Token | cable |
Token ID | 7862 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
isable | +0.746 |
ォ | +0.721 |
ERG | +0.714 |
ident | +0.714 |
OLD | +0.622 |
Neg logit contributions | |
enburg | -0.997 |
Directions | -0.943 |
Dracula | -0.932 |
warr | -0.859 |
etheless | -0.846 |
Token | traps |
Token ID | 20348 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
isable | +0.090 |
ォ | +0.087 |
ident | +0.086 |
ERG | +0.086 |
OLD | +0.075 |
Neg logit contributions | |
enburg | -0.110 |
Directions | -0.104 |
Dracula | -0.103 |
warr | -0.095 |
etheless | -0.093 |
Token | and |
Token ID | 290 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
enburg | +0.121 |
Directions | +0.115 |
Dracula | +0.113 |
warr | +0.104 |
etheless | +0.103 |
Neg logit contributions | |
isable | -0.095 |
ォ | -0.091 |
ident | -0.091 |
ERG | -0.090 |
OLD | -0.079 |
Token | an |
Token ID | 281 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
isable | +0.125 |
ォ | +0.121 |
ident | +0.120 |
ERG | +0.120 |
OLD | +0.106 |
Neg logit contributions | |
enburg | -0.147 |
Directions | -0.139 |
Dracula | -0.137 |
warr | -0.126 |
etheless | -0.124 |
Token | animal |
Token ID | 5044 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
isable | +0.087 |
ォ | +0.084 |
ident | +0.083 |
ERG | +0.083 |
OLD | +0.074 |
Neg logit contributions | |
enburg | -0.097 |
Directions | -0.092 |
Dracula | -0.090 |
warr | -0.083 |
etheless | -0.082 |
Token | carc |
Token ID | 36756 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
isable | +2.868 |
ident | +2.757 |
ォ | +2.752 |
ERG | +2.709 |
camp | +2.489 |
Neg logit contributions | |
enburg | -2.920 |
Directions | -2.755 |
Dracula | -2.680 |
etheless | -2.465 |
warr | -2.457 |
Token | ass |
Token ID | 562 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
isable | +4.602 |
ident | +4.454 |
ォ | +4.378 |
ERG | +4.326 |
abba | +3.777 |
Neg logit contributions | |
enburg | -6.909 |
Directions | -6.645 |
Dracula | -6.429 |
warr | -6.092 |
nces | -5.972 |
Token | as |
Token ID | 355 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
isable | +0.725 |
ォ | +0.701 |
ERG | +0.696 |
ident | +0.693 |
OLD | +0.611 |
Neg logit contributions | |
enburg | -0.863 |
Directions | -0.815 |
Dracula | -0.805 |
warr | -0.738 |
etheless | -0.728 |
Token | bait |
Token ID | 26536 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
enburg | +0.934 |
Directions | +0.889 |
Dracula | +0.884 |
warr | +0.814 |
nces | +0.802 |
Neg logit contributions | |
isable | -0.559 |
ォ | -0.538 |
ERG | -0.533 |
ident | -0.528 |
OLD | -0.455 |
Token | , |
Token ID | 11 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
isable | +0.672 |
ォ | +0.650 |
ERG | +0.642 |
ident | +0.641 |
OLD | +0.558 |
Neg logit contributions | |
enburg | -0.910 |
Directions | -0.865 |
Dracula | -0.854 |
warr | -0.785 |
etheless | -0.774 |
Token | they |
Token ID | 484 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
isable | +0.145 |
ォ | +0.140 |
ident | +0.138 |
ERG | +0.138 |
OLD | +0.121 |
Neg logit contributions | |
enburg | -0.175 |
Directions | -0.166 |
Dracula | -0.164 |
warr | -0.150 |
etheless | -0.148 |
Token | is |
Token ID | 318 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
isable | +2.347 |
ident | +2.245 |
ERG | +2.229 |
ォ | +2.210 |
camp | +1.984 |
Neg logit contributions | |
enburg | -3.085 |
Directions | -2.928 |
Dracula | -2.852 |
warr | -2.676 |
etheless | -2.660 |
Token | determined |
Token ID | 5295 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
isable | +0.371 |
ォ | +0.355 |
ident | +0.355 |
ERG | +0.354 |
camp | +0.313 |
Neg logit contributions | |
enburg | -0.416 |
Directions | -0.393 |
Dracula | -0.385 |
warr | -0.355 |
etheless | -0.351 |
Token | by |
Token ID | 416 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
isable | +0.511 |
ォ | +0.495 |
ident | +0.494 |
ERG | +0.492 |
OLD | +0.442 |
Neg logit contributions | |
enburg | -0.440 |
Directions | -0.411 |
Dracula | -0.404 |
warr | -0.366 |
etheless | -0.361 |
Token | how |
Token ID | 703 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
enburg | +0.839 |
Directions | +0.796 |
Dracula | +0.787 |
warr | +0.720 |
etheless | +0.708 |
Neg logit contributions | |
isable | -0.679 |
ォ | -0.663 |
ERG | -0.651 |
ident | -0.650 |
OLD | -0.573 |
Token | the |
Token ID | 262 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
isable | +1.741 |
ERG | +1.673 |
ident | +1.672 |
ォ | +1.657 |
camp | +1.488 |
Neg logit contributions | |
enburg | -1.996 |
Directions | -1.894 |
Dracula | -1.836 |
warr | -1.709 |
etheless | -1.702 |
Token | carc |
Token ID | 36756 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
isable | +0.472 |
ォ | +0.446 |
ident | +0.444 |
ERG | +0.442 |
OLD | +0.363 |
Neg logit contributions | |
enburg | -1.046 |
Directions | -1.000 |
Dracula | -0.986 |
warr | -0.928 |
etheless | -0.921 |
Token | ass |
Token ID | 562 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
isable | +4.478 |
ident | +4.310 |
ERG | +4.231 |
ォ | +4.214 |
abba | +3.678 |
Neg logit contributions | |
enburg | -6.955 |
Directions | -6.683 |
Dracula | -6.469 |
warr | -6.178 |
etheless | -6.037 |
Token | is |
Token ID | 318 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
isable | +1.316 |
ォ | +1.268 |
ident | +1.261 |
ERG | +1.257 |
OLD | +1.099 |
Neg logit contributions | |
enburg | -1.697 |
Directions | -1.607 |
Dracula | -1.581 |
warr | -1.463 |
etheless | -1.447 |
Token | scored |
Token ID | 7781 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
isable | +0.424 |
ォ | +0.412 |
ERG | +0.407 |
ident | +0.407 |
OLD | +0.357 |
Neg logit contributions | |
enburg | -0.522 |
Directions | -0.494 |
Dracula | -0.488 |
warr | -0.448 |
etheless | -0.443 |
Token | for |
Token ID | 329 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
isable | +0.049 |
ォ | +0.048 |
ERG | +0.047 |
ident | +0.047 |
OLD | +0.042 |
Neg logit contributions | |
enburg | -0.059 |
Directions | -0.056 |
Dracula | -0.055 |
warr | -0.051 |
etheless | -0.050 |
Token | things |
Token ID | 1243 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
isable | +0.360 |
ォ | +0.349 |
ERG | +0.345 |
ident | +0.345 |
OLD | +0.303 |
Neg logit contributions | |
enburg | -0.433 |
Directions | -0.409 |
Dracula | -0.404 |
warr | -0.371 |
etheless | -0.365 |
Token | back |
Token ID | 736 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
enburg | +0.217 |
Directions | +0.205 |
Dracula | +0.202 |
warr | +0.187 |
etheless | +0.185 |
Neg logit contributions | |
isable | -0.160 |
ォ | -0.155 |
ident | -0.154 |
ERG | -0.153 |
OLD | -0.133 |
Token | ' |
Token ID | 6 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
isable | +0.296 |
ident | +0.285 |
ォ | +0.284 |
ERG | +0.284 |
camp | +0.251 |
Neg logit contributions | |
enburg | -0.329 |
Directions | -0.310 |
Dracula | -0.305 |
warr | -0.281 |
etheless | -0.279 |
Token | the |
Token ID | 262 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
isable | +0.526 |
ident | +0.506 |
ォ | +0.502 |
ERG | +0.502 |
camp | +0.443 |
Neg logit contributions | |
enburg | -0.610 |
Directions | -0.574 |
Dracula | -0.565 |
warr | -0.522 |
etheless | -0.518 |
Token | web |
Token ID | 3992 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
isable | +0.376 |
ォ | +0.362 |
ident | +0.362 |
ERG | +0.360 |
camp | +0.319 |
Neg logit contributions | |
enburg | -0.407 |
Directions | -0.383 |
Dracula | -0.376 |
warr | -0.346 |
etheless | -0.344 |
Token | A |
Token ID | 317 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
isable | +1.293 |
ident | +1.241 |
ォ | +1.240 |
ERG | +1.234 |
OLD | +1.081 |
Neg logit contributions | |
enburg | -1.621 |
Directions | -1.526 |
Dracula | -1.509 |
warr | -1.397 |
etheless | -1.383 |
Token | Downing |
Token ID | 31355 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
isable | +1.285 |
ォ | +1.242 |
ident | +1.240 |
ERG | +1.236 |
OLD | +1.099 |
Neg logit contributions | |
enburg | -1.354 |
Directions | -1.274 |
Dracula | -1.252 |
warr | -1.149 |
etheless | -1.133 |
Token | Street |
Token ID | 3530 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
isable | +4.922 |
ident | +4.826 |
ERG | +4.740 |
ォ | +4.570 |
camp | +4.392 |
Neg logit contributions | |
enburg | -5.940 |
Dracula | -5.614 |
Directions | -5.613 |
warr | -5.392 |
nces | -5.229 |
Token | petition |
Token ID | 8853 |
Feature activation | +4.62e-03 |
Pos logit contributions | |
enburg | +0.520 |
Directions | +0.490 |
Dracula | +0.482 |
warr | +0.443 |
etheless | +0.437 |
Neg logit contributions | |
isable | -0.460 |
ォ | -0.445 |
ident | -0.442 |
ERG | -0.441 |
OLD | -0.390 |
Token | is |
Token ID | 318 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
isable | +0.009 |
ォ | +0.009 |
ident | +0.009 |
ERG | +0.009 |
OLD | +0.008 |
Neg logit contributions | |
enburg | -0.009 |
Directions | -0.009 |
Dracula | -0.009 |
warr | -0.008 |
etheless | -0.008 |
Token | calling |
Token ID | 4585 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
enburg | +0.912 |
Directions | +0.852 |
Dracula | +0.836 |
warr | +0.756 |
etheless | +0.743 |
Neg logit contributions | |
isable | -1.094 |
ォ | -1.063 |
ident | -1.057 |
ERG | -1.055 |
OLD | -0.950 |
Token | for |
Token ID | 329 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
isable | +0.321 |
ォ | +0.310 |
ident | +0.308 |
ERG | +0.307 |
OLD | +0.269 |
Neg logit contributions | |
enburg | -0.408 |
Directions | -0.386 |
Dracula | -0.380 |
warr | -0.351 |
etheless | -0.347 |
Token | performance |
Token ID | 2854 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
enburg | +0.247 |
Directions | +0.232 |
Dracula | +0.229 |
warr | +0.209 |
etheless | +0.206 |
Neg logit contributions | |
isable | -0.236 |
ォ | -0.229 |
ident | -0.227 |
ERG | -0.227 |
OLD | -0.201 |
Token | , |
Token ID | 11 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
enburg | +0.193 |
Directions | +0.184 |
Dracula | +0.181 |
warr | +0.168 |
etheless | +0.166 |
Neg logit contributions | |
isable | -0.141 |
ォ | -0.136 |
ident | -0.135 |
ERG | -0.135 |
OLD | -0.117 |
Token | Hein |
Token ID | 26431 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
isable | +0.430 |
ident | +0.414 |
ERG | +0.414 |
ォ | +0.411 |
camp | +0.366 |
Neg logit contributions | |
enburg | -0.515 |
Directions | -0.487 |
Dracula | -0.477 |
warr | -0.442 |
etheless | -0.439 |
Token | rich |
Token ID | 7527 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
enburg | +0.126 |
Directions | +0.116 |
Dracula | +0.114 |
warr | +0.101 |
etheless | +0.099 |
Neg logit contributions | |
isable | -0.177 |
ォ | -0.173 |
ident | -0.172 |
ERG | -0.171 |
OLD | -0.155 |
Token | H |
Token ID | 367 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
isable | +1.313 |
ident | +1.266 |
ォ | +1.263 |
ERG | +1.257 |
camp | +1.131 |
Neg logit contributions | |
enburg | -1.406 |
Directions | -1.333 |
Dracula | -1.300 |
warr | -1.224 |
etheless | -1.219 |
Token | imm |
Token ID | 8608 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
isable | +1.102 |
ident | +1.069 |
ERG | +1.054 |
ォ | +1.045 |
OLD | +0.938 |
Neg logit contributions | |
enburg | -1.282 |
Directions | -1.226 |
Dracula | -1.205 |
warr | -1.125 |
etheless | -1.119 |
Token | ler |
Token ID | 1754 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ident | +4.083 |
isable | +4.082 |
ERG | +3.965 |
ォ | +3.870 |
OLD | +3.457 |
Neg logit contributions | |
enburg | -5.940 |
Directions | -5.800 |
Dracula | -5.648 |
warr | -5.433 |
etheless | -5.335 |
Token | declared |
Token ID | 6875 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
isable | +0.337 |
ident | +0.328 |
ERG | +0.327 |
ォ | +0.325 |
camp | +0.296 |
Neg logit contributions | |
enburg | -0.295 |
Directions | -0.279 |
Dracula | -0.272 |
warr | -0.252 |
etheless | -0.248 |
Token | the |
Token ID | 262 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
enburg | +0.486 |
Directions | +0.459 |
Dracula | +0.451 |
warr | +0.417 |
etheless | +0.412 |
Neg logit contributions | |
isable | -0.396 |
ォ | -0.381 |
ident | -0.380 |
ERG | -0.379 |
OLD | -0.333 |
Token | unit |
Token ID | 4326 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
isable | +0.068 |
ident | +0.066 |
ERG | +0.065 |
ォ | +0.065 |
camp | +0.059 |
Neg logit contributions | |
enburg | -0.072 |
Directions | -0.068 |
Dracula | -0.067 |
warr | -0.061 |
etheless | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
isable | +0.361 |
ォ | +0.346 |
ident | +0.345 |
ERG | +0.345 |
OLD | +0.303 |
Neg logit contributions | |
enburg | -0.434 |
Directions | -0.410 |
Dracula | -0.403 |
warr | -0.372 |
etheless | -0.369 |
Token | K |
Token ID | 42 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
enburg | +0.284 |
Directions | +0.270 |
Dracula | +0.266 |
warr | +0.246 |
etheless | +0.242 |
Neg logit contributions | |
isable | -0.217 |
ォ | -0.210 |
ident | -0.207 |
ERG | -0.207 |
OLD | -0.180 |
Token | and |
Token ID | 290 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
enburg | +0.714 |
Directions | +0.675 |
Dracula | +0.666 |
warr | +0.613 |
etheless | +0.604 |
Neg logit contributions | |
isable | -0.568 |
ォ | -0.550 |
ident | -0.544 |
ERG | -0.541 |
OLD | -0.475 |
Token | Kare |
Token ID | 49396 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
isable | +0.202 |
ォ | +0.195 |
ident | +0.194 |
ERG | +0.194 |
OLD | +0.171 |
Neg logit contributions | |
enburg | -0.226 |
Directions | -0.212 |
Dracula | -0.209 |
warr | -0.192 |
etheless | -0.190 |
Token | ena |
Token ID | 8107 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
enburg | +2.126 |
Directions | +1.999 |
Dracula | +1.971 |
warr | +1.802 |
nces | +1.788 |
Neg logit contributions | |
isable | -1.819 |
ォ | -1.778 |
ident | -1.750 |
ERG | -1.748 |
OLD | -1.545 |
Token | sh |
Token ID | 427 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
isable | +1.193 |
ォ | +1.151 |
ident | +1.146 |
ERG | +1.145 |
camp | +1.009 |
Neg logit contributions | |
enburg | -1.404 |
Directions | -1.325 |
Dracula | -1.300 |
warr | -1.205 |
etheless | -1.197 |
Token | imm |
Token ID | 8608 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
enburg | +1.589 |
Directions | +1.482 |
Dracula | +1.470 |
warr | +1.340 |
nces | +1.339 |
Neg logit contributions | |
isable | -1.256 |
ォ | -1.221 |
ERG | -1.212 |
ident | -1.190 |
OLD | -1.053 |
Token | ying |
Token ID | 1112 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
isable | +4.177 |
ident | +3.993 |
ォ | +3.970 |
ERG | +3.899 |
abba | +3.460 |
Neg logit contributions | |
enburg | -5.949 |
Directions | -5.678 |
Dracula | -5.604 |
etheless | -5.307 |
warr | -5.260 |
Token | to |
Token ID | 284 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
enburg | +0.760 |
Directions | +0.720 |
Dracula | +0.711 |
warr | +0.653 |
etheless | +0.643 |
Neg logit contributions | |
isable | -0.616 |
ォ | -0.599 |
ident | -0.591 |
ERG | -0.591 |
OLD | -0.518 |
Token | Cham |
Token ID | 37790 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
enburg | +0.048 |
Directions | +0.045 |
Dracula | +0.045 |
warr | +0.041 |
nces | +0.041 |
Neg logit contributions | |
isable | -0.039 |
ォ | -0.038 |
ERG | -0.038 |
ident | -0.038 |
OLD | -0.033 |
Token | ak |
Token ID | 461 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
enburg | +0.404 |
Directions | +0.381 |
Dracula | +0.375 |
warr | +0.346 |
etheless | +0.341 |
Neg logit contributions | |
isable | -0.327 |
ォ | -0.315 |
ERG | -0.313 |
ident | -0.312 |
OLD | -0.275 |
Token | Chall |
Token ID | 10772 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
enburg | +0.441 |
Directions | +0.416 |
Dracula | +0.410 |
warr | +0.376 |
etheless | +0.371 |
Neg logit contributions | |
isable | -0.389 |
ォ | -0.377 |
ident | -0.374 |
ERG | -0.373 |
OLD | -0.330 |
Token | notifications |
Token ID | 19605 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
isable | +0.504 |
ォ | +0.487 |
ident | +0.482 |
ERG | +0.482 |
OLD | +0.420 |
Neg logit contributions | |
enburg | -0.679 |
Directions | -0.644 |
Dracula | -0.633 |
warr | -0.587 |
etheless | -0.578 |
Token | of |
Token ID | 286 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
isable | +0.482 |
ォ | +0.471 |
ident | +0.463 |
ERG | +0.462 |
abba | +0.408 |
Neg logit contributions | |
enburg | -0.544 |
Directions | -0.516 |
Dracula | -0.505 |
warr | -0.463 |
etheless | -0.456 |
Token | future |
Token ID | 2003 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
isable | +0.360 |
ォ | +0.348 |
ident | +0.346 |
ERG | +0.345 |
OLD | +0.306 |
Neg logit contributions | |
enburg | -0.392 |
Directions | -0.370 |
Dracula | -0.364 |
warr | -0.334 |
etheless | -0.329 |
Token | posts |
Token ID | 6851 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
isable | +0.433 |
ォ | +0.419 |
ident | +0.417 |
ERG | +0.417 |
OLD | +0.371 |
Neg logit contributions | |
enburg | -0.435 |
Directions | -0.409 |
Dracula | -0.403 |
warr | -0.368 |
etheless | -0.364 |
Token | , |
Token ID | 11 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
isable | +2.026 |
ident | +1.958 |
ERG | +1.947 |
ォ | +1.943 |
camp | +1.698 |
Neg logit contributions | |
enburg | -2.596 |
Directions | -2.444 |
Dracula | -2.426 |
warr | -2.255 |
etheless | -2.225 |
Token | click |
Token ID | 3904 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
isable | +0.352 |
ォ | +0.339 |
ident | +0.335 |
ERG | +0.335 |
OLD | +0.289 |
Neg logit contributions | |
enburg | -0.520 |
Directions | -0.493 |
Dracula | -0.487 |
warr | -0.452 |
etheless | -0.446 |
Token | on |
Token ID | 319 |
Feature activation | +6.80e-04 |
Pos logit contributions | |
isable | +3.625 |
ERG | +3.437 |
ident | +3.430 |
ォ | +3.390 |
OLD | +2.981 |
Neg logit contributions | |
enburg | -5.914 |
Directions | -5.493 |
Dracula | -5.477 |
etheless | -5.185 |
warr | -5.147 |
Token | the |
Token ID | 262 |
Feature activation | -9.82e-04 |
Pos logit contributions | |
isable | +0.001 |
ォ | +0.001 |
ident | +0.001 |
ERG | +0.001 |
abba | +0.001 |
Neg logit contributions | |
enburg | -0.002 |
Directions | -0.001 |
Dracula | -0.001 |
warr | -0.001 |
etheless | -0.001 |
Token | � |
Token ID | 564 |
Feature activation | +8.73e-02 |
Pos logit contributions | |
enburg | +0.002 |
Directions | +0.002 |
Dracula | +0.002 |
warr | +0.002 |
etheless | +0.002 |
Neg logit contributions | |
isable | -0.002 |
ォ | -0.002 |
ident | -0.002 |
ERG | -0.002 |
abba | -0.001 |
Token | � |
Token ID | 250 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
isable | +0.179 |
ォ | +0.174 |
ident | +0.172 |
ERG | +0.172 |
OLD | +0.154 |
Neg logit contributions | |
enburg | -0.175 |
Directions | -0.164 |
Dracula | -0.161 |
warr | -0.147 |
etheless | -0.145 |
Token | Follow |
Token ID | 7155 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
isable | +0.132 |
ォ | +0.128 |
ident | +0.126 |
ERG | +0.126 |
OLD | +0.111 |
Neg logit contributions | |
enburg | -0.160 |
Directions | -0.152 |
Dracula | -0.149 |
warr | -0.137 |
etheless | -0.135 |
Token | all |
Token ID | 477 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
enburg | +1.046 |
Directions | +0.989 |
Dracula | +0.974 |
warr | +0.890 |
nces | +0.878 |
Neg logit contributions | |
isable | -0.916 |
ォ | -0.889 |
ident | -0.873 |
ERG | -0.872 |
OLD | -0.771 |
Token | Money |
Token ID | 12911 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
enburg | +0.035 |
Directions | +0.034 |
Dracula | +0.033 |
warr | +0.030 |
etheless | +0.030 |
Neg logit contributions | |
isable | -0.030 |
ォ | -0.029 |
ident | -0.028 |
ERG | -0.028 |
OLD | -0.025 |
Token | Morning |
Token ID | 14410 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
isable | +0.855 |
ォ | +0.827 |
ident | +0.819 |
ERG | +0.817 |
OLD | +0.715 |
Neg logit contributions | |
enburg | -1.089 |
Directions | -1.031 |
Dracula | -1.016 |
warr | -0.937 |
etheless | -0.924 |
Token | content |
Token ID | 2695 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
enburg | +0.754 |
Directions | +0.712 |
Dracula | +0.699 |
warr | +0.639 |
etheless | +0.628 |
Neg logit contributions | |
isable | -0.711 |
ォ | -0.688 |
ident | -0.681 |
ERG | -0.679 |
OLD | -0.603 |
Token | , |
Token ID | 11 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
isable | +1.929 |
ident | +1.861 |
ERG | +1.847 |
ォ | +1.840 |
camp | +1.608 |
Neg logit contributions | |
enburg | -2.802 |
Directions | -2.652 |
Dracula | -2.626 |
warr | -2.458 |
etheless | -2.430 |
Token | click |
Token ID | 3904 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
isable | +0.492 |
ォ | +0.471 |
ident | +0.470 |
ERG | +0.469 |
OLD | +0.401 |
Neg logit contributions | |
enburg | -0.791 |
Directions | -0.751 |
Dracula | -0.741 |
warr | -0.691 |
etheless | -0.685 |
Token | here |
Token ID | 994 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
isable | +3.464 |
ident | +3.298 |
ERG | +3.296 |
ォ | +3.223 |
ety | +2.850 |
Neg logit contributions | |
enburg | -5.763 |
Directions | -5.354 |
Dracula | -5.333 |
etheless | -5.107 |
warr | -5.065 |
Token |
|
Token ID | 198 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
isable | +0.320 |
ォ | +0.308 |
ident | +0.307 |
ERG | +0.306 |
OLD | +0.268 |
Neg logit contributions | |
enburg | -0.407 |
Directions | -0.383 |
Dracula | -0.378 |
warr | -0.350 |
etheless | -0.346 |
Token |
|
Token ID | 198 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
isable | +0.208 |
ident | +0.201 |
ERG | +0.199 |
ォ | +0.197 |
OLD | +0.173 |
Neg logit contributions | |
enburg | -0.312 |
Directions | -0.297 |
Dracula | -0.292 |
warr | -0.284 |
etheless | -0.280 |
Token | About |
Token ID | 8585 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
enburg | +0.180 |
Directions | +0.170 |
Dracula | +0.167 |
warr | +0.154 |
etheless | +0.152 |
Neg logit contributions | |
isable | -0.149 |
ォ | -0.144 |
ident | -0.143 |
ERG | -0.143 |
OLD | -0.126 |
Token | Money |
Token ID | 12911 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
enburg | +2.422 |
Directions | +2.328 |
Dracula | +2.288 |
warr | +2.064 |
etheless | +2.035 |
Neg logit contributions | |
isable | -2.206 |
ォ | -2.141 |
ERG | -2.098 |
ident | -2.083 |
OLD | -1.869 |
Token | raising |
Token ID | 8620 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
enburg | +0.054 |
Directions | +0.052 |
Dracula | +0.051 |
warr | +0.047 |
etheless | +0.046 |
Neg logit contributions | |
isable | -0.043 |
ォ | -0.041 |
ident | -0.041 |
ERG | -0.041 |
OLD | -0.036 |
Token | real |
Token ID | 1103 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
isable | +0.656 |
ォ | +0.636 |
ident | +0.634 |
ERG | +0.631 |
camp | +0.568 |
Neg logit contributions | |
enburg | -0.586 |
Directions | -0.550 |
Dracula | -0.538 |
warr | -0.491 |
etheless | -0.486 |
Token | doubts |
Token ID | 17188 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
enburg | +1.338 |
Directions | +1.254 |
Dracula | +1.236 |
warr | +1.128 |
nces | +1.110 |
Neg logit contributions | |
isable | -1.209 |
ォ | -1.181 |
ERG | -1.169 |
ident | -1.166 |
OLD | -1.034 |
Token | about |
Token ID | 546 |
Feature activation | -6.81e-02 |
Pos logit contributions | |
isable | +0.257 |
ォ | +0.249 |
ident | +0.247 |
ERG | +0.247 |
OLD | +0.217 |
Neg logit contributions | |
enburg | -0.308 |
Directions | -0.291 |
Dracula | -0.287 |
warr | -0.264 |
etheless | -0.261 |
Token | their |
Token ID | 511 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
enburg | +0.158 |
Directions | +0.150 |
Dracula | +0.148 |
warr | +0.137 |
etheless | +0.135 |
Neg logit contributions | |
isable | -0.118 |
ォ | -0.113 |
ident | -0.113 |
ERG | -0.112 |
OLD | -0.098 |
Token | impartial |
Token ID | 32521 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
isable | +0.025 |
ォ | +0.024 |
ident | +0.024 |
ERG | +0.024 |
OLD | +0.021 |
Neg logit contributions | |
enburg | -0.025 |
Directions | -0.024 |
Dracula | -0.023 |
warr | -0.021 |
etheless | -0.021 |
Token | ity |
Token ID | 414 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
isable | +3.474 |
ident | +3.384 |
ERG | +3.266 |
ォ | +3.236 |
camp | +2.862 |
Neg logit contributions | |
enburg | -5.731 |
Directions | -5.480 |
Dracula | -5.433 |
warr | -5.084 |
etheless | -5.078 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
isable | +1.697 |
ident | +1.635 |
ォ | +1.629 |
ERG | +1.621 |
camp | +1.421 |
Neg logit contributions | |
enburg | -2.253 |
Directions | -2.133 |
Dracula | -2.101 |
warr | -1.956 |
etheless | -1.921 |
Token | For |
Token ID | 1114 |
Feature activation | -1.29e-03 |
Pos logit contributions | |
isable | +0.335 |
ォ | +0.324 |
ident | +0.321 |
ERG | +0.320 |
OLD | +0.280 |
Neg logit contributions | |
enburg | -0.435 |
Directions | -0.411 |
Dracula | -0.405 |
warr | -0.375 |
etheless | -0.370 |
Token | instance |
Token ID | 4554 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
enburg | +0.002 |
Directions | +0.002 |
Dracula | +0.002 |
warr | +0.002 |
etheless | +0.002 |
Neg logit contributions | |
isable | -0.003 |
ォ | -0.003 |
ident | -0.003 |
ERG | -0.003 |
OLD | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
isable | +0.667 |
ォ | +0.642 |
ident | +0.639 |
ERG | +0.637 |
OLD | +0.551 |
Neg logit contributions | |
enburg | -0.994 |
Directions | -0.949 |
Dracula | -0.933 |
warr | -0.874 |
etheless | -0.862 |
Token | f |
Token ID | 277 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
isable | +2.165 |
ident | +2.096 |
ォ | +2.087 |
ERG | +2.063 |
camp | +1.830 |
Neg logit contributions | |
enburg | -2.497 |
Directions | -2.367 |
Dracula | -2.327 |
warr | -2.139 |
nces | -2.110 |
Token | ait |
Token ID | 4548 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
enburg | +0.461 |
Directions | +0.435 |
Dracula | +0.427 |
warr | +0.395 |
nces | +0.394 |
Neg logit contributions | |
isable | -0.289 |
ォ | -0.285 |
ERG | -0.277 |
ident | -0.273 |
OLD | -0.239 |
Token | , |
Token ID | 11 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
isable | +0.786 |
ォ | +0.759 |
ident | +0.756 |
ERG | +0.753 |
OLD | +0.666 |
Neg logit contributions | |
enburg | -0.870 |
Directions | -0.819 |
Dracula | -0.806 |
warr | -0.740 |
etheless | -0.731 |
Token | il |
Token ID | 4229 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
isable | +0.301 |
ォ | +0.292 |
ident | +0.289 |
ERG | +0.289 |
OLD | +0.255 |
Neg logit contributions | |
enburg | -0.349 |
Directions | -0.329 |
Dracula | -0.324 |
warr | -0.298 |
etheless | -0.294 |
Token | s |
Token ID | 82 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
isable | +1.142 |
ォ | +1.101 |
ident | +1.099 |
ERG | +1.091 |
camp | +0.964 |
Neg logit contributions | |
enburg | -1.305 |
Directions | -1.227 |
Dracula | -1.211 |
warr | -1.112 |
etheless | -1.101 |
Token | ont |
Token ID | 39585 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
isable | +1.034 |
ォ | +0.998 |
ident | +0.996 |
ERG | +0.989 |
camp | +0.875 |
Neg logit contributions | |
enburg | -1.179 |
Directions | -1.109 |
Dracula | -1.092 |
warr | -1.003 |
etheless | -0.991 |
Token | sci |
Token ID | 20681 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
isable | +4.314 |
ident | +4.187 |
ォ | +4.114 |
ERG | +4.052 |
camp | +3.754 |
Neg logit contributions | |
enburg | -4.768 |
Directions | -4.435 |
Dracula | -4.364 |
warr | -4.005 |
nces | -3.991 |
Token | em |
Token ID | 368 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
enburg | +0.616 |
Directions | +0.577 |
Dracula | +0.565 |
nces | +0.520 |
warr | +0.518 |
Neg logit contributions | |
isable | -0.510 |
ォ | -0.508 |
ERG | -0.497 |
ident | -0.485 |
OLD | -0.438 |
Token | ment |
Token ID | 434 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
isable | +1.528 |
ident | +1.472 |
ォ | +1.448 |
ERG | +1.440 |
abba | +1.254 |
Neg logit contributions | |
enburg | -2.065 |
Directions | -1.972 |
Dracula | -1.941 |
warr | -1.807 |
etheless | -1.768 |
Token | br |
Token ID | 865 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
enburg | +0.499 |
Directions | +0.472 |
Dracula | +0.464 |
warr | +0.426 |
nces | +0.421 |
Neg logit contributions | |
isable | -0.431 |
ォ | -0.421 |
ERG | -0.416 |
ident | -0.412 |
OLD | -0.367 |
Token | is |
Token ID | 271 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
isable | +0.760 |
ォ | +0.737 |
ERG | +0.729 |
ident | +0.725 |
OLD | +0.636 |
Neg logit contributions | |
enburg | -1.022 |
Directions | -0.966 |
Dracula | -0.952 |
warr | -0.878 |
etheless | -0.869 |
Token | of |
Token ID | 286 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
isable | +1.195 |
ォ | +1.152 |
ident | +1.146 |
ERG | +1.141 |
OLD | +0.998 |
Neg logit contributions | |
enburg | -1.536 |
Directions | -1.453 |
Dracula | -1.433 |
warr | -1.324 |
etheless | -1.310 |
Token | today |
Token ID | 1909 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
isable | +0.235 |
ォ | +0.228 |
ident | +0.226 |
ERG | +0.225 |
OLD | +0.199 |
Neg logit contributions | |
enburg | -0.276 |
Directions | -0.261 |
Dracula | -0.257 |
warr | -0.237 |
etheless | -0.234 |
Token | 's |
Token ID | 338 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
isable | +1.308 |
ォ | +1.263 |
ident | +1.252 |
ERG | +1.250 |
OLD | +1.089 |
Neg logit contributions | |
enburg | -1.771 |
Directions | -1.679 |
Dracula | -1.655 |
warr | -1.532 |
etheless | -1.513 |
Token | photos |
Token ID | 5205 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
isable | +1.065 |
ident | +1.022 |
ォ | +1.021 |
ERG | +1.016 |
camp | +0.913 |
Neg logit contributions | |
enburg | -1.193 |
Directions | -1.123 |
Dracula | -1.112 |
warr | -1.031 |
etheless | -1.019 |
Token | , |
Token ID | 11 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
isable | +0.860 |
ォ | +0.825 |
ident | +0.823 |
ERG | +0.818 |
OLD | +0.708 |
Neg logit contributions | |
enburg | -1.264 |
Directions | -1.197 |
Dracula | -1.182 |
warr | -1.097 |
etheless | -1.088 |
Token | click |
Token ID | 3904 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
isable | +0.554 |
ォ | +0.528 |
ident | +0.528 |
ERG | +0.524 |
camp | +0.446 |
Neg logit contributions | |
enburg | -0.936 |
Directions | -0.891 |
Dracula | -0.879 |
warr | -0.820 |
etheless | -0.815 |
Token | here |
Token ID | 994 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
isable | +3.408 |
ERG | +3.176 |
ident | +3.170 |
ォ | +3.145 |
camp | +2.846 |
Neg logit contributions | |
enburg | -5.581 |
Directions | -5.194 |
Dracula | -5.173 |
etheless | -4.930 |
warr | -4.922 |
Token | : |
Token ID | 25 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
isable | +0.081 |
ォ | +0.078 |
ident | +0.077 |
ERG | +0.077 |
OLD | +0.068 |
Neg logit contributions | |
enburg | -0.100 |
Directions | -0.095 |
Dracula | -0.093 |
warr | -0.086 |
etheless | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | -6.80e-03 |
Pos logit contributions | |
enburg | +0.631 |
Directions | +0.599 |
Dracula | +0.589 |
warr | +0.540 |
etheless | +0.533 |
Neg logit contributions | |
isable | -0.517 |
ォ | -0.500 |
ident | -0.495 |
ERG | -0.492 |
OLD | -0.433 |
Token |
|
Token ID | 198 |
Feature activation | -8.56e-02 |
Pos logit contributions | |
enburg | +0.016 |
Directions | +0.015 |
Dracula | +0.015 |
warr | +0.015 |
etheless | +0.014 |
Neg logit contributions | |
isable | -0.011 |
ident | -0.011 |
ERG | -0.011 |
ォ | -0.011 |
OLD | -0.009 |
Token | Follow |
Token ID | 7155 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
enburg | +0.184 |
Directions | +0.173 |
Dracula | +0.170 |
warr | +0.155 |
nces | +0.153 |
Neg logit contributions | |
isable | -0.164 |
ォ | -0.158 |
ident | -0.156 |
ERG | -0.156 |
abba | -0.138 |
Token | poll |
Token ID | 3278 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
isable | +0.453 |
ォ | +0.440 |
ident | +0.438 |
ERG | +0.436 |
OLD | +0.389 |
Neg logit contributions | |
enburg | -0.477 |
Directions | -0.448 |
Dracula | -0.441 |
warr | -0.406 |
etheless | -0.400 |
Token | average |
Token ID | 2811 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
isable | +0.961 |
ォ | +0.933 |
ident | +0.932 |
ERG | +0.923 |
camp | +0.814 |
Neg logit contributions | |
enburg | -1.342 |
Directions | -1.277 |
Dracula | -1.258 |
warr | -1.178 |
etheless | -1.159 |
Token | . |
Token ID | 13 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
enburg | +0.272 |
Directions | +0.258 |
Dracula | +0.254 |
warr | +0.235 |
etheless | +0.232 |
Neg logit contributions | |
isable | -0.205 |
ォ | -0.198 |
ident | -0.197 |
ERG | -0.196 |
OLD | -0.171 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
enburg | +0.733 |
Directions | +0.695 |
Dracula | +0.682 |
warr | +0.615 |
nces | +0.611 |
Neg logit contributions | |
isable | -0.662 |
ォ | -0.637 |
ident | -0.631 |
ERG | -0.630 |
abba | -0.559 |
Token | H |
Token ID | 39 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
enburg | +0.336 |
Directions | +0.318 |
Dracula | +0.313 |
warr | +0.288 |
etheless | +0.284 |
Neg logit contributions | |
isable | -0.267 |
ォ | -0.258 |
ident | -0.256 |
ERG | -0.255 |
OLD | -0.223 |
Token | omew |
Token ID | 28030 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
enburg | +0.108 |
Directions | +0.102 |
Dracula | +0.100 |
warr | +0.093 |
nces | +0.091 |
Neg logit contributions | |
isable | -0.078 |
ォ | -0.075 |
ident | -0.074 |
ERG | -0.073 |
camp | -0.064 |
Token | orld |
Token ID | 1764 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
isable | +2.475 |
ident | +2.387 |
ォ | +2.323 |
ERG | +2.284 |
OLD | +1.881 |
Neg logit contributions | |
enburg | -6.407 |
Directions | -6.148 |
Dracula | -6.062 |
warr | -5.749 |
etheless | -5.688 |
Token | : |
Token ID | 25 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
isable | +1.392 |
ォ | +1.340 |
ident | +1.340 |
ERG | +1.338 |
OLD | +1.181 |
Neg logit contributions | |
enburg | -1.579 |
Directions | -1.481 |
Dracula | -1.458 |
warr | -1.356 |
etheless | -1.340 |
Token | Game |
Token ID | 3776 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
enburg | +0.648 |
Directions | +0.612 |
Dracula | +0.604 |
warr | +0.548 |
etheless | +0.542 |
Neg logit contributions | |
isable | -0.586 |
ォ | -0.567 |
ident | -0.561 |
ERG | -0.561 |
OLD | -0.496 |
Token | of |
Token ID | 286 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
enburg | +1.803 |
Directions | +1.705 |
Dracula | +1.671 |
warr | +1.527 |
nces | +1.498 |
Neg logit contributions | |
isable | -1.569 |
ォ | -1.523 |
ident | -1.511 |
ERG | -1.500 |
OLD | -1.331 |
Token | the |
Token ID | 262 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
isable | +0.303 |
ォ | +0.293 |
ident | +0.291 |
ERG | +0.290 |
OLD | +0.256 |
Neg logit contributions | |
enburg | -0.353 |
Directions | -0.332 |
Dracula | -0.326 |
warr | -0.301 |
etheless | -0.298 |
Token | spared |
Token ID | 31348 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
enburg | +0.097 |
Directions | +0.092 |
Dracula | +0.090 |
warr | +0.082 |
etheless | +0.081 |
Neg logit contributions | |
isable | -0.091 |
ォ | -0.088 |
ident | -0.087 |
ERG | -0.087 |
OLD | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +8.65e-02 |
Pos logit contributions | |
enburg | +0.292 |
Directions | +0.276 |
Dracula | +0.272 |
warr | +0.250 |
etheless | +0.247 |
Neg logit contributions | |
isable | -0.246 |
ォ | -0.239 |
ident | -0.236 |
ERG | -0.236 |
OLD | -0.208 |
Token | As |
Token ID | 1081 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
isable | +0.160 |
ォ | +0.154 |
ident | +0.154 |
ERG | +0.153 |
OLD | +0.135 |
Neg logit contributions | |
enburg | -0.190 |
Directions | -0.179 |
Dracula | -0.176 |
warr | -0.164 |
etheless | -0.162 |
Token | for |
Token ID | 329 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
enburg | +0.798 |
Directions | +0.752 |
Dracula | +0.741 |
warr | +0.681 |
etheless | +0.671 |
Neg logit contributions | |
isable | -0.702 |
ォ | -0.681 |
ident | -0.674 |
ERG | -0.674 |
OLD | -0.595 |
Token | the |
Token ID | 262 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
enburg | +0.731 |
Directions | +0.690 |
Dracula | +0.679 |
warr | +0.625 |
etheless | +0.616 |
Neg logit contributions | |
isable | -0.632 |
ォ | -0.612 |
ident | -0.607 |
ERG | -0.606 |
OLD | -0.535 |
Token | spo |
Token ID | 15106 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
enburg | +0.037 |
Directions | +0.035 |
Dracula | +0.035 |
warr | +0.032 |
etheless | +0.031 |
Neg logit contributions | |
isable | -0.034 |
ォ | -0.032 |
ident | -0.032 |
ERG | -0.032 |
OLD | -0.028 |
Token | ils |
Token ID | 4487 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
isable | +1.639 |
ERG | +1.468 |
ident | +1.462 |
ォ | +1.351 |
OLD | +1.038 |
Neg logit contributions | |
enburg | -7.251 |
Directions | -7.005 |
Dracula | -6.921 |
warr | -6.688 |
etheless | -6.675 |
Token | of |
Token ID | 286 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
isable | +1.099 |
ォ | +1.054 |
ident | +1.051 |
ERG | +1.048 |
OLD | +0.919 |
Neg logit contributions | |
enburg | -1.426 |
Directions | -1.352 |
Dracula | -1.327 |
warr | -1.229 |
etheless | -1.218 |
Token | the |
Token ID | 262 |
Feature activation | -7.33e-02 |
Pos logit contributions | |
enburg | +0.219 |
Directions | +0.205 |
Dracula | +0.201 |
warr | +0.183 |
etheless | +0.180 |
Neg logit contributions | |
isable | -0.244 |
ォ | -0.237 |
ident | -0.236 |
ERG | -0.235 |
OLD | -0.211 |
Token | war |
Token ID | 1175 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
enburg | +0.147 |
Directions | +0.138 |
Dracula | +0.136 |
warr | +0.124 |
etheless | +0.122 |
Neg logit contributions | |
isable | -0.150 |
ォ | -0.146 |
ident | -0.145 |
ERG | -0.145 |
OLD | -0.129 |
Token | , |
Token ID | 11 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
isable | +0.073 |
ォ | +0.071 |
ident | +0.070 |
ERG | +0.070 |
OLD | +0.061 |
Neg logit contributions | |
enburg | -0.096 |
Directions | -0.091 |
Dracula | -0.089 |
warr | -0.083 |
etheless | -0.081 |
Token | But |
Token ID | 1537 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
enburg | +0.189 |
Directions | +0.178 |
Dracula | +0.176 |
warr | +0.161 |
nces | +0.160 |
Neg logit contributions | |
isable | -0.146 |
ォ | -0.142 |
ident | -0.139 |
ERG | -0.139 |
abba | -0.122 |
Token | the |
Token ID | 262 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
isable | +0.150 |
ォ | +0.145 |
ident | +0.144 |
ERG | +0.143 |
OLD | +0.126 |
Neg logit contributions | |
enburg | -0.181 |
Directions | -0.171 |
Dracula | -0.168 |
warr | -0.155 |
etheless | -0.153 |
Token | people |
Token ID | 661 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
enburg | +0.307 |
Directions | +0.290 |
Dracula | +0.286 |
warr | +0.262 |
etheless | +0.258 |
Neg logit contributions | |
isable | -0.273 |
ォ | -0.266 |
ident | -0.263 |
ERG | -0.263 |
OLD | -0.232 |
Token | scoring |
Token ID | 9689 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
enburg | +1.451 |
Directions | +1.368 |
Dracula | +1.352 |
warr | +1.235 |
etheless | +1.214 |
Neg logit contributions | |
isable | -1.280 |
ォ | -1.248 |
ident | -1.226 |
ERG | -1.225 |
OLD | -1.079 |
Token | the |
Token ID | 262 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
enburg | +1.107 |
Directions | +1.048 |
Dracula | +1.045 |
warr | +0.957 |
nces | +0.934 |
Neg logit contributions | |
isable | -0.809 |
ォ | -0.791 |
ERG | -0.776 |
ident | -0.776 |
OLD | -0.675 |
Token | carc |
Token ID | 36756 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
isable | +0.052 |
ォ | +0.050 |
ERG | +0.050 |
ident | +0.050 |
OLD | +0.042 |
Neg logit contributions | |
enburg | -0.085 |
Directions | -0.080 |
Dracula | -0.080 |
warr | -0.074 |
etheless | -0.073 |
Token | asses |
Token ID | 13978 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
isable | +3.381 |
ident | +3.285 |
ERG | +3.125 |
ォ | +3.111 |
abba | +2.749 |
Neg logit contributions | |
enburg | -5.431 |
Directions | -5.293 |
Dracula | -5.114 |
warr | -5.001 |
etheless | -4.809 |
Token | are |
Token ID | 389 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
isable | +0.066 |
ォ | +0.064 |
ident | +0.063 |
ERG | +0.063 |
OLD | +0.055 |
Neg logit contributions | |
enburg | -0.085 |
Directions | -0.081 |
Dracula | -0.080 |
warr | -0.073 |
etheless | -0.072 |
Token | employees |
Token ID | 4409 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
enburg | +0.627 |
Directions | +0.591 |
Dracula | +0.582 |
warr | +0.535 |
etheless | +0.528 |
Neg logit contributions | |
isable | -0.555 |
ォ | -0.537 |
ident | -0.533 |
ERG | -0.532 |
OLD | -0.470 |
Token | of |
Token ID | 286 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
isable | +0.270 |
ォ | +0.262 |
ident | +0.259 |
ERG | +0.259 |
OLD | +0.227 |
Neg logit contributions | |
enburg | -0.334 |
Directions | -0.315 |
Dracula | -0.311 |
warr | -0.286 |
etheless | -0.282 |
Token | the |
Token ID | 262 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
isable | +0.332 |
ォ | +0.322 |
ident | +0.318 |
ERG | +0.318 |
OLD | +0.279 |
Neg logit contributions | |
enburg | -0.407 |
Directions | -0.385 |
Dracula | -0.380 |
warr | -0.350 |
etheless | -0.345 |
Token | " |
Token ID | 1 |
Feature activation | +8.24e-02 |
Pos logit contributions | |
isable | +1.859 |
ident | +1.800 |
ERG | +1.792 |
ォ | +1.790 |
OLD | +1.597 |
Neg logit contributions | |
enburg | -1.992 |
Directions | -1.876 |
Dracula | -1.835 |
warr | -1.710 |
etheless | -1.701 |
Token | or |
Token ID | 393 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
isable | +0.154 |
ォ | +0.149 |
ident | +0.148 |
ERG | +0.147 |
OLD | +0.130 |
Neg logit contributions | |
enburg | -0.180 |
Directions | -0.170 |
Dracula | -0.168 |
warr | -0.154 |
etheless | -0.152 |
Token | " |
Token ID | 366 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
isable | +0.864 |
ォ | +0.831 |
ident | +0.831 |
ERG | +0.826 |
camp | +0.728 |
Neg logit contributions | |
enburg | -1.068 |
Directions | -1.009 |
Dracula | -0.991 |
warr | -0.922 |
etheless | -0.909 |
Token | The |
Token ID | 464 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
isable | +1.292 |
ident | +1.255 |
ERG | +1.247 |
ォ | +1.246 |
OLD | +1.100 |
Neg logit contributions | |
enburg | -1.586 |
Directions | -1.502 |
Dracula | -1.473 |
warr | -1.377 |
etheless | -1.356 |
Token | Fl |
Token ID | 1610 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
isable | +0.593 |
ォ | +0.573 |
ident | +0.571 |
ERG | +0.570 |
OLD | +0.508 |
Neg logit contributions | |
enburg | -0.590 |
Directions | -0.553 |
Dracula | -0.542 |
warr | -0.500 |
etheless | -0.494 |
Token | im |
Token ID | 320 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
isable | +0.349 |
ident | +0.336 |
ォ | +0.334 |
ERG | +0.333 |
abba | +0.295 |
Neg logit contributions | |
enburg | -0.415 |
Directions | -0.395 |
Dracula | -0.388 |
warr | -0.362 |
etheless | -0.356 |
Token | Flam |
Token ID | 33104 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
isable | +3.992 |
ident | +3.846 |
ォ | +3.832 |
ERG | +3.786 |
camp | +3.383 |
Neg logit contributions | |
enburg | -4.780 |
Directions | -4.565 |
Dracula | -4.432 |
etheless | -4.220 |
warr | -4.200 |
Token | Bros |
Token ID | 14266 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
enburg | +1.865 |
Directions | +1.750 |
Dracula | +1.731 |
warr | +1.572 |
nces | +1.566 |
Neg logit contributions | |
isable | -1.575 |
ォ | -1.527 |
ERG | -1.505 |
ident | -1.499 |
OLD | -1.336 |
Token | " |
Token ID | 1 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
enburg | +2.559 |
Directions | +2.420 |
Dracula | +2.394 |
nces | +2.166 |
warr | +2.162 |
Neg logit contributions | |
isable | -2.137 |
ォ | -2.072 |
ERG | -2.040 |
ident | -2.022 |
OLD | -1.793 |
Token | or |
Token ID | 393 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
enburg | +0.673 |
Directions | +0.637 |
Dracula | +0.627 |
warr | +0.574 |
nces | +0.568 |
Neg logit contributions | |
isable | -0.554 |
ォ | -0.538 |
ERG | -0.530 |
ident | -0.530 |
OLD | -0.466 |
Token | " |
Token ID | 366 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
isable | +0.465 |
ォ | +0.448 |
ident | +0.446 |
ERG | +0.445 |
OLD | +0.391 |
Neg logit contributions | |
enburg | -0.565 |
Directions | -0.534 |
Dracula | -0.525 |
warr | -0.486 |
etheless | -0.480 |
Token | the |
Token ID | 262 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
enburg | +0.352 |
Directions | +0.333 |
Dracula | +0.328 |
warr | +0.303 |
etheless | +0.299 |
Neg logit contributions | |
isable | -0.283 |
ォ | -0.274 |
ident | -0.271 |
ERG | -0.271 |
OLD | -0.238 |
Token | timing |
Token ID | 10576 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
enburg | +0.358 |
Directions | +0.338 |
Dracula | +0.332 |
warr | +0.305 |
etheless | +0.301 |
Neg logit contributions | |
isable | -0.324 |
ォ | -0.314 |
ident | -0.311 |
ERG | -0.311 |
OLD | -0.275 |
Token | was |
Token ID | 373 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
isable | +0.562 |
ォ | +0.540 |
ident | +0.540 |
ERG | +0.538 |
OLD | +0.476 |
Neg logit contributions | |
enburg | -0.639 |
Directions | -0.602 |
Dracula | -0.590 |
warr | -0.544 |
etheless | -0.541 |
Token | a |
Token ID | 257 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
enburg | +0.330 |
Directions | +0.311 |
Dracula | +0.306 |
warr | +0.282 |
etheless | +0.279 |
Neg logit contributions | |
isable | -0.280 |
ォ | -0.270 |
ident | -0.269 |
ERG | -0.268 |
OLD | -0.236 |
Token | bit |
Token ID | 1643 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
isable | +0.303 |
ォ | +0.294 |
ident | +0.292 |
ERG | +0.291 |
OLD | +0.261 |
Neg logit contributions | |
enburg | -0.280 |
Directions | -0.263 |
Dracula | -0.257 |
warr | -0.235 |
etheless | -0.231 |
Token | premature |
Token ID | 19905 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
enburg | +0.978 |
Directions | +0.925 |
Dracula | +0.917 |
warr | +0.845 |
etheless | +0.831 |
Neg logit contributions | |
isable | -0.724 |
ォ | -0.703 |
ident | -0.692 |
ERG | -0.689 |
abba | -0.601 |
Token | . |
Token ID | 13 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
isable | +3.489 |
ident | +3.304 |
ERG | +3.243 |
ォ | +3.161 |
camp | +2.870 |
Neg logit contributions | |
enburg | -5.281 |
Directions | -5.020 |
Dracula | -4.893 |
warr | -4.644 |
etheless | -4.595 |
Token | In |
Token ID | 554 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
isable | +0.060 |
ォ | +0.058 |
ident | +0.058 |
ERG | +0.058 |
OLD | +0.051 |
Neg logit contributions | |
enburg | -0.076 |
Directions | -0.071 |
Dracula | -0.070 |
warr | -0.065 |
etheless | -0.064 |
Token | 1968 |
Token ID | 15963 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
isable | +0.044 |
ォ | +0.042 |
ident | +0.042 |
ERG | +0.042 |
OLD | +0.036 |
Neg logit contributions | |
enburg | -0.056 |
Directions | -0.053 |
Dracula | -0.052 |
warr | -0.048 |
etheless | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
enburg | +0.984 |
Directions | +0.930 |
Dracula | +0.925 |
warr | +0.844 |
nces | +0.834 |
Neg logit contributions | |
isable | -0.752 |
ォ | -0.736 |
ERG | -0.720 |
ident | -0.719 |
abba | -0.631 |
Token | Martin |
Token ID | 5780 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
isable | +0.354 |
ident | +0.340 |
ERG | +0.339 |
ォ | +0.338 |
OLD | +0.297 |
Neg logit contributions | |
enburg | -0.434 |
Directions | -0.410 |
Dracula | -0.402 |
warr | -0.373 |
etheless | -0.370 |
Token | Enjoy |
Token ID | 18179 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
isable | +0.334 |
ォ | +0.324 |
ident | +0.322 |
ERG | +0.321 |
OLD | +0.283 |
Neg logit contributions | |
enburg | -0.398 |
Directions | -0.375 |
Dracula | -0.370 |
warr | -0.342 |
etheless | -0.338 |
Token | ! |
Token ID | 0 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
enburg | +0.373 |
Directions | +0.352 |
Dracula | +0.346 |
warr | +0.318 |
etheless | +0.314 |
Neg logit contributions | |
isable | -0.333 |
ォ | -0.322 |
ident | -0.320 |
ERG | -0.319 |
OLD | -0.282 |
Token |
|
Token ID | 198 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
enburg | +0.560 |
Directions | +0.529 |
Dracula | +0.520 |
warr | +0.478 |
etheless | +0.472 |
Neg logit contributions | |
isable | -0.488 |
ォ | -0.472 |
ident | -0.468 |
ERG | -0.468 |
OLD | -0.413 |
Token |
|
Token ID | 198 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
isable | +0.106 |
ォ | +0.102 |
ident | +0.101 |
ERG | +0.101 |
OLD | +0.088 |
Neg logit contributions | |
enburg | -0.139 |
Directions | -0.131 |
Dracula | -0.129 |
warr | -0.119 |
etheless | -0.118 |
Token | ( |
Token ID | 7 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
enburg | +0.285 |
Directions | +0.269 |
Dracula | +0.265 |
warr | +0.244 |
etheless | +0.241 |
Neg logit contributions | |
isable | -0.241 |
ォ | -0.233 |
ident | -0.232 |
ERG | -0.231 |
OLD | -0.204 |
Token | Click |
Token ID | 8164 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
enburg | +0.139 |
Directions | +0.131 |
Dracula | +0.129 |
warr | +0.119 |
etheless | +0.117 |
Neg logit contributions | |
isable | -0.113 |
ォ | -0.110 |
ident | -0.109 |
ERG | -0.109 |
OLD | -0.096 |
Token | any |
Token ID | 597 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
isable | +3.686 |
ident | +3.547 |
ERG | +3.545 |
ォ | +3.530 |
ety | +3.140 |
Neg logit contributions | |
enburg | -4.985 |
Directions | -4.626 |
Dracula | -4.596 |
etheless | -4.407 |
warr | -4.335 |
Token | of |
Token ID | 286 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
isable | +0.607 |
ォ | +0.584 |
ident | +0.581 |
ERG | +0.579 |
OLD | +0.508 |
Neg logit contributions | |
enburg | -0.767 |
Directions | -0.725 |
Dracula | -0.715 |
warr | -0.661 |
etheless | -0.653 |
Token | the |
Token ID | 262 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
isable | +0.522 |
ォ | +0.503 |
ident | +0.500 |
ERG | +0.498 |
OLD | +0.437 |
Neg logit contributions | |
enburg | -0.668 |
Directions | -0.630 |
Dracula | -0.620 |
warr | -0.575 |
etheless | -0.571 |
Token | images |
Token ID | 4263 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
isable | +0.060 |
ォ | +0.058 |
ident | +0.058 |
ERG | +0.057 |
OLD | +0.051 |
Neg logit contributions | |
enburg | -0.071 |
Directions | -0.067 |
Dracula | -0.066 |
warr | -0.061 |
etheless | -0.060 |
Token | below |
Token ID | 2174 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
isable | +0.473 |
ォ | +0.455 |
ident | +0.453 |
ERG | +0.451 |
OLD | +0.397 |
Neg logit contributions | |
enburg | -0.585 |
Directions | -0.551 |
Dracula | -0.543 |
warr | -0.505 |
etheless | -0.501 |
Token | film |
Token ID | 2646 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
isable | +0.929 |
ォ | +0.891 |
ident | +0.890 |
ERG | +0.886 |
camp | +0.785 |
Neg logit contributions | |
enburg | -1.048 |
Directions | -0.990 |
Dracula | -0.972 |
warr | -0.895 |
etheless | -0.887 |
Token | . |
Token ID | 13 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
enburg | +0.423 |
Directions | +0.399 |
Dracula | +0.395 |
warr | +0.363 |
etheless | +0.358 |
Neg logit contributions | |
isable | -0.337 |
ォ | -0.329 |
ERG | -0.323 |
ident | -0.323 |
OLD | -0.285 |
Token | That |
Token ID | 1320 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
isable | +0.394 |
ォ | +0.381 |
ident | +0.379 |
ERG | +0.377 |
OLD | +0.330 |
Neg logit contributions | |
enburg | -0.513 |
Directions | -0.483 |
Dracula | -0.476 |
warr | -0.446 |
etheless | -0.440 |
Token | [ |
Token ID | 685 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
isable | +0.619 |
ォ | +0.594 |
ident | +0.593 |
ERG | +0.591 |
camp | +0.517 |
Neg logit contributions | |
enburg | -0.796 |
Directions | -0.754 |
Dracula | -0.740 |
warr | -0.686 |
etheless | -0.680 |
Token | F |
Token ID | 37 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
enburg | +0.474 |
Directions | +0.446 |
Dracula | +0.439 |
warr | +0.403 |
etheless | +0.397 |
Neg logit contributions | |
isable | -0.426 |
ォ | -0.413 |
ident | -0.408 |
ERG | -0.408 |
OLD | -0.361 |
Token | inn |
Token ID | 3732 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
enburg | +0.749 |
Directions | +0.702 |
Dracula | +0.688 |
warr | +0.643 |
nces | +0.633 |
Neg logit contributions | |
isable | -0.525 |
ォ | -0.524 |
ident | -0.502 |
ERG | -0.498 |
abba | -0.437 |
Token | ] |
Token ID | 60 |
Feature activation | +9.72e-02 |
Pos logit contributions | |
isable | +3.647 |
ident | +3.545 |
ERG | +3.455 |
ォ | +3.436 |
camp | +3.112 |
Neg logit contributions | |
enburg | -4.765 |
Directions | -4.608 |
Dracula | -4.450 |
warr | -4.299 |
etheless | -4.278 |
Token | the |
Token ID | 262 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
isable | +0.177 |
ident | +0.170 |
ォ | +0.170 |
ERG | +0.169 |
camp | +0.150 |
Neg logit contributions | |
enburg | -0.216 |
Directions | -0.205 |
Dracula | -0.201 |
warr | -0.187 |
etheless | -0.186 |
Token | best |
Token ID | 1266 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
enburg | +0.062 |
Directions | +0.059 |
Dracula | +0.058 |
warr | +0.053 |
etheless | +0.052 |
Neg logit contributions | |
isable | -0.056 |
ォ | -0.054 |
ident | -0.054 |
ERG | -0.054 |
OLD | -0.047 |
Token | character |
Token ID | 2095 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
enburg | +0.836 |
Directions | +0.791 |
Dracula | +0.781 |
warr | +0.720 |
etheless | +0.708 |
Neg logit contributions | |
isable | -0.633 |
ォ | -0.616 |
ERG | -0.609 |
ident | -0.607 |
OLD | -0.531 |
Token | I |
Token ID | 314 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
enburg | +0.522 |
Directions | +0.493 |
Dracula | +0.485 |
warr | +0.447 |
etheless | +0.441 |
Neg logit contributions | |
isable | -0.441 |
ォ | -0.426 |
ident | -0.423 |
ERG | -0.422 |
OLD | -0.371 |
Token | knew |
Token ID | 2993 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
isable | +0.403 |
ォ | +0.392 |
ident | +0.389 |
ERG | +0.387 |
OLD | +0.339 |
Neg logit contributions | |
enburg | -0.511 |
Directions | -0.485 |
Dracula | -0.477 |
warr | -0.442 |
etheless | -0.436 |
Token | that |
Token ID | 326 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
isable | +0.375 |
ident | +0.363 |
ォ | +0.360 |
ERG | +0.359 |
camp | +0.318 |
Neg logit contributions | |
enburg | -0.470 |
Directions | -0.444 |
Dracula | -0.435 |
etheless | -0.403 |
warr | -0.402 |
Token | if |
Token ID | 611 |
Feature activation | -4.05e-03 |
Pos logit contributions | |
isable | +0.200 |
ident | +0.193 |
ォ | +0.191 |
ERG | +0.191 |
camp | +0.169 |
Neg logit contributions | |
enburg | -0.246 |
Directions | -0.232 |
Dracula | -0.227 |
etheless | -0.210 |
warr | -0.210 |
Token | we |
Token ID | 356 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
enburg | +0.009 |
Directions | +0.008 |
Dracula | +0.008 |
warr | +0.008 |
etheless | +0.008 |
Neg logit contributions | |
isable | -0.007 |
ォ | -0.007 |
ident | -0.007 |
ERG | -0.007 |
OLD | -0.006 |
Token | fl |
Token ID | 781 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
isable | +0.052 |
ォ | +0.050 |
ident | +0.050 |
ERG | +0.050 |
OLD | +0.044 |
Neg logit contributions | |
enburg | -0.060 |
Directions | -0.056 |
Dracula | -0.056 |
warr | -0.051 |
etheless | -0.050 |
Token | unk |
Token ID | 2954 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
enburg | +1.278 |
Directions | +1.202 |
Dracula | +1.189 |
warr | +1.091 |
etheless | +1.081 |
Neg logit contributions | |
isable | -0.976 |
ォ | -0.944 |
ERG | -0.934 |
ident | -0.930 |
OLD | -0.813 |
Token | ed |
Token ID | 276 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
isable | +2.739 |
ident | +2.669 |
ォ | +2.649 |
ERG | +2.645 |
camp | +2.232 |
Neg logit contributions | |
enburg | -5.636 |
Directions | -5.499 |
Dracula | -5.358 |
warr | -5.092 |
etheless | -5.065 |
Token | out |
Token ID | 503 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
isable | +0.068 |
ォ | +0.066 |
ident | +0.065 |
ERG | +0.065 |
OLD | +0.058 |
Neg logit contributions | |
enburg | -0.080 |
Directions | -0.076 |
Dracula | -0.075 |
warr | -0.069 |
etheless | -0.068 |
Token | of |
Token ID | 286 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
enburg | +0.331 |
Directions | +0.312 |
Dracula | +0.308 |
warr | +0.284 |
etheless | +0.280 |
Neg logit contributions | |
isable | -0.265 |
ォ | -0.256 |
ident | -0.254 |
ERG | -0.253 |
OLD | -0.222 |
Token | school |
Token ID | 1524 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
enburg | +0.174 |
Directions | +0.163 |
Dracula | +0.161 |
warr | +0.147 |
etheless | +0.146 |
Neg logit contributions | |
isable | -0.167 |
ォ | -0.162 |
ident | -0.161 |
ERG | -0.161 |
OLD | -0.143 |
Token | or |
Token ID | 393 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
isable | +0.521 |
ォ | +0.500 |
ident | +0.499 |
ERG | +0.498 |
camp | +0.435 |
Neg logit contributions | |
enburg | -0.704 |
Directions | -0.670 |
Dracula | -0.659 |
warr | -0.611 |
etheless | -0.606 |
Token |
|
Token ID | 198 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
isable | +0.128 |
ident | +0.123 |
ERG | +0.122 |
ォ | +0.122 |
OLD | +0.106 |
Neg logit contributions | |
enburg | -0.185 |
Directions | -0.176 |
Dracula | -0.173 |
warr | -0.166 |
etheless | -0.163 |
Token | Add |
Token ID | 4550 |
Feature activation | +9.80e-02 |
Pos logit contributions | |
enburg | +0.247 |
Directions | +0.235 |
Dracula | +0.231 |
warr | +0.213 |
nces | +0.210 |
Neg logit contributions | |
isable | -0.178 |
ォ | -0.171 |
ident | -0.168 |
ERG | -0.168 |
abba | -0.146 |
Token | Imm |
Token ID | 9543 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
isable | +0.190 |
ォ | +0.183 |
ident | +0.182 |
ERG | +0.182 |
OLD | +0.161 |
Neg logit contributions | |
enburg | -0.208 |
Directions | -0.195 |
Dracula | -0.192 |
warr | -0.177 |
etheless | -0.174 |
Token | ersion |
Token ID | 6900 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
isable | +2.503 |
ident | +2.390 |
ERG | +2.359 |
ォ | +2.335 |
OLD | +1.963 |
Neg logit contributions | |
enburg | -4.655 |
Directions | -4.472 |
Dracula | -4.417 |
warr | -4.161 |
etheless | -4.132 |
Token | G |
Token ID | 402 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
enburg | +1.725 |
Directions | +1.647 |
Dracula | +1.613 |
warr | +1.467 |
nces | +1.461 |
Neg logit contributions | |
isable | -1.499 |
ォ | -1.467 |
ident | -1.443 |
ERG | -1.438 |
OLD | -1.273 |
Token | imm |
Token ID | 8608 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
enburg | +0.625 |
Directions | +0.594 |
Dracula | +0.585 |
warr | +0.538 |
nces | +0.533 |
Neg logit contributions | |
isable | -0.442 |
ォ | -0.433 |
ident | -0.422 |
ERG | -0.422 |
OLD | -0.363 |
Token | icks |
Token ID | 3378 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
isable | +4.122 |
ident | +3.971 |
ERG | +3.952 |
ォ | +3.891 |
weight | +3.548 |
Neg logit contributions | |
enburg | -4.244 |
Directions | -4.053 |
Dracula | -4.015 |
warr | -3.784 |
etheless | -3.746 |
Token | wherever |
Token ID | 14530 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
enburg | +0.270 |
Directions | +0.255 |
Dracula | +0.252 |
warr | +0.232 |
etheless | +0.229 |
Neg logit contributions | |
isable | -0.218 |
ォ | -0.211 |
ident | -0.209 |
ERG | -0.209 |
OLD | -0.183 |
Token | possible |
Token ID | 1744 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
isable | +0.054 |
ident | +0.052 |
ERG | +0.052 |
ォ | +0.052 |
camp | +0.045 |
Neg logit contributions | |
enburg | -0.073 |
Directions | -0.068 |
Dracula | -0.068 |
warr | -0.063 |
etheless | -0.063 |
Token | – |
Token ID | 784 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
enburg | +0.860 |
Directions | +0.817 |
Dracula | +0.801 |
warr | +0.731 |
etheless | +0.727 |
Neg logit contributions | |
isable | -0.715 |
ォ | -0.696 |
ident | -0.686 |
ERG | -0.686 |
OLD | -0.605 |
Token | if |
Token ID | 611 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
enburg | +0.449 |
Directions | +0.425 |
Dracula | +0.419 |
warr | +0.386 |
etheless | +0.381 |
Neg logit contributions | |
isable | -0.365 |
ォ | -0.353 |
ident | -0.350 |
ERG | -0.349 |
OLD | -0.307 |
Token | Vol |
Token ID | 4709 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
enburg | +0.218 |
Directions | +0.204 |
Dracula | +0.200 |
warr | +0.182 |
etheless | +0.179 |
Neg logit contributions | |
isable | -0.238 |
ォ | -0.231 |
ident | -0.229 |
ERG | -0.229 |
OLD | -0.205 |
Token | atile |
Token ID | 12610 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
enburg | +0.924 |
Directions | +0.879 |
Dracula | +0.868 |
warr | +0.810 |
etheless | +0.804 |
Neg logit contributions | |
isable | -0.639 |
ident | -0.620 |
ォ | -0.618 |
ERG | -0.607 |
abba | -0.530 |
Token | Evolution |
Token ID | 15815 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
isable | +0.138 |
ォ | +0.133 |
ident | +0.132 |
ERG | +0.131 |
OLD | +0.114 |
Neg logit contributions | |
enburg | -0.192 |
Directions | -0.182 |
Dracula | -0.179 |
warr | -0.166 |
etheless | -0.164 |
Token | ( |
Token ID | 357 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
isable | +1.334 |
ォ | +1.297 |
ident | +1.293 |
ERG | +1.288 |
OLD | +1.155 |
Neg logit contributions | |
enburg | -1.218 |
Directions | -1.142 |
Dracula | -1.123 |
warr | -1.022 |
etheless | -1.013 |
Token | MA |
Token ID | 5673 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
isable | +0.188 |
ォ | +0.181 |
ERG | +0.181 |
ident | +0.180 |
OLD | +0.156 |
Neg logit contributions | |
enburg | -0.271 |
Directions | -0.257 |
Dracula | -0.254 |
warr | -0.236 |
etheless | -0.233 |
Token | VEN |
Token ID | 28290 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
isable | +0.356 |
ォ | +0.344 |
ident | +0.341 |
ERG | +0.330 |
abba | +0.294 |
Neg logit contributions | |
enburg | -0.429 |
Directions | -0.405 |
Dracula | -0.395 |
warr | -0.361 |
nces | -0.357 |
Token | ) |
Token ID | 8 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
isable | +3.343 |
ERG | +3.193 |
ident | +3.190 |
ォ | +3.168 |
camp | +2.782 |
Neg logit contributions | |
enburg | -5.041 |
Directions | -4.824 |
Dracula | -4.757 |
warr | -4.433 |
etheless | -4.427 |
Token | mission |
Token ID | 4365 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
isable | +0.828 |
ォ | +0.802 |
ident | +0.800 |
ERG | +0.799 |
camp | +0.714 |
Neg logit contributions | |
enburg | -0.824 |
Directions | -0.775 |
Dracula | -0.764 |
warr | -0.695 |
etheless | -0.691 |
Token | , |
Token ID | 11 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
isable | +0.568 |
ォ | +0.544 |
ident | +0.543 |
ERG | +0.541 |
camp | +0.471 |
Neg logit contributions | |
enburg | -0.776 |
Directions | -0.736 |
Dracula | -0.725 |
warr | -0.673 |
etheless | -0.665 |
Token | which |
Token ID | 543 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
isable | +0.522 |
ERG | +0.501 |
ident | +0.500 |
ォ | +0.497 |
camp | +0.435 |
Neg logit contributions | |
enburg | -0.743 |
Directions | -0.705 |
Dracula | -0.691 |
warr | -0.645 |
etheless | -0.642 |
Token | until |
Token ID | 1566 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
enburg | +0.502 |
Directions | +0.474 |
Dracula | +0.467 |
warr | +0.430 |
etheless | +0.426 |
Neg logit contributions | |
isable | -0.434 |
ォ | -0.420 |
ident | -0.417 |
ERG | -0.416 |
OLD | -0.367 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
enburg | +0.254 |
Directions | +0.240 |
Dracula | +0.237 |
warr | +0.218 |
etheless | +0.216 |
Neg logit contributions | |
isable | -0.199 |
ォ | -0.192 |
ident | -0.191 |
ERG | -0.190 |
abba | -0.166 |
Token | Anaheim |
Token ID | 31100 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
enburg | +0.248 |
Directions | +0.233 |
Dracula | +0.229 |
warr | +0.210 |
etheless | +0.207 |
Neg logit contributions | |
isable | -0.242 |
ォ | -0.235 |
ident | -0.233 |
ERG | -0.232 |
OLD | -0.208 |
Token | Ducks |
Token ID | 27622 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
enburg | +0.569 |
Directions | +0.528 |
Dracula | +0.518 |
warr | +0.464 |
etheless | +0.456 |
Neg logit contributions | |
isable | -0.768 |
ォ | -0.748 |
ident | -0.744 |
ERG | -0.743 |
OLD | -0.673 |
Token | . |
Token ID | 13 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
enburg | +1.339 |
Directions | +1.260 |
Dracula | +1.245 |
warr | +1.138 |
etheless | +1.122 |
Neg logit contributions | |
isable | -1.121 |
ォ | -1.090 |
ident | -1.075 |
ERG | -1.075 |
OLD | -0.946 |
Token | Justin |
Token ID | 10799 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
isable | +0.063 |
ォ | +0.061 |
ident | +0.060 |
ERG | +0.060 |
OLD | +0.053 |
Neg logit contributions | |
enburg | -0.078 |
Directions | -0.074 |
Dracula | -0.073 |
warr | -0.068 |
etheless | -0.067 |
Token | Faul |
Token ID | 44760 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
isable | +1.276 |
ォ | +1.244 |
ident | +1.236 |
ERG | +1.234 |
OLD | +1.121 |
Neg logit contributions | |
enburg | -0.873 |
Directions | -0.807 |
Dracula | -0.791 |
warr | -0.704 |
etheless | -0.689 |
Token | k |
Token ID | 74 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
enburg | +4.243 |
Directions | +3.968 |
Dracula | +3.882 |
warr | +3.412 |
nces | +3.362 |
Neg logit contributions | |
isable | -5.349 |
ォ | -5.292 |
ERG | -5.236 |
ident | -5.100 |
OLD | -4.694 |
Token | , |
Token ID | 11 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
enburg | +1.232 |
Directions | +1.158 |
Dracula | +1.144 |
warr | +1.049 |
etheless | +1.038 |
Neg logit contributions | |
isable | -1.014 |
ォ | -0.982 |
ident | -0.971 |
ERG | -0.970 |
OLD | -0.852 |
Token | one |
Token ID | 530 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
enburg | +0.266 |
Directions | +0.251 |
Dracula | +0.247 |
warr | +0.228 |
etheless | +0.226 |
Neg logit contributions | |
isable | -0.223 |
ォ | -0.215 |
ident | -0.214 |
ERG | -0.214 |
OLD | -0.189 |
Token | of |
Token ID | 286 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
isable | +0.394 |
ォ | +0.381 |
ident | +0.378 |
ERG | +0.377 |
abba | +0.332 |
Neg logit contributions | |
enburg | -0.477 |
Directions | -0.449 |
Dracula | -0.444 |
warr | -0.409 |
etheless | -0.403 |
Token | the |
Token ID | 262 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
isable | +0.231 |
ォ | +0.223 |
ident | +0.221 |
ERG | +0.221 |
OLD | +0.193 |
Neg logit contributions | |
enburg | -0.293 |
Directions | -0.278 |
Dracula | -0.273 |
warr | -0.252 |
etheless | -0.249 |
Token | of |
Token ID | 286 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
enburg | +1.289 |
Directions | +1.213 |
Dracula | +1.200 |
warr | +1.093 |
etheless | +1.076 |
Neg logit contributions | |
isable | -1.151 |
ォ | -1.111 |
ERG | -1.099 |
ident | -1.099 |
OLD | -0.973 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
enburg | +0.320 |
Directions | +0.301 |
Dracula | +0.297 |
warr | +0.271 |
etheless | +0.267 |
Neg logit contributions | |
isable | -0.304 |
ォ | -0.294 |
ident | -0.292 |
ERG | -0.292 |
OLD | -0.259 |
Token | Sp |
Token ID | 1338 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
enburg | +0.373 |
Directions | +0.352 |
Dracula | +0.346 |
warr | +0.315 |
etheless | +0.310 |
Neg logit contributions | |
isable | -0.373 |
ォ | -0.362 |
ident | -0.359 |
ERG | -0.359 |
OLD | -0.320 |
Token | ires |
Token ID | 2387 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
enburg | +1.041 |
Directions | +0.970 |
Dracula | +0.961 |
warr | +0.865 |
nces | +0.865 |
Neg logit contributions | |
isable | -0.908 |
ォ | -0.897 |
ERG | -0.879 |
ident | -0.863 |
OLD | -0.776 |
Token | vs |
Token ID | 3691 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
enburg | +2.741 |
Directions | +2.602 |
Dracula | +2.565 |
warr | +2.369 |
nces | +2.340 |
Neg logit contributions | |
isable | -1.795 |
ォ | -1.702 |
ident | -1.692 |
ERG | -1.686 |
abba | -1.448 |
Token | Ank |
Token ID | 24792 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
isable | +0.677 |
ォ | +0.658 |
ident | +0.655 |
ERG | +0.652 |
OLD | +0.582 |
Neg logit contributions | |
enburg | -0.661 |
Directions | -0.617 |
Dracula | -0.603 |
warr | -0.558 |
etheless | -0.552 |
Token | le |
Token ID | 293 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
enburg | +5.120 |
Dracula | +4.696 |
Directions | +4.646 |
nces | +4.192 |
warr | +4.187 |
Neg logit contributions | |
ォ | -4.370 |
isable | -4.362 |
ERG | -4.255 |
ident | -4.160 |
OLD | -3.761 |
Token | Shank |
Token ID | 41595 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
enburg | +0.891 |
Directions | +0.839 |
Dracula | +0.830 |
warr | +0.755 |
etheless | +0.745 |
Neg logit contributions | |
isable | -0.769 |
ォ | -0.743 |
ERG | -0.734 |
ident | -0.734 |
OLD | -0.647 |
Token | er |
Token ID | 263 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
enburg | +0.850 |
Directions | +0.804 |
Dracula | +0.796 |
warr | +0.730 |
etheless | +0.721 |
Neg logit contributions | |
isable | -0.628 |
ォ | -0.604 |
ERG | -0.595 |
ident | -0.594 |
OLD | -0.518 |
Token |
|
Token ID | 198 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
enburg | +0.084 |
Directions | +0.080 |
Dracula | +0.079 |
warr | +0.072 |
etheless | +0.071 |
Neg logit contributions | |
isable | -0.068 |
ォ | -0.066 |
ERG | -0.065 |
ident | -0.065 |
OLD | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
enburg | +0.517 |
Directions | +0.491 |
Dracula | +0.483 |
warr | +0.454 |
etheless | +0.448 |
Neg logit contributions | |
isable | -0.373 |
ォ | -0.358 |
ident | -0.357 |
ERG | -0.356 |
OLD | -0.309 |
Token | is |
Token ID | 318 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
isable | +0.865 |
ォ | +0.836 |
ERG | +0.833 |
ident | +0.831 |
camp | +0.734 |
Neg logit contributions | |
enburg | -1.116 |
Directions | -1.058 |
Dracula | -1.045 |
warr | -0.969 |
etheless | -0.956 |
Token | currently |
Token ID | 3058 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
enburg | +0.349 |
Directions | +0.330 |
Dracula | +0.324 |
warr | +0.298 |
etheless | +0.294 |
Neg logit contributions | |
isable | -0.306 |
ォ | -0.296 |
ident | -0.294 |
ERG | -0.293 |
OLD | -0.259 |
Token | over |
Token ID | 625 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
enburg | +0.838 |
Directions | +0.789 |
Dracula | +0.777 |
warr | +0.712 |
etheless | +0.702 |
Neg logit contributions | |
isable | -0.778 |
ォ | -0.754 |
ident | -0.748 |
ERG | -0.746 |
OLD | -0.662 |
Token | a |
Token ID | 257 |
Feature activation | -9.47e-02 |
Pos logit contributions | |
isable | +0.669 |
ォ | +0.644 |
ident | +0.642 |
ERG | +0.637 |
camp | +0.557 |
Neg logit contributions | |
enburg | -0.877 |
Directions | -0.832 |
Dracula | -0.820 |
warr | -0.762 |
etheless | -0.749 |
Token | deep |
Token ID | 2769 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
enburg | +0.207 |
Directions | +0.196 |
Dracula | +0.193 |
warr | +0.178 |
nces | +0.176 |
Neg logit contributions | |
isable | -0.177 |
ォ | -0.170 |
ident | -0.170 |
ERG | -0.169 |
camp | -0.150 |
Token | gou |
Token ID | 50162 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
enburg | +0.840 |
Directions | +0.787 |
Dracula | +0.776 |
warr | +0.709 |
etheless | +0.699 |
Neg logit contributions | |
isable | -0.794 |
ォ | -0.772 |
ERG | -0.766 |
ident | -0.765 |
OLD | -0.678 |
Token | ge |
Token ID | 469 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
enburg | +3.622 |
Directions | +3.172 |
Dracula | +3.128 |
nces | +2.739 |
etheless | +2.705 |
Neg logit contributions | |
isable | -5.437 |
ォ | -5.368 |
ERG | -5.345 |
ident | -5.169 |
OLD | -4.881 |
Token | in |
Token ID | 287 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
enburg | +2.493 |
Directions | +2.353 |
Dracula | +2.316 |
warr | +2.115 |
etheless | +2.089 |
Neg logit contributions | |
isable | -2.171 |
ォ | -2.105 |
ERG | -2.084 |
ident | -2.079 |
OLD | -1.839 |
Token | which |
Token ID | 543 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
enburg | +0.610 |
Directions | +0.577 |
Dracula | +0.568 |
warr | +0.523 |
etheless | +0.515 |
Neg logit contributions | |
isable | -0.482 |
ォ | -0.466 |
ident | -0.462 |
ERG | -0.461 |
OLD | -0.403 |
Token | the |
Token ID | 262 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
enburg | +0.830 |
Directions | +0.783 |
Dracula | +0.772 |
warr | +0.709 |
etheless | +0.699 |
Neg logit contributions | |
isable | -0.711 |
ォ | -0.687 |
ident | -0.682 |
ERG | -0.681 |
OLD | -0.599 |
Token | bedrock |
Token ID | 47856 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
enburg | +0.038 |
Directions | +0.036 |
Dracula | +0.035 |
warr | +0.032 |
etheless | +0.032 |
Neg logit contributions | |
isable | -0.035 |
ォ | -0.034 |
ident | -0.034 |
ERG | -0.034 |
camp | -0.030 |
Token | This |
Token ID | 770 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
isable | +0.197 |
ォ | +0.191 |
ident | +0.191 |
ERG | +0.190 |
OLD | +0.168 |
Neg logit contributions | |
enburg | -0.232 |
Directions | -0.218 |
Dracula | -0.215 |
warr | -0.200 |
etheless | -0.197 |
Token | week |
Token ID | 1285 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
isable | +0.498 |
ォ | +0.478 |
ident | +0.478 |
ERG | +0.475 |
camp | +0.414 |
Neg logit contributions | |
enburg | -0.697 |
Directions | -0.664 |
Dracula | -0.653 |
warr | -0.607 |
etheless | -0.599 |
Token | , |
Token ID | 11 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
isable | +0.861 |
ident | +0.830 |
ォ | +0.830 |
ERG | +0.825 |
OLD | +0.722 |
Neg logit contributions | |
enburg | -1.156 |
Directions | -1.096 |
Dracula | -1.080 |
warr | -1.003 |
etheless | -0.988 |
Token | Alex |
Token ID | 4422 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
isable | +0.392 |
ident | +0.379 |
ERG | +0.377 |
ォ | +0.376 |
camp | +0.334 |
Neg logit contributions | |
enburg | -0.442 |
Directions | -0.416 |
Dracula | -0.410 |
warr | -0.378 |
etheless | -0.372 |
Token | ei |
Token ID | 20295 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
isable | +0.393 |
ォ | +0.378 |
ident | +0.376 |
ERG | +0.374 |
OLD | +0.325 |
Neg logit contributions | |
enburg | -0.540 |
Directions | -0.514 |
Dracula | -0.506 |
warr | -0.469 |
etheless | -0.463 |
Token | Sut |
Token ID | 45220 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
enburg | +0.322 |
Directions | +0.301 |
Dracula | +0.295 |
warr | +0.268 |
etheless | +0.263 |
Neg logit contributions | |
isable | -0.372 |
ォ | -0.361 |
ident | -0.359 |
ERG | -0.358 |
OLD | -0.322 |
Token | uga |
Token ID | 30302 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
enburg | +4.809 |
Directions | +4.424 |
Dracula | +4.414 |
warr | +3.968 |
nces | +3.959 |
Neg logit contributions | |
isable | -4.106 |
ォ | -4.058 |
ERG | -3.988 |
ident | -3.924 |
OLD | -3.521 |
Token | , |
Token ID | 11 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
enburg | +1.014 |
Directions | +0.958 |
Dracula | +0.943 |
warr | +0.869 |
etheless | +0.857 |
Neg logit contributions | |
isable | -0.838 |
ォ | -0.811 |
ident | -0.804 |
ERG | -0.802 |
OLD | -0.706 |
Token | an |
Token ID | 281 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
enburg | +0.766 |
Directions | +0.724 |
Dracula | +0.713 |
warr | +0.659 |
etheless | +0.649 |
Neg logit contributions | |
isable | -0.615 |
ォ | -0.594 |
ident | -0.591 |
ERG | -0.589 |
OLD | -0.517 |
Token | anti |
Token ID | 3098 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
enburg | +0.609 |
Directions | +0.575 |
Dracula | +0.567 |
warr | +0.522 |
etheless | +0.515 |
Neg logit contributions | |
isable | -0.513 |
ォ | -0.493 |
ident | -0.492 |
ERG | -0.490 |
OLD | -0.432 |
Token | - |
Token ID | 12 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
isable | +0.425 |
ォ | +0.408 |
ident | +0.404 |
ERG | +0.403 |
OLD | +0.344 |
Neg logit contributions | |
enburg | -0.709 |
Directions | -0.675 |
Dracula | -0.666 |
warr | -0.621 |
etheless | -0.614 |
Token | 89 |
Token ID | 9919 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
isable | +1.068 |
ident | +1.025 |
ERG | +1.011 |
ォ | +1.005 |
camp | +0.907 |
Neg logit contributions | |
enburg | -1.421 |
Directions | -1.348 |
Dracula | -1.316 |
warr | -1.250 |
etheless | -1.229 |
Token | % |
Token ID | 4 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
enburg | +0.477 |
Directions | +0.449 |
Dracula | +0.442 |
warr | +0.404 |
etheless | +0.398 |
Neg logit contributions | |
isable | -0.463 |
ォ | -0.449 |
ident | -0.445 |
ERG | -0.445 |
OLD | -0.396 |
Token | to |
Token ID | 284 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
isable | +1.035 |
ォ | +1.000 |
ident | +0.999 |
ERG | +0.990 |
camp | +0.879 |
Neg logit contributions | |
enburg | -1.230 |
Directions | -1.167 |
Dracula | -1.145 |
warr | -1.061 |
etheless | -1.051 |
Token | 65 |
Token ID | 6135 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
enburg | +1.214 |
Directions | +1.148 |
Dracula | +1.130 |
warr | +1.044 |
etheless | +1.030 |
Neg logit contributions | |
isable | -0.972 |
ォ | -0.938 |
ident | -0.931 |
ERG | -0.929 |
OLD | -0.815 |
Token | %. |
Token ID | 7225 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
enburg | +1.590 |
Directions | +1.501 |
Dracula | +1.478 |
warr | +1.346 |
nces | +1.344 |
Neg logit contributions | |
isable | -1.326 |
ォ | -1.285 |
ERG | -1.271 |
ident | -1.266 |
OLD | -1.123 |
Token | Pent |
Token ID | 9696 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
isable | +0.591 |
ォ | +0.579 |
ident | +0.573 |
ERG | +0.570 |
OLD | +0.499 |
Neg logit contributions | |
enburg | -0.769 |
Directions | -0.718 |
Dracula | -0.708 |
warr | -0.677 |
etheless | -0.664 |
Token | ec |
Token ID | 721 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
enburg | +4.966 |
Directions | +4.609 |
Dracula | +4.538 |
nces | +4.166 |
etheless | +4.110 |
Neg logit contributions | |
isable | -3.651 |
ォ | -3.613 |
ERG | -3.518 |
ident | -3.462 |
OLD | -3.123 |
Token | ost |
Token ID | 455 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
enburg | +0.947 |
Directions | +0.901 |
Dracula | +0.889 |
warr | +0.828 |
etheless | +0.820 |
Neg logit contributions | |
isable | -0.546 |
ォ | -0.525 |
ERG | -0.518 |
ident | -0.517 |
OLD | -0.440 |
Token | als |
Token ID | 874 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
enburg | +2.169 |
Directions | +2.058 |
Dracula | +2.023 |
warr | +1.858 |
nces | +1.842 |
Neg logit contributions | |
isable | -1.731 |
ォ | -1.676 |
ERG | -1.659 |
ident | -1.648 |
OLD | -1.444 |
Token | are |
Token ID | 389 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
enburg | +1.593 |
Directions | +1.506 |
Dracula | +1.487 |
warr | +1.365 |
etheless | +1.350 |
Neg logit contributions | |
isable | -1.250 |
ォ | -1.210 |
ERG | -1.198 |
ident | -1.192 |
OLD | -1.048 |
Token | poorer |
Token ID | 26647 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
enburg | +0.897 |
Directions | +0.846 |
Dracula | +0.831 |
warr | +0.765 |
etheless | +0.754 |
Neg logit contributions | |
isable | -0.817 |
ォ | -0.790 |
ident | -0.785 |
ERG | -0.783 |
OLD | -0.694 |
Token | 8 |
Token ID | 23 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
enburg | +0.283 |
Directions | +0.266 |
Dracula | +0.261 |
warr | +0.240 |
etheless | +0.237 |
Neg logit contributions | |
isable | -0.272 |
ォ | -0.264 |
ident | -0.262 |
ERG | -0.262 |
OLD | -0.233 |
Token | m |
Token ID | 76 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
enburg | +1.075 |
Directions | +1.001 |
Dracula | +0.982 |
warr | +0.883 |
etheless | +0.869 |
Neg logit contributions | |
isable | -1.362 |
ォ | -1.325 |
ERG | -1.317 |
ident | -1.316 |
OLD | -1.188 |
Token | followers |
Token ID | 10569 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
enburg | +0.396 |
Directions | +0.374 |
Dracula | +0.369 |
warr | +0.339 |
etheless | +0.334 |
Neg logit contributions | |
isable | -0.339 |
ォ | -0.328 |
ident | -0.325 |
ERG | -0.325 |
OLD | -0.286 |
Token | . |
Token ID | 13 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
enburg | +0.678 |
Directions | +0.642 |
Dracula | +0.632 |
warr | +0.583 |
etheless | +0.575 |
Neg logit contributions | |
isable | -0.549 |
ォ | -0.531 |
ident | -0.526 |
ERG | -0.525 |
OLD | -0.461 |
Token | As |
Token ID | 1081 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
enburg | +0.508 |
Directions | +0.479 |
Dracula | +0.472 |
warr | +0.438 |
etheless | +0.431 |
Neg logit contributions | |
isable | -0.412 |
ォ | -0.399 |
ident | -0.396 |
ERG | -0.395 |
OLD | -0.347 |
Token | Pent |
Token ID | 9696 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
enburg | +0.844 |
Directions | +0.796 |
Dracula | +0.784 |
warr | +0.719 |
etheless | +0.710 |
Neg logit contributions | |
isable | -0.753 |
ォ | -0.729 |
ident | -0.723 |
ERG | -0.722 |
OLD | -0.639 |
Token | ec |
Token ID | 721 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
enburg | +4.950 |
Directions | +4.584 |
Dracula | +4.512 |
nces | +4.137 |
etheless | +4.056 |
Neg logit contributions | |
isable | -3.612 |
ォ | -3.585 |
ERG | -3.481 |
ident | -3.439 |
OLD | -3.096 |
Token | ost |
Token ID | 455 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
enburg | +1.161 |
Directions | +1.105 |
Dracula | +1.090 |
warr | +1.015 |
etheless | +1.005 |
Neg logit contributions | |
isable | -0.671 |
ォ | -0.645 |
ERG | -0.636 |
ident | -0.636 |
OLD | -0.540 |
Token | als |
Token ID | 874 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
enburg | +2.113 |
Directions | +2.004 |
Dracula | +1.964 |
warr | +1.799 |
nces | +1.786 |
Neg logit contributions | |
isable | -1.770 |
ォ | -1.715 |
ERG | -1.703 |
ident | -1.690 |
OLD | -1.484 |
Token | go |
Token ID | 467 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
enburg | +1.683 |
Directions | +1.589 |
Dracula | +1.569 |
warr | +1.436 |
etheless | +1.415 |
Neg logit contributions | |
isable | -1.424 |
ォ | -1.385 |
ERG | -1.368 |
ident | -1.364 |
OLD | -1.205 |
Token | , |
Token ID | 11 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
isable | +1.017 |
ォ | +0.979 |
ident | +0.973 |
ERG | +0.972 |
OLD | +0.844 |
Neg logit contributions | |
enburg | -1.427 |
Directions | -1.351 |
Dracula | -1.331 |
warr | -1.237 |
etheless | -1.220 |
Token | and |
Token ID | 290 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
enburg | +0.091 |
Directions | +0.086 |
Dracula | +0.085 |
warr | +0.078 |
etheless | +0.077 |
Neg logit contributions | |
isable | -0.070 |
ォ | -0.068 |
ident | -0.067 |
ERG | -0.067 |
OLD | -0.059 |
Token | be |
Token ID | 307 |
Feature activation | +3.85e-03 |
Pos logit contributions | |
enburg | +0.189 |
Directions | +0.179 |
Dracula | +0.176 |
warr | +0.162 |
etheless | +0.160 |
Neg logit contributions | |
isable | -0.155 |
ォ | -0.150 |
ident | -0.149 |
ERG | -0.148 |
OLD | -0.130 |
Token | able |
Token ID | 1498 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
isable | +0.008 |
ォ | +0.007 |
ident | +0.007 |
ERG | +0.007 |
OLD | +0.006 |
Neg logit contributions | |
enburg | -0.008 |
Directions | -0.008 |
Dracula | -0.007 |
warr | -0.007 |
etheless | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
enburg | +0.447 |
Directions | +0.417 |
Dracula | +0.409 |
warr | +0.378 |
nces | +0.368 |
Neg logit contributions | |
isable | -0.438 |
ォ | -0.432 |
ERG | -0.423 |
ident | -0.420 |
OLD | -0.379 |
Token | fully |
Token ID | 3938 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
enburg | +0.908 |
Directions | +0.859 |
Dracula | +0.846 |
warr | +0.781 |
etheless | +0.772 |
Neg logit contributions | |
isable | -0.726 |
ォ | -0.700 |
ident | -0.697 |
ERG | -0.694 |
OLD | -0.608 |
Token | express |
Token ID | 4911 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
enburg | +1.092 |
Directions | +1.029 |
Dracula | +1.013 |
warr | +0.931 |
etheless | +0.916 |
Neg logit contributions | |
isable | -0.948 |
ォ | -0.923 |
ERG | -0.910 |
ident | -0.907 |
OLD | -0.805 |
Token | their |
Token ID | 511 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
enburg | +4.449 |
Directions | +4.217 |
Dracula | +4.143 |
warr | +3.803 |
nces | +3.665 |
Neg logit contributions | |
ォ | -4.000 |
isable | -3.941 |
ERG | -3.829 |
ident | -3.733 |
OLD | -3.368 |
Token | outrage |
Token ID | 11616 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
enburg | +0.498 |
Directions | +0.467 |
Dracula | +0.459 |
warr | +0.419 |
etheless | +0.414 |
Neg logit contributions | |
isable | -0.509 |
ォ | -0.493 |
ident | -0.491 |
ERG | -0.490 |
OLD | -0.437 |
Token | , |
Token ID | 11 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
isable | +0.591 |
ォ | +0.569 |
ident | +0.568 |
ERG | +0.566 |
OLD | +0.497 |
Neg logit contributions | |
enburg | -0.711 |
Directions | -0.671 |
Dracula | -0.660 |
warr | -0.609 |
etheless | -0.601 |
Token | their |
Token ID | 511 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
enburg | +0.250 |
Directions | +0.236 |
Dracula | +0.232 |
warr | +0.214 |
etheless | +0.210 |
Neg logit contributions | |
isable | -0.224 |
ォ | -0.218 |
ERG | -0.215 |
ident | -0.214 |
OLD | -0.190 |
Token | plan |
Token ID | 1410 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
enburg | +0.133 |
Directions | +0.125 |
Dracula | +0.123 |
warr | +0.112 |
etheless | +0.111 |
Neg logit contributions | |
isable | -0.133 |
ォ | -0.129 |
ident | -0.128 |
ERG | -0.128 |
OLD | -0.114 |
Token | Mc |
Token ID | 1982 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
isable | +0.156 |
ォ | +0.152 |
ident | +0.151 |
ERG | +0.151 |
OLD | +0.136 |
Neg logit contributions | |
enburg | -0.121 |
Directions | -0.113 |
Dracula | -0.111 |
warr | -0.099 |
etheless | -0.097 |
Token | Con |
Token ID | 3103 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
enburg | +0.698 |
Directions | +0.657 |
Dracula | +0.648 |
warr | +0.585 |
nces | +0.575 |
Neg logit contributions | |
isable | -0.705 |
ォ | -0.682 |
ERG | -0.677 |
ident | -0.677 |
OLD | -0.605 |
Token | aug |
Token ID | 7493 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
enburg | +0.523 |
Directions | +0.498 |
Dracula | +0.491 |
warr | +0.460 |
etheless | +0.452 |
Neg logit contributions | |
isable | -0.430 |
ident | -0.415 |
ォ | -0.414 |
ERG | -0.412 |
abba | -0.363 |
Token | hey |
Token ID | 20342 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
isable | +0.312 |
ident | +0.298 |
ERG | +0.297 |
ォ | +0.296 |
abba | +0.264 |
Neg logit contributions | |
enburg | -0.331 |
Directions | -0.319 |
Dracula | -0.315 |
warr | -0.295 |
etheless | -0.290 |
Token | True |
Token ID | 6407 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
enburg | +1.388 |
Directions | +1.323 |
Dracula | +1.305 |
warr | +1.171 |
nces | +1.149 |
Neg logit contributions | |
isable | -1.261 |
ォ | -1.227 |
ident | -1.210 |
ERG | -1.209 |
OLD | -1.071 |
Token | Detective |
Token ID | 20220 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
enburg | +0.762 |
Directions | +0.690 |
Dracula | +0.665 |
warr | +0.523 |
nces | +0.515 |
Neg logit contributions | |
isable | -2.121 |
ォ | -2.081 |
ERG | -2.067 |
ident | -2.061 |
OLD | -1.919 |
Token | HBO |
Token ID | 18804 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
enburg | +4.078 |
Directions | +3.978 |
Dracula | +3.904 |
warr | +3.323 |
Neighborhood | +3.314 |
Neg logit contributions | |
isable | -4.250 |
ォ | -4.176 |
ident | -4.070 |
ERG | -4.048 |
abba | -3.681 |
Token | Matthew |
Token ID | 9308 |
Feature activation | +2.15e-03 |
Pos logit contributions | |
enburg | +1.319 |
Directions | +1.256 |
Dracula | +1.239 |
warr | +1.112 |
etheless | +1.092 |
Neg logit contributions | |
isable | -1.208 |
ォ | -1.174 |
ident | -1.159 |
ERG | -1.156 |
OLD | -1.026 |
Token | Rh |
Token ID | 10323 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
isable | +0.005 |
ォ | +0.005 |
ident | +0.005 |
ERG | +0.005 |
OLD | +0.004 |
Neg logit contributions | |
enburg | -0.004 |
Directions | -0.004 |
Dracula | -0.004 |
warr | -0.003 |
etheless | -0.003 |
Token | ys |
Token ID | 893 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
enburg | +1.785 |
Directions | +1.595 |
Dracula | +1.549 |
nces | +1.385 |
warr | +1.324 |
Neg logit contributions | |
isable | -2.555 |
ォ | -2.553 |
ERG | -2.537 |
ident | -2.465 |
OLD | -2.298 |
Token | The |
Token ID | 383 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
enburg | +1.381 |
Directions | +1.303 |
Dracula | +1.283 |
warr | +1.174 |
etheless | +1.158 |
Neg logit contributions | |
isable | -1.244 |
ォ | -1.206 |
ident | -1.195 |
ERG | -1.193 |
OLD | -1.056 |
Token | and |
Token ID | 290 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
enburg | +1.064 |
Directions | +1.006 |
Dracula | +0.992 |
warr | +0.912 |
etheless | +0.900 |
Neg logit contributions | |
isable | -0.861 |
ォ | -0.835 |
ERG | -0.825 |
ident | -0.824 |
OLD | -0.723 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
isable | +0.069 |
ident | +0.065 |
ォ | +0.065 |
ERG | +0.064 |
camp | +0.055 |
Neg logit contributions | |
enburg | -0.135 |
Directions | -0.129 |
Dracula | -0.126 |
warr | -0.119 |
etheless | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
enburg | +0.222 |
Directions | +0.210 |
Dracula | +0.206 |
warr | +0.192 |
etheless | +0.190 |
Neg logit contributions | |
isable | -0.170 |
ォ | -0.163 |
ident | -0.163 |
ERG | -0.162 |
camp | -0.142 |
Token | number |
Token ID | 1271 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
enburg | +0.056 |
Directions | +0.053 |
Dracula | +0.051 |
warr | +0.047 |
etheless | +0.046 |
Neg logit contributions | |
isable | -0.063 |
ident | -0.061 |
ォ | -0.060 |
ERG | -0.060 |
camp | -0.054 |
Token | of |
Token ID | 286 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
enburg | +1.173 |
Directions | +1.092 |
Dracula | +1.089 |
warr | +0.988 |
nces | +0.978 |
Neg logit contributions | |
isable | -1.018 |
ォ | -0.990 |
ERG | -0.975 |
ident | -0.974 |
abba | -0.861 |
Token | Pent |
Token ID | 9696 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
enburg | +0.363 |
Directions | +0.343 |
Dracula | +0.336 |
warr | +0.310 |
etheless | +0.307 |
Neg logit contributions | |
isable | -0.325 |
ォ | -0.314 |
ident | -0.313 |
ERG | -0.311 |
camp | -0.276 |
Token | ec |
Token ID | 721 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
enburg | +4.810 |
Directions | +4.471 |
Dracula | +4.400 |
nces | +4.057 |
etheless | +3.965 |
Neg logit contributions | |
isable | -3.458 |
ォ | -3.427 |
ERG | -3.338 |
ident | -3.282 |
OLD | -2.947 |
Token | ost |
Token ID | 455 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
enburg | +1.081 |
Directions | +1.027 |
Dracula | +1.013 |
warr | +0.943 |
etheless | +0.934 |
Neg logit contributions | |
isable | -0.627 |
ォ | -0.604 |
ERG | -0.595 |
ident | -0.593 |
OLD | -0.506 |
Token | als |
Token ID | 874 |
Feature activation | -6.69e-01 |
Pos logit contributions | |
enburg | +2.317 |
Directions | +2.176 |
Dracula | +2.143 |
nces | +1.961 |
warr | +1.957 |
Neg logit contributions | |
isable | -1.931 |
ォ | -1.905 |
ERG | -1.887 |
ident | -1.838 |
abba | -1.638 |
Token | rose |
Token ID | 8278 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
enburg | +1.362 |
Directions | +1.282 |
Dracula | +1.260 |
warr | +1.154 |
etheless | +1.137 |
Neg logit contributions | |
isable | -1.349 |
ォ | -1.306 |
ident | -1.299 |
ERG | -1.296 |
OLD | -1.155 |
Token | from |
Token ID | 422 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
enburg | +1.228 |
Directions | +1.157 |
Dracula | +1.133 |
warr | +1.041 |
etheless | +1.030 |
Neg logit contributions | |
isable | -1.207 |
ident | -1.163 |
ォ | -1.157 |
ERG | -1.154 |
camp | -1.033 |
Token | , |
Token ID | 11 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
enburg | +1.095 |
Directions | +1.038 |
Dracula | +1.022 |
warr | +0.947 |
etheless | +0.935 |
Neg logit contributions | |
isable | -0.814 |
ォ | -0.785 |
ident | -0.780 |
ERG | -0.777 |
OLD | -0.678 |
Token | as |
Token ID | 355 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
isable | +0.396 |
ident | +0.382 |
ォ | +0.380 |
ERG | +0.379 |
camp | +0.337 |
Neg logit contributions | |
enburg | -0.493 |
Directions | -0.468 |
Dracula | -0.457 |
warr | -0.427 |
etheless | -0.422 |
Token | both |
Token ID | 1111 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
enburg | +0.643 |
Directions | +0.608 |
Dracula | +0.598 |
warr | +0.552 |
etheless | +0.546 |
Neg logit contributions | |
isable | -0.538 |
ォ | -0.520 |
ident | -0.517 |
ERG | -0.516 |
OLD | -0.455 |
Token | Brad |
Token ID | 8114 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
enburg | +0.231 |
Directions | +0.215 |
Dracula | +0.212 |
warr | +0.195 |
etheless | +0.193 |
Neg logit contributions | |
isable | -0.240 |
ォ | -0.233 |
ident | -0.233 |
ERG | -0.231 |
camp | -0.210 |
Token | en |
Token ID | 268 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
isable | +0.389 |
ォ | +0.376 |
ident | +0.372 |
ERG | +0.372 |
OLD | +0.326 |
Neg logit contributions | |
enburg | -0.484 |
Directions | -0.456 |
Dracula | -0.450 |
warr | -0.414 |
etheless | -0.408 |
Token | Holt |
Token ID | 28330 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
isable | +0.474 |
ォ | +0.464 |
ident | +0.461 |
ERG | +0.460 |
OLD | +0.421 |
Neg logit contributions | |
enburg | -0.290 |
Directions | -0.266 |
Dracula | -0.260 |
warr | -0.229 |
etheless | -0.225 |
Token | by |
Token ID | 1525 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
enburg | +3.344 |
Directions | +3.068 |
Dracula | +3.020 |
warr | +2.664 |
nces | +2.613 |
Neg logit contributions | |
isable | -4.899 |
ォ | -4.749 |
ERG | -4.697 |
ident | -4.696 |
abba | -4.269 |
Token | and |
Token ID | 290 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
enburg | +0.770 |
Directions | +0.730 |
Dracula | +0.719 |
warr | +0.664 |
etheless | +0.655 |
Neg logit contributions | |
isable | -0.617 |
ォ | -0.597 |
ident | -0.592 |
ERG | -0.590 |
OLD | -0.518 |
Token | Sergei |
Token ID | 34724 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
enburg | +0.403 |
Directions | +0.379 |
Dracula | +0.373 |
warr | +0.343 |
etheless | +0.339 |
Neg logit contributions | |
isable | -0.364 |
ォ | -0.353 |
ident | -0.351 |
ERG | -0.350 |
OLD | -0.310 |
Token | Bob |
Token ID | 5811 |
Feature activation | +8.15e-03 |
Pos logit contributions | |
enburg | +1.010 |
Directions | +0.956 |
Dracula | +0.951 |
warr | +0.862 |
nces | +0.841 |
Neg logit contributions | |
isable | -0.941 |
ォ | -0.913 |
ident | -0.902 |
ERG | -0.901 |
OLD | -0.800 |
Token | ro |
Token ID | 305 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
isable | +0.015 |
ォ | +0.015 |
ERG | +0.015 |
ident | +0.015 |
OLD | +0.013 |
Neg logit contributions | |
enburg | -0.018 |
Directions | -0.017 |
Dracula | -0.017 |
warr | -0.015 |
etheless | -0.015 |
Token | ). |
Token ID | 737 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
enburg | +0.549 |
Directions | +0.520 |
Dracula | +0.513 |
warr | +0.475 |
etheless | +0.469 |
Neg logit contributions | |
isable | -0.406 |
ォ | -0.392 |
ident | -0.389 |
ERG | -0.388 |
OLD | -0.338 |
Token | IV |
Token ID | 8363 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
isable | +0.414 |
ォ | +0.400 |
ident | +0.397 |
ERG | +0.397 |
OLD | +0.350 |
Neg logit contributions | |
enburg | -0.487 |
Directions | -0.458 |
Dracula | -0.452 |
warr | -0.417 |
etheless | -0.412 |
Token | . |
Token ID | 13 |
Feature activation | -9.28e-02 |
Pos logit contributions | |
enburg | +0.942 |
Directions | +0.909 |
Dracula | +0.886 |
warr | +0.805 |
nces | +0.800 |
Neg logit contributions | |
isable | -0.811 |
ォ | -0.791 |
ident | -0.778 |
ERG | -0.774 |
OLD | -0.687 |
Token | Mult |
Token ID | 7854 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
enburg | +0.197 |
Directions | +0.186 |
Dracula | +0.183 |
warr | +0.167 |
etheless | +0.165 |
Neg logit contributions | |
isable | -0.180 |
ォ | -0.174 |
ident | -0.172 |
ERG | -0.172 |
OLD | -0.153 |
Token | if |
Token ID | 361 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
isable | +0.537 |
ォ | +0.530 |
ERG | +0.519 |
ident | +0.508 |
OLD | +0.452 |
Neg logit contributions | |
enburg | -0.750 |
Directions | -0.708 |
Dracula | -0.699 |
warr | -0.642 |
nces | -0.636 |
Token | acet |
Token ID | 23253 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
enburg | +0.791 |
Directions | +0.743 |
Dracula | +0.734 |
warr | +0.670 |
etheless | +0.663 |
Neg logit contributions | |
isable | -0.603 |
ォ | -0.596 |
ERG | -0.581 |
ident | -0.575 |
OLD | -0.505 |
Token | ed |
Token ID | 276 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
enburg | +5.082 |
Directions | +4.868 |
Dracula | +4.783 |
warr | +4.430 |
nces | +4.382 |
Neg logit contributions | |
isable | -2.942 |
ォ | -2.876 |
ERG | -2.797 |
ident | -2.731 |
abba | -2.393 |
Token | study |
Token ID | 2050 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
enburg | +2.371 |
Directions | +2.263 |
Dracula | +2.225 |
warr | +2.039 |
nces | +2.007 |
Neg logit contributions | |
isable | -1.834 |
ォ | -1.791 |
ERG | -1.762 |
ident | -1.751 |
abba | -1.545 |
Token | of |
Token ID | 286 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
enburg | +0.523 |
Directions | +0.496 |
Dracula | +0.487 |
warr | +0.448 |
etheless | +0.442 |
Neg logit contributions | |
isable | -0.437 |
ォ | -0.423 |
ERG | -0.418 |
ident | -0.418 |
OLD | -0.369 |
Token | the |
Token ID | 262 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
isable | +0.392 |
ォ | +0.377 |
ident | +0.376 |
ERG | +0.374 |
camp | +0.329 |
Neg logit contributions | |
enburg | -0.477 |
Directions | -0.450 |
Dracula | -0.443 |
warr | -0.410 |
etheless | -0.406 |
Token | br |
Token ID | 865 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
isable | +0.141 |
ォ | +0.136 |
ident | +0.135 |
ERG | +0.135 |
OLD | +0.119 |
Neg logit contributions | |
enburg | -0.162 |
Directions | -0.153 |
Dracula | -0.150 |
warr | -0.138 |
etheless | -0.137 |
Token | . |
Token ID | 13 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
enburg | +0.880 |
Directions | +0.832 |
Dracula | +0.822 |
warr | +0.749 |
etheless | +0.745 |
Neg logit contributions | |
isable | -0.740 |
ォ | -0.718 |
ERG | -0.711 |
ident | -0.710 |
OLD | -0.626 |
Token |
|
Token ID | 198 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
isable | +0.070 |
ォ | +0.068 |
ident | +0.067 |
ERG | +0.067 |
OLD | +0.059 |
Neg logit contributions | |
enburg | -0.091 |
Directions | -0.086 |
Dracula | -0.085 |
warr | -0.079 |
etheless | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
isable | +0.277 |
ident | +0.267 |
ERG | +0.266 |
ォ | +0.265 |
OLD | +0.231 |
Neg logit contributions | |
enburg | -0.408 |
Directions | -0.387 |
Dracula | -0.381 |
warr | -0.368 |
etheless | -0.362 |
Token | Thanks |
Token ID | 9690 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
enburg | +0.093 |
Directions | +0.088 |
Dracula | +0.087 |
warr | +0.080 |
etheless | +0.079 |
Neg logit contributions | |
isable | -0.068 |
ォ | -0.065 |
ident | -0.065 |
ERG | -0.065 |
OLD | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | -8.84e-02 |
Pos logit contributions | |
enburg | +0.580 |
Directions | +0.548 |
Dracula | +0.540 |
warr | +0.493 |
etheless | +0.487 |
Neg logit contributions | |
isable | -0.525 |
ォ | -0.508 |
ERG | -0.505 |
ident | -0.504 |
OLD | -0.446 |
Token | Adobe |
Token ID | 21771 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
enburg | +0.193 |
Directions | +0.182 |
Dracula | +0.179 |
warr | +0.166 |
etheless | +0.163 |
Neg logit contributions | |
isable | -0.165 |
ォ | -0.160 |
ident | -0.159 |
ERG | -0.158 |
OLD | -0.139 |
Token | AIR |
Token ID | 31600 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
enburg | +3.810 |
Directions | +3.603 |
Dracula | +3.559 |
etheless | +3.170 |
warr | +3.158 |
Neg logit contributions | |
isable | -4.255 |
ォ | -4.151 |
ident | -4.042 |
ERG | -4.020 |
abba | -3.663 |
Token | for |
Token ID | 329 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
enburg | +0.868 |
Directions | +0.820 |
Dracula | +0.808 |
warr | +0.749 |
etheless | +0.742 |
Neg logit contributions | |
isable | -0.644 |
ォ | -0.621 |
ERG | -0.617 |
ident | -0.616 |
OLD | -0.540 |
Token | Android |
Token ID | 5565 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
enburg | +0.038 |
Directions | +0.036 |
Dracula | +0.035 |
warr | +0.032 |
etheless | +0.032 |
Neg logit contributions | |
isable | -0.042 |
ident | -0.041 |
ERG | -0.040 |
ォ | -0.040 |
camp | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
isable | +0.245 |
ォ | +0.237 |
ident | +0.236 |
ERG | +0.235 |
OLD | +0.208 |
Neg logit contributions | |
enburg | -0.277 |
Directions | -0.260 |
Dracula | -0.257 |
warr | -0.236 |
etheless | -0.233 |
Token | some |
Token ID | 617 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
isable | +0.550 |
ident | +0.532 |
ERG | +0.531 |
ォ | +0.526 |
camp | +0.470 |
Neg logit contributions | |
enburg | -0.563 |
Directions | -0.525 |
Dracula | -0.518 |
warr | -0.478 |
etheless | -0.470 |
Token | Great |
Token ID | 3878 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
enburg | +0.220 |
Directions | +0.206 |
Dracula | +0.202 |
warr | +0.185 |
etheless | +0.182 |
Neg logit contributions | |
isable | -0.234 |
ォ | -0.227 |
ident | -0.227 |
ERG | -0.225 |
camp | -0.203 |
Token | Eight |
Token ID | 18087 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
enburg | +0.977 |
Directions | +0.923 |
Dracula | +0.909 |
warr | +0.848 |
nces | +0.835 |
Neg logit contributions | |
isable | -0.788 |
ident | -0.769 |
ォ | -0.765 |
ERG | -0.761 |
camp | -0.671 |
Token | sent |
Token ID | 1908 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
isable | +0.583 |
ォ | +0.567 |
ERG | +0.563 |
ident | +0.563 |
camp | +0.502 |
Neg logit contributions | |
enburg | -0.628 |
Directions | -0.594 |
Dracula | -0.585 |
warr | -0.540 |
etheless | -0.532 |
Token | a |
Token ID | 257 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
isable | +0.339 |
ォ | +0.328 |
ERG | +0.326 |
ident | +0.326 |
OLD | +0.287 |
Neg logit contributions | |
enburg | -0.410 |
Directions | -0.387 |
Dracula | -0.380 |
warr | -0.352 |
etheless | -0.347 |
Token | powerful |
Token ID | 3665 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
enburg | +0.245 |
Directions | +0.231 |
Dracula | +0.226 |
warr | +0.208 |
nces | +0.206 |
Neg logit contributions | |
isable | -0.235 |
ォ | -0.227 |
ident | -0.226 |
ERG | -0.225 |
camp | -0.201 |
Token | wr |
Token ID | 1319 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
isable | +0.304 |
ident | +0.294 |
ォ | +0.293 |
ERG | +0.293 |
camp | +0.266 |
Neg logit contributions | |
enburg | -0.272 |
Directions | -0.255 |
Dracula | -0.249 |
warr | -0.229 |
nces | -0.229 |
Token | ister |
Token ID | 1694 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
enburg | +3.350 |
Directions | +2.864 |
Dracula | +2.857 |
mble | +2.437 |
warr | +2.412 |
Neg logit contributions | |
ォ | -4.681 |
ERG | -4.676 |
isable | -4.672 |
ident | -4.328 |
camp | -4.138 |
Token | beyond |
Token ID | 3675 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
enburg | +1.975 |
Directions | +1.856 |
Dracula | +1.836 |
warr | +1.684 |
etheless | +1.665 |
Neg logit contributions | |
isable | -1.607 |
ォ | -1.559 |
ERG | -1.537 |
ident | -1.536 |
OLD | -1.355 |
Token | the |
Token ID | 262 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
enburg | +0.232 |
Directions | +0.219 |
Dracula | +0.215 |
warr | +0.197 |
etheless | +0.195 |
Neg logit contributions | |
isable | -0.210 |
ォ | -0.204 |
ident | -0.202 |
ERG | -0.202 |
OLD | -0.178 |
Token | reach |
Token ID | 3151 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
enburg | +1.060 |
Directions | +0.996 |
Dracula | +0.978 |
warr | +0.900 |
nces | +0.891 |
Neg logit contributions | |
isable | -1.014 |
ォ | -0.978 |
ident | -0.976 |
ERG | -0.965 |
camp | -0.867 |
Token | of |
Token ID | 286 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
enburg | +0.981 |
Directions | +0.922 |
Dracula | +0.909 |
warr | +0.832 |
etheless | +0.825 |
Neg logit contributions | |
isable | -0.832 |
ォ | -0.805 |
ERG | -0.801 |
ident | -0.796 |
OLD | -0.702 |
Token | it |
Token ID | 340 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
enburg | +0.272 |
Directions | +0.256 |
Dracula | +0.252 |
warr | +0.232 |
etheless | +0.229 |
Neg logit contributions | |
isable | -0.244 |
ォ | -0.236 |
ident | -0.234 |
ERG | -0.234 |
OLD | -0.207 |
Token | himself |
Token ID | 2241 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
enburg | +0.077 |
Directions | +0.073 |
Dracula | +0.072 |
warr | +0.066 |
etheless | +0.066 |
Neg logit contributions | |
isable | -0.062 |
ォ | -0.060 |
ident | -0.060 |
ERG | -0.060 |
OLD | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
enburg | +0.420 |
Directions | +0.397 |
Dracula | +0.392 |
warr | +0.359 |
etheless | +0.355 |
Neg logit contributions | |
isable | -0.343 |
ォ | -0.331 |
ident | -0.328 |
ERG | -0.328 |
OLD | -0.287 |
Token | His |
Token ID | 2399 |
Feature activation | -9.35e-02 |
Pos logit contributions | |
isable | +0.083 |
ォ | +0.081 |
ident | +0.080 |
ERG | +0.080 |
OLD | +0.071 |
Neg logit contributions | |
enburg | -0.089 |
Directions | -0.083 |
Dracula | -0.082 |
warr | -0.076 |
etheless | -0.075 |
Token | face |
Token ID | 1986 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
enburg | +0.197 |
Directions | +0.185 |
Dracula | +0.182 |
warr | +0.167 |
etheless | +0.165 |
Neg logit contributions | |
isable | -0.182 |
ォ | -0.177 |
ident | -0.175 |
ERG | -0.175 |
OLD | -0.155 |
Token | cont |
Token ID | 542 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
enburg | +1.754 |
Directions | +1.640 |
Dracula | +1.629 |
warr | +1.459 |
nces | +1.443 |
Neg logit contributions | |
isable | -1.895 |
ォ | -1.861 |
ERG | -1.828 |
ident | -1.817 |
abba | -1.631 |
Token | orted |
Token ID | 9741 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
enburg | +4.260 |
Dracula | +3.971 |
Directions | +3.920 |
nces | +3.530 |
etheless | +3.474 |
Neg logit contributions | |
isable | -3.641 |
ォ | -3.624 |
ERG | -3.577 |
ident | -3.399 |
camp | -3.078 |
Token | slightly |
Token ID | 4622 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
enburg | +0.958 |
Directions | +0.904 |
Dracula | +0.892 |
warr | +0.818 |
etheless | +0.807 |
Neg logit contributions | |
isable | -0.832 |
ォ | -0.808 |
ident | -0.799 |
ERG | -0.798 |
OLD | -0.704 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
enburg | +0.793 |
Directions | +0.747 |
Dracula | +0.739 |
warr | +0.678 |
etheless | +0.670 |
Neg logit contributions | |
isable | -0.646 |
ォ | -0.626 |
ident | -0.619 |
ERG | -0.618 |
OLD | -0.543 |
Token | � |
Token ID | 564 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
enburg | +0.272 |
Directions | +0.256 |
Dracula | +0.251 |
warr | +0.232 |
etheless | +0.229 |
Neg logit contributions | |
isable | -0.263 |
ォ | -0.256 |
ident | -0.254 |
ERG | -0.253 |
OLD | -0.226 |
Token | � |
Token ID | 250 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
enburg | +0.444 |
Directions | +0.415 |
Dracula | +0.409 |
warr | +0.371 |
etheless | +0.365 |
Neg logit contributions | |
isable | -0.489 |
ォ | -0.472 |
ident | -0.471 |
ERG | -0.469 |
abba | -0.420 |
Token | is |
Token ID | 318 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
isable | +0.006 |
ォ | +0.005 |
ident | +0.005 |
ERG | +0.005 |
OLD | +0.005 |
Neg logit contributions | |
enburg | -0.007 |
Directions | -0.006 |
Dracula | -0.006 |
warr | -0.006 |
etheless | -0.006 |
Token | strictly |
Token ID | 14084 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
enburg | +0.432 |
Directions | +0.400 |
Dracula | +0.392 |
warr | +0.350 |
etheless | +0.343 |
Neg logit contributions | |
isable | -0.610 |
ォ | -0.596 |
ident | -0.591 |
ERG | -0.590 |
OLD | -0.536 |
Token | prohibited |
Token ID | 12244 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
enburg | +0.452 |
Directions | +0.413 |
Dracula | +0.403 |
warr | +0.351 |
etheless | +0.342 |
Neg logit contributions | |
isable | -0.848 |
ォ | -0.829 |
ident | -0.824 |
ERG | -0.823 |
OLD | -0.755 |
Token | without |
Token ID | 1231 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
enburg | +1.023 |
Directions | +0.966 |
Dracula | +0.950 |
warr | +0.874 |
etheless | +0.861 |
Neg logit contributions | |
isable | -0.872 |
ォ | -0.848 |
ERG | -0.837 |
ident | -0.837 |
OLD | -0.738 |
Token | the |
Token ID | 262 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
enburg | +0.102 |
Directions | +0.096 |
Dracula | +0.094 |
warr | +0.086 |
etheless | +0.085 |
Neg logit contributions | |
isable | -0.105 |
ォ | -0.102 |
ERG | -0.101 |
ident | -0.101 |
OLD | -0.090 |
Token | express |
Token ID | 4911 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
enburg | +0.719 |
Directions | +0.675 |
Dracula | +0.662 |
warr | +0.602 |
nces | +0.590 |
Neg logit contributions | |
isable | -0.758 |
ォ | -0.742 |
ERG | -0.733 |
ident | -0.731 |
OLD | -0.653 |
Token | written |
Token ID | 3194 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
enburg | +2.696 |
Directions | +2.515 |
Dracula | +2.382 |
warr | +2.031 |
nces | +1.950 |
Neg logit contributions | |
ォ | -5.264 |
isable | -5.164 |
ERG | -5.130 |
ident | -5.069 |
abba | -4.636 |
Token | permission |
Token ID | 7170 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
enburg | +1.175 |
Directions | +1.099 |
Dracula | +1.068 |
warr | +0.953 |
nces | +0.924 |
Neg logit contributions | |
isable | -1.640 |
ォ | -1.624 |
ERG | -1.599 |
ident | -1.592 |
OLD | -1.444 |
Token | of |
Token ID | 286 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
enburg | +1.967 |
Directions | +1.859 |
Dracula | +1.830 |
warr | +1.688 |
etheless | +1.647 |
Neg logit contributions | |
isable | -1.738 |
ォ | -1.702 |
ERG | -1.667 |
ident | -1.659 |
abba | -1.466 |
Token | Money |
Token ID | 12911 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
isable | +0.069 |
ォ | +0.067 |
ident | +0.066 |
ERG | +0.066 |
OLD | +0.058 |
Neg logit contributions | |
enburg | -0.083 |
Directions | -0.079 |
Dracula | -0.078 |
warr | -0.071 |
etheless | -0.071 |
Token | Morning |
Token ID | 14410 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
isable | +0.488 |
ォ | +0.472 |
ident | +0.471 |
ERG | +0.469 |
OLD | +0.415 |
Neg logit contributions | |
enburg | -0.597 |
Directions | -0.560 |
Dracula | -0.558 |
warr | -0.523 |
etheless | -0.513 |
Token | play |
Token ID | 711 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
isable | +0.356 |
ォ | +0.344 |
ident | +0.342 |
ERG | +0.341 |
camp | +0.301 |
Neg logit contributions | |
enburg | -0.401 |
Directions | -0.378 |
Dracula | -0.372 |
warr | -0.342 |
etheless | -0.338 |
Token | enough |
Token ID | 1576 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
enburg | +0.181 |
Directions | +0.172 |
Dracula | +0.169 |
warr | +0.156 |
etheless | +0.154 |
Neg logit contributions | |
isable | -0.145 |
ォ | -0.140 |
ident | -0.139 |
ERG | -0.139 |
OLD | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
enburg | +0.663 |
Directions | +0.623 |
Dracula | +0.615 |
warr | +0.562 |
etheless | +0.556 |
Neg logit contributions | |
isable | -0.588 |
ォ | -0.572 |
ident | -0.565 |
ERG | -0.565 |
OLD | -0.500 |
Token | Brad |
Token ID | 8114 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
isable | +0.369 |
ォ | +0.360 |
ident | +0.357 |
ERG | +0.357 |
OLD | +0.316 |
Neg logit contributions | |
enburg | -0.424 |
Directions | -0.397 |
Dracula | -0.391 |
warr | -0.366 |
etheless | -0.362 |
Token | en |
Token ID | 268 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
isable | +0.291 |
ォ | +0.280 |
ERG | +0.276 |
ident | +0.275 |
abba | +0.239 |
Neg logit contributions | |
enburg | -0.406 |
Directions | -0.384 |
Dracula | -0.378 |
warr | -0.350 |
nces | -0.344 |
Token | Holt |
Token ID | 28330 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
isable | +0.193 |
ォ | +0.188 |
ERG | +0.186 |
ident | +0.186 |
abba | +0.170 |
Neg logit contributions | |
enburg | -0.108 |
Directions | -0.100 |
Dracula | -0.098 |
warr | -0.086 |
etheless | -0.083 |
Token | by |
Token ID | 1525 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
enburg | +3.212 |
Directions | +2.934 |
Dracula | +2.900 |
warr | +2.529 |
nces | +2.484 |
Neg logit contributions | |
isable | -4.840 |
ォ | -4.687 |
ERG | -4.620 |
ident | -4.616 |
abba | -4.210 |
Token | lacks |
Token ID | 16523 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
enburg | +0.800 |
Directions | +0.756 |
Dracula | +0.746 |
warr | +0.689 |
etheless | +0.680 |
Neg logit contributions | |
isable | -0.589 |
ォ | -0.568 |
ident | -0.563 |
ERG | -0.561 |
OLD | -0.490 |
Token | the |
Token ID | 262 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
enburg | +0.155 |
Directions | +0.147 |
Dracula | +0.144 |
warr | +0.133 |
etheless | +0.131 |
Neg logit contributions | |
isable | -0.131 |
ォ | -0.127 |
ident | -0.126 |
ERG | -0.126 |
OLD | -0.111 |
Token | stamina |
Token ID | 32312 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
enburg | +0.233 |
Directions | +0.219 |
Dracula | +0.216 |
warr | +0.198 |
etheless | +0.195 |
Neg logit contributions | |
isable | -0.215 |
ォ | -0.208 |
ident | -0.207 |
ERG | -0.206 |
OLD | -0.182 |
Token | to |
Token ID | 284 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
isable | +0.806 |
ident | +0.773 |
ォ | +0.772 |
ERG | +0.770 |
camp | +0.679 |
Neg logit contributions | |
enburg | -0.999 |
Directions | -0.948 |
Dracula | -0.926 |
warr | -0.860 |
etheless | -0.850 |
Token |
|
Token ID | 198 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
enburg | +0.495 |
Directions | +0.470 |
Dracula | +0.463 |
warr | +0.434 |
etheless | +0.428 |
Neg logit contributions | |
isable | -0.358 |
ォ | -0.345 |
ident | -0.343 |
ERG | -0.342 |
OLD | -0.298 |
Token | With |
Token ID | 3152 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
enburg | +0.842 |
Directions | +0.798 |
Dracula | +0.787 |
warr | +0.724 |
nces | +0.717 |
Neg logit contributions | |
isable | -0.583 |
ォ | -0.560 |
ident | -0.552 |
ERG | -0.551 |
abba | -0.477 |
Token | Josh |
Token ID | 8518 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
isable | +0.180 |
ォ | +0.175 |
ident | +0.174 |
ERG | +0.173 |
OLD | +0.153 |
Neg logit contributions | |
enburg | -0.203 |
Directions | -0.191 |
Dracula | -0.188 |
warr | -0.173 |
etheless | -0.171 |
Token | Doct |
Token ID | 32155 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
enburg | +0.841 |
Directions | +0.789 |
Dracula | +0.776 |
warr | +0.707 |
etheless | +0.698 |
Neg logit contributions | |
isable | -0.876 |
ォ | -0.850 |
ident | -0.846 |
ERG | -0.843 |
OLD | -0.754 |
Token | son |
Token ID | 1559 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
enburg | +1.004 |
Directions | +0.951 |
Dracula | +0.932 |
warr | +0.850 |
nces | +0.836 |
Neg logit contributions | |
isable | -0.851 |
ォ | -0.822 |
ERG | -0.808 |
ident | -0.804 |
OLD | -0.712 |
Token | finally |
Token ID | 3443 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
enburg | +1.804 |
Directions | +1.694 |
Dracula | +1.669 |
warr | +1.525 |
nces | +1.496 |
Neg logit contributions | |
isable | -1.695 |
ォ | -1.648 |
ident | -1.631 |
ERG | -1.624 |
abba | -1.442 |
Token | becoming |
Token ID | 5033 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
enburg | +4.154 |
Directions | +3.879 |
Dracula | +3.865 |
warr | +3.495 |
nces | +3.412 |
Neg logit contributions | |
isable | -3.765 |
ォ | -3.697 |
ERG | -3.619 |
ident | -3.614 |
abba | -3.235 |
Token | a |
Token ID | 257 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
enburg | +1.512 |
Directions | +1.435 |
Dracula | +1.408 |
warr | +1.291 |
etheless | +1.270 |
Neg logit contributions | |
isable | -1.286 |
ォ | -1.252 |
ERG | -1.235 |
ident | -1.231 |
abba | -1.088 |
Token | force |
Token ID | 2700 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
enburg | +0.870 |
Directions | +0.821 |
Dracula | +0.807 |
warr | +0.739 |
etheless | +0.728 |
Neg logit contributions | |
isable | -0.813 |
ォ | -0.790 |
ident | -0.782 |
ERG | -0.782 |
OLD | -0.694 |
Token | in |
Token ID | 287 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
enburg | +0.039 |
Directions | +0.037 |
Dracula | +0.036 |
warr | +0.033 |
etheless | +0.033 |
Neg logit contributions | |
isable | -0.032 |
ォ | -0.031 |
ident | -0.031 |
ERG | -0.031 |
OLD | -0.027 |
Token | Washington |
Token ID | 2669 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
enburg | +0.898 |
Directions | +0.855 |
Dracula | +0.835 |
warr | +0.770 |
nces | +0.752 |
Neg logit contributions | |
isable | -0.797 |
ォ | -0.778 |
ERG | -0.766 |
ident | -0.765 |
OLD | -0.676 |
Token | When |
Token ID | 1649 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
isable | +0.549 |
ォ | +0.531 |
ident | +0.529 |
ERG | +0.529 |
OLD | +0.463 |
Neg logit contributions | |
enburg | -0.717 |
Directions | -0.676 |
Dracula | -0.667 |
warr | -0.619 |
etheless | -0.612 |
Token | we |
Token ID | 356 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
enburg | +0.037 |
Directions | +0.035 |
Dracula | +0.034 |
warr | +0.032 |
etheless | +0.031 |
Neg logit contributions | |
isable | -0.032 |
ォ | -0.031 |
ident | -0.031 |
ERG | -0.030 |
OLD | -0.027 |
Token | watched |
Token ID | 7342 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
isable | +0.010 |
ォ | +0.010 |
ident | +0.010 |
ERG | +0.010 |
OLD | +0.009 |
Neg logit contributions | |
enburg | -0.012 |
Directions | -0.012 |
Dracula | -0.011 |
warr | -0.011 |
etheless | -0.010 |
Token | this |
Token ID | 428 |
Feature activation | +6.64e-02 |
Pos logit contributions | |
isable | +1.357 |
ident | +1.316 |
ERG | +1.314 |
ォ | +1.305 |
camp | +1.176 |
Neg logit contributions | |
enburg | -1.638 |
Directions | -1.562 |
Dracula | -1.504 |
warr | -1.461 |
etheless | -1.442 |
Token | little |
Token ID | 1310 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
isable | +0.140 |
ident | +0.135 |
ォ | +0.135 |
ERG | +0.135 |
camp | +0.122 |
Neg logit contributions | |
enburg | -0.129 |
Directions | -0.122 |
Dracula | -0.117 |
warr | -0.110 |
etheless | -0.109 |
Token | gag |
Token ID | 32261 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
enburg | +0.925 |
Directions | +0.825 |
Dracula | +0.794 |
warr | +0.662 |
etheless | +0.641 |
Neg logit contributions | |
isable | -2.486 |
ォ | -2.434 |
ident | -2.423 |
ERG | -2.422 |
OLD | -2.241 |
Token | with |
Token ID | 351 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
enburg | +3.822 |
Directions | +3.568 |
Dracula | +3.538 |
nces | +3.140 |
warr | +3.131 |
Neg logit contributions | |
isable | -4.096 |
ォ | -4.014 |
ident | -3.926 |
ERG | -3.919 |
OLD | -3.520 |
Token | Erin |
Token ID | 28894 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
isable | +0.400 |
ォ | +0.390 |
ident | +0.388 |
ERG | +0.388 |
OLD | +0.351 |
Neg logit contributions | |
enburg | -0.284 |
Directions | -0.263 |
Dracula | -0.258 |
warr | -0.231 |
etheless | -0.227 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
enburg | +0.647 |
Directions | +0.618 |
Dracula | +0.606 |
warr | +0.581 |
etheless | +0.568 |
Neg logit contributions | |
isable | -0.461 |
ident | -0.445 |
ERG | -0.440 |
ォ | -0.438 |
camp | -0.389 |
Token | Jill |
Token ID | 22772 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
isable | +0.201 |
ォ | +0.194 |
ident | +0.194 |
ERG | +0.193 |
camp | +0.172 |
Neg logit contributions | |
enburg | -0.207 |
Directions | -0.194 |
Dracula | -0.192 |
warr | -0.177 |
etheless | -0.175 |
Token | ian |
Token ID | 666 |
Feature activation | +6.04e-03 |
Pos logit contributions | |
isable | +0.681 |
ident | +0.661 |
ERG | +0.652 |
ォ | +0.650 |
camp | +0.563 |
Neg logit contributions | |
enburg | -0.954 |
Directions | -0.923 |
Dracula | -0.906 |
warr | -0.859 |
etheless | -0.846 |
Token | v |
Token ID | 410 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
enburg | +0.421 |
Directions | +0.396 |
Dracula | +0.390 |
warr | +0.359 |
etheless | +0.354 |
Neg logit contributions | |
isable | -0.376 |
ォ | -0.363 |
ident | -0.361 |
ERG | -0.360 |
OLD | -0.319 |
Token | 3 |
Token ID | 18 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
enburg | +0.864 |
Directions | +0.810 |
Dracula | +0.796 |
warr | +0.725 |
etheless | +0.714 |
Neg logit contributions | |
isable | -0.932 |
ォ | -0.904 |
ident | -0.899 |
ERG | -0.897 |
OLD | -0.803 |
Token | . |
Token ID | 13 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
enburg | +0.559 |
Directions | +0.529 |
Dracula | +0.521 |
warr | +0.481 |
etheless | +0.475 |
Neg logit contributions | |
isable | -0.445 |
ォ | -0.430 |
ident | -0.427 |
ERG | -0.426 |
OLD | -0.374 |
Token | 7 |
Token ID | 22 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
enburg | +0.543 |
Directions | +0.510 |
Dracula | +0.501 |
warr | +0.461 |
etheless | +0.454 |
Neg logit contributions | |
isable | -0.574 |
ォ | -0.558 |
ident | -0.555 |
ERG | -0.554 |
OLD | -0.497 |
Token | ( |
Token ID | 357 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
enburg | +0.974 |
Directions | +0.922 |
Dracula | +0.907 |
warr | +0.834 |
etheless | +0.823 |
Neg logit contributions | |
isable | -0.798 |
ォ | -0.774 |
ident | -0.765 |
ERG | -0.763 |
OLD | -0.670 |
Token | Product |
Token ID | 15667 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
enburg | +0.457 |
Directions | +0.431 |
Dracula | +0.425 |
warr | +0.391 |
etheless | +0.386 |
Neg logit contributions | |
isable | -0.384 |
ォ | -0.371 |
ident | -0.369 |
ERG | -0.368 |
OLD | -0.324 |
Token | is |
Token ID | 318 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
enburg | +3.976 |
Directions | +3.775 |
Dracula | +3.705 |
warr | +3.362 |
etheless | +3.230 |
Neg logit contributions | |
isable | -3.841 |
ォ | -3.778 |
ident | -3.711 |
ERG | -3.709 |
OLD | -3.328 |
Token | delivered |
Token ID | 6793 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
isable | +0.384 |
ォ | +0.369 |
ident | +0.368 |
ERG | +0.368 |
OLD | +0.327 |
Neg logit contributions | |
enburg | -0.414 |
Directions | -0.389 |
Dracula | -0.383 |
warr | -0.352 |
etheless | -0.349 |
Token | as |
Token ID | 355 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
isable | +1.242 |
ident | +1.189 |
ォ | +1.187 |
ERG | +1.183 |
OLD | +1.048 |
Neg logit contributions | |
enburg | -1.456 |
Directions | -1.373 |
Dracula | -1.357 |
warr | -1.252 |
etheless | -1.238 |
Token | a |
Token ID | 257 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
isable | +0.228 |
ォ | +0.220 |
ident | +0.219 |
ERG | +0.218 |
OLD | +0.193 |
Neg logit contributions | |
enburg | -0.267 |
Directions | -0.251 |
Dracula | -0.247 |
warr | -0.228 |
etheless | -0.225 |
Token | direct |
Token ID | 1277 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
enburg | +0.063 |
Directions | +0.060 |
Dracula | +0.059 |
warr | +0.054 |
etheless | +0.053 |
Neg logit contributions | |
isable | -0.059 |
ォ | -0.058 |
ident | -0.057 |
ERG | -0.057 |
OLD | -0.051 |
Token | contacted |
Token ID | 11237 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
enburg | +0.660 |
Directions | +0.623 |
Dracula | +0.614 |
warr | +0.564 |
nces | +0.556 |
Neg logit contributions | |
isable | -0.547 |
ォ | -0.530 |
ERG | -0.524 |
ident | -0.523 |
OLD | -0.461 |
Token | 60 |
Token ID | 3126 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
enburg | +0.128 |
Directions | +0.121 |
Dracula | +0.119 |
warr | +0.110 |
etheless | +0.108 |
Neg logit contributions | |
isable | -0.110 |
ォ | -0.107 |
ident | -0.106 |
ERG | -0.106 |
OLD | -0.093 |
Token | people |
Token ID | 661 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
enburg | +0.120 |
Directions | +0.114 |
Dracula | +0.112 |
warr | +0.104 |
etheless | +0.102 |
Neg logit contributions | |
isable | -0.088 |
ォ | -0.085 |
ERG | -0.084 |
ident | -0.084 |
abba | -0.073 |
Token | daily |
Token ID | 4445 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
enburg | +0.720 |
Directions | +0.679 |
Dracula | +0.669 |
warr | +0.615 |
etheless | +0.607 |
Neg logit contributions | |
isable | -0.616 |
ォ | -0.596 |
ident | -0.591 |
ERG | -0.590 |
OLD | -0.520 |
Token | to |
Token ID | 284 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
enburg | +0.500 |
Directions | +0.472 |
Dracula | +0.465 |
warr | +0.429 |
etheless | +0.423 |
Neg logit contributions | |
isable | -0.422 |
ォ | -0.407 |
ident | -0.406 |
ERG | -0.404 |
OLD | -0.356 |
Token | express |
Token ID | 4911 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
enburg | +0.276 |
Directions | +0.262 |
Dracula | +0.257 |
warr | +0.239 |
etheless | +0.239 |
Neg logit contributions | |
isable | -0.203 |
ident | -0.195 |
ォ | -0.194 |
ERG | -0.192 |
camp | -0.171 |
Token | her |
Token ID | 607 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
enburg | +3.745 |
Directions | +3.518 |
Dracula | +3.464 |
warr | +3.143 |
nces | +2.994 |
Neg logit contributions | |
ォ | -4.075 |
isable | -4.073 |
ERG | -3.946 |
ident | -3.885 |
abba | -3.565 |
Token | support |
Token ID | 1104 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
enburg | +0.711 |
Directions | +0.668 |
Dracula | +0.657 |
warr | +0.600 |
etheless | +0.590 |
Neg logit contributions | |
isable | -0.702 |
ォ | -0.683 |
ERG | -0.676 |
ident | -0.676 |
OLD | -0.602 |
Token | for |
Token ID | 329 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
enburg | +0.382 |
Directions | +0.360 |
Dracula | +0.355 |
warr | +0.325 |
etheless | +0.321 |
Neg logit contributions | |
isable | -0.315 |
ォ | -0.309 |
ident | -0.302 |
ERG | -0.302 |
abba | -0.266 |
Token | the |
Token ID | 262 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
isable | +0.390 |
ォ | +0.377 |
ident | +0.374 |
ERG | +0.374 |
OLD | +0.331 |
Neg logit contributions | |
enburg | -0.436 |
Directions | -0.411 |
Dracula | -0.405 |
warr | -0.371 |
etheless | -0.366 |
Token | Vermont |
Token ID | 16033 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
isable | +0.279 |
ォ | +0.272 |
ERG | +0.269 |
ident | +0.269 |
OLD | +0.241 |
Neg logit contributions | |
enburg | -0.253 |
Directions | -0.238 |
Dracula | -0.233 |
warr | -0.212 |
etheless | -0.208 |
Token | NBC |
Token ID | 9716 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
enburg | +0.291 |
Directions | +0.276 |
Dracula | +0.271 |
warr | +0.249 |
etheless | +0.245 |
Neg logit contributions | |
isable | -0.251 |
ォ | -0.244 |
ident | -0.240 |
ERG | -0.239 |
OLD | -0.212 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
enburg | +1.761 |
Directions | +1.677 |
Dracula | +1.637 |
warr | +1.492 |
etheless | +1.469 |
Neg logit contributions | |
isable | -1.760 |
ォ | -1.705 |
ident | -1.677 |
ERG | -1.671 |
abba | -1.507 |
Token | while |
Token ID | 981 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
isable | +0.438 |
ident | +0.423 |
ォ | +0.423 |
ERG | +0.422 |
OLD | +0.371 |
Neg logit contributions | |
enburg | -0.516 |
Directions | -0.483 |
Dracula | -0.478 |
warr | -0.442 |
nces | -0.435 |
Token | CBS |
Token ID | 11133 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
enburg | +1.157 |
Directions | +1.089 |
Dracula | +1.072 |
warr | +0.984 |
etheless | +0.970 |
Neg logit contributions | |
isable | -1.053 |
ォ | -1.020 |
ident | -1.013 |
ERG | -1.011 |
OLD | -0.895 |
Token | ' |
Token ID | 6 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
enburg | +1.618 |
Directions | +1.526 |
Dracula | +1.504 |
warr | +1.366 |
etheless | +1.347 |
Neg logit contributions | |
isable | -1.584 |
ォ | -1.538 |
ident | -1.516 |
ERG | -1.507 |
abba | -1.353 |
Token | anchor |
Token ID | 18021 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
isable | +1.202 |
ォ | +1.163 |
ident | +1.162 |
ERG | +1.159 |
camp | +1.026 |
Neg logit contributions | |
enburg | -1.358 |
Directions | -1.271 |
Dracula | -1.248 |
warr | -1.162 |
etheless | -1.144 |
Token | Scott |
Token ID | 4746 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
isable | +2.849 |
ident | +2.744 |
ERG | +2.722 |
ォ | +2.721 |
camp | +2.452 |
Neg logit contributions | |
enburg | -3.403 |
Directions | -3.163 |
Dracula | -3.067 |
warr | -2.916 |
etheless | -2.910 |
Token | Pel |
Token ID | 12903 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
isable | +0.830 |
ォ | +0.783 |
ident | +0.774 |
ERG | +0.767 |
OLD | +0.642 |
Neg logit contributions | |
enburg | -1.750 |
Directions | -1.672 |
Dracula | -1.659 |
warr | -1.534 |
etheless | -1.515 |
Token | ley |
Token ID | 1636 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
isable | +0.232 |
ident | +0.222 |
ォ | +0.221 |
ERG | +0.216 |
abba | +0.186 |
Neg logit contributions | |
enburg | -0.391 |
Directions | -0.378 |
Dracula | -0.377 |
warr | -0.354 |
nces | -0.347 |
Token | is |
Token ID | 318 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
isable | +0.859 |
ォ | +0.832 |
ident | +0.827 |
ERG | +0.826 |
OLD | +0.722 |
Neg logit contributions | |
enburg | -1.086 |
Directions | -1.029 |
Dracula | -1.014 |
warr | -0.937 |
etheless | -0.927 |
Token | leading |
Token ID | 3756 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
enburg | +0.859 |
Directions | +0.803 |
Dracula | +0.790 |
warr | +0.714 |
etheless | +0.703 |
Neg logit contributions | |
isable | -0.987 |
ォ | -0.960 |
ident | -0.952 |
ERG | -0.950 |
OLD | -0.854 |
Token | both |
Token ID | 1111 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
isable | +0.659 |
ォ | +0.633 |
ident | +0.632 |
ERG | +0.631 |
OLD | +0.559 |
Neg logit contributions | |
enburg | -0.700 |
Directions | -0.656 |
Dracula | -0.645 |
warr | -0.593 |
etheless | -0.589 |
Token | , |
Token ID | 11 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
enburg | +0.142 |
Directions | +0.135 |
Dracula | +0.133 |
warr | +0.123 |
etheless | +0.122 |
Neg logit contributions | |
isable | -0.107 |
ォ | -0.103 |
ident | -0.102 |
ERG | -0.102 |
OLD | -0.089 |
Token | for |
Token ID | 329 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
isable | +0.698 |
ォ | +0.669 |
ident | +0.669 |
ERG | +0.666 |
camp | +0.581 |
Neg logit contributions | |
enburg | -0.921 |
Directions | -0.870 |
Dracula | -0.856 |
warr | -0.793 |
etheless | -0.789 |
Token | instance |
Token ID | 4554 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
enburg | +0.202 |
Directions | +0.188 |
Dracula | +0.184 |
warr | +0.165 |
etheless | +0.162 |
Neg logit contributions | |
isable | -0.265 |
ォ | -0.258 |
ident | -0.256 |
ERG | -0.256 |
OLD | -0.231 |
Token | B |
Token ID | 347 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
isable | +0.985 |
ォ | +0.946 |
ident | +0.941 |
ERG | +0.936 |
OLD | +0.813 |
Neg logit contributions | |
enburg | -1.406 |
Directions | -1.331 |
Dracula | -1.312 |
warr | -1.219 |
etheless | -1.206 |
Token | isc |
Token ID | 2304 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
isable | +1.054 |
ォ | +1.018 |
ident | +1.010 |
ERG | +1.008 |
OLD | +0.883 |
Neg logit contributions | |
enburg | -1.353 |
Directions | -1.280 |
Dracula | -1.261 |
warr | -1.165 |
etheless | -1.150 |
Token | uit |
Token ID | 5013 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
isable | +1.953 |
ident | +1.883 |
ォ | +1.831 |
ERG | +1.827 |
abba | +1.546 |
Neg logit contributions | |
enburg | -4.329 |
Directions | -4.177 |
Dracula | -4.087 |
warr | -3.930 |
etheless | -3.870 |
Token | , |
Token ID | 11 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
enburg | +0.717 |
Directions | +0.678 |
Dracula | +0.669 |
warr | +0.617 |
nces | +0.607 |
Neg logit contributions | |
isable | -0.555 |
ォ | -0.539 |
ident | -0.531 |
ERG | -0.530 |
OLD | -0.464 |
Token | you |
Token ID | 345 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
isable | +0.635 |
ォ | +0.608 |
ident | +0.608 |
ERG | +0.605 |
camp | +0.527 |
Neg logit contributions | |
enburg | -0.858 |
Directions | -0.811 |
Dracula | -0.797 |
warr | -0.740 |
etheless | -0.735 |
Token | will |
Token ID | 481 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
isable | +0.338 |
ident | +0.323 |
ォ | +0.323 |
ERG | +0.321 |
camp | +0.277 |
Neg logit contributions | |
enburg | -0.503 |
Directions | -0.477 |
Dracula | -0.470 |
warr | -0.437 |
etheless | -0.433 |
Token | see |
Token ID | 766 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
enburg | +0.425 |
Directions | +0.403 |
Dracula | +0.396 |
warr | +0.367 |
etheless | +0.363 |
Neg logit contributions | |
isable | -0.313 |
ォ | -0.300 |
ident | -0.300 |
ERG | -0.298 |
camp | -0.259 |
Token | and |
Token ID | 290 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
isable | +1.129 |
ident | +1.087 |
ォ | +1.079 |
ERG | +1.079 |
camp | +0.944 |
Neg logit contributions | |
enburg | -1.444 |
Directions | -1.359 |
Dracula | -1.342 |
warr | -1.241 |
etheless | -1.228 |
Token | the |
Token ID | 262 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
isable | +0.409 |
ォ | +0.394 |
ident | +0.393 |
ERG | +0.391 |
OLD | +0.344 |
Neg logit contributions | |
enburg | -0.486 |
Directions | -0.457 |
Dracula | -0.451 |
warr | -0.414 |
etheless | -0.410 |
Token | CRC |
Token ID | 45623 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
isable | +0.709 |
ォ | +0.681 |
ident | +0.681 |
ERG | +0.676 |
camp | +0.602 |
Neg logit contributions | |
enburg | -0.784 |
Directions | -0.733 |
Dracula | -0.724 |
warr | -0.664 |
etheless | -0.658 |
Token | - |
Token ID | 12 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
isable | +0.639 |
ォ | +0.617 |
ident | +0.616 |
ERG | +0.613 |
OLD | +0.547 |
Neg logit contributions | |
enburg | -0.652 |
Directions | -0.611 |
Dracula | -0.602 |
warr | -0.551 |
etheless | -0.546 |
Token | 32 |
Token ID | 2624 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
isable | +0.923 |
ident | +0.902 |
ERG | +0.895 |
ォ | +0.890 |
OLD | +0.804 |
Neg logit contributions | |
enburg | -0.812 |
Directions | -0.760 |
Dracula | -0.749 |
warr | -0.687 |
etheless | -0.675 |
Token | checks |
Token ID | 8794 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
isable | +0.364 |
ident | +0.350 |
ォ | +0.349 |
ERG | +0.348 |
OLD | +0.306 |
Neg logit contributions | |
enburg | -0.436 |
Directions | -0.411 |
Dracula | -0.406 |
warr | -0.374 |
etheless | -0.370 |
Token | um |
Token ID | 388 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
isable | +3.499 |
ident | +3.424 |
ERG | +3.365 |
ォ | +3.317 |
camp | +3.040 |
Neg logit contributions | |
enburg | -3.050 |
Directions | -2.829 |
Dracula | -2.773 |
warr | -2.538 |
etheless | -2.497 |
Token | for |
Token ID | 329 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
isable | +1.583 |
ident | +1.526 |
ォ | +1.515 |
ERG | +1.508 |
camp | +1.332 |
Neg logit contributions | |
enburg | -1.829 |
Directions | -1.719 |
Dracula | -1.694 |
warr | -1.561 |
etheless | -1.546 |
Token | integrity |
Token ID | 11540 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
isable | +0.084 |
ォ | +0.081 |
ident | +0.081 |
ERG | +0.081 |
OLD | +0.072 |
Neg logit contributions | |
enburg | -0.087 |
Directions | -0.082 |
Dracula | -0.080 |
warr | -0.074 |
etheless | -0.073 |
Token | .[ |
Token ID | 3693 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
isable | +3.051 |
ident | +2.948 |
ォ | +2.924 |
ERG | +2.901 |
camp | +2.590 |
Neg logit contributions | |
enburg | -3.256 |
Directions | -3.056 |
Dracula | -3.005 |
warr | -2.766 |
etheless | -2.749 |
Token | 10 |
Token ID | 940 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
enburg | +0.215 |
Directions | +0.202 |
Dracula | +0.198 |
warr | +0.180 |
etheless | +0.177 |
Neg logit contributions | |
isable | -0.241 |
ォ | -0.234 |
ident | -0.232 |
ERG | -0.232 |
OLD | -0.208 |
Token | the |
Token ID | 262 |
Feature activation | +8.00e-02 |
Pos logit contributions | |
enburg | +0.133 |
Directions | +0.126 |
Dracula | +0.124 |
warr | +0.114 |
etheless | +0.113 |
Neg logit contributions | |
isable | -0.112 |
ォ | -0.108 |
ident | -0.108 |
ERG | -0.108 |
OLD | -0.095 |
Token | US |
Token ID | 1294 |
Feature activation | +6.70e-01 |
Pos logit contributions | |
isable | +0.153 |
ォ | +0.148 |
ident | +0.147 |
ERG | +0.146 |
OLD | +0.129 |
Neg logit contributions | |
enburg | -0.172 |
Directions | -0.162 |
Dracula | -0.159 |
warr | -0.146 |
etheless | -0.144 |
Token | world |
Token ID | 995 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
isable | +1.388 |
ォ | +1.339 |
ident | +1.338 |
ERG | +1.333 |
OLD | +1.195 |
Neg logit contributions | |
enburg | -1.331 |
Directions | -1.247 |
Dracula | -1.229 |
warr | -1.119 |
etheless | -1.103 |
Token | champions |
Token ID | 7827 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
isable | +1.877 |
ォ | +1.806 |
ident | +1.802 |
ERG | +1.783 |
camp | +1.597 |
Neg logit contributions | |
enburg | -2.117 |
Directions | -2.003 |
Dracula | -1.959 |
warr | -1.809 |
nces | -1.796 |
Token | T |
Token ID | 309 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
isable | +0.246 |
ォ | +0.238 |
ident | +0.236 |
ERG | +0.236 |
OLD | +0.208 |
Neg logit contributions | |
enburg | -0.282 |
Directions | -0.266 |
Dracula | -0.261 |
warr | -0.241 |
etheless | -0.238 |
Token | erence |
Token ID | 1945 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
isable | +0.766 |
ォ | +0.735 |
ERG | +0.724 |
ident | +0.724 |
OLD | +0.599 |
Neg logit contributions | |
enburg | -1.537 |
Directions | -1.468 |
Dracula | -1.451 |
warr | -1.353 |
nces | -1.341 |
Token | Crawford |
Token ID | 24499 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
isable | +3.004 |
ident | +2.825 |
ERG | +2.812 |
ォ | +2.726 |
camp | +2.558 |
Neg logit contributions | |
enburg | -4.683 |
Directions | -4.371 |
Dracula | -4.293 |
etheless | -4.065 |
warr | -4.022 |
Token | and |
Token ID | 290 |
Feature activation | -9.06e-03 |
Pos logit contributions | |
enburg | +0.133 |
Directions | +0.125 |
Dracula | +0.123 |
warr | +0.113 |
etheless | +0.112 |
Neg logit contributions | |
isable | -0.116 |
ォ | -0.112 |
ident | -0.112 |
ERG | -0.111 |
OLD | -0.098 |
Token | G |
Token ID | 402 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
enburg | +0.019 |
Directions | +0.018 |
Dracula | +0.018 |
warr | +0.016 |
etheless | +0.016 |
Neg logit contributions | |
isable | -0.018 |
ォ | -0.017 |
ident | -0.017 |
ERG | -0.017 |
OLD | -0.015 |
Token | erv |
Token ID | 712 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
isable | +0.280 |
ォ | +0.271 |
ERG | +0.264 |
ident | +0.261 |
OLD | +0.214 |
Neg logit contributions | |
enburg | -0.664 |
Directions | -0.634 |
Dracula | -0.627 |
warr | -0.587 |
nces | -0.579 |
Token | ont |
Token ID | 756 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
isable | +0.857 |
ォ | +0.835 |
ERG | +0.826 |
ident | +0.813 |
OLD | +0.720 |
Neg logit contributions | |
enburg | -1.041 |
Directions | -0.977 |
Dracula | -0.966 |
warr | -0.873 |
nces | -0.867 |
Token | � |
Token ID | 447 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
isable | +0.944 |
ォ | +0.915 |
ident | +0.909 |
ERG | +0.907 |
OLD | +0.806 |
Neg logit contributions | |
enburg | -0.978 |
Directions | -0.920 |
Dracula | -0.905 |
warr | -0.828 |
etheless | -0.816 |
Token | � |
Token ID | 251 |
Feature activation | -1.86e-03 |
Pos logit contributions | |
isable | +0.593 |
ォ | +0.583 |
ERG | +0.579 |
ident | +0.578 |
OLD | +0.533 |
Neg logit contributions | |
enburg | -0.259 |
Directions | -0.234 |
Dracula | -0.226 |
warr | -0.194 |
etheless | -0.188 |
Token | and |
Token ID | 290 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
enburg | +0.004 |
Directions | +0.004 |
Dracula | +0.004 |
warr | +0.003 |
etheless | +0.003 |
Neg logit contributions | |
isable | -0.004 |
ォ | -0.004 |
ident | -0.004 |
ERG | -0.004 |
OLD | -0.003 |
Token | � |
Token ID | 564 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
isable | +0.045 |
ォ | +0.043 |
ident | +0.043 |
ERG | +0.043 |
OLD | +0.038 |
Neg logit contributions | |
enburg | -0.042 |
Directions | -0.039 |
Dracula | -0.039 |
warr | -0.035 |
etheless | -0.035 |
Token | � |
Token ID | 250 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
isable | +0.399 |
ォ | +0.392 |
ERG | +0.389 |
ident | +0.387 |
OLD | +0.346 |
Neg logit contributions | |
enburg | -0.441 |
Directions | -0.420 |
Dracula | -0.406 |
warr | -0.383 |
etheless | -0.379 |
Token | im |
Token ID | 320 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
isable | +0.426 |
ォ | +0.411 |
ident | +0.411 |
ERG | +0.408 |
OLD | +0.360 |
Neg logit contributions | |
enburg | -0.485 |
Directions | -0.459 |
Dracula | -0.450 |
warr | -0.416 |
etheless | -0.409 |
Token | moral |
Token ID | 41996 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
isable | +2.847 |
ident | +2.756 |
ォ | +2.732 |
ERG | +2.721 |
abba | +2.392 |
Neg logit contributions | |
enburg | -3.486 |
Directions | -3.338 |
Dracula | -3.255 |
warr | -3.052 |
etheless | -2.980 |
Token | � |
Token ID | 447 |
Feature activation | +1.76e-03 |
Pos logit contributions | |
isable | +1.088 |
ォ | +1.056 |
ident | +1.051 |
ERG | +1.046 |
OLD | +0.932 |
Neg logit contributions | |
enburg | -1.082 |
Directions | -1.018 |
Dracula | -1.000 |
warr | -0.918 |
etheless | -0.902 |
Token | � |
Token ID | 251 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
isable | +0.005 |
ォ | +0.005 |
ERG | +0.005 |
ident | +0.005 |
OLD | +0.004 |
Neg logit contributions | |
enburg | -0.002 |
Directions | -0.002 |
Dracula | -0.002 |
warr | -0.002 |
etheless | -0.002 |
Token | imagination |
Token ID | 13843 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
enburg | +0.217 |
Directions | +0.203 |
Dracula | +0.200 |
warr | +0.182 |
etheless | +0.179 |
Neg logit contributions | |
isable | -0.233 |
ォ | -0.227 |
ident | -0.225 |
ERG | -0.225 |
OLD | -0.201 |
Token | of |
Token ID | 286 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
enburg | +0.530 |
Directions | +0.501 |
Dracula | +0.493 |
warr | +0.456 |
etheless | +0.449 |
Neg logit contributions | |
isable | -0.432 |
ォ | -0.417 |
ident | -0.415 |
ERG | -0.414 |
OLD | -0.363 |
Token | in |
Token ID | 287 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
enburg | +0.505 |
Directions | +0.471 |
Dracula | +0.466 |
warr | +0.425 |
etheless | +0.421 |
Neg logit contributions | |
isable | -0.523 |
ォ | -0.507 |
ident | -0.507 |
ERG | -0.505 |
OLD | -0.451 |
Token | Canada |
Token ID | 3340 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
enburg | +0.199 |
Directions | +0.186 |
Dracula | +0.183 |
warr | +0.165 |
etheless | +0.162 |
Neg logit contributions | |
isable | -0.243 |
ォ | -0.236 |
ident | -0.235 |
ERG | -0.235 |
OLD | -0.211 |
Token | found |
Token ID | 1043 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
isable | +0.853 |
ident | +0.837 |
ォ | +0.833 |
ERG | +0.821 |
camp | +0.751 |
Neg logit contributions | |
enburg | -0.765 |
Directions | -0.714 |
Dracula | -0.703 |
warr | -0.645 |
etheless | -0.639 |
Token | 73 |
Token ID | 8854 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
isable | +0.144 |
ident | +0.139 |
ォ | +0.138 |
ERG | +0.138 |
camp | +0.121 |
Neg logit contributions | |
enburg | -0.187 |
Directions | -0.177 |
Dracula | -0.174 |
warr | -0.162 |
etheless | -0.159 |
Token | per |
Token ID | 583 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
enburg | +0.444 |
Directions | +0.417 |
Dracula | +0.410 |
warr | +0.374 |
etheless | +0.369 |
Neg logit contributions | |
isable | -0.440 |
ォ | -0.428 |
ident | -0.423 |
ERG | -0.423 |
OLD | -0.377 |
Token | cent |
Token ID | 1247 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
isable | +0.696 |
ident | +0.631 |
ERG | +0.548 |
ォ | +0.457 |
camp | +0.350 |
Neg logit contributions | |
enburg | -5.077 |
Directions | -4.946 |
Dracula | -4.931 |
warr | -4.835 |
etheless | -4.640 |
Token | of |
Token ID | 286 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
isable | +2.246 |
ident | +2.209 |
ERG | +2.175 |
ォ | +2.156 |
camp | +1.907 |
Neg logit contributions | |
enburg | -3.155 |
Directions | -3.011 |
Dracula | -2.957 |
nces | -2.754 |
warr | -2.749 |
Token | Saskatchewan |
Token ID | 31346 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
enburg | +0.749 |
Directions | +0.704 |
Dracula | +0.693 |
warr | +0.634 |
etheless | +0.625 |
Neg logit contributions | |
isable | -0.718 |
ォ | -0.696 |
ident | -0.692 |
ERG | -0.690 |
OLD | -0.613 |
Token | respondents |
Token ID | 14502 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
isable | +0.576 |
ォ | +0.557 |
ident | +0.555 |
ERG | +0.553 |
OLD | +0.490 |
Neg logit contributions | |
enburg | -0.617 |
Directions | -0.581 |
Dracula | -0.572 |
warr | -0.525 |
etheless | -0.519 |
Token | weren |
Token ID | 6304 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
enburg | +0.481 |
Directions | +0.454 |
Dracula | +0.446 |
warr | +0.411 |
etheless | +0.406 |
Neg logit contributions | |
isable | -0.415 |
ォ | -0.401 |
ident | -0.401 |
ERG | -0.398 |
OLD | -0.352 |
Token | � |
Token ID | 447 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
enburg | +2.913 |
Directions | +2.709 |
Dracula | +2.681 |
nces | +2.379 |
warr | +2.320 |
Neg logit contributions | |
isable | -2.806 |
ォ | -2.713 |
ERG | -2.671 |
ident | -2.655 |
ufact | -2.385 |
Token | also |
Token ID | 635 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
enburg | +0.539 |
Directions | +0.508 |
Dracula | +0.500 |
warr | +0.459 |
etheless | +0.452 |
Neg logit contributions | |
isable | -0.497 |
ォ | -0.482 |
ident | -0.478 |
ERG | -0.477 |
OLD | -0.423 |
Token | need |
Token ID | 761 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
enburg | +0.180 |
Directions | +0.168 |
Dracula | +0.165 |
warr | +0.149 |
etheless | +0.147 |
Neg logit contributions | |
isable | -0.215 |
ォ | -0.209 |
ident | -0.208 |
ERG | -0.207 |
OLD | -0.187 |
Token | to |
Token ID | 284 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
enburg | +0.641 |
Directions | +0.605 |
Dracula | +0.595 |
warr | +0.546 |
nces | +0.537 |
Neg logit contributions | |
isable | -0.587 |
ォ | -0.569 |
ERG | -0.565 |
ident | -0.564 |
OLD | -0.500 |
Token | detach |
Token ID | 48224 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
enburg | +0.066 |
Directions | +0.063 |
Dracula | +0.062 |
warr | +0.058 |
etheless | +0.057 |
Neg logit contributions | |
isable | -0.046 |
ォ | -0.044 |
ident | -0.044 |
ERG | -0.044 |
OLD | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
enburg | +0.697 |
Directions | +0.660 |
Dracula | +0.651 |
warr | +0.600 |
nces | +0.590 |
Neg logit contributions | |
isable | -0.551 |
ォ | -0.533 |
ERG | -0.527 |
ident | -0.527 |
OLD | -0.462 |
Token | cables |
Token ID | 18018 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
enburg | +0.153 |
Directions | +0.144 |
Dracula | +0.142 |
warr | +0.130 |
etheless | +0.128 |
Neg logit contributions | |
isable | -0.139 |
ォ | -0.135 |
ident | -0.134 |
ERG | -0.133 |
OLD | -0.118 |
Token | leading |
Token ID | 3756 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
isable | +0.582 |
ident | +0.562 |
ERG | +0.558 |
ォ | +0.558 |
camp | +0.494 |
Neg logit contributions | |
enburg | -0.676 |
Directions | -0.633 |
Dracula | -0.625 |
warr | -0.578 |
etheless | -0.572 |
Token | to |
Token ID | 284 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
enburg | +0.905 |
Directions | +0.857 |
Dracula | +0.840 |
warr | +0.768 |
nces | +0.755 |
Neg logit contributions | |
isable | -0.859 |
ォ | -0.834 |
ident | -0.824 |
ERG | -0.824 |
OLD | -0.734 |
Token | the |
Token ID | 262 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
isable | +0.348 |
ォ | +0.336 |
ident | +0.333 |
ERG | +0.332 |
OLD | +0.291 |
Neg logit contributions | |
enburg | -0.452 |
Directions | -0.428 |
Dracula | -0.421 |
warr | -0.389 |
etheless | -0.385 |
Token | traction |
Token ID | 23692 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
enburg | +0.113 |
Directions | +0.106 |
Dracula | +0.104 |
warr | +0.096 |
etheless | +0.094 |
Neg logit contributions | |
isable | -0.106 |
ォ | -0.102 |
ident | -0.102 |
ERG | -0.101 |
OLD | -0.090 |
Token | control |
Token ID | 1630 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
enburg | +1.179 |
Directions | +1.110 |
Dracula | +1.095 |
warr | +0.998 |
etheless | +0.977 |
Neg logit contributions | |
isable | -1.123 |
ォ | -1.093 |
ERG | -1.079 |
ident | -1.078 |
abba | -0.959 |
Token | be |
Token ID | 307 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
enburg | +0.519 |
Directions | +0.491 |
Dracula | +0.484 |
warr | +0.446 |
etheless | +0.439 |
Neg logit contributions | |
isable | -0.417 |
ォ | -0.403 |
ERG | -0.399 |
ident | -0.399 |
OLD | -0.350 |
Token | honest |
Token ID | 5508 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
isable | +0.088 |
ォ | +0.086 |
ident | +0.085 |
ERG | +0.085 |
OLD | +0.076 |
Neg logit contributions | |
enburg | -0.090 |
Directions | -0.085 |
Dracula | -0.083 |
warr | -0.076 |
etheless | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
isable | +0.978 |
ォ | +0.939 |
ident | +0.936 |
ERG | +0.932 |
OLD | +0.810 |
Neg logit contributions | |
enburg | -1.389 |
Directions | -1.314 |
Dracula | -1.295 |
warr | -1.206 |
etheless | -1.193 |
Token | a |
Token ID | 257 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
enburg | +0.013 |
Directions | +0.012 |
Dracula | +0.012 |
warr | +0.011 |
etheless | +0.011 |
Neg logit contributions | |
isable | -0.011 |
ォ | -0.010 |
ident | -0.010 |
ERG | -0.010 |
OLD | -0.009 |
Token | secondary |
Token ID | 9233 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
enburg | +0.066 |
Directions | +0.062 |
Dracula | +0.061 |
warr | +0.056 |
etheless | +0.056 |
Neg logit contributions | |
isable | -0.055 |
ォ | -0.053 |
ident | -0.053 |
ERG | -0.052 |
OLD | -0.046 |
Token | motivation |
Token ID | 14052 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
isable | +0.537 |
ォ | +0.515 |
ident | +0.514 |
ERG | +0.511 |
camp | +0.447 |
Neg logit contributions | |
enburg | -0.698 |
Directions | -0.660 |
Dracula | -0.650 |
warr | -0.600 |
etheless | -0.594 |
Token | is |
Token ID | 318 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
isable | +0.432 |
ォ | +0.418 |
ident | +0.414 |
ERG | +0.413 |
OLD | +0.361 |
Neg logit contributions | |
enburg | -0.560 |
Directions | -0.530 |
Dracula | -0.523 |
warr | -0.483 |
etheless | -0.476 |
Token | business |
Token ID | 1597 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
enburg | +0.377 |
Directions | +0.356 |
Dracula | +0.350 |
warr | +0.322 |
etheless | +0.317 |
Neg logit contributions | |
isable | -0.326 |
ォ | -0.316 |
ident | -0.313 |
ERG | -0.313 |
OLD | -0.276 |
Token | related |
Token ID | 3519 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
enburg | +0.384 |
Directions | +0.363 |
Dracula | +0.357 |
warr | +0.329 |
etheless | +0.324 |
Neg logit contributions | |
isable | -0.318 |
ォ | -0.309 |
ident | -0.305 |
ERG | -0.304 |
OLD | -0.268 |
Token | . |
Token ID | 13 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
enburg | +0.544 |
Directions | +0.515 |
Dracula | +0.507 |
warr | +0.468 |
etheless | +0.461 |
Neg logit contributions | |
isable | -0.431 |
ォ | -0.418 |
ident | -0.413 |
ERG | -0.413 |
OLD | -0.362 |
Token | PJ |
Token ID | 44941 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
isable | +0.517 |
ォ | +0.499 |
ident | +0.498 |
ERG | +0.496 |
OLD | +0.436 |
Neg logit contributions | |
enburg | -0.660 |
Directions | -0.619 |
Dracula | -0.612 |
warr | -0.573 |
etheless | -0.566 |
Token | announced |
Token ID | 3414 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
isable | +0.264 |
ォ | +0.256 |
ident | +0.255 |
ERG | +0.254 |
OLD | +0.225 |
Neg logit contributions | |
enburg | -0.293 |
Directions | -0.277 |
Dracula | -0.272 |
warr | -0.250 |
etheless | -0.247 |
Token | a |
Token ID | 257 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
isable | +0.882 |
ォ | +0.859 |
ident | +0.854 |
ERG | +0.852 |
camp | +0.754 |
Neg logit contributions | |
enburg | -1.067 |
Directions | -1.005 |
Dracula | -0.987 |
warr | -0.915 |
etheless | -0.910 |
Token | search |
Token ID | 2989 |
Feature activation | -5.24e-03 |
Pos logit contributions | |
isable | +0.479 |
ォ | +0.465 |
ident | +0.461 |
ERG | +0.460 |
OLD | +0.407 |
Neg logit contributions | |
enburg | -0.532 |
Directions | -0.501 |
Dracula | -0.493 |
warr | -0.453 |
etheless | -0.447 |
Token | for |
Token ID | 329 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
enburg | +0.012 |
Directions | +0.011 |
Dracula | +0.011 |
warr | +0.010 |
etheless | +0.010 |
Neg logit contributions | |
isable | -0.009 |
ォ | -0.009 |
ident | -0.009 |
ERG | -0.009 |
OLD | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
enburg | +0.456 |
Directions | +0.431 |
Dracula | +0.425 |
warr | +0.392 |
etheless | +0.387 |
Neg logit contributions | |
isable | -0.379 |
ォ | -0.367 |
ident | -0.365 |
ERG | -0.364 |
OLD | -0.320 |
Token | management |
Token ID | 4542 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
isable | +0.295 |
ォ | +0.286 |
ident | +0.285 |
ERG | +0.283 |
camp | +0.252 |
Neg logit contributions | |
enburg | -0.314 |
Directions | -0.296 |
Dracula | -0.291 |
warr | -0.267 |
nces | -0.264 |
Token | team |
Token ID | 1074 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
isable | +0.752 |
ォ | +0.725 |
ident | +0.723 |
ERG | +0.720 |
camp | +0.639 |
Neg logit contributions | |
enburg | -0.846 |
Directions | -0.801 |
Dracula | -0.789 |
warr | -0.723 |
etheless | -0.718 |
Token | after |
Token ID | 706 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
enburg | +1.119 |
Directions | +1.054 |
Dracula | +1.036 |
warr | +0.954 |
etheless | +0.940 |
Neg logit contributions | |
isable | -0.954 |
ォ | -0.922 |
ERG | -0.911 |
ident | -0.910 |
OLD | -0.804 |
Token | insisting |
Token ID | 23630 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
enburg | +0.500 |
Directions | +0.472 |
Dracula | +0.463 |
warr | +0.426 |
etheless | +0.420 |
Neg logit contributions | |
isable | -0.466 |
ォ | -0.453 |
ident | -0.450 |
ERG | -0.449 |
OLD | -0.398 |
Token | he |
Token ID | 339 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
enburg | +0.625 |
Directions | +0.588 |
Dracula | +0.578 |
warr | +0.531 |
etheless | +0.524 |
Neg logit contributions | |
isable | -0.581 |
ォ | -0.564 |
ident | -0.560 |
ERG | -0.558 |
OLD | -0.496 |
Token | would |
Token ID | 561 |
Feature activation | +9.55e-02 |
Pos logit contributions | |
isable | +0.151 |
ォ | +0.146 |
ident | +0.146 |
ERG | +0.144 |
camp | +0.128 |
Neg logit contributions | |
enburg | -0.180 |
Directions | -0.170 |
Dracula | -0.168 |
warr | -0.155 |
etheless | -0.154 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
isable | +0.097 |
ident | +0.094 |
ォ | +0.094 |
ERG | +0.094 |
camp | +0.085 |
Neg logit contributions | |
enburg | -0.080 |
Directions | -0.075 |
Dracula | -0.073 |
warr | -0.067 |
etheless | -0.066 |
Token | " |
Token ID | 366 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
enburg | +0.231 |
Directions | +0.218 |
Dracula | +0.215 |
warr | +0.198 |
etheless | +0.196 |
Neg logit contributions | |
isable | -0.185 |
ォ | -0.178 |
ident | -0.178 |
ERG | -0.177 |
OLD | -0.156 |
Token | to |
Token ID | 1462 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
isable | +1.039 |
ident | +1.014 |
ォ | +0.994 |
ERG | +0.988 |
OLD | +0.872 |
Neg logit contributions | |
enburg | -1.397 |
Directions | -1.325 |
Dracula | -1.306 |
warr | -1.220 |
etheless | -1.198 |
Token | sign |
Token ID | 1051 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
enburg | +0.815 |
Directions | +0.771 |
Dracula | +0.760 |
warr | +0.702 |
etheless | +0.692 |
Neg logit contributions | |
isable | -0.629 |
ォ | -0.608 |
ident | -0.601 |
ERG | -0.601 |
OLD | -0.526 |
Token | a |
Token ID | 257 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
isable | +0.208 |
ォ | +0.200 |
ident | +0.199 |
ERG | +0.198 |
OLD | +0.173 |
Neg logit contributions | |
enburg | -0.280 |
Directions | -0.265 |
Dracula | -0.261 |
warr | -0.242 |
etheless | -0.240 |
Token | nond |
Token ID | 30745 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
enburg | +0.281 |
Directions | +0.263 |
Dracula | +0.259 |
warr | +0.235 |
etheless | +0.231 |
Neg logit contributions | |
isable | -0.315 |
ォ | -0.306 |
ident | -0.304 |
ERG | -0.303 |
OLD | -0.272 |
Token | is |
Token ID | 271 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
isable | +0.945 |
ォ | +0.911 |
ident | +0.910 |
ERG | +0.905 |
OLD | +0.806 |
Neg logit contributions | |
enburg | -0.958 |
Directions | -0.905 |
Dracula | -0.889 |
warr | -0.813 |
etheless | -0.803 |
Token | closure |
Token ID | 17966 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
isable | +0.025 |
ォ | +0.025 |
ERG | +0.024 |
ident | +0.024 |
OLD | +0.021 |
Neg logit contributions | |
enburg | -0.036 |
Directions | -0.034 |
Dracula | -0.034 |
warr | -0.031 |
etheless | -0.031 |
Token | agreement |
Token ID | 4381 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
enburg | +0.060 |
Directions | +0.055 |
Dracula | +0.054 |
nces | +0.049 |
warr | +0.049 |
Neg logit contributions | |
isable | -0.078 |
ident | -0.076 |
ERG | -0.075 |
ォ | -0.075 |
camp | -0.069 |
Token | before |
Token ID | 878 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
enburg | +1.049 |
Directions | +0.989 |
Dracula | +0.972 |
warr | +0.895 |
etheless | +0.886 |
Neg logit contributions | |
isable | -0.916 |
ident | -0.882 |
ォ | -0.882 |
ERG | -0.878 |
OLD | -0.776 |
Token | we |
Token ID | 356 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
enburg | +0.603 |
Directions | +0.566 |
Dracula | +0.558 |
warr | +0.513 |
etheless | +0.508 |
Neg logit contributions | |
isable | -0.568 |
ident | -0.548 |
ォ | -0.544 |
ERG | -0.544 |
camp | -0.485 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
enburg | +0.151 |
Directions | +0.143 |
Dracula | +0.141 |
warr | +0.130 |
etheless | +0.127 |
Neg logit contributions | |
isable | -0.128 |
ォ | -0.124 |
ident | -0.122 |
ERG | -0.122 |
OLD | -0.108 |
Token | , |
Token ID | 11 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
isable | +0.217 |
ォ | +0.209 |
ident | +0.208 |
ERG | +0.207 |
OLD | +0.181 |
Neg logit contributions | |
enburg | -0.275 |
Directions | -0.260 |
Dracula | -0.256 |
warr | -0.237 |
etheless | -0.234 |
Token | in |
Token ID | 287 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
enburg | +0.197 |
Directions | +0.186 |
Dracula | +0.184 |
warr | +0.168 |
etheless | +0.165 |
Neg logit contributions | |
isable | -0.177 |
ォ | -0.173 |
ident | -0.171 |
ERG | -0.170 |
OLD | -0.151 |
Token | some |
Token ID | 617 |
Feature activation | -4.88e-03 |
Pos logit contributions | |
enburg | +0.312 |
Directions | +0.294 |
Dracula | +0.290 |
warr | +0.266 |
etheless | +0.262 |
Neg logit contributions | |
isable | -0.277 |
ォ | -0.269 |
ident | -0.267 |
ERG | -0.266 |
OLD | -0.235 |
Token | unthinkable |
Token ID | 39593 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
enburg | +0.009 |
Directions | +0.009 |
Dracula | +0.009 |
warr | +0.008 |
etheless | +0.008 |
Neg logit contributions | |
isable | -0.010 |
ォ | -0.010 |
ident | -0.010 |
ERG | -0.010 |
OLD | -0.009 |
Token | way |
Token ID | 835 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
isable | +0.563 |
ident | +0.540 |
ォ | +0.538 |
ERG | +0.535 |
camp | +0.481 |
Neg logit contributions | |
enburg | -0.539 |
Directions | -0.506 |
Dracula | -0.494 |
warr | -0.455 |
etheless | -0.450 |
Token | . |
Token ID | 13 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
isable | +0.028 |
ォ | +0.027 |
ident | +0.027 |
ERG | +0.027 |
OLD | +0.023 |
Neg logit contributions | |
enburg | -0.035 |
Directions | -0.033 |
Dracula | -0.033 |
warr | -0.030 |
etheless | -0.030 |
Token | So |
Token ID | 1406 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
isable | +0.065 |
ォ | +0.062 |
ident | +0.062 |
ERG | +0.062 |
OLD | +0.054 |
Neg logit contributions | |
enburg | -0.078 |
Directions | -0.073 |
Dracula | -0.072 |
warr | -0.067 |
etheless | -0.066 |
Token | we |
Token ID | 356 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
isable | +0.612 |
ォ | +0.588 |
ident | +0.586 |
ERG | +0.585 |
OLD | +0.512 |
Neg logit contributions | |
enburg | -0.791 |
Directions | -0.748 |
Dracula | -0.736 |
warr | -0.682 |
etheless | -0.675 |
Token | can |
Token ID | 460 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
isable | +0.193 |
ident | +0.185 |
ォ | +0.185 |
ERG | +0.184 |
camp | +0.161 |
Neg logit contributions | |
enburg | -0.258 |
Directions | -0.245 |
Dracula | -0.241 |
warr | -0.224 |
etheless | -0.222 |
Token | make |
Token ID | 787 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
enburg | +0.629 |
Directions | +0.595 |
Dracula | +0.587 |
warr | +0.542 |
etheless | +0.535 |
Neg logit contributions | |
isable | -0.482 |
ォ | -0.465 |
ident | -0.461 |
ERG | -0.461 |
OLD | -0.402 |
Token | part |
Token ID | 636 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
isable | +0.347 |
ォ | +0.337 |
ident | +0.333 |
ERG | +0.333 |
OLD | +0.291 |
Neg logit contributions | |
enburg | -0.445 |
Directions | -0.421 |
Dracula | -0.416 |
warr | -0.383 |
etheless | -0.378 |
Token | of |
Token ID | 286 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
enburg | +0.274 |
Directions | +0.258 |
Dracula | +0.255 |
warr | +0.234 |
nces | +0.230 |
Neg logit contributions | |
isable | -0.235 |
ォ | -0.228 |
ERG | -0.225 |
ident | -0.225 |
OLD | -0.198 |
Token | their |
Token ID | 511 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
enburg | +0.143 |
Directions | +0.135 |
Dracula | +0.133 |
warr | +0.122 |
etheless | +0.120 |
Neg logit contributions | |
isable | -0.119 |
ォ | -0.115 |
ERG | -0.114 |
ident | -0.114 |
OLD | -0.100 |
Token | hockey |
Token ID | 12217 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
enburg | +0.167 |
Directions | +0.158 |
Dracula | +0.155 |
warr | +0.142 |
etheless | +0.140 |
Neg logit contributions | |
isable | -0.152 |
ォ | -0.147 |
ident | -0.146 |
ERG | -0.146 |
OLD | -0.129 |
Token | team |
Token ID | 1074 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
isable | +0.429 |
ォ | +0.415 |
ident | +0.412 |
ERG | +0.411 |
OLD | +0.361 |
Neg logit contributions | |
enburg | -0.525 |
Directions | -0.496 |
Dracula | -0.489 |
warr | -0.451 |
etheless | -0.445 |
Token | . |
Token ID | 13 |
Feature activation | +6.63e-02 |
Pos logit contributions | |
enburg | +0.897 |
Directions | +0.842 |
Dracula | +0.832 |
warr | +0.763 |
etheless | +0.749 |
Neg logit contributions | |
isable | -0.733 |
ォ | -0.716 |
ERG | -0.703 |
ident | -0.699 |
OLD | -0.622 |
Token | It |
Token ID | 632 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
isable | +0.131 |
ォ | +0.127 |
ident | +0.126 |
ERG | +0.126 |
OLD | +0.112 |
Neg logit contributions | |
enburg | -0.138 |
Directions | -0.129 |
Dracula | -0.127 |
warr | -0.117 |
etheless | -0.116 |
Token | 's |
Token ID | 338 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
enburg | +0.023 |
Directions | +0.022 |
Dracula | +0.022 |
warr | +0.020 |
nces | +0.020 |
Neg logit contributions | |
isable | -0.018 |
ォ | -0.018 |
ERG | -0.018 |
ident | -0.018 |
OLD | -0.015 |
Token | been |
Token ID | 587 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
isable | +0.553 |
ォ | +0.535 |
ident | +0.530 |
ERG | +0.529 |
OLD | +0.464 |
Neg logit contributions | |
enburg | -0.695 |
Directions | -0.657 |
Dracula | -0.647 |
warr | -0.597 |
etheless | -0.589 |
Token | a |
Token ID | 257 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
isable | +0.613 |
ォ | +0.591 |
ident | +0.587 |
ERG | +0.586 |
OLD | +0.514 |
Neg logit contributions | |
enburg | -0.773 |
Directions | -0.731 |
Dracula | -0.720 |
warr | -0.665 |
etheless | -0.657 |
Token | short |
Token ID | 1790 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
isable | +0.080 |
ォ | +0.077 |
ident | +0.077 |
ERG | +0.077 |
OLD | +0.068 |
Neg logit contributions | |
enburg | -0.088 |
Directions | -0.083 |
Dracula | -0.082 |
warr | -0.075 |
etheless | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
enburg | +0.040 |
Directions | +0.038 |
Dracula | +0.038 |
warr | +0.035 |
etheless | +0.034 |
Neg logit contributions | |
isable | -0.032 |
ォ | -0.030 |
ident | -0.030 |
ERG | -0.030 |
OLD | -0.026 |
Token | best |
Token ID | 1266 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
enburg | +0.088 |
Directions | +0.083 |
Dracula | +0.081 |
warr | +0.075 |
etheless | +0.073 |
Neg logit contributions | |
isable | -0.079 |
ォ | -0.076 |
ident | -0.076 |
ERG | -0.076 |
OLD | -0.067 |
Token | people |
Token ID | 661 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
enburg | +1.456 |
Directions | +1.382 |
Dracula | +1.363 |
warr | +1.263 |
nces | +1.243 |
Neg logit contributions | |
isable | -0.893 |
ォ | -0.872 |
ERG | -0.853 |
ident | -0.849 |
abba | -0.730 |
Token | to |
Token ID | 284 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
enburg | +1.002 |
Directions | +0.945 |
Dracula | +0.931 |
warr | +0.856 |
etheless | +0.844 |
Neg logit contributions | |
isable | -0.851 |
ォ | -0.824 |
ident | -0.817 |
ERG | -0.815 |
OLD | -0.718 |
Token | cover |
Token ID | 3002 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
enburg | +0.546 |
Directions | +0.515 |
Dracula | +0.507 |
warr | +0.467 |
etheless | +0.462 |
Neg logit contributions | |
isable | -0.476 |
ォ | -0.460 |
ident | -0.458 |
ERG | -0.457 |
OLD | -0.403 |
Token | his |
Token ID | 465 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
enburg | +1.070 |
Directions | +1.002 |
Dracula | +0.989 |
warr | +0.900 |
nces | +0.882 |
Neg logit contributions | |
isable | -1.048 |
ォ | -1.026 |
ident | -1.006 |
ERG | -1.006 |
OLD | -0.899 |
Token | gaps |
Token ID | 17332 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
enburg | +0.198 |
Directions | +0.186 |
Dracula | +0.184 |
warr | +0.169 |
etheless | +0.165 |
Neg logit contributions | |
isable | -0.177 |
ォ | -0.171 |
ERG | -0.169 |
ident | -0.169 |
OLD | -0.150 |
Token | in |
Token ID | 287 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
isable | +0.521 |
ォ | +0.503 |
ident | +0.502 |
ERG | +0.500 |
camp | +0.437 |
Neg logit contributions | |
enburg | -0.677 |
Directions | -0.641 |
Dracula | -0.631 |
warr | -0.587 |
etheless | -0.580 |
Token | knowledge |
Token ID | 3725 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
enburg | +0.518 |
Directions | +0.487 |
Dracula | +0.479 |
warr | +0.439 |
etheless | +0.433 |
Neg logit contributions | |
isable | -0.501 |
ォ | -0.485 |
ident | -0.483 |
ERG | -0.481 |
OLD | -0.428 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
isable | +0.582 |
ォ | +0.561 |
ident | +0.558 |
ERG | +0.556 |
OLD | +0.486 |
Neg logit contributions | |
enburg | -0.762 |
Directions | -0.721 |
Dracula | -0.710 |
warr | -0.658 |
etheless | -0.649 |
Token | experience |
Token ID | 1998 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
enburg | +0.195 |
Directions | +0.182 |
Dracula | +0.179 |
warr | +0.163 |
etheless | +0.162 |
Neg logit contributions | |
isable | -0.215 |
ォ | -0.208 |
ident | -0.208 |
ERG | -0.207 |
camp | -0.186 |
Token | known |
Token ID | 1900 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
enburg | +0.400 |
Directions | +0.375 |
Dracula | +0.369 |
warr | +0.338 |
etheless | +0.334 |
Neg logit contributions | |
isable | -0.402 |
ォ | -0.388 |
ident | -0.387 |
ERG | -0.385 |
OLD | -0.343 |
Token | you |
Token ID | 345 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
enburg | +0.936 |
Directions | +0.884 |
Dracula | +0.870 |
warr | +0.802 |
etheless | +0.793 |
Neg logit contributions | |
isable | -0.778 |
ォ | -0.750 |
ident | -0.748 |
ERG | -0.745 |
OLD | -0.655 |
Token | , |
Token ID | 11 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
enburg | +0.370 |
Directions | +0.349 |
Dracula | +0.344 |
warr | +0.316 |
etheless | +0.312 |
Neg logit contributions | |
isable | -0.320 |
ォ | -0.310 |
ident | -0.308 |
ERG | -0.307 |
OLD | -0.271 |
Token | you |
Token ID | 345 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
enburg | +0.349 |
Directions | +0.329 |
Dracula | +0.325 |
warr | +0.299 |
etheless | +0.296 |
Neg logit contributions | |
isable | -0.288 |
ォ | -0.278 |
ident | -0.277 |
ERG | -0.276 |
OLD | -0.242 |
Token | have |
Token ID | 423 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
enburg | +0.131 |
Directions | +0.124 |
Dracula | +0.122 |
warr | +0.112 |
etheless | +0.111 |
Neg logit contributions | |
isable | -0.110 |
ォ | -0.106 |
ident | -0.105 |
ERG | -0.105 |
OLD | -0.092 |
Token | taught |
Token ID | 7817 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
isable | +0.206 |
ident | +0.199 |
ォ | +0.199 |
ERG | +0.197 |
camp | +0.176 |
Neg logit contributions | |
enburg | -0.207 |
Directions | -0.194 |
Dracula | -0.191 |
warr | -0.175 |
etheless | -0.174 |
Token | you |
Token ID | 345 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
enburg | +0.889 |
Directions | +0.840 |
Dracula | +0.827 |
warr | +0.760 |
etheless | +0.749 |
Neg logit contributions | |
isable | -0.767 |
ォ | -0.744 |
ident | -0.735 |
ERG | -0.734 |
OLD | -0.648 |
Token | or |
Token ID | 393 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
isable | +0.039 |
ォ | +0.038 |
ident | +0.037 |
ERG | +0.037 |
OLD | +0.033 |
Neg logit contributions | |
enburg | -0.049 |
Directions | -0.046 |
Dracula | -0.045 |
warr | -0.042 |
etheless | -0.041 |
Token | who |
Token ID | 508 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
enburg | +0.063 |
Directions | +0.059 |
Dracula | +0.058 |
warr | +0.054 |
etheless | +0.053 |
Neg logit contributions | |
isable | -0.055 |
ォ | -0.053 |
ident | -0.053 |
ERG | -0.053 |
OLD | -0.047 |
Token | you |
Token ID | 345 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
enburg | +0.367 |
Directions | +0.347 |
Dracula | +0.342 |
warr | +0.315 |
etheless | +0.311 |
Neg logit contributions | |
isable | -0.309 |
ォ | -0.300 |
ident | -0.298 |
ERG | -0.296 |
OLD | -0.261 |
Token | have |
Token ID | 423 |
Feature activation | -8.16e-03 |
Pos logit contributions | |
enburg | +0.052 |
Directions | +0.049 |
Dracula | +0.048 |
warr | +0.044 |
etheless | +0.044 |
Neg logit contributions | |
isable | -0.043 |
ォ | -0.041 |
ident | -0.041 |
ERG | -0.041 |
OLD | -0.036 |
Token | hess |
Token ID | 33979 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
enburg | +1.639 |
Directions | +1.541 |
Dracula | +1.517 |
warr | +1.389 |
etheless | +1.371 |
Neg logit contributions | |
isable | -1.507 |
ォ | -1.464 |
ident | -1.449 |
ERG | -1.449 |
OLD | -1.284 |
Token | TV |
Token ID | 3195 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
isable | +0.936 |
ォ | +0.902 |
ident | +0.899 |
ERG | +0.895 |
OLD | +0.783 |
Neg logit contributions | |
enburg | -1.197 |
Directions | -1.132 |
Dracula | -1.117 |
warr | -1.033 |
etheless | -1.019 |
Token | ! |
Token ID | 0 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
enburg | +1.074 |
Directions | +1.018 |
Dracula | +1.001 |
warr | +0.918 |
etheless | +0.907 |
Neg logit contributions | |
isable | -0.869 |
ォ | -0.845 |
ident | -0.834 |
ERG | -0.832 |
OLD | -0.731 |
Token |
|
Token ID | 198 |
Feature activation | -3.07e-03 |
Pos logit contributions | |
enburg | +0.799 |
Directions | +0.760 |
Dracula | +0.747 |
warr | +0.684 |
etheless | +0.675 |
Neg logit contributions | |
isable | -0.637 |
ォ | -0.618 |
ERG | -0.610 |
ident | -0.610 |
OLD | -0.534 |
Token |
|
Token ID | 198 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
enburg | +0.007 |
Directions | +0.007 |
Dracula | +0.007 |
warr | +0.007 |
etheless | +0.006 |
Neg logit contributions | |
isable | -0.005 |
ident | -0.005 |
ERG | -0.005 |
ォ | -0.005 |
OLD | -0.004 |
Token | Free |
Token ID | 11146 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
enburg | +0.405 |
Directions | +0.382 |
Dracula | +0.376 |
warr | +0.345 |
nces | +0.341 |
Neg logit contributions | |
isable | -0.325 |
ォ | -0.315 |
ERG | -0.310 |
ident | -0.309 |
abba | -0.272 |
Token | variations |
Token ID | 13991 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
enburg | +0.941 |
Directions | +0.888 |
Dracula | +0.871 |
warr | +0.796 |
etheless | +0.783 |
Neg logit contributions | |
isable | -0.904 |
ォ | -0.883 |
ident | -0.872 |
ERG | -0.872 |
OLD | -0.775 |
Token | in |
Token ID | 287 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
enburg | +1.392 |
Directions | +1.320 |
Dracula | +1.299 |
warr | +1.194 |
etheless | +1.177 |
Neg logit contributions | |
isable | -1.113 |
ォ | -1.090 |
ident | -1.073 |
ERG | -1.073 |
OLD | -0.940 |
Token | analysis |
Token ID | 3781 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
isable | +0.074 |
ォ | +0.071 |
ident | +0.071 |
ERG | +0.071 |
OLD | +0.063 |
Neg logit contributions | |
enburg | -0.081 |
Directions | -0.076 |
Dracula | -0.075 |
warr | -0.069 |
etheless | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
isable | +0.225 |
ォ | +0.213 |
ident | +0.213 |
ERG | +0.213 |
camp | +0.187 |
Neg logit contributions | |
enburg | -0.294 |
Directions | -0.276 |
Dracula | -0.274 |
warr | -0.254 |
etheless | -0.251 |
Token |
|
Token ID | 198 |
Feature activation | -8.78e-02 |
Pos logit contributions | |
isable | +0.118 |
ident | +0.113 |
ォ | +0.112 |
ERG | +0.112 |
OLD | +0.098 |
Neg logit contributions | |
enburg | -0.169 |
Directions | -0.160 |
Dracula | -0.157 |
warr | -0.150 |
etheless | -0.148 |
Token | got |
Token ID | 1392 |
Feature activation | +5.65e-03 |
Pos logit contributions | |
enburg | +0.158 |
Directions | +0.150 |
Dracula | +0.147 |
warr | +0.137 |
etheless | +0.135 |
Neg logit contributions | |
isable | -0.134 |
ident | -0.130 |
ォ | -0.129 |
ERG | -0.127 |
camp | -0.115 |
Token | it |
Token ID | 340 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
isable | +0.010 |
ォ | +0.010 |
ident | +0.010 |
ERG | +0.010 |
OLD | +0.009 |
Neg logit contributions | |
enburg | -0.013 |
Directions | -0.012 |
Dracula | -0.012 |
warr | -0.011 |
etheless | -0.011 |
Token | up |
Token ID | 510 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
enburg | +0.552 |
Directions | +0.521 |
Dracula | +0.513 |
warr | +0.470 |
etheless | +0.463 |
Neg logit contributions | |
isable | -0.486 |
ォ | -0.472 |
ERG | -0.467 |
ident | -0.467 |
OLD | -0.412 |
Token | on |
Token ID | 319 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
enburg | +1.001 |
Directions | +0.947 |
Dracula | +0.933 |
warr | +0.856 |
etheless | +0.843 |
Neg logit contributions | |
isable | -0.837 |
ォ | -0.813 |
ERG | -0.804 |
ident | -0.803 |
OLD | -0.705 |
Token | some |
Token ID | 617 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
enburg | +0.745 |
Directions | +0.704 |
Dracula | +0.694 |
warr | +0.639 |
etheless | +0.629 |
Neg logit contributions | |
isable | -0.609 |
ォ | -0.592 |
ERG | -0.585 |
ident | -0.584 |
abba | -0.512 |
Token | vinyl |
Token ID | 23275 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
enburg | +0.325 |
Directions | +0.307 |
Dracula | +0.302 |
warr | +0.277 |
etheless | +0.273 |
Neg logit contributions | |
isable | -0.290 |
ォ | -0.281 |
ident | -0.279 |
ERG | -0.278 |
OLD | -0.246 |
Token | floor |
Token ID | 4314 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
enburg | +2.548 |
Directions | +2.403 |
Dracula | +2.384 |
warr | +2.180 |
nces | +2.133 |
Neg logit contributions | |
isable | -2.228 |
ォ | -2.168 |
ERG | -2.155 |
ident | -2.144 |
OLD | -1.885 |
Token | ing |
Token ID | 278 |
Feature activation | -9.59e-01 |
Pos logit contributions | |
enburg | +0.807 |
Directions | +0.759 |
Dracula | +0.747 |
warr | +0.684 |
etheless | +0.673 |
Neg logit contributions | |
isable | -0.773 |
ォ | -0.750 |
ident | -0.744 |
ERG | -0.743 |
OLD | -0.660 |
Token | it |
Token ID | 340 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
enburg | +2.103 |
Directions | +1.990 |
Dracula | +1.965 |
warr | +1.800 |
nces | +1.770 |
Neg logit contributions | |
isable | -1.770 |
ォ | -1.718 |
ERG | -1.703 |
ident | -1.699 |
OLD | -1.490 |
Token | was |
Token ID | 373 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
enburg | +0.135 |
Directions | +0.128 |
Dracula | +0.125 |
warr | +0.115 |
etheless | +0.114 |
Neg logit contributions | |
isable | -0.123 |
ォ | -0.119 |
ident | -0.118 |
ERG | -0.118 |
OLD | -0.105 |
Token | game |
Token ID | 983 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
isable | +0.046 |
ォ | +0.045 |
ident | +0.044 |
ERG | +0.044 |
OLD | +0.039 |
Neg logit contributions | |
enburg | -0.059 |
Directions | -0.056 |
Dracula | -0.055 |
warr | -0.051 |
etheless | -0.050 |
Token | op |
Token ID | 404 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
isable | +0.148 |
ォ | +0.145 |
ERG | +0.144 |
ident | +0.142 |
OLD | +0.128 |
Neg logit contributions | |
enburg | -0.138 |
Directions | -0.129 |
Dracula | -0.126 |
warr | -0.114 |
nces | -0.114 |
Token | ards |
Token ID | 1371 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
isable | +0.398 |
ident | +0.379 |
ォ | +0.374 |
ERG | +0.374 |
OLD | +0.296 |
Neg logit contributions | |
enburg | -1.096 |
Directions | -1.057 |
Dracula | -1.042 |
warr | -1.000 |
etheless | -0.983 |
Token | in |
Token ID | 287 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
enburg | +0.557 |
Directions | +0.525 |
Dracula | +0.518 |
warr | +0.475 |
etheless | +0.468 |
Neg logit contributions | |
isable | -0.489 |
ォ | -0.475 |
ERG | -0.469 |
ident | -0.469 |
OLD | -0.415 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
enburg | +0.225 |
Directions | +0.212 |
Dracula | +0.208 |
warr | +0.192 |
etheless | +0.190 |
Neg logit contributions | |
isable | -0.199 |
ォ | -0.192 |
ident | -0.191 |
ERG | -0.190 |
OLD | -0.168 |
Token | Hindu |
Token ID | 16397 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
enburg | +0.237 |
Directions | +0.224 |
Dracula | +0.220 |
warr | +0.204 |
etheless | +0.202 |
Neg logit contributions | |
isable | -0.189 |
ォ | -0.181 |
ident | -0.181 |
ERG | -0.180 |
camp | -0.159 |
Token | Kush |
Token ID | 18995 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
enburg | +1.126 |
Directions | +1.051 |
Dracula | +1.032 |
warr | +0.934 |
etheless | +0.915 |
Neg logit contributions | |
isable | -1.287 |
ォ | -1.254 |
ident | -1.242 |
ERG | -1.242 |
OLD | -1.118 |
Token | for |
Token ID | 329 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
enburg | +2.482 |
Directions | +2.332 |
Dracula | +2.289 |
warr | +2.078 |
nces | +2.042 |
Neg logit contributions | |
isable | -2.493 |
ォ | -2.433 |
ERG | -2.393 |
ident | -2.391 |
OLD | -2.135 |
Token | the |
Token ID | 262 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
enburg | +0.663 |
Directions | +0.631 |
Dracula | +0.623 |
warr | +0.570 |
etheless | +0.561 |
Neg logit contributions | |
isable | -0.526 |
ォ | -0.515 |
ident | -0.503 |
ERG | -0.502 |
OLD | -0.443 |
Token | BBC |
Token ID | 7823 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
isable | +0.239 |
ォ | +0.231 |
ident | +0.229 |
ERG | +0.228 |
OLD | +0.200 |
Neg logit contributions | |
enburg | -0.310 |
Directions | -0.294 |
Dracula | -0.289 |
warr | -0.267 |
etheless | -0.264 |
Token | Two |
Token ID | 4930 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
isable | +0.651 |
ォ | +0.627 |
ident | +0.627 |
ERG | +0.624 |
camp | +0.550 |
Neg logit contributions | |
enburg | -0.763 |
Directions | -0.716 |
Dracula | -0.704 |
warr | -0.652 |
etheless | -0.643 |
Token | series |
Token ID | 2168 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
isable | +0.355 |
ォ | +0.343 |
ident | +0.342 |
ERG | +0.341 |
OLD | +0.303 |
Neg logit contributions | |
enburg | -0.371 |
Directions | -0.348 |
Dracula | -0.342 |
warr | -0.314 |
etheless | -0.310 |
Token | but |
Token ID | 475 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
isable | +0.085 |
ォ | +0.082 |
ident | +0.082 |
ERG | +0.081 |
OLD | +0.071 |
Neg logit contributions | |
enburg | -0.115 |
Directions | -0.109 |
Dracula | -0.108 |
warr | -0.100 |
etheless | -0.099 |
Token | I |
Token ID | 314 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
enburg | +0.369 |
Directions | +0.347 |
Dracula | +0.343 |
warr | +0.316 |
etheless | +0.313 |
Neg logit contributions | |
isable | -0.313 |
ォ | -0.301 |
ident | -0.301 |
ERG | -0.300 |
OLD | -0.265 |
Token | never |
Token ID | 1239 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
enburg | +0.091 |
Directions | +0.086 |
Dracula | +0.084 |
warr | +0.078 |
etheless | +0.077 |
Neg logit contributions | |
isable | -0.077 |
ident | -0.075 |
ォ | -0.074 |
ERG | -0.074 |
camp | -0.066 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
enburg | +2.711 |
Directions | +2.530 |
Dracula | +2.502 |
warr | +2.255 |
etheless | +2.230 |
Neg logit contributions | |
isable | -2.989 |
ォ | -2.926 |
ERG | -2.899 |
ident | -2.865 |
OLD | -2.589 |
Token | it |
Token ID | 340 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
isable | +0.378 |
ident | +0.365 |
ERG | +0.363 |
ォ | +0.362 |
camp | +0.321 |
Neg logit contributions | |
enburg | -0.451 |
Directions | -0.423 |
Dracula | -0.418 |
warr | -0.386 |
etheless | -0.383 |
Token | listed |
Token ID | 5610 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
enburg | +0.925 |
Directions | +0.871 |
Dracula | +0.856 |
warr | +0.782 |
etheless | +0.772 |
Neg logit contributions | |
isable | -0.873 |
ォ | -0.850 |
ERG | -0.840 |
ident | -0.839 |
OLD | -0.746 |
Token | for |
Token ID | 329 |
Feature activation | +4.84e-02 |
Pos logit contributions | |
enburg | +2.425 |
Directions | +2.292 |
Dracula | +2.250 |
warr | +2.048 |
etheless | +2.031 |
Neg logit contributions | |
isable | -2.147 |
ォ | -2.109 |
ERG | -2.071 |
ident | -2.063 |
OLD | -1.830 |
Token | the |
Token ID | 262 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
isable | +0.094 |
ident | +0.091 |
ォ | +0.091 |
ERG | +0.091 |
camp | +0.081 |
Neg logit contributions | |
enburg | -0.101 |
Directions | -0.095 |
Dracula | -0.094 |
warr | -0.086 |
nces | -0.085 |
Token | coax |
Token ID | 37700 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
isable | +0.517 |
ident | +0.498 |
ERG | +0.497 |
ォ | +0.497 |
camp | +0.444 |
Neg logit contributions | |
enburg | -0.560 |
Directions | -0.522 |
Dracula | -0.518 |
warr | -0.475 |
etheless | -0.472 |
Token | connection |
Token ID | 4637 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
enburg | +1.664 |
Directions | +1.576 |
Dracula | +1.550 |
warr | +1.423 |
etheless | +1.409 |
Neg logit contributions | |
isable | -1.358 |
ォ | -1.323 |
ident | -1.295 |
ERG | -1.295 |
abba | -1.141 |
Token | . |
Token ID | 13 |
Feature activation | +8.67e-02 |
Pos logit contributions | |
isable | +0.635 |
ident | +0.609 |
ォ | +0.605 |
ERG | +0.604 |
camp | +0.533 |
Neg logit contributions | |
enburg | -0.818 |
Directions | -0.770 |
Dracula | -0.761 |
warr | -0.705 |
etheless | -0.697 |
Token | slay |
Token ID | 42596 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
enburg | +0.108 |
Directions | +0.102 |
Dracula | +0.101 |
warr | +0.093 |
etheless | +0.092 |
Neg logit contributions | |
isable | -0.083 |
ォ | -0.080 |
ident | -0.079 |
ERG | -0.079 |
OLD | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
enburg | +0.491 |
Directions | +0.465 |
Dracula | +0.459 |
warr | +0.422 |
etheless | +0.417 |
Neg logit contributions | |
isable | -0.386 |
ォ | -0.376 |
ident | -0.371 |
ERG | -0.370 |
OLD | -0.324 |
Token | even |
Token ID | 772 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
enburg | +0.172 |
Directions | +0.162 |
Dracula | +0.159 |
warr | +0.146 |
etheless | +0.144 |
Neg logit contributions | |
isable | -0.163 |
ォ | -0.157 |
ident | -0.156 |
ERG | -0.156 |
OLD | -0.139 |
Token | if |
Token ID | 611 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
isable | +0.502 |
ident | +0.480 |
ォ | +0.477 |
ERG | +0.476 |
camp | +0.421 |
Neg logit contributions | |
enburg | -0.630 |
Directions | -0.596 |
Dracula | -0.582 |
warr | -0.540 |
etheless | -0.538 |
Token | I |
Token ID | 314 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
isable | +0.109 |
ォ | +0.104 |
ident | +0.104 |
ERG | +0.104 |
OLD | +0.091 |
Neg logit contributions | |
enburg | -0.136 |
Directions | -0.128 |
Dracula | -0.126 |
warr | -0.116 |
etheless | -0.116 |
Token | myself |
Token ID | 3589 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
enburg | +0.480 |
Directions | +0.453 |
Dracula | +0.445 |
warr | +0.410 |
etheless | +0.405 |
Neg logit contributions | |
isable | -0.409 |
ォ | -0.393 |
ident | -0.393 |
ERG | -0.390 |
OLD | -0.345 |
Token | am |
Token ID | 716 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
enburg | +2.770 |
Directions | +2.617 |
Dracula | +2.607 |
warr | +2.379 |
etheless | +2.339 |
Neg logit contributions | |
isable | -2.326 |
ォ | -2.315 |
ERG | -2.266 |
ident | -2.228 |
OLD | -1.976 |
Token | slain |
Token ID | 27029 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
enburg | +0.772 |
Directions | +0.728 |
Dracula | +0.719 |
warr | +0.659 |
etheless | +0.648 |
Neg logit contributions | |
isable | -0.694 |
ォ | -0.677 |
ident | -0.669 |
ERG | -0.669 |
OLD | -0.590 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
isable | +0.988 |
ォ | +0.947 |
ident | +0.945 |
ERG | +0.941 |
OLD | +0.825 |
Neg logit contributions | |
enburg | -1.272 |
Directions | -1.203 |
Dracula | -1.183 |
warr | -1.098 |
etheless | -1.082 |
Token |
|
Token ID | 198 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
enburg | +0.359 |
Directions | +0.340 |
Dracula | +0.335 |
warr | +0.309 |
etheless | +0.305 |
Neg logit contributions | |
isable | -0.277 |
ォ | -0.268 |
ident | -0.265 |
ERG | -0.265 |
OLD | -0.232 |
Token |
|
Token ID | 198 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
enburg | +0.238 |
Directions | +0.226 |
Dracula | +0.222 |
warr | +0.211 |
etheless | +0.209 |
Neg logit contributions | |
isable | -0.167 |
ident | -0.160 |
ォ | -0.160 |
ERG | -0.159 |
OLD | -0.138 |
Token | some |
Token ID | 617 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
enburg | +0.461 |
Directions | +0.438 |
Dracula | +0.431 |
warr | +0.398 |
etheless | +0.394 |
Neg logit contributions | |
isable | -0.359 |
ident | -0.344 |
ォ | -0.344 |
ERG | -0.343 |
camp | -0.300 |
Token | 20 |
Token ID | 1160 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
enburg | +0.547 |
Directions | +0.518 |
Dracula | +0.511 |
warr | +0.469 |
etheless | +0.466 |
Neg logit contributions | |
isable | -0.466 |
ident | -0.447 |
ERG | -0.446 |
ォ | -0.445 |
camp | -0.394 |
Token | security |
Token ID | 2324 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
enburg | +0.488 |
Directions | +0.463 |
Dracula | +0.456 |
warr | +0.421 |
etheless | +0.417 |
Neg logit contributions | |
isable | -0.380 |
ォ | -0.366 |
ident | -0.364 |
ERG | -0.363 |
OLD | -0.319 |
Token | guards |
Token ID | 10942 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
isable | +1.344 |
ident | +1.284 |
ォ | +1.282 |
ERG | +1.280 |
camp | +1.096 |
Neg logit contributions | |
enburg | -2.205 |
Directions | -2.098 |
Dracula | -2.071 |
warr | -1.925 |
etheless | -1.911 |
Token | . |
Token ID | 13 |
Feature activation | -9.64e-02 |
Pos logit contributions | |
isable | +0.148 |
ident | +0.142 |
ERG | +0.142 |
ォ | +0.141 |
camp | +0.125 |
Neg logit contributions | |
enburg | -0.172 |
Directions | -0.162 |
Dracula | -0.159 |
warr | -0.146 |
etheless | -0.145 |
Token | LG |
Token ID | 17370 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
enburg | +0.217 |
Directions | +0.205 |
Dracula | +0.202 |
warr | +0.186 |
etheless | +0.184 |
Neg logit contributions | |
isable | -0.174 |
ォ | -0.168 |
ident | -0.167 |
ERG | -0.167 |
OLD | -0.146 |
Token | should |
Token ID | 815 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
enburg | +2.435 |
Directions | +2.297 |
Dracula | +2.252 |
warr | +2.078 |
etheless | +2.037 |
Neg logit contributions | |
isable | -2.225 |
ォ | -2.162 |
ident | -2.128 |
ERG | -2.120 |
OLD | -1.879 |
Token | have |
Token ID | 423 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
isable | +0.172 |
ォ | +0.165 |
ident | +0.165 |
ERG | +0.164 |
OLD | +0.142 |
Neg logit contributions | |
enburg | -0.236 |
Directions | -0.224 |
Dracula | -0.221 |
warr | -0.204 |
etheless | -0.202 |
Token | placed |
Token ID | 4624 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
isable | +0.243 |
ident | +0.234 |
ォ | +0.234 |
ERG | +0.232 |
camp | +0.207 |
Neg logit contributions | |
enburg | -0.254 |
Directions | -0.240 |
Dracula | -0.235 |
warr | -0.215 |
etheless | -0.213 |
Token | more |
Token ID | 517 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
enburg | +0.587 |
Directions | +0.554 |
Dracula | +0.545 |
warr | +0.502 |
etheless | +0.495 |
Neg logit contributions | |
isable | -0.502 |
ォ | -0.486 |
ident | -0.481 |
ERG | -0.481 |
OLD | -0.424 |
Token | security |
Token ID | 2324 |
Feature activation | +9.68e-01 |
Pos logit contributions | |
enburg | +0.547 |
Directions | +0.518 |
Dracula | +0.511 |
warr | +0.472 |
etheless | +0.466 |
Neg logit contributions | |
isable | -0.407 |
ォ | -0.394 |
ident | -0.390 |
ERG | -0.389 |
OLD | -0.339 |