Token | , |
Token ID | 11 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
shoulder | +0.665 |
handc | +0.661 |
Bench | +0.651 |
� | +0.616 |
� | +0.616 |
Neg logit contributions | |
alks | -0.625 |
alt | -0.589 |
iley | -0.557 |
MH | -0.545 |
azeera | -0.542 |
Token | " |
Token ID | 366 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
alks | +0.061 |
alt | +0.057 |
iley | +0.054 |
MH | +0.053 |
azeera | +0.053 |
Neg logit contributions | |
shoulder | -0.061 |
handc | -0.061 |
Bench | -0.060 |
� | -0.057 |
� | -0.057 |
Token | Okay |
Token ID | 16454 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
shoulder | +1.013 |
handc | +1.008 |
Bench | +0.997 |
� | +0.947 |
� | +0.946 |
Neg logit contributions | |
alks | -1.013 |
alt | -0.952 |
iley | -0.907 |
MH | -0.891 |
azeera | -0.885 |
Token | , |
Token ID | 11 |
Feature activation | -9.82e-03 |
Pos logit contributions | |
alks | +0.071 |
alt | +0.067 |
iley | +0.063 |
MH | +0.062 |
azeera | +0.061 |
Neg logit contributions | |
shoulder | -0.078 |
handc | -0.077 |
Bench | -0.076 |
� | -0.072 |
� | -0.072 |
Token | why |
Token ID | 1521 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
shoulder | +0.025 |
handc | +0.025 |
Bench | +0.024 |
� | +0.023 |
� | +0.023 |
Neg logit contributions | |
alks | -0.026 |
alt | -0.025 |
iley | -0.024 |
MH | -0.023 |
azeera | -0.023 |
Token | don |
Token ID | 836 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
alks | +0.529 |
alt | +0.499 |
iley | +0.472 |
MH | +0.462 |
ic | +0.456 |
Neg logit contributions | |
shoulder | -0.587 |
handc | -0.586 |
Bench | -0.576 |
� | -0.548 |
� | -0.547 |
Token | 't |
Token ID | 470 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
alks | +3.024 |
alt | +2.938 |
iley | +2.719 |
ic | +2.603 |
MH | +2.543 |
Neg logit contributions | |
handc | -5.638 |
shoulder | -5.502 |
Bench | -5.347 |
stitching | -5.278 |
fencing | -5.141 |
Token | you |
Token ID | 345 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
alks | +2.375 |
alt | +2.247 |
iley | +2.131 |
MH | +2.102 |
ic | +2.088 |
Neg logit contributions | |
handc | -2.376 |
shoulder | -2.376 |
Bench | -2.320 |
� | -2.224 |
� | -2.221 |
Token | figure |
Token ID | 3785 |
Feature activation | -7.24e-02 |
Pos logit contributions | |
alks | +2.353 |
alt | +2.235 |
iley | +2.112 |
MH | +2.093 |
ic | +2.083 |
Neg logit contributions | |
handc | -2.286 |
shoulder | -2.285 |
Bench | -2.249 |
� | -2.185 |
� | -2.179 |
Token | that |
Token ID | 326 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
shoulder | +0.190 |
handc | +0.189 |
Bench | +0.186 |
� | +0.176 |
� | +0.176 |
Neg logit contributions | |
alks | -0.188 |
alt | -0.177 |
iley | -0.168 |
MH | -0.165 |
ic | -0.163 |
Token | out |
Token ID | 503 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
alks | +0.177 |
alt | +0.168 |
iley | +0.159 |
MH | +0.156 |
ic | +0.155 |
Neg logit contributions | |
handc | -0.201 |
shoulder | -0.201 |
Bench | -0.198 |
� | -0.189 |
� | -0.189 |
Token | the |
Token ID | 262 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
alks | +0.444 |
alt | +0.421 |
iley | +0.401 |
MH | +0.391 |
azeera | +0.388 |
Neg logit contributions | |
shoulder | -0.407 |
handc | -0.405 |
Bench | -0.398 |
� | -0.375 |
� | -0.374 |
Token | April |
Token ID | 3035 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
shoulder | +0.383 |
handc | +0.379 |
Bench | +0.372 |
� | +0.349 |
� | +0.348 |
Neg logit contributions | |
alks | -0.402 |
alt | -0.379 |
iley | -0.362 |
azeera | -0.351 |
MH | -0.347 |
Token | 1986 |
Token ID | 12113 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
shoulder | +0.506 |
handc | +0.496 |
Bench | +0.492 |
� | +0.461 |
� | +0.459 |
Neg logit contributions | |
alks | -0.526 |
alt | -0.495 |
iley | -0.474 |
azeera | -0.465 |
MH | -0.454 |
Token | launch |
Token ID | 4219 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
shoulder | +0.508 |
handc | +0.499 |
Bench | +0.493 |
� | +0.464 |
� | +0.463 |
Neg logit contributions | |
alks | -0.447 |
alt | -0.418 |
iley | -0.398 |
azeera | -0.388 |
iens | -0.378 |
Token | vehicle |
Token ID | 4038 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
alks | +0.085 |
alt | +0.080 |
iley | +0.075 |
MH | +0.073 |
azeera | +0.073 |
Neg logit contributions | |
shoulder | -0.103 |
handc | -0.102 |
Bench | -0.100 |
� | -0.095 |
� | -0.095 |
Token | accident |
Token ID | 5778 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
alks | +1.171 |
alt | +1.104 |
iley | +1.038 |
MH | +1.023 |
ic | +1.005 |
Neg logit contributions | |
handc | -1.444 |
shoulder | -1.441 |
Bench | -1.420 |
� | -1.360 |
� | -1.356 |
Token | , |
Token ID | 11 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
alks | +3.998 |
alt | +3.803 |
MH | +3.642 |
iley | +3.610 |
ic | +3.533 |
Neg logit contributions | |
handc | -3.963 |
shoulder | -3.897 |
Bench | -3.819 |
� | -3.670 |
� | -3.648 |
Token | Per |
Token ID | 2448 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
alks | +0.437 |
alt | +0.415 |
iley | +0.394 |
MH | +0.387 |
ic | +0.383 |
Neg logit contributions | |
shoulder | -0.394 |
handc | -0.394 |
Bench | -0.386 |
� | -0.365 |
� | -0.364 |
Token | kin |
Token ID | 5116 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
shoulder | +0.333 |
handc | +0.332 |
Bench | +0.327 |
� | +0.310 |
� | +0.309 |
Neg logit contributions | |
alks | -0.305 |
alt | -0.287 |
iley | -0.272 |
MH | -0.267 |
ic | -0.263 |
Token | - |
Token ID | 12 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
shoulder | +1.359 |
handc | +1.340 |
Bench | +1.328 |
� | +1.255 |
� | +1.248 |
Neg logit contributions | |
alks | -1.282 |
alt | -1.196 |
iley | -1.146 |
azeera | -1.119 |
MH | -1.119 |
Token | El |
Token ID | 9527 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
shoulder | +0.083 |
handc | +0.082 |
Bench | +0.081 |
� | +0.077 |
� | +0.077 |
Neg logit contributions | |
alks | -0.073 |
alt | -0.068 |
iley | -0.065 |
MH | -0.063 |
azeera | -0.063 |
Token | to |
Token ID | 284 |
Feature activation | -7.74e-03 |
Pos logit contributions | |
handc | +0.254 |
shoulder | +0.254 |
Bench | +0.250 |
� | +0.237 |
� | +0.236 |
Neg logit contributions | |
alks | -0.294 |
alt | -0.280 |
iley | -0.267 |
MH | -0.262 |
ic | -0.260 |
Token | guarantee |
Token ID | 9149 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
shoulder | +0.018 |
handc | +0.018 |
Bench | +0.018 |
� | +0.017 |
� | +0.017 |
Neg logit contributions | |
alks | -0.022 |
alt | -0.021 |
iley | -0.020 |
MH | -0.020 |
azeera | -0.020 |
Token | safe |
Token ID | 3338 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
alks | +0.115 |
alt | +0.109 |
iley | +0.103 |
MH | +0.101 |
azeera | +0.100 |
Neg logit contributions | |
shoulder | -0.115 |
handc | -0.115 |
Bench | -0.113 |
� | -0.106 |
� | -0.106 |
Token | de |
Token ID | 390 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
shoulder | +0.017 |
handc | +0.017 |
Bench | +0.017 |
� | +0.016 |
� | +0.016 |
Neg logit contributions | |
alks | -0.020 |
alt | -0.019 |
iley | -0.018 |
MH | -0.018 |
azeera | -0.018 |
Token | ref |
Token ID | 5420 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
shoulder | +0.370 |
handc | +0.370 |
Bench | +0.363 |
� | +0.345 |
� | +0.345 |
Neg logit contributions | |
alks | -0.337 |
alt | -0.315 |
iley | -0.300 |
MH | -0.295 |
azeera | -0.292 |
Token | eren |
Token ID | 14226 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
alks | +1.823 |
alt | +1.715 |
iley | +1.604 |
ic | +1.560 |
MH | +1.555 |
Neg logit contributions | |
handc | -2.369 |
shoulder | -2.368 |
Bench | -2.323 |
� | -2.220 |
� | -2.219 |
Token | cing |
Token ID | 2259 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
alks | +3.307 |
alt | +3.179 |
iley | +2.880 |
ic | +2.856 |
iens | +2.800 |
Neg logit contributions | |
handc | -4.701 |
shoulder | -4.699 |
Bench | -4.618 |
� | -4.387 |
� | -4.387 |
Token | of |
Token ID | 286 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
alks | +0.693 |
alt | +0.659 |
iley | +0.627 |
MH | +0.617 |
ic | +0.610 |
Neg logit contributions | |
shoulder | -0.615 |
handc | -0.615 |
Bench | -0.603 |
� | -0.572 |
� | -0.570 |
Token | the |
Token ID | 262 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
alks | +0.363 |
alt | +0.344 |
iley | +0.328 |
MH | +0.318 |
azeera | +0.317 |
Neg logit contributions | |
shoulder | -0.348 |
handc | -0.345 |
Bench | -0.341 |
� | -0.318 |
� | -0.317 |
Token | pointer |
Token ID | 17562 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
alks | +0.236 |
alt | +0.223 |
iley | +0.212 |
MH | +0.208 |
azeera | +0.206 |
Neg logit contributions | |
shoulder | -0.221 |
handc | -0.219 |
Bench | -0.216 |
� | -0.202 |
� | -0.201 |
Token | within |
Token ID | 1626 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
alks | +0.725 |
alt | +0.687 |
iley | +0.653 |
MH | +0.640 |
azeera | +0.635 |
Neg logit contributions | |
shoulder | -0.660 |
handc | -0.657 |
Bench | -0.646 |
� | -0.607 |
� | -0.607 |
Token | make |
Token ID | 787 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
shoulder | +0.018 |
handc | +0.018 |
Bench | +0.018 |
� | +0.017 |
� | +0.017 |
Neg logit contributions | |
alks | -0.022 |
alt | -0.021 |
iley | -0.020 |
MH | -0.019 |
ic | -0.019 |
Token | reproductive |
Token ID | 18391 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
alks | +0.338 |
alt | +0.322 |
iley | +0.307 |
MH | +0.302 |
azeera | +0.298 |
Neg logit contributions | |
shoulder | -0.257 |
handc | -0.256 |
Bench | -0.250 |
� | -0.235 |
� | -0.235 |
Token | choices |
Token ID | 7747 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
shoulder | +1.133 |
handc | +1.118 |
Bench | +1.081 |
� | +1.003 |
� | +0.998 |
Neg logit contributions | |
alks | -2.089 |
alt | -2.002 |
iley | -1.922 |
MH | -1.885 |
azeera | -1.883 |
Token | . |
Token ID | 13 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
shoulder | +0.371 |
handc | +0.370 |
Bench | +0.361 |
� | +0.341 |
� | +0.340 |
Neg logit contributions | |
alks | -0.426 |
alt | -0.403 |
iley | -0.384 |
MH | -0.377 |
azeera | -0.372 |
Token | Why |
Token ID | 4162 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
alks | +0.247 |
alt | +0.234 |
iley | +0.222 |
MH | +0.218 |
ic | +0.215 |
Neg logit contributions | |
shoulder | -0.252 |
handc | -0.251 |
Bench | -0.246 |
� | -0.234 |
� | -0.233 |
Token | don |
Token ID | 836 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
alks | +0.862 |
alt | +0.817 |
iley | +0.778 |
MH | +0.765 |
azeera | +0.754 |
Neg logit contributions | |
shoulder | -0.748 |
handc | -0.745 |
Bench | -0.731 |
� | -0.688 |
� | -0.687 |
Token | � |
Token ID | 447 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
alks | +4.070 |
alt | +3.962 |
iley | +3.773 |
ic | +3.673 |
MH | +3.625 |
Neg logit contributions | |
handc | -3.880 |
shoulder | -3.776 |
Bench | -3.619 |
stitching | -3.553 |
fencing | -3.406 |
Token | � |
Token ID | 247 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
alks | +0.053 |
alt | +0.050 |
iley | +0.048 |
MH | +0.047 |
azeera | +0.046 |
Neg logit contributions | |
shoulder | -0.045 |
handc | -0.044 |
Bench | -0.044 |
� | -0.041 |
� | -0.041 |
Token | t |
Token ID | 83 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
alks | +0.066 |
alt | +0.063 |
iley | +0.061 |
MH | +0.060 |
azeera | +0.059 |
Neg logit contributions | |
shoulder | -0.041 |
handc | -0.041 |
Bench | -0.040 |
� | -0.038 |
� | -0.037 |
Token | you |
Token ID | 345 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
alks | +0.674 |
alt | +0.638 |
iley | +0.606 |
MH | +0.593 |
azeera | +0.587 |
Neg logit contributions | |
shoulder | -0.649 |
handc | -0.648 |
Bench | -0.635 |
� | -0.600 |
� | -0.600 |
Token | trust |
Token ID | 3774 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
alks | +0.864 |
alt | +0.818 |
iley | +0.776 |
MH | +0.761 |
ic | +0.752 |
Neg logit contributions | |
shoulder | -0.852 |
handc | -0.849 |
Bench | -0.835 |
� | -0.791 |
� | -0.789 |
Token | slower |
Token ID | 13611 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
alks | +1.340 |
alt | +1.277 |
iley | +1.219 |
MH | +1.199 |
ic | +1.186 |
Neg logit contributions | |
shoulder | -1.016 |
handc | -1.013 |
Bench | -0.992 |
� | -0.932 |
� | -0.930 |
Token | pacing |
Token ID | 37572 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
alks | +2.023 |
alt | +1.940 |
iley | +1.850 |
MH | +1.830 |
ic | +1.809 |
Neg logit contributions | |
handc | -1.515 |
shoulder | -1.515 |
Bench | -1.489 |
� | -1.411 |
� | -1.404 |
Token | and |
Token ID | 290 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
alks | +1.890 |
alt | +1.804 |
iley | +1.704 |
MH | +1.674 |
ic | +1.649 |
Neg logit contributions | |
handc | -2.065 |
shoulder | -2.058 |
Bench | -2.022 |
� | -1.939 |
� | -1.934 |
Token | spreads |
Token ID | 26779 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
alks | +0.935 |
alt | +0.887 |
iley | +0.844 |
MH | +0.831 |
ic | +0.821 |
Neg logit contributions | |
shoulder | -0.834 |
handc | -0.833 |
Bench | -0.818 |
� | -0.774 |
� | -0.773 |
Token | its |
Token ID | 663 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
alks | +0.342 |
alt | +0.325 |
iley | +0.310 |
MH | +0.304 |
ic | +0.301 |
Neg logit contributions | |
shoulder | -0.291 |
handc | -0.290 |
Bench | -0.285 |
� | -0.269 |
� | -0.268 |
Token | attention |
Token ID | 3241 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
alks | +1.568 |
alt | +1.493 |
iley | +1.423 |
MH | +1.407 |
ic | +1.394 |
Neg logit contributions | |
handc | -1.329 |
shoulder | -1.328 |
Bench | -1.305 |
� | -1.240 |
� | -1.235 |
Token | across |
Token ID | 1973 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
alks | +4.238 |
alt | +4.108 |
iley | +3.875 |
ic | +3.841 |
MH | +3.824 |
Neg logit contributions | |
handc | -3.466 |
Bench | -3.418 |
shoulder | -3.398 |
� | -3.214 |
� | -3.206 |
Token | multiple |
Token ID | 3294 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
alks | +0.857 |
alt | +0.816 |
iley | +0.778 |
MH | +0.765 |
ic | +0.756 |
Neg logit contributions | |
shoulder | -0.682 |
handc | -0.679 |
Bench | -0.666 |
� | -0.627 |
� | -0.625 |
Token | different |
Token ID | 1180 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
alks | +1.062 |
alt | +1.004 |
iley | +0.951 |
MH | +0.933 |
ic | +0.921 |
Neg logit contributions | |
shoulder | -1.112 |
handc | -1.110 |
Bench | -1.091 |
� | -1.036 |
� | -1.035 |
Token | perspectives |
Token ID | 22582 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
shoulder | +0.671 |
handc | +0.664 |
Bench | +0.652 |
� | +0.616 |
stitching | +0.612 |
Neg logit contributions | |
alks | -0.657 |
alt | -0.622 |
iley | -0.590 |
MH | -0.574 |
azeera | -0.569 |
Token | of |
Token ID | 286 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
alks | +0.886 |
alt | +0.840 |
iley | +0.795 |
MH | +0.780 |
ic | +0.771 |
Neg logit contributions | |
handc | -0.903 |
shoulder | -0.901 |
Bench | -0.885 |
� | -0.841 |
� | -0.840 |
Token | of |
Token ID | 286 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
shoulder | +1.137 |
handc | +1.127 |
Bench | +1.110 |
� | +1.047 |
� | +1.046 |
Neg logit contributions | |
alks | -1.241 |
alt | -1.176 |
iley | -1.112 |
azeera | -1.110 |
MH | -1.095 |
Token | what |
Token ID | 644 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
shoulder | +0.230 |
handc | +0.229 |
Bench | +0.223 |
� | +0.208 |
� | +0.206 |
Neg logit contributions | |
alks | -0.314 |
alt | -0.298 |
iley | -0.285 |
azeera | -0.279 |
MH | -0.278 |
Token | happened |
Token ID | 3022 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
shoulder | +0.726 |
handc | +0.721 |
Bench | +0.708 |
� | +0.665 |
� | +0.659 |
Neg logit contributions | |
alks | -0.790 |
alt | -0.745 |
iley | -0.708 |
azeera | -0.702 |
MH | -0.686 |
Token | offers |
Token ID | 4394 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
alks | +1.223 |
alt | +1.162 |
iley | +1.107 |
MH | +1.088 |
ic | +1.078 |
Neg logit contributions | |
handc | -1.086 |
shoulder | -1.085 |
Bench | -1.065 |
� | -1.009 |
� | -1.005 |
Token | an |
Token ID | 281 |
Feature activation | +7.73e-01 |
Pos logit contributions | |
alks | +0.412 |
alt | +0.391 |
iley | +0.373 |
MH | +0.364 |
azeera | +0.363 |
Neg logit contributions | |
shoulder | -0.348 |
handc | -0.346 |
Bench | -0.338 |
� | -0.319 |
� | -0.317 |
Token | opportunity |
Token ID | 3663 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
alks | +2.187 |
alt | +2.080 |
iley | +1.980 |
MH | +1.950 |
ic | +1.932 |
Neg logit contributions | |
shoulder | -1.835 |
handc | -1.830 |
Bench | -1.800 |
� | -1.698 |
� | -1.696 |
Token | of |
Token ID | 286 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
alks | +4.094 |
alt | +3.924 |
iley | +3.725 |
MH | +3.667 |
ic | +3.627 |
Neg logit contributions | |
handc | -3.673 |
shoulder | -3.643 |
Bench | -3.560 |
� | -3.416 |
� | -3.393 |
Token | sorts |
Token ID | 10524 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
shoulder | +0.091 |
handc | +0.091 |
Bench | +0.089 |
� | +0.083 |
� | +0.083 |
Neg logit contributions | |
alks | -0.113 |
alt | -0.107 |
iley | -0.102 |
MH | -0.100 |
azeera | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
alks | +1.416 |
alt | +1.350 |
iley | +1.284 |
MH | +1.260 |
ic | +1.248 |
Neg logit contributions | |
handc | -1.263 |
shoulder | -1.259 |
Bench | -1.234 |
� | -1.172 |
� | -1.167 |
Token | There |
Token ID | 1318 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
shoulder | +0.582 |
Bench | +0.577 |
handc | +0.576 |
� | +0.543 |
� | +0.541 |
Neg logit contributions | |
alks | -0.569 |
alt | -0.537 |
iley | -0.510 |
azeera | -0.501 |
MH | -0.493 |
Token | has |
Token ID | 468 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
alks | +0.416 |
alt | +0.394 |
iley | +0.374 |
MH | +0.367 |
ic | +0.362 |
Neg logit contributions | |
handc | -0.402 |
shoulder | -0.402 |
Bench | -0.395 |
� | -0.374 |
� | -0.373 |
Token | cool |
Token ID | 3608 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
shoulder | +0.221 |
handc | +0.221 |
Bench | +0.216 |
� | +0.204 |
� | +0.204 |
Neg logit contributions | |
alks | -0.264 |
alt | -0.251 |
iley | -0.239 |
MH | -0.235 |
ic | -0.232 |
Token | enough |
Token ID | 1576 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
shoulder | +2.004 |
handc | +1.976 |
Bench | +1.940 |
� | +1.833 |
� | +1.828 |
Neg logit contributions | |
alks | -2.150 |
alt | -2.034 |
iley | -1.937 |
MH | -1.908 |
azeera | -1.868 |
Token | to |
Token ID | 284 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
shoulder | +1.764 |
handc | +1.739 |
Bench | +1.701 |
� | +1.596 |
� | +1.595 |
Neg logit contributions | |
alks | -2.105 |
alt | -1.987 |
iley | -1.893 |
MH | -1.873 |
azeera | -1.840 |
Token | catch |
Token ID | 4929 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
shoulder | +0.361 |
handc | +0.359 |
Bench | +0.351 |
� | +0.330 |
� | +0.329 |
Neg logit contributions | |
alks | -0.387 |
alt | -0.366 |
iley | -0.347 |
MH | -0.342 |
azeera | -0.337 |
Token | his |
Token ID | 465 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
alks | +0.837 |
alt | +0.796 |
iley | +0.758 |
MH | +0.744 |
ic | +0.735 |
Neg logit contributions | |
shoulder | -0.700 |
handc | -0.698 |
Bench | -0.685 |
� | -0.645 |
� | -0.643 |
Token | attention |
Token ID | 3241 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
alks | +0.177 |
alt | +0.168 |
iley | +0.160 |
MH | +0.157 |
azeera | +0.155 |
Neg logit contributions | |
shoulder | -0.148 |
handc | -0.147 |
Bench | -0.144 |
� | -0.136 |
� | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
alks | +3.504 |
alt | +3.390 |
iley | +3.179 |
ic | +3.148 |
MH | +3.102 |
Neg logit contributions | |
handc | -4.069 |
Bench | -4.048 |
shoulder | -4.008 |
� | -3.893 |
� | -3.864 |
Token |
|
Token ID | 198 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
alks | +0.519 |
alt | +0.491 |
iley | +0.465 |
MH | +0.455 |
azeera | +0.450 |
Neg logit contributions | |
shoulder | -0.534 |
handc | -0.532 |
Bench | -0.524 |
� | -0.497 |
� | -0.496 |
Token |
|
Token ID | 198 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
alks | +0.564 |
alt | +0.536 |
iley | +0.505 |
MH | +0.499 |
ic | +0.491 |
Neg logit contributions | |
handc | -0.604 |
shoulder | -0.600 |
Bench | -0.586 |
� | -0.559 |
� | -0.557 |
Token | The |
Token ID | 464 |
Feature activation | +8.51e-02 |
Pos logit contributions | |
alks | +0.548 |
alt | +0.517 |
iley | +0.490 |
MH | +0.481 |
ic | +0.475 |
Neg logit contributions | |
shoulder | -0.572 |
handc | -0.570 |
Bench | -0.561 |
� | -0.534 |
� | -0.533 |
Token | truth |
Token ID | 3872 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
alks | +0.234 |
alt | +0.222 |
iley | +0.212 |
MH | +0.207 |
azeera | +0.205 |
Neg logit contributions | |
shoulder | -0.210 |
handc | -0.209 |
Bench | -0.206 |
� | -0.194 |
� | -0.193 |
Token | turn |
Token ID | 1210 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
alks | +0.276 |
alt | +0.261 |
iley | +0.248 |
MH | +0.244 |
azeera | +0.241 |
Neg logit contributions | |
shoulder | -0.250 |
handc | -0.249 |
Bench | -0.244 |
� | -0.230 |
� | -0.230 |
Token | this |
Token ID | 428 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
shoulder | +0.044 |
handc | +0.044 |
Bench | +0.043 |
� | +0.041 |
� | +0.041 |
Neg logit contributions | |
alks | -0.050 |
alt | -0.047 |
iley | -0.045 |
MH | -0.044 |
azeera | -0.043 |
Token | incident |
Token ID | 4519 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
shoulder | +0.251 |
handc | +0.250 |
Bench | +0.245 |
� | +0.232 |
� | +0.232 |
Neg logit contributions | |
alks | -0.256 |
alt | -0.242 |
iley | -0.229 |
MH | -0.225 |
azeera | -0.222 |
Token | into |
Token ID | 656 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
alks | +0.619 |
alt | +0.583 |
iley | +0.550 |
MH | +0.538 |
ic | +0.531 |
Neg logit contributions | |
shoulder | -0.745 |
handc | -0.743 |
Bench | -0.731 |
� | -0.697 |
� | -0.695 |
Token | an |
Token ID | 281 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
shoulder | +0.389 |
handc | +0.388 |
Bench | +0.380 |
� | +0.357 |
� | +0.356 |
Neg logit contributions | |
alks | -0.481 |
alt | -0.458 |
iley | -0.436 |
MH | -0.428 |
azeera | -0.423 |
Token | accident |
Token ID | 5778 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
alks | +0.400 |
alt | +0.378 |
iley | +0.357 |
MH | +0.348 |
azeera | +0.344 |
Neg logit contributions | |
shoulder | -0.442 |
handc | -0.441 |
Bench | -0.433 |
� | -0.411 |
� | -0.410 |
Token | ( |
Token ID | 357 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
alks | +3.642 |
alt | +3.473 |
iley | +3.319 |
MH | +3.306 |
ic | +3.237 |
Neg logit contributions | |
shoulder | -3.975 |
handc | -3.963 |
Bench | -3.915 |
� | -3.782 |
� | -3.769 |
Token | Cost |
Token ID | 13729 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
shoulder | +0.405 |
handc | +0.404 |
Bench | +0.398 |
� | +0.379 |
� | +0.379 |
Neg logit contributions | |
alks | -0.370 |
alt | -0.346 |
iley | -0.329 |
MH | -0.323 |
azeera | -0.319 |
Token | of |
Token ID | 286 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
shoulder | +1.304 |
handc | +1.291 |
Bench | +1.273 |
� | +1.200 |
� | +1.200 |
Neg logit contributions | |
alks | -1.329 |
alt | -1.247 |
iley | -1.185 |
MH | -1.159 |
azeera | -1.156 |
Token | repairs |
Token ID | 20097 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
shoulder | +0.144 |
handc | +0.143 |
Bench | +0.141 |
� | +0.132 |
� | +0.132 |
Neg logit contributions | |
alks | -0.165 |
alt | -0.157 |
iley | -0.150 |
MH | -0.146 |
azeera | -0.145 |
Token | in |
Token ID | 287 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
alks | +0.989 |
alt | +0.939 |
iley | +0.891 |
MH | +0.876 |
ic | +0.865 |
Neg logit contributions | |
shoulder | -0.937 |
handc | -0.935 |
Bench | -0.918 |
� | -0.871 |
� | -0.868 |
Token | %. |
Token ID | 7225 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
shoulder | +0.539 |
handc | +0.537 |
Bench | +0.528 |
� | +0.498 |
� | +0.498 |
Neg logit contributions | |
alks | -0.598 |
alt | -0.566 |
iley | -0.539 |
MH | -0.528 |
azeera | -0.523 |
Token | How |
Token ID | 1374 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
alks | +0.904 |
alt | +0.858 |
iley | +0.811 |
MH | +0.804 |
ic | +0.790 |
Neg logit contributions | |
handc | -0.930 |
shoulder | -0.925 |
Bench | -0.895 |
� | -0.858 |
� | -0.853 |
Token | does |
Token ID | 857 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
shoulder | +0.104 |
handc | +0.104 |
Bench | +0.102 |
� | +0.096 |
� | +0.096 |
Neg logit contributions | |
alks | -0.117 |
alt | -0.110 |
iley | -0.105 |
MH | -0.103 |
azeera | -0.102 |
Token | asset |
Token ID | 11171 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
alks | +1.975 |
alt | +1.880 |
iley | +1.784 |
MH | +1.782 |
ic | +1.756 |
Neg logit contributions | |
shoulder | -1.835 |
handc | -1.833 |
Bench | -1.793 |
� | -1.723 |
� | -1.709 |
Token | ownership |
Token ID | 9238 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
alks | +0.089 |
alt | +0.085 |
iley | +0.080 |
MH | +0.079 |
ic | +0.078 |
Neg logit contributions | |
shoulder | -0.079 |
handc | -0.079 |
Bench | -0.078 |
� | -0.074 |
� | -0.074 |
Token | reflect |
Token ID | 4079 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
alks | +0.248 |
alt | +0.236 |
iley | +0.226 |
MH | +0.222 |
ic | +0.220 |
Neg logit contributions | |
shoulder | -0.188 |
handc | -0.187 |
Bench | -0.184 |
� | -0.173 |
� | -0.173 |
Token | in |
Token ID | 287 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
alks | +4.276 |
alt | +4.111 |
MH | +3.980 |
iley | +3.959 |
ic | +3.921 |
Neg logit contributions | |
handc | -3.217 |
shoulder | -3.143 |
Bench | -3.114 |
� | -2.996 |
� | -2.964 |
Token | terms |
Token ID | 2846 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
alks | +0.547 |
alt | +0.520 |
iley | +0.495 |
MH | +0.487 |
ic | +0.481 |
Neg logit contributions | |
handc | -0.483 |
shoulder | -0.483 |
Bench | -0.473 |
� | -0.447 |
� | -0.446 |
Token | of |
Token ID | 286 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
alks | +1.213 |
alt | +1.160 |
iley | +1.110 |
MH | +1.097 |
ic | +1.080 |
Neg logit contributions | |
handc | -1.116 |
shoulder | -1.109 |
Bench | -1.087 |
� | -1.039 |
� | -1.034 |
Token | income |
Token ID | 3739 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
shoulder | +0.537 |
handc | +0.534 |
Bench | +0.523 |
� | +0.488 |
� | +0.487 |
Neg logit contributions | |
alks | -0.799 |
alt | -0.763 |
iley | -0.729 |
MH | -0.717 |
azeera | -0.710 |
Token | levels |
Token ID | 2974 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
shoulder | +0.609 |
handc | +0.605 |
Bench | +0.595 |
� | +0.558 |
� | +0.558 |
Neg logit contributions | |
alks | -0.767 |
alt | -0.729 |
iley | -0.694 |
MH | -0.681 |
azeera | -0.676 |
Token | , |
Token ID | 11 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
alks | +0.975 |
alt | +0.921 |
iley | +0.872 |
MH | +0.854 |
ic | +0.844 |
Neg logit contributions | |
shoulder | -1.045 |
handc | -1.044 |
Bench | -1.025 |
� | -0.975 |
� | -0.973 |
Token | but |
Token ID | 475 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
alks | +0.775 |
alt | +0.732 |
iley | +0.694 |
MH | +0.686 |
ic | +0.680 |
Neg logit contributions | |
handc | -0.712 |
shoulder | -0.709 |
Bench | -0.698 |
� | -0.664 |
� | -0.663 |
Token | Israeli |
Token ID | 6085 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
alks | +0.059 |
alt | +0.056 |
iley | +0.053 |
MH | +0.052 |
ic | +0.051 |
Neg logit contributions | |
shoulder | -0.054 |
handc | -0.054 |
Bench | -0.053 |
� | -0.050 |
� | -0.050 |
Token | media |
Token ID | 2056 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
alks | +1.103 |
alt | +1.045 |
iley | +0.989 |
MH | +0.974 |
ic | +0.962 |
Neg logit contributions | |
handc | -1.062 |
shoulder | -1.061 |
Bench | -1.041 |
� | -0.989 |
� | -0.989 |
Token | reports |
Token ID | 3136 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
alks | +1.618 |
alt | +1.519 |
iley | +1.428 |
MH | +1.403 |
ic | +1.383 |
Neg logit contributions | |
shoulder | -2.008 |
handc | -2.007 |
Bench | -1.976 |
� | -1.889 |
� | -1.885 |
Token | confirm |
Token ID | 6216 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
alks | +0.891 |
alt | +0.844 |
iley | +0.800 |
MH | +0.787 |
ic | +0.777 |
Neg logit contributions | |
shoulder | -0.877 |
handc | -0.875 |
Bench | -0.861 |
� | -0.818 |
� | -0.816 |
Token | the |
Token ID | 262 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
alks | +3.569 |
alt | +3.304 |
MH | +3.114 |
iley | +3.114 |
ic | +3.090 |
Neg logit contributions | |
handc | -4.026 |
Bench | -3.971 |
shoulder | -3.950 |
� | -3.847 |
� | -3.827 |
Token | meeting |
Token ID | 3249 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
alks | +1.083 |
alt | +1.031 |
iley | +0.983 |
MH | +0.971 |
ic | +0.961 |
Neg logit contributions | |
shoulder | -0.806 |
handc | -0.806 |
Bench | -0.792 |
� | -0.745 |
� | -0.745 |
Token | including |
Token ID | 1390 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
alks | +1.498 |
alt | +1.424 |
iley | +1.343 |
MH | +1.335 |
ic | +1.321 |
Neg logit contributions | |
handc | -1.386 |
shoulder | -1.381 |
Bench | -1.357 |
� | -1.290 |
� | -1.290 |
Token | a |
Token ID | 257 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
alks | +0.631 |
alt | +0.599 |
iley | +0.570 |
MH | +0.560 |
ic | +0.554 |
Neg logit contributions | |
shoulder | -0.559 |
handc | -0.558 |
Bench | -0.547 |
� | -0.518 |
� | -0.517 |
Token | number |
Token ID | 1271 |
Feature activation | +9.35e-02 |
Pos logit contributions | |
alks | +0.155 |
alt | +0.147 |
iley | +0.141 |
MH | +0.138 |
azeera | +0.137 |
Neg logit contributions | |
shoulder | -0.130 |
handc | -0.129 |
Bench | -0.126 |
� | -0.118 |
� | -0.118 |
Token | is |
Token ID | 318 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
alks | +0.597 |
alt | +0.566 |
iley | +0.537 |
MH | +0.526 |
ic | +0.521 |
Neg logit contributions | |
handc | -0.580 |
shoulder | -0.578 |
Bench | -0.569 |
� | -0.539 |
� | -0.537 |
Token | no |
Token ID | 645 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
alks | +0.425 |
alt | +0.404 |
iley | +0.384 |
MH | +0.378 |
ic | +0.373 |
Neg logit contributions | |
shoulder | -0.375 |
handc | -0.374 |
Bench | -0.367 |
� | -0.347 |
� | -0.346 |
Token | one |
Token ID | 530 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
alks | +0.737 |
alt | +0.704 |
iley | +0.673 |
MH | +0.662 |
azeera | +0.655 |
Neg logit contributions | |
shoulder | -0.506 |
handc | -0.504 |
Bench | -0.492 |
� | -0.460 |
� | -0.459 |
Token | I |
Token ID | 314 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
alks | +0.482 |
alt | +0.458 |
iley | +0.436 |
MH | +0.428 |
ic | +0.422 |
Neg logit contributions | |
shoulder | -0.426 |
handc | -0.425 |
Bench | -0.417 |
� | -0.393 |
� | -0.393 |
Token | would |
Token ID | 561 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
alks | +0.442 |
alt | +0.420 |
iley | +0.399 |
MH | +0.391 |
ic | +0.388 |
Neg logit contributions | |
shoulder | -0.408 |
handc | -0.408 |
Bench | -0.401 |
� | -0.379 |
� | -0.378 |
Token | rather |
Token ID | 2138 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
alks | +0.987 |
alt | +0.939 |
iley | +0.888 |
MH | +0.872 |
ic | +0.867 |
Neg logit contributions | |
shoulder | -0.955 |
handc | -0.954 |
Bench | -0.944 |
� | -0.901 |
� | -0.897 |
Token | be |
Token ID | 307 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
alks | +3.626 |
alt | +3.445 |
iley | +3.303 |
ic | +3.243 |
MH | +3.209 |
Neg logit contributions | |
Bench | -3.714 |
handc | -3.701 |
shoulder | -3.672 |
� | -3.623 |
� | -3.545 |
Token | working |
Token ID | 1762 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
alks | +0.597 |
alt | +0.569 |
iley | +0.545 |
MH | +0.537 |
ic | +0.532 |
Neg logit contributions | |
shoulder | -0.486 |
handc | -0.486 |
Bench | -0.479 |
� | -0.455 |
� | -0.451 |
Token | alongside |
Token ID | 7848 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
shoulder | +0.483 |
handc | +0.482 |
Bench | +0.471 |
� | +0.441 |
� | +0.440 |
Neg logit contributions | |
alks | -0.705 |
alt | -0.673 |
iley | -0.644 |
MH | -0.634 |
ic | -0.627 |
Token | . |
Token ID | 13 |
Feature activation | +6.60e-02 |
Pos logit contributions | |
alks | +0.751 |
alt | +0.711 |
iley | +0.672 |
MH | +0.659 |
ic | +0.652 |
Neg logit contributions | |
handc | -0.844 |
shoulder | -0.843 |
Bench | -0.830 |
� | -0.794 |
� | -0.791 |
Token |
|
Token ID | 198 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
alks | +0.169 |
alt | +0.160 |
iley | +0.151 |
MH | +0.149 |
ic | +0.147 |
Neg logit contributions | |
handc | -0.175 |
shoulder | -0.175 |
Bench | -0.171 |
� | -0.162 |
� | -0.162 |
Token | up |
Token ID | 510 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
alks | +0.315 |
alt | +0.298 |
iley | +0.283 |
MH | +0.277 |
azeera | +0.274 |
Neg logit contributions | |
shoulder | -0.300 |
handc | -0.298 |
Bench | -0.292 |
� | -0.275 |
� | -0.275 |
Token | and |
Token ID | 290 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
alks | +1.271 |
alt | +1.215 |
iley | +1.151 |
MH | +1.124 |
ic | +1.117 |
Neg logit contributions | |
handc | -1.238 |
shoulder | -1.226 |
Bench | -1.218 |
� | -1.167 |
� | -1.154 |
Token | figure |
Token ID | 3785 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
shoulder | +0.430 |
handc | +0.429 |
Bench | +0.421 |
� | +0.397 |
� | +0.397 |
Neg logit contributions | |
alks | -0.448 |
alt | -0.425 |
iley | -0.403 |
MH | -0.396 |
azeera | -0.390 |
Token | out |
Token ID | 503 |
Feature activation | -8.22e-03 |
Pos logit contributions | |
shoulder | +1.086 |
handc | +1.073 |
Bench | +1.054 |
� | +0.996 |
� | +0.993 |
Neg logit contributions | |
alks | -1.096 |
alt | -1.039 |
iley | -0.984 |
azeera | -0.960 |
MH | -0.958 |
Token | how |
Token ID | 703 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
shoulder | +0.019 |
handc | +0.019 |
Bench | +0.019 |
� | +0.018 |
� | +0.018 |
Neg logit contributions | |
alks | -0.023 |
alt | -0.022 |
iley | -0.021 |
MH | -0.021 |
azeera | -0.021 |
Token | long |
Token ID | 890 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
shoulder | +0.572 |
handc | +0.565 |
Bench | +0.555 |
� | +0.522 |
� | +0.521 |
Neg logit contributions | |
alks | -0.628 |
alt | -0.592 |
iley | -0.562 |
azeera | -0.551 |
MH | -0.550 |
Token | it |
Token ID | 340 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
alks | +3.814 |
alt | +3.614 |
iley | +3.431 |
MH | +3.414 |
ic | +3.383 |
Neg logit contributions | |
handc | -3.690 |
shoulder | -3.624 |
Bench | -3.603 |
� | -3.448 |
� | -3.421 |
Token | took |
Token ID | 1718 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
alks | +0.975 |
alt | +0.927 |
iley | +0.877 |
MH | +0.860 |
ic | +0.856 |
Neg logit contributions | |
handc | -0.926 |
shoulder | -0.924 |
Bench | -0.911 |
� | -0.865 |
� | -0.862 |
Token | the |
Token ID | 262 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
alks | +1.451 |
alt | +1.386 |
iley | +1.314 |
MH | +1.303 |
ic | +1.286 |
Neg logit contributions | |
handc | -1.235 |
shoulder | -1.228 |
Bench | -1.208 |
� | -1.147 |
� | -1.142 |
Token | dent |
Token ID | 18794 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
alks | +1.105 |
alt | +1.047 |
iley | +0.993 |
MH | +0.975 |
ic | +0.966 |
Neg logit contributions | |
handc | -1.098 |
shoulder | -1.097 |
Bench | -1.078 |
� | -1.025 |
� | -1.022 |
Token | ition |
Token ID | 653 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
alks | +0.835 |
alt | +0.783 |
iley | +0.735 |
MH | +0.719 |
ic | +0.706 |
Neg logit contributions | |
shoulder | -1.119 |
handc | -1.116 |
Bench | -1.099 |
� | -1.049 |
� | -1.047 |
Token | c |
Token ID | 66 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
alks | +1.332 |
alt | +1.255 |
iley | +1.182 |
MH | +1.158 |
ic | +1.146 |
Neg logit contributions | |
shoulder | -1.533 |
handc | -1.530 |
Bench | -1.505 |
� | -1.436 |
� | -1.431 |
Token | has |
Token ID | 468 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
alks | +1.769 |
alt | +1.675 |
iley | +1.587 |
MH | +1.562 |
ic | +1.543 |
Neg logit contributions | |
handc | -1.726 |
shoulder | -1.725 |
Bench | -1.692 |
� | -1.611 |
� | -1.607 |
Token | positive |
Token ID | 3967 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
alks | +0.541 |
alt | +0.513 |
iley | +0.491 |
MH | +0.480 |
azeera | +0.476 |
Neg logit contributions | |
shoulder | -0.458 |
handc | -0.455 |
Bench | -0.446 |
� | -0.418 |
� | -0.418 |
Token | implications |
Token ID | 10939 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
alks | +0.702 |
alt | +0.662 |
iley | +0.629 |
MH | +0.614 |
azeera | +0.607 |
Neg logit contributions | |
shoulder | -0.725 |
handc | -0.721 |
Bench | -0.708 |
� | -0.669 |
� | -0.665 |
Token | for |
Token ID | 329 |
Feature activation | -3.80e-04 |
Pos logit contributions | |
alks | +0.424 |
alt | +0.404 |
iley | +0.385 |
MH | +0.378 |
ic | +0.374 |
Neg logit contributions | |
shoulder | -0.340 |
handc | -0.340 |
Bench | -0.332 |
� | -0.313 |
� | -0.312 |
Token | life |
Token ID | 1204 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
shoulder | +0.001 |
handc | +0.001 |
Bench | +0.001 |
� | +0.001 |
� | +0.001 |
Neg logit contributions | |
alks | -0.001 |
alt | -0.001 |
iley | -0.001 |
azeera | -0.001 |
MH | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
alks | +3.464 |
alt | +3.301 |
iley | +3.121 |
MH | +3.097 |
ic | +3.063 |
Neg logit contributions | |
handc | -3.925 |
shoulder | -3.903 |
Bench | -3.811 |
� | -3.683 |
� | -3.612 |
Token | � |
Token ID | 447 |
Feature activation | +3.10e-03 |
Pos logit contributions | |
alks | +0.422 |
alt | +0.401 |
iley | +0.381 |
MH | +0.374 |
ic | +0.369 |
Neg logit contributions | |
shoulder | -0.384 |
handc | -0.383 |
Bench | -0.376 |
� | -0.355 |
� | -0.355 |
Token | � |
Token ID | 251 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
alks | +0.008 |
alt | +0.008 |
iley | +0.007 |
MH | +0.007 |
ic | +0.007 |
Neg logit contributions | |
shoulder | -0.008 |
handc | -0.008 |
Bench | -0.008 |
� | -0.007 |
� | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
alks | +0.100 |
alt | +0.094 |
iley | +0.090 |
MH | +0.088 |
azeera | +0.088 |
Neg logit contributions | |
shoulder | -0.086 |
handc | -0.086 |
Bench | -0.085 |
� | -0.080 |
� | -0.080 |
Token |
|
Token ID | 198 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
shoulder | +0.875 |
handc | +0.870 |
Bench | +0.859 |
� | +0.812 |
� | +0.811 |
Neg logit contributions | |
alks | -0.907 |
alt | -0.857 |
iley | -0.815 |
MH | -0.797 |
azeera | -0.792 |
Token | our |
Token ID | 674 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
alks | +0.689 |
alt | +0.656 |
iley | +0.624 |
MH | +0.615 |
ic | +0.608 |
Neg logit contributions | |
shoulder | -0.596 |
handc | -0.596 |
Bench | -0.584 |
� | -0.554 |
� | -0.554 |
Token | hard |
Token ID | 1327 |
Feature activation | -9.13e-02 |
Pos logit contributions | |
alks | +0.504 |
alt | +0.481 |
iley | +0.460 |
MH | +0.453 |
ic | +0.448 |
Neg logit contributions | |
handc | -0.375 |
shoulder | -0.375 |
Bench | -0.367 |
� | -0.346 |
� | -0.345 |
Token | work |
Token ID | 670 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
shoulder | +0.169 |
handc | +0.168 |
Bench | +0.164 |
� | +0.151 |
� | +0.151 |
Neg logit contributions | |
alks | -0.308 |
alt | -0.295 |
iley | -0.283 |
MH | -0.279 |
azeera | -0.276 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
shoulder | +1.410 |
handc | +1.395 |
Bench | +1.375 |
� | +1.290 |
� | +1.290 |
Neg logit contributions | |
alks | -1.502 |
alt | -1.413 |
iley | -1.347 |
MH | -1.319 |
azeera | -1.308 |
Token | Why |
Token ID | 4162 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
alks | +0.281 |
alt | +0.266 |
iley | +0.252 |
MH | +0.247 |
azeera | +0.245 |
Neg logit contributions | |
shoulder | -0.274 |
handc | -0.273 |
Bench | -0.270 |
� | -0.254 |
� | -0.254 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
alks | +1.458 |
alt | +1.382 |
iley | +1.312 |
MH | +1.288 |
ic | +1.272 |
Neg logit contributions | |
shoulder | -1.364 |
handc | -1.360 |
Bench | -1.335 |
� | -1.262 |
� | -1.259 |
Token | 't |
Token ID | 470 |
Feature activation | +7.77e-01 |
Pos logit contributions | |
alks | +2.604 |
alt | +2.507 |
iley | +2.330 |
MH | +2.268 |
ic | +2.246 |
Neg logit contributions | |
handc | -4.857 |
shoulder | -4.782 |
Bench | -4.631 |
stitching | -4.527 |
� | -4.476 |
Token | we |
Token ID | 356 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
alks | +2.035 |
alt | +1.930 |
iley | +1.834 |
MH | +1.808 |
ic | +1.785 |
Neg logit contributions | |
handc | -1.989 |
shoulder | -1.988 |
Bench | -1.949 |
� | -1.859 |
� | -1.855 |
Token | do |
Token ID | 466 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
alks | +1.827 |
alt | +1.743 |
iley | +1.654 |
MH | +1.629 |
ic | +1.612 |
Neg logit contributions | |
handc | -1.713 |
shoulder | -1.710 |
Bench | -1.681 |
� | -1.610 |
� | -1.607 |
Token | this |
Token ID | 428 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
shoulder | +0.126 |
handc | +0.125 |
Bench | +0.123 |
� | +0.115 |
� | +0.115 |
Neg logit contributions | |
alks | -0.149 |
alt | -0.142 |
iley | -0.135 |
MH | -0.132 |
azeera | -0.131 |
Token | hike |
Token ID | 18390 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
alks | +0.835 |
alt | +0.793 |
iley | +0.754 |
MH | +0.741 |
ic | +0.732 |
Neg logit contributions | |
shoulder | -0.764 |
handc | -0.763 |
Bench | -0.749 |
� | -0.710 |
� | -0.708 |
Token | him |
Token ID | 683 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
alks | +0.384 |
alt | +0.365 |
iley | +0.348 |
MH | +0.343 |
ic | +0.339 |
Neg logit contributions | |
handc | -0.323 |
shoulder | -0.323 |
Bench | -0.317 |
� | -0.299 |
� | -0.299 |
Token | ; |
Token ID | 26 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
shoulder | +0.609 |
handc | +0.606 |
Bench | +0.596 |
� | +0.563 |
� | +0.562 |
Neg logit contributions | |
alks | -0.622 |
alt | -0.588 |
iley | -0.559 |
MH | -0.548 |
azeera | -0.541 |
Token | Despite |
Token ID | 7945 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
shoulder | +0.453 |
handc | +0.451 |
Bench | +0.443 |
� | +0.417 |
� | +0.416 |
Neg logit contributions | |
alks | -0.529 |
alt | -0.503 |
iley | -0.479 |
MH | -0.470 |
ic | -0.465 |
Token | taking |
Token ID | 2263 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
shoulder | +1.539 |
handc | +1.517 |
Bench | +1.490 |
� | +1.383 |
stitching | +1.379 |
Neg logit contributions | |
alks | -1.970 |
alt | -1.873 |
iley | -1.785 |
azeera | -1.758 |
MH | -1.754 |
Token | a |
Token ID | 257 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
alks | +0.402 |
alt | +0.382 |
iley | +0.364 |
MH | +0.356 |
azeera | +0.353 |
Neg logit contributions | |
shoulder | -0.337 |
handc | -0.334 |
Bench | -0.328 |
� | -0.309 |
� | -0.307 |
Token | career |
Token ID | 3451 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
alks | +0.735 |
alt | +0.700 |
iley | +0.668 |
MH | +0.657 |
ic | +0.649 |
Neg logit contributions | |
shoulder | -0.583 |
handc | -0.582 |
Bench | -0.571 |
� | -0.537 |
� | -0.536 |
Token | highest |
Token ID | 4511 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
alks | +3.723 |
alt | +3.641 |
iley | +3.455 |
ic | +3.378 |
MH | +3.376 |
Neg logit contributions | |
handc | -3.592 |
shoulder | -3.511 |
Bench | -3.504 |
� | -3.370 |
� | -3.364 |
Token | second |
Token ID | 1218 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
alks | +1.014 |
alt | +0.975 |
iley | +0.930 |
MH | +0.917 |
ic | +0.909 |
Neg logit contributions | |
handc | -0.728 |
shoulder | -0.727 |
Bench | -0.713 |
� | -0.673 |
� | -0.672 |
Token | place |
Token ID | 1295 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
alks | +0.274 |
alt | +0.261 |
iley | +0.248 |
MH | +0.244 |
ic | +0.241 |
Neg logit contributions | |
shoulder | -0.226 |
handc | -0.226 |
Bench | -0.221 |
� | -0.208 |
� | -0.208 |
Token | along |
Token ID | 1863 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
alks | +0.362 |
alt | +0.344 |
iley | +0.328 |
MH | +0.323 |
ic | +0.319 |
Neg logit contributions | |
shoulder | -0.289 |
handc | -0.288 |
Bench | -0.283 |
� | -0.266 |
� | -0.265 |
Token | with |
Token ID | 351 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
alks | +1.069 |
alt | +1.019 |
iley | +0.975 |
MH | +0.961 |
ic | +0.952 |
Neg logit contributions | |
handc | -0.765 |
shoulder | -0.760 |
Bench | -0.743 |
� | -0.703 |
� | -0.699 |
Token | , |
Token ID | 11 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
alks | +0.963 |
alt | +0.914 |
iley | +0.868 |
MH | +0.852 |
ic | +0.842 |
Neg logit contributions | |
shoulder | -0.910 |
handc | -0.908 |
Bench | -0.892 |
� | -0.844 |
� | -0.843 |
Token | including |
Token ID | 1390 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
alks | +0.125 |
alt | +0.119 |
iley | +0.114 |
MH | +0.112 |
azeera | +0.111 |
Neg logit contributions | |
shoulder | -0.085 |
handc | -0.084 |
Bench | -0.082 |
� | -0.077 |
� | -0.076 |
Token | an |
Token ID | 281 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
shoulder | +0.918 |
handc | +0.914 |
Bench | +0.894 |
stitching | +0.834 |
� | +0.827 |
Neg logit contributions | |
alks | -1.191 |
alt | -1.125 |
iley | -1.079 |
azeera | -1.056 |
MH | -1.045 |
Token | image |
Token ID | 2939 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
shoulder | +0.253 |
handc | +0.252 |
Bench | +0.245 |
� | +0.229 |
� | +0.228 |
Neg logit contributions | |
alks | -0.328 |
alt | -0.309 |
iley | -0.298 |
azeera | -0.290 |
MH | -0.288 |
Token | falsely |
Token ID | 24566 |
Feature activation | -3.58e-03 |
Pos logit contributions | |
alks | +0.038 |
alt | +0.036 |
iley | +0.034 |
MH | +0.033 |
azeera | +0.033 |
Neg logit contributions | |
shoulder | -0.032 |
handc | -0.031 |
Bench | -0.031 |
� | -0.029 |
� | -0.029 |
Token | suggesting |
Token ID | 9524 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
shoulder | +0.008 |
handc | +0.008 |
Bench | +0.008 |
� | +0.008 |
� | +0.008 |
Neg logit contributions | |
alks | -0.010 |
alt | -0.010 |
iley | -0.009 |
MH | -0.009 |
azeera | -0.009 |
Token | Kurt |
Token ID | 20642 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
alks | +3.461 |
alt | +3.282 |
iley | +3.119 |
MH | +3.107 |
ic | +3.027 |
Neg logit contributions | |
handc | -3.950 |
shoulder | -3.931 |
Bench | -3.890 |
� | -3.728 |
� | -3.717 |
Token | Cob |
Token ID | 14828 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
alks | +0.734 |
alt | +0.690 |
iley | +0.656 |
MH | +0.639 |
ic | +0.633 |
Neg logit contributions | |
shoulder | -0.778 |
handc | -0.773 |
Bench | -0.763 |
� | -0.721 |
� | -0.718 |
Token | ain |
Token ID | 391 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
alks | +0.316 |
alt | +0.284 |
iley | +0.279 |
MH | +0.278 |
azeera | +0.277 |
Neg logit contributions | |
shoulder | -0.351 |
handc | -0.350 |
Bench | -0.346 |
� | -0.323 |
� | -0.321 |
Token | had |
Token ID | 550 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
alks | +0.153 |
alt | +0.146 |
iley | +0.138 |
MH | +0.136 |
azeera | +0.136 |
Neg logit contributions | |
shoulder | -0.129 |
handc | -0.128 |
Bench | -0.126 |
� | -0.119 |
� | -0.118 |
Token | predicted |
Token ID | 11001 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
alks | +0.452 |
alt | +0.429 |
iley | +0.407 |
azeera | +0.397 |
MH | +0.395 |
Neg logit contributions | |
shoulder | -0.389 |
handc | -0.384 |
Bench | -0.376 |
� | -0.354 |
� | -0.353 |
Token | Square |
Token ID | 9276 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
alks | +0.462 |
alt | +0.438 |
iley | +0.417 |
MH | +0.409 |
ic | +0.404 |
Neg logit contributions | |
shoulder | -0.418 |
handc | -0.417 |
Bench | -0.410 |
� | -0.387 |
� | -0.386 |
Token | Enix |
Token ID | 41361 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
alks | +0.355 |
alt | +0.337 |
iley | +0.323 |
MH | +0.316 |
azeera | +0.314 |
Neg logit contributions | |
shoulder | -0.278 |
handc | -0.276 |
Bench | -0.272 |
� | -0.255 |
� | -0.254 |
Token | has |
Token ID | 468 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
alks | +1.055 |
alt | +1.000 |
iley | +0.953 |
MH | +0.933 |
azeera | +0.924 |
Neg logit contributions | |
shoulder | -0.956 |
handc | -0.951 |
Bench | -0.935 |
� | -0.881 |
� | -0.878 |
Token | yet |
Token ID | 1865 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
alks | +1.184 |
alt | +1.125 |
iley | +1.069 |
MH | +1.053 |
ic | +1.041 |
Neg logit contributions | |
handc | -1.046 |
shoulder | -1.046 |
Bench | -1.026 |
� | -0.971 |
� | -0.969 |
Token | to |
Token ID | 284 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
alks | +0.817 |
alt | +0.775 |
iley | +0.739 |
MH | +0.726 |
ic | +0.720 |
Neg logit contributions | |
handc | -0.722 |
shoulder | -0.718 |
Bench | -0.712 |
� | -0.675 |
� | -0.674 |
Token | confirm |
Token ID | 6216 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
alks | +0.525 |
alt | +0.499 |
iley | +0.475 |
MH | +0.466 |
ic | +0.460 |
Neg logit contributions | |
shoulder | -0.462 |
handc | -0.460 |
Bench | -0.452 |
� | -0.426 |
� | -0.425 |
Token | a |
Token ID | 257 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
alks | +3.904 |
alt | +3.707 |
MH | +3.537 |
iley | +3.498 |
ic | +3.450 |
Neg logit contributions | |
handc | -3.500 |
shoulder | -3.433 |
Bench | -3.387 |
� | -3.315 |
� | -3.247 |
Token | release |
Token ID | 2650 |
Feature activation | +9.99e-02 |
Pos logit contributions | |
alks | +1.038 |
alt | +0.986 |
iley | +0.936 |
MH | +0.922 |
ic | +0.910 |
Neg logit contributions | |
shoulder | -0.934 |
handc | -0.933 |
Bench | -0.916 |
� | -0.867 |
� | -0.866 |
Token | date |
Token ID | 3128 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
alks | +0.269 |
alt | +0.254 |
iley | +0.242 |
MH | +0.236 |
azeera | +0.235 |
Neg logit contributions | |
shoulder | -0.253 |
handc | -0.252 |
Bench | -0.248 |
� | -0.234 |
� | -0.232 |
Token | for |
Token ID | 329 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
alks | +1.340 |
alt | +1.276 |
iley | +1.202 |
MH | +1.188 |
ic | +1.175 |
Neg logit contributions | |
shoulder | -1.292 |
handc | -1.289 |
Bench | -1.271 |
� | -1.219 |
� | -1.214 |
Token | either |
Token ID | 2035 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
alks | +0.648 |
alt | +0.615 |
iley | +0.585 |
MH | +0.576 |
ic | +0.569 |
Neg logit contributions | |
shoulder | -0.565 |
handc | -0.564 |
Bench | -0.553 |
� | -0.523 |
� | -0.522 |
Token | ," |
Token ID | 553 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
shoulder | +0.395 |
handc | +0.393 |
Bench | +0.385 |
� | +0.363 |
� | +0.362 |
Neg logit contributions | |
alks | -0.441 |
alt | -0.418 |
iley | -0.397 |
MH | -0.390 |
azeera | -0.385 |
Token | he |
Token ID | 339 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
alks | +0.403 |
alt | +0.380 |
iley | +0.358 |
MH | +0.351 |
ic | +0.346 |
Neg logit contributions | |
shoulder | -0.469 |
handc | -0.468 |
Bench | -0.461 |
� | -0.439 |
� | -0.438 |
Token | said |
Token ID | 531 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
alks | +0.706 |
alt | +0.673 |
iley | +0.640 |
MH | +0.628 |
ic | +0.622 |
Neg logit contributions | |
shoulder | -0.594 |
handc | -0.592 |
Bench | -0.582 |
� | -0.549 |
� | -0.548 |
Token | . |
Token ID | 13 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
alks | +0.581 |
alt | +0.551 |
iley | +0.522 |
MH | +0.512 |
ic | +0.506 |
Neg logit contributions | |
shoulder | -0.592 |
handc | -0.590 |
Bench | -0.581 |
� | -0.552 |
� | -0.551 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
alks | +0.746 |
alt | +0.707 |
iley | +0.669 |
MH | +0.657 |
ic | +0.648 |
Neg logit contributions | |
shoulder | -0.752 |
handc | -0.750 |
Bench | -0.736 |
� | -0.699 |
� | -0.698 |
Token | Meg |
Token ID | 42672 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
alks | +0.130 |
alt | +0.122 |
iley | +0.116 |
MH | +0.114 |
ic | +0.112 |
Neg logit contributions | |
shoulder | -0.137 |
handc | -0.136 |
Bench | -0.134 |
� | -0.128 |
� | -0.128 |
Token | ach |
Token ID | 620 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
alks | +3.709 |
alt | +3.565 |
iley | +3.361 |
MH | +3.285 |
ic | +3.238 |
Neg logit contributions | |
handc | -3.586 |
shoulder | -3.567 |
Bench | -3.439 |
� | -3.340 |
� | -3.305 |
Token | urch |
Token ID | 2575 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
alks | +0.211 |
alt | +0.201 |
iley | +0.192 |
MH | +0.189 |
ic | +0.187 |
Neg logit contributions | |
shoulder | -0.172 |
handc | -0.170 |
Bench | -0.169 |
� | -0.157 |
� | -0.157 |
Token | founder |
Token ID | 9119 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
alks | +1.844 |
alt | +1.749 |
iley | +1.658 |
MH | +1.630 |
ic | +1.608 |
Neg logit contributions | |
shoulder | -1.796 |
handc | -1.791 |
Bench | -1.755 |
� | -1.670 |
� | -1.665 |
Token | Rev |
Token ID | 5416 |
Feature activation | -6.19e-02 |
Pos logit contributions | |
shoulder | +0.895 |
handc | +0.890 |
Bench | +0.884 |
� | +0.825 |
� | +0.823 |
Neg logit contributions | |
alks | -0.899 |
alt | -0.849 |
iley | -0.803 |
azeera | -0.786 |
MH | -0.782 |
Token | . |
Token ID | 13 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
shoulder | +0.175 |
handc | +0.174 |
Bench | +0.172 |
� | +0.164 |
� | +0.163 |
Neg logit contributions | |
alks | -0.147 |
alt | -0.138 |
iley | -0.130 |
MH | -0.128 |
ic | -0.126 |
Token | disappointed |
Token ID | 11679 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
alks | +0.888 |
alt | +0.846 |
iley | +0.808 |
MH | +0.794 |
ic | +0.786 |
Neg logit contributions | |
shoulder | -0.705 |
handc | -0.704 |
Bench | -0.690 |
� | -0.651 |
� | -0.648 |
Token | but |
Token ID | 475 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
alks | +0.822 |
alt | +0.780 |
iley | +0.741 |
MH | +0.727 |
ic | +0.719 |
Neg logit contributions | |
handc | -0.797 |
shoulder | -0.794 |
Bench | -0.780 |
� | -0.739 |
� | -0.737 |
Token | Canadians |
Token ID | 14008 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
alks | +0.310 |
alt | +0.295 |
iley | +0.281 |
MH | +0.276 |
ic | +0.273 |
Neg logit contributions | |
shoulder | -0.254 |
handc | -0.253 |
Bench | -0.248 |
� | -0.234 |
� | -0.233 |
Token | as |
Token ID | 355 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
shoulder | +0.350 |
handc | +0.349 |
Bench | +0.342 |
� | +0.322 |
� | +0.322 |
Neg logit contributions | |
alks | -0.429 |
alt | -0.408 |
iley | -0.389 |
MH | -0.383 |
ic | -0.378 |
Token | a |
Token ID | 257 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
alks | +0.633 |
alt | +0.607 |
iley | +0.582 |
MH | +0.573 |
ic | +0.568 |
Neg logit contributions | |
handc | -0.393 |
shoulder | -0.391 |
Bench | -0.383 |
� | -0.359 |
� | -0.354 |
Token | whole |
Token ID | 2187 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
alks | +1.056 |
alt | +1.007 |
iley | +0.957 |
MH | +0.941 |
ic | +0.933 |
Neg logit contributions | |
handc | -0.882 |
shoulder | -0.881 |
Bench | -0.864 |
� | -0.817 |
� | -0.814 |
Token | are |
Token ID | 389 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
alks | +3.449 |
alt | +3.346 |
iley | +3.190 |
ic | +3.102 |
MH | +3.075 |
Neg logit contributions | |
handc | -3.649 |
Bench | -3.598 |
shoulder | -3.570 |
� | -3.424 |
� | -3.378 |
Token | , |
Token ID | 11 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
alks | +1.329 |
alt | +1.270 |
iley | +1.217 |
MH | +1.192 |
ic | +1.184 |
Neg logit contributions | |
handc | -1.116 |
shoulder | -1.113 |
Bench | -1.098 |
� | -1.039 |
� | -1.032 |
Token | � |
Token ID | 447 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
alks | +0.657 |
alt | +0.629 |
iley | +0.600 |
MH | +0.590 |
ic | +0.586 |
Neg logit contributions | |
handc | -0.509 |
shoulder | -0.507 |
Bench | -0.498 |
� | -0.470 |
� | -0.467 |
Token | � |
Token ID | 251 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
alks | +0.083 |
alt | +0.079 |
iley | +0.075 |
MH | +0.074 |
azeera | +0.073 |
Neg logit contributions | |
shoulder | -0.070 |
handc | -0.070 |
Bench | -0.069 |
� | -0.065 |
� | -0.065 |
Token | said |
Token ID | 531 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
shoulder | +0.064 |
handc | +0.064 |
Bench | +0.062 |
� | +0.059 |
� | +0.059 |
Neg logit contributions | |
alks | -0.072 |
alt | -0.069 |
iley | -0.065 |
MH | -0.064 |
ic | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
shoulder | +0.649 |
handc | +0.641 |
Bench | +0.634 |
� | +0.596 |
� | +0.593 |
Neg logit contributions | |
alks | -0.646 |
alt | -0.609 |
iley | -0.576 |
azeera | -0.562 |
MH | -0.560 |
Token | general |
Token ID | 2276 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
shoulder | +0.814 |
handc | +0.807 |
Bench | +0.797 |
� | +0.754 |
� | +0.750 |
Neg logit contributions | |
alks | -0.725 |
alt | -0.681 |
iley | -0.644 |
MH | -0.625 |
azeera | -0.620 |
Token | public |
Token ID | 1171 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
alks | +0.867 |
alt | +0.818 |
iley | +0.774 |
MH | +0.758 |
ic | +0.748 |
Neg logit contributions | |
shoulder | -0.946 |
handc | -0.944 |
Bench | -0.928 |
� | -0.882 |
� | -0.880 |
Token | to |
Token ID | 284 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
alks | +0.029 |
alt | +0.027 |
iley | +0.026 |
MH | +0.025 |
ic | +0.025 |
Neg logit contributions | |
shoulder | -0.027 |
handc | -0.027 |
Bench | -0.027 |
� | -0.025 |
� | -0.025 |
Token | its |
Token ID | 663 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
alks | +0.205 |
alt | +0.195 |
iley | +0.186 |
MH | +0.183 |
ic | +0.181 |
Neg logit contributions | |
shoulder | -0.160 |
handc | -0.160 |
Bench | -0.157 |
� | -0.148 |
� | -0.148 |
Token | detriment |
Token ID | 42959 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
alks | +0.723 |
alt | +0.692 |
iley | +0.657 |
MH | +0.648 |
ic | +0.644 |
Neg logit contributions | |
handc | -0.649 |
shoulder | -0.646 |
Bench | -0.638 |
� | -0.603 |
� | -0.602 |
Token | . |
Token ID | 13 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
alks | +3.301 |
alt | +3.182 |
iley | +3.010 |
ic | +2.909 |
MH | +2.896 |
Neg logit contributions | |
handc | -3.845 |
shoulder | -3.798 |
Bench | -3.774 |
� | -3.605 |
� | -3.566 |
Token |
|
Token ID | 198 |
Feature activation | -9.22e-02 |
Pos logit contributions | |
alks | +0.143 |
alt | +0.135 |
iley | +0.128 |
MH | +0.126 |
ic | +0.124 |
Neg logit contributions | |
shoulder | -0.145 |
handc | -0.145 |
Bench | -0.142 |
� | -0.135 |
� | -0.135 |
Token |
|
Token ID | 198 |
Feature activation | -8.82e-02 |
Pos logit contributions | |
shoulder | +0.240 |
handc | +0.240 |
Bench | +0.235 |
� | +0.223 |
� | +0.223 |
Neg logit contributions | |
alks | -0.240 |
alt | -0.227 |
iley | -0.215 |
MH | -0.211 |
ic | -0.209 |
Token | Last |
Token ID | 5956 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
shoulder | +0.234 |
handc | +0.232 |
Bench | +0.230 |
� | +0.219 |
� | +0.219 |
Neg logit contributions | |
alks | -0.225 |
alt | -0.212 |
iley | -0.201 |
MH | -0.198 |
azeera | -0.196 |
Token | night |
Token ID | 1755 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
alks | +1.112 |
alt | +1.064 |
iley | +1.019 |
MH | +1.005 |
ic | +0.996 |
Neg logit contributions | |
shoulder | -0.701 |
handc | -0.700 |
Bench | -0.683 |
� | -0.639 |
� | -0.637 |
Token |
|
Token ID | 198 |
Feature activation | +9.09e-02 |
Pos logit contributions | |
alks | +0.116 |
alt | +0.110 |
iley | +0.104 |
MH | +0.102 |
ic | +0.101 |
Neg logit contributions | |
handc | -0.118 |
shoulder | -0.118 |
Bench | -0.116 |
� | -0.110 |
� | -0.109 |
Token | pur |
Token ID | 14225 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
alks | +0.226 |
alt | +0.213 |
iley | +0.202 |
MH | +0.198 |
ic | +0.195 |
Neg logit contributions | |
shoulder | -0.247 |
handc | -0.247 |
Bench | -0.243 |
� | -0.231 |
� | -0.230 |
Token | ak |
Token ID | 461 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
shoulder | +0.218 |
handc | +0.217 |
Bench | +0.214 |
� | +0.204 |
� | +0.204 |
Neg logit contributions | |
alks | -0.158 |
alt | -0.146 |
iley | -0.137 |
MH | -0.135 |
ic | -0.133 |
Token | ushi |
Token ID | 17731 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
shoulder | +0.566 |
handc | +0.564 |
Bench | +0.556 |
� | +0.528 |
� | +0.527 |
Neg logit contributions | |
alks | -0.499 |
alt | -0.470 |
iley | -0.444 |
MH | -0.436 |
ic | -0.429 |
Token | Profile |
Token ID | 13118 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
alks | +0.294 |
alt | +0.277 |
iley | +0.262 |
MH | +0.257 |
ic | +0.253 |
Neg logit contributions | |
shoulder | -0.328 |
handc | -0.326 |
Bench | -0.321 |
� | -0.305 |
� | -0.305 |
Token | Joined |
Token ID | 16709 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
Bench | +2.862 |
shoulder | +2.857 |
handc | +2.791 |
� | +2.637 |
� | +2.607 |
Neg logit contributions | |
alks | -3.580 |
alt | -3.423 |
iley | -3.265 |
azeera | -3.190 |
iens | -3.176 |
Token | August |
Token ID | 2932 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
Bench | +3.956 |
shoulder | +3.925 |
handc | +3.691 |
� | +3.541 |
� | +3.478 |
Neg logit contributions | |
alks | -6.218 |
alt | -5.894 |
azeera | -5.850 |
iley | -5.744 |
iens | -5.593 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
shoulder | +0.966 |
Bench | +0.952 |
handc | +0.948 |
� | +0.896 |
� | +0.893 |
Neg logit contributions | |
alks | -1.009 |
alt | -0.953 |
iley | -0.908 |
azeera | -0.887 |
MH | -0.878 |
Token | United |
Token ID | 1578 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
shoulder | +0.812 |
handc | +0.802 |
Bench | +0.799 |
� | +0.750 |
� | +0.749 |
Neg logit contributions | |
alks | -0.918 |
alt | -0.870 |
iley | -0.828 |
azeera | -0.811 |
MH | -0.806 |
Token | States |
Token ID | 1829 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
shoulder | +0.843 |
handc | +0.836 |
Bench | +0.827 |
� | +0.780 |
� | +0.778 |
Neg logit contributions | |
alks | -0.961 |
alt | -0.907 |
iley | -0.869 |
MH | -0.849 |
azeera | -0.843 |
Token | 3 |
Token ID | 513 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
shoulder | +1.697 |
Bench | +1.677 |
handc | +1.668 |
� | +1.558 |
� | +1.555 |
Neg logit contributions | |
alks | -2.105 |
alt | -1.997 |
iley | -1.903 |
azeera | -1.876 |
MH | -1.852 |
Token |
|
Token ID | 198 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
shoulder | +0.213 |
handc | +0.212 |
Bench | +0.209 |
� | +0.198 |
� | +0.197 |
Neg logit contributions | |
alks | -0.218 |
alt | -0.206 |
iley | -0.195 |
MH | -0.191 |
azeera | -0.189 |
Token | heart |
Token ID | 11499 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
alks | +0.168 |
alt | +0.158 |
iley | +0.150 |
MH | +0.147 |
ic | +0.145 |
Neg logit contributions | |
shoulder | -0.175 |
handc | -0.175 |
Bench | -0.172 |
� | -0.163 |
� | -0.163 |
Token | l |
Token ID | 75 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
alks | +1.238 |
alt | +1.179 |
iley | +1.111 |
MH | +1.089 |
ic | +1.075 |
Neg logit contributions | |
shoulder | -1.255 |
handc | -1.254 |
Bench | -1.232 |
� | -1.165 |
� | -1.161 |
Token | xp |
Token ID | 42372 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
shoulder | +0.813 |
handc | +0.811 |
Bench | +0.798 |
� | +0.761 |
� | +0.761 |
Neg logit contributions | |
alks | -0.645 |
alt | -0.606 |
iley | -0.571 |
MH | -0.559 |
ic | -0.550 |
Token | Profile |
Token ID | 13118 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
alks | +0.600 |
alt | +0.565 |
iley | +0.532 |
MH | +0.519 |
ic | +0.512 |
Neg logit contributions | |
shoulder | -0.740 |
handc | -0.738 |
Bench | -0.725 |
� | -0.692 |
� | -0.690 |
Token | Joined |
Token ID | 16709 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
Bench | +3.155 |
shoulder | +3.150 |
handc | +3.090 |
� | +2.924 |
� | +2.902 |
Neg logit contributions | |
alks | -4.014 |
alt | -3.812 |
iley | -3.636 |
azeera | -3.578 |
iens | -3.536 |
Token | September |
Token ID | 2693 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
Bench | +3.543 |
shoulder | +3.531 |
handc | +3.330 |
� | +3.161 |
� | +3.129 |
Neg logit contributions | |
alks | -5.617 |
alt | -5.321 |
azeera | -5.271 |
iley | -5.164 |
iens | -4.995 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
shoulder | +1.014 |
Bench | +0.998 |
handc | +0.997 |
� | +0.939 |
� | +0.938 |
Neg logit contributions | |
alks | -1.048 |
alt | -0.992 |
iley | -0.943 |
azeera | -0.920 |
MH | -0.914 |
Token | United |
Token ID | 1578 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
shoulder | +0.404 |
handc | +0.402 |
Bench | +0.396 |
� | +0.374 |
� | +0.373 |
Neg logit contributions | |
alks | -0.442 |
alt | -0.419 |
iley | -0.398 |
MH | -0.391 |
azeera | -0.386 |
Token | States |
Token ID | 1829 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
shoulder | +0.825 |
handc | +0.815 |
Bench | +0.810 |
� | +0.763 |
� | +0.759 |
Neg logit contributions | |
alks | -0.964 |
alt | -0.907 |
iley | -0.873 |
MH | -0.850 |
azeera | -0.849 |
Token | 10 |
Token ID | 838 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
shoulder | +1.345 |
handc | +1.327 |
Bench | +1.323 |
� | +1.232 |
� | +1.230 |
Neg logit contributions | |
alks | -1.685 |
alt | -1.599 |
iley | -1.524 |
azeera | -1.498 |
MH | -1.487 |
Token | prohibited |
Token ID | 12244 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
alks | +0.380 |
alt | +0.362 |
iley | +0.345 |
MH | +0.341 |
ic | +0.338 |
Neg logit contributions | |
handc | -0.303 |
shoulder | -0.302 |
Bench | -0.296 |
� | -0.281 |
� | -0.279 |
Token | had |
Token ID | 550 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
shoulder | +1.388 |
handc | +1.376 |
Bench | +1.350 |
� | +1.269 |
� | +1.267 |
Neg logit contributions | |
alks | -1.598 |
alt | -1.521 |
iley | -1.439 |
MH | -1.415 |
azeera | -1.402 |
Token | been |
Token ID | 587 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
shoulder | +0.865 |
handc | +0.855 |
Bench | +0.837 |
� | +0.788 |
� | +0.786 |
Neg logit contributions | |
alks | -1.049 |
alt | -0.997 |
iley | -0.946 |
MH | -0.929 |
azeera | -0.926 |
Token | called |
Token ID | 1444 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
shoulder | +1.156 |
handc | +1.146 |
Bench | +1.125 |
� | +1.054 |
� | +1.053 |
Neg logit contributions | |
alks | -1.495 |
alt | -1.421 |
iley | -1.356 |
MH | -1.331 |
azeera | -1.323 |
Token | into |
Token ID | 656 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
shoulder | +2.234 |
handc | +2.200 |
Bench | +2.181 |
� | +2.068 |
� | +2.066 |
Neg logit contributions | |
alks | -1.993 |
alt | -1.861 |
iley | -1.757 |
MH | -1.719 |
azeera | -1.712 |
Token | question |
Token ID | 1808 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
shoulder | +0.523 |
handc | +0.519 |
Bench | +0.502 |
� | +0.452 |
� | +0.450 |
Neg logit contributions | |
alks | -1.421 |
alt | -1.367 |
iley | -1.320 |
MH | -1.302 |
ic | -1.291 |
Token | , |
Token ID | 11 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
shoulder | +4.002 |
handc | +3.897 |
Bench | +3.866 |
� | +3.599 |
� | +3.582 |
Neg logit contributions | |
alks | -4.915 |
alt | -4.644 |
iley | -4.410 |
MH | -4.324 |
iens | -4.307 |
Token | Dr |
Token ID | 1583 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
shoulder | +0.605 |
handc | +0.603 |
Bench | +0.592 |
� | +0.556 |
� | +0.556 |
Neg logit contributions | |
alks | -0.711 |
alt | -0.675 |
iley | -0.643 |
MH | -0.629 |
azeera | -0.624 |
Token | Naz |
Token ID | 12819 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
alks | +0.431 |
alt | +0.392 |
iley | +0.356 |
MH | +0.344 |
ic | +0.336 |
Neg logit contributions | |
shoulder | -1.035 |
handc | -1.034 |
Bench | -1.020 |
� | -0.984 |
� | -0.982 |
Token | ir |
Token ID | 343 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
alks | +1.002 |
alt | +0.935 |
iley | +0.862 |
MH | +0.832 |
ic | +0.819 |
Neg logit contributions | |
handc | -1.804 |
shoulder | -1.791 |
Bench | -1.771 |
� | -1.702 |
� | -1.701 |
Token | - |
Token ID | 12 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
alks | +0.638 |
alt | +0.605 |
iley | +0.567 |
MH | +0.559 |
ic | +0.550 |
Neg logit contributions | |
handc | -0.725 |
shoulder | -0.723 |
Bench | -0.710 |
� | -0.677 |
� | -0.676 |
Token | . |
Token ID | 13 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
alks | +0.920 |
alt | +0.877 |
iley | +0.830 |
MH | +0.812 |
ic | +0.805 |
Neg logit contributions | |
handc | -0.957 |
shoulder | -0.954 |
Bench | -0.939 |
� | -0.899 |
� | -0.893 |
Token |
|
Token ID | 198 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
alks | +0.146 |
alt | +0.138 |
iley | +0.131 |
MH | +0.129 |
ic | +0.127 |
Neg logit contributions | |
shoulder | -0.148 |
handc | -0.147 |
Bench | -0.144 |
� | -0.137 |
� | -0.137 |
Token |
|
Token ID | 198 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
alks | +0.415 |
alt | +0.394 |
iley | +0.371 |
MH | +0.366 |
ic | +0.360 |
Neg logit contributions | |
handc | -0.437 |
shoulder | -0.434 |
Bench | -0.424 |
� | -0.401 |
� | -0.400 |
Token | Continue |
Token ID | 29453 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
alks | +0.407 |
alt | +0.383 |
iley | +0.362 |
MH | +0.355 |
ic | +0.350 |
Neg logit contributions | |
shoulder | -0.451 |
handc | -0.449 |
Bench | -0.442 |
� | -0.421 |
� | -0.420 |
Token | Reading |
Token ID | 11725 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
alks | +0.376 |
alt | +0.357 |
iley | +0.339 |
MH | +0.334 |
ic | +0.329 |
Neg logit contributions | |
shoulder | -0.346 |
handc | -0.345 |
Bench | -0.338 |
� | -0.321 |
� | -0.320 |
Token | Below |
Token ID | 10383 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
shoulder | +1.007 |
Bench | +0.999 |
handc | +0.996 |
� | +0.933 |
� | +0.933 |
Neg logit contributions | |
alks | -0.992 |
alt | -0.935 |
iley | -0.886 |
azeera | -0.870 |
MH | -0.865 |
Token | Advertisement |
Token ID | 39711 |
Feature activation | -8.66e-01 |
Pos logit contributions | |
Bench | +3.404 |
shoulder | +3.272 |
handc | +3.155 |
� | +2.973 |
Waterloo | +2.889 |
Neg logit contributions | |
alks | -5.023 |
alt | -4.686 |
azeera | -4.616 |
iley | -4.452 |
orno | -4.435 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
Bench | +2.053 |
shoulder | +2.035 |
handc | +1.975 |
� | +1.881 |
� | +1.852 |
Neg logit contributions | |
alks | -2.440 |
alt | -2.299 |
azeera | -2.201 |
iley | -2.186 |
ic | -2.121 |
Token | There |
Token ID | 1858 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
alks | +0.135 |
alt | +0.127 |
iley | +0.120 |
MH | +0.118 |
ic | +0.117 |
Neg logit contributions | |
shoulder | -0.142 |
handc | -0.142 |
Bench | -0.139 |
� | -0.133 |
� | -0.133 |
Token | are |
Token ID | 389 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
alks | +0.085 |
alt | +0.081 |
iley | +0.077 |
MH | +0.075 |
ic | +0.074 |
Neg logit contributions | |
shoulder | -0.077 |
handc | -0.077 |
Bench | -0.076 |
� | -0.071 |
� | -0.071 |
Token | many |
Token ID | 867 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
alks | +1.710 |
alt | +1.627 |
iley | +1.546 |
MH | +1.525 |
ic | +1.505 |
Neg logit contributions | |
shoulder | -1.463 |
handc | -1.462 |
Bench | -1.433 |
� | -1.358 |
� | -1.355 |
Token | their |
Token ID | 511 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
shoulder | +0.241 |
handc | +0.241 |
Bench | +0.236 |
� | +0.222 |
� | +0.221 |
Neg logit contributions | |
alks | -0.315 |
alt | -0.300 |
iley | -0.286 |
MH | -0.282 |
ic | -0.279 |
Token | life |
Token ID | 1204 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
shoulder | +0.410 |
handc | +0.409 |
Bench | +0.401 |
� | +0.380 |
� | +0.379 |
Neg logit contributions | |
alks | -0.419 |
alt | -0.396 |
iley | -0.376 |
MH | -0.369 |
ic | -0.364 |
Token | from |
Token ID | 422 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
alks | +1.505 |
alt | +1.440 |
iley | +1.361 |
MH | +1.348 |
ic | +1.340 |
Neg logit contributions | |
handc | -1.631 |
shoulder | -1.628 |
Bench | -1.617 |
� | -1.547 |
� | -1.529 |
Token | the |
Token ID | 262 |
Feature activation | -7.31e-03 |
Pos logit contributions | |
shoulder | +1.188 |
handc | +1.175 |
Bench | +1.154 |
� | +1.081 |
� | +1.081 |
Neg logit contributions | |
alks | -1.399 |
alt | -1.321 |
iley | -1.264 |
azeera | -1.231 |
MH | -1.230 |
Token | polar |
Token ID | 13559 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
shoulder | +0.018 |
handc | +0.018 |
Bench | +0.018 |
� | +0.016 |
� | +0.016 |
Neg logit contributions | |
alks | -0.020 |
alt | -0.019 |
iley | -0.018 |
MH | -0.018 |
azeera | -0.018 |
Token | opposite |
Token ID | 6697 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
shoulder | +0.699 |
handc | +0.694 |
Bench | +0.679 |
� | +0.636 |
� | +0.634 |
Neg logit contributions | |
alks | -0.973 |
alt | -0.927 |
iley | -0.886 |
MH | -0.872 |
azeera | -0.862 |
Token | vantage |
Token ID | 47929 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
shoulder | +4.030 |
handc | +3.862 |
Bench | +3.797 |
stitching | +3.586 |
� | +3.499 |
Neg logit contributions | |
alks | -4.385 |
alt | -4.006 |
iley | -3.882 |
azeera | -3.836 |
MH | -3.757 |
Token | point |
Token ID | 966 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
shoulder | +0.961 |
handc | +0.948 |
Bench | +0.936 |
� | +0.892 |
� | +0.887 |
Neg logit contributions | |
alks | -0.697 |
alt | -0.645 |
iley | -0.611 |
MH | -0.592 |
azeera | -0.590 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
shoulder | +0.886 |
handc | +0.876 |
Bench | +0.861 |
� | +0.813 |
� | +0.812 |
Neg logit contributions | |
alks | -0.925 |
alt | -0.871 |
iley | -0.830 |
MH | -0.811 |
azeera | -0.808 |
Token | For |
Token ID | 1114 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
alks | +0.403 |
alt | +0.380 |
iley | +0.360 |
MH | +0.354 |
ic | +0.349 |
Neg logit contributions | |
handc | -0.430 |
shoulder | -0.430 |
Bench | -0.420 |
� | -0.400 |
� | -0.399 |
Token | Bob |
Token ID | 5811 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
shoulder | +0.634 |
handc | +0.631 |
Bench | +0.621 |
� | +0.583 |
� | +0.581 |
Neg logit contributions | |
alks | -0.753 |
alt | -0.717 |
iley | -0.682 |
MH | -0.669 |
azeera | -0.664 |
Token | many |
Token ID | 867 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
alks | +0.085 |
alt | +0.081 |
iley | +0.077 |
MH | +0.075 |
ic | +0.075 |
Neg logit contributions | |
shoulder | -0.073 |
handc | -0.073 |
Bench | -0.072 |
� | -0.068 |
� | -0.068 |
Token | economic |
Token ID | 3034 |
Feature activation | -7.93e-03 |
Pos logit contributions | |
shoulder | +0.383 |
handc | +0.381 |
Bench | +0.372 |
� | +0.347 |
� | +0.346 |
Neg logit contributions | |
alks | -0.622 |
alt | -0.595 |
iley | -0.570 |
MH | -0.562 |
ic | -0.556 |
Token | indicators |
Token ID | 21337 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
handc | +0.017 |
shoulder | +0.017 |
Bench | +0.016 |
� | +0.015 |
� | +0.015 |
Neg logit contributions | |
alks | -0.024 |
alt | -0.023 |
iley | -0.022 |
MH | -0.022 |
ic | -0.022 |
Token | that |
Token ID | 326 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
shoulder | +2.130 |
handc | +2.100 |
Bench | +2.087 |
� | +1.973 |
� | +1.967 |
Neg logit contributions | |
alks | -2.195 |
alt | -2.070 |
iley | -1.953 |
azeera | -1.930 |
MH | -1.914 |
Token | inter |
Token ID | 987 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
shoulder | +0.312 |
handc | +0.311 |
Bench | +0.305 |
� | +0.288 |
� | +0.287 |
Neg logit contributions | |
alks | -0.353 |
alt | -0.335 |
iley | -0.319 |
MH | -0.313 |
ic | -0.309 |
Token | play |
Token ID | 1759 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
shoulder | +1.533 |
handc | +1.516 |
Bench | +1.494 |
� | +1.420 |
� | +1.413 |
Neg logit contributions | |
alks | -1.496 |
alt | -1.410 |
iley | -1.344 |
MH | -1.329 |
azeera | -1.315 |
Token | to |
Token ID | 284 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
shoulder | +3.157 |
handc | +3.062 |
Bench | +2.995 |
� | +2.831 |
� | +2.817 |
Neg logit contributions | |
alks | -5.007 |
alt | -4.746 |
iley | -4.498 |
MH | -4.482 |
iens | -4.466 |
Token | create |
Token ID | 2251 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
handc | +0.246 |
shoulder | +0.246 |
Bench | +0.242 |
� | +0.227 |
� | +0.227 |
Neg logit contributions | |
alks | -0.322 |
alt | -0.308 |
iley | -0.294 |
MH | -0.289 |
ic | -0.287 |
Token | a |
Token ID | 257 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
alks | +0.565 |
alt | +0.538 |
iley | +0.517 |
MH | +0.510 |
ic | +0.505 |
Neg logit contributions | |
handc | -0.443 |
shoulder | -0.439 |
Bench | -0.432 |
� | -0.409 |
� | -0.406 |
Token | judgement |
Token ID | 23071 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
alks | +0.757 |
alt | +0.724 |
iley | +0.697 |
MH | +0.686 |
ic | +0.679 |
Neg logit contributions | |
handc | -0.587 |
shoulder | -0.582 |
Bench | -0.570 |
� | -0.542 |
� | -0.540 |
Token | on |
Token ID | 319 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
alks | +0.743 |
alt | +0.704 |
iley | +0.666 |
MH | +0.651 |
ic | +0.645 |
Neg logit contributions | |
handc | -0.846 |
shoulder | -0.844 |
Bench | -0.828 |
� | -0.790 |
� | -0.790 |
Token |
|
Token ID | 198 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
shoulder | +0.012 |
handc | +0.012 |
Bench | +0.011 |
� | +0.011 |
� | +0.011 |
Neg logit contributions | |
alks | -0.012 |
alt | -0.011 |
iley | -0.011 |
MH | -0.010 |
ic | -0.010 |
Token | re |
Token ID | 260 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
alks | +0.453 |
alt | +0.428 |
iley | +0.405 |
MH | +0.397 |
ic | +0.392 |
Neg logit contributions | |
shoulder | -0.473 |
handc | -0.472 |
Bench | -0.464 |
� | -0.441 |
� | -0.440 |
Token | i |
Token ID | 72 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
shoulder | +0.462 |
handc | +0.460 |
Bench | +0.453 |
� | +0.433 |
� | +0.432 |
Neg logit contributions | |
alks | -0.386 |
alt | -0.358 |
iley | -0.340 |
MH | -0.336 |
azeera | -0.330 |
Token | Profile |
Token ID | 13118 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
alks | +1.130 |
alt | +1.075 |
iley | +1.013 |
MH | +0.986 |
ic | +0.975 |
Neg logit contributions | |
shoulder | -1.243 |
handc | -1.241 |
Bench | -1.219 |
� | -1.151 |
� | -1.150 |
Token | Blog |
Token ID | 14001 |
Feature activation | -8.47e-01 |
Pos logit contributions | |
shoulder | +2.826 |
Bench | +2.822 |
handc | +2.778 |
� | +2.634 |
� | +2.609 |
Neg logit contributions | |
alks | -3.611 |
alt | -3.435 |
iley | -3.288 |
azeera | -3.221 |
iens | -3.200 |
Token | Joined |
Token ID | 16709 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
shoulder | +2.097 |
handc | +2.071 |
Bench | +2.067 |
� | +1.950 |
� | +1.937 |
Neg logit contributions | |
alks | -2.301 |
alt | -2.170 |
iley | -2.076 |
MH | -2.028 |
azeera | -2.021 |
Token | October |
Token ID | 3267 |
Feature activation | +7.98e-02 |
Pos logit contributions | |
Bench | +3.215 |
shoulder | +3.213 |
handc | +3.083 |
� | +2.913 |
� | +2.892 |
Neg logit contributions | |
alks | -4.820 |
alt | -4.580 |
azeera | -4.470 |
iley | -4.436 |
iens | -4.305 |
Token | 2002 |
Token ID | 6244 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
alks | +0.212 |
alt | +0.201 |
iley | +0.191 |
MH | +0.189 |
ic | +0.186 |
Neg logit contributions | |
handc | -0.205 |
shoulder | -0.203 |
Bench | -0.198 |
� | -0.189 |
� | -0.188 |
Token | United |
Token ID | 1578 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
shoulder | +0.403 |
handc | +0.401 |
Bench | +0.395 |
� | +0.373 |
� | +0.373 |
Neg logit contributions | |
alks | -0.440 |
alt | -0.417 |
iley | -0.396 |
MH | -0.388 |
azeera | -0.384 |
Token | States |
Token ID | 1829 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
shoulder | +0.871 |
handc | +0.862 |
Bench | +0.856 |
� | +0.806 |
� | +0.803 |
Neg logit contributions | |
alks | -1.011 |
alt | -0.952 |
iley | -0.915 |
MH | -0.891 |
azeera | -0.890 |
Token | 3 |
Token ID | 513 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
shoulder | +1.343 |
handc | +1.328 |
Bench | +1.320 |
� | +1.233 |
� | +1.230 |
Neg logit contributions | |
alks | -1.668 |
alt | -1.584 |
iley | -1.510 |
azeera | -1.480 |
MH | -1.475 |
Token | of |
Token ID | 286 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
shoulder | +0.079 |
handc | +0.079 |
Bench | +0.078 |
� | +0.074 |
� | +0.074 |
Neg logit contributions | |
alks | -0.072 |
alt | -0.069 |
iley | -0.065 |
MH | -0.064 |
ic | -0.063 |
Token | Dal |
Token ID | 12348 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
shoulder | +0.459 |
handc | +0.456 |
Bench | +0.452 |
� | +0.429 |
� | +0.428 |
Neg logit contributions | |
alks | -0.400 |
alt | -0.377 |
iley | -0.357 |
MH | -0.347 |
azeera | -0.343 |
Token | K |
Token ID | 42 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
shoulder | +1.559 |
Bench | +1.548 |
handc | +1.547 |
� | +1.476 |
� | +1.475 |
Neg logit contributions | |
alks | -0.864 |
alt | -0.788 |
iley | -0.735 |
MH | -0.729 |
azeera | -0.713 |
Token | Profile |
Token ID | 13118 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
shoulder | +1.287 |
handc | +1.280 |
Bench | +1.269 |
� | +1.205 |
� | +1.202 |
Neg logit contributions | |
alks | -1.032 |
alt | -0.965 |
iley | -0.911 |
MH | -0.895 |
ic | -0.884 |
Token | Blog |
Token ID | 14001 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
Bench | +2.564 |
shoulder | +2.563 |
handc | +2.522 |
� | +2.382 |
� | +2.361 |
Neg logit contributions | |
alks | -3.386 |
alt | -3.219 |
iley | -3.075 |
azeera | -3.017 |
MH | -2.999 |
Token | Joined |
Token ID | 16709 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
shoulder | +1.717 |
Bench | +1.699 |
handc | +1.691 |
� | +1.596 |
� | +1.586 |
Neg logit contributions | |
alks | -1.899 |
alt | -1.791 |
iley | -1.709 |
azeera | -1.675 |
MH | -1.668 |
Token | June |
Token ID | 2795 |
Feature activation | -2.75e-03 |
Pos logit contributions | |
Bench | +3.153 |
shoulder | +3.142 |
handc | +2.994 |
� | +2.810 |
� | +2.798 |
Neg logit contributions | |
alks | -4.867 |
alt | -4.622 |
azeera | -4.538 |
iley | -4.469 |
iens | -4.326 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
shoulder | +0.007 |
handc | +0.007 |
Bench | +0.007 |
� | +0.007 |
� | +0.007 |
Neg logit contributions | |
alks | -0.007 |
alt | -0.007 |
iley | -0.006 |
MH | -0.006 |
azeera | -0.006 |
Token | Canada |
Token ID | 3340 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
shoulder | +0.558 |
handc | +0.553 |
Bench | +0.549 |
� | +0.516 |
� | +0.515 |
Neg logit contributions | |
alks | -0.619 |
alt | -0.586 |
iley | -0.558 |
azeera | -0.545 |
MH | -0.544 |
Token | 16 |
Token ID | 1467 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
shoulder | +0.325 |
handc | +0.324 |
Bench | +0.318 |
� | +0.299 |
� | +0.298 |
Neg logit contributions | |
alks | -0.400 |
alt | -0.380 |
iley | -0.363 |
MH | -0.356 |
ic | -0.352 |
Token | 10 |
Token ID | 940 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
shoulder | +1.379 |
handc | +1.369 |
Bench | +1.356 |
� | +1.282 |
� | +1.278 |
Neg logit contributions | |
alks | -1.553 |
alt | -1.476 |
iley | -1.404 |
MH | -1.372 |
azeera | -1.368 |
Token | 2002 |
Token ID | 6244 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
alks | +0.212 |
alt | +0.201 |
iley | +0.191 |
MH | +0.189 |
ic | +0.186 |
Neg logit contributions | |
handc | -0.205 |
shoulder | -0.203 |
Bench | -0.198 |
� | -0.189 |
� | -0.188 |
Token | United |
Token ID | 1578 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
shoulder | +0.403 |
handc | +0.401 |
Bench | +0.395 |
� | +0.373 |
� | +0.373 |
Neg logit contributions | |
alks | -0.440 |
alt | -0.417 |
iley | -0.396 |
MH | -0.388 |
azeera | -0.384 |
Token | States |
Token ID | 1829 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
shoulder | +0.871 |
handc | +0.862 |
Bench | +0.856 |
� | +0.806 |
� | +0.803 |
Neg logit contributions | |
alks | -1.011 |
alt | -0.952 |
iley | -0.915 |
MH | -0.891 |
azeera | -0.890 |
Token | 3 |
Token ID | 513 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
shoulder | +1.343 |
handc | +1.328 |
Bench | +1.320 |
� | +1.233 |
� | +1.230 |
Neg logit contributions | |
alks | -1.668 |
alt | -1.584 |
iley | -1.510 |
azeera | -1.480 |
MH | -1.475 |
Token | 472 |
Token ID | 37856 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
shoulder | +0.998 |
handc | +0.988 |
Bench | +0.981 |
� | +0.928 |
� | +0.923 |
Neg logit contributions | |
alks | -1.171 |
alt | -1.113 |
iley | -1.060 |
MH | -1.039 |
azeera | -1.037 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
handc | +0.573 |
shoulder | +0.572 |
Bench | +0.559 |
� | +0.535 |
� | +0.534 |
Neg logit contributions | |
alks | -0.431 |
alt | -0.404 |
iley | -0.379 |
MH | -0.372 |
ic | -0.366 |
Token | # |
Token ID | 1303 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
shoulder | +3.288 |
Bench | +3.252 |
handc | +3.173 |
� | +2.921 |
stitching | +2.921 |
Neg logit contributions | |
alks | -4.597 |
alt | -4.349 |
iley | -4.203 |
azeera | -4.149 |
iens | -4.086 |
Token | 11 |
Token ID | 1157 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
shoulder | +1.835 |
handc | +1.825 |
Bench | +1.807 |
� | +1.732 |
� | +1.724 |
Neg logit contributions | |
alks | -1.191 |
alt | -1.104 |
iley | -1.038 |
MH | -1.009 |
azeera | -1.003 |
Token | On |
Token ID | 1550 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
shoulder | +0.341 |
handc | +0.338 |
Bench | +0.336 |
� | +0.316 |
� | +0.316 |
Neg logit contributions | |
alks | -0.289 |
alt | -0.271 |
iley | -0.256 |
azeera | -0.249 |
MH | -0.247 |
Token | September |
Token ID | 2693 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
shoulder | +1.089 |
handc | +1.080 |
Bench | +1.065 |
� | +0.991 |
� | +0.989 |
Neg logit contributions | |
alks | -1.473 |
alt | -1.399 |
iley | -1.336 |
MH | -1.311 |
azeera | -1.307 |
Token | 20 |
Token ID | 1160 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
alks | +0.826 |
alt | +0.783 |
iley | +0.742 |
MH | +0.732 |
ic | +0.721 |
Neg logit contributions | |
handc | -0.811 |
shoulder | -0.810 |
Bench | -0.791 |
� | -0.753 |
� | -0.751 |
Token | . |
Token ID | 13 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
alks | +0.884 |
alt | +0.833 |
iley | +0.786 |
MH | +0.769 |
ic | +0.759 |
Neg logit contributions | |
shoulder | -1.019 |
handc | -1.016 |
Bench | -1.001 |
� | -0.951 |
� | -0.949 |
Token | ( |
Token ID | 357 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
alks | +0.832 |
alt | +0.787 |
iley | +0.745 |
MH | +0.730 |
ic | +0.721 |
Neg logit contributions | |
shoulder | -0.864 |
handc | -0.861 |
Bench | -0.848 |
� | -0.804 |
� | -0.802 |
Token | See |
Token ID | 6214 |
Feature activation | +7.30e-01 |
Pos logit contributions | |
alks | +0.603 |
alt | +0.568 |
iley | +0.538 |
MH | +0.525 |
azeera | +0.520 |
Neg logit contributions | |
shoulder | -0.684 |
handc | -0.681 |
Bench | -0.674 |
� | -0.641 |
� | -0.640 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
alks | +2.003 |
alt | +1.901 |
iley | +1.808 |
MH | +1.786 |
ic | +1.761 |
Neg logit contributions | |
handc | -1.792 |
shoulder | -1.788 |
Bench | -1.742 |
� | -1.667 |
� | -1.663 |
Token | slideshow |
Token ID | 48828 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
shoulder | +0.319 |
handc | +0.316 |
Bench | +0.314 |
� | +0.291 |
� | +0.291 |
Neg logit contributions | |
alks | -0.368 |
alt | -0.350 |
iley | -0.333 |
MH | -0.325 |
azeera | -0.324 |
Token | below |
Token ID | 2174 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
shoulder | +1.130 |
handc | +1.114 |
Bench | +1.107 |
stitching | +1.028 |
� | +1.021 |
Neg logit contributions | |
alks | -1.362 |
alt | -1.282 |
iley | -1.225 |
azeera | -1.213 |
MH | -1.195 |
Token | for |
Token ID | 329 |
Feature activation | +2.97e-03 |
Pos logit contributions | |
shoulder | +3.461 |
Bench | +3.391 |
handc | +3.343 |
stitching | +3.146 |
� | +3.055 |
Neg logit contributions | |
alks | -4.400 |
alt | -4.076 |
azeera | -3.984 |
iley | -3.900 |
iens | -3.859 |
Token | some |
Token ID | 617 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
alks | +0.008 |
alt | +0.008 |
iley | +0.008 |
MH | +0.007 |
azeera | +0.007 |
Neg logit contributions | |
shoulder | -0.007 |
handc | -0.007 |
Bench | -0.007 |
� | -0.006 |
� | -0.006 |
Token | great |
Token ID | 1049 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
shoulder | +0.368 |
handc | +0.365 |
Bench | +0.360 |
� | +0.334 |
stitching | +0.333 |
Neg logit contributions | |
alks | -0.449 |
alt | -0.428 |
iley | -0.408 |
MH | -0.397 |
azeera | -0.396 |
Token | Honey |
Token ID | 21788 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
shoulder | +0.933 |
handc | +0.921 |
Bench | +0.918 |
stitching | +0.852 |
� | +0.849 |
Neg logit contributions | |
alks | -0.974 |
alt | -0.927 |
iley | -0.884 |
azeera | -0.854 |
MH | -0.850 |
Token | comb |
Token ID | 24011 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
shoulder | +0.962 |
handc | +0.960 |
Bench | +0.946 |
� | +0.903 |
� | +0.902 |
Neg logit contributions | |
alks | -0.697 |
alt | -0.652 |
iley | -0.612 |
MH | -0.598 |
ic | -0.588 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
shoulder | +0.966 |
Bench | +0.952 |
handc | +0.948 |
� | +0.896 |
� | +0.893 |
Neg logit contributions | |
alks | -1.009 |
alt | -0.953 |
iley | -0.908 |
azeera | -0.887 |
MH | -0.878 |
Token | United |
Token ID | 1578 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
shoulder | +0.812 |
handc | +0.802 |
Bench | +0.799 |
� | +0.750 |
� | +0.749 |
Neg logit contributions | |
alks | -0.918 |
alt | -0.870 |
iley | -0.828 |
azeera | -0.811 |
MH | -0.806 |
Token | States |
Token ID | 1829 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
shoulder | +0.843 |
handc | +0.836 |
Bench | +0.827 |
� | +0.780 |
� | +0.778 |
Neg logit contributions | |
alks | -0.961 |
alt | -0.907 |
iley | -0.869 |
MH | -0.849 |
azeera | -0.843 |
Token | 3 |
Token ID | 513 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
shoulder | +1.697 |
Bench | +1.677 |
handc | +1.668 |
� | +1.558 |
� | +1.555 |
Neg logit contributions | |
alks | -2.105 |
alt | -1.997 |
iley | -1.903 |
azeera | -1.876 |
MH | -1.852 |
Token | 240 |
Token ID | 16102 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
shoulder | +1.005 |
handc | +0.996 |
Bench | +0.990 |
� | +0.939 |
� | +0.936 |
Neg logit contributions | |
alks | -0.956 |
alt | -0.904 |
iley | -0.857 |
MH | -0.838 |
azeera | -0.837 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
handc | +0.119 |
shoulder | +0.119 |
Bench | +0.116 |
� | +0.111 |
� | +0.111 |
Neg logit contributions | |
alks | -0.089 |
alt | -0.083 |
iley | -0.078 |
MH | -0.077 |
ic | -0.076 |
Token | # |
Token ID | 1303 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
shoulder | +3.430 |
Bench | +3.393 |
handc | +3.301 |
� | +3.052 |
� | +3.048 |
Neg logit contributions | |
alks | -4.383 |
alt | -4.160 |
iley | -4.003 |
azeera | -3.981 |
iens | -3.895 |
Token | 10 |
Token ID | 940 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
shoulder | +1.430 |
handc | +1.418 |
Bench | +1.410 |
� | +1.346 |
� | +1.339 |
Neg logit contributions | |
alks | -1.411 |
alt | -1.327 |
iley | -1.260 |
azeera | -1.236 |
MH | -1.235 |
Token | Happy |
Token ID | 14628 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
alks | +0.137 |
alt | +0.129 |
iley | +0.122 |
MH | +0.118 |
azeera | +0.118 |
Neg logit contributions | |
shoulder | -0.168 |
handc | -0.167 |
Bench | -0.165 |
� | -0.157 |
� | -0.157 |
Token | it |
Token ID | 340 |
Feature activation | +9.72e-03 |
Pos logit contributions | |
shoulder | +0.494 |
handc | +0.491 |
Bench | +0.483 |
� | +0.454 |
� | +0.454 |
Neg logit contributions | |
alks | -0.562 |
alt | -0.534 |
iley | -0.508 |
MH | -0.497 |
azeera | -0.493 |
Token | worked |
Token ID | 3111 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
alks | +0.027 |
alt | +0.026 |
iley | +0.025 |
MH | +0.024 |
ic | +0.024 |
Neg logit contributions | |
shoulder | -0.023 |
handc | -0.023 |
Bench | -0.023 |
� | -0.021 |
� | -0.021 |
Token | lights |
Token ID | 7588 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
shoulder | +0.481 |
handc | +0.475 |
Bench | +0.466 |
� | +0.439 |
stitching | +0.436 |
Neg logit contributions | |
alks | -0.506 |
alt | -0.478 |
iley | -0.455 |
MH | -0.444 |
azeera | -0.443 |
Token | were |
Token ID | 547 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
shoulder | +0.397 |
handc | +0.396 |
Bench | +0.387 |
� | +0.365 |
� | +0.364 |
Neg logit contributions | |
alks | -0.405 |
alt | -0.381 |
iley | -0.362 |
MH | -0.355 |
azeera | -0.353 |
Token | out |
Token ID | 503 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
alks | +0.545 |
alt | +0.516 |
iley | +0.491 |
MH | +0.481 |
azeera | +0.477 |
Neg logit contributions | |
shoulder | -0.494 |
handc | -0.492 |
Bench | -0.481 |
� | -0.454 |
� | -0.453 |
Token | !" |
Token ID | 2474 |
Feature activation | -8.17e-03 |
Pos logit contributions | |
alks | +0.950 |
alt | +0.904 |
iley | +0.854 |
MH | +0.842 |
ic | +0.834 |
Neg logit contributions | |
handc | -0.919 |
shoulder | -0.919 |
Bench | -0.904 |
� | -0.859 |
� | -0.859 |
Token | By |
Token ID | 2750 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
shoulder | +0.021 |
handc | +0.021 |
Bench | +0.021 |
� | +0.020 |
� | +0.020 |
Neg logit contributions | |
alks | -0.021 |
alt | -0.020 |
iley | -0.019 |
MH | -0.019 |
azeera | -0.019 |
Token | now |
Token ID | 783 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
shoulder | +0.836 |
handc | +0.830 |
Bench | +0.818 |
stitching | +0.760 |
� | +0.759 |
Neg logit contributions | |
alks | -0.961 |
alt | -0.903 |
iley | -0.864 |
azeera | -0.847 |
MH | -0.839 |
Token | I |
Token ID | 314 |
Feature activation | -1.80e-04 |
Pos logit contributions | |
shoulder | +3.596 |
handc | +3.503 |
Bench | +3.441 |
� | +3.225 |
Thornton | +3.205 |
Neg logit contributions | |
alks | -4.306 |
alt | -4.031 |
azeera | -3.844 |
iley | -3.817 |
iens | -3.650 |
Token | was |
Token ID | 373 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
shoulder | +0.000 |
handc | +0.000 |
Bench | +0.000 |
� | +0.000 |
� | +0.000 |
Neg logit contributions | |
alks | -0.001 |
alt | -0.001 |
iley | -0.000 |
MH | -0.000 |
azeera | -0.000 |
Token | doing |
Token ID | 1804 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
shoulder | +0.083 |
handc | +0.083 |
Bench | +0.081 |
� | +0.076 |
� | +0.076 |
Neg logit contributions | |
alks | -0.100 |
alt | -0.095 |
iley | -0.090 |
MH | -0.089 |
azeera | -0.088 |
Token | 180 |
Token ID | 11546 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
shoulder | +1.647 |
handc | +1.625 |
Bench | +1.592 |
� | +1.493 |
� | +1.477 |
Neg logit contributions | |
alks | -2.004 |
alt | -1.890 |
iley | -1.812 |
MH | -1.765 |
azeera | -1.765 |
Token | knots |
Token ID | 33620 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
alks | +0.482 |
alt | +0.459 |
iley | +0.436 |
MH | +0.429 |
ic | +0.423 |
Neg logit contributions | |
handc | -0.446 |
shoulder | -0.445 |
Bench | -0.437 |
� | -0.414 |
� | -0.414 |
Token | June |
Token ID | 2795 |
Feature activation | -2.75e-03 |
Pos logit contributions | |
Bench | +3.153 |
shoulder | +3.142 |
handc | +2.994 |
� | +2.810 |
� | +2.798 |
Neg logit contributions | |
alks | -4.867 |
alt | -4.622 |
azeera | -4.538 |
iley | -4.469 |
iens | -4.326 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
shoulder | +0.007 |
handc | +0.007 |
Bench | +0.007 |
� | +0.007 |
� | +0.007 |
Neg logit contributions | |
alks | -0.007 |
alt | -0.007 |
iley | -0.006 |
MH | -0.006 |
azeera | -0.006 |
Token | Canada |
Token ID | 3340 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
shoulder | +0.558 |
handc | +0.553 |
Bench | +0.549 |
� | +0.516 |
� | +0.515 |
Neg logit contributions | |
alks | -0.619 |
alt | -0.586 |
iley | -0.558 |
azeera | -0.545 |
MH | -0.544 |
Token | 16 |
Token ID | 1467 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
shoulder | +0.325 |
handc | +0.324 |
Bench | +0.318 |
� | +0.299 |
� | +0.298 |
Neg logit contributions | |
alks | -0.400 |
alt | -0.380 |
iley | -0.363 |
MH | -0.356 |
ic | -0.352 |
Token | 10 |
Token ID | 940 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
shoulder | +1.379 |
handc | +1.369 |
Bench | +1.356 |
� | +1.282 |
� | +1.278 |
Neg logit contributions | |
alks | -1.553 |
alt | -1.476 |
iley | -1.404 |
MH | -1.372 |
azeera | -1.368 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
shoulder | +1.437 |
handc | +1.433 |
Bench | +1.413 |
� | +1.347 |
� | +1.345 |
Neg logit contributions | |
alks | -1.122 |
alt | -1.054 |
iley | -0.991 |
MH | -0.968 |
ic | -0.954 |
Token | # |
Token ID | 1303 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
shoulder | +3.261 |
Bench | +3.235 |
handc | +3.160 |
� | +2.913 |
� | +2.911 |
Neg logit contributions | |
alks | -4.539 |
alt | -4.278 |
iley | -4.135 |
azeera | -4.085 |
iens | -4.039 |
Token | 14 |
Token ID | 1415 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
shoulder | +1.441 |
handc | +1.437 |
Bench | +1.412 |
� | +1.341 |
� | +1.338 |
Neg logit contributions | |
alks | -1.510 |
alt | -1.416 |
iley | -1.352 |
MH | -1.328 |
azeera | -1.316 |
Token | On |
Token ID | 1550 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
shoulder | +0.183 |
handc | +0.182 |
Bench | +0.180 |
� | +0.171 |
� | +0.170 |
Neg logit contributions | |
alks | -0.135 |
alt | -0.126 |
iley | -0.118 |
azeera | -0.114 |
MH | -0.114 |
Token | September |
Token ID | 2693 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
shoulder | +0.920 |
handc | +0.915 |
Bench | +0.899 |
� | +0.838 |
� | +0.834 |
Neg logit contributions | |
alks | -1.287 |
alt | -1.226 |
iley | -1.169 |
MH | -1.149 |
azeera | -1.141 |
Token | 20 |
Token ID | 1160 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
alks | +0.679 |
alt | +0.651 |
iley | +0.625 |
MH | +0.616 |
ic | +0.610 |
Neg logit contributions | |
shoulder | -0.362 |
handc | -0.361 |
Bench | -0.352 |
� | -0.325 |
� | -0.324 |
Token |
|
Token ID | 198 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
shoulder | +1.338 |
handc | +1.331 |
Bench | +1.329 |
� | +1.220 |
stitching | +1.211 |
Neg logit contributions | |
alks | -1.363 |
alt | -1.292 |
iley | -1.240 |
MH | -1.203 |
azeera | -1.202 |
Token |
|
Token ID | 198 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
alks | +0.253 |
alt | +0.241 |
iley | +0.227 |
MH | +0.223 |
ic | +0.220 |
Neg logit contributions | |
handc | -0.265 |
shoulder | -0.265 |
Bench | -0.259 |
� | -0.246 |
� | -0.246 |
Token | I |
Token ID | 40 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
alks | +0.497 |
alt | +0.474 |
iley | +0.448 |
MH | +0.437 |
ic | +0.432 |
Neg logit contributions | |
shoulder | -0.481 |
handc | -0.480 |
Bench | -0.469 |
� | -0.443 |
� | -0.441 |
Token | don |
Token ID | 836 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
shoulder | +0.917 |
handc | +0.915 |
Bench | +0.899 |
� | +0.849 |
� | +0.847 |
Neg logit contributions | |
alks | -0.949 |
alt | -0.897 |
iley | -0.852 |
MH | -0.838 |
azeera | -0.827 |
Token | 't |
Token ID | 470 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
shoulder | +0.403 |
handc | +0.402 |
Bench | +0.396 |
� | +0.378 |
� | +0.377 |
Neg logit contributions | |
alks | -0.296 |
alt | -0.277 |
iley | -0.260 |
MH | -0.254 |
ic | -0.250 |
Token | expect |
Token ID | 1607 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
shoulder | +1.128 |
handc | +1.126 |
Bench | +1.099 |
� | +1.029 |
� | +1.025 |
Neg logit contributions | |
alks | -1.157 |
alt | -1.093 |
iley | -1.037 |
MH | -1.019 |
azeera | -1.015 |
Token | Manning |
Token ID | 15281 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
shoulder | +3.671 |
handc | +3.652 |
Bench | +3.602 |
Thornton | +3.291 |
stitching | +3.280 |
Neg logit contributions | |
alks | -4.174 |
alt | -3.958 |
azeera | -3.812 |
iley | -3.781 |
MH | -3.702 |
Token | to |
Token ID | 284 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
shoulder | +0.468 |
handc | +0.467 |
Bench | +0.458 |
� | +0.434 |
� | +0.433 |
Neg logit contributions | |
alks | -0.541 |
alt | -0.515 |
iley | -0.490 |
MH | -0.481 |
ic | -0.476 |
Token | be |
Token ID | 307 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
shoulder | +0.507 |
handc | +0.506 |
Bench | +0.496 |
� | +0.469 |
� | +0.467 |
Neg logit contributions | |
alks | -0.564 |
alt | -0.535 |
iley | -0.508 |
MH | -0.499 |
ic | -0.493 |
Token | awful |
Token ID | 12659 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
shoulder | +0.780 |
handc | +0.778 |
Bench | +0.762 |
� | +0.711 |
� | +0.709 |
Neg logit contributions | |
alks | -0.942 |
alt | -0.893 |
iley | -0.851 |
MH | -0.836 |
azeera | -0.833 |
Token | this |
Token ID | 428 |
Feature activation | -8.35e-02 |
Pos logit contributions | |
shoulder | +0.115 |
handc | +0.115 |
Bench | +0.113 |
� | +0.108 |
� | +0.108 |
Neg logit contributions | |
alks | -0.129 |
alt | -0.123 |
iley | -0.116 |
MH | -0.115 |
ic | -0.114 |
Token | signatures |
Token ID | 17239 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
shoulder | +0.298 |
handc | +0.297 |
Bench | +0.291 |
� | +0.275 |
� | +0.274 |
Neg logit contributions | |
alks | -0.324 |
alt | -0.307 |
iley | -0.293 |
MH | -0.286 |
ic | -0.284 |
Token | will |
Token ID | 481 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
shoulder | +0.715 |
handc | +0.710 |
Bench | +0.696 |
� | +0.653 |
� | +0.651 |
Neg logit contributions | |
alks | -0.835 |
alt | -0.789 |
iley | -0.753 |
azeera | -0.737 |
MH | -0.736 |
Token | be |
Token ID | 307 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
alks | +0.492 |
alt | +0.467 |
iley | +0.442 |
MH | +0.434 |
ic | +0.429 |
Neg logit contributions | |
shoulder | -0.488 |
handc | -0.487 |
Bench | -0.479 |
� | -0.456 |
� | -0.454 |
Token | called |
Token ID | 1444 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
alks | +0.716 |
alt | +0.680 |
iley | +0.644 |
MH | +0.635 |
ic | +0.627 |
Neg logit contributions | |
handc | -0.682 |
shoulder | -0.681 |
Bench | -0.669 |
� | -0.636 |
� | -0.634 |
Token | into |
Token ID | 656 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
shoulder | +0.822 |
handc | +0.815 |
Bench | +0.802 |
� | +0.756 |
� | +0.754 |
Neg logit contributions | |
alks | -0.845 |
alt | -0.796 |
iley | -0.758 |
MH | -0.736 |
azeera | -0.734 |
Token | question |
Token ID | 1808 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
shoulder | +0.565 |
handc | +0.562 |
Bench | +0.550 |
� | +0.510 |
� | +0.508 |
Neg logit contributions | |
alks | -0.927 |
alt | -0.887 |
iley | -0.850 |
MH | -0.835 |
azeera | -0.829 |
Token | , |
Token ID | 11 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
shoulder | +3.555 |
handc | +3.513 |
Bench | +3.443 |
stitching | +3.204 |
� | +3.186 |
Neg logit contributions | |
alks | -4.284 |
alt | -4.027 |
iley | -3.831 |
azeera | -3.754 |
MH | -3.706 |
Token | and |
Token ID | 290 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
shoulder | +0.451 |
handc | +0.450 |
Bench | +0.441 |
� | +0.415 |
� | +0.414 |
Neg logit contributions | |
alks | -0.539 |
alt | -0.512 |
iley | -0.488 |
MH | -0.479 |
azeera | -0.474 |
Token | rejected |
Token ID | 8606 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
shoulder | +0.817 |
handc | +0.814 |
Bench | +0.798 |
� | +0.750 |
� | +0.748 |
Neg logit contributions | |
alks | -0.934 |
alt | -0.889 |
iley | -0.846 |
MH | -0.826 |
azeera | -0.823 |
Token | . |
Token ID | 13 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
alks | +0.568 |
alt | +0.539 |
iley | +0.511 |
MH | +0.504 |
ic | +0.497 |
Neg logit contributions | |
shoulder | -0.544 |
handc | -0.544 |
Bench | -0.534 |
� | -0.508 |
� | -0.506 |
Token |
|
Token ID | 198 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
alks | +1.287 |
alt | +1.219 |
MH | +1.146 |
iley | +1.146 |
ic | +1.120 |
Neg logit contributions | |
handc | -1.348 |
shoulder | -1.336 |
Bench | -1.292 |
� | -1.238 |
� | -1.231 |
Token | uk |
Token ID | 2724 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
shoulder | +0.513 |
handc | +0.512 |
Bench | +0.505 |
� | +0.485 |
� | +0.484 |
Neg logit contributions | |
alks | -0.297 |
alt | -0.276 |
iley | -0.256 |
MH | -0.249 |
ic | -0.244 |
Token | / |
Token ID | 14 |
Feature activation | -6.14e-03 |
Pos logit contributions | |
alks | +0.426 |
alt | +0.396 |
iley | +0.366 |
MH | +0.356 |
ic | +0.348 |
Neg logit contributions | |
handc | -0.766 |
shoulder | -0.764 |
Bench | -0.749 |
� | -0.720 |
� | -0.718 |
Token | in |
Token ID | 259 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
shoulder | +0.016 |
handc | +0.016 |
Bench | +0.016 |
� | +0.015 |
� | +0.015 |
Neg logit contributions | |
alks | -0.016 |
alt | -0.015 |
iley | -0.014 |
MH | -0.014 |
ic | -0.014 |
Token | coming |
Token ID | 4976 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
shoulder | +0.864 |
handc | +0.860 |
Bench | +0.846 |
� | +0.803 |
� | +0.802 |
Neg logit contributions | |
alks | -0.850 |
alt | -0.804 |
iley | -0.763 |
MH | -0.748 |
ic | -0.738 |
Token | / |
Token ID | 14 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
shoulder | +1.859 |
handc | +1.840 |
Bench | +1.823 |
� | +1.753 |
� | +1.747 |
Neg logit contributions | |
alks | -1.196 |
alt | -1.107 |
iley | -1.046 |
MH | -1.024 |
ic | -1.006 |
Token | article |
Token ID | 20205 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
alks | +0.060 |
alt | +0.057 |
iley | +0.054 |
MH | +0.053 |
ic | +0.052 |
Neg logit contributions | |
handc | -0.057 |
shoulder | -0.057 |
Bench | -0.056 |
� | -0.053 |
� | -0.053 |
Token | 35 |
Token ID | 2327 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
shoulder | +2.760 |
handc | +2.658 |
Bench | +2.631 |
� | +2.502 |
� | +2.478 |
Neg logit contributions | |
alks | -4.978 |
alt | -4.747 |
azeera | -4.522 |
MH | -4.482 |
ic | -4.448 |
Token | 79 |
Token ID | 3720 |
Feature activation | +9.02e-02 |
Pos logit contributions | |
shoulder | +0.196 |
handc | +0.196 |
Bench | +0.190 |
� | +0.178 |
� | +0.177 |
Neg logit contributions | |
alks | -0.287 |
alt | -0.273 |
iley | -0.261 |
MH | -0.258 |
ic | -0.255 |
Token | 23 |
Token ID | 1954 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
alks | +0.251 |
alt | +0.239 |
iley | +0.227 |
MH | +0.223 |
ic | +0.220 |
Neg logit contributions | |
shoulder | -0.219 |
handc | -0.219 |
Bench | -0.214 |
� | -0.202 |
� | -0.201 |
Token | 12 |
Token ID | 1065 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
alks | +0.705 |
alt | +0.671 |
iley | +0.637 |
MH | +0.624 |
ic | +0.617 |
Neg logit contributions | |
shoulder | -0.642 |
handc | -0.642 |
Bench | -0.626 |
� | -0.590 |
� | -0.589 |
Token | . |
Token ID | 13 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
alks | +0.491 |
alt | +0.465 |
iley | +0.440 |
MH | +0.433 |
ic | +0.426 |
Neg logit contributions | |
handc | -0.542 |
shoulder | -0.538 |
Bench | -0.525 |
� | -0.503 |
� | -0.501 |
Token | Th |
Token ID | 817 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
alks | +0.590 |
alt | +0.560 |
iley | +0.531 |
MH | +0.523 |
ic | +0.515 |
Neg logit contributions | |
shoulder | -0.576 |
handc | -0.575 |
Bench | -0.563 |
� | -0.536 |
� | -0.534 |
Token | ING |
Token ID | 2751 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
alks | +0.131 |
alt | +0.124 |
iley | +0.117 |
MH | +0.114 |
ic | +0.113 |
Neg logit contributions | |
shoulder | -0.149 |
handc | -0.149 |
Bench | -0.146 |
� | -0.139 |
� | -0.139 |
Token | Premiere |
Token ID | 44151 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
alks | +0.120 |
alt | +0.114 |
iley | +0.108 |
MH | +0.106 |
ic | +0.104 |
Neg logit contributions | |
shoulder | -0.116 |
handc | -0.116 |
Bench | -0.114 |
� | -0.108 |
� | -0.108 |
Token | ( |
Token ID | 357 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
alks | +0.351 |
alt | +0.333 |
iley | +0.316 |
MH | +0.310 |
ic | +0.306 |
Neg logit contributions | |
shoulder | -0.351 |
handc | -0.351 |
Bench | -0.343 |
� | -0.326 |
� | -0.326 |
Token | by |
Token ID | 1525 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
alks | +0.038 |
alt | +0.036 |
iley | +0.034 |
MH | +0.033 |
ic | +0.033 |
Neg logit contributions | |
shoulder | -0.041 |
handc | -0.041 |
Bench | -0.040 |
� | -0.038 |
� | -0.038 |
Token | now |
Token ID | 783 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
alks | +0.183 |
alt | +0.173 |
iley | +0.164 |
MH | +0.161 |
ic | +0.159 |
Neg logit contributions | |
shoulder | -0.176 |
handc | -0.175 |
Bench | -0.172 |
� | -0.163 |
� | -0.163 |
Token | you |
Token ID | 345 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
shoulder | +3.668 |
handc | +3.601 |
Bench | +3.595 |
� | +3.349 |
� | +3.329 |
Neg logit contributions | |
alks | -4.063 |
alt | -3.806 |
azeera | -3.628 |
iley | -3.596 |
iens | -3.526 |
Token | have |
Token ID | 423 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
shoulder | +0.447 |
handc | +0.446 |
Bench | +0.437 |
� | +0.412 |
� | +0.411 |
Neg logit contributions | |
alks | -0.519 |
alt | -0.493 |
iley | -0.470 |
MH | -0.461 |
ic | -0.456 |
Token | probably |
Token ID | 2192 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
shoulder | +1.496 |
handc | +1.483 |
Bench | +1.468 |
� | +1.377 |
� | +1.368 |
Neg logit contributions | |
alks | -1.706 |
alt | -1.607 |
iley | -1.528 |
MH | -1.495 |
azeera | -1.491 |
Token | figured |
Token ID | 11638 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
shoulder | +2.504 |
handc | +2.473 |
Bench | +2.438 |
� | +2.273 |
� | +2.240 |
Neg logit contributions | |
alks | -2.762 |
alt | -2.586 |
iley | -2.460 |
azeera | -2.441 |
iens | -2.418 |
Token | out |
Token ID | 503 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
shoulder | +2.285 |
handc | +2.249 |
Bench | +2.230 |
� | +2.101 |
� | +2.087 |
Neg logit contributions | |
alks | -2.572 |
alt | -2.444 |
iley | -2.305 |
azeera | -2.279 |
MH | -2.243 |
Token | care |
Token ID | 6651 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
shoulder | +0.363 |
handc | +0.362 |
Bench | +0.356 |
� | +0.338 |
� | +0.337 |
Neg logit contributions | |
alks | -0.353 |
alt | -0.334 |
iley | -0.317 |
MH | -0.310 |
ic | -0.306 |
Token | er |
Token ID | 263 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
shoulder | +1.058 |
handc | +1.055 |
Bench | +1.041 |
� | +0.995 |
� | +0.993 |
Neg logit contributions | |
alks | -0.645 |
alt | -0.597 |
iley | -0.561 |
MH | -0.544 |
azeera | -0.534 |
Token | teachers |
Token ID | 7799 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
alks | +0.454 |
alt | +0.432 |
iley | +0.410 |
MH | +0.403 |
ic | +0.399 |
Neg logit contributions | |
shoulder | -0.408 |
handc | -0.408 |
Bench | -0.400 |
� | -0.379 |
� | -0.378 |
Token | . |
Token ID | 13 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
shoulder | +0.511 |
handc | +0.509 |
Bench | +0.499 |
� | +0.470 |
� | +0.469 |
Neg logit contributions | |
alks | -0.619 |
alt | -0.589 |
iley | -0.561 |
MH | -0.551 |
azeera | -0.544 |
Token | The |
Token ID | 383 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
alks | +1.021 |
alt | +0.966 |
iley | +0.906 |
MH | +0.904 |
ic | +0.882 |
Neg logit contributions | |
handc | -1.161 |
shoulder | -1.153 |
Bench | -1.111 |
� | -1.073 |
� | -1.069 |
Token | opposite |
Token ID | 6697 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
alks | +0.385 |
alt | +0.367 |
iley | +0.349 |
MH | +0.344 |
ic | +0.340 |
Neg logit contributions | |
shoulder | -0.307 |
handc | -0.307 |
Bench | -0.300 |
� | -0.284 |
� | -0.283 |
Token | may |
Token ID | 743 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
shoulder | +3.636 |
handc | +3.563 |
Bench | +3.514 |
stitching | +3.272 |
� | +3.255 |
Neg logit contributions | |
alks | -4.111 |
alt | -3.861 |
iley | -3.678 |
azeera | -3.650 |
MH | -3.592 |
Token | be |
Token ID | 307 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
alks | +0.575 |
alt | +0.544 |
iley | +0.515 |
MH | +0.505 |
ic | +0.498 |
Neg logit contributions | |
shoulder | -0.598 |
handc | -0.597 |
Bench | -0.587 |
� | -0.558 |
� | -0.556 |
Token | true |
Token ID | 2081 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
alks | +0.899 |
alt | +0.862 |
iley | +0.827 |
MH | +0.817 |
ic | +0.809 |
Neg logit contributions | |
shoulder | -0.587 |
handc | -0.587 |
Bench | -0.574 |
� | -0.538 |
� | -0.534 |
Token | if |
Token ID | 611 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
handc | +0.433 |
shoulder | +0.433 |
Bench | +0.426 |
� | +0.405 |
� | +0.404 |
Neg logit contributions | |
alks | -0.412 |
alt | -0.391 |
iley | -0.370 |
MH | -0.363 |
ic | -0.359 |
Token | teachers |
Token ID | 7799 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
alks | +0.034 |
alt | +0.032 |
iley | +0.031 |
MH | +0.030 |
ic | +0.030 |
Neg logit contributions | |
shoulder | -0.029 |
handc | -0.029 |
Bench | -0.029 |
� | -0.027 |
� | -0.027 |
Token | J |
Token ID | 41 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
alks | +0.161 |
alt | +0.152 |
iley | +0.144 |
MH | +0.141 |
ic | +0.139 |
Neg logit contributions | |
shoulder | -0.167 |
handc | -0.167 |
Bench | -0.164 |
� | -0.156 |
� | -0.156 |
Token | aaaa |
Token ID | 24794 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
alks | +0.783 |
alt | +0.739 |
iley | +0.696 |
MH | +0.680 |
ic | +0.671 |
Neg logit contributions | |
shoulder | -0.921 |
handc | -0.919 |
Bench | -0.904 |
� | -0.860 |
� | -0.858 |
Token | asper |
Token ID | 32981 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
alks | +1.789 |
alt | +1.689 |
iley | +1.583 |
MH | +1.520 |
ic | +1.498 |
Neg logit contributions | |
handc | -2.446 |
shoulder | -2.435 |
Bench | -2.385 |
� | -2.276 |
� | -2.266 |
Token | Profile |
Token ID | 13118 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
alks | +0.201 |
alt | +0.189 |
iley | +0.178 |
MH | +0.175 |
ic | +0.172 |
Neg logit contributions | |
shoulder | -0.245 |
handc | -0.244 |
Bench | -0.241 |
� | -0.229 |
� | -0.229 |
Token | Blog |
Token ID | 14001 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
shoulder | +2.979 |
Bench | +2.977 |
handc | +2.917 |
� | +2.770 |
� | +2.748 |
Neg logit contributions | |
alks | -3.817 |
alt | -3.627 |
iley | -3.470 |
azeera | -3.404 |
MH | -3.368 |
Token | Joined |
Token ID | 16709 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
shoulder | +1.891 |
Bench | +1.870 |
handc | +1.860 |
� | +1.760 |
� | +1.746 |
Neg logit contributions | |
alks | -2.098 |
alt | -1.982 |
iley | -1.891 |
azeera | -1.853 |
MH | -1.845 |
Token | April |
Token ID | 3035 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
Bench | +3.012 |
shoulder | +3.001 |
handc | +2.867 |
� | +2.700 |
� | +2.689 |
Neg logit contributions | |
alks | -4.596 |
alt | -4.365 |
azeera | -4.275 |
iley | -4.222 |
iens | -4.082 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
shoulder | +0.236 |
handc | +0.234 |
Bench | +0.232 |
� | +0.219 |
� | +0.218 |
Neg logit contributions | |
alks | -0.246 |
alt | -0.232 |
iley | -0.221 |
MH | -0.216 |
azeera | -0.214 |
Token | United |
Token ID | 1578 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
shoulder | +0.324 |
handc | +0.322 |
Bench | +0.318 |
� | +0.299 |
� | +0.299 |
Neg logit contributions | |
alks | -0.360 |
alt | -0.341 |
iley | -0.325 |
MH | -0.318 |
azeera | -0.316 |
Token | States |
Token ID | 1829 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
shoulder | +0.590 |
handc | +0.584 |
Bench | +0.580 |
� | +0.546 |
� | +0.544 |
Neg logit contributions | |
alks | -0.686 |
alt | -0.646 |
iley | -0.620 |
MH | -0.604 |
azeera | -0.603 |
Token | 5 |
Token ID | 642 |
Feature activation | -9.47e-02 |
Pos logit contributions | |
shoulder | +1.139 |
handc | +1.129 |
Bench | +1.119 |
� | +1.046 |
� | +1.044 |
Neg logit contributions | |
alks | -1.403 |
alt | -1.333 |
iley | -1.270 |
azeera | -1.243 |
MH | -1.242 |
Token | both |
Token ID | 1111 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
shoulder | +1.600 |
handc | +1.577 |
Bench | +1.555 |
� | +1.466 |
� | +1.458 |
Neg logit contributions | |
alks | -1.838 |
alt | -1.741 |
iley | -1.669 |
MH | -1.626 |
azeera | -1.614 |
Token | . |
Token ID | 13 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
shoulder | +0.217 |
handc | +0.215 |
Bench | +0.212 |
� | +0.200 |
� | +0.200 |
Neg logit contributions | |
alks | -0.223 |
alt | -0.211 |
iley | -0.200 |
MH | -0.196 |
azeera | -0.194 |
Token | I |
Token ID | 314 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
shoulder | +1.405 |
Bench | +1.401 |
handc | +1.383 |
� | +1.312 |
� | +1.312 |
Neg logit contributions | |
alks | -1.457 |
alt | -1.376 |
iley | -1.312 |
azeera | -1.292 |
MH | -1.265 |
Token | did |
Token ID | 750 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
shoulder | +0.786 |
handc | +0.779 |
Bench | +0.765 |
� | +0.721 |
� | +0.719 |
Neg logit contributions | |
alks | -0.864 |
alt | -0.816 |
iley | -0.778 |
MH | -0.763 |
azeera | -0.759 |
Token | not |
Token ID | 407 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
shoulder | +1.105 |
handc | +1.097 |
Bench | +1.076 |
� | +1.012 |
� | +1.009 |
Neg logit contributions | |
alks | -1.267 |
alt | -1.198 |
iley | -1.141 |
MH | -1.120 |
azeera | -1.107 |
Token | expect |
Token ID | 1607 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
shoulder | +0.450 |
handc | +0.448 |
Bench | +0.439 |
� | +0.413 |
� | +0.412 |
Neg logit contributions | |
alks | -0.510 |
alt | -0.483 |
iley | -0.460 |
MH | -0.452 |
azeera | -0.446 |
Token | to |
Token ID | 284 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
shoulder | +3.415 |
handc | +3.360 |
Bench | +3.311 |
� | +3.106 |
� | +3.076 |
Neg logit contributions | |
alks | -4.178 |
alt | -3.932 |
iley | -3.790 |
azeera | -3.747 |
MH | -3.663 |
Token | be |
Token ID | 307 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
shoulder | +1.288 |
handc | +1.274 |
Bench | +1.239 |
� | +1.158 |
� | +1.158 |
Neg logit contributions | |
alks | -1.476 |
alt | -1.393 |
iley | -1.322 |
MH | -1.304 |
azeera | -1.295 |
Token | confronted |
Token ID | 16675 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
shoulder | +1.547 |
handc | +1.527 |
Bench | +1.495 |
� | +1.399 |
stitching | +1.395 |
Neg logit contributions | |
alks | -1.808 |
alt | -1.712 |
iley | -1.627 |
azeera | -1.605 |
MH | -1.590 |
Token | almost |
Token ID | 2048 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
shoulder | +1.366 |
handc | +1.350 |
Bench | +1.315 |
� | +1.231 |
� | +1.230 |
Neg logit contributions | |
alks | -1.928 |
alt | -1.837 |
iley | -1.754 |
MH | -1.730 |
azeera | -1.710 |
Token | immediately |
Token ID | 3393 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
alks | +0.404 |
alt | +0.385 |
iley | +0.367 |
MH | +0.363 |
ic | +0.361 |
Neg logit contributions | |
handc | -0.320 |
shoulder | -0.318 |
Bench | -0.316 |
� | -0.300 |
� | -0.297 |
Token | been |
Token ID | 587 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
shoulder | +0.131 |
handc | +0.131 |
Bench | +0.128 |
� | +0.122 |
� | +0.121 |
Neg logit contributions | |
alks | -0.158 |
alt | -0.151 |
iley | -0.144 |
MH | -0.142 |
ic | -0.141 |
Token | closed |
Token ID | 4838 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
alks | +0.164 |
alt | +0.154 |
iley | +0.145 |
MH | +0.142 |
ic | +0.141 |
Neg logit contributions | |
handc | -0.201 |
shoulder | -0.199 |
Bench | -0.196 |
� | -0.189 |
� | -0.189 |
Token | . |
Token ID | 13 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
shoulder | +0.104 |
handc | +0.104 |
Bench | +0.102 |
� | +0.097 |
� | +0.097 |
Neg logit contributions | |
alks | -0.106 |
alt | -0.101 |
iley | -0.096 |
MH | -0.094 |
ic | -0.092 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +6.70e-03 |
Pos logit contributions | |
shoulder | +1.669 |
handc | +1.660 |
Bench | +1.652 |
� | +1.546 |
� | +1.541 |
Neg logit contributions | |
alks | -1.950 |
alt | -1.856 |
iley | -1.769 |
MH | -1.725 |
azeera | -1.718 |
Token | ( |
Token ID | 7 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
alks | +0.017 |
alt | +0.016 |
iley | +0.015 |
MH | +0.015 |
ic | +0.015 |
Neg logit contributions | |
shoulder | -0.018 |
handc | -0.018 |
Bench | -0.018 |
� | -0.017 |
� | -0.017 |
Token | J |
Token ID | 41 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
alks | +0.796 |
alt | +0.748 |
iley | +0.704 |
MH | +0.689 |
ic | +0.679 |
Neg logit contributions | |
shoulder | -0.987 |
handc | -0.985 |
Bench | -0.969 |
� | -0.924 |
� | -0.922 |
Token | TA |
Token ID | 5603 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
alks | +2.004 |
alt | +1.899 |
iley | +1.775 |
MH | +1.740 |
azeera | +1.730 |
Neg logit contributions | |
handc | -2.417 |
shoulder | -2.412 |
Bench | -2.360 |
� | -2.260 |
� | -2.252 |
Token | ) |
Token ID | 8 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
alks | +0.912 |
alt | +0.859 |
iley | +0.811 |
MH | +0.794 |
ic | +0.783 |
Neg logit contributions | |
shoulder | -1.061 |
handc | -1.059 |
Bench | -1.041 |
� | -0.991 |
� | -0.989 |
Token | — |
Token ID | 851 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
alks | +1.537 |
alt | +1.441 |
iley | +1.350 |
MH | +1.340 |
ic | +1.313 |
Neg logit contributions | |
handc | -1.902 |
shoulder | -1.897 |
Bench | -1.852 |
� | -1.780 |
� | -1.775 |
Token | Finland |
Token ID | 17837 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
alks | +1.572 |
alt | +1.485 |
iley | +1.399 |
MH | +1.394 |
ic | +1.364 |
Neg logit contributions | |
handc | -1.698 |
shoulder | -1.686 |
Bench | -1.634 |
� | -1.572 |
� | -1.568 |
Token | � |
Token ID | 447 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
alks | +0.117 |
alt | +0.111 |
iley | +0.106 |
MH | +0.104 |
azeera | +0.103 |
Neg logit contributions | |
shoulder | -0.102 |
handc | -0.101 |
Bench | -0.099 |
� | -0.094 |
� | -0.093 |
Token | since |
Token ID | 1201 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
handc | +0.359 |
shoulder | +0.358 |
Bench | +0.351 |
� | +0.332 |
� | +0.330 |
Neg logit contributions | |
alks | -0.420 |
alt | -0.400 |
iley | -0.381 |
MH | -0.375 |
ic | -0.371 |
Token | 1901 |
Token ID | 39923 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
alks | +0.244 |
alt | +0.232 |
iley | +0.221 |
MH | +0.217 |
ic | +0.214 |
Neg logit contributions | |
shoulder | -0.217 |
handc | -0.217 |
Bench | -0.213 |
� | -0.201 |
� | -0.200 |
Token | . |
Token ID | 13 |
Feature activation | +9.27e-03 |
Pos logit contributions | |
shoulder | +1.038 |
handc | +1.031 |
Bench | +1.017 |
� | +0.961 |
� | +0.960 |
Neg logit contributions | |
alks | -1.057 |
alt | -0.999 |
iley | -0.949 |
MH | -0.928 |
azeera | -0.919 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
alks | +0.024 |
alt | +0.023 |
iley | +0.022 |
MH | +0.022 |
ic | +0.021 |
Neg logit contributions | |
handc | -0.024 |
shoulder | -0.024 |
Bench | -0.023 |
� | -0.022 |
� | -0.022 |
Token | A |
Token ID | 32 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
shoulder | +0.432 |
handc | +0.431 |
Bench | +0.425 |
� | +0.407 |
� | +0.407 |
Neg logit contributions | |
alks | -0.410 |
alt | -0.384 |
iley | -0.366 |
MH | -0.360 |
azeera | -0.355 |
Token | bicycl |
Token ID | 48094 |
Feature activation | +8.17e-01 |
Pos logit contributions | |
alks | +0.202 |
alt | +0.191 |
iley | +0.182 |
MH | +0.179 |
ic | +0.177 |
Neg logit contributions | |
shoulder | -0.179 |
handc | -0.178 |
Bench | -0.175 |
� | -0.165 |
� | -0.165 |
Token | ist |
Token ID | 396 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
alks | +1.505 |
alt | +1.404 |
iley | +1.294 |
MH | +1.248 |
ic | +1.228 |
Neg logit contributions | |
handc | -2.747 |
shoulder | -2.746 |
Bench | -2.706 |
� | -2.608 |
� | -2.601 |
Token | was |
Token ID | 373 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
alks | +0.394 |
alt | +0.373 |
iley | +0.355 |
MH | +0.347 |
azeera | +0.346 |
Neg logit contributions | |
shoulder | -0.347 |
handc | -0.345 |
Bench | -0.337 |
� | -0.315 |
� | -0.315 |
Token | taken |
Token ID | 2077 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
shoulder | +0.098 |
handc | +0.097 |
Bench | +0.095 |
� | +0.090 |
� | +0.090 |
Neg logit contributions | |
alks | -0.088 |
alt | -0.083 |
iley | -0.078 |
MH | -0.076 |
azeera | -0.076 |
Token | to |
Token ID | 284 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
alks | +0.872 |
alt | +0.835 |
iley | +0.793 |
MH | +0.783 |
ic | +0.773 |
Neg logit contributions | |
shoulder | -0.836 |
handc | -0.835 |
Bench | -0.827 |
� | -0.795 |
� | -0.785 |
Token | the |
Token ID | 262 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
shoulder | +0.176 |
handc | +0.176 |
Bench | +0.172 |
� | +0.164 |
� | +0.163 |
Neg logit contributions | |
alks | -0.189 |
alt | -0.179 |
iley | -0.170 |
MH | -0.168 |
ic | -0.165 |
Token | how |
Token ID | 703 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
alks | +0.828 |
alt | +0.788 |
iley | +0.750 |
MH | +0.740 |
ic | +0.731 |
Neg logit contributions | |
handc | -0.722 |
shoulder | -0.720 |
Bench | -0.708 |
� | -0.672 |
� | -0.670 |
Token | non |
Token ID | 1729 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
shoulder | +0.239 |
handc | +0.239 |
Bench | +0.234 |
� | +0.220 |
� | +0.220 |
Neg logit contributions | |
alks | -0.276 |
alt | -0.262 |
iley | -0.249 |
MH | -0.244 |
azeera | -0.242 |
Token | chal |
Token ID | 38009 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
shoulder | +0.339 |
handc | +0.338 |
Bench | +0.332 |
� | +0.314 |
� | +0.313 |
Neg logit contributions | |
alks | -0.361 |
alt | -0.342 |
iley | -0.325 |
MH | -0.319 |
ic | -0.315 |
Token | antly |
Token ID | 3875 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
alks | +0.243 |
alt | +0.227 |
iley | +0.213 |
MH | +0.209 |
azeera | +0.206 |
Neg logit contributions | |
shoulder | -0.337 |
handc | -0.336 |
Bench | -0.331 |
� | -0.316 |
� | -0.316 |
Token | some |
Token ID | 617 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
alks | +0.217 |
alt | +0.206 |
iley | +0.195 |
MH | +0.193 |
ic | +0.190 |
Neg logit contributions | |
handc | -0.201 |
shoulder | -0.201 |
Bench | -0.197 |
� | -0.187 |
� | -0.187 |
Token | say |
Token ID | 910 |
Feature activation | +7.75e-01 |
Pos logit contributions | |
alks | +1.683 |
alt | +1.605 |
iley | +1.505 |
MH | +1.498 |
ic | +1.481 |
Neg logit contributions | |
handc | -1.621 |
shoulder | -1.609 |
Bench | -1.587 |
� | -1.524 |
� | -1.515 |
Token | that |
Token ID | 326 |
Feature activation | +7.39e-03 |
Pos logit contributions | |
alks | +1.999 |
alt | +1.902 |
iley | +1.798 |
MH | +1.782 |
ic | +1.758 |
Neg logit contributions | |
handc | -2.014 |
shoulder | -2.003 |
Bench | -1.979 |
� | -1.897 |
� | -1.892 |
Token | it |
Token ID | 340 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
alks | +0.020 |
alt | +0.019 |
iley | +0.018 |
MH | +0.018 |
ic | +0.018 |
Neg logit contributions | |
shoulder | -0.018 |
handc | -0.018 |
Bench | -0.018 |
� | -0.017 |
� | -0.017 |
Token | doesn |
Token ID | 1595 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
alks | +0.518 |
alt | +0.492 |
iley | +0.469 |
MH | +0.461 |
ic | +0.457 |
Neg logit contributions | |
shoulder | -0.455 |
handc | -0.455 |
Bench | -0.447 |
� | -0.424 |
� | -0.422 |
Token | 't |
Token ID | 470 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
alks | +0.659 |
alt | +0.624 |
iley | +0.582 |
MH | +0.564 |
ic | +0.559 |
Neg logit contributions | |
handc | -1.034 |
shoulder | -1.022 |
Bench | -1.000 |
stitching | -0.960 |
� | -0.955 |
Token | matter |
Token ID | 2300 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
alks | +0.278 |
alt | +0.264 |
iley | +0.251 |
MH | +0.247 |
ic | +0.244 |
Neg logit contributions | |
shoulder | -0.261 |
handc | -0.261 |
Bench | -0.257 |
� | -0.244 |
� | -0.243 |
Token | new |
Token ID | 649 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
alks | +0.481 |
alt | +0.458 |
iley | +0.436 |
MH | +0.429 |
ic | +0.424 |
Neg logit contributions | |
shoulder | -0.407 |
handc | -0.407 |
Bench | -0.399 |
� | -0.377 |
� | -0.376 |
Token | version |
Token ID | 2196 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
shoulder | +0.930 |
handc | +0.921 |
Bench | +0.903 |
� | +0.854 |
� | +0.852 |
Neg logit contributions | |
alks | -0.922 |
alt | -0.867 |
iley | -0.821 |
MH | -0.800 |
azeera | -0.796 |
Token | of |
Token ID | 286 |
Feature activation | -2.23e-03 |
Pos logit contributions | |
shoulder | +0.792 |
handc | +0.785 |
Bench | +0.771 |
� | +0.721 |
� | +0.720 |
Neg logit contributions | |
alks | -1.032 |
alt | -0.979 |
iley | -0.933 |
azeera | -0.912 |
MH | -0.911 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
shoulder | +0.005 |
handc | +0.005 |
Bench | +0.005 |
� | +0.005 |
� | +0.005 |
Neg logit contributions | |
alks | -0.006 |
alt | -0.006 |
iley | -0.006 |
MH | -0.005 |
azeera | -0.005 |
Token | H |
Token ID | 367 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
shoulder | +0.282 |
handc | +0.280 |
Bench | +0.275 |
� | +0.258 |
� | +0.258 |
Neg logit contributions | |
alks | -0.311 |
alt | -0.294 |
iley | -0.280 |
MH | -0.272 |
azeera | -0.272 |
Token | 1 |
Token ID | 16 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
alks | +0.381 |
alt | +0.359 |
iley | +0.335 |
MH | +0.325 |
ic | +0.322 |
Neg logit contributions | |
shoulder | -0.567 |
handc | -0.566 |
Bench | -0.560 |
� | -0.535 |
� | -0.532 |
Token | N |
Token ID | 45 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
alks | +2.005 |
alt | +1.927 |
iley | +1.794 |
ic | +1.771 |
MH | +1.770 |
Neg logit contributions | |
handc | -2.497 |
shoulder | -2.491 |
Bench | -2.467 |
� | -2.366 |
� | -2.350 |
Token | 1 |
Token ID | 16 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
alks | +0.284 |
alt | +0.266 |
iley | +0.243 |
MH | +0.235 |
ic | +0.233 |
Neg logit contributions | |
shoulder | -0.523 |
handc | -0.523 |
Bench | -0.516 |
� | -0.495 |
� | -0.494 |
Token | strain |
Token ID | 14022 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
alks | +1.034 |
alt | +0.977 |
iley | +0.921 |
MH | +0.902 |
ic | +0.893 |
Neg logit contributions | |
handc | -1.225 |
shoulder | -1.225 |
Bench | -1.206 |
� | -1.149 |
� | -1.146 |
Token | which |
Token ID | 543 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
shoulder | +0.432 |
handc | +0.429 |
Bench | +0.421 |
� | +0.398 |
� | +0.397 |
Neg logit contributions | |
alks | -0.450 |
alt | -0.425 |
iley | -0.403 |
azeera | -0.393 |
MH | -0.391 |
Token | caused |
Token ID | 4073 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
alks | +1.294 |
alt | +1.227 |
iley | +1.169 |
MH | +1.150 |
ic | +1.137 |
Neg logit contributions | |
handc | -1.193 |
shoulder | -1.193 |
Bench | -1.170 |
� | -1.108 |
� | -1.106 |
Token | enjoying |
Token ID | 13226 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
alks | +0.890 |
alt | +0.849 |
iley | +0.810 |
MH | +0.797 |
ic | +0.789 |
Neg logit contributions | |
shoulder | -0.689 |
handc | -0.688 |
Bench | -0.674 |
� | -0.636 |
� | -0.634 |
Token | its |
Token ID | 663 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
alks | +1.284 |
alt | +1.220 |
iley | +1.162 |
MH | +1.144 |
ic | +1.132 |
Neg logit contributions | |
shoulder | -1.103 |
handc | -1.102 |
Bench | -1.081 |
� | -1.022 |
� | -1.022 |
Token | very |
Token ID | 845 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
alks | +1.132 |
alt | +1.082 |
iley | +1.034 |
MH | +1.018 |
ic | +1.007 |
Neg logit contributions | |
shoulder | -0.793 |
handc | -0.791 |
Bench | -0.774 |
� | -0.726 |
� | -0.724 |
Token | own |
Token ID | 898 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
alks | +0.496 |
alt | +0.470 |
iley | +0.448 |
MH | +0.441 |
azeera | +0.436 |
Neg logit contributions | |
shoulder | -0.419 |
handc | -0.417 |
Bench | -0.408 |
� | -0.383 |
� | -0.382 |
Token | network |
Token ID | 3127 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
alks | +1.067 |
alt | +1.010 |
iley | +0.959 |
MH | +0.940 |
ic | +0.927 |
Neg logit contributions | |
shoulder | -1.048 |
handc | -1.045 |
Bench | -1.027 |
� | -0.972 |
� | -0.970 |
Token | renaissance |
Token ID | 48917 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
alks | +0.788 |
alt | +0.752 |
iley | +0.712 |
MH | +0.701 |
ic | +0.695 |
Neg logit contributions | |
handc | -0.756 |
shoulder | -0.753 |
Bench | -0.740 |
� | -0.705 |
� | -0.704 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
alks | +2.253 |
alt | +2.143 |
iley | +2.020 |
MH | +1.994 |
ic | +1.986 |
Neg logit contributions | |
handc | -2.315 |
shoulder | -2.303 |
Bench | -2.278 |
� | -2.188 |
� | -2.169 |
Token | After |
Token ID | 2293 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
alks | +0.272 |
alt | +0.257 |
iley | +0.244 |
MH | +0.239 |
ic | +0.236 |
Neg logit contributions | |
shoulder | -0.273 |
handc | -0.273 |
Bench | -0.268 |
� | -0.254 |
� | -0.253 |
Token | a |
Token ID | 257 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
alks | +1.518 |
alt | +1.449 |
iley | +1.379 |
MH | +1.362 |
ic | +1.350 |
Neg logit contributions | |
handc | -1.217 |
shoulder | -1.212 |
Bench | -1.188 |
� | -1.134 |
� | -1.131 |
Token | period |
Token ID | 2278 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
shoulder | +0.067 |
handc | +0.067 |
Bench | +0.066 |
� | +0.061 |
� | +0.061 |
Neg logit contributions | |
alks | -0.084 |
alt | -0.079 |
iley | -0.076 |
MH | -0.074 |
azeera | -0.074 |
Token | of |
Token ID | 286 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
alks | +0.823 |
alt | +0.783 |
iley | +0.747 |
MH | +0.734 |
ic | +0.726 |
Neg logit contributions | |
shoulder | -0.674 |
handc | -0.673 |
Bench | -0.660 |
� | -0.622 |
� | -0.620 |
Token | Che |
Token ID | 7376 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
alks | +0.885 |
alt | +0.841 |
iley | +0.799 |
MH | +0.789 |
ic | +0.777 |
Neg logit contributions | |
shoulder | -0.719 |
handc | -0.717 |
Bench | -0.699 |
� | -0.666 |
� | -0.661 |
Token | f |
Token ID | 69 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
alks | +0.582 |
alt | +0.555 |
iley | +0.529 |
MH | +0.517 |
ic | +0.511 |
Neg logit contributions | |
shoulder | -0.458 |
handc | -0.458 |
Bench | -0.445 |
� | -0.422 |
� | -0.421 |
Token | 2015 |
Token ID | 1853 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
shoulder | +0.070 |
handc | +0.070 |
Bench | +0.069 |
� | +0.065 |
� | +0.064 |
Neg logit contributions | |
alks | -0.100 |
alt | -0.095 |
iley | -0.091 |
MH | -0.090 |
ic | -0.089 |
Token |
|
Token ID | 198 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
shoulder | +0.091 |
handc | +0.090 |
Bench | +0.088 |
� | +0.081 |
� | +0.081 |
Neg logit contributions | |
alks | -0.167 |
alt | -0.160 |
iley | -0.153 |
MH | -0.151 |
ic | -0.150 |
Token |
|
Token ID | 198 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
shoulder | +0.539 |
handc | +0.537 |
Bench | +0.529 |
� | +0.500 |
� | +0.500 |
Neg logit contributions | |
alks | -0.549 |
alt | -0.519 |
iley | -0.493 |
MH | -0.483 |
azeera | -0.478 |
Token | * |
Token ID | 9 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
shoulder | +0.307 |
handc | +0.305 |
Bench | +0.303 |
� | +0.289 |
� | +0.289 |
Neg logit contributions | |
alks | -0.251 |
alt | -0.233 |
iley | -0.221 |
MH | -0.218 |
ic | -0.215 |
Token | Master |
Token ID | 5599 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
alks | +0.026 |
alt | +0.024 |
iley | +0.023 |
MH | +0.023 |
azeera | +0.022 |
Neg logit contributions | |
shoulder | -0.028 |
handc | -0.028 |
Bench | -0.027 |
� | -0.026 |
� | -0.026 |
Token | che |
Token ID | 2395 |
Feature activation | -7.90e-02 |
Pos logit contributions | |
alks | +0.952 |
alt | +0.902 |
iley | +0.858 |
MH | +0.844 |
ic | +0.832 |
Neg logit contributions | |
shoulder | -0.829 |
handc | -0.827 |
Bench | -0.809 |
� | -0.768 |
� | -0.763 |
Token | f |
Token ID | 69 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
shoulder | +0.177 |
handc | +0.176 |
Bench | +0.172 |
� | +0.162 |
� | +0.162 |
Neg logit contributions | |
alks | -0.235 |
alt | -0.225 |
iley | -0.214 |
MH | -0.211 |
ic | -0.208 |
Token | NZ |
Token ID | 26905 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
alks | +0.234 |
alt | +0.223 |
iley | +0.212 |
MH | +0.208 |
azeera | +0.206 |
Neg logit contributions | |
shoulder | -0.189 |
handc | -0.189 |
Bench | -0.185 |
� | -0.173 |
� | -0.173 |
Token | fails |
Token ID | 10143 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
shoulder | +0.355 |
handc | +0.352 |
Bench | +0.347 |
� | +0.323 |
� | +0.323 |
Neg logit contributions | |
alks | -0.421 |
alt | -0.400 |
iley | -0.381 |
MH | -0.373 |
azeera | -0.369 |
Token | . |
Token ID | 13 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
shoulder | +0.669 |
handc | +0.667 |
Bench | +0.655 |
� | +0.618 |
� | +0.617 |
Neg logit contributions | |
alks | -0.721 |
alt | -0.683 |
iley | -0.649 |
MH | -0.636 |
azeera | -0.629 |
Token | But |
Token ID | 887 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
shoulder | +0.080 |
Bench | +0.079 |
handc | +0.079 |
� | +0.075 |
� | +0.074 |
Neg logit contributions | |
alks | -0.082 |
alt | -0.077 |
iley | -0.074 |
azeera | -0.072 |
MH | -0.071 |
Token | , |
Token ID | 11 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
alks | +1.215 |
alt | +1.153 |
iley | +1.096 |
MH | +1.076 |
ic | +1.063 |
Neg logit contributions | |
shoulder | -1.100 |
handc | -1.097 |
Bench | -1.077 |
� | -1.017 |
� | -1.015 |
Token | if |
Token ID | 611 |
Feature activation | -1.70e-03 |
Pos logit contributions | |
alks | +1.035 |
alt | +0.983 |
iley | +0.935 |
MH | +0.920 |
ic | +0.909 |
Neg logit contributions | |
shoulder | -0.930 |
handc | -0.929 |
Bench | -0.911 |
� | -0.863 |
� | -0.862 |
Token | I |
Token ID | 314 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
shoulder | +0.004 |
handc | +0.004 |
Bench | +0.004 |
� | +0.004 |
� | +0.004 |
Neg logit contributions | |
alks | -0.005 |
alt | -0.004 |
iley | -0.004 |
azeera | -0.004 |
MH | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
alks | +0.915 |
alt | +0.877 |
iley | +0.829 |
MH | +0.816 |
ic | +0.807 |
Neg logit contributions | |
handc | -0.863 |
shoulder | -0.862 |
Bench | -0.849 |
� | -0.813 |
� | -0.812 |
Token | to |
Token ID | 284 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
alks | +0.471 |
alt | +0.448 |
iley | +0.427 |
MH | +0.420 |
ic | +0.415 |
Neg logit contributions | |
shoulder | -0.394 |
handc | -0.393 |
Bench | -0.385 |
� | -0.363 |
� | -0.363 |
Token | take |
Token ID | 1011 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
shoulder | +1.016 |
handc | +1.012 |
Bench | +0.984 |
� | +0.918 |
� | +0.909 |
Neg logit contributions | |
alks | -1.302 |
alt | -1.236 |
iley | -1.182 |
MH | -1.156 |
azeera | -1.154 |
Token | a |
Token ID | 257 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
alks | +1.011 |
alt | +0.961 |
iley | +0.914 |
MH | +0.899 |
ic | +0.888 |
Neg logit contributions | |
shoulder | -0.866 |
handc | -0.864 |
Bench | -0.847 |
� | -0.799 |
� | -0.798 |
Token | guess |
Token ID | 4724 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
alks | +0.252 |
alt | +0.239 |
iley | +0.227 |
MH | +0.222 |
ic | +0.219 |
Neg logit contributions | |
shoulder | -0.234 |
handc | -0.233 |
Bench | -0.229 |
� | -0.217 |
� | -0.216 |
Token | as |
Token ID | 355 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
alks | +0.297 |
alt | +0.280 |
iley | +0.265 |
MH | +0.260 |
ic | +0.257 |
Neg logit contributions | |
shoulder | -0.316 |
handc | -0.316 |
Bench | -0.310 |
� | -0.295 |
� | -0.294 |
Token | to |
Token ID | 284 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
alks | +0.372 |
alt | +0.353 |
iley | +0.335 |
MH | +0.329 |
ic | +0.325 |
Neg logit contributions | |
shoulder | -0.332 |
handc | -0.331 |
Bench | -0.325 |
� | -0.307 |
� | -0.306 |
Token | do |
Token ID | 466 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
alks | +0.238 |
alt | +0.226 |
iley | +0.214 |
MH | +0.210 |
ic | +0.208 |
Neg logit contributions | |
shoulder | -0.234 |
handc | -0.234 |
Bench | -0.230 |
� | -0.218 |
� | -0.217 |
Token | with |
Token ID | 351 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
handc | +0.525 |
shoulder | +0.520 |
Bench | +0.512 |
� | +0.479 |
� | +0.478 |
Neg logit contributions | |
alks | -0.736 |
alt | -0.704 |
iley | -0.674 |
MH | -0.664 |
ic | -0.659 |
Token | the |
Token ID | 262 |
Feature activation | +8.89e-02 |
Pos logit contributions | |
alks | +1.191 |
alt | +1.134 |
iley | +1.083 |
MH | +1.070 |
ic | +1.054 |
Neg logit contributions | |
handc | -1.032 |
shoulder | -1.027 |
Bench | -1.008 |
� | -0.959 |
� | -0.958 |
Token | fact |
Token ID | 1109 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
alks | +0.249 |
alt | +0.237 |
iley | +0.226 |
MH | +0.223 |
ic | +0.220 |
Neg logit contributions | |
handc | -0.213 |
shoulder | -0.212 |
Bench | -0.208 |
� | -0.197 |
� | -0.197 |
Token | that |
Token ID | 326 |
Feature activation | -9.44e-02 |
Pos logit contributions | |
alks | +1.678 |
alt | +1.579 |
iley | +1.495 |
MH | +1.465 |
ic | +1.440 |
Neg logit contributions | |
handc | -2.009 |
shoulder | -2.000 |
Bench | -1.967 |
� | -1.878 |
� | -1.877 |
Token | the |
Token ID | 262 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
shoulder | +0.233 |
handc | +0.232 |
Bench | +0.228 |
� | +0.216 |
� | +0.215 |
Neg logit contributions | |
alks | -0.259 |
alt | -0.246 |
iley | -0.234 |
MH | -0.229 |
ic | -0.227 |
Token | 16 |
Token ID | 1467 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
shoulder | +0.278 |
handc | +0.278 |
Bench | +0.272 |
� | +0.257 |
� | +0.257 |
Neg logit contributions | |
alks | -0.337 |
alt | -0.321 |
iley | -0.306 |
MH | -0.301 |
ic | -0.297 |
Token | - |
Token ID | 12 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
shoulder | +1.178 |
handc | +1.165 |
Bench | +1.151 |
� | +1.090 |
� | +1.085 |
Neg logit contributions | |
alks | -1.159 |
alt | -1.094 |
iley | -1.038 |
MH | -1.013 |
azeera | -1.008 |
Token | 35 |
Token ID | 2327 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
alks | +1.184 |
alt | +1.135 |
iley | +1.084 |
MH | +1.065 |
ic | +1.048 |
Neg logit contributions | |
handc | -0.838 |
shoulder | -0.832 |
Bench | -0.817 |
� | -0.762 |
� | -0.761 |
Token | f |
Token ID | 277 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
alks | +0.106 |
alt | +0.099 |
iley | +0.092 |
MH | +0.090 |
ic | +0.088 |
Neg logit contributions | |
shoulder | -0.172 |
handc | -0.171 |
Bench | -0.169 |
� | -0.162 |
� | -0.161 |
Token | exactly |
Token ID | 3446 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
shoulder | +0.322 |
handc | +0.322 |
Bench | +0.316 |
� | +0.297 |
� | +0.296 |
Neg logit contributions | |
alks | -0.357 |
alt | -0.338 |
iley | -0.320 |
MH | -0.315 |
azeera | -0.314 |
Token | that |
Token ID | 326 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
alks | +0.886 |
alt | +0.841 |
iley | +0.800 |
MH | +0.785 |
ic | +0.776 |
Neg logit contributions | |
shoulder | -0.803 |
handc | -0.801 |
Bench | -0.786 |
� | -0.743 |
� | -0.741 |
Token | clinic |
Token ID | 15760 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
alks | +1.381 |
alt | +1.311 |
iley | +1.244 |
MH | +1.226 |
ic | +1.209 |
Neg logit contributions | |
shoulder | -1.297 |
handc | -1.295 |
Bench | -1.272 |
� | -1.206 |
� | -1.201 |
Token | was |
Token ID | 373 |
Feature activation | +9.27e-01 |
Pos logit contributions | |
alks | +1.120 |
alt | +1.064 |
iley | +1.016 |
MH | +0.998 |
azeera | +0.988 |
Neg logit contributions | |
shoulder | -0.909 |
handc | -0.904 |
Bench | -0.888 |
� | -0.835 |
� | -0.834 |
Token | is |
Token ID | 318 |
Feature activation | +6.80e-01 |
Pos logit contributions | |
alks | +2.353 |
alt | +2.243 |
iley | +2.112 |
MH | +2.075 |
ic | +2.055 |
Neg logit contributions | |
handc | -2.451 |
shoulder | -2.438 |
Bench | -2.395 |
� | -2.280 |
� | -2.273 |
Token | not |
Token ID | 407 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
alks | +1.715 |
alt | +1.622 |
iley | +1.544 |
MH | +1.521 |
ic | +1.507 |
Neg logit contributions | |
handc | -1.801 |
shoulder | -1.797 |
Bench | -1.785 |
� | -1.697 |
� | -1.684 |
Token | totally |
Token ID | 6635 |
Feature activation | +9.33e-01 |
Pos logit contributions | |
alks | +1.637 |
alt | +1.554 |
iley | +1.475 |
MH | +1.461 |
ic | +1.436 |
Neg logit contributions | |
shoulder | -1.584 |
handc | -1.581 |
Bench | -1.573 |
� | -1.486 |
� | -1.479 |
Token | clear |
Token ID | 1598 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
alks | +2.069 |
alt | +1.944 |
iley | +1.817 |
MH | +1.798 |
ic | +1.764 |
Neg logit contributions | |
shoulder | -2.771 |
handc | -2.761 |
Bench | -2.743 |
� | -2.613 |
� | -2.604 |
Token | ; |
Token ID | 26 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
alks | +1.554 |
alt | +1.471 |
iley | +1.391 |
MH | +1.364 |
ic | +1.352 |
Neg logit contributions | |
handc | -1.772 |
shoulder | -1.770 |
Bench | -1.747 |
� | -1.662 |
� | -1.654 |
Token | the |
Token ID | 262 |
Feature activation | +1.42e-03 |
Pos logit contributions | |
alks | +0.447 |
alt | +0.424 |
iley | +0.403 |
MH | +0.395 |
ic | +0.391 |
Neg logit contributions | |
shoulder | -0.403 |
handc | -0.402 |
Bench | -0.395 |
� | -0.372 |
� | -0.371 |
Token | D |
Token ID | 360 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
alks | +0.004 |
alt | +0.004 |
iley | +0.003 |
MH | +0.003 |
ic | +0.003 |
Neg logit contributions | |
shoulder | -0.004 |
handc | -0.004 |
Bench | -0.004 |
� | -0.003 |
� | -0.003 |
Token | series |
Token ID | 2168 |
Feature activation | +3.82e-04 |
Pos logit contributions | |
shoulder | +0.406 |
handc | +0.406 |
Bench | +0.398 |
� | +0.377 |
� | +0.376 |
Neg logit contributions | |
alks | -0.452 |
alt | -0.430 |
iley | -0.409 |
MH | -0.402 |
ic | -0.397 |
Token | of |
Token ID | 286 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
alks | +0.001 |
alt | +0.001 |
iley | +0.001 |
MH | +0.001 |
ic | +0.001 |
Neg logit contributions | |
handc | -0.001 |
shoulder | -0.001 |
Bench | -0.001 |
� | -0.001 |
� | -0.001 |
Token | currency |
Token ID | 7395 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
shoulder | +0.063 |
handc | +0.063 |
Bench | +0.061 |
� | +0.058 |
� | +0.058 |
Neg logit contributions | |
alks | -0.076 |
alt | -0.072 |
iley | -0.069 |
MH | -0.068 |
ic | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
shoulder | +1.130 |
handc | +1.126 |
Bench | +1.107 |
� | +1.051 |
� | +1.049 |
Neg logit contributions | |
alks | -0.997 |
alt | -0.940 |
iley | -0.887 |
MH | -0.866 |
azeera | -0.857 |
Token |
|
Token ID | 198 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
shoulder | +0.822 |
handc | +0.821 |
Bench | +0.804 |
� | +0.762 |
� | +0.760 |
Neg logit contributions | |
alks | -0.870 |
alt | -0.825 |
iley | -0.782 |
MH | -0.770 |
ic | -0.759 |
Token |
|
Token ID | 198 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
shoulder | +0.176 |
handc | +0.176 |
Bench | +0.172 |
� | +0.163 |
� | +0.163 |
Neg logit contributions | |
alks | -0.175 |
alt | -0.166 |
iley | -0.157 |
MH | -0.155 |
ic | -0.152 |
Token | $ |
Token ID | 3 |
Feature activation | -7.28e-02 |
Pos logit contributions | |
alks | +0.157 |
alt | +0.148 |
iley | +0.139 |
MH | +0.136 |
ic | +0.134 |
Neg logit contributions | |
shoulder | -0.190 |
handc | -0.190 |
Bench | -0.186 |
� | -0.177 |
� | -0.177 |
Token | 20 |
Token ID | 1238 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
handc | +0.175 |
shoulder | +0.175 |
Bench | +0.171 |
� | +0.161 |
� | +0.161 |
Neg logit contributions | |
alks | -0.204 |
alt | -0.195 |
iley | -0.185 |
MH | -0.182 |
ic | -0.180 |
Token | – |
Token ID | 784 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
shoulder | +0.659 |
handc | +0.654 |
Bench | +0.641 |
� | +0.597 |
� | +0.596 |
Neg logit contributions | |
alks | -1.007 |
alt | -0.960 |
iley | -0.921 |
MH | -0.905 |
azeera | -0.897 |
Token | yellow |
Token ID | 7872 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
shoulder | +0.771 |
handc | +0.765 |
Bench | +0.752 |
� | +0.713 |
� | +0.710 |
Neg logit contributions | |
alks | -0.692 |
alt | -0.650 |
iley | -0.617 |
MH | -0.601 |
azeera | -0.595 |
Token | ( |
Token ID | 357 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
alks | +0.345 |
alt | +0.327 |
iley | +0.309 |
MH | +0.303 |
ic | +0.299 |
Neg logit contributions | |
shoulder | -0.359 |
handc | -0.359 |
Bench | -0.352 |
� | -0.335 |
� | -0.334 |
Token | . |
Token ID | 13 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
alks | +1.182 |
alt | +1.131 |
iley | +1.064 |
MH | +1.048 |
ic | +1.033 |
Neg logit contributions | |
handc | -1.356 |
shoulder | -1.343 |
Bench | -1.306 |
� | -1.251 |
� | -1.249 |
Token | x |
Token ID | 87 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
handc | +0.047 |
shoulder | +0.047 |
Bench | +0.046 |
� | +0.043 |
� | +0.043 |
Neg logit contributions | |
alks | -0.061 |
alt | -0.059 |
iley | -0.056 |
MH | -0.055 |
ic | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
alks | +1.209 |
alt | +1.161 |
iley | +1.095 |
MH | +1.068 |
ic | +1.058 |
Neg logit contributions | |
handc | -1.379 |
shoulder | -1.366 |
Bench | -1.330 |
� | -1.275 |
� | -1.273 |
Token | 85 |
Token ID | 5332 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
alks | +0.117 |
alt | +0.114 |
iley | +0.109 |
ic | +0.106 |
MH | +0.105 |
Neg logit contributions | |
handc | -0.067 |
shoulder | -0.067 |
Bench | -0.065 |
� | -0.061 |
� | -0.060 |
Token | 114 |
Token ID | 17342 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
shoulder | +0.432 |
handc | +0.427 |
Bench | +0.422 |
� | +0.392 |
� | +0.390 |
Neg logit contributions | |
alks | -0.673 |
alt | -0.642 |
iley | -0.615 |
MH | -0.603 |
azeera | -0.601 |
Token | 16 |
Token ID | 1433 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
alks | +0.423 |
alt | +0.402 |
iley | +0.382 |
MH | +0.377 |
ic | +0.372 |
Neg logit contributions | |
handc | -0.406 |
shoulder | -0.403 |
Bench | -0.392 |
� | -0.370 |
stitching | -0.369 |
Token | 21 |
Token ID | 2481 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
shoulder | +0.787 |
handc | +0.782 |
Bench | +0.772 |
� | +0.729 |
� | +0.728 |
Neg logit contributions | |
alks | -0.878 |
alt | -0.833 |
iley | -0.793 |
MH | -0.776 |
azeera | -0.772 |
Token | x |
Token ID | 2124 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
handc | +0.109 |
shoulder | +0.109 |
Bench | +0.106 |
� | +0.101 |
� | +0.101 |
Neg logit contributions | |
alks | -0.118 |
alt | -0.112 |
iley | -0.107 |
MH | -0.105 |
ic | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -7.29e-02 |
Pos logit contributions | |
alks | +1.330 |
alt | +1.275 |
iley | +1.197 |
MH | +1.181 |
ic | +1.165 |
Neg logit contributions | |
handc | -1.557 |
shoulder | -1.537 |
Bench | -1.494 |
� | -1.435 |
� | -1.432 |
Token | x |
Token ID | 87 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
handc | +0.164 |
shoulder | +0.164 |
Bench | +0.159 |
� | +0.150 |
� | +0.149 |
Neg logit contributions | |
alks | -0.217 |
alt | -0.208 |
iley | -0.198 |
MH | -0.194 |
ic | -0.192 |
Token | . |
Token ID | 13 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
alks | +1.254 |
alt | +1.201 |
iley | +1.129 |
MH | +1.111 |
ic | +1.095 |
Neg logit contributions | |
handc | -1.454 |
shoulder | -1.438 |
Bench | -1.396 |
� | -1.337 |
stitching | -1.337 |
Token | the |
Token ID | 262 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
shoulder | +1.485 |
handc | +1.463 |
Bench | +1.458 |
� | +1.354 |
� | +1.352 |
Neg logit contributions | |
alks | -1.835 |
alt | -1.742 |
iley | -1.655 |
azeera | -1.621 |
MH | -1.610 |
Token | 30 |
Token ID | 1542 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
shoulder | +1.285 |
handc | +1.272 |
Bench | +1.252 |
� | +1.165 |
stitching | +1.161 |
Neg logit contributions | |
alks | -1.662 |
alt | -1.581 |
iley | -1.509 |
MH | -1.476 |
azeera | -1.468 |
Token | lenses |
Token ID | 18405 |
Feature activation | -4.93e-01 |
Pos logit contributions | |
shoulder | +0.692 |
handc | +0.687 |
Bench | +0.675 |
� | +0.635 |
� | +0.632 |
Neg logit contributions | |
alks | -0.808 |
alt | -0.768 |
iley | -0.731 |
MH | -0.715 |
azeera | -0.711 |
Token | were |
Token ID | 547 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
shoulder | +1.159 |
handc | +1.137 |
Bench | +1.128 |
stitching | +1.045 |
� | +1.035 |
Neg logit contributions | |
alks | -1.429 |
alt | -1.359 |
iley | -1.292 |
azeera | -1.271 |
MH | -1.262 |
Token | less |
Token ID | 1342 |
Feature activation | -9.47e-02 |
Pos logit contributions | |
shoulder | +0.255 |
handc | +0.253 |
Bench | +0.249 |
� | +0.232 |
� | +0.231 |
Neg logit contributions | |
alks | -0.321 |
alt | -0.304 |
iley | -0.291 |
MH | -0.284 |
azeera | -0.283 |
Token | than |
Token ID | 621 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
shoulder | +0.281 |
handc | +0.278 |
Bench | +0.274 |
� | +0.261 |
� | +0.261 |
Neg logit contributions | |
alks | -0.216 |
alt | -0.201 |
iley | -0.189 |
MH | -0.184 |
azeera | -0.182 |
Token | 9 |
Token ID | 860 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
shoulder | +0.386 |
handc | +0.383 |
Bench | +0.376 |
� | +0.351 |
� | +0.350 |
Neg logit contributions | |
alks | -0.505 |
alt | -0.480 |
iley | -0.458 |
MH | -0.448 |
azeera | -0.446 |
Token | or |
Token ID | 393 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
shoulder | +0.226 |
handc | +0.225 |
Bench | +0.221 |
� | +0.210 |
� | +0.210 |
Neg logit contributions | |
alks | -0.217 |
alt | -0.206 |
iley | -0.195 |
MH | -0.191 |
ic | -0.188 |
Token | more |
Token ID | 517 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
alks | +0.729 |
alt | +0.698 |
iley | +0.665 |
MH | +0.659 |
ic | +0.651 |
Neg logit contributions | |
handc | -0.539 |
shoulder | -0.538 |
Bench | -0.527 |
� | -0.500 |
� | -0.499 |
Token | than |
Token ID | 621 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
shoulder | +1.677 |
handc | +1.655 |
Bench | +1.630 |
stitching | +1.535 |
� | +1.532 |
Neg logit contributions | |
alks | -1.675 |
alt | -1.576 |
iley | -1.492 |
azeera | -1.450 |
iens | -1.441 |
Token | 20 |
Token ID | 1160 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
alks | +0.985 |
alt | +0.940 |
iley | +0.897 |
MH | +0.888 |
ic | +0.876 |
Neg logit contributions | |
handc | -0.779 |
shoulder | -0.776 |
Bench | -0.762 |
� | -0.723 |
� | -0.723 |
Token | who |
Token ID | 508 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
shoulder | +0.776 |
handc | +0.773 |
Bench | +0.759 |
� | +0.712 |
� | +0.709 |
Neg logit contributions | |
alks | -0.993 |
alt | -0.944 |
iley | -0.901 |
MH | -0.883 |
azeera | -0.877 |
Token | the |
Token ID | 262 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
alks | +0.130 |
alt | +0.123 |
iley | +0.117 |
MH | +0.115 |
ic | +0.114 |
Neg logit contributions | |
shoulder | -0.115 |
handc | -0.115 |
Bench | -0.113 |
� | -0.106 |
� | -0.106 |
Token | Koch |
Token ID | 17009 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
alks | +0.345 |
alt | +0.326 |
iley | +0.307 |
MH | +0.302 |
ic | +0.298 |
Neg logit contributions | |
shoulder | -0.394 |
handc | -0.394 |
Bench | -0.387 |
� | -0.368 |
� | -0.368 |
Token | brothers |
Token ID | 9397 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
shoulder | +0.410 |
handc | +0.406 |
Bench | +0.399 |
� | +0.375 |
� | +0.373 |
Neg logit contributions | |
alks | -0.544 |
alt | -0.518 |
iley | -0.495 |
MH | -0.488 |
azeera | -0.481 |
Token | are |
Token ID | 389 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
shoulder | +0.542 |
handc | +0.537 |
Bench | +0.525 |
� | +0.498 |
� | +0.494 |
Neg logit contributions | |
alks | -0.594 |
alt | -0.564 |
iley | -0.536 |
MH | -0.529 |
azeera | -0.522 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
alks | +1.496 |
alt | +1.419 |
iley | +1.343 |
MH | +1.312 |
ic | +1.303 |
Neg logit contributions | |
handc | -1.634 |
shoulder | -1.631 |
Bench | -1.608 |
� | -1.538 |
� | -1.530 |
Token | With |
Token ID | 2080 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
shoulder | +0.468 |
Bench | +0.466 |
handc | +0.462 |
� | +0.436 |
� | +0.435 |
Neg logit contributions | |
alks | -0.462 |
alt | -0.435 |
iley | -0.415 |
azeera | -0.405 |
MH | -0.400 |
Token | 25 |
Token ID | 1679 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
alks | +1.229 |
alt | +1.167 |
iley | +1.109 |
MH | +1.091 |
ic | +1.077 |
Neg logit contributions | |
shoulder | -1.116 |
handc | -1.114 |
Bench | -1.092 |
� | -1.034 |
� | -1.032 |
Token | percent |
Token ID | 1411 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
shoulder | +0.712 |
handc | +0.708 |
Bench | +0.699 |
� | +0.665 |
� | +0.663 |
Neg logit contributions | |
alks | -0.571 |
alt | -0.536 |
iley | -0.504 |
MH | -0.490 |
azeera | -0.485 |
Token | of |
Token ID | 286 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
alks | +0.171 |
alt | +0.163 |
iley | +0.155 |
MH | +0.153 |
ic | +0.151 |
Neg logit contributions | |
shoulder | -0.140 |
handc | -0.140 |
Bench | -0.137 |
� | -0.129 |
� | -0.129 |
Token | GW |
Token ID | 27164 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
shoulder | +1.430 |
handc | +1.417 |
Bench | +1.403 |
� | +1.297 |
stitching | +1.297 |
Neg logit contributions | |
alks | -1.510 |
alt | -1.430 |
iley | -1.354 |
azeera | -1.314 |
MH | -1.303 |
Token | works |
Token ID | 2499 |
Feature activation | +9.87e-02 |
Pos logit contributions | |
alks | +0.017 |
alt | +0.016 |
iley | +0.016 |
MH | +0.015 |
ic | +0.015 |
Neg logit contributions | |
shoulder | -0.014 |
handc | -0.013 |
Bench | -0.013 |
� | -0.012 |
� | -0.012 |
Token | somehow |
Token ID | 7599 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
alks | +0.265 |
alt | +0.252 |
iley | +0.240 |
MH | +0.235 |
ic | +0.233 |
Neg logit contributions | |
handc | -0.247 |
shoulder | -0.247 |
Bench | -0.243 |
� | -0.230 |
� | -0.229 |
Token | . |
Token ID | 13 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
shoulder | +0.087 |
handc | +0.087 |
Bench | +0.085 |
� | +0.081 |
� | +0.080 |
Neg logit contributions | |
alks | -0.094 |
alt | -0.089 |
iley | -0.085 |
MH | -0.083 |
ic | -0.082 |
Token | � |
Token ID | 447 |
Feature activation | +7.44e-02 |
Pos logit contributions | |
alks | +0.279 |
alt | +0.265 |
iley | +0.251 |
MH | +0.248 |
ic | +0.245 |
Neg logit contributions | |
handc | -0.265 |
shoulder | -0.265 |
Bench | -0.259 |
� | -0.245 |
� | -0.244 |
Token | � |
Token ID | 251 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
alks | +0.207 |
alt | +0.197 |
iley | +0.188 |
MH | +0.184 |
ic | +0.182 |
Neg logit contributions | |
shoulder | -0.180 |
handc | -0.180 |
Bench | -0.176 |
� | -0.166 |
� | -0.166 |
Token | Never |
Token ID | 7236 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
shoulder | +0.106 |
handc | +0.106 |
Bench | +0.104 |
� | +0.099 |
� | +0.099 |
Neg logit contributions | |
alks | -0.108 |
alt | -0.103 |
iley | -0.097 |
MH | -0.095 |
azeera | -0.095 |
Token | Mind |
Token ID | 10175 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
shoulder | +0.466 |
handc | +0.464 |
Bench | +0.456 |
� | +0.431 |
� | +0.430 |
Neg logit contributions | |
alks | -0.500 |
alt | -0.474 |
iley | -0.450 |
MH | -0.442 |
ic | -0.436 |
Token | the |
Token ID | 262 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
alks | +1.416 |
alt | +1.341 |
iley | +1.274 |
MH | +1.254 |
ic | +1.236 |
Neg logit contributions | |
handc | -1.392 |
shoulder | -1.379 |
Bench | -1.345 |
� | -1.285 |
� | -1.281 |
Token | Boll |
Token ID | 43479 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
alks | +0.208 |
alt | +0.197 |
iley | +0.186 |
MH | +0.182 |
ic | +0.180 |
Neg logit contributions | |
shoulder | -0.222 |
handc | -0.221 |
Bench | -0.218 |
� | -0.207 |
� | -0.206 |
Token | ocks |
Token ID | 3320 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
alks | +0.028 |
alt | +0.026 |
iley | +0.024 |
MH | +0.024 |
ic | +0.023 |
Neg logit contributions | |
shoulder | -0.040 |
handc | -0.040 |
Bench | -0.039 |
� | -0.038 |
� | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
shoulder | +0.449 |
handc | +0.447 |
Bench | +0.440 |
� | +0.416 |
� | +0.416 |
Neg logit contributions | |
alks | -0.447 |
alt | -0.423 |
iley | -0.401 |
MH | -0.394 |
azeera | -0.388 |
Token | fall |
Token ID | 2121 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
alks | +0.702 |
alt | +0.665 |
iley | +0.632 |
MH | +0.620 |
ic | +0.612 |
Neg logit contributions | |
shoulder | -0.674 |
handc | -0.672 |
Bench | -0.660 |
� | -0.624 |
� | -0.623 |
Token | in |
Token ID | 287 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
alks | +0.850 |
alt | +0.805 |
iley | +0.762 |
MH | +0.748 |
ic | +0.740 |
Neg logit contributions | |
shoulder | -0.889 |
handc | -0.889 |
Bench | -0.874 |
� | -0.832 |
� | -0.829 |
Token | the |
Token ID | 262 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
alks | +0.778 |
alt | +0.742 |
iley | +0.707 |
MH | +0.697 |
ic | +0.690 |
Neg logit contributions | |
handc | -0.610 |
shoulder | -0.610 |
Bench | -0.598 |
� | -0.565 |
� | -0.563 |
Token | middle |
Token ID | 3504 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
alks | +0.548 |
alt | +0.522 |
iley | +0.498 |
MH | +0.491 |
ic | +0.486 |
Neg logit contributions | |
shoulder | -0.420 |
handc | -0.420 |
Bench | -0.411 |
� | -0.387 |
� | -0.386 |
Token | of |
Token ID | 286 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
alks | +0.588 |
alt | +0.561 |
iley | +0.532 |
MH | +0.525 |
ic | +0.520 |
Neg logit contributions | |
handc | -0.496 |
shoulder | -0.494 |
Bench | -0.486 |
� | -0.459 |
� | -0.458 |
Token | the |
Token ID | 262 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
shoulder | +0.497 |
handc | +0.494 |
Bench | +0.485 |
� | +0.454 |
� | +0.453 |
Neg logit contributions | |
alks | -0.623 |
alt | -0.594 |
iley | -0.567 |
MH | -0.555 |
azeera | -0.550 |
Token | day |
Token ID | 1110 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
shoulder | +0.503 |
handc | +0.498 |
Bench | +0.491 |
� | +0.460 |
� | +0.458 |
Neg logit contributions | |
alks | -0.553 |
alt | -0.524 |
iley | -0.499 |
MH | -0.487 |
azeera | -0.482 |
Token | in |
Token ID | 287 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
alks | +1.582 |
alt | +1.506 |
iley | +1.426 |
MH | +1.400 |
ic | +1.388 |
Neg logit contributions | |
handc | -1.556 |
shoulder | -1.554 |
Bench | -1.527 |
� | -1.455 |
� | -1.445 |
Token | a |
Token ID | 257 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
alks | +0.915 |
alt | +0.873 |
iley | +0.831 |
MH | +0.821 |
ic | +0.812 |
Neg logit contributions | |
handc | -0.738 |
shoulder | -0.738 |
Bench | -0.723 |
� | -0.684 |
� | -0.682 |
Token | city |
Token ID | 1748 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
alks | +0.549 |
alt | +0.524 |
iley | +0.498 |
MH | +0.490 |
ic | +0.486 |
Neg logit contributions | |
shoulder | -0.482 |
handc | -0.482 |
Bench | -0.473 |
� | -0.449 |
� | -0.448 |
Token | with |
Token ID | 351 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
alks | +0.331 |
alt | +0.314 |
iley | +0.299 |
MH | +0.294 |
ic | +0.290 |
Neg logit contributions | |
shoulder | -0.299 |
handc | -0.298 |
Bench | -0.293 |
� | -0.277 |
� | -0.276 |
Token | Mont |
Token ID | 5575 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
alks | +0.651 |
alt | +0.616 |
iley | +0.584 |
MH | +0.578 |
ic | +0.567 |
Neg logit contributions | |
handc | -0.651 |
shoulder | -0.649 |
Bench | -0.631 |
� | -0.605 |
� | -0.603 |
Token | rose |
Token ID | 13698 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
alks | +0.930 |
alt | +0.886 |
iley | +0.843 |
MH | +0.825 |
ic | +0.816 |
Neg logit contributions | |
handc | -0.779 |
shoulder | -0.779 |
Bench | -0.760 |
� | -0.722 |
� | -0.720 |
Token | Today |
Token ID | 6288 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
shoulder | +0.019 |
handc | +0.019 |
Bench | +0.019 |
� | +0.017 |
� | +0.017 |
Neg logit contributions | |
alks | -0.023 |
alt | -0.022 |
iley | -0.021 |
MH | -0.021 |
ic | -0.021 |
Token | : |
Token ID | 25 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
shoulder | +0.770 |
handc | +0.766 |
Bench | +0.758 |
� | +0.716 |
� | +0.712 |
Neg logit contributions | |
alks | -0.711 |
alt | -0.671 |
iley | -0.637 |
MH | -0.622 |
azeera | -0.621 |
Token | Brass |
Token ID | 32309 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
shoulder | +0.533 |
handc | +0.531 |
Bench | +0.524 |
� | +0.495 |
� | +0.495 |
Neg logit contributions | |
alks | -0.526 |
alt | -0.498 |
iley | -0.472 |
MH | -0.462 |
azeera | -0.457 |
Token | erie |
Token ID | 18287 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
shoulder | +0.447 |
handc | +0.445 |
Bench | +0.439 |
� | +0.406 |
� | +0.405 |
Neg logit contributions | |
alks | -0.669 |
alt | -0.637 |
iley | -0.612 |
MH | -0.601 |
azeera | -0.597 |
Token | Max |
Token ID | 5436 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
Bench | +2.885 |
shoulder | +2.872 |
handc | +2.839 |
� | +2.657 |
� | +2.636 |
Neg logit contributions | |
alks | -2.720 |
alt | -2.544 |
iley | -2.428 |
azeera | -2.396 |
MH | -2.333 |
Token | and |
Token ID | 290 |
Feature activation | -7.27e-02 |
Pos logit contributions | |
alks | +0.194 |
alt | +0.182 |
iley | +0.173 |
MH | +0.169 |
azeera | +0.167 |
Neg logit contributions | |
shoulder | -0.221 |
handc | -0.220 |
Bench | -0.217 |
� | -0.206 |
� | -0.205 |
Token | Julie |
Token ID | 21946 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
shoulder | +0.188 |
handc | +0.188 |
Bench | +0.185 |
� | +0.175 |
� | +0.175 |
Neg logit contributions | |
alks | -0.190 |
alt | -0.180 |
iley | -0.171 |
MH | -0.167 |
ic | -0.165 |
Token | Previously |
Token ID | 29875 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
alks | +0.590 |
alt | +0.562 |
iley | +0.533 |
MH | +0.524 |
ic | +0.517 |
Neg logit contributions | |
shoulder | -0.545 |
handc | -0.544 |
Bench | -0.529 |
� | -0.506 |
� | -0.504 |
Token | : |
Token ID | 25 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
alks | +1.390 |
alt | +1.314 |
iley | +1.227 |
MH | +1.216 |
ic | +1.191 |
Neg logit contributions | |
handc | -1.785 |
shoulder | -1.776 |
Bench | -1.721 |
� | -1.662 |
� | -1.657 |
Token | a |
Token ID | 257 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
shoulder | +0.096 |
handc | +0.095 |
Bench | +0.094 |
� | +0.088 |
� | +0.088 |
Neg logit contributions | |
alks | -0.123 |
alt | -0.117 |
iley | -0.112 |
MH | -0.110 |
ic | -0.109 |
Token | rise |
Token ID | 4485 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
shoulder | +0.323 |
handc | +0.322 |
Bench | +0.315 |
� | +0.297 |
� | +0.297 |
Neg logit contributions | |
alks | -0.412 |
alt | -0.393 |
iley | -0.375 |
MH | -0.368 |
ic | -0.364 |
Token | until |
Token ID | 1566 |
Feature activation | -5.73e-01 |
Pos logit contributions | |
alks | +0.323 |
alt | +0.307 |
iley | +0.293 |
MH | +0.288 |
ic | +0.284 |
Neg logit contributions | |
shoulder | -0.278 |
handc | -0.277 |
Bench | -0.272 |
� | -0.257 |
� | -0.256 |
Token | around |
Token ID | 1088 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
shoulder | +1.388 |
handc | +1.381 |
Bench | +1.359 |
� | +1.262 |
stitching | +1.261 |
Neg logit contributions | |
alks | -1.601 |
alt | -1.526 |
iley | -1.452 |
azeera | -1.420 |
MH | -1.411 |
Token | mid |
Token ID | 3095 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
shoulder | +1.297 |
handc | +1.288 |
Bench | +1.263 |
� | +1.177 |
stitching | +1.174 |
Neg logit contributions | |
alks | -1.545 |
alt | -1.471 |
iley | -1.397 |
azeera | -1.366 |
MH | -1.360 |
Token | - |
Token ID | 12 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
shoulder | +0.720 |
handc | +0.718 |
Bench | +0.705 |
� | +0.668 |
� | +0.666 |
Neg logit contributions | |
alks | -0.715 |
alt | -0.674 |
iley | -0.640 |
MH | -0.626 |
azeera | -0.620 |
Token | 2015 |
Token ID | 4626 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
handc | +2.635 |
shoulder | +2.635 |
Bench | +2.586 |
� | +2.466 |
� | +2.459 |
Neg logit contributions | |
alks | -3.095 |
alt | -2.866 |
iley | -2.777 |
MH | -2.736 |
azeera | -2.733 |
Token | at |
Token ID | 379 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
shoulder | +0.758 |
handc | +0.755 |
Bench | +0.743 |
� | +0.702 |
� | +0.700 |
Neg logit contributions | |
alks | -0.752 |
alt | -0.709 |
iley | -0.673 |
MH | -0.660 |
ic | -0.652 |
Token | the |
Token ID | 262 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
alks | +0.431 |
alt | +0.410 |
iley | +0.390 |
MH | +0.384 |
ic | +0.379 |
Neg logit contributions | |
shoulder | -0.368 |
handc | -0.367 |
Bench | -0.360 |
� | -0.340 |
� | -0.339 |
Token | earliest |
Token ID | 14555 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
shoulder | +0.040 |
handc | +0.040 |
Bench | +0.040 |
� | +0.037 |
� | +0.037 |
Neg logit contributions | |
alks | -0.045 |
alt | -0.043 |
iley | -0.041 |
MH | -0.040 |
ic | -0.040 |
Token | - |
Token ID | 532 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
shoulder | +0.100 |
handc | +0.099 |
Bench | +0.098 |
� | +0.092 |
� | +0.092 |
Neg logit contributions | |
alks | -0.099 |
alt | -0.094 |
iley | -0.089 |
MH | -0.087 |
azeera | -0.086 |
Token | June |
Token ID | 2795 |
Feature activation | -2.75e-03 |
Pos logit contributions | |
Bench | +3.153 |
shoulder | +3.142 |
handc | +2.994 |
� | +2.810 |
� | +2.798 |
Neg logit contributions | |
alks | -4.867 |
alt | -4.622 |
azeera | -4.538 |
iley | -4.469 |
iens | -4.326 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
shoulder | +0.007 |
handc | +0.007 |
Bench | +0.007 |
� | +0.007 |
� | +0.007 |
Neg logit contributions | |
alks | -0.007 |
alt | -0.007 |
iley | -0.006 |
MH | -0.006 |
azeera | -0.006 |
Token | Canada |
Token ID | 3340 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
shoulder | +0.558 |
handc | +0.553 |
Bench | +0.549 |
� | +0.516 |
� | +0.515 |
Neg logit contributions | |
alks | -0.619 |
alt | -0.586 |
iley | -0.558 |
azeera | -0.545 |
MH | -0.544 |
Token | 16 |
Token ID | 1467 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
shoulder | +0.325 |
handc | +0.324 |
Bench | +0.318 |
� | +0.299 |
� | +0.298 |
Neg logit contributions | |
alks | -0.400 |
alt | -0.380 |
iley | -0.363 |
MH | -0.356 |
ic | -0.352 |
Token | 10 |
Token ID | 940 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
shoulder | +1.379 |
handc | +1.369 |
Bench | +1.356 |
� | +1.282 |
� | +1.278 |
Neg logit contributions | |
alks | -1.553 |
alt | -1.476 |
iley | -1.404 |
MH | -1.372 |
azeera | -1.368 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
shoulder | +1.437 |
handc | +1.433 |
Bench | +1.413 |
� | +1.347 |
� | +1.345 |
Neg logit contributions | |
alks | -1.122 |
alt | -1.054 |
iley | -0.991 |
MH | -0.968 |
ic | -0.954 |
Token | # |
Token ID | 1303 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
shoulder | +3.261 |
Bench | +3.235 |
handc | +3.160 |
� | +2.913 |
� | +2.911 |
Neg logit contributions | |
alks | -4.539 |
alt | -4.278 |
iley | -4.135 |
azeera | -4.085 |
iens | -4.039 |
Token | 14 |
Token ID | 1415 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
shoulder | +1.441 |
handc | +1.437 |
Bench | +1.412 |
� | +1.341 |
� | +1.338 |
Neg logit contributions | |
alks | -1.510 |
alt | -1.416 |
iley | -1.352 |
MH | -1.328 |
azeera | -1.316 |
Token | On |
Token ID | 1550 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
shoulder | +0.183 |
handc | +0.182 |
Bench | +0.180 |
� | +0.171 |
� | +0.170 |
Neg logit contributions | |
alks | -0.135 |
alt | -0.126 |
iley | -0.118 |
azeera | -0.114 |
MH | -0.114 |
Token | September |
Token ID | 2693 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
shoulder | +0.920 |
handc | +0.915 |
Bench | +0.899 |
� | +0.838 |
� | +0.834 |
Neg logit contributions | |
alks | -1.287 |
alt | -1.226 |
iley | -1.169 |
MH | -1.149 |
azeera | -1.141 |
Token | 20 |
Token ID | 1160 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
alks | +0.679 |
alt | +0.651 |
iley | +0.625 |
MH | +0.616 |
ic | +0.610 |
Neg logit contributions | |
shoulder | -0.362 |
handc | -0.361 |
Bench | -0.352 |
� | -0.325 |
� | -0.324 |
Token | J |
Token ID | 41 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
alks | +0.161 |
alt | +0.152 |
iley | +0.144 |
MH | +0.141 |
ic | +0.139 |
Neg logit contributions | |
shoulder | -0.167 |
handc | -0.167 |
Bench | -0.164 |
� | -0.156 |
� | -0.156 |
Token | aaaa |
Token ID | 24794 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
alks | +0.783 |
alt | +0.739 |
iley | +0.696 |
MH | +0.680 |
ic | +0.671 |
Neg logit contributions | |
shoulder | -0.921 |
handc | -0.919 |
Bench | -0.904 |
� | -0.860 |
� | -0.858 |
Token | asper |
Token ID | 32981 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
alks | +1.789 |
alt | +1.689 |
iley | +1.583 |
MH | +1.520 |
ic | +1.498 |
Neg logit contributions | |
handc | -2.446 |
shoulder | -2.435 |
Bench | -2.385 |
� | -2.276 |
� | -2.266 |
Token | Profile |
Token ID | 13118 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
alks | +0.201 |
alt | +0.189 |
iley | +0.178 |
MH | +0.175 |
ic | +0.172 |
Neg logit contributions | |
shoulder | -0.245 |
handc | -0.244 |
Bench | -0.241 |
� | -0.229 |
� | -0.229 |
Token | Blog |
Token ID | 14001 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
shoulder | +2.979 |
Bench | +2.977 |
handc | +2.917 |
� | +2.770 |
� | +2.748 |
Neg logit contributions | |
alks | -3.817 |
alt | -3.627 |
iley | -3.470 |
azeera | -3.404 |
MH | -3.368 |
Token | Joined |
Token ID | 16709 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
shoulder | +1.891 |
Bench | +1.870 |
handc | +1.860 |
� | +1.760 |
� | +1.746 |
Neg logit contributions | |
alks | -2.098 |
alt | -1.982 |
iley | -1.891 |
azeera | -1.853 |
MH | -1.845 |
Token | April |
Token ID | 3035 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
Bench | +3.012 |
shoulder | +3.001 |
handc | +2.867 |
� | +2.700 |
� | +2.689 |
Neg logit contributions | |
alks | -4.596 |
alt | -4.365 |
azeera | -4.275 |
iley | -4.222 |
iens | -4.082 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
shoulder | +0.236 |
handc | +0.234 |
Bench | +0.232 |
� | +0.219 |
� | +0.218 |
Neg logit contributions | |
alks | -0.246 |
alt | -0.232 |
iley | -0.221 |
MH | -0.216 |
azeera | -0.214 |
Token | United |
Token ID | 1578 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
shoulder | +0.324 |
handc | +0.322 |
Bench | +0.318 |
� | +0.299 |
� | +0.299 |
Neg logit contributions | |
alks | -0.360 |
alt | -0.341 |
iley | -0.325 |
MH | -0.318 |
azeera | -0.316 |
Token | States |
Token ID | 1829 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
shoulder | +0.590 |
handc | +0.584 |
Bench | +0.580 |
� | +0.546 |
� | +0.544 |
Neg logit contributions | |
alks | -0.686 |
alt | -0.646 |
iley | -0.620 |
MH | -0.604 |
azeera | -0.603 |
Token | 5 |
Token ID | 642 |
Feature activation | -9.47e-02 |
Pos logit contributions | |
shoulder | +1.139 |
handc | +1.129 |
Bench | +1.119 |
� | +1.046 |
� | +1.044 |
Neg logit contributions | |
alks | -1.403 |
alt | -1.333 |
iley | -1.270 |
azeera | -1.243 |
MH | -1.242 |
Token | difference |
Token ID | 3580 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
shoulder | +0.476 |
handc | +0.473 |
Bench | +0.463 |
� | +0.433 |
� | +0.430 |
Neg logit contributions | |
alks | -0.637 |
alt | -0.606 |
iley | -0.578 |
MH | -0.567 |
azeera | -0.562 |
Token | Walker |
Token ID | 10120 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
shoulder | +0.284 |
handc | +0.283 |
Bench | +0.278 |
� | +0.263 |
� | +0.263 |
Neg logit contributions | |
alks | -0.297 |
alt | -0.282 |
iley | -0.267 |
MH | -0.262 |
ic | -0.259 |
Token | saw |
Token ID | 2497 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
alks | +0.210 |
alt | +0.199 |
iley | +0.189 |
MH | +0.186 |
ic | +0.184 |
Neg logit contributions | |
shoulder | -0.194 |
handc | -0.194 |
Bench | -0.190 |
� | -0.180 |
� | -0.179 |
Token | between |
Token ID | 1022 |
Feature activation | +6.09e-03 |
Pos logit contributions | |
shoulder | +0.134 |
handc | +0.134 |
Bench | +0.132 |
� | +0.125 |
� | +0.124 |
Neg logit contributions | |
alks | -0.148 |
alt | -0.140 |
iley | -0.133 |
MH | -0.131 |
ic | -0.129 |
Token | the |
Token ID | 262 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
alks | +0.017 |
alt | +0.016 |
iley | +0.015 |
MH | +0.015 |
ic | +0.015 |
Neg logit contributions | |
shoulder | -0.015 |
handc | -0.015 |
Bench | -0.014 |
� | -0.014 |
� | -0.014 |
Token | pair |
Token ID | 5166 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
shoulder | +0.915 |
handc | +0.905 |
Bench | +0.890 |
� | +0.822 |
stitching | +0.817 |
Neg logit contributions | |
alks | -1.194 |
alt | -1.138 |
iley | -1.085 |
azeera | -1.063 |
MH | -1.062 |
Token | 's |
Token ID | 338 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
shoulder | +2.582 |
handc | +2.526 |
Bench | +2.475 |
stitching | +2.317 |
� | +2.280 |
Neg logit contributions | |
alks | -3.177 |
alt | -3.027 |
iley | -2.893 |
azeera | -2.858 |
MH | -2.821 |
Token | defect |
Token ID | 11855 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
shoulder | +0.724 |
handc | +0.715 |
Bench | +0.703 |
� | +0.653 |
stitching | +0.651 |
Neg logit contributions | |
alks | -0.893 |
alt | -0.850 |
iley | -0.811 |
MH | -0.793 |
azeera | -0.790 |
Token | ions |
Token ID | 507 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
shoulder | +0.146 |
handc | +0.145 |
Bench | +0.143 |
� | +0.138 |
� | +0.137 |
Neg logit contributions | |
alks | -0.068 |
alt | -0.062 |
iley | -0.057 |
MH | -0.055 |
ic | -0.054 |
Token | was |
Token ID | 373 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
alks | +0.824 |
alt | +0.784 |
iley | +0.746 |
MH | +0.734 |
ic | +0.725 |
Neg logit contributions | |
shoulder | -0.706 |
handc | -0.705 |
Bench | -0.692 |
� | -0.655 |
� | -0.653 |
Token | that |
Token ID | 326 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
shoulder | +0.464 |
handc | +0.462 |
Bench | +0.453 |
� | +0.427 |
� | +0.425 |
Neg logit contributions | |
alks | -0.509 |
alt | -0.484 |
iley | -0.459 |
MH | -0.450 |
azeera | -0.446 |