Token | 14 |
Token ID | 1415 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
digitally | +0.611 |
MTV | +0.602 |
spaced | +0.601 |
patches | +0.596 |
THR | +0.596 |
Neg logit contributions | |
anyahu | -0.871 |
jriwal | -0.865 |
DonaldTrump | -0.840 |
uphem | -0.829 |
owa | -0.806 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
anyahu | +0.383 |
jriwal | +0.378 |
DonaldTrump | +0.370 |
uphem | +0.362 |
owa | +0.353 |
Neg logit contributions | |
digitally | -0.358 |
spaced | -0.354 |
MTV | -0.351 |
THR | -0.351 |
CST | -0.347 |
Token |
|
Token ID | 198 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
anyahu | +0.782 |
jriwal | +0.774 |
DonaldTrump | +0.761 |
uphem | +0.750 |
owa | +0.736 |
Neg logit contributions | |
digitally | -0.351 |
spaced | -0.345 |
MTV | -0.340 |
THR | -0.337 |
patches | -0.336 |
Token |
|
Token ID | 198 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
digitally | +0.209 |
MTV | +0.204 |
spaced | +0.203 |
THR | +0.202 |
CST | +0.201 |
Neg logit contributions | |
anyahu | -0.537 |
jriwal | -0.535 |
DonaldTrump | -0.525 |
uphem | -0.514 |
owa | -0.503 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
digitally | +1.129 |
spaced | +1.110 |
MTV | +1.109 |
patches | +1.100 |
CST | +1.095 |
Neg logit contributions | |
anyahu | -1.721 |
jriwal | -1.706 |
DonaldTrump | -1.671 |
uphem | -1.640 |
owa | -1.596 |
Token | Share |
Token ID | 8734 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
anyahu | +0.916 |
jriwal | +0.906 |
DonaldTrump | +0.898 |
uphem | +0.888 |
owa | +0.864 |
Neg logit contributions | |
digitally | -0.672 |
THR | -0.660 |
spaced | -0.655 |
MTV | -0.649 |
Warwick | -0.648 |
Token | this |
Token ID | 428 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
anyahu | +6.862 |
DonaldTrump | +6.647 |
jriwal | +6.603 |
uphem | +6.389 |
owa | +6.370 |
Neg logit contributions | |
spaced | -4.917 |
THR | -4.911 |
CEPT | -4.886 |
Warwick | -4.803 |
Constantin | -4.762 |
Token | comment |
Token ID | 2912 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
anyahu | +0.075 |
jriwal | +0.074 |
DonaldTrump | +0.073 |
uphem | +0.072 |
owa | +0.070 |
Neg logit contributions | |
digitally | -0.036 |
spaced | -0.035 |
MTV | -0.035 |
CST | -0.034 |
THR | -0.034 |
Token | on |
Token ID | 319 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
anyahu | +1.959 |
jriwal | +1.947 |
DonaldTrump | +1.909 |
uphem | +1.870 |
owa | +1.831 |
Neg logit contributions | |
digitally | -0.974 |
spaced | -0.962 |
MTV | -0.955 |
THR | -0.948 |
patches | -0.943 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
digitally | +0.546 |
MTV | +0.536 |
spaced | +0.532 |
CST | +0.531 |
dot | +0.525 |
Neg logit contributions | |
anyahu | -1.539 |
jriwal | -1.530 |
DonaldTrump | -1.493 |
uphem | -1.470 |
owa | -1.454 |
Token | Share |
Token ID | 8734 |
Feature activation | +6.57e+00 |
Pos logit contributions | |
anyahu | +1.542 |
jriwal | +1.520 |
DonaldTrump | +1.506 |
uphem | +1.470 |
owa | +1.443 |
Neg logit contributions | |
digitally | -0.990 |
spaced | -0.970 |
THR | -0.964 |
MTV | -0.963 |
patches | -0.947 |
Token | : |
Token ID | 25 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
anyahu | +1.610 |
jriwal | +1.593 |
DonaldTrump | +1.567 |
uphem | +1.524 |
owa | +1.502 |
Neg logit contributions | |
digitally | -1.376 |
spaced | -1.349 |
MTV | -1.339 |
CST | -1.337 |
THR | -1.332 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
anyahu | +1.265 |
jriwal | +1.252 |
DonaldTrump | +1.232 |
uphem | +1.207 |
owa | +1.183 |
Neg logit contributions | |
digitally | -0.713 |
spaced | -0.701 |
MTV | -0.694 |
THR | -0.688 |
patches | -0.688 |
Token |
|
Token ID | 198 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
anyahu | +2.378 |
jriwal | +2.342 |
DonaldTrump | +2.316 |
uphem | +2.276 |
owa | +2.241 |
Neg logit contributions | |
digitally | -1.110 |
spaced | -1.093 |
MTV | -1.071 |
patches | -1.067 |
CST | -1.059 |
Token | 13 |
Token ID | 1485 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
anyahu | +1.861 |
jriwal | +1.832 |
DonaldTrump | +1.816 |
uphem | +1.765 |
owa | +1.731 |
Neg logit contributions | |
digitally | -1.345 |
spaced | -1.324 |
MTV | -1.308 |
patches | -1.304 |
CST | -1.302 |
Token | PH |
Token ID | 9370 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
anyahu | +0.408 |
jriwal | +0.404 |
DonaldTrump | +0.397 |
uphem | +0.389 |
owa | +0.382 |
Neg logit contributions | |
digitally | -0.215 |
spaced | -0.211 |
MTV | -0.210 |
THR | -0.208 |
patches | -0.207 |
Token | OTOS |
Token ID | 33291 |
Feature activation | +8.48e+00 |
Pos logit contributions | |
anyahu | +3.120 |
jriwal | +3.080 |
DonaldTrump | +3.047 |
uphem | +2.987 |
owa | +2.916 |
Neg logit contributions | |
digitally | -1.587 |
spaced | -1.566 |
MTV | -1.546 |
CST | -1.537 |
patches | -1.532 |
Token | Donald |
Token ID | 3759 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
DonaldTrump | +6.395 |
anyahu | +6.026 |
uphem | +5.725 |
owa | +5.720 |
jriwal | +5.598 |
Neg logit contributions | |
spaced | -5.728 |
digitally | -5.657 |
patches | -5.515 |
breakup | -5.416 |
CST | -5.392 |
Token | Trump |
Token ID | 1301 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
anyahu | +1.553 |
jriwal | +1.525 |
DonaldTrump | +1.512 |
uphem | +1.426 |
owa | +1.413 |
Neg logit contributions | |
digitally | -2.146 |
spaced | -2.133 |
CST | -2.100 |
patches | -2.096 |
MTV | -2.092 |
Token | 's |
Token ID | 338 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
anyahu | +0.707 |
jriwal | +0.699 |
DonaldTrump | +0.690 |
uphem | +0.674 |
owa | +0.661 |
Neg logit contributions | |
digitally | -0.436 |
spaced | -0.430 |
MTV | -0.426 |
CST | -0.422 |
THR | -0.421 |
Token | potential |
Token ID | 2785 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
anyahu | +1.745 |
jriwal | +1.722 |
DonaldTrump | +1.700 |
uphem | +1.657 |
owa | +1.629 |
Neg logit contributions | |
digitally | -1.228 |
spaced | -1.207 |
MTV | -1.196 |
THR | -1.190 |
CST | -1.188 |
Token | running |
Token ID | 2491 |
Feature activation | +7.24e-01 |
Pos logit contributions | |
anyahu | +0.304 |
jriwal | +0.302 |
DonaldTrump | +0.296 |
uphem | +0.291 |
owa | +0.284 |
Neg logit contributions | |
digitally | -0.163 |
spaced | -0.160 |
MTV | -0.160 |
THR | -0.158 |
patches | -0.157 |
Token | 55 |
Token ID | 2816 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
digitally | +0.713 |
MTV | +0.705 |
spaced | +0.701 |
THR | +0.696 |
CST | +0.693 |
Neg logit contributions | |
anyahu | -1.254 |
jriwal | -1.251 |
DonaldTrump | -1.206 |
uphem | -1.196 |
owa | -1.164 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
anyahu | +0.099 |
jriwal | +0.098 |
DonaldTrump | +0.096 |
uphem | +0.093 |
owa | +0.091 |
Neg logit contributions | |
digitally | -0.088 |
spaced | -0.087 |
MTV | -0.086 |
THR | -0.086 |
patches | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
anyahu | +0.854 |
jriwal | +0.844 |
DonaldTrump | +0.831 |
uphem | +0.818 |
owa | +0.803 |
Neg logit contributions | |
digitally | -0.376 |
spaced | -0.368 |
MTV | -0.364 |
THR | -0.360 |
patches | -0.360 |
Token |
|
Token ID | 198 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
digitally | +0.298 |
MTV | +0.292 |
spaced | +0.290 |
THR | +0.289 |
CST | +0.287 |
Neg logit contributions | |
anyahu | -0.778 |
jriwal | -0.774 |
DonaldTrump | -0.759 |
uphem | -0.744 |
owa | -0.727 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
digitally | +1.260 |
MTV | +1.241 |
spaced | +1.240 |
patches | +1.225 |
CST | +1.224 |
Neg logit contributions | |
anyahu | -1.923 |
jriwal | -1.910 |
DonaldTrump | -1.865 |
uphem | -1.832 |
owa | -1.782 |
Token | Share |
Token ID | 8734 |
Feature activation | +7.61e+00 |
Pos logit contributions | |
anyahu | +1.000 |
jriwal | +0.986 |
DonaldTrump | +0.977 |
uphem | +0.961 |
owa | +0.939 |
Neg logit contributions | |
digitally | -0.679 |
spaced | -0.663 |
THR | -0.662 |
MTV | -0.656 |
patches | -0.654 |
Token | this |
Token ID | 428 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
anyahu | +6.560 |
jriwal | +6.360 |
uphem | +6.231 |
DonaldTrump | +6.188 |
owa | +6.136 |
Neg logit contributions | |
CEPT | -4.040 |
Constantin | -3.990 |
MTV | -3.964 |
spaced | -3.950 |
THR | -3.938 |
Token | comment |
Token ID | 2912 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
digitally | +0.165 |
spaced | +0.162 |
MTV | +0.161 |
THR | +0.160 |
patches | +0.159 |
Neg logit contributions | |
anyahu | -0.322 |
jriwal | -0.319 |
DonaldTrump | -0.313 |
uphem | -0.307 |
owa | -0.302 |
Token | on |
Token ID | 319 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
anyahu | +2.555 |
jriwal | +2.523 |
DonaldTrump | +2.481 |
uphem | +2.439 |
owa | +2.396 |
Neg logit contributions | |
digitally | -1.303 |
spaced | -1.291 |
MTV | -1.276 |
patches | -1.264 |
THR | -1.260 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
digitally | +0.004 |
MTV | +0.004 |
spaced | +0.004 |
THR | +0.004 |
CST | +0.004 |
Neg logit contributions | |
anyahu | -0.011 |
jriwal | -0.011 |
DonaldTrump | -0.011 |
uphem | -0.011 |
owa | -0.011 |
Token | Share |
Token ID | 8734 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
anyahu | +2.244 |
jriwal | +2.176 |
DonaldTrump | +2.175 |
uphem | +2.119 |
owa | +2.099 |
Neg logit contributions | |
digitally | -1.579 |
spaced | -1.542 |
THR | -1.518 |
MTV | -1.517 |
patches | -1.515 |
Token | 27 |
Token ID | 1983 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
digitally | +1.030 |
MTV | +1.020 |
spaced | +1.012 |
THR | +1.006 |
CST | +1.001 |
Neg logit contributions | |
jriwal | -1.733 |
anyahu | -1.731 |
DonaldTrump | -1.657 |
uphem | -1.647 |
owa | -1.598 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
anyahu | +0.613 |
jriwal | +0.601 |
DonaldTrump | +0.592 |
uphem | +0.578 |
owa | +0.564 |
Neg logit contributions | |
digitally | -0.566 |
spaced | -0.557 |
MTV | -0.553 |
THR | -0.551 |
patches | -0.550 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
anyahu | +1.292 |
jriwal | +1.276 |
DonaldTrump | +1.257 |
uphem | +1.237 |
owa | +1.216 |
Neg logit contributions | |
digitally | -0.580 |
spaced | -0.569 |
MTV | -0.561 |
THR | -0.556 |
patches | -0.555 |
Token |
|
Token ID | 198 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
digitally | +0.487 |
MTV | +0.479 |
THR | +0.473 |
spaced | +0.473 |
CST | +0.469 |
Neg logit contributions | |
anyahu | -1.326 |
jriwal | -1.321 |
DonaldTrump | -1.296 |
uphem | -1.268 |
owa | -1.237 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
digitally | +1.412 |
MTV | +1.400 |
spaced | +1.390 |
patches | +1.378 |
CST | +1.377 |
Neg logit contributions | |
jriwal | -2.314 |
anyahu | -2.308 |
DonaldTrump | -2.245 |
uphem | -2.210 |
owa | -2.143 |
Token | Share |
Token ID | 8734 |
Feature activation | +7.36e+00 |
Pos logit contributions | |
anyahu | +1.126 |
jriwal | +1.106 |
DonaldTrump | +1.093 |
uphem | +1.073 |
owa | +1.052 |
Neg logit contributions | |
digitally | -0.743 |
spaced | -0.726 |
THR | -0.722 |
MTV | -0.720 |
patches | -0.717 |
Token | this |
Token ID | 428 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
anyahu | +6.405 |
jriwal | +6.180 |
uphem | +6.086 |
DonaldTrump | +6.055 |
owa | +6.006 |
Neg logit contributions | |
CEPT | -3.829 |
Constantin | -3.813 |
spaced | -3.804 |
MTV | -3.786 |
digitally | -3.786 |
Token | comment |
Token ID | 2912 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
anyahu | +0.174 |
jriwal | +0.172 |
DonaldTrump | +0.169 |
uphem | +0.166 |
owa | +0.163 |
Neg logit contributions | |
digitally | -0.090 |
spaced | -0.088 |
MTV | -0.088 |
THR | -0.087 |
patches | -0.087 |
Token | on |
Token ID | 319 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
anyahu | +2.695 |
jriwal | +2.658 |
DonaldTrump | +2.616 |
uphem | +2.573 |
owa | +2.530 |
Neg logit contributions | |
digitally | -1.384 |
spaced | -1.372 |
MTV | -1.355 |
patches | -1.342 |
THR | -1.337 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
digitally | +0.023 |
MTV | +0.023 |
spaced | +0.023 |
THR | +0.023 |
CST | +0.022 |
Neg logit contributions | |
anyahu | -0.060 |
jriwal | -0.060 |
DonaldTrump | -0.059 |
uphem | -0.058 |
owa | -0.057 |
Token | Share |
Token ID | 8734 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
anyahu | +1.966 |
jriwal | +1.915 |
DonaldTrump | +1.906 |
uphem | +1.861 |
owa | +1.839 |
Neg logit contributions | |
digitally | -1.323 |
spaced | -1.296 |
MTV | -1.274 |
THR | -1.272 |
patches | -1.271 |
Token | Hide |
Token ID | 10415 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
digitally | +0.257 |
MTV | +0.252 |
spaced | +0.251 |
THR | +0.249 |
CST | +0.248 |
Neg logit contributions | |
anyahu | -0.482 |
jriwal | -0.479 |
DonaldTrump | -0.468 |
uphem | -0.460 |
owa | -0.450 |
Token | Caption |
Token ID | 11260 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
anyahu | +1.414 |
jriwal | +1.390 |
DonaldTrump | +1.355 |
uphem | +1.309 |
owa | +1.271 |
Neg logit contributions | |
digitally | -1.976 |
spaced | -1.957 |
MTV | -1.946 |
THR | -1.936 |
patches | -1.934 |
Token | 4 |
Token ID | 604 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
anyahu | +4.400 |
DonaldTrump | +4.353 |
jriwal | +4.228 |
uphem | +4.186 |
owa | +4.164 |
Neg logit contributions | |
digitally | -3.006 |
spaced | -3.002 |
patches | -2.931 |
breakup | -2.897 |
checkpoints | -2.869 |
Token | of |
Token ID | 286 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
anyahu | +0.521 |
jriwal | +0.519 |
DonaldTrump | +0.507 |
uphem | +0.497 |
owa | +0.484 |
Neg logit contributions | |
digitally | -0.355 |
MTV | -0.348 |
spaced | -0.347 |
THR | -0.345 |
patches | -0.343 |
Token | 15 |
Token ID | 1315 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
anyahu | +0.950 |
jriwal | +0.944 |
DonaldTrump | +0.921 |
uphem | +0.907 |
owa | +0.884 |
Neg logit contributions | |
digitally | -0.554 |
MTV | -0.545 |
spaced | -0.543 |
THR | -0.540 |
CST | -0.536 |
Token | Photos |
Token ID | 9434 |
Feature activation | +7.23e+00 |
Pos logit contributions | |
anyahu | +0.826 |
jriwal | +0.817 |
DonaldTrump | +0.798 |
uphem | +0.781 |
owa | +0.761 |
Neg logit contributions | |
digitally | -0.673 |
MTV | -0.664 |
spaced | -0.660 |
THR | -0.658 |
CST | -0.655 |
Token | : |
Token ID | 25 |
Feature activation | +4.42e+00 |
Pos logit contributions | |
anyahu | +5.766 |
DonaldTrump | +5.564 |
uphem | +5.542 |
jriwal | +5.531 |
owa | +5.343 |
Neg logit contributions | |
spaced | -4.377 |
digitally | -4.345 |
patches | -4.324 |
breakup | -4.304 |
CST | -4.254 |
Token | Carly |
Token ID | 35982 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
anyahu | +3.625 |
DonaldTrump | +3.584 |
jriwal | +3.532 |
uphem | +3.448 |
owa | +3.429 |
Neg logit contributions | |
digitally | -2.727 |
spaced | -2.719 |
patches | -2.631 |
CST | -2.575 |
MTV | -2.573 |
Token | Fiorina |
Token ID | 49028 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
anyahu | +1.063 |
jriwal | +1.049 |
DonaldTrump | +1.017 |
uphem | +0.983 |
owa | +0.957 |
Neg logit contributions | |
digitally | -1.310 |
spaced | -1.296 |
MTV | -1.283 |
THR | -1.278 |
patches | -1.272 |
Token | 's |
Token ID | 338 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
anyahu | +1.677 |
jriwal | +1.657 |
DonaldTrump | +1.627 |
uphem | +1.592 |
owa | +1.570 |
Neg logit contributions | |
digitally | -1.085 |
spaced | -1.069 |
MTV | -1.058 |
CST | -1.048 |
THR | -1.047 |
Token | political |
Token ID | 1964 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
anyahu | +1.269 |
jriwal | +1.253 |
DonaldTrump | +1.233 |
uphem | +1.204 |
owa | +1.185 |
Neg logit contributions | |
digitally | -0.831 |
spaced | -0.816 |
MTV | -0.809 |
THR | -0.804 |
patches | -0.803 |
Token | 19 |
Token ID | 1129 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
digitally | +0.810 |
MTV | +0.800 |
spaced | +0.796 |
THR | +0.790 |
CST | +0.786 |
Neg logit contributions | |
anyahu | -1.307 |
jriwal | -1.305 |
DonaldTrump | -1.253 |
uphem | -1.242 |
owa | -1.207 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
anyahu | +0.134 |
jriwal | +0.133 |
DonaldTrump | +0.130 |
uphem | +0.127 |
owa | +0.124 |
Neg logit contributions | |
digitally | -0.120 |
spaced | -0.118 |
MTV | -0.117 |
THR | -0.117 |
patches | -0.116 |
Token |
|
Token ID | 198 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
anyahu | +1.290 |
jriwal | +1.274 |
DonaldTrump | +1.256 |
uphem | +1.235 |
owa | +1.214 |
Neg logit contributions | |
digitally | -0.581 |
spaced | -0.569 |
MTV | -0.561 |
THR | -0.556 |
patches | -0.556 |
Token |
|
Token ID | 198 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
digitally | +0.232 |
MTV | +0.227 |
spaced | +0.226 |
THR | +0.225 |
CST | +0.223 |
Neg logit contributions | |
anyahu | -0.600 |
jriwal | -0.597 |
DonaldTrump | -0.586 |
uphem | -0.574 |
owa | -0.561 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
digitally | +1.074 |
MTV | +1.058 |
spaced | +1.056 |
CST | +1.043 |
patches | +1.043 |
Neg logit contributions | |
anyahu | -1.662 |
jriwal | -1.656 |
DonaldTrump | -1.612 |
uphem | -1.585 |
owa | -1.541 |
Token | Share |
Token ID | 8734 |
Feature activation | +7.23e+00 |
Pos logit contributions | |
anyahu | +1.240 |
jriwal | +1.216 |
DonaldTrump | +1.206 |
uphem | +1.182 |
owa | +1.159 |
Neg logit contributions | |
digitally | -0.827 |
spaced | -0.809 |
THR | -0.805 |
MTV | -0.800 |
patches | -0.799 |
Token | this |
Token ID | 428 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
anyahu | +6.305 |
jriwal | +6.095 |
uphem | +5.998 |
DonaldTrump | +5.965 |
owa | +5.908 |
Neg logit contributions | |
CEPT | -3.754 |
Constantin | -3.746 |
spaced | -3.727 |
MTV | -3.717 |
digitally | -3.706 |
Token | comment |
Token ID | 2912 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
digitally | +0.076 |
spaced | +0.074 |
MTV | +0.074 |
THR | +0.074 |
patches | +0.073 |
Neg logit contributions | |
anyahu | -0.148 |
jriwal | -0.146 |
DonaldTrump | -0.144 |
uphem | -0.141 |
owa | -0.138 |
Token | on |
Token ID | 319 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
anyahu | +2.269 |
jriwal | +2.240 |
DonaldTrump | +2.204 |
uphem | +2.167 |
owa | +2.128 |
Neg logit contributions | |
digitally | -1.144 |
spaced | -1.132 |
MTV | -1.119 |
patches | -1.108 |
THR | -1.105 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
digitally | +0.075 |
MTV | +0.073 |
spaced | +0.073 |
THR | +0.072 |
CST | +0.072 |
Neg logit contributions | |
anyahu | -0.194 |
jriwal | -0.193 |
DonaldTrump | -0.189 |
uphem | -0.187 |
owa | -0.183 |
Token | Share |
Token ID | 8734 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
anyahu | +1.918 |
jriwal | +1.867 |
DonaldTrump | +1.860 |
uphem | +1.814 |
owa | +1.793 |
Neg logit contributions | |
digitally | -1.300 |
spaced | -1.274 |
MTV | -1.252 |
patches | -1.250 |
THR | -1.248 |
Token | something |
Token ID | 1223 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
digitally | +0.159 |
spaced | +0.155 |
MTV | +0.154 |
THR | +0.152 |
CST | +0.151 |
Neg logit contributions | |
anyahu | -0.319 |
jriwal | -0.317 |
DonaldTrump | -0.310 |
uphem | -0.305 |
owa | -0.299 |
Token | that |
Token ID | 326 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
anyahu | +0.990 |
jriwal | +0.981 |
DonaldTrump | +0.962 |
uphem | +0.944 |
owa | +0.928 |
Neg logit contributions | |
digitally | -0.542 |
spaced | -0.530 |
MTV | -0.529 |
THR | -0.522 |
CST | -0.521 |
Token | actually |
Token ID | 1682 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
digitally | +0.192 |
spaced | +0.189 |
MTV | +0.188 |
THR | +0.186 |
patches | +0.185 |
Neg logit contributions | |
anyahu | -0.300 |
jriwal | -0.297 |
DonaldTrump | -0.291 |
uphem | -0.285 |
owa | -0.279 |
Token | worked |
Token ID | 3111 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
digitally | +0.891 |
spaced | +0.877 |
MTV | +0.869 |
THR | +0.864 |
patches | +0.863 |
Neg logit contributions | |
anyahu | -1.422 |
jriwal | -1.407 |
DonaldTrump | -1.382 |
uphem | -1.355 |
owa | -1.320 |
Token | out |
Token ID | 503 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
digitally | +0.911 |
spaced | +0.885 |
MTV | +0.878 |
THR | +0.874 |
patches | +0.868 |
Neg logit contributions | |
anyahu | -2.310 |
jriwal | -2.294 |
DonaldTrump | -2.262 |
uphem | -2.222 |
owa | -2.173 |
Token | Administrator |
Token ID | 22998 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
digitally | +0.186 |
spaced | +0.182 |
MTV | +0.181 |
THR | +0.180 |
patches | +0.179 |
Neg logit contributions | |
anyahu | -0.317 |
jriwal | -0.314 |
DonaldTrump | -0.308 |
uphem | -0.302 |
owa | -0.296 |
Token | Laws |
Token ID | 21153 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
jriwal | +6.642 |
anyahu | +6.555 |
DonaldTrump | +6.301 |
uphem | +6.297 |
owa | +6.280 |
Neg logit contributions | |
spaced | -3.122 |
digitally | -3.096 |
breakup | -3.081 |
CEPT | -2.974 |
COMPLE | -2.962 |
Token | change |
Token ID | 1487 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
digitally | +0.263 |
spaced | +0.259 |
MTV | +0.257 |
THR | +0.255 |
patches | +0.254 |
Neg logit contributions | |
anyahu | -0.423 |
jriwal | -0.418 |
DonaldTrump | -0.411 |
uphem | -0.403 |
owa | -0.394 |
Token | depending |
Token ID | 6906 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
anyahu | +0.199 |
jriwal | +0.198 |
DonaldTrump | +0.194 |
uphem | +0.190 |
owa | +0.186 |
Neg logit contributions | |
digitally | -0.114 |
spaced | -0.112 |
MTV | -0.111 |
THR | -0.110 |
CST | -0.110 |
Token | on |
Token ID | 319 |
Feature activation | +9.78e-01 |
Pos logit contributions | |
digitally | +0.006 |
spaced | +0.006 |
MTV | +0.006 |
THR | +0.006 |
CST | +0.006 |
Neg logit contributions | |
anyahu | -0.012 |
jriwal | -0.012 |
DonaldTrump | -0.012 |
uphem | -0.011 |
owa | -0.011 |
Token | who |
Token ID | 508 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
anyahu | +0.922 |
jriwal | +0.918 |
DonaldTrump | +0.899 |
uphem | +0.883 |
owa | +0.868 |
Neg logit contributions | |
digitally | -0.494 |
spaced | -0.485 |
MTV | -0.481 |
THR | -0.476 |
CST | -0.473 |
Token | Share |
Token ID | 8734 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
anyahu | +0.916 |
jriwal | +0.906 |
DonaldTrump | +0.898 |
uphem | +0.888 |
owa | +0.864 |
Neg logit contributions | |
digitally | -0.672 |
THR | -0.660 |
spaced | -0.655 |
MTV | -0.649 |
Warwick | -0.648 |
Token | this |
Token ID | 428 |
Feature activation | +7.55e-02 |
Pos logit contributions | |
anyahu | +6.862 |
DonaldTrump | +6.647 |
jriwal | +6.603 |
uphem | +6.389 |
owa | +6.370 |
Neg logit contributions | |
spaced | -4.917 |
THR | -4.911 |
CEPT | -4.886 |
Warwick | -4.803 |
Constantin | -4.762 |
Token | comment |
Token ID | 2912 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
anyahu | +0.075 |
jriwal | +0.074 |
DonaldTrump | +0.073 |
uphem | +0.072 |
owa | +0.070 |
Neg logit contributions | |
digitally | -0.036 |
spaced | -0.035 |
MTV | -0.035 |
CST | -0.034 |
THR | -0.034 |
Token | on |
Token ID | 319 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
anyahu | +1.959 |
jriwal | +1.947 |
DonaldTrump | +1.909 |
uphem | +1.870 |
owa | +1.831 |
Neg logit contributions | |
digitally | -0.974 |
spaced | -0.962 |
MTV | -0.955 |
THR | -0.948 |
patches | -0.943 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
digitally | +0.546 |
MTV | +0.536 |
spaced | +0.532 |
CST | +0.531 |
dot | +0.525 |
Neg logit contributions | |
anyahu | -1.539 |
jriwal | -1.530 |
DonaldTrump | -1.493 |
uphem | -1.470 |
owa | -1.454 |
Token | Share |
Token ID | 8734 |
Feature activation | +6.57e+00 |
Pos logit contributions | |
anyahu | +1.542 |
jriwal | +1.520 |
DonaldTrump | +1.506 |
uphem | +1.470 |
owa | +1.443 |
Neg logit contributions | |
digitally | -0.990 |
spaced | -0.970 |
THR | -0.964 |
MTV | -0.963 |
patches | -0.947 |
Token | this |
Token ID | 428 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
anyahu | +6.201 |
jriwal | +6.016 |
DonaldTrump | +6.011 |
uphem | +5.835 |
owa | +5.658 |
Neg logit contributions | |
spaced | -3.234 |
digitally | -3.194 |
THR | -3.185 |
MTV | -3.164 |
Warwick | -3.145 |
Token | comment |
Token ID | 2912 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
digitally | +0.556 |
spaced | +0.544 |
MTV | +0.540 |
THR | +0.534 |
CST | +0.533 |
Neg logit contributions | |
anyahu | -1.132 |
jriwal | -1.122 |
DonaldTrump | -1.103 |
uphem | -1.083 |
owa | -1.060 |
Token | on |
Token ID | 319 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
anyahu | +2.337 |
jriwal | +2.315 |
DonaldTrump | +2.274 |
uphem | +2.233 |
owa | +2.190 |
Neg logit contributions | |
digitally | -1.128 |
spaced | -1.110 |
MTV | -1.103 |
THR | -1.092 |
patches | -1.087 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
digitally | +0.307 |
spaced | +0.301 |
MTV | +0.300 |
THR | +0.296 |
CST | +0.296 |
Neg logit contributions | |
anyahu | -0.601 |
jriwal | -0.596 |
DonaldTrump | -0.585 |
uphem | -0.575 |
owa | -0.563 |
Token |
|
Token ID | 198 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
anyahu | +0.680 |
jriwal | +0.674 |
DonaldTrump | +0.664 |
uphem | +0.652 |
owa | +0.640 |
Neg logit contributions | |
digitally | -0.321 |
spaced | -0.312 |
MTV | -0.309 |
THR | -0.308 |
patches | -0.305 |
Token | thread |
Token ID | 4704 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
digitally | +0.511 |
spaced | +0.501 |
MTV | +0.499 |
THR | +0.494 |
patches | +0.491 |
Neg logit contributions | |
anyahu | -0.923 |
jriwal | -0.913 |
DonaldTrump | -0.895 |
uphem | -0.879 |
owa | -0.863 |
Token | is |
Token ID | 318 |
Feature activation | -5.94e-03 |
Pos logit contributions | |
digitally | +0.241 |
spaced | +0.238 |
MTV | +0.237 |
THR | +0.235 |
CST | +0.234 |
Neg logit contributions | |
anyahu | -0.369 |
jriwal | -0.365 |
DonaldTrump | -0.357 |
uphem | -0.351 |
owa | -0.344 |
Token | here |
Token ID | 994 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
digitally | +0.003 |
spaced | +0.003 |
MTV | +0.003 |
THR | +0.003 |
patches | +0.003 |
Neg logit contributions | |
anyahu | -0.005 |
jriwal | -0.005 |
DonaldTrump | -0.005 |
uphem | -0.005 |
owa | -0.005 |
Token | : |
Token ID | 25 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
digitally | +0.614 |
spaced | +0.602 |
MTV | +0.598 |
THR | +0.592 |
CST | +0.592 |
Neg logit contributions | |
anyahu | -1.046 |
jriwal | -1.036 |
DonaldTrump | -1.014 |
uphem | -0.996 |
owa | -0.973 |
Token | ( |
Token ID | 357 |
Feature activation | -9.53e-01 |
Pos logit contributions | |
anyahu | +0.935 |
jriwal | +0.923 |
DonaldTrump | +0.906 |
uphem | +0.886 |
owa | +0.870 |
Neg logit contributions | |
digitally | -0.728 |
spaced | -0.717 |
MTV | -0.712 |
THR | -0.706 |
patches | -0.701 |
Token | https |
Token ID | 3740 |
Feature activation | +6.34e+00 |
Pos logit contributions | |
digitally | +0.577 |
spaced | +0.568 |
MTV | +0.566 |
patches | +0.561 |
THR | +0.561 |
Neg logit contributions | |
anyahu | -0.814 |
jriwal | -0.808 |
DonaldTrump | -0.790 |
uphem | -0.774 |
owa | -0.754 |
Token | :// |
Token ID | 1378 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
anyahu | +4.839 |
DonaldTrump | +4.683 |
jriwal | +4.678 |
uphem | +4.541 |
owa | +4.508 |
Neg logit contributions | |
spaced | -4.257 |
digitally | -4.245 |
patches | -4.123 |
MTV | -4.117 |
CST | -4.101 |
Token | www |
Token ID | 2503 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
anyahu | +1.369 |
jriwal | +1.351 |
DonaldTrump | +1.313 |
uphem | +1.288 |
owa | +1.263 |
Neg logit contributions | |
digitally | -1.433 |
spaced | -1.413 |
MTV | -1.408 |
THR | -1.397 |
CST | -1.392 |
Token | . |
Token ID | 13 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
digitally | +0.169 |
MTV | +0.166 |
spaced | +0.165 |
THR | +0.165 |
CST | +0.163 |
Neg logit contributions | |
anyahu | -0.327 |
jriwal | -0.326 |
DonaldTrump | -0.319 |
uphem | -0.312 |
owa | -0.305 |
Token | cor |
Token ID | 10215 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
anyahu | +0.541 |
jriwal | +0.533 |
DonaldTrump | +0.522 |
uphem | +0.509 |
owa | +0.497 |
Neg logit contributions | |
digitally | -0.541 |
spaced | -0.533 |
MTV | -0.531 |
THR | -0.527 |
patches | -0.525 |
Token | vette |
Token ID | 33573 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
anyahu | +0.458 |
jriwal | +0.452 |
DonaldTrump | +0.441 |
uphem | +0.431 |
owa | +0.420 |
Neg logit contributions | |
digitally | -0.477 |
spaced | -0.471 |
MTV | -0.470 |
THR | -0.467 |
CST | -0.465 |
Token | recognition |
Token ID | 9465 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
anyahu | +0.474 |
jriwal | +0.469 |
DonaldTrump | +0.460 |
uphem | +0.451 |
owa | +0.441 |
Neg logit contributions | |
digitally | -0.296 |
spaced | -0.291 |
MTV | -0.289 |
THR | -0.287 |
patches | -0.286 |
Token | technology |
Token ID | 3037 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
anyahu | +0.518 |
jriwal | +0.513 |
DonaldTrump | +0.503 |
uphem | +0.494 |
owa | +0.483 |
Neg logit contributions | |
digitally | -0.295 |
spaced | -0.289 |
MTV | -0.288 |
THR | -0.285 |
patches | -0.285 |
Token | : |
Token ID | 25 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
anyahu | +0.612 |
jriwal | +0.606 |
DonaldTrump | +0.595 |
uphem | +0.583 |
owa | +0.571 |
Neg logit contributions | |
digitally | -0.364 |
spaced | -0.357 |
MTV | -0.355 |
THR | -0.352 |
patches | -0.351 |
Token |
|
Token ID | 198 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
anyahu | +1.706 |
jriwal | +1.693 |
DonaldTrump | +1.668 |
uphem | +1.641 |
owa | +1.607 |
Neg logit contributions | |
digitally | -0.849 |
spaced | -0.840 |
MTV | -0.817 |
CST | -0.811 |
patches | -0.810 |
Token |
|
Token ID | 198 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
anyahu | +2.409 |
jriwal | +2.373 |
DonaldTrump | +2.346 |
uphem | +2.310 |
owa | +2.271 |
Neg logit contributions | |
digitally | -1.138 |
spaced | -1.124 |
MTV | -1.099 |
patches | -1.096 |
CST | -1.085 |
Token | https |
Token ID | 5450 |
Feature activation | +6.22e+00 |
Pos logit contributions | |
anyahu | +1.662 |
jriwal | +1.642 |
DonaldTrump | +1.622 |
uphem | +1.584 |
owa | +1.549 |
Neg logit contributions | |
digitally | -1.130 |
spaced | -1.117 |
MTV | -1.101 |
CST | -1.094 |
THR | -1.093 |
Token | :// |
Token ID | 1378 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
anyahu | +4.802 |
jriwal | +4.634 |
DonaldTrump | +4.633 |
uphem | +4.520 |
owa | +4.440 |
Neg logit contributions | |
spaced | -4.181 |
digitally | -4.159 |
patches | -4.063 |
CST | -4.042 |
MTV | -4.042 |
Token | www |
Token ID | 2503 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
anyahu | +1.830 |
jriwal | +1.810 |
DonaldTrump | +1.770 |
uphem | +1.733 |
owa | +1.691 |
Neg logit contributions | |
digitally | -1.857 |
spaced | -1.843 |
MTV | -1.825 |
CST | -1.814 |
THR | -1.812 |
Token | . |
Token ID | 13 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
anyahu | +1.037 |
jriwal | +1.027 |
DonaldTrump | +1.009 |
uphem | +0.990 |
owa | +0.969 |
Neg logit contributions | |
digitally | -0.579 |
spaced | -0.568 |
MTV | -0.565 |
THR | -0.559 |
patches | -0.558 |
Token | eff |
Token ID | 14822 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
anyahu | +1.257 |
jriwal | +1.246 |
DonaldTrump | +1.227 |
uphem | +1.206 |
owa | +1.183 |
Neg logit contributions | |
digitally | -0.577 |
spaced | -0.565 |
MTV | -0.560 |
THR | -0.554 |
CST | -0.554 |
Token | . |
Token ID | 13 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
anyahu | +0.376 |
jriwal | +0.373 |
DonaldTrump | +0.366 |
uphem | +0.359 |
owa | +0.351 |
Neg logit contributions | |
digitally | -0.197 |
spaced | -0.192 |
MTV | -0.192 |
THR | -0.190 |
patches | -0.189 |
Token | 10 |
Token ID | 940 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
digitally | +0.424 |
spaced | +0.417 |
MTV | +0.413 |
THR | +0.409 |
patches | +0.408 |
Neg logit contributions | |
anyahu | -0.890 |
jriwal | -0.880 |
DonaldTrump | -0.863 |
uphem | -0.849 |
owa | -0.834 |
Token | . |
Token ID | 13 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
anyahu | +2.469 |
jriwal | +2.416 |
DonaldTrump | +2.396 |
uphem | +2.358 |
owa | +2.319 |
Neg logit contributions | |
digitally | -1.595 |
spaced | -1.594 |
MTV | -1.563 |
patches | -1.559 |
CST | -1.553 |
Token | 13 |
Token ID | 1485 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
anyahu | +2.738 |
jriwal | +2.688 |
DonaldTrump | +2.661 |
uphem | +2.620 |
owa | +2.588 |
Neg logit contributions | |
digitally | -1.231 |
spaced | -1.202 |
patches | -1.190 |
MTV | -1.180 |
CST | -1.177 |
Token | 71 |
Token ID | 4869 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
anyahu | +1.756 |
jriwal | +1.725 |
DonaldTrump | +1.672 |
uphem | +1.649 |
owa | +1.606 |
Neg logit contributions | |
digitally | -2.083 |
spaced | -2.050 |
MTV | -2.046 |
patches | -2.038 |
THR | -2.034 |
Token | / |
Token ID | 14 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
anyahu | +2.384 |
jriwal | +2.334 |
DonaldTrump | +2.298 |
uphem | +2.257 |
owa | +2.213 |
Neg logit contributions | |
digitally | -1.667 |
spaced | -1.642 |
MTV | -1.631 |
patches | -1.620 |
THR | -1.611 |
Token | journal |
Token ID | 24891 |
Feature activation | +6.16e+00 |
Pos logit contributions | |
anyahu | +0.281 |
jriwal | +0.278 |
DonaldTrump | +0.272 |
uphem | +0.266 |
owa | +0.258 |
Neg logit contributions | |
digitally | -0.220 |
spaced | -0.216 |
MTV | -0.216 |
THR | -0.215 |
patches | -0.214 |
Token | . |
Token ID | 13 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
anyahu | +5.126 |
jriwal | +4.953 |
uphem | +4.917 |
DonaldTrump | +4.893 |
owa | +4.882 |
Neg logit contributions | |
spaced | -3.691 |
digitally | -3.685 |
MTV | -3.616 |
CST | -3.604 |
patches | -3.570 |
Token | p |
Token ID | 79 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
anyahu | +1.112 |
jriwal | +1.099 |
DonaldTrump | +1.074 |
uphem | +1.042 |
owa | +1.011 |
Neg logit contributions | |
digitally | -1.253 |
spaced | -1.237 |
MTV | -1.232 |
THR | -1.225 |
patches | -1.223 |
Token | one |
Token ID | 505 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
anyahu | +0.925 |
jriwal | +0.915 |
DonaldTrump | +0.896 |
uphem | +0.871 |
owa | +0.847 |
Neg logit contributions | |
digitally | -0.874 |
spaced | -0.863 |
MTV | -0.859 |
THR | -0.854 |
patches | -0.851 |
Token | . |
Token ID | 13 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
anyahu | +2.144 |
jriwal | +2.120 |
DonaldTrump | +2.084 |
uphem | +2.046 |
owa | +2.003 |
Neg logit contributions | |
digitally | -1.201 |
spaced | -1.180 |
MTV | -1.172 |
THR | -1.161 |
patches | -1.159 |
Token | 00 |
Token ID | 405 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
digitally | +0.516 |
spaced | +0.508 |
MTV | +0.506 |
THR | +0.500 |
patches | +0.497 |
Neg logit contributions | |
anyahu | -0.876 |
jriwal | -0.874 |
DonaldTrump | -0.856 |
uphem | -0.833 |
owa | -0.811 |
Token | : |
Token ID | 25 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
anyahu | +1.501 |
jriwal | +1.484 |
DonaldTrump | +1.458 |
uphem | +1.430 |
owa | +1.400 |
Neg logit contributions | |
digitally | -0.883 |
spaced | -0.868 |
MTV | -0.862 |
THR | -0.854 |
patches | -0.852 |
Token |
|
Token ID | 198 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
anyahu | +1.995 |
jriwal | +1.973 |
DonaldTrump | +1.944 |
uphem | +1.912 |
owa | +1.878 |
Neg logit contributions | |
digitally | -0.891 |
spaced | -0.872 |
MTV | -0.863 |
patches | -0.853 |
THR | -0.852 |
Token |
|
Token ID | 198 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
anyahu | +3.358 |
jriwal | +3.300 |
DonaldTrump | +3.266 |
uphem | +3.219 |
owa | +3.181 |
Neg logit contributions | |
digitally | -1.686 |
spaced | -1.658 |
MTV | -1.620 |
patches | -1.619 |
CST | -1.595 |
Token | Show |
Token ID | 15307 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
anyahu | +2.249 |
jriwal | +2.217 |
DonaldTrump | +2.175 |
uphem | +2.124 |
owa | +2.079 |
Neg logit contributions | |
digitally | -1.913 |
spaced | -1.870 |
MTV | -1.863 |
THR | -1.850 |
CST | -1.838 |
Token | nested |
Token ID | 28376 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
anyahu | +0.420 |
jriwal | +0.414 |
DonaldTrump | +0.407 |
uphem | +0.398 |
owa | +0.389 |
Neg logit contributions | |
digitally | -0.304 |
spaced | -0.299 |
MTV | -0.298 |
THR | -0.295 |
patches | -0.294 |
Token | quote |
Token ID | 9577 |
Feature activation | +5.79e+00 |
Pos logit contributions | |
anyahu | +4.051 |
jriwal | +4.016 |
DonaldTrump | +3.895 |
uphem | +3.876 |
owa | +3.836 |
Neg logit contributions | |
digitally | -1.380 |
Warwick | -1.349 |
MTV | -1.348 |
THR | -1.318 |
Constantin | -1.309 |
Token | + |
Token ID | 1343 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
anyahu | +4.445 |
jriwal | +4.440 |
DonaldTrump | +4.306 |
owa | +4.237 |
uphem | +4.215 |
Neg logit contributions | |
digitally | -3.731 |
MTV | -3.699 |
spaced | -3.684 |
CST | -3.649 |
Warwick | -3.648 |
Token |
|
Token ID | 198 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
anyahu | +2.007 |
jriwal | +1.977 |
DonaldTrump | +1.960 |
uphem | +1.921 |
owa | +1.887 |
Neg logit contributions | |
digitally | -0.919 |
spaced | -0.903 |
MTV | -0.887 |
patches | -0.882 |
CST | -0.873 |
Token |
|
Token ID | 198 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
digitally | +0.061 |
MTV | +0.059 |
spaced | +0.059 |
THR | +0.059 |
CST | +0.058 |
Neg logit contributions | |
anyahu | -0.154 |
jriwal | -0.153 |
DonaldTrump | -0.150 |
uphem | -0.147 |
owa | -0.144 |
Token | On |
Token ID | 2202 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
digitally | +0.122 |
spaced | +0.120 |
MTV | +0.120 |
THR | +0.119 |
patches | +0.118 |
Neg logit contributions | |
anyahu | -0.159 |
jriwal | -0.157 |
DonaldTrump | -0.154 |
uphem | -0.151 |
owa | -0.147 |
Token | September |
Token ID | 2693 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
anyahu | +0.676 |
jriwal | +0.669 |
DonaldTrump | +0.657 |
uphem | +0.645 |
owa | +0.632 |
Neg logit contributions | |
digitally | -0.379 |
spaced | -0.373 |
MTV | -0.370 |
THR | -0.367 |
patches | -0.365 |
Token | full |
Token ID | 1336 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
digitally | +0.194 |
spaced | +0.190 |
MTV | +0.189 |
THR | +0.187 |
patches | +0.187 |
Neg logit contributions | |
anyahu | -0.398 |
jriwal | -0.394 |
DonaldTrump | -0.387 |
uphem | -0.380 |
owa | -0.372 |
Token | lawsuit |
Token ID | 8087 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
anyahu | +1.053 |
jriwal | +1.044 |
DonaldTrump | +1.027 |
uphem | +1.004 |
owa | +0.989 |
Neg logit contributions | |
digitally | -0.533 |
spaced | -0.524 |
MTV | -0.517 |
CST | -0.513 |
THR | -0.512 |
Token | : |
Token ID | 25 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
anyahu | +0.761 |
jriwal | +0.753 |
DonaldTrump | +0.741 |
uphem | +0.726 |
owa | +0.711 |
Neg logit contributions | |
digitally | -0.467 |
spaced | -0.461 |
MTV | -0.457 |
CST | -0.452 |
THR | -0.452 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
anyahu | +1.455 |
jriwal | +1.442 |
DonaldTrump | +1.420 |
uphem | +1.392 |
owa | +1.363 |
Neg logit contributions | |
digitally | -0.840 |
spaced | -0.830 |
MTV | -0.817 |
patches | -0.809 |
THR | -0.808 |
Token |
|
Token ID | 198 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
anyahu | +1.495 |
jriwal | +1.480 |
DonaldTrump | +1.457 |
uphem | +1.432 |
owa | +1.405 |
Neg logit contributions | |
digitally | -0.643 |
spaced | -0.630 |
MTV | -0.624 |
THR | -0.617 |
patches | -0.616 |
Token | https |
Token ID | 5450 |
Feature activation | +5.64e+00 |
Pos logit contributions | |
anyahu | +0.959 |
jriwal | +0.948 |
DonaldTrump | +0.933 |
uphem | +0.912 |
owa | +0.891 |
Neg logit contributions | |
digitally | -0.719 |
spaced | -0.711 |
MTV | -0.704 |
THR | -0.700 |
patches | -0.699 |
Token | :// |
Token ID | 1378 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
anyahu | +4.473 |
jriwal | +4.362 |
DonaldTrump | +4.318 |
uphem | +4.218 |
owa | +4.151 |
Neg logit contributions | |
digitally | -3.694 |
spaced | -3.686 |
MTV | -3.603 |
CST | -3.596 |
patches | -3.584 |
Token | www |
Token ID | 2503 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
anyahu | +1.863 |
jriwal | +1.840 |
DonaldTrump | +1.801 |
uphem | +1.766 |
owa | +1.724 |
Neg logit contributions | |
digitally | -1.896 |
spaced | -1.888 |
MTV | -1.870 |
CST | -1.854 |
THR | -1.853 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
digitally | +0.258 |
THR | +0.252 |
MTV | +0.252 |
spaced | +0.251 |
CST | +0.248 |
Neg logit contributions | |
anyahu | -0.520 |
jriwal | -0.518 |
DonaldTrump | -0.507 |
uphem | -0.496 |
owa | -0.485 |
Token | eff |
Token ID | 14822 |
Feature activation | +6.94e-02 |
Pos logit contributions | |
anyahu | +1.410 |
jriwal | +1.398 |
DonaldTrump | +1.378 |
uphem | +1.361 |
owa | +1.334 |
Neg logit contributions | |
digitally | -0.636 |
spaced | -0.625 |
MTV | -0.619 |
CST | -0.613 |
THR | -0.611 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
anyahu | +0.067 |
jriwal | +0.066 |
DonaldTrump | +0.065 |
uphem | +0.064 |
owa | +0.062 |
Neg logit contributions | |
digitally | -0.035 |
spaced | -0.034 |
MTV | -0.034 |
patches | -0.033 |
THR | -0.033 |
Token | what |
Token ID | 644 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
anyahu | +1.773 |
jriwal | +1.749 |
DonaldTrump | +1.722 |
uphem | +1.686 |
owa | +1.658 |
Neg logit contributions | |
digitally | -0.978 |
spaced | -0.959 |
MTV | -0.952 |
THR | -0.945 |
patches | -0.943 |
Token | that |
Token ID | 326 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
anyahu | +0.844 |
jriwal | +0.834 |
DonaldTrump | +0.820 |
uphem | +0.804 |
owa | +0.793 |
Neg logit contributions | |
digitally | -0.438 |
spaced | -0.427 |
MTV | -0.425 |
THR | -0.423 |
CST | -0.421 |
Token | 's |
Token ID | 338 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
anyahu | +0.748 |
jriwal | +0.736 |
DonaldTrump | +0.722 |
uphem | +0.709 |
owa | +0.694 |
Neg logit contributions | |
digitally | -0.596 |
spaced | -0.586 |
MTV | -0.582 |
THR | -0.580 |
CST | -0.579 |
Token | enough |
Token ID | 1576 |
Feature activation | +5.25e-01 |
Pos logit contributions | |
anyahu | +0.859 |
jriwal | +0.850 |
DonaldTrump | +0.835 |
uphem | +0.819 |
owa | +0.802 |
Neg logit contributions | |
digitally | -0.471 |
spaced | -0.462 |
MTV | -0.460 |
THR | -0.455 |
patches | -0.454 |
Token | . |
Token ID | 13 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
anyahu | +0.476 |
jriwal | +0.471 |
DonaldTrump | +0.462 |
uphem | +0.453 |
owa | +0.444 |
Neg logit contributions | |
digitally | -0.290 |
spaced | -0.285 |
MTV | -0.283 |
THR | -0.281 |
patches | -0.280 |
Token |
|
Token ID | 198 |
Feature activation | +5.63e+00 |
Pos logit contributions | |
anyahu | +0.347 |
jriwal | +0.343 |
DonaldTrump | +0.337 |
uphem | +0.330 |
owa | +0.323 |
Neg logit contributions | |
digitally | -0.230 |
spaced | -0.226 |
MTV | -0.225 |
THR | -0.223 |
patches | -0.222 |
Token |
|
Token ID | 198 |
Feature activation | +5.16e+00 |
Pos logit contributions | |
anyahu | +5.214 |
DonaldTrump | +5.112 |
jriwal | +5.056 |
uphem | +4.987 |
owa | +4.922 |
Neg logit contributions | |
digitally | -2.880 |
spaced | -2.865 |
patches | -2.788 |
MTV | -2.761 |
CST | -2.712 |
Token | Tw |
Token ID | 5080 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
anyahu | +4.006 |
jriwal | +3.932 |
DonaldTrump | +3.888 |
uphem | +3.790 |
owa | +3.742 |
Neg logit contributions | |
digitally | -3.487 |
spaced | -3.431 |
MTV | -3.413 |
THR | -3.408 |
CST | -3.399 |
Token | itch |
Token ID | 2007 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
anyahu | +0.438 |
jriwal | +0.433 |
DonaldTrump | +0.424 |
uphem | +0.415 |
owa | +0.406 |
Neg logit contributions | |
digitally | -0.335 |
spaced | -0.330 |
MTV | -0.328 |
THR | -0.326 |
patches | -0.325 |
Token | : |
Token ID | 25 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
anyahu | +2.142 |
jriwal | +2.113 |
DonaldTrump | +2.083 |
uphem | +2.029 |
owa | +1.991 |
Neg logit contributions | |
digitally | -1.447 |
spaced | -1.423 |
MTV | -1.411 |
THR | -1.404 |
patches | -1.395 |
Token | http |
Token ID | 2638 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
anyahu | +1.107 |
jriwal | +1.097 |
DonaldTrump | +1.078 |
uphem | +1.055 |
owa | +1.036 |
Neg logit contributions | |
digitally | -0.660 |
spaced | -0.647 |
MTV | -0.640 |
patches | -0.636 |
THR | -0.635 |
Token | ') |
Token ID | 11537 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
anyahu | +0.364 |
jriwal | +0.360 |
DonaldTrump | +0.353 |
uphem | +0.345 |
owa | +0.337 |
Neg logit contributions | |
digitally | -0.265 |
spaced | -0.261 |
MTV | -0.260 |
THR | -0.258 |
patches | -0.257 |
Token | | |
Token ID | 930 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
anyahu | +0.753 |
jriwal | +0.745 |
DonaldTrump | +0.731 |
uphem | +0.716 |
owa | +0.701 |
Neg logit contributions | |
digitally | -0.487 |
spaced | -0.480 |
MTV | -0.477 |
THR | -0.473 |
patches | -0.472 |
Token | out |
Token ID | 503 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
digitally | +0.255 |
spaced | +0.249 |
MTV | +0.245 |
THR | +0.243 |
patches | +0.242 |
Neg logit contributions | |
anyahu | -0.785 |
jriwal | -0.779 |
DonaldTrump | -0.768 |
uphem | -0.754 |
owa | -0.740 |
Token | - |
Token ID | 12 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
anyahu | +1.950 |
jriwal | +1.923 |
DonaldTrump | +1.890 |
uphem | +1.850 |
owa | +1.810 |
Neg logit contributions | |
digitally | -1.395 |
spaced | -1.373 |
MTV | -1.365 |
THR | -1.352 |
patches | -1.352 |
Token | null |
Token ID | 8423 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
digitally | +0.751 |
MTV | +0.737 |
THR | +0.734 |
spaced | +0.733 |
dot | +0.730 |
Neg logit contributions | |
jriwal | -1.159 |
anyahu | -1.150 |
DonaldTrump | -1.127 |
uphem | -1.097 |
owa | -1.063 |
Token | # |
Token ID | 1303 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
anyahu | +0.063 |
jriwal | +0.062 |
DonaldTrump | +0.061 |
uphem | +0.060 |
owa | +0.058 |
Neg logit contributions | |
digitally | -0.043 |
spaced | -0.043 |
MTV | -0.043 |
THR | -0.042 |
patches | -0.042 |
Token | Draw |
Token ID | 15315 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
anyahu | +4.688 |
jriwal | +4.597 |
DonaldTrump | +4.583 |
uphem | +4.433 |
owa | +4.392 |
Neg logit contributions | |
digitally | -3.175 |
spaced | -3.162 |
MTV | -3.142 |
Warwick | -3.094 |
CST | -3.074 |
Token | string |
Token ID | 4731 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
anyahu | +0.630 |
jriwal | +0.627 |
DonaldTrump | +0.614 |
uphem | +0.602 |
owa | +0.592 |
Neg logit contributions | |
digitally | -0.371 |
MTV | -0.366 |
spaced | -0.366 |
CST | -0.361 |
THR | -0.361 |
Token | > |
Token ID | 1875 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
anyahu | +1.119 |
jriwal | +1.109 |
DonaldTrump | +1.087 |
uphem | +1.062 |
owa | +1.045 |
Neg logit contributions | |
digitally | -0.671 |
spaced | -0.665 |
MTV | -0.663 |
THR | -0.654 |
CST | -0.653 |
Token | align |
Token ID | 10548 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
anyahu | +0.700 |
jriwal | +0.692 |
DonaldTrump | +0.680 |
uphem | +0.666 |
owa | +0.653 |
Neg logit contributions | |
digitally | -0.468 |
spaced | -0.462 |
MTV | -0.460 |
THR | -0.454 |
CST | -0.453 |
Token | ement |
Token ID | 972 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
digitally | +0.547 |
spaced | +0.539 |
MTV | +0.535 |
THR | +0.532 |
patches | +0.531 |
Neg logit contributions | |
anyahu | -0.756 |
jriwal | -0.750 |
DonaldTrump | -0.733 |
uphem | -0.719 |
owa | -0.701 |
Token | ] |
Token ID | 60 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
anyahu | +0.435 |
jriwal | +0.430 |
DonaldTrump | +0.420 |
uphem | +0.410 |
owa | +0.398 |
Neg logit contributions | |
digitally | -0.431 |
spaced | -0.426 |
MTV | -0.424 |
THR | -0.421 |
patches | -0.420 |
Token | any |
Token ID | 597 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
anyahu | +0.493 |
jriwal | +0.487 |
DonaldTrump | +0.477 |
uphem | +0.467 |
owa | +0.456 |
Neg logit contributions | |
digitally | -0.380 |
spaced | -0.374 |
MTV | -0.373 |
THR | -0.370 |
CST | -0.368 |
Token | text |
Token ID | 2420 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
anyahu | +1.905 |
jriwal | +1.880 |
DonaldTrump | +1.852 |
uphem | +1.831 |
owa | +1.803 |
Neg logit contributions | |
digitally | -0.560 |
MTV | -0.545 |
Warwick | -0.538 |
THR | -0.537 |
CST | -0.532 |
Token | or |
Token ID | 393 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
anyahu | +1.796 |
jriwal | +1.778 |
DonaldTrump | +1.743 |
uphem | +1.718 |
owa | +1.679 |
Neg logit contributions | |
digitally | -0.997 |
MTV | -0.979 |
spaced | -0.972 |
THR | -0.968 |
CST | -0.966 |
Token | empty |
Token ID | 6565 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
anyahu | +0.848 |
jriwal | +0.837 |
DonaldTrump | +0.822 |
uphem | +0.808 |
owa | +0.791 |
Neg logit contributions | |
digitally | -0.510 |
MTV | -0.500 |
spaced | -0.498 |
THR | -0.495 |
CST | -0.493 |
Token | Example |
Token ID | 17934 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
digitally | +0.483 |
spaced | +0.477 |
MTV | +0.474 |
THR | +0.471 |
patches | +0.470 |
Neg logit contributions | |
anyahu | -0.556 |
jriwal | -0.549 |
DonaldTrump | -0.538 |
uphem | -0.524 |
owa | -0.512 |
Token | : |
Token ID | 25 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
anyahu | +4.608 |
jriwal | +4.550 |
DonaldTrump | +4.438 |
uphem | +4.379 |
owa | +4.334 |
Neg logit contributions | |
digitally | -3.246 |
spaced | -3.181 |
MTV | -3.179 |
Warwick | -3.164 |
THR | -3.160 |
Token | /* |
Token ID | 11900 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
anyahu | +2.313 |
jriwal | +2.280 |
DonaldTrump | +2.212 |
uphem | +2.201 |
owa | +2.157 |
Neg logit contributions | |
digitally | -1.900 |
MTV | -1.868 |
spaced | -1.853 |
CST | -1.849 |
THR | -1.848 |
Token | vim |
Token ID | 43907 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
anyahu | +3.793 |
jriwal | +3.738 |
DonaldTrump | +3.684 |
uphem | +3.618 |
owa | +3.582 |
Neg logit contributions | |
digitally | -2.620 |
spaced | -2.564 |
MTV | -2.552 |
CST | -2.551 |
Warwick | -2.537 |
Token | : |
Token ID | 25 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
anyahu | +1.431 |
jriwal | +1.412 |
DonaldTrump | +1.384 |
uphem | +1.359 |
owa | +1.335 |
Neg logit contributions | |
digitally | -0.899 |
spaced | -0.882 |
MTV | -0.878 |
THR | -0.871 |
CST | -0.866 |
Token | set |
Token ID | 900 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
anyahu | +1.884 |
jriwal | +1.856 |
DonaldTrump | +1.820 |
uphem | +1.788 |
owa | +1.757 |
Neg logit contributions | |
digitally | -1.388 |
spaced | -1.359 |
MTV | -1.357 |
THR | -1.351 |
CST | -1.346 |
Token | getting |
Token ID | 1972 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
anyahu | +1.087 |
jriwal | +1.079 |
DonaldTrump | +1.061 |
uphem | +1.041 |
owa | +1.025 |
Neg logit contributions | |
digitally | -0.463 |
spaced | -0.456 |
MTV | -0.453 |
CST | -0.452 |
THR | -0.450 |
Token | a |
Token ID | 257 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
anyahu | +0.472 |
jriwal | +0.468 |
DonaldTrump | +0.460 |
uphem | +0.451 |
owa | +0.442 |
Neg logit contributions | |
digitally | -0.251 |
spaced | -0.247 |
CST | -0.245 |
MTV | -0.245 |
THR | -0.243 |
Token | cross |
Token ID | 3272 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
anyahu | +0.292 |
jriwal | +0.289 |
DonaldTrump | +0.284 |
uphem | +0.279 |
owa | +0.273 |
Neg logit contributions | |
digitally | -0.166 |
spaced | -0.164 |
MTV | -0.163 |
THR | -0.161 |
patches | -0.161 |
Token | - |
Token ID | 12 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
anyahu | +0.797 |
jriwal | +0.793 |
DonaldTrump | +0.774 |
uphem | +0.763 |
owa | +0.749 |
Neg logit contributions | |
digitally | -0.479 |
spaced | -0.470 |
MTV | -0.468 |
CST | -0.467 |
THR | -0.466 |
Token | f |
Token ID | 69 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
digitally | +0.867 |
spaced | +0.852 |
MTV | +0.852 |
THR | +0.839 |
patches | +0.839 |
Neg logit contributions | |
anyahu | -1.164 |
jriwal | -1.146 |
DonaldTrump | -1.123 |
uphem | -1.099 |
owa | -1.070 |
Token | ert |
Token ID | 861 |
Feature activation | +5.47e+00 |
Pos logit contributions | |
digitally | +1.652 |
MTV | +1.623 |
spaced | +1.622 |
THR | +1.605 |
CST | +1.604 |
Neg logit contributions | |
anyahu | -2.234 |
jriwal | -2.200 |
DonaldTrump | -2.169 |
uphem | -2.107 |
owa | -2.052 |
Token | il |
Token ID | 346 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
anyahu | +3.774 |
jriwal | +3.712 |
DonaldTrump | +3.594 |
uphem | +3.517 |
owa | +3.444 |
Neg logit contributions | |
Warwick | -3.954 |
CST | -3.932 |
CEPT | -3.920 |
THR | -3.913 |
digitally | -3.908 |
Token | ization |
Token ID | 1634 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
digitally | +0.234 |
spaced | +0.230 |
MTV | +0.229 |
THR | +0.227 |
patches | +0.227 |
Neg logit contributions | |
anyahu | -0.314 |
jriwal | -0.311 |
DonaldTrump | -0.305 |
uphem | -0.298 |
owa | -0.291 |
Token | of |
Token ID | 286 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
anyahu | +0.827 |
jriwal | +0.824 |
DonaldTrump | +0.803 |
uphem | +0.789 |
owa | +0.774 |
Neg logit contributions | |
digitally | -0.535 |
spaced | -0.528 |
MTV | -0.524 |
CST | -0.523 |
THR | -0.521 |
Token | forms |
Token ID | 5107 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
digitally | +0.227 |
spaced | +0.222 |
MTV | +0.221 |
THR | +0.219 |
patches | +0.218 |
Neg logit contributions | |
anyahu | -0.446 |
jriwal | -0.442 |
DonaldTrump | -0.434 |
uphem | -0.426 |
owa | -0.417 |
Token | throughout |
Token ID | 3690 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
digitally | +0.125 |
spaced | +0.123 |
MTV | +0.122 |
THR | +0.121 |
patches | +0.121 |
Neg logit contributions | |
anyahu | -0.197 |
jriwal | -0.195 |
DonaldTrump | -0.192 |
uphem | -0.188 |
owa | -0.184 |
Token | looks |
Token ID | 3073 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
anyahu | +0.319 |
jriwal | +0.317 |
DonaldTrump | +0.310 |
uphem | +0.305 |
owa | +0.296 |
Neg logit contributions | |
digitally | -0.184 |
MTV | -0.181 |
spaced | -0.181 |
THR | -0.180 |
CST | -0.178 |
Token | different |
Token ID | 1180 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
anyahu | +0.333 |
jriwal | +0.330 |
DonaldTrump | +0.323 |
uphem | +0.317 |
owa | +0.313 |
Neg logit contributions | |
digitally | -0.187 |
spaced | -0.184 |
MTV | -0.184 |
THR | -0.183 |
CST | -0.182 |
Token | . |
Token ID | 13 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
digitally | +0.481 |
spaced | +0.469 |
MTV | +0.466 |
THR | +0.462 |
patches | +0.460 |
Neg logit contributions | |
anyahu | -0.891 |
jriwal | -0.882 |
DonaldTrump | -0.866 |
uphem | -0.851 |
owa | -0.830 |
Token | George |
Token ID | 4502 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
digitally | +0.335 |
MTV | +0.333 |
THR | +0.332 |
CST | +0.329 |
spaced | +0.329 |
Neg logit contributions | |
anyahu | -0.529 |
jriwal | -0.527 |
DonaldTrump | -0.510 |
uphem | -0.505 |
owa | -0.490 |
Token | W |
Token ID | 370 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
anyahu | +1.123 |
jriwal | +1.112 |
DonaldTrump | +1.092 |
uphem | +1.071 |
owa | +1.048 |
Neg logit contributions | |
digitally | -0.656 |
spaced | -0.645 |
MTV | -0.641 |
THR | -0.635 |
patches | -0.633 |
Token | . |
Token ID | 13 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
anyahu | +0.959 |
jriwal | +0.949 |
DonaldTrump | +0.932 |
uphem | +0.914 |
owa | +0.895 |
Neg logit contributions | |
digitally | -0.527 |
spaced | -0.517 |
MTV | -0.514 |
THR | -0.510 |
patches | -0.508 |
Token | Bush |
Token ID | 5511 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
anyahu | +3.733 |
DonaldTrump | +3.656 |
jriwal | +3.584 |
uphem | +3.482 |
owa | +3.434 |
Neg logit contributions | |
digitally | -4.054 |
spaced | -4.019 |
MTV | -3.953 |
patches | -3.945 |
THR | -3.937 |
Token | was |
Token ID | 373 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
anyahu | +0.537 |
jriwal | +0.531 |
DonaldTrump | +0.522 |
uphem | +0.510 |
owa | +0.500 |
Neg logit contributions | |
digitally | -0.357 |
spaced | -0.352 |
MTV | -0.350 |
THR | -0.347 |
patches | -0.346 |
Token | president |
Token ID | 1893 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
digitally | +0.071 |
spaced | +0.070 |
MTV | +0.070 |
THR | +0.069 |
CST | +0.069 |
Neg logit contributions | |
anyahu | -0.109 |
jriwal | -0.108 |
DonaldTrump | -0.106 |
uphem | -0.104 |
owa | -0.101 |
Token | -- |
Token ID | 438 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
anyahu | +1.764 |
jriwal | +1.744 |
DonaldTrump | +1.720 |
uphem | +1.676 |
owa | +1.650 |
Neg logit contributions | |
digitally | -1.066 |
spaced | -1.052 |
MTV | -1.043 |
THR | -1.034 |
patches | -1.033 |
Token | and |
Token ID | 290 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
digitally | +0.491 |
spaced | +0.483 |
MTV | +0.482 |
THR | +0.477 |
CST | +0.475 |
Neg logit contributions | |
anyahu | -0.692 |
jriwal | -0.688 |
DonaldTrump | -0.669 |
uphem | -0.658 |
owa | -0.640 |
Token | lawyer |
Token ID | 6853 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
anyahu | +0.384 |
jriwal | +0.380 |
DonaldTrump | +0.372 |
uphem | +0.364 |
owa | +0.356 |
Neg logit contributions | |
digitally | -0.288 |
spaced | -0.284 |
MTV | -0.284 |
THR | -0.281 |
CST | -0.280 |
Token | during |
Token ID | 1141 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
digitally | +0.603 |
spaced | +0.589 |
MTV | +0.584 |
THR | +0.582 |
patches | +0.577 |
Neg logit contributions | |
anyahu | -1.034 |
jriwal | -1.027 |
DonaldTrump | -1.010 |
uphem | -0.988 |
owa | -0.962 |
Token | the |
Token ID | 262 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
anyahu | +0.616 |
jriwal | +0.610 |
DonaldTrump | +0.600 |
uphem | +0.588 |
owa | +0.576 |
Neg logit contributions | |
digitally | -0.329 |
spaced | -0.323 |
MTV | -0.320 |
THR | -0.317 |
patches | -0.317 |
Token | George |
Token ID | 4502 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
anyahu | +1.006 |
jriwal | +0.993 |
DonaldTrump | +0.981 |
uphem | +0.956 |
owa | +0.940 |
Neg logit contributions | |
digitally | -0.597 |
spaced | -0.587 |
MTV | -0.583 |
THR | -0.578 |
patches | -0.577 |
Token | W |
Token ID | 370 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
anyahu | +0.362 |
jriwal | +0.360 |
DonaldTrump | +0.352 |
uphem | +0.347 |
owa | +0.339 |
Neg logit contributions | |
digitally | -0.180 |
spaced | -0.176 |
MTV | -0.176 |
THR | -0.174 |
patches | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
anyahu | +0.504 |
jriwal | +0.501 |
DonaldTrump | +0.491 |
uphem | +0.482 |
owa | +0.471 |
Neg logit contributions | |
digitally | -0.271 |
spaced | -0.266 |
MTV | -0.265 |
THR | -0.263 |
patches | -0.261 |
Token | Bush |
Token ID | 5511 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
anyahu | +3.556 |
DonaldTrump | +3.493 |
jriwal | +3.370 |
uphem | +3.256 |
owa | +3.255 |
Neg logit contributions | |
digitally | -4.195 |
spaced | -4.159 |
MTV | -4.068 |
THR | -4.055 |
patches | -4.053 |
Token | administration |
Token ID | 3662 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
digitally | +0.875 |
spaced | +0.862 |
MTV | +0.857 |
THR | +0.851 |
patches | +0.848 |
Neg logit contributions | |
anyahu | -1.227 |
jriwal | -1.214 |
DonaldTrump | -1.190 |
uphem | -1.165 |
owa | -1.138 |
Token | , |
Token ID | 11 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
digitally | +1.134 |
spaced | +1.107 |
MTV | +1.104 |
THR | +1.096 |
patches | +1.090 |
Neg logit contributions | |
anyahu | -2.005 |
jriwal | -1.991 |
DonaldTrump | -1.955 |
uphem | -1.922 |
owa | -1.869 |
Token | said |
Token ID | 531 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
anyahu | +0.230 |
jriwal | +0.227 |
DonaldTrump | +0.223 |
uphem | +0.218 |
owa | +0.214 |
Neg logit contributions | |
digitally | -0.143 |
spaced | -0.140 |
MTV | -0.140 |
THR | -0.138 |
CST | -0.138 |
Token | there |
Token ID | 612 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
anyahu | +0.527 |
jriwal | +0.522 |
DonaldTrump | +0.512 |
uphem | +0.502 |
owa | +0.492 |
Neg logit contributions | |
digitally | -0.308 |
spaced | -0.302 |
MTV | -0.300 |
patches | -0.297 |
THR | -0.297 |
Token | Share |
Token ID | 8734 |
Feature activation | +7.61e+00 |
Pos logit contributions | |
anyahu | +1.000 |
jriwal | +0.986 |
DonaldTrump | +0.977 |
uphem | +0.961 |
owa | +0.939 |
Neg logit contributions | |
digitally | -0.679 |
spaced | -0.663 |
THR | -0.662 |
MTV | -0.656 |
patches | -0.654 |
Token | this |
Token ID | 428 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
anyahu | +6.560 |
jriwal | +6.360 |
uphem | +6.231 |
DonaldTrump | +6.188 |
owa | +6.136 |
Neg logit contributions | |
CEPT | -4.040 |
Constantin | -3.990 |
MTV | -3.964 |
spaced | -3.950 |
THR | -3.938 |
Token | comment |
Token ID | 2912 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
digitally | +0.165 |
spaced | +0.162 |
MTV | +0.161 |
THR | +0.160 |
patches | +0.159 |
Neg logit contributions | |
anyahu | -0.322 |
jriwal | -0.319 |
DonaldTrump | -0.313 |
uphem | -0.307 |
owa | -0.302 |
Token | on |
Token ID | 319 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
anyahu | +2.555 |
jriwal | +2.523 |
DonaldTrump | +2.481 |
uphem | +2.439 |
owa | +2.396 |
Neg logit contributions | |
digitally | -1.303 |
spaced | -1.291 |
MTV | -1.276 |
patches | -1.264 |
THR | -1.260 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
digitally | +0.004 |
MTV | +0.004 |
spaced | +0.004 |
THR | +0.004 |
CST | +0.004 |
Neg logit contributions | |
anyahu | -0.011 |
jriwal | -0.011 |
DonaldTrump | -0.011 |
uphem | -0.011 |
owa | -0.011 |
Token | Share |
Token ID | 8734 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
anyahu | +2.244 |
jriwal | +2.176 |
DonaldTrump | +2.175 |
uphem | +2.119 |
owa | +2.099 |
Neg logit contributions | |
digitally | -1.579 |
spaced | -1.542 |
THR | -1.518 |
MTV | -1.517 |
patches | -1.515 |
Token | this |
Token ID | 428 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
anyahu | +5.087 |
jriwal | +4.947 |
DonaldTrump | +4.857 |
uphem | +4.834 |
owa | +4.739 |
Neg logit contributions | |
spaced | -2.624 |
digitally | -2.618 |
MTV | -2.612 |
Warwick | -2.592 |
THR | -2.571 |
Token | comment |
Token ID | 2912 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
digitally | +0.679 |
spaced | +0.663 |
MTV | +0.657 |
THR | +0.650 |
CST | +0.649 |
Neg logit contributions | |
anyahu | -1.406 |
jriwal | -1.393 |
DonaldTrump | -1.370 |
uphem | -1.346 |
owa | -1.316 |
Token | on |
Token ID | 319 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
anyahu | +2.526 |
jriwal | +2.496 |
DonaldTrump | +2.453 |
uphem | +2.412 |
owa | +2.370 |
Neg logit contributions | |
digitally | -1.251 |
spaced | -1.237 |
MTV | -1.224 |
THR | -1.210 |
patches | -1.210 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
digitally | +0.387 |
MTV | +0.380 |
spaced | +0.376 |
THR | +0.375 |
CST | +0.373 |
Neg logit contributions | |
anyahu | -0.797 |
jriwal | -0.794 |
DonaldTrump | -0.775 |
uphem | -0.765 |
owa | -0.745 |
Token |
|
Token ID | 198 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
anyahu | +0.109 |
jriwal | +0.108 |
DonaldTrump | +0.106 |
uphem | +0.104 |
owa | +0.102 |
Neg logit contributions | |
digitally | -0.049 |
spaced | -0.048 |
MTV | -0.047 |
THR | -0.047 |
patches | -0.047 |
Token | algorithm |
Token ID | 11862 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
digitally | +0.282 |
spaced | +0.277 |
MTV | +0.274 |
THR | +0.272 |
patches | +0.272 |
Neg logit contributions | |
anyahu | -0.537 |
jriwal | -0.531 |
DonaldTrump | -0.522 |
uphem | -0.512 |
owa | -0.501 |
Token | to |
Token ID | 284 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
digitally | +0.871 |
spaced | +0.852 |
MTV | +0.839 |
patches | +0.838 |
THR | +0.837 |
Neg logit contributions | |
anyahu | -1.680 |
jriwal | -1.660 |
DonaldTrump | -1.642 |
uphem | -1.606 |
owa | -1.571 |
Token | all |
Token ID | 477 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
anyahu | +0.212 |
jriwal | +0.210 |
DonaldTrump | +0.206 |
uphem | +0.202 |
owa | +0.198 |
Neg logit contributions | |
digitally | -0.117 |
spaced | -0.115 |
MTV | -0.114 |
THR | -0.113 |
patches | -0.113 |
Token | parts |
Token ID | 3354 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
anyahu | +0.414 |
jriwal | +0.410 |
DonaldTrump | +0.403 |
uphem | +0.395 |
owa | +0.387 |
Neg logit contributions | |
digitally | -0.213 |
spaced | -0.208 |
MTV | -0.208 |
THR | -0.206 |
CST | -0.204 |
Token | for |
Token ID | 329 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
anyahu | +0.962 |
jriwal | +0.954 |
DonaldTrump | +0.936 |
uphem | +0.916 |
owa | +0.898 |
Neg logit contributions | |
digitally | -0.611 |
MTV | -0.602 |
spaced | -0.599 |
THR | -0.596 |
CST | -0.594 |
Token | ( |
Token ID | 357 |
Feature activation | -6.71e+00 |
Pos logit contributions | |
anyahu | +1.008 |
jriwal | +0.998 |
DonaldTrump | +0.979 |
uphem | +0.960 |
owa | +0.940 |
Neg logit contributions | |
digitally | -0.616 |
spaced | -0.605 |
MTV | -0.604 |
THR | -0.598 |
CST | -0.595 |
Token | part |
Token ID | 3911 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
digitally | +3.214 |
spaced | +3.197 |
MTV | +3.118 |
dot | +3.106 |
patches | +3.081 |
Neg logit contributions | |
anyahu | -6.587 |
jriwal | -6.491 |
DonaldTrump | -6.449 |
uphem | -6.303 |
owa | -6.109 |
Token | in |
Token ID | 287 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
digitally | +0.729 |
spaced | +0.718 |
MTV | +0.714 |
THR | +0.708 |
patches | +0.707 |
Neg logit contributions | |
anyahu | -1.023 |
jriwal | -1.013 |
DonaldTrump | -0.992 |
uphem | -0.972 |
owa | -0.948 |
Token | parts |
Token ID | 3354 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
digitally | +0.233 |
spaced | +0.227 |
MTV | +0.224 |
THR | +0.221 |
patches | +0.221 |
Neg logit contributions | |
anyahu | -0.742 |
jriwal | -0.736 |
DonaldTrump | -0.725 |
uphem | -0.714 |
owa | -0.700 |
Token | ) |
Token ID | 8 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
digitally | +0.217 |
spaced | +0.213 |
MTV | +0.212 |
THR | +0.210 |
patches | +0.210 |
Neg logit contributions | |
anyahu | -0.317 |
jriwal | -0.313 |
DonaldTrump | -0.307 |
uphem | -0.301 |
owa | -0.294 |
Token | return |
Token ID | 1441 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
digitally | +0.358 |
spaced | +0.353 |
MTV | +0.352 |
THR | +0.349 |
patches | +0.349 |
Neg logit contributions | |
anyahu | -0.375 |
jriwal | -0.371 |
DonaldTrump | -0.362 |
uphem | -0.353 |
owa | -0.344 |
Token | � |
Token ID | 447 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
anyahu | +1.086 |
jriwal | +1.074 |
DonaldTrump | +1.054 |
uphem | +1.033 |
owa | +1.009 |
Neg logit contributions | |
digitally | -0.731 |
spaced | -0.719 |
MTV | -0.716 |
THR | -0.710 |
patches | -0.708 |
Token | � |
Token ID | 247 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
digitally | +0.425 |
spaced | +0.409 |
MTV | +0.404 |
THR | +0.399 |
patches | +0.392 |
Neg logit contributions | |
anyahu | -2.126 |
jriwal | -2.121 |
DonaldTrump | -2.087 |
uphem | -2.055 |
owa | -2.022 |
Token | t |
Token ID | 83 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
digitally | +0.572 |
spaced | +0.566 |
MTV | +0.561 |
THR | +0.560 |
patches | +0.557 |
Neg logit contributions | |
anyahu | -0.666 |
jriwal | -0.659 |
DonaldTrump | -0.645 |
uphem | -0.629 |
owa | -0.614 |
Token | take |
Token ID | 1011 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
digitally | +0.600 |
spaced | +0.586 |
MTV | +0.585 |
THR | +0.579 |
patches | +0.575 |
Neg logit contributions | |
anyahu | -1.168 |
jriwal | -1.157 |
DonaldTrump | -1.136 |
uphem | -1.112 |
owa | -1.089 |
Token | my |
Token ID | 616 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
digitally | +0.185 |
spaced | +0.181 |
MTV | +0.178 |
THR | +0.177 |
patches | +0.177 |
Neg logit contributions | |
anyahu | -0.465 |
jriwal | -0.461 |
DonaldTrump | -0.454 |
uphem | -0.446 |
owa | -0.438 |
Token | word |
Token ID | 1573 |
Feature activation | -6.46e+00 |
Pos logit contributions | |
digitally | +0.212 |
spaced | +0.204 |
MTV | +0.202 |
THR | +0.197 |
patches | +0.196 |
Neg logit contributions | |
anyahu | -1.049 |
jriwal | -1.041 |
DonaldTrump | -1.026 |
uphem | -1.011 |
owa | -0.995 |
Token | for |
Token ID | 329 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
digitally | +3.182 |
patches | +3.054 |
MTV | +3.042 |
dot | +3.023 |
spaced | +3.021 |
Neg logit contributions | |
anyahu | -6.146 |
jriwal | -6.050 |
DonaldTrump | -6.008 |
uphem | -5.792 |
owa | -5.668 |
Token | it |
Token ID | 340 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
digitally | +1.406 |
MTV | +1.385 |
spaced | +1.373 |
THR | +1.352 |
patches | +1.346 |
Neg logit contributions | |
anyahu | -3.534 |
jriwal | -3.498 |
DonaldTrump | -3.415 |
uphem | -3.360 |
owa | -3.328 |
Token | . |
Token ID | 13 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
digitally | +1.087 |
MTV | +1.053 |
spaced | +1.050 |
patches | +1.038 |
THR | +1.035 |
Neg logit contributions | |
anyahu | -2.037 |
jriwal | -2.024 |
DonaldTrump | -1.979 |
uphem | -1.941 |
owa | -1.901 |
Token | Just |
Token ID | 2329 |
Feature activation | -7.14e-01 |
Pos logit contributions | |
digitally | +0.224 |
MTV | +0.221 |
spaced | +0.219 |
THR | +0.219 |
CST | +0.217 |
Neg logit contributions | |
anyahu | -0.390 |
jriwal | -0.387 |
DonaldTrump | -0.377 |
uphem | -0.370 |
owa | -0.363 |
Token | watch |
Token ID | 2342 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
digitally | +0.373 |
spaced | +0.366 |
MTV | +0.364 |
THR | +0.362 |
patches | +0.360 |
Neg logit contributions | |
anyahu | -0.673 |
jriwal | -0.666 |
DonaldTrump | -0.654 |
uphem | -0.641 |
owa | -0.628 |
Token |
|
Token ID | 198 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
anyahu | +0.072 |
jriwal | +0.072 |
DonaldTrump | +0.070 |
uphem | +0.069 |
owa | +0.068 |
Neg logit contributions | |
digitally | -0.041 |
spaced | -0.041 |
MTV | -0.040 |
THR | -0.040 |
patches | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
digitally | +0.080 |
MTV | +0.078 |
spaced | +0.077 |
THR | +0.077 |
CST | +0.076 |
Neg logit contributions | |
anyahu | -0.204 |
jriwal | -0.203 |
DonaldTrump | -0.199 |
uphem | -0.195 |
owa | -0.191 |
Token | Go |
Token ID | 5247 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
anyahu | +0.038 |
jriwal | +0.038 |
DonaldTrump | +0.036 |
uphem | +0.035 |
owa | +0.034 |
Neg logit contributions | |
digitally | -0.065 |
spaced | -0.064 |
MTV | -0.064 |
THR | -0.063 |
CST | -0.063 |
Token | of |
Token ID | 1659 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
anyahu | +1.142 |
jriwal | +1.108 |
DonaldTrump | +1.076 |
uphem | +1.061 |
owa | +1.043 |
Neg logit contributions | |
spaced | -1.394 |
digitally | -1.385 |
MTV | -1.377 |
THR | -1.365 |
patches | -1.363 |
Token | s |
Token ID | 82 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
anyahu | +0.065 |
jriwal | +0.064 |
DonaldTrump | +0.063 |
uphem | +0.062 |
owa | +0.060 |
Neg logit contributions | |
digitally | -0.050 |
spaced | -0.049 |
MTV | -0.049 |
THR | -0.049 |
patches | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
digitally | +0.918 |
MTV | +0.897 |
THR | +0.890 |
spaced | +0.885 |
Warwick | +0.881 |
Neg logit contributions | |
anyahu | -2.444 |
jriwal | -2.418 |
DonaldTrump | -2.379 |
uphem | -2.326 |
owa | -2.282 |
Token |
|
Token ID | 198 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
THR | +1.917 |
MTV | +1.912 |
digitally | +1.887 |
dot | +1.847 |
CST | +1.810 |
Neg logit contributions | |
jriwal | -7.517 |
anyahu | -7.401 |
DonaldTrump | -7.215 |
uphem | -7.089 |
owa | -6.855 |
Token | C |
Token ID | 34 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
digitally | +3.074 |
MTV | +3.072 |
Warwick | +3.027 |
THR | +3.015 |
COMPLE | +3.006 |
Neg logit contributions | |
jriwal | -4.499 |
anyahu | -4.451 |
DonaldTrump | -4.197 |
uphem | -4.194 |
owa | -4.029 |
Token | offer |
Token ID | 47895 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
digitally | +0.969 |
spaced | +0.959 |
patches | +0.948 |
MTV | +0.947 |
THR | +0.943 |
Neg logit contributions | |
anyahu | -0.977 |
jriwal | -0.964 |
DonaldTrump | -0.949 |
uphem | -0.913 |
owa | -0.882 |
Token | is |
Token ID | 318 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
digitally | +0.872 |
spaced | +0.854 |
MTV | +0.851 |
patches | +0.847 |
THR | +0.844 |
Neg logit contributions | |
anyahu | -1.271 |
jriwal | -1.267 |
DonaldTrump | -1.235 |
uphem | -1.203 |
owa | -1.169 |
Token | seen |
Token ID | 1775 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
digitally | +0.765 |
spaced | +0.753 |
MTV | +0.740 |
THR | +0.735 |
patches | +0.730 |
Neg logit contributions | |
anyahu | -1.318 |
jriwal | -1.306 |
DonaldTrump | -1.282 |
uphem | -1.255 |
owa | -1.223 |
Token | , |
Token ID | 11 |
Feature activation | +8.84e-01 |
Pos logit contributions | |
anyahu | +2.630 |
jriwal | +2.601 |
DonaldTrump | +2.566 |
uphem | +2.495 |
owa | +2.466 |
Neg logit contributions | |
digitally | -1.586 |
spaced | -1.569 |
MTV | -1.566 |
THR | -1.558 |
CST | -1.544 |
Token | please |
Token ID | 3387 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
anyahu | +0.820 |
jriwal | +0.811 |
DonaldTrump | +0.797 |
uphem | +0.781 |
owa | +0.765 |
Neg logit contributions | |
digitally | -0.470 |
spaced | -0.462 |
MTV | -0.459 |
THR | -0.455 |
patches | -0.454 |
Token | help |
Token ID | 1037 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
digitally | +0.804 |
spaced | +0.785 |
MTV | +0.780 |
THR | +0.775 |
patches | +0.770 |
Neg logit contributions | |
anyahu | -1.569 |
jriwal | -1.556 |
DonaldTrump | -1.527 |
uphem | -1.496 |
owa | -1.459 |
Token | spread |
Token ID | 4104 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
digitally | +0.488 |
spaced | +0.476 |
MTV | +0.472 |
THR | +0.467 |
patches | +0.465 |
Neg logit contributions | |
anyahu | -1.293 |
jriwal | -1.282 |
DonaldTrump | -1.261 |
uphem | -1.240 |
owa | -1.217 |
Token | the |
Token ID | 262 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
digitally | +0.847 |
spaced | +0.830 |
MTV | +0.825 |
THR | +0.817 |
patches | +0.814 |
Neg logit contributions | |
anyahu | -1.710 |
jriwal | -1.695 |
DonaldTrump | -1.665 |
uphem | -1.635 |
owa | -1.602 |
Token | word |
Token ID | 1573 |
Feature activation | -6.25e+00 |
Pos logit contributions | |
digitally | +0.388 |
spaced | +0.370 |
MTV | +0.367 |
patches | +0.361 |
THR | +0.356 |
Neg logit contributions | |
anyahu | -2.062 |
jriwal | -2.048 |
DonaldTrump | -2.017 |
uphem | -1.990 |
owa | -1.955 |
Token | about |
Token ID | 546 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
digitally | +3.205 |
THR | +3.024 |
MTV | +3.002 |
spaced | +2.981 |
dot | +2.974 |
Neg logit contributions | |
anyahu | -5.980 |
jriwal | -5.962 |
DonaldTrump | -5.884 |
uphem | -5.800 |
owa | -5.588 |
Token | this |
Token ID | 428 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
digitally | +1.203 |
spaced | +1.175 |
MTV | +1.170 |
THR | +1.162 |
patches | +1.158 |
Neg logit contributions | |
anyahu | -2.372 |
jriwal | -2.353 |
DonaldTrump | -2.302 |
uphem | -2.268 |
owa | -2.222 |
Token | campaign |
Token ID | 1923 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
digitally | +0.367 |
spaced | +0.361 |
MTV | +0.359 |
THR | +0.356 |
patches | +0.356 |
Neg logit contributions | |
anyahu | -0.530 |
jriwal | -0.525 |
DonaldTrump | -0.514 |
uphem | -0.504 |
owa | -0.492 |
Token | to |
Token ID | 284 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
digitally | +0.325 |
spaced | +0.321 |
MTV | +0.319 |
THR | +0.315 |
patches | +0.314 |
Neg logit contributions | |
anyahu | -0.586 |
jriwal | -0.580 |
DonaldTrump | -0.570 |
uphem | -0.558 |
owa | -0.548 |
Token | your |
Token ID | 534 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
digitally | +0.514 |
spaced | +0.502 |
MTV | +0.499 |
THR | +0.493 |
patches | +0.492 |
Neg logit contributions | |
anyahu | -1.241 |
jriwal | -1.229 |
DonaldTrump | -1.211 |
uphem | -1.190 |
owa | -1.167 |
Token | yr |
Token ID | 2417 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
digitally | +1.422 |
spaced | +1.402 |
MTV | +1.397 |
THR | +1.387 |
patches | +1.383 |
Neg logit contributions | |
anyahu | -1.521 |
jriwal | -1.502 |
DonaldTrump | -1.470 |
uphem | -1.432 |
owa | -1.385 |
Token | ds |
Token ID | 9310 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
digitally | +2.097 |
MTV | +2.078 |
spaced | +2.073 |
patches | +2.065 |
THR | +2.063 |
Neg logit contributions | |
anyahu | -1.708 |
jriwal | -1.696 |
DonaldTrump | -1.653 |
uphem | -1.594 |
owa | -1.533 |
Token | Cover |
Token ID | 17546 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
digitally | +0.742 |
spaced | +0.727 |
MTV | +0.726 |
THR | +0.718 |
patches | +0.717 |
Neg logit contributions | |
anyahu | -1.189 |
jriwal | -1.178 |
DonaldTrump | -1.154 |
uphem | -1.132 |
owa | -1.103 |
Token | ) |
Token ID | 8 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
digitally | +1.114 |
MTV | +1.095 |
spaced | +1.083 |
patches | +1.073 |
THR | +1.069 |
Neg logit contributions | |
anyahu | -1.700 |
jriwal | -1.694 |
DonaldTrump | -1.659 |
uphem | -1.619 |
owa | -1.571 |
Token | ( |
Token ID | 357 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
digitally | +0.519 |
spaced | +0.507 |
MTV | +0.505 |
THR | +0.500 |
patches | +0.498 |
Neg logit contributions | |
anyahu | -1.034 |
jriwal | -1.025 |
DonaldTrump | -1.006 |
uphem | -0.988 |
owa | -0.968 |
Token | via |
Token ID | 8869 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
digitally | +0.766 |
spaced | +0.752 |
MTV | +0.752 |
THR | +0.743 |
patches | +0.740 |
Neg logit contributions | |
anyahu | -1.039 |
jriwal | -1.030 |
DonaldTrump | -1.005 |
uphem | -0.986 |
owa | -0.961 |
Token | Sound |
Token ID | 9506 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
MTV | +3.479 |
digitally | +3.394 |
THR | +3.358 |
spaced | +3.287 |
CST | +3.266 |
Neg logit contributions | |
anyahu | -5.500 |
jriwal | -5.446 |
DonaldTrump | -5.288 |
uphem | -5.197 |
owa | -4.996 |
Token | Cloud |
Token ID | 18839 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
digitally | +0.751 |
MTV | +0.740 |
spaced | +0.736 |
patches | +0.726 |
THR | +0.725 |
Neg logit contributions | |
anyahu | -1.547 |
jriwal | -1.527 |
DonaldTrump | -1.501 |
uphem | -1.474 |
owa | -1.432 |
Token | ) |
Token ID | 8 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
digitally | +0.019 |
spaced | +0.019 |
MTV | +0.019 |
THR | +0.019 |
patches | +0.019 |
Neg logit contributions | |
anyahu | -0.029 |
jriwal | -0.029 |
DonaldTrump | -0.028 |
uphem | -0.027 |
owa | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
digitally | +0.358 |
spaced | +0.350 |
MTV | +0.349 |
THR | +0.345 |
patches | +0.343 |
Neg logit contributions | |
anyahu | -0.820 |
jriwal | -0.813 |
DonaldTrump | -0.799 |
uphem | -0.785 |
owa | -0.770 |
Token |
|
Token ID | 198 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
anyahu | +0.592 |
jriwal | +0.589 |
DonaldTrump | +0.577 |
uphem | +0.567 |
owa | +0.556 |
Neg logit contributions | |
digitally | -0.243 |
MTV | -0.238 |
spaced | -0.237 |
THR | -0.235 |
CST | -0.233 |
Token | / |
Token ID | 14 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
anyahu | +0.531 |
jriwal | +0.525 |
DonaldTrump | +0.515 |
uphem | +0.504 |
owa | +0.493 |
Neg logit contributions | |
digitally | -0.368 |
spaced | -0.362 |
MTV | -0.361 |
THR | -0.358 |
patches | -0.357 |
Token | AUT |
Token ID | 39371 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
digitally | +1.807 |
THR | +1.742 |
CST | +1.727 |
spaced | +1.711 |
MTV | +1.710 |
Neg logit contributions | |
jriwal | -3.474 |
anyahu | -3.452 |
DonaldTrump | -3.396 |
uphem | -3.299 |
owa | -3.234 |
Token | OC |
Token ID | 4503 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
digitally | +0.066 |
THR | +0.065 |
CST | +0.065 |
COMPLE | +0.065 |
MTV | +0.064 |
Neg logit contributions | |
jriwal | -0.060 |
anyahu | -0.060 |
DonaldTrump | -0.056 |
uphem | -0.056 |
owa | -0.051 |
Token | R |
Token ID | 49 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
digitally | +1.695 |
spaced | +1.679 |
MTV | +1.677 |
THR | +1.668 |
CST | +1.649 |
Neg logit contributions | |
anyahu | -2.247 |
jriwal | -2.233 |
DonaldTrump | -2.181 |
uphem | -2.128 |
owa | -2.069 |
Token | OP |
Token ID | 3185 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
digitally | +0.628 |
THR | +0.615 |
MTV | +0.613 |
spaced | +0.612 |
CST | +0.605 |
Neg logit contributions | |
anyahu | -1.057 |
jriwal | -1.046 |
DonaldTrump | -1.026 |
uphem | -1.003 |
owa | -0.978 |
Token | / |
Token ID | 14 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
digitally | +0.992 |
THR | +0.988 |
MTV | +0.955 |
CST | +0.954 |
patches | +0.951 |
Neg logit contributions | |
anyahu | -1.849 |
jriwal | -1.833 |
DonaldTrump | -1.822 |
uphem | -1.780 |
owa | -1.697 |
Token | h |
Token ID | 71 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
THR | +3.001 |
digitally | +2.994 |
spaced | +2.923 |
STAT | +2.892 |
dot | +2.888 |
Neg logit contributions | |
anyahu | -5.286 |
jriwal | -5.218 |
DonaldTrump | -5.128 |
uphem | -5.024 |
owa | -4.710 |
Token | 342 |
Token ID | 31575 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
digitally | +0.779 |
spaced | +0.766 |
MTV | +0.762 |
THR | +0.756 |
patches | +0.754 |
Neg logit contributions | |
anyahu | -1.188 |
jriwal | -1.176 |
DonaldTrump | -1.153 |
uphem | -1.129 |
owa | -1.103 |
Token | / |
Token ID | 14 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
digitally | +0.550 |
spaced | +0.535 |
MTV | +0.534 |
THR | +0.531 |
patches | +0.528 |
Neg logit contributions | |
anyahu | -0.946 |
jriwal | -0.942 |
DonaldTrump | -0.927 |
uphem | -0.897 |
owa | -0.874 |
Token | UST |
Token ID | 7759 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
digitally | +2.570 |
MTV | +2.526 |
spaced | +2.511 |
THR | +2.506 |
CST | +2.491 |
Neg logit contributions | |
anyahu | -2.997 |
jriwal | -2.975 |
DonaldTrump | -2.855 |
uphem | -2.829 |
owa | -2.720 |
Token | r |
Token ID | 81 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
digitally | +0.840 |
MTV | +0.824 |
spaced | +0.820 |
THR | +0.819 |
CST | +0.813 |
Neg logit contributions | |
anyahu | -1.279 |
jriwal | -1.268 |
DonaldTrump | -1.240 |
uphem | -1.215 |
owa | -1.179 |
Token | . |
Token ID | 13 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
anyahu | +1.079 |
jriwal | +1.067 |
DonaldTrump | +1.050 |
uphem | +1.027 |
owa | +1.007 |
Neg logit contributions | |
digitally | -0.640 |
MTV | -0.627 |
spaced | -0.626 |
CST | -0.620 |
THR | -0.618 |
Token | Size |
Token ID | 10699 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
anyahu | +0.301 |
jriwal | +0.298 |
DonaldTrump | +0.293 |
uphem | +0.286 |
owa | +0.280 |
Neg logit contributions | |
digitally | -0.199 |
spaced | -0.195 |
MTV | -0.194 |
THR | -0.192 |
CST | -0.192 |
Token | () |
Token ID | 3419 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
anyahu | +0.870 |
jriwal | +0.862 |
DonaldTrump | +0.843 |
uphem | +0.825 |
owa | +0.808 |
Neg logit contributions | |
digitally | -0.631 |
spaced | -0.620 |
MTV | -0.619 |
CST | -0.613 |
patches | -0.612 |
Token | ; |
Token ID | 2162 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
anyahu | +1.309 |
jriwal | +1.291 |
DonaldTrump | +1.262 |
uphem | +1.236 |
owa | +1.213 |
Neg logit contributions | |
digitally | -0.983 |
MTV | -0.966 |
spaced | -0.965 |
CST | -0.955 |
THR | -0.953 |
Token | for |
Token ID | 329 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
anyahu | +0.852 |
jriwal | +0.843 |
DonaldTrump | +0.824 |
uphem | +0.810 |
owa | +0.793 |
Neg logit contributions | |
digitally | -0.600 |
MTV | -0.589 |
spaced | -0.588 |
THR | -0.585 |
CST | -0.582 |
Token | ( |
Token ID | 7 |
Feature activation | -5.74e+00 |
Pos logit contributions | |
digitally | +0.642 |
spaced | +0.631 |
MTV | +0.628 |
THR | +0.623 |
patches | +0.621 |
Neg logit contributions | |
anyahu | -0.963 |
jriwal | -0.954 |
DonaldTrump | -0.935 |
uphem | -0.915 |
owa | -0.893 |
Token | size |
Token ID | 2546 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
spaced | +3.256 |
digitally | +3.232 |
dot | +3.205 |
MTV | +3.202 |
CST | +3.122 |
Neg logit contributions | |
anyahu | -5.110 |
jriwal | -5.066 |
DonaldTrump | -5.015 |
uphem | -4.865 |
owa | -4.750 |
Token | _ |
Token ID | 62 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
anyahu | +0.342 |
jriwal | +0.340 |
DonaldTrump | +0.332 |
uphem | +0.325 |
owa | +0.315 |
Neg logit contributions | |
digitally | -0.262 |
MTV | -0.258 |
spaced | -0.257 |
THR | -0.256 |
patches | -0.255 |
Token | t |
Token ID | 83 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
anyahu | +0.136 |
jriwal | +0.135 |
DonaldTrump | +0.132 |
uphem | +0.129 |
owa | +0.125 |
Neg logit contributions | |
digitally | -0.122 |
spaced | -0.120 |
MTV | -0.120 |
THR | -0.119 |
patches | -0.119 |
Token | u |
Token ID | 334 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
digitally | +1.184 |
spaced | +1.172 |
MTV | +1.157 |
patches | +1.151 |
THR | +1.144 |
Neg logit contributions | |
anyahu | -2.464 |
jriwal | -2.440 |
DonaldTrump | -2.409 |
uphem | -2.364 |
owa | -2.300 |
Token | Part |
Token ID | 7841 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
digitally | +1.176 |
spaced | +1.170 |
MTV | +1.166 |
THR | +1.156 |
patches | +1.155 |
Neg logit contributions | |
anyahu | -1.207 |
jriwal | -1.188 |
DonaldTrump | -1.176 |
uphem | -1.137 |
owa | -1.097 |
Token | analysis |
Token ID | 3781 |
Feature activation | -9.97e-01 |
Pos logit contributions | |
digitally | +0.427 |
spaced | +0.417 |
MTV | +0.415 |
THR | +0.410 |
patches | +0.410 |
Neg logit contributions | |
anyahu | -0.854 |
jriwal | -0.846 |
DonaldTrump | -0.831 |
uphem | -0.816 |
owa | -0.800 |
Token | methods |
Token ID | 5050 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
digitally | +0.468 |
spaced | +0.456 |
MTV | +0.451 |
patches | +0.450 |
THR | +0.448 |
Neg logit contributions | |
anyahu | -0.997 |
jriwal | -0.988 |
DonaldTrump | -0.971 |
uphem | -0.952 |
owa | -0.934 |
Token | , |
Token ID | 11 |
Feature activation | -9.70e-01 |
Pos logit contributions | |
digitally | +0.667 |
spaced | +0.654 |
MTV | +0.650 |
THR | +0.645 |
patches | +0.643 |
Neg logit contributions | |
anyahu | -1.199 |
jriwal | -1.188 |
DonaldTrump | -1.167 |
uphem | -1.144 |
owa | -1.120 |
Token | says |
Token ID | 1139 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
digitally | +0.512 |
spaced | +0.503 |
MTV | +0.499 |
THR | +0.495 |
patches | +0.494 |
Neg logit contributions | |
anyahu | -0.909 |
jriwal | -0.900 |
DonaldTrump | -0.884 |
uphem | -0.867 |
owa | -0.847 |
Token | T |
Token ID | 309 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
anyahu | +0.628 |
jriwal | +0.621 |
DonaldTrump | +0.611 |
uphem | +0.600 |
owa | +0.589 |
Neg logit contributions | |
digitally | -0.327 |
spaced | -0.321 |
MTV | -0.317 |
patches | -0.314 |
THR | -0.314 |
Token | omm |
Token ID | 2002 |
Feature activation | -5.73e+00 |
Pos logit contributions | |
anyahu | +0.127 |
jriwal | +0.126 |
DonaldTrump | +0.122 |
uphem | +0.120 |
owa | +0.118 |
Neg logit contributions | |
digitally | -0.118 |
spaced | -0.116 |
MTV | -0.115 |
THR | -0.115 |
CST | -0.114 |
Token | as |
Token ID | 292 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
MTV | +3.703 |
CST | +3.642 |
THR | +3.639 |
Warwick | +3.639 |
Carnegie | +3.615 |
Neg logit contributions | |
jriwal | -4.568 |
anyahu | -4.533 |
DonaldTrump | -4.477 |
uphem | -4.284 |
owa | -4.094 |
Token | o |
Token ID | 78 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
MTV | +1.685 |
digitally | +1.685 |
dot | +1.674 |
patches | +1.666 |
THR | +1.665 |
Neg logit contributions | |
jriwal | -1.703 |
anyahu | -1.677 |
DonaldTrump | -1.658 |
uphem | -1.596 |
owa | -1.507 |
Token | Tre |
Token ID | 4700 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
digitally | +1.923 |
MTV | +1.914 |
spaced | +1.889 |
THR | +1.887 |
Warwick | +1.880 |
Neg logit contributions | |
anyahu | -2.950 |
jriwal | -2.940 |
DonaldTrump | -2.871 |
uphem | -2.797 |
owa | -2.717 |
Token | u |
Token ID | 84 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
digitally | +1.825 |
MTV | +1.809 |
spaced | +1.803 |
THR | +1.790 |
patches | +1.788 |
Neg logit contributions | |
anyahu | -1.710 |
jriwal | -1.683 |
DonaldTrump | -1.664 |
uphem | -1.592 |
owa | -1.545 |
Token | , |
Token ID | 11 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
digitally | +0.367 |
spaced | +0.361 |
MTV | +0.359 |
THR | +0.357 |
patches | +0.356 |
Neg logit contributions | |
anyahu | -0.475 |
jriwal | -0.470 |
DonaldTrump | -0.460 |
uphem | -0.450 |
owa | -0.439 |
Token | lo |
Token ID | 5439 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
THR | +1.275 |
spaced | +1.251 |
MTV | +1.242 |
patches | +1.238 |
CST | +1.238 |
Neg logit contributions | |
jriwal | -1.611 |
DonaldTrump | -1.579 |
anyahu | -1.560 |
uphem | -1.494 |
owa | -1.459 |
Token | " |
Token ID | 366 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
anyahu | +1.220 |
jriwal | +1.183 |
DonaldTrump | +1.166 |
owa | +1.157 |
uphem | +1.154 |
Neg logit contributions | |
digitally | -0.304 |
spaced | -0.297 |
Warwick | -0.272 |
MTV | -0.270 |
breakup | -0.267 |
Token | 25 |
Token ID | 1495 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
anyahu | +0.308 |
jriwal | +0.306 |
DonaldTrump | +0.299 |
uphem | +0.294 |
owa | +0.287 |
Neg logit contributions | |
digitally | -0.177 |
spaced | -0.174 |
MTV | -0.173 |
THR | -0.172 |
CST | -0.171 |
Token | 56 |
Token ID | 3980 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
digitally | +0.539 |
MTV | +0.529 |
spaced | +0.529 |
THR | +0.525 |
patches | +0.522 |
Neg logit contributions | |
anyahu | -0.898 |
jriwal | -0.895 |
DonaldTrump | -0.874 |
uphem | -0.855 |
owa | -0.831 |
Token | "] |
Token ID | 8973 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
digitally | +0.317 |
MTV | +0.313 |
spaced | +0.312 |
patches | +0.311 |
THR | +0.311 |
Neg logit contributions | |
anyahu | -0.310 |
jriwal | -0.310 |
DonaldTrump | -0.302 |
uphem | -0.292 |
owa | -0.282 |
Token | [ |
Token ID | 685 |
Feature activation | -5.70e+00 |
Pos logit contributions | |
digitally | +0.755 |
THR | +0.753 |
MTV | +0.753 |
CST | +0.745 |
spaced | +0.741 |
Neg logit contributions | |
jriwal | -1.330 |
anyahu | -1.315 |
DonaldTrump | -1.284 |
uphem | -1.259 |
owa | -1.224 |
Token | Black |
Token ID | 9915 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
MTV | +4.876 |
CEPT | +4.854 |
digitally | +4.847 |
CST | +4.818 |
patches | +4.809 |
Neg logit contributions | |
jriwal | -3.469 |
uphem | -3.228 |
anyahu | -3.087 |
DonaldTrump | -3.043 |
owa | -2.869 |
Token | E |
Token ID | 36 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
digitally | +0.866 |
spaced | +0.849 |
MTV | +0.844 |
THR | +0.837 |
patches | +0.834 |
Neg logit contributions | |
anyahu | -1.600 |
jriwal | -1.588 |
DonaldTrump | -1.558 |
uphem | -1.528 |
owa | -1.495 |
Token | lo |
Token ID | 5439 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
spaced | +2.015 |
digitally | +2.014 |
CST | +2.001 |
THR | +1.999 |
dot | +1.979 |
Neg logit contributions | |
jriwal | -2.712 |
DonaldTrump | -2.649 |
anyahu | -2.601 |
uphem | -2.492 |
owa | -2.446 |
Token | " |
Token ID | 366 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
anyahu | +1.178 |
jriwal | +1.168 |
DonaldTrump | +1.153 |
uphem | +1.136 |
owa | +1.119 |
Neg logit contributions | |
digitally | -0.204 |
spaced | -0.195 |
MTV | -0.192 |
THR | -0.187 |
patches | -0.186 |
Token | 25 |
Token ID | 1495 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
digitally | +0.528 |
MTV | +0.520 |
THR | +0.517 |
spaced | +0.516 |
CST | +0.515 |
Neg logit contributions | |
jriwal | -0.899 |
anyahu | -0.899 |
DonaldTrump | -0.871 |
uphem | -0.860 |
owa | -0.835 |
Token | = |
Token ID | 796 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
digitally | +0.124 |
spaced | +0.122 |
MTV | +0.121 |
THR | +0.121 |
patches | +0.120 |
Neg logit contributions | |
anyahu | -0.134 |
jriwal | -0.132 |
DonaldTrump | -0.129 |
uphem | -0.126 |
owa | -0.123 |
Token | 1 |
Token ID | 352 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
anyahu | +0.803 |
jriwal | +0.792 |
DonaldTrump | +0.781 |
uphem | +0.763 |
owa | +0.751 |
Neg logit contributions | |
digitally | -0.486 |
spaced | -0.477 |
MTV | -0.476 |
patches | -0.470 |
THR | -0.469 |
Token | Try |
Token ID | 9993 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
digitally | +0.678 |
spaced | +0.667 |
MTV | +0.661 |
THR | +0.657 |
patches | +0.654 |
Neg logit contributions | |
anyahu | -1.031 |
jriwal | -1.022 |
DonaldTrump | -1.003 |
uphem | -0.982 |
owa | -0.957 |
Token | { |
Token ID | 1391 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
anyahu | +0.503 |
jriwal | +0.496 |
DonaldTrump | +0.484 |
uphem | +0.471 |
owa | +0.458 |
Neg logit contributions | |
digitally | -0.570 |
spaced | -0.563 |
MTV | -0.561 |
THR | -0.558 |
patches | -0.557 |
Token | for |
Token ID | 329 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
digitally | +0.440 |
spaced | +0.433 |
MTV | +0.430 |
THR | +0.428 |
CST | +0.426 |
Neg logit contributions | |
anyahu | -0.636 |
jriwal | -0.629 |
DonaldTrump | -0.617 |
uphem | -0.604 |
owa | -0.589 |
Token | ($ |
Token ID | 7198 |
Feature activation | -5.70e+00 |
Pos logit contributions | |
digitally | +1.152 |
spaced | +1.133 |
MTV | +1.128 |
THR | +1.120 |
patches | +1.116 |
Neg logit contributions | |
anyahu | -1.664 |
jriwal | -1.650 |
DonaldTrump | -1.614 |
uphem | -1.581 |
owa | -1.541 |
Token | i |
Token ID | 72 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
digitally | +3.840 |
MTV | +3.822 |
dot | +3.796 |
spaced | +3.786 |
patches | +3.755 |
Neg logit contributions | |
jriwal | -4.352 |
anyahu | -4.340 |
DonaldTrump | -4.220 |
uphem | -4.156 |
owa | -3.929 |
Token | ; |
Token ID | 2162 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
digitally | +0.287 |
spaced | +0.283 |
MTV | +0.282 |
THR | +0.280 |
patches | +0.280 |
Neg logit contributions | |
anyahu | -0.294 |
jriwal | -0.291 |
DonaldTrump | -0.285 |
uphem | -0.278 |
owa | -0.269 |
Token | $ |
Token ID | 720 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
digitally | +0.083 |
spaced | +0.082 |
MTV | +0.081 |
THR | +0.081 |
CST | +0.081 |
Neg logit contributions | |
anyahu | -0.109 |
jriwal | -0.108 |
DonaldTrump | -0.106 |
uphem | -0.104 |
owa | -0.101 |
Token | i |
Token ID | 72 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
digitally | +1.740 |
spaced | +1.715 |
MTV | +1.713 |
patches | +1.701 |
dot | +1.699 |
Neg logit contributions | |
anyahu | -1.842 |
jriwal | -1.822 |
DonaldTrump | -1.790 |
uphem | -1.755 |
owa | -1.677 |
Token | - |
Token ID | 532 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
digitally | +0.235 |
spaced | +0.231 |
MTV | +0.230 |
patches | +0.229 |
THR | +0.228 |
Neg logit contributions | |
anyahu | -0.361 |
jriwal | -0.359 |
DonaldTrump | -0.353 |
uphem | -0.346 |
owa | -0.336 |
Token | I |
Token ID | 40 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
digitally | +0.491 |
spaced | +0.482 |
MTV | +0.480 |
THR | +0.476 |
patches | +0.474 |
Neg logit contributions | |
anyahu | -0.825 |
jriwal | -0.818 |
DonaldTrump | -0.801 |
uphem | -0.786 |
owa | -0.768 |
Token | laugh |
Token ID | 6487 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
digitally | +0.289 |
spaced | +0.283 |
MTV | +0.282 |
THR | +0.280 |
patches | +0.279 |
Neg logit contributions | |
anyahu | -0.487 |
jriwal | -0.482 |
DonaldTrump | -0.472 |
uphem | -0.464 |
owa | -0.455 |
Token | lightly |
Token ID | 15376 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
digitally | +0.326 |
spaced | +0.320 |
MTV | +0.319 |
THR | +0.316 |
CST | +0.315 |
Neg logit contributions | |
anyahu | -0.575 |
jriwal | -0.569 |
DonaldTrump | -0.559 |
uphem | -0.549 |
owa | -0.537 |
Token | , |
Token ID | 11 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
digitally | +0.114 |
spaced | +0.111 |
MTV | +0.111 |
THR | +0.110 |
CST | +0.110 |
Neg logit contributions | |
anyahu | -0.193 |
jriwal | -0.191 |
DonaldTrump | -0.187 |
uphem | -0.184 |
owa | -0.180 |
Token | pull |
Token ID | 2834 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
digitally | +0.630 |
spaced | +0.615 |
MTV | +0.612 |
THR | +0.605 |
CST | +0.603 |
Neg logit contributions | |
anyahu | -1.568 |
jriwal | -1.554 |
DonaldTrump | -1.528 |
uphem | -1.504 |
owa | -1.476 |
Token | her |
Token ID | 607 |
Feature activation | -5.69e+00 |
Pos logit contributions | |
anyahu | +0.266 |
jriwal | +0.264 |
DonaldTrump | +0.259 |
uphem | +0.257 |
owa | +0.254 |
Neg logit contributions | |
digitally | -0.089 |
spaced | -0.087 |
MTV | -0.087 |
CST | -0.086 |
THR | -0.085 |
Token | towards |
Token ID | 3371 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
digitally | +2.582 |
spaced | +2.512 |
patches | +2.479 |
MTV | +2.473 |
THR | +2.438 |
Neg logit contributions | |
anyahu | -5.805 |
jriwal | -5.780 |
DonaldTrump | -5.752 |
uphem | -5.511 |
owa | -5.352 |
Token | me |
Token ID | 502 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
digitally | +0.884 |
spaced | +0.861 |
MTV | +0.849 |
THR | +0.836 |
patches | +0.834 |
Neg logit contributions | |
anyahu | -3.389 |
jriwal | -3.372 |
DonaldTrump | -3.314 |
uphem | -3.249 |
owa | -3.202 |
Token | and |
Token ID | 290 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
digitally | +0.930 |
spaced | +0.912 |
MTV | +0.905 |
THR | +0.897 |
patches | +0.895 |
Neg logit contributions | |
anyahu | -1.611 |
jriwal | -1.596 |
DonaldTrump | -1.569 |
uphem | -1.535 |
owa | -1.501 |
Token | kiss |
Token ID | 9245 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
digitally | +1.141 |
spaced | +1.121 |
MTV | +1.111 |
THR | +1.103 |
patches | +1.096 |
Neg logit contributions | |
anyahu | -2.076 |
jriwal | -2.060 |
DonaldTrump | -2.030 |
uphem | -1.976 |
owa | -1.929 |
Token | her |
Token ID | 607 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
digitally | +0.704 |
spaced | +0.684 |
MTV | +0.678 |
THR | +0.669 |
patches | +0.665 |
Neg logit contributions | |
anyahu | -2.188 |
jriwal | -2.171 |
DonaldTrump | -2.138 |
uphem | -2.101 |
owa | -2.063 |
Token | � |
Token ID | 447 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
digitally | +0.055 |
spaced | +0.054 |
MTV | +0.053 |
THR | +0.053 |
patches | +0.053 |
Neg logit contributions | |
anyahu | -0.093 |
jriwal | -0.092 |
DonaldTrump | -0.090 |
uphem | -0.088 |
owa | -0.087 |
Token | � |
Token ID | 247 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
digitally | +0.271 |
spaced | +0.261 |
MTV | +0.258 |
THR | +0.253 |
patches | +0.251 |
Neg logit contributions | |
anyahu | -1.264 |
jriwal | -1.258 |
DonaldTrump | -1.238 |
uphem | -1.219 |
owa | -1.199 |
Token | ll |
Token ID | 297 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
digitally | +0.451 |
spaced | +0.445 |
MTV | +0.443 |
THR | +0.440 |
patches | +0.439 |
Neg logit contributions | |
anyahu | -0.471 |
jriwal | -0.465 |
DonaldTrump | -0.455 |
uphem | -0.443 |
owa | -0.431 |
Token | take |
Token ID | 1011 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
digitally | +0.605 |
spaced | +0.592 |
MTV | +0.588 |
THR | +0.584 |
patches | +0.583 |
Neg logit contributions | |
anyahu | -1.071 |
jriwal | -1.060 |
DonaldTrump | -1.041 |
uphem | -1.019 |
owa | -0.991 |
Token | your |
Token ID | 534 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
digitally | +0.127 |
spaced | +0.124 |
MTV | +0.123 |
THR | +0.122 |
CST | +0.121 |
Neg logit contributions | |
anyahu | -0.263 |
jriwal | -0.261 |
DonaldTrump | -0.256 |
uphem | -0.252 |
owa | -0.247 |
Token | word |
Token ID | 1573 |
Feature activation | -5.64e+00 |
Pos logit contributions | |
digitally | +0.096 |
spaced | +0.094 |
MTV | +0.093 |
patches | +0.093 |
THR | +0.093 |
Neg logit contributions | |
anyahu | -0.161 |
jriwal | -0.159 |
DonaldTrump | -0.156 |
uphem | -0.153 |
owa | -0.149 |
Token | for |
Token ID | 329 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
digitally | +2.844 |
dot | +2.714 |
patches | +2.707 |
spaced | +2.692 |
MTV | +2.659 |
Neg logit contributions | |
anyahu | -5.393 |
DonaldTrump | -5.345 |
jriwal | -5.332 |
uphem | -5.128 |
owa | -4.925 |
Token | it |
Token ID | 340 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
digitally | +0.449 |
spaced | +0.439 |
MTV | +0.436 |
patches | +0.431 |
THR | +0.431 |
Neg logit contributions | |
anyahu | -1.037 |
jriwal | -1.027 |
DonaldTrump | -1.009 |
uphem | -0.992 |
owa | -0.973 |
Token | . |
Token ID | 13 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
digitally | +0.915 |
spaced | +0.892 |
MTV | +0.886 |
THR | +0.880 |
patches | +0.879 |
Neg logit contributions | |
anyahu | -1.673 |
jriwal | -1.659 |
DonaldTrump | -1.630 |
uphem | -1.598 |
owa | -1.558 |
Token | Perhaps |
Token ID | 8673 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
digitally | +0.375 |
MTV | +0.368 |
spaced | +0.367 |
THR | +0.364 |
patches | +0.361 |
Neg logit contributions | |
anyahu | -0.793 |
jriwal | -0.786 |
DonaldTrump | -0.769 |
uphem | -0.756 |
owa | -0.740 |
Token | there |
Token ID | 612 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
digitally | +0.138 |
spaced | +0.135 |
MTV | +0.134 |
THR | +0.133 |
CST | +0.132 |
Neg logit contributions | |
anyahu | -0.298 |
jriwal | -0.295 |
DonaldTrump | -0.290 |
uphem | -0.285 |
owa | -0.280 |
Token | , |
Token ID | 11 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
anyahu | +0.920 |
jriwal | +0.910 |
DonaldTrump | +0.893 |
uphem | +0.876 |
owa | +0.860 |
Neg logit contributions | |
digitally | -0.558 |
spaced | -0.548 |
MTV | -0.545 |
THR | -0.541 |
patches | -0.539 |
Token | San |
Token ID | 2986 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
anyahu | +1.001 |
jriwal | +0.991 |
DonaldTrump | +0.974 |
uphem | +0.959 |
owa | +0.939 |
Neg logit contributions | |
digitally | -0.565 |
spaced | -0.556 |
MTV | -0.552 |
THR | -0.549 |
patches | -0.546 |
Token | i |
Token ID | 72 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
digitally | +0.588 |
MTV | +0.579 |
spaced | +0.579 |
THR | +0.574 |
patches | +0.573 |
Neg logit contributions | |
anyahu | -0.683 |
jriwal | -0.675 |
DonaldTrump | -0.663 |
uphem | -0.644 |
owa | -0.625 |
Token | , |
Token ID | 11 |
Feature activation | -8.47e-02 |
Pos logit contributions | |
digitally | +0.059 |
spaced | +0.058 |
MTV | +0.058 |
THR | +0.057 |
patches | +0.057 |
Neg logit contributions | |
anyahu | -0.093 |
jriwal | -0.092 |
DonaldTrump | -0.090 |
uphem | -0.089 |
owa | -0.087 |
Token | T |
Token ID | 309 |
Feature activation | -4.79e-03 |
Pos logit contributions | |
digitally | +0.045 |
spaced | +0.044 |
MTV | +0.044 |
THR | +0.043 |
patches | +0.043 |
Neg logit contributions | |
anyahu | -0.079 |
jriwal | -0.078 |
DonaldTrump | -0.077 |
uphem | -0.075 |
owa | -0.074 |
Token | omm |
Token ID | 2002 |
Feature activation | -5.55e+00 |
Pos logit contributions | |
digitally | +0.003 |
spaced | +0.003 |
MTV | +0.003 |
THR | +0.003 |
patches | +0.003 |
Neg logit contributions | |
anyahu | -0.004 |
jriwal | -0.004 |
DonaldTrump | -0.003 |
uphem | -0.003 |
owa | -0.003 |
Token | i |
Token ID | 72 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
MTV | +3.700 |
dot | +3.619 |
THR | +3.601 |
digitally | +3.597 |
Warwick | +3.595 |
Neg logit contributions | |
jriwal | -4.293 |
DonaldTrump | -4.284 |
anyahu | -4.276 |
uphem | -4.004 |
owa | -3.806 |
Token | L |
Token ID | 406 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
digitally | +0.600 |
spaced | +0.591 |
MTV | +0.591 |
THR | +0.584 |
CST | +0.583 |
Neg logit contributions | |
anyahu | -0.961 |
jriwal | -0.953 |
DonaldTrump | -0.936 |
uphem | -0.914 |
owa | -0.890 |
Token | ä |
Token ID | 11033 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
digitally | +0.493 |
spaced | +0.487 |
MTV | +0.486 |
patches | +0.482 |
THR | +0.481 |
Neg logit contributions | |
anyahu | -0.516 |
jriwal | -0.509 |
DonaldTrump | -0.500 |
uphem | -0.483 |
owa | -0.468 |
Token | nt |
Token ID | 429 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
anyahu | +0.716 |
jriwal | +0.707 |
DonaldTrump | +0.687 |
uphem | +0.674 |
owa | +0.657 |
Neg logit contributions | |
digitally | -0.721 |
spaced | -0.710 |
MTV | -0.707 |
THR | -0.703 |
patches | -0.701 |
Token | inen |
Token ID | 42326 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
anyahu | +0.413 |
jriwal | +0.408 |
DonaldTrump | +0.397 |
uphem | +0.389 |
owa | +0.381 |
Neg logit contributions | |
digitally | -0.385 |
spaced | -0.379 |
MTV | -0.377 |
THR | -0.375 |
patches | -0.374 |
Token | during |
Token ID | 1141 |
Feature activation | +9.93e-01 |
Pos logit contributions | |
anyahu | +1.052 |
jriwal | +1.043 |
DonaldTrump | +1.020 |
uphem | +1.001 |
owa | +0.985 |
Neg logit contributions | |
digitally | -0.641 |
MTV | -0.630 |
spaced | -0.629 |
CST | -0.623 |
THR | -0.623 |
Token | their |
Token ID | 511 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
anyahu | +0.929 |
jriwal | +0.920 |
DonaldTrump | +0.904 |
uphem | +0.887 |
owa | +0.869 |
Neg logit contributions | |
digitally | -0.519 |
spaced | -0.509 |
MTV | -0.506 |
THR | -0.502 |
patches | -0.500 |
Token | performance |
Token ID | 2854 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
anyahu | +0.723 |
jriwal | +0.715 |
DonaldTrump | +0.701 |
uphem | +0.685 |
owa | +0.669 |
Neg logit contributions | |
digitally | -0.554 |
spaced | -0.546 |
MTV | -0.543 |
THR | -0.539 |
patches | -0.538 |
Token | , |
Token ID | 11 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
anyahu | +0.837 |
jriwal | +0.829 |
DonaldTrump | +0.813 |
uphem | +0.798 |
owa | +0.782 |
Neg logit contributions | |
digitally | -0.475 |
spaced | -0.466 |
MTV | -0.465 |
THR | -0.460 |
CST | -0.458 |
Token | but |
Token ID | 475 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
anyahu | +0.595 |
jriwal | +0.590 |
DonaldTrump | +0.578 |
uphem | +0.568 |
owa | +0.557 |
Neg logit contributions | |
digitally | -0.338 |
spaced | -0.331 |
MTV | -0.330 |
THR | -0.327 |
CST | -0.326 |
Token | rather |
Token ID | 2138 |
Feature activation | -5.55e+00 |
Pos logit contributions | |
digitally | +0.640 |
spaced | +0.629 |
MTV | +0.623 |
patches | +0.621 |
THR | +0.619 |
Neg logit contributions | |
anyahu | -1.140 |
jriwal | -1.128 |
DonaldTrump | -1.112 |
uphem | -1.089 |
owa | -1.066 |
Token | a |
Token ID | 257 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
digitally | +2.649 |
spaced | +2.624 |
patches | +2.619 |
MTV | +2.573 |
THR | +2.549 |
Neg logit contributions | |
anyahu | -5.420 |
jriwal | -5.360 |
DonaldTrump | -5.314 |
uphem | -5.219 |
owa | -5.083 |
Token | different |
Token ID | 1180 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
digitally | +0.862 |
spaced | +0.847 |
MTV | +0.836 |
patches | +0.829 |
THR | +0.827 |
Neg logit contributions | |
anyahu | -1.582 |
jriwal | -1.563 |
DonaldTrump | -1.535 |
uphem | -1.509 |
owa | -1.475 |
Token | stance |
Token ID | 12046 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
digitally | +0.410 |
spaced | +0.402 |
MTV | +0.400 |
THR | +0.395 |
patches | +0.394 |
Neg logit contributions | |
anyahu | -0.820 |
jriwal | -0.812 |
DonaldTrump | -0.798 |
uphem | -0.785 |
owa | -0.768 |
Token | on |
Token ID | 319 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
anyahu | +0.699 |
jriwal | +0.694 |
DonaldTrump | +0.680 |
uphem | +0.668 |
owa | +0.654 |
Neg logit contributions | |
digitally | -0.382 |
spaced | -0.375 |
MTV | -0.374 |
THR | -0.371 |
CST | -0.370 |
Token | how |
Token ID | 703 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
anyahu | +0.426 |
jriwal | +0.423 |
DonaldTrump | +0.416 |
uphem | +0.408 |
owa | +0.399 |
Neg logit contributions | |
digitally | -0.231 |
spaced | -0.227 |
MTV | -0.225 |
THR | -0.224 |
patches | -0.222 |
Token | Bounty |
Token ID | 38166 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
anyahu | +0.415 |
jriwal | +0.409 |
DonaldTrump | +0.402 |
uphem | +0.393 |
owa | +0.385 |
Neg logit contributions | |
digitally | -0.297 |
spaced | -0.292 |
MTV | -0.289 |
THR | -0.287 |
patches | -0.287 |
Token | gate |
Token ID | 10494 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
digitally | +0.645 |
MTV | +0.633 |
spaced | +0.633 |
THR | +0.628 |
patches | +0.627 |
Neg logit contributions | |
anyahu | -0.954 |
jriwal | -0.944 |
DonaldTrump | -0.928 |
uphem | -0.905 |
owa | -0.882 |
Token | Lit |
Token ID | 25659 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
anyahu | +1.332 |
jriwal | +1.319 |
DonaldTrump | +1.289 |
uphem | +1.268 |
owa | +1.244 |
Neg logit contributions | |
digitally | -0.911 |
spaced | -0.899 |
MTV | -0.885 |
THR | -0.880 |
patches | -0.877 |
Token | igation |
Token ID | 7065 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
digitally | +0.303 |
MTV | +0.301 |
spaced | +0.300 |
THR | +0.299 |
CST | +0.299 |
Neg logit contributions | |
anyahu | -0.168 |
jriwal | -0.166 |
DonaldTrump | -0.163 |
uphem | -0.154 |
owa | -0.148 |
Token | Q |
Token ID | 1195 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
anyahu | +1.201 |
jriwal | +1.188 |
DonaldTrump | +1.163 |
uphem | +1.146 |
owa | +1.124 |
Neg logit contributions | |
digitally | -0.707 |
spaced | -0.697 |
MTV | -0.687 |
patches | -0.682 |
THR | -0.680 |
Token | & |
Token ID | 5 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
anyahu | +1.649 |
jriwal | +1.633 |
DonaldTrump | +1.608 |
uphem | +1.584 |
owa | +1.556 |
Neg logit contributions | |
digitally | -0.799 |
spaced | -0.787 |
MTV | -0.773 |
patches | -0.766 |
CST | -0.765 |
Token | As |
Token ID | 1722 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
MTV | +2.488 |
digitally | +2.469 |
patches | +2.453 |
dot | +2.414 |
THR | +2.408 |
Neg logit contributions | |
jriwal | -5.427 |
anyahu | -5.390 |
DonaldTrump | -5.280 |
uphem | -5.169 |
owa | -5.059 |
Token | . |
Token ID | 13 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
anyahu | +0.178 |
jriwal | +0.177 |
DonaldTrump | +0.173 |
uphem | +0.170 |
owa | +0.167 |
Neg logit contributions | |
digitally | -0.099 |
spaced | -0.097 |
MTV | -0.097 |
THR | -0.096 |
patches | -0.096 |
Token |
|
Token ID | 198 |
Feature activation | +2.62e-03 |
Pos logit contributions | |
digitally | +0.106 |
spaced | +0.103 |
MTV | +0.103 |
THR | +0.102 |
CST | +0.102 |
Neg logit contributions | |
anyahu | -0.202 |
jriwal | -0.201 |
DonaldTrump | -0.197 |
uphem | -0.193 |
owa | -0.189 |
Token |
|
Token ID | 198 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
anyahu | +0.003 |
jriwal | +0.003 |
DonaldTrump | +0.003 |
uphem | +0.003 |
owa | +0.003 |
Neg logit contributions | |
digitally | -0.001 |
MTV | -0.001 |
spaced | -0.001 |
THR | -0.001 |
CST | -0.001 |
Token | Maybe |
Token ID | 13300 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
anyahu | +0.228 |
jriwal | +0.226 |
DonaldTrump | +0.221 |
uphem | +0.216 |
owa | +0.211 |
Neg logit contributions | |
digitally | -0.187 |
MTV | -0.184 |
spaced | -0.184 |
THR | -0.182 |
patches | -0.181 |
Token | D |
Token ID | 360 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
digitally | +0.285 |
spaced | +0.279 |
MTV | +0.276 |
THR | +0.273 |
patches | +0.273 |
Neg logit contributions | |
anyahu | -0.644 |
jriwal | -0.638 |
DonaldTrump | -0.627 |
uphem | -0.616 |
owa | -0.605 |
Token | IV |
Token ID | 3824 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
digitally | +0.300 |
spaced | +0.297 |
MTV | +0.296 |
patches | +0.293 |
THR | +0.292 |
Neg logit contributions | |
anyahu | -0.303 |
jriwal | -0.299 |
DonaldTrump | -0.292 |
uphem | -0.284 |
owa | -0.275 |
Token | IDE |
Token ID | 14114 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
digitally | +2.032 |
MTV | +1.995 |
THR | +1.989 |
spaced | +1.979 |
CST | +1.965 |
Neg logit contributions | |
anyahu | -2.962 |
jriwal | -2.948 |
DonaldTrump | -2.885 |
uphem | -2.828 |
owa | -2.759 |
Token | AND |
Token ID | 5357 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
digitally | +1.411 |
spaced | +1.384 |
MTV | +1.384 |
THR | +1.380 |
CST | +1.359 |
Neg logit contributions | |
anyahu | -2.475 |
jriwal | -2.448 |
DonaldTrump | -2.397 |
uphem | -2.361 |
owa | -2.306 |
Token | CON |
Token ID | 7102 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
digitally | +1.778 |
THR | +1.773 |
MTV | +1.758 |
spaced | +1.737 |
CST | +1.726 |
Neg logit contributions | |
anyahu | -3.588 |
jriwal | -3.554 |
DonaldTrump | -3.468 |
uphem | -3.421 |
owa | -3.326 |
Token | QU |
Token ID | 10917 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
digitally | +0.268 |
spaced | +0.263 |
MTV | +0.260 |
patches | +0.259 |
THR | +0.257 |
Neg logit contributions | |
anyahu | -0.521 |
jriwal | -0.516 |
DonaldTrump | -0.508 |
uphem | -0.499 |
owa | -0.489 |
Token | ER |
Token ID | 1137 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
THR | +1.885 |
digitally | +1.873 |
MTV | +1.862 |
CEPT | +1.851 |
CST | +1.822 |
Neg logit contributions | |
anyahu | -6.102 |
jriwal | -6.083 |
DonaldTrump | -5.909 |
uphem | -5.869 |
owa | -5.732 |
Token | … |
Token ID | 1399 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
digitally | +2.122 |
MTV | +2.090 |
THR | +2.080 |
spaced | +2.070 |
CST | +2.046 |
Neg logit contributions | |
anyahu | -3.687 |
jriwal | -3.649 |
DonaldTrump | -3.569 |
uphem | -3.506 |
owa | -3.421 |
Token | how |
Token ID | 703 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
digitally | +1.134 |
spaced | +1.112 |
MTV | +1.111 |
THR | +1.104 |
patches | +1.092 |
Neg logit contributions | |
anyahu | -1.972 |
jriwal | -1.952 |
DonaldTrump | -1.914 |
uphem | -1.878 |
owa | -1.837 |
Token | did |
Token ID | 750 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
digitally | +0.500 |
spaced | +0.491 |
MTV | +0.488 |
THR | +0.484 |
patches | +0.482 |
Neg logit contributions | |
anyahu | -0.927 |
jriwal | -0.918 |
DonaldTrump | -0.902 |
uphem | -0.885 |
owa | -0.866 |
Token | you |
Token ID | 345 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
digitally | +0.426 |
spaced | +0.416 |
MTV | +0.410 |
patches | +0.406 |
THR | +0.405 |
Neg logit contributions | |
anyahu | -1.186 |
jriwal | -1.177 |
DonaldTrump | -1.159 |
uphem | -1.139 |
owa | -1.119 |
Token |
|
Token ID | 198 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
anyahu | +0.523 |
jriwal | +0.520 |
DonaldTrump | +0.510 |
uphem | +0.501 |
owa | +0.491 |
Neg logit contributions | |
digitally | -0.214 |
spaced | -0.208 |
MTV | -0.208 |
THR | -0.206 |
CST | -0.205 |
Token | Take |
Token ID | 12322 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
anyahu | +0.406 |
jriwal | +0.402 |
DonaldTrump | +0.393 |
uphem | +0.386 |
owa | +0.377 |
Neg logit contributions | |
digitally | -0.274 |
spaced | -0.269 |
MTV | -0.268 |
THR | -0.266 |
patches | -0.265 |
Token | GL |
Token ID | 10188 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
digitally | +0.858 |
spaced | +0.842 |
MTV | +0.834 |
THR | +0.828 |
patches | +0.827 |
Neg logit contributions | |
anyahu | -1.909 |
jriwal | -1.900 |
DonaldTrump | -1.866 |
uphem | -1.829 |
owa | -1.789 |
Token | a |
Token ID | 64 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
digitally | +0.618 |
spaced | +0.606 |
MTV | +0.605 |
THR | +0.602 |
patches | +0.601 |
Neg logit contributions | |
anyahu | -0.841 |
jriwal | -0.831 |
DonaldTrump | -0.814 |
uphem | -0.796 |
owa | -0.775 |
Token | DOS |
Token ID | 35178 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
digitally | +0.654 |
spaced | +0.639 |
MTV | +0.636 |
THR | +0.632 |
CST | +0.626 |
Neg logit contributions | |
anyahu | -1.498 |
jriwal | -1.486 |
DonaldTrump | -1.457 |
uphem | -1.432 |
owa | -1.399 |
Token | ' |
Token ID | 6 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
digitally | +2.212 |
THR | +2.140 |
spaced | +2.132 |
MTV | +2.131 |
patches | +2.108 |
Neg logit contributions | |
anyahu | -4.534 |
jriwal | -4.522 |
DonaldTrump | -4.443 |
uphem | -4.334 |
owa | -4.213 |
Token | escape |
Token ID | 6654 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
digitally | +3.152 |
spaced | +3.115 |
patches | +3.098 |
MTV | +3.078 |
dot | +3.065 |
Neg logit contributions | |
anyahu | -4.895 |
jriwal | -4.883 |
DonaldTrump | -4.776 |
uphem | -4.617 |
owa | -4.481 |
Token | advice |
Token ID | 5608 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
digitally | +1.042 |
spaced | +1.022 |
THR | +1.018 |
MTV | +1.017 |
patches | +1.016 |
Neg logit contributions | |
anyahu | -1.508 |
jriwal | -1.486 |
DonaldTrump | -1.467 |
uphem | -1.418 |
owa | -1.386 |
Token | ( |
Token ID | 357 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
digitally | +1.652 |
spaced | +1.599 |
THR | +1.596 |
MTV | +1.586 |
patches | +1.580 |
Neg logit contributions | |
anyahu | -2.945 |
jriwal | -2.919 |
DonaldTrump | -2.883 |
uphem | -2.785 |
owa | -2.723 |
Token | 28 |
Token ID | 2078 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
digitally | +0.717 |
MTV | +0.706 |
THR | +0.701 |
spaced | +0.700 |
CST | +0.694 |
Neg logit contributions | |
jriwal | -1.245 |
anyahu | -1.242 |
DonaldTrump | -1.203 |
uphem | -1.182 |
owa | -1.150 |
Token | ) |
Token ID | 8 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
digitally | +0.542 |
THR | +0.533 |
MTV | +0.531 |
CST | +0.531 |
spaced | +0.528 |
Neg logit contributions | |
jriwal | -0.923 |
anyahu | -0.919 |
DonaldTrump | -0.909 |
uphem | -0.878 |
owa | -0.853 |
Token | Rect |
Token ID | 45474 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
digitally | +1.078 |
spaced | +1.060 |
MTV | +1.053 |
THR | +1.048 |
patches | +1.044 |
Neg logit contributions | |
anyahu | -1.280 |
jriwal | -1.270 |
DonaldTrump | -1.240 |
uphem | -1.210 |
owa | -1.179 |
Token | angle |
Token ID | 9248 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
digitally | +0.209 |
spaced | +0.206 |
MTV | +0.205 |
THR | +0.203 |
patches | +0.203 |
Neg logit contributions | |
anyahu | -0.293 |
jriwal | -0.290 |
DonaldTrump | -0.284 |
uphem | -0.278 |
owa | -0.272 |
Token | ( |
Token ID | 7 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
digitally | +0.171 |
spaced | +0.169 |
MTV | +0.168 |
THR | +0.167 |
patches | +0.167 |
Neg logit contributions | |
anyahu | -0.199 |
jriwal | -0.197 |
DonaldTrump | -0.192 |
uphem | -0.188 |
owa | -0.183 |
Token | ( |
Token ID | 357 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
digitally | +1.828 |
spaced | +1.799 |
MTV | +1.777 |
CST | +1.772 |
THR | +1.772 |
Neg logit contributions | |
anyahu | -2.825 |
jriwal | -2.823 |
DonaldTrump | -2.726 |
uphem | -2.701 |
owa | -2.619 |
Token | New |
Token ID | 3791 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
digitally | +1.071 |
spaced | +1.057 |
MTV | +1.045 |
THR | +1.039 |
CST | +1.037 |
Neg logit contributions | |
anyahu | -1.539 |
jriwal | -1.528 |
DonaldTrump | -1.490 |
uphem | -1.464 |
owa | -1.426 |
Token | - |
Token ID | 12 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
digitally | +1.646 |
spaced | +1.617 |
MTV | +1.599 |
CST | +1.596 |
dot | +1.591 |
Neg logit contributions | |
jriwal | -2.589 |
anyahu | -2.587 |
DonaldTrump | -2.507 |
uphem | -2.468 |
owa | -2.390 |
Token | Object |
Token ID | 10267 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
digitally | +3.096 |
spaced | +3.050 |
dot | +3.033 |
MTV | +3.009 |
patches | +2.985 |
Neg logit contributions | |
jriwal | -4.963 |
anyahu | -4.900 |
DonaldTrump | -4.690 |
uphem | -4.658 |
owa | -4.516 |
Token | Drawing |
Token ID | 40027 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
digitally | +0.445 |
spaced | +0.437 |
MTV | +0.435 |
THR | +0.431 |
CST | +0.430 |
Neg logit contributions | |
anyahu | -0.752 |
jriwal | -0.747 |
DonaldTrump | -0.729 |
uphem | -0.716 |
owa | -0.700 |
Token | . |
Token ID | 13 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
anyahu | +0.609 |
jriwal | +0.603 |
DonaldTrump | +0.593 |
uphem | +0.580 |
owa | +0.566 |
Neg logit contributions | |
digitally | -0.445 |
spaced | -0.438 |
MTV | -0.436 |
THR | -0.433 |
patches | -0.431 |
Token | Solid |
Token ID | 46933 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
anyahu | +0.235 |
jriwal | +0.231 |
DonaldTrump | +0.228 |
uphem | +0.221 |
owa | +0.215 |
Neg logit contributions | |
digitally | -0.224 |
spaced | -0.221 |
MTV | -0.220 |
THR | -0.219 |
patches | -0.218 |
Token | Br |
Token ID | 9414 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
anyahu | +0.340 |
jriwal | +0.334 |
DonaldTrump | +0.329 |
uphem | +0.319 |
owa | +0.312 |
Neg logit contributions | |
digitally | -0.318 |
spaced | -0.314 |
MTV | -0.313 |
THR | -0.311 |
patches | -0.309 |
Token | s |
Token ID | 82 |
Feature activation | +5.25e-01 |
Pos logit contributions | |
digitally | +0.107 |
spaced | +0.106 |
MTV | +0.106 |
THR | +0.105 |
patches | +0.105 |
Neg logit contributions | |
anyahu | -0.157 |
jriwal | -0.157 |
DonaldTrump | -0.154 |
uphem | -0.150 |
owa | -0.145 |
Token | )} |
Token ID | 38165 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
anyahu | +0.385 |
jriwal | +0.383 |
DonaldTrump | +0.374 |
uphem | +0.364 |
owa | +0.353 |
Neg logit contributions | |
digitally | -0.382 |
spaced | -0.377 |
MTV | -0.377 |
THR | -0.377 |
CST | -0.373 |
Token | 9 |
Token ID | 860 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
anyahu | +0.688 |
jriwal | +0.681 |
DonaldTrump | +0.669 |
uphem | +0.656 |
owa | +0.642 |
Neg logit contributions | |
digitally | -0.416 |
spaced | -0.409 |
MTV | -0.406 |
THR | -0.403 |
patches | -0.402 |
Token | . |
Token ID | 13 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
anyahu | +1.745 |
jriwal | +1.725 |
DonaldTrump | +1.696 |
uphem | +1.664 |
owa | +1.630 |
Neg logit contributions | |
digitally | -0.980 |
spaced | -0.963 |
MTV | -0.956 |
THR | -0.946 |
patches | -0.945 |
Token | B |
Token ID | 347 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
digitally | +0.247 |
spaced | +0.242 |
MTV | +0.241 |
THR | +0.239 |
patches | +0.238 |
Neg logit contributions | |
anyahu | -0.447 |
jriwal | -0.443 |
DonaldTrump | -0.435 |
uphem | -0.426 |
owa | -0.416 |
Token | g |
Token ID | 70 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
digitally | +1.716 |
spaced | +1.711 |
THR | +1.691 |
MTV | +1.685 |
patches | +1.678 |
Neg logit contributions | |
anyahu | -1.979 |
jriwal | -1.975 |
DonaldTrump | -1.938 |
uphem | -1.874 |
owa | -1.804 |
Token | 5 |
Token ID | 20 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
digitally | +2.685 |
spaced | +2.662 |
patches | +2.630 |
THR | +2.614 |
dot | +2.595 |
Neg logit contributions | |
jriwal | -5.219 |
DonaldTrump | -5.086 |
anyahu | -5.080 |
uphem | -4.907 |
owa | -4.757 |
Token | { |
Token ID | 1391 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
digitally | +1.979 |
spaced | +1.965 |
MTV | +1.949 |
THR | +1.946 |
patches | +1.932 |
Neg logit contributions | |
jriwal | -2.100 |
anyahu | -2.096 |
DonaldTrump | -2.032 |
uphem | -1.983 |
owa | -1.908 |
Token | ( |
Token ID | 7 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
jriwal | +0.097 |
anyahu | +0.096 |
DonaldTrump | +0.094 |
uphem | +0.092 |
owa | +0.090 |
Neg logit contributions | |
digitally | -0.058 |
THR | -0.057 |
MTV | -0.057 |
spaced | -0.056 |
CST | -0.056 |
Token | 8 |
Token ID | 23 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
digitally | +1.704 |
MTV | +1.685 |
THR | +1.681 |
CST | +1.674 |
spaced | +1.673 |
Neg logit contributions | |
anyahu | -2.430 |
jriwal | -2.429 |
DonaldTrump | -2.347 |
uphem | -2.310 |
owa | -2.241 |
Token | 65 |
Token ID | 2996 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
digitally | +1.416 |
spaced | +1.403 |
MTV | +1.402 |
patches | +1.401 |
THR | +1.387 |
Neg logit contributions | |
anyahu | -2.175 |
jriwal | -2.171 |
DonaldTrump | -2.133 |
uphem | -2.084 |
owa | -2.014 |
Token | ague |
Token ID | 2064 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
anyahu | +1.084 |
jriwal | +1.062 |
DonaldTrump | +1.025 |
uphem | +1.013 |
owa | +0.979 |
Neg logit contributions | |
digitally | -1.446 |
spaced | -1.430 |
MTV | -1.429 |
patches | -1.413 |
THR | -1.410 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
anyahu | +0.067 |
jriwal | +0.067 |
DonaldTrump | +0.065 |
uphem | +0.064 |
owa | +0.063 |
Neg logit contributions | |
digitally | -0.038 |
spaced | -0.038 |
MTV | -0.038 |
THR | -0.037 |
patches | -0.037 |
Token | La |
Token ID | 4689 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
digitally | +1.538 |
MTV | +1.524 |
THR | +1.514 |
spaced | +1.504 |
CST | +1.490 |
Neg logit contributions | |
jriwal | -1.765 |
anyahu | -1.761 |
DonaldTrump | -1.695 |
uphem | -1.659 |
owa | -1.603 |
Token | Feb |
Token ID | 15146 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
anyahu | +0.356 |
jriwal | +0.352 |
DonaldTrump | +0.344 |
uphem | +0.337 |
owa | +0.329 |
Neg logit contributions | |
digitally | -0.262 |
spaced | -0.258 |
MTV | -0.257 |
THR | -0.255 |
CST | -0.254 |
Token | re |
Token ID | 260 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
digitally | +0.368 |
spaced | +0.363 |
MTV | +0.362 |
THR | +0.362 |
patches | +0.359 |
Neg logit contributions | |
anyahu | -0.238 |
jriwal | -0.236 |
DonaldTrump | -0.230 |
uphem | -0.219 |
owa | -0.211 |
Token | via |
Token ID | 2884 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
digitally | +0.771 |
spaced | +0.754 |
MTV | +0.751 |
THR | +0.742 |
CST | +0.739 |
Neg logit contributions | |
anyahu | -1.481 |
jriwal | -1.474 |
DonaldTrump | -1.442 |
uphem | -1.413 |
owa | -1.380 |
Token | text |
Token ID | 2420 |
Feature activation | -9.47e-02 |
Pos logit contributions | |
digitally | +1.799 |
MTV | +1.732 |
spaced | +1.715 |
THR | +1.707 |
Skype | +1.683 |
Neg logit contributions | |
anyahu | -6.317 |
jriwal | -6.266 |
DonaldTrump | -6.111 |
uphem | -5.970 |
owa | -5.878 |
Token | , |
Token ID | 11 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
digitally | +0.037 |
spaced | +0.036 |
MTV | +0.036 |
patches | +0.035 |
THR | +0.035 |
Neg logit contributions | |
anyahu | -0.101 |
jriwal | -0.100 |
DonaldTrump | -0.099 |
uphem | -0.097 |
owa | -0.096 |
Token | � |
Token ID | 564 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
digitally | +1.106 |
spaced | +1.076 |
MTV | +1.074 |
THR | +1.068 |
CST | +1.057 |
Neg logit contributions | |
anyahu | -1.881 |
jriwal | -1.867 |
DonaldTrump | -1.827 |
uphem | -1.789 |
owa | -1.755 |
Token | � |
Token ID | 250 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
digitally | +0.635 |
spaced | +0.619 |
MTV | +0.614 |
THR | +0.607 |
patches | +0.600 |
Neg logit contributions | |
anyahu | -1.880 |
jriwal | -1.870 |
DonaldTrump | -1.831 |
uphem | -1.805 |
owa | -1.775 |
Token | Can |
Token ID | 6090 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
digitally | +0.435 |
spaced | +0.429 |
MTV | +0.427 |
THR | +0.424 |
patches | +0.422 |
Neg logit contributions | |
anyahu | -0.565 |
jriwal | -0.560 |
DonaldTrump | -0.547 |
uphem | -0.536 |
owa | -0.523 |
Token | : |
Token ID | 25 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
anyahu | +1.501 |
jriwal | +1.484 |
DonaldTrump | +1.458 |
uphem | +1.430 |
owa | +1.400 |
Neg logit contributions | |
digitally | -0.883 |
spaced | -0.868 |
MTV | -0.862 |
THR | -0.854 |
patches | -0.852 |
Token |
|
Token ID | 198 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
anyahu | +1.995 |
jriwal | +1.973 |
DonaldTrump | +1.944 |
uphem | +1.912 |
owa | +1.878 |
Neg logit contributions | |
digitally | -0.891 |
spaced | -0.872 |
MTV | -0.863 |
patches | -0.853 |
THR | -0.852 |
Token |
|
Token ID | 198 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
anyahu | +3.358 |
jriwal | +3.300 |
DonaldTrump | +3.266 |
uphem | +3.219 |
owa | +3.181 |
Neg logit contributions | |
digitally | -1.686 |
spaced | -1.658 |
MTV | -1.620 |
patches | -1.619 |
CST | -1.595 |
Token | Show |
Token ID | 15307 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
anyahu | +2.249 |
jriwal | +2.217 |
DonaldTrump | +2.175 |
uphem | +2.124 |
owa | +2.079 |
Neg logit contributions | |
digitally | -1.913 |
spaced | -1.870 |
MTV | -1.863 |
THR | -1.850 |
CST | -1.838 |
Token | nested |
Token ID | 28376 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
anyahu | +0.420 |
jriwal | +0.414 |
DonaldTrump | +0.407 |
uphem | +0.398 |
owa | +0.389 |
Neg logit contributions | |
digitally | -0.304 |
spaced | -0.299 |
MTV | -0.298 |
THR | -0.295 |
patches | -0.294 |
Token | quote |
Token ID | 9577 |
Feature activation | +5.79e+00 |
Pos logit contributions | |
anyahu | +4.051 |
jriwal | +4.016 |
DonaldTrump | +3.895 |
uphem | +3.876 |
owa | +3.836 |
Neg logit contributions | |
digitally | -1.380 |
Warwick | -1.349 |
MTV | -1.348 |
THR | -1.318 |
Constantin | -1.309 |
Token | + |
Token ID | 1343 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
anyahu | +4.445 |
jriwal | +4.440 |
DonaldTrump | +4.306 |
owa | +4.237 |
uphem | +4.215 |
Neg logit contributions | |
digitally | -3.731 |
MTV | -3.699 |
spaced | -3.684 |
CST | -3.649 |
Warwick | -3.648 |
Token |
|
Token ID | 198 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
anyahu | +2.007 |
jriwal | +1.977 |
DonaldTrump | +1.960 |
uphem | +1.921 |
owa | +1.887 |
Neg logit contributions | |
digitally | -0.919 |
spaced | -0.903 |
MTV | -0.887 |
patches | -0.882 |
CST | -0.873 |
Token |
|
Token ID | 198 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
digitally | +0.061 |
MTV | +0.059 |
spaced | +0.059 |
THR | +0.059 |
CST | +0.058 |
Neg logit contributions | |
anyahu | -0.154 |
jriwal | -0.153 |
DonaldTrump | -0.150 |
uphem | -0.147 |
owa | -0.144 |
Token | On |
Token ID | 2202 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
digitally | +0.122 |
spaced | +0.120 |
MTV | +0.120 |
THR | +0.119 |
patches | +0.118 |
Neg logit contributions | |
anyahu | -0.159 |
jriwal | -0.157 |
DonaldTrump | -0.154 |
uphem | -0.151 |
owa | -0.147 |
Token | September |
Token ID | 2693 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
anyahu | +0.676 |
jriwal | +0.669 |
DonaldTrump | +0.657 |
uphem | +0.645 |
owa | +0.632 |
Neg logit contributions | |
digitally | -0.379 |
spaced | -0.373 |
MTV | -0.370 |
THR | -0.367 |
patches | -0.365 |
Token | square |
Token ID | 6616 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
digitally | +0.076 |
spaced | +0.075 |
dot | +0.073 |
patches | +0.073 |
THR | +0.072 |
Neg logit contributions | |
anyahu | -0.158 |
jriwal | -0.156 |
DonaldTrump | -0.154 |
uphem | -0.151 |
owa | -0.147 |
Token | struct |
Token ID | 2878 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
digitally | +0.947 |
spaced | +0.931 |
patches | +0.919 |
dot | +0.918 |
MTV | +0.917 |
Neg logit contributions | |
anyahu | -1.676 |
jriwal | -1.649 |
DonaldTrump | -1.631 |
uphem | -1.583 |
owa | -1.547 |
Token | uring |
Token ID | 870 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
anyahu | +2.254 |
jriwal | +2.220 |
DonaldTrump | +2.157 |
uphem | +2.122 |
owa | +2.061 |
Neg logit contributions | |
digitally | -2.393 |
spaced | -2.365 |
MTV | -2.357 |
CST | -2.339 |
THR | -2.333 |
Token | kernel |
Token ID | 9720 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
anyahu | +1.932 |
jriwal | +1.914 |
DonaldTrump | +1.866 |
uphem | +1.834 |
owa | +1.797 |
Neg logit contributions | |
digitally | -1.351 |
spaced | -1.332 |
MTV | -1.324 |
THR | -1.317 |
CST | -1.317 |
Token | rect |
Token ID | 13621 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
anyahu | +1.565 |
jriwal | +1.551 |
DonaldTrump | +1.512 |
uphem | +1.498 |
owa | +1.464 |
Neg logit contributions | |
digitally | -0.842 |
MTV | -0.826 |
spaced | -0.823 |
THR | -0.815 |
Warwick | -0.811 |
Token | K |
Token ID | 42 |
Feature activation | +4.76e+00 |
Pos logit contributions | |
anyahu | +1.696 |
jriwal | +1.674 |
DonaldTrump | +1.642 |
uphem | +1.604 |
owa | +1.566 |
Neg logit contributions | |
digitally | -1.394 |
spaced | -1.374 |
MTV | -1.365 |
THR | -1.357 |
patches | -1.353 |
Token | ernel |
Token ID | 7948 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
anyahu | +2.886 |
jriwal | +2.839 |
DonaldTrump | +2.746 |
uphem | +2.696 |
owa | +2.619 |
Neg logit contributions | |
digitally | -4.067 |
spaced | -4.023 |
MTV | -3.994 |
THR | -3.981 |
CST | -3.977 |
Token | = |
Token ID | 796 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
anyahu | +1.347 |
jriwal | +1.326 |
DonaldTrump | +1.295 |
uphem | +1.263 |
owa | +1.227 |
Neg logit contributions | |
digitally | -1.496 |
spaced | -1.478 |
MTV | -1.469 |
THR | -1.460 |
patches | -1.459 |
Token | c |
Token ID | 269 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
anyahu | +0.927 |
jriwal | +0.918 |
DonaldTrump | +0.901 |
uphem | +0.883 |
owa | +0.865 |
Neg logit contributions | |
digitally | -0.540 |
spaced | -0.530 |
MTV | -0.527 |
THR | -0.522 |
patches | -0.521 |
Token | v |
Token ID | 85 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
digitally | +0.418 |
spaced | +0.412 |
MTV | +0.411 |
patches | +0.408 |
THR | +0.407 |
Neg logit contributions | |
anyahu | -0.419 |
jriwal | -0.416 |
DonaldTrump | -0.407 |
uphem | -0.394 |
owa | -0.381 |
Token | 2 |
Token ID | 17 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
anyahu | +2.043 |
jriwal | +2.004 |
DonaldTrump | +1.984 |
uphem | +1.942 |
owa | +1.903 |
Neg logit contributions | |
digitally | -1.291 |
spaced | -1.273 |
MTV | -1.263 |
patches | -1.252 |
CST | -1.251 |
Token | the |
Token ID | 262 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
anyahu | +0.166 |
jriwal | +0.165 |
DonaldTrump | +0.162 |
uphem | +0.159 |
owa | +0.156 |
Neg logit contributions | |
digitally | -0.087 |
spaced | -0.086 |
MTV | -0.085 |
THR | -0.084 |
patches | -0.084 |
Token | answer |
Token ID | 3280 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
anyahu | +0.089 |
jriwal | +0.088 |
DonaldTrump | +0.087 |
uphem | +0.085 |
owa | +0.083 |
Neg logit contributions | |
digitally | -0.062 |
spaced | -0.061 |
MTV | -0.061 |
THR | -0.060 |
patches | -0.060 |
Token | is |
Token ID | 318 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
anyahu | +1.912 |
jriwal | +1.892 |
DonaldTrump | +1.842 |
uphem | +1.815 |
owa | +1.775 |
Neg logit contributions | |
digitally | -1.275 |
spaced | -1.260 |
MTV | -1.247 |
THR | -1.240 |
CST | -1.236 |
Token | affirm |
Token ID | 16266 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
anyahu | +0.237 |
jriwal | +0.235 |
DonaldTrump | +0.230 |
uphem | +0.226 |
owa | +0.221 |
Neg logit contributions | |
digitally | -0.153 |
spaced | -0.151 |
MTV | -0.150 |
THR | -0.149 |
patches | -0.148 |
Token | - |
Token ID | 12 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
anyahu | +2.035 |
jriwal | +2.000 |
DonaldTrump | +1.963 |
uphem | +1.920 |
owa | +1.889 |
Neg logit contributions | |
digitally | -1.546 |
spaced | -1.528 |
patches | -1.517 |
MTV | -1.515 |
THR | -1.511 |
Token | at |
Token ID | 379 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
anyahu | +1.835 |
jriwal | +1.809 |
DonaldTrump | +1.775 |
uphem | +1.736 |
owa | +1.686 |
Neg logit contributions | |
digitally | -1.576 |
spaced | -1.555 |
MTV | -1.543 |
THR | -1.538 |
patches | -1.536 |
Token | ive |
Token ID | 425 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
anyahu | +3.799 |
jriwal | +3.713 |
DonaldTrump | +3.594 |
uphem | +3.534 |
owa | +3.503 |
Neg logit contributions | |
digitally | -3.749 |
spaced | -3.671 |
THR | -3.643 |
MTV | -3.639 |
patches | -3.631 |
Token | , |
Token ID | 11 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
anyahu | +2.167 |
jriwal | +2.142 |
DonaldTrump | +2.107 |
uphem | +2.058 |
owa | +2.022 |
Neg logit contributions | |
digitally | -1.324 |
spaced | -1.310 |
MTV | -1.298 |
THR | -1.288 |
patches | -1.285 |
Token | do |
Token ID | 466 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
anyahu | +0.290 |
jriwal | +0.287 |
DonaldTrump | +0.282 |
uphem | +0.277 |
owa | +0.271 |
Neg logit contributions | |
digitally | -0.158 |
spaced | -0.155 |
MTV | -0.154 |
THR | -0.152 |
patches | -0.152 |
Token | you |
Token ID | 345 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
anyahu | +0.339 |
jriwal | +0.336 |
DonaldTrump | +0.331 |
uphem | +0.326 |
owa | +0.320 |
Neg logit contributions | |
digitally | -0.096 |
spaced | -0.093 |
MTV | -0.092 |
THR | -0.091 |
patches | -0.090 |
Token | want |
Token ID | 765 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
anyahu | +0.102 |
jriwal | +0.101 |
DonaldTrump | +0.099 |
uphem | +0.097 |
owa | +0.094 |
Neg logit contributions | |
digitally | -0.062 |
spaced | -0.060 |
MTV | -0.060 |
THR | -0.059 |
patches | -0.059 |
Token | The |
Token ID | 383 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
digitally | +0.149 |
spaced | +0.147 |
MTV | +0.146 |
THR | +0.145 |
patches | +0.144 |
Neg logit contributions | |
anyahu | -0.222 |
jriwal | -0.220 |
DonaldTrump | -0.216 |
uphem | -0.211 |
owa | -0.206 |
Token | terrain |
Token ID | 15510 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
digitally | +0.012 |
spaced | +0.012 |
MTV | +0.012 |
THR | +0.012 |
CST | +0.012 |
Neg logit contributions | |
anyahu | -0.022 |
jriwal | -0.022 |
DonaldTrump | -0.021 |
uphem | -0.021 |
owa | -0.020 |
Token | would |
Token ID | 561 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
digitally | +0.251 |
spaced | +0.247 |
MTV | +0.246 |
THR | +0.244 |
CST | +0.243 |
Neg logit contributions | |
anyahu | -0.386 |
jriwal | -0.381 |
DonaldTrump | -0.374 |
uphem | -0.366 |
owa | -0.358 |
Token | either |
Token ID | 2035 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
digitally | +0.392 |
spaced | +0.384 |
MTV | +0.380 |
THR | +0.377 |
patches | +0.377 |
Neg logit contributions | |
anyahu | -0.701 |
jriwal | -0.693 |
DonaldTrump | -0.682 |
uphem | -0.668 |
owa | -0.653 |
Token | a |
Token ID | 257 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
anyahu | +0.437 |
jriwal | +0.433 |
DonaldTrump | +0.424 |
uphem | +0.416 |
owa | +0.408 |
Neg logit contributions | |
digitally | -0.257 |
spaced | -0.253 |
MTV | -0.252 |
THR | -0.249 |
CST | -0.249 |
Token | ) |
Token ID | 8 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
digitally | +0.016 |
spaced | +0.016 |
MTV | +0.016 |
THR | +0.016 |
patches | +0.015 |
Neg logit contributions | |
anyahu | -0.023 |
jriwal | -0.023 |
DonaldTrump | -0.023 |
uphem | -0.022 |
owa | -0.022 |
Token | still |
Token ID | 991 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
anyahu | +4.725 |
jriwal | +4.678 |
DonaldTrump | +4.589 |
uphem | +4.500 |
owa | +4.471 |
Neg logit contributions | |
digitally | -2.747 |
CST | -2.720 |
MTV | -2.713 |
spaced | -2.700 |
Warwick | -2.683 |
Token | be |
Token ID | 307 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
anyahu | +0.832 |
jriwal | +0.824 |
DonaldTrump | +0.809 |
uphem | +0.792 |
owa | +0.778 |
Neg logit contributions | |
digitally | -0.492 |
spaced | -0.484 |
MTV | -0.483 |
CST | -0.477 |
THR | -0.477 |
Token | too |
Token ID | 1165 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
digitally | +0.290 |
spaced | +0.285 |
MTV | +0.283 |
THR | +0.281 |
patches | +0.280 |
Neg logit contributions | |
anyahu | -0.516 |
jriwal | -0.511 |
DonaldTrump | -0.502 |
uphem | -0.492 |
owa | -0.482 |
Token | random |
Token ID | 4738 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
digitally | +0.011 |
MTV | +0.011 |
THR | +0.011 |
CST | +0.011 |
spaced | +0.011 |
Neg logit contributions | |
anyahu | -0.019 |
jriwal | -0.019 |
DonaldTrump | -0.019 |
uphem | -0.018 |
owa | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +9.56e-02 |
Pos logit contributions | |
digitally | +1.291 |
spaced | +1.266 |
patches | +1.246 |
MTV | +1.237 |
THR | +1.223 |
Neg logit contributions | |
anyahu | -2.409 |
jriwal | -2.373 |
DonaldTrump | -2.348 |
uphem | -2.300 |
owa | -2.235 |
Token |
|
Token ID | 198 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
digitally | +0.394 |
spaced | +0.385 |
MTV | +0.383 |
THR | +0.379 |
patches | +0.377 |
Neg logit contributions | |
anyahu | -0.938 |
jriwal | -0.929 |
DonaldTrump | -0.914 |
uphem | -0.898 |
owa | -0.881 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
digitally | +0.058 |
MTV | +0.056 |
spaced | +0.056 |
THR | +0.056 |
CST | +0.055 |
Neg logit contributions | |
anyahu | -0.146 |
jriwal | -0.145 |
DonaldTrump | -0.142 |
uphem | -0.139 |
owa | -0.136 |
Token | 2012 |
Token ID | 6999 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
anyahu | +1.405 |
DonaldTrump | +1.374 |
jriwal | +1.368 |
uphem | +1.337 |
owa | +1.315 |
Neg logit contributions | |
digitally | -0.723 |
spaced | -0.713 |
patches | -0.701 |
THR | -0.696 |
CST | -0.693 |
Token | Louis |
Token ID | 5593 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
digitally | +0.285 |
spaced | +0.278 |
MTV | +0.277 |
THR | +0.274 |
patches | +0.272 |
Neg logit contributions | |
anyahu | -0.700 |
jriwal | -0.694 |
DonaldTrump | -0.683 |
uphem | -0.671 |
owa | -0.658 |
Token | C |
Token ID | 327 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
anyahu | +0.121 |
jriwal | +0.120 |
DonaldTrump | +0.118 |
uphem | +0.116 |
owa | +0.113 |
Neg logit contributions | |
digitally | -0.047 |
spaced | -0.046 |
MTV | -0.046 |
THR | -0.045 |
patches | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +5.40e+00 |
Pos logit contributions | |
anyahu | +0.228 |
jriwal | +0.226 |
DonaldTrump | +0.221 |
uphem | +0.217 |
owa | +0.212 |
Neg logit contributions | |
digitally | -0.117 |
MTV | -0.114 |
spaced | -0.114 |
THR | -0.114 |
CST | -0.113 |
Token | K |
Token ID | 42 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
anyahu | +3.962 |
DonaldTrump | +3.854 |
jriwal | +3.801 |
uphem | +3.713 |
owa | +3.654 |
Neg logit contributions | |
digitally | -3.821 |
spaced | -3.810 |
patches | -3.745 |
MTV | -3.699 |
THR | -3.687 |
Token | . |
Token ID | 13 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
anyahu | +2.461 |
jriwal | +2.428 |
DonaldTrump | +2.393 |
uphem | +2.350 |
owa | +2.305 |
Neg logit contributions | |
digitally | -1.469 |
spaced | -1.451 |
MTV | -1.427 |
patches | -1.420 |
THR | -1.408 |
Token |
|
Token ID | 198 |
Feature activation | +6.99e-02 |
Pos logit contributions | |
anyahu | +0.294 |
jriwal | +0.292 |
DonaldTrump | +0.287 |
uphem | +0.282 |
owa | +0.277 |
Neg logit contributions | |
digitally | -0.132 |
spaced | -0.129 |
THR | -0.126 |
patches | -0.126 |
MTV | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
anyahu | +0.073 |
jriwal | +0.073 |
DonaldTrump | +0.071 |
uphem | +0.070 |
owa | +0.068 |
Neg logit contributions | |
digitally | -0.029 |
MTV | -0.029 |
spaced | -0.029 |
THR | -0.028 |
CST | -0.028 |
Token | 2012 |
Token ID | 6999 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
anyahu | +1.873 |
DonaldTrump | +1.829 |
jriwal | +1.810 |
uphem | +1.779 |
owa | +1.749 |
Neg logit contributions | |
digitally | -0.990 |
spaced | -0.979 |
patches | -0.964 |
THR | -0.949 |
CST | -0.945 |
Token | that |
Token ID | 326 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
digitally | +0.018 |
spaced | +0.018 |
MTV | +0.018 |
THR | +0.018 |
patches | +0.018 |
Neg logit contributions | |
anyahu | -0.038 |
jriwal | -0.037 |
DonaldTrump | -0.037 |
uphem | -0.036 |
owa | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
anyahu | +1.290 |
jriwal | +1.279 |
DonaldTrump | +1.254 |
uphem | +1.232 |
owa | +1.213 |
Neg logit contributions | |
digitally | -0.719 |
MTV | -0.708 |
spaced | -0.706 |
THR | -0.694 |
CST | -0.694 |
Token | air |
Token ID | 1633 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
anyahu | +0.034 |
jriwal | +0.033 |
DonaldTrump | +0.033 |
uphem | +0.032 |
owa | +0.031 |
Neg logit contributions | |
digitally | -0.020 |
spaced | -0.019 |
MTV | -0.019 |
THR | -0.019 |
patches | -0.019 |
Token | in |
Token ID | 287 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
anyahu | +0.283 |
jriwal | +0.280 |
DonaldTrump | +0.275 |
uphem | +0.269 |
owa | +0.264 |
Neg logit contributions | |
digitally | -0.180 |
spaced | -0.177 |
MTV | -0.177 |
THR | -0.175 |
CST | -0.174 |
Token | clusions |
Token ID | 11539 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
anyahu | +0.924 |
jriwal | +0.916 |
DonaldTrump | +0.899 |
uphem | +0.882 |
owa | +0.864 |
Neg logit contributions | |
digitally | -0.509 |
spaced | -0.499 |
MTV | -0.497 |
THR | -0.492 |
CST | -0.490 |
Token | recovered |
Token ID | 11911 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
anyahu | +1.792 |
jriwal | +1.777 |
DonaldTrump | +1.736 |
uphem | +1.707 |
owa | +1.677 |
Neg logit contributions | |
digitally | -1.097 |
MTV | -1.082 |
spaced | -1.074 |
CST | -1.067 |
THR | -1.065 |
Token | in |
Token ID | 287 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
anyahu | +1.046 |
jriwal | +1.036 |
DonaldTrump | +1.017 |
uphem | +0.999 |
owa | +0.981 |
Neg logit contributions | |
digitally | -0.521 |
spaced | -0.514 |
MTV | -0.512 |
THR | -0.504 |
CST | -0.503 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
anyahu | +1.537 |
jriwal | +1.528 |
DonaldTrump | +1.493 |
uphem | +1.471 |
owa | +1.445 |
Neg logit contributions | |
digitally | -0.865 |
MTV | -0.849 |
spaced | -0.848 |
CST | -0.834 |
THR | -0.834 |
Token | ice |
Token ID | 4771 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
anyahu | +0.942 |
jriwal | +0.934 |
DonaldTrump | +0.915 |
uphem | +0.897 |
owa | +0.883 |
Neg logit contributions | |
digitally | -0.538 |
spaced | -0.528 |
MTV | -0.526 |
THR | -0.519 |
CST | -0.517 |
Token | cores |
Token ID | 21758 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
anyahu | +0.251 |
jriwal | +0.248 |
DonaldTrump | +0.243 |
uphem | +0.239 |
owa | +0.235 |
Neg logit contributions | |
digitally | -0.146 |
spaced | -0.144 |
MTV | -0.143 |
THR | -0.142 |
CST | -0.142 |
Token | have |
Token ID | 423 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
anyahu | +0.317 |
jriwal | +0.314 |
DonaldTrump | +0.307 |
uphem | +0.301 |
owa | +0.295 |
Neg logit contributions | |
digitally | -0.202 |
spaced | -0.198 |
MTV | -0.198 |
THR | -0.196 |
CST | -0.195 |
Token | educ |
Token ID | 2711 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
digitally | +0.056 |
spaced | +0.055 |
MTV | +0.055 |
THR | +0.054 |
patches | +0.054 |
Neg logit contributions | |
anyahu | -0.100 |
jriwal | -0.099 |
DonaldTrump | -0.097 |
uphem | -0.095 |
owa | -0.093 |
Token | ations |
Token ID | 602 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
digitally | +1.015 |
spaced | +1.000 |
MTV | +0.997 |
THR | +0.996 |
patches | +0.988 |
Neg logit contributions | |
anyahu | -1.270 |
jriwal | -1.259 |
DonaldTrump | -1.237 |
uphem | -1.203 |
owa | -1.167 |
Token | or |
Token ID | 393 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
digitally | +0.133 |
spaced | +0.131 |
MTV | +0.130 |
THR | +0.129 |
patches | +0.129 |
Neg logit contributions | |
anyahu | -0.211 |
jriwal | -0.209 |
DonaldTrump | -0.206 |
uphem | -0.201 |
owa | -0.196 |
Token | are |
Token ID | 389 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
digitally | +0.100 |
spaced | +0.098 |
MTV | +0.098 |
THR | +0.097 |
patches | +0.097 |
Neg logit contributions | |
anyahu | -0.155 |
jriwal | -0.154 |
DonaldTrump | -0.151 |
uphem | -0.148 |
owa | -0.144 |
Token | promoted |
Token ID | 13722 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
anyahu | +0.850 |
jriwal | +0.843 |
DonaldTrump | +0.827 |
uphem | +0.812 |
owa | +0.797 |
Neg logit contributions | |
digitally | -0.504 |
MTV | -0.498 |
spaced | -0.497 |
THR | -0.493 |
CST | -0.492 |
Token | retire |
Token ID | 8058 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
digitally | +0.230 |
spaced | +0.225 |
MTV | +0.223 |
THR | +0.222 |
patches | +0.222 |
Neg logit contributions | |
anyahu | -0.408 |
jriwal | -0.404 |
DonaldTrump | -0.398 |
uphem | -0.390 |
owa | -0.380 |
Token | later |
Token ID | 1568 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
anyahu | +1.332 |
jriwal | +1.322 |
DonaldTrump | +1.294 |
uphem | +1.265 |
owa | +1.248 |
Neg logit contributions | |
digitally | -0.765 |
MTV | -0.757 |
spaced | -0.753 |
CST | -0.747 |
THR | -0.747 |
Token | . |
Token ID | 13 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
digitally | +0.152 |
spaced | +0.149 |
MTV | +0.148 |
THR | +0.147 |
patches | +0.146 |
Neg logit contributions | |
anyahu | -0.270 |
jriwal | -0.267 |
DonaldTrump | -0.263 |
uphem | -0.258 |
owa | -0.252 |
Token | However |
Token ID | 2102 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
anyahu | +0.375 |
jriwal | +0.372 |
DonaldTrump | +0.364 |
uphem | +0.357 |
owa | +0.350 |
Neg logit contributions | |
digitally | -0.205 |
spaced | -0.201 |
MTV | -0.201 |
THR | -0.199 |
CST | -0.198 |
Token | , |
Token ID | 11 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
digitally | +0.272 |
spaced | +0.268 |
MTV | +0.265 |
patches | +0.263 |
THR | +0.263 |
Neg logit contributions | |
anyahu | -0.470 |
jriwal | -0.466 |
DonaldTrump | -0.458 |
uphem | -0.448 |
owa | -0.438 |
Token | if |
Token ID | 611 |
Feature activation | +7.01e-01 |
Pos logit contributions | |
digitally | +0.036 |
spaced | +0.035 |
MTV | +0.035 |
THR | +0.035 |
patches | +0.035 |
Neg logit contributions | |
anyahu | -0.070 |
jriwal | -0.069 |
DonaldTrump | -0.068 |
uphem | -0.067 |
owa | -0.065 |
Token | . |
Token ID | 764 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
digitally | +0.418 |
THR | +0.406 |
MTV | +0.406 |
spaced | +0.404 |
dot | +0.402 |
Neg logit contributions | |
anyahu | -1.144 |
jriwal | -1.144 |
DonaldTrump | -1.118 |
uphem | -1.094 |
owa | -1.073 |
Token | cl |
Token ID | 537 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
anyahu | +0.854 |
jriwal | +0.847 |
DonaldTrump | +0.829 |
uphem | +0.818 |
owa | +0.802 |
Neg logit contributions | |
digitally | -0.406 |
MTV | -0.398 |
spaced | -0.397 |
THR | -0.392 |
CST | -0.390 |
Token | ojure |
Token ID | 32511 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
anyahu | +0.250 |
jriwal | +0.247 |
DonaldTrump | +0.243 |
uphem | +0.234 |
owa | +0.226 |
Neg logit contributions | |
digitally | -0.253 |
spaced | -0.249 |
MTV | -0.248 |
THR | -0.247 |
patches | -0.247 |
Token | / |
Token ID | 1220 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
anyahu | +0.062 |
jriwal | +0.062 |
DonaldTrump | +0.061 |
uphem | +0.059 |
owa | +0.058 |
Neg logit contributions | |
digitally | -0.039 |
MTV | -0.038 |
spaced | -0.038 |
THR | -0.038 |
CST | -0.038 |
Token | java |
Token ID | 20129 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
digitally | +0.089 |
spaced | +0.087 |
MTV | +0.086 |
THR | +0.086 |
CST | +0.085 |
Neg logit contributions | |
anyahu | -0.174 |
jriwal | -0.172 |
DonaldTrump | -0.169 |
uphem | -0.166 |
owa | -0.163 |
Token | . |
Token ID | 764 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
anyahu | +1.272 |
jriwal | +1.262 |
DonaldTrump | +1.242 |
uphem | +1.217 |
owa | +1.196 |
Neg logit contributions | |
digitally | -0.683 |
spaced | -0.669 |
MTV | -0.667 |
THR | -0.658 |
CST | -0.654 |
Token | j |
Token ID | 474 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
anyahu | +1.787 |
jriwal | +1.777 |
DonaldTrump | +1.734 |
uphem | +1.704 |
owa | +1.681 |
Neg logit contributions | |
digitally | -1.140 |
spaced | -1.120 |
MTV | -1.117 |
THR | -1.108 |
CST | -1.105 |
Token | db |
Token ID | 9945 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
anyahu | +0.089 |
jriwal | +0.088 |
DonaldTrump | +0.086 |
uphem | +0.085 |
owa | +0.083 |
Neg logit contributions | |
digitally | -0.067 |
spaced | -0.066 |
MTV | -0.066 |
THR | -0.065 |
CST | -0.065 |
Token | c |
Token ID | 66 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
anyahu | +2.054 |
jriwal | +2.022 |
DonaldTrump | +1.983 |
uphem | +1.936 |
owa | +1.912 |
Neg logit contributions | |
digitally | -1.680 |
spaced | -1.655 |
MTV | -1.647 |
THR | -1.636 |
CST | -1.631 |
Token | | |
Token ID | 930 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
anyahu | +0.267 |
jriwal | +0.265 |
DonaldTrump | +0.260 |
uphem | +0.255 |
owa | +0.249 |
Neg logit contributions | |
digitally | -0.159 |
spaced | -0.156 |
MTV | -0.155 |
THR | -0.154 |
CST | -0.153 |
Token | 10 |
Token ID | 838 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
anyahu | +0.345 |
jriwal | +0.340 |
DonaldTrump | +0.336 |
uphem | +0.329 |
owa | +0.323 |
Neg logit contributions | |
digitally | -0.189 |
spaced | -0.186 |
MTV | -0.184 |
patches | -0.182 |
THR | -0.182 |
Token | Academy |
Token ID | 8581 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
anyahu | +0.574 |
jriwal | +0.566 |
DonaldTrump | +0.555 |
uphem | +0.534 |
owa | +0.520 |
Neg logit contributions | |
digitally | -0.794 |
spaced | -0.787 |
MTV | -0.780 |
patches | -0.779 |
THR | -0.776 |
Token | Awards |
Token ID | 15434 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
anyahu | +1.597 |
jriwal | +1.581 |
DonaldTrump | +1.561 |
uphem | +1.520 |
owa | +1.496 |
Neg logit contributions | |
digitally | -0.978 |
spaced | -0.972 |
MTV | -0.961 |
patches | -0.954 |
THR | -0.948 |
Token | . |
Token ID | 13 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
digitally | +0.248 |
spaced | +0.243 |
MTV | +0.242 |
THR | +0.240 |
patches | +0.239 |
Neg logit contributions | |
anyahu | -0.440 |
jriwal | -0.436 |
DonaldTrump | -0.427 |
uphem | -0.420 |
owa | -0.410 |
Token | It |
Token ID | 632 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
anyahu | +0.074 |
jriwal | +0.073 |
DonaldTrump | +0.071 |
uphem | +0.070 |
owa | +0.068 |
Neg logit contributions | |
digitally | -0.061 |
spaced | -0.060 |
MTV | -0.060 |
THR | -0.060 |
patches | -0.059 |
Token | � |
Token ID | 447 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
anyahu | +0.479 |
jriwal | +0.474 |
DonaldTrump | +0.465 |
uphem | +0.455 |
owa | +0.445 |
Neg logit contributions | |
digitally | -0.339 |
spaced | -0.333 |
MTV | -0.332 |
THR | -0.329 |
patches | -0.328 |
Token | � |
Token ID | 247 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
anyahu | +0.817 |
jriwal | +0.809 |
DonaldTrump | +0.798 |
uphem | +0.785 |
owa | +0.772 |
Neg logit contributions | |
digitally | -0.257 |
spaced | -0.250 |
MTV | -0.248 |
THR | -0.244 |
CST | -0.244 |
Token | s |
Token ID | 82 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
anyahu | +0.906 |
jriwal | +0.895 |
DonaldTrump | +0.878 |
uphem | +0.858 |
owa | +0.836 |
Neg logit contributions | |
digitally | -0.680 |
spaced | -0.670 |
MTV | -0.666 |
THR | -0.661 |
patches | -0.661 |
Token | good |
Token ID | 922 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
anyahu | +1.309 |
jriwal | +1.300 |
DonaldTrump | +1.277 |
uphem | +1.252 |
owa | +1.232 |
Neg logit contributions | |
digitally | -0.672 |
spaced | -0.660 |
MTV | -0.652 |
patches | -0.649 |
THR | -0.649 |
Token | for |
Token ID | 329 |
Feature activation | +8.25e-01 |
Pos logit contributions | |
digitally | +0.204 |
spaced | +0.200 |
MTV | +0.199 |
THR | +0.197 |
patches | +0.197 |
Neg logit contributions | |
anyahu | -0.374 |
jriwal | -0.370 |
DonaldTrump | -0.364 |
uphem | -0.357 |
owa | -0.349 |
Token | our |
Token ID | 674 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
anyahu | +0.776 |
jriwal | +0.769 |
DonaldTrump | +0.761 |
uphem | +0.742 |
owa | +0.732 |
Neg logit contributions | |
digitally | -0.417 |
spaced | -0.410 |
patches | -0.405 |
CST | -0.400 |
MTV | -0.400 |
Token | brand |
Token ID | 4508 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
anyahu | +0.109 |
jriwal | +0.108 |
DonaldTrump | +0.106 |
uphem | +0.103 |
owa | +0.102 |
Neg logit contributions | |
digitally | -0.058 |
spaced | -0.057 |
MTV | -0.056 |
patches | -0.056 |
THR | -0.056 |
Token | two |
Token ID | 734 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
digitally | +0.172 |
spaced | +0.170 |
MTV | +0.169 |
THR | +0.167 |
patches | +0.167 |
Neg logit contributions | |
anyahu | -0.232 |
jriwal | -0.229 |
DonaldTrump | -0.225 |
uphem | -0.220 |
owa | -0.214 |
Token | bodies |
Token ID | 5920 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
digitally | +0.331 |
spaced | +0.323 |
MTV | +0.320 |
patches | +0.318 |
THR | +0.318 |
Neg logit contributions | |
anyahu | -0.653 |
jriwal | -0.646 |
DonaldTrump | -0.635 |
uphem | -0.623 |
owa | -0.609 |
Token | are |
Token ID | 389 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
anyahu | +0.709 |
jriwal | +0.702 |
DonaldTrump | +0.687 |
uphem | +0.673 |
owa | +0.658 |
Neg logit contributions | |
digitally | -0.504 |
spaced | -0.496 |
MTV | -0.496 |
THR | -0.491 |
CST | -0.490 |
Token | willing |
Token ID | 4684 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
anyahu | +0.305 |
jriwal | +0.302 |
DonaldTrump | +0.297 |
uphem | +0.291 |
owa | +0.285 |
Neg logit contributions | |
digitally | -0.175 |
MTV | -0.172 |
spaced | -0.172 |
THR | -0.171 |
CST | -0.170 |
Token | to |
Token ID | 284 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
digitally | +0.153 |
spaced | +0.150 |
MTV | +0.149 |
THR | +0.148 |
CST | +0.147 |
Neg logit contributions | |
anyahu | -0.294 |
jriwal | -0.291 |
DonaldTrump | -0.286 |
uphem | -0.281 |
owa | -0.275 |
Token | compromise |
Token ID | 13110 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
anyahu | +0.592 |
jriwal | +0.588 |
DonaldTrump | +0.575 |
uphem | +0.566 |
owa | +0.555 |
Neg logit contributions | |
digitally | -0.334 |
spaced | -0.329 |
MTV | -0.328 |
CST | -0.324 |
THR | -0.324 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
anyahu | +2.399 |
jriwal | +2.370 |
DonaldTrump | +2.325 |
uphem | +2.283 |
owa | +2.240 |
Neg logit contributions | |
spaced | -1.376 |
MTV | -1.376 |
digitally | -1.373 |
THR | -1.358 |
CST | -1.348 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
digitally | +0.008 |
spaced | +0.008 |
MTV | +0.008 |
THR | +0.007 |
CST | +0.007 |
Neg logit contributions | |
anyahu | -0.015 |
jriwal | -0.015 |
DonaldTrump | -0.015 |
uphem | -0.014 |
owa | -0.014 |
Token | Alt |
Token ID | 29161 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
digitally | +0.366 |
MTV | +0.362 |
spaced | +0.360 |
THR | +0.358 |
patches | +0.355 |
Neg logit contributions | |
anyahu | -0.478 |
jriwal | -0.474 |
DonaldTrump | -0.459 |
uphem | -0.451 |
owa | -0.440 |
Token | ria |
Token ID | 7496 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
anyahu | +0.206 |
jriwal | +0.203 |
DonaldTrump | +0.199 |
uphem | +0.194 |
owa | +0.190 |
Neg logit contributions | |
digitally | -0.173 |
spaced | -0.171 |
MTV | -0.170 |
THR | -0.169 |
CST | -0.168 |
Token | shares |
Token ID | 7303 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
anyahu | +0.543 |
jriwal | +0.537 |
DonaldTrump | +0.528 |
uphem | +0.517 |
owa | +0.506 |
Neg logit contributions | |
digitally | -0.342 |
spaced | -0.336 |
MTV | -0.334 |
THR | -0.332 |
patches | -0.331 |
Token |
|
Token ID | 198 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
digitally | +0.168 |
spaced | +0.165 |
MTV | +0.164 |
THR | +0.163 |
CST | +0.162 |
Neg logit contributions | |
anyahu | -0.337 |
jriwal | -0.335 |
DonaldTrump | -0.328 |
uphem | -0.322 |
owa | -0.315 |
Token |
|
Token ID | 198 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
anyahu | +0.032 |
jriwal | +0.032 |
DonaldTrump | +0.032 |
uphem | +0.031 |
owa | +0.030 |
Neg logit contributions | |
digitally | -0.013 |
MTV | -0.013 |
spaced | -0.012 |
THR | -0.012 |
CST | -0.012 |
Token | The |
Token ID | 464 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
anyahu | +0.097 |
jriwal | +0.096 |
DonaldTrump | +0.093 |
uphem | +0.092 |
owa | +0.089 |
Neg logit contributions | |
digitally | -0.073 |
spaced | -0.072 |
MTV | -0.071 |
THR | -0.071 |
patches | -0.070 |
Token | dilemma |
Token ID | 26566 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
digitally | +0.377 |
spaced | +0.371 |
MTV | +0.369 |
THR | +0.367 |
CST | +0.364 |
Neg logit contributions | |
anyahu | -0.577 |
jriwal | -0.572 |
DonaldTrump | -0.562 |
uphem | -0.550 |
owa | -0.537 |
Token | is |
Token ID | 318 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
anyahu | +0.414 |
jriwal | +0.409 |
DonaldTrump | +0.402 |
uphem | +0.394 |
owa | +0.385 |
Neg logit contributions | |
digitally | -0.260 |
spaced | -0.256 |
MTV | -0.254 |
THR | -0.252 |
patches | -0.251 |
Token | that |
Token ID | 326 |
Feature activation | +7.30e-01 |
Pos logit contributions | |
anyahu | +0.077 |
jriwal | +0.076 |
DonaldTrump | +0.075 |
uphem | +0.073 |
owa | +0.072 |
Neg logit contributions | |
digitally | -0.040 |
spaced | -0.039 |
MTV | -0.039 |
THR | -0.039 |
patches | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
anyahu | +0.720 |
jriwal | +0.713 |
DonaldTrump | +0.701 |
uphem | +0.688 |
owa | +0.675 |
Neg logit contributions | |
digitally | -0.345 |
spaced | -0.338 |
MTV | -0.336 |
THR | -0.332 |
patches | -0.331 |
Token | Garg |
Token ID | 39013 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
anyahu | +0.045 |
jriwal | +0.045 |
DonaldTrump | +0.044 |
uphem | +0.043 |
owa | +0.042 |
Neg logit contributions | |
digitally | -0.029 |
spaced | -0.029 |
MTV | -0.028 |
THR | -0.028 |
CST | -0.028 |
Token | ant |
Token ID | 415 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
anyahu | +0.734 |
jriwal | +0.728 |
DonaldTrump | +0.717 |
uphem | +0.694 |
owa | +0.671 |
Neg logit contributions | |
digitally | -0.632 |
spaced | -0.622 |
MTV | -0.621 |
patches | -0.616 |
THR | -0.615 |
Token | ua |
Token ID | 6413 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
anyahu | +1.233 |
jriwal | +1.214 |
DonaldTrump | +1.179 |
uphem | +1.138 |
owa | +1.095 |
Neg logit contributions | |
digitally | -1.970 |
spaced | -1.949 |
MTV | -1.943 |
THR | -1.933 |
patches | -1.929 |
Token | � |
Token ID | 447 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
digitally | +0.573 |
spaced | +0.563 |
MTV | +0.559 |
THR | +0.557 |
patches | +0.554 |
Neg logit contributions | |
anyahu | -0.867 |
jriwal | -0.859 |
DonaldTrump | -0.844 |
uphem | -0.828 |
owa | -0.804 |
Token | inventive |
Token ID | 47602 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
digitally | +0.301 |
spaced | +0.295 |
MTV | +0.293 |
THR | +0.290 |
patches | +0.289 |
Neg logit contributions | |
anyahu | -0.690 |
jriwal | -0.684 |
DonaldTrump | -0.673 |
uphem | -0.662 |
owa | -0.649 |
Token | coll |
Token ID | 2927 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
digitally | +0.116 |
spaced | +0.113 |
MTV | +0.113 |
THR | +0.112 |
CST | +0.111 |
Neg logit contributions | |
anyahu | -0.252 |
jriwal | -0.249 |
DonaldTrump | -0.246 |
uphem | -0.241 |
owa | -0.237 |
Token | ages |
Token ID | 1095 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
anyahu | +1.305 |
jriwal | +1.289 |
DonaldTrump | +1.256 |
uphem | +1.215 |
owa | +1.179 |
Neg logit contributions | |
digitally | -1.514 |
spaced | -1.494 |
MTV | -1.488 |
patches | -1.480 |
THR | -1.476 |
Token | now |
Token ID | 783 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
anyahu | +0.774 |
jriwal | +0.767 |
DonaldTrump | +0.755 |
uphem | +0.740 |
owa | +0.726 |
Neg logit contributions | |
digitally | -0.447 |
MTV | -0.441 |
spaced | -0.441 |
THR | -0.438 |
CST | -0.437 |
Token | making |
Token ID | 1642 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
anyahu | +0.011 |
jriwal | +0.011 |
DonaldTrump | +0.011 |
uphem | +0.011 |
owa | +0.011 |
Neg logit contributions | |
digitally | -0.007 |
spaced | -0.007 |
MTV | -0.007 |
THR | -0.006 |
patches | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
anyahu | +0.438 |
jriwal | +0.435 |
DonaldTrump | +0.428 |
uphem | +0.420 |
owa | +0.412 |
Neg logit contributions | |
digitally | -0.246 |
spaced | -0.243 |
MTV | -0.240 |
THR | -0.240 |
CST | -0.239 |
Token | rounds |
Token ID | 9196 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
anyahu | +0.554 |
jriwal | +0.550 |
DonaldTrump | +0.540 |
uphem | +0.532 |
owa | +0.520 |
Neg logit contributions | |
digitally | -0.317 |
spaced | -0.314 |
THR | -0.311 |
MTV | -0.310 |
CST | -0.310 |
Token | of |
Token ID | 286 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
digitally | +0.340 |
spaced | +0.333 |
MTV | +0.331 |
THR | +0.327 |
patches | +0.326 |
Neg logit contributions | |
anyahu | -0.780 |
jriwal | -0.773 |
DonaldTrump | -0.760 |
uphem | -0.747 |
owa | -0.732 |
Token | the |
Token ID | 262 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
digitally | +0.276 |
spaced | +0.272 |
MTV | +0.269 |
THR | +0.267 |
patches | +0.267 |
Neg logit contributions | |
anyahu | -0.559 |
jriwal | -0.554 |
DonaldTrump | -0.546 |
uphem | -0.536 |
owa | -0.525 |
Token | internet |
Token ID | 5230 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
digitally | +0.229 |
spaced | +0.225 |
MTV | +0.222 |
THR | +0.221 |
patches | +0.220 |
Neg logit contributions | |
anyahu | -0.613 |
jriwal | -0.608 |
DonaldTrump | -0.599 |
uphem | -0.590 |
owa | -0.578 |
Token | . |
Token ID | 13 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
digitally | +0.395 |
spaced | +0.389 |
MTV | +0.387 |
THR | +0.383 |
CST | +0.382 |
Neg logit contributions | |
anyahu | -0.707 |
jriwal | -0.700 |
DonaldTrump | -0.688 |
uphem | -0.674 |
owa | -0.660 |
Token | of |
Token ID | 286 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
anyahu | +0.862 |
jriwal | +0.853 |
DonaldTrump | +0.835 |
uphem | +0.818 |
owa | +0.801 |
Neg logit contributions | |
digitally | -0.624 |
spaced | -0.614 |
MTV | -0.611 |
THR | -0.606 |
patches | -0.605 |
Token | Shiva |
Token ID | 33995 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
anyahu | +2.757 |
jriwal | +2.756 |
DonaldTrump | +2.679 |
uphem | +2.635 |
owa | +2.602 |
Neg logit contributions | |
digitally | -1.611 |
spaced | -1.574 |
MTV | -1.560 |
THR | -1.546 |
patches | -1.545 |
Token | and |
Token ID | 290 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
anyahu | +1.831 |
jriwal | +1.819 |
DonaldTrump | +1.777 |
uphem | +1.746 |
owa | +1.718 |
Neg logit contributions | |
digitally | -1.123 |
spaced | -1.100 |
MTV | -1.096 |
THR | -1.084 |
patches | -1.081 |
Token | normally |
Token ID | 7685 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
anyahu | +1.051 |
jriwal | +1.044 |
DonaldTrump | +1.020 |
uphem | +1.002 |
owa | +0.984 |
Neg logit contributions | |
digitally | -0.636 |
spaced | -0.624 |
MTV | -0.621 |
THR | -0.616 |
patches | -0.612 |
Token | none |
Token ID | 4844 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
anyahu | +1.526 |
jriwal | +1.515 |
DonaldTrump | +1.478 |
uphem | +1.453 |
owa | +1.431 |
Neg logit contributions | |
digitally | -0.959 |
MTV | -0.937 |
spaced | -0.934 |
THR | -0.927 |
patches | -0.921 |
Token | of |
Token ID | 286 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
anyahu | +0.795 |
jriwal | +0.788 |
DonaldTrump | +0.773 |
uphem | +0.757 |
owa | +0.741 |
Neg logit contributions | |
digitally | -0.491 |
spaced | -0.482 |
MTV | -0.480 |
THR | -0.476 |
patches | -0.474 |
Token | the |
Token ID | 262 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
anyahu | +0.119 |
jriwal | +0.118 |
DonaldTrump | +0.116 |
uphem | +0.113 |
owa | +0.111 |
Neg logit contributions | |
digitally | -0.060 |
spaced | -0.058 |
MTV | -0.058 |
THR | -0.057 |
patches | -0.057 |
Token | Vish |
Token ID | 36900 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
anyahu | +0.432 |
jriwal | +0.429 |
DonaldTrump | +0.420 |
uphem | +0.412 |
owa | +0.405 |
Neg logit contributions | |
digitally | -0.229 |
spaced | -0.225 |
MTV | -0.224 |
THR | -0.222 |
CST | -0.220 |
Token | nu |
Token ID | 28803 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
anyahu | +0.801 |
jriwal | +0.789 |
DonaldTrump | +0.766 |
uphem | +0.746 |
owa | +0.724 |
Neg logit contributions | |
digitally | -1.144 |
spaced | -1.130 |
MTV | -1.125 |
THR | -1.120 |
CST | -1.118 |
Token | temple |
Token ID | 12505 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
anyahu | +0.935 |
jriwal | +0.928 |
DonaldTrump | +0.908 |
uphem | +0.891 |
owa | +0.875 |
Neg logit contributions | |
digitally | -0.570 |
spaced | -0.559 |
MTV | -0.558 |
THR | -0.552 |
CST | -0.549 |
Token | have |
Token ID | 423 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
anyahu | +0.302 |
jriwal | +0.299 |
DonaldTrump | +0.293 |
uphem | +0.287 |
owa | +0.281 |
Neg logit contributions | |
digitally | -0.186 |
spaced | -0.183 |
MTV | -0.182 |
THR | -0.181 |
CST | -0.180 |
Token | priced |
Token ID | 30883 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
anyahu | +0.115 |
jriwal | +0.114 |
DonaldTrump | +0.112 |
uphem | +0.109 |
owa | +0.106 |
Neg logit contributions | |
digitally | -0.106 |
spaced | -0.105 |
MTV | -0.105 |
THR | -0.104 |
patches | -0.104 |
Token | tickets |
Token ID | 8587 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
anyahu | +1.047 |
jriwal | +1.034 |
DonaldTrump | +1.012 |
uphem | +0.988 |
owa | +0.965 |
Neg logit contributions | |
digitally | -0.902 |
spaced | -0.890 |
MTV | -0.886 |
THR | -0.880 |
CST | -0.877 |
Token | in |
Token ID | 287 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
anyahu | +1.767 |
jriwal | +1.751 |
DonaldTrump | +1.718 |
uphem | +1.684 |
owa | +1.658 |
Neg logit contributions | |
digitally | -1.002 |
MTV | -0.998 |
spaced | -0.993 |
CST | -0.983 |
THR | -0.979 |
Token | the |
Token ID | 262 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
anyahu | +1.262 |
jriwal | +1.254 |
DonaldTrump | +1.231 |
uphem | +1.208 |
owa | +1.186 |
Neg logit contributions | |
digitally | -0.679 |
spaced | -0.665 |
MTV | -0.664 |
THR | -0.656 |
CST | -0.655 |
Token | room |
Token ID | 2119 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
anyahu | +0.581 |
jriwal | +0.575 |
DonaldTrump | +0.564 |
uphem | +0.553 |
owa | +0.541 |
Neg logit contributions | |
digitally | -0.351 |
spaced | -0.345 |
MTV | -0.343 |
THR | -0.340 |
patches | -0.339 |
Token | , |
Token ID | 11 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
anyahu | +0.212 |
jriwal | +0.210 |
DonaldTrump | +0.207 |
uphem | +0.203 |
owa | +0.198 |
Neg logit contributions | |
digitally | -0.121 |
spaced | -0.119 |
MTV | -0.119 |
THR | -0.117 |
patches | -0.117 |
Token | � |
Token ID | 447 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
anyahu | +0.168 |
jriwal | +0.166 |
DonaldTrump | +0.163 |
uphem | +0.160 |
owa | +0.156 |
Neg logit contributions | |
digitally | -0.102 |
spaced | -0.100 |
MTV | -0.099 |
THR | -0.099 |
patches | -0.098 |
Token | � |
Token ID | 251 |
Feature activation | +5.63e-03 |
Pos logit contributions | |
anyahu | +0.032 |
jriwal | +0.032 |
DonaldTrump | +0.031 |
uphem | +0.031 |
owa | +0.030 |
Neg logit contributions | |
digitally | -0.021 |
spaced | -0.020 |
MTV | -0.020 |
THR | -0.020 |
CST | -0.020 |
Token | H |
Token ID | 367 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
anyahu | +0.005 |
jriwal | +0.005 |
DonaldTrump | +0.005 |
uphem | +0.005 |
owa | +0.005 |
Neg logit contributions | |
digitally | -0.003 |
spaced | -0.003 |
MTV | -0.003 |
THR | -0.003 |
patches | -0.003 |
Token | ower |
Token ID | 789 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
anyahu | +0.482 |
jriwal | +0.467 |
DonaldTrump | +0.449 |
uphem | +0.414 |
owa | +0.387 |
Neg logit contributions | |
digitally | -1.552 |
spaced | -1.535 |
MTV | -1.530 |
THR | -1.528 |
patches | -1.525 |
Token | ton |
Token ID | 1122 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
digitally | +0.250 |
MTV | +0.248 |
spaced | +0.247 |
THR | +0.246 |
patches | +0.245 |
Neg logit contributions | |
anyahu | -0.186 |
jriwal | -0.186 |
DonaldTrump | -0.181 |
uphem | -0.176 |
owa | -0.168 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
anyahu | +0.857 |
jriwal | +0.849 |
DonaldTrump | +0.835 |
uphem | +0.819 |
owa | +0.803 |
Neg logit contributions | |
digitally | -0.449 |
spaced | -0.441 |
MTV | -0.437 |
THR | -0.433 |
patches | -0.433 |
Token |
|
Token ID | 198 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
digitally | +0.066 |
spaced | +0.065 |
MTV | +0.064 |
THR | +0.064 |
patches | +0.063 |
Neg logit contributions | |
anyahu | -0.139 |
jriwal | -0.138 |
DonaldTrump | -0.135 |
uphem | -0.133 |
owa | -0.130 |
Token |
|
Token ID | 198 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
anyahu | +0.368 |
jriwal | +0.366 |
DonaldTrump | +0.359 |
uphem | +0.352 |
owa | +0.345 |
Neg logit contributions | |
digitally | -0.148 |
MTV | -0.144 |
spaced | -0.144 |
THR | -0.143 |
CST | -0.141 |
Token | So |
Token ID | 2396 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
anyahu | +0.235 |
jriwal | +0.232 |
DonaldTrump | +0.226 |
uphem | +0.222 |
owa | +0.216 |
Neg logit contributions | |
digitally | -0.191 |
spaced | -0.188 |
MTV | -0.188 |
THR | -0.186 |
patches | -0.185 |
Token | , |
Token ID | 11 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
anyahu | +0.375 |
jriwal | +0.372 |
DonaldTrump | +0.364 |
uphem | +0.358 |
owa | +0.350 |
Neg logit contributions | |
digitally | -0.215 |
spaced | -0.211 |
MTV | -0.210 |
THR | -0.208 |
CST | -0.207 |
Token | Sanders |
Token ID | 5831 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
digitally | +0.539 |
MTV | +0.528 |
spaced | +0.527 |
THR | +0.521 |
CST | +0.517 |
Neg logit contributions | |
anyahu | -0.969 |
jriwal | -0.960 |
DonaldTrump | -0.941 |
uphem | -0.924 |
owa | -0.903 |
Token | knows |
Token ID | 4206 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
anyahu | +0.127 |
jriwal | +0.125 |
DonaldTrump | +0.123 |
uphem | +0.120 |
owa | +0.118 |
Neg logit contributions | |
digitally | -0.084 |
spaced | -0.083 |
MTV | -0.082 |
THR | -0.081 |
patches | -0.081 |
Token | what |
Token ID | 644 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
digitally | +0.055 |
spaced | +0.054 |
MTV | +0.054 |
THR | +0.053 |
patches | +0.053 |
Neg logit contributions | |
anyahu | -0.110 |
jriwal | -0.109 |
DonaldTrump | -0.107 |
uphem | -0.105 |
owa | -0.103 |
Token | year |
Token ID | 614 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
anyahu | +0.117 |
jriwal | +0.115 |
DonaldTrump | +0.114 |
uphem | +0.111 |
owa | +0.109 |
Neg logit contributions | |
digitally | -0.060 |
spaced | -0.059 |
MTV | -0.058 |
THR | -0.058 |
patches | -0.058 |
Token | it |
Token ID | 340 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
digitally | +0.356 |
MTV | +0.349 |
spaced | +0.348 |
THR | +0.343 |
patches | +0.342 |
Neg logit contributions | |
anyahu | -0.711 |
jriwal | -0.705 |
DonaldTrump | -0.691 |
uphem | -0.678 |
owa | -0.663 |
Token | is |
Token ID | 318 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
anyahu | +1.510 |
jriwal | +1.492 |
DonaldTrump | +1.465 |
uphem | +1.432 |
owa | +1.402 |
Neg logit contributions | |
digitally | -1.125 |
spaced | -1.110 |
MTV | -1.105 |
THR | -1.096 |
patches | -1.093 |
Token | , |
Token ID | 11 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
THR | +1.583 |
digitally | +1.583 |
MTV | +1.568 |
spaced | +1.566 |
CST | +1.557 |
Neg logit contributions | |
anyahu | -3.061 |
jriwal | -3.053 |
DonaldTrump | -2.968 |
uphem | -2.929 |
owa | -2.840 |
Token | AND |
Token ID | 5357 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
digitally | +0.343 |
spaced | +0.337 |
MTV | +0.336 |
THR | +0.333 |
CST | +0.331 |
Neg logit contributions | |
anyahu | -0.682 |
jriwal | -0.677 |
DonaldTrump | -0.664 |
uphem | -0.652 |
owa | -0.638 |
Token | AN |
Token ID | 3537 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
digitally | +0.629 |
MTV | +0.617 |
spaced | +0.617 |
THR | +0.615 |
CST | +0.609 |
Neg logit contributions | |
anyahu | -1.261 |
jriwal | -1.254 |
DonaldTrump | -1.225 |
uphem | -1.204 |
owa | -1.178 |
Token | P |
Token ID | 47 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
digitally | +0.944 |
MTV | +0.929 |
spaced | +0.928 |
THR | +0.920 |
CST | +0.913 |
Neg logit contributions | |
anyahu | -1.903 |
jriwal | -1.890 |
DonaldTrump | -1.853 |
uphem | -1.819 |
owa | -1.780 |
Token | FAC |
Token ID | 44216 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
anyahu | +1.709 |
jriwal | +1.689 |
DonaldTrump | +1.657 |
uphem | +1.622 |
owa | +1.588 |
Neg logit contributions | |
digitally | -1.202 |
spaced | -1.184 |
MTV | -1.175 |
patches | -1.164 |
CST | -1.161 |
Token | IL |
Token ID | 4146 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
digitally | +0.396 |
spaced | +0.389 |
MTV | +0.386 |
patches | +0.383 |
THR | +0.382 |
Neg logit contributions | |
anyahu | -0.553 |
jriwal | -0.546 |
DonaldTrump | -0.537 |
uphem | -0.525 |
owa | -0.512 |
Token | ITIES |
Token ID | 30383 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
digitally | +1.676 |
MTV | +1.673 |
THR | +1.666 |
spaced | +1.648 |
CST | +1.640 |
Neg logit contributions | |
anyahu | -3.136 |
jriwal | -3.103 |
uphem | -2.982 |
DonaldTrump | -2.974 |
owa | -2.905 |
Token | ARE |
Token ID | 15986 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
anyahu | +0.181 |
jriwal | +0.179 |
DonaldTrump | +0.176 |
uphem | +0.173 |
owa | +0.170 |
Neg logit contributions | |
digitally | -0.093 |
spaced | -0.091 |
MTV | -0.091 |
patches | -0.090 |
THR | -0.089 |
Token | TO |
Token ID | 5390 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
digitally | +0.587 |
spaced | +0.576 |
MTV | +0.572 |
THR | +0.566 |
patches | +0.564 |
Neg logit contributions | |
anyahu | -1.224 |
jriwal | -1.213 |
DonaldTrump | -1.192 |
uphem | -1.170 |
owa | -1.147 |
Token | BE |
Token ID | 9348 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
digitally | +0.591 |
spaced | +0.582 |
MTV | +0.579 |
THR | +0.575 |
patches | +0.573 |
Neg logit contributions | |
anyahu | -0.815 |
jriwal | -0.807 |
DonaldTrump | -0.791 |
uphem | -0.774 |
owa | -0.756 |
Token | SE |
Token ID | 7946 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
digitally | +0.493 |
spaced | +0.483 |
MTV | +0.479 |
patches | +0.475 |
THR | +0.473 |
Neg logit contributions | |
anyahu | -0.962 |
jriwal | -0.951 |
DonaldTrump | -0.937 |
uphem | -0.919 |
owa | -0.900 |
Token | was |
Token ID | 373 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
digitally | +0.747 |
spaced | +0.732 |
MTV | +0.728 |
patches | +0.722 |
THR | +0.721 |
Neg logit contributions | |
anyahu | -1.090 |
jriwal | -1.081 |
DonaldTrump | -1.060 |
uphem | -1.037 |
owa | -1.010 |
Token | designed |
Token ID | 3562 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
anyahu | +1.010 |
jriwal | +0.996 |
DonaldTrump | +0.979 |
uphem | +0.963 |
owa | +0.943 |
Neg logit contributions | |
digitally | -0.660 |
MTV | -0.653 |
spaced | -0.653 |
THR | -0.651 |
patches | -0.648 |
Token | by |
Token ID | 416 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
anyahu | +0.574 |
jriwal | +0.571 |
DonaldTrump | +0.559 |
uphem | +0.550 |
owa | +0.540 |
Neg logit contributions | |
digitally | -0.314 |
spaced | -0.314 |
MTV | -0.314 |
THR | -0.310 |
CST | -0.309 |
Token | up |
Token ID | 510 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
digitally | +0.142 |
spaced | +0.139 |
MTV | +0.139 |
THR | +0.137 |
patches | +0.137 |
Neg logit contributions | |
anyahu | -0.286 |
jriwal | -0.283 |
DonaldTrump | -0.278 |
uphem | -0.273 |
owa | -0.268 |
Token | - |
Token ID | 12 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
anyahu | +0.793 |
jriwal | +0.784 |
DonaldTrump | +0.774 |
uphem | +0.754 |
owa | +0.744 |
Neg logit contributions | |
digitally | -0.463 |
spaced | -0.458 |
MTV | -0.455 |
THR | -0.454 |
patches | -0.449 |
Token | and |
Token ID | 392 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
digitally | +0.783 |
spaced | +0.767 |
MTV | +0.762 |
patches | +0.753 |
THR | +0.751 |
Neg logit contributions | |
anyahu | -1.325 |
jriwal | -1.315 |
DonaldTrump | -1.288 |
uphem | -1.266 |
owa | -1.235 |
Token | - |
Token ID | 12 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
digitally | +2.358 |
spaced | +2.309 |
dot | +2.240 |
MTV | +2.198 |
THR | +2.198 |
Neg logit contributions | |
anyahu | -4.201 |
jriwal | -4.181 |
DonaldTrump | -4.113 |
uphem | -4.014 |
owa | -3.856 |
Token | coming |
Token ID | 4976 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
digitally | +0.722 |
spaced | +0.713 |
MTV | +0.708 |
patches | +0.703 |
CST | +0.702 |
Neg logit contributions | |
anyahu | -0.511 |
jriwal | -0.505 |
DonaldTrump | -0.489 |
uphem | -0.474 |
owa | -0.456 |
Token | fashion |
Token ID | 6977 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
anyahu | +1.102 |
jriwal | +1.087 |
DonaldTrump | +1.069 |
uphem | +1.050 |
owa | +1.035 |
Neg logit contributions | |
digitally | -0.673 |
spaced | -0.665 |
MTV | -0.663 |
THR | -0.662 |
patches | -0.655 |
Token | designer |
Token ID | 11915 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
digitally | +0.408 |
spaced | +0.401 |
MTV | +0.399 |
THR | +0.396 |
patches | +0.394 |
Neg logit contributions | |
anyahu | -0.661 |
jriwal | -0.654 |
DonaldTrump | -0.642 |
uphem | -0.630 |
owa | -0.616 |
Token | Amanda |
Token ID | 23040 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
anyahu | +1.421 |
jriwal | +1.402 |
DonaldTrump | +1.383 |
uphem | +1.360 |
owa | +1.338 |
Neg logit contributions | |
digitally | -0.730 |
spaced | -0.720 |
MTV | -0.707 |
THR | -0.704 |
patches | -0.703 |
Token | Ariel |
Token ID | 33364 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
anyahu | +0.852 |
jriwal | +0.841 |
DonaldTrump | +0.826 |
uphem | +0.809 |
owa | +0.792 |
Neg logit contributions | |
digitally | -0.538 |
spaced | -0.529 |
MTV | -0.526 |
THR | -0.522 |
patches | -0.520 |
Token | is |
Token ID | 318 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
anyahu | +0.751 |
jriwal | +0.745 |
DonaldTrump | +0.733 |
uphem | +0.716 |
owa | +0.697 |
Neg logit contributions | |
digitally | -0.522 |
spaced | -0.513 |
MTV | -0.512 |
THR | -0.507 |
patches | -0.506 |
Token | following |
Token ID | 1708 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
digitally | +0.150 |
spaced | +0.147 |
MTV | +0.146 |
THR | +0.145 |
patches | +0.144 |
Neg logit contributions | |
anyahu | -0.313 |
jriwal | -0.311 |
DonaldTrump | -0.305 |
uphem | -0.300 |
owa | -0.294 |
Token | in |
Token ID | 287 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
digitally | +0.346 |
spaced | +0.339 |
MTV | +0.337 |
THR | +0.335 |
patches | +0.333 |
Neg logit contributions | |
anyahu | -0.667 |
jriwal | -0.662 |
DonaldTrump | -0.650 |
uphem | -0.638 |
owa | -0.625 |
Token | the |
Token ID | 262 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
digitally | +0.521 |
spaced | +0.515 |
MTV | +0.514 |
THR | +0.510 |
patches | +0.508 |
Neg logit contributions | |
anyahu | -0.705 |
jriwal | -0.700 |
DonaldTrump | -0.688 |
uphem | -0.671 |
owa | -0.657 |
Token | footsteps |
Token ID | 27146 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
digitally | +0.097 |
spaced | +0.094 |
MTV | +0.094 |
THR | +0.093 |
patches | +0.092 |
Neg logit contributions | |
anyahu | -0.244 |
jriwal | -0.242 |
DonaldTrump | -0.238 |
uphem | -0.234 |
owa | -0.230 |
Token | of |
Token ID | 286 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
digitally | +2.533 |
spaced | +2.497 |
MTV | +2.483 |
THR | +2.468 |
patches | +2.441 |
Neg logit contributions | |
anyahu | -4.044 |
jriwal | -3.996 |
DonaldTrump | -3.974 |
uphem | -3.867 |
owa | -3.785 |
Token | Ariel |
Token ID | 33364 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
digitally | +0.312 |
spaced | +0.306 |
MTV | +0.305 |
THR | +0.302 |
patches | +0.301 |
Neg logit contributions | |
anyahu | -0.522 |
jriwal | -0.517 |
DonaldTrump | -0.507 |
uphem | -0.497 |
owa | -0.487 |
Token | Sharon |
Token ID | 26918 |
Feature activation | -3.41e-03 |
Pos logit contributions | |
anyahu | +1.136 |
jriwal | +1.122 |
DonaldTrump | +1.102 |
uphem | +1.082 |
owa | +1.062 |
Neg logit contributions | |
digitally | -0.662 |
spaced | -0.652 |
MTV | -0.645 |
THR | -0.641 |
CST | -0.640 |
Token | because |
Token ID | 780 |
Feature activation | -9.10e-01 |
Pos logit contributions | |
digitally | +0.002 |
spaced | +0.002 |
MTV | +0.002 |
THR | +0.002 |
patches | +0.002 |
Neg logit contributions | |
anyahu | -0.003 |
jriwal | -0.003 |
DonaldTrump | -0.003 |
uphem | -0.003 |
owa | -0.003 |
Token | Israeli |
Token ID | 6085 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
digitally | +0.472 |
spaced | +0.464 |
MTV | +0.461 |
THR | +0.457 |
patches | +0.456 |
Neg logit contributions | |
anyahu | -0.856 |
jriwal | -0.849 |
DonaldTrump | -0.835 |
uphem | -0.818 |
owa | -0.800 |
Token | combat |
Token ID | 5249 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
digitally | +1.115 |
spaced | +1.079 |
MTV | +1.077 |
THR | +1.071 |
patches | +1.064 |
Neg logit contributions | |
anyahu | -1.970 |
jriwal | -1.950 |
DonaldTrump | -1.915 |
uphem | -1.868 |
owa | -1.836 |
Token | this |
Token ID | 428 |
Feature activation | +6.54e-01 |
Pos logit contributions | |
anyahu | +0.549 |
jriwal | +0.543 |
DonaldTrump | +0.534 |
uphem | +0.525 |
owa | +0.512 |
Neg logit contributions | |
digitally | -0.267 |
spaced | -0.264 |
MTV | -0.263 |
THR | -0.260 |
CST | -0.259 |
Token | wave |
Token ID | 6769 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
anyahu | +0.580 |
jriwal | +0.574 |
DonaldTrump | +0.563 |
uphem | +0.552 |
owa | +0.540 |
Neg logit contributions | |
digitally | -0.378 |
spaced | -0.370 |
MTV | -0.368 |
THR | -0.365 |
patches | -0.364 |
Token | of |
Token ID | 286 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
anyahu | +0.150 |
jriwal | +0.149 |
DonaldTrump | +0.146 |
uphem | +0.143 |
owa | +0.140 |
Neg logit contributions | |
digitally | -0.105 |
spaced | -0.102 |
MTV | -0.102 |
THR | -0.101 |
patches | -0.101 |
Token | misinformation |
Token ID | 32805 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
anyahu | +0.637 |
jriwal | +0.632 |
DonaldTrump | +0.620 |
uphem | +0.608 |
owa | +0.594 |
Neg logit contributions | |
digitally | -0.388 |
spaced | -0.383 |
MTV | -0.382 |
THR | -0.378 |
CST | -0.377 |
Token | is |
Token ID | 318 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
digitally | +0.029 |
spaced | +0.028 |
MTV | +0.028 |
THR | +0.028 |
patches | +0.028 |
Neg logit contributions | |
anyahu | -0.044 |
jriwal | -0.043 |
DonaldTrump | -0.042 |
uphem | -0.041 |
owa | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
digitally | +1.571 |
spaced | +1.535 |
MTV | +1.519 |
THR | +1.509 |
patches | +1.506 |
Neg logit contributions | |
anyahu | -2.718 |
jriwal | -2.685 |
DonaldTrump | -2.647 |
uphem | -2.580 |
owa | -2.519 |
Token | talk |
Token ID | 1561 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
digitally | +1.466 |
spaced | +1.428 |
MTV | +1.415 |
THR | +1.407 |
patches | +1.397 |
Neg logit contributions | |
anyahu | -2.431 |
jriwal | -2.400 |
DonaldTrump | -2.361 |
uphem | -2.301 |
owa | -2.243 |
Token | with |
Token ID | 351 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
anyahu | +0.843 |
jriwal | +0.832 |
DonaldTrump | +0.817 |
uphem | +0.801 |
owa | +0.781 |
Neg logit contributions | |
digitally | -0.579 |
spaced | -0.575 |
MTV | -0.569 |
CST | -0.565 |
THR | -0.564 |
Token | the |
Token ID | 262 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
anyahu | +0.546 |
jriwal | +0.541 |
DonaldTrump | +0.533 |
uphem | +0.523 |
owa | +0.512 |
Neg logit contributions | |
digitally | -0.274 |
spaced | -0.269 |
MTV | -0.267 |
THR | -0.265 |
CST | -0.264 |
Token | non |
Token ID | 1729 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
anyahu | +0.124 |
jriwal | +0.123 |
DonaldTrump | +0.121 |
uphem | +0.119 |
owa | +0.116 |
Neg logit contributions | |
digitally | -0.066 |
spaced | -0.064 |
MTV | -0.064 |
THR | -0.063 |
patches | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
anyahu | +0.283 |
jriwal | +0.280 |
DonaldTrump | +0.275 |
uphem | +0.270 |
owa | +0.265 |
Neg logit contributions | |
digitally | -0.163 |
spaced | -0.160 |
MTV | -0.160 |
THR | -0.158 |
CST | -0.158 |
Token | But |
Token ID | 887 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
digitally | +0.535 |
MTV | +0.526 |
spaced | +0.525 |
THR | +0.520 |
CST | +0.517 |
Neg logit contributions | |
anyahu | -0.900 |
jriwal | -0.893 |
DonaldTrump | -0.872 |
uphem | -0.857 |
owa | -0.837 |
Token | we |
Token ID | 356 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
anyahu | +0.177 |
jriwal | +0.175 |
DonaldTrump | +0.173 |
uphem | +0.169 |
owa | +0.166 |
Neg logit contributions | |
digitally | -0.088 |
spaced | -0.086 |
MTV | -0.085 |
THR | -0.085 |
CST | -0.085 |
Token | can |
Token ID | 460 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
digitally | +0.518 |
spaced | +0.510 |
MTV | +0.505 |
THR | +0.502 |
patches | +0.498 |
Neg logit contributions | |
anyahu | -0.699 |
jriwal | -0.695 |
DonaldTrump | -0.681 |
uphem | -0.663 |
owa | -0.646 |
Token | 't |
Token ID | 470 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
anyahu | +0.399 |
jriwal | +0.396 |
DonaldTrump | +0.389 |
uphem | +0.382 |
owa | +0.375 |
Neg logit contributions | |
digitally | -0.180 |
spaced | -0.177 |
MTV | -0.176 |
THR | -0.174 |
patches | -0.173 |
Token | entirely |
Token ID | 5000 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
digitally | +0.355 |
spaced | +0.348 |
MTV | +0.346 |
THR | +0.343 |
patches | +0.341 |
Neg logit contributions | |
anyahu | -0.630 |
jriwal | -0.624 |
DonaldTrump | -0.613 |
uphem | -0.600 |
owa | -0.587 |
Token | shy |
Token ID | 15800 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
digitally | +1.717 |
spaced | +1.679 |
MTV | +1.663 |
THR | +1.657 |
patches | +1.647 |
Neg logit contributions | |
anyahu | -2.827 |
jriwal | -2.792 |
DonaldTrump | -2.762 |
uphem | -2.661 |
owa | -2.596 |
Token | away |
Token ID | 1497 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
anyahu | +1.036 |
jriwal | +1.021 |
DonaldTrump | +0.998 |
uphem | +0.988 |
owa | +0.969 |
Neg logit contributions | |
digitally | -0.484 |
spaced | -0.481 |
MTV | -0.473 |
THR | -0.470 |
CST | -0.469 |
Token | from |
Token ID | 422 |
Feature activation | -8.22e-01 |
Pos logit contributions | |
digitally | +0.797 |
MTV | +0.771 |
spaced | +0.770 |
patches | +0.759 |
THR | +0.756 |
Neg logit contributions | |
anyahu | -2.025 |
jriwal | -2.008 |
DonaldTrump | -1.971 |
uphem | -1.940 |
owa | -1.901 |
Token | pointing |
Token ID | 10609 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
digitally | +0.409 |
spaced | +0.403 |
MTV | +0.400 |
THR | +0.397 |
CST | +0.396 |
Neg logit contributions | |
anyahu | -0.787 |
jriwal | -0.779 |
DonaldTrump | -0.767 |
uphem | -0.752 |
owa | -0.737 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
anyahu | +1.531 |
jriwal | +1.514 |
DonaldTrump | +1.495 |
uphem | +1.473 |
owa | +1.446 |
Neg logit contributions | |
digitally | -0.587 |
spaced | -0.583 |
CST | -0.574 |
MTV | -0.571 |
THR | -0.569 |