Token | � |
Token ID | 447 |
Feature activation | +9.83e-02 |
Pos logit contributions | |
angles | +0.657 |
Supervisor | +0.599 |
ortment | +0.595 |
mble | +0.593 |
aina | +0.590 |
Neg logit contributions | |
ield | -0.653 |
vine | -0.651 |
het | -0.639 |
794 | -0.636 |
PF | -0.633 |
Token | � |
Token ID | 251 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
angles | +0.225 |
ortment | +0.205 |
Supervisor | +0.205 |
mble | +0.204 |
aina | +0.203 |
Neg logit contributions | |
ield | -0.196 |
vine | -0.195 |
het | -0.192 |
794 | -0.190 |
PF | -0.189 |
Token | might |
Token ID | 1244 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
angles | +0.073 |
Supervisor | +0.066 |
ortment | +0.066 |
mble | +0.066 |
aina | +0.066 |
Neg logit contributions | |
ield | -0.073 |
vine | -0.073 |
het | -0.072 |
794 | -0.071 |
PF | -0.071 |
Token | be |
Token ID | 307 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
angles | +0.720 |
ortment | +0.657 |
Supervisor | +0.657 |
mble | +0.653 |
aina | +0.650 |
Neg logit contributions | |
ield | -0.649 |
vine | -0.647 |
het | -0.635 |
794 | -0.631 |
PF | -0.628 |
Token | more |
Token ID | 517 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
angles | +0.765 |
Supervisor | +0.708 |
ortment | +0.701 |
aina | +0.700 |
mble | +0.698 |
Neg logit contributions | |
ield | -0.627 |
vine | -0.624 |
het | -0.610 |
794 | -0.607 |
PF | -0.604 |
Token | apt |
Token ID | 15409 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
angles | +1.152 |
Supervisor | +1.069 |
ortment | +1.060 |
aina | +1.057 |
mble | +1.051 |
Neg logit contributions | |
ield | -0.842 |
vine | -0.840 |
het | -0.817 |
794 | -0.816 |
PF | -0.810 |
Token | , |
Token ID | 11 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
angles | +4.698 |
Supervisor | +4.387 |
aina | +4.325 |
ortment | +4.283 |
mble | +4.260 |
Neg logit contributions | |
vine | -5.103 |
ield | -5.100 |
PF | -4.999 |
794 | -4.975 |
het | -4.915 |
Token | especially |
Token ID | 2592 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
angles | +0.730 |
Supervisor | +0.674 |
ortment | +0.671 |
mble | +0.666 |
aina | +0.666 |
Neg logit contributions | |
ield | -0.551 |
vine | -0.548 |
het | -0.535 |
794 | -0.531 |
PF | -0.529 |
Token | since |
Token ID | 1201 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
ield | +0.365 |
vine | +0.361 |
het | +0.355 |
794 | +0.352 |
PF | +0.347 |
Neg logit contributions | |
angles | -0.492 |
ortment | -0.453 |
mble | -0.450 |
Supervisor | -0.448 |
aina | -0.447 |
Token | Land |
Token ID | 6379 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
ield | +0.438 |
vine | +0.437 |
het | +0.429 |
794 | +0.425 |
PF | +0.421 |
Neg logit contributions | |
angles | -0.486 |
ortment | -0.443 |
Supervisor | -0.439 |
mble | -0.439 |
aina | -0.438 |
Token | rie |
Token ID | 5034 |
Feature activation | -6.31e-02 |
Pos logit contributions | |
ield | +0.714 |
vine | +0.710 |
het | +0.695 |
794 | +0.691 |
PF | +0.682 |
Neg logit contributions | |
angles | -0.650 |
ortment | -0.587 |
Supervisor | -0.586 |
mble | -0.583 |
aina | -0.576 |
Token | vibrant |
Token ID | 21266 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
ield | +1.137 |
vine | +1.133 |
het | +1.110 |
794 | +1.099 |
PF | +1.088 |
Neg logit contributions | |
angles | -1.612 |
Supervisor | -1.483 |
ortment | -1.483 |
mble | -1.477 |
aina | -1.471 |
Token | sexual |
Token ID | 3206 |
Feature activation | +8.46e-03 |
Pos logit contributions | |
angles | +1.187 |
ortment | +1.087 |
Supervisor | +1.084 |
mble | +1.082 |
aina | +1.076 |
Neg logit contributions | |
ield | -0.956 |
vine | -0.950 |
794 | -0.930 |
het | -0.930 |
PF | -0.929 |
Token | appet |
Token ID | 16422 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
angles | +0.020 |
ortment | +0.018 |
Supervisor | +0.018 |
aina | +0.018 |
mble | +0.018 |
Neg logit contributions | |
ield | -0.016 |
vine | -0.016 |
het | -0.016 |
794 | -0.016 |
PF | -0.016 |
Token | ites |
Token ID | 2737 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
angles | +0.401 |
ortment | +0.366 |
Supervisor | +0.365 |
aina | +0.364 |
mble | +0.361 |
Neg logit contributions | |
ield | -0.313 |
vine | -0.312 |
PF | -0.306 |
het | -0.305 |
794 | -0.303 |
Token | and |
Token ID | 290 |
Feature activation | +7.67e-02 |
Pos logit contributions | |
angles | +0.819 |
ortment | +0.750 |
Supervisor | +0.748 |
aina | +0.743 |
mble | +0.742 |
Neg logit contributions | |
ield | -0.733 |
vine | -0.731 |
het | -0.714 |
794 | -0.712 |
PF | -0.708 |
Token | apt |
Token ID | 15409 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
angles | +0.182 |
ortment | +0.167 |
Supervisor | +0.167 |
mble | +0.166 |
aina | +0.165 |
Neg logit contributions | |
ield | -0.146 |
vine | -0.145 |
het | -0.143 |
794 | -0.142 |
PF | -0.141 |
Token | itudes |
Token ID | 10455 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
angles | +4.730 |
ortment | +4.243 |
Supervisor | +4.122 |
aina | +4.112 |
mble | +4.090 |
Neg logit contributions | |
vine | -5.206 |
ield | -5.185 |
PF | -5.138 |
794 | -5.131 |
Finch | -5.035 |
Token | past |
Token ID | 1613 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
angles | +2.504 |
ortment | +2.299 |
Supervisor | +2.289 |
aina | +2.272 |
mble | +2.253 |
Neg logit contributions | |
ield | -2.411 |
vine | -2.407 |
794 | -2.347 |
het | -2.337 |
PF | -2.335 |
Token | a |
Token ID | 257 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
ield | +0.526 |
vine | +0.524 |
het | +0.514 |
794 | +0.510 |
PF | +0.506 |
Neg logit contributions | |
angles | -0.648 |
ortment | -0.593 |
Supervisor | -0.593 |
mble | -0.590 |
aina | -0.588 |
Token | certain |
Token ID | 1728 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
ield | +0.090 |
vine | +0.090 |
het | +0.088 |
794 | +0.088 |
PF | +0.087 |
Neg logit contributions | |
angles | -0.101 |
Supervisor | -0.092 |
ortment | -0.092 |
mble | -0.092 |
aina | -0.091 |
Token | age |
Token ID | 2479 |
Feature activation | +8.69e-02 |
Pos logit contributions | |
angles | +0.201 |
ortment | +0.186 |
Supervisor | +0.186 |
mble | +0.185 |
aina | +0.184 |
Neg logit contributions | |
ield | -0.127 |
vine | -0.126 |
het | -0.123 |
794 | -0.122 |
PF | -0.121 |
Token | not |
Token ID | 407 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
angles | +0.554 |
Supervisor | +0.499 |
ortment | +0.498 |
mble | +0.495 |
aina | +0.493 |
Neg logit contributions | |
ield | -0.631 |
vine | -0.628 |
het | -0.618 |
794 | -0.614 |
PF | -0.610 |
Token | only |
Token ID | 691 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
ield | +0.383 |
vine | +0.381 |
het | +0.373 |
794 | +0.370 |
PF | +0.367 |
Neg logit contributions | |
angles | -0.544 |
ortment | -0.501 |
Supervisor | -0.501 |
mble | -0.498 |
aina | -0.497 |
Token | includes |
Token ID | 3407 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
angles | +0.825 |
ortment | +0.741 |
Supervisor | +0.739 |
mble | +0.736 |
aina | +0.730 |
Neg logit contributions | |
ield | -1.038 |
vine | -1.034 |
het | -1.018 |
794 | -1.012 |
PF | -1.006 |
Token | the |
Token ID | 262 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
angles | +0.947 |
Supervisor | +0.867 |
ortment | +0.867 |
mble | +0.862 |
aina | +0.859 |
Neg logit contributions | |
ield | -0.770 |
vine | -0.766 |
het | -0.752 |
794 | -0.746 |
PF | -0.741 |
Token | iconic |
Token ID | 14133 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
angles | +1.283 |
Supervisor | +1.176 |
ortment | +1.174 |
mble | +1.167 |
aina | +1.164 |
Neg logit contributions | |
ield | -1.103 |
vine | -1.098 |
het | -1.078 |
794 | -1.068 |
PF | -1.062 |
Token | Falling |
Token ID | 42914 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
angles | +1.107 |
Supervisor | +1.006 |
ortment | +1.005 |
mble | +0.999 |
aina | +0.995 |
Neg logit contributions | |
ield | -1.070 |
vine | -1.065 |
het | -1.047 |
794 | -1.039 |
PF | -1.033 |
Token | water |
Token ID | 7050 |
Feature activation | +9.08e-01 |
Pos logit contributions | |
angles | +4.637 |
Supervisor | +4.196 |
ortment | +4.099 |
Buff | +4.093 |
aina | +4.086 |
Neg logit contributions | |
ield | -4.625 |
PF | -4.580 |
het | -4.569 |
794 | -4.561 |
vine | -4.521 |
Token | , |
Token ID | 11 |
Feature activation | +5.35e-01 |
Pos logit contributions | |
angles | +1.888 |
Supervisor | +1.715 |
ortment | +1.703 |
mble | +1.691 |
aina | +1.691 |
Neg logit contributions | |
ield | -1.992 |
vine | -1.982 |
het | -1.950 |
794 | -1.935 |
PF | -1.926 |
Token | aka |
Token ID | 22430 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
angles | +1.243 |
Supervisor | +1.138 |
ortment | +1.133 |
mble | +1.127 |
aina | +1.125 |
Neg logit contributions | |
ield | -1.047 |
vine | -1.042 |
het | -1.023 |
794 | -1.011 |
PF | -1.007 |
Token | The |
Token ID | 383 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
ield | +0.223 |
vine | +0.222 |
het | +0.218 |
794 | +0.217 |
PF | +0.215 |
Neg logit contributions | |
angles | -0.233 |
ortment | -0.213 |
Supervisor | -0.212 |
mble | -0.211 |
aina | -0.210 |
Token | Edgar |
Token ID | 29166 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
angles | +0.818 |
ortment | +0.742 |
Supervisor | +0.742 |
mble | +0.737 |
aina | +0.735 |
Neg logit contributions | |
ield | -0.818 |
vine | -0.814 |
het | -0.801 |
794 | -0.795 |
PF | -0.790 |
Token | As |
Token ID | 1081 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
angles | +0.275 |
ortment | +0.250 |
mble | +0.250 |
Supervisor | +0.249 |
aina | +0.248 |
Neg logit contributions | |
ield | -0.272 |
vine | -0.270 |
het | -0.266 |
794 | -0.264 |
PF | -0.262 |
Token | Coinbase |
Token ID | 45802 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
angles | +0.935 |
Supervisor | +0.858 |
ortment | +0.858 |
mble | +0.852 |
aina | +0.850 |
Neg logit contributions | |
ield | -0.735 |
vine | -0.732 |
het | -0.717 |
794 | -0.712 |
PF | -0.707 |
Token | co |
Token ID | 763 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
ield | +0.879 |
vine | +0.876 |
het | +0.863 |
794 | +0.853 |
PF | +0.850 |
Neg logit contributions | |
angles | -0.886 |
ortment | -0.805 |
mble | -0.802 |
aina | -0.795 |
Supervisor | -0.787 |
Token | - |
Token ID | 12 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
angles | +0.507 |
ortment | +0.467 |
Supervisor | +0.466 |
mble | +0.464 |
aina | +0.463 |
Neg logit contributions | |
ield | -0.349 |
vine | -0.347 |
het | -0.340 |
794 | -0.337 |
PF | -0.334 |
Token | founder |
Token ID | 15454 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
ield | +0.734 |
vine | +0.728 |
het | +0.710 |
794 | +0.700 |
PF | +0.694 |
Neg logit contributions | |
angles | -1.533 |
ortment | -1.431 |
Supervisor | -1.428 |
mble | -1.426 |
aina | -1.419 |
Token | Fred |
Token ID | 8559 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
vine | +1.312 |
ield | +1.307 |
het | +1.302 |
Finch | +1.282 |
794 | +1.266 |
Neg logit contributions | |
angles | -1.366 |
mble | -1.264 |
ortment | -1.251 |
aina | -1.234 |
Supervisor | -1.226 |
Token | Eh |
Token ID | 31480 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
angles | +4.234 |
Buff | +3.764 |
aina | +3.756 |
Supervisor | +3.728 |
ortment | +3.661 |
Neg logit contributions | |
ield | -4.882 |
vine | -4.816 |
ournals | -4.805 |
PF | -4.802 |
794 | -4.730 |
Token | rs |
Token ID | 3808 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
angles | +0.005 |
Supervisor | +0.005 |
ortment | +0.005 |
aina | +0.005 |
mble | +0.005 |
Neg logit contributions | |
ield | -0.005 |
vine | -0.005 |
het | -0.005 |
794 | -0.005 |
PF | -0.005 |
Token | am |
Token ID | 321 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
angles | +2.224 |
Supervisor | +2.044 |
ortment | +2.034 |
aina | +2.025 |
mble | +1.995 |
Neg logit contributions | |
vine | -1.665 |
ield | -1.664 |
ordes | -1.607 |
794 | -1.602 |
PF | -1.602 |
Token | emphasized |
Token ID | 20047 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
ield | +2.264 |
vine | +2.233 |
het | +2.225 |
794 | +2.177 |
PF | +2.169 |
Neg logit contributions | |
angles | -2.453 |
mble | -2.237 |
ortment | -2.228 |
Supervisor | -2.222 |
aina | -2.197 |
Token | in |
Token ID | 287 |
Feature activation | +8.49e-03 |
Pos logit contributions | |
ield | +0.805 |
vine | +0.799 |
het | +0.791 |
794 | +0.780 |
PF | +0.776 |
Neg logit contributions | |
angles | -0.886 |
ortment | -0.807 |
mble | -0.800 |
Supervisor | -0.797 |
aina | -0.797 |
Token | it |
Token ID | 340 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
ield | +0.067 |
vine | +0.066 |
het | +0.065 |
794 | +0.065 |
PF | +0.064 |
Neg logit contributions | |
angles | -0.070 |
ortment | -0.064 |
Supervisor | -0.063 |
mble | -0.063 |
aina | -0.063 |
Token | comes |
Token ID | 2058 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
angles | +0.555 |
ortment | +0.499 |
Supervisor | +0.499 |
mble | +0.495 |
aina | +0.495 |
Neg logit contributions | |
ield | -0.629 |
vine | -0.626 |
het | -0.616 |
794 | -0.612 |
PF | -0.609 |
Token | up |
Token ID | 510 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
angles | +1.788 |
ortment | +1.633 |
Supervisor | +1.630 |
mble | +1.626 |
aina | +1.616 |
Neg logit contributions | |
ield | -1.563 |
vine | -1.559 |
het | -1.531 |
794 | -1.521 |
PF | -1.507 |
Token | in |
Token ID | 287 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
angles | +0.316 |
ortment | +0.288 |
Supervisor | +0.287 |
mble | +0.287 |
aina | +0.286 |
Neg logit contributions | |
ield | -0.282 |
vine | -0.281 |
het | -0.276 |
794 | -0.274 |
PF | -0.271 |
Token | my |
Token ID | 616 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
angles | +0.745 |
Supervisor | +0.683 |
ortment | +0.681 |
mble | +0.677 |
aina | +0.676 |
Neg logit contributions | |
ield | -0.611 |
vine | -0.608 |
het | -0.596 |
794 | -0.593 |
PF | -0.588 |
Token | mind |
Token ID | 2000 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
angles | +1.537 |
Supervisor | +1.399 |
ortment | +1.388 |
mble | +1.379 |
aina | +1.379 |
Neg logit contributions | |
ield | -1.646 |
vine | -1.645 |
het | -1.616 |
794 | -1.604 |
PF | -1.600 |
Token | . |
Token ID | 13 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
angles | +4.304 |
Supervisor | +3.965 |
ortment | +3.902 |
aina | +3.890 |
mble | +3.869 |
Neg logit contributions | |
ield | -4.596 |
vine | -4.552 |
794 | -4.455 |
het | -4.449 |
PF | -4.444 |
Token | A |
Token ID | 317 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
angles | +0.281 |
Supervisor | +0.256 |
ortment | +0.256 |
mble | +0.253 |
aina | +0.253 |
Neg logit contributions | |
ield | -0.273 |
vine | -0.272 |
het | -0.267 |
794 | -0.265 |
PF | -0.263 |
Token | voice |
Token ID | 3809 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
ield | +0.038 |
vine | +0.037 |
het | +0.037 |
794 | +0.036 |
PF | +0.036 |
Neg logit contributions | |
angles | -0.053 |
ortment | -0.049 |
Supervisor | -0.049 |
mble | -0.049 |
aina | -0.049 |
Token | in |
Token ID | 287 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
angles | +0.090 |
ortment | +0.082 |
Supervisor | +0.082 |
mble | +0.081 |
aina | +0.081 |
Neg logit contributions | |
ield | -0.086 |
vine | -0.085 |
het | -0.084 |
794 | -0.083 |
PF | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
ield | +0.274 |
vine | +0.274 |
het | +0.270 |
794 | +0.266 |
PF | +0.264 |
Neg logit contributions | |
angles | -0.355 |
ortment | -0.328 |
mble | -0.325 |
Supervisor | -0.324 |
aina | -0.323 |
Token | , |
Token ID | 11 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
angles | +0.545 |
ortment | +0.497 |
Supervisor | +0.496 |
mble | +0.494 |
aina | +0.492 |
Neg logit contributions | |
ield | -0.491 |
vine | -0.489 |
het | -0.480 |
794 | -0.477 |
PF | -0.473 |
Token | and |
Token ID | 290 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
vine | +0.416 |
ield | +0.416 |
het | +0.409 |
794 | +0.405 |
PF | +0.402 |
Neg logit contributions | |
angles | -0.507 |
ortment | -0.464 |
Supervisor | -0.462 |
mble | -0.461 |
aina | -0.458 |
Token | as |
Token ID | 355 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
vine | +0.138 |
ield | +0.137 |
het | +0.135 |
794 | +0.134 |
PF | +0.132 |
Neg logit contributions | |
angles | -0.163 |
ortment | -0.149 |
mble | -0.147 |
Supervisor | -0.147 |
aina | -0.147 |
Token | I |
Token ID | 314 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
vine | +0.197 |
ield | +0.197 |
het | +0.193 |
794 | +0.191 |
PF | +0.190 |
Neg logit contributions | |
angles | -0.248 |
ortment | -0.226 |
mble | -0.225 |
Supervisor | -0.225 |
aina | -0.224 |
Token | am |
Token ID | 716 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ield | +0.548 |
vine | +0.542 |
het | +0.535 |
794 | +0.530 |
drip | +0.526 |
Neg logit contributions | |
angles | -0.562 |
ortment | -0.510 |
mble | -0.509 |
Supervisor | -0.506 |
aina | -0.500 |
Token | apt |
Token ID | 15409 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
ield | +0.236 |
vine | +0.235 |
het | +0.230 |
794 | +0.228 |
PF | +0.227 |
Neg logit contributions | |
angles | -0.315 |
ortment | -0.289 |
Supervisor | -0.289 |
mble | -0.288 |
aina | -0.286 |
Token | to |
Token ID | 284 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
angles | +4.209 |
mble | +3.788 |
ortment | +3.787 |
aina | +3.786 |
Supervisor | +3.775 |
Neg logit contributions | |
ield | -4.550 |
vine | -4.504 |
794 | -4.478 |
het | -4.453 |
PF | -4.416 |
Token | do |
Token ID | 466 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
ield | +0.218 |
vine | +0.216 |
het | +0.212 |
794 | +0.211 |
PF | +0.208 |
Neg logit contributions | |
angles | -0.269 |
ortment | -0.246 |
Supervisor | -0.245 |
mble | -0.245 |
aina | -0.242 |
Token | when |
Token ID | 618 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
vine | +0.758 |
ield | +0.757 |
het | +0.741 |
794 | +0.734 |
PF | +0.729 |
Neg logit contributions | |
angles | -0.858 |
ortment | -0.789 |
mble | -0.779 |
Supervisor | -0.778 |
aina | -0.775 |
Token | I |
Token ID | 314 |
Feature activation | -8.23e-02 |
Pos logit contributions | |
angles | +1.347 |
Supervisor | +1.235 |
ortment | +1.230 |
mble | +1.222 |
aina | +1.219 |
Neg logit contributions | |
ield | -1.224 |
vine | -1.210 |
het | -1.193 |
794 | -1.186 |
PF | -1.176 |
Token | have |
Token ID | 423 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
ield | +0.163 |
vine | +0.162 |
het | +0.159 |
794 | +0.158 |
PF | +0.156 |
Neg logit contributions | |
angles | -0.190 |
ortment | -0.174 |
Supervisor | -0.173 |
mble | -0.172 |
aina | -0.171 |
Token | the |
Token ID | 262 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
ield | +1.374 |
vine | +1.367 |
het | +1.335 |
794 | +1.326 |
drip | +1.319 |
Neg logit contributions | |
angles | -1.751 |
ortment | -1.600 |
Supervisor | -1.596 |
mble | -1.590 |
aina | -1.581 |
Token | Coconut |
Token ID | 46943 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
angles | +0.463 |
Supervisor | +0.428 |
ortment | +0.427 |
mble | +0.425 |
aina | +0.424 |
Neg logit contributions | |
ield | -0.319 |
vine | -0.318 |
het | -0.311 |
794 | -0.308 |
PF | -0.306 |
Token | Bay |
Token ID | 4696 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
angles | +0.941 |
Supervisor | +0.855 |
ortment | +0.855 |
aina | +0.849 |
mble | +0.847 |
Neg logit contributions | |
ield | -0.921 |
vine | -0.913 |
het | -0.897 |
794 | -0.896 |
PF | -0.891 |
Token | Resort |
Token ID | 28345 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
ield | +0.730 |
vine | +0.726 |
het | +0.715 |
794 | +0.706 |
PF | +0.704 |
Neg logit contributions | |
angles | -0.626 |
ortment | -0.562 |
Supervisor | -0.561 |
mble | -0.560 |
aina | -0.554 |
Token | & |
Token ID | 1222 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
angles | +1.636 |
Supervisor | +1.486 |
ortment | +1.484 |
mble | +1.473 |
aina | +1.469 |
Neg logit contributions | |
ield | -1.664 |
vine | -1.660 |
het | -1.630 |
794 | -1.620 |
PF | -1.610 |
Token | Spa |
Token ID | 40208 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
angles | +0.521 |
ortment | +0.467 |
mble | +0.465 |
Supervisor | +0.462 |
aina | +0.461 |
Neg logit contributions | |
ield | -0.571 |
vine | -0.571 |
het | -0.561 |
794 | -0.556 |
PF | -0.553 |
Token | in |
Token ID | 287 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
angles | +4.297 |
Supervisor | +3.966 |
ortment | +3.918 |
aina | +3.877 |
Buff | +3.876 |
Neg logit contributions | |
vine | -4.417 |
ield | -4.376 |
794 | -4.296 |
het | -4.294 |
PF | -4.281 |
Token | St |
Token ID | 520 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
angles | +1.086 |
Supervisor | +0.989 |
ortment | +0.986 |
mble | +0.979 |
aina | +0.979 |
Neg logit contributions | |
ield | -1.054 |
vine | -1.049 |
het | -1.031 |
794 | -1.025 |
PF | -1.018 |
Token | Lucia |
Token ID | 48158 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
angles | +0.894 |
Supervisor | +0.806 |
ortment | +0.805 |
mble | +0.800 |
aina | +0.796 |
Neg logit contributions | |
ield | -1.017 |
vine | -1.012 |
het | -0.997 |
794 | -0.989 |
PF | -0.983 |
Token | from |
Token ID | 422 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
angles | +0.473 |
ortment | +0.431 |
Supervisor | +0.431 |
mble | +0.428 |
aina | +0.426 |
Neg logit contributions | |
ield | -0.428 |
vine | -0.426 |
het | -0.418 |
794 | -0.415 |
PF | -0.413 |
Token | £ |
Token ID | 4248 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
angles | +1.203 |
Supervisor | +1.102 |
ortment | +1.101 |
mble | +1.095 |
aina | +1.092 |
Neg logit contributions | |
ield | -1.023 |
vine | -1.020 |
het | -1.001 |
794 | -0.991 |
PF | -0.986 |
Token | see |
Token ID | 766 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
angles | +0.396 |
ortment | +0.361 |
Supervisor | +0.360 |
mble | +0.358 |
aina | +0.357 |
Neg logit contributions | |
ield | -0.368 |
vine | -0.366 |
het | -0.360 |
794 | -0.357 |
PF | -0.355 |
Token | inside |
Token ID | 2641 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
angles | +0.054 |
ortment | +0.049 |
Supervisor | +0.049 |
mble | +0.049 |
aina | +0.049 |
Neg logit contributions | |
ield | -0.049 |
vine | -0.049 |
het | -0.048 |
794 | -0.047 |
PF | -0.047 |
Token | of |
Token ID | 286 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
ield | +1.350 |
vine | +1.343 |
het | +1.321 |
794 | +1.312 |
PF | +1.307 |
Neg logit contributions | |
angles | -1.596 |
ortment | -1.455 |
mble | -1.451 |
Supervisor | -1.449 |
aina | -1.442 |
Token | it |
Token ID | 340 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
angles | +0.772 |
Supervisor | +0.709 |
ortment | +0.708 |
mble | +0.701 |
aina | +0.700 |
Neg logit contributions | |
ield | -0.655 |
vine | -0.653 |
het | -0.640 |
794 | -0.635 |
PF | -0.629 |
Token | . |
Token ID | 13 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
angles | +0.761 |
ortment | +0.694 |
Supervisor | +0.694 |
mble | +0.690 |
aina | +0.688 |
Neg logit contributions | |
ield | -0.684 |
vine | -0.681 |
het | -0.668 |
794 | -0.664 |
PF | -0.659 |
Token | Take |
Token ID | 7214 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
angles | +0.632 |
Supervisor | +0.580 |
ortment | +0.577 |
mble | +0.571 |
aina | +0.570 |
Neg logit contributions | |
ield | -0.602 |
vine | -0.599 |
het | -0.588 |
794 | -0.585 |
PF | -0.580 |
Token | a |
Token ID | 257 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
angles | +4.563 |
Supervisor | +4.297 |
ortment | +4.209 |
aina | +4.202 |
mble | +4.160 |
Neg logit contributions | |
ield | -4.098 |
vine | -4.070 |
het | -3.986 |
794 | -3.931 |
ournals | -3.905 |
Token | look |
Token ID | 804 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
angles | +0.867 |
Supervisor | +0.791 |
ortment | +0.790 |
mble | +0.785 |
aina | +0.784 |
Neg logit contributions | |
ield | -0.807 |
vine | -0.804 |
het | -0.790 |
794 | -0.784 |
PF | -0.778 |
Token | at |
Token ID | 379 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
angles | +0.555 |
Supervisor | +0.506 |
ortment | +0.506 |
mble | +0.503 |
aina | +0.502 |
Neg logit contributions | |
ield | -0.490 |
vine | -0.488 |
het | -0.479 |
794 | -0.476 |
PF | -0.472 |
Token | the |
Token ID | 262 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
ield | +0.952 |
vine | +0.950 |
het | +0.930 |
794 | +0.925 |
PF | +0.920 |
Neg logit contributions | |
angles | -1.228 |
ortment | -1.125 |
mble | -1.120 |
Supervisor | -1.117 |
aina | -1.113 |
Token | map |
Token ID | 3975 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
angles | +0.115 |
Supervisor | +0.105 |
ortment | +0.105 |
mble | +0.104 |
aina | +0.104 |
Neg logit contributions | |
ield | -0.101 |
vine | -0.101 |
het | -0.099 |
794 | -0.098 |
PF | -0.098 |
Token | by |
Token ID | 416 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
ield | +0.125 |
vine | +0.125 |
het | +0.122 |
794 | +0.121 |
PF | +0.120 |
Neg logit contributions | |
angles | -0.177 |
Supervisor | -0.163 |
ortment | -0.163 |
mble | -0.162 |
aina | -0.162 |
Token | talking |
Token ID | 3375 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
angles | +0.223 |
ortment | +0.204 |
mble | +0.203 |
Supervisor | +0.203 |
aina | +0.202 |
Neg logit contributions | |
ield | -0.183 |
vine | -0.182 |
het | -0.178 |
794 | -0.177 |
PF | -0.176 |
Token | about |
Token ID | 546 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
vine | +0.615 |
ield | +0.615 |
het | +0.597 |
794 | +0.595 |
PF | +0.587 |
Neg logit contributions | |
angles | -0.962 |
ortment | -0.890 |
mble | -0.888 |
aina | -0.886 |
Supervisor | -0.884 |
Token | the |
Token ID | 262 |
Feature activation | -7.91e-03 |
Pos logit contributions | |
angles | +0.285 |
Supervisor | +0.262 |
ortment | +0.262 |
mble | +0.260 |
aina | +0.259 |
Neg logit contributions | |
ield | -0.217 |
vine | -0.216 |
het | -0.212 |
794 | -0.210 |
PF | -0.208 |
Token | seemingly |
Token ID | 9775 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
ield | +0.015 |
vine | +0.015 |
het | +0.015 |
794 | +0.015 |
PF | +0.015 |
Neg logit contributions | |
angles | -0.018 |
ortment | -0.017 |
Supervisor | -0.017 |
mble | -0.017 |
aina | -0.017 |
Token | incompetent |
Token ID | 33840 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
angles | +1.703 |
Supervisor | +1.569 |
ortment | +1.552 |
aina | +1.540 |
mble | +1.535 |
Neg logit contributions | |
vine | -1.415 |
ield | -1.409 |
het | -1.377 |
PF | -1.375 |
794 | -1.369 |
Token | investigation |
Token ID | 3645 |
Feature activation | +8.82e-02 |
Pos logit contributions | |
angles | +4.191 |
Supervisor | +3.991 |
ortment | +3.831 |
mble | +3.801 |
aina | +3.766 |
Neg logit contributions | |
ield | -4.368 |
vine | -4.367 |
ournals | -4.299 |
het | -4.265 |
PF | -4.264 |
Token | of |
Token ID | 286 |
Feature activation | +9.08e-03 |
Pos logit contributions | |
angles | +0.209 |
Supervisor | +0.191 |
ortment | +0.191 |
mble | +0.190 |
aina | +0.189 |
Neg logit contributions | |
ield | -0.169 |
vine | -0.168 |
het | -0.165 |
794 | -0.164 |
PF | -0.163 |
Token | the |
Token ID | 262 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
angles | +0.019 |
ortment | +0.018 |
Supervisor | +0.018 |
mble | +0.018 |
aina | +0.017 |
Neg logit contributions | |
ield | -0.019 |
vine | -0.019 |
het | -0.019 |
794 | -0.019 |
PF | -0.019 |
Token | Las |
Token ID | 10123 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
ield | +0.280 |
vine | +0.279 |
het | +0.275 |
794 | +0.273 |
PF | +0.270 |
Neg logit contributions | |
angles | -0.292 |
ortment | -0.265 |
mble | -0.264 |
Supervisor | -0.263 |
aina | -0.262 |
Token | Vegas |
Token ID | 9621 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
vine | +0.491 |
ield | +0.491 |
het | +0.484 |
794 | +0.480 |
PF | +0.477 |
Neg logit contributions | |
angles | -0.260 |
ortment | -0.227 |
mble | -0.226 |
Supervisor | -0.225 |
aina | -0.221 |
Token | regular |
Token ID | 16338 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
angles | +1.341 |
ortment | +1.217 |
mble | +1.214 |
aina | +1.210 |
Supervisor | +1.208 |
Neg logit contributions | |
ield | -1.142 |
vine | -1.124 |
het | -1.114 |
794 | -1.107 |
PF | -1.106 |
Token | s |
Token ID | 82 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
angles | +1.553 |
Supervisor | +1.408 |
ortment | +1.404 |
mble | +1.394 |
aina | +1.384 |
Neg logit contributions | |
ield | -1.626 |
vine | -1.618 |
het | -1.596 |
PF | -1.579 |
794 | -1.576 |
Token | theme |
Token ID | 7505 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
ield | +0.644 |
vine | +0.641 |
het | +0.631 |
794 | +0.627 |
PF | +0.621 |
Neg logit contributions | |
angles | -0.686 |
ortment | -0.621 |
Supervisor | -0.621 |
mble | -0.618 |
aina | -0.616 |
Token | represents |
Token ID | 6870 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
angles | +0.267 |
Supervisor | +0.241 |
ortment | +0.241 |
mble | +0.239 |
aina | +0.238 |
Neg logit contributions | |
ield | -0.290 |
vine | -0.289 |
het | -0.285 |
794 | -0.283 |
PF | -0.281 |
Token | the |
Token ID | 262 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
angles | +0.508 |
Supervisor | +0.467 |
ortment | +0.467 |
mble | +0.464 |
aina | +0.463 |
Neg logit contributions | |
ield | -0.397 |
vine | -0.395 |
het | -0.387 |
794 | -0.384 |
PF | -0.381 |
Token | unorthodox |
Token ID | 49830 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
angles | +0.385 |
Supervisor | +0.355 |
ortment | +0.355 |
mble | +0.352 |
aina | +0.352 |
Neg logit contributions | |
ield | -0.278 |
vine | -0.277 |
het | -0.271 |
794 | -0.268 |
PF | -0.266 |
Token | collections |
Token ID | 17268 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
angles | +4.665 |
Supervisor | +4.284 |
ortment | +4.284 |
aina | +4.205 |
mble | +4.196 |
Neg logit contributions | |
vine | -3.914 |
ield | -3.907 |
het | -3.878 |
ournals | -3.841 |
794 | -3.780 |
Token | of |
Token ID | 286 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
angles | +1.633 |
Supervisor | +1.499 |
ortment | +1.497 |
mble | +1.477 |
aina | +1.475 |
Neg logit contributions | |
ield | -1.441 |
vine | -1.437 |
het | -1.400 |
794 | -1.391 |
PF | -1.386 |
Token | soldiers |
Token ID | 5795 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
angles | +0.402 |
ortment | +0.367 |
Supervisor | +0.367 |
mble | +0.365 |
aina | +0.364 |
Neg logit contributions | |
ield | -0.340 |
vine | -0.338 |
het | -0.332 |
794 | -0.330 |
PF | -0.327 |
Token | who |
Token ID | 508 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
angles | +0.403 |
Supervisor | +0.369 |
ortment | +0.369 |
mble | +0.366 |
aina | +0.365 |
Neg logit contributions | |
ield | -0.334 |
vine | -0.333 |
het | -0.327 |
794 | -0.324 |
PF | -0.322 |
Token | sometimes |
Token ID | 3360 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
ield | +0.107 |
vine | +0.106 |
het | +0.104 |
794 | +0.103 |
PF | +0.102 |
Neg logit contributions | |
angles | -0.117 |
Supervisor | -0.107 |
ortment | -0.107 |
mble | -0.106 |
aina | -0.106 |
Token | , |
Token ID | 11 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
ield | +0.317 |
vine | +0.317 |
het | +0.311 |
794 | +0.309 |
PF | +0.307 |
Neg logit contributions | |
angles | -0.325 |
ortment | -0.295 |
Supervisor | -0.294 |
mble | -0.293 |
aina | -0.292 |
Token | that |
Token ID | 326 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
angles | +1.269 |
Supervisor | +1.171 |
ortment | +1.154 |
mble | +1.149 |
aina | +1.143 |
Neg logit contributions | |
ield | -1.101 |
vine | -1.093 |
het | -1.068 |
794 | -1.066 |
PF | -1.056 |
Token | 's |
Token ID | 338 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
angles | +0.270 |
Supervisor | +0.243 |
ortment | +0.243 |
mble | +0.241 |
aina | +0.240 |
Neg logit contributions | |
ield | -0.325 |
vine | -0.323 |
het | -0.318 |
794 | -0.316 |
PF | -0.314 |
Token | the |
Token ID | 262 |
Feature activation | +5.69e-02 |
Pos logit contributions | |
angles | +0.638 |
Supervisor | +0.592 |
ortment | +0.586 |
mble | +0.585 |
aina | +0.581 |
Neg logit contributions | |
ield | -0.456 |
vine | -0.454 |
het | -0.443 |
794 | -0.440 |
PF | -0.436 |
Token | old |
Token ID | 1468 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
angles | +0.137 |
Supervisor | +0.126 |
ortment | +0.125 |
mble | +0.125 |
aina | +0.125 |
Neg logit contributions | |
ield | -0.106 |
vine | -0.106 |
het | -0.104 |
794 | -0.103 |
PF | -0.102 |
Token | Stan |
Token ID | 7299 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
angles | +0.993 |
Supervisor | +0.919 |
ortment | +0.905 |
aina | +0.905 |
mble | +0.904 |
Neg logit contributions | |
ield | -0.697 |
vine | -0.690 |
het | -0.682 |
794 | -0.674 |
PF | -0.670 |
Token | Lee |
Token ID | 5741 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
angles | +4.374 |
Supervisor | +4.060 |
Buff | +3.931 |
bilt | +3.889 |
ortment | +3.878 |
Neg logit contributions | |
ield | -4.176 |
PF | -4.048 |
het | -4.024 |
794 | -3.992 |
vine | -3.971 |
Token | quote |
Token ID | 9577 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
angles | +1.449 |
Supervisor | +1.339 |
ortment | +1.310 |
aina | +1.300 |
mble | +1.298 |
Neg logit contributions | |
ield | -1.263 |
vine | -1.248 |
het | -1.238 |
794 | -1.226 |
PF | -1.223 |
Token | : |
Token ID | 25 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
angles | +0.597 |
Supervisor | +0.544 |
ortment | +0.543 |
mble | +0.540 |
aina | +0.537 |
Neg logit contributions | |
ield | -0.575 |
vine | -0.572 |
het | -0.562 |
794 | -0.558 |
PF | -0.555 |
Token | Thing |
Token ID | 21561 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
angles | +0.130 |
Supervisor | +0.118 |
ortment | +0.118 |
mble | +0.118 |
aina | +0.117 |
Neg logit contributions | |
ield | -0.128 |
vine | -0.127 |
het | -0.125 |
794 | -0.124 |
PF | -0.124 |
Token | versus |
Token ID | 9051 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
angles | +1.456 |
Supervisor | +1.319 |
ortment | +1.312 |
mble | +1.301 |
aina | +1.299 |
Neg logit contributions | |
ield | -1.470 |
vine | -1.467 |
het | -1.439 |
794 | -1.433 |
PF | -1.425 |
Token | details |
Token ID | 3307 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
angles | +0.659 |
Supervisor | +0.612 |
ortment | +0.608 |
aina | +0.605 |
mble | +0.605 |
Neg logit contributions | |
ield | -0.415 |
vine | -0.412 |
het | -0.403 |
794 | -0.399 |
PF | -0.396 |
Token | below |
Token ID | 2174 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
angles | +0.120 |
ortment | +0.110 |
Supervisor | +0.110 |
mble | +0.109 |
aina | +0.109 |
Neg logit contributions | |
ield | -0.091 |
vine | -0.091 |
het | -0.089 |
794 | -0.088 |
PF | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
angles | +0.160 |
ortment | +0.146 |
Supervisor | +0.146 |
mble | +0.145 |
aina | +0.144 |
Neg logit contributions | |
ield | -0.145 |
vine | -0.144 |
het | -0.142 |
794 | -0.141 |
PF | -0.140 |
Token |
|
Token ID | 198 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
ield | +0.402 |
vine | +0.401 |
het | +0.394 |
794 | +0.391 |
PF | +0.389 |
Neg logit contributions | |
angles | -0.424 |
ortment | -0.384 |
mble | -0.384 |
Supervisor | -0.383 |
aina | -0.381 |
Token |
|
Token ID | 198 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
ield | +0.419 |
vine | +0.419 |
794 | +0.411 |
het | +0.409 |
PF | +0.405 |
Neg logit contributions | |
angles | -0.498 |
mble | -0.474 |
ortment | -0.460 |
unlaw | -0.457 |
aina | -0.453 |
Token | ■ |
Token ID | 29316 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
ield | +0.430 |
vine | +0.429 |
het | +0.421 |
794 | +0.420 |
PF | +0.418 |
Neg logit contributions | |
angles | -0.475 |
ortment | -0.434 |
Supervisor | -0.434 |
mble | -0.432 |
aina | -0.429 |
Token | Opening |
Token ID | 25522 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
angles | +4.189 |
Supervisor | +3.872 |
ortment | +3.815 |
aina | +3.728 |
Buff | +3.725 |
Neg logit contributions | |
ield | -4.379 |
vine | -4.368 |
het | -4.271 |
ournals | -4.242 |
drip | -4.229 |
Token | Theme |
Token ID | 26729 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
angles | +1.070 |
Supervisor | +0.971 |
ortment | +0.970 |
mble | +0.964 |
aina | +0.960 |
Neg logit contributions | |
ield | -1.081 |
vine | -1.077 |
het | -1.058 |
794 | -1.051 |
PF | -1.044 |
Token | Movie |
Token ID | 15875 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
angles | +2.197 |
Supervisor | +2.016 |
ortment | +2.002 |
mble | +1.977 |
aina | +1.969 |
Neg logit contributions | |
ield | -2.271 |
vine | -2.258 |
het | -2.215 |
794 | -2.192 |
PF | -2.184 |
Token | The |
Token ID | 383 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
angles | +1.410 |
Supervisor | +1.284 |
ortment | +1.280 |
mble | +1.272 |
aina | +1.265 |
Neg logit contributions | |
ield | -1.444 |
vine | -1.436 |
het | -1.412 |
794 | -1.398 |
PF | -1.392 |
Token | opening |
Token ID | 4756 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
angles | +1.134 |
Supervisor | +1.039 |
ortment | +1.036 |
mble | +1.032 |
aina | +1.029 |
Neg logit contributions | |
ield | -0.986 |
vine | -0.982 |
het | -0.963 |
794 | -0.954 |
PF | -0.950 |
Token | from |
Token ID | 422 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
ield | +0.610 |
vine | +0.606 |
het | +0.596 |
794 | +0.590 |
PF | +0.586 |
Neg logit contributions | |
angles | -0.644 |
ortment | -0.588 |
Supervisor | -0.585 |
mble | -0.585 |
aina | -0.581 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
angles | +0.631 |
Supervisor | +0.577 |
ortment | +0.576 |
mble | +0.573 |
aina | +0.572 |
Neg logit contributions | |
ield | -0.539 |
vine | -0.535 |
het | -0.526 |
794 | -0.522 |
PF | -0.518 |
Token | Harvard |
Token ID | 11131 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
angles | +0.234 |
Supervisor | +0.211 |
ortment | +0.211 |
mble | +0.209 |
aina | +0.208 |
Neg logit contributions | |
ield | -0.259 |
vine | -0.258 |
het | -0.254 |
794 | -0.252 |
PF | -0.251 |
Token | - |
Token ID | 12 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
angles | +0.004 |
Supervisor | +0.004 |
ortment | +0.004 |
mble | +0.004 |
aina | +0.004 |
Neg logit contributions | |
ield | -0.004 |
vine | -0.004 |
het | -0.004 |
794 | -0.004 |
PF | -0.004 |
Token | Smith |
Token ID | 17919 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
angles | +0.865 |
ortment | +0.767 |
aina | +0.767 |
Supervisor | +0.754 |
mble | +0.751 |
Neg logit contributions | |
vine | -0.820 |
ield | -0.816 |
het | -0.809 |
drip | -0.802 |
794 | -0.801 |
Token | sonian |
Token ID | 35202 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
angles | +0.395 |
ortment | +0.269 |
Supervisor | +0.267 |
mble | +0.262 |
aina | +0.254 |
Neg logit contributions | |
ield | -2.332 |
vine | -2.327 |
het | -2.303 |
794 | -2.294 |
PF | -2.286 |
Token | Center |
Token ID | 3337 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
angles | +3.123 |
Supervisor | +2.832 |
mble | +2.758 |
aina | +2.621 |
Buff | +2.603 |
Neg logit contributions | |
794 | -5.382 |
vine | -5.376 |
het | -5.367 |
ield | -5.336 |
PF | -5.303 |
Token | for |
Token ID | 329 |
Feature activation | -6.43e-03 |
Pos logit contributions | |
angles | +1.858 |
Supervisor | +1.680 |
ortment | +1.677 |
mble | +1.649 |
aina | +1.646 |
Neg logit contributions | |
ield | -2.068 |
vine | -2.067 |
het | -2.030 |
794 | -2.017 |
PF | -2.007 |
Token | Ast |
Token ID | 8304 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ield | +0.015 |
vine | +0.015 |
het | +0.014 |
794 | +0.014 |
PF | +0.014 |
Neg logit contributions | |
angles | -0.013 |
Supervisor | -0.012 |
ortment | -0.012 |
mble | -0.011 |
aina | -0.011 |
Token | roph |
Token ID | 10051 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
angles | +2.167 |
Supervisor | +1.937 |
ortment | +1.933 |
aina | +1.930 |
mble | +1.888 |
Neg logit contributions | |
vine | -2.601 |
794 | -2.565 |
ield | -2.560 |
PF | -2.536 |
drip | -2.526 |
Token | ysics |
Token ID | 23154 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
angles | +4.050 |
ortment | +3.717 |
aina | +3.647 |
Supervisor | +3.628 |
mble | +3.593 |
Neg logit contributions | |
vine | -3.192 |
ield | -3.167 |
794 | -3.149 |
PF | -3.123 |
drip | -3.115 |
Token | the |
Token ID | 262 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
ield | +0.106 |
vine | +0.106 |
het | +0.104 |
794 | +0.103 |
PF | +0.103 |
Neg logit contributions | |
angles | -0.115 |
ortment | -0.104 |
mble | -0.104 |
Supervisor | -0.104 |
aina | -0.103 |
Token | video |
Token ID | 2008 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
angles | +0.199 |
ortment | +0.181 |
mble | +0.180 |
Supervisor | +0.180 |
aina | +0.179 |
Neg logit contributions | |
ield | -0.193 |
vine | -0.192 |
het | -0.188 |
794 | -0.187 |
PF | -0.185 |
Token | start |
Token ID | 923 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
ield | +0.464 |
vine | +0.461 |
het | +0.448 |
794 | +0.444 |
PF | +0.437 |
Neg logit contributions | |
angles | -0.968 |
ortment | -0.904 |
Supervisor | -0.899 |
mble | -0.897 |
aina | -0.894 |
Token | - |
Token ID | 12 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
ield | +0.717 |
vine | +0.713 |
het | +0.701 |
794 | +0.693 |
PF | +0.682 |
Neg logit contributions | |
angles | -1.019 |
ortment | -0.940 |
Supervisor | -0.940 |
aina | -0.935 |
mble | -0.933 |
Token | up |
Token ID | 929 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
angles | +0.525 |
Supervisor | +0.470 |
ortment | +0.469 |
mble | +0.466 |
aina | +0.464 |
Neg logit contributions | |
ield | -0.643 |
vine | -0.641 |
het | -0.630 |
794 | -0.627 |
PF | -0.624 |
Token | Magn |
Token ID | 14872 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
ield | +0.289 |
vine | +0.289 |
het | +0.282 |
794 | +0.280 |
PF | +0.278 |
Neg logit contributions | |
angles | -0.311 |
ortment | -0.284 |
Supervisor | -0.282 |
mble | -0.282 |
aina | -0.281 |
Token | ify |
Token ID | 1958 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
angles | +5.234 |
aina | +4.798 |
ortment | +4.706 |
mble | +4.674 |
Supervisor | +4.659 |
Neg logit contributions | |
ield | -3.338 |
794 | -3.330 |
vine | -3.321 |
PF | -3.295 |
Finch | -3.261 |
Token | . |
Token ID | 13 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
angles | +0.425 |
ortment | +0.389 |
Supervisor | +0.388 |
mble | +0.386 |
aina | +0.385 |
Neg logit contributions | |
ield | -0.368 |
vine | -0.367 |
het | -0.360 |
794 | -0.357 |
PF | -0.355 |
Token |
|
Token ID | 198 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
angles | +0.102 |
mble | +0.093 |
ortment | +0.093 |
Supervisor | +0.092 |
aina | +0.092 |
Neg logit contributions | |
ield | -0.095 |
vine | -0.095 |
het | -0.094 |
794 | -0.093 |
PF | -0.092 |
Token |
|
Token ID | 198 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
angles | +0.385 |
mble | +0.357 |
ortment | +0.353 |
Supervisor | +0.350 |
aina | +0.348 |
Neg logit contributions | |
ield | -0.350 |
vine | -0.349 |
het | -0.342 |
794 | -0.342 |
PF | -0.338 |
Token | Advertisement |
Token ID | 4723 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
angles | +0.280 |
Supervisor | +0.256 |
ortment | +0.256 |
aina | +0.254 |
mble | +0.254 |
Neg logit contributions | |
ield | -0.216 |
vine | -0.214 |
het | -0.210 |
794 | -0.208 |
PF | -0.206 |
Token | , |
Token ID | 11 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
angles | +0.080 |
Supervisor | +0.073 |
ortment | +0.073 |
mble | +0.073 |
aina | +0.073 |
Neg logit contributions | |
ield | -0.071 |
vine | -0.070 |
het | -0.069 |
794 | -0.069 |
PF | -0.068 |
Token | � |
Token ID | 447 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
ield | +0.621 |
vine | +0.618 |
het | +0.607 |
794 | +0.602 |
PF | +0.597 |
Neg logit contributions | |
angles | -0.802 |
ortment | -0.736 |
Supervisor | -0.735 |
mble | -0.732 |
aina | -0.729 |
Token | � |
Token ID | 251 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ield | +0.029 |
vine | +0.029 |
het | +0.028 |
794 | +0.028 |
PF | +0.028 |
Neg logit contributions | |
angles | -0.033 |
ortment | -0.030 |
Supervisor | -0.030 |
mble | -0.030 |
aina | -0.029 |
Token | poll |
Token ID | 3278 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
angles | +0.241 |
Supervisor | +0.222 |
ortment | +0.221 |
aina | +0.220 |
mble | +0.219 |
Neg logit contributions | |
ield | -0.198 |
vine | -0.197 |
het | -0.192 |
794 | -0.191 |
PF | -0.190 |
Token | ster |
Token ID | 1706 |
Feature activation | +7.17e-01 |
Pos logit contributions | |
angles | +0.480 |
Supervisor | +0.441 |
ortment | +0.438 |
aina | +0.436 |
mble | +0.435 |
Neg logit contributions | |
ield | -0.402 |
vine | -0.400 |
het | -0.392 |
794 | -0.389 |
PF | -0.387 |
Token | Stan |
Token ID | 7299 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
angles | +1.480 |
Supervisor | +1.363 |
ortment | +1.340 |
aina | +1.335 |
mble | +1.327 |
Neg logit contributions | |
ield | -1.581 |
vine | -1.560 |
het | -1.536 |
794 | -1.526 |
PF | -1.523 |
Token | Greenberg |
Token ID | 46702 |
Feature activation | -8.66e-02 |
Pos logit contributions | |
angles | +2.783 |
Supervisor | +2.476 |
aina | +2.418 |
mble | +2.407 |
ortment | +2.402 |
Neg logit contributions | |
ield | -5.680 |
794 | -5.592 |
PF | -5.589 |
vine | -5.572 |
het | -5.556 |
Token | told |
Token ID | 1297 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
ield | +0.184 |
vine | +0.183 |
het | +0.180 |
794 | +0.179 |
PF | +0.177 |
Neg logit contributions | |
angles | -0.187 |
ortment | -0.170 |
Supervisor | -0.170 |
mble | -0.169 |
aina | -0.168 |
Token | left |
Token ID | 1364 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
angles | +0.053 |
Supervisor | +0.048 |
ortment | +0.048 |
mble | +0.048 |
aina | +0.048 |
Neg logit contributions | |
ield | -0.044 |
vine | -0.044 |
het | -0.043 |
794 | -0.042 |
PF | -0.042 |
Token | - |
Token ID | 12 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
vine | +1.274 |
ield | +1.268 |
het | +1.242 |
794 | +1.232 |
PF | +1.218 |
Neg logit contributions | |
angles | -1.698 |
ortment | -1.585 |
mble | -1.571 |
Supervisor | -1.566 |
aina | -1.552 |
Token | leaning |
Token ID | 25909 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
angles | +0.348 |
Supervisor | +0.309 |
ortment | +0.308 |
aina | +0.306 |
mble | +0.305 |
Neg logit contributions | |
ield | -0.444 |
vine | -0.441 |
het | -0.434 |
794 | -0.432 |
PF | -0.431 |
Token | the |
Token ID | 262 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ield | +0.413 |
vine | +0.411 |
het | +0.403 |
794 | +0.400 |
PF | +0.397 |
Neg logit contributions | |
angles | -0.518 |
ortment | -0.475 |
Supervisor | -0.473 |
mble | -0.473 |
aina | -0.470 |
Token | Soon |
Token ID | 15894 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
ield | +0.313 |
vine | +0.312 |
het | +0.306 |
794 | +0.303 |
PF | +0.301 |
Neg logit contributions | |
angles | -0.399 |
ortment | -0.367 |
Supervisor | -0.366 |
mble | -0.364 |
aina | -0.363 |
Token | ers |
Token ID | 364 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
angles | +0.412 |
ortment | +0.372 |
Supervisor | +0.371 |
mble | +0.370 |
aina | +0.367 |
Neg logit contributions | |
ield | -0.478 |
vine | -0.475 |
het | -0.468 |
794 | -0.464 |
PF | -0.462 |
Token | ' |
Token ID | 6 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
angles | +0.595 |
Supervisor | +0.546 |
ortment | +0.546 |
mble | +0.543 |
aina | +0.542 |
Neg logit contributions | |
ield | -0.446 |
vine | -0.444 |
het | -0.435 |
794 | -0.431 |
PF | -0.428 |
Token | own |
Token ID | 898 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
angles | +1.473 |
Supervisor | +1.357 |
ortment | +1.350 |
aina | +1.340 |
mble | +1.338 |
Neg logit contributions | |
ield | -1.131 |
vine | -1.130 |
het | -1.104 |
794 | -1.096 |
PF | -1.086 |
Token | inept |
Token ID | 43891 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
angles | +0.355 |
Supervisor | +0.326 |
ortment | +0.326 |
mble | +0.324 |
aina | +0.323 |
Neg logit contributions | |
ield | -0.278 |
vine | -0.277 |
het | -0.271 |
794 | -0.269 |
PF | -0.267 |
Token | offense |
Token ID | 6907 |
Feature activation | -8.01e-01 |
Pos logit contributions | |
angles | +4.375 |
Supervisor | +3.928 |
ortment | +3.918 |
mble | +3.901 |
aina | +3.881 |
Neg logit contributions | |
vine | -4.155 |
het | -4.060 |
ield | -4.060 |
PF | -4.047 |
794 | -4.044 |
Token | and |
Token ID | 290 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
ield | +1.501 |
vine | +1.495 |
het | +1.473 |
794 | +1.458 |
PF | +1.446 |
Neg logit contributions | |
angles | -1.916 |
ortment | -1.759 |
Supervisor | -1.751 |
mble | -1.750 |
aina | -1.743 |
Token | OU |
Token ID | 47070 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ield | +0.146 |
vine | +0.146 |
het | +0.143 |
794 | +0.142 |
PF | +0.141 |
Neg logit contributions | |
angles | -0.182 |
ortment | -0.167 |
Supervisor | -0.166 |
mble | -0.166 |
aina | -0.165 |
Token | 's |
Token ID | 338 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
ield | +2.333 |
vine | +2.307 |
het | +2.266 |
PF | +2.261 |
794 | +2.249 |
Neg logit contributions | |
angles | -2.100 |
ortment | -1.912 |
mble | -1.904 |
Supervisor | -1.885 |
aina | -1.881 |
Token | front |
Token ID | 2166 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
ield | +0.122 |
vine | +0.121 |
het | +0.119 |
794 | +0.118 |
PF | +0.117 |
Neg logit contributions | |
angles | -0.164 |
ortment | -0.151 |
Supervisor | -0.150 |
mble | -0.150 |
aina | -0.149 |
Token | satisfaction |
Token ID | 14676 |
Feature activation | -5.73e-01 |
Pos logit contributions | |
ield | +1.373 |
vine | +1.363 |
het | +1.340 |
794 | +1.328 |
PF | +1.319 |
Neg logit contributions | |
angles | -1.631 |
ortment | -1.492 |
Supervisor | -1.488 |
mble | -1.487 |
aina | -1.476 |
Token | . |
Token ID | 13 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
ield | +1.128 |
vine | +1.119 |
het | +1.101 |
794 | +1.092 |
PF | +1.084 |
Neg logit contributions | |
angles | -1.327 |
ortment | -1.213 |
Supervisor | -1.211 |
mble | -1.207 |
aina | -1.202 |
Token |
|
Token ID | 198 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
angles | +0.525 |
Supervisor | +0.478 |
ortment | +0.476 |
mble | +0.472 |
aina | +0.471 |
Neg logit contributions | |
ield | -0.507 |
vine | -0.505 |
het | -0.496 |
794 | -0.492 |
PF | -0.490 |
Token |
|
Token ID | 198 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
angles | +1.011 |
mble | +0.923 |
ortment | +0.922 |
Supervisor | +0.918 |
aina | +0.911 |
Neg logit contributions | |
ield | -0.957 |
vine | -0.954 |
het | -0.937 |
794 | -0.932 |
PF | -0.924 |
Token | Final |
Token ID | 19006 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
angles | +0.708 |
Supervisor | +0.644 |
ortment | +0.642 |
aina | +0.639 |
mble | +0.637 |
Neg logit contributions | |
ield | -0.618 |
vine | -0.615 |
het | -0.605 |
794 | -0.597 |
PF | -0.592 |
Token | Take |
Token ID | 7214 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
angles | +0.067 |
ortment | +0.060 |
Supervisor | +0.060 |
mble | +0.060 |
aina | +0.060 |
Neg logit contributions | |
ield | -0.079 |
vine | -0.079 |
het | -0.078 |
794 | -0.077 |
PF | -0.077 |
Token | away |
Token ID | 8272 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
angles | +4.143 |
Supervisor | +3.785 |
ortment | +3.741 |
aina | +3.685 |
Buff | +3.637 |
Neg logit contributions | |
vine | -4.339 |
ield | -4.333 |
het | -4.243 |
794 | -4.213 |
ournals | -4.200 |
Token | : |
Token ID | 25 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
angles | +1.640 |
Supervisor | +1.492 |
ortment | +1.486 |
mble | +1.472 |
aina | +1.471 |
Neg logit contributions | |
ield | -1.604 |
vine | -1.595 |
het | -1.569 |
794 | -1.556 |
PF | -1.547 |
Token |
|
Token ID | 198 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
angles | +0.804 |
Supervisor | +0.739 |
ortment | +0.734 |
aina | +0.727 |
mble | +0.727 |
Neg logit contributions | |
ield | -0.682 |
vine | -0.680 |
het | -0.666 |
794 | -0.660 |
PF | -0.657 |
Token |
|
Token ID | 198 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
ield | +0.026 |
vine | +0.026 |
794 | +0.025 |
het | +0.025 |
PF | +0.025 |
Neg logit contributions | |
angles | -0.029 |
mble | -0.027 |
ortment | -0.027 |
aina | -0.026 |
Supervisor | -0.026 |
Token | In |
Token ID | 818 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
ield | +0.214 |
vine | +0.213 |
het | +0.209 |
794 | +0.207 |
PF | +0.205 |
Neg logit contributions | |
angles | -0.289 |
Supervisor | -0.265 |
ortment | -0.265 |
mble | -0.263 |
aina | -0.263 |
Token |
|
Token ID | 198 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
ield | +0.275 |
vine | +0.273 |
794 | +0.268 |
het | +0.268 |
PF | +0.265 |
Neg logit contributions | |
angles | -0.306 |
mble | -0.286 |
ortment | -0.282 |
aina | -0.279 |
Supervisor | -0.279 |
Token | Because |
Token ID | 8128 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
ield | +0.320 |
vine | +0.318 |
het | +0.312 |
794 | +0.309 |
PF | +0.307 |
Neg logit contributions | |
angles | -0.405 |
ortment | -0.372 |
Supervisor | -0.372 |
mble | -0.370 |
aina | -0.368 |
Token | system |
Token ID | 1080 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
vine | +1.432 |
ield | +1.429 |
het | +1.402 |
794 | +1.397 |
Finch | +1.389 |
Neg logit contributions | |
angles | -1.733 |
ortment | -1.594 |
aina | -1.593 |
mble | -1.584 |
Supervisor | -1.566 |
Token | () |
Token ID | 3419 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
ield | +0.312 |
vine | +0.310 |
het | +0.304 |
794 | +0.303 |
PF | +0.300 |
Neg logit contributions | |
angles | -0.437 |
ortment | -0.405 |
Supervisor | -0.402 |
mble | -0.401 |
aina | -0.400 |
Token | only |
Token ID | 691 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
angles | +0.505 |
ortment | +0.457 |
Supervisor | +0.456 |
mble | +0.455 |
aina | +0.453 |
Neg logit contributions | |
ield | -0.543 |
vine | -0.540 |
het | -0.533 |
794 | -0.529 |
PF | -0.526 |
Token | takes |
Token ID | 2753 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
angles | +0.521 |
ortment | +0.471 |
Supervisor | +0.469 |
mble | +0.468 |
aina | +0.465 |
Neg logit contributions | |
ield | -0.597 |
vine | -0.592 |
het | -0.585 |
794 | -0.582 |
PF | -0.577 |
Token | one |
Token ID | 530 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
angles | +4.546 |
Supervisor | +4.218 |
Buff | +4.064 |
aina | +4.024 |
ortment | +4.011 |
Neg logit contributions | |
vine | -3.992 |
ield | -3.875 |
het | -3.861 |
ournals | -3.826 |
794 | -3.772 |
Token | argument |
Token ID | 4578 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
angles | +2.551 |
Supervisor | +2.377 |
ortment | +2.344 |
mble | +2.328 |
aina | +2.324 |
Neg logit contributions | |
vine | -1.851 |
ield | -1.835 |
het | -1.793 |
794 | -1.780 |
PF | -1.767 |
Token | , |
Token ID | 11 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
angles | +0.300 |
ortment | +0.273 |
Supervisor | +0.273 |
mble | +0.271 |
aina | +0.270 |
Neg logit contributions | |
ield | -0.289 |
vine | -0.288 |
het | -0.283 |
794 | -0.281 |
PF | -0.279 |
Token | building |
Token ID | 2615 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
ield | +0.218 |
vine | +0.217 |
het | +0.214 |
794 | +0.212 |
PF | +0.211 |
Neg logit contributions | |
angles | -0.253 |
ortment | -0.231 |
Supervisor | -0.230 |
mble | -0.230 |
aina | -0.229 |
Token | the |
Token ID | 262 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
ield | +0.610 |
vine | +0.607 |
het | +0.596 |
794 | +0.592 |
PF | +0.588 |
Neg logit contributions | |
angles | -0.752 |
ortment | -0.690 |
Supervisor | -0.688 |
mble | -0.685 |
aina | -0.684 |
Token | available |
Token ID | 1695 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
angles | +2.229 |
Supervisor | +2.043 |
ortment | +2.034 |
aina | +2.019 |
mble | +2.017 |
Neg logit contributions | |
ield | -2.021 |
vine | -2.004 |
het | -1.964 |
794 | -1.948 |
PF | -1.943 |
Token | at |
Token ID | 379 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
angles | +0.416 |
ortment | +0.381 |
Supervisor | +0.380 |
mble | +0.379 |
aina | +0.376 |
Neg logit contributions | |
ield | -0.353 |
vine | -0.351 |
het | -0.345 |
794 | -0.343 |
PF | -0.340 |
Token | the |
Token ID | 262 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
angles | +0.146 |
ortment | +0.134 |
Supervisor | +0.134 |
mble | +0.133 |
aina | +0.132 |
Neg logit contributions | |
ield | -0.124 |
vine | -0.123 |
het | -0.121 |
794 | -0.120 |
PF | -0.119 |
Token | gallery |
Token ID | 15604 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
angles | +1.067 |
Supervisor | +0.979 |
ortment | +0.977 |
mble | +0.973 |
aina | +0.969 |
Neg logit contributions | |
ield | -0.871 |
vine | -0.864 |
het | -0.848 |
794 | -0.840 |
PF | -0.837 |
Token | .) |
Token ID | 2014 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
angles | +0.427 |
Supervisor | +0.392 |
ortment | +0.391 |
mble | +0.389 |
aina | +0.388 |
Neg logit contributions | |
ield | -0.346 |
vine | -0.345 |
het | -0.338 |
794 | -0.335 |
PF | -0.333 |
Token | ■ |
Token ID | 34252 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
angles | +0.744 |
Supervisor | +0.676 |
ortment | +0.675 |
mble | +0.670 |
aina | +0.668 |
Neg logit contributions | |
ield | -0.748 |
vine | -0.745 |
het | -0.732 |
794 | -0.726 |
PF | -0.722 |
Token | Event |
Token ID | 8558 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
angles | +4.042 |
Supervisor | +3.746 |
ortment | +3.680 |
aina | +3.609 |
Buff | +3.604 |
Neg logit contributions | |
ield | -4.417 |
vine | -4.373 |
het | -4.317 |
ournals | -4.280 |
drip | -4.253 |
Token | Qu |
Token ID | 2264 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
angles | +1.277 |
Supervisor | +1.149 |
ortment | +1.142 |
mble | +1.133 |
aina | +1.122 |
Neg logit contributions | |
ield | -1.697 |
vine | -1.692 |
het | -1.668 |
794 | -1.656 |
PF | -1.646 |
Token | ests |
Token ID | 3558 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
angles | +1.348 |
Supervisor | +1.266 |
ortment | +1.261 |
mble | +1.261 |
aina | +1.253 |
Neg logit contributions | |
ield | -0.572 |
vine | -0.560 |
het | -0.549 |
794 | -0.538 |
PF | -0.530 |
Token | and |
Token ID | 290 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
angles | +0.954 |
ortment | +0.865 |
Supervisor | +0.864 |
mble | +0.859 |
aina | +0.856 |
Neg logit contributions | |
ield | -0.968 |
vine | -0.964 |
het | -0.948 |
794 | -0.941 |
PF | -0.935 |
Token | Special |
Token ID | 6093 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
angles | +1.054 |
Supervisor | +0.962 |
ortment | +0.962 |
mble | +0.956 |
aina | +0.952 |
Neg logit contributions | |
ield | -0.946 |
vine | -0.941 |
het | -0.924 |
794 | -0.917 |
PF | -0.911 |
Token | ability |
Token ID | 2694 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
angles | +0.208 |
ortment | +0.191 |
Supervisor | +0.191 |
mble | +0.189 |
aina | +0.189 |
Neg logit contributions | |
ield | -0.166 |
vine | -0.166 |
het | -0.162 |
794 | -0.161 |
PF | -0.160 |
Token | to |
Token ID | 284 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
angles | +1.245 |
ortment | +1.133 |
Supervisor | +1.128 |
mble | +1.124 |
aina | +1.118 |
Neg logit contributions | |
ield | -1.174 |
vine | -1.170 |
het | -1.148 |
794 | -1.142 |
PF | -1.134 |
Token | hold |
Token ID | 1745 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
angles | +0.291 |
Supervisor | +0.266 |
ortment | +0.266 |
mble | +0.265 |
aina | +0.263 |
Neg logit contributions | |
ield | -0.259 |
vine | -0.257 |
het | -0.253 |
794 | -0.251 |
PF | -0.249 |
Token | equally |
Token ID | 8603 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
angles | +0.853 |
ortment | +0.781 |
Supervisor | +0.781 |
mble | +0.777 |
aina | +0.774 |
Neg logit contributions | |
ield | -0.690 |
vine | -0.687 |
het | -0.674 |
794 | -0.669 |
PF | -0.664 |
Token | in |
Token ID | 287 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
angles | +0.519 |
ortment | +0.475 |
Supervisor | +0.475 |
mble | +0.472 |
aina | +0.471 |
Neg logit contributions | |
ield | -0.427 |
vine | -0.425 |
het | -0.417 |
794 | -0.414 |
PF | -0.411 |
Token | mind |
Token ID | 2000 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
angles | +1.249 |
ortment | +1.139 |
Supervisor | +1.139 |
mble | +1.133 |
aina | +1.128 |
Neg logit contributions | |
vine | -0.954 |
ield | -0.952 |
794 | -0.923 |
het | -0.921 |
PF | -0.918 |
Token | totally |
Token ID | 6635 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
angles | +4.572 |
Supervisor | +4.108 |
ortment | +4.097 |
aina | +4.051 |
mble | +4.048 |
Neg logit contributions | |
vine | -3.988 |
ield | -3.902 |
794 | -3.815 |
PF | -3.790 |
ournals | -3.756 |
Token | opposing |
Token ID | 12330 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
angles | +0.037 |
Supervisor | +0.034 |
ortment | +0.034 |
mble | +0.034 |
aina | +0.034 |
Neg logit contributions | |
ield | -0.030 |
vine | -0.029 |
het | -0.029 |
794 | -0.029 |
PF | -0.028 |
Token | ideas |
Token ID | 4213 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
ield | +0.138 |
vine | +0.137 |
het | +0.136 |
794 | +0.134 |
PF | +0.133 |
Neg logit contributions | |
angles | -0.133 |
ortment | -0.121 |
mble | -0.121 |
Supervisor | -0.120 |
aina | -0.120 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
ield | +0.105 |
vine | +0.104 |
het | +0.103 |
794 | +0.102 |
PF | +0.101 |
Neg logit contributions | |
angles | -0.117 |
mble | -0.107 |
ortment | -0.107 |
Supervisor | -0.107 |
aina | -0.106 |
Token | is |
Token ID | 318 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ield | +0.260 |
vine | +0.259 |
het | +0.254 |
794 | +0.252 |
PF | +0.251 |
Neg logit contributions | |
angles | -0.302 |
ortment | -0.276 |
Supervisor | -0.276 |
mble | -0.274 |
aina | -0.273 |
Token | room |
Token ID | 2119 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
ield | +1.755 |
vine | +1.746 |
het | +1.719 |
794 | +1.691 |
PF | +1.682 |
Neg logit contributions | |
angles | -2.225 |
ortment | -2.042 |
Supervisor | -2.029 |
mble | -2.015 |
aina | -2.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
ield | +0.320 |
vine | +0.318 |
het | +0.312 |
794 | +0.310 |
PF | +0.308 |
Neg logit contributions | |
angles | -0.368 |
ortment | -0.336 |
Supervisor | -0.336 |
mble | -0.334 |
aina | -0.333 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ield | +0.354 |
vine | +0.353 |
het | +0.347 |
794 | +0.344 |
PF | +0.342 |
Neg logit contributions | |
angles | -0.371 |
ortment | -0.337 |
Supervisor | -0.336 |
mble | -0.335 |
aina | -0.333 |
Token |
|
Token ID | 198 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
ield | +0.263 |
vine | +0.263 |
794 | +0.258 |
het | +0.257 |
PF | +0.254 |
Neg logit contributions | |
angles | -0.302 |
mble | -0.284 |
ortment | -0.278 |
aina | -0.274 |
unlaw | -0.274 |
Token | Rob |
Token ID | 14350 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ield | +0.445 |
vine | +0.442 |
het | +0.435 |
794 | +0.430 |
PF | +0.426 |
Neg logit contributions | |
angles | -0.552 |
Supervisor | -0.506 |
ortment | -0.505 |
mble | -0.502 |
aina | -0.502 |
Token | Harley |
Token ID | 27440 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
ield | +0.255 |
vine | +0.255 |
het | +0.250 |
794 | +0.247 |
PF | +0.245 |
Neg logit contributions | |
angles | -0.289 |
ortment | -0.263 |
Supervisor | -0.263 |
mble | -0.262 |
aina | -0.260 |
Token | is |
Token ID | 318 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
ield | +5.706 |
vine | +5.445 |
het | +5.417 |
794 | +5.406 |
Finch | +5.358 |
Neg logit contributions | |
angles | -5.638 |
Supervisor | -5.177 |
mble | -5.127 |
ortment | -5.097 |
atown | -5.086 |
Token | at |
Token ID | 379 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
ield | +0.486 |
vine | +0.484 |
het | +0.475 |
794 | +0.471 |
PF | +0.468 |
Neg logit contributions | |
angles | -0.585 |
ortment | -0.535 |
Supervisor | -0.534 |
mble | -0.532 |
aina | -0.530 |
Token | blind |
Token ID | 7770 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
ield | +1.611 |
vine | +1.604 |
het | +1.575 |
794 | +1.560 |
PF | +1.549 |
Neg logit contributions | |
angles | -2.188 |
ortment | -2.012 |
Supervisor | -2.010 |
mble | -2.000 |
aina | -1.991 |
Token | - |
Token ID | 12 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
angles | +1.087 |
ortment | +0.951 |
Supervisor | +0.945 |
aina | +0.940 |
mble | +0.927 |
Neg logit contributions | |
ield | -1.599 |
vine | -1.598 |
794 | -1.579 |
het | -1.573 |
PF | -1.562 |
Token | side |
Token ID | 1589 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
vine | +1.782 |
ield | +1.781 |
het | +1.742 |
PF | +1.727 |
794 | +1.722 |
Neg logit contributions | |
angles | -1.189 |
mble | -1.099 |
Supervisor | -1.095 |
ortment | -1.087 |
aina | -1.064 |
Token | . |
Token ID | 13 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
angles | +0.247 |
ortment | +0.227 |
mble | +0.226 |
Supervisor | +0.224 |
aina | +0.224 |
Neg logit contributions | |
ield | -0.198 |
vine | -0.195 |
het | -0.193 |
794 | -0.191 |
PF | -0.190 |
Token | Congress |
Token ID | 3162 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
ield | +0.519 |
vine | +0.515 |
het | +0.507 |
794 | +0.503 |
PF | +0.502 |
Neg logit contributions | |
angles | -0.572 |
ortment | -0.520 |
Supervisor | -0.518 |
mble | -0.518 |
aina | -0.515 |
Token | to |
Token ID | 284 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
ield | +2.969 |
vine | +2.885 |
het | +2.858 |
794 | +2.816 |
PF | +2.801 |
Neg logit contributions | |
angles | -3.877 |
Supervisor | -3.561 |
ortment | -3.546 |
aina | -3.512 |
mble | -3.507 |
Token | broaden |
Token ID | 44870 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
angles | +0.055 |
ortment | +0.050 |
mble | +0.050 |
Supervisor | +0.050 |
aina | +0.050 |
Neg logit contributions | |
ield | -0.046 |
vine | -0.045 |
het | -0.044 |
794 | -0.044 |
PF | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
ield | +0.297 |
vine | +0.294 |
het | +0.289 |
794 | +0.287 |
PF | +0.286 |
Neg logit contributions | |
angles | -0.351 |
ortment | -0.320 |
Supervisor | -0.318 |
mble | -0.317 |
aina | -0.316 |
Token | leng |
Token ID | 40038 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
ield | +0.454 |
vine | +0.450 |
het | +0.439 |
794 | +0.436 |
PF | +0.432 |
Neg logit contributions | |
angles | -0.680 |
Supervisor | -0.626 |
mble | -0.626 |
ortment | -0.626 |
aina | -0.623 |
Token | then |
Token ID | 8524 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
het | +2.936 |
vine | +2.904 |
ield | +2.879 |
794 | +2.575 |
ordes | +2.533 |
Neg logit contributions | |
angles | -8.336 |
ortment | -7.983 |
unlaw | -7.977 |
Supervisor | -7.945 |
mble | -7.830 |
Token | the |
Token ID | 262 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
ield | +2.179 |
vine | +2.130 |
het | +2.112 |
PF | +2.109 |
794 | +2.093 |
Neg logit contributions | |
angles | -2.706 |
ortment | -2.466 |
mble | -2.447 |
aina | -2.427 |
Supervisor | -2.417 |
Token | tax |
Token ID | 1687 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
ield | +0.628 |
vine | +0.621 |
het | +0.612 |
794 | +0.608 |
PF | +0.604 |
Neg logit contributions | |
angles | -0.752 |
ortment | -0.687 |
Supervisor | -0.684 |
mble | -0.684 |
aina | -0.680 |
Token | credits |
Token ID | 10824 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
ield | +0.091 |
vine | +0.091 |
het | +0.089 |
794 | +0.089 |
PF | +0.088 |
Neg logit contributions | |
angles | -0.106 |
Supervisor | -0.097 |
ortment | -0.097 |
mble | -0.096 |
aina | -0.096 |
Token | for |
Token ID | 329 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
ield | +0.246 |
vine | +0.245 |
het | +0.241 |
794 | +0.239 |
PF | +0.238 |
Neg logit contributions | |
angles | -0.278 |
ortment | -0.253 |
mble | -0.252 |
Supervisor | -0.251 |
aina | -0.250 |
Token | address |
Token ID | 2209 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
ield | +2.555 |
vine | +2.510 |
het | +2.485 |
794 | +2.442 |
drip | +2.432 |
Neg logit contributions | |
angles | -3.002 |
mble | -2.804 |
unlaw | -2.758 |
aina | -2.723 |
arnaev | -2.712 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ield | +0.862 |
vine | +0.861 |
het | +0.837 |
794 | +0.833 |
PF | +0.820 |
Neg logit contributions | |
angles | -1.148 |
mble | -1.075 |
ortment | -1.060 |
aina | -1.049 |
unlaw | -1.047 |
Token | Please |
Token ID | 4222 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
ield | +2.337 |
vine | +2.295 |
Finch | +2.266 |
794 | +2.253 |
het | +2.253 |
Neg logit contributions | |
angles | -2.891 |
mble | -2.773 |
ortment | -2.668 |
aina | -2.666 |
arnaev | -2.633 |
Token | re |
Token ID | 302 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
ield | +2.289 |
het | +2.134 |
vine | +2.127 |
794 | +2.100 |
Finch | +2.018 |
Neg logit contributions | |
angles | -5.248 |
mble | -5.052 |
aina | -4.919 |
ortment | -4.913 |
arnaev | -4.865 |
Token | - |
Token ID | 12 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ield | +0.342 |
vine | +0.337 |
het | +0.331 |
794 | +0.330 |
PF | +0.322 |
Neg logit contributions | |
angles | -0.516 |
mble | -0.487 |
ortment | -0.483 |
Supervisor | -0.479 |
unlaw | -0.478 |
Token | enter |
Token ID | 9255 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ield | +1.835 |
vine | +1.784 |
794 | +1.733 |
het | +1.733 |
PF | +1.665 |
Neg logit contributions | |
angles | -3.362 |
unlaw | -3.204 |
ortment | -3.178 |
mble | -3.168 |
Supervisor | -3.160 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
vine | +3.704 |
ield | +3.588 |
794 | +3.512 |
het | +3.457 |
PF | +3.330 |
Neg logit contributions | |
mble | -6.423 |
mathemat | -6.395 |
angles | -6.379 |
unlaw | -6.336 |
ortment | -6.050 |
Token | You |
Token ID | 921 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
vine | +1.608 |
ield | +1.604 |
794 | +1.552 |
het | +1.545 |
Finch | +1.541 |
Neg logit contributions | |
angles | -2.889 |
mble | -2.765 |
unlaw | -2.698 |
aina | -2.689 |
ortment | -2.684 |
Token | must |
Token ID | 1276 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ield | +3.256 |
vine | +3.211 |
het | +3.199 |
Finch | +3.146 |
drip | +3.136 |
Neg logit contributions | |
angles | -1.823 |
unlaw | -1.653 |
mble | -1.642 |
ortment | -1.639 |
aina | -1.596 |
Token | select |
Token ID | 2922 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ield | +1.971 |
vine | +1.893 |
het | +1.864 |
drip | +1.840 |
PF | +1.826 |
Neg logit contributions | |
angles | -2.851 |
mble | -2.706 |
aina | -2.665 |
unlaw | -2.651 |
ortment | -2.640 |
Token | a |
Token ID | 257 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ield | +2.206 |
vine | +2.190 |
het | +2.092 |
PF | +2.047 |
794 | +2.047 |
Neg logit contributions | |
angles | -3.842 |
mble | -3.628 |
unlaw | -3.583 |
ortment | -3.580 |
aina | -3.529 |
Token | . |
Token ID | 13 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
ield | +0.011 |
vine | +0.011 |
het | +0.011 |
794 | +0.011 |
PF | +0.011 |
Neg logit contributions | |
angles | -0.013 |
Supervisor | -0.012 |
ortment | -0.012 |
mble | -0.012 |
aina | -0.011 |
Token | As |
Token ID | 1081 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
angles | +0.347 |
Supervisor | +0.315 |
ortment | +0.314 |
mble | +0.311 |
aina | +0.311 |
Neg logit contributions | |
ield | -0.358 |
vine | -0.356 |
het | -0.350 |
794 | -0.347 |
PF | -0.345 |
Token | Thomas |
Token ID | 5658 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
angles | +0.912 |
Supervisor | +0.840 |
ortment | +0.837 |
mble | +0.831 |
aina | +0.830 |
Neg logit contributions | |
ield | -0.704 |
vine | -0.700 |
het | -0.684 |
794 | -0.681 |
PF | -0.674 |
Token | C |
Token ID | 327 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
angles | +1.337 |
Supervisor | +1.217 |
ortment | +1.209 |
aina | +1.201 |
mble | +1.196 |
Neg logit contributions | |
ield | -1.327 |
vine | -1.325 |
het | -1.298 |
794 | -1.294 |
PF | -1.287 |
Token | . |
Token ID | 13 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
angles | +0.089 |
Supervisor | +0.081 |
ortment | +0.081 |
mble | +0.081 |
aina | +0.080 |
Neg logit contributions | |
ield | -0.076 |
vine | -0.075 |
het | -0.074 |
794 | -0.073 |
PF | -0.073 |
Token | Tig |
Token ID | 14189 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ield | +0.907 |
vine | +0.906 |
het | +0.891 |
794 | +0.884 |
PF | +0.877 |
Neg logit contributions | |
angles | -0.881 |
mble | -0.800 |
ortment | -0.798 |
Supervisor | -0.793 |
aina | -0.789 |
Token | he |
Token ID | 258 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
het | +4.456 |
ield | +4.423 |
vine | +4.345 |
ordes | +4.092 |
Finch | +3.966 |
Neg logit contributions | |
angles | -5.686 |
mble | -5.372 |
unlaw | -5.345 |
Supervisor | -5.315 |
ortment | -5.299 |
Token | , |
Token ID | 11 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
ield | +0.653 |
vine | +0.648 |
het | +0.639 |
794 | +0.630 |
PF | +0.627 |
Neg logit contributions | |
angles | -0.631 |
ortment | -0.573 |
Supervisor | -0.570 |
mble | -0.569 |
aina | -0.566 |
Token | a |
Token ID | 257 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
ield | +0.061 |
vine | +0.061 |
het | +0.060 |
794 | +0.059 |
PF | +0.059 |
Neg logit contributions | |
angles | -0.076 |
Supervisor | -0.070 |
ortment | -0.069 |
mble | -0.069 |
aina | -0.069 |
Token | U |
Token ID | 471 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
ield | +0.493 |
vine | +0.490 |
het | +0.482 |
794 | +0.479 |
PF | +0.476 |
Neg logit contributions | |
angles | -0.551 |
ortment | -0.502 |
Supervisor | -0.499 |
mble | -0.499 |
aina | -0.497 |
Token | . |
Token ID | 13 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
ield | +1.226 |
vine | +1.221 |
794 | +1.197 |
het | +1.193 |
PF | +1.183 |
Neg logit contributions | |
angles | -1.651 |
mble | -1.562 |
ortment | -1.536 |
unlaw | -1.533 |
Supervisor | -1.511 |
Token |
|
Token ID | 198 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
ield | +0.606 |
vine | +0.603 |
het | +0.592 |
794 | +0.588 |
PF | +0.585 |
Neg logit contributions | |
angles | -0.635 |
ortment | -0.577 |
Supervisor | -0.576 |
mble | -0.574 |
aina | -0.571 |
Token |
|
Token ID | 198 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
vine | +0.798 |
ield | +0.798 |
794 | +0.782 |
het | +0.777 |
PF | +0.769 |
Neg logit contributions | |
angles | -0.958 |
mble | -0.913 |
ortment | -0.886 |
unlaw | -0.885 |
aina | -0.873 |
Token | 6 |
Token ID | 21 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ield | +0.730 |
vine | +0.727 |
het | +0.712 |
794 | +0.708 |
PF | +0.704 |
Neg logit contributions | |
angles | -0.964 |
ortment | -0.886 |
Supervisor | -0.885 |
mble | -0.882 |
aina | -0.877 |
Token | . |
Token ID | 13 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
ield | +0.192 |
vine | +0.191 |
794 | +0.187 |
het | +0.186 |
PF | +0.185 |
Neg logit contributions | |
angles | -0.241 |
mble | -0.224 |
ortment | -0.222 |
aina | -0.220 |
Supervisor | -0.219 |
Token | Rob |
Token ID | 3851 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
angles | +0.584 |
Supervisor | +0.528 |
ortment | +0.525 |
aina | +0.522 |
mble | +0.519 |
Neg logit contributions | |
ield | -0.654 |
vine | -0.653 |
het | -0.641 |
794 | -0.638 |
PF | -0.632 |
Token | Harley |
Token ID | 27440 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ield | +0.417 |
vine | +0.413 |
het | +0.402 |
PF | +0.390 |
794 | +0.389 |
Neg logit contributions | |
angles | -0.866 |
ortment | -0.808 |
Supervisor | -0.805 |
mble | -0.804 |
aina | -0.796 |
Token | ( |
Token ID | 357 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ield | +4.283 |
vine | +4.149 |
PF | +4.100 |
794 | +4.016 |
Finch | +4.008 |
Neg logit contributions | |
angles | -5.760 |
atown | -5.279 |
mble | -5.273 |
ortment | -5.240 |
Supervisor | -5.217 |
Token | 152 |
Token ID | 17827 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
ield | +0.246 |
vine | +0.245 |
het | +0.240 |
794 | +0.239 |
PF | +0.237 |
Neg logit contributions | |
angles | -0.320 |
ortment | -0.294 |
Supervisor | -0.293 |
mble | -0.292 |
aina | -0.291 |
Token | ) |
Token ID | 8 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
ield | +0.420 |
vine | +0.416 |
het | +0.409 |
794 | +0.407 |
PF | +0.403 |
Neg logit contributions | |
angles | -0.621 |
mble | -0.578 |
ortment | -0.577 |
Supervisor | -0.571 |
aina | -0.571 |
Token | ( |
Token ID | 357 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
ield | +1.183 |
vine | +1.178 |
794 | +1.160 |
PF | +1.149 |
het | +1.140 |
Neg logit contributions | |
angles | -1.644 |
mble | -1.550 |
ortment | -1.520 |
unlaw | -1.505 |
aina | -1.501 |
Token | 31 |
Token ID | 3132 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
ield | +0.458 |
vine | +0.456 |
het | +0.447 |
794 | +0.446 |
PF | +0.442 |
Neg logit contributions | |
angles | -0.584 |
ortment | -0.537 |
mble | -0.536 |
Supervisor | -0.536 |
aina | -0.532 |
Token | address |
Token ID | 2209 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
ield | +2.439 |
vine | +2.399 |
het | +2.376 |
794 | +2.341 |
drip | +2.311 |
Neg logit contributions | |
angles | -2.748 |
mble | -2.537 |
unlaw | -2.500 |
aina | -2.477 |
ortment | -2.465 |
Token | . |
Token ID | 13 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ield | +0.906 |
vine | +0.903 |
het | +0.879 |
794 | +0.876 |
PF | +0.859 |
Neg logit contributions | |
angles | -1.206 |
mble | -1.126 |
ortment | -1.112 |
aina | -1.100 |
unlaw | -1.100 |
Token | Please |
Token ID | 4222 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
ield | +2.362 |
vine | +2.317 |
Finch | +2.279 |
794 | +2.271 |
het | +2.270 |
Neg logit contributions | |
angles | -2.927 |
mble | -2.795 |
ortment | -2.685 |
aina | -2.677 |
arnaev | -2.647 |
Token | re |
Token ID | 302 |
Feature activation | -9.83e-02 |
Pos logit contributions | |
ield | +2.100 |
vine | +1.986 |
het | +1.979 |
794 | +1.954 |
Finch | +1.880 |
Neg logit contributions | |
angles | -4.684 |
mble | -4.491 |
ortment | -4.377 |
aina | -4.377 |
arnaev | -4.333 |
Token | - |
Token ID | 12 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ield | +0.169 |
vine | +0.168 |
het | +0.164 |
794 | +0.164 |
PF | +0.160 |
Neg logit contributions | |
angles | -0.252 |
mble | -0.236 |
ortment | -0.235 |
Supervisor | -0.233 |
unlaw | -0.232 |
Token | enter |
Token ID | 9255 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
ield | +1.790 |
vine | +1.745 |
het | +1.694 |
794 | +1.692 |
PF | +1.631 |
Neg logit contributions | |
angles | -3.296 |
unlaw | -3.119 |
ortment | -3.094 |
mble | -3.077 |
Supervisor | -3.072 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
vine | +3.643 |
ield | +3.548 |
794 | +3.460 |
het | +3.422 |
Finch | +3.277 |
Neg logit contributions | |
angles | -6.298 |
mble | -6.281 |
unlaw | -6.225 |
mathemat | -6.223 |
ortment | -5.940 |
Token | You |
Token ID | 921 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ield | +1.651 |
vine | +1.651 |
794 | +1.599 |
het | +1.592 |
Finch | +1.575 |
Neg logit contributions | |
angles | -3.020 |
mble | -2.897 |
unlaw | -2.829 |
aina | -2.811 |
ortment | -2.798 |
Token | must |
Token ID | 1276 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ield | +3.270 |
vine | +3.242 |
het | +3.217 |
Finch | +3.162 |
794 | +3.162 |
Neg logit contributions | |
angles | -1.867 |
unlaw | -1.708 |
mble | -1.695 |
ortment | -1.676 |
aina | -1.644 |
Token | select |
Token ID | 2922 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ield | +1.837 |
vine | +1.761 |
het | +1.731 |
794 | +1.701 |
drip | +1.700 |
Neg logit contributions | |
angles | -2.663 |
mble | -2.525 |
aina | -2.488 |
unlaw | -2.470 |
ortment | -2.466 |
Token | a |
Token ID | 257 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ield | +2.078 |
vine | +2.067 |
het | +1.976 |
794 | +1.936 |
PF | +1.926 |
Neg logit contributions | |
angles | -3.538 |
mble | -3.319 |
unlaw | -3.286 |
ortment | -3.282 |
aina | -3.243 |
Token | ground |
Token ID | 2323 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
ield | +0.207 |
vine | +0.206 |
het | +0.202 |
794 | +0.201 |
PF | +0.199 |
Neg logit contributions | |
angles | -0.238 |
Supervisor | -0.217 |
ortment | -0.217 |
mble | -0.216 |
aina | -0.215 |
Token | . |
Token ID | 13 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
vine | +1.325 |
ield | +1.316 |
het | +1.290 |
794 | +1.276 |
PF | +1.261 |
Neg logit contributions | |
angles | -1.711 |
Supervisor | -1.569 |
ortment | -1.563 |
aina | -1.552 |
mble | -1.549 |
Token |
|
Token ID | 198 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
angles | +0.028 |
Supervisor | +0.026 |
ortment | +0.026 |
mble | +0.025 |
aina | +0.025 |
Neg logit contributions | |
ield | -0.028 |
vine | -0.027 |
het | -0.027 |
794 | -0.027 |
PF | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
angles | +0.260 |
mble | +0.239 |
ortment | +0.238 |
Supervisor | +0.236 |
aina | +0.235 |
Neg logit contributions | |
ield | -0.240 |
vine | -0.240 |
het | -0.235 |
794 | -0.235 |
PF | -0.232 |
Token | On |
Token ID | 2202 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
ield | +0.063 |
vine | +0.063 |
het | +0.062 |
794 | +0.061 |
PF | +0.061 |
Neg logit contributions | |
angles | -0.079 |
ortment | -0.072 |
Supervisor | -0.072 |
mble | -0.071 |
aina | -0.071 |
Token | grass |
Token ID | 8701 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
angles | +0.408 |
Supervisor | +0.374 |
ortment | +0.373 |
mble | +0.371 |
aina | +0.370 |
Neg logit contributions | |
ield | -0.344 |
vine | -0.342 |
het | -0.335 |
794 | -0.333 |
PF | -0.331 |
Token | y |
Token ID | 88 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
vine | +4.580 |
het | +4.532 |
ield | +4.495 |
drip | +4.366 |
794 | +4.339 |
Neg logit contributions | |
angles | -5.296 |
ortment | -4.881 |
Supervisor | -4.854 |
unlaw | -4.822 |
arnaev | -4.819 |
Token | kn |
Token ID | 638 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
ield | +2.964 |
vine | +2.945 |
het | +2.921 |
drip | +2.850 |
Finch | +2.838 |
Neg logit contributions | |
angles | -3.727 |
ortment | -3.478 |
mble | -3.428 |
Supervisor | -3.408 |
aina | -3.385 |
Token | olls |
Token ID | 33421 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
ield | +1.198 |
vine | +1.185 |
het | +1.160 |
794 | +1.143 |
PF | +1.134 |
Neg logit contributions | |
angles | -2.227 |
Supervisor | -2.076 |
ortment | -2.072 |
mble | -2.064 |
aina | -2.052 |
Token | fathers |
Token ID | 17150 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
ield | +1.759 |
vine | +1.750 |
het | +1.729 |
794 | +1.700 |
PF | +1.681 |
Neg logit contributions | |
angles | -2.016 |
ortment | -1.839 |
mble | -1.832 |
Supervisor | -1.826 |
aina | -1.823 |
Token | and |
Token ID | 290 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
ield | +0.833 |
vine | +0.831 |
het | +0.817 |
794 | +0.806 |
PF | +0.800 |
Neg logit contributions | |
angles | -0.943 |
Supervisor | -0.862 |
ortment | -0.861 |
mble | -0.857 |
aina | -0.850 |
Token | flower |
Token ID | 25547 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ield | +0.717 |
vine | +0.714 |
het | +0.697 |
794 | +0.688 |
PF | +0.682 |
Neg logit contributions | |
angles | -1.287 |
ortment | -1.196 |
Supervisor | -1.196 |
mble | -1.191 |
aina | -1.183 |
Token | Seeds |
Token ID | 41833 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
vine | +1.727 |
ield | +1.720 |
het | +1.660 |
794 | +1.633 |
Finch | +1.626 |
Neg logit contributions | |
angles | -3.235 |
mble | -3.065 |
ortment | -3.023 |
Supervisor | -3.018 |
aina | -2.979 |
Token | , |
Token ID | 837 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
ield | +1.761 |
vine | +1.755 |
het | +1.723 |
794 | +1.698 |
PF | +1.696 |
Neg logit contributions | |
angles | -2.031 |
mble | -1.873 |
Supervisor | -1.857 |
ortment | -1.844 |
aina | -1.837 |
Token | Brown |
Token ID | 4373 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
angles | +0.132 |
Supervisor | +0.120 |
ortment | +0.120 |
mble | +0.119 |
aina | +0.119 |
Neg logit contributions | |
ield | -0.123 |
vine | -0.123 |
het | -0.121 |
794 | -0.120 |
PF | -0.119 |
Token | S |
Token ID | 311 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
ield | +1.094 |
vine | +1.089 |
het | +1.067 |
794 | +1.053 |
PF | +1.051 |
Neg logit contributions | |
angles | -1.335 |
ortment | -1.226 |
mble | -1.226 |
Supervisor | -1.223 |
aina | -1.206 |
Token | esame |
Token ID | 34038 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
ield | +1.672 |
vine | +1.642 |
het | +1.620 |
794 | +1.593 |
PF | +1.570 |
Neg logit contributions | |
angles | -2.211 |
Supervisor | -2.046 |
mble | -2.040 |
ortment | -2.025 |
aina | -2.003 |
Token | Seeds |
Token ID | 41833 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
vine | +3.126 |
ield | +3.070 |
Finch | +2.937 |
het | +2.930 |
794 | +2.856 |
Neg logit contributions | |
angles | -6.659 |
mble | -6.512 |
ortment | -6.345 |
unlaw | -6.300 |
arnaev | -6.294 |
Token | , |
Token ID | 837 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
ield | +1.545 |
vine | +1.538 |
het | +1.509 |
794 | +1.489 |
PF | +1.487 |
Neg logit contributions | |
angles | -1.761 |
mble | -1.617 |
Supervisor | -1.610 |
ortment | -1.604 |
aina | -1.595 |
Token | Black |
Token ID | 2619 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
angles | +0.330 |
Supervisor | +0.300 |
ortment | +0.300 |
mble | +0.298 |
aina | +0.297 |
Neg logit contributions | |
ield | -0.307 |
vine | -0.305 |
het | -0.300 |
794 | -0.298 |
PF | -0.296 |
Token | S |
Token ID | 311 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
ield | +0.135 |
vine | +0.134 |
het | +0.131 |
794 | +0.130 |
PF | +0.129 |
Neg logit contributions | |
angles | -0.172 |
ortment | -0.158 |
Supervisor | -0.158 |
mble | -0.157 |
aina | -0.157 |
Token | esame |
Token ID | 34038 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
ield | +1.423 |
vine | +1.389 |
het | +1.368 |
794 | +1.334 |
PF | +1.310 |
Neg logit contributions | |
angles | -2.767 |
Supervisor | -2.599 |
mble | -2.590 |
ortment | -2.576 |
aina | -2.548 |
Token | internal |
Token ID | 5387 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
ield | +0.019 |
vine | +0.019 |
het | +0.019 |
794 | +0.019 |
PF | +0.019 |
Neg logit contributions | |
angles | -0.027 |
Supervisor | -0.025 |
ortment | -0.025 |
mble | -0.024 |
aina | -0.024 |
Token | conflict |
Token ID | 5358 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
ield | +0.827 |
vine | +0.823 |
het | +0.806 |
794 | +0.801 |
PF | +0.792 |
Neg logit contributions | |
angles | -1.211 |
ortment | -1.114 |
Supervisor | -1.114 |
mble | -1.111 |
aina | -1.105 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
angles | +0.104 |
Supervisor | +0.096 |
ortment | +0.095 |
aina | +0.094 |
mble | +0.094 |
Neg logit contributions | |
ield | -0.094 |
vine | -0.094 |
het | -0.092 |
794 | -0.091 |
PF | -0.091 |
Token | For |
Token ID | 1114 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
ield | +0.296 |
vine | +0.295 |
het | +0.290 |
794 | +0.288 |
PF | +0.286 |
Neg logit contributions | |
angles | -0.316 |
mble | -0.288 |
ortment | -0.287 |
Supervisor | -0.286 |
aina | -0.285 |
Token | two |
Token ID | 734 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
ield | +0.421 |
vine | +0.419 |
het | +0.410 |
794 | +0.406 |
PF | +0.403 |
Neg logit contributions | |
angles | -0.633 |
Supervisor | -0.584 |
ortment | -0.583 |
mble | -0.580 |
aina | -0.579 |
Token | long |
Token ID | 890 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ield | +1.219 |
vine | +1.201 |
het | +1.189 |
794 | +1.173 |
PF | +1.150 |
Neg logit contributions | |
angles | -2.463 |
ortment | -2.307 |
mble | -2.304 |
Supervisor | -2.284 |
aina | -2.283 |
Token | years |
Token ID | 812 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
ield | +2.827 |
het | +2.740 |
highs | +2.694 |
794 | +2.692 |
drip | +2.690 |
Neg logit contributions | |
angles | -6.800 |
mble | -6.560 |
arnaev | -6.471 |
ortment | -6.447 |
aina | -6.400 |
Token | I |
Token ID | 314 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
ield | +1.105 |
vine | +1.103 |
het | +1.089 |
794 | +1.080 |
PF | +1.070 |
Neg logit contributions | |
angles | -1.202 |
ortment | -1.100 |
mble | -1.090 |
Supervisor | -1.083 |
aina | -1.082 |
Token | went |
Token ID | 1816 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
ield | +0.674 |
vine | +0.667 |
het | +0.663 |
794 | +0.654 |
PF | +0.648 |
Neg logit contributions | |
angles | -0.673 |
ortment | -0.612 |
mble | -0.611 |
Supervisor | -0.610 |
aina | -0.601 |
Token | through |
Token ID | 832 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
angles | +0.915 |
Supervisor | +0.834 |
ortment | +0.832 |
mble | +0.827 |
aina | +0.825 |
Neg logit contributions | |
ield | -0.822 |
vine | -0.818 |
het | -0.803 |
794 | -0.797 |
PF | -0.792 |
Token | weekly |
Token ID | 10273 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
ield | +0.806 |
vine | +0.803 |
het | +0.788 |
794 | +0.781 |
PF | +0.774 |
Neg logit contributions | |
angles | -1.104 |
ortment | -1.016 |
Supervisor | -1.014 |
mble | -1.012 |
aina | -1.007 |
Token | twice |
Token ID | 5403 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
angles | +0.069 |
ortment | +0.063 |
Supervisor | +0.063 |
mble | +0.062 |
aina | +0.062 |
Neg logit contributions | |
ield | -0.058 |
vine | -0.058 |
het | -0.057 |
794 | -0.057 |
PF | -0.056 |
Token | as |
Token ID | 355 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ield | +0.945 |
vine | +0.944 |
het | +0.919 |
794 | +0.915 |
PF | +0.907 |
Neg logit contributions | |
angles | -1.359 |
ortment | -1.258 |
mble | -1.253 |
Supervisor | -1.248 |
aina | -1.246 |
Token | wide |
Token ID | 3094 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
angles | +0.451 |
Supervisor | +0.415 |
ortment | +0.414 |
mble | +0.411 |
aina | +0.411 |
Neg logit contributions | |
ield | -0.330 |
vine | -0.329 |
het | -0.322 |
794 | -0.319 |
PF | -0.317 |
Token | , |
Token ID | 11 |
Feature activation | -7.16e-02 |
Pos logit contributions | |
angles | +1.078 |
Supervisor | +0.991 |
ortment | +0.985 |
aina | +0.980 |
mble | +0.979 |
Neg logit contributions | |
ield | -0.850 |
vine | -0.847 |
het | -0.833 |
794 | -0.824 |
PF | -0.817 |
Token | and |
Token ID | 290 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
ield | +0.142 |
vine | +0.141 |
het | +0.139 |
794 | +0.138 |
PF | +0.137 |
Neg logit contributions | |
angles | -0.165 |
Supervisor | -0.151 |
ortment | -0.150 |
mble | -0.149 |
aina | -0.149 |
Token | leng |
Token ID | 40038 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
angles | +0.181 |
ortment | +0.166 |
Supervisor | +0.165 |
mble | +0.165 |
aina | +0.164 |
Neg logit contributions | |
ield | -0.152 |
vine | -0.151 |
het | -0.149 |
794 | -0.147 |
PF | -0.146 |
Token | then |
Token ID | 8524 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
ield | +2.617 |
het | +2.585 |
vine | +2.535 |
794 | +2.347 |
ordes | +2.286 |
Neg logit contributions | |
angles | -7.173 |
ortment | -6.859 |
Supervisor | -6.808 |
unlaw | -6.795 |
mble | -6.789 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
ield | +1.752 |
vine | +1.719 |
het | +1.706 |
794 | +1.680 |
PF | +1.674 |
Neg logit contributions | |
angles | -2.267 |
ortment | -2.096 |
mble | -2.083 |
Supervisor | -2.077 |
aina | -2.075 |
Token | so |
Token ID | 523 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
angles | +0.285 |
ortment | +0.260 |
Supervisor | +0.260 |
mble | +0.259 |
aina | +0.258 |
Neg logit contributions | |
ield | -0.239 |
vine | -0.238 |
het | -0.234 |
794 | -0.232 |
PF | -0.230 |
Token | it |
Token ID | 340 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
ield | +0.460 |
vine | +0.458 |
het | +0.449 |
794 | +0.446 |
PF | +0.442 |
Neg logit contributions | |
angles | -0.550 |
ortment | -0.504 |
mble | -0.501 |
Supervisor | -0.501 |
aina | -0.499 |
Token | meets |
Token ID | 11185 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
angles | +0.634 |
ortment | +0.569 |
Supervisor | +0.568 |
mble | +0.566 |
aina | +0.563 |
Neg logit contributions | |
ield | -0.775 |
vine | -0.772 |
het | -0.761 |
794 | -0.756 |
PF | -0.751 |
Token | address |
Token ID | 2209 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
ield | +2.503 |
vine | +2.458 |
het | +2.445 |
794 | +2.408 |
drip | +2.374 |
Neg logit contributions | |
angles | -2.803 |
mble | -2.575 |
unlaw | -2.538 |
aina | -2.529 |
ortment | -2.512 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ield | +0.803 |
vine | +0.800 |
het | +0.779 |
794 | +0.775 |
PF | +0.761 |
Neg logit contributions | |
angles | -1.061 |
mble | -0.986 |
ortment | -0.977 |
aina | -0.967 |
unlaw | -0.964 |
Token | Please |
Token ID | 4222 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
ield | +2.186 |
vine | +2.147 |
Finch | +2.111 |
het | +2.104 |
794 | +2.096 |
Neg logit contributions | |
angles | -2.678 |
mble | -2.541 |
ortment | -2.450 |
aina | -2.445 |
arnaev | -2.409 |
Token | re |
Token ID | 302 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
ield | +2.165 |
vine | +2.037 |
het | +2.033 |
794 | +1.995 |
Finch | +1.944 |
Neg logit contributions | |
angles | -4.803 |
mble | -4.605 |
ortment | -4.498 |
aina | -4.496 |
atown | -4.428 |
Token | - |
Token ID | 12 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
ield | +0.113 |
vine | +0.111 |
het | +0.109 |
794 | +0.109 |
PF | +0.106 |
Neg logit contributions | |
angles | -0.167 |
mble | -0.157 |
ortment | -0.156 |
Supervisor | -0.155 |
unlaw | -0.154 |
Token | enter |
Token ID | 9255 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
ield | +1.817 |
vine | +1.768 |
het | +1.717 |
794 | +1.716 |
PF | +1.658 |
Neg logit contributions | |
angles | -3.298 |
unlaw | -3.124 |
ortment | -3.106 |
mble | -3.082 |
Supervisor | -3.079 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
vine | +3.628 |
ield | +3.530 |
794 | +3.431 |
het | +3.401 |
Finch | +3.275 |
Neg logit contributions | |
angles | -6.152 |
mble | -6.115 |
unlaw | -6.048 |
mathemat | -6.032 |
ortment | -5.814 |
Token | You |
Token ID | 921 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
vine | +1.608 |
ield | +1.607 |
794 | +1.546 |
het | +1.539 |
Finch | +1.530 |
Neg logit contributions | |
angles | -2.885 |
mble | -2.740 |
unlaw | -2.684 |
aina | -2.680 |
ortment | -2.670 |
Token | must |
Token ID | 1276 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ield | +3.420 |
vine | +3.378 |
het | +3.358 |
Finch | +3.306 |
794 | +3.294 |
Neg logit contributions | |
angles | -1.907 |
unlaw | -1.738 |
mble | -1.716 |
ortment | -1.709 |
aina | -1.670 |
Token | select |
Token ID | 2922 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
ield | +2.091 |
vine | +2.001 |
het | +1.963 |
drip | +1.949 |
794 | +1.924 |
Neg logit contributions | |
angles | -3.068 |
mble | -2.910 |
aina | -2.870 |
unlaw | -2.866 |
ortment | -2.841 |
Token | a |
Token ID | 257 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ield | +2.184 |
vine | +2.178 |
het | +2.074 |
794 | +2.029 |
PF | +2.021 |
Neg logit contributions | |
angles | -3.819 |
mble | -3.603 |
unlaw | -3.566 |
ortment | -3.557 |
aina | -3.498 |
Token | address |
Token ID | 2209 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
ield | +2.534 |
vine | +2.483 |
het | +2.458 |
794 | +2.428 |
drip | +2.403 |
Neg logit contributions | |
angles | -2.909 |
mble | -2.686 |
unlaw | -2.651 |
aina | -2.630 |
ortment | -2.611 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ield | +0.897 |
vine | +0.893 |
het | +0.867 |
794 | +0.863 |
PF | +0.847 |
Neg logit contributions | |
angles | -1.191 |
mble | -1.108 |
ortment | -1.095 |
aina | -1.086 |
unlaw | -1.084 |
Token | Please |
Token ID | 4222 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
ield | +2.297 |
vine | +2.253 |
Finch | +2.230 |
794 | +2.214 |
het | +2.208 |
Neg logit contributions | |
angles | -2.878 |
mble | -2.762 |
aina | -2.636 |
ortment | -2.632 |
arnaev | -2.605 |
Token | re |
Token ID | 302 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
ield | +2.166 |
het | +2.040 |
vine | +2.037 |
794 | +2.003 |
Finch | +1.945 |
Neg logit contributions | |
angles | -4.844 |
mble | -4.655 |
ortment | -4.543 |
aina | -4.537 |
arnaev | -4.470 |
Token | - |
Token ID | 12 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ield | +0.267 |
vine | +0.263 |
het | +0.258 |
794 | +0.257 |
PF | +0.252 |
Neg logit contributions | |
angles | -0.399 |
mble | -0.375 |
ortment | -0.372 |
Supervisor | -0.370 |
unlaw | -0.369 |
Token | enter |
Token ID | 9255 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
ield | +1.764 |
vine | +1.710 |
794 | +1.663 |
het | +1.663 |
PF | +1.602 |
Neg logit contributions | |
angles | -3.188 |
unlaw | -3.022 |
ortment | -3.002 |
mble | -2.991 |
Supervisor | -2.985 |
Token | . |
Token ID | 13 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
vine | +3.613 |
ield | +3.499 |
794 | +3.421 |
het | +3.372 |
PF | +3.250 |
Neg logit contributions | |
angles | -6.103 |
mble | -6.086 |
mathemat | -6.007 |
unlaw | -6.002 |
ortment | -5.760 |
Token | You |
Token ID | 921 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ield | +1.638 |
vine | +1.637 |
794 | +1.580 |
het | +1.569 |
Finch | +1.566 |
Neg logit contributions | |
angles | -2.966 |
mble | -2.842 |
unlaw | -2.771 |
aina | -2.757 |
ortment | -2.740 |
Token | must |
Token ID | 1276 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ield | +3.402 |
vine | +3.361 |
het | +3.342 |
Finch | +3.291 |
794 | +3.278 |
Neg logit contributions | |
angles | -1.904 |
unlaw | -1.730 |
mble | -1.721 |
ortment | -1.707 |
aina | -1.668 |
Token | select |
Token ID | 2922 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ield | +2.061 |
vine | +1.983 |
het | +1.945 |
drip | +1.927 |
794 | +1.905 |
Neg logit contributions | |
angles | -3.026 |
mble | -2.874 |
aina | -2.830 |
unlaw | -2.823 |
ortment | -2.800 |
Token | a |
Token ID | 257 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ield | +2.193 |
vine | +2.187 |
het | +2.078 |
794 | +2.035 |
PF | +2.030 |
Neg logit contributions | |
angles | -3.841 |
mble | -3.633 |
unlaw | -3.588 |
ortment | -3.577 |
aina | -3.522 |
Token | foul |
Token ID | 15626 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
angles | +0.628 |
Supervisor | +0.579 |
ortment | +0.574 |
aina | +0.574 |
mble | +0.571 |
Neg logit contributions | |
ield | -0.485 |
vine | -0.482 |
het | -0.473 |
794 | -0.470 |
PF | -0.464 |
Token | trouble |
Token ID | 5876 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
angles | +1.702 |
ortment | +1.560 |
Supervisor | +1.559 |
mble | +1.548 |
aina | +1.548 |
Neg logit contributions | |
ield | -1.315 |
vine | -1.309 |
het | -1.283 |
794 | -1.275 |
PF | -1.264 |
Token | while |
Token ID | 981 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
ield | +0.682 |
vine | +0.677 |
het | +0.667 |
794 | +0.661 |
PF | +0.657 |
Neg logit contributions | |
angles | -0.739 |
ortment | -0.674 |
Supervisor | -0.672 |
mble | -0.671 |
aina | -0.664 |
Token | defending |
Token ID | 11749 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
ield | +0.055 |
vine | +0.054 |
het | +0.053 |
794 | +0.053 |
PF | +0.053 |
Neg logit contributions | |
angles | -0.064 |
ortment | -0.058 |
mble | -0.058 |
Supervisor | -0.058 |
aina | -0.057 |
Token | Frank |
Token ID | 5278 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
ield | +0.621 |
vine | +0.617 |
het | +0.606 |
794 | +0.600 |
PF | +0.597 |
Neg logit contributions | |
angles | -0.795 |
ortment | -0.731 |
Supervisor | -0.728 |
mble | -0.726 |
aina | -0.723 |
Token | Kam |
Token ID | 12670 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
angles | +0.393 |
Supervisor | +0.356 |
ortment | +0.356 |
aina | +0.354 |
mble | +0.353 |
Neg logit contributions | |
ield | -0.390 |
vine | -0.388 |
het | -0.382 |
794 | -0.380 |
PF | -0.377 |
Token | insky |
Token ID | 19870 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
ield | +5.584 |
vine | +5.497 |
het | +5.407 |
PF | +5.339 |
794 | +5.331 |
Neg logit contributions | |
angles | -4.137 |
Supervisor | -3.818 |
mble | -3.817 |
ortment | -3.807 |
unlaw | -3.743 |
Token | of |
Token ID | 286 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
ield | +1.000 |
vine | +0.987 |
het | +0.977 |
794 | +0.966 |
PF | +0.965 |
Neg logit contributions | |
angles | -1.042 |
ortment | -0.954 |
mble | -0.947 |
Supervisor | -0.945 |
aina | -0.936 |
Token | Wisconsin |
Token ID | 9279 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
angles | +0.268 |
ortment | +0.244 |
Supervisor | +0.244 |
mble | +0.242 |
aina | +0.241 |
Neg logit contributions | |
ield | -0.253 |
vine | -0.252 |
het | -0.248 |
794 | -0.246 |
PF | -0.244 |
Token | .[ |
Token ID | 3693 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
ield | +2.158 |
vine | +2.107 |
het | +2.084 |
PF | +2.071 |
794 | +2.063 |
Neg logit contributions | |
angles | -2.440 |
ortment | -2.261 |
mble | -2.238 |
Supervisor | -2.221 |
aina | -2.193 |
Token | 125 |
Token ID | 11623 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
angles | +1.676 |
ortment | +1.523 |
Supervisor | +1.515 |
aina | +1.508 |
mble | +1.496 |
Neg logit contributions | |
ield | -1.505 |
vine | -1.490 |
het | -1.477 |
794 | -1.470 |
PF | -1.447 |
Token | Recommend |
Token ID | 41248 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
ield | +2.427 |
PF | +2.412 |
vine | +2.409 |
794 | +2.367 |
het | +2.273 |
Neg logit contributions | |
angles | -3.203 |
mble | -3.092 |
ortment | -3.075 |
unlaw | -2.988 |
aina | -2.987 |
Token | Share |
Token ID | 8734 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
ield | +1.568 |
vine | +1.566 |
het | +1.536 |
794 | +1.525 |
PF | +1.514 |
Neg logit contributions | |
angles | -1.706 |
Supervisor | -1.553 |
ortment | -1.550 |
aina | -1.539 |
mble | -1.536 |
Token | this |
Token ID | 428 |
Feature activation | -1.65e-05 |
Pos logit contributions | |
vine | +0.835 |
ield | +0.824 |
het | +0.808 |
794 | +0.805 |
PF | +0.799 |
Neg logit contributions | |
angles | -1.346 |
mble | -1.268 |
ortment | -1.262 |
aina | -1.254 |
Supervisor | -1.244 |
Token | comment |
Token ID | 2912 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
vine | +0.000 |
ield | +0.000 |
het | +0.000 |
PF | +0.000 |
794 | +0.000 |
Neg logit contributions | |
angles | -0.000 |
mble | -0.000 |
ortment | -0.000 |
Supervisor | -0.000 |
aina | -0.000 |
Token | on |
Token ID | 319 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
vine | +0.926 |
ield | +0.919 |
PF | +0.891 |
het | +0.888 |
drip | +0.887 |
Neg logit contributions | |
angles | -0.828 |
mble | -0.773 |
ortment | -0.772 |
unlaw | -0.739 |
Supervisor | -0.739 |
Token | Facebook |
Token ID | 3203 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
vine | +0.441 |
ield | +0.439 |
het | +0.429 |
PF | +0.426 |
794 | +0.426 |
Neg logit contributions | |
angles | -0.557 |
ortment | -0.515 |
mble | -0.513 |
Supervisor | -0.510 |
aina | -0.506 |
Token | Share |
Token ID | 8734 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
vine | +4.010 |
drip | +3.722 |
ield | +3.718 |
794 | +3.641 |
het | +3.637 |
Neg logit contributions | |
angles | -5.812 |
mble | -5.610 |
ortment | -5.480 |
arnaev | -5.455 |
bilt | -5.453 |
Token | this |
Token ID | 428 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
vine | +1.569 |
ield | +1.541 |
794 | +1.495 |
het | +1.485 |
PF | +1.483 |
Neg logit contributions | |
angles | -2.381 |
mble | -2.246 |
ortment | -2.217 |
Supervisor | -2.202 |
aina | -2.195 |
Token | comment |
Token ID | 2912 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
vine | +0.825 |
ield | +0.819 |
het | +0.796 |
794 | +0.793 |
PF | +0.790 |
Neg logit contributions | |
angles | -1.194 |
mble | -1.114 |
ortment | -1.108 |
Supervisor | -1.090 |
aina | -1.081 |
Token | on |
Token ID | 319 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
vine | +1.133 |
ield | +1.127 |
794 | +1.097 |
het | +1.093 |
PF | +1.088 |
Neg logit contributions | |
angles | -1.028 |
mble | -0.976 |
unlaw | -0.956 |
ortment | -0.953 |
aina | -0.922 |
Token | Twitter |
Token ID | 3009 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
vine | +0.585 |
ield | +0.576 |
het | +0.559 |
PF | +0.555 |
794 | +0.554 |
Neg logit contributions | |
angles | -1.279 |
ortment | -1.200 |
mble | -1.199 |
aina | -1.185 |
Supervisor | -1.184 |
Token | be |
Token ID | 307 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
ield | +0.791 |
vine | +0.785 |
het | +0.773 |
794 | +0.767 |
PF | +0.762 |
Neg logit contributions | |
angles | -0.795 |
ortment | -0.722 |
Supervisor | -0.722 |
mble | -0.720 |
aina | -0.713 |
Token | located |
Token ID | 5140 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
ield | +0.139 |
vine | +0.138 |
het | +0.136 |
794 | +0.135 |
PF | +0.134 |
Neg logit contributions | |
angles | -0.150 |
ortment | -0.136 |
Supervisor | -0.136 |
mble | -0.136 |
aina | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ield | +0.396 |
vine | +0.395 |
het | +0.388 |
794 | +0.385 |
PF | +0.382 |
Neg logit contributions | |
angles | -0.441 |
ortment | -0.402 |
Supervisor | -0.401 |
mble | -0.400 |
aina | -0.398 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
ield | +0.293 |
vine | +0.293 |
het | +0.292 |
794 | +0.287 |
PF | +0.286 |
Neg logit contributions | |
angles | -0.319 |
mble | -0.293 |
ortment | -0.289 |
aina | -0.287 |
Supervisor | -0.285 |
Token | The |
Token ID | 464 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
angles | +1.328 |
ortment | +1.209 |
Supervisor | +1.209 |
aina | +1.205 |
mble | +1.200 |
Neg logit contributions | |
ield | -1.059 |
vine | -1.050 |
het | -1.030 |
794 | -1.020 |
Finch | -1.008 |
Token | thorn |
Token ID | 38999 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
ield | +0.864 |
vine | +0.861 |
het | +0.846 |
794 | +0.839 |
PF | +0.834 |
Neg logit contributions | |
angles | -0.947 |
ortment | -0.862 |
Supervisor | -0.856 |
mble | -0.855 |
aina | -0.852 |
Token | y |
Token ID | 88 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
vine | +4.458 |
ield | +4.359 |
het | +4.212 |
Finch | +4.145 |
794 | +4.137 |
Neg logit contributions | |
angles | -5.247 |
mble | -4.874 |
Supervisor | -4.828 |
ortment | -4.802 |
aina | -4.747 |
Token | debates |
Token ID | 15389 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
ield | +1.204 |
vine | +1.190 |
het | +1.165 |
794 | +1.151 |
PF | +1.133 |
Neg logit contributions | |
angles | -1.944 |
ortment | -1.808 |
mble | -1.794 |
Supervisor | -1.793 |
aina | -1.789 |
Token | around |
Token ID | 1088 |
Feature activation | +1.71e-03 |
Pos logit contributions | |
ield | +1.349 |
vine | +1.339 |
het | +1.316 |
794 | +1.304 |
drip | +1.288 |
Neg logit contributions | |
angles | -1.476 |
ortment | -1.355 |
mble | -1.343 |
Supervisor | -1.338 |
aina | -1.333 |
Token | encryption |
Token ID | 15835 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
angles | +0.004 |
ortment | +0.004 |
Supervisor | +0.004 |
mble | +0.004 |
aina | +0.004 |
Neg logit contributions | |
ield | -0.003 |
vine | -0.003 |
het | -0.003 |
794 | -0.003 |
PF | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
ield | +0.837 |
vine | +0.834 |
het | +0.819 |
794 | +0.812 |
PF | +0.805 |
Neg logit contributions | |
angles | -0.919 |
ortment | -0.839 |
Supervisor | -0.836 |
mble | -0.835 |
aina | -0.830 |
Token | India |
Token ID | 3794 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
angles | +1.280 |
mble | +1.198 |
ortment | +1.198 |
Supervisor | +1.189 |
aina | +1.183 |
Neg logit contributions | |
ield | -0.492 |
vine | -0.487 |
het | -0.474 |
794 | -0.464 |
PF | -0.461 |
Token | photo |
Token ID | 4590 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
ield | +0.736 |
vine | +0.731 |
het | +0.717 |
794 | +0.712 |
PF | +0.707 |
Neg logit contributions | |
angles | -0.982 |
ortment | -0.902 |
mble | -0.900 |
Supervisor | -0.894 |
aina | -0.893 |
Token | ) |
Token ID | 8 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ield | +1.205 |
vine | +1.200 |
het | +1.175 |
794 | +1.160 |
PF | +1.157 |
Neg logit contributions | |
angles | -1.706 |
ortment | -1.567 |
mble | -1.558 |
Supervisor | -1.558 |
aina | -1.550 |
Token |
|
Token ID | 198 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
ield | +0.297 |
vine | +0.296 |
het | +0.291 |
794 | +0.289 |
PF | +0.287 |
Neg logit contributions | |
angles | -0.315 |
ortment | -0.287 |
Supervisor | -0.286 |
mble | -0.285 |
aina | -0.284 |
Token |
|
Token ID | 198 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
ield | +1.096 |
vine | +1.090 |
794 | +1.078 |
PF | +1.073 |
het | +1.071 |
Neg logit contributions | |
angles | -1.270 |
mble | -1.203 |
ortment | -1.175 |
unlaw | -1.157 |
aina | -1.155 |
Token | Del |
Token ID | 13856 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
ield | +1.174 |
vine | +1.166 |
het | +1.144 |
794 | +1.139 |
PF | +1.133 |
Neg logit contributions | |
angles | -1.495 |
ortment | -1.377 |
Supervisor | -1.371 |
mble | -1.371 |
aina | -1.360 |
Token | hi |
Token ID | 5303 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
ield | +4.078 |
het | +4.047 |
vine | +3.939 |
794 | +3.903 |
PF | +3.825 |
Neg logit contributions | |
angles | -5.309 |
mble | -5.074 |
ortment | -5.023 |
Supervisor | -4.973 |
unlaw | -4.870 |
Token | Law |
Token ID | 3854 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
ield | +3.196 |
vine | +3.124 |
het | +3.099 |
794 | +3.074 |
PF | +3.049 |
Neg logit contributions | |
angles | -3.324 |
ortment | -3.027 |
mble | -3.002 |
aina | -2.980 |
atown | -2.894 |
Token | Minister |
Token ID | 4139 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ield | +1.395 |
vine | +1.377 |
het | +1.364 |
794 | +1.342 |
PF | +1.342 |
Neg logit contributions | |
angles | -1.493 |
ortment | -1.362 |
mble | -1.338 |
aina | -1.337 |
unlaw | -1.313 |
Token | J |
Token ID | 449 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
ield | +2.400 |
vine | +2.379 |
het | +2.363 |
Finch | +2.322 |
794 | +2.311 |
Neg logit contributions | |
angles | -3.162 |
mble | -2.928 |
ortment | -2.891 |
atown | -2.861 |
aina | -2.845 |
Token | it |
Token ID | 270 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
ield | +0.798 |
het | +0.792 |
vine | +0.788 |
PF | +0.771 |
794 | +0.769 |
Neg logit contributions | |
angles | -0.691 |
Supervisor | -0.630 |
mble | -0.628 |
ortment | -0.626 |
unlaw | -0.612 |
Token | mix |
Token ID | 5022 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
ield | +0.108 |
vine | +0.107 |
het | +0.104 |
794 | +0.103 |
PF | +0.100 |
Neg logit contributions | |
angles | -0.273 |
mble | -0.257 |
Supervisor | -0.256 |
ortment | -0.255 |
aina | -0.255 |
Token | well |
Token ID | 880 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
angles | +1.607 |
ortment | +1.495 |
Supervisor | +1.493 |
mble | +1.491 |
aina | +1.484 |
Neg logit contributions | |
ield | -0.780 |
vine | -0.774 |
het | -0.757 |
794 | -0.745 |
PF | -0.739 |
Token | until |
Token ID | 1566 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
ield | +1.284 |
vine | +1.270 |
het | +1.221 |
794 | +1.206 |
drip | +1.197 |
Neg logit contributions | |
angles | -2.223 |
mble | -2.120 |
ortment | -2.082 |
unlaw | -2.080 |
Supervisor | -2.075 |
Token | the |
Token ID | 262 |
Feature activation | -9.70e-02 |
Pos logit contributions | |
angles | +0.571 |
ortment | +0.523 |
Supervisor | +0.523 |
mble | +0.522 |
aina | +0.519 |
Neg logit contributions | |
ield | -0.446 |
vine | -0.445 |
het | -0.436 |
794 | -0.432 |
PF | -0.429 |
Token | mixture |
Token ID | 11710 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
ield | +0.154 |
vine | +0.154 |
het | +0.150 |
794 | +0.149 |
PF | +0.148 |
Neg logit contributions | |
angles | -0.260 |
Supervisor | -0.241 |
mble | -0.240 |
aina | -0.240 |
ortment | -0.240 |
Token | feels |
Token ID | 5300 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
angles | +0.707 |
ortment | +0.649 |
Supervisor | +0.648 |
mble | +0.647 |
aina | +0.642 |
Neg logit contributions | |
ield | -0.593 |
vine | -0.591 |
het | -0.579 |
794 | -0.574 |
PF | -0.570 |
Token | like |
Token ID | 588 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
vine | +2.544 |
drip | +2.536 |
ield | +2.509 |
het | +2.473 |
794 | +2.463 |
Neg logit contributions | |
angles | -6.779 |
mble | -6.450 |
ortment | -6.289 |
unlaw | -6.264 |
arnaev | -6.242 |
Token | it |
Token ID | 340 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
ield | +0.427 |
vine | +0.425 |
het | +0.417 |
794 | +0.415 |
drip | +0.411 |
Neg logit contributions | |
angles | -0.515 |
mble | -0.475 |
ortment | -0.469 |
aina | -0.465 |
Supervisor | -0.465 |
Token | will |
Token ID | 481 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
angles | +0.842 |
mble | +0.756 |
ortment | +0.755 |
Supervisor | +0.748 |
aina | +0.744 |
Neg logit contributions | |
ield | -1.165 |
vine | -1.161 |
het | -1.143 |
794 | -1.134 |
PF | -1.126 |
Token | hold |
Token ID | 1745 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ield | +1.535 |
vine | +1.502 |
het | +1.475 |
drip | +1.466 |
794 | +1.453 |
Neg logit contributions | |
angles | -1.115 |
mble | -1.025 |
unlaw | -1.015 |
ortment | -1.011 |
Supervisor | -1.005 |
Token | together |
Token ID | 1978 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
ield | +0.401 |
vine | +0.398 |
het | +0.391 |
794 | +0.389 |
PF | +0.388 |
Neg logit contributions | |
angles | -0.314 |
ortment | -0.282 |
mble | -0.281 |
Supervisor | -0.280 |
aina | -0.279 |
Token | mix |
Token ID | 5022 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
ield | +0.158 |
vine | +0.157 |
het | +0.154 |
794 | +0.152 |
PF | +0.151 |
Neg logit contributions | |
angles | -0.219 |
ortment | -0.201 |
Supervisor | -0.201 |
mble | -0.200 |
aina | -0.200 |
Token | well |
Token ID | 880 |
Feature activation | -9.01e-01 |
Pos logit contributions | |
angles | +1.424 |
Supervisor | +1.317 |
ortment | +1.313 |
mble | +1.311 |
aina | +1.301 |
Neg logit contributions | |
ield | -0.942 |
vine | -0.937 |
het | -0.912 |
794 | -0.909 |
PF | -0.901 |
Token | until |
Token ID | 1566 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
ield | +1.721 |
vine | +1.703 |
het | +1.679 |
794 | +1.666 |
PF | +1.649 |
Neg logit contributions | |
angles | -2.128 |
ortment | -1.964 |
Supervisor | -1.946 |
aina | -1.943 |
mble | -1.940 |
Token | the |
Token ID | 262 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
angles | +0.442 |
Supervisor | +0.408 |
ortment | +0.408 |
mble | +0.405 |
aina | +0.404 |
Neg logit contributions | |
ield | -0.294 |
vine | -0.291 |
het | -0.286 |
794 | -0.283 |
PF | -0.281 |
Token | mixture |
Token ID | 11710 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
ield | +0.239 |
vine | +0.238 |
het | +0.233 |
794 | +0.231 |
PF | +0.228 |
Neg logit contributions | |
angles | -0.365 |
ortment | -0.336 |
Supervisor | -0.336 |
mble | -0.335 |
aina | -0.334 |
Token | feels |
Token ID | 5300 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
angles | +0.808 |
Supervisor | +0.725 |
ortment | +0.723 |
mble | +0.715 |
aina | +0.715 |
Neg logit contributions | |
ield | -0.927 |
vine | -0.925 |
het | -0.911 |
794 | -0.905 |
PF | -0.899 |
Token | like |
Token ID | 588 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
ield | +3.367 |
vine | +3.304 |
drip | +3.300 |
het | +3.213 |
794 | +3.194 |
Neg logit contributions | |
angles | -5.945 |
mble | -5.556 |
aina | -5.498 |
Supervisor | -5.470 |
ortment | -5.458 |
Token | it |
Token ID | 340 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
ield | +0.487 |
vine | +0.485 |
het | +0.476 |
794 | +0.472 |
PF | +0.468 |
Neg logit contributions | |
angles | -0.623 |
ortment | -0.572 |
Supervisor | -0.571 |
mble | -0.568 |
aina | -0.567 |
Token | will |
Token ID | 481 |
Feature activation | -6.86e-01 |
Pos logit contributions | |
angles | +0.926 |
ortment | +0.822 |
Supervisor | +0.822 |
aina | +0.816 |
mble | +0.810 |
Neg logit contributions | |
ield | -1.155 |
vine | -1.153 |
het | -1.134 |
794 | -1.127 |
PF | -1.121 |
Token | hold |
Token ID | 1745 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
ield | +1.374 |
vine | +1.364 |
het | +1.338 |
794 | +1.330 |
PF | +1.318 |
Neg logit contributions | |
angles | -1.564 |
Supervisor | -1.430 |
ortment | -1.428 |
mble | -1.424 |
aina | -1.420 |
Token | together |
Token ID | 1978 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
ield | +0.368 |
vine | +0.367 |
het | +0.361 |
794 | +0.359 |
PF | +0.357 |
Neg logit contributions | |
angles | -0.310 |
ortment | -0.278 |
Supervisor | -0.278 |
mble | -0.276 |
aina | -0.275 |
Token |
|
Token ID | 198 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
angles | +0.076 |
Supervisor | +0.069 |
ortment | +0.069 |
mble | +0.068 |
aina | +0.068 |
Neg logit contributions | |
ield | -0.074 |
vine | -0.073 |
het | -0.072 |
794 | -0.071 |
PF | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
angles | +0.700 |
mble | +0.643 |
ortment | +0.640 |
Supervisor | +0.636 |
aina | +0.632 |
Neg logit contributions | |
ield | -0.650 |
vine | -0.648 |
het | -0.636 |
794 | -0.634 |
PF | -0.628 |
Token | " |
Token ID | 1 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
angles | +0.490 |
ortment | +0.442 |
Supervisor | +0.442 |
aina | +0.439 |
mble | +0.437 |
Neg logit contributions | |
ield | -0.489 |
vine | -0.486 |
het | -0.478 |
794 | -0.473 |
Finch | -0.470 |
Token | It |
Token ID | 1026 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
ield | +0.493 |
vine | +0.491 |
het | +0.481 |
794 | +0.477 |
PF | +0.473 |
Neg logit contributions | |
angles | -0.611 |
Supervisor | -0.559 |
ortment | -0.559 |
mble | -0.555 |
aina | -0.554 |
Token | 's |
Token ID | 338 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
angles | +0.603 |
Supervisor | +0.541 |
ortment | +0.540 |
mble | +0.535 |
aina | +0.535 |
Neg logit contributions | |
ield | -0.738 |
vine | -0.736 |
het | -0.723 |
794 | -0.720 |
PF | -0.716 |
Token | grass |
Token ID | 8701 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
angles | +0.468 |
Supervisor | +0.431 |
ortment | +0.429 |
mble | +0.427 |
aina | +0.426 |
Neg logit contributions | |
ield | -0.351 |
vine | -0.349 |
het | -0.342 |
794 | -0.339 |
PF | -0.337 |
Token | - |
Token ID | 12 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
ield | +3.611 |
vine | +3.601 |
het | +3.597 |
drip | +3.503 |
Finch | +3.434 |
Neg logit contributions | |
angles | -5.621 |
ortment | -5.247 |
arnaev | -5.200 |
mble | -5.187 |
Supervisor | -5.187 |
Token | roots |
Token ID | 19150 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
ield | +1.324 |
vine | +1.320 |
het | +1.295 |
794 | +1.282 |
PF | +1.272 |
Neg logit contributions | |
angles | -1.664 |
Supervisor | -1.527 |
ortment | -1.526 |
mble | -1.517 |
aina | -1.512 |
Token | ," |
Token ID | 553 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
vine | +3.538 |
ield | +3.515 |
het | +3.479 |
drip | +3.410 |
Finch | +3.385 |
Neg logit contributions | |
angles | -4.576 |
ortment | -4.173 |
mble | -4.157 |
aina | -4.126 |
arnaev | -4.108 |
Token | he |
Token ID | 339 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
ield | +0.976 |
vine | +0.973 |
het | +0.956 |
794 | +0.945 |
PF | +0.934 |
Neg logit contributions | |
angles | -1.358 |
ortment | -1.255 |
mble | -1.249 |
Supervisor | -1.241 |
aina | -1.239 |
Token | says |
Token ID | 1139 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
angles | +0.480 |
Supervisor | +0.439 |
ortment | +0.438 |
mble | +0.435 |
aina | +0.435 |
Neg logit contributions | |
ield | -0.412 |
vine | -0.410 |
het | -0.402 |
794 | -0.400 |
PF | -0.397 |
Token | reading |
Token ID | 3555 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
angles | +0.390 |
mble | +0.353 |
ortment | +0.350 |
aina | +0.344 |
Supervisor | +0.342 |
Neg logit contributions | |
vine | -0.473 |
ield | -0.468 |
het | -0.463 |
794 | -0.461 |
drip | -0.456 |
Token | the |
Token ID | 262 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
vine | +1.000 |
ield | +0.987 |
Finch | +0.984 |
794 | +0.973 |
het | +0.971 |
Neg logit contributions | |
angles | -1.619 |
mble | -1.541 |
ortment | -1.504 |
arnaev | -1.489 |
aina | -1.485 |
Token | main |
Token ID | 1388 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
ield | +0.236 |
vine | +0.235 |
het | +0.228 |
794 | +0.224 |
Finch | +0.224 |
Neg logit contributions | |
angles | -0.533 |
mble | -0.498 |
ortment | -0.495 |
Supervisor | -0.494 |
aina | -0.492 |
Token | story |
Token ID | 1621 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
angles | +0.216 |
Supervisor | +0.201 |
ortment | +0.200 |
mble | +0.200 |
aina | +0.199 |
Neg logit contributions | |
vine | -0.119 |
ield | -0.118 |
het | -0.115 |
794 | -0.114 |
PF | -0.113 |
Token | Please |
Token ID | 4222 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
ield | +1.303 |
vine | +1.298 |
het | +1.266 |
794 | +1.264 |
PF | +1.260 |
Neg logit contributions | |
angles | -1.439 |
mble | -1.312 |
ortment | -1.304 |
aina | -1.300 |
Supervisor | -1.287 |
Token | verify |
Token ID | 11767 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
ield | +1.789 |
vine | +1.769 |
het | +1.762 |
794 | +1.725 |
drip | +1.695 |
Neg logit contributions | |
angles | -2.257 |
Supervisor | -2.083 |
mble | -2.083 |
aina | -2.040 |
atown | -2.027 |
Token | you |
Token ID | 345 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ield | +3.396 |
vine | +3.394 |
het | +3.281 |
794 | +3.198 |
drip | +3.116 |
Neg logit contributions | |
angles | -5.891 |
mble | -5.502 |
unlaw | -5.489 |
ortment | -5.316 |
bilt | -5.284 |
Token | 're |
Token ID | 821 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
ield | +2.185 |
vine | +2.091 |
het | +2.060 |
drip | +1.983 |
794 | +1.970 |
Neg logit contributions | |
angles | -4.728 |
unlaw | -4.399 |
mble | -4.395 |
Supervisor | -4.311 |
atown | -4.308 |
Token | not |
Token ID | 407 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ield | +1.633 |
vine | +1.620 |
het | +1.619 |
Finch | +1.554 |
794 | +1.547 |
Neg logit contributions | |
angles | -2.200 |
ortment | -1.988 |
mble | -1.987 |
unlaw | -1.961 |
atown | -1.958 |
Token | a |
Token ID | 257 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
ield | +1.634 |
vine | +1.608 |
het | +1.578 |
PF | +1.532 |
794 | +1.532 |
Neg logit contributions | |
angles | -2.859 |
ortment | -2.661 |
mble | -2.657 |
unlaw | -2.592 |
aina | -2.590 |
Token | robot |
Token ID | 9379 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
ield | +1.024 |
het | +1.013 |
vine | +1.002 |
drip | +0.987 |
794 | +0.984 |
Neg logit contributions | |
angles | -2.153 |
ortment | -1.992 |
mble | -1.980 |
aina | -1.960 |
Supervisor | -1.945 |
Token | over |
Token ID | 625 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
ield | +0.297 |
vine | +0.296 |
het | +0.290 |
794 | +0.288 |
PF | +0.287 |
Neg logit contributions | |
angles | -0.384 |
ortment | -0.351 |
mble | -0.349 |
aina | -0.347 |
Supervisor | -0.346 |
Token | 15 |
Token ID | 1315 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
angles | +0.250 |
Supervisor | +0.228 |
ortment | +0.228 |
mble | +0.227 |
aina | +0.226 |
Neg logit contributions | |
ield | -0.218 |
vine | -0.217 |
het | -0.213 |
794 | -0.212 |
PF | -0.210 |
Token | insurance |
Token ID | 5096 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
ield | +0.136 |
vine | +0.135 |
het | +0.133 |
794 | +0.132 |
PF | +0.131 |
Neg logit contributions | |
angles | -0.165 |
ortment | -0.150 |
Supervisor | -0.150 |
mble | -0.150 |
aina | -0.149 |
Token | agencies |
Token ID | 5942 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
ield | +0.910 |
vine | +0.906 |
het | +0.885 |
794 | +0.883 |
PF | +0.879 |
Neg logit contributions | |
angles | -1.004 |
ortment | -0.912 |
aina | -0.907 |
mble | -0.905 |
Supervisor | -0.904 |
Token | that |
Token ID | 326 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
angles | +1.444 |
Supervisor | +1.317 |
ortment | +1.315 |
mble | +1.305 |
aina | +1.303 |
Neg logit contributions | |
ield | -1.338 |
vine | -1.332 |
het | -1.308 |
794 | -1.298 |
PF | -1.290 |
Token | hire |
Token ID | 11078 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
angles | +0.229 |
ortment | +0.208 |
mble | +0.207 |
Supervisor | +0.207 |
aina | +0.206 |
Neg logit contributions | |
ield | -0.211 |
vine | -0.210 |
het | -0.206 |
794 | -0.205 |
PF | -0.204 |
Token | V |
Token ID | 569 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
angles | +2.489 |
Supervisor | +2.304 |
ortment | +2.267 |
mble | +2.252 |
aina | +2.247 |
Neg logit contributions | |
ield | -2.433 |
vine | -2.411 |
het | -2.374 |
794 | -2.355 |
PF | -2.333 |
Token | AS |
Token ID | 1921 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
angles | +0.689 |
mble | +0.629 |
ortment | +0.629 |
Supervisor | +0.627 |
aina | +0.621 |
Neg logit contributions | |
ield | -0.644 |
vine | -0.638 |
het | -0.630 |
794 | -0.624 |
PF | -0.622 |
Token | graduates |
Token ID | 19087 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
angles | +0.616 |
ortment | +0.559 |
Supervisor | +0.558 |
mble | +0.556 |
aina | +0.553 |
Neg logit contributions | |
ield | -0.605 |
vine | -0.602 |
het | -0.592 |
794 | -0.588 |
PF | -0.584 |
Token | right |
Token ID | 826 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
angles | +0.945 |
Supervisor | +0.864 |
ortment | +0.863 |
mble | +0.856 |
aina | +0.855 |
Neg logit contributions | |
ield | -0.835 |
vine | -0.831 |
het | -0.815 |
794 | -0.810 |
PF | -0.804 |
Token | out |
Token ID | 503 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
ield | +2.193 |
vine | +2.192 |
het | +2.148 |
PF | +2.134 |
794 | +2.132 |
Neg logit contributions | |
angles | -2.709 |
mble | -2.530 |
ortment | -2.522 |
aina | -2.485 |
Supervisor | -2.483 |
Token | them |
Token ID | 606 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
ield | +0.649 |
vine | +0.647 |
het | +0.636 |
PF | +0.630 |
794 | +0.625 |
Neg logit contributions | |
angles | -0.781 |
mble | -0.736 |
ortment | -0.724 |
unlaw | -0.717 |
aina | -0.708 |
Token | in |
Token ID | 287 |
Feature activation | -8.93e-02 |
Pos logit contributions | |
vine | +0.056 |
ield | +0.056 |
het | +0.055 |
794 | +0.054 |
PF | +0.053 |
Neg logit contributions | |
angles | -0.062 |
mble | -0.058 |
ortment | -0.057 |
aina | -0.056 |
Supervisor | -0.056 |
Token | their |
Token ID | 511 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
vine | +0.179 |
ield | +0.179 |
het | +0.176 |
794 | +0.174 |
PF | +0.171 |
Neg logit contributions | |
angles | -0.203 |
mble | -0.188 |
ortment | -0.187 |
Supervisor | -0.184 |
aina | -0.183 |
Token | language |
Token ID | 3303 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
angles | +0.388 |
ortment | +0.349 |
Supervisor | +0.348 |
mble | +0.348 |
aina | +0.345 |
Neg logit contributions | |
ield | -0.446 |
vine | -0.446 |
het | -0.438 |
794 | -0.435 |
PF | -0.432 |
Token | . |
Token ID | 13 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
angles | +0.550 |
Supervisor | +0.504 |
ortment | +0.503 |
mble | +0.500 |
aina | +0.498 |
Neg logit contributions | |
ield | -0.475 |
vine | -0.473 |
het | -0.463 |
794 | -0.460 |
PF | -0.457 |
Token | - |
Token ID | 532 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
angles | +0.397 |
ortment | +0.359 |
mble | +0.359 |
Supervisor | +0.356 |
aina | +0.355 |
Neg logit contributions | |
vine | -0.413 |
ield | -0.412 |
het | -0.406 |
794 | -0.403 |
PF | -0.399 |
Token | Katherine |
Token ID | 32719 |
Feature activation | +8.26e-01 |
Pos logit contributions | |
angles | +2.110 |
Supervisor | +1.905 |
ortment | +1.887 |
aina | +1.873 |
mble | +1.863 |
Neg logit contributions | |
ield | -2.709 |
vine | -2.683 |
794 | -2.633 |
het | -2.625 |
PF | -2.613 |
Token | d |
Token ID | 288 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
angles | +1.705 |
Supervisor | +1.548 |
ortment | +1.539 |
aina | +1.523 |
mble | +1.522 |
Neg logit contributions | |
ield | -1.822 |
vine | -1.813 |
het | -1.781 |
794 | -1.780 |
PF | -1.768 |
Token | ' |
Token ID | 6 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
angles | +2.754 |
Supervisor | +2.516 |
ortment | +2.500 |
aina | +2.498 |
mble | +2.460 |
Neg logit contributions | |
ield | -2.123 |
vine | -2.115 |
het | -2.073 |
794 | -2.058 |
PF | -2.057 |
Token | Ent |
Token ID | 14539 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
angles | +0.280 |
ortment | +0.257 |
Supervisor | +0.257 |
mble | +0.256 |
aina | +0.255 |
Neg logit contributions | |
ield | -0.224 |
vine | -0.223 |
het | -0.218 |
794 | -0.217 |
PF | -0.215 |
Token | rem |
Token ID | 2787 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ield | +0.362 |
vine | +0.360 |
het | +0.356 |
794 | +0.349 |
ordes | +0.344 |
Neg logit contributions | |
angles | -0.364 |
Supervisor | -0.337 |
ortment | -0.335 |
mble | -0.335 |
unlaw | -0.328 |
Token | better |
Token ID | 1365 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
vine | +0.348 |
ield | +0.347 |
het | +0.341 |
794 | +0.338 |
PF | +0.337 |
Neg logit contributions | |
angles | -0.459 |
ortment | -0.425 |
Supervisor | -0.419 |
mble | -0.419 |
aina | -0.419 |
Token | ! |
Token ID | 0 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
ield | +0.031 |
vine | +0.031 |
het | +0.031 |
794 | +0.030 |
PF | +0.030 |
Neg logit contributions | |
angles | -0.036 |
ortment | -0.033 |
Supervisor | -0.033 |
mble | -0.033 |
aina | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
angles | +0.240 |
Supervisor | +0.219 |
ortment | +0.218 |
mble | +0.217 |
aina | +0.216 |
Neg logit contributions | |
ield | -0.229 |
vine | -0.228 |
het | -0.224 |
794 | -0.223 |
PF | -0.221 |
Token |
|
Token ID | 198 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
angles | +0.073 |
mble | +0.068 |
ortment | +0.067 |
Supervisor | +0.066 |
aina | +0.066 |
Neg logit contributions | |
ield | -0.064 |
vine | -0.064 |
794 | -0.063 |
het | -0.063 |
PF | -0.062 |
Token | -- |
Token ID | 438 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
ield | +0.103 |
vine | +0.103 |
het | +0.101 |
794 | +0.100 |
PF | +0.099 |
Neg logit contributions | |
angles | -0.129 |
ortment | -0.118 |
Supervisor | -0.118 |
mble | -0.117 |
aina | -0.117 |
Token | Chris |
Token ID | 5180 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
angles | +2.414 |
Supervisor | +2.208 |
ortment | +2.183 |
aina | +2.179 |
mble | +2.173 |
Neg logit contributions | |
ield | -2.309 |
vine | -2.302 |
het | -2.253 |
794 | -2.239 |
PF | -2.207 |
Token | " |
Token ID | 366 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
angles | +2.458 |
Supervisor | +2.245 |
ortment | +2.223 |
aina | +2.213 |
mble | +2.206 |
Neg logit contributions | |
ield | -2.440 |
vine | -2.433 |
794 | -2.388 |
het | -2.383 |
PF | -2.375 |
Token | D |
Token ID | 35 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
angles | +0.052 |
Supervisor | +0.048 |
ortment | +0.048 |
mble | +0.048 |
aina | +0.048 |
Neg logit contributions | |
ield | -0.042 |
vine | -0.042 |
het | -0.041 |
794 | -0.041 |
PF | -0.041 |
Token | ingo |
Token ID | 32735 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
angles | +0.432 |
ortment | +0.395 |
Supervisor | +0.395 |
mble | +0.392 |
aina | +0.391 |
Neg logit contributions | |
ield | -0.377 |
vine | -0.375 |
het | -0.368 |
794 | -0.365 |
PF | -0.363 |
Token | " |
Token ID | 1 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
ield | +1.413 |
vine | +1.408 |
het | +1.386 |
794 | +1.375 |
PF | +1.370 |
Neg logit contributions | |
angles | -1.495 |
ortment | -1.367 |
mble | -1.362 |
Supervisor | -1.358 |
aina | -1.350 |
Token | Davies |
Token ID | 26618 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
ield | +0.508 |
vine | +0.507 |
het | +0.497 |
794 | +0.494 |
PF | +0.491 |
Neg logit contributions | |
angles | -0.493 |
ortment | -0.447 |
mble | -0.446 |
Supervisor | -0.445 |
aina | -0.442 |
Token | flight |
Token ID | 5474 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
ield | +1.635 |
vine | +1.626 |
het | +1.596 |
794 | +1.595 |
PF | +1.579 |
Neg logit contributions | |
angles | -1.858 |
ortment | -1.711 |
mble | -1.695 |
aina | -1.695 |
atown | -1.674 |
Token | control |
Token ID | 1630 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
ield | +0.307 |
vine | +0.305 |
het | +0.300 |
794 | +0.298 |
PF | +0.296 |
Neg logit contributions | |
angles | -0.321 |
ortment | -0.292 |
Supervisor | -0.291 |
mble | -0.290 |
aina | -0.289 |
Token | primary |
Token ID | 4165 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
angles | +1.226 |
Supervisor | +1.119 |
ortment | +1.106 |
aina | +1.100 |
mble | +1.100 |
Neg logit contributions | |
ield | -1.077 |
vine | -1.073 |
het | -1.053 |
794 | -1.047 |
PF | -1.040 |
Token | computers |
Token ID | 9061 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
angles | +2.228 |
Supervisor | +2.027 |
ortment | +2.000 |
aina | +1.988 |
mble | +1.987 |
Neg logit contributions | |
ield | -2.119 |
vine | -2.111 |
het | -2.061 |
794 | -2.057 |
PF | -2.055 |
Token | then |
Token ID | 788 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
angles | +0.581 |
ortment | +0.531 |
mble | +0.530 |
Supervisor | +0.524 |
aina | +0.524 |
Neg logit contributions | |
ield | -0.581 |
vine | -0.577 |
het | -0.569 |
794 | -0.565 |
PF | -0.558 |
Token | reacted |
Token ID | 22798 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
angles | +1.175 |
ortment | +1.065 |
Supervisor | +1.063 |
mble | +1.060 |
aina | +1.055 |
Neg logit contributions | |
ield | -1.239 |
vine | -1.232 |
het | -1.213 |
794 | -1.205 |
PF | -1.195 |
Token | to |
Token ID | 284 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
angles | +2.684 |
Supervisor | +2.446 |
ortment | +2.432 |
aina | +2.397 |
mble | +2.395 |
Neg logit contributions | |
vine | -2.565 |
ield | -2.554 |
het | -2.493 |
794 | -2.491 |
PF | -2.476 |
Token | the |
Token ID | 262 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
ield | +0.195 |
vine | +0.194 |
het | +0.190 |
794 | +0.188 |
PF | +0.187 |
Neg logit contributions | |
angles | -0.248 |
ortment | -0.227 |
Supervisor | -0.227 |
mble | -0.226 |
aina | -0.225 |
Token | angle |
Token ID | 9848 |
Feature activation | -8.01e-03 |
Pos logit contributions | |
ield | +0.112 |
vine | +0.111 |
het | +0.109 |
794 | +0.108 |
PF | +0.108 |
Neg logit contributions | |
angles | -0.149 |
ortment | -0.138 |
mble | -0.137 |
Supervisor | -0.137 |
aina | -0.136 |
Token | - |
Token ID | 12 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
ield | +0.015 |
vine | +0.015 |
het | +0.015 |
794 | +0.015 |
PF | +0.015 |
Neg logit contributions | |
angles | -0.019 |
ortment | -0.017 |
Supervisor | -0.017 |
mble | -0.017 |
aina | -0.017 |
Token | of |
Token ID | 1659 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
ield | +0.165 |
vine | +0.164 |
het | +0.159 |
794 | +0.157 |
PF | +0.155 |
Neg logit contributions | |
angles | -0.431 |
Supervisor | -0.406 |
ortment | -0.405 |
mble | -0.404 |
aina | -0.402 |
Token | � |
Token ID | 447 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
angles | +0.385 |
Supervisor | +0.351 |
ortment | +0.351 |
mble | +0.349 |
aina | +0.348 |
Neg logit contributions | |
ield | -0.344 |
vine | -0.342 |
het | -0.336 |
794 | -0.333 |
PF | -0.331 |
Token | � |
Token ID | 251 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
angles | +0.168 |
ortment | +0.154 |
Supervisor | +0.154 |
mble | +0.153 |
aina | +0.152 |
Neg logit contributions | |
ield | -0.149 |
vine | -0.148 |
het | -0.145 |
794 | -0.144 |
PF | -0.143 |
Token | that |
Token ID | 326 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
angles | +0.396 |
Supervisor | +0.362 |
ortment | +0.361 |
mble | +0.359 |
aina | +0.358 |
Neg logit contributions | |
ield | -0.345 |
vine | -0.344 |
het | -0.337 |
794 | -0.334 |
PF | -0.331 |
Token | takes |
Token ID | 2753 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
ield | +0.140 |
vine | +0.140 |
het | +0.138 |
794 | +0.137 |
PF | +0.136 |
Neg logit contributions | |
angles | -0.122 |
ortment | -0.109 |
mble | -0.109 |
Supervisor | -0.109 |
aina | -0.108 |
Token | into |
Token ID | 656 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
angles | +3.676 |
Supervisor | +3.415 |
ortment | +3.351 |
aina | +3.313 |
mble | +3.298 |
Neg logit contributions | |
vine | -3.367 |
ield | -3.347 |
het | -3.291 |
794 | -3.244 |
ournals | -3.223 |
Token | account |
Token ID | 1848 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
angles | +2.123 |
Supervisor | +1.939 |
ortment | +1.908 |
mble | +1.900 |
aina | +1.892 |
Neg logit contributions | |
vine | -2.517 |
ield | -2.508 |
794 | -2.447 |
PF | -2.444 |
het | -2.441 |
Token | the |
Token ID | 262 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
angles | +3.698 |
Supervisor | +3.478 |
ortment | +3.400 |
mble | +3.375 |
aina | +3.348 |
Neg logit contributions | |
ield | -3.088 |
vine | -3.057 |
het | -3.007 |
794 | -2.981 |
PF | -2.955 |
Token | feedback |
Token ID | 7538 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
angles | +0.062 |
Supervisor | +0.057 |
ortment | +0.057 |
mble | +0.056 |
aina | +0.056 |
Neg logit contributions | |
ield | -0.049 |
vine | -0.049 |
het | -0.048 |
794 | -0.048 |
PF | -0.047 |
Token | submitted |
Token ID | 8948 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
ield | +0.812 |
vine | +0.810 |
het | +0.793 |
794 | +0.788 |
PF | +0.782 |
Neg logit contributions | |
angles | -0.940 |
ortment | -0.859 |
Supervisor | -0.856 |
mble | -0.854 |
aina | -0.851 |
Token | by |
Token ID | 416 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
angles | +0.230 |
ortment | +0.211 |
Supervisor | +0.211 |
mble | +0.209 |
aina | +0.209 |
Neg logit contributions | |
ield | -0.193 |
vine | -0.192 |
het | -0.188 |
794 | -0.187 |
PF | -0.185 |
Token | Alf |
Token ID | 17136 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
angles | +0.135 |
Supervisor | +0.124 |
ortment | +0.124 |
mble | +0.123 |
aina | +0.123 |
Neg logit contributions | |
ield | -0.103 |
vine | -0.103 |
het | -0.101 |
794 | -0.100 |
PF | -0.099 |
Token | shapes |
Token ID | 15268 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
ield | +0.039 |
vine | +0.039 |
het | +0.039 |
794 | +0.038 |
PF | +0.038 |
Neg logit contributions | |
angles | -0.036 |
ortment | -0.032 |
Supervisor | -0.032 |
mble | -0.032 |
aina | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
ield | +1.140 |
vine | +1.130 |
het | +1.123 |
794 | +1.107 |
PF | +1.097 |
Neg logit contributions | |
angles | -1.379 |
ortment | -1.265 |
Supervisor | -1.264 |
mble | -1.261 |
aina | -1.254 |
Token | start |
Token ID | 923 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
angles | +0.063 |
ortment | +0.057 |
Supervisor | +0.057 |
mble | +0.057 |
aina | +0.057 |
Neg logit contributions | |
ield | -0.054 |
vine | -0.053 |
het | -0.052 |
794 | -0.052 |
PF | -0.052 |
Token | with |
Token ID | 351 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
ield | +0.706 |
vine | +0.698 |
het | +0.689 |
794 | +0.685 |
PF | +0.673 |
Neg logit contributions | |
angles | -0.722 |
Supervisor | -0.664 |
ortment | -0.662 |
aina | -0.659 |
mble | -0.653 |
Token | , |
Token ID | 11 |
Feature activation | -8.70e-02 |
Pos logit contributions | |
ield | +0.589 |
vine | +0.584 |
het | +0.576 |
794 | +0.571 |
PF | +0.566 |
Neg logit contributions | |
angles | -0.673 |
ortment | -0.616 |
Supervisor | -0.615 |
mble | -0.612 |
aina | -0.610 |
Token | so |
Token ID | 523 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
ield | +0.161 |
vine | +0.160 |
het | +0.157 |
794 | +0.156 |
PF | +0.155 |
Neg logit contributions | |
angles | -0.212 |
Supervisor | -0.195 |
ortment | -0.194 |
mble | -0.192 |
aina | -0.192 |
Token | the |
Token ID | 262 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
ield | +0.303 |
vine | +0.302 |
het | +0.297 |
794 | +0.294 |
PF | +0.292 |
Neg logit contributions | |
angles | -0.355 |
ortment | -0.325 |
Supervisor | -0.324 |
mble | -0.323 |
aina | -0.322 |
Token | fake |
Token ID | 8390 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
ield | +0.172 |
vine | +0.171 |
het | +0.168 |
794 | +0.167 |
PF | +0.165 |
Neg logit contributions | |
angles | -0.197 |
ortment | -0.180 |
Supervisor | -0.179 |
mble | -0.178 |
aina | -0.178 |
Token | data |
Token ID | 1366 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
angles | +0.232 |
ortment | +0.214 |
Supervisor | +0.213 |
mble | +0.212 |
aina | +0.212 |
Neg logit contributions | |
ield | -0.170 |
vine | -0.169 |
het | -0.165 |
794 | -0.164 |
PF | -0.162 |
Token | include |
Token ID | 2291 |
Feature activation | +7.13e-02 |
Pos logit contributions | |
angles | +0.464 |
ortment | +0.417 |
Supervisor | +0.416 |
mble | +0.413 |
aina | +0.412 |
Neg logit contributions | |
ield | -0.574 |
vine | -0.573 |
het | -0.563 |
794 | -0.560 |
PF | -0.556 |
Token | galaxies |
Token ID | 27982 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
angles | +0.169 |
Supervisor | +0.155 |
ortment | +0.155 |
mble | +0.154 |
aina | +0.154 |
Neg logit contributions | |
ield | -0.136 |
vine | -0.136 |
het | -0.133 |
794 | -0.132 |
PF | -0.131 |
Token | � |
Token ID | 447 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
angles | +0.106 |
ortment | +0.097 |
Supervisor | +0.097 |
mble | +0.096 |
aina | +0.096 |
Neg logit contributions | |
ield | -0.095 |
vine | -0.094 |
het | -0.093 |
794 | -0.092 |
PF | -0.091 |
Token | � |
Token ID | 247 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
angles | +0.242 |
ortment | +0.221 |
Supervisor | +0.220 |
mble | +0.220 |
aina | +0.218 |
Neg logit contributions | |
ield | -0.219 |
vine | -0.217 |
het | -0.213 |
794 | -0.212 |
PF | -0.210 |
Token | st |
Token ID | 336 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
angles | +0.811 |
ortment | +0.724 |
Supervisor | +0.722 |
mble | +0.720 |
aina | +0.715 |
Neg logit contributions | |
ield | -1.094 |
vine | -1.090 |
het | -1.074 |
794 | -1.070 |
PF | -1.060 |
Token | up |
Token ID | 929 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
angles | +0.627 |
ortment | +0.567 |
Supervisor | +0.567 |
mble | +0.563 |
aina | +0.560 |
Neg logit contributions | |
ield | -0.693 |
vine | -0.690 |
het | -0.679 |
794 | -0.674 |
PF | -0.670 |
Token | end |
Token ID | 437 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
ield | +0.394 |
vine | +0.392 |
het | +0.385 |
794 | +0.381 |
PF | +0.379 |
Neg logit contributions | |
angles | -0.477 |
Supervisor | -0.438 |
ortment | -0.437 |
mble | -0.435 |
aina | -0.434 |
Token | ous |
Token ID | 516 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
angles | +0.936 |
Supervisor | +0.855 |
ortment | +0.855 |
mble | +0.851 |
aina | +0.848 |
Neg logit contributions | |
ield | -0.752 |
vine | -0.749 |
het | -0.735 |
794 | -0.731 |
PF | -0.725 |
Token | imaging |
Token ID | 19560 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
angles | +1.446 |
ortment | +1.327 |
Supervisor | +1.325 |
mble | +1.323 |
aina | +1.312 |
Neg logit contributions | |
ield | -1.186 |
vine | -1.183 |
het | -1.159 |
794 | -1.154 |
PF | -1.146 |
Token | , |
Token ID | 11 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
angles | +2.162 |
Supervisor | +1.970 |
ortment | +1.969 |
mble | +1.957 |
aina | +1.937 |
Neg logit contributions | |
ield | -2.089 |
vine | -2.084 |
het | -2.046 |
794 | -2.039 |
PF | -2.028 |
Token | gorgeous |
Token ID | 18857 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
angles | +0.744 |
Supervisor | +0.685 |
ortment | +0.685 |
mble | +0.679 |
aina | +0.678 |
Neg logit contributions | |
ield | -0.572 |
vine | -0.568 |
het | -0.556 |
794 | -0.553 |
PF | -0.550 |
Token | midrange |
Token ID | 49260 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
angles | +1.926 |
Supervisor | +1.773 |
ortment | +1.766 |
mble | +1.760 |
aina | +1.748 |
Neg logit contributions | |
ield | -1.508 |
vine | -1.500 |
794 | -1.469 |
het | -1.468 |
PF | -1.464 |
Token | , |
Token ID | 11 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
angles | +0.698 |
Supervisor | +0.635 |
ortment | +0.635 |
mble | +0.632 |
aina | +0.628 |
Neg logit contributions | |
ield | -0.649 |
vine | -0.646 |
het | -0.635 |
794 | -0.631 |
PF | -0.627 |
Token | a |
Token ID | 257 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
vine | +0.690 |
ield | +0.689 |
het | +0.676 |
794 | +0.668 |
PF | +0.663 |
Neg logit contributions | |
angles | -0.890 |
ortment | -0.818 |
Supervisor | -0.818 |
mble | -0.815 |
aina | -0.813 |
Token | microscope |
Token ID | 36396 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
angles | +0.365 |
Supervisor | +0.329 |
ortment | +0.328 |
mble | +0.326 |
aina | +0.324 |
Neg logit contributions | |
ield | -0.436 |
vine | -0.432 |
het | -0.428 |
794 | -0.425 |
PF | -0.422 |
Token | and |
Token ID | 290 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
angles | +2.639 |
Supervisor | +2.392 |
mble | +2.355 |
ortment | +2.331 |
aina | +2.313 |
Neg logit contributions | |
ield | -3.023 |
vine | -2.964 |
het | -2.956 |
ordes | -2.946 |
PF | -2.934 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
angles | +0.401 |
ortment | +0.366 |
mble | +0.364 |
Supervisor | +0.363 |
aina | +0.362 |
Neg logit contributions | |
ield | -0.403 |
vine | -0.402 |
het | -0.394 |
794 | -0.391 |
PF | -0.388 |
Token | the |
Token ID | 262 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
angles | +0.265 |
Supervisor | +0.243 |
ortment | +0.243 |
mble | +0.241 |
aina | +0.240 |
Neg logit contributions | |
ield | -0.203 |
vine | -0.202 |
het | -0.198 |
794 | -0.196 |
PF | -0.195 |
Token | tell |
Token ID | 1560 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
angles | +0.155 |
Supervisor | +0.142 |
ortment | +0.142 |
mble | +0.141 |
aina | +0.141 |
Neg logit contributions | |
ield | -0.126 |
vine | -0.125 |
het | -0.123 |
794 | -0.122 |
PF | -0.121 |
Token | tale |
Token ID | 29429 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
angles | +0.222 |
ortment | +0.202 |
Supervisor | +0.202 |
mble | +0.201 |
aina | +0.200 |
Neg logit contributions | |
ield | -0.210 |
vine | -0.209 |
het | -0.205 |
794 | -0.204 |
PF | -0.203 |
Token | dent |
Token ID | 18794 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
angles | +0.674 |
Supervisor | +0.615 |
ortment | +0.614 |
mble | +0.610 |
aina | +0.609 |
Neg logit contributions | |
ield | -0.583 |
vine | -0.579 |
het | -0.570 |
794 | -0.565 |
PF | -0.562 |
Token | in |
Token ID | 259 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
angles | +1.252 |
ortment | +1.128 |
aina | +1.120 |
Supervisor | +1.116 |
mble | +1.113 |
Neg logit contributions | |
ield | -1.284 |
vine | -1.275 |
het | -1.259 |
PF | -1.252 |
794 | -1.248 |
Token | layers |
Token ID | 11685 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
ield | +0.571 |
vine | +0.571 |
het | +0.558 |
794 | +0.555 |
PF | +0.550 |
Neg logit contributions | |
angles | -0.614 |
ortment | -0.560 |
Supervisor | -0.560 |
mble | -0.557 |
aina | -0.554 |
Token | , |
Token ID | 11 |
Feature activation | -7.05e-03 |
Pos logit contributions | |
angles | +0.862 |
Supervisor | +0.785 |
ortment | +0.785 |
mble | +0.779 |
aina | +0.776 |
Neg logit contributions | |
ield | -0.797 |
vine | -0.790 |
het | -0.777 |
794 | -0.771 |
PF | -0.768 |
Token | it |
Token ID | 340 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
ield | +0.776 |
vine | +0.771 |
het | +0.761 |
794 | +0.754 |
PF | +0.751 |
Neg logit contributions | |
angles | -0.865 |
mble | -0.792 |
ortment | -0.790 |
Supervisor | -0.788 |
aina | -0.781 |
Token | Y |
Token ID | 575 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
angles | +0.645 |
Supervisor | +0.590 |
aina | +0.586 |
ortment | +0.585 |
mble | +0.581 |
Neg logit contributions | |
vine | -0.566 |
ield | -0.566 |
het | -0.552 |
794 | -0.549 |
PF | -0.545 |
Token | OW |
Token ID | 3913 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
ield | +1.135 |
vine | +1.107 |
het | +1.097 |
PF | +1.086 |
794 | +1.084 |
Neg logit contributions | |
angles | -1.352 |
ortment | -1.246 |
Supervisor | -1.240 |
mble | -1.236 |
aina | -1.225 |
Token | is |
Token ID | 318 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
angles | +0.936 |
Supervisor | +0.850 |
ortment | +0.844 |
aina | +0.843 |
mble | +0.839 |
Neg logit contributions | |
ield | -0.891 |
vine | -0.890 |
het | -0.872 |
794 | -0.867 |
PF | -0.860 |
Token | a |
Token ID | 257 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
angles | +0.829 |
Supervisor | +0.762 |
ortment | +0.757 |
mble | +0.754 |
aina | +0.752 |
Neg logit contributions | |
ield | -0.684 |
vine | -0.682 |
het | -0.669 |
794 | -0.663 |
PF | -0.658 |
Token | free |
Token ID | 1479 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
angles | +0.092 |
Supervisor | +0.085 |
ortment | +0.084 |
mble | +0.084 |
aina | +0.084 |
Neg logit contributions | |
ield | -0.073 |
vine | -0.073 |
het | -0.072 |
794 | -0.071 |
PF | -0.071 |
Token | local |
Token ID | 1957 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
ield | +0.104 |
vine | +0.103 |
het | +0.101 |
794 | +0.101 |
PF | +0.100 |
Neg logit contributions | |
angles | -0.118 |
Supervisor | -0.108 |
ortment | -0.108 |
mble | -0.107 |
aina | -0.107 |
Token | fitness |
Token ID | 13547 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
angles | +1.344 |
ortment | +1.236 |
Supervisor | +1.235 |
aina | +1.229 |
mble | +1.226 |
Neg logit contributions | |
vine | -1.188 |
ield | -1.184 |
het | -1.163 |
PF | -1.152 |
794 | -1.150 |
Token | initiative |
Token ID | 10219 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
angles | +0.381 |
Supervisor | +0.350 |
ortment | +0.349 |
mble | +0.347 |
aina | +0.346 |
Neg logit contributions | |
ield | -0.313 |
vine | -0.312 |
het | -0.306 |
794 | -0.303 |
PF | -0.302 |
Token | inspired |
Token ID | 7867 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
ield | +1.190 |
vine | +1.181 |
het | +1.159 |
794 | +1.156 |
PF | +1.141 |
Neg logit contributions | |
angles | -1.433 |
ortment | -1.312 |
mble | -1.304 |
Supervisor | -1.299 |
aina | -1.298 |
Token | by |
Token ID | 416 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
ield | +0.051 |
vine | +0.051 |
het | +0.050 |
794 | +0.050 |
PF | +0.049 |
Neg logit contributions | |
angles | -0.064 |
Supervisor | -0.059 |
ortment | -0.059 |
mble | -0.059 |
aina | -0.059 |
Token | do |
Token ID | 466 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
angles | +0.376 |
Supervisor | +0.345 |
ortment | +0.343 |
aina | +0.341 |
mble | +0.340 |
Neg logit contributions | |
ield | -0.325 |
vine | -0.324 |
het | -0.319 |
794 | -0.318 |
PF | -0.314 |
Token | you |
Token ID | 345 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
ield | +0.693 |
vine | +0.691 |
het | +0.676 |
794 | +0.670 |
PF | +0.665 |
Neg logit contributions | |
angles | -0.848 |
ortment | -0.779 |
Supervisor | -0.772 |
mble | -0.772 |
aina | -0.767 |
Token | maintain |
Token ID | 5529 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
ield | +1.402 |
vine | +1.372 |
het | +1.349 |
drip | +1.336 |
794 | +1.330 |
Neg logit contributions | |
angles | -2.027 |
mble | -1.875 |
ortment | -1.874 |
aina | -1.852 |
Supervisor | -1.849 |
Token | a |
Token ID | 257 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
ield | +0.680 |
vine | +0.678 |
het | +0.661 |
794 | +0.656 |
PF | +0.650 |
Neg logit contributions | |
angles | -0.986 |
ortment | -0.911 |
Supervisor | -0.905 |
mble | -0.903 |
aina | -0.902 |
Token | semblance |
Token ID | 45960 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
ield | +0.883 |
vine | +0.878 |
het | +0.859 |
794 | +0.855 |
PF | +0.850 |
Neg logit contributions | |
angles | -1.069 |
ortment | -0.977 |
mble | -0.968 |
aina | -0.968 |
Supervisor | -0.966 |
Token | of |
Token ID | 286 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
vine | +1.217 |
ield | +1.211 |
het | +1.189 |
794 | +1.174 |
Finch | +1.158 |
Neg logit contributions | |
angles | -1.743 |
mble | -1.614 |
ortment | -1.612 |
aina | -1.588 |
unlaw | -1.574 |
Token | normal |
Token ID | 3487 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
angles | +0.373 |
ortment | +0.343 |
Supervisor | +0.342 |
mble | +0.341 |
aina | +0.340 |
Neg logit contributions | |
ield | -0.276 |
vine | -0.275 |
het | -0.270 |
794 | -0.267 |
PF | -0.265 |
Token | ? |
Token ID | 30 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
angles | +0.255 |
ortment | +0.228 |
Supervisor | +0.228 |
mble | +0.227 |
aina | +0.226 |
Neg logit contributions | |
ield | -0.316 |
vine | -0.315 |
het | -0.310 |
794 | -0.308 |
PF | -0.306 |
Token | Given |
Token ID | 11259 |
Feature activation | -9.52e-01 |
Pos logit contributions | |
ield | +0.600 |
vine | +0.599 |
het | +0.588 |
794 | +0.582 |
Finch | +0.579 |
Neg logit contributions | |
angles | -0.678 |
mble | -0.622 |
ortment | -0.616 |
aina | -0.613 |
Supervisor | -0.610 |
Token | the |
Token ID | 262 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
ield | +1.657 |
vine | +1.656 |
het | +1.618 |
794 | +1.600 |
Finch | +1.586 |
Neg logit contributions | |
angles | -2.407 |
ortment | -2.219 |
Supervisor | -2.197 |
mble | -2.196 |
aina | -2.196 |
Token | world |
Token ID | 995 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
ield | +0.795 |
vine | +0.791 |
het | +0.775 |
794 | +0.769 |
PF | +0.761 |
Neg logit contributions | |
angles | -1.011 |
ortment | -0.928 |
Supervisor | -0.922 |
mble | -0.922 |
aina | -0.916 |
Token | result |
Token ID | 1255 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
ield | +0.026 |
vine | +0.025 |
het | +0.025 |
794 | +0.025 |
PF | +0.025 |
Neg logit contributions | |
angles | -0.029 |
Supervisor | -0.026 |
ortment | -0.026 |
mble | -0.026 |
aina | -0.026 |
Token | of |
Token ID | 286 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
angles | +0.091 |
mble | +0.084 |
ortment | +0.083 |
Supervisor | +0.083 |
aina | +0.083 |
Neg logit contributions | |
ield | -0.070 |
vine | -0.070 |
het | -0.069 |
794 | -0.068 |
PF | -0.068 |
Token | supply |
Token ID | 5127 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ield | +0.400 |
vine | +0.397 |
het | +0.390 |
794 | +0.385 |
PF | +0.382 |
Neg logit contributions | |
angles | -0.529 |
ortment | -0.486 |
mble | -0.485 |
Supervisor | -0.482 |
aina | -0.481 |
Token | - |
Token ID | 12 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
vine | +2.399 |
ield | +2.398 |
het | +2.335 |
794 | +2.318 |
drip | +2.315 |
Neg logit contributions | |
angles | -3.239 |
mble | -3.024 |
ortment | -3.000 |
aina | -2.964 |
arnaev | -2.935 |
Token | and |
Token ID | 392 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ield | +0.668 |
vine | +0.665 |
het | +0.656 |
794 | +0.652 |
PF | +0.649 |
Neg logit contributions | |
angles | -0.444 |
Supervisor | -0.392 |
ortment | -0.392 |
mble | -0.389 |
aina | -0.387 |
Token | - |
Token ID | 12 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
ield | +0.246 |
vine | +0.244 |
het | +0.239 |
794 | +0.237 |
PF | +0.234 |
Neg logit contributions | |
angles | -0.339 |
ortment | -0.312 |
Supervisor | -0.312 |
mble | -0.312 |
aina | -0.309 |
Token | demand |
Token ID | 28550 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
ield | +0.391 |
vine | +0.387 |
het | +0.380 |
794 | +0.376 |
PF | +0.374 |
Neg logit contributions | |
angles | -0.793 |
ortment | -0.734 |
Supervisor | -0.734 |
aina | -0.731 |
mble | -0.729 |
Token | fundamentals |
Token ID | 33099 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
ield | +1.674 |
vine | +1.661 |
het | +1.638 |
794 | +1.618 |
drip | +1.601 |
Neg logit contributions | |
angles | -2.050 |
ortment | -1.914 |
mble | -1.899 |
aina | -1.888 |
Supervisor | -1.872 |
Token | . |
Token ID | 13 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
ield | +0.776 |
vine | +0.773 |
het | +0.762 |
794 | +0.754 |
PF | +0.748 |
Neg logit contributions | |
angles | -0.914 |
ortment | -0.838 |
mble | -0.835 |
Supervisor | -0.830 |
aina | -0.829 |
Token | More |
Token ID | 3125 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
angles | +0.052 |
ortment | +0.048 |
Supervisor | +0.048 |
mble | +0.047 |
aina | +0.047 |
Neg logit contributions | |
ield | -0.047 |
vine | -0.046 |
het | -0.046 |
794 | -0.045 |
PF | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
ield | +0.398 |
vine | +0.396 |
het | +0.389 |
794 | +0.385 |
PF | +0.382 |
Neg logit contributions | |
angles | -0.504 |
ortment | -0.463 |
Supervisor | -0.461 |
mble | -0.460 |
aina | -0.458 |
Token | groups |
Token ID | 2628 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
angles | +1.309 |
Supervisor | +1.200 |
ortment | +1.192 |
mble | +1.182 |
aina | +1.182 |
Neg logit contributions | |
ield | -1.233 |
vine | -1.232 |
het | -1.210 |
PF | -1.199 |
794 | -1.198 |
Token | , |
Token ID | 11 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
angles | +1.781 |
Supervisor | +1.645 |
aina | +1.622 |
ortment | +1.613 |
mble | +1.605 |
Neg logit contributions | |
ield | -1.604 |
vine | -1.593 |
het | -1.546 |
794 | -1.542 |
PF | -1.540 |
Token | the |
Token ID | 262 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
angles | +0.262 |
Supervisor | +0.241 |
ortment | +0.238 |
mble | +0.237 |
aina | +0.237 |
Neg logit contributions | |
ield | -0.230 |
vine | -0.229 |
het | -0.225 |
794 | -0.223 |
PF | -0.222 |
Token | United |
Token ID | 1578 |
Feature activation | -8.71e-02 |
Pos logit contributions | |
angles | +0.092 |
Supervisor | +0.083 |
ortment | +0.083 |
mble | +0.083 |
aina | +0.082 |
Neg logit contributions | |
ield | -0.091 |
vine | -0.091 |
het | -0.089 |
794 | -0.088 |
PF | -0.088 |
Token | States |
Token ID | 1829 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
ield | +0.264 |
vine | +0.263 |
het | +0.260 |
794 | +0.258 |
PF | +0.257 |
Neg logit contributions | |
angles | -0.110 |
ortment | -0.092 |
Supervisor | -0.092 |
mble | -0.091 |
aina | -0.090 |
Token | Senate |
Token ID | 3845 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
angles | +0.581 |
Supervisor | +0.528 |
ortment | +0.524 |
mble | +0.521 |
aina | +0.520 |
Neg logit contributions | |
ield | -0.626 |
vine | -0.625 |
het | -0.614 |
794 | -0.610 |
PF | -0.606 |
Token | , |
Token ID | 11 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
ield | +1.450 |
vine | +1.432 |
het | +1.409 |
794 | +1.403 |
PF | +1.391 |
Neg logit contributions | |
angles | -1.627 |
ortment | -1.490 |
Supervisor | -1.471 |
mble | -1.469 |
aina | -1.465 |
Token | and |
Token ID | 290 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
angles | +0.602 |
Supervisor | +0.556 |
ortment | +0.549 |
mble | +0.548 |
aina | +0.547 |
Neg logit contributions | |
ield | -0.538 |
vine | -0.537 |
het | -0.525 |
794 | -0.522 |
PF | -0.518 |
Token | the |
Token ID | 262 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
angles | +0.642 |
Supervisor | +0.589 |
ortment | +0.586 |
mble | +0.584 |
aina | +0.581 |
Neg logit contributions | |
ield | -0.546 |
vine | -0.544 |
het | -0.534 |
794 | -0.530 |
PF | -0.526 |
Token | American |
Token ID | 1605 |
Feature activation | +8.11e-03 |
Pos logit contributions | |
ield | +0.355 |
vine | +0.352 |
het | +0.347 |
794 | +0.345 |
PF | +0.342 |
Neg logit contributions | |
angles | -0.378 |
ortment | -0.344 |
Supervisor | -0.341 |
mble | -0.341 |
aina | -0.340 |
Token | people |
Token ID | 661 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
angles | +0.017 |
ortment | +0.015 |
Supervisor | +0.015 |
mble | +0.015 |
aina | +0.015 |
Neg logit contributions | |
ield | -0.018 |
vine | -0.018 |
het | -0.017 |
794 | -0.017 |
PF | -0.017 |
Token | George |
Token ID | 4502 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
ield | +0.808 |
vine | +0.807 |
het | +0.795 |
794 | +0.791 |
Finch | +0.789 |
Neg logit contributions | |
angles | -0.930 |
ortment | -0.855 |
mble | -0.846 |
aina | -0.840 |
Supervisor | -0.833 |
Token | . |
Token ID | 13 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
angles | +0.061 |
ortment | +0.055 |
mble | +0.055 |
Supervisor | +0.055 |
aina | +0.054 |
Neg logit contributions | |
ield | -0.063 |
vine | -0.063 |
het | -0.062 |
794 | -0.061 |
PF | -0.061 |
Token | How |
Token ID | 1374 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
ield | +0.479 |
vine | +0.477 |
het | +0.469 |
794 | +0.466 |
PF | +0.463 |
Neg logit contributions | |
angles | -0.541 |
mble | -0.496 |
ortment | -0.495 |
aina | -0.490 |
Supervisor | -0.489 |
Token | are |
Token ID | 389 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
angles | +0.434 |
Supervisor | +0.396 |
ortment | +0.395 |
mble | +0.393 |
aina | +0.392 |
Neg logit contributions | |
ield | -0.386 |
vine | -0.385 |
het | -0.378 |
794 | -0.376 |
PF | -0.372 |
Token | you |
Token ID | 345 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
ield | +0.077 |
vine | +0.077 |
het | +0.075 |
794 | +0.075 |
PF | +0.074 |
Neg logit contributions | |
angles | -0.095 |
ortment | -0.088 |
Supervisor | -0.087 |
mble | -0.087 |
aina | -0.087 |
Token | doing |
Token ID | 1804 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
ield | +0.774 |
vine | +0.772 |
het | +0.766 |
794 | +0.750 |
Finch | +0.748 |
Neg logit contributions | |
angles | -1.012 |
ortment | -0.939 |
mble | -0.934 |
Supervisor | -0.917 |
aina | -0.917 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
ield | +1.243 |
vine | +1.237 |
het | +1.216 |
794 | +1.203 |
Finch | +1.201 |
Neg logit contributions | |
angles | -1.342 |
ortment | -1.240 |
mble | -1.221 |
Supervisor | -1.211 |
aina | -1.211 |
Token | My |
Token ID | 2011 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
ield | +0.850 |
vine | +0.845 |
het | +0.837 |
794 | +0.826 |
PF | +0.822 |
Neg logit contributions | |
angles | -0.928 |
mble | -0.852 |
ortment | -0.847 |
aina | -0.841 |
Supervisor | -0.838 |
Token | answer |
Token ID | 3280 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
angles | +0.693 |
ortment | +0.635 |
Supervisor | +0.634 |
mble | +0.631 |
aina | +0.629 |
Neg logit contributions | |
ield | -0.563 |
vine | -0.560 |
het | -0.550 |
794 | -0.545 |
PF | -0.542 |
Token | was |
Token ID | 373 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
angles | +0.573 |
ortment | +0.512 |
Supervisor | +0.511 |
mble | +0.508 |
aina | +0.506 |
Neg logit contributions | |
ield | -0.743 |
vine | -0.740 |
het | -0.729 |
794 | -0.725 |
PF | -0.721 |
Token | short |
Token ID | 1790 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ield | +0.224 |
vine | +0.223 |
het | +0.220 |
794 | +0.218 |
PF | +0.216 |
Neg logit contributions | |
angles | -0.268 |
ortment | -0.245 |
mble | -0.245 |
Supervisor | -0.244 |
aina | -0.243 |
Token | part |
Token ID | 636 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
angles | +1.121 |
Supervisor | +1.029 |
ortment | +1.023 |
mble | +1.018 |
aina | +1.015 |
Neg logit contributions | |
ield | -0.983 |
vine | -0.982 |
het | -0.960 |
794 | -0.955 |
PF | -0.947 |
Token | , |
Token ID | 11 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
angles | +0.255 |
Supervisor | +0.233 |
ortment | +0.233 |
mble | +0.232 |
aina | +0.231 |
Neg logit contributions | |
ield | -0.219 |
vine | -0.218 |
het | -0.214 |
794 | -0.213 |
PF | -0.211 |
Token | but |
Token ID | 475 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
ield | +0.365 |
vine | +0.363 |
het | +0.356 |
794 | +0.353 |
PF | +0.350 |
Neg logit contributions | |
angles | -0.489 |
Supervisor | -0.449 |
ortment | -0.449 |
mble | -0.446 |
aina | -0.445 |
Token | my |
Token ID | 616 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
angles | +1.288 |
Supervisor | +1.180 |
ortment | +1.173 |
mble | +1.170 |
aina | +1.163 |
Neg logit contributions | |
ield | -1.107 |
vine | -1.100 |
het | -1.077 |
794 | -1.072 |
PF | -1.064 |
Token | advice |
Token ID | 5608 |
Feature activation | +6.37e-02 |
Pos logit contributions | |
ield | +0.110 |
vine | +0.110 |
het | +0.108 |
794 | +0.107 |
PF | +0.106 |
Neg logit contributions | |
angles | -0.133 |
ortment | -0.122 |
Supervisor | -0.122 |
mble | -0.121 |
aina | -0.121 |
Token | is |
Token ID | 318 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
angles | +0.134 |
ortment | +0.121 |
Supervisor | +0.121 |
mble | +0.120 |
aina | +0.120 |
Neg logit contributions | |
ield | -0.139 |
vine | -0.139 |
het | -0.137 |
794 | -0.135 |
PF | -0.135 |
Token | to |
Token ID | 284 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ield | +0.690 |
vine | +0.687 |
het | +0.673 |
794 | +0.668 |
PF | +0.659 |
Neg logit contributions | |
angles | -0.867 |
ortment | -0.792 |
mble | -0.789 |
Supervisor | -0.788 |
aina | -0.785 |
Token | skip |
Token ID | 14267 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
ield | +0.205 |
vine | +0.202 |
het | +0.199 |
794 | +0.198 |
PF | +0.196 |
Neg logit contributions | |
angles | -0.263 |
ortment | -0.242 |
mble | -0.241 |
Supervisor | -0.241 |
aina | -0.240 |
Token | it |
Token ID | 340 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
ield | +0.496 |
vine | +0.493 |
het | +0.485 |
794 | +0.481 |
PF | +0.478 |
Neg logit contributions | |
angles | -0.600 |
ortment | -0.550 |
Supervisor | -0.547 |
mble | -0.545 |
aina | -0.543 |
Token | only |
Token ID | 691 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
angles | +0.508 |
Supervisor | +0.465 |
ortment | +0.464 |
mble | +0.461 |
aina | +0.461 |
Neg logit contributions | |
ield | -0.438 |
vine | -0.436 |
het | -0.427 |
794 | -0.425 |
PF | -0.422 |
Token | if |
Token ID | 611 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
angles | +0.448 |
ortment | +0.411 |
Supervisor | +0.411 |
mble | +0.409 |
aina | +0.407 |
Neg logit contributions | |
ield | -0.343 |
vine | -0.341 |
het | -0.335 |
794 | -0.332 |
PF | -0.330 |
Token | slim |
Token ID | 18862 |
Feature activation | +9.58e-01 |
Pos logit contributions | |
ield | +0.058 |
vine | +0.057 |
het | +0.056 |
794 | +0.056 |
PF | +0.056 |
Neg logit contributions | |
angles | -0.063 |
ortment | -0.058 |
Supervisor | -0.058 |
mble | -0.057 |
aina | -0.057 |
Token | y |
Token ID | 88 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
angles | +2.180 |
mble | +2.008 |
Supervisor | +2.005 |
ortment | +2.003 |
aina | +1.989 |
Neg logit contributions | |
ield | -1.890 |
vine | -1.883 |
het | -1.850 |
794 | -1.841 |
PF | -1.834 |
Token | smooth |
Token ID | 7209 |
Feature activation | -8.62e-01 |
Pos logit contributions | |
ield | +0.445 |
vine | +0.443 |
het | +0.436 |
794 | +0.431 |
PF | +0.428 |
Neg logit contributions | |
angles | -0.486 |
ortment | -0.441 |
Supervisor | -0.439 |
mble | -0.437 |
aina | -0.437 |
Token | m |
Token ID | 285 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
ield | +1.591 |
vine | +1.587 |
het | +1.558 |
794 | +1.536 |
PF | +1.523 |
Neg logit contributions | |
angles | -2.096 |
ortment | -1.917 |
Supervisor | -1.917 |
mble | -1.907 |
aina | -1.899 |
Token | over |
Token ID | 2502 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
ield | +0.501 |
vine | +0.499 |
het | +0.490 |
794 | +0.487 |
PF | +0.484 |
Neg logit contributions | |
angles | -0.527 |
Supervisor | -0.479 |
ortment | -0.479 |
mble | -0.476 |
aina | -0.474 |
Token | named |
Token ID | 3706 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
angles | +0.288 |
Supervisor | +0.264 |
ortment | +0.263 |
mble | +0.261 |
aina | +0.261 |
Neg logit contributions | |
ield | -0.252 |
vine | -0.251 |
het | -0.246 |
794 | -0.245 |
PF | -0.244 |
Token | Dim |
Token ID | 14048 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
vine | +1.373 |
ield | +1.369 |
het | +1.359 |
Finch | +1.349 |
794 | +1.337 |
Neg logit contributions | |
angles | -1.198 |
ortment | -1.077 |
mble | -1.076 |
aina | -1.064 |
Supervisor | -1.057 |
Token | itri |
Token ID | 49510 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
angles | +3.396 |
aina | +3.065 |
mble | +3.056 |
ortment | +3.043 |
Supervisor | +3.031 |
Neg logit contributions | |
vine | -3.474 |
ield | -3.468 |
794 | -3.425 |
PF | -3.423 |
het | -3.380 |
Token | to |
Token ID | 284 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
angles | +1.813 |
ortment | +1.648 |
Supervisor | +1.643 |
aina | +1.634 |
mble | +1.628 |
Neg logit contributions | |
ield | -1.773 |
vine | -1.766 |
het | -1.730 |
794 | -1.728 |
PF | -1.719 |
Token | a |
Token ID | 257 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
angles | +0.210 |
ortment | +0.193 |
Supervisor | +0.193 |
mble | +0.192 |
aina | +0.191 |
Neg logit contributions | |
ield | -0.163 |
vine | -0.162 |
het | -0.159 |
794 | -0.158 |
PF | -0.156 |
Token | r |
Token ID | 374 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
ield | +0.342 |
vine | +0.340 |
het | +0.334 |
794 | +0.331 |
PF | +0.329 |
Neg logit contributions | |
angles | -0.373 |
ortment | -0.339 |
Supervisor | -0.338 |
mble | -0.337 |
aina | -0.336 |
Token | against |
Token ID | 1028 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
angles | +0.001 |
ortment | +0.001 |
Supervisor | +0.001 |
mble | +0.001 |
aina | +0.001 |
Neg logit contributions | |
ield | -0.001 |
vine | -0.001 |
het | -0.001 |
794 | -0.001 |
PF | -0.001 |
Token | them |
Token ID | 606 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
angles | +0.229 |
ortment | +0.208 |
mble | +0.206 |
Supervisor | +0.205 |
aina | +0.205 |
Neg logit contributions | |
ield | -0.218 |
vine | -0.218 |
het | -0.214 |
794 | -0.212 |
PF | -0.211 |
Token | that |
Token ID | 326 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
ield | +0.255 |
vine | +0.254 |
het | +0.250 |
794 | +0.248 |
PF | +0.246 |
Neg logit contributions | |
angles | -0.284 |
ortment | -0.259 |
Supervisor | -0.258 |
mble | -0.257 |
aina | -0.256 |
Token | causes |
Token ID | 5640 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
vine | +0.215 |
ield | +0.214 |
het | +0.210 |
794 | +0.208 |
PF | +0.207 |
Neg logit contributions | |
angles | -0.230 |
ortment | -0.209 |
mble | -0.207 |
aina | -0.207 |
Supervisor | -0.206 |
Token | them |
Token ID | 606 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
ield | +1.359 |
vine | +1.356 |
het | +1.329 |
794 | +1.312 |
Finch | +1.304 |
Neg logit contributions | |
angles | -1.574 |
ortment | -1.441 |
mble | -1.432 |
aina | -1.425 |
Supervisor | -1.415 |
Token | harm |
Token ID | 4419 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ield | +0.276 |
vine | +0.275 |
het | +0.270 |
794 | +0.267 |
PF | +0.265 |
Neg logit contributions | |
angles | -0.323 |
ortment | -0.295 |
Supervisor | -0.295 |
mble | -0.293 |
aina | -0.292 |
Token | are |
Token ID | 389 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
vine | +3.049 |
ield | +3.049 |
het | +2.999 |
794 | +2.961 |
PF | +2.930 |
Neg logit contributions | |
angles | -3.538 |
mble | -3.202 |
ortment | -3.196 |
Supervisor | -3.182 |
aina | -3.168 |
Token | often |
Token ID | 1690 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
angles | +0.619 |
Supervisor | +0.571 |
ortment | +0.570 |
mble | +0.567 |
aina | +0.566 |
Neg logit contributions | |
ield | -0.424 |
vine | -0.421 |
het | -0.412 |
794 | -0.409 |
PF | -0.406 |
Token | urged |
Token ID | 11643 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
angles | +1.067 |
Supervisor | +0.979 |
ortment | +0.974 |
mble | +0.967 |
aina | +0.966 |
Neg logit contributions | |
ield | -0.945 |
vine | -0.941 |
het | -0.923 |
794 | -0.918 |
PF | -0.912 |
Token | to |
Token ID | 284 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
angles | +0.661 |
Supervisor | +0.604 |
ortment | +0.603 |
mble | +0.599 |
aina | +0.598 |
Neg logit contributions | |
ield | -0.614 |
vine | -0.611 |
het | -0.599 |
794 | -0.597 |
PF | -0.592 |
Token | sue |
Token ID | 20889 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
angles | +0.080 |
ortment | +0.073 |
Supervisor | +0.073 |
mble | +0.073 |
aina | +0.072 |
Neg logit contributions | |
ield | -0.071 |
vine | -0.071 |
het | -0.070 |
794 | -0.069 |
PF | -0.069 |
Token | low |
Token ID | 1877 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
ield | +0.753 |
vine | +0.751 |
het | +0.737 |
794 | +0.732 |
PF | +0.727 |
Neg logit contributions | |
angles | -0.833 |
ortment | -0.759 |
Supervisor | -0.754 |
mble | -0.753 |
aina | -0.750 |
Token | - |
Token ID | 12 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
angles | +0.050 |
ortment | +0.046 |
Supervisor | +0.046 |
mble | +0.046 |
aina | +0.046 |
Neg logit contributions | |
ield | -0.041 |
vine | -0.041 |
het | -0.040 |
794 | -0.040 |
PF | -0.040 |
Token | profile |
Token ID | 13317 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
ield | +0.026 |
vine | +0.026 |
het | +0.025 |
794 | +0.025 |
PF | +0.025 |
Neg logit contributions | |
angles | -0.035 |
ortment | -0.033 |
Supervisor | -0.033 |
mble | -0.032 |
aina | -0.032 |
Token | comedian |
Token ID | 23139 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
angles | +0.856 |
Supervisor | +0.785 |
ortment | +0.782 |
mble | +0.778 |
aina | +0.775 |
Neg logit contributions | |
ield | -0.736 |
vine | -0.734 |
het | -0.720 |
794 | -0.715 |
PF | -0.711 |
Token | has |
Token ID | 468 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
ield | +0.794 |
vine | +0.791 |
het | +0.779 |
794 | +0.771 |
Finch | +0.765 |
Neg logit contributions | |
angles | -0.808 |
ortment | -0.736 |
mble | -0.730 |
Supervisor | -0.728 |
aina | -0.725 |
Token | long |
Token ID | 890 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ield | +0.610 |
vine | +0.607 |
het | +0.598 |
794 | +0.593 |
PF | +0.587 |
Neg logit contributions | |
angles | -0.656 |
ortment | -0.595 |
aina | -0.592 |
mble | -0.591 |
Supervisor | -0.591 |
Token | been |
Token ID | 587 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
ield | +3.108 |
vine | +3.057 |
het | +3.015 |
drip | +3.002 |
Finch | +2.986 |
Neg logit contributions | |
angles | -3.017 |
mble | -2.745 |
Supervisor | -2.715 |
ortment | -2.705 |
aina | -2.697 |
Token | on |
Token ID | 319 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
ield | +0.482 |
vine | +0.480 |
het | +0.470 |
794 | +0.466 |
PF | +0.463 |
Neg logit contributions | |
angles | -0.642 |
ortment | -0.589 |
Supervisor | -0.589 |
mble | -0.586 |
aina | -0.584 |
Token | the |
Token ID | 262 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
angles | +0.470 |
Supervisor | +0.432 |
ortment | +0.429 |
mble | +0.427 |
aina | +0.426 |
Neg logit contributions | |
ield | -0.409 |
vine | -0.404 |
het | -0.396 |
794 | -0.396 |
PF | -0.394 |
Token | periphery |
Token ID | 46628 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
ield | +0.391 |
vine | +0.389 |
het | +0.382 |
794 | +0.379 |
PF | +0.376 |
Neg logit contributions | |
angles | -0.504 |
Supervisor | -0.462 |
ortment | -0.462 |
mble | -0.459 |
aina | -0.458 |
Token | of |
Token ID | 286 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
ield | +0.784 |
vine | +0.782 |
het | +0.769 |
794 | +0.759 |
PF | +0.753 |
Neg logit contributions | |
angles | -0.988 |
Supervisor | -0.904 |
ortment | -0.904 |
mble | -0.901 |
aina | -0.897 |
Token | who |
Token ID | 508 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
ield | +0.528 |
vine | +0.525 |
het | +0.515 |
794 | +0.511 |
PF | +0.506 |
Neg logit contributions | |
angles | -0.734 |
ortment | -0.677 |
Supervisor | -0.675 |
mble | -0.672 |
aina | -0.670 |
Token | clearly |
Token ID | 4084 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
ield | +0.311 |
vine | +0.309 |
het | +0.304 |
794 | +0.302 |
PF | +0.300 |
Neg logit contributions | |
angles | -0.348 |
ortment | -0.317 |
Supervisor | -0.316 |
mble | -0.315 |
aina | -0.314 |
Token | see |
Token ID | 766 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
angles | +1.150 |
Supervisor | +1.055 |
ortment | +1.049 |
aina | +1.044 |
mble | +1.043 |
Neg logit contributions | |
vine | -0.976 |
ield | -0.976 |
het | -0.953 |
794 | -0.945 |
PF | -0.942 |
Token | a |
Token ID | 257 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
angles | +1.154 |
Supervisor | +1.065 |
ortment | +1.052 |
aina | +1.048 |
mble | +1.041 |
Neg logit contributions | |
vine | -0.950 |
ield | -0.949 |
het | -0.921 |
794 | -0.915 |
PF | -0.906 |
Token | decades |
Token ID | 4647 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
angles | +0.322 |
Supervisor | +0.297 |
ortment | +0.295 |
mble | +0.294 |
aina | +0.293 |
Neg logit contributions | |
vine | -0.254 |
ield | -0.253 |
het | -0.247 |
794 | -0.245 |
PF | -0.243 |
Token | long |
Token ID | 890 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
angles | +0.596 |
Supervisor | +0.547 |
ortment | +0.542 |
aina | +0.541 |
mble | +0.540 |
Neg logit contributions | |
vine | -0.464 |
ield | -0.462 |
het | -0.451 |
794 | -0.449 |
PF | -0.446 |
Token | decline |
Token ID | 7794 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
ield | +3.044 |
vine | +2.996 |
drip | +2.953 |
het | +2.948 |
794 | +2.937 |
Neg logit contributions | |
angles | -3.672 |
ortment | -3.378 |
mble | -3.359 |
arnaev | -3.333 |
aina | -3.328 |
Token | in |
Token ID | 287 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
ield | +0.611 |
vine | +0.609 |
het | +0.599 |
794 | +0.595 |
PF | +0.590 |
Neg logit contributions | |
angles | -0.619 |
Supervisor | -0.562 |
ortment | -0.562 |
mble | -0.559 |
aina | -0.556 |
Token | financial |
Token ID | 3176 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
ield | +0.966 |
vine | +0.955 |
het | +0.944 |
794 | +0.940 |
PF | +0.928 |
Neg logit contributions | |
angles | -1.151 |
ortment | -1.055 |
mble | -1.049 |
Supervisor | -1.048 |
aina | -1.037 |
Token | and |
Token ID | 290 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
ield | +0.606 |
vine | +0.601 |
het | +0.592 |
794 | +0.588 |
PF | +0.585 |
Neg logit contributions | |
angles | -0.674 |
ortment | -0.616 |
mble | -0.615 |
Supervisor | -0.615 |
aina | -0.608 |
Token | employment |
Token ID | 7184 |
Feature activation | -4.80e-03 |
Pos logit contributions | |
ield | +0.323 |
vine | +0.318 |
het | +0.314 |
794 | +0.310 |
PF | +0.308 |
Neg logit contributions | |
angles | -0.417 |
ortment | -0.382 |
mble | -0.381 |
Supervisor | -0.380 |
aina | -0.378 |
Token | . |
Token ID | 13 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
vine | +1.846 |
ield | +1.846 |
794 | +1.811 |
het | +1.802 |
PF | +1.789 |
Neg logit contributions | |
angles | -2.571 |
Supervisor | -2.370 |
ortment | -2.364 |
mble | -2.356 |
aina | -2.347 |
Token | jpg |
Token ID | 9479 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
ield | +0.918 |
vine | +0.914 |
PF | +0.888 |
794 | +0.888 |
het | +0.886 |
Neg logit contributions | |
angles | -1.626 |
Supervisor | -1.520 |
mble | -1.509 |
ortment | -1.505 |
aina | -1.501 |
Token |
|
Token ID | 198 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
ield | +1.204 |
vine | +1.200 |
794 | +1.174 |
het | +1.172 |
PF | +1.170 |
Neg logit contributions | |
angles | -1.402 |
mble | -1.326 |
ortment | -1.285 |
aina | -1.280 |
unlaw | -1.278 |
Token |
|
Token ID | 198 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
ield | +1.471 |
vine | +1.464 |
794 | +1.452 |
het | +1.430 |
PF | +1.426 |
Neg logit contributions | |
angles | -1.828 |
mble | -1.788 |
unlaw | -1.714 |
ortment | -1.708 |
aina | -1.689 |
Token | Email |
Token ID | 15333 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ield | +0.691 |
vine | +0.671 |
het | +0.642 |
PF | +0.641 |
794 | +0.640 |
Neg logit contributions | |
angles | -2.976 |
mble | -2.822 |
Supervisor | -2.817 |
aina | -2.814 |
ortment | -2.812 |
Token |
|
Token ID | 198 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
vine | +2.093 |
ield | +2.075 |
794 | +2.045 |
het | +2.018 |
PF | +2.014 |
Neg logit contributions | |
angles | -2.648 |
mble | -2.588 |
ortment | -2.480 |
unlaw | -2.478 |
arnaev | -2.422 |
Token |
|
Token ID | 198 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
vine | +2.834 |
ield | +2.802 |
794 | +2.781 |
het | +2.725 |
PF | +2.724 |
Neg logit contributions | |
mble | -3.852 |
angles | -3.843 |
unlaw | -3.665 |
mathemat | -3.663 |
ortment | -3.633 |
Token | Dis |
Token ID | 7279 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
vine | +2.786 |
ield | +2.764 |
794 | +2.725 |
PF | +2.721 |
het | +2.682 |
Neg logit contributions | |
angles | -3.671 |
mble | -3.501 |
ortment | -3.472 |
Supervisor | -3.436 |
aina | -3.381 |
Token | ruption |
Token ID | 6417 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
vine | +1.275 |
ield | +1.267 |
het | +1.221 |
794 | +1.174 |
ordes | +1.147 |
Neg logit contributions | |
angles | -3.205 |
mble | -3.090 |
Supervisor | -3.089 |
ortment | -3.047 |
unlaw | -2.995 |
Token | is |
Token ID | 318 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
vine | +2.894 |
ield | +2.856 |
het | +2.798 |
794 | +2.756 |
PF | +2.734 |
Neg logit contributions | |
mble | -3.336 |
angles | -3.313 |
unlaw | -3.147 |
ortment | -3.130 |
Supervisor | -3.061 |
Token | perhaps |
Token ID | 3737 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
vine | +0.710 |
ield | +0.695 |
het | +0.676 |
PF | +0.666 |
794 | +0.666 |
Neg logit contributions | |
angles | -1.864 |
mble | -1.763 |
ortment | -1.740 |
Supervisor | -1.733 |
aina | -1.725 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
ield | +1.011 |
vine | +1.011 |
het | +0.997 |
794 | +0.985 |
Finch | +0.972 |
Neg logit contributions | |
angles | -1.337 |
ortment | -1.224 |
mble | -1.217 |
aina | -1.212 |
Supervisor | -1.200 |
Token | is |
Token ID | 318 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
angles | +0.268 |
ortment | +0.243 |
Supervisor | +0.243 |
mble | +0.242 |
aina | +0.241 |
Neg logit contributions | |
ield | -0.273 |
vine | -0.272 |
het | -0.267 |
794 | -0.265 |
PF | -0.264 |
Token | investigating |
Token ID | 10240 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
angles | +0.237 |
ortment | +0.216 |
Supervisor | +0.215 |
mble | +0.215 |
aina | +0.213 |
Neg logit contributions | |
ield | -0.221 |
vine | -0.219 |
het | -0.216 |
794 | -0.214 |
PF | -0.213 |
Token | . |
Token ID | 13 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
angles | +1.065 |
Supervisor | +0.983 |
mble | +0.971 |
aina | +0.965 |
ortment | +0.962 |
Neg logit contributions | |
ield | -0.918 |
vine | -0.907 |
het | -0.895 |
794 | -0.888 |
PF | -0.884 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
angles | +0.081 |
ortment | +0.073 |
mble | +0.073 |
Supervisor | +0.073 |
aina | +0.073 |
Neg logit contributions | |
ield | -0.081 |
vine | -0.081 |
het | -0.080 |
794 | -0.079 |
PF | -0.079 |
Token | Long |
Token ID | 14617 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
angles | +1.690 |
ortment | +1.534 |
Supervisor | +1.534 |
aina | +1.532 |
mble | +1.522 |
Neg logit contributions | |
ield | -1.349 |
vine | -1.334 |
het | -1.311 |
794 | -1.295 |
drip | -1.287 |
Token | view |
Token ID | 1177 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
vine | +3.373 |
ield | +3.367 |
het | +3.279 |
794 | +3.234 |
PF | +3.227 |
Neg logit contributions | |
angles | -3.848 |
mble | -3.591 |
ortment | -3.513 |
Supervisor | -3.466 |
aina | -3.460 |
Token | , |
Token ID | 11 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
angles | +0.970 |
Supervisor | +0.881 |
ortment | +0.878 |
mble | +0.872 |
aina | +0.869 |
Neg logit contributions | |
ield | -1.003 |
vine | -1.000 |
het | -0.983 |
794 | -0.976 |
PF | -0.970 |
Token | Wash |
Token ID | 22892 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
angles | +0.915 |
Supervisor | +0.842 |
ortment | +0.834 |
mble | +0.828 |
aina | +0.828 |
Neg logit contributions | |
ield | -0.853 |
vine | -0.848 |
het | -0.834 |
794 | -0.827 |
PF | -0.823 |
Token | ., |
Token ID | 1539 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
angles | +1.293 |
Supervisor | +1.175 |
ortment | +1.174 |
mble | +1.167 |
aina | +1.163 |
Neg logit contributions | |
ield | -1.268 |
vine | -1.264 |
het | -1.241 |
794 | -1.233 |
PF | -1.226 |
Token | police |
Token ID | 1644 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
angles | +1.800 |
Supervisor | +1.641 |
ortment | +1.622 |
mble | +1.613 |
aina | +1.608 |
Neg logit contributions | |
ield | -1.973 |
vine | -1.967 |
het | -1.933 |
794 | -1.918 |
PF | -1.910 |