648 |
interference | -3.583 |
Token | a |
Token ID | 257 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
gements | +2.315 |
attest | +2.248 |
emed | +2.129 |
Blossom | +2.085 |
respectively | +2.074 |
Neg logit contributions | |
disrupting | -2.203 |
disruption | -2.098 |
obstruction | -2.066 |
enforcing | -2.025 |
interference | -2.003 |
Token | uj |
Token ID | 23577 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
gements | +1.864 |
attest | +1.819 |
emed | +1.737 |
Blossom | +1.693 |
feature | +1.683 |
Neg logit contributions | |
disrupting | -1.663 |
disruption | -1.580 |
obstruction | -1.545 |
enforcing | -1.522 |
interference | -1.497 |
Token | our |
Token ID | 454 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
gements | +6.070 |
attest | +5.858 |
emed | +5.664 |
feature | +5.504 |
Blossom | +5.448 |
Neg logit contributions | |
disrupting | -5.829 |
disruption | -5.521 |
obstruction | -5.400 |
HUD | -5.305 |
enforcing | -5.281 |
Token | d |
Token ID | 67 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
gements | +5.276 |
attest | +5.030 |
emed | +4.706 |
feature | +4.645 |
respectively | +4.577 |
Neg logit contributions | |
disrupting | -7.970 |
disruption | -7.637 |
obstruction | -7.575 |
enforcing | -7.435 |
HUD | -7.404 |
Token | ' |
Token ID | 6 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
gements | +5.388 |
attest | +5.257 |
emed | +4.897 |
respectively | +4.879 |
feature | +4.878 |
Neg logit contributions | |
disrupting | -7.411 |
disruption | -7.048 |
obstruction | -6.967 |
enforcing | -6.836 |
HUD | -6.830 |
Token | h |
Token ID | 71 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
gements | +9.358 |
respectively | +8.342 |
attest | +8.115 |
emed | +8.024 |
feature | +8.015 |
Neg logit contributions | |
disrupting | -19.155 |
HUD | -18.754 |
disruption | -18.098 |
thrott | -17.734 |
enforcing | -17.448 |
Token | ui |
Token ID | 9019 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
gements | +4.506 |
attest | +4.260 |
emed | +4.098 |
feature | +3.932 |
respectively | +3.900 |
Neg logit contributions | |
disrupting | -6.943 |
disruption | -6.653 |
obstruction | -6.569 |
enforcing | -6.478 |
HUD | -6.444 |
Token | , |
Token ID | 11 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
gements | +2.138 |
attest | +2.069 |
emed | +1.979 |
Blossom | +1.928 |
feature | +1.917 |
Neg logit contributions | |
disrupting | -1.981 |
disruption | -1.881 |
obstruction | -1.859 |
enforcing | -1.826 |
interference | -1.786 |
Token | merc |
Token ID | 11991 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
gements | +0.904 |
attest | +0.870 |
emed | +0.842 |
creen | +0.819 |
Blossom | +0.814 |
Neg logit contributions | |
disrupting | -0.814 |
disruption | -0.772 |
obstruction | -0.761 |
enforcing | -0.748 |
interference | -0.734 |
Token | red |
Token ID | 445 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
gements | +6.126 |
attest | +5.899 |
emed | +5.664 |
feature | +5.438 |
respectively | +5.404 |
Neg logit contributions | |
disrupting | -7.292 |
disruption | -6.929 |
obstruction | -6.799 |
enforcing | -6.704 |
interference | -6.617 |
Token | possible |
Token ID | 1744 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
disrupting | +2.386 |
disruption | +2.347 |
obstruction | +2.318 |
interference | +2.254 |
enforcing | +2.243 |
Neg logit contributions | |
gements | -2.079 |
attest | -1.918 |
emed | -1.905 |
creen | -1.833 |
Blossom | -1.800 |
Token | objection |
Token ID | 19630 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
disrupting | +2.425 |
disruption | +2.394 |
obstruction | +2.345 |
interference | +2.294 |
enforcing | +2.252 |
Neg logit contributions | |
gements | -2.375 |
attest | -2.200 |
emed | -2.186 |
creen | -2.118 |
berry | -2.054 |
Token | to |
Token ID | 284 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
disrupting | +1.163 |
obstruction | +1.108 |
disruption | +1.106 |
enforcing | +1.087 |
interference | +1.070 |
Neg logit contributions | |
gements | -1.235 |
attest | -1.178 |
emed | -1.135 |
creen | -1.107 |
Blossom | -1.085 |
Token | how |
Token ID | 703 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
gements | +1.447 |
attest | +1.404 |
emed | +1.356 |
creen | +1.306 |
Blossom | +1.297 |
Neg logit contributions | |
disrupting | -1.304 |
disruption | -1.236 |
obstruction | -1.209 |
enforcing | -1.202 |
interference | -1.178 |
Token | the |
Token ID | 262 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
disrupting | +2.961 |
disruption | +2.833 |
obstruction | +2.788 |
enforcing | +2.714 |
interference | +2.688 |
Neg logit contributions | |
gements | -3.281 |
attest | -3.149 |
emed | -3.039 |
creen | -2.960 |
Blossom | -2.897 |
Token | Hug |
Token ID | 12905 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
disrupting | +0.992 |
disruption | +0.962 |
obstruction | +0.945 |
enforcing | +0.927 |
interference | +0.925 |
Neg logit contributions | |
gements | -0.794 |
attest | -0.754 |
emed | -0.738 |
creen | -0.705 |
Blossom | -0.694 |
Token | os |
Token ID | 418 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
gements | +12.195 |
attest | +11.501 |
feature | +11.012 |
emed | +10.718 |
Fest | +10.365 |
Neg logit contributions | |
disrupting | -16.290 |
obstruction | -15.849 |
disruption | -15.585 |
enforcing | -15.329 |
activity | -15.262 |
Token | shook |
Token ID | 14682 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
gements | +5.575 |
attest | +5.427 |
emed | +5.127 |
feature | +5.022 |
Blossom | +4.931 |
Neg logit contributions | |
disrupting | -7.147 |
disruption | -6.816 |
obstruction | -6.726 |
enforcing | -6.640 |
interference | -6.537 |
Token | out |
Token ID | 503 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
gements | +2.365 |
attest | +2.290 |
emed | +2.211 |
Blossom | +2.157 |
creen | +2.149 |
Neg logit contributions | |
disrupting | -1.409 |
disruption | -1.320 |
obstruction | -1.287 |
enforcing | -1.260 |
interference | -1.237 |
Token | this |
Token ID | 428 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
disrupting | +0.425 |
disruption | +0.399 |
obstruction | +0.393 |
enforcing | +0.391 |
interference | +0.382 |
Neg logit contributions | |
gements | -0.452 |
attest | -0.430 |
emed | -0.416 |
Blossom | -0.404 |
creen | -0.398 |
Token | year |
Token ID | 614 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
disrupting | +0.792 |
disruption | +0.769 |
obstruction | +0.750 |
interference | +0.740 |
enforcing | +0.733 |
Neg logit contributions | |
gements | -0.626 |
attest | -0.569 |
emed | -0.560 |
creen | -0.536 |
Blossom | -0.531 |
Token | Pan |
Token ID | 5961 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
gements | +2.021 |
attest | +1.941 |
emed | +1.866 |
feature | +1.788 |
Blossom | +1.785 |
Neg logit contributions | |
disrupting | -2.027 |
disruption | -1.930 |
obstruction | -1.900 |
enforcing | -1.868 |
interference | -1.845 |
Token | statue |
Token ID | 15207 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
disrupting | +2.533 |
obstruction | +2.430 |
disruption | +2.428 |
interference | +2.349 |
enforcing | +2.333 |
Neg logit contributions | |
gements | -2.203 |
attest | -1.992 |
emed | -1.989 |
creen | -1.838 |
respectively | -1.818 |
Token | in |
Token ID | 287 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
disrupting | +0.502 |
disruption | +0.470 |
obstruction | +0.462 |
enforcing | +0.460 |
interference | +0.448 |
Neg logit contributions | |
gements | -0.540 |
attest | -0.501 |
emed | -0.498 |
Blossom | -0.472 |
creen | -0.471 |
Token | London |
Token ID | 3576 |
Feature activation | -6.43e-01 |
Pos logit contributions | |
gements | +0.589 |
attest | +0.570 |
emed | +0.547 |
creen | +0.522 |
respectively | +0.517 |
Neg logit contributions | |
disrupting | -0.506 |
disruption | -0.481 |
obstruction | -0.474 |
enforcing | -0.463 |
interference | -0.461 |
Token | 's |
Token ID | 338 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
gements | +2.784 |
attest | +2.658 |
emed | +2.574 |
Blossom | +2.460 |
creen | +2.458 |
Neg logit contributions | |
disrupting | -2.645 |
disruption | -2.496 |
obstruction | -2.457 |
enforcing | -2.435 |
interference | -2.388 |
Token | Kens |
Token ID | 29018 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
gements | +2.135 |
attest | +2.061 |
emed | +1.979 |
creen | +1.904 |
feature | +1.886 |
Neg logit contributions | |
disrupting | -2.023 |
disruption | -1.923 |
obstruction | -1.889 |
enforcing | -1.856 |
interference | -1.841 |
Token | ington |
Token ID | 9557 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
gements | +7.642 |
emed | +7.285 |
Blossom | +6.840 |
berry | +6.772 |
creen | +6.688 |
Neg logit contributions | |
disrupting | -20.465 |
obstruction | -19.997 |
disruption | -19.606 |
thrott | -19.515 |
enforcing | -19.434 |
Token | Gardens |
Token ID | 29330 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
gements | +6.531 |
attest | +6.413 |
emed | +6.071 |
feature | +5.984 |
Blossom | +5.927 |
Neg logit contributions | |
disrupting | -7.333 |
disruption | -6.971 |
obstruction | -6.856 |
enforcing | -6.786 |
interference | -6.694 |
Token | has |
Token ID | 468 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
gements | +5.563 |
attest | +5.456 |
emed | +5.162 |
feature | +5.068 |
Blossom | +4.997 |
Neg logit contributions | |
disrupting | -6.265 |
disruption | -5.969 |
obstruction | -5.874 |
enforcing | -5.799 |
interference | -5.716 |
Token | been |
Token ID | 587 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
gements | +3.953 |
attest | +3.780 |
emed | +3.614 |
creen | +3.456 |
Blossom | +3.448 |
Neg logit contributions | |
disrupting | -4.813 |
disruption | -4.580 |
obstruction | -4.515 |
enforcing | -4.449 |
interference | -4.402 |
Token | at |
Token ID | 379 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
gements | +0.545 |
attest | +0.521 |
emed | +0.504 |
Blossom | +0.486 |
creen | +0.479 |
Neg logit contributions | |
disrupting | -0.480 |
disruption | -0.446 |
obstruction | -0.441 |
enforcing | -0.438 |
interference | -0.426 |
Token | their |
Token ID | 511 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
gements | +2.178 |
attest | +2.102 |
emed | +2.034 |
creen | +1.961 |
Blossom | +1.945 |
Neg logit contributions | |
disrupting | -2.051 |
disruption | -1.947 |
obstruction | -1.907 |
enforcing | -1.896 |
interference | -1.868 |
Token | animals |
Token ID | 4695 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
gements | +3.380 |
attest | +3.276 |
emed | +3.146 |
Blossom | +3.038 |
feature | +3.030 |
Neg logit contributions | |
disrupting | -3.146 |
disruption | -2.989 |
obstruction | -2.938 |
enforcing | -2.895 |
interference | -2.858 |
Token | roam |
Token ID | 35563 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
gements | +0.923 |
attest | +0.854 |
emed | +0.843 |
Blossom | +0.799 |
creen | +0.796 |
Neg logit contributions | |
disrupting | -1.183 |
disruption | -1.128 |
obstruction | -1.113 |
enforcing | -1.106 |
interference | -1.096 |
Token | as |
Token ID | 355 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
gements | +0.109 |
attest | +0.102 |
emed | +0.099 |
Blossom | +0.094 |
creen | +0.093 |
Neg logit contributions | |
disrupting | -0.140 |
disruption | -0.133 |
obstruction | -0.131 |
enforcing | -0.131 |
interference | -0.130 |
Token | they |
Token ID | 484 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
disrupting | +1.210 |
disruption | +1.152 |
obstruction | +1.143 |
enforcing | +1.120 |
interference | +1.102 |
Neg logit contributions | |
gements | -1.230 |
attest | -1.152 |
emed | -1.108 |
creen | -1.072 |
Blossom | -1.071 |
Token | gra |
Token ID | 7933 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
gements | +1.965 |
attest | +1.821 |
emed | +1.777 |
Blossom | +1.704 |
creen | +1.679 |
Neg logit contributions | |
disrupting | -2.988 |
disruption | -2.853 |
obstruction | -2.829 |
enforcing | -2.806 |
interference | -2.764 |
Token | ze |
Token ID | 2736 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
gements | +8.346 |
attest | +8.289 |
berry | +7.875 |
respectively | +7.700 |
emed | +7.689 |
Neg logit contributions | |
disrupting | -19.779 |
disruption | -19.406 |
obstruction | -19.014 |
enforcing | -18.643 |
interference | -18.216 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
gements | +2.536 |
attest | +2.438 |
emed | +2.347 |
Blossom | +2.260 |
feature | +2.255 |
Neg logit contributions | |
disrupting | -2.443 |
disruption | -2.304 |
obstruction | -2.271 |
enforcing | -2.249 |
interference | -2.208 |
Token | for |
Token ID | 329 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
disrupting | +0.734 |
disruption | +0.700 |
obstruction | +0.684 |
enforcing | +0.678 |
interference | +0.671 |
Neg logit contributions | |
gements | -0.629 |
emed | -0.582 |
attest | -0.582 |
Blossom | -0.552 |
creen | -0.552 |
Token | age |
Token ID | 496 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
gements | +1.415 |
attest | +1.372 |
emed | +1.318 |
Blossom | +1.282 |
creen | +1.264 |
Neg logit contributions | |
disrupting | -1.392 |
disruption | -1.328 |
obstruction | -1.302 |
enforcing | -1.291 |
interference | -1.267 |
Token | like |
Token ID | 588 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
gements | +2.560 |
attest | +2.460 |
emed | +2.371 |
Blossom | +2.291 |
creen | +2.284 |
Neg logit contributions | |
disrupting | -2.372 |
disruption | -2.238 |
obstruction | -2.203 |
enforcing | -2.180 |
interference | -2.144 |
Token | like |
Token ID | 588 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
disrupting | +1.406 |
disruption | +1.314 |
enforcing | +1.286 |
obstruction | +1.281 |
thrott | +1.271 |
Neg logit contributions | |
gements | -1.484 |
attest | -1.414 |
emed | -1.375 |
Blossom | -1.300 |
creen | -1.289 |
Token | Netflix |
Token ID | 12074 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
gements | +1.053 |
attest | +1.019 |
emed | +0.975 |
creen | +0.920 |
Blossom | +0.919 |
Neg logit contributions | |
disrupting | -1.288 |
disruption | -1.233 |
obstruction | -1.207 |
enforcing | -1.192 |
thrott | -1.177 |
Token | , |
Token ID | 11 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
gements | +3.250 |
attest | +3.159 |
emed | +3.011 |
feature | +2.913 |
Blossom | +2.900 |
Neg logit contributions | |
disrupting | -3.447 |
disruption | -3.279 |
obstruction | -3.219 |
enforcing | -3.182 |
interference | -3.133 |
Token | YouTube |
Token ID | 7444 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
gements | +0.127 |
attest | +0.122 |
emed | +0.118 |
creen | +0.112 |
Blossom | +0.111 |
Neg logit contributions | |
disrupting | -0.134 |
disruption | -0.127 |
obstruction | -0.124 |
enforcing | -0.122 |
thrott | -0.121 |
Token | , |
Token ID | 11 |
Feature activation | -7.95e-02 |
Pos logit contributions | |
gements | +1.406 |
attest | +1.383 |
emed | +1.319 |
creen | +1.262 |
Blossom | +1.261 |
Neg logit contributions | |
disrupting | -1.323 |
disruption | -1.258 |
obstruction | -1.228 |
enforcing | -1.207 |
interference | -1.192 |
Token | Hulu |
Token ID | 39739 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
gements | +0.321 |
attest | +0.313 |
emed | +0.301 |
creen | +0.286 |
Blossom | +0.282 |
Neg logit contributions | |
disrupting | -0.347 |
disruption | -0.330 |
obstruction | -0.321 |
enforcing | -0.316 |
interference | -0.312 |
Token | , |
Token ID | 11 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
respectively | +11.492 |
gements | +11.463 |
feature | +11.305 |
attest | +11.224 |
Blossom | +10.775 |
Neg logit contributions | |
disrupting | -15.299 |
obstruction | -14.936 |
HUD | -14.806 |
disruption | -14.712 |
enforcing | -14.379 |
Token | and |
Token ID | 290 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
gements | +1.606 |
attest | +1.564 |
emed | +1.500 |
creen | +1.434 |
Blossom | +1.423 |
Neg logit contributions | |
disrupting | -1.578 |
disruption | -1.499 |
obstruction | -1.467 |
enforcing | -1.441 |
interference | -1.419 |
Token | Amazon |
Token ID | 6186 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
gements | +2.630 |
attest | +2.556 |
emed | +2.444 |
creen | +2.341 |
Blossom | +2.340 |
Neg logit contributions | |
disrupting | -2.712 |
disruption | -2.580 |
obstruction | -2.535 |
enforcing | -2.492 |
interference | -2.456 |
Token | Prime |
Token ID | 5537 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
gements | +5.067 |
attest | +4.928 |
emed | +4.677 |
feature | +4.571 |
respectively | +4.538 |
Neg logit contributions | |
disrupting | -5.747 |
disruption | -5.471 |
obstruction | -5.402 |
enforcing | -5.331 |
interference | -5.250 |
Token | among |
Token ID | 1871 |
Feature activation | -7.24e-01 |
Pos logit contributions | |
gements | +10.928 |
attest | +10.889 |
respectively | +10.719 |
feature | +10.434 |
emed | +10.250 |
Neg logit contributions | |
disrupting | -13.362 |
obstruction | -13.094 |
HUD | -12.921 |
disruption | -12.684 |
enforcing | -12.511 |
Token | Louisiana |
Token ID | 13340 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
disrupting | +1.887 |
disruption | +1.806 |
obstruction | +1.766 |
enforcing | +1.746 |
interference | +1.711 |
Neg logit contributions | |
gements | -2.533 |
attest | -2.427 |
emed | -2.321 |
creen | -2.300 |
Blossom | -2.234 |
Token | and |
Token ID | 290 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
gements | +7.030 |
attest | +6.960 |
emed | +6.522 |
feature | +6.488 |
respectively | +6.371 |
Neg logit contributions | |
disrupting | -7.676 |
disruption | -7.369 |
obstruction | -7.199 |
enforcing | -7.056 |
HUD | -7.023 |
Token | talk |
Token ID | 1561 |
Feature activation | +6.70e-03 |
Pos logit contributions | |
disrupting | +0.634 |
disruption | +0.605 |
enforcing | +0.594 |
obstruction | +0.592 |
interference | +0.581 |
Neg logit contributions | |
gements | -0.428 |
attest | -0.400 |
emed | -0.392 |
creen | -0.374 |
Blossom | -0.372 |
Token | to |
Token ID | 284 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
disrupting | +0.027 |
disruption | +0.026 |
obstruction | +0.026 |
enforcing | +0.026 |
interference | +0.025 |
Neg logit contributions | |
gements | -0.028 |
attest | -0.027 |
emed | -0.026 |
creen | -0.026 |
Blossom | -0.026 |
Token | [ |
Token ID | 685 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
gements | +1.311 |
attest | +1.264 |
emed | +1.205 |
creen | +1.178 |
respectively | +1.160 |
Neg logit contributions | |
disrupting | -1.394 |
disruption | -1.333 |
obstruction | -1.312 |
enforcing | -1.299 |
interference | -1.277 |
Token | Louis |
Token ID | 32117 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
gements | +3.942 |
attest | +3.845 |
emed | +3.648 |
feature | +3.560 |
Blossom | +3.541 |
Neg logit contributions | |
disrupting | -4.008 |
disruption | -3.808 |
obstruction | -3.750 |
enforcing | -3.702 |
interference | -3.643 |
Token | ian |
Token ID | 666 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
gements | +7.452 |
attest | +6.826 |
feature | +6.293 |
berry | +6.203 |
emed | +5.882 |
Neg logit contributions | |
disrupting | -20.797 |
disruption | -19.966 |
obstruction | -19.958 |
enforcing | -19.509 |
thrott | -19.237 |
Token | ians |
Token ID | 1547 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
gements | +5.797 |
attest | +5.638 |
emed | +5.384 |
feature | +5.222 |
Blossom | +5.189 |
Neg logit contributions | |
disrupting | -6.696 |
disruption | -6.368 |
obstruction | -6.261 |
enforcing | -6.188 |
interference | -6.078 |
Token | ], |
Token ID | 4357 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
gements | +4.236 |
attest | +4.139 |
emed | +3.942 |
feature | +3.843 |
Blossom | +3.809 |
Neg logit contributions | |
disrupting | -4.395 |
disruption | -4.183 |
obstruction | -4.114 |
enforcing | -4.050 |
interference | -3.994 |
Token | they |
Token ID | 484 |
Feature activation | -9.35e-01 |
Pos logit contributions | |
gements | +0.978 |
attest | +0.934 |
emed | +0.911 |
creen | +0.873 |
Blossom | +0.861 |
Neg logit contributions | |
disrupting | -1.210 |
disruption | -1.159 |
obstruction | -1.136 |
enforcing | -1.126 |
interference | -1.108 |
Token | � |
Token ID | 447 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
gements | +3.361 |
attest | +3.225 |
emed | +3.077 |
Blossom | +2.965 |
feature | +2.949 |
Neg logit contributions | |
disrupting | -4.448 |
disruption | -4.251 |
obstruction | -4.190 |
enforcing | -4.149 |
interference | -4.083 |
Token | crisis |
Token ID | 4902 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
disruption | +4.876 |
disrupting | +4.641 |
obstruction | +4.508 |
interference | +4.239 |
punishing | +4.115 |
Neg logit contributions | |
gements | -9.159 |
emed | -8.697 |
attest | -8.478 |
Blossom | -8.183 |
creen | -8.143 |
Token | — |
Token ID | 851 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
disrupting | +2.943 |
disruption | +2.755 |
enforcing | +2.673 |
obstruction | +2.650 |
interference | +2.588 |
Neg logit contributions | |
gements | -3.186 |
attest | -3.021 |
emed | -2.946 |
Blossom | -2.819 |
feature | -2.771 |
Token | and |
Token ID | 290 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
disrupting | +4.079 |
disruption | +3.842 |
obstruction | +3.665 |
enforcing | +3.649 |
interference | +3.573 |
Neg logit contributions | |
gements | -4.464 |
emed | -4.253 |
attest | -4.236 |
creen | -3.988 |
Blossom | -3.979 |
Token | win |
Token ID | 1592 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
gements | +2.181 |
attest | +2.100 |
emed | +2.027 |
creen | +1.938 |
Blossom | +1.937 |
Neg logit contributions | |
disrupting | -2.201 |
disruption | -2.091 |
obstruction | -2.048 |
enforcing | -2.026 |
interference | -1.992 |
Token | a |
Token ID | 257 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
gements | +5.648 |
attest | +5.509 |
emed | +5.267 |
feature | +5.146 |
Blossom | +5.105 |
Neg logit contributions | |
disrupting | -5.182 |
disruption | -4.901 |
obstruction | -4.825 |
enforcing | -4.756 |
interference | -4.671 |
Token | Nobel |
Token ID | 20715 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
gements | +4.116 |
attest | +3.993 |
emed | +3.782 |
feature | +3.667 |
Blossom | +3.638 |
Neg logit contributions | |
disrupting | -5.333 |
disruption | -5.092 |
obstruction | -5.021 |
enforcing | -4.964 |
interference | -4.894 |
Token | prize |
Token ID | 11596 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
gements | +12.461 |
attest | +12.131 |
feature | +11.671 |
respectively | +11.656 |
emed | +11.539 |
Neg logit contributions | |
disrupting | -14.953 |
HUD | -14.747 |
inventoryQuantity | -14.243 |
obstruction | -14.069 |
disruption | -14.006 |
Token | in |
Token ID | 287 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
attest | +9.245 |
gements | +9.226 |
emed | +8.686 |
respectively | +8.651 |
feature | +8.548 |
Neg logit contributions | |
disrupting | -11.267 |
disruption | -10.888 |
obstruction | -10.734 |
HUD | -10.682 |
interference | -10.416 |
Token | the |
Token ID | 262 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
gements | +2.538 |
attest | +2.468 |
emed | +2.375 |
Blossom | +2.306 |
creen | +2.299 |
Neg logit contributions | |
disrupting | -1.915 |
disruption | -1.807 |
obstruction | -1.768 |
enforcing | -1.740 |
interference | -1.709 |
Token | process |
Token ID | 1429 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
gements | +1.988 |
attest | +1.935 |
emed | +1.860 |
creen | +1.802 |
Blossom | +1.793 |
Neg logit contributions | |
disrupting | -1.641 |
disruption | -1.562 |
obstruction | -1.523 |
enforcing | -1.500 |
interference | -1.481 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
disrupting | +1.815 |
disruption | +1.705 |
enforcing | +1.685 |
obstruction | +1.664 |
interference | +1.634 |
Neg logit contributions | |
gements | -1.882 |
attest | -1.814 |
emed | -1.736 |
Blossom | -1.704 |
creen | -1.684 |
Token | and |
Token ID | 290 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
gements | +0.242 |
attest | +0.232 |
emed | +0.224 |
Blossom | +0.216 |
creen | +0.213 |
Neg logit contributions | |
disrupting | -0.236 |
disruption | -0.222 |
enforcing | -0.218 |
obstruction | -0.216 |
interference | -0.212 |
Token | services |
Token ID | 2594 |
Feature activation | -9.14e-02 |
Pos logit contributions | |
disrupting | +1.695 |
disruption | +1.625 |
enforcing | +1.583 |
obstruction | +1.578 |
interference | +1.551 |
Neg logit contributions | |
gements | -1.658 |
attest | -1.558 |
emed | -1.554 |
creen | -1.482 |
faced | -1.466 |
Token | to |
Token ID | 284 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
gements | +0.390 |
attest | +0.371 |
emed | +0.358 |
Blossom | +0.346 |
creen | +0.342 |
Neg logit contributions | |
disrupting | -0.385 |
disruption | -0.362 |
enforcing | -0.356 |
obstruction | -0.354 |
interference | -0.346 |
Token | 10 |
Token ID | 838 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
gements | +3.272 |
attest | +3.182 |
emed | +3.064 |
feature | +2.952 |
creen | +2.951 |
Neg logit contributions | |
disrupting | -2.917 |
disruption | -2.757 |
obstruction | -2.701 |
enforcing | -2.684 |
interference | -2.625 |
Token | million |
Token ID | 1510 |
Feature activation | -9.64e-01 |
Pos logit contributions | |
disrupting | +1.458 |
disruption | +1.368 |
enforcing | +1.346 |
obstruction | +1.338 |
interference | +1.288 |
Neg logit contributions | |
gements | -1.805 |
attest | -1.753 |
emed | -1.692 |
creen | -1.658 |
Blossom | -1.607 |
Token | Hait |
Token ID | 19456 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
gements | +3.597 |
attest | +3.485 |
emed | +3.316 |
feature | +3.197 |
Blossom | +3.180 |
Neg logit contributions | |
disrupting | -4.442 |
disruption | -4.240 |
obstruction | -4.174 |
enforcing | -4.132 |
interference | -4.067 |
Token | ians |
Token ID | 1547 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
gements | +6.324 |
emed | +5.853 |
attest | +5.312 |
berry | +5.206 |
ean | +5.091 |
Neg logit contributions | |
disrupting | -21.189 |
obstruction | -20.776 |
disruption | -20.580 |
thrott | -20.472 |
enforcing | -20.214 |
Token | and |
Token ID | 290 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
gements | +4.459 |
attest | +4.360 |
emed | +4.138 |
feature | +4.038 |
respectively | +4.017 |
Neg logit contributions | |
disrupting | -4.738 |
disruption | -4.515 |
obstruction | -4.446 |
enforcing | -4.371 |
interference | -4.316 |
Token | increase |
Token ID | 2620 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
disrupting | +0.147 |
disruption | +0.140 |
obstruction | +0.136 |
enforcing | +0.136 |
interference | +0.133 |
Neg logit contributions | |
gements | -0.150 |
attest | -0.143 |
emed | -0.140 |
creen | -0.135 |
Blossom | -0.133 |
Token | investment |
Token ID | 4896 |
Feature activation | +8.65e-01 |
Pos logit contributions | |
disrupting | +2.063 |
disruption | +2.001 |
enforcing | +1.917 |
obstruction | +1.903 |
interference | +1.888 |
Neg logit contributions | |
gements | -2.813 |
attest | -2.684 |
emed | -2.656 |
creen | -2.574 |
faced | -2.482 |
Token | in |
Token ID | 287 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
disrupting | +3.269 |
disruption | +3.133 |
enforcing | +3.027 |
obstruction | +2.978 |
interference | +2.961 |
Neg logit contributions | |
gements | -3.884 |
emed | -3.686 |
attest | -3.670 |
creen | -3.536 |
faced | -3.456 |
Token | the |
Token ID | 262 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
gements | +2.205 |
attest | +2.125 |
emed | +2.051 |
creen | +1.974 |
Blossom | +1.967 |
Neg logit contributions | |
disrupting | -2.358 |
disruption | -2.244 |
obstruction | -2.204 |
enforcing | -2.188 |
interference | -2.146 |
Token | age |
Token ID | 2479 |
Feature activation | +8.51e-02 |
Pos logit contributions | |
gements | +0.646 |
attest | +0.627 |
emed | +0.609 |
creen | +0.583 |
Blossom | +0.578 |
Neg logit contributions | |
disrupting | -0.656 |
disruption | -0.633 |
obstruction | -0.618 |
enforcing | -0.610 |
interference | -0.605 |
Token | of |
Token ID | 286 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
disrupting | +0.306 |
disruption | +0.291 |
enforcing | +0.281 |
obstruction | +0.281 |
interference | +0.274 |
Neg logit contributions | |
gements | -0.404 |
attest | -0.392 |
emed | -0.386 |
Blossom | -0.371 |
creen | -0.368 |
Token | Jamie |
Token ID | 17826 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
gements | +1.027 |
attest | +0.988 |
emed | +0.954 |
creen | +0.910 |
Blossom | +0.909 |
Neg logit contributions | |
disrupting | -1.013 |
disruption | -0.973 |
obstruction | -0.949 |
enforcing | -0.940 |
interference | -0.926 |
Token | and |
Token ID | 290 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
gements | +6.419 |
attest | +6.328 |
emed | +6.006 |
feature | +5.948 |
Blossom | +5.905 |
Neg logit contributions | |
disrupting | -5.192 |
disruption | -4.925 |
obstruction | -4.859 |
enforcing | -4.728 |
HUD | -4.665 |
Token | Claire |
Token ID | 20758 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
disrupting | +1.862 |
disruption | +1.792 |
obstruction | +1.741 |
enforcing | +1.726 |
interference | +1.715 |
Neg logit contributions | |
gements | -1.614 |
attest | -1.542 |
emed | -1.484 |
creen | -1.430 |
otes | -1.386 |
Token | on |
Token ID | 319 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
attest | +15.605 |
gements | +15.272 |
respectively | +15.094 |
feature | +14.824 |
Blossom | +14.469 |
Neg logit contributions | |
disrupting | -10.568 |
obstruction | -10.181 |
disruption | -10.128 |
HUD | -10.074 |
activity | -9.666 |
Token | their |
Token ID | 511 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
disrupting | +0.247 |
disruption | +0.234 |
enforcing | +0.231 |
obstruction | +0.229 |
interference | +0.226 |
Neg logit contributions | |
gements | -0.279 |
attest | -0.269 |
emed | -0.259 |
Blossom | -0.247 |
creen | -0.247 |
Token | reunion |
Token ID | 30321 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
gements | +0.638 |
attest | +0.607 |
emed | +0.596 |
creen | +0.569 |
Blossom | +0.563 |
Neg logit contributions | |
disrupting | -0.639 |
disruption | -0.613 |
obstruction | -0.602 |
enforcing | -0.594 |
interference | -0.588 |
Token | night |
Token ID | 1755 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
gements | +7.197 |
attest | +7.128 |
feature | +6.646 |
emed | +6.633 |
respectively | +6.493 |
Neg logit contributions | |
disrupting | -7.780 |
disruption | -7.373 |
obstruction | -7.283 |
enforcing | -7.181 |
HUD | -7.100 |
Token | is |
Token ID | 318 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
disrupting | +1.574 |
disruption | +1.502 |
enforcing | +1.459 |
obstruction | +1.450 |
interference | +1.430 |
Neg logit contributions | |
gements | -1.589 |
emed | -1.473 |
attest | -1.468 |
Blossom | -1.425 |
creen | -1.395 |
Token | multi |
Token ID | 5021 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
disrupting | +1.245 |
disruption | +1.196 |
obstruction | +1.169 |
enforcing | +1.122 |
interference | +1.121 |
Neg logit contributions | |
gements | -1.474 |
attest | -1.406 |
emed | -1.375 |
creen | -1.325 |
maid | -1.296 |
Token | - |
Token ID | 12 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
gements | +1.335 |
attest | +1.303 |
emed | +1.249 |
feature | +1.219 |
Blossom | +1.211 |
Neg logit contributions | |
disrupting | -1.095 |
disruption | -1.033 |
obstruction | -1.014 |
enforcing | -1.000 |
interference | -0.981 |
Token | use |
Token ID | 1904 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
gements | +1.851 |
attest | +1.816 |
emed | +1.693 |
Blossom | +1.668 |
creen | +1.643 |
Neg logit contributions | |
disrupting | -1.976 |
disruption | -1.878 |
obstruction | -1.856 |
enforcing | -1.821 |
interference | -1.796 |
Token | trails |
Token ID | 19196 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
disrupting | +1.140 |
disruption | +1.100 |
obstruction | +1.079 |
interference | +1.031 |
enforcing | +1.017 |
Neg logit contributions | |
gements | -1.439 |
attest | -1.373 |
emed | -1.316 |
creen | -1.280 |
maid | -1.253 |
Token | from |
Token ID | 422 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
gements | +2.719 |
attest | +2.633 |
emed | +2.538 |
Blossom | +2.450 |
feature | +2.446 |
Neg logit contributions | |
disrupting | -2.200 |
disruption | -2.064 |
obstruction | -2.028 |
enforcing | -2.001 |
interference | -1.957 |
Token | Claire |
Token ID | 20758 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
gements | +0.951 |
attest | +0.922 |
emed | +0.885 |
creen | +0.850 |
feature | +0.841 |
Neg logit contributions | |
disrupting | -0.839 |
disruption | -0.795 |
obstruction | -0.780 |
enforcing | -0.763 |
interference | -0.755 |
Token | ville |
Token ID | 4244 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
gements | +13.303 |
berry | +13.165 |
Blossom | +13.090 |
attest | +12.857 |
emed | +12.627 |
Neg logit contributions | |
disrupting | -13.460 |
obstruction | -12.687 |
disruption | -12.656 |
enforcing | -12.507 |
thrott | -12.337 |
Token | Reserv |
Token ID | 40425 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
gements | +0.052 |
attest | +0.050 |
emed | +0.048 |
creen | +0.046 |
feature | +0.045 |
Neg logit contributions | |
disrupting | -0.048 |
disruption | -0.045 |
obstruction | -0.045 |
enforcing | -0.044 |
interference | -0.043 |
Token | oir |
Token ID | 10840 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
gements | +3.155 |
attest | +2.930 |
emed | +2.686 |
feature | +2.519 |
Blossom | +2.472 |
Neg logit contributions | |
disrupting | -9.271 |
disruption | -8.939 |
obstruction | -8.845 |
enforcing | -8.787 |
interference | -8.682 |
Token | in |
Token ID | 287 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
gements | +3.071 |
attest | +2.972 |
emed | +2.851 |
Blossom | +2.748 |
creen | +2.746 |
Neg logit contributions | |
disrupting | -2.817 |
disruption | -2.669 |
obstruction | -2.628 |
enforcing | -2.586 |
interference | -2.535 |
Token | Bram |
Token ID | 35315 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
gements | +0.575 |
attest | +0.556 |
emed | +0.535 |
creen | +0.516 |
feature | +0.508 |
Neg logit contributions | |
disrupting | -0.452 |
disruption | -0.430 |
obstruction | -0.420 |
enforcing | -0.412 |
interference | -0.406 |
Token | principles |
Token ID | 7811 |
Feature activation | -9.21e-01 |
Pos logit contributions | |
gements | +3.708 |
attest | +3.600 |
emed | +3.462 |
Blossom | +3.356 |
creen | +3.351 |
Neg logit contributions | |
disrupting | -3.111 |
disruption | -2.958 |
obstruction | -2.913 |
enforcing | -2.866 |
interference | -2.820 |
Token | as |
Token ID | 355 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
gements | +4.192 |
attest | +4.091 |
emed | +3.920 |
feature | +3.823 |
Blossom | +3.803 |
Neg logit contributions | |
disrupting | -3.467 |
disruption | -3.272 |
obstruction | -3.213 |
enforcing | -3.168 |
interference | -3.110 |
Token | for |
Token ID | 329 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
gements | +1.349 |
attest | +1.312 |
emed | +1.262 |
creen | +1.219 |
Blossom | +1.219 |
Neg logit contributions | |
disrupting | -1.140 |
disruption | -1.076 |
obstruction | -1.062 |
enforcing | -1.055 |
interference | -1.028 |
Token | the |
Token ID | 262 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
gements | +1.174 |
attest | +1.158 |
emed | +1.106 |
creen | +1.069 |
Blossom | +1.064 |
Neg logit contributions | |
disrupting | -1.026 |
disruption | -0.972 |
obstruction | -0.956 |
enforcing | -0.952 |
interference | -0.926 |
Token | Nobel |
Token ID | 20715 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
disrupting | +0.513 |
disruption | +0.496 |
obstruction | +0.488 |
enforcing | +0.480 |
interference | +0.476 |
Neg logit contributions | |
gements | -0.464 |
attest | -0.455 |
emed | -0.442 |
creen | -0.421 |
Blossom | -0.411 |
Token | Pri |
Token ID | 4389 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
gements | +12.798 |
attest | +12.188 |
respectively | +11.598 |
feature | +11.548 |
emed | +11.487 |
Neg logit contributions | |
disrupting | -13.448 |
HUD | -12.694 |
disruption | -12.631 |
obstruction | -12.428 |
thrott | -12.186 |
Token | zes |
Token ID | 12271 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
gements | +9.173 |
emed | +8.531 |
otes | +7.829 |
attest | +7.752 |
creen | +7.163 |
Neg logit contributions | |
disrupting | -19.165 |
disruption | -18.418 |
obstruction | -18.287 |
interfering | -18.119 |
interference | -17.829 |
Token | that |
Token ID | 326 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
gements | +8.806 |
attest | +8.641 |
emed | +8.137 |
respectively | +8.095 |
feature | +8.062 |
Neg logit contributions | |
disrupting | -9.077 |
disruption | -8.643 |
obstruction | -8.446 |
HUD | -8.215 |
interference | -8.200 |
Token | have |
Token ID | 423 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
gements | +6.265 |
attest | +6.107 |
emed | +5.761 |
feature | +5.628 |
Blossom | +5.555 |
Neg logit contributions | |
disrupting | -7.765 |
disruption | -7.407 |
obstruction | -7.299 |
enforcing | -7.200 |
interference | -7.103 |
Token | been |
Token ID | 587 |
Feature activation | -9.89e-01 |
Pos logit contributions | |
gements | +7.400 |
attest | +7.222 |
emed | +6.796 |
feature | +6.666 |
respectively | +6.582 |
Neg logit contributions | |
disrupting | -9.786 |
disruption | -9.387 |
obstruction | -9.252 |
enforcing | -9.085 |
HUD | -9.021 |
Token | awarded |
Token ID | 11343 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
gements | +4.669 |
attest | +4.560 |
emed | +4.378 |
feature | +4.264 |
Blossom | +4.257 |
Neg logit contributions | |
disrupting | -3.576 |
disruption | -3.357 |
obstruction | -3.299 |
enforcing | -3.270 |
interference | -3.191 |
Token | the |
Token ID | 262 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
disrupting | +0.886 |
disruption | +0.847 |
obstruction | +0.822 |
enforcing | +0.807 |
interference | +0.806 |
Neg logit contributions | |
gements | -1.078 |
attest | -1.045 |
emed | -0.993 |
creen | -0.964 |
Blossom | -0.961 |
Token | galaxy |
Token ID | 16161 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
disrupting | +1.733 |
disruption | +1.653 |
obstruction | +1.629 |
enforcing | +1.591 |
interference | +1.585 |
Neg logit contributions | |
gements | -2.104 |
attest | -2.035 |
emed | -1.950 |
creen | -1.883 |
Blossom | -1.863 |
Token | , |
Token ID | 11 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
gements | +1.414 |
attest | +1.369 |
emed | +1.303 |
creen | +1.256 |
feature | +1.253 |
Neg logit contributions | |
disrupting | -1.423 |
disruption | -1.345 |
obstruction | -1.320 |
enforcing | -1.309 |
interference | -1.290 |
Token | and |
Token ID | 290 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
gements | +0.814 |
attest | +0.788 |
emed | +0.757 |
creen | +0.728 |
Blossom | +0.725 |
Neg logit contributions | |
disrupting | -0.731 |
disruption | -0.680 |
obstruction | -0.673 |
enforcing | -0.668 |
interference | -0.655 |
Token | the |
Token ID | 262 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
gements | +2.247 |
attest | +2.189 |
emed | +2.093 |
creen | +2.015 |
Blossom | +2.013 |
Neg logit contributions | |
disrupting | -2.166 |
disruption | -2.048 |
obstruction | -2.018 |
enforcing | -1.991 |
interference | -1.964 |
Token | Milky |
Token ID | 34822 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
gements | +1.182 |
attest | +1.153 |
emed | +1.104 |
creen | +1.063 |
Blossom | +1.060 |
Neg logit contributions | |
disrupting | -1.120 |
disruption | -1.068 |
obstruction | -1.050 |
enforcing | -1.030 |
interference | -1.025 |
Token | Way |
Token ID | 6378 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
feature | +4.610 |
Blossom | +4.324 |
ingly | +3.883 |
otes | +3.778 |
pair | +3.772 |
Neg logit contributions | |
disrupting | -21.701 |
enforcing | -20.973 |
disruption | -20.820 |
inventoryQuantity | -20.581 |
HUD | -20.575 |
Token | is |
Token ID | 318 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
gements | +4.480 |
attest | +4.357 |
emed | +4.136 |
feature | +4.029 |
Blossom | +3.993 |
Neg logit contributions | |
disrupting | -5.167 |
disruption | -4.921 |
obstruction | -4.844 |
enforcing | -4.789 |
interference | -4.714 |
Token | at |
Token ID | 379 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
gements | +2.276 |
attest | +2.212 |
emed | +2.108 |
Blossom | +2.032 |
feature | +2.026 |
Neg logit contributions | |
disrupting | -2.312 |
disruption | -2.186 |
obstruction | -2.163 |
enforcing | -2.128 |
interference | -2.099 |
Token | least |
Token ID | 1551 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
disrupting | +5.001 |
disruption | +4.773 |
interference | +4.734 |
obstruction | +4.711 |
enforcing | +4.532 |
Neg logit contributions | |
gements | -6.508 |
attest | -6.383 |
emed | -6.024 |
Blossom | -5.865 |
creen | -5.861 |
Token | 100 |
Token ID | 1802 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
gements | +2.890 |
attest | +2.816 |
emed | +2.695 |
feature | +2.620 |
Blossom | +2.609 |
Neg logit contributions | |
disrupting | -2.585 |
disruption | -2.446 |
obstruction | -2.405 |
enforcing | -2.371 |
interference | -2.332 |
Token | famous |
Token ID | 5863 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
gements | +5.880 |
attest | +5.732 |
emed | +5.469 |
feature | +5.337 |
Blossom | +5.296 |
Neg logit contributions | |
disrupting | -5.825 |
disruption | -5.526 |
obstruction | -5.439 |
enforcing | -5.369 |
interference | -5.280 |
Token | Renaissance |
Token ID | 29396 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
gements | +8.253 |
attest | +8.228 |
feature | +7.798 |
emed | +7.753 |
Blossom | +7.632 |
Neg logit contributions | |
disrupting | -8.788 |
disruption | -8.334 |
obstruction | -8.180 |
HUD | -8.147 |
enforcing | -8.125 |
Token | artist |
Token ID | 6802 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
attest | +9.645 |
gements | +9.417 |
feature | +9.244 |
emed | +9.203 |
respectively | +8.935 |
Neg logit contributions | |
disrupting | -10.657 |
HUD | -10.224 |
disruption | -10.114 |
obstruction | -9.991 |
enforcing | -9.975 |
Token | , |
Token ID | 11 |
Feature activation | -6.47e-01 |
Pos logit contributions | |
gements | +1.481 |
attest | +1.413 |
emed | +1.354 |
creen | +1.301 |
Blossom | +1.300 |
Neg logit contributions | |
disrupting | -1.407 |
disruption | -1.315 |
obstruction | -1.294 |
enforcing | -1.293 |
interference | -1.264 |
Token | whose |
Token ID | 3025 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
gements | +2.874 |
attest | +2.768 |
emed | +2.677 |
creen | +2.579 |
feature | +2.564 |
Neg logit contributions | |
disrupting | -2.595 |
disruption | -2.425 |
enforcing | -2.389 |
obstruction | -2.387 |
interference | -2.325 |
Token | names |
Token ID | 3891 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
gements | +7.207 |
attest | +7.159 |
feature | +6.744 |
emed | +6.669 |
Blossom | +6.564 |
Neg logit contributions | |
disrupting | -7.712 |
disruption | -7.283 |
obstruction | -7.170 |
enforcing | -7.138 |
HUD | -7.123 |
Token | he |
Token ID | 339 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
attest | +14.065 |
feature | +13.289 |
gements | +13.244 |
Blossom | +12.817 |
emed | +12.781 |
Neg logit contributions | |
disrupting | -13.305 |
HUD | -12.778 |
disruption | -12.747 |
obstruction | -12.593 |
enforcing | -12.202 |
Token | had |
Token ID | 550 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
gements | +3.527 |
attest | +3.406 |
emed | +3.248 |
Blossom | +3.162 |
feature | +3.147 |
Neg logit contributions | |
disrupting | -3.612 |
disruption | -3.414 |
obstruction | -3.358 |
enforcing | -3.339 |
interference | -3.269 |
Token | found |
Token ID | 1043 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
gements | +4.448 |
attest | +4.301 |
emed | +4.128 |
feature | +3.981 |
Blossom | +3.981 |
Neg logit contributions | |
disrupting | -4.358 |
disruption | -4.119 |
obstruction | -4.057 |
enforcing | -4.013 |
interference | -3.945 |
Token | in |
Token ID | 287 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
gements | +3.021 |
attest | +2.893 |
emed | +2.799 |
creen | +2.709 |
Blossom | +2.708 |
Neg logit contributions | |
disrupting | -2.509 |
disruption | -2.333 |
obstruction | -2.308 |
enforcing | -2.292 |
interference | -2.235 |
Token | an |
Token ID | 281 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
gements | +1.868 |
attest | +1.788 |
emed | +1.737 |
creen | +1.685 |
feature | +1.666 |
Neg logit contributions | |
disrupting | -1.457 |
disruption | -1.370 |
obstruction | -1.352 |
enforcing | -1.337 |
interference | -1.328 |
Token | , |
Token ID | 11 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
gements | +4.593 |
attest | +4.466 |
emed | +4.263 |
feature | +4.119 |
Blossom | +4.110 |
Neg logit contributions | |
disrupting | -4.763 |
disruption | -4.527 |
obstruction | -4.448 |
enforcing | -4.400 |
interference | -4.322 |
Token | returning |
Token ID | 8024 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
gements | +2.521 |
attest | +2.443 |
emed | +2.349 |
creen | +2.265 |
Blossom | +2.241 |
Neg logit contributions | |
disrupting | -2.531 |
disruption | -2.398 |
obstruction | -2.350 |
enforcing | -2.338 |
interference | -2.289 |
Token | on |
Token ID | 319 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
gements | +5.479 |
attest | +5.348 |
emed | +5.124 |
feature | +4.997 |
Blossom | +4.971 |
Neg logit contributions | |
disrupting | -4.502 |
disruption | -4.249 |
obstruction | -4.171 |
enforcing | -4.112 |
interference | -4.037 |
Token | New |
Token ID | 968 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
gements | +0.966 |
attest | +0.943 |
emed | +0.894 |
creen | +0.856 |
Blossom | +0.852 |
Neg logit contributions | |
disrupting | -1.071 |
disruption | -1.022 |
obstruction | -1.003 |
enforcing | -0.990 |
interference | -0.971 |
Token | Year |
Token ID | 6280 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
gements | +4.894 |
attest | +4.460 |
emed | +4.068 |
feature | +4.053 |
Blossom | +3.963 |
Neg logit contributions | |
disrupting | -14.872 |
disruption | -14.358 |
obstruction | -14.250 |
enforcing | -14.057 |
HUD | -14.011 |
Token | 's |
Token ID | 338 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
gements | +8.837 |
attest | +8.537 |
feature | +8.106 |
Blossom | +8.039 |
respectively | +8.037 |
Neg logit contributions | |
disrupting | -9.087 |
disruption | -8.627 |
obstruction | -8.569 |
HUD | -8.554 |
enforcing | -8.461 |
Token | Day |
Token ID | 3596 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
gements | +10.347 |
attest | +10.042 |
respectively | +9.634 |
feature | +9.600 |
Blossom | +9.483 |
Neg logit contributions | |
disrupting | -16.103 |
HUD | -15.713 |
obstruction | -15.573 |
disruption | -15.246 |
enforcing | -15.244 |
Token | . |
Token ID | 13 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
gements | +5.708 |
attest | +5.557 |
emed | +5.291 |
feature | +5.153 |
Blossom | +5.113 |
Neg logit contributions | |
disrupting | -6.024 |
disruption | -5.726 |
obstruction | -5.638 |
enforcing | -5.565 |
interference | -5.479 |
Token |
|
Token ID | 198 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
gements | +1.290 |
attest | +1.258 |
emed | +1.198 |
creen | +1.158 |
feature | +1.144 |
Neg logit contributions | |
disrupting | -1.016 |
disruption | -0.966 |
obstruction | -0.947 |
enforcing | -0.938 |
HUD | -0.915 |
Token |
|
Token ID | 198 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
gements | +1.357 |
attest | +1.325 |
emed | +1.271 |
feature | +1.243 |
Blossom | +1.235 |
Neg logit contributions | |
disrupting | -1.112 |
disruption | -1.048 |
obstruction | -1.029 |
enforcing | -1.014 |
interference | -0.995 |
Token | C |
Token ID | 34 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
gements | +2.723 |
attest | +2.667 |
emed | +2.504 |
feature | +2.451 |
Blossom | +2.442 |
Neg logit contributions | |
disrupting | -2.730 |
disruption | -2.602 |
obstruction | -2.552 |
HUD | -2.535 |
enforcing | -2.518 |
Token | the |
Token ID | 262 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
gements | +2.604 |
attest | +2.495 |
emed | +2.412 |
creen | +2.317 |
feature | +2.310 |
Neg logit contributions | |
disrupting | -2.437 |
disruption | -2.314 |
obstruction | -2.268 |
enforcing | -2.247 |
interference | -2.209 |
Token | Playboy |
Token ID | 43995 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
gements | +1.038 |
attest | +0.993 |
emed | +0.963 |
creen | +0.935 |
respectively | +0.917 |
Neg logit contributions | |
disrupting | -0.850 |
disruption | -0.808 |
obstruction | -0.785 |
enforcing | -0.778 |
interference | -0.767 |
Token | empire |
Token ID | 13735 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
attest | +9.615 |
gements | +9.570 |
feature | +9.230 |
emed | +9.154 |
Blossom | +9.086 |
Neg logit contributions | |
disrupting | -8.511 |
disruption | -7.955 |
obstruction | -7.894 |
enforcing | -7.738 |
HUD | -7.641 |
Token | [ |
Token ID | 685 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
disrupting | +2.484 |
disruption | +2.322 |
enforcing | +2.318 |
obstruction | +2.277 |
interference | +2.244 |
Neg logit contributions | |
gements | -2.462 |
attest | -2.325 |
emed | -2.242 |
feature | -2.144 |
Blossom | -2.136 |
Token | under |
Token ID | 4625 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
gements | +4.688 |
attest | +4.556 |
emed | +4.341 |
feature | +4.207 |
Blossom | +4.177 |
Neg logit contributions | |
disrupting | -5.491 |
disruption | -5.237 |
obstruction | -5.157 |
enforcing | -5.092 |
interference | -5.016 |
Token | Hugh |
Token ID | 25464 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
gements | +1.619 |
attest | +1.572 |
emed | +1.502 |
feature | +1.446 |
creen | +1.424 |
Neg logit contributions | |
disrupting | -1.475 |
disruption | -1.409 |
obstruction | -1.374 |
enforcing | -1.359 |
interference | -1.351 |
Token | He |
Token ID | 679 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
attest | +13.655 |
gements | +13.372 |
respectively | +13.328 |
Blossom | +13.297 |
feature | +13.195 |
Neg logit contributions | |
disrupting | -13.441 |
obstruction | -12.572 |
enforcing | -12.561 |
disruption | -12.503 |
inventoryQuantity | -12.152 |
Token | f |
Token ID | 69 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
gements | +2.563 |
attest | +2.497 |
emed | +2.375 |
feature | +2.310 |
Blossom | +2.298 |
Neg logit contributions | |
disrupting | -2.680 |
disruption | -2.547 |
obstruction | -2.507 |
enforcing | -2.475 |
interference | -2.436 |
Token | ner |
Token ID | 1008 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
disrupting | +1.740 |
obstruction | +1.637 |
disruption | +1.626 |
enforcing | +1.626 |
interference | +1.577 |
Neg logit contributions | |
gements | -1.893 |
attest | -1.814 |
Blossom | -1.675 |
emed | -1.670 |
respectively | -1.636 |
Token | , |
Token ID | 11 |
Feature activation | +1.63e-03 |
Pos logit contributions | |
gements | +0.553 |
attest | +0.535 |
emed | +0.510 |
Blossom | +0.485 |
feature | +0.483 |
Neg logit contributions | |
disrupting | -0.616 |
disruption | -0.586 |
obstruction | -0.574 |
enforcing | -0.570 |
interference | -0.565 |
Token | whose |
Token ID | 3025 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
disrupting | +0.007 |
disruption | +0.006 |
obstruction | +0.006 |
enforcing | +0.006 |
interference | +0.006 |
Neg logit contributions | |
gements | -0.007 |
attest | -0.007 |
emed | -0.007 |
creen | -0.006 |
feature | -0.006 |
Token | of |
Token ID | 286 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
gements | +2.770 |
attest | +2.710 |
emed | +2.586 |
feature | +2.541 |
respectively | +2.533 |
Neg logit contributions | |
disrupting | -2.419 |
disruption | -2.291 |
obstruction | -2.250 |
enforcing | -2.211 |
interference | -2.169 |
Token | two |
Token ID | 734 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
gements | +0.605 |
attest | +0.574 |
emed | +0.562 |
creen | +0.551 |
Blossom | +0.543 |
Neg logit contributions | |
disrupting | -0.612 |
disruption | -0.579 |
obstruction | -0.567 |
enforcing | -0.563 |
interference | -0.555 |
Token | noun |
Token ID | 23227 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
gements | +2.023 |
attest | +1.932 |
emed | +1.868 |
creen | +1.824 |
Blossom | +1.805 |
Neg logit contributions | |
disrupting | -2.281 |
disruption | -2.163 |
obstruction | -2.133 |
enforcing | -2.108 |
interference | -2.082 |
Token | s |
Token ID | 82 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
gements | +3.193 |
attest | +3.047 |
emed | +2.778 |
feature | +2.651 |
respectively | +2.604 |
Neg logit contributions | |
disrupting | -8.233 |
disruption | -7.951 |
obstruction | -7.856 |
enforcing | -7.788 |
interference | -7.703 |
Token | or |
Token ID | 393 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
gements | +2.843 |
attest | +2.728 |
emed | +2.615 |
Blossom | +2.547 |
creen | +2.525 |
Neg logit contributions | |
disrupting | -2.819 |
disruption | -2.640 |
obstruction | -2.605 |
enforcing | -2.597 |
interference | -2.538 |
Token | adject |
Token ID | 31129 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
gements | +2.187 |
attest | +2.114 |
emed | +2.066 |
Blossom | +2.012 |
creen | +2.009 |
Neg logit contributions | |
disrupting | -2.356 |
disruption | -2.227 |
obstruction | -2.203 |
enforcing | -2.173 |
interference | -2.143 |
Token | ives |
Token ID | 1083 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
gements | +6.051 |
attest | +5.663 |
otes | +5.570 |
emed | +5.337 |
respectively | +5.276 |
Neg logit contributions | |
disrupting | -20.746 |
disruption | -20.358 |
obstruction | -19.867 |
inventoryQuantity | -19.756 |
interference | -19.401 |
Token | , |
Token ID | 11 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
gements | +7.898 |
attest | +7.874 |
respectively | +7.517 |
emed | +7.422 |
feature | +7.383 |
Neg logit contributions | |
disrupting | -8.317 |
disruption | -8.108 |
obstruction | -7.936 |
HUD | -7.828 |
enforcing | -7.672 |
Token | representing |
Token ID | 10200 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
gements | +2.771 |
attest | +2.664 |
emed | +2.574 |
creen | +2.493 |
Blossom | +2.491 |
Neg logit contributions | |
disrupting | -2.752 |
disruption | -2.586 |
obstruction | -2.546 |
enforcing | -2.537 |
interference | -2.480 |
Token | qualities |
Token ID | 14482 |
Feature activation | -8.40e-01 |
Pos logit contributions | |
gements | +2.984 |
attest | +2.869 |
emed | +2.783 |
creen | +2.712 |
Blossom | +2.695 |
Neg logit contributions | |
disrupting | -2.740 |
disruption | -2.596 |
obstruction | -2.546 |
enforcing | -2.519 |
interference | -2.483 |
Token | such |
Token ID | 884 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
gements | +3.561 |
attest | +3.448 |
emed | +3.303 |
Blossom | +3.200 |
feature | +3.193 |
Neg logit contributions | |
disrupting | -3.473 |
disruption | -3.276 |
obstruction | -3.229 |
enforcing | -3.202 |
interference | -3.142 |
Token | ing |
Token ID | 278 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
gements | +2.624 |
attest | +2.521 |
emed | +2.390 |
creen | +2.286 |
Blossom | +2.274 |
Neg logit contributions | |
disrupting | -3.478 |
disruption | -3.343 |
obstruction | -3.269 |
enforcing | -3.234 |
thrott | -3.193 |
Token | house |
Token ID | 4803 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
gements | +3.375 |
attest | +3.276 |
emed | +3.142 |
Blossom | +3.062 |
feature | +3.029 |
Neg logit contributions | |
disrupting | -3.066 |
disruption | -2.906 |
obstruction | -2.858 |
enforcing | -2.773 |
interference | -2.769 |
Token | reactors |
Token ID | 28502 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
gements | +4.812 |
attest | +4.710 |
emed | +4.513 |
feature | +4.421 |
Blossom | +4.392 |
Neg logit contributions | |
disrupting | -3.633 |
disruption | -3.416 |
obstruction | -3.351 |
enforcing | -3.303 |
interference | -3.238 |
Token | . |
Token ID | 13 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
gements | +0.165 |
attest | +0.157 |
emed | +0.152 |
Blossom | +0.147 |
feature | +0.145 |
Neg logit contributions | |
disrupting | -0.159 |
disruption | -0.149 |
enforcing | -0.146 |
obstruction | -0.146 |
thrott | -0.142 |
Token | The |
Token ID | 383 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
gements | +1.543 |
attest | +1.496 |
emed | +1.442 |
creen | +1.392 |
Blossom | +1.388 |
Neg logit contributions | |
disrupting | -1.280 |
disruption | -1.211 |
obstruction | -1.184 |
enforcing | -1.167 |
interference | -1.146 |
Token | Gerald |
Token ID | 27027 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
gements | +0.628 |
attest | +0.605 |
emed | +0.588 |
creen | +0.566 |
Blossom | +0.562 |
Neg logit contributions | |
disrupting | -0.524 |
disruption | -0.502 |
obstruction | -0.488 |
enforcing | -0.480 |
interference | -0.472 |
Token | Ford |
Token ID | 8092 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
feature | +12.006 |
attest | +11.986 |
gements | +11.803 |
emed | +11.476 |
pair | +11.405 |
Neg logit contributions | |
disrupting | -12.713 |
HUD | -12.507 |
activity | -12.227 |
disruption | -12.048 |
obstruction | -11.926 |
Token | - |
Token ID | 12 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
gements | +0.074 |
attest | +0.071 |
emed | +0.069 |
creen | +0.067 |
Blossom | +0.065 |
Neg logit contributions | |
disrupting | -0.059 |
disruption | -0.057 |
obstruction | -0.055 |
enforcing | -0.054 |
interference | -0.053 |
Token | class |
Token ID | 4871 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
gements | +1.664 |
attest | +1.616 |
emed | +1.519 |
creen | +1.465 |
Blossom | +1.460 |
Neg logit contributions | |
disrupting | -1.960 |
disruption | -1.872 |
obstruction | -1.839 |
enforcing | -1.815 |
HUD | -1.793 |
Token | ( |
Token ID | 357 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
gements | +1.070 |
attest | +1.028 |
emed | +0.996 |
creen | +0.966 |
Blossom | +0.956 |
Neg logit contributions | |
disrupting | -0.732 |
disruption | -0.695 |
obstruction | -0.676 |
enforcing | -0.662 |
interference | -0.651 |
Token | CV |
Token ID | 33538 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
gements | +2.497 |
attest | +2.444 |
emed | +2.324 |
feature | +2.261 |
Blossom | +2.255 |
Neg logit contributions | |
disrupting | -2.153 |
disruption | -2.040 |
obstruction | -2.003 |
enforcing | -1.969 |
HUD | -1.944 |
Token | A |
Token ID | 32 |
Feature activation | -7.41e-01 |
Pos logit contributions | |
gements | +3.039 |
attest | +2.976 |
emed | +2.817 |
feature | +2.753 |
Blossom | +2.751 |
Neg logit contributions | |
disrupting | -3.198 |
disruption | -3.043 |
obstruction | -2.986 |
enforcing | -2.948 |
HUD | -2.911 |
Token | ward |
Token ID | 904 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
gements | +3.623 |
attest | +3.545 |
emed | +3.408 |
Blossom | +3.315 |
respectively | +3.296 |
Neg logit contributions | |
disrupting | -2.580 |
disruption | -2.433 |
obstruction | -2.355 |
enforcing | -2.330 |
interference | -2.282 |
Token | ed |
Token ID | 276 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
gements | +9.419 |
attest | +9.082 |
emed | +8.621 |
feature | +8.466 |
respectively | +8.260 |
Neg logit contributions | |
disrupting | -11.175 |
disruption | -10.490 |
obstruction | -10.482 |
HUD | -10.298 |
enforcing | -10.257 |
Token | to |
Token ID | 284 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
gements | +7.371 |
attest | +7.106 |
emed | +6.745 |
feature | +6.659 |
respectively | +6.620 |
Neg logit contributions | |
disrupting | -7.702 |
disruption | -7.300 |
obstruction | -7.206 |
enforcing | -7.109 |
HUD | -7.005 |
Token | 79 |
Token ID | 9225 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
gements | +2.182 |
attest | +2.112 |
emed | +2.038 |
creen | +1.949 |
Blossom | +1.949 |
Neg logit contributions | |
disrupting | -2.322 |
disruption | -2.212 |
obstruction | -2.172 |
enforcing | -2.142 |
interference | -2.107 |
Token | Laure |
Token ID | 13826 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
disrupting | +2.620 |
disruption | +2.500 |
obstruction | +2.459 |
enforcing | +2.428 |
interference | +2.378 |
Neg logit contributions | |
gements | -2.096 |
attest | -2.033 |
emed | -1.939 |
creen | -1.919 |
Blossom | -1.874 |
Token | ates |
Token ID | 689 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
gements | +9.692 |
emed | +8.290 |
otes | +7.861 |
attest | +7.614 |
ingly | +7.470 |
Neg logit contributions | |
obstruction | -17.851 |
disrupting | -17.848 |
inventoryQuantity | -17.624 |
disruption | -17.332 |
enforcing | -16.893 |
Token | since |
Token ID | 1201 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
gements | +10.844 |
attest | +10.590 |
emed | +10.137 |
feature | +10.110 |
respectively | +10.084 |
Neg logit contributions | |
disrupting | -10.029 |
disruption | -9.681 |
obstruction | -9.558 |
HUD | -9.464 |
interference | -9.300 |
Token | 1969 |
Token ID | 16450 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
gements | +0.087 |
attest | +0.085 |
emed | +0.083 |
creen | +0.080 |
Blossom | +0.078 |
Neg logit contributions | |
disrupting | -0.088 |
disruption | -0.084 |
obstruction | -0.081 |
enforcing | -0.081 |
interference | -0.079 |
Token |
|
Token ID | 198 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
gements | +1.234 |
attest | +1.199 |
emed | +1.160 |
Blossom | +1.115 |
creen | +1.115 |
Neg logit contributions | |
disrupting | -1.045 |
disruption | -0.982 |
enforcing | -0.963 |
obstruction | -0.960 |
interference | -0.935 |
Token |
|
Token ID | 198 |
Feature activation | -8.60e-01 |
Pos logit contributions | |
gements | +1.764 |
attest | +1.723 |
emed | +1.656 |
feature | +1.624 |
Blossom | +1.609 |
Neg logit contributions | |
disrupting | -1.500 |
disruption | -1.415 |
obstruction | -1.389 |
enforcing | -1.368 |
interference | -1.343 |
Token | to |
Token ID | 284 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
disrupting | +1.296 |
disruption | +1.239 |
obstruction | +1.219 |
enforcing | +1.216 |
interference | +1.207 |
Neg logit contributions | |
gements | -1.372 |
attest | -1.321 |
emed | -1.268 |
Blossom | -1.253 |
feature | -1.232 |
Token | funding |
Token ID | 4918 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
gements | +1.937 |
attest | +1.857 |
emed | +1.805 |
Blossom | +1.739 |
creen | +1.735 |
Neg logit contributions | |
disrupting | -1.432 |
disruption | -1.355 |
obstruction | -1.315 |
enforcing | -1.315 |
interference | -1.281 |
Token | , |
Token ID | 11 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
gements | +0.435 |
attest | +0.418 |
emed | +0.405 |
Blossom | +0.390 |
creen | +0.386 |
Neg logit contributions | |
disrupting | -0.413 |
disruption | -0.394 |
obstruction | -0.386 |
enforcing | -0.383 |
interference | -0.379 |
Token | requiring |
Token ID | 10616 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
disrupting | +0.155 |
disruption | +0.145 |
enforcing | +0.144 |
obstruction | +0.142 |
interference | +0.139 |
Neg logit contributions | |
gements | -0.184 |
attest | -0.174 |
emed | -0.173 |
creen | -0.164 |
Blossom | -0.163 |
Token | F |
Token ID | 376 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
gements | +2.284 |
attest | +2.180 |
emed | +2.142 |
creen | +2.069 |
Blossom | +2.058 |
Neg logit contributions | |
disrupting | -1.796 |
disruption | -1.701 |
obstruction | -1.668 |
enforcing | -1.650 |
interference | -1.618 |
Token | annie |
Token ID | 42883 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
gements | +0.122 |
attest | +0.118 |
Blossom | +0.113 |
emed | +0.113 |
creen | +0.112 |
Neg logit contributions | |
disrupting | -0.089 |
disruption | -0.084 |
obstruction | -0.082 |
HUD | -0.081 |
enforcing | -0.081 |
Token | Mae |
Token ID | 34673 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
gements | +10.064 |
attest | +9.626 |
feature | +9.253 |
respectively | +9.177 |
berry | +8.914 |
Neg logit contributions | |
disrupting | -15.634 |
obstruction | -15.266 |
disruption | -15.117 |
enforcing | -14.603 |
punishing | -14.445 |
Token | and |
Token ID | 290 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
gements | +1.398 |
attest | +1.346 |
emed | +1.318 |
creen | +1.259 |
Blossom | +1.248 |
Neg logit contributions | |
disrupting | -1.344 |
disruption | -1.282 |
obstruction | -1.251 |
enforcing | -1.245 |
interference | -1.225 |
Token | Freddie |
Token ID | 31658 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
gements | +3.426 |
attest | +3.323 |
emed | +3.152 |
feature | +3.051 |
Blossom | +3.031 |
Neg logit contributions | |
disrupting | -4.277 |
disruption | -4.080 |
obstruction | -4.021 |
enforcing | -3.977 |
interference | -3.916 |
Token | Mac |
Token ID | 4100 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
gements | +5.915 |
attest | +5.831 |
feature | +5.669 |
Blossom | +5.606 |
respectively | +5.582 |
Neg logit contributions | |
disrupting | -10.670 |
obstruction | -10.328 |
disruption | -10.280 |
enforcing | -10.189 |
punishing | -9.978 |
Token | to |
Token ID | 284 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
disrupting | +1.032 |
disruption | +0.992 |
enforcing | +0.976 |
obstruction | +0.969 |
thrott | +0.950 |
Neg logit contributions | |
gements | -1.133 |
emed | -1.077 |
attest | -1.065 |
creen | -1.021 |
Blossom | -1.012 |
Token | ad |
Token ID | 324 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
disrupting | +2.640 |
disruption | +2.513 |
interference | +2.463 |
enforcing | +2.421 |
obstruction | +2.410 |
Neg logit contributions | |
gements | -3.388 |
attest | -3.299 |
emed | -3.155 |
feature | -3.091 |
respectively | -3.062 |
Token | amia |
Token ID | 49442 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
gements | +2.755 |
attest | +2.718 |
feature | +2.582 |
creen | +2.576 |
emed | +2.572 |
Neg logit contributions | |
disrupting | -1.479 |
disruption | -1.373 |
obstruction | -1.333 |
enforcing | -1.316 |
interference | -1.298 |
Token | N |
Token ID | 399 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
gements | +2.365 |
attest | +2.302 |
emed | +2.202 |
creen | +2.165 |
feature | +2.146 |
Neg logit contributions | |
disrupting | -1.925 |
disruption | -1.810 |
obstruction | -1.768 |
enforcing | -1.764 |
interference | -1.728 |
Token | uts |
Token ID | 5500 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
gements | +1.147 |
attest | +1.123 |
feature | +1.052 |
respectively | +1.046 |
emed | +1.043 |
Neg logit contributions | |
disrupting | -1.016 |
disruption | -0.960 |
obstruction | -0.935 |
enforcing | -0.935 |
interference | -0.919 |
Token | , |
Token ID | 837 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
gements | +4.381 |
attest | +4.280 |
emed | +4.093 |
feature | +3.986 |
Blossom | +3.957 |
Neg logit contributions | |
disrupting | -3.748 |
disruption | -3.539 |
obstruction | -3.476 |
enforcing | -3.431 |
interference | -3.371 |
Token | Hazel |
Token ID | 42805 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
gements | +0.656 |
attest | +0.633 |
emed | +0.611 |
creen | +0.591 |
feature | +0.587 |
Neg logit contributions | |
disrupting | -0.605 |
disruption | -0.572 |
obstruction | -0.556 |
enforcing | -0.553 |
interference | -0.548 |
Token | nuts |
Token ID | 31381 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
gements | +9.087 |
berry | +8.725 |
emed | +8.445 |
attest | +8.278 |
Blossom | +8.244 |
Neg logit contributions | |
disrupting | -17.147 |
obstruction | -16.655 |
disruption | -16.577 |
HUD | -16.245 |
enforcing | -16.084 |
Token | ( |
Token ID | 357 |
Feature activation | -8.20e-01 |
Pos logit contributions | |
gements | +3.053 |
attest | +2.981 |
emed | +2.844 |
feature | +2.764 |
creen | +2.735 |
Neg logit contributions | |
disrupting | -2.572 |
disruption | -2.422 |
obstruction | -2.379 |
enforcing | -2.353 |
interference | -2.313 |
Token | Fil |
Token ID | 11928 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
gements | +3.195 |
attest | +3.111 |
emed | +2.935 |
feature | +2.849 |
Blossom | +2.833 |
Neg logit contributions | |
disrupting | -3.659 |
disruption | -3.482 |
obstruction | -3.420 |
enforcing | -3.393 |
interference | -3.346 |
Token | ber |
Token ID | 527 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
gements | +4.560 |
attest | +4.431 |
emed | +4.199 |
feature | +4.075 |
Blossom | +4.045 |
Neg logit contributions | |
disrupting | -5.597 |
disruption | -5.340 |
obstruction | -5.259 |
enforcing | -5.198 |
interference | -5.127 |
Token | ts |
Token ID | 912 |
Feature activation | -9.31e-01 |
Pos logit contributions | |
disrupting | +0.733 |
disruption | +0.692 |
enforcing | +0.679 |
obstruction | +0.679 |
interference | +0.669 |
Neg logit contributions | |
gements | -0.745 |
attest | -0.725 |
emed | -0.698 |
creen | -0.672 |
Blossom | -0.661 |
Token | En |
Token ID | 2039 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
disrupting | +1.746 |
disruption | +1.659 |
enforcing | +1.620 |
obstruction | +1.614 |
interference | +1.592 |
Neg logit contributions | |
gements | -2.234 |
attest | -2.140 |
emed | -2.102 |
creen | -2.052 |
Blossom | -2.004 |
Token | bridge |
Token ID | 9458 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
disrupting | +0.364 |
disruption | +0.350 |
obstruction | +0.344 |
enforcing | +0.338 |
interference | +0.336 |
Neg logit contributions | |
gements | -0.240 |
attest | -0.236 |
emed | -0.214 |
Blossom | -0.210 |
feature | -0.210 |
Token | � |
Token ID | 447 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
disrupting | +5.878 |
disruption | +5.481 |
obstruction | +5.268 |
enforcing | +5.265 |
interference | +5.218 |
Neg logit contributions | |
gements | -8.427 |
emed | -8.117 |
attest | -8.088 |
creen | -7.980 |
Blossom | -7.690 |
Token | � |
Token ID | 247 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
disrupting | +2.807 |
obstruction | +2.706 |
disruption | +2.692 |
HUD | +2.668 |
enforcing | +2.661 |
Neg logit contributions | |
gements | -2.025 |
attest | -2.007 |
emed | -1.875 |
creen | -1.818 |
respectively | -1.798 |
Token | s |
Token ID | 82 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
gements | +0.185 |
attest | +0.177 |
emed | +0.161 |
feature | +0.151 |
Blossom | +0.151 |
Neg logit contributions | |
disrupting | -0.489 |
disruption | -0.473 |
obstruction | -0.467 |
enforcing | -0.464 |
interference | -0.458 |
Token | Line |
Token ID | 6910 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
disrupting | +4.106 |
disruption | +3.972 |
obstruction | +3.850 |
interference | +3.783 |
enforcing | +3.762 |
Neg logit contributions | |
gements | -5.051 |
emed | -4.717 |
attest | -4.715 |
creen | -4.576 |
Blossom | -4.427 |
Token | 9 |
Token ID | 860 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
disrupting | +6.704 |
disruption | +6.663 |
obstruction | +6.382 |
interference | +6.276 |
enforcing | +6.100 |
Neg logit contributions | |
gements | -12.083 |
attest | -11.286 |
emed | -11.124 |
creen | -10.977 |
Blossom | -10.636 |
Token | pipeline |
Token ID | 11523 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
disrupting | +4.839 |
disruption | +4.679 |
obstruction | +4.525 |
interference | +4.465 |
enforcing | +4.374 |
Neg logit contributions | |
gements | -6.395 |
emed | -5.979 |
attest | -5.963 |
creen | -5.813 |
Blossom | -5.684 |
Token | was |
Token ID | 373 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
disrupting | +7.805 |
disruption | +7.550 |
obstruction | +7.276 |
interference | +7.266 |
enforcing | +7.118 |
Neg logit contributions | |
gements | -8.795 |
emed | -8.152 |
creen | -7.781 |
attest | -7.715 |
Blossom | -7.701 |
Token | � |
Token ID | 564 |
Feature activation | +7.29e-01 |
Pos logit contributions | |
disrupting | +3.521 |
obstruction | +3.297 |
disruption | +3.272 |
interference | +3.200 |
enforcing | +3.192 |
Neg logit contributions | |
gements | -5.160 |
attest | -4.869 |
emed | -4.778 |
Blossom | -4.684 |
creen | -4.611 |
Token | � |
Token ID | 250 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
disrupting | +3.162 |
obstruction | +3.028 |
disruption | +3.012 |
enforcing | +2.991 |
HUD | +2.954 |
Neg logit contributions | |
gements | -2.788 |
attest | -2.732 |
emed | -2.529 |
creen | -2.526 |
feature | -2.475 |
Token | , |
Token ID | 11 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
disrupting | +0.984 |
disruption | +0.914 |
enforcing | +0.905 |
obstruction | +0.903 |
interference | +0.888 |
Neg logit contributions | |
gements | -1.106 |
attest | -1.053 |
emed | -1.001 |
Blossom | -0.967 |
creen | -0.957 |
Token | but |
Token ID | 475 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
disrupting | +1.773 |
disruption | +1.632 |
obstruction | +1.618 |
enforcing | +1.606 |
interference | +1.569 |
Neg logit contributions | |
gements | -2.162 |
attest | -2.063 |
emed | -2.031 |
creen | -1.950 |
Blossom | -1.921 |
Token | there |
Token ID | 612 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
disrupting | +0.044 |
disruption | +0.042 |
obstruction | +0.041 |
enforcing | +0.040 |
interference | +0.040 |
Neg logit contributions | |
gements | -0.050 |
attest | -0.047 |
emed | -0.045 |
creen | -0.044 |
feature | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
disrupting | +1.768 |
disruption | +1.678 |
obstruction | +1.654 |
enforcing | +1.627 |
interference | +1.596 |
Neg logit contributions | |
gements | -2.103 |
attest | -1.957 |
emed | -1.853 |
creen | -1.847 |
Blossom | -1.813 |
Token | no |
Token ID | 645 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
disrupting | +0.925 |
disruption | +0.898 |
obstruction | +0.877 |
enforcing | +0.858 |
interference | +0.857 |
Neg logit contributions | |
gements | -0.931 |
attest | -0.852 |
emed | -0.835 |
creen | -0.805 |
respectively | -0.777 |
Token | wind |
Token ID | 2344 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
disrupting | +3.342 |
disruption | +3.258 |
obstruction | +3.220 |
interference | +3.107 |
enforcing | +3.086 |
Neg logit contributions | |
gements | -3.564 |
attest | -3.263 |
emed | -3.218 |
creen | -3.120 |
respectively | -3.029 |
Token | , |
Token ID | 11 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
disrupting | +10.049 |
disruption | +9.268 |
interference | +9.058 |
obstruction | +8.967 |
interfering | +8.787 |
Neg logit contributions | |
gements | -12.558 |
attest | -11.806 |
emed | -11.407 |
Blossom | -11.103 |
otes | -10.787 |
Token | which |
Token ID | 543 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
disrupting | +4.040 |
disruption | +3.839 |
obstruction | +3.785 |
interference | +3.721 |
enforcing | +3.633 |
Neg logit contributions | |
gements | -4.466 |
attest | -4.206 |
emed | -4.186 |
creen | -3.988 |
Blossom | -3.925 |
Token | meant |
Token ID | 4001 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
gements | +3.324 |
attest | +3.209 |
emed | +3.066 |
feature | +2.969 |
Blossom | +2.962 |
Neg logit contributions | |
disrupting | -3.419 |
disruption | -3.243 |
obstruction | -3.197 |
enforcing | -3.148 |
interference | -3.105 |
Token | good |
Token ID | 922 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
disrupting | +0.380 |
disruption | +0.363 |
obstruction | +0.355 |
enforcing | +0.347 |
interference | +0.343 |
Neg logit contributions | |
gements | -0.499 |
attest | -0.479 |
emed | -0.466 |
creen | -0.451 |
Blossom | -0.442 |
Token | launch |
Token ID | 4219 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
gements | +1.795 |
attest | +1.728 |
emed | +1.669 |
creen | +1.623 |
Blossom | +1.604 |
Neg logit contributions | |
disrupting | -1.275 |
disruption | -1.210 |
obstruction | -1.194 |
enforcing | -1.164 |
interference | -1.154 |
Token | % |
Token ID | 4 |
Feature activation | +7.61e-01 |
Pos logit contributions | |
disrupting | +6.155 |
disruption | +5.887 |
enforcing | +5.875 |
obstruction | +5.819 |
interference | +5.783 |
Neg logit contributions | |
gements | -5.011 |
attest | -4.673 |
creen | -4.597 |
emed | -4.512 |
Blossom | -4.333 |
Token | copper |
Token ID | 15317 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
disrupting | +3.118 |
disruption | +3.008 |
obstruction | +2.991 |
interference | +2.942 |
enforcing | +2.930 |
Neg logit contributions | |
gements | -3.162 |
attest | -2.878 |
emed | -2.870 |
creen | -2.777 |
Blossom | -2.730 |
Token | with |
Token ID | 351 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
disrupting | +2.767 |
disruption | +2.662 |
obstruction | +2.607 |
interference | +2.584 |
enforcing | +2.558 |
Neg logit contributions | |
gements | -2.685 |
attest | -2.518 |
emed | -2.454 |
creen | -2.340 |
Blossom | -2.333 |
Token | the |
Token ID | 262 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
disrupting | +0.500 |
disruption | +0.474 |
obstruction | +0.467 |
enforcing | +0.461 |
interference | +0.458 |
Neg logit contributions | |
gements | -0.587 |
attest | -0.562 |
emed | -0.547 |
creen | -0.528 |
Blossom | -0.521 |
Token | remainder |
Token ID | 17675 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
disrupting | +0.877 |
disruption | +0.837 |
obstruction | +0.830 |
interference | +0.810 |
enforcing | +0.809 |
Neg logit contributions | |
gements | -1.043 |
attest | -0.994 |
emed | -0.977 |
creen | -0.941 |
Blossom | -0.920 |
Token | tin |
Token ID | 19783 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
disrupting | +1.086 |
disruption | +1.011 |
enforcing | +1.000 |
obstruction | +0.999 |
interference | +0.974 |
Neg logit contributions | |
gements | -1.313 |
attest | -1.240 |
emed | -1.199 |
creen | -1.153 |
Blossom | -1.152 |
Token | and |
Token ID | 290 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
disrupting | +8.039 |
disruption | +7.583 |
obstruction | +7.477 |
interference | +7.436 |
interfering | +7.399 |
Neg logit contributions | |
gements | -7.888 |
attest | -7.194 |
emed | -6.793 |
Blossom | -6.769 |
Gift | -6.695 |
Token | zinc |
Token ID | 31861 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
disrupting | +0.797 |
disruption | +0.753 |
obstruction | +0.742 |
interference | +0.731 |
enforcing | +0.731 |
Neg logit contributions | |
gements | -0.847 |
attest | -0.789 |
emed | -0.758 |
Blossom | -0.733 |
creen | -0.733 |
Token | ) |
Token ID | 8 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
disrupting | +0.925 |
disruption | +0.874 |
obstruction | +0.867 |
enforcing | +0.861 |
interference | +0.855 |
Neg logit contributions | |
gements | -0.812 |
attest | -0.758 |
emed | -0.739 |
Blossom | -0.720 |
creen | -0.706 |
Token | as |
Token ID | 355 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
gements | +1.430 |
attest | +1.365 |
emed | +1.328 |
Blossom | +1.282 |
creen | +1.273 |
Neg logit contributions | |
disrupting | -1.296 |
disruption | -1.226 |
obstruction | -1.204 |
enforcing | -1.196 |
interference | -1.186 |
Token | a |
Token ID | 257 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
gements | +3.200 |
attest | +3.094 |
emed | +2.977 |
Blossom | +2.887 |
feature | +2.875 |
Neg logit contributions | |
disrupting | -2.840 |
disruption | -2.686 |
obstruction | -2.644 |
enforcing | -2.609 |
interference | -2.568 |
Token | Georgia |
Token ID | 7859 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
disrupting | +3.980 |
disruption | +3.742 |
obstruction | +3.687 |
enforcing | +3.647 |
interference | +3.556 |
Neg logit contributions | |
gements | -5.070 |
attest | -4.825 |
emed | -4.824 |
creen | -4.548 |
Blossom | -4.508 |
Token | , |
Token ID | 11 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
gements | +0.899 |
attest | +0.863 |
emed | +0.829 |
Blossom | +0.801 |
creen | +0.800 |
Neg logit contributions | |
disrupting | -0.869 |
disruption | -0.824 |
obstruction | -0.815 |
enforcing | -0.799 |
interference | -0.795 |
Token | where |
Token ID | 810 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
disrupting | +0.634 |
disruption | +0.592 |
obstruction | +0.588 |
enforcing | +0.585 |
interference | +0.569 |
Neg logit contributions | |
gements | -0.695 |
attest | -0.663 |
emed | -0.651 |
creen | -0.621 |
Blossom | -0.617 |
Token | Democrat |
Token ID | 9755 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
gements | +1.256 |
attest | +1.210 |
emed | +1.182 |
creen | +1.136 |
Blossom | +1.119 |
Neg logit contributions | |
disrupting | -1.041 |
disruption | -0.986 |
obstruction | -0.972 |
enforcing | -0.952 |
interference | -0.941 |
Token | Jon |
Token ID | 5966 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
gements | +1.425 |
attest | +1.374 |
emed | +1.339 |
creen | +1.286 |
feature | +1.272 |
Neg logit contributions | |
disrupting | -1.310 |
disruption | -1.243 |
obstruction | -1.215 |
enforcing | -1.199 |
interference | -1.195 |
Token | Os |
Token ID | 8834 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
gements | +1.677 |
attest | +1.643 |
emed | +1.587 |
feature | +1.541 |
creen | +1.531 |
Neg logit contributions | |
disrupting | -1.040 |
disruption | -0.975 |
obstruction | -0.946 |
enforcing | -0.933 |
interference | -0.924 |
Token | so |
Token ID | 568 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
disrupting | +4.073 |
disruption | +3.827 |
HUD | +3.670 |
enforcing | +3.447 |
interference | +3.278 |
Neg logit contributions | |
gements | -15.434 |
attest | -14.650 |
emed | -14.619 |
Blossom | -14.282 |
creen | -14.004 |
Token | ff |
Token ID | 487 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
disrupting | +4.442 |
disruption | +4.267 |
enforcing | +4.229 |
obstruction | +4.197 |
HUD | +4.181 |
Neg logit contributions | |
gements | -8.192 |
attest | -8.044 |
Blossom | -7.779 |
maid | -7.756 |
creen | -7.621 |
Token | narrowly |
Token ID | 26167 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
disrupting | +2.657 |
obstruction | +2.518 |
disruption | +2.501 |
thrott | +2.453 |
enforcing | +2.446 |
Neg logit contributions | |
gements | -2.721 |
attest | -2.674 |
emed | -2.571 |
Blossom | -2.508 |
creen | -2.461 |
Token | missed |
Token ID | 6825 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
gements | +2.451 |
attest | +2.398 |
emed | +2.292 |
Blossom | +2.211 |
feature | +2.210 |
Neg logit contributions | |
disrupting | -1.932 |
disruption | -1.813 |
obstruction | -1.801 |
enforcing | -1.759 |
thrott | -1.735 |
Token | winning |
Token ID | 5442 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
disrupting | +0.979 |
disruption | +0.905 |
obstruction | +0.890 |
enforcing | +0.885 |
interference | +0.859 |
Neg logit contributions | |
gements | -1.473 |
attest | -1.420 |
emed | -1.387 |
creen | -1.333 |
Blossom | -1.319 |
Token | who |
Token ID | 508 |
Feature activation | -9.41e-01 |
Pos logit contributions | |
gements | +2.602 |
attest | +2.491 |
emed | +2.427 |
Blossom | +2.332 |
feature | +2.311 |
Neg logit contributions | |
disrupting | -2.243 |
disruption | -2.113 |
obstruction | -2.058 |
enforcing | -2.044 |
interference | -2.005 |
Token | appeared |
Token ID | 4120 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
gements | +3.949 |
attest | +3.775 |
emed | +3.654 |
Blossom | +3.525 |
feature | +3.492 |
Neg logit contributions | |
disrupting | -3.963 |
disruption | -3.764 |
obstruction | -3.680 |
enforcing | -3.629 |
interference | -3.589 |
Token | onstage |
Token ID | 38500 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
gements | +2.899 |
attest | +2.794 |
emed | +2.703 |
Blossom | +2.616 |
feature | +2.596 |
Neg logit contributions | |
disrupting | -2.507 |
disruption | -2.363 |
obstruction | -2.311 |
enforcing | -2.284 |
interference | -2.244 |
Token | weren |
Token ID | 6304 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
gements | +0.814 |
attest | +0.764 |
emed | +0.751 |
Blossom | +0.720 |
creen | +0.710 |
Neg logit contributions | |
disrupting | -0.820 |
disruption | -0.773 |
obstruction | -0.755 |
enforcing | -0.750 |
interference | -0.735 |
Token | 't |
Token ID | 470 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
gements | +1.440 |
attest | +1.375 |
emed | +1.273 |
feature | +1.217 |
Blossom | +1.199 |
Neg logit contributions | |
disrupting | -3.638 |
disruption | -3.510 |
obstruction | -3.464 |
enforcing | -3.430 |
interference | -3.392 |
Token | Apple |
Token ID | 4196 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
disrupting | +0.247 |
disruption | +0.230 |
obstruction | +0.225 |
enforcing | +0.224 |
interference | +0.218 |
Neg logit contributions | |
gements | -0.292 |
attest | -0.281 |
emed | -0.273 |
creen | -0.262 |
Blossom | -0.262 |
Token | representatives |
Token ID | 10826 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
disrupting | +8.020 |
disruption | +7.692 |
obstruction | +7.421 |
interference | +7.297 |
enforcing | +7.184 |
Neg logit contributions | |
gements | -8.081 |
emed | -7.863 |
attest | -7.677 |
creen | -7.264 |
faced | -7.140 |
Token | . |
Token ID | 13 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
gements | +2.505 |
attest | +2.427 |
emed | +2.326 |
Blossom | +2.238 |
feature | +2.236 |
Neg logit contributions | |
disrupting | -2.586 |
disruption | -2.431 |
obstruction | -2.390 |
enforcing | -2.389 |
interference | -2.332 |
Token | An |
Token ID | 1052 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
disrupting | +0.914 |
disruption | +0.859 |
enforcing | +0.837 |
obstruction | +0.832 |
interference | +0.824 |
Neg logit contributions | |
gements | -1.238 |
attest | -1.203 |
emed | -1.156 |
creen | -1.106 |
feature | -1.094 |
Token | Apple |
Token ID | 4196 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
disrupting | +0.473 |
disruption | +0.452 |
obstruction | +0.449 |
interference | +0.441 |
enforcing | +0.438 |
Neg logit contributions | |
gements | -0.443 |
attest | -0.425 |
emed | -0.413 |
creen | -0.399 |
Blossom | -0.392 |
Token | spokesperson |
Token ID | 11821 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
disrupting | +4.077 |
disruption | +3.937 |
obstruction | +3.781 |
enforcing | +3.728 |
interference | +3.693 |
Neg logit contributions | |
gements | -7.449 |
attest | -6.983 |
emed | -6.972 |
creen | -6.630 |
otes | -6.552 |
Token | if |
Token ID | 611 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
disrupting | +0.643 |
disruption | +0.607 |
obstruction | +0.604 |
enforcing | +0.594 |
interference | +0.581 |
Neg logit contributions | |
gements | -0.682 |
attest | -0.655 |
emed | -0.635 |
Blossom | -0.616 |
creen | -0.610 |
Token | there |
Token ID | 612 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
gements | +0.953 |
attest | +0.911 |
emed | +0.882 |
Blossom | +0.853 |
creen | +0.850 |
Neg logit contributions | |
disrupting | -0.878 |
disruption | -0.835 |
obstruction | -0.831 |
enforcing | -0.811 |
interference | -0.800 |
Token | is |
Token ID | 318 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
disrupting | +2.105 |
disruption | +2.023 |
obstruction | +2.020 |
enforcing | +1.960 |
interference | +1.932 |
Neg logit contributions | |
gements | -2.211 |
attest | -2.094 |
emed | -2.031 |
creen | -1.996 |
Blossom | -1.995 |
Token | real |
Token ID | 1103 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
disrupting | +3.243 |
disruption | +3.220 |
obstruction | +3.146 |
interference | +3.057 |
enforcing | +2.944 |
Neg logit contributions | |
gements | -3.796 |
attest | -3.487 |
emed | -3.428 |
creen | -3.312 |
Blossom | -3.303 |
Token | evidence |
Token ID | 2370 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
disruption | +4.900 |
disrupting | +4.825 |
obstruction | +4.789 |
interference | +4.569 |
enforcing | +4.330 |
Neg logit contributions | |
gements | -6.050 |
emed | -5.560 |
attest | -5.487 |
Blossom | -5.255 |
creen | -5.244 |
Token | it |
Token ID | 340 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
disrupting | +4.504 |
obstruction | +4.192 |
disruption | +4.151 |
enforcing | +4.059 |
interference | +3.984 |
Neg logit contributions | |
gements | -6.711 |
emed | -6.269 |
attest | -6.165 |
creen | -6.127 |
Blossom | -6.112 |
Token | will |
Token ID | 481 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
disrupting | +1.437 |
disruption | +1.353 |
obstruction | +1.353 |
enforcing | +1.318 |
thrott | +1.310 |
Neg logit contributions | |
gements | -1.305 |
attest | -1.256 |
emed | -1.223 |
creen | -1.168 |
Blossom | -1.164 |
Token | cost |
Token ID | 1575 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
gements | +5.341 |
attest | +5.188 |
emed | +4.960 |
feature | +4.805 |
Blossom | +4.796 |
Neg logit contributions | |
disrupting | -5.379 |
disruption | -5.108 |
obstruction | -5.026 |
enforcing | -4.954 |
interference | -4.879 |
Token | them |
Token ID | 606 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
gements | +1.668 |
attest | +1.595 |
emed | +1.526 |
creen | +1.469 |
Blossom | +1.461 |
Neg logit contributions | |
disrupting | -2.191 |
disruption | -2.101 |
obstruction | -2.082 |
enforcing | -2.043 |
interference | -2.017 |
Token | their |
Token ID | 511 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
gements | +0.224 |
attest | +0.211 |
emed | +0.210 |
creen | +0.199 |
Blossom | +0.197 |
Neg logit contributions | |
disrupting | -0.202 |
disruption | -0.193 |
obstruction | -0.192 |
enforcing | -0.187 |
interference | -0.184 |
Token | jobs |
Token ID | 3946 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
gements | +2.274 |
attest | +2.149 |
emed | +2.105 |
creen | +2.022 |
Blossom | +1.991 |
Neg logit contributions | |
disrupting | -2.409 |
disruption | -2.302 |
obstruction | -2.280 |
enforcing | -2.237 |
interference | -2.185 |
Token | . |
Token ID | 13 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
gements | +0.550 |
attest | +0.527 |
emed | +0.509 |
Blossom | +0.495 |
creen | +0.487 |
Neg logit contributions | |
disrupting | -0.573 |
disruption | -0.540 |
enforcing | -0.532 |
obstruction | -0.531 |
interference | -0.516 |
Token |
|
Token ID | 198 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
gements | +1.279 |
attest | +1.239 |
emed | +1.197 |
creen | +1.152 |
feature | +1.148 |
Neg logit contributions | |
disrupting | -1.033 |
disruption | -0.975 |
obstruction | -0.957 |
enforcing | -0.951 |
HUD | -0.931 |
Token |
|
Token ID | 198 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
gements | +0.671 |
attest | +0.655 |
emed | +0.628 |
feature | +0.612 |
Blossom | +0.609 |
Neg logit contributions | |
disrupting | -0.535 |
disruption | -0.504 |
obstruction | -0.495 |
enforcing | -0.488 |
interference | -0.479 |
Token | Thus |
Token ID | 19093 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
gements | +1.248 |
attest | +1.224 |
emed | +1.154 |
Blossom | +1.133 |
feature | +1.127 |
Neg logit contributions | |
disrupting | -1.268 |
disruption | -1.206 |
HUD | -1.194 |
obstruction | -1.183 |
enforcing | -1.172 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
disrupting | +2.274 |
disruption | +2.141 |
obstruction | +2.081 |
enforcing | +2.079 |
interference | +2.039 |
Neg logit contributions | |
gements | -2.543 |
attest | -2.435 |
emed | -2.386 |
Blossom | -2.350 |
creen | -2.298 |
Token | the |
Token ID | 262 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
gements | +0.580 |
attest | +0.562 |
emed | +0.553 |
Blossom | +0.532 |
creen | +0.529 |
Neg logit contributions | |
disrupting | -0.526 |
disruption | -0.497 |
obstruction | -0.484 |
enforcing | -0.482 |
interference | -0.475 |
Token | occupation |
Token ID | 13755 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
disrupting | +0.904 |
disruption | +0.869 |
obstruction | +0.838 |
enforcing | +0.821 |
interference | +0.820 |
Neg logit contributions | |
gements | -1.078 |
attest | -1.029 |
emed | -1.022 |
Blossom | -0.982 |
creen | -0.979 |
Token | structure |
Token ID | 4645 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
disrupting | +3.343 |
disruption | +3.231 |
obstruction | +3.099 |
enforcing | +3.083 |
interference | +3.024 |
Neg logit contributions | |
gements | -3.892 |
attest | -3.738 |
emed | -3.679 |
Blossom | -3.609 |
creen | -3.576 |
Token | for |
Token ID | 329 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
disrupting | +3.558 |
disruption | +3.334 |
enforcing | +3.241 |
obstruction | +3.209 |
interference | +3.149 |
Neg logit contributions | |
gements | -4.639 |
attest | -4.457 |
emed | -4.420 |
Blossom | -4.284 |
creen | -4.229 |
Token | the |
Token ID | 262 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
gements | +1.324 |
attest | +1.282 |
emed | +1.246 |
Blossom | +1.206 |
creen | +1.197 |
Neg logit contributions | |
disrupting | -1.236 |
disruption | -1.172 |
obstruction | -1.139 |
enforcing | -1.138 |
interference | -1.110 |
Token | three |
Token ID | 1115 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
disrupting | +0.608 |
disruption | +0.579 |
obstruction | +0.559 |
enforcing | +0.553 |
interference | +0.543 |
Neg logit contributions | |
gements | -0.758 |
attest | -0.729 |
emed | -0.718 |
Blossom | -0.692 |
creen | -0.687 |
Token | keeps |
Token ID | 7622 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
gements | +5.872 |
attest | +5.718 |
emed | +5.421 |
feature | +5.324 |
Blossom | +5.238 |
Neg logit contributions | |
disrupting | -7.090 |
disruption | -6.784 |
obstruction | -6.662 |
enforcing | -6.551 |
interference | -6.495 |
Token | working |
Token ID | 1762 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
gements | +5.878 |
attest | +5.756 |
emed | +5.550 |
feature | +5.460 |
Blossom | +5.420 |
Neg logit contributions | |
disrupting | -3.394 |
disruption | -3.181 |
obstruction | -3.098 |
enforcing | -3.028 |
interference | -2.974 |
Token | .) |
Token ID | 2014 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
disrupting | +1.351 |
disruption | +1.282 |
obstruction | +1.277 |
enforcing | +1.273 |
interference | +1.259 |
Neg logit contributions | |
gements | -0.901 |
attest | -0.867 |
emed | -0.818 |
Blossom | -0.793 |
respectively | -0.764 |
Token | There |
Token ID | 1318 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
gements | +1.964 |
attest | +1.904 |
emed | +1.823 |
creen | +1.753 |
feature | +1.745 |
Neg logit contributions | |
disrupting | -1.727 |
disruption | -1.634 |
obstruction | -1.602 |
enforcing | -1.597 |
interference | -1.558 |
Token | was |
Token ID | 373 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
disrupting | +0.771 |
disruption | +0.740 |
obstruction | +0.736 |
enforcing | +0.732 |
interference | +0.714 |
Neg logit contributions | |
gements | -0.526 |
attest | -0.501 |
emed | -0.468 |
Blossom | -0.467 |
creen | -0.461 |
Token | some |
Token ID | 617 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
disrupting | +3.291 |
disruption | +3.203 |
obstruction | +3.096 |
interference | +3.086 |
enforcing | +3.043 |
Neg logit contributions | |
gements | -3.399 |
attest | -3.255 |
emed | -3.112 |
creen | -3.045 |
berry | -2.971 |
Token | # |
Token ID | 1303 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
disrupting | +7.550 |
disruption | +7.529 |
interference | +7.229 |
obstruction | +7.189 |
enforcing | +6.874 |
Neg logit contributions | |
gements | -8.035 |
attest | -7.298 |
emed | -7.181 |
creen | -6.934 |
otes | -6.860 |
Token | design |
Token ID | 26124 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
gements | +1.007 |
attest | +0.988 |
emed | +0.931 |
Blossom | +0.906 |
feature | +0.903 |
Neg logit contributions | |
disrupting | -0.971 |
disruption | -0.925 |
obstruction | -0.908 |
enforcing | -0.895 |
interference | -0.884 |
Token | space |
Token ID | 13200 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
gements | +0.976 |
attest | +0.958 |
emed | +0.910 |
Blossom | +0.883 |
creen | +0.881 |
Neg logit contributions | |
disrupting | -0.923 |
disruption | -0.882 |
obstruction | -0.858 |
enforcing | -0.845 |
interference | -0.842 |
Token | open |
Token ID | 1280 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
disrupting | +5.612 |
disruption | +5.441 |
interference | +5.203 |
obstruction | +5.184 |
enforcing | +5.091 |
Neg logit contributions | |
gements | -5.912 |
attest | -5.735 |
emed | -5.443 |
Blossom | -5.336 |
creen | -5.289 |
Token | to |
Token ID | 284 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
disrupting | +1.396 |
disruption | +1.331 |
enforcing | +1.302 |
obstruction | +1.298 |
interference | +1.277 |
Neg logit contributions | |
gements | -1.614 |
attest | -1.596 |
emed | -1.523 |
Blossom | -1.504 |
creen | -1.488 |
Token | enna |
Token ID | 13713 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
disrupting | +4.530 |
onz | +4.295 |
obstruction | +4.261 |
disruption | +4.253 |
activity | +4.220 |
Neg logit contributions | |
attest | -6.030 |
gements | -5.786 |
Blossom | -5.677 |
feature | -5.515 |
faced | -5.407 |
Token | some |
Token ID | 617 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
disrupting | +2.318 |
disruption | +2.259 |
obstruction | +2.173 |
interference | +2.150 |
enforcing | +2.112 |
Neg logit contributions | |
gements | -3.120 |
attest | -2.965 |
emed | -2.939 |
creen | -2.845 |
Blossom | -2.795 |
Token | time |
Token ID | 640 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
disrupting | +4.183 |
disruption | +4.138 |
interference | +3.960 |
obstruction | +3.898 |
enforcing | +3.816 |
Neg logit contributions | |
gements | -4.782 |
emed | -4.411 |
attest | -4.392 |
creen | -4.209 |
Blossom | -4.140 |
Token | in |
Token ID | 287 |
Feature activation | +5.79e-01 |
Pos logit contributions | |
disrupting | +5.365 |
disruption | +5.172 |
enforcing | +5.073 |
interference | +4.861 |
obstruction | +4.829 |
Neg logit contributions | |
gements | -7.716 |
attest | -7.250 |
emed | -7.218 |
Blossom | -7.034 |
creen | -6.808 |
Token | the |
Token ID | 262 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
disrupting | +2.181 |
disruption | +2.070 |
enforcing | +2.009 |
obstruction | +2.006 |
interference | +1.972 |
Neg logit contributions | |
gements | -2.686 |
attest | -2.547 |
emed | -2.457 |
creen | -2.417 |
Blossom | -2.353 |
Token | car |
Token ID | 1097 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
disrupting | +2.330 |
disruption | +2.245 |
obstruction | +2.173 |
enforcing | +2.133 |
thrott | +2.120 |
Neg logit contributions | |
gements | -2.590 |
attest | -2.445 |
emed | -2.395 |
creen | -2.340 |
faced | -2.273 |
Token | on |
Token ID | 319 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
disrupting | +5.912 |
disruption | +5.705 |
enforcing | +5.482 |
obstruction | +5.377 |
interference | +5.360 |
Neg logit contributions | |
gements | -7.624 |
attest | -7.206 |
emed | -7.060 |
creen | -6.629 |
Blossom | -6.619 |
Token | Sunday |
Token ID | 3502 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
disrupting | +1.627 |
disruption | +1.569 |
obstruction | +1.513 |
enforcing | +1.501 |
interference | +1.492 |
Neg logit contributions | |
gements | -1.609 |
attest | -1.536 |
emed | -1.485 |
creen | -1.410 |
Blossom | -1.403 |
Token | , |
Token ID | 11 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
gements | +5.649 |
attest | +5.530 |
emed | +5.220 |
feature | +5.185 |
respectively | +5.182 |
Neg logit contributions | |
disrupting | -6.469 |
disruption | -6.154 |
obstruction | -6.090 |
enforcing | -5.994 |
interference | -5.905 |
Token | they |
Token ID | 484 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
gements | +1.105 |
attest | +1.069 |
emed | +1.040 |
creen | +0.998 |
Blossom | +0.993 |
Neg logit contributions | |
disrupting | -1.033 |
disruption | -0.986 |
obstruction | -0.965 |
enforcing | -0.952 |
interference | -0.937 |
Token | made |
Token ID | 925 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
gements | +1.186 |
attest | +1.107 |
emed | +1.064 |
Blossom | +1.027 |
creen | +0.995 |
Neg logit contributions | |
disrupting | -1.849 |
disruption | -1.776 |
obstruction | -1.755 |
enforcing | -1.739 |
thrott | -1.707 |
Token | following |
Token ID | 1708 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
disrupting | +3.118 |
disruption | +2.911 |
obstruction | +2.889 |
enforcing | +2.881 |
interference | +2.816 |
Neg logit contributions | |
gements | -3.426 |
emed | -3.278 |
attest | -3.254 |
creen | -3.153 |
feature | -3.020 |
Token | Margaret |
Token ID | 19579 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
disrupting | +1.697 |
disruption | +1.638 |
obstruction | +1.602 |
interference | +1.551 |
enforcing | +1.546 |
Neg logit contributions | |
gements | -2.234 |
attest | -2.134 |
emed | -2.128 |
creen | -2.038 |
feature | -1.973 |
Token | Thatcher |
Token ID | 31706 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
gements | +4.005 |
attest | +3.901 |
emed | +3.709 |
feature | +3.613 |
Blossom | +3.594 |
Neg logit contributions | |
disrupting | -4.298 |
disruption | -4.088 |
obstruction | -4.027 |
enforcing | -3.976 |
interference | -3.913 |
Token | , |
Token ID | 11 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
disrupting | +1.694 |
disruption | +1.580 |
obstruction | +1.558 |
enforcing | +1.539 |
interference | +1.501 |
Neg logit contributions | |
gements | -2.206 |
attest | -2.140 |
emed | -2.127 |
creen | -2.014 |
feature | -1.986 |
Token | told |
Token ID | 1297 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
disrupting | +0.498 |
disruption | +0.471 |
obstruction | +0.467 |
enforcing | +0.460 |
interference | +0.451 |
Neg logit contributions | |
gements | -0.518 |
attest | -0.488 |
emed | -0.486 |
creen | -0.463 |
Blossom | -0.453 |
Token | the |
Token ID | 262 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
disrupting | +0.217 |
disruption | +0.208 |
obstruction | +0.202 |
enforcing | +0.200 |
interference | +0.199 |
Neg logit contributions | |
gements | -0.255 |
attest | -0.243 |
emed | -0.236 |
creen | -0.226 |
berry | -0.223 |
Token | BBC |
Token ID | 7823 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
disrupting | +0.413 |
disruption | +0.397 |
obstruction | +0.383 |
enforcing | +0.379 |
interference | +0.376 |
Neg logit contributions | |
gements | -0.452 |
attest | -0.432 |
emed | -0.427 |
creen | -0.404 |
respectively | -0.395 |
Token | . |
Token ID | 13 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
disrupting | +1.646 |
disruption | +1.561 |
obstruction | +1.502 |
enforcing | +1.499 |
interference | +1.498 |
Neg logit contributions | |
gements | -2.154 |
attest | -2.092 |
emed | -2.033 |
creen | -1.947 |
Blossom | -1.895 |
Token |
|
Token ID | 198 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
disrupting | +2.095 |
disruption | +1.982 |
obstruction | +1.945 |
HUD | +1.929 |
enforcing | +1.924 |
Neg logit contributions | |
gements | -3.518 |
attest | -3.441 |
emed | -3.297 |
creen | -3.191 |
feature | -3.130 |
Token |
|
Token ID | 198 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
disrupting | +1.940 |
disruption | +1.829 |
obstruction | +1.794 |
enforcing | +1.765 |
HUD | +1.764 |
Neg logit contributions | |
gements | -3.267 |
attest | -3.190 |
emed | -3.018 |
creen | -2.941 |
respectively | -2.906 |
Token | Asked |
Token ID | 18932 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
disrupting | +1.305 |
HUD | +1.289 |
disruption | +1.242 |
obstruction | +1.219 |
interference | +1.206 |
Neg logit contributions | |
gements | -1.559 |
attest | -1.549 |
emed | -1.422 |
feature | -1.416 |
respectively | -1.411 |
Token | behind |
Token ID | 2157 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
disrupting | +3.896 |
disruption | +3.709 |
obstruction | +3.642 |
enforcing | +3.589 |
interference | +3.547 |
Neg logit contributions | |
gements | -4.392 |
emed | -4.091 |
attest | -4.083 |
creen | -3.830 |
Blossom | -3.824 |
Token | the |
Token ID | 262 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
disrupting | +1.169 |
disruption | +1.107 |
obstruction | +1.089 |
enforcing | +1.075 |
interference | +1.067 |
Neg logit contributions | |
gements | -1.468 |
attest | -1.416 |
emed | -1.370 |
creen | -1.315 |
respectively | -1.309 |
Token | 361 |
Token ID | 47744 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
disrupting | +0.811 |
disruption | +0.773 |
obstruction | +0.768 |
enforcing | +0.753 |
interference | +0.741 |
Neg logit contributions | |
gements | -0.833 |
attest | -0.800 |
emed | -0.775 |
creen | -0.742 |
respectively | -0.731 |
Token | University |
Token ID | 2059 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
disrupting | +0.997 |
disruption | +0.945 |
obstruction | +0.934 |
enforcing | +0.915 |
interference | +0.909 |
Neg logit contributions | |
gements | -1.182 |
attest | -1.141 |
emed | -1.088 |
creen | -1.061 |
respectively | -1.033 |
Token | courthouse |
Token ID | 33625 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
disrupting | +1.313 |
disruption | +1.251 |
obstruction | +1.211 |
interference | +1.185 |
enforcing | +1.171 |
Neg logit contributions | |
gements | -1.525 |
attest | -1.457 |
emed | -1.424 |
creen | -1.377 |
Blossom | -1.330 |
Token | , |
Token ID | 11 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
gements | +2.755 |
attest | +2.631 |
emed | +2.569 |
Blossom | +2.458 |
creen | +2.451 |
Neg logit contributions | |
disrupting | -2.624 |
disruption | -2.492 |
obstruction | -2.459 |
enforcing | -2.417 |
interference | -2.381 |
Token | where |
Token ID | 810 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
gements | +3.295 |
attest | +3.183 |
emed | +3.069 |
feature | +2.946 |
Blossom | +2.945 |
Neg logit contributions | |
disrupting | -3.030 |
disruption | -2.862 |
obstruction | -2.817 |
enforcing | -2.784 |
interference | -2.735 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
gements | +3.537 |
attest | +3.429 |
emed | +3.300 |
feature | +3.178 |
Blossom | +3.162 |
Neg logit contributions | |
disrupting | -3.355 |
disruption | -3.184 |
obstruction | -3.135 |
enforcing | -3.091 |
interference | -3.032 |
Token | city |
Token ID | 1748 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
gements | +0.532 |
attest | +0.510 |
emed | +0.494 |
creen | +0.474 |
Blossom | +0.467 |
Neg logit contributions | |
disrupting | -0.564 |
disruption | -0.541 |
obstruction | -0.534 |
enforcing | -0.522 |
interference | -0.515 |
Token | 's |
Token ID | 338 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
disrupting | +3.198 |
disruption | +3.040 |
obstruction | +3.015 |
enforcing | +3.007 |
interference | +2.938 |
Neg logit contributions | |
gements | -3.562 |
emed | -3.322 |
attest | -3.321 |
Blossom | -3.150 |
creen | -3.123 |
Token | most |
Token ID | 749 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
gements | +0.952 |
attest | +0.915 |
emed | +0.885 |
creen | +0.854 |
Blossom | +0.843 |
Neg logit contributions | |
disrupting | -0.853 |
disruption | -0.813 |
obstruction | -0.803 |
enforcing | -0.793 |
interference | -0.777 |
Token |
|
Token ID | 198 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
disrupting | +1.289 |
disruption | +1.212 |
obstruction | +1.192 |
enforcing | +1.173 |
interference | +1.151 |
Neg logit contributions | |
gements | -1.940 |
attest | -1.895 |
emed | -1.802 |
creen | -1.745 |
Blossom | -1.740 |
Token | Read |
Token ID | 5569 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
gements | +0.138 |
attest | +0.136 |
emed | +0.124 |
feature | +0.124 |
Blossom | +0.123 |
Neg logit contributions | |
disrupting | -0.123 |
HUD | -0.118 |
disruption | -0.116 |
obstruction | -0.116 |
interference | -0.112 |
Token | More |
Token ID | 3125 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
disrupting | +1.049 |
disruption | +0.999 |
obstruction | +0.972 |
enforcing | +0.951 |
interference | +0.930 |
Neg logit contributions | |
gements | -1.205 |
attest | -1.185 |
emed | -1.139 |
creen | -1.097 |
Blossom | -1.091 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
disrupting | +4.870 |
obstruction | +4.670 |
HUD | +4.655 |
disruption | +4.606 |
interference | +4.387 |
Neg logit contributions | |
gements | -6.064 |
attest | -5.991 |
emed | -5.771 |
respectively | -5.525 |
creen | -5.252 |
Token | H |
Token ID | 39 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
gements | +1.823 |
attest | +1.793 |
emed | +1.672 |
feature | +1.650 |
Blossom | +1.647 |
Neg logit contributions | |
disrupting | -1.822 |
disruption | -1.727 |
HUD | -1.701 |
obstruction | -1.696 |
enforcing | -1.677 |
Token | ats |
Token ID | 1381 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
gements | +8.951 |
attest | +8.704 |
emed | +8.560 |
feature | +8.255 |
creen | +8.105 |
Neg logit contributions | |
disrupting | -6.803 |
disruption | -6.393 |
obstruction | -6.275 |
enforcing | -6.181 |
interference | -5.975 |
Token | une |
Token ID | 1726 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
gements | +5.834 |
attest | +5.683 |
emed | +5.423 |
feature | +5.274 |
Blossom | +5.249 |
Neg logit contributions | |
disrupting | -5.552 |
disruption | -5.265 |
obstruction | -5.172 |
enforcing | -5.109 |
interference | -5.019 |
Token | Mik |
Token ID | 17722 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
disrupting | +0.517 |
disruption | +0.484 |
obstruction | +0.467 |
enforcing | +0.458 |
interference | +0.449 |
Neg logit contributions | |
gements | -1.101 |
attest | -1.060 |
emed | -1.026 |
creen | -1.024 |
Blossom | -1.004 |
Token | u |
Token ID | 84 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
disrupting | +0.700 |
disruption | +0.640 |
obstruction | +0.629 |
enforcing | +0.628 |
HUD | +0.622 |
Neg logit contributions | |
gements | -1.000 |
attest | -0.991 |
emed | -0.940 |
respectively | -0.938 |
creen | -0.925 |
Token | : |
Token ID | 25 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
gements | +0.066 |
attest | +0.063 |
emed | +0.061 |
creen | +0.059 |
respectively | +0.058 |
Neg logit contributions | |
disrupting | -0.063 |
disruption | -0.059 |
obstruction | -0.058 |
enforcing | -0.058 |
interference | -0.056 |
Token | Project |
Token ID | 4935 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
gements | +1.139 |
attest | +1.110 |
emed | +1.060 |
creen | +1.016 |
feature | +1.007 |
Neg logit contributions | |
disrupting | -1.094 |
disruption | -1.039 |
obstruction | -1.024 |
enforcing | -1.015 |
interference | -0.995 |
Token | some |
Token ID | 617 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
gements | +1.655 |
attest | +1.593 |
emed | +1.527 |
creen | +1.468 |
Blossom | +1.460 |
Neg logit contributions | |
disrupting | -1.614 |
disruption | -1.538 |
obstruction | -1.501 |
enforcing | -1.493 |
interference | -1.469 |
Token | guy |
Token ID | 3516 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
disrupting | +0.980 |
disruption | +0.945 |
obstruction | +0.928 |
enforcing | +0.918 |
interference | +0.906 |
Neg logit contributions | |
gements | -0.720 |
attest | -0.680 |
emed | -0.635 |
Blossom | -0.617 |
creen | -0.616 |
Token | says |
Token ID | 1139 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
disrupting | +1.736 |
disruption | +1.617 |
enforcing | +1.614 |
obstruction | +1.590 |
thrott | +1.566 |
Neg logit contributions | |
gements | -1.714 |
attest | -1.637 |
emed | -1.602 |
Blossom | -1.539 |
creen | -1.524 |
Token | that |
Token ID | 326 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
gements | +2.164 |
attest | +2.094 |
emed | +2.012 |
Blossom | +1.953 |
feature | +1.937 |
Neg logit contributions | |
disrupting | -2.010 |
disruption | -1.905 |
obstruction | -1.873 |
enforcing | -1.847 |
interference | -1.819 |
Token | he |
Token ID | 339 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
gements | +2.111 |
attest | +2.046 |
emed | +1.965 |
Blossom | +1.900 |
feature | +1.890 |
Neg logit contributions | |
disrupting | -1.892 |
disruption | -1.787 |
obstruction | -1.752 |
enforcing | -1.738 |
interference | -1.707 |
Token | has |
Token ID | 468 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
gements | +1.054 |
attest | +1.028 |
emed | +0.986 |
Blossom | +0.963 |
feature | +0.954 |
Neg logit contributions | |
disrupting | -0.891 |
disruption | -0.836 |
obstruction | -0.825 |
enforcing | -0.816 |
interference | -0.797 |
Token | one |
Token ID | 530 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
gements | +3.396 |
attest | +3.297 |
emed | +3.159 |
Blossom | +3.051 |
feature | +3.047 |
Neg logit contributions | |
disrupting | -3.153 |
disruption | -2.991 |
obstruction | -2.944 |
enforcing | -2.896 |
interference | -2.857 |
Token | , |
Token ID | 11 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
gements | +4.922 |
attest | +4.809 |
emed | +4.597 |
feature | +4.496 |
Blossom | +4.457 |
Neg logit contributions | |
disrupting | -4.341 |
disruption | -4.106 |
obstruction | -4.033 |
enforcing | -3.978 |
interference | -3.906 |
Token | and |
Token ID | 290 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
gements | +1.118 |
attest | +1.088 |
emed | +1.050 |
Blossom | +1.019 |
creen | +1.015 |
Neg logit contributions | |
disrupting | -0.859 |
disruption | -0.801 |
obstruction | -0.787 |
enforcing | -0.782 |
interference | -0.763 |
Token | there |
Token ID | 612 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
gements | +3.102 |
attest | +3.014 |
emed | +2.880 |
feature | +2.786 |
Blossom | +2.773 |
Neg logit contributions | |
disrupting | -3.148 |
disruption | -2.990 |
obstruction | -2.937 |
enforcing | -2.908 |
interference | -2.860 |
Token | � |
Token ID | 447 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
gements | +0.993 |
attest | +0.957 |
emed | +0.915 |
Blossom | +0.892 |
creen | +0.888 |
Neg logit contributions | |
disrupting | -0.977 |
disruption | -0.928 |
obstruction | -0.915 |
enforcing | -0.908 |
interference | -0.890 |
Token | a |
Token ID | 257 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
disrupting | +1.446 |
disruption | +1.355 |
enforcing | +1.325 |
obstruction | +1.317 |
interference | +1.274 |
Neg logit contributions | |
gements | -2.249 |
attest | -2.173 |
emed | -2.099 |
creen | -2.048 |
Blossom | -2.041 |
Token | statement |
Token ID | 2643 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
disrupting | +0.695 |
disruption | +0.672 |
obstruction | +0.644 |
enforcing | +0.638 |
interference | +0.631 |
Neg logit contributions | |
gements | -0.723 |
attest | -0.699 |
emed | -0.676 |
Blossom | -0.651 |
creen | -0.647 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
gements | +3.548 |
attest | +3.455 |
emed | +3.291 |
feature | +3.205 |
Blossom | +3.179 |
Neg logit contributions | |
disrupting | -3.677 |
disruption | -3.497 |
obstruction | -3.442 |
enforcing | -3.392 |
interference | -3.343 |
Token | according |
Token ID | 1864 |
Feature activation | +6.69e-02 |
Pos logit contributions | |
gements | +0.533 |
attest | +0.516 |
emed | +0.500 |
creen | +0.484 |
Blossom | +0.483 |
Neg logit contributions | |
disrupting | -0.374 |
disruption | -0.345 |
obstruction | -0.339 |
enforcing | -0.339 |
interference | -0.327 |
Token | to |
Token ID | 284 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
disrupting | +0.286 |
disruption | +0.273 |
obstruction | +0.269 |
enforcing | +0.266 |
interference | +0.262 |
Neg logit contributions | |
gements | -0.271 |
attest | -0.262 |
emed | -0.250 |
feature | -0.239 |
creen | -0.239 |
Token | Variety |
Token ID | 39143 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
disrupting | +1.700 |
disruption | +1.622 |
obstruction | +1.576 |
enforcing | +1.562 |
interference | +1.553 |
Neg logit contributions | |
gements | -1.916 |
attest | -1.820 |
emed | -1.767 |
creen | -1.699 |
berry | -1.680 |
Token | . |
Token ID | 13 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
gements | +3.283 |
attest | +3.190 |
emed | +3.033 |
feature | +2.943 |
Blossom | +2.926 |
Neg logit contributions | |
disrupting | -3.718 |
disruption | -3.540 |
obstruction | -3.486 |
enforcing | -3.445 |
interference | -3.392 |
Token | � |
Token ID | 564 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
gements | +0.343 |
attest | +0.336 |
emed | +0.321 |
creen | +0.309 |
feature | +0.308 |
Neg logit contributions | |
disrupting | -0.240 |
disruption | -0.227 |
obstruction | -0.223 |
enforcing | -0.219 |
interference | -0.216 |
Token | � |
Token ID | 250 |
Feature activation | +8.17e-01 |
Pos logit contributions | |
disrupting | +1.000 |
disruption | +0.957 |
obstruction | +0.939 |
enforcing | +0.931 |
interference | +0.906 |
Neg logit contributions | |
gements | -0.874 |
attest | -0.859 |
emed | -0.809 |
creen | -0.793 |
feature | -0.784 |
Token | We |
Token ID | 1135 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
disrupting | +3.255 |
disruption | +3.114 |
HUD | +3.089 |
obstruction | +3.071 |
enforcing | +3.011 |
Neg logit contributions | |
gements | -3.409 |
attest | -3.338 |
emed | -3.144 |
Blossom | -2.990 |
respectively | -2.987 |
Token | wish |
Token ID | 4601 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
disrupting | +6.161 |
disruption | +5.851 |
obstruction | +5.778 |
enforcing | +5.737 |
thrott | +5.672 |
Neg logit contributions | |
gements | -4.485 |
Blossom | -4.126 |
emed | -4.067 |
attest | -4.038 |
otes | -3.819 |
Token |
|
Token ID | 198 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
gements | +0.445 |
attest | +0.435 |
emed | +0.417 |
feature | +0.406 |
Blossom | +0.404 |
Neg logit contributions | |
disrupting | -0.357 |
disruption | -0.337 |
obstruction | -0.331 |
enforcing | -0.326 |
interference | -0.320 |
Token |
|
Token ID | 198 |
Feature activation | -8.26e-02 |
Pos logit contributions | |
gements | +0.168 |
attest | +0.164 |
emed | +0.157 |
feature | +0.152 |
Blossom | +0.152 |
Neg logit contributions | |
disrupting | -0.126 |
disruption | -0.119 |
obstruction | -0.117 |
enforcing | -0.115 |
interference | -0.113 |
Token | Some |
Token ID | 4366 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
gements | +0.352 |
attest | +0.344 |
emed | +0.322 |
respectively | +0.315 |
feature | +0.315 |
Neg logit contributions | |
disrupting | -0.334 |
disruption | -0.318 |
HUD | -0.317 |
obstruction | -0.313 |
enforcing | -0.309 |
Token | of |
Token ID | 286 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
disrupting | +2.882 |
disruption | +2.753 |
obstruction | +2.680 |
enforcing | +2.625 |
interference | +2.603 |
Neg logit contributions | |
gements | -3.239 |
attest | -3.011 |
emed | -2.998 |
creen | -2.884 |
Blossom | -2.875 |
Token | the |
Token ID | 262 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
gements | +0.175 |
attest | +0.168 |
emed | +0.163 |
creen | +0.158 |
Blossom | +0.156 |
Neg logit contributions | |
disrupting | -0.148 |
disruption | -0.141 |
obstruction | -0.137 |
enforcing | -0.136 |
interference | -0.132 |
Token | deaths |
Token ID | 7040 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
disrupting | +0.446 |
disruption | +0.426 |
obstruction | +0.412 |
enforcing | +0.404 |
interference | +0.395 |
Neg logit contributions | |
gements | -0.606 |
attest | -0.578 |
emed | -0.567 |
creen | -0.551 |
Blossom | -0.548 |
Token | may |
Token ID | 743 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
attest | +6.954 |
gements | +6.930 |
emed | +6.498 |
feature | +6.488 |
respectively | +6.330 |
Neg logit contributions | |
disrupting | -8.188 |
disruption | -7.834 |
obstruction | -7.754 |
enforcing | -7.618 |
HUD | -7.604 |
Token | also |
Token ID | 635 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
gements | +3.660 |
attest | +3.410 |
emed | +3.232 |
Blossom | +3.108 |
creen | +3.037 |
Neg logit contributions | |
disrupting | -7.258 |
disruption | -6.988 |
obstruction | -6.888 |
enforcing | -6.811 |
interference | -6.742 |
Token | have |
Token ID | 423 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
gements | +3.276 |
attest | +3.028 |
emed | +2.890 |
Blossom | +2.777 |
creen | +2.709 |
Neg logit contributions | |
disrupting | -6.695 |
disruption | -6.451 |
obstruction | -6.362 |
enforcing | -6.285 |
interference | -6.229 |
Token | been |
Token ID | 587 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
gements | +1.788 |
attest | +1.654 |
emed | +1.576 |
Blossom | +1.518 |
creen | +1.510 |
Neg logit contributions | |
disrupting | -2.969 |
disruption | -2.851 |
obstruction | -2.810 |
enforcing | -2.767 |
interference | -2.743 |
Token | a |
Token ID | 257 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
disrupting | +1.260 |
disruption | +1.178 |
obstruction | +1.160 |
enforcing | +1.141 |
interference | +1.119 |
Neg logit contributions | |
gements | -1.671 |
attest | -1.583 |
emed | -1.558 |
Blossom | -1.502 |
creen | -1.472 |
Token | bitter |
Token ID | 12922 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
gements | +1.599 |
attest | +1.525 |
emed | +1.496 |
creen | +1.439 |
respectively | +1.404 |
Neg logit contributions | |
disrupting | -1.508 |
disruption | -1.440 |
obstruction | -1.407 |
enforcing | -1.385 |
interference | -1.374 |
Token | finish |
Token ID | 5461 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
gements | +6.797 |
attest | +6.662 |
emed | +6.344 |
feature | +6.232 |
Blossom | +6.191 |
Neg logit contributions | |
disrupting | -5.969 |
disruption | -5.635 |
obstruction | -5.546 |
enforcing | -5.481 |
interference | -5.364 |
Token | reminds |
Token ID | 17603 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
gements | +0.583 |
attest | +0.549 |
emed | +0.548 |
creen | +0.518 |
Blossom | +0.510 |
Neg logit contributions | |
disrupting | -0.608 |
disruption | -0.572 |
obstruction | -0.561 |
enforcing | -0.558 |
thrott | -0.553 |
Token | me |
Token ID | 502 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
gements | +5.933 |
attest | +5.776 |
emed | +5.393 |
feature | +5.262 |
respectively | +5.210 |
Neg logit contributions | |
disrupting | -10.467 |
disruption | -10.021 |
obstruction | -9.873 |
enforcing | -9.815 |
HUD | -9.698 |
Token | of |
Token ID | 286 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
gements | +0.324 |
attest | +0.315 |
emed | +0.304 |
creen | +0.293 |
Blossom | +0.293 |
Neg logit contributions | |
disrupting | -0.234 |
disruption | -0.219 |
obstruction | -0.217 |
enforcing | -0.213 |
interference | -0.210 |
Token | grape |
Token ID | 30777 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
gements | +2.773 |
attest | +2.681 |
emed | +2.587 |
creen | +2.497 |
feature | +2.490 |
Neg logit contributions | |
disrupting | -2.421 |
disruption | -2.292 |
obstruction | -2.250 |
enforcing | -2.214 |
interference | -2.183 |
Token | fruit |
Token ID | 34711 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
attest | +9.597 |
gements | +9.493 |
Blossom | +9.084 |
berry | +9.017 |
emed | +8.824 |
Neg logit contributions | |
disrupting | -11.581 |
disruption | -10.998 |
obstruction | -10.874 |
enforcing | -10.859 |
HUD | -10.709 |
Token | r |
Token ID | 374 |
Feature activation | -8.13e-01 |
Pos logit contributions | |
gements | +5.876 |
attest | +5.747 |
emed | +5.468 |
feature | +5.341 |
Blossom | +5.323 |
Neg logit contributions | |
disrupting | -5.715 |
disruption | -5.420 |
obstruction | -5.335 |
enforcing | -5.269 |
interference | -5.170 |
Token | ind |
Token ID | 521 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
gements | +3.292 |
attest | +3.212 |
emed | +3.064 |
feature | +2.980 |
Blossom | +2.969 |
Neg logit contributions | |
disrupting | -3.470 |
disruption | -3.292 |
obstruction | -3.238 |
enforcing | -3.193 |
interference | -3.155 |
Token | . |
Token ID | 13 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
gements | +6.164 |
attest | +6.036 |
emed | +5.683 |
feature | +5.560 |
Blossom | +5.537 |
Neg logit contributions | |
disrupting | -7.152 |
disruption | -6.800 |
obstruction | -6.725 |
enforcing | -6.647 |
interference | -6.528 |
Token |
|
Token ID | 198 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
gements | +2.459 |
attest | +2.388 |
emed | +2.295 |
feature | +2.213 |
creen | +2.209 |
Neg logit contributions | |
disrupting | -1.990 |
disruption | -1.881 |
obstruction | -1.843 |
enforcing | -1.831 |
HUD | -1.801 |
Token | through |
Token ID | 832 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
disrupting | +5.141 |
disruption | +4.838 |
obstruction | +4.741 |
interference | +4.673 |
enforcing | +4.650 |
Neg logit contributions | |
gements | -6.946 |
attest | -6.522 |
emed | -6.362 |
creen | -6.126 |
Blossom | -6.043 |
Token | her |
Token ID | 607 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
disrupting | +4.086 |
disruption | +3.917 |
obstruction | +3.757 |
interference | +3.711 |
enforcing | +3.685 |
Neg logit contributions | |
gements | -5.076 |
attest | -4.912 |
emed | -4.760 |
creen | -4.586 |
respectively | -4.460 |
Token | family |
Token ID | 1641 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
disrupting | +3.217 |
disruption | +3.107 |
obstruction | +2.992 |
interference | +2.957 |
enforcing | +2.918 |
Neg logit contributions | |
gements | -2.977 |
attest | -2.821 |
emed | -2.791 |
creen | -2.640 |
respectively | -2.551 |
Token | brunch |
Token ID | 50219 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
disrupting | +1.015 |
disruption | +0.964 |
obstruction | +0.932 |
enforcing | +0.920 |
interference | +0.906 |
Neg logit contributions | |
gements | -1.233 |
attest | -1.175 |
emed | -1.166 |
creen | -1.104 |
respectively | -1.086 |
Token | at |
Token ID | 379 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
gements | +1.225 |
attest | +1.187 |
emed | +1.157 |
creen | +1.096 |
Blossom | +1.093 |
Neg logit contributions | |
disrupting | -1.044 |
disruption | -0.991 |
obstruction | -0.959 |
enforcing | -0.950 |
interference | -0.932 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
disrupting | +3.697 |
disruption | +3.521 |
obstruction | +3.391 |
enforcing | +3.354 |
interference | +3.346 |
Neg logit contributions | |
gements | -4.308 |
attest | -4.234 |
emed | -4.087 |
respectively | -3.856 |
creen | -3.829 |
Token | ih |
Token ID | 4449 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
gements | +6.287 |
emed | +5.876 |
attest | +5.720 |
berry | +5.577 |
Blossom | +5.467 |
Neg logit contributions | |
disrupting | -10.424 |
disruption | -10.095 |
obstruction | -9.922 |
enforcing | -9.811 |
interfering | -9.528 |
Token | ana |
Token ID | 2271 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
gements | +4.624 |
emed | +3.968 |
attest | +3.847 |
berry | +3.729 |
feature | +3.698 |
Neg logit contributions | |
disrupting | -15.669 |
disruption | -15.077 |
enforcing | -14.939 |
obstruction | -14.593 |
interfering | -14.588 |
Token | restaurant |
Token ID | 7072 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
gements | +1.811 |
attest | +1.762 |
emed | +1.699 |
creen | +1.639 |
feature | +1.627 |
Neg logit contributions | |
disrupting | -1.439 |
disruption | -1.352 |
obstruction | -1.318 |
enforcing | -1.308 |
interference | -1.281 |
Token | in |
Token ID | 287 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
disrupting | +1.834 |
disruption | +1.687 |
enforcing | +1.669 |
obstruction | +1.644 |
interference | +1.598 |
Neg logit contributions | |
gements | -2.403 |
attest | -2.290 |
emed | -2.255 |
creen | -2.118 |
Blossom | -2.104 |
Token | LA |
Token ID | 9131 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
disrupting | +1.572 |
disruption | +1.484 |
obstruction | +1.460 |
enforcing | +1.459 |
interference | +1.416 |
Neg logit contributions | |
gements | -1.983 |
attest | -1.910 |
emed | -1.856 |
creen | -1.759 |
feature | -1.746 |
Token | of |
Token ID | 286 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
gements | +1.012 |
attest | +0.985 |
emed | +0.946 |
feature | +0.921 |
Blossom | +0.920 |
Neg logit contributions | |
disrupting | -0.783 |
disruption | -0.738 |
obstruction | -0.725 |
enforcing | -0.716 |
interference | -0.700 |
Token | Natural |
Token ID | 12068 |
Feature activation | -8.13e-01 |
Pos logit contributions | |
gements | +1.186 |
attest | +1.139 |
emed | +1.098 |
creen | +1.072 |
feature | +1.038 |
Neg logit contributions | |
disrupting | -1.240 |
disruption | -1.183 |
obstruction | -1.160 |
enforcing | -1.140 |
interference | -1.123 |
Token | History |
Token ID | 7443 |
Feature activation | -8.39e-02 |
Pos logit contributions | |
gements | +2.805 |
attest | +2.719 |
emed | +2.597 |
creen | +2.485 |
Blossom | +2.483 |
Neg logit contributions | |
disrupting | -3.967 |
disruption | -3.805 |
obstruction | -3.751 |
enforcing | -3.690 |
interference | -3.643 |
Token | ( |
Token ID | 357 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
gements | +0.370 |
attest | +0.354 |
emed | +0.342 |
creen | +0.329 |
Blossom | +0.327 |
Neg logit contributions | |
disrupting | -0.337 |
disruption | -0.316 |
obstruction | -0.314 |
enforcing | -0.308 |
interference | -0.303 |
Token | which |
Token ID | 4758 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
gements | +4.841 |
attest | +4.727 |
emed | +4.535 |
feature | +4.427 |
Blossom | +4.398 |
Neg logit contributions | |
disrupting | -3.801 |
disruption | -3.582 |
obstruction | -3.515 |
enforcing | -3.464 |
interference | -3.399 |
Token | houses |
Token ID | 7777 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
gements | +5.483 |
attest | +5.361 |
emed | +5.069 |
feature | +4.987 |
respectively | +4.907 |
Neg logit contributions | |
disrupting | -6.040 |
disruption | -5.749 |
obstruction | -5.650 |
enforcing | -5.573 |
interference | -5.502 |
Token | the |
Token ID | 262 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
gements | +8.070 |
attest | +7.968 |
feature | +7.511 |
emed | +7.477 |
Blossom | +7.298 |
Neg logit contributions | |
disrupting | -8.363 |
disruption | -7.897 |
obstruction | -7.775 |
enforcing | -7.716 |
HUD | -7.678 |
Token | Prot |
Token ID | 5038 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
gements | +1.849 |
attest | +1.771 |
emed | +1.725 |
creen | +1.647 |
Blossom | +1.639 |
Neg logit contributions | |
disrupting | -2.042 |
disruption | -1.960 |
obstruction | -1.932 |
enforcing | -1.886 |
interference | -1.877 |
Token | oc |
Token ID | 420 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
gements | +4.037 |
attest | +3.952 |
emed | +3.786 |
Blossom | +3.705 |
feature | +3.697 |
Neg logit contributions | |
disrupting | -2.804 |
disruption | -2.631 |
obstruction | -2.574 |
enforcing | -2.528 |
interference | -2.481 |
Token | er |
Token ID | 263 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
disrupting | +6.578 |
disruption | +6.517 |
activity | +6.250 |
oxic | +6.240 |
obstruction | +6.088 |
Neg logit contributions | |
attest | -6.601 |
gements | -6.589 |
soever | -6.165 |
Blossom | -6.122 |
emed | -5.855 |
Token | at |
Token ID | 265 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
gements | +1.060 |
attest | +1.036 |
emed | +0.992 |
Blossom | +0.964 |
creen | +0.960 |
Neg logit contributions | |
disrupting | -0.697 |
disruption | -0.654 |
obstruction | -0.639 |
enforcing | -0.621 |
interference | -0.615 |
Token | reported |
Token ID | 2098 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
disrupting | +1.054 |
enforcing | +0.988 |
disruption | +0.977 |
obstruction | +0.973 |
interference | +0.942 |
Neg logit contributions | |
gements | -1.315 |
attest | -1.260 |
emed | -1.219 |
Blossom | -1.194 |
creen | -1.178 |
Token | to |
Token ID | 284 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
disrupting | +1.833 |
enforcing | +1.727 |
disruption | +1.697 |
obstruction | +1.676 |
interference | +1.642 |
Neg logit contributions | |
gements | -2.238 |
attest | -2.131 |
emed | -2.088 |
Blossom | -2.068 |
creen | -2.040 |
Token | hold |
Token ID | 1745 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
disrupting | +1.805 |
disruption | +1.760 |
enforcing | +1.758 |
interference | +1.711 |
obstruction | +1.710 |
Neg logit contributions | |
gements | -1.117 |
attest | -1.000 |
emed | -0.985 |
creen | -0.957 |
Blossom | -0.949 |
Token | that |
Token ID | 326 |
Feature activation | -8.26e-02 |
Pos logit contributions | |
gements | +2.156 |
attest | +2.095 |
emed | +2.018 |
Blossom | +1.971 |
creen | +1.947 |
Neg logit contributions | |
disrupting | -1.657 |
disruption | -1.568 |
obstruction | -1.540 |
enforcing | -1.533 |
interference | -1.501 |
Token | the |
Token ID | 262 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
gements | +0.385 |
attest | +0.372 |
emed | +0.357 |
Blossom | +0.349 |
creen | +0.348 |
Neg logit contributions | |
disrupting | -0.311 |
disruption | -0.296 |
enforcing | -0.292 |
obstruction | -0.292 |
interference | -0.288 |
Token | Four |
Token ID | 6675 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
disrupting | +1.707 |
disruption | +1.663 |
obstruction | +1.634 |
interference | +1.623 |
enforcing | +1.615 |
Neg logit contributions | |
gements | -1.852 |
attest | -1.786 |
emed | -1.710 |
creen | -1.689 |
maid | -1.651 |
Token | teenth |
Token ID | 20283 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
gements | +2.989 |
attest | +2.470 |
emed | +2.142 |
Blossom | +2.103 |
feature | +2.044 |
Neg logit contributions | |
disrupting | -14.582 |
disruption | -14.108 |
obstruction | -13.918 |
enforcing | -13.690 |
interference | -13.580 |
Token | Amendment |
Token ID | 8441 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
gements | +7.056 |
attest | +7.056 |
feature | +6.577 |
emed | +6.514 |
respectively | +6.194 |
Neg logit contributions | |
disrupting | -14.946 |
HUD | -14.180 |
disruption | -14.124 |
obstruction | -13.908 |
thrott | -13.899 |
Token | 's |
Token ID | 338 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
gements | +0.725 |
attest | +0.704 |
emed | +0.675 |
creen | +0.656 |
Blossom | +0.652 |
Neg logit contributions | |
disrupting | -0.704 |
disruption | -0.672 |
obstruction | -0.661 |
interference | -0.659 |
enforcing | -0.656 |
Token | equal |
Token ID | 4961 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
disrupting | +0.096 |
disruption | +0.093 |
obstruction | +0.091 |
interference | +0.091 |
enforcing | +0.091 |
Neg logit contributions | |
gements | -0.101 |
attest | -0.097 |
emed | -0.092 |
creen | -0.091 |
Blossom | -0.090 |
Token | protection |
Token ID | 4800 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
gements | +1.652 |
attest | +1.607 |
emed | +1.550 |
Blossom | +1.508 |
creen | +1.506 |
Neg logit contributions | |
disrupting | -1.083 |
disruption | -1.019 |
obstruction | -0.999 |
enforcing | -0.980 |
interference | -0.966 |
Token | # |
Token ID | 2 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
gements | +0.944 |
attest | +0.917 |
emed | +0.864 |
creen | +0.836 |
Blossom | +0.836 |
Neg logit contributions | |
disrupting | -0.960 |
disruption | -0.915 |
obstruction | -0.906 |
enforcing | -0.885 |
interference | -0.876 |
Token | ] |
Token ID | 60 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
disrupting | +0.751 |
disruption | +0.719 |
obstruction | +0.711 |
enforcing | +0.694 |
interference | +0.685 |
Neg logit contributions | |
gements | -0.741 |
attest | -0.716 |
emed | -0.683 |
creen | -0.658 |
Blossom | -0.651 |
Token | & |
Token ID | 1222 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
gements | +2.028 |
attest | +1.976 |
emed | +1.883 |
feature | +1.809 |
Blossom | +1.809 |
Neg logit contributions | |
disrupting | -2.120 |
disruption | -2.021 |
obstruction | -1.989 |
enforcing | -1.959 |
interference | -1.928 |
Token | / |
Token ID | 1220 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
disrupting | +0.506 |
disruption | +0.484 |
obstruction | +0.479 |
enforcing | +0.462 |
interference | +0.459 |
Neg logit contributions | |
gements | -0.592 |
attest | -0.574 |
emed | -0.551 |
creen | -0.534 |
Blossom | -0.527 |
Token | @ |
Token ID | 31 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
disrupting | +4.123 |
disruption | +3.937 |
obstruction | +3.908 |
HUD | +3.808 |
enforcing | +3.805 |
Neg logit contributions | |
gements | -4.232 |
attest | -4.076 |
emed | -3.763 |
Blossom | -3.689 |
ingly | -3.652 |
Token | Keys |
Token ID | 26363 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
disrupting | +3.068 |
disruption | +2.955 |
obstruction | +2.933 |
interference | +2.838 |
enforcing | +2.778 |
Neg logit contributions | |
gements | -3.119 |
attest | -3.056 |
emed | -2.877 |
creen | -2.778 |
respectively | -2.720 |
Token | [ |
Token ID | 58 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
gements | +7.577 |
attest | +7.342 |
emed | +7.009 |
feature | +6.937 |
respectively | +6.914 |
Neg logit contributions | |
disrupting | -8.260 |
disruption | -7.825 |
obstruction | -7.716 |
enforcing | -7.623 |
interference | -7.494 |
Token | step |
Token ID | 9662 |
Feature activation | +9.89e-02 |
Pos logit contributions | |
gements | +4.284 |
attest | +4.173 |
emed | +3.983 |
feature | +3.878 |
Blossom | +3.853 |
Neg logit contributions | |
disrupting | -4.202 |
disruption | -3.986 |
obstruction | -3.920 |
enforcing | -3.871 |
interference | -3.806 |
Token | Old |
Token ID | 19620 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
disrupting | +0.443 |
disruption | +0.423 |
obstruction | +0.414 |
enforcing | +0.409 |
HUD | +0.408 |
Neg logit contributions | |
gements | -0.378 |
attest | -0.367 |
emed | -0.347 |
Blossom | -0.336 |
creen | -0.336 |
Token | ], |
Token ID | 4357 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
gements | +0.467 |
attest | +0.453 |
emed | +0.428 |
Blossom | +0.415 |
creen | +0.414 |
Neg logit contributions | |
disrupting | -0.466 |
disruption | -0.445 |
obstruction | -0.437 |
enforcing | -0.430 |
HUD | -0.426 |
Token | 1 |
Token ID | 352 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
gements | +0.541 |
attest | +0.523 |
emed | +0.508 |
creen | +0.487 |
Blossom | +0.476 |
Neg logit contributions | |
disrupting | -0.534 |
disruption | -0.510 |
obstruction | -0.504 |
enforcing | -0.486 |
interference | -0.486 |