Token | email |
Token ID | 3053 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
uncle | +0.189 |
oker | +0.185 |
ecast | +0.177 |
enegger | +0.174 |
Liberties | +0.174 |
Neg logit contributions | |
Malk | -0.210 |
Andreas | -0.162 |
Manson | -0.155 |
Goldberg | -0.152 |
twists | -0.152 |
Token | to |
Token ID | 284 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
Malk | +1.186 |
Andreas | +0.926 |
Manson | +0.890 |
Goldberg | +0.877 |
twists | +0.876 |
Neg logit contributions | |
uncle | -0.945 |
oker | -0.919 |
ecast | -0.881 |
enegger | -0.869 |
Liberties | -0.866 |
Token | get |
Token ID | 651 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
Malk | +0.329 |
Andreas | +0.251 |
Manson | +0.240 |
Goldberg | +0.236 |
twists | +0.236 |
Neg logit contributions | |
uncle | -0.307 |
oker | -0.301 |
ecast | -0.288 |
Liberties | -0.284 |
enegger | -0.284 |
Token | out |
Token ID | 503 |
Feature activation | +6.63e-02 |
Pos logit contributions | |
Malk | +1.067 |
Andreas | +0.819 |
Manson | +0.785 |
Goldberg | +0.774 |
twists | +0.770 |
Neg logit contributions | |
uncle | -0.948 |
oker | -0.926 |
ecast | -0.883 |
enegger | -0.876 |
Liberties | -0.872 |
Token | . |
Token ID | 13 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
Malk | +0.172 |
Andreas | +0.134 |
Manson | +0.129 |
Goldberg | +0.127 |
twists | +0.127 |
Neg logit contributions | |
uncle | -0.136 |
oker | -0.132 |
ecast | -0.127 |
enegger | -0.124 |
Liberties | -0.124 |
Token | Either |
Token ID | 15467 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
uncle | +0.078 |
oker | +0.076 |
ecast | +0.072 |
enegger | +0.071 |
Liberties | +0.071 |
Neg logit contributions | |
Malk | -0.107 |
Andreas | -0.084 |
Manson | -0.081 |
Goldberg | -0.080 |
twists | -0.080 |
Token | way |
Token ID | 835 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
Malk | +3.656 |
Andreas | +2.570 |
Manson | +2.378 |
Goldberg | +2.351 |
twists | +2.322 |
Neg logit contributions | |
uncle | -5.179 |
oker | -5.020 |
enegger | -4.936 |
ecast | -4.919 |
iscopal | -4.867 |
Token | , |
Token ID | 11 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
uncle | +0.563 |
oker | +0.553 |
ecast | +0.525 |
Liberties | +0.519 |
bnb | +0.517 |
Neg logit contributions | |
Malk | -0.749 |
Andreas | -0.588 |
Manson | -0.566 |
Goldberg | -0.560 |
twists | -0.559 |
Token | Kam |
Token ID | 12670 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
uncle | +0.034 |
oker | +0.033 |
ecast | +0.032 |
enegger | +0.031 |
Liberties | +0.031 |
Neg logit contributions | |
Malk | -0.039 |
Andreas | -0.030 |
Manson | -0.028 |
Goldberg | -0.028 |
twists | -0.028 |
Token | al |
Token ID | 282 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
Malk | +0.125 |
Andreas | +0.095 |
Manson | +0.090 |
Goldberg | +0.089 |
twists | +0.088 |
Neg logit contributions | |
uncle | -0.120 |
oker | -0.118 |
enegger | -0.114 |
ecast | -0.113 |
Liberties | -0.112 |
Token | had |
Token ID | 550 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
Malk | +1.293 |
Andreas | +1.005 |
Manson | +0.961 |
twists | +0.950 |
Goldberg | +0.948 |
Neg logit contributions | |
uncle | -1.073 |
oker | -1.048 |
ecast | -1.002 |
enegger | -0.984 |
Liberties | -0.984 |
Token | for |
Token ID | 329 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
uncle | +0.991 |
oker | +0.966 |
ecast | +0.925 |
enegger | +0.906 |
bnb | +0.905 |
Neg logit contributions | |
Malk | -1.181 |
Andreas | -0.913 |
Manson | -0.877 |
twists | -0.861 |
Goldberg | -0.859 |
Token | Stephen |
Token ID | 7970 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
Malk | +0.290 |
Andreas | +0.226 |
Manson | +0.217 |
Goldberg | +0.214 |
twists | +0.214 |
Neg logit contributions | |
uncle | -0.237 |
oker | -0.231 |
ecast | -0.223 |
enegger | -0.219 |
Liberties | -0.218 |
Token | 's |
Token ID | 338 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
Malk | +1.060 |
Andreas | +0.853 |
Manson | +0.824 |
Goldberg | +0.813 |
twists | +0.812 |
Neg logit contributions | |
uncle | -0.637 |
oker | -0.620 |
ecast | -0.588 |
Liberties | -0.577 |
enegger | -0.577 |
Token | self |
Token ID | 2116 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
uncle | +0.350 |
oker | +0.343 |
ecast | +0.329 |
Liberties | +0.324 |
enegger | +0.324 |
Neg logit contributions | |
Malk | -0.377 |
Andreas | -0.288 |
Manson | -0.276 |
Goldberg | -0.271 |
twists | -0.271 |
Token | - |
Token ID | 12 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Malk | +0.556 |
Andreas | +0.448 |
Manson | +0.433 |
Goldberg | +0.427 |
twists | +0.427 |
Neg logit contributions | |
uncle | -0.324 |
oker | -0.316 |
ecast | -0.298 |
Liberties | -0.292 |
enegger | -0.292 |
Token | destruct |
Token ID | 35678 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
Malk | +0.489 |
Andreas | +0.380 |
Manson | +0.364 |
twists | +0.357 |
Goldberg | +0.357 |
Neg logit contributions | |
uncle | -0.416 |
oker | -0.408 |
ecast | -0.393 |
Liberties | -0.388 |
bnb | -0.387 |
Token | ive |
Token ID | 425 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
Malk | +4.167 |
Andreas | +3.185 |
Manson | +3.027 |
twists | +3.026 |
Goldberg | +3.005 |
Neg logit contributions | |
uncle | -4.127 |
oker | -3.964 |
enegger | -3.848 |
Liberties | -3.836 |
bnb | -3.828 |
Token | melancholy |
Token ID | 47886 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
Malk | +0.861 |
Andreas | +0.654 |
Manson | +0.625 |
twists | +0.617 |
Goldberg | +0.615 |
Neg logit contributions | |
uncle | -0.850 |
oker | -0.831 |
ecast | -0.802 |
enegger | -0.792 |
bnb | -0.790 |
Token | , |
Token ID | 11 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
uncle | +0.530 |
oker | +0.516 |
ecast | +0.490 |
Liberties | +0.481 |
enegger | +0.481 |
Neg logit contributions | |
Malk | -0.830 |
Andreas | -0.664 |
Manson | -0.640 |
Goldberg | -0.632 |
twists | -0.631 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
Malk | +0.112 |
Andreas | +0.086 |
Manson | +0.083 |
twists | +0.082 |
Goldberg | +0.082 |
Neg logit contributions | |
uncle | -0.098 |
oker | -0.095 |
ecast | -0.092 |
enegger | -0.090 |
Liberties | -0.090 |
Token | their |
Token ID | 511 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
Malk | +0.034 |
Andreas | +0.026 |
Manson | +0.025 |
Goldberg | +0.025 |
twists | +0.025 |
Neg logit contributions | |
uncle | -0.029 |
oker | -0.029 |
ecast | -0.027 |
enegger | -0.027 |
Liberties | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
Malk | +0.200 |
Andreas | +0.151 |
Manson | +0.144 |
Goldberg | +0.142 |
twists | +0.141 |
Neg logit contributions | |
uncle | -0.200 |
oker | -0.196 |
ecast | -0.188 |
Liberties | -0.185 |
enegger | -0.185 |
Token | allowed |
Token ID | 3142 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
uncle | +0.337 |
oker | +0.330 |
ecast | +0.316 |
Liberties | +0.312 |
enegger | +0.312 |
Neg logit contributions | |
Malk | -0.352 |
Andreas | -0.268 |
Manson | -0.256 |
Goldberg | -0.252 |
twists | -0.252 |
Token | to |
Token ID | 284 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
Malk | +0.601 |
Andreas | +0.472 |
Manson | +0.454 |
Goldberg | +0.448 |
twists | +0.447 |
Neg logit contributions | |
uncle | -0.452 |
oker | -0.441 |
ecast | -0.421 |
Liberties | -0.414 |
enegger | -0.414 |
Token | self |
Token ID | 2116 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Malk | +0.386 |
Andreas | +0.285 |
Manson | +0.271 |
Goldberg | +0.265 |
twists | +0.265 |
Neg logit contributions | |
uncle | -0.442 |
oker | -0.434 |
ecast | -0.418 |
enegger | -0.412 |
Liberties | -0.412 |
Token | - |
Token ID | 12 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
Malk | +0.291 |
Andreas | +0.232 |
Manson | +0.224 |
Goldberg | +0.221 |
twists | +0.221 |
Neg logit contributions | |
uncle | -0.192 |
oker | -0.187 |
ecast | -0.178 |
Liberties | -0.174 |
enegger | -0.174 |
Token | destruct |
Token ID | 35678 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
Malk | +0.546 |
Andreas | +0.399 |
Manson | +0.377 |
Goldberg | +0.369 |
twists | +0.369 |
Neg logit contributions | |
uncle | -0.661 |
oker | -0.650 |
ecast | -0.627 |
enegger | -0.620 |
Liberties | -0.620 |
Token | by |
Token ID | 416 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
Malk | +4.365 |
Andreas | +3.382 |
Manson | +3.192 |
twists | +3.178 |
Goldberg | +3.173 |
Neg logit contributions | |
uncle | -3.909 |
oker | -3.815 |
ecast | -3.671 |
Liberties | -3.643 |
enegger | -3.610 |
Token | the |
Token ID | 262 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
Malk | +0.989 |
Andreas | +0.762 |
Manson | +0.729 |
Goldberg | +0.717 |
twists | +0.716 |
Neg logit contributions | |
uncle | -0.881 |
oker | -0.861 |
ecast | -0.826 |
enegger | -0.813 |
Liberties | -0.812 |
Token | creation |
Token ID | 6282 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Malk | +0.046 |
Andreas | +0.034 |
Manson | +0.032 |
Goldberg | +0.031 |
twists | +0.031 |
Neg logit contributions | |
uncle | -0.054 |
oker | -0.053 |
ecast | -0.051 |
Liberties | -0.050 |
enegger | -0.050 |
Token | of |
Token ID | 286 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
uncle | +0.212 |
oker | +0.207 |
ecast | +0.197 |
bnb | +0.193 |
Liberties | +0.192 |
Neg logit contributions | |
Malk | -0.297 |
Andreas | -0.235 |
Manson | -0.227 |
Goldberg | -0.223 |
twists | -0.223 |
Token | the |
Token ID | 262 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
Malk | +0.343 |
Andreas | +0.265 |
Manson | +0.254 |
Goldberg | +0.250 |
twists | +0.250 |
Neg logit contributions | |
uncle | -0.297 |
oker | -0.290 |
ecast | -0.278 |
Liberties | -0.273 |
enegger | -0.273 |
Token | 4 |
Token ID | 604 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Malk | +0.563 |
Andreas | +0.437 |
Manson | +0.419 |
twists | +0.413 |
Goldberg | +0.413 |
Neg logit contributions | |
uncle | -0.471 |
oker | -0.460 |
ecast | -0.439 |
enegger | -0.434 |
Liberties | -0.432 |
Token | 34 |
Token ID | 2682 |
Feature activation | +5.10e-03 |
Pos logit contributions | |
Malk | +0.392 |
Andreas | +0.312 |
Manson | +0.299 |
Goldberg | +0.296 |
twists | +0.295 |
Neg logit contributions | |
uncle | -0.265 |
oker | -0.255 |
ecast | -0.248 |
enegger | -0.245 |
Liberties | -0.241 |
Token | - |
Token ID | 12 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
Malk | +0.015 |
Andreas | +0.012 |
Manson | +0.011 |
Goldberg | +0.011 |
twists | +0.011 |
Neg logit contributions | |
uncle | -0.009 |
oker | -0.009 |
ecast | -0.008 |
enegger | -0.008 |
Liberties | -0.008 |
Token | lb |
Token ID | 23160 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Malk | +0.859 |
Andreas | +0.638 |
Manson | +0.611 |
twists | +0.603 |
Goldberg | +0.600 |
Neg logit contributions | |
uncle | -0.920 |
oker | -0.910 |
enegger | -0.877 |
ecast | -0.875 |
Liberties | -0.869 |
Token | .- |
Token ID | 7874 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
Malk | +3.403 |
Andreas | +2.727 |
Manson | +2.604 |
twists | +2.575 |
Goldberg | +2.569 |
Neg logit contributions | |
uncle | -2.254 |
oker | -2.190 |
enegger | -2.138 |
ecast | -2.138 |
Liberties | -2.108 |
Token | ft |
Token ID | 701 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
Malk | +0.564 |
Andreas | +0.402 |
Manson | +0.378 |
Goldberg | +0.370 |
twists | +0.370 |
Neg logit contributions | |
uncle | -0.777 |
oker | -0.763 |
ecast | -0.739 |
enegger | -0.734 |
Liberties | -0.731 |
Token | . |
Token ID | 13 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
Malk | +4.771 |
Andreas | +3.809 |
Manson | +3.659 |
Goldberg | +3.643 |
twists | +3.596 |
Neg logit contributions | |
uncle | -3.355 |
oker | -3.254 |
ecast | -3.151 |
enegger | -3.147 |
Liberties | -3.111 |
Token | 6 |
Token ID | 718 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
Malk | +1.542 |
Andreas | +1.195 |
Manson | +1.142 |
Goldberg | +1.130 |
twists | +1.127 |
Neg logit contributions | |
uncle | -1.346 |
oker | -1.313 |
ecast | -1.262 |
enegger | -1.260 |
Liberties | -1.240 |
Token | . |
Token ID | 13 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
Malk | +1.105 |
Andreas | +0.875 |
Manson | +0.838 |
twists | +0.824 |
Goldberg | +0.823 |
Neg logit contributions | |
uncle | -0.844 |
oker | -0.813 |
enegger | -0.783 |
ecast | -0.780 |
Liberties | -0.772 |
Token | 2 |
Token ID | 17 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
Malk | +1.379 |
Andreas | +1.049 |
Manson | +0.999 |
twists | +0.970 |
Goldberg | +0.969 |
Neg logit contributions | |
uncle | -1.391 |
oker | -1.350 |
ecast | -1.323 |
enegger | -1.302 |
bnb | -1.288 |
Token | - |
Token ID | 12 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
Malk | +0.268 |
Andreas | +0.215 |
Manson | +0.208 |
twists | +0.206 |
Goldberg | +0.205 |
Neg logit contributions | |
uncle | -0.164 |
oker | -0.159 |
ecast | -0.151 |
enegger | -0.150 |
Liberties | -0.148 |
Token |
|
Token ID | 198 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
Malk | +0.716 |
Andreas | +0.570 |
Manson | +0.542 |
Goldberg | +0.538 |
twists | +0.534 |
Neg logit contributions | |
uncle | -0.530 |
oker | -0.520 |
ecast | -0.500 |
enegger | -0.500 |
bnb | -0.493 |
Token | Col |
Token ID | 5216 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
Malk | +0.739 |
Andreas | +0.567 |
Manson | +0.541 |
twists | +0.534 |
Goldberg | +0.533 |
Neg logit contributions | |
uncle | -0.661 |
oker | -0.652 |
enegger | -0.626 |
ecast | -0.625 |
Liberties | -0.620 |
Token | on |
Token ID | 261 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
Malk | +0.588 |
Andreas | +0.459 |
Manson | +0.442 |
Goldberg | +0.436 |
twists | +0.433 |
Neg logit contributions | |
uncle | -0.472 |
oker | -0.461 |
ecast | -0.441 |
Liberties | -0.438 |
enegger | -0.436 |
Token | ize |
Token ID | 1096 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
uncle | +0.053 |
oker | +0.051 |
ecast | +0.048 |
enegger | +0.048 |
Liberties | +0.048 |
Neg logit contributions | |
Malk | -0.070 |
Andreas | -0.055 |
Manson | -0.053 |
twists | -0.052 |
Goldberg | -0.052 |
Token | unch |
Token ID | 16258 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
Malk | +0.341 |
Andreas | +0.264 |
Manson | +0.252 |
Goldberg | +0.249 |
twists | +0.248 |
Neg logit contributions | |
uncle | -0.297 |
oker | -0.290 |
ecast | -0.278 |
enegger | -0.273 |
Liberties | -0.273 |
Token | arted |
Token ID | 19112 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
Malk | +0.915 |
Andreas | +0.678 |
Manson | +0.643 |
Goldberg | +0.629 |
twists | +0.619 |
Neg logit contributions | |
uncle | -1.028 |
oker | -0.995 |
ecast | -0.977 |
enegger | -0.975 |
bnb | -0.964 |
Token | planets |
Token ID | 14705 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
Malk | +3.459 |
Andreas | +2.488 |
twists | +2.358 |
Manson | +2.311 |
Goldberg | +2.278 |
Neg logit contributions | |
uncle | -4.794 |
oker | -4.657 |
ecast | -4.546 |
enegger | -4.486 |
bnb | -4.444 |
Token | and |
Token ID | 290 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
Malk | +1.185 |
Andreas | +0.923 |
Manson | +0.884 |
twists | +0.874 |
Goldberg | +0.871 |
Neg logit contributions | |
uncle | -0.983 |
oker | -0.959 |
ecast | -0.916 |
enegger | -0.907 |
Liberties | -0.901 |
Token | collect |
Token ID | 2824 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Malk | +0.301 |
Andreas | +0.228 |
Manson | +0.218 |
Goldberg | +0.214 |
twists | +0.213 |
Neg logit contributions | |
uncle | -0.295 |
oker | -0.289 |
ecast | -0.277 |
Liberties | -0.273 |
enegger | -0.273 |
Token | gifts |
Token ID | 13201 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Malk | +2.369 |
Andreas | +1.770 |
Manson | +1.699 |
twists | +1.692 |
Goldberg | +1.675 |
Neg logit contributions | |
uncle | -2.441 |
oker | -2.380 |
ecast | -2.293 |
enegger | -2.262 |
bnb | -2.257 |
Token | from |
Token ID | 422 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
Malk | +0.394 |
Andreas | +0.308 |
Manson | +0.296 |
twists | +0.292 |
Goldberg | +0.292 |
Neg logit contributions | |
uncle | -0.310 |
oker | -0.304 |
ecast | -0.290 |
enegger | -0.285 |
Liberties | -0.284 |
Token | . |
Token ID | 13 |
Feature activation | -9.84e-02 |
Pos logit contributions | |
uncle | +0.430 |
oker | +0.419 |
ecast | +0.400 |
Liberties | +0.393 |
enegger | +0.392 |
Neg logit contributions | |
Malk | -0.588 |
Andreas | -0.464 |
Manson | -0.446 |
Goldberg | -0.440 |
twists | -0.439 |
Token |
|
Token ID | 198 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
uncle | +0.184 |
oker | +0.179 |
ecast | +0.171 |
enegger | +0.168 |
Liberties | +0.167 |
Neg logit contributions | |
Malk | -0.273 |
Andreas | -0.217 |
Manson | -0.209 |
Goldberg | -0.206 |
twists | -0.205 |
Token |
|
Token ID | 198 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
uncle | +0.084 |
oker | +0.082 |
ecast | +0.078 |
enegger | +0.078 |
bnb | +0.077 |
Neg logit contributions | |
Malk | -0.118 |
Andreas | -0.093 |
Manson | -0.089 |
Goldberg | -0.088 |
twists | -0.088 |
Token | Follow |
Token ID | 7155 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
Malk | +0.142 |
Andreas | +0.110 |
Manson | +0.105 |
Goldberg | +0.104 |
twists | +0.103 |
Neg logit contributions | |
uncle | -0.120 |
oker | -0.117 |
ecast | -0.113 |
enegger | -0.111 |
Liberties | -0.111 |
Token | @ |
Token ID | 2488 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
uncle | +0.032 |
oker | +0.031 |
ecast | +0.030 |
enegger | +0.029 |
Liberties | +0.029 |
Neg logit contributions | |
Malk | -0.034 |
Andreas | -0.026 |
Manson | -0.025 |
Goldberg | -0.024 |
twists | -0.024 |
Token | greg |
Token ID | 9903 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
uncle | +1.845 |
oker | +1.798 |
ecast | +1.714 |
bnb | +1.681 |
Liberties | +1.678 |
Neg logit contributions | |
Malk | -2.262 |
Andreas | -1.757 |
Manson | -1.698 |
Goldberg | -1.665 |
twists | -1.659 |
Token | ory |
Token ID | 652 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
Malk | +4.286 |
Andreas | +3.263 |
Manson | +3.187 |
twists | +3.110 |
Goldberg | +3.067 |
Neg logit contributions | |
uncle | -3.613 |
Liberties | -3.541 |
oker | -3.536 |
覚醒 | -3.504 |
enegger | -3.488 |
Token | k |
Token ID | 74 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
Malk | +1.659 |
Andreas | +1.286 |
Manson | +1.238 |
twists | +1.213 |
Goldberg | +1.210 |
Neg logit contributions | |
uncle | -1.440 |
oker | -1.407 |
Liberties | -1.359 |
ecast | -1.358 |
enegger | -1.338 |
Token | ort |
Token ID | 419 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
Malk | +2.355 |
Andreas | +1.864 |
Manson | +1.779 |
Goldberg | +1.767 |
twists | +1.741 |
Neg logit contributions | |
uncle | -1.749 |
oker | -1.707 |
ecast | -1.678 |
Liberties | -1.658 |
bnb | -1.645 |
Token | e |
Token ID | 68 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
Malk | +1.071 |
Andreas | +0.874 |
Manson | +0.843 |
twists | +0.836 |
Goldberg | +0.834 |
Neg logit contributions | |
uncle | -0.575 |
oker | -0.553 |
ecast | -0.525 |
enegger | -0.516 |
Liberties | -0.512 |
Token | on |
Token ID | 319 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
Malk | +0.536 |
Andreas | +0.419 |
Manson | +0.402 |
Goldberg | +0.397 |
twists | +0.395 |
Neg logit contributions | |
uncle | -0.426 |
oker | -0.416 |
ecast | -0.397 |
Liberties | -0.391 |
enegger | -0.391 |
Token | deploying |
Token ID | 29682 |
Feature activation | +6.70e-01 |
Pos logit contributions | |
uncle | +0.797 |
oker | +0.779 |
ecast | +0.747 |
Liberties | +0.735 |
enegger | +0.734 |
Neg logit contributions | |
Malk | -0.910 |
Andreas | -0.702 |
Manson | -0.673 |
Goldberg | -0.662 |
twists | -0.661 |
Token | them |
Token ID | 606 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
Malk | +1.555 |
Andreas | +1.184 |
Manson | +1.121 |
twists | +1.108 |
Goldberg | +1.108 |
Neg logit contributions | |
uncle | -1.555 |
oker | -1.525 |
ecast | -1.465 |
enegger | -1.447 |
bnb | -1.437 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
Malk | +0.661 |
Andreas | +0.522 |
Manson | +0.499 |
Goldberg | +0.492 |
twists | +0.492 |
Neg logit contributions | |
uncle | -0.512 |
oker | -0.500 |
ecast | -0.476 |
enegger | -0.468 |
Liberties | -0.468 |
Token | In |
Token ID | 554 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
uncle | +0.214 |
oker | +0.209 |
ecast | +0.199 |
enegger | +0.196 |
bnb | +0.194 |
Neg logit contributions | |
Malk | -0.304 |
Andreas | -0.242 |
Manson | -0.232 |
Goldberg | -0.229 |
twists | -0.228 |
Token | Bitcoin |
Token ID | 6185 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Malk | +0.286 |
Andreas | +0.217 |
Manson | +0.207 |
Goldberg | +0.204 |
twists | +0.203 |
Neg logit contributions | |
uncle | -0.286 |
oker | -0.280 |
ecast | -0.269 |
enegger | -0.266 |
Liberties | -0.264 |
Token | Core |
Token ID | 7231 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Malk | +3.957 |
Andreas | +3.111 |
Goldberg | +2.950 |
twists | +2.932 |
Manson | +2.910 |
Neg logit contributions | |
uncle | -3.245 |
oker | -3.152 |
ecast | -3.056 |
enegger | -3.032 |
bnb | -2.958 |
Token | we |
Token ID | 356 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Malk | +4.267 |
Andreas | +3.360 |
Manson | +3.132 |
twists | +3.129 |
Goldberg | +3.120 |
Neg logit contributions | |
uncle | -3.709 |
oker | -3.585 |
ecast | -3.484 |
enegger | -3.455 |
bnb | -3.357 |
Token | should |
Token ID | 815 |
Feature activation | -9.53e-02 |
Pos logit contributions | |
Malk | +0.403 |
Andreas | +0.311 |
Manson | +0.298 |
Goldberg | +0.293 |
twists | +0.293 |
Neg logit contributions | |
uncle | -0.354 |
oker | -0.346 |
ecast | -0.332 |
enegger | -0.326 |
Liberties | -0.326 |
Token | keep |
Token ID | 1394 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
uncle | +0.214 |
oker | +0.209 |
ecast | +0.201 |
enegger | +0.198 |
Liberties | +0.198 |
Neg logit contributions | |
Malk | -0.228 |
Andreas | -0.174 |
Manson | -0.167 |
Goldberg | -0.164 |
twists | -0.164 |
Token | patches |
Token ID | 16082 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
Malk | +0.373 |
Andreas | +0.286 |
Manson | +0.272 |
Goldberg | +0.268 |
twists | +0.268 |
Neg logit contributions | |
uncle | -0.347 |
oker | -0.339 |
ecast | -0.325 |
enegger | -0.322 |
bnb | -0.319 |
Token | ready |
Token ID | 3492 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
Malk | +1.750 |
Andreas | +1.358 |
Manson | +1.298 |
twists | +1.281 |
Goldberg | +1.280 |
Neg logit contributions | |
uncle | -1.561 |
oker | -1.506 |
ecast | -1.451 |
enegger | -1.441 |
Liberties | -1.440 |
Token | seemed |
Token ID | 3947 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
uncle | +1.266 |
oker | +1.240 |
ecast | +1.191 |
enegger | +1.174 |
Liberties | +1.165 |
Neg logit contributions | |
Malk | -1.431 |
Andreas | -1.102 |
Manson | -1.051 |
Goldberg | -1.036 |
twists | -1.027 |
Token | intent |
Token ID | 6824 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
uncle | +1.035 |
oker | +1.013 |
ecast | +0.970 |
Liberties | +0.944 |
enegger | +0.942 |
Neg logit contributions | |
Malk | -1.327 |
Andreas | -1.038 |
Manson | -0.994 |
twists | -0.984 |
Goldberg | -0.983 |
Token | on |
Token ID | 319 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
uncle | +0.772 |
oker | +0.768 |
ecast | +0.735 |
bnb | +0.717 |
enegger | +0.714 |
Neg logit contributions | |
Malk | -0.826 |
Andreas | -0.632 |
Manson | -0.601 |
Goldberg | -0.597 |
PRESS | -0.590 |
Token | creating |
Token ID | 4441 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
Malk | +0.270 |
Andreas | +0.208 |
Manson | +0.200 |
twists | +0.196 |
Goldberg | +0.196 |
Neg logit contributions | |
uncle | -0.238 |
oker | -0.233 |
ecast | -0.223 |
Liberties | -0.220 |
enegger | -0.220 |
Token | a |
Token ID | 257 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
Malk | +0.658 |
Andreas | +0.507 |
Manson | +0.485 |
Goldberg | +0.477 |
twists | +0.476 |
Neg logit contributions | |
uncle | -0.585 |
oker | -0.573 |
ecast | -0.548 |
enegger | -0.540 |
Liberties | -0.540 |
Token | stark |
Token ID | 19278 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Malk | +0.405 |
Andreas | +0.304 |
Manson | +0.291 |
Goldberg | +0.285 |
twists | +0.285 |
Neg logit contributions | |
uncle | -0.415 |
oker | -0.406 |
ecast | -0.391 |
Liberties | -0.386 |
enegger | -0.385 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Malk | +4.154 |
Andreas | +3.202 |
Manson | +3.122 |
Goldberg | +3.058 |
twists | +3.049 |
Neg logit contributions | |
uncle | -3.777 |
oker | -3.582 |
ecast | -3.546 |
Liberties | -3.459 |
bnb | -3.458 |
Token | disturbing |
Token ID | 14851 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
Malk | +0.379 |
Andreas | +0.285 |
Manson | +0.273 |
Goldberg | +0.268 |
twists | +0.268 |
Neg logit contributions | |
uncle | -0.387 |
oker | -0.379 |
ecast | -0.365 |
Liberties | -0.360 |
enegger | -0.359 |
Token | portrait |
Token ID | 18560 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Malk | +1.167 |
Andreas | +0.887 |
Manson | +0.853 |
Goldberg | +0.837 |
twists | +0.836 |
Neg logit contributions | |
uncle | -1.129 |
oker | -1.093 |
ecast | -1.062 |
Liberties | -1.046 |
bnb | -1.040 |
Token | of |
Token ID | 286 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
uncle | +0.208 |
oker | +0.203 |
ecast | +0.192 |
enegger | +0.189 |
bnb | +0.189 |
Neg logit contributions | |
Malk | -0.291 |
Andreas | -0.230 |
Manson | -0.220 |
twists | -0.218 |
Goldberg | -0.218 |
Token | a |
Token ID | 257 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
Malk | +0.319 |
Andreas | +0.244 |
Manson | +0.234 |
twists | +0.229 |
Goldberg | +0.229 |
Neg logit contributions | |
uncle | -0.296 |
oker | -0.289 |
ecast | -0.279 |
enegger | -0.274 |
Liberties | -0.274 |
Token | : |
Token ID | 25 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
uncle | +0.155 |
oker | +0.152 |
ecast | +0.145 |
enegger | +0.143 |
Liberties | +0.142 |
Neg logit contributions | |
Malk | -0.207 |
Andreas | -0.163 |
Manson | -0.157 |
Goldberg | -0.155 |
twists | -0.154 |
Token | � |
Token ID | 564 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
Malk | +0.502 |
Andreas | +0.390 |
Manson | +0.375 |
Goldberg | +0.371 |
twists | +0.369 |
Neg logit contributions | |
uncle | -0.404 |
oker | -0.398 |
ecast | -0.381 |
enegger | -0.376 |
bnb | -0.371 |
Token | � |
Token ID | 250 |
Feature activation | -7.38e-03 |
Pos logit contributions | |
Malk | +0.371 |
Andreas | +0.280 |
Manson | +0.269 |
Goldberg | +0.268 |
twists | +0.266 |
Neg logit contributions | |
uncle | -0.355 |
oker | -0.352 |
ecast | -0.339 |
enegger | -0.334 |
Liberties | -0.332 |
Token | you |
Token ID | 5832 |
Feature activation | +9.29e-03 |
Pos logit contributions | |
uncle | +0.015 |
oker | +0.015 |
ecast | +0.015 |
enegger | +0.015 |
Liberties | +0.014 |
Neg logit contributions | |
Malk | -0.019 |
Andreas | -0.014 |
Manson | -0.014 |
Goldberg | -0.014 |
twists | -0.014 |
Token | are |
Token ID | 389 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
Malk | +0.023 |
Andreas | +0.018 |
Manson | +0.017 |
twists | +0.017 |
Goldberg | +0.017 |
Neg logit contributions | |
uncle | -0.020 |
oker | -0.020 |
ecast | -0.019 |
enegger | -0.019 |
Liberties | -0.019 |
Token | either |
Token ID | 2035 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Malk | +1.075 |
Andreas | +0.828 |
Manson | +0.797 |
Goldberg | +0.782 |
twists | +0.777 |
Neg logit contributions | |
uncle | -0.928 |
oker | -0.911 |
ecast | -0.866 |
enegger | -0.859 |
Liberties | -0.857 |
Token | with |
Token ID | 351 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
Malk | +4.143 |
Andreas | +3.143 |
Manson | +3.018 |
Goldberg | +2.964 |
twists | +2.924 |
Neg logit contributions | |
uncle | -3.770 |
oker | -3.732 |
ecast | -3.550 |
enegger | -3.549 |
Liberties | -3.543 |
Token | us |
Token ID | 514 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
uncle | +0.418 |
oker | +0.408 |
ecast | +0.392 |
enegger | +0.387 |
Liberties | +0.385 |
Neg logit contributions | |
Malk | -0.471 |
Andreas | -0.362 |
Manson | -0.347 |
Goldberg | -0.342 |
twists | -0.340 |
Token | or |
Token ID | 393 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
Malk | +1.351 |
Andreas | +1.053 |
Manson | +1.013 |
Goldberg | +0.994 |
twists | +0.993 |
Neg logit contributions | |
uncle | -1.071 |
oker | -1.045 |
ecast | -1.003 |
enegger | -0.984 |
Liberties | -0.983 |
Token | with |
Token ID | 351 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
Malk | +0.689 |
Andreas | +0.544 |
Manson | +0.523 |
Goldberg | +0.516 |
twists | +0.515 |
Neg logit contributions | |
uncle | -0.488 |
oker | -0.480 |
ecast | -0.457 |
enegger | -0.452 |
Liberties | -0.448 |
Token | the |
Token ID | 262 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
uncle | +0.148 |
oker | +0.145 |
ecast | +0.139 |
enegger | +0.138 |
Liberties | +0.137 |
Neg logit contributions | |
Malk | -0.142 |
Andreas | -0.107 |
Manson | -0.102 |
Goldberg | -0.100 |
twists | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
Malk | +1.290 |
Andreas | +1.007 |
Manson | +0.970 |
twists | +0.954 |
Goldberg | +0.949 |
Neg logit contributions | |
uncle | -1.082 |
oker | -1.055 |
ecast | -1.018 |
enegger | -1.002 |
Liberties | -0.997 |
Token | 4 |
Token ID | 604 |
Feature activation | -8.08e-02 |
Pos logit contributions | |
Malk | +0.369 |
Andreas | +0.287 |
Manson | +0.276 |
twists | +0.272 |
Goldberg | +0.271 |
Neg logit contributions | |
uncle | -0.308 |
oker | -0.301 |
ecast | -0.288 |
enegger | -0.284 |
Liberties | -0.283 |
Token | 34 |
Token ID | 2682 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
uncle | +0.153 |
oker | +0.150 |
ecast | +0.145 |
enegger | +0.142 |
bnb | +0.140 |
Neg logit contributions | |
Malk | -0.223 |
Andreas | -0.177 |
Manson | -0.170 |
Goldberg | -0.168 |
twists | -0.168 |
Token | lb |
Token ID | 18360 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
uncle | +0.303 |
oker | +0.296 |
ecast | +0.285 |
enegger | +0.281 |
Liberties | +0.281 |
Neg logit contributions | |
Malk | -0.312 |
Andreas | -0.237 |
Manson | -0.226 |
twists | -0.223 |
Goldberg | -0.223 |
Token | .- |
Token ID | 7874 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
Malk | +2.929 |
Andreas | +2.356 |
Manson | +2.268 |
twists | +2.238 |
Goldberg | +2.218 |
Neg logit contributions | |
uncle | -2.039 |
oker | -1.988 |
ecast | -1.924 |
enegger | -1.907 |
覚醒 | -1.874 |
Token | ft |
Token ID | 701 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Malk | +0.350 |
Andreas | +0.249 |
Manson | +0.228 |
twists | +0.228 |
Goldberg | +0.222 |
Neg logit contributions | |
uncle | -0.529 |
oker | -0.519 |
enegger | -0.508 |
ecast | -0.505 |
Liberties | -0.499 |
Token | . |
Token ID | 13 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
Malk | +4.776 |
Andreas | +3.835 |
Manson | +3.707 |
twists | +3.658 |
Goldberg | +3.652 |
Neg logit contributions | |
uncle | -3.041 |
oker | -2.946 |
ecast | -2.832 |
Liberties | -2.768 |
enegger | -2.767 |
Token | of |
Token ID | 286 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Malk | +0.950 |
Andreas | +0.744 |
Manson | +0.715 |
twists | +0.706 |
Goldberg | +0.705 |
Neg logit contributions | |
uncle | -0.742 |
oker | -0.721 |
ecast | -0.692 |
enegger | -0.684 |
Liberties | -0.676 |
Token | torque |
Token ID | 26415 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
Malk | +0.431 |
Andreas | +0.324 |
Manson | +0.308 |
twists | +0.307 |
Goldberg | +0.305 |
Neg logit contributions | |
uncle | -0.450 |
oker | -0.441 |
ecast | -0.425 |
enegger | -0.421 |
Liberties | -0.417 |
Token | . |
Token ID | 13 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
Malk | +0.418 |
Andreas | +0.336 |
Manson | +0.324 |
twists | +0.320 |
Goldberg | +0.320 |
Neg logit contributions | |
uncle | -0.269 |
oker | -0.261 |
ecast | -0.249 |
enegger | -0.244 |
Liberties | -0.244 |
Token |
|
Token ID | 198 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
Malk | +0.835 |
Andreas | +0.667 |
Manson | +0.644 |
twists | +0.635 |
Goldberg | +0.635 |
Neg logit contributions | |
uncle | -0.539 |
oker | -0.525 |
ecast | -0.499 |
Liberties | -0.490 |
enegger | -0.489 |
Token | y |
Token ID | 88 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
Malk | +0.104 |
Andreas | +0.079 |
Manson | +0.075 |
Goldberg | +0.073 |
twists | +0.073 |
Neg logit contributions | |
uncle | -0.107 |
oker | -0.105 |
ecast | -0.101 |
enegger | -0.100 |
Liberties | -0.100 |
Token | atta |
Token ID | 25014 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Malk | +0.710 |
Andreas | +0.599 |
Manson | +0.579 |
Goldberg | +0.569 |
twists | +0.568 |
Neg logit contributions | |
uncle | -0.258 |
oker | -0.252 |
enegger | -0.238 |
ecast | -0.236 |
Liberties | -0.235 |
Token | travelled |
Token ID | 21650 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
Malk | +0.384 |
Andreas | +0.296 |
Manson | +0.282 |
Goldberg | +0.279 |
twists | +0.278 |
Neg logit contributions | |
uncle | -0.341 |
oker | -0.333 |
ecast | -0.319 |
Liberties | -0.315 |
bnb | -0.314 |
Token | to |
Token ID | 284 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
uncle | +0.927 |
oker | +0.906 |
ecast | +0.863 |
enegger | +0.849 |
Liberties | +0.848 |
Neg logit contributions | |
Malk | -1.239 |
Andreas | -0.975 |
Manson | -0.937 |
Goldberg | -0.924 |
twists | -0.922 |
Token | Mog |
Token ID | 30926 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
uncle | +0.659 |
oker | +0.644 |
ecast | +0.615 |
Liberties | +0.605 |
enegger | +0.605 |
Neg logit contributions | |
Malk | -0.840 |
Andreas | -0.658 |
Manson | -0.632 |
Goldberg | -0.622 |
twists | -0.622 |
Token | adish |
Token ID | 48563 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
Malk | +0.453 |
Andreas | +0.359 |
Manson | +0.345 |
Goldberg | +0.341 |
twists | +0.339 |
Neg logit contributions | |
uncle | -0.324 |
oker | -0.315 |
ecast | -0.302 |
Liberties | -0.297 |
enegger | -0.297 |
Token | u |
Token ID | 84 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
Malk | +4.257 |
Andreas | +3.401 |
Manson | +3.205 |
Goldberg | +3.159 |
Mai | +3.118 |
Neg logit contributions | |
uncle | -3.411 |
oker | -3.404 |
enegger | -3.333 |
ecast | -3.249 |
bnb | -3.200 |
Token | to |
Token ID | 284 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Malk | +0.475 |
Andreas | +0.371 |
Manson | +0.354 |
Goldberg | +0.350 |
twists | +0.348 |
Neg logit contributions | |
uncle | -0.389 |
oker | -0.380 |
ecast | -0.364 |
enegger | -0.359 |
Liberties | -0.356 |
Token | discuss |
Token ID | 2112 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
Malk | +0.273 |
Andreas | +0.213 |
Manson | +0.204 |
Goldberg | +0.201 |
twists | +0.201 |
Neg logit contributions | |
uncle | -0.225 |
oker | -0.220 |
ecast | -0.211 |
enegger | -0.208 |
bnb | -0.207 |
Token | the |
Token ID | 262 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
Malk | +0.318 |
Andreas | +0.244 |
Manson | +0.232 |
Goldberg | +0.230 |
twists | +0.229 |
Neg logit contributions | |
uncle | -0.289 |
oker | -0.282 |
ecast | -0.271 |
enegger | -0.268 |
Liberties | -0.266 |
Token | issue |
Token ID | 2071 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
uncle | +0.106 |
oker | +0.103 |
ecast | +0.099 |
enegger | +0.098 |
bnb | +0.098 |
Neg logit contributions | |
Malk | -0.121 |
Andreas | -0.093 |
Manson | -0.089 |
Goldberg | -0.088 |
twists | -0.088 |
Token | Next |
Token ID | 7406 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
uncle | +0.397 |
oker | +0.388 |
ecast | +0.372 |
enegger | +0.367 |
Liberties | +0.364 |
Neg logit contributions | |
Malk | -0.467 |
Andreas | -0.363 |
Manson | -0.347 |
Goldberg | -0.343 |
twists | -0.341 |
Token | E |
Token ID | 36 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
uncle | +0.981 |
oker | +0.955 |
ecast | +0.910 |
enegger | +0.900 |
Liberties | +0.895 |
Neg logit contributions | |
Malk | -1.417 |
Andreas | -1.124 |
Manson | -1.077 |
Goldberg | -1.070 |
twists | -1.066 |
Token | ra |
Token ID | 430 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
uncle | +0.857 |
oker | +0.836 |
ecast | +0.798 |
Liberties | +0.784 |
enegger | +0.783 |
Neg logit contributions | |
Malk | -1.164 |
Andreas | -0.917 |
Manson | -0.882 |
Goldberg | -0.870 |
twists | -0.869 |
Token | , |
Token ID | 11 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
uncle | +0.424 |
oker | +0.413 |
ecast | +0.395 |
enegger | +0.389 |
Liberties | +0.388 |
Neg logit contributions | |
Malk | -0.563 |
Andreas | -0.442 |
Manson | -0.425 |
Goldberg | -0.420 |
twists | -0.419 |
Token | in |
Token ID | 287 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
uncle | +0.416 |
oker | +0.407 |
ecast | +0.390 |
enegger | +0.384 |
Liberties | +0.383 |
Neg logit contributions | |
Malk | -0.482 |
Andreas | -0.374 |
Manson | -0.357 |
Goldberg | -0.352 |
twists | -0.352 |
Token | stark |
Token ID | 19278 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
uncle | +0.212 |
oker | +0.207 |
ecast | +0.199 |
enegger | +0.196 |
Liberties | +0.195 |
Neg logit contributions | |
Malk | -0.223 |
Andreas | -0.171 |
Manson | -0.163 |
Goldberg | -0.161 |
twists | -0.160 |
Token | contrast |
Token ID | 6273 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
Malk | +2.873 |
Andreas | +1.979 |
Goldberg | +1.848 |
twists | +1.848 |
Manson | +1.831 |
Neg logit contributions | |
uncle | -4.748 |
oker | -4.661 |
ecast | -4.559 |
bnb | -4.444 |
enegger | -4.443 |
Token | to |
Token ID | 284 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
uncle | +0.439 |
oker | +0.427 |
ecast | +0.406 |
enegger | +0.400 |
Liberties | +0.398 |
Neg logit contributions | |
Malk | -0.636 |
Andreas | -0.504 |
Manson | -0.486 |
Goldberg | -0.480 |
twists | -0.479 |
Token | the |
Token ID | 262 |
Feature activation | -5.06e-03 |
Pos logit contributions | |
uncle | +0.111 |
oker | +0.108 |
ecast | +0.104 |
enegger | +0.102 |
Liberties | +0.102 |
Neg logit contributions | |
Malk | -0.121 |
Andreas | -0.093 |
Manson | -0.089 |
Goldberg | -0.088 |
twists | -0.088 |
Token | localized |
Token ID | 36618 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
uncle | +0.011 |
oker | +0.011 |
ecast | +0.010 |
enegger | +0.010 |
Liberties | +0.010 |
Neg logit contributions | |
Malk | -0.013 |
Andreas | -0.010 |
Manson | -0.009 |
twists | -0.009 |
Goldberg | -0.009 |
Token | alternatives |
Token ID | 14693 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
uncle | +0.924 |
oker | +0.903 |
ecast | +0.868 |
enegger | +0.856 |
Liberties | +0.851 |
Neg logit contributions | |
Malk | -0.990 |
Andreas | -0.759 |
Manson | -0.724 |
twists | -0.716 |
Goldberg | -0.714 |
Token | 2015 |
Token ID | 1853 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
uncle | +0.219 |
oker | +0.215 |
ecast | +0.206 |
Liberties | +0.203 |
enegger | +0.202 |
Neg logit contributions | |
Malk | -0.239 |
Andreas | -0.183 |
Manson | -0.175 |
Goldberg | -0.172 |
twists | -0.172 |
Token | : |
Token ID | 25 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
uncle | +0.644 |
oker | +0.629 |
ecast | +0.596 |
Liberties | +0.585 |
enegger | +0.584 |
Neg logit contributions | |
Malk | -0.983 |
Andreas | -0.785 |
Manson | -0.757 |
Goldberg | -0.747 |
twists | -0.745 |
Token |
|
Token ID | 198 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
uncle | +0.694 |
oker | +0.681 |
ecast | +0.647 |
Liberties | +0.640 |
enegger | +0.635 |
Neg logit contributions | |
Malk | -0.858 |
Andreas | -0.666 |
Manson | -0.642 |
Goldberg | -0.633 |
twists | -0.633 |
Token |
|
Token ID | 198 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
Malk | +0.131 |
Andreas | +0.105 |
Manson | +0.100 |
Goldberg | +0.099 |
twists | +0.099 |
Neg logit contributions | |
uncle | -0.094 |
oker | -0.092 |
ecast | -0.088 |
enegger | -0.087 |
bnb | -0.086 |
Token | Jo |
Token ID | 9908 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Malk | +0.267 |
Andreas | +0.205 |
Manson | +0.196 |
Goldberg | +0.192 |
twists | +0.192 |
Neg logit contributions | |
uncle | -0.245 |
oker | -0.240 |
ecast | -0.231 |
enegger | -0.227 |
Liberties | -0.227 |
Token | aquin |
Token ID | 48734 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
Malk | +0.505 |
Andreas | +0.405 |
Manson | +0.393 |
Goldberg | +0.387 |
twists | +0.383 |
Neg logit contributions | |
uncle | -0.311 |
oker | -0.300 |
ecast | -0.286 |
enegger | -0.282 |
Liberties | -0.282 |
Token | Ben |
Token ID | 3932 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
Malk | +3.681 |
Andreas | +2.757 |
Manson | +2.594 |
Goldberg | +2.585 |
Mai | +2.419 |
Neg logit contributions | |
enegger | -4.107 |
uncle | -4.051 |
ecast | -3.878 |
bnb | -3.876 |
oker | -3.873 |
Token | oit |
Token ID | 30711 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
uncle | +0.113 |
oker | +0.110 |
ecast | +0.107 |
enegger | +0.106 |
Liberties | +0.105 |
Neg logit contributions | |
Malk | -0.094 |
Andreas | -0.069 |
Manson | -0.066 |
Goldberg | -0.065 |
twists | -0.064 |
Token | : |
Token ID | 25 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
Malk | +1.169 |
Andreas | +0.911 |
Manson | +0.872 |
Goldberg | +0.859 |
twists | +0.858 |
Neg logit contributions | |
uncle | -0.963 |
oker | -0.938 |
ecast | -0.902 |
enegger | -0.887 |
Liberties | -0.884 |
Token | 1 |
Token ID | 352 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Malk | +0.669 |
Andreas | +0.524 |
Manson | +0.502 |
Goldberg | +0.495 |
twists | +0.494 |
Neg logit contributions | |
uncle | -0.525 |
oker | -0.512 |
ecast | -0.493 |
enegger | -0.483 |
bnb | -0.481 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
Malk | +0.322 |
Andreas | +0.259 |
Manson | +0.250 |
Goldberg | +0.246 |
twists | +0.246 |
Neg logit contributions | |
uncle | -0.201 |
oker | -0.196 |
ecast | -0.187 |
enegger | -0.184 |
Liberties | -0.183 |
Token | you |
Token ID | 345 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
uncle | +0.091 |
oker | +0.089 |
ecast | +0.085 |
enegger | +0.083 |
Liberties | +0.083 |
Neg logit contributions | |
Malk | -0.119 |
Andreas | -0.094 |
Manson | -0.090 |
Goldberg | -0.089 |
twists | -0.089 |
Token | must |
Token ID | 1276 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
uncle | +0.042 |
oker | +0.041 |
ecast | +0.039 |
Liberties | +0.039 |
enegger | +0.039 |
Neg logit contributions | |
Malk | -0.044 |
Andreas | -0.033 |
Manson | -0.032 |
twists | -0.031 |
Goldberg | -0.031 |
Token | make |
Token ID | 787 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
uncle | +0.767 |
oker | +0.752 |
ecast | +0.720 |
enegger | +0.712 |
Liberties | +0.711 |
Neg logit contributions | |
Malk | -0.854 |
Andreas | -0.658 |
Manson | -0.632 |
Goldberg | -0.623 |
twists | -0.622 |
Token | it |
Token ID | 340 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
uncle | +0.324 |
oker | +0.317 |
ecast | +0.305 |
enegger | +0.300 |
bnb | +0.298 |
Neg logit contributions | |
Malk | -0.373 |
Andreas | -0.289 |
Manson | -0.278 |
Goldberg | -0.273 |
twists | -0.273 |
Token | . |
Token ID | 13 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
Malk | +0.449 |
Andreas | +0.345 |
Manson | +0.332 |
Goldberg | +0.327 |
twists | +0.324 |
Neg logit contributions | |
uncle | -0.409 |
oker | -0.401 |
ecast | -0.384 |
Liberties | -0.379 |
enegger | -0.378 |
Token | Either |
Token ID | 15467 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
uncle | +0.341 |
oker | +0.332 |
ecast | +0.317 |
enegger | +0.310 |
bnb | +0.308 |
Neg logit contributions | |
Malk | -0.515 |
Andreas | -0.411 |
Manson | -0.396 |
Goldberg | -0.391 |
twists | -0.388 |
Token | way |
Token ID | 835 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
Malk | +3.790 |
Andreas | +2.880 |
Manson | +2.760 |
Goldberg | +2.718 |
twists | +2.702 |
Neg logit contributions | |
uncle | -3.740 |
oker | -3.669 |
ecast | -3.565 |
enegger | -3.558 |
Liberties | -3.493 |
Token | , |
Token ID | 11 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
uncle | +0.826 |
oker | +0.803 |
ecast | +0.764 |
Liberties | +0.752 |
enegger | +0.752 |
Neg logit contributions | |
Malk | -1.134 |
Andreas | -0.893 |
Manson | -0.854 |
Goldberg | -0.850 |
twists | -0.841 |
Token | my |
Token ID | 616 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
uncle | +0.075 |
oker | +0.073 |
ecast | +0.070 |
enegger | +0.069 |
Liberties | +0.069 |
Neg logit contributions | |
Malk | -0.091 |
Andreas | -0.071 |
Manson | -0.068 |
Goldberg | -0.067 |
twists | -0.067 |
Token | first |
Token ID | 717 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
uncle | +0.187 |
oker | +0.183 |
ecast | +0.177 |
enegger | +0.174 |
Liberties | +0.174 |
Neg logit contributions | |
Malk | -0.186 |
Andreas | -0.140 |
Manson | -0.134 |
Goldberg | -0.132 |
twists | -0.132 |
Token | three |
Token ID | 1115 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
uncle | +0.154 |
oker | +0.151 |
ecast | +0.146 |
enegger | +0.144 |
Liberties | +0.144 |
Neg logit contributions | |
Malk | -0.131 |
Andreas | -0.096 |
Manson | -0.091 |
Goldberg | -0.090 |
twists | -0.090 |
Token | her |
Token ID | 607 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
Malk | +0.148 |
Andreas | +0.115 |
Manson | +0.110 |
twists | +0.108 |
Goldberg | +0.108 |
Neg logit contributions | |
uncle | -0.122 |
oker | -0.119 |
ecast | -0.114 |
enegger | -0.113 |
Liberties | -0.112 |
Token | departure |
Token ID | 12928 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
Malk | +0.294 |
Andreas | +0.226 |
Manson | +0.217 |
twists | +0.214 |
Goldberg | +0.213 |
Neg logit contributions | |
uncle | -0.263 |
oker | -0.257 |
ecast | -0.247 |
enegger | -0.244 |
Liberties | -0.243 |
Token | time |
Token ID | 640 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
Malk | +0.939 |
Andreas | +0.719 |
Manson | +0.686 |
twists | +0.681 |
Goldberg | +0.677 |
Neg logit contributions | |
uncle | -0.862 |
oker | -0.844 |
ecast | -0.810 |
enegger | -0.808 |
Liberties | -0.794 |
Token | carefully |
Token ID | 7773 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
uncle | +0.355 |
oker | +0.347 |
ecast | +0.334 |
enegger | +0.329 |
Liberties | +0.328 |
Neg logit contributions | |
Malk | -0.366 |
Andreas | -0.279 |
Manson | -0.266 |
Goldberg | -0.262 |
twists | -0.262 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-03 |
Pos logit contributions | |
Malk | +0.451 |
Andreas | +0.360 |
Manson | +0.348 |
twists | +0.344 |
Goldberg | +0.342 |
Neg logit contributions | |
uncle | -0.306 |
oker | -0.298 |
ecast | -0.282 |
enegger | -0.280 |
Liberties | -0.279 |
Token | Either |
Token ID | 15467 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Malk | +0.005 |
Andreas | +0.004 |
Manson | +0.003 |
Goldberg | +0.003 |
twists | +0.003 |
Neg logit contributions | |
uncle | -0.003 |
oker | -0.003 |
ecast | -0.003 |
enegger | -0.003 |
bnb | -0.003 |
Token | she |
Token ID | 673 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
Malk | +3.862 |
Andreas | +2.939 |
Manson | +2.800 |
Goldberg | +2.779 |
twists | +2.767 |
Neg logit contributions | |
uncle | -3.729 |
oker | -3.644 |
enegger | -3.570 |
ecast | -3.544 |
覚醒 | -3.508 |
Token | gets |
Token ID | 3011 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
uncle | +0.264 |
oker | +0.258 |
ecast | +0.247 |
enegger | +0.244 |
Liberties | +0.243 |
Neg logit contributions | |
Malk | -0.298 |
Andreas | -0.230 |
Manson | -0.220 |
twists | -0.217 |
Goldberg | -0.217 |
Token | on |
Token ID | 319 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
uncle | +0.490 |
oker | +0.480 |
ecast | +0.459 |
enegger | +0.453 |
Liberties | +0.453 |
Neg logit contributions | |
Malk | -0.530 |
Andreas | -0.406 |
Manson | -0.388 |
twists | -0.382 |
Goldberg | -0.382 |
Token | the |
Token ID | 262 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
uncle | +0.601 |
oker | +0.588 |
ecast | +0.563 |
enegger | +0.557 |
Liberties | +0.554 |
Neg logit contributions | |
Malk | -0.674 |
Andreas | -0.519 |
Manson | -0.496 |
Goldberg | -0.489 |
twists | -0.487 |
Token | London |
Token ID | 3576 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
uncle | +0.468 |
oker | +0.458 |
ecast | +0.440 |
Liberties | +0.434 |
enegger | +0.433 |
Neg logit contributions | |
Malk | -0.490 |
Andreas | -0.373 |
Manson | -0.357 |
Goldberg | -0.351 |
twists | -0.350 |
Token | record |
Token ID | 1700 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
uncle | +0.178 |
oker | +0.174 |
ecast | +0.167 |
enegger | +0.164 |
Liberties | +0.163 |
Neg logit contributions | |
Malk | -0.225 |
Andreas | -0.177 |
Manson | -0.170 |
twists | -0.167 |
Goldberg | -0.167 |
Token | at |
Token ID | 379 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
uncle | +0.439 |
oker | +0.428 |
ecast | +0.410 |
enegger | +0.403 |
Liberties | +0.402 |
Neg logit contributions | |
Malk | -0.584 |
Andreas | -0.460 |
Manson | -0.442 |
twists | -0.436 |
Goldberg | -0.435 |
Token | the |
Token ID | 262 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
Malk | +0.303 |
Andreas | +0.235 |
Manson | +0.225 |
twists | +0.221 |
Goldberg | +0.221 |
Neg logit contributions | |
uncle | -0.264 |
oker | -0.256 |
ecast | -0.249 |
enegger | -0.245 |
bnb | -0.242 |
Token | Tour |
Token ID | 9852 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
uncle | +0.057 |
oker | +0.055 |
ecast | +0.054 |
bnb | +0.052 |
enegger | +0.052 |
Neg logit contributions | |
Malk | -0.065 |
Andreas | -0.051 |
Manson | -0.049 |
twists | -0.048 |
Goldberg | -0.048 |
Token | Down |
Token ID | 5588 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
Malk | +1.673 |
Andreas | +1.288 |
Manson | +1.220 |
twists | +1.206 |
Goldberg | +1.199 |
Neg logit contributions | |
uncle | -1.603 |
oker | -1.561 |
ecast | -1.521 |
enegger | -1.494 |
bnb | -1.486 |
Token | Under |
Token ID | 4698 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Malk | +0.955 |
Andreas | +0.765 |
Manson | +0.711 |
twists | +0.711 |
Goldberg | +0.708 |
Neg logit contributions | |
uncle | -0.867 |
oker | -0.838 |
ecast | -0.817 |
enegger | -0.807 |
bnb | -0.804 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
Malk | +4.265 |
Andreas | +3.427 |
Manson | +3.269 |
twists | +3.243 |
Goldberg | +3.177 |
Neg logit contributions | |
uncle | -3.203 |
oker | -3.075 |
ecast | -3.009 |
enegger | -2.935 |
bnb | -2.896 |
Token | He |
Token ID | 679 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
Malk | +0.374 |
Andreas | +0.291 |
Manson | +0.279 |
Goldberg | +0.274 |
twists | +0.273 |
Neg logit contributions | |
uncle | -0.310 |
oker | -0.303 |
ecast | -0.292 |
enegger | -0.286 |
Liberties | -0.285 |
Token | had |
Token ID | 550 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
uncle | +0.108 |
oker | +0.105 |
ecast | +0.101 |
Liberties | +0.100 |
enegger | +0.100 |
Neg logit contributions | |
Malk | -0.116 |
Andreas | -0.088 |
Manson | -0.085 |
twists | -0.083 |
Goldberg | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
uncle | +0.240 |
oker | +0.234 |
ecast | +0.225 |
enegger | +0.221 |
Liberties | +0.221 |
Neg logit contributions | |
Malk | -0.268 |
Andreas | -0.207 |
Manson | -0.197 |
twists | -0.195 |
Goldberg | -0.194 |
Token | good |
Token ID | 922 |
Feature activation | +2.44e-03 |
Pos logit contributions | |
uncle | +0.217 |
oker | +0.211 |
ecast | +0.204 |
enegger | +0.201 |
Liberties | +0.200 |
Neg logit contributions | |
Malk | -0.224 |
Andreas | -0.171 |
Manson | -0.163 |
twists | -0.161 |
Goldberg | -0.160 |
Token | June |
Token ID | 2795 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
Malk | +0.404 |
Andreas | +0.293 |
Manson | +0.276 |
twists | +0.270 |
Goldberg | +0.270 |
Neg logit contributions | |
uncle | -0.516 |
oker | -0.504 |
ecast | -0.490 |
enegger | -0.483 |
Liberties | -0.481 |
Token | 2007 |
Token ID | 4343 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Malk | +0.888 |
Andreas | +0.654 |
Manson | +0.619 |
Goldberg | +0.609 |
twists | +0.604 |
Neg logit contributions | |
uncle | -1.055 |
oker | -1.033 |
ecast | -1.003 |
enegger | -0.984 |
Liberties | -0.982 |
Token | , |
Token ID | 11 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
Malk | +0.550 |
Andreas | +0.428 |
Manson | +0.411 |
Goldberg | +0.405 |
twists | +0.404 |
Neg logit contributions | |
uncle | -0.449 |
oker | -0.439 |
ecast | -0.420 |
Liberties | -0.413 |
enegger | -0.413 |
Token | Singapore |
Token ID | 12551 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
uncle | +0.136 |
oker | +0.133 |
ecast | +0.127 |
Liberties | +0.125 |
enegger | +0.125 |
Neg logit contributions | |
Malk | -0.155 |
Andreas | -0.119 |
Manson | -0.114 |
Goldberg | -0.112 |
twists | -0.112 |
Token | and |
Token ID | 290 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
Malk | +1.889 |
Andreas | +1.479 |
Manson | +1.408 |
Goldberg | +1.402 |
twists | +1.386 |
Neg logit contributions | |
uncle | -1.542 |
oker | -1.496 |
ecast | -1.439 |
bnb | -1.420 |
enegger | -1.418 |
Token | Brune |
Token ID | 46070 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
uncle | +0.085 |
oker | +0.083 |
ecast | +0.080 |
enegger | +0.078 |
Liberties | +0.078 |
Neg logit contributions | |
Malk | -0.088 |
Andreas | -0.067 |
Manson | -0.064 |
Goldberg | -0.063 |
twists | -0.063 |
Token | i |
Token ID | 72 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
Malk | +3.346 |
Andreas | +2.465 |
Manson | +2.321 |
Goldberg | +2.306 |
twists | +2.264 |
Neg logit contributions | |
uncle | -4.169 |
oker | -4.070 |
enegger | -3.954 |
ecast | -3.883 |
bnb | -3.864 |
Token | celebrated |
Token ID | 13943 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
uncle | +0.357 |
oker | +0.349 |
ecast | +0.334 |
Liberties | +0.329 |
enegger | +0.328 |
Neg logit contributions | |
Malk | -0.425 |
Andreas | -0.329 |
Manson | -0.316 |
Goldberg | -0.311 |
twists | -0.311 |
Token | the |
Token ID | 262 |
Feature activation | -9.09e-02 |
Pos logit contributions | |
Malk | +1.134 |
Andreas | +0.879 |
Manson | +0.839 |
Goldberg | +0.830 |
twists | +0.824 |
Neg logit contributions | |
uncle | -0.981 |
oker | -0.959 |
ecast | -0.926 |
enegger | -0.905 |
bnb | -0.904 |
Token | 40 |
Token ID | 2319 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
uncle | +0.175 |
oker | +0.170 |
ecast | +0.163 |
enegger | +0.159 |
Liberties | +0.159 |
Neg logit contributions | |
Malk | -0.247 |
Andreas | -0.196 |
Manson | -0.188 |
Goldberg | -0.186 |
twists | -0.186 |
Token | th |
Token ID | 400 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
Malk | +0.335 |
Andreas | +0.256 |
Manson | +0.244 |
Goldberg | +0.241 |
twists | +0.241 |
Neg logit contributions | |
uncle | -0.311 |
oker | -0.305 |
ecast | -0.294 |
enegger | -0.289 |
Liberties | -0.288 |
Token | � |
Token ID | 447 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
uncle | +0.007 |
oker | +0.007 |
ecast | +0.007 |
enegger | +0.006 |
bnb | +0.006 |
Neg logit contributions | |
Malk | -0.012 |
Andreas | -0.009 |
Manson | -0.009 |
twists | -0.009 |
Goldberg | -0.009 |
Token | � |
Token ID | 247 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
uncle | +0.298 |
oker | +0.292 |
ecast | +0.282 |
Liberties | +0.278 |
enegger | +0.278 |
Neg logit contributions | |
Malk | -0.237 |
Andreas | -0.172 |
Manson | -0.162 |
twists | -0.159 |
Goldberg | -0.159 |
Token | t |
Token ID | 83 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
Malk | +0.048 |
Andreas | +0.037 |
Manson | +0.036 |
twists | +0.035 |
Goldberg | +0.035 |
Neg logit contributions | |
uncle | -0.041 |
oker | -0.041 |
ecast | -0.039 |
Liberties | -0.038 |
enegger | -0.038 |
Token | ideal |
Token ID | 7306 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
uncle | +0.477 |
oker | +0.468 |
ecast | +0.449 |
Liberties | +0.442 |
enegger | +0.442 |
Neg logit contributions | |
Malk | -0.496 |
Andreas | -0.377 |
Manson | -0.360 |
Goldberg | -0.355 |
twists | -0.354 |
Token | , |
Token ID | 11 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
Malk | +0.412 |
Andreas | +0.328 |
Manson | +0.316 |
twists | +0.311 |
Goldberg | +0.311 |
Neg logit contributions | |
uncle | -0.279 |
oker | -0.271 |
ecast | -0.259 |
Liberties | -0.254 |
enegger | -0.253 |
Token | either |
Token ID | 2035 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Malk | +0.385 |
Andreas | +0.294 |
Manson | +0.281 |
Goldberg | +0.276 |
twists | +0.276 |
Neg logit contributions | |
uncle | -0.364 |
oker | -0.357 |
ecast | -0.342 |
Liberties | -0.337 |
enegger | -0.337 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
Malk | +4.374 |
Andreas | +3.486 |
Manson | +3.367 |
twists | +3.313 |
Goldberg | +3.290 |
Neg logit contributions | |
uncle | -3.100 |
oker | -3.021 |
Liberties | -2.877 |
ecast | -2.876 |
enegger | -2.811 |
Token | ( |
Token ID | 357 |
Feature activation | +7.01e-02 |
Pos logit contributions | |
Malk | +0.331 |
Andreas | +0.263 |
Manson | +0.253 |
Goldberg | +0.250 |
twists | +0.249 |
Neg logit contributions | |
uncle | -0.227 |
oker | -0.221 |
ecast | -0.211 |
enegger | -0.206 |
bnb | -0.206 |
Token | Both |
Token ID | 10265 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
Malk | +0.173 |
Andreas | +0.133 |
Manson | +0.128 |
twists | +0.126 |
Goldberg | +0.125 |
Neg logit contributions | |
uncle | -0.152 |
oker | -0.149 |
ecast | -0.143 |
Liberties | -0.141 |
enegger | -0.141 |
Token | of |
Token ID | 286 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
Malk | +0.752 |
Andreas | +0.567 |
Manson | +0.542 |
Goldberg | +0.531 |
twists | +0.531 |
Neg logit contributions | |
uncle | -0.771 |
oker | -0.754 |
ecast | -0.727 |
enegger | -0.715 |
Liberties | -0.714 |
Token | which |
Token ID | 543 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
Malk | +0.131 |
Andreas | +0.098 |
Manson | +0.094 |
Goldberg | +0.092 |
twists | +0.092 |
Neg logit contributions | |
uncle | -0.138 |
oker | -0.135 |
ecast | -0.131 |
enegger | -0.129 |
Liberties | -0.128 |
Token | aire |
Token ID | 7626 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
Malk | +1.959 |
Andreas | +1.548 |
Manson | +1.468 |
Goldberg | +1.440 |
twists | +1.440 |
Neg logit contributions | |
uncle | -1.613 |
oker | -1.583 |
ecast | -1.514 |
Liberties | -1.496 |
enegger | -1.492 |
Token | ser |
Token ID | 1055 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
Malk | +0.752 |
Andreas | +0.586 |
Manson | +0.559 |
Goldberg | +0.549 |
twists | +0.547 |
Neg logit contributions | |
uncle | -0.631 |
oker | -0.617 |
ecast | -0.592 |
enegger | -0.582 |
Liberties | -0.580 |
Token | ait |
Token ID | 4548 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
uncle | +0.025 |
oker | +0.024 |
ecast | +0.023 |
Liberties | +0.023 |
enegger | +0.023 |
Neg logit contributions | |
Malk | -0.028 |
Andreas | -0.021 |
Manson | -0.020 |
twists | -0.020 |
Goldberg | -0.020 |
Token | ill |
Token ID | 2801 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Malk | +1.510 |
Andreas | +1.169 |
Manson | +1.120 |
Goldberg | +1.102 |
twists | +1.099 |
Neg logit contributions | |
uncle | -1.285 |
oker | -1.257 |
ecast | -1.205 |
enegger | -1.189 |
Liberties | -1.185 |
Token | é |
Token ID | 2634 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
Malk | +2.781 |
Andreas | +2.209 |
Manson | +2.125 |
Goldberg | +2.091 |
twists | +2.070 |
Neg logit contributions | |
uncle | -1.935 |
oker | -1.866 |
ecast | -1.817 |
Liberties | -1.790 |
enegger | -1.787 |
Token | git |
Token ID | 18300 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Malk | +1.001 |
Andreas | +0.779 |
Manson | +0.749 |
Goldberg | +0.733 |
twists | +0.732 |
Neg logit contributions | |
uncle | -0.820 |
oker | -0.802 |
ecast | -0.766 |
enegger | -0.759 |
Liberties | -0.757 |
Token | ime |
Token ID | 524 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Malk | +4.183 |
Andreas | +3.284 |
Manson | +3.143 |
Goldberg | +3.117 |
Mai | +3.077 |
Neg logit contributions | |
uncle | -3.177 |
oker | -3.097 |
ecast | -3.037 |
Liberties | -2.996 |
bnb | -2.969 |
Token | et |
Token ID | 2123 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
Malk | +0.453 |
Andreas | +0.354 |
Manson | +0.339 |
Goldberg | +0.334 |
twists | +0.333 |
Neg logit contributions | |
uncle | -0.366 |
oker | -0.357 |
ecast | -0.343 |
enegger | -0.337 |
Liberties | -0.336 |
Token | br |
Token ID | 865 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
uncle | +0.680 |
oker | +0.665 |
ecast | +0.637 |
Liberties | +0.627 |
enegger | +0.627 |
Neg logit contributions | |
Malk | -0.754 |
Andreas | -0.579 |
Manson | -0.554 |
Goldberg | -0.545 |
twists | -0.545 |
Token | ider |
Token ID | 1304 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
Malk | +2.660 |
Andreas | +2.089 |
Manson | +1.985 |
Goldberg | +1.956 |
OTH | +1.950 |
Neg logit contributions | |
uncle | -2.043 |
oker | -2.022 |
Liberties | -1.968 |
enegger | -1.950 |
ecast | -1.946 |
Token | ait |
Token ID | 4548 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Malk | +0.320 |
Andreas | +0.248 |
Manson | +0.237 |
Goldberg | +0.234 |
twists | +0.233 |
Neg logit contributions | |
uncle | -0.273 |
oker | -0.266 |
ecast | -0.257 |
Liberties | -0.252 |
enegger | -0.251 |
Token | oil |
Token ID | 3056 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
uncle | +0.816 |
oker | +0.793 |
ecast | +0.759 |
enegger | +0.748 |
Liberties | +0.744 |
Neg logit contributions | |
Malk | -1.148 |
Andreas | -0.907 |
Manson | -0.871 |
Goldberg | -0.861 |
twists | -0.859 |
Token | firm |
Token ID | 4081 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Malk | +0.945 |
Andreas | +0.744 |
Manson | +0.711 |
twists | +0.708 |
Goldberg | +0.706 |
Neg logit contributions | |
uncle | -0.669 |
oker | -0.648 |
ecast | -0.630 |
enegger | -0.622 |
Liberties | -0.618 |
Token | Uk |
Token ID | 5065 |
Feature activation | -8.30e-02 |
Pos logit contributions | |
uncle | +2.200 |
oker | +2.152 |
ecast | +2.049 |
Liberties | +2.025 |
enegger | +2.010 |
Neg logit contributions | |
Malk | -2.735 |
Andreas | -2.139 |
Manson | -2.061 |
twists | -2.030 |
Goldberg | -2.024 |
Token | r |
Token ID | 81 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
uncle | +0.170 |
oker | +0.165 |
ecast | +0.159 |
Liberties | +0.156 |
enegger | +0.156 |
Neg logit contributions | |
Malk | -0.215 |
Andreas | -0.168 |
Manson | -0.160 |
Goldberg | -0.159 |
twists | -0.158 |
Token | na |
Token ID | 2616 |
Feature activation | +4.57e-03 |
Pos logit contributions | |
Malk | +0.608 |
Andreas | +0.471 |
Manson | +0.449 |
Goldberg | +0.446 |
twists | +0.442 |
Neg logit contributions | |
uncle | -0.483 |
oker | -0.469 |
ecast | -0.453 |
enegger | -0.448 |
Liberties | -0.444 |
Token | ft |
Token ID | 701 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Malk | +0.011 |
Andreas | +0.009 |
Manson | +0.008 |
Goldberg | +0.008 |
twists | +0.008 |
Neg logit contributions | |
uncle | -0.010 |
oker | -0.010 |
ecast | -0.009 |
enegger | -0.009 |
Liberties | -0.009 |
Token | a |
Token ID | 64 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Malk | +4.198 |
Andreas | +3.251 |
Goldberg | +3.100 |
Manson | +3.074 |
twists | +3.044 |
Neg logit contributions | |
uncle | -3.277 |
oker | -3.133 |
ecast | -3.037 |
enegger | -3.032 |
Liberties | -2.979 |
Token | in |
Token ID | 287 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
Malk | +0.335 |
Andreas | +0.264 |
Manson | +0.252 |
Goldberg | +0.250 |
twists | +0.248 |
Neg logit contributions | |
uncle | -0.236 |
oker | -0.228 |
ecast | -0.219 |
enegger | -0.215 |
Liberties | -0.214 |
Token | the |
Token ID | 262 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
uncle | +0.043 |
oker | +0.042 |
ecast | +0.040 |
enegger | +0.040 |
bnb | +0.039 |
Neg logit contributions | |
Malk | -0.051 |
Andreas | -0.039 |
Manson | -0.037 |
Goldberg | -0.037 |
twists | -0.037 |
Token | Kh |
Token ID | 5311 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Malk | +0.047 |
Andreas | +0.036 |
Manson | +0.035 |
Goldberg | +0.034 |
twists | +0.034 |
Neg logit contributions | |
uncle | -0.040 |
oker | -0.039 |
ecast | -0.038 |
enegger | -0.037 |
bnb | -0.037 |
Token | mel |
Token ID | 17694 |
Feature activation | +1.93e-03 |
Pos logit contributions | |
Malk | +0.795 |
Andreas | +0.629 |
Manson | +0.602 |
Goldberg | +0.593 |
twists | +0.589 |
Neg logit contributions | |
uncle | -0.516 |
oker | -0.495 |
ecast | -0.479 |
enegger | -0.477 |
Liberties | -0.473 |
Token | GO |
Token ID | 10351 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Malk | +1.292 |
Andreas | +1.009 |
Manson | +0.961 |
Goldberg | +0.950 |
PRESS | +0.947 |
Neg logit contributions | |
uncle | -1.025 |
oker | -1.002 |
ecast | -0.974 |
Liberties | -0.955 |
enegger | -0.953 |
Token | AL |
Token ID | 1847 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
uncle | +0.025 |
oker | +0.024 |
ecast | +0.023 |
Liberties | +0.023 |
enegger | +0.023 |
Neg logit contributions | |
Malk | -0.027 |
Andreas | -0.021 |
Manson | -0.020 |
Goldberg | -0.020 |
PRESS | -0.020 |
Token | ! |
Token ID | 0 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
uncle | +0.274 |
oker | +0.267 |
ecast | +0.258 |
Liberties | +0.253 |
enegger | +0.253 |
Neg logit contributions | |
Malk | -0.303 |
Andreas | -0.233 |
Manson | -0.223 |
Goldberg | -0.219 |
twists | -0.218 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Malk | +1.466 |
Andreas | +1.158 |
Manson | +1.105 |
Goldberg | +1.091 |
twists | +1.083 |
Neg logit contributions | |
uncle | -1.087 |
oker | -1.054 |
ecast | -1.018 |
enegger | -0.997 |
Liberties | -0.995 |
Token | . |
Token ID | 13 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
uncle | +1.897 |
oker | +1.844 |
ecast | +1.739 |
Liberties | +1.706 |
enegger | +1.701 |
Neg logit contributions | |
Malk | -3.379 |
Andreas | -2.734 |
Manson | -2.643 |
Goldberg | -2.612 |
twists | -2.610 |
Token | twitter |
Token ID | 6956 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
Malk | +0.202 |
Andreas | +0.168 |
Manson | +0.162 |
twists | +0.161 |
Goldberg | +0.160 |
Neg logit contributions | |
uncle | -0.078 |
oker | -0.075 |
enegger | -0.071 |
ecast | -0.070 |
Liberties | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +7.62e-02 |
Pos logit contributions | |
uncle | +3.119 |
oker | +3.020 |
Liberties | +2.783 |
ecast | +2.772 |
bnb | +2.711 |
Neg logit contributions | |
Malk | -6.450 |
Andreas | -5.229 |
Manson | -5.089 |
twists | -5.065 |
Goldberg | -5.056 |
Token | com |
Token ID | 785 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
Malk | +0.173 |
Andreas | +0.131 |
Manson | +0.124 |
Goldberg | +0.122 |
twists | +0.121 |
Neg logit contributions | |
uncle | -0.179 |
oker | -0.176 |
enegger | -0.169 |
ecast | -0.169 |
Liberties | -0.166 |
Token | / |
Token ID | 14 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
uncle | +1.607 |
oker | +1.562 |
ecast | +1.470 |
Liberties | +1.449 |
enegger | +1.433 |
Neg logit contributions | |
Malk | -2.923 |
Andreas | -2.367 |
Manson | -2.291 |
Goldberg | -2.264 |
twists | -2.263 |
Token | E |
Token ID | 36 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
Malk | +0.702 |
Andreas | +0.548 |
Manson | +0.524 |
twists | +0.517 |
Goldberg | +0.516 |
Neg logit contributions | |
uncle | -0.576 |
oker | -0.562 |
ecast | -0.540 |
enegger | -0.537 |
Liberties | -0.536 |
Token | q |
Token ID | 80 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Malk | +0.897 |
Andreas | +0.699 |
Manson | +0.664 |
twists | +0.656 |
Goldberg | +0.654 |
Neg logit contributions | |
uncle | -0.773 |
oker | -0.751 |
Liberties | -0.722 |
ecast | -0.722 |
enegger | -0.722 |
Token | fin |
Token ID | 957 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
uncle | +0.846 |
oker | +0.829 |
ecast | +0.801 |
enegger | +0.790 |
Liberties | +0.789 |
Neg logit contributions | |
Malk | -0.746 |
Andreas | -0.554 |
Manson | -0.525 |
Goldberg | -0.516 |
twists | -0.516 |
Token | isher |
Token ID | 4828 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
Malk | +0.159 |
Andreas | +0.120 |
Manson | +0.114 |
Goldberg | +0.110 |
twists | +0.109 |
Neg logit contributions | |
uncle | -0.185 |
oker | -0.178 |
Liberties | -0.177 |
enegger | -0.177 |
ecast | -0.176 |
Token | " |
Token ID | 1 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
Malk | +0.838 |
Andreas | +0.665 |
Manson | +0.637 |
Goldberg | +0.630 |
twists | +0.628 |
Neg logit contributions | |
uncle | -0.616 |
oker | -0.597 |
ecast | -0.574 |
enegger | -0.564 |
Liberties | -0.562 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
uncle | +0.673 |
oker | +0.655 |
ecast | +0.619 |
Liberties | +0.607 |
enegger | +0.607 |
Neg logit contributions | |
Malk | -1.142 |
Andreas | -0.920 |
Manson | -0.889 |
twists | -0.877 |
Goldberg | -0.877 |
Token | . |
Token ID | 13 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
uncle | +2.003 |
oker | +1.949 |
ecast | +1.828 |
Liberties | +1.801 |
enegger | +1.786 |
Neg logit contributions | |
Malk | -3.663 |
Andreas | -2.967 |
Manson | -2.873 |
twists | -2.843 |
Goldberg | -2.840 |
Token | twitter |
Token ID | 6956 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
uncle | +0.251 |
oker | +0.242 |
enegger | +0.227 |
ecast | +0.224 |
bnb | +0.218 |
Neg logit contributions | |
Malk | -0.651 |
Andreas | -0.545 |
Manson | -0.526 |
twists | -0.521 |
Goldberg | -0.519 |
Token | . |
Token ID | 13 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
uncle | +3.132 |
oker | +3.031 |
ecast | +2.807 |
Liberties | +2.796 |
bnb | +2.726 |
Neg logit contributions | |
Malk | -6.357 |
Andreas | -5.154 |
Manson | -5.019 |
twists | -4.991 |
Goldberg | -4.970 |
Token | com |
Token ID | 785 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
uncle | +0.235 |
oker | +0.231 |
enegger | +0.222 |
ecast | +0.222 |
bnb | +0.218 |
Neg logit contributions | |
Malk | -0.228 |
Andreas | -0.173 |
Manson | -0.164 |
Goldberg | -0.162 |
twists | -0.160 |
Token | / |
Token ID | 14 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
uncle | +1.636 |
oker | +1.589 |
ecast | +1.499 |
Liberties | +1.476 |
enegger | +1.460 |
Neg logit contributions | |
Malk | -2.967 |
Andreas | -2.402 |
Manson | -2.326 |
twists | -2.297 |
Goldberg | -2.296 |
Token | Eh |
Token ID | 43894 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
Malk | +0.409 |
Andreas | +0.319 |
Manson | +0.305 |
twists | +0.301 |
Goldberg | +0.300 |
Neg logit contributions | |
uncle | -0.336 |
oker | -0.328 |
ecast | -0.315 |
enegger | -0.313 |
Liberties | -0.313 |
Token | U |
Token ID | 52 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
Malk | +1.784 |
Andreas | +1.391 |
Manson | +1.326 |
twists | +1.308 |
Goldberg | +1.307 |
Neg logit contributions | |
uncle | -1.476 |
oker | -1.430 |
enegger | -1.384 |
ecast | -1.383 |
Liberties | -1.383 |
Token | ️ |
Token ID | 37929 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
Malk | +0.704 |
Andreas | +0.546 |
Manson | +0.524 |
Goldberg | +0.514 |
twists | +0.514 |
Neg logit contributions | |
uncle | -0.579 |
oker | -0.567 |
ecast | -0.543 |
enegger | -0.538 |
Liberties | -0.535 |
Token | Be |
Token ID | 1355 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
Malk | +0.542 |
Andreas | +0.423 |
Manson | +0.407 |
Goldberg | +0.399 |
twists | +0.398 |
Neg logit contributions | |
uncle | -0.431 |
oker | -0.422 |
ecast | -0.404 |
enegger | -0.397 |
Liberties | -0.396 |
Token | safe |
Token ID | 3338 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
Malk | +0.088 |
Andreas | +0.067 |
Manson | +0.064 |
Goldberg | +0.063 |
twists | +0.063 |
Neg logit contributions | |
uncle | -0.081 |
oker | -0.079 |
ecast | -0.076 |
enegger | -0.075 |
Liberties | -0.075 |
Token | pic |
Token ID | 8301 |
Feature activation | -9.97e-01 |
Pos logit contributions | |
uncle | +0.312 |
oker | +0.305 |
ecast | +0.291 |
enegger | +0.286 |
Liberties | +0.286 |
Neg logit contributions | |
Malk | -0.406 |
Andreas | -0.318 |
Manson | -0.306 |
Goldberg | -0.301 |
twists | -0.301 |
Token | . |
Token ID | 13 |
Feature activation | +7.55e-03 |
Pos logit contributions | |
uncle | +1.672 |
oker | +1.624 |
ecast | +1.535 |
Liberties | +1.504 |
enegger | +1.502 |
Neg logit contributions | |
Malk | -2.950 |
Andreas | -2.386 |
Manson | -2.306 |
Goldberg | -2.278 |
twists | -2.276 |
Token | twitter |
Token ID | 6956 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
Malk | +0.025 |
Andreas | +0.021 |
Manson | +0.020 |
twists | +0.020 |
Goldberg | +0.020 |
Neg logit contributions | |
uncle | -0.010 |
oker | -0.009 |
enegger | -0.009 |
ecast | -0.009 |
Liberties | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
uncle | +3.103 |
oker | +3.006 |
Liberties | +2.781 |
ecast | +2.770 |
bnb | +2.702 |
Neg logit contributions | |
Malk | -6.298 |
Andreas | -5.102 |
Manson | -4.957 |
twists | -4.939 |
Goldberg | -4.921 |
Token | com |
Token ID | 785 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
Malk | +0.081 |
Andreas | +0.061 |
Manson | +0.058 |
Goldberg | +0.057 |
twists | +0.056 |
Neg logit contributions | |
uncle | -0.085 |
oker | -0.083 |
enegger | -0.080 |
ecast | -0.080 |
Liberties | -0.079 |
Token | / |
Token ID | 14 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
uncle | +1.501 |
oker | +1.458 |
ecast | +1.377 |
Liberties | +1.352 |
enegger | +1.346 |
Neg logit contributions | |
Malk | -2.667 |
Andreas | -2.157 |
Manson | -2.086 |
Goldberg | -2.060 |
twists | -2.059 |
Token | 74 |
Token ID | 4524 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
Malk | +0.707 |
Andreas | +0.552 |
Manson | +0.528 |
twists | +0.521 |
Goldberg | +0.519 |
Neg logit contributions | |
uncle | -0.580 |
oker | -0.566 |
ecast | -0.544 |
enegger | -0.541 |
Liberties | -0.540 |
Token | E |
Token ID | 36 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
Malk | +0.636 |
Andreas | +0.498 |
Manson | +0.476 |
twists | +0.469 |
Goldberg | +0.468 |
Neg logit contributions | |
uncle | -0.511 |
oker | -0.498 |
ecast | -0.478 |
enegger | -0.477 |
Liberties | -0.476 |
Token | Deb |
Token ID | 16587 |
Feature activation | -6.45e-02 |
Pos logit contributions | |
uncle | +0.652 |
oker | +0.636 |
ecast | +0.604 |
Liberties | +0.593 |
enegger | +0.591 |
Neg logit contributions | |
Malk | -1.025 |
Andreas | -0.819 |
Manson | -0.790 |
Goldberg | -0.780 |
twists | -0.779 |
Token | ate |
Token ID | 378 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
uncle | +0.094 |
oker | +0.090 |
enegger | +0.086 |
ecast | +0.085 |
Liberties | +0.084 |
Neg logit contributions | |
Malk | -0.203 |
Andreas | -0.168 |
Manson | -0.161 |
Goldberg | -0.160 |
twists | -0.160 |
Token | 2016 |
Token ID | 5304 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
uncle | +0.082 |
oker | +0.081 |
ecast | +0.078 |
Liberties | +0.076 |
enegger | +0.076 |
Neg logit contributions | |
Malk | -0.083 |
Andreas | -0.063 |
Manson | -0.060 |
Goldberg | -0.059 |
twists | -0.059 |
Token | pic |
Token ID | 8301 |
Feature activation | -9.21e-01 |
Pos logit contributions | |
uncle | +0.558 |
oker | +0.543 |
ecast | +0.518 |
Liberties | +0.509 |
enegger | +0.509 |
Neg logit contributions | |
Malk | -0.786 |
Andreas | -0.623 |
Manson | -0.599 |
twists | -0.591 |
Goldberg | -0.591 |
Token | . |
Token ID | 13 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
uncle | +1.541 |
oker | +1.497 |
ecast | +1.414 |
Liberties | +1.386 |
enegger | +1.383 |
Neg logit contributions | |
Malk | -2.729 |
Andreas | -2.207 |
Manson | -2.134 |
Goldberg | -2.107 |
twists | -2.106 |
Token | twitter |
Token ID | 6956 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
uncle | +0.050 |
oker | +0.048 |
enegger | +0.045 |
ecast | +0.045 |
Liberties | +0.044 |
Neg logit contributions | |
Malk | -0.133 |
Andreas | -0.112 |
Manson | -0.108 |
twists | -0.107 |
Goldberg | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
uncle | +3.076 |
oker | +2.974 |
Liberties | +2.753 |
ecast | +2.750 |
bnb | +2.678 |
Neg logit contributions | |
Malk | -6.306 |
Andreas | -5.115 |
Manson | -4.984 |
twists | -4.956 |
Goldberg | -4.930 |
Token | com |
Token ID | 785 |
Feature activation | -9.31e-01 |
Pos logit contributions | |
uncle | +0.104 |
oker | +0.102 |
ecast | +0.098 |
enegger | +0.098 |
Liberties | +0.097 |
Neg logit contributions | |
Malk | -0.103 |
Andreas | -0.078 |
Manson | -0.074 |
Goldberg | -0.073 |
twists | -0.072 |
Token | / |
Token ID | 14 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
uncle | +1.546 |
oker | +1.503 |
ecast | +1.418 |
Liberties | +1.394 |
enegger | +1.384 |
Neg logit contributions | |
Malk | -2.764 |
Andreas | -2.236 |
Manson | -2.163 |
Goldberg | -2.136 |
twists | -2.135 |
Token | F |
Token ID | 37 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
Malk | +0.375 |
Andreas | +0.293 |
Manson | +0.280 |
twists | +0.276 |
Goldberg | +0.276 |
Neg logit contributions | |
uncle | -0.309 |
oker | -0.301 |
ecast | -0.289 |
enegger | -0.287 |
Liberties | -0.287 |
Token | c |
Token ID | 66 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
Malk | +1.465 |
Andreas | +1.139 |
Manson | +1.083 |
twists | +1.070 |
Goldberg | +1.065 |
Neg logit contributions | |
uncle | -1.279 |
oker | -1.243 |
Liberties | -1.201 |
enegger | -1.200 |
ecast | -1.195 |
Token | ju |
Token ID | 7544 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
Malk | +0.588 |
Andreas | +0.454 |
Manson | +0.433 |
twists | +0.427 |
Goldberg | +0.425 |
Neg logit contributions | |
uncle | -0.522 |
oker | -0.509 |
ecast | -0.489 |
enegger | -0.481 |
Liberties | -0.481 |
Token | illet |
Token ID | 32512 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
Malk | +2.089 |
Andreas | +1.678 |
Manson | +1.608 |
twists | +1.595 |
Goldberg | +1.591 |
Neg logit contributions | |
uncle | -1.388 |
oker | -1.322 |
ecast | -1.278 |
enegger | -1.253 |
Liberties | -1.252 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
uncle | +0.387 |
oker | +0.378 |
ecast | +0.362 |
Liberties | +0.356 |
enegger | +0.356 |
Neg logit contributions | |
Malk | -0.445 |
Andreas | -0.344 |
Manson | -0.329 |
Goldberg | -0.324 |
twists | -0.324 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Malk | +0.655 |
Andreas | +0.517 |
Manson | +0.495 |
Goldberg | +0.485 |
twists | +0.484 |
Neg logit contributions | |
uncle | -0.492 |
oker | -0.480 |
ecast | -0.459 |
enegger | -0.450 |
Liberties | -0.448 |
Token | . |
Token ID | 13 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
uncle | +1.978 |
oker | +1.922 |
ecast | +1.818 |
Liberties | +1.780 |
enegger | +1.779 |
Neg logit contributions | |
Malk | -3.484 |
Andreas | -2.817 |
Manson | -2.723 |
Goldberg | -2.689 |
twists | -2.686 |
Token | twitter |
Token ID | 6956 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
uncle | +0.068 |
oker | +0.066 |
enegger | +0.062 |
ecast | +0.060 |
覚醒 | +0.060 |
Neg logit contributions | |
Malk | -0.170 |
Andreas | -0.142 |
Manson | -0.137 |
twists | -0.136 |
PRESS | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
uncle | +3.113 |
oker | +3.015 |
ecast | +2.790 |
Liberties | +2.786 |
bnb | +2.716 |
Neg logit contributions | |
Malk | -6.261 |
Andreas | -5.062 |
Manson | -4.930 |
twists | -4.908 |
Goldberg | -4.894 |
Token | com |
Token ID | 785 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
uncle | +0.071 |
oker | +0.069 |
enegger | +0.067 |
ecast | +0.066 |
Liberties | +0.065 |
Neg logit contributions | |
Malk | -0.066 |
Andreas | -0.050 |
Manson | -0.048 |
Goldberg | -0.047 |
twists | -0.046 |
Token | / |
Token ID | 14 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
uncle | +1.481 |
oker | +1.438 |
ecast | +1.359 |
Liberties | +1.334 |
enegger | +1.327 |
Neg logit contributions | |
Malk | -2.632 |
Andreas | -2.129 |
Manson | -2.059 |
Goldberg | -2.033 |
twists | -2.033 |
Token | u |
Token ID | 84 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
Malk | +0.634 |
Andreas | +0.496 |
Manson | +0.473 |
twists | +0.467 |
Goldberg | +0.466 |
Neg logit contributions | |
uncle | -0.522 |
oker | -0.509 |
ecast | -0.489 |
enegger | -0.486 |
Liberties | -0.485 |
Token | I |
Token ID | 40 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
Malk | +1.114 |
Andreas | +0.869 |
Manson | +0.828 |
twists | +0.818 |
Goldberg | +0.817 |
Neg logit contributions | |
uncle | -0.920 |
oker | -0.891 |
ecast | -0.858 |
Liberties | -0.858 |
enegger | -0.857 |
Token | 8 |
Token ID | 23 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
Malk | +0.539 |
Andreas | +0.420 |
Manson | +0.402 |
twists | +0.395 |
Goldberg | +0.395 |
Neg logit contributions | |
uncle | -0.446 |
oker | -0.434 |
ecast | -0.417 |
Liberties | -0.416 |
enegger | -0.415 |
Token | b |
Token ID | 65 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
uncle | +0.463 |
oker | +0.451 |
ecast | +0.433 |
enegger | +0.428 |
Liberties | +0.428 |
Neg logit contributions | |
Malk | -0.574 |
Andreas | -0.448 |
Manson | -0.430 |
Goldberg | -0.423 |
twists | -0.423 |
Token | ET |
Token ID | 2767 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
Malk | +1.123 |
Andreas | +0.875 |
Manson | +0.839 |
twists | +0.825 |
Goldberg | +0.822 |
Neg logit contributions | |
uncle | -0.930 |
oker | -0.905 |
ecast | -0.872 |
Liberties | -0.869 |
enegger | -0.868 |
Token | pic |
Token ID | 8301 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
uncle | +0.473 |
oker | +0.461 |
ecast | +0.444 |
Liberties | +0.437 |
enegger | +0.436 |
Neg logit contributions | |
Malk | -0.535 |
Andreas | -0.413 |
Manson | -0.395 |
Goldberg | -0.388 |
twists | -0.387 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
uncle | +1.379 |
oker | +1.339 |
ecast | +1.271 |
enegger | +1.245 |
Liberties | +1.241 |
Neg logit contributions | |
Malk | -2.381 |
Andreas | -1.924 |
Manson | -1.858 |
Goldberg | -1.834 |
twists | -1.830 |
Token | twitter |
Token ID | 6956 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
uncle | +0.187 |
oker | +0.180 |
enegger | +0.171 |
ecast | +0.167 |
覚醒 | +0.164 |
Neg logit contributions | |
Malk | -0.478 |
Andreas | -0.401 |
Manson | -0.386 |
twists | -0.383 |
Goldberg | -0.381 |
Token | . |
Token ID | 13 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
uncle | +3.075 |
oker | +2.974 |
ecast | +2.751 |
Liberties | +2.745 |
bnb | +2.681 |
Neg logit contributions | |
Malk | -6.251 |
Andreas | -5.070 |
Manson | -4.933 |
twists | -4.905 |
Goldberg | -4.888 |
Token | com |
Token ID | 785 |
Feature activation | -8.40e-01 |
Pos logit contributions | |
uncle | +0.069 |
oker | +0.068 |
enegger | +0.065 |
ecast | +0.065 |
Liberties | +0.064 |
Neg logit contributions | |
Malk | -0.066 |
Andreas | -0.050 |
Manson | -0.047 |
Goldberg | -0.046 |
twists | -0.046 |
Token | / |
Token ID | 14 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
uncle | +1.404 |
oker | +1.364 |
ecast | +1.289 |
Liberties | +1.265 |
enegger | +1.260 |
Neg logit contributions | |
Malk | -2.488 |
Andreas | -2.012 |
Manson | -1.946 |
Goldberg | -1.922 |
twists | -1.920 |
Token | W |
Token ID | 54 |
Feature activation | -8.32e-02 |
Pos logit contributions | |
Malk | +0.670 |
Andreas | +0.524 |
Manson | +0.500 |
twists | +0.494 |
Goldberg | +0.492 |
Neg logit contributions | |
uncle | -0.549 |
oker | -0.535 |
ecast | -0.513 |
enegger | -0.511 |
Liberties | -0.510 |
Token | 8 |
Token ID | 23 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
uncle | +0.179 |
oker | +0.174 |
ecast | +0.167 |
Liberties | +0.166 |
enegger | +0.166 |
Neg logit contributions | |
Malk | -0.206 |
Andreas | -0.159 |
Manson | -0.152 |
twists | -0.150 |
Goldberg | -0.150 |
Token | Deb |
Token ID | 16587 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
uncle | +0.566 |
oker | +0.552 |
ecast | +0.525 |
Liberties | +0.517 |
enegger | +0.515 |
Neg logit contributions | |
Malk | -0.879 |
Andreas | -0.701 |
Manson | -0.676 |
Goldberg | -0.667 |
twists | -0.667 |
Token | ates |
Token ID | 689 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
uncle | +0.259 |
oker | +0.250 |
enegger | +0.238 |
ecast | +0.236 |
Liberties | +0.234 |
Neg logit contributions | |
Malk | -0.523 |
Andreas | -0.431 |
Manson | -0.413 |
Goldberg | -0.409 |
twists | -0.409 |
Token | 2016 |
Token ID | 5304 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
Malk | +0.381 |
Andreas | +0.299 |
Manson | +0.285 |
Goldberg | +0.284 |
twists | +0.283 |
Neg logit contributions | |
uncle | -0.296 |
oker | -0.287 |
ecast | -0.276 |
enegger | -0.272 |
Liberties | -0.271 |
Token | pic |
Token ID | 8301 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
uncle | +0.522 |
oker | +0.508 |
ecast | +0.486 |
enegger | +0.478 |
Liberties | +0.477 |
Neg logit contributions | |
Malk | -0.717 |
Andreas | -0.566 |
Manson | -0.543 |
twists | -0.537 |
Goldberg | -0.537 |
Token | . |
Token ID | 13 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
uncle | +1.351 |
oker | +1.313 |
ecast | +1.242 |
enegger | +1.216 |
Liberties | +1.216 |
Neg logit contributions | |
Malk | -2.370 |
Andreas | -1.917 |
Manson | -1.852 |
Goldberg | -1.829 |
twists | -1.827 |
Token | twitter |
Token ID | 6956 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
uncle | +0.042 |
oker | +0.040 |
enegger | +0.038 |
ecast | +0.037 |
Liberties | +0.036 |
Neg logit contributions | |
Malk | -0.109 |
Andreas | -0.092 |
Manson | -0.088 |
twists | -0.088 |
Goldberg | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
uncle | +3.028 |
oker | +2.928 |
Liberties | +2.711 |
ecast | +2.710 |
bnb | +2.635 |
Neg logit contributions | |
Malk | -6.183 |
Andreas | -5.017 |
Manson | -4.887 |
twists | -4.863 |
Goldberg | -4.834 |
Token | com |
Token ID | 785 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
Malk | +0.086 |
Andreas | +0.065 |
Manson | +0.062 |
Goldberg | +0.061 |
twists | +0.060 |
Neg logit contributions | |
uncle | -0.089 |
oker | -0.087 |
enegger | -0.084 |
ecast | -0.084 |
Liberties | -0.082 |
Token | / |
Token ID | 14 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
uncle | +1.548 |
oker | +1.504 |
ecast | +1.420 |
Liberties | +1.395 |
enegger | +1.386 |
Neg logit contributions | |
Malk | -2.759 |
Andreas | -2.232 |
Manson | -2.159 |
Goldberg | -2.131 |
twists | -2.131 |
Token | j |
Token ID | 73 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
Malk | +0.694 |
Andreas | +0.542 |
Manson | +0.518 |
twists | +0.511 |
Goldberg | +0.510 |
Neg logit contributions | |
uncle | -0.572 |
oker | -0.557 |
ecast | -0.535 |
enegger | -0.533 |
Liberties | -0.532 |
Token | Ax |
Token ID | 31554 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Malk | +1.058 |
Andreas | +0.825 |
Manson | +0.789 |
twists | +0.778 |
Goldberg | +0.776 |
Neg logit contributions | |
uncle | -0.875 |
oker | -0.850 |
ecast | -0.819 |
Liberties | -0.816 |
enegger | -0.815 |
Token | media |
Token ID | 2056 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
uncle | +0.174 |
oker | +0.170 |
ecast | +0.163 |
enegger | +0.161 |
Liberties | +0.160 |
Neg logit contributions | |
Malk | -0.186 |
Andreas | -0.142 |
Manson | -0.136 |
Goldberg | -0.134 |
twists | -0.134 |
Token | bus |
Token ID | 1323 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Malk | +1.454 |
Andreas | +1.093 |
Manson | +1.042 |
twists | +1.040 |
Goldberg | +1.028 |
Neg logit contributions | |
uncle | -1.402 |
oker | -1.360 |
ecast | -1.311 |
enegger | -1.300 |
Liberties | -1.293 |
Token | ) |
Token ID | 8 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
uncle | +0.592 |
oker | +0.577 |
ecast | +0.553 |
enegger | +0.542 |
Liberties | +0.542 |
Neg logit contributions | |
Malk | -0.794 |
Andreas | -0.625 |
Manson | -0.600 |
twists | -0.593 |
Goldberg | -0.591 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
Malk | +0.505 |
Andreas | +0.399 |
Manson | +0.383 |
twists | +0.378 |
Goldberg | +0.378 |
Neg logit contributions | |
uncle | -0.363 |
oker | -0.352 |
ecast | -0.339 |
enegger | -0.331 |
bnb | -0.330 |
Token | . |
Token ID | 13 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
uncle | +1.731 |
oker | +1.683 |
ecast | +1.588 |
Liberties | +1.558 |
enegger | +1.553 |
Neg logit contributions | |
Malk | -3.077 |
Andreas | -2.489 |
Manson | -2.407 |
Goldberg | -2.378 |
twists | -2.377 |
Token | twitter |
Token ID | 6956 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
uncle | +0.051 |
oker | +0.049 |
enegger | +0.046 |
ecast | +0.045 |
覚醒 | +0.044 |
Neg logit contributions | |
Malk | -0.129 |
Andreas | -0.107 |
Manson | -0.104 |
twists | -0.102 |
Goldberg | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
uncle | +3.038 |
oker | +2.942 |
ecast | +2.722 |
Liberties | +2.719 |
bnb | +2.654 |
Neg logit contributions | |
Malk | -6.133 |
Andreas | -4.969 |
Manson | -4.835 |
twists | -4.813 |
Goldberg | -4.794 |
Token | com |
Token ID | 785 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
Malk | +0.050 |
Andreas | +0.038 |
Manson | +0.036 |
Goldberg | +0.036 |
twists | +0.035 |
Neg logit contributions | |
uncle | -0.053 |
oker | -0.052 |
enegger | -0.050 |
ecast | -0.050 |
bnb | -0.049 |
Token | / |
Token ID | 14 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
uncle | +1.340 |
oker | +1.302 |
ecast | +1.231 |
Liberties | +1.207 |
enegger | +1.204 |
Neg logit contributions | |
Malk | -2.363 |
Andreas | -1.911 |
Manson | -1.847 |
Goldberg | -1.824 |
twists | -1.823 |
Token | z |
Token ID | 89 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
Malk | +0.851 |
Andreas | +0.664 |
Manson | +0.634 |
twists | +0.626 |
Goldberg | +0.624 |
Neg logit contributions | |
uncle | -0.700 |
oker | -0.683 |
ecast | -0.656 |
enegger | -0.653 |
Liberties | -0.653 |
Token | J |
Token ID | 41 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
Malk | +0.830 |
Andreas | +0.647 |
Manson | +0.618 |
twists | +0.609 |
Goldberg | +0.608 |
Neg logit contributions | |
uncle | -0.685 |
oker | -0.667 |
ecast | -0.642 |
Liberties | -0.639 |
enegger | -0.636 |
Token | never |
Token ID | 1239 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
Malk | +0.146 |
Andreas | +0.110 |
Manson | +0.105 |
twists | +0.104 |
Goldberg | +0.103 |
Neg logit contributions | |
uncle | -0.150 |
oker | -0.147 |
ecast | -0.141 |
enegger | -0.140 |
bnb | -0.140 |
Token | called |
Token ID | 1444 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
Malk | +1.259 |
Andreas | +0.973 |
twists | +0.938 |
Manson | +0.934 |
Goldberg | +0.924 |
Neg logit contributions | |
uncle | -1.141 |
oker | -1.108 |
ecast | -1.066 |
enegger | -1.055 |
bnb | -1.054 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Malk | +1.434 |
Andreas | +1.096 |
Manson | +1.049 |
Goldberg | +1.041 |
twists | +1.031 |
Neg logit contributions | |
uncle | -1.342 |
oker | -1.303 |
ecast | -1.267 |
enegger | -1.253 |
bnb | -1.240 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Malk | +0.386 |
Andreas | +0.306 |
Manson | +0.294 |
Goldberg | +0.291 |
twists | +0.288 |
Neg logit contributions | |
uncle | -0.268 |
oker | -0.262 |
ecast | -0.251 |
enegger | -0.245 |
bnb | -0.245 |
Token | . |
Token ID | 13 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
uncle | +1.778 |
oker | +1.729 |
ecast | +1.630 |
Liberties | +1.600 |
enegger | +1.594 |
Neg logit contributions | |
Malk | -3.183 |
Andreas | -2.576 |
Manson | -2.491 |
Goldberg | -2.461 |
twists | -2.461 |
Token | twitter |
Token ID | 6956 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
Malk | +0.263 |
Andreas | +0.220 |
Manson | +0.212 |
twists | +0.210 |
Goldberg | +0.209 |
Neg logit contributions | |
uncle | -0.104 |
oker | -0.100 |
enegger | -0.094 |
ecast | -0.092 |
覚醒 | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
uncle | +2.994 |
oker | +2.897 |
ecast | +2.678 |
Liberties | +2.677 |
bnb | +2.603 |
Neg logit contributions | |
Malk | -6.117 |
Andreas | -4.963 |
Manson | -4.828 |
twists | -4.804 |
Goldberg | -4.787 |
Token | com |
Token ID | 785 |
Feature activation | -8.96e-01 |
Pos logit contributions | |
Malk | +0.229 |
Andreas | +0.173 |
Manson | +0.164 |
Goldberg | +0.162 |
twists | +0.160 |
Neg logit contributions | |
uncle | -0.241 |
oker | -0.237 |
enegger | -0.229 |
ecast | -0.227 |
bnb | -0.224 |
Token | / |
Token ID | 14 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
uncle | +1.486 |
oker | +1.444 |
ecast | +1.362 |
Liberties | +1.339 |
enegger | +1.329 |
Neg logit contributions | |
Malk | -2.663 |
Andreas | -2.156 |
Manson | -2.085 |
Goldberg | -2.059 |
twists | -2.059 |
Token | CS |
Token ID | 7902 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
Malk | +0.639 |
Andreas | +0.499 |
Manson | +0.477 |
twists | +0.471 |
Goldberg | +0.469 |
Neg logit contributions | |
uncle | -0.525 |
oker | -0.513 |
ecast | -0.492 |
enegger | -0.489 |
Liberties | -0.489 |
Token | 3 |
Token ID | 18 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
Malk | +1.601 |
Andreas | +1.244 |
Manson | +1.184 |
twists | +1.169 |
Goldberg | +1.165 |
Neg logit contributions | |
uncle | -1.375 |
oker | -1.335 |
Liberties | -1.287 |
ecast | -1.285 |
enegger | -1.284 |
Token | ed |
Token ID | 276 |
Feature activation | +5.43e-01 |
Pos logit contributions | |
Malk | +0.145 |
Andreas | +0.113 |
Manson | +0.108 |
twists | +0.106 |
Goldberg | +0.106 |
Neg logit contributions | |
uncle | -0.124 |
oker | -0.121 |
ecast | -0.116 |
enegger | -0.114 |
Liberties | -0.114 |
Token | Out |
Token ID | 7975 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
Malk | +1.356 |
Andreas | +1.053 |
Manson | +0.995 |
twists | +0.988 |
Goldberg | +0.979 |
Neg logit contributions | |
uncle | -1.149 |
oker | -1.125 |
ecast | -1.084 |
enegger | -1.073 |
Liberties | -1.068 |
Token | Atlanta |
Token ID | 43482 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
uncle | +0.144 |
oker | +0.140 |
ecast | +0.135 |
enegger | +0.132 |
Liberties | +0.132 |
Neg logit contributions | |
Malk | -0.189 |
Andreas | -0.148 |
Manson | -0.142 |
twists | -0.140 |
Goldberg | -0.140 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
uncle | +0.722 |
oker | +0.704 |
ecast | +0.673 |
enegger | +0.661 |
Liberties | +0.661 |
Neg logit contributions | |
Malk | -0.952 |
Andreas | -0.748 |
Manson | -0.718 |
Goldberg | -0.708 |
twists | -0.707 |
Token | . |
Token ID | 13 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
uncle | +1.773 |
oker | +1.724 |
ecast | +1.621 |
Liberties | +1.595 |
enegger | +1.583 |
Neg logit contributions | |
Malk | -3.217 |
Andreas | -2.604 |
Manson | -2.519 |
Goldberg | -2.492 |
twists | -2.492 |
Token | twitter |
Token ID | 6956 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
Malk | +0.214 |
Andreas | +0.180 |
Manson | +0.173 |
twists | +0.171 |
Goldberg | +0.170 |
Neg logit contributions | |
uncle | -0.085 |
oker | -0.082 |
enegger | -0.077 |
ecast | -0.076 |
覚醒 | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
uncle | +2.935 |
oker | +2.842 |
Liberties | +2.626 |
ecast | +2.622 |
bnb | +2.564 |
Neg logit contributions | |
Malk | -5.903 |
Andreas | -4.783 |
Manson | -4.647 |
twists | -4.620 |
Goldberg | -4.617 |
Token | com |
Token ID | 785 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
Malk | +0.103 |
Andreas | +0.078 |
Manson | +0.074 |
Goldberg | +0.073 |
twists | +0.072 |
Neg logit contributions | |
uncle | -0.107 |
oker | -0.105 |
enegger | -0.101 |
ecast | -0.101 |
Liberties | -0.099 |
Token | / |
Token ID | 14 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
uncle | +1.411 |
oker | +1.371 |
ecast | +1.293 |
Liberties | +1.271 |
enegger | +1.263 |
Neg logit contributions | |
Malk | -2.520 |
Andreas | -2.039 |
Manson | -1.972 |
Goldberg | -1.948 |
twists | -1.946 |
Token | xx |
Token ID | 5324 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Malk | +0.708 |
Andreas | +0.553 |
Manson | +0.529 |
twists | +0.521 |
Goldberg | +0.520 |
Neg logit contributions | |
uncle | -0.583 |
oker | -0.568 |
ecast | -0.546 |
enegger | -0.543 |
Liberties | -0.543 |
Token | A |
Token ID | 32 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
Malk | +2.663 |
Andreas | +2.085 |
Manson | +1.986 |
OTH | +1.972 |
twists | +1.964 |
Neg logit contributions | |
uncle | -2.179 |
oker | -2.115 |
enegger | -2.050 |
Liberties | -2.048 |
ecast | -2.040 |
Token | Roman |
Token ID | 32454 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
uncle | +1.557 |
oker | +1.523 |
ecast | +1.461 |
enegger | +1.429 |
Liberties | +1.428 |
Neg logit contributions | |
Malk | -1.903 |
Andreas | -1.479 |
Manson | -1.418 |
twists | -1.400 |
Goldberg | -1.398 |
Token | F |
Token ID | 37 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
uncle | +1.006 |
oker | +0.985 |
ecast | +0.940 |
enegger | +0.923 |
Liberties | +0.922 |
Neg logit contributions | |
Malk | -1.308 |
Andreas | -1.026 |
Manson | -0.986 |
Goldberg | -0.971 |
twists | -0.971 |
Token | lynn |
Token ID | 12935 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
Malk | +0.099 |
Andreas | +0.079 |
Manson | +0.075 |
Goldberg | +0.074 |
twists | +0.074 |
Neg logit contributions | |
uncle | -0.075 |
oker | -0.072 |
ecast | -0.070 |
Liberties | -0.069 |
enegger | -0.069 |
Token | pic |
Token ID | 8301 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
uncle | +1.183 |
oker | +1.155 |
ecast | +1.104 |
Liberties | +1.085 |
enegger | +1.085 |
Neg logit contributions | |
Malk | -1.504 |
Andreas | -1.176 |
Manson | -1.130 |
Goldberg | -1.113 |
twists | -1.112 |
Token | . |
Token ID | 13 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
uncle | +1.736 |
oker | +1.689 |
ecast | +1.593 |
Liberties | +1.562 |
enegger | +1.557 |
Neg logit contributions | |
Malk | -3.105 |
Andreas | -2.512 |
Manson | -2.429 |
Goldberg | -2.402 |
twists | -2.401 |
Token | twitter |
Token ID | 6956 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
Malk | +0.133 |
Andreas | +0.112 |
Manson | +0.108 |
twists | +0.106 |
Goldberg | +0.106 |
Neg logit contributions | |
uncle | -0.053 |
oker | -0.051 |
enegger | -0.048 |
ecast | -0.047 |
覚醒 | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
uncle | +2.885 |
oker | +2.792 |
ecast | +2.582 |
Liberties | +2.576 |
bnb | +2.515 |
Neg logit contributions | |
Malk | -5.823 |
Andreas | -4.714 |
Manson | -4.591 |
twists | -4.566 |
Goldberg | -4.553 |
Token | com |
Token ID | 785 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
Malk | +0.150 |
Andreas | +0.114 |
Manson | +0.107 |
Goldberg | +0.106 |
twists | +0.105 |
Neg logit contributions | |
uncle | -0.159 |
oker | -0.156 |
enegger | -0.151 |
ecast | -0.150 |
覚醒 | -0.148 |
Token | / |
Token ID | 14 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
uncle | +1.317 |
oker | +1.279 |
ecast | +1.209 |
Liberties | +1.186 |
enegger | +1.181 |
Neg logit contributions | |
Malk | -2.330 |
Andreas | -1.884 |
Manson | -1.822 |
Goldberg | -1.799 |
twists | -1.798 |
Token | gy |
Token ID | 1360 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
Malk | +0.955 |
Andreas | +0.746 |
Manson | +0.712 |
twists | +0.703 |
Goldberg | +0.701 |
Neg logit contributions | |
uncle | -0.786 |
oker | -0.766 |
ecast | -0.736 |
enegger | -0.733 |
Liberties | -0.733 |
Token | d |
Token ID | 67 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Malk | +1.159 |
Andreas | +0.905 |
Manson | +0.863 |
twists | +0.852 |
Goldberg | +0.849 |
Neg logit contributions | |
uncle | -0.962 |
oker | -0.937 |
ecast | -0.901 |
Liberties | -0.899 |
enegger | -0.898 |
Token | The |
Token ID | 464 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
Malk | +1.292 |
Andreas | +1.004 |
Goldberg | +0.956 |
Manson | +0.955 |
twists | +0.948 |
Neg logit contributions | |
uncle | -1.016 |
oker | -0.982 |
ecast | -0.944 |
enegger | -0.936 |
Liberties | -0.927 |
Token | Deb |
Token ID | 16587 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
Malk | +1.024 |
Andreas | +0.793 |
Manson | +0.759 |
Goldberg | +0.752 |
twists | +0.748 |
Neg logit contributions | |
uncle | -0.810 |
oker | -0.786 |
ecast | -0.758 |
enegger | -0.748 |
Liberties | -0.743 |
Token | ates |
Token ID | 689 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
Malk | +0.166 |
Andreas | +0.136 |
Manson | +0.131 |
Goldberg | +0.130 |
twists | +0.129 |
Neg logit contributions | |
uncle | -0.083 |
oker | -0.080 |
ecast | -0.075 |
enegger | -0.074 |
Liberties | -0.074 |
Token | pic |
Token ID | 8301 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
Malk | +0.189 |
Andreas | +0.150 |
Manson | +0.144 |
Goldberg | +0.143 |
twists | +0.143 |
Neg logit contributions | |
uncle | -0.128 |
oker | -0.124 |
ecast | -0.119 |
enegger | -0.116 |
Liberties | -0.116 |
Token | . |
Token ID | 13 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
uncle | +1.190 |
oker | +1.156 |
ecast | +1.097 |
enegger | +1.074 |
Liberties | +1.070 |
Neg logit contributions | |
Malk | -2.051 |
Andreas | -1.658 |
Manson | -1.601 |
Goldberg | -1.579 |
twists | -1.576 |
Token | twitter |
Token ID | 6956 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
Malk | +0.115 |
Andreas | +0.096 |
Manson | +0.092 |
twists | +0.092 |
Goldberg | +0.091 |
Neg logit contributions | |
uncle | -0.046 |
oker | -0.044 |
enegger | -0.041 |
ecast | -0.040 |
覚醒 | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
uncle | +2.769 |
oker | +2.682 |
ecast | +2.494 |
Liberties | +2.480 |
bnb | +2.427 |
Neg logit contributions | |
Malk | -5.464 |
Andreas | -4.430 |
Manson | -4.309 |
twists | -4.279 |
Goldberg | -4.261 |
Token | com |
Token ID | 785 |
Feature activation | -7.41e-01 |
Pos logit contributions | |
Malk | +0.242 |
Andreas | +0.184 |
Manson | +0.173 |
Goldberg | +0.171 |
twists | +0.170 |
Neg logit contributions | |
uncle | -0.254 |
oker | -0.250 |
enegger | -0.240 |
ecast | -0.239 |
覚醒 | -0.236 |
Token | / |
Token ID | 14 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
uncle | +1.256 |
oker | +1.220 |
ecast | +1.155 |
enegger | +1.132 |
Liberties | +1.129 |
Neg logit contributions | |
Malk | -2.182 |
Andreas | -1.763 |
Manson | -1.703 |
Goldberg | -1.682 |
twists | -1.680 |
Token | O |
Token ID | 46 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
Malk | +0.965 |
Andreas | +0.753 |
Manson | +0.719 |
twists | +0.710 |
Goldberg | +0.708 |
Neg logit contributions | |
uncle | -0.795 |
oker | -0.776 |
ecast | -0.745 |
enegger | -0.742 |
Liberties | -0.742 |
Token | JV |
Token ID | 41697 |
Feature activation | -6.84e-03 |
Pos logit contributions | |
Malk | +1.774 |
Andreas | +1.380 |
Manson | +1.306 |
twists | +1.292 |
OTH | +1.292 |
Neg logit contributions | |
uncle | -1.563 |
oker | -1.511 |
Liberties | -1.463 |
enegger | -1.458 |
ecast | -1.455 |
Token | old |
Token ID | 1468 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
Malk | +0.134 |
Andreas | +0.103 |
Manson | +0.098 |
twists | +0.097 |
Goldberg | +0.097 |
Neg logit contributions | |
uncle | -0.123 |
oker | -0.121 |
ecast | -0.116 |
enegger | -0.114 |
Liberties | -0.114 |
Token | and |
Token ID | 290 |
Feature activation | +2.32e-03 |
Pos logit contributions | |
Malk | +0.177 |
Andreas | +0.138 |
Manson | +0.132 |
twists | +0.130 |
Goldberg | +0.129 |
Neg logit contributions | |
uncle | -0.153 |
oker | -0.150 |
ecast | -0.143 |
enegger | -0.141 |
bnb | -0.140 |
Token | new |
Token ID | 649 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
Malk | +0.005 |
Andreas | +0.003 |
Manson | +0.003 |
Goldberg | +0.003 |
twists | +0.003 |
Neg logit contributions | |
uncle | -0.006 |
oker | -0.006 |
ecast | -0.006 |
Liberties | -0.006 |
enegger | -0.006 |
Token | kinds |
Token ID | 6982 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
Malk | +0.758 |
Andreas | +0.578 |
Manson | +0.550 |
twists | +0.545 |
Goldberg | +0.538 |
Neg logit contributions | |
uncle | -0.760 |
oker | -0.746 |
ecast | -0.714 |
enegger | -0.707 |
bnb | -0.702 |
Token | of |
Token ID | 286 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
Malk | +0.906 |
Andreas | +0.713 |
Manson | +0.685 |
twists | +0.675 |
Goldberg | +0.673 |
Neg logit contributions | |
uncle | -0.695 |
oker | -0.682 |
ecast | -0.651 |
Liberties | -0.640 |
enegger | -0.640 |
Token | Pent |
Token ID | 9696 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Malk | +0.104 |
Andreas | +0.080 |
Manson | +0.077 |
twists | +0.076 |
Goldberg | +0.075 |
Neg logit contributions | |
uncle | -0.096 |
oker | -0.094 |
ecast | -0.090 |
enegger | -0.089 |
bnb | -0.089 |
Token | ec |
Token ID | 721 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
uncle | +2.265 |
oker | +2.182 |
ecast | +2.046 |
Liberties | +1.976 |
enegger | +1.964 |
Neg logit contributions | |
Malk | -5.834 |
Andreas | -4.835 |
Manson | -4.711 |
Goldberg | -4.657 |
twists | -4.655 |
Token | ost |
Token ID | 455 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
uncle | +1.217 |
oker | +1.196 |
ecast | +1.157 |
Liberties | +1.142 |
enegger | +1.140 |
Neg logit contributions | |
Malk | -0.857 |
Andreas | -0.604 |
Manson | -0.568 |
Goldberg | -0.556 |
twists | -0.555 |
Token | al |
Token ID | 282 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
uncle | +0.290 |
oker | +0.284 |
ecast | +0.272 |
enegger | +0.270 |
bnb | +0.269 |
Neg logit contributions | |
Malk | -0.312 |
Andreas | -0.239 |
Manson | -0.228 |
twists | -0.225 |
Goldberg | -0.224 |
Token | ism |
Token ID | 1042 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
uncle | +0.469 |
oker | +0.459 |
ecast | +0.439 |
enegger | +0.434 |
bnb | +0.432 |
Neg logit contributions | |
Malk | -0.530 |
Andreas | -0.409 |
Manson | -0.391 |
twists | -0.387 |
Goldberg | -0.384 |
Token | . |
Token ID | 13 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
uncle | +0.796 |
oker | +0.777 |
ecast | +0.740 |
enegger | +0.727 |
Liberties | +0.727 |
Neg logit contributions | |
Malk | -1.133 |
Andreas | -0.899 |
Manson | -0.865 |
twists | -0.853 |
Goldberg | -0.852 |
Token | the |
Token ID | 262 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
uncle | +0.613 |
oker | +0.600 |
ecast | +0.575 |
Liberties | +0.566 |
enegger | +0.566 |
Neg logit contributions | |
Malk | -0.672 |
Andreas | -0.516 |
Manson | -0.493 |
Goldberg | -0.485 |
twists | -0.485 |
Token | backend |
Token ID | 30203 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
uncle | +0.282 |
oker | +0.276 |
ecast | +0.265 |
Liberties | +0.261 |
enegger | +0.260 |
Neg logit contributions | |
Malk | -0.305 |
Andreas | -0.234 |
Manson | -0.223 |
Goldberg | -0.220 |
twists | -0.219 |
Token | and |
Token ID | 290 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
uncle | +1.038 |
oker | +1.016 |
ecast | +0.970 |
Liberties | +0.955 |
bnb | +0.953 |
Neg logit contributions | |
Malk | -1.282 |
Andreas | -0.997 |
Manson | -0.961 |
Goldberg | -0.944 |
twists | -0.941 |
Token | the |
Token ID | 262 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
Malk | +0.218 |
Andreas | +0.167 |
Manson | +0.159 |
Goldberg | +0.157 |
twists | +0.156 |
Neg logit contributions | |
uncle | -0.206 |
oker | -0.202 |
ecast | -0.194 |
Liberties | -0.191 |
enegger | -0.191 |
Token | front |
Token ID | 2166 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
Malk | +0.061 |
Andreas | +0.047 |
Manson | +0.045 |
Goldberg | +0.044 |
twists | +0.044 |
Neg logit contributions | |
uncle | -0.057 |
oker | -0.056 |
ecast | -0.053 |
enegger | -0.053 |
Liberties | -0.052 |
Token | end |
Token ID | 437 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
Malk | +0.372 |
Andreas | +0.285 |
Manson | +0.272 |
Goldberg | +0.268 |
twists | +0.268 |
Neg logit contributions | |
uncle | -0.344 |
oker | -0.336 |
ecast | -0.323 |
enegger | -0.318 |
Liberties | -0.318 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
uncle | +3.525 |
oker | +3.464 |
ecast | +3.282 |
bnb | +3.279 |
Liberties | +3.269 |
Neg logit contributions | |
Malk | -4.276 |
Andreas | -3.263 |
Manson | -3.208 |
Goldberg | -3.121 |
twists | -3.078 |
Token |
|
Token ID | 198 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
Malk | +0.031 |
Andreas | +0.024 |
Manson | +0.023 |
Goldberg | +0.023 |
twists | +0.023 |
Neg logit contributions | |
uncle | -0.022 |
oker | -0.021 |
ecast | -0.020 |
enegger | -0.020 |
bnb | -0.020 |
Token | And |
Token ID | 1870 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
Malk | +0.061 |
Andreas | +0.047 |
Manson | +0.045 |
Goldberg | +0.044 |
twists | +0.044 |
Neg logit contributions | |
uncle | -0.056 |
oker | -0.055 |
ecast | -0.053 |
enegger | -0.053 |
Liberties | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
Malk | +0.772 |
Andreas | +0.590 |
Manson | +0.558 |
twists | +0.554 |
Goldberg | +0.550 |
Neg logit contributions | |
uncle | -0.757 |
oker | -0.743 |
ecast | -0.716 |
enegger | -0.706 |
bnb | -0.701 |
Token | of |
Token ID | 286 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Malk | +0.323 |
Andreas | +0.250 |
Manson | +0.239 |
twists | +0.236 |
Goldberg | +0.235 |
Neg logit contributions | |
uncle | -0.281 |
oker | -0.275 |
ecast | -0.264 |
enegger | -0.260 |
Liberties | -0.259 |
Token | Kaz |
Token ID | 16385 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
uncle | +0.372 |
oker | +0.366 |
ecast | +0.352 |
Liberties | +0.348 |
enegger | +0.347 |
Neg logit contributions | |
Malk | -0.341 |
Andreas | -0.254 |
Manson | -0.243 |
twists | -0.239 |
Goldberg | -0.237 |
Token | arov |
Token ID | 42737 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
Malk | +0.892 |
Andreas | +0.731 |
Manson | +0.704 |
Goldberg | +0.702 |
twists | +0.696 |
Neg logit contributions | |
uncle | -0.411 |
oker | -0.393 |
ecast | -0.372 |
Liberties | -0.369 |
bnb | -0.361 |
Token | , |
Token ID | 11 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
uncle | +0.707 |
oker | +0.689 |
ecast | +0.656 |
enegger | +0.647 |
Liberties | +0.646 |
Neg logit contributions | |
Malk | -0.931 |
Andreas | -0.731 |
Manson | -0.709 |
twists | -0.698 |
Goldberg | -0.695 |
Token | Vital |
Token ID | 28476 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
Malk | +0.037 |
Andreas | +0.028 |
Manson | +0.027 |
Goldberg | +0.027 |
twists | +0.027 |
Neg logit contributions | |
uncle | -0.032 |
oker | -0.031 |
ecast | -0.030 |
enegger | -0.030 |
Liberties | -0.029 |
Token | y |
Token ID | 88 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
uncle | +0.043 |
oker | +0.042 |
ecast | +0.040 |
enegger | +0.040 |
Liberties | +0.039 |
Neg logit contributions | |
Malk | -0.045 |
Andreas | -0.034 |
Manson | -0.032 |
Goldberg | -0.032 |
twists | -0.032 |
Token | Nem |
Token ID | 22547 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
uncle | +0.600 |
oker | +0.584 |
ecast | +0.561 |
enegger | +0.551 |
bnb | +0.549 |
Neg logit contributions | |
Malk | -0.716 |
Andreas | -0.554 |
Manson | -0.530 |
Goldberg | -0.523 |
twists | -0.519 |
Token | ov |
Token ID | 709 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
uncle | +3.306 |
oker | +3.240 |
ecast | +3.100 |
enegger | +3.028 |
Liberties | +3.005 |
Neg logit contributions | |
Malk | -4.515 |
Andreas | -3.558 |
Manson | -3.446 |
Goldberg | -3.396 |
twists | -3.389 |
Token | , |
Token ID | 11 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
uncle | +1.198 |
oker | +1.176 |
ecast | +1.120 |
enegger | +1.100 |
Liberties | +1.100 |
Neg logit contributions | |
Malk | -1.562 |
Andreas | -1.228 |
Manson | -1.186 |
Goldberg | -1.164 |
twists | -1.163 |
Token | V |
Token ID | 569 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
Malk | +0.068 |
Andreas | +0.053 |
Manson | +0.050 |
Goldberg | +0.049 |
twists | +0.049 |
Neg logit contributions | |
uncle | -0.060 |
oker | -0.059 |
ecast | -0.057 |
enegger | -0.056 |
Liberties | -0.056 |
Token | ad |
Token ID | 324 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
uncle | +0.970 |
oker | +0.949 |
ecast | +0.909 |
Liberties | +0.894 |
enegger | +0.894 |
Neg logit contributions | |
Malk | -1.123 |
Andreas | -0.867 |
Manson | -0.831 |
Goldberg | -0.818 |
twists | -0.817 |
Token | im |
Token ID | 320 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Malk | +1.109 |
Andreas | +0.912 |
Manson | +0.879 |
Goldberg | +0.871 |
twists | +0.863 |
Neg logit contributions | |
uncle | -0.524 |
oker | -0.501 |
ecast | -0.474 |
enegger | -0.468 |
Liberties | -0.466 |
Token | . |
Token ID | 13 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
Malk | +0.120 |
Andreas | +0.094 |
Manson | +0.090 |
twists | +0.089 |
Goldberg | +0.088 |
Neg logit contributions | |
uncle | -0.099 |
oker | -0.097 |
ecast | -0.092 |
enegger | -0.091 |
Liberties | -0.091 |
Token | ( |
Token ID | 357 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
uncle | +0.329 |
oker | +0.320 |
ecast | +0.304 |
enegger | +0.299 |
Liberties | +0.297 |
Neg logit contributions | |
Malk | -0.514 |
Andreas | -0.412 |
Manson | -0.397 |
Goldberg | -0.391 |
twists | -0.391 |
Token | what |
Token ID | 10919 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
uncle | +0.036 |
oker | +0.035 |
ecast | +0.034 |
enegger | +0.033 |
Liberties | +0.033 |
Neg logit contributions | |
Malk | -0.041 |
Andreas | -0.032 |
Manson | -0.030 |
twists | -0.030 |
Goldberg | -0.030 |
Token | actually |
Token ID | 1682 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
Malk | +0.466 |
Andreas | +0.363 |
Manson | +0.347 |
twists | +0.342 |
Goldberg | +0.341 |
Neg logit contributions | |
uncle | -0.395 |
oker | -0.387 |
ecast | -0.372 |
enegger | -0.366 |
Liberties | -0.364 |
Token | gets |
Token ID | 3011 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
Malk | +1.096 |
Andreas | +0.837 |
Manson | +0.794 |
twists | +0.787 |
Goldberg | +0.783 |
Neg logit contributions | |
uncle | -1.086 |
oker | -1.060 |
ecast | -1.023 |
enegger | -1.007 |
Liberties | -1.005 |
Token | loaded |
Token ID | 9639 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
uncle | +0.790 |
oker | +0.774 |
ecast | +0.738 |
Liberties | +0.734 |
enegger | +0.731 |
Neg logit contributions | |
Malk | -0.820 |
Andreas | -0.624 |
Manson | -0.595 |
Goldberg | -0.587 |
twists | -0.586 |
Token | in |
Token ID | 287 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
uncle | +3.411 |
oker | +3.341 |
ecast | +3.221 |
bnb | +3.181 |
enegger | +3.169 |
Neg logit contributions | |
Malk | -4.261 |
Andreas | -3.308 |
Manson | -3.189 |
Goldberg | -3.135 |
twists | -3.087 |
Token | depends |
Token ID | 8338 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
uncle | +1.062 |
oker | +1.039 |
ecast | +0.997 |
Liberties | +0.982 |
enegger | +0.981 |
Neg logit contributions | |
Malk | -1.139 |
Andreas | -0.870 |
Manson | -0.832 |
Goldberg | -0.819 |
twists | -0.818 |
Token | on |
Token ID | 319 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
uncle | +1.685 |
oker | +1.651 |
ecast | +1.583 |
bnb | +1.568 |
enegger | +1.558 |
Neg logit contributions | |
Malk | -1.687 |
Andreas | -1.268 |
Manson | -1.212 |
Goldberg | -1.199 |
twists | -1.182 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
uncle | +1.050 |
oker | +1.027 |
ecast | +0.985 |
Liberties | +0.970 |
enegger | +0.969 |
Neg logit contributions | |
Malk | -1.170 |
Andreas | -0.899 |
Manson | -0.861 |
Goldberg | -0.847 |
twists | -0.845 |
Token | OS |
Token ID | 7294 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
uncle | +0.814 |
oker | +0.797 |
ecast | +0.766 |
enegger | +0.756 |
Liberties | +0.753 |
Neg logit contributions | |
Malk | -0.809 |
Andreas | -0.611 |
Manson | -0.584 |
twists | -0.575 |
Goldberg | -0.574 |
Token | on |
Token ID | 319 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
Malk | +0.730 |
Andreas | +0.562 |
Manson | +0.540 |
Goldberg | +0.531 |
twists | +0.528 |
Neg logit contributions | |
uncle | -0.665 |
oker | -0.654 |
ecast | -0.634 |
enegger | -0.623 |
Liberties | -0.621 |
Token | Lu |
Token ID | 6026 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
Malk | +0.092 |
Andreas | +0.071 |
Manson | +0.069 |
Goldberg | +0.068 |
twists | +0.067 |
Neg logit contributions | |
uncle | -0.075 |
oker | -0.073 |
ecast | -0.071 |
enegger | -0.069 |
Liberties | -0.069 |
Token | ol |
Token ID | 349 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
uncle | +1.237 |
oker | +1.203 |
ecast | +1.157 |
enegger | +1.147 |
Liberties | +1.141 |
Neg logit contributions | |
Malk | -1.454 |
Andreas | -1.129 |
Manson | -1.082 |
Goldberg | -1.063 |
twists | -1.062 |
Token | Deng |
Token ID | 41985 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
uncle | +0.077 |
oker | +0.075 |
ecast | +0.073 |
enegger | +0.072 |
bnb | +0.072 |
Neg logit contributions | |
Malk | -0.074 |
Andreas | -0.056 |
Manson | -0.053 |
Goldberg | -0.052 |
twists | -0.052 |
Token | as |
Token ID | 355 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
Malk | +0.076 |
Andreas | +0.059 |
Manson | +0.057 |
twists | +0.056 |
Goldberg | +0.056 |
Neg logit contributions | |
uncle | -0.065 |
oker | -0.064 |
ecast | -0.061 |
Liberties | -0.060 |
enegger | -0.060 |
Token | belonging |
Token ID | 16686 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
uncle | +0.437 |
oker | +0.427 |
ecast | +0.408 |
Liberties | +0.401 |
enegger | +0.401 |
Neg logit contributions | |
Malk | -0.550 |
Andreas | -0.430 |
Manson | -0.413 |
Goldberg | -0.406 |
twists | -0.406 |
Token | to |
Token ID | 284 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
uncle | +3.055 |
oker | +2.934 |
ecast | +2.829 |
Liberties | +2.749 |
bnb | +2.746 |
Neg logit contributions | |
Malk | -4.603 |
Andreas | -3.652 |
Manson | -3.538 |
Goldberg | -3.455 |
twists | -3.454 |
Token | someone |
Token ID | 2130 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
Malk | +0.863 |
Andreas | +0.670 |
Manson | +0.643 |
Goldberg | +0.632 |
twists | +0.628 |
Neg logit contributions | |
uncle | -0.730 |
oker | -0.715 |
ecast | -0.689 |
enegger | -0.678 |
Liberties | -0.675 |
Token | else |
Token ID | 2073 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
uncle | +0.849 |
oker | +0.830 |
ecast | +0.795 |
Liberties | +0.782 |
enegger | +0.781 |
Neg logit contributions | |
Malk | -0.988 |
Andreas | -0.764 |
Manson | -0.732 |
Goldberg | -0.721 |
twists | -0.720 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
Malk | +0.100 |
Andreas | +0.080 |
Manson | +0.077 |
Goldberg | +0.076 |
twists | +0.076 |
Neg logit contributions | |
uncle | -0.066 |
oker | -0.064 |
ecast | -0.061 |
Liberties | -0.060 |
enegger | -0.060 |
Token | On |
Token ID | 1550 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
uncle | +0.352 |
oker | +0.343 |
ecast | +0.327 |
enegger | +0.321 |
Liberties | +0.319 |
Neg logit contributions | |
Malk | -0.516 |
Andreas | -0.410 |
Manson | -0.395 |
Goldberg | -0.390 |
twists | -0.388 |
Token | space |
Token ID | 2272 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
uncle | +0.183 |
oker | +0.180 |
ecast | +0.173 |
enegger | +0.171 |
bnb | +0.170 |
Neg logit contributions | |
Malk | -0.174 |
Andreas | -0.130 |
Manson | -0.125 |
twists | -0.123 |
Goldberg | -0.122 |
Token | , |
Token ID | 11 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
uncle | +0.712 |
oker | +0.698 |
ecast | +0.662 |
enegger | +0.651 |
Liberties | +0.650 |
Neg logit contributions | |
Malk | -1.023 |
Andreas | -0.809 |
Manson | -0.781 |
Goldberg | -0.770 |
twists | -0.766 |
Token | which |
Token ID | 543 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
Malk | +0.613 |
Andreas | +0.469 |
Manson | +0.446 |
twists | +0.440 |
Goldberg | +0.438 |
Neg logit contributions | |
uncle | -0.587 |
oker | -0.575 |
ecast | -0.553 |
Liberties | -0.544 |
enegger | -0.543 |
Token | is |
Token ID | 318 |
Feature activation | +9.74e-02 |
Pos logit contributions | |
Malk | +1.555 |
Andreas | +1.180 |
Manson | +1.120 |
twists | +1.114 |
Goldberg | +1.100 |
Neg logit contributions | |
uncle | -1.571 |
oker | -1.536 |
ecast | -1.479 |
enegger | -1.458 |
Liberties | -1.456 |
Token | the |
Token ID | 262 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
Malk | +0.229 |
Andreas | +0.174 |
Manson | +0.166 |
Goldberg | +0.164 |
twists | +0.163 |
Neg logit contributions | |
uncle | -0.223 |
oker | -0.219 |
ecast | -0.210 |
Liberties | -0.207 |
enegger | -0.206 |
Token | argument |
Token ID | 4578 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
uncle | +0.047 |
oker | +0.046 |
ecast | +0.044 |
Liberties | +0.044 |
enegger | +0.044 |
Neg logit contributions | |
Malk | -0.056 |
Andreas | -0.043 |
Manson | -0.042 |
Goldberg | -0.041 |
twists | -0.041 |
Token | for |
Token ID | 329 |
Feature activation | -8.56e-02 |
Pos logit contributions | |
uncle | +3.387 |
oker | +3.317 |
ecast | +3.152 |
bnb | +3.077 |
Liberties | +3.065 |
Neg logit contributions | |
Malk | -4.158 |
Andreas | -3.254 |
Manson | -3.141 |
Goldberg | -3.062 |
twists | -3.027 |
Token | a |
Token ID | 257 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
uncle | +0.176 |
oker | +0.172 |
ecast | +0.164 |
enegger | +0.162 |
Liberties | +0.162 |
Neg logit contributions | |
Malk | -0.220 |
Andreas | -0.172 |
Manson | -0.165 |
Goldberg | -0.163 |
twists | -0.162 |
Token | " |
Token ID | 366 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
uncle | +0.054 |
oker | +0.052 |
ecast | +0.050 |
enegger | +0.049 |
Liberties | +0.049 |
Neg logit contributions | |
Malk | -0.074 |
Andreas | -0.058 |
Manson | -0.056 |
Goldberg | -0.055 |
twists | -0.055 |
Token | : |
Token ID | 25 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
uncle | +0.127 |
oker | +0.123 |
ecast | +0.118 |
enegger | +0.115 |
bnb | +0.115 |
Neg logit contributions | |
Malk | -0.181 |
Andreas | -0.144 |
Manson | -0.139 |
Goldberg | -0.137 |
twists | -0.137 |
Token | set |
Token ID | 2617 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
Malk | +0.081 |
Andreas | +0.054 |
Manson | +0.051 |
Goldberg | +0.049 |
twists | +0.049 |
Neg logit contributions | |
uncle | -0.139 |
oker | -0.137 |
ecast | -0.132 |
Liberties | -0.131 |
enegger | -0.131 |
Token | to |
Token ID | 284 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
Malk | +0.005 |
Andreas | +0.004 |
Manson | +0.004 |
Goldberg | +0.003 |
twists | +0.003 |
Neg logit contributions | |
uncle | -0.004 |
oker | -0.004 |
ecast | -0.004 |
Liberties | -0.004 |
enegger | -0.003 |
Token | actually |
Token ID | 1682 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
uncle | +0.615 |
oker | +0.603 |
ecast | +0.580 |
enegger | +0.573 |
Liberties | +0.571 |
Neg logit contributions | |
Malk | -0.651 |
Andreas | -0.500 |
Manson | -0.477 |
Goldberg | -0.470 |
twists | -0.469 |
Token | communicate |
Token ID | 10996 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
Malk | +0.797 |
Andreas | +0.614 |
Manson | +0.586 |
Goldberg | +0.576 |
twists | +0.576 |
Neg logit contributions | |
uncle | -0.747 |
oker | -0.734 |
ecast | -0.705 |
enegger | -0.697 |
Liberties | -0.695 |
Token | on |
Token ID | 319 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
uncle | +0.495 |
oker | +0.484 |
ecast | +0.463 |
enegger | +0.456 |
Liberties | +0.455 |
Neg logit contributions | |
Malk | -0.589 |
Andreas | -0.458 |
Manson | -0.438 |
Goldberg | -0.431 |
twists | -0.431 |
Token | the |
Token ID | 262 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
uncle | +1.562 |
oker | +1.528 |
ecast | +1.465 |
Liberties | +1.442 |
enegger | +1.439 |
Neg logit contributions | |
Malk | -1.773 |
Andreas | -1.365 |
Manson | -1.309 |
Goldberg | -1.287 |
twists | -1.286 |
Token | subject |
Token ID | 2426 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
uncle | +0.897 |
oker | +0.878 |
ecast | +0.842 |
Liberties | +0.830 |
enegger | +0.828 |
Neg logit contributions | |
Malk | -0.975 |
Andreas | -0.746 |
Manson | -0.715 |
Goldberg | -0.703 |
twists | -0.702 |
Token | ) |
Token ID | 8 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
uncle | +3.119 |
oker | +3.048 |
ecast | +2.899 |
bnb | +2.870 |
Liberties | +2.842 |
Neg logit contributions | |
Malk | -4.419 |
Andreas | -3.478 |
Manson | -3.360 |
Goldberg | -3.320 |
twists | -3.311 |
Token | then |
Token ID | 788 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
Malk | +0.036 |
Andreas | +0.029 |
Manson | +0.027 |
Goldberg | +0.027 |
twists | +0.027 |
Neg logit contributions | |
uncle | -0.028 |
oker | -0.028 |
ecast | -0.027 |
enegger | -0.026 |
Liberties | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
uncle | +0.348 |
oker | +0.340 |
ecast | +0.326 |
enegger | +0.321 |
Liberties | +0.320 |
Neg logit contributions | |
Malk | -0.405 |
Andreas | -0.314 |
Manson | -0.300 |
Goldberg | -0.296 |
twists | -0.295 |
Token | sport |
Token ID | 6332 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
uncle | +0.253 |
oker | +0.248 |
ecast | +0.238 |
Liberties | +0.234 |
enegger | +0.234 |
Neg logit contributions | |
Malk | -0.262 |
Andreas | -0.199 |
Manson | -0.190 |
Goldberg | -0.187 |
twists | -0.186 |
Token | is |
Token ID | 318 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
Malk | +1.148 |
Andreas | +0.906 |
Manson | +0.868 |
twists | +0.859 |
Goldberg | +0.855 |
Neg logit contributions | |
uncle | -0.878 |
oker | -0.853 |
ecast | -0.820 |
Liberties | -0.808 |
enegger | -0.802 |
Token | has |
Token ID | 468 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
Malk | +0.155 |
Andreas | +0.118 |
Manson | +0.113 |
Goldberg | +0.111 |
twists | +0.111 |
Neg logit contributions | |
uncle | -0.147 |
oker | -0.144 |
ecast | -0.138 |
enegger | -0.136 |
Liberties | -0.136 |
Token | emerged |
Token ID | 9349 |
Feature activation | +9.72e-02 |
Pos logit contributions | |
Malk | +0.456 |
Andreas | +0.347 |
Manson | +0.331 |
Goldberg | +0.325 |
twists | +0.325 |
Neg logit contributions | |
uncle | -0.446 |
oker | -0.437 |
ecast | -0.420 |
enegger | -0.414 |
Liberties | -0.413 |
Token | is |
Token ID | 318 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
Malk | +0.237 |
Andreas | +0.183 |
Manson | +0.175 |
Goldberg | +0.172 |
twists | +0.172 |
Neg logit contributions | |
uncle | -0.213 |
oker | -0.209 |
ecast | -0.200 |
Liberties | -0.197 |
enegger | -0.197 |
Token | vag |
Token ID | 14334 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
Malk | +0.316 |
Andreas | +0.241 |
Manson | +0.231 |
Goldberg | +0.227 |
twists | +0.227 |
Neg logit contributions | |
uncle | -0.295 |
oker | -0.288 |
ecast | -0.276 |
Liberties | -0.272 |
enegger | -0.272 |
Token | us |
Token ID | 385 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
uncle | +0.740 |
oker | +0.722 |
ecast | +0.695 |
enegger | +0.690 |
Liberties | +0.689 |
Neg logit contributions | |
Malk | -0.871 |
Andreas | -0.680 |
Manson | -0.649 |
Goldberg | -0.640 |
twists | -0.637 |
Token | nerve |
Token ID | 16384 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
Malk | +0.514 |
Andreas | +0.406 |
Manson | +0.387 |
twists | +0.385 |
Goldberg | +0.384 |
Neg logit contributions | |
uncle | -0.395 |
oker | -0.380 |
ecast | -0.366 |
enegger | -0.365 |
Liberties | -0.364 |
Token | stimulation |
Token ID | 20087 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
uncle | +3.272 |
oker | +3.239 |
ecast | +3.058 |
enegger | +3.020 |
bnb | +3.018 |
Neg logit contributions | |
Malk | -4.103 |
Andreas | -3.167 |
Manson | -3.078 |
Goldberg | -3.015 |
twists | -2.971 |
Token | ( |
Token ID | 357 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
uncle | +0.774 |
oker | +0.760 |
ecast | +0.727 |
enegger | +0.716 |
Liberties | +0.715 |
Neg logit contributions | |
Malk | -0.856 |
Andreas | -0.656 |
Manson | -0.629 |
Goldberg | -0.618 |
twists | -0.617 |
Token | V |
Token ID | 53 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
uncle | +0.262 |
oker | +0.256 |
ecast | +0.247 |
enegger | +0.247 |
Liberties | +0.245 |
Neg logit contributions | |
Malk | -0.281 |
Andreas | -0.215 |
Manson | -0.205 |
Goldberg | -0.202 |
twists | -0.202 |
Token | NS |
Token ID | 8035 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
uncle | +0.900 |
oker | +0.876 |
ecast | +0.834 |
enegger | +0.819 |
Liberties | +0.819 |
Neg logit contributions | |
Malk | -1.369 |
Andreas | -1.093 |
Manson | -1.053 |
Goldberg | -1.039 |
twists | -1.038 |
Token | ), |
Token ID | 828 |
Feature activation | -5.84e-02 |
Pos logit contributions | |
Malk | +0.778 |
Andreas | +0.624 |
Manson | +0.600 |
twists | +0.592 |
Goldberg | +0.590 |
Neg logit contributions | |
uncle | -0.523 |
oker | -0.504 |
ecast | -0.483 |
Liberties | -0.477 |
enegger | -0.473 |
Token | think |
Token ID | 892 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
Malk | +0.379 |
Andreas | +0.290 |
Manson | +0.277 |
twists | +0.273 |
Goldberg | +0.272 |
Neg logit contributions | |
uncle | -0.355 |
oker | -0.346 |
ecast | -0.334 |
enegger | -0.328 |
Liberties | -0.328 |
Token | that |
Token ID | 326 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
Malk | +0.125 |
Andreas | +0.097 |
Manson | +0.093 |
Goldberg | +0.092 |
twists | +0.092 |
Neg logit contributions | |
uncle | -0.103 |
oker | -0.100 |
ecast | -0.096 |
enegger | -0.094 |
Liberties | -0.094 |
Token | one |
Token ID | 530 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
uncle | +0.313 |
oker | +0.305 |
ecast | +0.292 |
Liberties | +0.287 |
enegger | +0.287 |
Neg logit contributions | |
Malk | -0.400 |
Andreas | -0.313 |
Manson | -0.300 |
Goldberg | -0.296 |
twists | -0.296 |
Token | of |
Token ID | 286 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
uncle | +0.226 |
oker | +0.221 |
ecast | +0.211 |
enegger | +0.208 |
Liberties | +0.207 |
Neg logit contributions | |
Malk | -0.291 |
Andreas | -0.228 |
Manson | -0.219 |
twists | -0.215 |
Goldberg | -0.215 |
Token | the |
Token ID | 262 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Malk | +0.537 |
Andreas | +0.415 |
Manson | +0.398 |
Goldberg | +0.391 |
twists | +0.391 |
Neg logit contributions | |
uncle | -0.467 |
oker | -0.456 |
ecast | -0.438 |
enegger | -0.430 |
Liberties | -0.430 |
Token | things |
Token ID | 1243 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
Malk | +0.234 |
Andreas | +0.173 |
Manson | +0.164 |
Goldberg | +0.160 |
twists | +0.160 |
Neg logit contributions | |
uncle | -0.272 |
oker | -0.267 |
ecast | -0.257 |
Liberties | -0.254 |
enegger | -0.253 |
Token | that |
Token ID | 326 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
Malk | +2.534 |
Andreas | +1.996 |
Manson | +1.920 |
Goldberg | +1.882 |
twists | +1.876 |
Neg logit contributions | |
uncle | -2.080 |
oker | -2.029 |
ecast | -1.964 |
enegger | -1.929 |
Liberties | -1.913 |
Token | I |
Token ID | 314 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
Malk | +0.063 |
Andreas | +0.049 |
Manson | +0.048 |
Goldberg | +0.047 |
twists | +0.047 |
Neg logit contributions | |
uncle | -0.050 |
oker | -0.049 |
ecast | -0.047 |
enegger | -0.046 |
Liberties | -0.046 |
Token | brought |
Token ID | 3181 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
Malk | +0.735 |
Andreas | +0.562 |
Manson | +0.538 |
Goldberg | +0.529 |
twists | +0.529 |
Neg logit contributions | |
uncle | -0.684 |
oker | -0.669 |
ecast | -0.643 |
enegger | -0.632 |
Liberties | -0.632 |
Token | up |
Token ID | 510 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
uncle | +0.085 |
oker | +0.083 |
ecast | +0.079 |
enegger | +0.078 |
Liberties | +0.078 |
Neg logit contributions | |
Malk | -0.104 |
Andreas | -0.081 |
Manson | -0.077 |
Goldberg | -0.076 |
twists | -0.076 |
Token | in |
Token ID | 287 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
uncle | +0.553 |
oker | +0.541 |
ecast | +0.518 |
Liberties | +0.510 |
enegger | +0.509 |
Neg logit contributions | |
Malk | -0.641 |
Andreas | -0.494 |
Manson | -0.474 |
Goldberg | -0.466 |
twists | -0.465 |
Token | . |
Token ID | 13 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
Malk | +0.086 |
Andreas | +0.067 |
Manson | +0.064 |
Goldberg | +0.063 |
twists | +0.063 |
Neg logit contributions | |
uncle | -0.077 |
oker | -0.075 |
ecast | -0.072 |
Liberties | -0.071 |
enegger | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
uncle | +0.136 |
oker | +0.132 |
ecast | +0.126 |
bnb | +0.124 |
enegger | +0.124 |
Neg logit contributions | |
Malk | -0.193 |
Andreas | -0.153 |
Manson | -0.147 |
Goldberg | -0.146 |
twists | -0.144 |
Token |
|
Token ID | 198 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
uncle | +0.105 |
oker | +0.102 |
ecast | +0.098 |
enegger | +0.097 |
bnb | +0.096 |
Neg logit contributions | |
Malk | -0.150 |
Andreas | -0.119 |
Manson | -0.114 |
Goldberg | -0.113 |
twists | -0.113 |
Token | Os |
Token ID | 16748 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
Malk | +0.063 |
Andreas | +0.048 |
Manson | +0.046 |
Goldberg | +0.046 |
twists | +0.045 |
Neg logit contributions | |
uncle | -0.055 |
oker | -0.054 |
ecast | -0.052 |
enegger | -0.051 |
Liberties | -0.051 |
Token | so |
Token ID | 568 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
Malk | +0.355 |
Andreas | +0.259 |
Goldberg | +0.239 |
Manson | +0.239 |
twists | +0.237 |
Neg logit contributions | |
uncle | -0.473 |
oker | -0.461 |
ecast | -0.450 |
enegger | -0.448 |
Liberties | -0.447 |
Token | ff |
Token ID | 487 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
Malk | +1.275 |
Andreas | +1.002 |
Goldberg | +0.946 |
Manson | +0.945 |
twists | +0.944 |
Neg logit contributions | |
uncle | -1.047 |
oker | -1.015 |
ecast | -0.973 |
enegger | -0.972 |
覚醒 | -0.971 |
Token | hasn |
Token ID | 5818 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
Malk | +2.405 |
Andreas | +1.865 |
twists | +1.777 |
Manson | +1.773 |
Goldberg | +1.758 |
Neg logit contributions | |
uncle | -2.010 |
oker | -1.946 |
ecast | -1.877 |
Liberties | -1.848 |
bnb | -1.842 |
Token | � |
Token ID | 447 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Malk | +0.951 |
Andreas | +0.774 |
Manson | +0.748 |
twists | +0.737 |
Goldberg | +0.737 |
Neg logit contributions | |
uncle | -0.511 |
oker | -0.495 |
ecast | -0.478 |
enegger | -0.467 |
bnb | -0.460 |
Token | � |
Token ID | 247 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
Malk | +0.480 |
Andreas | +0.341 |
Manson | +0.323 |
Goldberg | +0.316 |
twists | +0.314 |
Neg logit contributions | |
uncle | -0.659 |
oker | -0.649 |
ecast | -0.628 |
enegger | -0.624 |
bnb | -0.617 |
Token | t |
Token ID | 83 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
Malk | +0.154 |
Andreas | +0.121 |
Manson | +0.115 |
twists | +0.114 |
Goldberg | +0.113 |
Neg logit contributions | |
uncle | -0.125 |
oker | -0.123 |
ecast | -0.118 |
enegger | -0.117 |
Liberties | -0.116 |
Token | followed |
Token ID | 3940 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
uncle | +0.399 |
oker | +0.390 |
ecast | +0.374 |
Liberties | +0.369 |
enegger | +0.368 |
Neg logit contributions | |
Malk | -0.434 |
Andreas | -0.333 |
Manson | -0.318 |
twists | -0.313 |
Goldberg | -0.313 |
Token | a |
Token ID | 257 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
uncle | +0.078 |
oker | +0.077 |
ecast | +0.074 |
Liberties | +0.073 |
enegger | +0.072 |
Neg logit contributions | |
Malk | -0.080 |
Andreas | -0.060 |
Manson | -0.058 |
Goldberg | -0.057 |
twists | -0.056 |
Token | few |
Token ID | 1178 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
Malk | +0.506 |
Andreas | +0.383 |
Manson | +0.366 |
Goldberg | +0.360 |
twists | +0.359 |
Neg logit contributions | |
uncle | -0.502 |
oker | -0.491 |
ecast | -0.472 |
enegger | -0.467 |
Liberties | -0.465 |
Token | fake |
Token ID | 8390 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
Malk | +1.257 |
Andreas | +0.946 |
Manson | +0.901 |
twists | +0.887 |
Goldberg | +0.885 |
Neg logit contributions | |
uncle | -1.303 |
oker | -1.278 |
ecast | -1.229 |
enegger | -1.212 |
Liberties | -1.210 |
Token | news |
Token ID | 1705 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Malk | +0.358 |
Andreas | +0.246 |
Manson | +0.230 |
Goldberg | +0.224 |
twists | +0.223 |
Neg logit contributions | |
uncle | -0.561 |
oker | -0.551 |
ecast | -0.534 |
Liberties | -0.527 |
enegger | -0.527 |
Token | items |
Token ID | 3709 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
Malk | +0.595 |
Andreas | +0.435 |
Manson | +0.413 |
twists | +0.406 |
Goldberg | +0.404 |
Neg logit contributions | |
uncle | -0.715 |
oker | -0.703 |
ecast | -0.678 |
Liberties | -0.667 |
enegger | -0.667 |
Token | myself |
Token ID | 3589 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Malk | +2.116 |
Andreas | +1.655 |
Manson | +1.579 |
twists | +1.561 |
Goldberg | +1.547 |
Neg logit contributions | |
uncle | -1.826 |
oker | -1.790 |
ecast | -1.725 |
enegger | -1.684 |
Liberties | -1.676 |
Token | ( |
Token ID | 357 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
Malk | +2.804 |
Andreas | +2.229 |
Manson | +2.131 |
twists | +2.102 |
Goldberg | +2.093 |
Neg logit contributions | |
uncle | -2.102 |
oker | -2.048 |
ecast | -1.972 |
enegger | -1.933 |
Liberties | -1.921 |
Token | not |
Token ID | 1662 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
uncle | +0.101 |
oker | +0.099 |
ecast | +0.095 |
Liberties | +0.093 |
enegger | +0.093 |
Neg logit contributions | |
Malk | -0.112 |
Andreas | -0.086 |
Manson | -0.082 |
Goldberg | -0.081 |
twists | -0.081 |
Token | often |
Token ID | 1690 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
uncle | +0.732 |
oker | +0.716 |
ecast | +0.687 |
Liberties | +0.677 |
enegger | +0.676 |
Neg logit contributions | |
Malk | -0.789 |
Andreas | -0.603 |
Manson | -0.577 |
Goldberg | -0.567 |
twists | -0.567 |
Token | but |
Token ID | 475 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
uncle | +0.245 |
oker | +0.240 |
ecast | +0.229 |
Liberties | +0.225 |
enegger | +0.225 |
Neg logit contributions | |
Malk | -0.324 |
Andreas | -0.255 |
Manson | -0.245 |
Goldberg | -0.241 |
twists | -0.241 |
Token | a |
Token ID | 257 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
uncle | +0.281 |
oker | +0.276 |
ecast | +0.264 |
Liberties | +0.260 |
enegger | +0.260 |
Neg logit contributions | |
Malk | -0.318 |
Andreas | -0.245 |
Manson | -0.235 |
Goldberg | -0.231 |
twists | -0.231 |
Token | . |
Token ID | 13 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
uncle | +0.853 |
oker | +0.834 |
ecast | +0.791 |
Liberties | +0.776 |
enegger | +0.776 |
Neg logit contributions | |
Malk | -1.244 |
Andreas | -0.986 |
Manson | -0.950 |
Goldberg | -0.939 |
twists | -0.936 |
Token |
|
Token ID | 198 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
Malk | +0.197 |
Andreas | +0.157 |
Manson | +0.151 |
Goldberg | +0.149 |
twists | +0.149 |
Neg logit contributions | |
uncle | -0.136 |
oker | -0.132 |
ecast | -0.126 |
enegger | -0.124 |
Liberties | -0.123 |
Token |
|
Token ID | 198 |
Feature activation | -9.37e-02 |
Pos logit contributions | |
uncle | +0.228 |
oker | +0.222 |
ecast | +0.213 |
enegger | +0.211 |
bnb | +0.209 |
Neg logit contributions | |
Malk | -0.330 |
Andreas | -0.264 |
Manson | -0.253 |
Goldberg | -0.249 |
twists | -0.249 |
Token | cd |
Token ID | 10210 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
uncle | +0.214 |
oker | +0.210 |
ecast | +0.202 |
Liberties | +0.199 |
enegger | +0.199 |
Neg logit contributions | |
Malk | -0.220 |
Andreas | -0.167 |
Manson | -0.159 |
twists | -0.156 |
Goldberg | -0.156 |
Token | / |
Token ID | 1220 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
uncle | +0.524 |
oker | +0.511 |
ecast | +0.487 |
enegger | +0.478 |
Liberties | +0.478 |
Neg logit contributions | |
Malk | -0.739 |
Andreas | -0.585 |
Manson | -0.563 |
twists | -0.555 |
Goldberg | -0.555 |
Token | etc |
Token ID | 14784 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
uncle | +0.546 |
oker | +0.535 |
ecast | +0.513 |
enegger | +0.506 |
Liberties | +0.506 |
Neg logit contributions | |
Malk | -0.583 |
Andreas | -0.446 |
Manson | -0.426 |
Goldberg | -0.418 |
twists | -0.418 |
Token | / |
Token ID | 14 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
Malk | +3.467 |
Andreas | +2.792 |
Manson | +2.695 |
twists | +2.659 |
Goldberg | +2.648 |
Neg logit contributions | |
uncle | -2.112 |
oker | -2.058 |
ecast | -1.952 |
Liberties | -1.916 |
enegger | -1.913 |
Token | pm |
Token ID | 4426 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Malk | +1.902 |
Andreas | +1.448 |
Manson | +1.377 |
twists | +1.361 |
Goldberg | +1.348 |
Neg logit contributions | |
uncle | -1.817 |
oker | -1.779 |
ecast | -1.718 |
Liberties | -1.702 |
enegger | -1.700 |
Token | / |
Token ID | 14 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
Malk | +0.354 |
Andreas | +0.280 |
Manson | +0.269 |
twists | +0.265 |
Goldberg | +0.265 |
Neg logit contributions | |
uncle | -0.259 |
oker | -0.252 |
ecast | -0.242 |
Liberties | -0.239 |
enegger | -0.238 |
Token | power |
Token ID | 6477 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
Malk | +1.979 |
Andreas | +1.511 |
Manson | +1.438 |
twists | +1.420 |
Goldberg | +1.414 |
Neg logit contributions | |
uncle | -1.884 |
oker | -1.843 |
ecast | -1.775 |
Liberties | -1.761 |
enegger | -1.757 |
Token | . |
Token ID | 13 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
Malk | +0.403 |
Andreas | +0.315 |
Manson | +0.303 |
twists | +0.299 |
Goldberg | +0.298 |
Neg logit contributions | |
uncle | -0.313 |
oker | -0.304 |
ecast | -0.293 |
Liberties | -0.289 |
enegger | -0.288 |
Token | Lab |
Token ID | 17822 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
uncle | +0.327 |
oker | +0.321 |
ecast | +0.308 |
enegger | +0.304 |
Liberties | +0.304 |
Neg logit contributions | |
Malk | -0.373 |
Andreas | -0.288 |
Manson | -0.275 |
Goldberg | -0.271 |
twists | -0.270 |
Token | els |
Token ID | 1424 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
uncle | +0.462 |
oker | +0.453 |
ecast | +0.430 |
enegger | +0.424 |
Liberties | +0.423 |
Neg logit contributions | |
Malk | -0.660 |
Andreas | -0.524 |
Manson | -0.502 |
Goldberg | -0.497 |
twists | -0.495 |
Token | : |
Token ID | 25 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
uncle | +0.614 |
oker | +0.599 |
ecast | +0.571 |
Liberties | +0.561 |
enegger | +0.561 |
Neg logit contributions | |
Malk | -0.833 |
Andreas | -0.656 |
Manson | -0.631 |
Goldberg | -0.622 |
twists | -0.622 |
Token | economics |
Token ID | 12446 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
Malk | +0.912 |
Andreas | +0.719 |
Manson | +0.688 |
Goldberg | +0.679 |
twists | +0.677 |
Neg logit contributions | |
uncle | -0.684 |
oker | -0.668 |
ecast | -0.639 |
enegger | -0.628 |
Liberties | -0.625 |
Token | , |
Token ID | 11 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
Malk | +0.662 |
Andreas | +0.519 |
Manson | +0.496 |
Goldberg | +0.490 |
twists | +0.489 |
Neg logit contributions | |
uncle | -0.521 |
oker | -0.508 |
ecast | -0.487 |
enegger | -0.479 |
Liberties | -0.479 |
Token | economy |
Token ID | 3773 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
Malk | +0.558 |
Andreas | +0.431 |
Manson | +0.412 |
Goldberg | +0.405 |
twists | +0.405 |
Neg logit contributions | |
uncle | -0.492 |
oker | -0.481 |
ecast | -0.462 |
enegger | -0.454 |
Liberties | -0.454 |
Token | , |
Token ID | 11 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
Malk | +2.513 |
Andreas | +1.965 |
Manson | +1.862 |
twists | +1.852 |
Goldberg | +1.847 |
Neg logit contributions | |
uncle | -2.034 |
oker | -1.983 |
ecast | -1.913 |
enegger | -1.892 |
Liberties | -1.878 |
Token | Edward |
Token ID | 10443 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
Malk | +0.846 |
Andreas | +0.654 |
Manson | +0.625 |
Goldberg | +0.616 |
twists | +0.615 |
Neg logit contributions | |
uncle | -0.737 |
oker | -0.720 |
ecast | -0.691 |
enegger | -0.680 |
Liberties | -0.678 |
Token | C |
Token ID | 327 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
uncle | +0.624 |
oker | +0.603 |
ecast | +0.566 |
enegger | +0.553 |
Liberties | +0.552 |
Neg logit contributions | |
Malk | -1.306 |
Andreas | -1.071 |
Manson | -1.038 |
Goldberg | -1.025 |
twists | -1.023 |
Token | . |
Token ID | 13 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
uncle | +0.045 |
oker | +0.044 |
ecast | +0.041 |
Liberties | +0.040 |
enegger | +0.040 |
Neg logit contributions | |
Malk | -0.079 |
Andreas | -0.064 |
Manson | -0.062 |
twists | -0.061 |
Goldberg | -0.061 |
Token | Har |
Token ID | 2113 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
uncle | +1.165 |
oker | +1.142 |
ecast | +1.097 |
Liberties | +1.082 |
enegger | +1.081 |
Neg logit contributions | |
Malk | -1.145 |
Andreas | -0.863 |
Manson | -0.823 |
Goldberg | -0.809 |
twists | -0.808 |
Token | in |
Token ID | 287 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
uncle | +0.597 |
oker | +0.585 |
ecast | +0.559 |
Liberties | +0.551 |
enegger | +0.550 |
Neg logit contributions | |
Malk | -0.676 |
Andreas | -0.520 |
Manson | -0.498 |
twists | -0.490 |
Goldberg | -0.490 |
Token | which |
Token ID | 543 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
Malk | +0.344 |
Andreas | +0.267 |
Manson | +0.256 |
Goldberg | +0.252 |
twists | +0.252 |
Neg logit contributions | |
uncle | -0.292 |
oker | -0.285 |
ecast | -0.273 |
enegger | -0.269 |
Liberties | -0.269 |
Token | his |
Token ID | 465 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Malk | +1.303 |
Andreas | +1.011 |
Manson | +0.959 |
Goldberg | +0.951 |
twists | +0.944 |
Neg logit contributions | |
uncle | -1.128 |
oker | -1.102 |
ecast | -1.062 |
enegger | -1.044 |
Liberties | -1.038 |
Token | administration |
Token ID | 3662 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
Malk | +0.237 |
Andreas | +0.180 |
Manson | +0.171 |
Goldberg | +0.169 |
twists | +0.168 |
Neg logit contributions | |
uncle | -0.236 |
oker | -0.231 |
ecast | -0.223 |
Liberties | -0.219 |
enegger | -0.219 |
Token | has |
Token ID | 468 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
Malk | +1.306 |
Andreas | +1.003 |
Manson | +0.959 |
twists | +0.950 |
Goldberg | +0.942 |
Neg logit contributions | |
uncle | -1.186 |
oker | -1.159 |
ecast | -1.113 |
Liberties | -1.096 |
enegger | -1.096 |
Token | plainly |
Token ID | 30723 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
Malk | +0.115 |
Andreas | +0.087 |
Manson | +0.083 |
Goldberg | +0.081 |
twists | +0.081 |
Neg logit contributions | |
uncle | -0.118 |
oker | -0.115 |
ecast | -0.111 |
Liberties | -0.109 |
enegger | -0.109 |
Token | , |
Token ID | 11 |
Feature activation | +8.76e-02 |
Pos logit contributions | |
Malk | +0.234 |
Andreas | +0.172 |
Manson | +0.163 |
Goldberg | +0.160 |
twists | +0.159 |
Neg logit contributions | |
uncle | -0.275 |
oker | -0.269 |
ecast | -0.260 |
Liberties | -0.256 |
enegger | -0.256 |
Token | if |
Token ID | 611 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
Malk | +0.204 |
Andreas | +0.154 |
Manson | +0.147 |
Goldberg | +0.145 |
twists | +0.145 |
Neg logit contributions | |
uncle | -0.203 |
oker | -0.198 |
ecast | -0.191 |
Liberties | -0.188 |
enegger | -0.188 |
Token | quietly |
Token ID | 12703 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
Malk | +0.198 |
Andreas | +0.151 |
Manson | +0.144 |
twists | +0.142 |
Goldberg | +0.142 |
Neg logit contributions | |
uncle | -0.188 |
oker | -0.185 |
ecast | -0.177 |
Liberties | -0.175 |
enegger | -0.175 |
Token | , |
Token ID | 11 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
Malk | +1.196 |
Andreas | +0.935 |
Manson | +0.898 |
Goldberg | +0.886 |
twists | +0.881 |
Neg logit contributions | |
uncle | -0.923 |
oker | -0.901 |
ecast | -0.862 |
enegger | -0.845 |
Liberties | -0.842 |
Token | failed |
Token ID | 4054 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
Malk | +1.221 |
Andreas | +0.912 |
Manson | +0.869 |
twists | +0.854 |
Goldberg | +0.853 |
Neg logit contributions | |
uncle | -1.308 |
oker | -1.282 |
ecast | -1.231 |
Liberties | -1.216 |
enegger | -1.215 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Malk | +0.610 |
Andreas | +0.475 |
Manson | +0.456 |
Goldberg | +0.450 |
twists | +0.447 |
Neg logit contributions | |
uncle | -0.499 |
oker | -0.488 |
ecast | -0.467 |
bnb | -0.458 |
enegger | -0.457 |
Token | city |
Token ID | 1748 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Malk | +0.256 |
Andreas | +0.197 |
Manson | +0.189 |
Goldberg | +0.186 |
twists | +0.186 |
Neg logit contributions | |
uncle | -0.230 |
oker | -0.225 |
ecast | -0.216 |
bnb | -0.212 |
enegger | -0.211 |
Token | , |
Token ID | 11 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Malk | +0.298 |
Andreas | +0.237 |
Manson | +0.228 |
twists | +0.225 |
Goldberg | +0.225 |
Neg logit contributions | |
uncle | -0.208 |
oker | -0.202 |
ecast | -0.194 |
enegger | -0.189 |
Liberties | -0.189 |
Token | or |
Token ID | 393 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
Malk | +0.291 |
Andreas | +0.224 |
Manson | +0.214 |
Goldberg | +0.211 |
twists | +0.211 |
Neg logit contributions | |
uncle | -0.258 |
oker | -0.252 |
ecast | -0.242 |
enegger | -0.238 |
Liberties | -0.237 |
Token | parked |
Token ID | 19584 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
Malk | +0.550 |
Andreas | +0.423 |
Manson | +0.405 |
Goldberg | +0.398 |
twists | +0.398 |
Neg logit contributions | |
uncle | -0.501 |
oker | -0.491 |
ecast | -0.471 |
Liberties | -0.463 |
enegger | -0.462 |
Token | in |
Token ID | 287 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
Malk | +0.685 |
Andreas | +0.525 |
Manson | +0.500 |
twists | +0.493 |
Goldberg | +0.493 |
Neg logit contributions | |
uncle | -0.645 |
oker | -0.632 |
ecast | -0.606 |
enegger | -0.598 |
Liberties | -0.596 |
Token | front |
Token ID | 2166 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
Malk | +0.572 |
Andreas | +0.441 |
Manson | +0.423 |
twists | +0.416 |
Goldberg | +0.416 |
Neg logit contributions | |
uncle | -0.510 |
oker | -0.497 |
ecast | -0.477 |
enegger | -0.467 |
bnb | -0.466 |
Token | of |
Token ID | 286 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
Malk | +1.157 |
Andreas | +0.915 |
Manson | +0.879 |
Goldberg | +0.866 |
twists | +0.865 |
Neg logit contributions | |
uncle | -0.844 |
oker | -0.823 |
ecast | -0.784 |
enegger | -0.772 |
Liberties | -0.770 |
Token | buildings |
Token ID | 6832 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Malk | +0.309 |
Andreas | +0.238 |
Manson | +0.228 |
Goldberg | +0.224 |
twists | +0.224 |
Neg logit contributions | |
uncle | -0.275 |
oker | -0.269 |
ecast | -0.258 |
enegger | -0.253 |
bnb | -0.253 |
Token | like |
Token ID | 588 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
Malk | +0.652 |
Andreas | +0.513 |
Manson | +0.493 |
Goldberg | +0.485 |
twists | +0.485 |
Neg logit contributions | |
uncle | -0.492 |
oker | -0.480 |
ecast | -0.459 |
enegger | -0.451 |
Liberties | -0.450 |
Token | Sur |
Token ID | 4198 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Malk | +0.887 |
Andreas | +0.685 |
Manson | +0.657 |
Goldberg | +0.648 |
twists | +0.644 |
Neg logit contributions | |
uncle | -0.766 |
oker | -0.748 |
ecast | -0.717 |
bnb | -0.703 |
enegger | -0.703 |
Token |
|
Token ID | 198 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
Malk | +0.155 |
Andreas | +0.124 |
Manson | +0.119 |
Goldberg | +0.117 |
PRESS | +0.114 |
Neg logit contributions | |
uncle | -0.110 |
oker | -0.107 |
ecast | -0.103 |
enegger | -0.100 |
bnb | -0.100 |
Token |
|
Token ID | 198 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
Malk | +0.452 |
Andreas | +0.359 |
Manson | +0.342 |
Goldberg | +0.337 |
twists | +0.334 |
Neg logit contributions | |
uncle | -0.336 |
oker | -0.330 |
ecast | -0.316 |
enegger | -0.315 |
bnb | -0.313 |
Token | Th |
Token ID | 817 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
Malk | +0.527 |
Andreas | +0.408 |
Manson | +0.387 |
OTH | +0.380 |
Goldberg | +0.378 |
Neg logit contributions | |
uncle | -0.473 |
oker | -0.469 |
ecast | -0.452 |
enegger | -0.451 |
bnb | -0.444 |
Token | ur |
Token ID | 333 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
uncle | +0.650 |
oker | +0.634 |
ecast | +0.610 |
enegger | +0.604 |
Liberties | +0.599 |
Neg logit contributions | |
Malk | -0.852 |
Andreas | -0.672 |
Manson | -0.645 |
Goldberg | -0.633 |
twists | -0.629 |
Token | rock |
Token ID | 10823 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
uncle | +1.730 |
oker | +1.687 |
ecast | +1.608 |
enegger | +1.575 |
Liberties | +1.574 |
Neg logit contributions | |
Malk | -2.373 |
Andreas | -1.872 |
Manson | -1.803 |
twists | -1.779 |
Goldberg | -1.779 |
Token | Council |
Token ID | 4281 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
uncle | +0.217 |
oker | +0.211 |
ecast | +0.202 |
enegger | +0.199 |
Liberties | +0.198 |
Neg logit contributions | |
Malk | -0.264 |
Andreas | -0.206 |
Manson | -0.198 |
Goldberg | -0.194 |
twists | -0.194 |
Token | said |
Token ID | 531 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
Malk | +0.478 |
Andreas | +0.371 |
Manson | +0.354 |
twists | +0.349 |
Goldberg | +0.347 |
Neg logit contributions | |
uncle | -0.416 |
oker | -0.406 |
ecast | -0.392 |
enegger | -0.386 |
bnb | -0.382 |
Token | workers |
Token ID | 3259 |
Feature activation | -5.21e-02 |
Pos logit contributions | |
Malk | +1.178 |
Andreas | +0.922 |
Manson | +0.874 |
twists | +0.859 |
Goldberg | +0.855 |
Neg logit contributions | |
uncle | -1.048 |
oker | -1.026 |
enegger | -0.983 |
ecast | -0.983 |
bnb | -0.963 |
Token | had |
Token ID | 550 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
uncle | +0.118 |
oker | +0.116 |
ecast | +0.112 |
Liberties | +0.110 |
enegger | +0.110 |
Neg logit contributions | |
Malk | -0.123 |
Andreas | -0.094 |
Manson | -0.090 |
Goldberg | -0.088 |
twists | -0.088 |
Token | begun |
Token ID | 9258 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
uncle | +0.633 |
oker | +0.620 |
ecast | +0.594 |
enegger | +0.586 |
Liberties | +0.585 |
Neg logit contributions | |
Malk | -0.689 |
Andreas | -0.529 |
Manson | -0.505 |
twists | -0.497 |
Goldberg | -0.496 |
Token | removing |
Token ID | 10829 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
Malk | +1.950 |
Andreas | +1.519 |
Manson | +1.453 |
twists | +1.435 |
Goldberg | +1.424 |
Neg logit contributions | |
uncle | -1.693 |
oker | -1.654 |
enegger | -1.593 |
ecast | -1.572 |
Liberties | -1.554 |
Token | bree |
Token ID | 20780 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
Malk | +0.393 |
Andreas | +0.298 |
Manson | +0.284 |
Goldberg | +0.280 |
twists | +0.280 |
Neg logit contributions | |
uncle | -0.387 |
oker | -0.379 |
ecast | -0.364 |
enegger | -0.359 |
Liberties | -0.358 |
Token | zy |
Token ID | 7357 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
uncle | +1.290 |
oker | +1.266 |
ecast | +1.209 |
Liberties | +1.188 |
enegger | +1.186 |
Neg logit contributions | |
Malk | -1.492 |
Andreas | -1.150 |
Manson | -1.102 |
Goldberg | -1.086 |
twists | -1.083 |
Token | read |
Token ID | 1100 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
uncle | +0.075 |
oker | +0.073 |
ecast | +0.070 |
enegger | +0.069 |
Liberties | +0.069 |
Neg logit contributions | |
Malk | -0.081 |
Andreas | -0.062 |
Manson | -0.059 |
Goldberg | -0.058 |
twists | -0.058 |
Token | , |
Token ID | 11 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
Malk | +0.715 |
Andreas | +0.563 |
Manson | +0.540 |
twists | +0.532 |
Goldberg | +0.532 |
Neg logit contributions | |
uncle | -0.545 |
oker | -0.531 |
ecast | -0.510 |
Liberties | -0.501 |
enegger | -0.499 |
Token | how |
Token ID | 703 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
Malk | +0.089 |
Andreas | +0.069 |
Manson | +0.066 |
twists | +0.065 |
Goldberg | +0.065 |
Neg logit contributions | |
uncle | -0.076 |
oker | -0.075 |
ecast | -0.072 |
enegger | -0.070 |
Liberties | -0.070 |
Token | deliberate |
Token ID | 18988 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
uncle | +0.134 |
oker | +0.131 |
ecast | +0.126 |
enegger | +0.124 |
Liberties | +0.124 |
Neg logit contributions | |
Malk | -0.156 |
Andreas | -0.121 |
Manson | -0.116 |
twists | -0.115 |
Goldberg | -0.114 |
Token | was |
Token ID | 373 |
Feature activation | +7.03e-02 |
Pos logit contributions | |
Malk | +2.227 |
Andreas | +1.730 |
Manson | +1.659 |
twists | +1.643 |
Goldberg | +1.637 |
Neg logit contributions | |
uncle | -1.918 |
oker | -1.865 |
ecast | -1.798 |
enegger | -1.787 |
Liberties | -1.760 |
Token | that |
Token ID | 326 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Malk | +0.178 |
Andreas | +0.138 |
Manson | +0.132 |
Goldberg | +0.130 |
twists | +0.130 |
Neg logit contributions | |
uncle | -0.149 |
oker | -0.146 |
ecast | -0.140 |
enegger | -0.138 |
Liberties | -0.137 |
Token | ? |
Token ID | 30 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
Malk | +0.034 |
Andreas | +0.026 |
Manson | +0.025 |
Goldberg | +0.025 |
twists | +0.025 |
Neg logit contributions | |
uncle | -0.028 |
oker | -0.028 |
ecast | -0.027 |
enegger | -0.026 |
Liberties | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
Malk | +1.399 |
Andreas | +1.116 |
Manson | +1.075 |
Goldberg | +1.061 |
twists | +1.051 |
Neg logit contributions | |
uncle | -0.945 |
oker | -0.919 |
ecast | -0.882 |
enegger | -0.866 |
bnb | -0.857 |
Token |
|
Token ID | 198 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
Malk | +0.495 |
Andreas | +0.393 |
Manson | +0.374 |
Goldberg | +0.371 |
twists | +0.367 |
Neg logit contributions | |
uncle | -0.365 |
oker | -0.357 |
enegger | -0.344 |
ecast | -0.344 |
bnb | -0.338 |
Token | they |
Token ID | 484 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
Malk | +0.807 |
Andreas | +0.625 |
Manson | +0.599 |
Goldberg | +0.592 |
twists | +0.591 |
Neg logit contributions | |
uncle | -0.680 |
oker | -0.663 |
ecast | -0.635 |
enegger | -0.627 |
Liberties | -0.624 |
Token | are |
Token ID | 389 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
uncle | +0.650 |
oker | +0.636 |
ecast | +0.610 |
enegger | +0.600 |
Liberties | +0.600 |
Neg logit contributions | |
Malk | -0.699 |
Andreas | -0.534 |
Manson | -0.511 |
Goldberg | -0.503 |
twists | -0.501 |
Token | making |
Token ID | 1642 |
Feature activation | -8.36e-02 |
Pos logit contributions | |
uncle | +0.284 |
oker | +0.278 |
ecast | +0.266 |
Liberties | +0.261 |
enegger | +0.261 |
Neg logit contributions | |
Malk | -0.341 |
Andreas | -0.265 |
Manson | -0.254 |
Goldberg | -0.250 |
twists | -0.250 |
Token | an |
Token ID | 281 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
uncle | +0.193 |
oker | +0.189 |
ecast | +0.182 |
enegger | +0.179 |
Liberties | +0.179 |
Neg logit contributions | |
Malk | -0.195 |
Andreas | -0.147 |
Manson | -0.141 |
Goldberg | -0.138 |
twists | -0.138 |
Token | informed |
Token ID | 7981 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
uncle | +0.293 |
oker | +0.287 |
ecast | +0.274 |
enegger | +0.272 |
bnb | +0.271 |
Neg logit contributions | |
Malk | -0.379 |
Andreas | -0.297 |
Manson | -0.283 |
twists | -0.283 |
Goldberg | -0.282 |
Token | decision |
Token ID | 2551 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
uncle | +0.761 |
oker | +0.744 |
ecast | +0.717 |
enegger | +0.708 |
Liberties | +0.707 |
Neg logit contributions | |
Malk | -0.709 |
Andreas | -0.528 |
Manson | -0.503 |
Goldberg | -0.496 |
twists | -0.495 |
Token | come |
Token ID | 1282 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
Malk | +1.305 |
Andreas | +1.019 |
Manson | +0.974 |
twists | +0.965 |
Goldberg | +0.960 |
Neg logit contributions | |
uncle | -1.067 |
oker | -1.038 |
ecast | -0.994 |
Liberties | -0.982 |
enegger | -0.975 |
Token | Election |
Token ID | 14219 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
Malk | +1.567 |
Andreas | +1.101 |
Manson | +1.032 |
twists | +1.019 |
Goldberg | +1.015 |
Neg logit contributions | |
uncle | -2.219 |
oker | -2.178 |
ecast | -2.105 |
Liberties | -2.087 |
enegger | -2.082 |
Token | Day |
Token ID | 3596 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
uncle | +0.066 |
oker | +0.063 |
ecast | +0.060 |
enegger | +0.059 |
bnb | +0.058 |
Neg logit contributions | |
Malk | -0.121 |
Andreas | -0.098 |
Manson | -0.095 |
twists | -0.094 |
Goldberg | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
Malk | +0.414 |
Andreas | +0.331 |
Manson | +0.319 |
twists | +0.316 |
Goldberg | +0.316 |
Neg logit contributions | |
uncle | -0.264 |
oker | -0.255 |
ecast | -0.244 |
enegger | -0.238 |
Liberties | -0.238 |
Token | Imagine |
Token ID | 18450 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
Malk | +0.112 |
Andreas | +0.089 |
Manson | +0.085 |
Goldberg | +0.085 |
twists | +0.084 |
Neg logit contributions | |
uncle | -0.076 |
oker | -0.074 |
ecast | -0.070 |
enegger | -0.069 |
bnb | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
uncle | +0.084 |
oker | +0.082 |
ecast | +0.078 |
Liberties | +0.077 |
enegger | +0.077 |
Neg logit contributions | |
Malk | -0.095 |
Andreas | -0.073 |
Manson | -0.070 |
twists | -0.069 |
Goldberg | -0.069 |
Token | private |
Token ID | 2839 |
Feature activation | -4.88e-04 |
Pos logit contributions | |
Malk | +0.012 |
Andreas | +0.009 |
Manson | +0.009 |
Goldberg | +0.009 |
twists | +0.009 |
Neg logit contributions | |
uncle | -0.013 |
oker | -0.013 |
ecast | -0.012 |
Liberties | -0.012 |
enegger | -0.012 |
Token | clinic |
Token ID | 15760 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
uncle | +0.001 |
oker | +0.001 |
ecast | +0.001 |
enegger | +0.001 |
Liberties | +0.001 |
Neg logit contributions | |
Malk | -0.001 |
Andreas | -0.001 |
Manson | -0.001 |
Goldberg | -0.001 |
twists | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
Malk | +0.348 |
Andreas | +0.271 |
Manson | +0.260 |
Goldberg | +0.257 |
twists | +0.256 |
Neg logit contributions | |
uncle | -0.282 |
oker | -0.276 |
ecast | -0.264 |
Liberties | -0.259 |
enegger | -0.259 |
Token | the |
Token ID | 262 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
uncle | +0.076 |
oker | +0.074 |
ecast | +0.071 |
Liberties | +0.070 |
enegger | +0.070 |
Neg logit contributions | |
Malk | -0.091 |
Andreas | -0.071 |
Manson | -0.068 |
Goldberg | -0.067 |
twists | -0.067 |
Token | GP |
Token ID | 14714 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
uncle | +0.135 |
oker | +0.133 |
ecast | +0.127 |
Liberties | +0.126 |
enegger | +0.126 |
Neg logit contributions | |
Malk | -0.117 |
Andreas | -0.086 |
Manson | -0.082 |
twists | -0.080 |
Goldberg | -0.080 |
Token | replied |
Token ID | 8712 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
uncle | +0.637 |
oker | +0.624 |
ecast | +0.599 |
Liberties | +0.589 |
enegger | +0.589 |
Neg logit contributions | |
Malk | -0.685 |
Andreas | -0.523 |
Manson | -0.501 |
Goldberg | -0.492 |
twists | -0.492 |
Token | : |
Token ID | 25 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
uncle | +0.862 |
oker | +0.840 |
ecast | +0.800 |
Liberties | +0.785 |
enegger | +0.784 |
Neg logit contributions | |
Malk | -1.230 |
Andreas | -0.974 |
Manson | -0.938 |
Goldberg | -0.926 |
twists | -0.924 |
Token | � |
Token ID | 564 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
uncle | +0.433 |
oker | +0.420 |
ecast | +0.398 |
enegger | +0.392 |
Liberties | +0.390 |
Neg logit contributions | |
Malk | -0.616 |
Andreas | -0.488 |
Manson | -0.469 |
Goldberg | -0.465 |
twists | -0.464 |
Token | � |
Token ID | 250 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
uncle | +0.113 |
oker | +0.110 |
ecast | +0.105 |
enegger | +0.104 |
Liberties | +0.103 |
Neg logit contributions | |
Malk | -0.131 |
Andreas | -0.101 |
Manson | -0.097 |
twists | -0.095 |
Goldberg | -0.095 |
Token | A |
Token ID | 32 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
uncle | +0.088 |
oker | +0.087 |
ecast | +0.083 |
enegger | +0.082 |
Liberties | +0.082 |
Neg logit contributions | |
Malk | -0.109 |
Andreas | -0.085 |
Manson | -0.081 |
Goldberg | -0.080 |
twists | -0.080 |
Token | a |
Token ID | 257 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
Malk | +0.296 |
Andreas | +0.230 |
Manson | +0.221 |
Goldberg | +0.218 |
twists | +0.217 |
Neg logit contributions | |
uncle | -0.245 |
oker | -0.240 |
ecast | -0.229 |
Liberties | -0.225 |
enegger | -0.225 |
Token | scandal |
Token ID | 10731 |
Feature activation | -8.67e-02 |
Pos logit contributions | |
uncle | +0.320 |
oker | +0.313 |
ecast | +0.302 |
enegger | +0.297 |
Liberties | +0.297 |
Neg logit contributions | |
Malk | -0.316 |
Andreas | -0.238 |
Manson | -0.227 |
Goldberg | -0.224 |
twists | -0.223 |
Token | as |
Token ID | 355 |
Feature activation | +4.70e-06 |
Pos logit contributions | |
uncle | +0.194 |
oker | +0.190 |
ecast | +0.182 |
enegger | +0.179 |
Liberties | +0.179 |
Neg logit contributions | |
Malk | -0.208 |
Andreas | -0.159 |
Manson | -0.152 |
Goldberg | -0.149 |
twists | -0.149 |
Token | the |
Token ID | 262 |
Feature activation | -2.71e-02 |
Token | initial |
Token ID | 4238 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
uncle | +0.067 |
oker | +0.066 |
ecast | +0.064 |
Liberties | +0.063 |
enegger | +0.063 |
Neg logit contributions | |
Malk | -0.058 |
Andreas | -0.043 |
Manson | -0.041 |
Goldberg | -0.040 |
twists | -0.040 |
Token | targeting |
Token ID | 10822 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
uncle | +0.290 |
oker | +0.284 |
ecast | +0.274 |
Liberties | +0.270 |
enegger | +0.270 |
Neg logit contributions | |
Malk | -0.252 |
Andreas | -0.186 |
Manson | -0.177 |
Goldberg | -0.173 |
twists | -0.173 |
Token | . |
Token ID | 13 |
Feature activation | -8.47e-02 |
Pos logit contributions | |
uncle | +0.281 |
oker | +0.275 |
ecast | +0.264 |
enegger | +0.260 |
Liberties | +0.260 |
Neg logit contributions | |
Malk | -0.294 |
Andreas | -0.224 |
Manson | -0.214 |
Goldberg | -0.211 |
twists | -0.211 |
Token | Ms |
Token ID | 6997 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
uncle | +0.165 |
oker | +0.160 |
ecast | +0.153 |
enegger | +0.150 |
bnb | +0.150 |
Neg logit contributions | |
Malk | -0.229 |
Andreas | -0.181 |
Manson | -0.174 |
Goldberg | -0.172 |
twists | -0.171 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Malk | +0.343 |
Andreas | +0.275 |
Manson | +0.266 |
Goldberg | +0.262 |
twists | +0.258 |
Neg logit contributions | |
uncle | -0.215 |
oker | -0.209 |
ecast | -0.199 |
enegger | -0.196 |
bnb | -0.196 |
Token | Lerner |
Token ID | 49750 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
uncle | +0.246 |
oker | +0.241 |
ecast | +0.231 |
Liberties | +0.228 |
enegger | +0.227 |
Neg logit contributions | |
Malk | -0.273 |
Andreas | -0.210 |
Manson | -0.201 |
twists | -0.198 |
Goldberg | -0.198 |
Token | , |
Token ID | 11 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
Malk | +0.474 |
Andreas | +0.368 |
Manson | +0.353 |
twists | +0.348 |
Goldberg | +0.347 |
Neg logit contributions | |
uncle | -0.397 |
oker | -0.388 |
ecast | -0.372 |
Liberties | -0.365 |
enegger | -0.365 |
Token | using |
Token ID | 1262 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
Malk | +1.834 |
Andreas | +1.399 |
Manson | +1.337 |
twists | +1.315 |
Goldberg | +1.310 |
Neg logit contributions | |
uncle | -1.736 |
oker | -1.693 |
ecast | -1.626 |
Liberties | -1.615 |
bnb | -1.602 |
Token | a |
Token ID | 257 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
Malk | +0.739 |
Andreas | +0.570 |
Manson | +0.546 |
Goldberg | +0.537 |
twists | +0.537 |
Neg logit contributions | |
uncle | -0.650 |
oker | -0.635 |
ecast | -0.609 |
Liberties | -0.599 |
enegger | -0.599 |
Token | Model |
Token ID | 9104 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
Malk | +0.213 |
Andreas | +0.162 |
Manson | +0.155 |
Goldberg | +0.152 |
twists | +0.152 |
Neg logit contributions | |
uncle | -0.202 |
oker | -0.198 |
ecast | -0.190 |
enegger | -0.187 |
Liberties | -0.187 |
Token | 1903 |
Token ID | 41625 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
Malk | +0.515 |
Andreas | +0.396 |
Manson | +0.378 |
Goldberg | +0.372 |
twists | +0.371 |
Neg logit contributions | |
uncle | -0.462 |
oker | -0.450 |
ecast | -0.435 |
enegger | -0.429 |
Liberties | -0.428 |
Token | , |
Token ID | 11 |
Feature activation | -3.90e-03 |
Pos logit contributions | |
Malk | +0.108 |
Andreas | +0.085 |
Manson | +0.081 |
twists | +0.080 |
Goldberg | +0.080 |
Neg logit contributions | |
uncle | -0.086 |
oker | -0.084 |
ecast | -0.080 |
enegger | -0.080 |
Liberties | -0.080 |
Token | instead |
Token ID | 2427 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
uncle | +0.009 |
oker | +0.008 |
ecast | +0.008 |
enegger | +0.008 |
Liberties | +0.008 |
Neg logit contributions | |
Malk | -0.010 |
Andreas | -0.007 |
Manson | -0.007 |
Goldberg | -0.007 |
twists | -0.007 |
Token | of |
Token ID | 286 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
uncle | +0.916 |
oker | +0.894 |
ecast | +0.844 |
enegger | +0.823 |
bnb | +0.823 |
Neg logit contributions | |
Malk | -1.339 |
Andreas | -1.064 |
Manson | -1.024 |
twists | -1.015 |
Goldberg | -1.012 |
Token | a |
Token ID | 257 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
Malk | +0.153 |
Andreas | +0.118 |
Manson | +0.113 |
Goldberg | +0.111 |
twists | +0.111 |
Neg logit contributions | |
uncle | -0.134 |
oker | -0.130 |
ecast | -0.125 |
Liberties | -0.123 |
enegger | -0.123 |
Token | 1903 |
Token ID | 41625 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
Malk | +0.039 |
Andreas | +0.029 |
Manson | +0.027 |
Goldberg | +0.027 |
twists | +0.027 |
Neg logit contributions | |
uncle | -0.042 |
oker | -0.041 |
ecast | -0.039 |
enegger | -0.039 |
Liberties | -0.039 |
Token | / |
Token ID | 14 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
uncle | +0.625 |
oker | +0.610 |
ecast | +0.582 |
enegger | +0.574 |
Liberties | +0.573 |
Neg logit contributions | |
Malk | -0.844 |
Andreas | -0.666 |
Manson | -0.640 |
Goldberg | -0.630 |
twists | -0.629 |
Token | A |
Token ID | 32 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
Malk | +0.767 |
Andreas | +0.605 |
Manson | +0.579 |
Goldberg | +0.571 |
twists | +0.569 |
Neg logit contributions | |
uncle | -0.575 |
oker | -0.562 |
ecast | -0.538 |
Liberties | -0.532 |
enegger | -0.530 |
Token | � |
Token ID | 447 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
Malk | +0.240 |
Andreas | +0.186 |
Manson | +0.178 |
Goldberg | +0.176 |
twists | +0.176 |
Neg logit contributions | |
uncle | -0.206 |
oker | -0.201 |
ecast | -0.193 |
Liberties | -0.191 |
enegger | -0.190 |
Token | � |
Token ID | 251 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Malk | +0.419 |
Andreas | +0.315 |
Manson | +0.300 |
Goldberg | +0.295 |
twists | +0.294 |
Neg logit contributions | |
uncle | -0.433 |
oker | -0.425 |
ecast | -0.409 |
enegger | -0.403 |
Liberties | -0.402 |
Token | books |
Token ID | 3835 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
Malk | +0.262 |
Andreas | +0.201 |
Manson | +0.192 |
twists | +0.189 |
Goldberg | +0.189 |
Neg logit contributions | |
uncle | -0.235 |
oker | -0.230 |
ecast | -0.221 |
Liberties | -0.218 |
enegger | -0.218 |
Token | , |
Token ID | 11 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
Malk | +0.864 |
Andreas | +0.666 |
Manson | +0.637 |
twists | +0.629 |
Goldberg | +0.625 |
Neg logit contributions | |
uncle | -0.778 |
oker | -0.761 |
ecast | -0.730 |
Liberties | -0.717 |
enegger | -0.716 |
Token | the |
Token ID | 262 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
uncle | +0.286 |
oker | +0.279 |
ecast | +0.267 |
enegger | +0.263 |
Liberties | +0.262 |
Neg logit contributions | |
Malk | -0.361 |
Andreas | -0.282 |
Manson | -0.271 |
twists | -0.267 |
Goldberg | -0.267 |
Token | hipp |
Token ID | 18568 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
Malk | +0.105 |
Andreas | +0.079 |
Manson | +0.076 |
twists | +0.075 |
Goldberg | +0.074 |
Neg logit contributions | |
uncle | -0.106 |
oker | -0.104 |
ecast | -0.100 |
enegger | -0.098 |
Liberties | -0.098 |
Token | ie |
Token ID | 494 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
uncle | +0.882 |
oker | +0.862 |
ecast | +0.823 |
enegger | +0.808 |
Liberties | +0.804 |
Neg logit contributions | |
Malk | -1.162 |
Andreas | -0.911 |
Manson | -0.877 |
Goldberg | -0.864 |
twists | -0.864 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
uncle | +1.360 |
oker | +1.330 |
ecast | +1.272 |
enegger | +1.254 |
Liberties | +1.253 |
Neg logit contributions | |
Malk | -1.437 |
Andreas | -1.091 |
Manson | -1.045 |
Goldberg | -1.026 |
twists | -1.020 |
Token | me |
Token ID | 502 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
uncle | +0.308 |
oker | +0.301 |
ecast | +0.289 |
enegger | +0.285 |
Liberties | +0.284 |
Neg logit contributions | |
Malk | -0.354 |
Andreas | -0.273 |
Manson | -0.262 |
Goldberg | -0.257 |
twists | -0.257 |
Token | thinks |
Token ID | 6834 |
Feature activation | +6.49e-03 |
Pos logit contributions | |
Malk | +0.635 |
Andreas | +0.486 |
Manson | +0.465 |
twists | +0.458 |
Goldberg | +0.457 |
Neg logit contributions | |
uncle | -0.592 |
oker | -0.579 |
ecast | -0.556 |
Liberties | -0.547 |
enegger | -0.547 |
Token | that |
Token ID | 326 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
Malk | +0.016 |
Andreas | +0.013 |
Manson | +0.012 |
Goldberg | +0.012 |
twists | +0.012 |
Neg logit contributions | |
uncle | -0.014 |
oker | -0.014 |
ecast | -0.013 |
enegger | -0.013 |
Liberties | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
Malk | +0.403 |
Andreas | +0.312 |
Manson | +0.298 |
twists | +0.294 |
Goldberg | +0.294 |
Neg logit contributions | |
uncle | -0.349 |
oker | -0.341 |
ecast | -0.327 |
enegger | -0.322 |
Liberties | -0.321 |
Token | outcome |
Token ID | 8055 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
uncle | +0.099 |
oker | +0.097 |
ecast | +0.093 |
Liberties | +0.092 |
enegger | +0.092 |
Neg logit contributions | |
Malk | -0.104 |
Andreas | -0.079 |
Manson | -0.075 |
Goldberg | -0.074 |
twists | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
Malk | +0.080 |
Andreas | +0.062 |
Manson | +0.060 |
twists | +0.059 |
Goldberg | +0.059 |
Neg logit contributions | |
uncle | -0.062 |
oker | -0.061 |
ecast | -0.058 |
enegger | -0.057 |
Liberties | -0.057 |
Token | be |
Token ID | 307 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
Malk | +0.092 |
Andreas | +0.069 |
Manson | +0.066 |
twists | +0.065 |
Goldberg | +0.065 |
Neg logit contributions | |
uncle | -0.096 |
oker | -0.094 |
ecast | -0.091 |
enegger | -0.090 |
Liberties | -0.090 |
Token | worse |
Token ID | 4785 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
uncle | +0.048 |
oker | +0.047 |
ecast | +0.045 |
enegger | +0.044 |
Liberties | +0.044 |
Neg logit contributions | |
Malk | -0.049 |
Andreas | -0.038 |
Manson | -0.036 |
twists | -0.035 |
Goldberg | -0.035 |
Token | than |
Token ID | 621 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
Malk | +2.132 |
Andreas | +1.643 |
twists | +1.560 |
Manson | +1.559 |
Goldberg | +1.531 |
Neg logit contributions | |
uncle | -1.939 |
oker | -1.893 |
ecast | -1.827 |
Liberties | -1.795 |
enegger | -1.789 |
Token | the |
Token ID | 262 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
uncle | +0.410 |
oker | +0.401 |
ecast | +0.384 |
Liberties | +0.378 |
enegger | +0.378 |
Neg logit contributions | |
Malk | -0.472 |
Andreas | -0.365 |
Manson | -0.349 |
Goldberg | -0.344 |
twists | -0.343 |
Token | partial |
Token ID | 13027 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
uncle | +0.228 |
oker | +0.223 |
ecast | +0.214 |
Liberties | +0.210 |
enegger | +0.210 |
Neg logit contributions | |
Malk | -0.251 |
Andreas | -0.192 |
Manson | -0.184 |
Goldberg | -0.181 |
twists | -0.181 |
Token | reactor |
Token ID | 21905 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
Malk | +0.806 |
Andreas | +0.579 |
Manson | +0.541 |
twists | +0.532 |
Goldberg | +0.530 |
Neg logit contributions | |
uncle | -1.100 |
oker | -1.076 |
ecast | -1.044 |
enegger | -1.032 |
Liberties | -1.029 |
Token | meltdown |
Token ID | 37824 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
Malk | +0.954 |
Andreas | +0.689 |
Manson | +0.644 |
twists | +0.636 |
Goldberg | +0.634 |
Neg logit contributions | |
uncle | -1.260 |
oker | -1.237 |
ecast | -1.197 |
Liberties | -1.187 |
enegger | -1.182 |
Token | at |
Token ID | 379 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
Malk | +0.430 |
Andreas | +0.329 |
Manson | +0.312 |
Goldberg | +0.308 |
twists | +0.308 |
Neg logit contributions | |
uncle | -0.416 |
oker | -0.405 |
ecast | -0.391 |
Liberties | -0.386 |
bnb | -0.385 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Malk | +0.189 |
Andreas | +0.145 |
Manson | +0.139 |
Goldberg | +0.137 |
twists | +0.137 |
Neg logit contributions | |
uncle | -0.171 |
oker | -0.167 |
ecast | -0.161 |
Liberties | -0.158 |
enegger | -0.158 |
Token | kind |
Token ID | 1611 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
uncle | +0.298 |
oker | +0.292 |
ecast | +0.280 |
enegger | +0.276 |
Liberties | +0.276 |
Neg logit contributions | |
Malk | -0.315 |
Andreas | -0.240 |
Manson | -0.230 |
Goldberg | -0.226 |
twists | -0.226 |
Token | of |
Token ID | 286 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
Malk | +0.419 |
Andreas | +0.327 |
Manson | +0.313 |
twists | +0.310 |
Goldberg | +0.309 |
Neg logit contributions | |
uncle | -0.343 |
oker | -0.336 |
ecast | -0.323 |
enegger | -0.322 |
Liberties | -0.318 |
Token | beast |
Token ID | 13824 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
uncle | +0.410 |
oker | +0.402 |
ecast | +0.385 |
Liberties | +0.380 |
enegger | +0.379 |
Neg logit contributions | |
Malk | -0.435 |
Andreas | -0.332 |
Manson | -0.317 |
Goldberg | -0.312 |
twists | -0.311 |
Token | Steven |
Token ID | 8239 |
Feature activation | +4.99e-03 |
Pos logit contributions | |
uncle | +0.477 |
oker | +0.466 |
ecast | +0.445 |
enegger | +0.437 |
Liberties | +0.437 |
Neg logit contributions | |
Malk | -0.638 |
Andreas | -0.502 |
Manson | -0.484 |
Goldberg | -0.477 |
twists | -0.475 |
Token | Spielberg |
Token ID | 41151 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Malk | +0.013 |
Andreas | +0.010 |
Manson | +0.010 |
Goldberg | +0.009 |
twists | +0.009 |
Neg logit contributions | |
uncle | -0.010 |
oker | -0.010 |
ecast | -0.010 |
enegger | -0.010 |
bnb | -0.010 |
Token | might |
Token ID | 1244 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
uncle | +2.575 |
oker | +2.527 |
ecast | +2.422 |
enegger | +2.390 |
Liberties | +2.370 |
Neg logit contributions | |
Malk | -3.015 |
Andreas | -2.327 |
Manson | -2.229 |
Goldberg | -2.194 |
twists | -2.178 |
Token | feature |
Token ID | 3895 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
uncle | +0.412 |
oker | +0.404 |
ecast | +0.386 |
Liberties | +0.383 |
bnb | +0.380 |
Neg logit contributions | |
Malk | -0.459 |
Andreas | -0.355 |
Manson | -0.341 |
twists | -0.335 |
Goldberg | -0.333 |
Token | in |
Token ID | 287 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
uncle | +0.094 |
oker | +0.092 |
ecast | +0.088 |
enegger | +0.087 |
Liberties | +0.087 |
Neg logit contributions | |
Malk | -0.113 |
Andreas | -0.088 |
Manson | -0.084 |
Goldberg | -0.083 |
twists | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
uncle | +0.530 |
oker | +0.518 |
ecast | +0.495 |
Liberties | +0.487 |
enegger | +0.487 |
Neg logit contributions | |
Malk | -0.651 |
Andreas | -0.507 |
Manson | -0.486 |
twists | -0.479 |
Goldberg | -0.479 |
Token | movie |
Token ID | 3807 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
uncle | +0.679 |
oker | +0.665 |
ecast | +0.637 |
Liberties | +0.627 |
enegger | +0.627 |
Neg logit contributions | |
Malk | -0.759 |
Andreas | -0.584 |
Manson | -0.559 |
Goldberg | -0.550 |
twists | -0.549 |
Token | v |
Token ID | 85 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
Malk | +2.151 |
Andreas | +1.643 |
Manson | +1.548 |
twists | +1.531 |
Goldberg | +1.524 |
Neg logit contributions | |
uncle | -2.064 |
oker | -1.999 |
ecast | -1.942 |
Liberties | -1.930 |
enegger | -1.928 |
Token | 2 |
Token ID | 17 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
uncle | +0.788 |
oker | +0.777 |
ecast | +0.741 |
bnb | +0.728 |
Liberties | +0.725 |
Neg logit contributions | |
Malk | -0.870 |
Andreas | -0.664 |
Manson | -0.639 |
Goldberg | -0.633 |
twists | -0.628 |
Token | # |
Token ID | 1303 |
Feature activation | +8.27e-02 |
Pos logit contributions | |
Malk | +0.761 |
Andreas | +0.595 |
Manson | +0.569 |
twists | +0.561 |
Goldberg | +0.560 |
Neg logit contributions | |
uncle | -0.614 |
oker | -0.598 |
ecast | -0.574 |
enegger | -0.566 |
Liberties | -0.566 |
Token | construct |
Token ID | 5678 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
Malk | +0.200 |
Andreas | +0.153 |
Manson | +0.146 |
twists | +0.144 |
Goldberg | +0.144 |
Neg logit contributions | |
uncle | -0.184 |
oker | -0.180 |
ecast | -0.173 |
enegger | -0.171 |
Liberties | -0.170 |
Token | the |
Token ID | 262 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
uncle | +0.441 |
oker | +0.431 |
ecast | +0.413 |
Liberties | +0.407 |
enegger | +0.406 |
Neg logit contributions | |
Malk | -0.497 |
Andreas | -0.382 |
Manson | -0.366 |
Goldberg | -0.360 |
twists | -0.360 |
Token | argument |
Token ID | 4578 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Malk | +0.603 |
Andreas | +0.461 |
Manson | +0.439 |
Goldberg | +0.433 |
twists | +0.433 |
Neg logit contributions | |
uncle | -0.567 |
oker | -0.553 |
ecast | -0.532 |
enegger | -0.523 |
Liberties | -0.522 |
Token | parse |
Token ID | 21136 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
uncle | +2.797 |
oker | +2.757 |
ecast | +2.643 |
bnb | +2.594 |
Liberties | +2.583 |
Neg logit contributions | |
Malk | -2.654 |
Andreas | -1.991 |
Manson | -1.896 |
Goldberg | -1.861 |
twists | -1.845 |
Token | and |
Token ID | 290 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
Malk | +0.411 |
Andreas | +0.323 |
Manson | +0.310 |
twists | +0.306 |
Goldberg | +0.306 |
Neg logit contributions | |
uncle | -0.309 |
oker | -0.300 |
ecast | -0.287 |
Liberties | -0.284 |
enegger | -0.282 |
Token | parse |
Token ID | 21136 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
Malk | +0.464 |
Andreas | +0.349 |
Manson | +0.332 |
twists | +0.327 |
Goldberg | +0.326 |
Neg logit contributions | |
uncle | -0.477 |
oker | -0.466 |
ecast | -0.450 |
enegger | -0.445 |
Liberties | -0.444 |
Token | the |
Token ID | 262 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
Malk | +0.044 |
Andreas | +0.034 |
Manson | +0.032 |
twists | +0.032 |
Goldberg | +0.032 |
Neg logit contributions | |
uncle | -0.041 |
oker | -0.040 |
ecast | -0.038 |
Liberties | -0.038 |
enegger | -0.038 |
Token | arguments |
Token ID | 7159 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
Malk | +0.419 |
Andreas | +0.318 |
Manson | +0.302 |
Goldberg | +0.299 |
twists | +0.298 |
Neg logit contributions | |
uncle | -0.404 |
oker | -0.393 |
ecast | -0.380 |
Liberties | -0.373 |
enegger | -0.372 |
Token |
|
Token ID | 198 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
uncle | +0.178 |
oker | +0.174 |
ecast | +0.167 |
enegger | +0.166 |
bnb | +0.164 |
Neg logit contributions | |
Malk | -0.254 |
Andreas | -0.202 |
Manson | -0.193 |
Goldberg | -0.191 |
twists | -0.191 |
Token | Nearly |
Token ID | 27927 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
uncle | +0.182 |
oker | +0.178 |
ecast | +0.172 |
Liberties | +0.169 |
enegger | +0.169 |
Neg logit contributions | |
Malk | -0.205 |
Andreas | -0.157 |
Manson | -0.151 |
Goldberg | -0.148 |
twists | -0.147 |
Token | 300 |
Token ID | 5867 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
uncle | +0.227 |
oker | +0.221 |
ecast | +0.212 |
enegger | +0.209 |
Liberties | +0.207 |
Neg logit contributions | |
Malk | -0.271 |
Andreas | -0.211 |
Manson | -0.202 |
Goldberg | -0.200 |
twists | -0.199 |
Token | wildfires |
Token ID | 48298 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
Malk | +1.275 |
Andreas | +0.974 |
Manson | +0.931 |
Goldberg | +0.916 |
twists | +0.916 |
Neg logit contributions | |
uncle | -1.196 |
oker | -1.171 |
ecast | -1.124 |
enegger | -1.106 |
Liberties | -1.104 |
Token | are |
Token ID | 389 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
uncle | +0.594 |
oker | +0.581 |
ecast | +0.558 |
Liberties | +0.549 |
enegger | +0.549 |
Neg logit contributions | |
Malk | -0.655 |
Andreas | -0.503 |
Manson | -0.481 |
Goldberg | -0.473 |
twists | -0.473 |
Token | burning |
Token ID | 9482 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
uncle | +0.337 |
oker | +0.329 |
ecast | +0.314 |
Liberties | +0.308 |
enegger | +0.308 |
Neg logit contributions | |
Malk | -0.456 |
Andreas | -0.359 |
Manson | -0.345 |
Goldberg | -0.340 |
twists | -0.340 |
Token | in |
Token ID | 287 |
Feature activation | -9.14e-02 |
Pos logit contributions | |
uncle | +2.507 |
oker | +2.470 |
ecast | +2.351 |
Liberties | +2.324 |
enegger | +2.320 |
Neg logit contributions | |
Malk | -2.956 |
Andreas | -2.273 |
Manson | -2.187 |
Goldberg | -2.171 |
twists | -2.162 |
Token | Alaska |
Token ID | 12926 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
uncle | +0.200 |
oker | +0.196 |
ecast | +0.188 |
Liberties | +0.185 |
enegger | +0.185 |
Neg logit contributions | |
Malk | -0.224 |
Andreas | -0.172 |
Manson | -0.165 |
Goldberg | -0.162 |
twists | -0.162 |
Token | right |
Token ID | 826 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
Malk | +0.987 |
Andreas | +0.777 |
Manson | +0.745 |
twists | +0.734 |
Goldberg | +0.734 |
Neg logit contributions | |
uncle | -0.744 |
oker | -0.726 |
ecast | -0.696 |
enegger | -0.684 |
Liberties | -0.681 |
Token | now |
Token ID | 783 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
Malk | +0.129 |
Andreas | +0.091 |
Manson | +0.085 |
twists | +0.083 |
Goldberg | +0.083 |
Neg logit contributions | |
uncle | -0.186 |
oker | -0.182 |
ecast | -0.176 |
Liberties | -0.174 |
enegger | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-03 |
Pos logit contributions | |
Malk | +0.639 |
Andreas | +0.505 |
Manson | +0.485 |
twists | +0.476 |
Goldberg | +0.475 |
Neg logit contributions | |
uncle | -0.469 |
oker | -0.457 |
ecast | -0.439 |
Liberties | -0.429 |
enegger | -0.428 |
Token | for |
Token ID | 329 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
Malk | +0.682 |
Andreas | +0.525 |
Manson | +0.499 |
Goldberg | +0.492 |
twists | +0.492 |
Neg logit contributions | |
uncle | -0.629 |
oker | -0.613 |
ecast | -0.591 |
enegger | -0.582 |
Liberties | -0.579 |
Token | an |
Token ID | 281 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Malk | +0.605 |
Andreas | +0.468 |
Manson | +0.448 |
Goldberg | +0.441 |
twists | +0.440 |
Neg logit contributions | |
uncle | -0.526 |
oker | -0.513 |
ecast | -0.492 |
enegger | -0.486 |
Liberties | -0.484 |
Token | impact |
Token ID | 2928 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
uncle | +0.248 |
oker | +0.243 |
ecast | +0.234 |
enegger | +0.231 |
bnb | +0.231 |
Neg logit contributions | |
Malk | -0.272 |
Andreas | -0.211 |
Manson | -0.200 |
twists | -0.198 |
Goldberg | -0.198 |
Token | on |
Token ID | 319 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
Malk | +1.256 |
Andreas | +0.982 |
Manson | +0.933 |
twists | +0.919 |
Goldberg | +0.914 |
Neg logit contributions | |
uncle | -1.108 |
oker | -1.081 |
ecast | -1.048 |
enegger | -1.038 |
bnb | -1.036 |
Token | the |
Token ID | 262 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
Malk | +0.581 |
Andreas | +0.449 |
Manson | +0.430 |
Goldberg | +0.423 |
twists | +0.423 |
Neg logit contributions | |
uncle | -0.505 |
oker | -0.493 |
ecast | -0.473 |
enegger | -0.465 |
Liberties | -0.465 |
Token | western |
Token ID | 8830 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
uncle | +0.034 |
oker | +0.033 |
ecast | +0.032 |
enegger | +0.031 |
Liberties | +0.031 |
Neg logit contributions | |
Malk | -0.044 |
Andreas | -0.035 |
Manson | -0.033 |
Goldberg | -0.033 |
twists | -0.033 |
Token | edge |
Token ID | 5743 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
uncle | +2.533 |
oker | +2.493 |
ecast | +2.383 |
Liberties | +2.360 |
enegger | +2.335 |
Neg logit contributions | |
Malk | -2.742 |
Andreas | -2.080 |
Manson | -1.993 |
Goldberg | -1.973 |
twists | -1.958 |
Token | of |
Token ID | 286 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
uncle | +0.441 |
oker | +0.431 |
ecast | +0.409 |
Liberties | +0.402 |
enegger | +0.401 |
Neg logit contributions | |
Malk | -0.611 |
Andreas | -0.481 |
Manson | -0.463 |
Goldberg | -0.458 |
twists | -0.456 |
Token | Mare |
Token ID | 36989 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
Malk | +0.831 |
Andreas | +0.637 |
Manson | +0.610 |
Goldberg | +0.599 |
twists | +0.599 |
Neg logit contributions | |
uncle | -0.761 |
oker | -0.744 |
ecast | -0.714 |
enegger | -0.703 |
Liberties | -0.703 |
Token | Tr |
Token ID | 833 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
Malk | +0.721 |
Andreas | +0.561 |
Manson | +0.537 |
Goldberg | +0.528 |
twists | +0.527 |
Neg logit contributions | |
uncle | -0.604 |
oker | -0.591 |
ecast | -0.566 |
enegger | -0.557 |
Liberties | -0.555 |
Token | anqu |
Token ID | 26184 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
Malk | +1.423 |
Andreas | +1.093 |
Manson | +1.038 |
Goldberg | +1.023 |
twists | +1.012 |
Neg logit contributions | |
uncle | -1.327 |
oker | -1.302 |
enegger | -1.280 |
ecast | -1.272 |
覚醒 | -1.254 |
Token | 1 |
Token ID | 352 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
uncle | +0.608 |
oker | +0.592 |
ecast | +0.564 |
Liberties | +0.553 |
bnb | +0.549 |
Neg logit contributions | |
Malk | -0.885 |
Andreas | -0.703 |
Manson | -0.679 |
Goldberg | -0.666 |
twists | -0.665 |
Token | / |
Token ID | 14 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
uncle | +0.850 |
oker | +0.826 |
ecast | +0.784 |
enegger | +0.775 |
bnb | +0.767 |
Neg logit contributions | |
Malk | -1.421 |
Andreas | -1.145 |
Manson | -1.104 |
twists | -1.095 |
Goldberg | -1.093 |
Token | 2 |
Token ID | 17 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
Malk | +0.233 |
Andreas | +0.173 |
Manson | +0.164 |
Goldberg | +0.161 |
twists | +0.161 |
Neg logit contributions | |
uncle | -0.260 |
oker | -0.256 |
ecast | -0.246 |
enegger | -0.246 |
bnb | -0.243 |
Token | hours |
Token ID | 2250 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
uncle | +0.175 |
oker | +0.172 |
ecast | +0.167 |
enegger | +0.165 |
Liberties | +0.165 |
Neg logit contributions | |
Malk | -0.102 |
Andreas | -0.068 |
Manson | -0.063 |
Goldberg | -0.062 |
twists | -0.062 |
Token | of |
Token ID | 286 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
Malk | +0.183 |
Andreas | +0.143 |
Manson | +0.137 |
Goldberg | +0.136 |
twists | +0.135 |
Neg logit contributions | |
uncle | -0.146 |
oker | -0.143 |
ecast | -0.137 |
enegger | -0.135 |
Liberties | -0.135 |
Token | sleep |
Token ID | 3993 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
uncle | +0.337 |
oker | +0.329 |
ecast | +0.317 |
enegger | +0.312 |
Liberties | +0.311 |
Neg logit contributions | |
Malk | -0.356 |
Andreas | -0.272 |
Manson | -0.260 |
Goldberg | -0.256 |
twists | -0.255 |
Token | a |
Token ID | 257 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
uncle | +2.428 |
oker | +2.373 |
ecast | +2.272 |
Liberties | +2.233 |
bnb | +2.232 |
Neg logit contributions | |
Malk | -3.003 |
Andreas | -2.335 |
Manson | -2.251 |
twists | -2.208 |
Goldberg | -2.205 |
Token | night |
Token ID | 1755 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
uncle | +0.392 |
oker | +0.382 |
ecast | +0.365 |
enegger | +0.360 |
Liberties | +0.358 |
Neg logit contributions | |
Malk | -0.537 |
Andreas | -0.424 |
Manson | -0.408 |
Goldberg | -0.402 |
twists | -0.402 |
Token | . |
Token ID | 13 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
uncle | +2.381 |
oker | +2.320 |
ecast | +2.192 |
bnb | +2.157 |
Liberties | +2.154 |
Neg logit contributions | |
Malk | -3.657 |
Andreas | -2.914 |
Manson | -2.813 |
twists | -2.773 |
Goldberg | -2.771 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
uncle | +0.314 |
oker | +0.305 |
ecast | +0.292 |
enegger | +0.285 |
bnb | +0.284 |
Neg logit contributions | |
Malk | -0.456 |
Andreas | -0.363 |
Manson | -0.349 |
Goldberg | -0.345 |
twists | -0.342 |
Token |
|
Token ID | 198 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
uncle | +0.276 |
oker | +0.269 |
ecast | +0.258 |
enegger | +0.256 |
bnb | +0.253 |
Neg logit contributions | |
Malk | -0.391 |
Andreas | -0.312 |
Manson | -0.298 |
Goldberg | -0.295 |
twists | -0.294 |