Token | sway |
Token ID | 20009 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
arcity | +0.058 |
Cage | +0.048 |
Hardware | +0.048 |
Sisters | +0.047 |
τ | +0.046 |
Neg logit contributions | |
anka | -0.044 |
plom | -0.043 |
glitter | -0.043 |
agen | -0.042 |
interpol | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
arcity | +1.164 |
Hardware | +0.955 |
Cage | +0.955 |
Sisters | +0.928 |
τ | +0.912 |
Neg logit contributions | |
anka | -0.942 |
plom | -0.914 |
glitter | -0.899 |
agen | -0.893 |
interpol | -0.857 |
Token | system |
Token ID | 1080 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
arcity | +1.326 |
Cage | +1.104 |
Hardware | +1.102 |
Sisters | +1.073 |
τ | +1.063 |
Neg logit contributions | |
anka | -0.902 |
plom | -0.873 |
glitter | -0.853 |
agen | -0.851 |
interpol | -0.809 |
Token | through |
Token ID | 832 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
arcity | +0.933 |
Cage | +0.771 |
Hardware | +0.770 |
Sisters | +0.749 |
τ | +0.738 |
Neg logit contributions | |
anka | -0.726 |
plom | -0.705 |
agen | -0.689 |
glitter | -0.687 |
interpol | -0.659 |
Token | ult |
Token ID | 3789 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
arcity | +1.074 |
Cage | +0.892 |
Hardware | +0.888 |
Sisters | +0.864 |
τ | +0.852 |
Neg logit contributions | |
anka | -0.822 |
plom | -0.800 |
glitter | -0.782 |
agen | -0.778 |
interpol | -0.749 |
Token | im |
Token ID | 320 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
arcity | +0.417 |
Cage | +0.352 |
Hardware | +0.349 |
Sisters | +0.342 |
τ | +0.333 |
Neg logit contributions | |
anka | -0.333 |
plom | -0.328 |
agen | -0.320 |
glitter | -0.311 |
interpol | -0.297 |
Token | at |
Token ID | 265 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
arcity | +4.910 |
Cage | +3.974 |
Hardware | +3.924 |
Sisters | +3.823 |
τ | +3.798 |
Neg logit contributions | |
anka | -4.453 |
plom | -4.289 |
glitter | -4.255 |
agen | -4.174 |
interpol | -4.096 |
Token | ums |
Token ID | 5700 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
arcity | +1.580 |
Cage | +1.243 |
Hardware | +1.224 |
Sisters | +1.191 |
τ | +1.160 |
Neg logit contributions | |
anka | -2.025 |
plom | -1.977 |
agen | -1.940 |
glitter | -1.930 |
interpol | -1.873 |
Token | , |
Token ID | 11 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
arcity | +2.132 |
Cage | +1.765 |
Hardware | +1.746 |
Sisters | +1.709 |
τ | +1.675 |
Neg logit contributions | |
anka | -1.784 |
plom | -1.732 |
agen | -1.690 |
glitter | -1.682 |
interpol | -1.620 |
Token | are |
Token ID | 389 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
arcity | +1.562 |
Cage | +1.298 |
Hardware | +1.293 |
Sisters | +1.258 |
τ | +1.237 |
Neg logit contributions | |
anka | -1.176 |
plom | -1.143 |
glitter | -1.123 |
agen | -1.109 |
interpol | -1.076 |
Token | based |
Token ID | 1912 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
arcity | +2.266 |
Cage | +1.882 |
Hardware | +1.875 |
Sisters | +1.828 |
τ | +1.797 |
Neg logit contributions | |
anka | -1.688 |
plom | -1.646 |
glitter | -1.612 |
agen | -1.605 |
interpol | -1.550 |
Token | , |
Token ID | 837 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
arcity | +2.869 |
Cage | +2.455 |
Hardware | +2.429 |
Sisters | +2.383 |
τ | +2.343 |
Neg logit contributions | |
anka | -1.828 |
plom | -1.774 |
agen | -1.703 |
glitter | -1.701 |
interpol | -1.620 |
Token | height |
Token ID | 6001 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
arcity | +0.415 |
Cage | +0.336 |
Hardware | +0.333 |
Sisters | +0.323 |
τ | +0.318 |
Neg logit contributions | |
anka | -0.410 |
plom | -0.397 |
glitter | -0.392 |
agen | -0.390 |
interpol | -0.379 |
Token | = |
Token ID | 796 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
arcity | +2.121 |
Cage | +1.675 |
Hardware | +1.655 |
Sisters | +1.595 |
τ | +1.559 |
Neg logit contributions | |
anka | -2.880 |
plom | -2.825 |
glitter | -2.753 |
agen | -2.751 |
interpol | -2.665 |
Token | 600 |
Token ID | 10053 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
arcity | +1.165 |
Cage | +0.997 |
Hardware | +0.986 |
Sisters | +0.969 |
τ | +0.951 |
Neg logit contributions | |
anka | -0.634 |
plom | -0.611 |
agen | -0.591 |
glitter | -0.585 |
interpol | -0.555 |
Token | ) |
Token ID | 1267 |
Feature activation | +7.93e-02 |
Pos logit contributions | |
arcity | +1.292 |
Cage | +1.077 |
Hardware | +1.066 |
Sisters | +1.046 |
τ | +1.027 |
Neg logit contributions | |
anka | -0.945 |
plom | -0.911 |
agen | -0.892 |
glitter | -0.886 |
interpol | -0.853 |
Token | gray |
Token ID | 12768 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
arcity | +0.569 |
Cage | +0.499 |
Hardware | +0.496 |
Sisters | +0.489 |
τ | +0.481 |
Neg logit contributions | |
anka | -0.175 |
plom | -0.169 |
agen | -0.159 |
glitter | -0.153 |
interpol | -0.141 |
Token | = |
Token ID | 796 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
arcity | +3.501 |
Cage | +2.778 |
Hardware | +2.755 |
Sisters | +2.626 |
Drive | +2.601 |
Neg logit contributions | |
anka | -5.511 |
plom | -5.471 |
glitter | -5.297 |
agen | -5.279 |
interpol | -5.121 |
Token | c |
Token ID | 269 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
arcity | +0.741 |
Cage | +0.620 |
Hardware | +0.617 |
Sisters | +0.604 |
τ | +0.596 |
Neg logit contributions | |
anka | -0.492 |
plom | -0.476 |
agen | -0.463 |
glitter | -0.462 |
interpol | -0.445 |
Token | v |
Token ID | 85 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
arcity | +0.712 |
Cage | +0.616 |
Hardware | +0.610 |
Sisters | +0.601 |
τ | +0.592 |
Neg logit contributions | |
anka | -0.318 |
plom | -0.303 |
agen | -0.293 |
glitter | -0.289 |
interpol | -0.274 |
Token | 2 |
Token ID | 17 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
anka | +0.152 |
plom | +0.146 |
agen | +0.143 |
glitter | +0.141 |
interpol | +0.137 |
Neg logit contributions | |
arcity | -0.205 |
Cage | -0.169 |
Hardware | -0.165 |
Sisters | -0.164 |
τ | -0.159 |
Token | . |
Token ID | 764 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
arcity | +1.641 |
Cage | +1.381 |
Hardware | +1.368 |
Sisters | +1.342 |
τ | +1.319 |
Neg logit contributions | |
anka | -1.118 |
plom | -1.080 |
agen | -1.051 |
glitter | -1.045 |
interpol | -1.002 |
Token | He |
Token ID | 679 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
anka | +0.425 |
plom | +0.410 |
agen | +0.402 |
glitter | +0.398 |
interpol | +0.384 |
Neg logit contributions | |
arcity | -0.488 |
Cage | -0.393 |
Hardware | -0.390 |
Sisters | -0.383 |
τ | -0.372 |
Token | was |
Token ID | 373 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
arcity | +0.142 |
Hardware | +0.117 |
Cage | +0.116 |
Sisters | +0.114 |
Drive | +0.111 |
Neg logit contributions | |
anka | -0.117 |
plom | -0.113 |
agen | -0.110 |
glitter | -0.110 |
interpol | -0.106 |
Token | given |
Token ID | 1813 |
Feature activation | -9.15e-02 |
Pos logit contributions | |
anka | +0.208 |
plom | +0.202 |
glitter | +0.197 |
agen | +0.194 |
interpol | +0.189 |
Neg logit contributions | |
arcity | -0.273 |
Hardware | -0.227 |
Cage | -0.224 |
Sisters | -0.219 |
τ | -0.215 |
Token | an |
Token ID | 281 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
anka | +0.390 |
plom | +0.374 |
glitter | +0.370 |
agen | +0.365 |
interpol | +0.354 |
Neg logit contributions | |
arcity | -0.479 |
Hardware | -0.397 |
Cage | -0.389 |
Sisters | -0.382 |
Drive | -0.376 |
Token | ult |
Token ID | 3789 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
arcity | +0.806 |
Cage | +0.679 |
Hardware | +0.678 |
Sisters | +0.661 |
τ | +0.652 |
Neg logit contributions | |
anka | -0.517 |
plom | -0.496 |
agen | -0.484 |
glitter | -0.479 |
interpol | -0.462 |
Token | im |
Token ID | 320 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
arcity | +0.472 |
Cage | +0.391 |
Hardware | +0.389 |
Sisters | +0.380 |
τ | +0.372 |
Neg logit contributions | |
anka | -0.394 |
plom | -0.383 |
agen | -0.374 |
glitter | -0.368 |
interpol | -0.355 |
Token | atum |
Token ID | 21307 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
arcity | +2.227 |
Cage | +1.329 |
Hardware | +1.191 |
Sisters | +1.113 |
τ | +1.112 |
Neg logit contributions | |
anka | -6.789 |
glitter | -6.693 |
plom | -6.676 |
agen | -6.552 |
interpol | -6.530 |
Token | and |
Token ID | 290 |
Feature activation | -9.57e-02 |
Pos logit contributions | |
anka | +0.189 |
plom | +0.183 |
agen | +0.178 |
glitter | +0.178 |
interpol | +0.171 |
Neg logit contributions | |
arcity | -0.241 |
Hardware | -0.199 |
Cage | -0.199 |
Sisters | -0.193 |
τ | -0.190 |
Token | a |
Token ID | 257 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
anka | +0.405 |
plom | +0.389 |
glitter | +0.381 |
agen | +0.377 |
interpol | +0.367 |
Neg logit contributions | |
arcity | -0.505 |
Hardware | -0.419 |
Cage | -0.408 |
Sisters | -0.404 |
τ | -0.395 |
Token | day |
Token ID | 1110 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
anka | +0.942 |
plom | +0.909 |
glitter | +0.893 |
agen | +0.881 |
interpol | +0.843 |
Neg logit contributions | |
arcity | -1.315 |
Hardware | -1.098 |
Cage | -1.082 |
Sisters | -1.059 |
τ | -1.045 |
Token | to |
Token ID | 284 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
anka | +0.203 |
plom | +0.196 |
agen | +0.192 |
glitter | +0.190 |
interpol | +0.182 |
Neg logit contributions | |
arcity | -0.290 |
Cage | -0.243 |
Hardware | -0.243 |
Sisters | -0.237 |
τ | -0.232 |
Token | ? |
Token ID | 30 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
arcity | +0.115 |
Cage | +0.092 |
Hardware | +0.092 |
Sisters | +0.089 |
Drive | +0.087 |
Neg logit contributions | |
anka | -0.116 |
plom | -0.112 |
agen | -0.111 |
glitter | -0.110 |
interpol | -0.106 |
Token | � |
Token ID | 447 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
arcity | +0.153 |
Cage | +0.122 |
Hardware | +0.122 |
Sisters | +0.118 |
τ | +0.117 |
Neg logit contributions | |
anka | -0.140 |
plom | -0.134 |
glitter | -0.133 |
agen | -0.133 |
interpol | -0.128 |
Token | � |
Token ID | 251 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
arcity | +0.918 |
Cage | +0.775 |
Hardware | +0.766 |
Sisters | +0.755 |
Drive | +0.738 |
Neg logit contributions | |
anka | -0.586 |
plom | -0.563 |
agen | -0.551 |
glitter | -0.549 |
interpol | -0.522 |
Token | is |
Token ID | 318 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
arcity | +0.424 |
Cage | +0.341 |
Hardware | +0.340 |
Sisters | +0.329 |
τ | +0.324 |
Neg logit contributions | |
anka | -0.402 |
plom | -0.387 |
glitter | -0.384 |
agen | -0.380 |
interpol | -0.368 |
Token | Sh |
Token ID | 911 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
arcity | +1.315 |
Hardware | +1.086 |
Cage | +1.085 |
Sisters | +1.055 |
τ | +1.040 |
Neg logit contributions | |
anka | -1.030 |
plom | -0.994 |
glitter | -0.980 |
agen | -0.971 |
interpol | -0.937 |
Token | ag |
Token ID | 363 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
anka | +0.083 |
plom | +0.079 |
agen | +0.078 |
glitter | +0.076 |
interpol | +0.073 |
Neg logit contributions | |
arcity | -0.108 |
Cage | -0.091 |
Hardware | -0.090 |
Sisters | -0.088 |
τ | -0.086 |
Token | gy |
Token ID | 1360 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
arcity | +4.845 |
Cage | +4.039 |
Hardware | +4.009 |
Sisters | +3.956 |
τ | +3.843 |
Neg logit contributions | |
anka | -4.064 |
plom | -4.016 |
agen | -3.896 |
glitter | -3.856 |
interpol | -3.757 |
Token | � |
Token ID | 447 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
arcity | +0.393 |
Cage | +0.312 |
Hardware | +0.310 |
Sisters | +0.301 |
τ | +0.297 |
Neg logit contributions | |
anka | -0.389 |
plom | -0.373 |
glitter | -0.370 |
agen | -0.367 |
interpol | -0.354 |
Token | � |
Token ID | 247 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
arcity | +2.025 |
Cage | +1.712 |
Hardware | +1.695 |
Sisters | +1.663 |
τ | +1.636 |
Neg logit contributions | |
anka | -1.318 |
plom | -1.273 |
agen | -1.237 |
glitter | -1.230 |
interpol | -1.178 |
Token | s |
Token ID | 82 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
arcity | +0.992 |
Cage | +0.810 |
Hardware | +0.805 |
Sisters | +0.784 |
τ | +0.768 |
Neg logit contributions | |
anka | -0.875 |
plom | -0.846 |
agen | -0.828 |
glitter | -0.825 |
interpol | -0.794 |
Token | most |
Token ID | 749 |
Feature activation | -9.04e-02 |
Pos logit contributions | |
arcity | +0.606 |
Cage | +0.504 |
Hardware | +0.503 |
Sisters | +0.490 |
τ | +0.484 |
Neg logit contributions | |
anka | -0.431 |
plom | -0.415 |
glitter | -0.406 |
agen | -0.405 |
interpol | -0.389 |
Token | News |
Token ID | 3000 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
arcity | +2.188 |
Cage | +1.791 |
Hardware | +1.769 |
Sisters | +1.730 |
τ | +1.693 |
Neg logit contributions | |
anka | -2.052 |
plom | -1.994 |
agen | -1.948 |
glitter | -1.939 |
interpol | -1.873 |
Token | Alert |
Token ID | 23276 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
arcity | +1.616 |
Cage | +1.344 |
Hardware | +1.330 |
Sisters | +1.305 |
τ | +1.279 |
Neg logit contributions | |
anka | -1.286 |
plom | -1.246 |
agen | -1.214 |
glitter | -1.210 |
interpol | -1.165 |
Token | s |
Token ID | 82 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
anka | +0.119 |
plom | +0.116 |
agen | +0.113 |
glitter | +0.112 |
interpol | +0.108 |
Neg logit contributions | |
arcity | -0.157 |
Cage | -0.131 |
Hardware | -0.130 |
Sisters | -0.126 |
τ | -0.125 |
Token | Keep |
Token ID | 9175 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
arcity | +0.135 |
Cage | +0.111 |
Hardware | +0.110 |
Sisters | +0.108 |
τ | +0.106 |
Neg logit contributions | |
anka | -0.119 |
plom | -0.116 |
agen | -0.113 |
glitter | -0.112 |
interpol | -0.108 |
Token | an |
Token ID | 281 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
anka | +1.210 |
plom | +1.181 |
agen | +1.149 |
glitter | +1.144 |
interpol | +1.102 |
Neg logit contributions | |
arcity | -1.597 |
Hardware | -1.341 |
Cage | -1.329 |
Sisters | -1.295 |
τ | -1.285 |
Token | eye |
Token ID | 4151 |
Feature activation | +9.55e-01 |
Pos logit contributions | |
arcity | +1.932 |
Cage | +1.656 |
Hardware | +1.646 |
Sisters | +1.614 |
τ | +1.592 |
Neg logit contributions | |
anka | -0.993 |
plom | -0.953 |
agen | -0.924 |
glitter | -0.916 |
interpol | -0.870 |
Token | out |
Token ID | 503 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
arcity | +4.553 |
Cage | +3.840 |
Sisters | +3.720 |
Hardware | +3.709 |
τ | +3.579 |
Neg logit contributions | |
anka | -4.289 |
plom | -4.191 |
agen | -4.079 |
glitter | -3.969 |
interpol | -3.890 |
Token | for |
Token ID | 329 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
arcity | +2.042 |
Cage | +1.695 |
Hardware | +1.672 |
Sisters | +1.644 |
τ | +1.607 |
Neg logit contributions | |
anka | -1.698 |
plom | -1.653 |
agen | -1.611 |
glitter | -1.600 |
interpol | -1.544 |
Token | an |
Token ID | 281 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
arcity | +2.325 |
Cage | +1.934 |
Hardware | +1.917 |
Sisters | +1.875 |
τ | +1.845 |
Neg logit contributions | |
anka | -1.804 |
plom | -1.746 |
agen | -1.702 |
glitter | -1.695 |
interpol | -1.630 |
Token | email |
Token ID | 3053 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
arcity | +0.785 |
Cage | +0.663 |
Hardware | +0.660 |
Sisters | +0.643 |
τ | +0.637 |
Neg logit contributions | |
anka | -0.518 |
plom | -0.498 |
agen | -0.488 |
glitter | -0.480 |
interpol | -0.463 |
Token | to |
Token ID | 284 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
arcity | +0.743 |
Cage | +0.623 |
Hardware | +0.622 |
Sisters | +0.605 |
τ | +0.599 |
Neg logit contributions | |
anka | -0.520 |
plom | -0.505 |
agen | -0.489 |
glitter | -0.488 |
interpol | -0.470 |
Token |
|
Token ID | 198 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
arcity | +0.757 |
Cage | +0.591 |
Hardware | +0.587 |
Sisters | +0.567 |
τ | +0.554 |
Neg logit contributions | |
anka | -0.933 |
plom | -0.907 |
agen | -0.890 |
glitter | -0.887 |
interpol | -0.859 |
Token |
|
Token ID | 198 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
arcity | +1.209 |
Cage | +0.939 |
Hardware | +0.925 |
Sisters | +0.901 |
τ | +0.881 |
Neg logit contributions | |
anka | -1.475 |
plom | -1.436 |
agen | -1.412 |
glitter | -1.403 |
interpol | -1.361 |
Token | Update |
Token ID | 10260 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
arcity | +1.041 |
Cage | +0.828 |
Sisters | +0.798 |
Hardware | +0.796 |
RI | +0.771 |
Neg logit contributions | |
anka | -1.071 |
plom | -1.043 |
agen | -1.027 |
glitter | -1.010 |
interpol | -0.974 |
Token | , |
Token ID | 11 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
arcity | +1.298 |
Cage | +1.052 |
Hardware | +1.044 |
Sisters | +1.015 |
τ | +0.998 |
Neg logit contributions | |
anka | -1.260 |
plom | -1.224 |
agen | -1.201 |
glitter | -1.192 |
interpol | -1.156 |
Token | 3 |
Token ID | 513 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
arcity | +1.369 |
Hardware | +1.094 |
Cage | +1.090 |
Sisters | +1.050 |
τ | +1.046 |
Neg logit contributions | |
anka | -1.361 |
plom | -1.316 |
glitter | -1.296 |
agen | -1.294 |
interpol | -1.259 |
Token | : |
Token ID | 25 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
arcity | +3.530 |
Cage | +2.765 |
Hardware | +2.717 |
Sisters | +2.670 |
τ | +2.579 |
Neg logit contributions | |
anka | -4.716 |
plom | -4.618 |
agen | -4.502 |
glitter | -4.483 |
interpol | -4.361 |
Token | 01 |
Token ID | 486 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
arcity | +4.305 |
Cage | +3.601 |
Hardware | +3.532 |
Sisters | +3.472 |
τ | +3.376 |
Neg logit contributions | |
anka | -4.527 |
plom | -4.428 |
agen | -4.294 |
glitter | -4.288 |
interpol | -4.132 |
Token | p |
Token ID | 279 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
arcity | +2.821 |
Cage | +2.284 |
Hardware | +2.259 |
Sisters | +2.201 |
τ | +2.156 |
Neg logit contributions | |
anka | -2.879 |
plom | -2.801 |
agen | -2.743 |
glitter | -2.730 |
interpol | -2.640 |
Token | . |
Token ID | 13 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
arcity | +1.414 |
Cage | +1.171 |
Hardware | +1.159 |
Sisters | +1.136 |
τ | +1.116 |
Neg logit contributions | |
anka | -1.066 |
plom | -1.024 |
agen | -1.009 |
glitter | -0.992 |
interpol | -0.956 |
Token | m |
Token ID | 76 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
arcity | +1.754 |
Cage | +1.439 |
Hardware | +1.422 |
Sisters | +1.390 |
τ | +1.362 |
Neg logit contributions | |
anka | -1.592 |
plom | -1.547 |
agen | -1.512 |
glitter | -1.503 |
interpol | -1.451 |
Token | . |
Token ID | 13 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
arcity | +0.913 |
Cage | +0.753 |
Hardware | +0.741 |
Sisters | +0.728 |
τ | +0.711 |
Neg logit contributions | |
anka | -0.830 |
plom | -0.806 |
glitter | -0.791 |
agen | -0.779 |
interpol | -0.760 |
Token | 20 |
Token ID | 1160 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
arcity | +0.115 |
Cage | +0.093 |
Hardware | +0.093 |
Drive | +0.089 |
Sisters | +0.089 |
Neg logit contributions | |
anka | -0.097 |
plom | -0.095 |
glitter | -0.094 |
agen | -0.094 |
interpol | -0.092 |
Token | cents |
Token ID | 16059 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
arcity | +0.546 |
Cage | +0.456 |
Hardware | +0.453 |
Sisters | +0.440 |
Drive | +0.432 |
Neg logit contributions | |
anka | -0.412 |
plom | -0.401 |
agen | -0.392 |
glitter | -0.388 |
interpol | -0.375 |
Token | per |
Token ID | 583 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
arcity | +1.507 |
Cage | +1.233 |
Hardware | +1.223 |
Sisters | +1.193 |
τ | +1.167 |
Neg logit contributions | |
anka | -1.352 |
plom | -1.323 |
agen | -1.300 |
glitter | -1.289 |
interpol | -1.249 |
Token | kil |
Token ID | 8769 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
arcity | +1.252 |
Hardware | +1.045 |
Cage | +1.043 |
Sisters | +1.018 |
τ | +1.001 |
Neg logit contributions | |
anka | -0.822 |
plom | -0.794 |
agen | -0.776 |
glitter | -0.770 |
interpol | -0.739 |
Token | ow |
Token ID | 322 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
arcity | +0.771 |
Cage | +0.609 |
Hardware | +0.601 |
Sisters | +0.588 |
τ | +0.567 |
Neg logit contributions | |
anka | -0.985 |
plom | -0.958 |
agen | -0.946 |
glitter | -0.931 |
interpol | -0.909 |
Token | att |
Token ID | 1078 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
arcity | +2.195 |
Cage | +1.826 |
Hardware | +1.811 |
Sisters | +1.778 |
τ | +1.733 |
Neg logit contributions | |
anka | -1.790 |
plom | -1.729 |
agen | -1.701 |
glitter | -1.670 |
interpol | -1.619 |
Token | - |
Token ID | 12 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
arcity | +4.794 |
Cage | +3.959 |
Hardware | +3.915 |
Sisters | +3.842 |
τ | +3.809 |
Neg logit contributions | |
anka | -4.014 |
plom | -3.936 |
glitter | -3.804 |
agen | -3.721 |
interpol | -3.606 |
Token | hour |
Token ID | 9769 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
arcity | +2.093 |
Cage | +1.708 |
Hardware | +1.684 |
Sisters | +1.649 |
τ | +1.599 |
Neg logit contributions | |
anka | -1.975 |
plom | -1.913 |
agen | -1.885 |
glitter | -1.851 |
interpol | -1.801 |
Token | , |
Token ID | 11 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
arcity | +1.850 |
Cage | +1.527 |
Hardware | +1.514 |
Sisters | +1.482 |
τ | +1.449 |
Neg logit contributions | |
anka | -1.529 |
plom | -1.482 |
agen | -1.461 |
glitter | -1.448 |
interpol | -1.404 |
Token | electricity |
Token ID | 8744 |
Feature activation | -8.99e-02 |
Pos logit contributions | |
arcity | +0.863 |
Hardware | +0.714 |
Cage | +0.710 |
Sisters | +0.696 |
τ | +0.684 |
Neg logit contributions | |
anka | -0.650 |
glitter | -0.625 |
plom | -0.624 |
agen | -0.617 |
interpol | -0.605 |
Token | from |
Token ID | 422 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
anka | +0.346 |
plom | +0.333 |
glitter | +0.330 |
agen | +0.330 |
interpol | +0.324 |
Neg logit contributions | |
arcity | -0.501 |
Cage | -0.419 |
Hardware | -0.418 |
Sisters | -0.405 |
τ | -0.398 |
Token | website |
Token ID | 3052 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
anka | +0.693 |
plom | +0.666 |
glitter | +0.664 |
agen | +0.663 |
interpol | +0.638 |
Neg logit contributions | |
arcity | -0.608 |
Hardware | -0.475 |
Cage | -0.466 |
Sisters | -0.452 |
τ | -0.451 |
Token | . |
Token ID | 13 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
anka | +1.150 |
plom | +1.103 |
glitter | +1.082 |
agen | +1.082 |
interpol | +1.037 |
Neg logit contributions | |
arcity | -1.588 |
Cage | -1.316 |
Hardware | -1.304 |
Sisters | -1.280 |
τ | -1.255 |
Token |
|
Token ID | 198 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
arcity | +0.284 |
Cage | +0.225 |
Hardware | +0.222 |
Sisters | +0.216 |
τ | +0.211 |
Neg logit contributions | |
anka | -0.333 |
plom | -0.324 |
agen | -0.317 |
glitter | -0.316 |
interpol | -0.306 |
Token |
|
Token ID | 198 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
anka | +0.247 |
plom | +0.238 |
agen | +0.236 |
glitter | +0.234 |
interpol | +0.226 |
Neg logit contributions | |
arcity | -0.224 |
Cage | -0.174 |
Hardware | -0.171 |
Sisters | -0.168 |
τ | -0.164 |
Token | M |
Token ID | 44 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
arcity | +0.091 |
Cage | +0.072 |
Sisters | +0.069 |
Hardware | +0.069 |
RI | +0.067 |
Neg logit contributions | |
anka | -0.097 |
plom | -0.094 |
agen | -0.092 |
glitter | -0.091 |
interpol | -0.088 |
Token | ISS |
Token ID | 16744 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
arcity | +0.259 |
Cage | +0.215 |
Hardware | +0.213 |
Sisters | +0.207 |
τ | +0.203 |
Neg logit contributions | |
anka | -0.218 |
plom | -0.210 |
agen | -0.207 |
glitter | -0.203 |
interpol | -0.195 |
Token | ING |
Token ID | 2751 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
arcity | +4.354 |
Cage | +3.606 |
Hardware | +3.552 |
Sisters | +3.494 |
RI | +3.433 |
Neg logit contributions | |
anka | -4.262 |
plom | -4.153 |
agen | -4.025 |
glitter | -4.014 |
interpol | -3.905 |
Token | 2 |
Token ID | 362 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
arcity | +1.312 |
Cage | +1.035 |
Hardware | +1.030 |
Sisters | +0.996 |
τ | +0.982 |
Neg logit contributions | |
anka | -1.430 |
plom | -1.390 |
glitter | -1.372 |
agen | -1.365 |
interpol | -1.321 |
Token | - |
Token ID | 12 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
arcity | +0.491 |
Cage | +0.392 |
Hardware | +0.389 |
Sisters | +0.372 |
τ | +0.370 |
Neg logit contributions | |
anka | -0.500 |
plom | -0.489 |
glitter | -0.481 |
agen | -0.477 |
interpol | -0.462 |
Token | Y |
Token ID | 56 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
arcity | +0.073 |
Cage | +0.058 |
Hardware | +0.057 |
Sisters | +0.056 |
τ | +0.055 |
Neg logit contributions | |
anka | -0.073 |
plom | -0.071 |
agen | -0.070 |
glitter | -0.069 |
interpol | -0.067 |
Token | EAR |
Token ID | 17133 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
arcity | +0.094 |
Hardware | +0.081 |
Cage | +0.081 |
Sisters | +0.079 |
τ | +0.079 |
Neg logit contributions | |
anka | -0.040 |
agen | -0.037 |
plom | -0.037 |
glitter | -0.036 |
interpol | -0.033 |
Token | other |
Token ID | 584 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
arcity | +0.251 |
Cage | +0.210 |
Hardware | +0.208 |
Sisters | +0.204 |
τ | +0.200 |
Neg logit contributions | |
anka | -0.185 |
plom | -0.179 |
agen | -0.175 |
glitter | -0.174 |
interpol | -0.167 |
Token | . |
Token ID | 13 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
anka | +0.495 |
plom | +0.474 |
glitter | +0.470 |
agen | +0.467 |
interpol | +0.443 |
Neg logit contributions | |
arcity | -0.640 |
Hardware | -0.524 |
Cage | -0.523 |
Sisters | -0.507 |
Drive | -0.501 |
Token | Glow |
Token ID | 45929 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
arcity | +0.901 |
Hardware | +0.715 |
Cage | +0.711 |
τ | +0.684 |
Sisters | +0.679 |
Neg logit contributions | |
anka | -0.840 |
plom | -0.812 |
glitter | -0.798 |
agen | -0.796 |
interpol | -0.763 |
Token | - |
Token ID | 12 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
anka | +0.726 |
plom | +0.700 |
glitter | +0.697 |
agen | +0.690 |
interpol | +0.653 |
Neg logit contributions | |
arcity | -0.881 |
Cage | -0.722 |
Hardware | -0.720 |
Sisters | -0.697 |
τ | -0.685 |
Token | in |
Token ID | 259 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
arcity | +0.810 |
Cage | +0.662 |
Hardware | +0.655 |
Sisters | +0.642 |
τ | +0.626 |
Neg logit contributions | |
anka | -0.745 |
plom | -0.723 |
agen | -0.711 |
glitter | -0.708 |
interpol | -0.677 |
Token | - |
Token ID | 12 |
Feature activation | +9.11e-01 |
Pos logit contributions | |
arcity | +0.578 |
Cage | +0.487 |
Hardware | +0.478 |
Sisters | +0.472 |
τ | +0.463 |
Neg logit contributions | |
anka | -0.427 |
plom | -0.416 |
agen | -0.403 |
glitter | -0.396 |
interpol | -0.385 |
Token | the |
Token ID | 1169 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
arcity | +3.023 |
Cage | +2.300 |
Hardware | +2.275 |
τ | +2.132 |
Drive | +2.111 |
Neg logit contributions | |
anka | -5.454 |
plom | -5.349 |
agen | -5.206 |
glitter | -5.200 |
interpol | -5.141 |
Token | - |
Token ID | 12 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
arcity | +1.831 |
Cage | +1.540 |
Hardware | +1.525 |
Sisters | +1.495 |
τ | +1.470 |
Neg logit contributions | |
anka | -1.267 |
plom | -1.225 |
agen | -1.193 |
glitter | -1.186 |
interpol | -1.137 |
Token | dark |
Token ID | 21953 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
arcity | +3.922 |
Cage | +3.367 |
Hardware | +3.322 |
Sisters | +3.235 |
τ | +3.206 |
Neg logit contributions | |
anka | -2.693 |
plom | -2.626 |
agen | -2.530 |
glitter | -2.508 |
interpol | -2.444 |
Token | football |
Token ID | 4346 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
arcity | +1.768 |
Cage | +1.467 |
Hardware | +1.458 |
Sisters | +1.420 |
τ | +1.399 |
Neg logit contributions | |
anka | -1.355 |
plom | -1.310 |
glitter | -1.282 |
agen | -1.278 |
interpol | -1.225 |
Token | s |
Token ID | 82 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
anka | +0.526 |
plom | +0.510 |
glitter | +0.507 |
agen | +0.497 |
interpol | +0.478 |
Neg logit contributions | |
arcity | -0.689 |
Cage | -0.567 |
Hardware | -0.566 |
Sisters | -0.546 |
Drive | -0.539 |
Token |
|
Token ID | 198 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
arcity | +0.767 |
Cage | +0.600 |
Hardware | +0.580 |
Sisters | +0.574 |
τ | +0.559 |
Neg logit contributions | |
anka | -0.905 |
plom | -0.880 |
agen | -0.864 |
glitter | -0.858 |
interpol | -0.830 |
Token | Does |
Token ID | 13921 |
Feature activation | -9.66e-02 |
Pos logit contributions | |
arcity | +0.700 |
Cage | +0.556 |
Sisters | +0.537 |
Hardware | +0.532 |
RI | +0.518 |
Neg logit contributions | |
anka | -0.736 |
plom | -0.717 |
agen | -0.704 |
glitter | -0.695 |
interpol | -0.672 |
Token | One |
Token ID | 1881 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
anka | +0.405 |
plom | +0.392 |
glitter | +0.385 |
agen | +0.383 |
interpol | +0.366 |
Neg logit contributions | |
arcity | -0.507 |
Hardware | -0.420 |
Cage | -0.420 |
Sisters | -0.406 |
τ | -0.404 |
Token | - |
Token ID | 12 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
arcity | +0.252 |
Hardware | +0.203 |
Cage | +0.201 |
Sisters | +0.195 |
τ | +0.195 |
Neg logit contributions | |
anka | -0.247 |
plom | -0.239 |
agen | -0.235 |
glitter | -0.234 |
interpol | -0.226 |
Token | on |
Token ID | 261 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
arcity | +1.155 |
Cage | +0.923 |
Hardware | +0.899 |
Sisters | +0.889 |
RI | +0.872 |
Neg logit contributions | |
anka | -1.099 |
plom | -1.073 |
agen | -1.051 |
glitter | -1.040 |
interpol | -1.005 |
Token | - |
Token ID | 12 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
arcity | +0.151 |
Cage | +0.126 |
Hardware | +0.125 |
Sisters | +0.123 |
τ | +0.122 |
Neg logit contributions | |
anka | -0.095 |
plom | -0.091 |
agen | -0.090 |
glitter | -0.088 |
interpol | -0.084 |
Token | One |
Token ID | 3198 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
arcity | +3.427 |
Cage | +2.649 |
Hardware | +2.630 |
Sisters | +2.520 |
τ | +2.461 |
Neg logit contributions | |
anka | -5.052 |
plom | -4.947 |
glitter | -4.846 |
agen | -4.839 |
interpol | -4.705 |
Token | Out |
Token ID | 3806 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
anka | +0.495 |
plom | +0.477 |
glitter | +0.468 |
agen | +0.457 |
interpol | +0.450 |
Neg logit contributions | |
arcity | -0.630 |
Hardware | -0.520 |
Cage | -0.508 |
τ | -0.504 |
Sisters | -0.498 |
Token | reach |
Token ID | 16250 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
anka | +0.183 |
plom | +0.178 |
agen | +0.173 |
glitter | +0.173 |
interpol | +0.166 |
Neg logit contributions | |
arcity | -0.193 |
Hardware | -0.157 |
Cage | -0.157 |
Sisters | -0.152 |
τ | -0.151 |
Token | Make |
Token ID | 6889 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
anka | +0.119 |
plom | +0.116 |
glitter | +0.115 |
agen | +0.113 |
interpol | +0.111 |
Neg logit contributions | |
arcity | -0.137 |
Hardware | -0.112 |
Cage | -0.110 |
τ | -0.108 |
Sisters | -0.107 |
Token | You |
Token ID | 921 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
arcity | +0.547 |
Hardware | +0.450 |
Cage | +0.446 |
τ | +0.431 |
Sisters | +0.431 |
Neg logit contributions | |
anka | -0.455 |
plom | -0.443 |
glitter | -0.430 |
agen | -0.427 |
interpol | -0.413 |
Token | draw |
Token ID | 19334 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
arcity | +0.872 |
Cage | +0.732 |
Hardware | +0.722 |
Sisters | +0.713 |
τ | +0.695 |
Neg logit contributions | |
anka | -0.629 |
plom | -0.610 |
agen | -0.596 |
glitter | -0.587 |
interpol | -0.565 |
Token | ing |
Token ID | 278 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
arcity | +2.137 |
Cage | +1.785 |
Hardware | +1.772 |
Sisters | +1.726 |
τ | +1.698 |
Neg logit contributions | |
anka | -1.647 |
plom | -1.601 |
glitter | -1.558 |
agen | -1.549 |
interpol | -1.493 |
Token | . |
Token ID | 13 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
arcity | +0.931 |
Cage | +0.770 |
Hardware | +0.760 |
Sisters | +0.748 |
τ | +0.733 |
Neg logit contributions | |
anka | -0.715 |
plom | -0.691 |
agen | -0.677 |
glitter | -0.670 |
interpol | -0.646 |
Token | String |
Token ID | 10100 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
arcity | +1.663 |
Cage | +1.269 |
Hardware | +1.236 |
Sisters | +1.210 |
τ | +1.166 |
Neg logit contributions | |
anka | -2.441 |
plom | -2.386 |
agen | -2.345 |
glitter | -2.332 |
interpol | -2.271 |
Token | Al |
Token ID | 2348 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
arcity | +0.992 |
Cage | +0.736 |
Hardware | +0.717 |
Sisters | +0.700 |
τ | +0.676 |
Neg logit contributions | |
anka | -1.657 |
plom | -1.619 |
agen | -1.596 |
glitter | -1.583 |
interpol | -1.546 |
Token | ignment |
Token ID | 16747 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
arcity | +3.460 |
Cage | +3.013 |
Hardware | +3.005 |
Sisters | +2.939 |
Drive | +2.912 |
Neg logit contributions | |
anka | -1.028 |
plom | -0.985 |
glitter | -0.959 |
agen | -0.910 |
interpol | -0.863 |
Token | ] |
Token ID | 60 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
arcity | +4.282 |
Hardware | +3.587 |
Cage | +3.583 |
Sisters | +3.452 |
τ | +3.418 |
Neg logit contributions | |
anka | -4.001 |
plom | -3.901 |
glitter | -3.789 |
agen | -3.761 |
interpol | -3.635 |
Token | :: |
Token ID | 3712 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
arcity | +1.437 |
Cage | +1.179 |
Hardware | +1.163 |
Sisters | +1.149 |
τ | +1.115 |
Neg logit contributions | |
anka | -1.108 |
plom | -1.073 |
agen | -1.054 |
glitter | -1.039 |
interpol | -1.005 |
Token | Near |
Token ID | 40640 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
arcity | +0.670 |
Cage | +0.520 |
Hardware | +0.508 |
Sisters | +0.504 |
RI | +0.488 |
Neg logit contributions | |
anka | -0.749 |
plom | -0.731 |
agen | -0.720 |
glitter | -0.710 |
interpol | -0.691 |
Token | } |
Token ID | 1782 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
arcity | +1.105 |
Cage | +0.899 |
Hardware | +0.887 |
Sisters | +0.867 |
τ | +0.851 |
Neg logit contributions | |
anka | -1.051 |
plom | -1.024 |
agen | -1.003 |
glitter | -1.000 |
interpol | -0.969 |
Token | } |
Token ID | 1782 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
arcity | +0.237 |
Hardware | +0.187 |
Cage | +0.186 |
Sisters | +0.182 |
τ | +0.181 |
Neg logit contributions | |
anka | -0.219 |
plom | -0.213 |
glitter | -0.212 |
agen | -0.211 |
interpol | -0.206 |
Token | be |
Token ID | 1350 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
anka | +0.178 |
plom | +0.171 |
agen | +0.169 |
glitter | +0.166 |
interpol | +0.161 |
Neg logit contributions | |
arcity | -0.215 |
Cage | -0.178 |
Hardware | -0.174 |
Sisters | -0.173 |
τ | -0.168 |
Token | 66 |
Token ID | 2791 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
arcity | +0.571 |
Cage | +0.474 |
Hardware | +0.465 |
Sisters | +0.461 |
τ | +0.449 |
Neg logit contributions | |
anka | -0.471 |
plom | -0.455 |
agen | -0.447 |
glitter | -0.442 |
interpol | -0.427 |
Token | IEEE |
Token ID | 40552 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
arcity | +1.149 |
Cage | +0.942 |
Hardware | +0.928 |
Sisters | +0.914 |
τ | +0.893 |
Neg logit contributions | |
anka | -1.029 |
plom | -0.993 |
agen | -0.975 |
glitter | -0.968 |
interpol | -0.934 |
Token | 802 |
Token ID | 33121 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
arcity | +1.276 |
Cage | +0.958 |
Hardware | +0.942 |
Sisters | +0.923 |
τ | +0.898 |
Neg logit contributions | |
anka | -1.823 |
plom | -1.775 |
agen | -1.747 |
glitter | -1.743 |
interpol | -1.712 |
Token | . |
Token ID | 13 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
arcity | +2.773 |
Cage | +2.302 |
Hardware | +2.279 |
Sisters | +2.231 |
τ | +2.189 |
Neg logit contributions | |
anka | -2.249 |
plom | -2.181 |
agen | -2.126 |
glitter | -2.116 |
interpol | -2.036 |
Token | 11 |
Token ID | 1157 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
arcity | +2.790 |
Hardware | +2.269 |
Cage | +2.261 |
Sisters | +2.170 |
τ | +2.133 |
Neg logit contributions | |
anka | -3.047 |
plom | -2.972 |
glitter | -2.909 |
agen | -2.872 |
interpol | -2.792 |
Token | b |
Token ID | 65 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
arcity | +4.507 |
Cage | +3.744 |
Hardware | +3.719 |
Sisters | +3.627 |
τ | +3.546 |
Neg logit contributions | |
anka | -3.820 |
plom | -3.720 |
glitter | -3.610 |
agen | -3.605 |
interpol | -3.459 |
Token | gn |
Token ID | 4593 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
arcity | +0.567 |
Cage | +0.459 |
Hardware | +0.443 |
Sisters | +0.440 |
τ | +0.433 |
Neg logit contributions | |
anka | -0.516 |
plom | -0.504 |
agen | -0.494 |
glitter | -0.489 |
interpol | -0.480 |
Token | E |
Token ID | 412 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
arcity | +0.557 |
Cage | +0.450 |
Hardware | +0.439 |
Sisters | +0.434 |
τ | +0.424 |
Neg logit contributions | |
anka | -0.522 |
plom | -0.506 |
agen | -0.498 |
glitter | -0.492 |
interpol | -0.482 |
Token | SS |
Token ID | 5432 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
arcity | +0.621 |
Cage | +0.501 |
Hardware | +0.493 |
Sisters | +0.486 |
τ | +0.475 |
Neg logit contributions | |
anka | -0.577 |
plom | -0.564 |
agen | -0.554 |
glitter | -0.545 |
interpol | -0.530 |
Token | ID |
Token ID | 2389 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
arcity | +0.439 |
Cage | +0.337 |
Hardware | +0.332 |
Sisters | +0.328 |
τ | +0.317 |
Neg logit contributions | |
anka | -0.522 |
plom | -0.507 |
agen | -0.503 |
glitter | -0.502 |
interpol | -0.487 |
Token | original |
Token ID | 2656 |
Feature activation | +7.07e-02 |
Pos logit contributions | |
anka | +0.886 |
plom | +0.852 |
agen | +0.824 |
glitter | +0.822 |
interpol | +0.778 |
Neg logit contributions | |
arcity | -1.788 |
Cage | -1.529 |
Hardware | -1.522 |
Sisters | -1.495 |
τ | -1.474 |
Token | image |
Token ID | 2939 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
arcity | +0.413 |
Cage | +0.350 |
Hardware | +0.347 |
Sisters | +0.341 |
τ | +0.336 |
Neg logit contributions | |
anka | -0.255 |
plom | -0.247 |
agen | -0.238 |
glitter | -0.237 |
interpol | -0.226 |
Token | Download |
Token ID | 10472 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
anka | +1.143 |
plom | +1.114 |
agen | +1.096 |
glitter | +1.090 |
interpol | +1.058 |
Neg logit contributions | |
arcity | -0.831 |
Cage | -0.640 |
Hardware | -0.633 |
Sisters | -0.610 |
τ | -0.596 |
Token | : |
Token ID | 25 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
anka | +1.221 |
plom | +1.190 |
glitter | +1.179 |
agen | +1.178 |
interpol | +1.133 |
Neg logit contributions | |
arcity | -1.089 |
Cage | -0.859 |
Hardware | -0.855 |
Sisters | -0.830 |
τ | -0.805 |
Token | Figure |
Token ID | 11291 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
arcity | +0.680 |
Hardware | +0.544 |
Cage | +0.541 |
Sisters | +0.527 |
Drive | +0.518 |
Neg logit contributions | |
anka | -0.630 |
plom | -0.610 |
glitter | -0.603 |
agen | -0.603 |
interpol | -0.581 |
Token | 1 |
Token ID | 352 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
anka | +0.756 |
plom | +0.717 |
agen | +0.714 |
glitter | +0.712 |
interpol | +0.694 |
Neg logit contributions | |
arcity | -0.815 |
Hardware | -0.649 |
Cage | -0.647 |
Sisters | -0.625 |
τ | -0.622 |
Token | . |
Token ID | 13 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
arcity | +4.121 |
Cage | +3.572 |
Hardware | +3.470 |
Sisters | +3.379 |
RI | +3.267 |
Neg logit contributions | |
anka | -4.050 |
plom | -3.975 |
agen | -3.816 |
glitter | -3.796 |
interpol | -3.661 |
Token | Maze |
Token ID | 33412 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
arcity | +3.343 |
Cage | +2.800 |
Hardware | +2.729 |
Sisters | +2.664 |
τ | +2.581 |
Neg logit contributions | |
anka | -3.441 |
plom | -3.373 |
agen | -3.261 |
glitter | -3.248 |
interpol | -3.138 |
Token | and |
Token ID | 290 |
Feature activation | -3.87e-03 |
Pos logit contributions | |
arcity | +2.969 |
Cage | +2.543 |
Hardware | +2.527 |
Sisters | +2.486 |
τ | +2.449 |
Neg logit contributions | |
anka | -1.457 |
plom | -1.396 |
agen | -1.352 |
glitter | -1.343 |
interpol | -1.274 |
Token | task |
Token ID | 4876 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
anka | +0.015 |
plom | +0.014 |
glitter | +0.014 |
agen | +0.014 |
interpol | +0.013 |
Neg logit contributions | |
arcity | -0.022 |
Hardware | -0.018 |
Cage | -0.018 |
Sisters | -0.018 |
τ | -0.018 |
Token | description |
Token ID | 6764 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
arcity | +0.186 |
Hardware | +0.155 |
Cage | +0.155 |
Sisters | +0.152 |
τ | +0.150 |
Neg logit contributions | |
anka | -0.120 |
plom | -0.115 |
agen | -0.113 |
glitter | -0.112 |
interpol | -0.108 |
Token | on |
Token ID | 319 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
arcity | +1.034 |
Cage | +0.860 |
Hardware | +0.849 |
Sisters | +0.829 |
τ | +0.815 |
Neg logit contributions | |
anka | -0.868 |
plom | -0.845 |
agen | -0.822 |
glitter | -0.818 |
interpol | -0.789 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
arcity | +2.135 |
Cage | +1.749 |
Hardware | +1.727 |
Sisters | +1.678 |
τ | +1.653 |
Neg logit contributions | |
anka | -2.163 |
plom | -2.103 |
agen | -2.059 |
glitter | -2.041 |
interpol | -1.972 |
Token | . |
Token ID | 13 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
arcity | +2.013 |
Cage | +1.688 |
Hardware | +1.664 |
Sisters | +1.623 |
τ | +1.594 |
Neg logit contributions | |
anka | -1.667 |
plom | -1.630 |
agen | -1.580 |
glitter | -1.578 |
interpol | -1.518 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
arcity | +0.515 |
Cage | +0.415 |
Hardware | +0.409 |
Sisters | +0.400 |
τ | +0.391 |
Neg logit contributions | |
anka | -0.547 |
plom | -0.532 |
agen | -0.521 |
glitter | -0.519 |
interpol | -0.502 |
Token | Q |
Token ID | 48 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
arcity | +0.182 |
Cage | +0.146 |
Sisters | +0.141 |
Hardware | +0.139 |
RI | +0.135 |
Neg logit contributions | |
anka | -0.189 |
plom | -0.184 |
agen | -0.181 |
glitter | -0.178 |
interpol | -0.172 |
Token | : |
Token ID | 25 |
Feature activation | +8.78e-01 |
Pos logit contributions | |
anka | +0.085 |
plom | +0.082 |
agen | +0.081 |
glitter | +0.079 |
interpol | +0.076 |
Neg logit contributions | |
arcity | -0.087 |
Cage | -0.070 |
Hardware | -0.070 |
Sisters | -0.068 |
τ | -0.067 |
Token | Sebastian |
Token ID | 26190 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
arcity | +4.307 |
Cage | +3.634 |
Hardware | +3.561 |
Sisters | +3.529 |
RI | +3.410 |
Neg logit contributions | |
anka | -3.846 |
plom | -3.786 |
agen | -3.675 |
glitter | -3.652 |
interpol | -3.534 |
Token | , |
Token ID | 11 |
Feature activation | -6.40e-02 |
Pos logit contributions | |
arcity | +0.723 |
Cage | +0.584 |
Hardware | +0.584 |
Sisters | +0.567 |
τ | +0.559 |
Neg logit contributions | |
anka | -0.657 |
plom | -0.634 |
agen | -0.624 |
glitter | -0.620 |
interpol | -0.601 |
Token | last |
Token ID | 938 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
anka | +0.262 |
plom | +0.251 |
agen | +0.246 |
glitter | +0.246 |
interpol | +0.237 |
Neg logit contributions | |
arcity | -0.349 |
Hardware | -0.285 |
Cage | -0.284 |
Sisters | -0.275 |
τ | -0.275 |
Token | year |
Token ID | 614 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
anka | +0.709 |
plom | +0.678 |
glitter | +0.674 |
agen | +0.665 |
interpol | +0.643 |
Neg logit contributions | |
arcity | -1.054 |
Cage | -0.869 |
Hardware | -0.866 |
Sisters | -0.845 |
τ | -0.836 |
Token | you |
Token ID | 345 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
arcity | +0.102 |
Hardware | +0.083 |
Cage | +0.083 |
Sisters | +0.080 |
τ | +0.080 |
Neg logit contributions | |
anka | -0.082 |
plom | -0.079 |
agen | -0.077 |
glitter | -0.077 |
interpol | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
arcity | +0.740 |
Cage | +0.614 |
Hardware | +0.609 |
Sisters | +0.594 |
τ | +0.585 |
Neg logit contributions | |
anka | -0.542 |
plom | -0.520 |
agen | -0.510 |
glitter | -0.507 |
interpol | -0.487 |
Token |
|
Token ID | 198 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
arcity | +0.757 |
Cage | +0.591 |
Hardware | +0.587 |
Sisters | +0.567 |
τ | +0.554 |
Neg logit contributions | |
anka | -0.933 |
plom | -0.907 |
agen | -0.890 |
glitter | -0.887 |
interpol | -0.859 |
Token |
|
Token ID | 198 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
arcity | +1.209 |
Cage | +0.939 |
Hardware | +0.925 |
Sisters | +0.901 |
τ | +0.881 |
Neg logit contributions | |
anka | -1.475 |
plom | -1.436 |
agen | -1.412 |
glitter | -1.403 |
interpol | -1.361 |
Token | Update |
Token ID | 10260 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
arcity | +1.041 |
Cage | +0.828 |
Sisters | +0.798 |
Hardware | +0.796 |
RI | +0.771 |
Neg logit contributions | |
anka | -1.071 |
plom | -1.043 |
agen | -1.027 |
glitter | -1.010 |
interpol | -0.974 |
Token | , |
Token ID | 11 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
arcity | +1.298 |
Cage | +1.052 |
Hardware | +1.044 |
Sisters | +1.015 |
τ | +0.998 |
Neg logit contributions | |
anka | -1.260 |
plom | -1.224 |
agen | -1.201 |
glitter | -1.192 |
interpol | -1.156 |
Token | 3 |
Token ID | 513 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
arcity | +1.369 |
Hardware | +1.094 |
Cage | +1.090 |
Sisters | +1.050 |
τ | +1.046 |
Neg logit contributions | |
anka | -1.361 |
plom | -1.316 |
glitter | -1.296 |
agen | -1.294 |
interpol | -1.259 |
Token | : |
Token ID | 25 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
arcity | +3.530 |
Cage | +2.765 |
Hardware | +2.717 |
Sisters | +2.670 |
τ | +2.579 |
Neg logit contributions | |
anka | -4.716 |
plom | -4.618 |
agen | -4.502 |
glitter | -4.483 |
interpol | -4.361 |
Token | 01 |
Token ID | 486 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
arcity | +4.305 |
Cage | +3.601 |
Hardware | +3.532 |
Sisters | +3.472 |
τ | +3.376 |
Neg logit contributions | |
anka | -4.527 |
plom | -4.428 |
agen | -4.294 |
glitter | -4.288 |
interpol | -4.132 |
Token | p |
Token ID | 279 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
arcity | +2.821 |
Cage | +2.284 |
Hardware | +2.259 |
Sisters | +2.201 |
τ | +2.156 |
Neg logit contributions | |
anka | -2.879 |
plom | -2.801 |
agen | -2.743 |
glitter | -2.730 |
interpol | -2.640 |
Token | . |
Token ID | 13 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
arcity | +1.414 |
Cage | +1.171 |
Hardware | +1.159 |
Sisters | +1.136 |
τ | +1.116 |
Neg logit contributions | |
anka | -1.066 |
plom | -1.024 |
agen | -1.009 |
glitter | -0.992 |
interpol | -0.956 |
Token | m |
Token ID | 76 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
arcity | +1.754 |
Cage | +1.439 |
Hardware | +1.422 |
Sisters | +1.390 |
τ | +1.362 |
Neg logit contributions | |
anka | -1.592 |
plom | -1.547 |
agen | -1.512 |
glitter | -1.503 |
interpol | -1.451 |
Token | G |
Token ID | 402 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
arcity | +0.460 |
Hardware | +0.367 |
Cage | +0.364 |
Sisters | +0.354 |
τ | +0.350 |
Neg logit contributions | |
anka | -0.445 |
glitter | -0.425 |
plom | -0.425 |
agen | -0.418 |
interpol | -0.401 |
Token | AL |
Token ID | 1847 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
arcity | +1.016 |
Cage | +0.806 |
Hardware | +0.794 |
Sisters | +0.774 |
τ | +0.755 |
Neg logit contributions | |
anka | -1.246 |
plom | -1.217 |
agen | -1.192 |
glitter | -1.187 |
interpol | -1.153 |
Token | AX |
Token ID | 25922 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
arcity | +0.243 |
Cage | +0.185 |
Hardware | +0.184 |
Sisters | +0.177 |
τ | +0.176 |
Neg logit contributions | |
anka | -0.308 |
glitter | -0.293 |
agen | -0.280 |
plom | -0.277 |
interpol | -0.276 |
Token | Y |
Token ID | 56 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
anka | +0.390 |
glitter | +0.375 |
plom | +0.369 |
agen | +0.363 |
interpol | +0.347 |
Neg logit contributions | |
arcity | -0.683 |
Cage | -0.558 |
Hardware | -0.550 |
Sisters | -0.530 |
Drive | -0.528 |
Token | VOL |
Token ID | 38570 |
Feature activation | +5.25e-01 |
Pos logit contributions | |
arcity | +0.462 |
Hardware | +0.376 |
Cage | +0.375 |
Sisters | +0.362 |
τ | +0.361 |
Neg logit contributions | |
anka | -0.350 |
plom | -0.338 |
glitter | -0.335 |
agen | -0.327 |
interpol | -0.319 |
Token | . |
Token ID | 13 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
arcity | +2.673 |
Cage | +2.219 |
Hardware | +2.196 |
Sisters | +2.151 |
τ | +2.113 |
Neg logit contributions | |
anka | -2.271 |
plom | -2.220 |
agen | -2.149 |
glitter | -2.142 |
interpol | -2.064 |
Token | 2 |
Token ID | 362 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
arcity | +3.730 |
Cage | +3.008 |
Hardware | +2.961 |
Sisters | +2.905 |
τ | +2.848 |
Neg logit contributions | |
anka | -4.515 |
plom | -4.413 |
agen | -4.303 |
glitter | -4.270 |
interpol | -4.136 |
Token | . |
Token ID | 13 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
arcity | +3.373 |
Cage | +2.806 |
Hardware | +2.761 |
Sisters | +2.724 |
τ | +2.658 |
Neg logit contributions | |
anka | -2.984 |
plom | -2.912 |
agen | -2.839 |
glitter | -2.797 |
interpol | -2.709 |
Token | However |
Token ID | 2102 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
arcity | +1.151 |
Cage | +0.924 |
Hardware | +0.920 |
Sisters | +0.893 |
τ | +0.881 |
Neg logit contributions | |
anka | -1.047 |
plom | -1.011 |
agen | -0.992 |
glitter | -0.991 |
interpol | -0.956 |
Token | , |
Token ID | 11 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
anka | +0.361 |
plom | +0.350 |
agen | +0.342 |
glitter | +0.341 |
interpol | +0.329 |
Neg logit contributions | |
arcity | -0.408 |
Cage | -0.335 |
Hardware | -0.332 |
Sisters | -0.325 |
τ | -0.318 |
Token | even |
Token ID | 772 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
anka | +0.203 |
glitter | +0.194 |
plom | +0.192 |
agen | +0.189 |
interpol | +0.185 |
Neg logit contributions | |
arcity | -0.276 |
Hardware | -0.227 |
Cage | -0.224 |
τ | -0.219 |
Sisters | -0.218 |
Token |
|
Token ID | 198 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
anka | +0.344 |
plom | +0.332 |
agen | +0.326 |
glitter | +0.324 |
interpol | +0.314 |
Neg logit contributions | |
arcity | -0.326 |
Hardware | -0.254 |
Cage | -0.253 |
Sisters | -0.244 |
τ | -0.243 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
anka | +0.008 |
plom | +0.008 |
agen | +0.008 |
glitter | +0.008 |
interpol | +0.008 |
Neg logit contributions | |
arcity | -0.008 |
Cage | -0.006 |
Hardware | -0.006 |
Sisters | -0.006 |
τ | -0.006 |
Token | U |
Token ID | 52 |
Feature activation | +4.12e-03 |
Pos logit contributions | |
arcity | +0.088 |
Cage | +0.069 |
Sisters | +0.067 |
Hardware | +0.066 |
RI | +0.065 |
Neg logit contributions | |
anka | -0.087 |
plom | -0.085 |
agen | -0.084 |
glitter | -0.082 |
interpol | -0.079 |
Token | PDATED |
Token ID | 49316 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
arcity | +0.022 |
Cage | +0.018 |
Hardware | +0.018 |
Sisters | +0.018 |
τ | +0.017 |
Neg logit contributions | |
anka | -0.017 |
plom | -0.017 |
agen | -0.016 |
glitter | -0.016 |
interpol | -0.016 |
Token | 2 |
Token ID | 362 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
arcity | +1.096 |
Cage | +0.874 |
Hardware | +0.872 |
Sisters | +0.842 |
τ | +0.831 |
Neg logit contributions | |
anka | -1.107 |
plom | -1.074 |
agen | -1.060 |
glitter | -1.052 |
interpol | -1.022 |
Token | : |
Token ID | 25 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
arcity | +2.094 |
Cage | +1.646 |
Hardware | +1.625 |
Sisters | +1.571 |
τ | +1.538 |
Neg logit contributions | |
anka | -2.639 |
plom | -2.572 |
agen | -2.526 |
glitter | -2.515 |
interpol | -2.440 |
Token | 35 |
Token ID | 2327 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
arcity | +4.055 |
Cage | +3.379 |
Hardware | +3.319 |
Sisters | +3.262 |
τ | +3.176 |
Neg logit contributions | |
anka | -4.152 |
plom | -4.057 |
agen | -3.937 |
glitter | -3.931 |
interpol | -3.789 |
Token | p |
Token ID | 279 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
arcity | +2.412 |
Cage | +1.944 |
Hardware | +1.927 |
Sisters | +1.872 |
τ | +1.834 |
Neg logit contributions | |
anka | -2.540 |
plom | -2.470 |
agen | -2.420 |
glitter | -2.409 |
interpol | -2.329 |
Token | . |
Token ID | 13 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
arcity | +1.488 |
Cage | +1.234 |
Hardware | +1.222 |
Sisters | +1.196 |
τ | +1.174 |
Neg logit contributions | |
anka | -1.173 |
plom | -1.134 |
agen | -1.110 |
glitter | -1.100 |
interpol | -1.059 |
Token | m |
Token ID | 76 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
arcity | +1.528 |
Cage | +1.254 |
Hardware | +1.239 |
Sisters | +1.211 |
τ | +1.187 |
Neg logit contributions | |
anka | -1.388 |
plom | -1.348 |
agen | -1.318 |
glitter | -1.310 |
interpol | -1.264 |
Token | .: |
Token ID | 11207 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
arcity | +0.786 |
Cage | +0.651 |
Hardware | +0.641 |
Sisters | +0.629 |
τ | +0.615 |
Neg logit contributions | |
anka | -0.714 |
plom | -0.693 |
glitter | -0.680 |
agen | -0.671 |
interpol | -0.653 |
Token | Volume |
Token ID | 31715 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
arcity | +1.696 |
Cage | +1.394 |
Hardware | +1.368 |
Sisters | +1.342 |
τ | +1.317 |
Neg logit contributions | |
anka | -1.496 |
plom | -1.447 |
agen | -1.427 |
glitter | -1.412 |
interpol | -1.368 |
Token | () |
Token ID | 3419 |
Feature activation | +9.74e-02 |
Pos logit contributions | |
arcity | +2.875 |
Cage | +2.343 |
Hardware | +2.299 |
Sisters | +2.257 |
τ | +2.206 |
Neg logit contributions | |
anka | -2.637 |
plom | -2.549 |
agen | -2.516 |
glitter | -2.496 |
interpol | -2.425 |
Token | ; |
Token ID | 2162 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
arcity | +0.499 |
Cage | +0.412 |
Hardware | +0.406 |
Sisters | +0.395 |
τ | +0.389 |
Neg logit contributions | |
anka | -0.422 |
plom | -0.408 |
glitter | -0.403 |
agen | -0.403 |
interpol | -0.392 |
Token | return |
Token ID | 1441 |
Feature activation | +5.21e-02 |
Pos logit contributions | |
arcity | +1.825 |
Hardware | +1.462 |
Cage | +1.460 |
Sisters | +1.411 |
τ | +1.395 |
Neg logit contributions | |
anka | -1.699 |
plom | -1.646 |
glitter | -1.629 |
agen | -1.619 |
interpol | -1.579 |
Token | f |
Token ID | 277 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
arcity | +0.264 |
Hardware | +0.214 |
Cage | +0.212 |
Sisters | +0.207 |
τ | +0.205 |
Neg logit contributions | |
anka | -0.229 |
glitter | -0.222 |
plom | -0.221 |
agen | -0.219 |
interpol | -0.216 |
Token | Mass |
Token ID | 20273 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
arcity | +2.077 |
Cage | +1.700 |
Hardware | +1.693 |
Sisters | +1.654 |
τ | +1.622 |
Neg logit contributions | |
anka | -1.874 |
plom | -1.826 |
agen | -1.771 |
glitter | -1.767 |
interpol | -1.703 |
Token | ; |
Token ID | 2162 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
arcity | +4.324 |
Cage | +3.576 |
Hardware | +3.546 |
Sisters | +3.465 |
τ | +3.420 |
Neg logit contributions | |
anka | -3.845 |
plom | -3.746 |
agen | -3.624 |
glitter | -3.606 |
interpol | -3.454 |
Token | } |
Token ID | 1782 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
arcity | +1.617 |
Cage | +1.291 |
Hardware | +1.264 |
Sisters | +1.234 |
τ | +1.206 |
Neg logit contributions | |
anka | -1.978 |
plom | -1.924 |
agen | -1.875 |
glitter | -1.864 |
interpol | -1.800 |
Token |
|
Token ID | 198 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
arcity | +0.537 |
Cage | +0.426 |
Hardware | +0.425 |
Sisters | +0.411 |
τ | +0.411 |
Neg logit contributions | |
anka | -0.513 |
plom | -0.496 |
glitter | -0.495 |
agen | -0.491 |
interpol | -0.480 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
arcity | +1.310 |
Cage | +1.021 |
Hardware | +1.003 |
Sisters | +0.980 |
τ | +0.956 |
Neg logit contributions | |
anka | -1.603 |
plom | -1.561 |
agen | -1.536 |
glitter | -1.526 |
interpol | -1.482 |
Token | Ass |
Token ID | 8021 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
arcity | +0.658 |
Cage | +0.525 |
Sisters | +0.507 |
Hardware | +0.496 |
RI | +0.486 |
Neg logit contributions | |
anka | -0.707 |
plom | -0.685 |
agen | -0.678 |
glitter | -0.668 |
interpol | -0.652 |
Token | performance |
Token ID | 2854 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
arcity | +1.173 |
Hardware | +0.977 |
Cage | +0.976 |
Sisters | +0.949 |
τ | +0.932 |
Neg logit contributions | |
anka | -0.864 |
plom | -0.833 |
glitter | -0.824 |
agen | -0.818 |
interpol | -0.791 |
Token | , |
Token ID | 11 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
anka | +0.545 |
plom | +0.523 |
agen | +0.517 |
glitter | +0.515 |
interpol | +0.496 |
Neg logit contributions | |
arcity | -0.749 |
Cage | -0.627 |
Hardware | -0.622 |
Sisters | -0.606 |
τ | -0.596 |
Token | would |
Token ID | 561 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
arcity | +1.034 |
Hardware | +0.865 |
Cage | +0.862 |
Sisters | +0.843 |
τ | +0.831 |
Neg logit contributions | |
anka | -0.686 |
plom | -0.663 |
glitter | -0.651 |
agen | -0.645 |
interpol | -0.623 |
Token | give |
Token ID | 1577 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
arcity | +0.629 |
Cage | +0.525 |
Hardware | +0.525 |
Sisters | +0.511 |
τ | +0.503 |
Neg logit contributions | |
anka | -0.472 |
plom | -0.462 |
glitter | -0.456 |
agen | -0.452 |
interpol | -0.438 |
Token | them |
Token ID | 606 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
arcity | +1.318 |
Cage | +1.065 |
Hardware | +1.064 |
Sisters | +1.035 |
τ | +1.015 |
Neg logit contributions | |
anka | -1.278 |
plom | -1.244 |
glitter | -1.220 |
agen | -1.219 |
interpol | -1.177 |
Token | enough |
Token ID | 1576 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
arcity | +0.845 |
Hardware | +0.702 |
Cage | +0.698 |
Sisters | +0.681 |
τ | +0.671 |
Neg logit contributions | |
anka | -0.649 |
plom | -0.629 |
glitter | -0.623 |
agen | -0.616 |
interpol | -0.597 |
Token | space |
Token ID | 2272 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
arcity | +4.807 |
Cage | +4.075 |
Hardware | +3.995 |
Sisters | +3.924 |
τ | +3.856 |
Neg logit contributions | |
anka | -3.376 |
plom | -3.278 |
agen | -3.174 |
glitter | -3.105 |
interpol | -2.991 |
Token | to |
Token ID | 284 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
arcity | +1.077 |
Cage | +0.876 |
Hardware | +0.863 |
Sisters | +0.844 |
τ | +0.827 |
Neg logit contributions | |
anka | -1.075 |
plom | -1.046 |
agen | -1.021 |
glitter | -1.016 |
interpol | -0.983 |
Token | sign |
Token ID | 1051 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
arcity | +1.203 |
Cage | +0.993 |
Hardware | +0.990 |
Sisters | +0.967 |
τ | +0.951 |
Neg logit contributions | |
anka | -0.951 |
plom | -0.926 |
glitter | -0.908 |
agen | -0.903 |
interpol | -0.874 |
Token | at |
Token ID | 379 |
Feature activation | -6.06e-06 |
Pos logit contributions | |
arcity | +0.456 |
Hardware | +0.377 |
Cage | +0.370 |
Sisters | +0.366 |
Drive | +0.360 |
Neg logit contributions | |
anka | -0.339 |
glitter | -0.330 |
plom | -0.328 |
agen | -0.323 |
interpol | -0.315 |
Token | least |
Token ID | 1551 |
Feature activation | +5.03e-01 |
Token | being |
Token ID | 852 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
anka | +0.531 |
glitter | +0.517 |
plom | +0.517 |
agen | +0.499 |
interpol | +0.490 |
Neg logit contributions | |
arcity | -0.762 |
Hardware | -0.636 |
Cage | -0.628 |
Sisters | -0.606 |
Drive | -0.606 |
Token | warned |
Token ID | 7728 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
arcity | +0.232 |
Hardware | +0.187 |
Cage | +0.187 |
Sisters | +0.180 |
τ | +0.179 |
Neg logit contributions | |
anka | -0.226 |
plom | -0.220 |
glitter | -0.217 |
agen | -0.214 |
interpol | -0.209 |
Token | to |
Token ID | 284 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
anka | +0.644 |
glitter | +0.619 |
plom | +0.618 |
agen | +0.617 |
interpol | +0.583 |
Neg logit contributions | |
arcity | -0.818 |
Hardware | -0.680 |
Cage | -0.665 |
Sisters | -0.649 |
τ | -0.644 |
Token | keep |
Token ID | 1394 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
arcity | +1.471 |
Hardware | +1.219 |
Cage | +1.212 |
Sisters | +1.184 |
τ | +1.164 |
Neg logit contributions | |
anka | -1.120 |
plom | -1.096 |
glitter | -1.086 |
agen | -1.073 |
interpol | -1.038 |
Token | an |
Token ID | 281 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
arcity | +0.505 |
Hardware | +0.417 |
Cage | +0.406 |
Sisters | +0.396 |
τ | +0.393 |
Neg logit contributions | |
anka | -0.398 |
glitter | -0.393 |
plom | -0.382 |
agen | -0.375 |
interpol | -0.369 |
Token | eye |
Token ID | 4151 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
arcity | +1.399 |
Hardware | +1.215 |
Cage | +1.210 |
Sisters | +1.181 |
τ | +1.178 |
Neg logit contributions | |
anka | -0.550 |
plom | -0.526 |
glitter | -0.512 |
agen | -0.510 |
interpol | -0.479 |
Token | out |
Token ID | 503 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
arcity | +4.244 |
Cage | +3.615 |
Sisters | +3.507 |
Hardware | +3.468 |
τ | +3.354 |
Neg logit contributions | |
anka | -3.763 |
plom | -3.673 |
agen | -3.580 |
glitter | -3.438 |
interpol | -3.409 |
Token | for |
Token ID | 329 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
arcity | +0.679 |
Cage | +0.564 |
Hardware | +0.559 |
Sisters | +0.546 |
τ | +0.537 |
Neg logit contributions | |
anka | -0.538 |
plom | -0.521 |
agen | -0.508 |
glitter | -0.507 |
interpol | -0.486 |
Token | two |
Token ID | 734 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
arcity | +1.369 |
Hardware | +1.129 |
Cage | +1.124 |
Sisters | +1.088 |
τ | +1.085 |
Neg logit contributions | |
anka | -1.059 |
plom | -1.016 |
glitter | -1.014 |
agen | -1.000 |
interpol | -0.962 |
Token | scams |
Token ID | 41718 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
arcity | +0.342 |
Hardware | +0.284 |
Cage | +0.281 |
τ | +0.274 |
Drive | +0.269 |
Neg logit contributions | |
anka | -0.243 |
glitter | -0.236 |
plom | -0.231 |
agen | -0.228 |
interpol | -0.222 |
Token | working |
Token ID | 1762 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
arcity | +1.322 |
Cage | +1.092 |
Hardware | +1.091 |
Sisters | +1.056 |
τ | +1.044 |
Neg logit contributions | |
anka | -1.030 |
plom | -0.996 |
glitter | -0.975 |
agen | -0.973 |
interpol | -0.935 |
Token | - |
Token ID | 12 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
anka | +2.732 |
plom | +2.570 |
agen | +2.537 |
glitter | +2.535 |
interpol | +2.445 |
Neg logit contributions | |
arcity | -5.279 |
Hardware | -4.430 |
Cage | -4.424 |
τ | -4.307 |
Sisters | -4.291 |
Token | and |
Token ID | 392 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
anka | +2.595 |
plom | +2.586 |
agen | +2.488 |
glitter | +2.435 |
interpol | +2.332 |
Neg logit contributions | |
arcity | -3.650 |
Cage | -3.071 |
Hardware | -2.978 |
Sisters | -2.893 |
Drive | -2.849 |
Token | - |
Token ID | 12 |
Feature activation | -8.24e-02 |
Pos logit contributions | |
anka | +1.645 |
plom | +1.584 |
agen | +1.548 |
glitter | +1.529 |
interpol | +1.468 |
Neg logit contributions | |
arcity | -2.827 |
Cage | -2.390 |
Hardware | -2.380 |
τ | -2.305 |
Sisters | -2.304 |
Token | soc |
Token ID | 35634 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
anka | +0.429 |
plom | +0.420 |
agen | +0.413 |
glitter | +0.407 |
interpol | +0.395 |
Neg logit contributions | |
arcity | -0.351 |
Cage | -0.276 |
Hardware | -0.274 |
Sisters | -0.264 |
τ | -0.256 |
Token | iety |
Token ID | 1905 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
anka | +0.538 |
plom | +0.525 |
agen | +0.479 |
glitter | +0.475 |
interpol | +0.432 |
Neg logit contributions | |
arcity | -1.385 |
Cage | -1.204 |
Hardware | -1.197 |
Sisters | -1.178 |
τ | -1.176 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
anka | +2.567 |
plom | +2.510 |
agen | +2.442 |
glitter | +2.425 |
interpol | +2.328 |
Neg logit contributions | |
arcity | -3.130 |
Cage | -2.578 |
Hardware | -2.544 |
Sisters | -2.499 |
τ | -2.448 |
Token | community |
Token ID | 2055 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
anka | +4.382 |
glitter | +4.286 |
mer | +4.105 |
plom | +4.045 |
agen | +4.034 |
Neg logit contributions | |
arcity | -6.278 |
Cage | -5.242 |
Hardware | -5.219 |
Drive | -5.083 |
τ | -4.999 |
Token | - |
Token ID | 12 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
anka | +2.832 |
plom | +2.696 |
agen | +2.652 |
glitter | +2.633 |
interpol | +2.541 |
Neg logit contributions | |
arcity | -5.248 |
Cage | -4.414 |
Hardware | -4.394 |
τ | -4.299 |
Sisters | -4.289 |
Token | organ |
Token ID | 9971 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
anka | +2.295 |
plom | +2.260 |
agen | +2.191 |
glitter | +2.163 |
interpol | +2.066 |
Neg logit contributions | |
arcity | -2.921 |
Cage | -2.446 |
Hardware | -2.363 |
Sisters | -2.343 |
Drive | -2.281 |
Token | isations |
Token ID | 38189 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
anka | +1.550 |
plom | +1.484 |
agen | +1.477 |
glitter | +1.475 |
interpol | +1.404 |
Neg logit contributions | |
arcity | -2.128 |
Cage | -1.773 |
Hardware | -1.759 |
Sisters | -1.715 |
τ | -1.693 |
Token | , |
Token ID | 11 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
anka | +1.143 |
plom | +1.111 |
agen | +1.086 |
glitter | +1.075 |
interpol | +1.034 |
Neg logit contributions | |
arcity | -1.429 |
Cage | -1.184 |
Hardware | -1.175 |
Sisters | -1.149 |
τ | -1.128 |
Token | marine |
Token ID | 16050 |
Feature activation | -6.42e-01 |
Pos logit contributions | |
glitter | +3.864 |
anka | +3.805 |
plom | +3.715 |
interpol | +3.689 |
agen | +3.676 |
Neg logit contributions | |
arcity | -4.742 |
Hardware | -3.853 |
Cage | -3.814 |
τ | -3.747 |
Drive | -3.733 |
Token | - |
Token ID | 12 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
anka | +2.340 |
plom | +2.265 |
agen | +2.248 |
glitter | +2.195 |
interpol | +2.091 |
Neg logit contributions | |
arcity | -3.668 |
Cage | -3.143 |
Hardware | -3.089 |
Sisters | -3.078 |
τ | -3.014 |
Token | biology |
Token ID | 43592 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
anka | +1.942 |
plom | +1.898 |
agen | +1.894 |
glitter | +1.802 |
interpol | +1.709 |
Neg logit contributions | |
arcity | -2.920 |
Cage | -2.486 |
Sisters | -2.461 |
Hardware | -2.446 |
Drive | -2.378 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
anka | +2.888 |
agen | +2.789 |
plom | +2.789 |
glitter | +2.708 |
interpol | +2.613 |
Neg logit contributions | |
arcity | -4.353 |
Hardware | -3.599 |
Cage | -3.571 |
Sisters | -3.496 |
τ | -3.470 |
Token | conservation |
Token ID | 14903 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
glitter | +4.317 |
anka | +4.288 |
plom | +4.200 |
interpol | +4.121 |
mer | +4.120 |
Neg logit contributions | |
arcity | -5.623 |
Hardware | -4.629 |
Cage | -4.521 |
Drive | -4.481 |
τ | -4.449 |
Token | , |
Token ID | 11 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
anka | +0.717 |
plom | +0.693 |
agen | +0.682 |
glitter | +0.675 |
interpol | +0.652 |
Neg logit contributions | |
arcity | -0.856 |
Cage | -0.708 |
Hardware | -0.703 |
Sisters | -0.686 |
τ | -0.676 |
Token | environment |
Token ID | 2858 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
glitter | +4.444 |
anka | +4.384 |
plom | +4.281 |
mer | +4.237 |
interpol | +4.223 |
Neg logit contributions | |
arcity | -5.863 |
Hardware | -4.847 |
Cage | -4.720 |
Drive | -4.702 |
τ | -4.654 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
anka | +2.107 |
plom | +2.027 |
agen | +2.014 |
glitter | +1.982 |
interpol | +1.917 |
Neg logit contributions | |
arcity | -2.807 |
Cage | -2.317 |
Hardware | -2.314 |
Sisters | -2.242 |
τ | -2.231 |
Token | oceans |
Token ID | 23744 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
glitter | +4.345 |
anka | +4.316 |
plom | +4.208 |
mer | +4.136 |
interpol | +4.125 |
Neg logit contributions | |
arcity | -5.721 |
Hardware | -4.745 |
Drive | -4.601 |
Cage | -4.601 |
τ | -4.539 |
Token | - |
Token ID | 12 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
anka | +2.166 |
plom | +2.096 |
agen | +2.045 |
glitter | +2.032 |
interpol | +1.950 |
Neg logit contributions | |
arcity | -2.939 |
Cage | -2.459 |
Hardware | -2.435 |
Sisters | -2.388 |
τ | -2.344 |
Token | and |
Token ID | 392 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
anka | +2.233 |
plom | +2.174 |
agen | +2.120 |
glitter | +2.079 |
interpol | +1.994 |
Neg logit contributions | |
arcity | -2.737 |
Hardware | -2.291 |
Cage | -2.288 |
Sisters | -2.231 |
τ | -2.157 |
Token | - |
Token ID | 12 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
anka | +2.322 |
agen | +2.254 |
plom | +2.236 |
glitter | +2.147 |
interpol | +2.043 |
Neg logit contributions | |
arcity | -4.134 |
Cage | -3.473 |
Hardware | -3.452 |
Sisters | -3.355 |
τ | -3.347 |
Token | and |
Token ID | 392 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
anka | +1.871 |
plom | +1.828 |
agen | +1.784 |
glitter | +1.766 |
interpol | +1.700 |
Neg logit contributions | |
arcity | -2.143 |
Cage | -1.763 |
Hardware | -1.733 |
Sisters | -1.702 |
τ | -1.659 |
Token | - |
Token ID | 12 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
anka | +1.062 |
plom | +1.023 |
agen | +1.001 |
glitter | +0.987 |
interpol | +0.942 |
Neg logit contributions | |
arcity | -1.647 |
Cage | -1.395 |
Hardware | -1.374 |
Sisters | -1.349 |
τ | -1.333 |
Token | f |
Token ID | 69 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
anka | +0.951 |
plom | +0.934 |
agen | +0.920 |
glitter | +0.909 |
interpol | +0.880 |
Neg logit contributions | |
arcity | -0.642 |
Cage | -0.493 |
Hardware | -0.487 |
Sisters | -0.475 |
τ | -0.451 |
Token | inance |
Token ID | 14149 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
anka | +0.182 |
plom | +0.177 |
agen | +0.173 |
glitter | +0.171 |
interpol | +0.164 |
Neg logit contributions | |
arcity | -0.213 |
Cage | -0.177 |
Hardware | -0.175 |
Sisters | -0.171 |
τ | -0.168 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
anka | +2.046 |
plom | +1.944 |
agen | +1.921 |
glitter | +1.899 |
interpol | +1.827 |
Neg logit contributions | |
arcity | -2.814 |
Hardware | -2.297 |
Cage | -2.288 |
τ | -2.217 |
Sisters | -2.214 |
Token | aust |
Token ID | 38132 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
anka | +4.465 |
glitter | +4.376 |
plom | +4.291 |
mer | +4.214 |
agen | +4.184 |
Neg logit contributions | |
arcity | -5.832 |
Hardware | -4.708 |
Cage | -4.679 |
Drive | -4.616 |
τ | -4.601 |
Token | ral |
Token ID | 1373 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
anka | +1.214 |
plom | +1.151 |
agen | +1.113 |
glitter | +1.084 |
interpol | +1.027 |
Neg logit contributions | |
arcity | -2.063 |
Cage | -1.796 |
Hardware | -1.796 |
Sisters | -1.747 |
τ | -1.736 |
Token | ia |
Token ID | 544 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
arcity | +0.177 |
Cage | +0.136 |
Hardware | +0.133 |
Sisters | +0.129 |
τ | +0.125 |
Neg logit contributions | |
anka | -0.288 |
plom | -0.282 |
agen | -0.277 |
glitter | -0.272 |
interpol | -0.264 |
Token |
|
Token ID | 198 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
anka | +0.923 |
plom | +0.898 |
agen | +0.881 |
glitter | +0.878 |
interpol | +0.853 |
Neg logit contributions | |
arcity | -0.764 |
Cage | -0.607 |
Hardware | -0.599 |
Sisters | -0.583 |
τ | -0.568 |
Token |
|
Token ID | 198 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
anka | +2.877 |
plom | +2.769 |
agen | +2.757 |
glitter | +2.711 |
mer | +2.658 |
Neg logit contributions | |
arcity | -3.314 |
conduc | -2.654 |
Cage | -2.530 |
Hardware | -2.473 |
Instr | -2.458 |
Token | : |
Token ID | 25 |
Feature activation | -8.82e-01 |
Pos logit contributions | |
arcity | +2.800 |
Cage | +2.200 |
Hardware | +2.160 |
Sisters | +2.106 |
τ | +2.021 |
Neg logit contributions | |
anka | -3.956 |
plom | -3.879 |
agen | -3.787 |
glitter | -3.775 |
interpol | -3.679 |
Token | beef |
Token ID | 12023 |
Feature activation | -7.63e-01 |
Pos logit contributions | |
anka | +4.293 |
glitter | +4.230 |
plom | +4.143 |
agen | +4.125 |
interpol | +4.049 |
Neg logit contributions | |
arcity | -3.935 |
Hardware | -3.057 |
Cage | -3.041 |
τ | -3.002 |
Drive | -2.966 |
Token | - |
Token ID | 12 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
anka | +3.030 |
plom | +2.911 |
agen | +2.896 |
glitter | +2.794 |
interpol | +2.689 |
Neg logit contributions | |
arcity | -4.139 |
Cage | -3.480 |
Hardware | -3.444 |
Sisters | -3.337 |
τ | -3.306 |
Token | c |
Token ID | 66 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
anka | +2.180 |
plom | +2.135 |
agen | +2.128 |
glitter | +2.048 |
interpol | +1.962 |
Neg logit contributions | |
arcity | -2.724 |
Cage | -2.251 |
Sisters | -2.236 |
Hardware | -2.212 |
τ | -2.149 |
Token | attle |
Token ID | 1999 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
anka | +0.962 |
plom | +0.935 |
agen | +0.932 |
glitter | +0.858 |
interpol | +0.834 |
Neg logit contributions | |
arcity | -1.282 |
Cage | -1.108 |
Hardware | -1.107 |
Sisters | -1.081 |
τ | -1.068 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
anka | +1.957 |
plom | +1.898 |
agen | +1.866 |
glitter | +1.804 |
interpol | +1.753 |
Neg logit contributions | |
arcity | -2.422 |
Cage | -2.010 |
Hardware | -1.996 |
Sisters | -1.964 |
τ | -1.937 |
Token | rural |
Token ID | 10016 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
anka | +4.936 |
glitter | +4.848 |
plom | +4.743 |
agen | +4.684 |
mer | +4.667 |
Neg logit contributions | |
arcity | -5.365 |
Hardware | -4.257 |
Cage | -4.205 |
τ | -4.179 |
Drive | -4.162 |
Token | , |
Token ID | 11 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
anka | +2.337 |
plom | +2.257 |
agen | +2.209 |
glitter | +2.180 |
interpol | +2.094 |
Neg logit contributions | |
arcity | -3.211 |
Cage | -2.678 |
Hardware | -2.660 |
Sisters | -2.605 |
τ | -2.566 |
Token | business |
Token ID | 1597 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
anka | +4.108 |
glitter | +4.028 |
plom | +3.925 |
agen | +3.852 |
diplomacy | +3.839 |
Neg logit contributions | |
arcity | -5.397 |
Hardware | -4.398 |
Cage | -4.329 |
Drive | -4.289 |
τ | -4.283 |
Token | - |
Token ID | 12 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
anka | +1.782 |
plom | +1.701 |
agen | +1.661 |
glitter | +1.634 |
interpol | +1.573 |
Neg logit contributions | |
arcity | -2.692 |
Cage | -2.269 |
Hardware | -2.243 |
Sisters | -2.192 |
τ | -2.154 |
Token | econom |
Token ID | 13926 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
plom | +1.600 |
anka | +1.586 |
agen | +1.557 |
glitter | +1.468 |
interpol | +1.368 |
Neg logit contributions | |
arcity | -3.758 |
Cage | -3.241 |
Sisters | -3.184 |
Hardware | -3.139 |
Drive | -3.059 |
Token | biology |
Token ID | 43592 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
anka | +1.942 |
plom | +1.898 |
agen | +1.894 |
glitter | +1.802 |
interpol | +1.709 |
Neg logit contributions | |
arcity | -2.920 |
Cage | -2.486 |
Sisters | -2.461 |
Hardware | -2.446 |
Drive | -2.378 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
anka | +2.888 |
agen | +2.789 |
plom | +2.789 |
glitter | +2.708 |
interpol | +2.613 |
Neg logit contributions | |
arcity | -4.353 |
Hardware | -3.599 |
Cage | -3.571 |
Sisters | -3.496 |
τ | -3.470 |
Token | conservation |
Token ID | 14903 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
glitter | +4.317 |
anka | +4.288 |
plom | +4.200 |
interpol | +4.121 |
mer | +4.120 |
Neg logit contributions | |
arcity | -5.623 |
Hardware | -4.629 |
Cage | -4.521 |
Drive | -4.481 |
τ | -4.449 |
Token | , |
Token ID | 11 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
anka | +0.717 |
plom | +0.693 |
agen | +0.682 |
glitter | +0.675 |
interpol | +0.652 |
Neg logit contributions | |
arcity | -0.856 |
Cage | -0.708 |
Hardware | -0.703 |
Sisters | -0.686 |
τ | -0.676 |
Token | environment |
Token ID | 2858 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
glitter | +4.444 |
anka | +4.384 |
plom | +4.281 |
mer | +4.237 |
interpol | +4.223 |
Neg logit contributions | |
arcity | -5.863 |
Hardware | -4.847 |
Cage | -4.720 |
Drive | -4.702 |
τ | -4.654 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
anka | +2.107 |
plom | +2.027 |
agen | +2.014 |
glitter | +1.982 |
interpol | +1.917 |
Neg logit contributions | |
arcity | -2.807 |
Cage | -2.317 |
Hardware | -2.314 |
Sisters | -2.242 |
τ | -2.231 |
Token | oceans |
Token ID | 23744 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
glitter | +4.345 |
anka | +4.316 |
plom | +4.208 |
mer | +4.136 |
interpol | +4.125 |
Neg logit contributions | |
arcity | -5.721 |
Hardware | -4.745 |
Drive | -4.601 |
Cage | -4.601 |
τ | -4.539 |
Token | - |
Token ID | 12 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
anka | +2.166 |
plom | +2.096 |
agen | +2.045 |
glitter | +2.032 |
interpol | +1.950 |
Neg logit contributions | |
arcity | -2.939 |
Cage | -2.459 |
Hardware | -2.435 |
Sisters | -2.388 |
τ | -2.344 |
Token | and |
Token ID | 392 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
anka | +2.233 |
plom | +2.174 |
agen | +2.120 |
glitter | +2.079 |
interpol | +1.994 |
Neg logit contributions | |
arcity | -2.737 |
Hardware | -2.291 |
Cage | -2.288 |
Sisters | -2.231 |
τ | -2.157 |
Token | - |
Token ID | 12 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
anka | +0.897 |
plom | +0.864 |
agen | +0.845 |
glitter | +0.833 |
interpol | +0.797 |
Neg logit contributions | |
arcity | -1.405 |
Cage | -1.194 |
Hardware | -1.187 |
Sisters | -1.157 |
τ | -1.148 |
Token | ree |
Token ID | 631 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
anka | +0.191 |
plom | +0.182 |
agen | +0.175 |
glitter | +0.171 |
interpol | +0.160 |
Neg logit contributions | |
arcity | -0.403 |
Cage | -0.350 |
Hardware | -0.350 |
Sisters | -0.342 |
τ | -0.335 |
Token | Topics |
Token ID | 25902 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
arcity | +0.737 |
Cage | +0.592 |
Hardware | +0.577 |
Sisters | +0.571 |
τ | +0.552 |
Neg logit contributions | |
anka | -0.758 |
plom | -0.737 |
agen | -0.723 |
glitter | -0.716 |
interpol | -0.691 |
Token | : |
Token ID | 25 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
arcity | +2.615 |
Cage | +2.055 |
Hardware | +2.020 |
Sisters | +1.968 |
τ | +1.895 |
Neg logit contributions | |
anka | -3.651 |
plom | -3.578 |
agen | -3.498 |
glitter | -3.485 |
interpol | -3.395 |
Token | marine |
Token ID | 16050 |
Feature activation | -6.42e-01 |
Pos logit contributions | |
glitter | +3.864 |
anka | +3.805 |
plom | +3.715 |
interpol | +3.689 |
agen | +3.676 |
Neg logit contributions | |
arcity | -4.742 |
Hardware | -3.853 |
Cage | -3.814 |
τ | -3.747 |
Drive | -3.733 |
Token | - |
Token ID | 12 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
anka | +2.340 |
plom | +2.265 |
agen | +2.248 |
glitter | +2.195 |
interpol | +2.091 |
Neg logit contributions | |
arcity | -3.668 |
Cage | -3.143 |
Hardware | -3.089 |
Sisters | -3.078 |
τ | -3.014 |
Token | biology |
Token ID | 43592 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
anka | +1.942 |
plom | +1.898 |
agen | +1.894 |
glitter | +1.802 |
interpol | +1.709 |
Neg logit contributions | |
arcity | -2.920 |
Cage | -2.486 |
Sisters | -2.461 |
Hardware | -2.446 |
Drive | -2.378 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
anka | +2.888 |
agen | +2.789 |
plom | +2.789 |
glitter | +2.708 |
interpol | +2.613 |
Neg logit contributions | |
arcity | -4.353 |
Hardware | -3.599 |
Cage | -3.571 |
Sisters | -3.496 |
τ | -3.470 |
Token | conservation |
Token ID | 14903 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
glitter | +4.317 |
anka | +4.288 |
plom | +4.200 |
interpol | +4.121 |
mer | +4.120 |
Neg logit contributions | |
arcity | -5.623 |
Hardware | -4.629 |
Cage | -4.521 |
Drive | -4.481 |
τ | -4.449 |
Token | , |
Token ID | 11 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
anka | +0.717 |
plom | +0.693 |
agen | +0.682 |
glitter | +0.675 |
interpol | +0.652 |
Neg logit contributions | |
arcity | -0.856 |
Cage | -0.708 |
Hardware | -0.703 |
Sisters | -0.686 |
τ | -0.676 |
Token | environment |
Token ID | 2858 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
glitter | +4.444 |
anka | +4.384 |
plom | +4.281 |
mer | +4.237 |
interpol | +4.223 |
Neg logit contributions | |
arcity | -5.863 |
Hardware | -4.847 |
Cage | -4.720 |
Drive | -4.702 |
τ | -4.654 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
anka | +2.107 |
plom | +2.027 |
agen | +2.014 |
glitter | +1.982 |
interpol | +1.917 |
Neg logit contributions | |
arcity | -2.807 |
Cage | -2.317 |
Hardware | -2.314 |
Sisters | -2.242 |
τ | -2.231 |
Token | oceans |
Token ID | 23744 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
glitter | +4.345 |
anka | +4.316 |
plom | +4.208 |
mer | +4.136 |
interpol | +4.125 |
Neg logit contributions | |
arcity | -5.721 |
Hardware | -4.745 |
Drive | -4.601 |
Cage | -4.601 |
τ | -4.539 |
Token | - |
Token ID | 12 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
anka | +2.166 |
plom | +2.096 |
agen | +2.045 |
glitter | +2.032 |
interpol | +1.950 |
Neg logit contributions | |
arcity | -2.939 |
Cage | -2.459 |
Hardware | -2.435 |
Sisters | -2.388 |
τ | -2.344 |
Token | and |
Token ID | 392 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
anka | +2.233 |
plom | +2.174 |
agen | +2.120 |
glitter | +2.079 |
interpol | +1.994 |
Neg logit contributions | |
arcity | -2.737 |
Hardware | -2.291 |
Cage | -2.288 |
Sisters | -2.231 |
τ | -2.157 |
Token | - |
Token ID | 12 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
anka | +0.897 |
plom | +0.864 |
agen | +0.845 |
glitter | +0.833 |
interpol | +0.797 |
Neg logit contributions | |
arcity | -1.405 |
Cage | -1.194 |
Hardware | -1.187 |
Sisters | -1.157 |
τ | -1.148 |
Token | ree |
Token ID | 631 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
anka | +0.191 |
plom | +0.182 |
agen | +0.175 |
glitter | +0.171 |
interpol | +0.160 |
Neg logit contributions | |
arcity | -0.403 |
Cage | -0.350 |
Hardware | -0.350 |
Sisters | -0.342 |
τ | -0.335 |
Token | fs |
Token ID | 9501 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
anka | +1.024 |
plom | +1.011 |
agen | +0.997 |
glitter | +0.975 |
interpol | +0.948 |
Neg logit contributions | |
arcity | -0.412 |
Cage | -0.274 |
Hardware | -0.267 |
Sisters | -0.267 |
τ | -0.241 |
Token | , |
Token ID | 11 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
anka | +0.050 |
plom | +0.049 |
glitter | +0.048 |
agen | +0.048 |
interpol | +0.046 |
Neg logit contributions | |
arcity | -0.054 |
Cage | -0.044 |
Hardware | -0.043 |
Sisters | -0.042 |
τ | -0.042 |
Token | water |
Token ID | 1660 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
anka | +3.965 |
glitter | +3.960 |
plom | +3.894 |
mer | +3.776 |
interpol | +3.761 |
Neg logit contributions | |
arcity | -5.844 |
Hardware | -4.889 |
Cage | -4.749 |
Drive | -4.735 |
τ | -4.672 |
Token | - |
Token ID | 12 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
anka | +1.051 |
plom | +1.013 |
agen | +1.004 |
glitter | +0.988 |
interpol | +0.955 |
Neg logit contributions | |
arcity | -1.360 |
Cage | -1.158 |
Hardware | -1.139 |
Sisters | -1.107 |
τ | -1.093 |
Token | poll |
Token ID | 30393 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
anka | +1.452 |
agen | +1.423 |
plom | +1.419 |
glitter | +1.366 |
interpol | +1.316 |
Neg logit contributions | |
arcity | -2.087 |
Cage | -1.768 |
Hardware | -1.749 |
Sisters | -1.749 |
τ | -1.681 |
Token | ution |
Token ID | 1009 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
anka | +0.966 |
agen | +0.919 |
plom | +0.859 |
glitter | +0.782 |
interpol | +0.739 |
Neg logit contributions | |
arcity | -2.834 |
Hardware | -2.589 |
Cage | -2.543 |
τ | -2.521 |
Sisters | -2.519 |
Token | , |
Token ID | 11 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
anka | +0.944 |
plom | +0.918 |
agen | +0.893 |
glitter | +0.893 |
interpol | +0.861 |
Neg logit contributions | |
arcity | -1.005 |
Cage | -0.822 |
Hardware | -0.808 |
Sisters | -0.791 |
τ | -0.773 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
anka | +2.567 |
plom | +2.510 |
agen | +2.442 |
glitter | +2.425 |
interpol | +2.328 |
Neg logit contributions | |
arcity | -3.130 |
Cage | -2.578 |
Hardware | -2.544 |
Sisters | -2.499 |
τ | -2.448 |
Token | community |
Token ID | 2055 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
anka | +4.382 |
glitter | +4.286 |
mer | +4.105 |
plom | +4.045 |
agen | +4.034 |
Neg logit contributions | |
arcity | -6.278 |
Cage | -5.242 |
Hardware | -5.219 |
Drive | -5.083 |
τ | -4.999 |
Token | - |
Token ID | 12 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
anka | +2.832 |
plom | +2.696 |
agen | +2.652 |
glitter | +2.633 |
interpol | +2.541 |
Neg logit contributions | |
arcity | -5.248 |
Cage | -4.414 |
Hardware | -4.394 |
τ | -4.299 |
Sisters | -4.289 |
Token | organ |
Token ID | 9971 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
anka | +2.295 |
plom | +2.260 |
agen | +2.191 |
glitter | +2.163 |
interpol | +2.066 |
Neg logit contributions | |
arcity | -2.921 |
Cage | -2.446 |
Hardware | -2.363 |
Sisters | -2.343 |
Drive | -2.281 |
Token | isations |
Token ID | 38189 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
anka | +1.550 |
plom | +1.484 |
agen | +1.477 |
glitter | +1.475 |
interpol | +1.404 |
Neg logit contributions | |
arcity | -2.128 |
Cage | -1.773 |
Hardware | -1.759 |
Sisters | -1.715 |
τ | -1.693 |
Token | , |
Token ID | 11 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
anka | +1.143 |
plom | +1.111 |
agen | +1.086 |
glitter | +1.075 |
interpol | +1.034 |
Neg logit contributions | |
arcity | -1.429 |
Cage | -1.184 |
Hardware | -1.175 |
Sisters | -1.149 |
τ | -1.128 |
Token | electricity |
Token ID | 8744 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
anka | +3.989 |
glitter | +3.921 |
mer | +3.747 |
plom | +3.704 |
agen | +3.667 |
Neg logit contributions | |
arcity | -5.944 |
Cage | -4.950 |
Hardware | -4.912 |
Drive | -4.796 |
τ | -4.734 |
Token | - |
Token ID | 12 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
anka | +1.103 |
plom | +1.057 |
agen | +1.046 |
glitter | +1.027 |
interpol | +0.983 |
Neg logit contributions | |
arcity | -1.645 |
Cage | -1.391 |
Hardware | -1.375 |
Sisters | -1.352 |
τ | -1.324 |
Token | energy |
Token ID | 22554 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
anka | +0.808 |
agen | +0.787 |
plom | +0.781 |
glitter | +0.770 |
interpol | +0.742 |
Neg logit contributions | |
arcity | -0.674 |
Cage | -0.544 |
Sisters | -0.529 |
Hardware | -0.526 |
τ | -0.498 |
Token | - |
Token ID | 12 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
anka | +1.178 |
plom | +1.124 |
agen | +1.121 |
glitter | +1.092 |
interpol | +1.060 |
Neg logit contributions | |
arcity | -1.786 |
Cage | -1.521 |
Hardware | -1.492 |
Sisters | -1.472 |
τ | -1.456 |
Token | and |
Token ID | 392 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
anka | +0.087 |
plom | +0.084 |
agen | +0.084 |
glitter | +0.082 |
interpol | +0.079 |
Neg logit contributions | |
arcity | -0.101 |
Cage | -0.085 |
Hardware | -0.083 |
Sisters | -0.082 |
τ | -0.080 |
Token | - |
Token ID | 12 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
anka | +3.030 |
plom | +2.911 |
agen | +2.896 |
glitter | +2.794 |
interpol | +2.689 |
Neg logit contributions | |
arcity | -4.139 |
Cage | -3.480 |
Hardware | -3.444 |
Sisters | -3.337 |
τ | -3.306 |
Token | c |
Token ID | 66 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
anka | +2.180 |
plom | +2.135 |
agen | +2.128 |
glitter | +2.048 |
interpol | +1.962 |
Neg logit contributions | |
arcity | -2.724 |
Cage | -2.251 |
Sisters | -2.236 |
Hardware | -2.212 |
τ | -2.149 |
Token | attle |
Token ID | 1999 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
anka | +0.962 |
plom | +0.935 |
agen | +0.932 |
glitter | +0.858 |
interpol | +0.834 |
Neg logit contributions | |
arcity | -1.282 |
Cage | -1.108 |
Hardware | -1.107 |
Sisters | -1.081 |
τ | -1.068 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
anka | +1.957 |
plom | +1.898 |
agen | +1.866 |
glitter | +1.804 |
interpol | +1.753 |
Neg logit contributions | |
arcity | -2.422 |
Cage | -2.010 |
Hardware | -1.996 |
Sisters | -1.964 |
τ | -1.937 |
Token | rural |
Token ID | 10016 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
anka | +4.936 |
glitter | +4.848 |
plom | +4.743 |
agen | +4.684 |
mer | +4.667 |
Neg logit contributions | |
arcity | -5.365 |
Hardware | -4.257 |
Cage | -4.205 |
τ | -4.179 |
Drive | -4.162 |
Token | , |
Token ID | 11 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
anka | +2.337 |
plom | +2.257 |
agen | +2.209 |
glitter | +2.180 |
interpol | +2.094 |
Neg logit contributions | |
arcity | -3.211 |
Cage | -2.678 |
Hardware | -2.660 |
Sisters | -2.605 |
τ | -2.566 |
Token | business |
Token ID | 1597 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
anka | +4.108 |
glitter | +4.028 |
plom | +3.925 |
agen | +3.852 |
diplomacy | +3.839 |
Neg logit contributions | |
arcity | -5.397 |
Hardware | -4.398 |
Cage | -4.329 |
Drive | -4.289 |
τ | -4.283 |
Token | - |
Token ID | 12 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
anka | +1.782 |
plom | +1.701 |
agen | +1.661 |
glitter | +1.634 |
interpol | +1.573 |
Neg logit contributions | |
arcity | -2.692 |
Cage | -2.269 |
Hardware | -2.243 |
Sisters | -2.192 |
τ | -2.154 |
Token | econom |
Token ID | 13926 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
plom | +1.600 |
anka | +1.586 |
agen | +1.557 |
glitter | +1.468 |
interpol | +1.368 |
Neg logit contributions | |
arcity | -3.758 |
Cage | -3.241 |
Sisters | -3.184 |
Hardware | -3.139 |
Drive | -3.059 |
Token | ics |
Token ID | 873 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
anka | +2.438 |
plom | +2.375 |
agen | +2.361 |
glitter | +2.284 |
interpol | +2.195 |
Neg logit contributions | |
arcity | -2.730 |
Hardware | -2.215 |
Cage | -2.211 |
Sisters | -2.159 |
τ | -2.126 |
Token | - |
Token ID | 12 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
anka | +2.322 |
agen | +2.254 |
plom | +2.236 |
glitter | +2.147 |
interpol | +2.043 |
Neg logit contributions | |
arcity | -4.134 |
Cage | -3.473 |
Hardware | -3.452 |
Sisters | -3.355 |
τ | -3.347 |
Token | , |
Token ID | 11 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
anka | +0.050 |
plom | +0.049 |
glitter | +0.048 |
agen | +0.048 |
interpol | +0.046 |
Neg logit contributions | |
arcity | -0.054 |
Cage | -0.044 |
Hardware | -0.043 |
Sisters | -0.042 |
τ | -0.042 |
Token | water |
Token ID | 1660 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
anka | +3.965 |
glitter | +3.960 |
plom | +3.894 |
mer | +3.776 |
interpol | +3.761 |
Neg logit contributions | |
arcity | -5.844 |
Hardware | -4.889 |
Cage | -4.749 |
Drive | -4.735 |
τ | -4.672 |
Token | - |
Token ID | 12 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
anka | +1.051 |
plom | +1.013 |
agen | +1.004 |
glitter | +0.988 |
interpol | +0.955 |
Neg logit contributions | |
arcity | -1.360 |
Cage | -1.158 |
Hardware | -1.139 |
Sisters | -1.107 |
τ | -1.093 |
Token | poll |
Token ID | 30393 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
anka | +1.452 |
agen | +1.423 |
plom | +1.419 |
glitter | +1.366 |
interpol | +1.316 |
Neg logit contributions | |
arcity | -2.087 |
Cage | -1.768 |
Hardware | -1.749 |
Sisters | -1.749 |
τ | -1.681 |
Token | ution |
Token ID | 1009 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
anka | +0.966 |
agen | +0.919 |
plom | +0.859 |
glitter | +0.782 |
interpol | +0.739 |
Neg logit contributions | |
arcity | -2.834 |
Hardware | -2.589 |
Cage | -2.543 |
τ | -2.521 |
Sisters | -2.519 |
Token | , |
Token ID | 11 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
anka | +0.944 |
plom | +0.918 |
agen | +0.893 |
glitter | +0.893 |
interpol | +0.861 |
Neg logit contributions | |
arcity | -1.005 |
Cage | -0.822 |
Hardware | -0.808 |
Sisters | -0.791 |
τ | -0.773 |
Token | environmental |
Token ID | 6142 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
anka | +3.781 |
glitter | +3.771 |
plom | +3.692 |
mer | +3.581 |
interpol | +3.572 |
Neg logit contributions | |
arcity | -5.452 |
Hardware | -4.558 |
Cage | -4.438 |
Drive | -4.392 |
τ | -4.363 |
Token | - |
Token ID | 12 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
anka | +0.512 |
plom | +0.498 |
glitter | +0.482 |
agen | +0.477 |
interpol | +0.461 |
Neg logit contributions | |
arcity | -0.683 |
Cage | -0.567 |
Hardware | -0.561 |
Sisters | -0.553 |
τ | -0.540 |
Token | impact |
Token ID | 48240 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
anka | +1.279 |
plom | +1.245 |
agen | +1.219 |
glitter | +1.192 |
interpol | +1.155 |
Neg logit contributions | |
arcity | -1.611 |
Cage | -1.348 |
Hardware | -1.318 |
Sisters | -1.298 |
Drive | -1.268 |
Token | , |
Token ID | 11 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
anka | +0.821 |
plom | +0.800 |
glitter | +0.781 |
agen | +0.778 |
interpol | +0.753 |
Neg logit contributions | |
arcity | -0.855 |
Cage | -0.699 |
Hardware | -0.684 |
Sisters | -0.672 |
τ | -0.658 |
Token | environmental |
Token ID | 6142 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
anka | +3.653 |
glitter | +3.636 |
plom | +3.565 |
interpol | +3.449 |
mer | +3.449 |
Neg logit contributions | |
arcity | -5.300 |
Hardware | -4.434 |
Cage | -4.329 |
Drive | -4.266 |
τ | -4.231 |
Token | S |
Token ID | 50 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
anka | +0.939 |
plom | +0.912 |
agen | +0.903 |
glitter | +0.893 |
interpol | +0.869 |
Neg logit contributions | |
arcity | -0.864 |
Cage | -0.683 |
Hardware | -0.672 |
Sisters | -0.654 |
τ | -0.635 |
Token | + |
Token ID | 10 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
arcity | +1.573 |
Cage | +1.338 |
Hardware | +1.326 |
Sisters | +1.301 |
τ | +1.282 |
Neg logit contributions | |
anka | -0.950 |
plom | -0.917 |
agen | -0.887 |
glitter | -0.882 |
interpol | -0.842 |
Token | \ |
Token ID | 59 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
arcity | +0.848 |
Cage | +0.683 |
Hardware | +0.673 |
Sisters | +0.658 |
τ | +0.641 |
Neg logit contributions | |
anka | -0.892 |
plom | -0.868 |
agen | -0.851 |
glitter | -0.846 |
interpol | -0.820 |
Token | s |
Token ID | 82 |
Feature activation | -9.28e-02 |
Pos logit contributions | |
anka | +1.658 |
plom | +1.609 |
agen | +1.599 |
glitter | +1.580 |
interpol | +1.539 |
Neg logit contributions | |
arcity | -1.478 |
Cage | -1.151 |
Hardware | -1.141 |
Sisters | -1.103 |
τ | -1.071 |
Token | + |
Token ID | 10 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
anka | +0.321 |
plom | +0.309 |
agen | +0.302 |
glitter | +0.299 |
interpol | +0.285 |
Neg logit contributions | |
arcity | -0.557 |
Cage | -0.472 |
Hardware | -0.468 |
Sisters | -0.460 |
τ | -0.452 |
Token | "( |
Token ID | 18109 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
arcity | +0.473 |
Cage | +0.389 |
Hardware | +0.384 |
Sisters | +0.376 |
τ | +0.367 |
Neg logit contributions | |
anka | -0.413 |
plom | -0.400 |
agen | -0.392 |
glitter | -0.390 |
interpol | -0.376 |
Token | \ |
Token ID | 59 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
anka | +4.673 |
agen | +4.517 |
plom | +4.491 |
glitter | +4.407 |
interpol | +4.283 |
Neg logit contributions | |
arcity | -4.643 |
Cage | -3.661 |
Hardware | -3.565 |
Sisters | -3.548 |
RI | -3.391 |
Token | S |
Token ID | 50 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
anka | +1.264 |
plom | +1.228 |
agen | +1.219 |
glitter | +1.206 |
interpol | +1.173 |
Neg logit contributions | |
arcity | -1.097 |
Cage | -0.845 |
Hardware | -0.830 |
Sisters | -0.810 |
τ | -0.785 |
Token | +) |
Token ID | 28988 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
arcity | +1.125 |
Cage | +0.957 |
Hardware | +0.948 |
Sisters | +0.929 |
τ | +0.915 |
Neg logit contributions | |
anka | -0.697 |
plom | -0.674 |
agen | -0.651 |
glitter | -0.649 |
interpol | -0.621 |
Token | "] |
Token ID | 8973 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
anka | +0.764 |
plom | +0.737 |
agen | +0.727 |
glitter | +0.721 |
interpol | +0.699 |
Neg logit contributions | |
arcity | -0.868 |
Cage | -0.707 |
Hardware | -0.699 |
Sisters | -0.682 |
τ | -0.666 |
Token | ' |
Token ID | 6 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
anka | +2.361 |
plom | +2.275 |
agen | +2.226 |
glitter | +2.214 |
interpol | +2.141 |
Neg logit contributions | |
arcity | -3.181 |
Cage | -2.629 |
Hardware | -2.595 |
Sisters | -2.561 |
τ | -2.493 |
Token | impact |
Token ID | 48240 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
anka | +1.279 |
plom | +1.245 |
agen | +1.219 |
glitter | +1.192 |
interpol | +1.155 |
Neg logit contributions | |
arcity | -1.611 |
Cage | -1.348 |
Hardware | -1.318 |
Sisters | -1.298 |
Drive | -1.268 |
Token | , |
Token ID | 11 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
anka | +0.821 |
plom | +0.800 |
glitter | +0.781 |
agen | +0.778 |
interpol | +0.753 |
Neg logit contributions | |
arcity | -0.855 |
Cage | -0.699 |
Hardware | -0.684 |
Sisters | -0.672 |
τ | -0.658 |
Token | environmental |
Token ID | 6142 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
anka | +3.653 |
glitter | +3.636 |
plom | +3.565 |
interpol | +3.449 |
mer | +3.449 |
Neg logit contributions | |
arcity | -5.300 |
Hardware | -4.434 |
Cage | -4.329 |
Drive | -4.266 |
τ | -4.231 |
Token | - |
Token ID | 12 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
anka | +1.039 |
plom | +1.007 |
agen | +0.977 |
glitter | +0.974 |
interpol | +0.934 |
Neg logit contributions | |
arcity | -1.479 |
Cage | -1.240 |
Hardware | -1.228 |
Sisters | -1.205 |
τ | -1.183 |
Token | management |
Token ID | 27604 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
anka | +0.963 |
plom | +0.934 |
agen | +0.912 |
glitter | +0.898 |
interpol | +0.868 |
Neg logit contributions | |
arcity | -1.159 |
Cage | -0.959 |
Hardware | -0.945 |
Sisters | -0.925 |
τ | -0.903 |
Token | , |
Token ID | 11 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
anka | +0.626 |
plom | +0.607 |
agen | +0.595 |
glitter | +0.586 |
interpol | +0.566 |
Neg logit contributions | |
arcity | -0.754 |
Cage | -0.624 |
Hardware | -0.617 |
Sisters | -0.603 |
τ | -0.592 |
Token | environmental |
Token ID | 6142 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
anka | +3.679 |
glitter | +3.654 |
plom | +3.584 |
mer | +3.467 |
interpol | +3.463 |
Neg logit contributions | |
arcity | -5.438 |
Hardware | -4.564 |
Cage | -4.451 |
Drive | -4.401 |
τ | -4.364 |
Token | - |
Token ID | 12 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
anka | +0.867 |
plom | +0.843 |
glitter | +0.816 |
agen | +0.813 |
interpol | +0.782 |
Neg logit contributions | |
arcity | -1.201 |
Cage | -1.001 |
Hardware | -0.991 |
Sisters | -0.975 |
τ | -0.955 |
Token | technology |
Token ID | 45503 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
anka | +0.700 |
plom | +0.681 |
agen | +0.665 |
glitter | +0.654 |
interpol | +0.633 |
Neg logit contributions | |
arcity | -0.841 |
Cage | -0.695 |
Hardware | -0.685 |
Sisters | -0.670 |
τ | -0.655 |
Token | , |
Token ID | 11 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
anka | +2.152 |
plom | +2.088 |
agen | +2.045 |
glitter | +2.025 |
interpol | +1.952 |
Neg logit contributions | |
arcity | -2.566 |
Cage | -2.109 |
Hardware | -2.087 |
Sisters | -2.037 |
τ | -2.001 |
Token | q |
Token ID | 10662 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
anka | +3.481 |
glitter | +3.449 |
plom | +3.394 |
mer | +3.268 |
interpol | +3.267 |
Neg logit contributions | |
arcity | -5.233 |
Hardware | -4.375 |
Cage | -4.286 |
Drive | -4.225 |
τ | -4.195 |
Token | ution |
Token ID | 1009 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
anka | +0.966 |
agen | +0.919 |
plom | +0.859 |
glitter | +0.782 |
interpol | +0.739 |
Neg logit contributions | |
arcity | -2.834 |
Hardware | -2.589 |
Cage | -2.543 |
τ | -2.521 |
Sisters | -2.519 |
Token | , |
Token ID | 11 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
anka | +0.944 |
plom | +0.918 |
agen | +0.893 |
glitter | +0.893 |
interpol | +0.861 |
Neg logit contributions | |
arcity | -1.005 |
Cage | -0.822 |
Hardware | -0.808 |
Sisters | -0.791 |
τ | -0.773 |
Token | environmental |
Token ID | 6142 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
anka | +3.781 |
glitter | +3.771 |
plom | +3.692 |
mer | +3.581 |
interpol | +3.572 |
Neg logit contributions | |
arcity | -5.452 |
Hardware | -4.558 |
Cage | -4.438 |
Drive | -4.392 |
τ | -4.363 |
Token | - |
Token ID | 12 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
anka | +0.512 |
plom | +0.498 |
glitter | +0.482 |
agen | +0.477 |
interpol | +0.461 |
Neg logit contributions | |
arcity | -0.683 |
Cage | -0.567 |
Hardware | -0.561 |
Sisters | -0.553 |
τ | -0.540 |
Token | impact |
Token ID | 48240 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
anka | +1.279 |
plom | +1.245 |
agen | +1.219 |
glitter | +1.192 |
interpol | +1.155 |
Neg logit contributions | |
arcity | -1.611 |
Cage | -1.348 |
Hardware | -1.318 |
Sisters | -1.298 |
Drive | -1.268 |
Token | , |
Token ID | 11 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
anka | +0.821 |
plom | +0.800 |
glitter | +0.781 |
agen | +0.778 |
interpol | +0.753 |
Neg logit contributions | |
arcity | -0.855 |
Cage | -0.699 |
Hardware | -0.684 |
Sisters | -0.672 |
τ | -0.658 |
Token | environmental |
Token ID | 6142 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
anka | +3.653 |
glitter | +3.636 |
plom | +3.565 |
interpol | +3.449 |
mer | +3.449 |
Neg logit contributions | |
arcity | -5.300 |
Hardware | -4.434 |
Cage | -4.329 |
Drive | -4.266 |
τ | -4.231 |
Token | - |
Token ID | 12 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
anka | +1.039 |
plom | +1.007 |
agen | +0.977 |
glitter | +0.974 |
interpol | +0.934 |
Neg logit contributions | |
arcity | -1.479 |
Cage | -1.240 |
Hardware | -1.228 |
Sisters | -1.205 |
τ | -1.183 |
Token | management |
Token ID | 27604 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
anka | +0.963 |
plom | +0.934 |
agen | +0.912 |
glitter | +0.898 |
interpol | +0.868 |
Neg logit contributions | |
arcity | -1.159 |
Cage | -0.959 |
Hardware | -0.945 |
Sisters | -0.925 |
τ | -0.903 |
Token | , |
Token ID | 11 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
anka | +0.626 |
plom | +0.607 |
agen | +0.595 |
glitter | +0.586 |
interpol | +0.566 |
Neg logit contributions | |
arcity | -0.754 |
Cage | -0.624 |
Hardware | -0.617 |
Sisters | -0.603 |
τ | -0.592 |
Token | environmental |
Token ID | 6142 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
anka | +3.679 |
glitter | +3.654 |
plom | +3.584 |
mer | +3.467 |
interpol | +3.463 |
Neg logit contributions | |
arcity | -5.438 |
Hardware | -4.564 |
Cage | -4.451 |
Drive | -4.401 |
τ | -4.364 |
Token | ale |
Token ID | 1000 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
anka | +0.201 |
plom | +0.190 |
agen | +0.187 |
glitter | +0.180 |
interpol | +0.169 |
Neg logit contributions | |
arcity | -0.462 |
Cage | -0.403 |
Hardware | -0.401 |
Sisters | -0.395 |
τ | -0.390 |
Token | wil |
Token ID | 39716 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
anka | +0.515 |
plom | +0.502 |
glitter | +0.490 |
agen | +0.488 |
interpol | +0.466 |
Neg logit contributions | |
arcity | -0.756 |
Cage | -0.631 |
Hardware | -0.629 |
Sisters | -0.612 |
Drive | -0.605 |
Token | kins |
Token ID | 5331 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
arcity | +0.588 |
Cage | +0.482 |
Hardware | +0.477 |
Sisters | +0.461 |
τ | +0.453 |
Neg logit contributions | |
anka | -0.574 |
plom | -0.556 |
agen | -0.546 |
glitter | -0.539 |
interpol | -0.520 |
Token | i |
Token ID | 72 |
Feature activation | -8.73e-02 |
Pos logit contributions | |
arcity | +0.494 |
Cage | +0.399 |
Hardware | +0.397 |
Sisters | +0.383 |
Drive | +0.375 |
Neg logit contributions | |
anka | -0.503 |
plom | -0.487 |
agen | -0.478 |
glitter | -0.474 |
interpol | -0.458 |
Token | ), |
Token ID | 828 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
anka | +0.451 |
plom | +0.441 |
agen | +0.434 |
glitter | +0.426 |
interpol | +0.411 |
Neg logit contributions | |
arcity | -0.379 |
Hardware | -0.300 |
Cage | -0.299 |
Sisters | -0.287 |
Drive | -0.281 |
Token | after |
Token ID | 706 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
anka | +1.041 |
plom | +0.995 |
agen | +0.981 |
glitter | +0.975 |
interpol | +0.936 |
Neg logit contributions | |
arcity | -1.632 |
Hardware | -1.367 |
Cage | -1.367 |
Sisters | -1.325 |
Drive | -1.309 |
Token | Australian |
Token ID | 6638 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
anka | +3.962 |
glitter | +3.868 |
agen | +3.816 |
plom | +3.774 |
mer | +3.663 |
Neg logit contributions | |
arcity | -4.922 |
Hardware | -4.129 |
Cage | -3.912 |
Drive | -3.857 |
τ | -3.850 |
Token | explorer |
Token ID | 39349 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
arcity | +0.814 |
Cage | +0.681 |
Hardware | +0.679 |
Sisters | +0.662 |
τ | +0.656 |
Neg logit contributions | |
anka | -0.518 |
plom | -0.499 |
glitter | -0.490 |
agen | -0.490 |
interpol | -0.468 |
Token | , |
Token ID | 11 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
arcity | +0.265 |
Hardware | +0.212 |
Cage | +0.210 |
Sisters | +0.202 |
τ | +0.202 |
Neg logit contributions | |
anka | -0.261 |
plom | -0.253 |
agen | -0.250 |
glitter | -0.248 |
interpol | -0.240 |
Token | natural |
Token ID | 3288 |
Feature activation | +6.68e-02 |
Pos logit contributions | |
arcity | +0.350 |
Hardware | +0.287 |
Cage | +0.285 |
Sisters | +0.276 |
τ | +0.273 |
Neg logit contributions | |
anka | -0.286 |
plom | -0.278 |
glitter | -0.273 |
agen | -0.272 |
interpol | -0.262 |
Token | ist |
Token ID | 396 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
arcity | +0.334 |
Cage | +0.274 |
Hardware | +0.272 |
Sisters | +0.265 |
τ | +0.260 |
Neg logit contributions | |
anka | -0.297 |
plom | -0.288 |
agen | -0.285 |
glitter | -0.282 |
interpol | -0.273 |
Token | 366 |
Token ID | 44856 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
anka | +0.024 |
plom | +0.023 |
agen | +0.023 |
glitter | +0.023 |
interpol | +0.022 |
Neg logit contributions | |
arcity | -0.025 |
Cage | -0.020 |
Hardware | -0.020 |
Sisters | -0.019 |
τ | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
arcity | +1.958 |
Cage | +1.630 |
Hardware | +1.607 |
Sisters | +1.576 |
τ | +1.549 |
Neg logit contributions | |
anka | -1.660 |
plom | -1.610 |
agen | -1.572 |
glitter | -1.560 |
interpol | -1.503 |
Token |
|
Token ID | 198 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
arcity | +0.547 |
Cage | +0.436 |
Hardware | +0.430 |
Sisters | +0.419 |
τ | +0.409 |
Neg logit contributions | |
anka | -0.637 |
plom | -0.621 |
agen | -0.608 |
glitter | -0.606 |
interpol | -0.587 |
Token |
|
Token ID | 198 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
arcity | +0.435 |
Cage | +0.338 |
Hardware | +0.333 |
Sisters | +0.324 |
τ | +0.318 |
Neg logit contributions | |
anka | -0.507 |
plom | -0.492 |
agen | -0.486 |
glitter | -0.482 |
interpol | -0.468 |
Token | Topics |
Token ID | 25902 |
Feature activation | +6.90e-01 |
Pos logit contributions | |
arcity | +0.346 |
Cage | +0.248 |
Hardware | +0.236 |
Sisters | +0.231 |
RI | +0.221 |
Neg logit contributions | |
anka | -0.687 |
plom | -0.671 |
agen | -0.662 |
glitter | -0.654 |
interpol | -0.637 |
Token | : |
Token ID | 25 |
Feature activation | -9.57e-01 |
Pos logit contributions | |
arcity | +2.675 |
Cage | +2.103 |
Hardware | +2.063 |
Sisters | +2.014 |
τ | +1.933 |
Neg logit contributions | |
anka | -3.783 |
plom | -3.715 |
agen | -3.627 |
glitter | -3.615 |
interpol | -3.523 |
Token | community |
Token ID | 2055 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
anka | +3.693 |
glitter | +3.657 |
agen | +3.460 |
plom | +3.455 |
mer | +3.446 |
Neg logit contributions | |
arcity | -5.211 |
Cage | -4.367 |
Hardware | -4.308 |
Drive | -4.164 |
τ | -4.151 |
Token | - |
Token ID | 12 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
anka | +2.732 |
plom | +2.570 |
agen | +2.537 |
glitter | +2.535 |
interpol | +2.445 |
Neg logit contributions | |
arcity | -5.279 |
Hardware | -4.430 |
Cage | -4.424 |
τ | -4.307 |
Sisters | -4.291 |
Token | and |
Token ID | 392 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
anka | +2.595 |
plom | +2.586 |
agen | +2.488 |
glitter | +2.435 |
interpol | +2.332 |
Neg logit contributions | |
arcity | -3.650 |
Cage | -3.071 |
Hardware | -2.978 |
Sisters | -2.893 |
Drive | -2.849 |
Token | - |
Token ID | 12 |
Feature activation | -8.24e-02 |
Pos logit contributions | |
anka | +1.645 |
plom | +1.584 |
agen | +1.548 |
glitter | +1.529 |
interpol | +1.468 |
Neg logit contributions | |
arcity | -2.827 |
Cage | -2.390 |
Hardware | -2.380 |
τ | -2.305 |
Sisters | -2.304 |
Token | soc |
Token ID | 35634 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
anka | +0.429 |
plom | +0.420 |
agen | +0.413 |
glitter | +0.407 |
interpol | +0.395 |
Neg logit contributions | |
arcity | -0.351 |
Cage | -0.276 |
Hardware | -0.274 |
Sisters | -0.264 |
τ | -0.256 |
Token | - |
Token ID | 12 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
anka | +1.178 |
plom | +1.124 |
agen | +1.121 |
glitter | +1.092 |
interpol | +1.060 |
Neg logit contributions | |
arcity | -1.786 |
Cage | -1.521 |
Hardware | -1.492 |
Sisters | -1.472 |
τ | -1.456 |
Token | and |
Token ID | 392 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
anka | +0.087 |
plom | +0.084 |
agen | +0.084 |
glitter | +0.082 |
interpol | +0.079 |
Neg logit contributions | |
arcity | -0.101 |
Cage | -0.085 |
Hardware | -0.083 |
Sisters | -0.082 |
τ | -0.080 |
Token | - |
Token ID | 12 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
anka | +0.573 |
plom | +0.550 |
agen | +0.542 |
glitter | +0.534 |
interpol | +0.512 |
Neg logit contributions | |
arcity | -0.898 |
Cage | -0.763 |
Hardware | -0.753 |
Sisters | -0.744 |
τ | -0.732 |
Token | ut |
Token ID | 315 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
arcity | +0.800 |
Cage | +0.600 |
Hardware | +0.589 |
Sisters | +0.576 |
τ | +0.548 |
Neg logit contributions | |
anka | -1.387 |
plom | -1.352 |
agen | -1.336 |
glitter | -1.325 |
interpol | -1.288 |
Token | ilities |
Token ID | 2410 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
arcity | +0.831 |
Cage | +0.680 |
Hardware | +0.672 |
Sisters | +0.656 |
τ | +0.643 |
Neg logit contributions | |
anka | -0.763 |
plom | -0.742 |
agen | -0.723 |
glitter | -0.722 |
interpol | -0.696 |
Token | , |
Token ID | 11 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
anka | +1.692 |
plom | +1.624 |
agen | +1.609 |
glitter | +1.569 |
interpol | +1.511 |
Neg logit contributions | |
arcity | -2.244 |
Cage | -1.842 |
Hardware | -1.824 |
Sisters | -1.778 |
τ | -1.763 |
Token | family |
Token ID | 1641 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
anka | +3.443 |
glitter | +3.368 |
plom | +3.217 |
mer | +3.196 |
agen | +3.168 |
Neg logit contributions | |
arcity | -5.354 |
Cage | -4.478 |
Hardware | -4.447 |
Drive | -4.340 |
τ | -4.299 |
Token | - |
Token ID | 12 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
anka | +1.185 |
plom | +1.131 |
agen | +1.117 |
glitter | +1.096 |
interpol | +1.054 |
Neg logit contributions | |
arcity | -1.757 |
Cage | -1.480 |
Hardware | -1.469 |
Sisters | -1.436 |
τ | -1.408 |
Token | and |
Token ID | 392 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
anka | +0.498 |
plom | +0.482 |
agen | +0.477 |
glitter | +0.469 |
interpol | +0.448 |
Neg logit contributions | |
arcity | -0.636 |
Cage | -0.531 |
Hardware | -0.521 |
Sisters | -0.510 |
Drive | -0.492 |
Token | - |
Token ID | 12 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
anka | +0.826 |
plom | +0.795 |
agen | +0.777 |
glitter | +0.769 |
interpol | +0.736 |
Neg logit contributions | |
arcity | -1.338 |
Cage | -1.132 |
Hardware | -1.127 |
τ | -1.091 |
Sisters | -1.087 |
Token | children |
Token ID | 17197 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
arcity | +1.350 |
Cage | +1.047 |
Hardware | +1.034 |
Sisters | +1.007 |
τ | +0.984 |
Neg logit contributions | |
anka | -1.928 |
plom | -1.887 |
agen | -1.839 |
glitter | -1.837 |
interpol | -1.801 |
Token | management |
Token ID | 27604 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
anka | +0.963 |
plom | +0.934 |
agen | +0.912 |
glitter | +0.898 |
interpol | +0.868 |
Neg logit contributions | |
arcity | -1.159 |
Cage | -0.959 |
Hardware | -0.945 |
Sisters | -0.925 |
τ | -0.903 |
Token | , |
Token ID | 11 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
anka | +0.626 |
plom | +0.607 |
agen | +0.595 |
glitter | +0.586 |
interpol | +0.566 |
Neg logit contributions | |
arcity | -0.754 |
Cage | -0.624 |
Hardware | -0.617 |
Sisters | -0.603 |
τ | -0.592 |
Token | environmental |
Token ID | 6142 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
anka | +3.679 |
glitter | +3.654 |
plom | +3.584 |
mer | +3.467 |
interpol | +3.463 |
Neg logit contributions | |
arcity | -5.438 |
Hardware | -4.564 |
Cage | -4.451 |
Drive | -4.401 |
τ | -4.364 |
Token | - |
Token ID | 12 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
anka | +0.867 |
plom | +0.843 |
glitter | +0.816 |
agen | +0.813 |
interpol | +0.782 |
Neg logit contributions | |
arcity | -1.201 |
Cage | -1.001 |
Hardware | -0.991 |
Sisters | -0.975 |
τ | -0.955 |
Token | technology |
Token ID | 45503 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
anka | +0.700 |
plom | +0.681 |
agen | +0.665 |
glitter | +0.654 |
interpol | +0.633 |
Neg logit contributions | |
arcity | -0.841 |
Cage | -0.695 |
Hardware | -0.685 |
Sisters | -0.670 |
τ | -0.655 |
Token | , |
Token ID | 11 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
anka | +2.152 |
plom | +2.088 |
agen | +2.045 |
glitter | +2.025 |
interpol | +1.952 |
Neg logit contributions | |
arcity | -2.566 |
Cage | -2.109 |
Hardware | -2.087 |
Sisters | -2.037 |
τ | -2.001 |
Token | q |
Token ID | 10662 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
anka | +3.481 |
glitter | +3.449 |
plom | +3.394 |
mer | +3.268 |
interpol | +3.267 |
Neg logit contributions | |
arcity | -5.233 |
Hardware | -4.375 |
Cage | -4.286 |
Drive | -4.225 |
τ | -4.195 |
Token | ld |
Token ID | 335 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
anka | +2.221 |
agen | +2.124 |
plom | +2.106 |
glitter | +1.911 |
interpol | +1.831 |
Neg logit contributions | |
arcity | -4.202 |
Hardware | -3.776 |
Cage | -3.725 |
FACE | -3.716 |
Sisters | -3.680 |
Token |
|
Token ID | 198 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
anka | +2.390 |
plom | +2.328 |
agen | +2.281 |
glitter | +2.260 |
interpol | +2.189 |
Neg logit contributions | |
arcity | -2.302 |
Cage | -1.863 |
Hardware | -1.839 |
Sisters | -1.792 |
τ | -1.754 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
anka | +2.196 |
plom | +2.112 |
agen | +2.097 |
glitter | +2.072 |
interpol | +2.014 |
Neg logit contributions | |
arcity | -2.329 |
Cage | -1.785 |
conduc | -1.770 |
Hardware | -1.745 |
Sisters | -1.722 |
Token | First |
Token ID | 5962 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
anka | +0.664 |
plom | +0.647 |
agen | +0.631 |
glitter | +0.622 |
interpol | +0.599 |
Neg logit contributions | |
arcity | -0.700 |
Cage | -0.566 |
Hardware | -0.553 |
Sisters | -0.546 |
RI | -0.533 |
Token | them |
Token ID | 606 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
arcity | +1.296 |
Hardware | +1.076 |
Cage | +1.064 |
Sisters | +1.037 |
τ | +1.030 |
Neg logit contributions | |
anka | -0.953 |
plom | -0.915 |
glitter | -0.913 |
agen | -0.897 |
interpol | -0.870 |
Token | ". |
Token ID | 1911 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
arcity | +0.750 |
Cage | +0.621 |
Hardware | +0.618 |
Sisters | +0.603 |
τ | +0.593 |
Neg logit contributions | |
anka | -0.595 |
plom | -0.577 |
agen | -0.563 |
glitter | -0.561 |
interpol | -0.540 |
Token |
|
Token ID | 198 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
arcity | +0.535 |
Cage | +0.419 |
Hardware | +0.415 |
Sisters | +0.403 |
τ | +0.394 |
Neg logit contributions | |
anka | -0.655 |
plom | -0.637 |
agen | -0.625 |
glitter | -0.621 |
interpol | -0.602 |
Token |
|
Token ID | 198 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
arcity | +0.879 |
Cage | +0.684 |
Hardware | +0.675 |
Sisters | +0.657 |
τ | +0.643 |
Neg logit contributions | |
anka | -1.061 |
plom | -1.031 |
agen | -1.015 |
glitter | -1.009 |
interpol | -0.979 |
Token | Topics |
Token ID | 25902 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
arcity | +0.737 |
Cage | +0.592 |
Hardware | +0.577 |
Sisters | +0.571 |
τ | +0.552 |
Neg logit contributions | |
anka | -0.758 |
plom | -0.737 |
agen | -0.723 |
glitter | -0.716 |
interpol | -0.691 |
Token | : |
Token ID | 25 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
arcity | +2.615 |
Cage | +2.055 |
Hardware | +2.020 |
Sisters | +1.968 |
τ | +1.895 |
Neg logit contributions | |
anka | -3.651 |
plom | -3.578 |
agen | -3.498 |
glitter | -3.485 |
interpol | -3.395 |
Token | marine |
Token ID | 16050 |
Feature activation | -6.42e-01 |
Pos logit contributions | |
glitter | +3.864 |
anka | +3.805 |
plom | +3.715 |
interpol | +3.689 |
agen | +3.676 |
Neg logit contributions | |
arcity | -4.742 |
Hardware | -3.853 |
Cage | -3.814 |
τ | -3.747 |
Drive | -3.733 |
Token | - |
Token ID | 12 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
anka | +2.340 |
plom | +2.265 |
agen | +2.248 |
glitter | +2.195 |
interpol | +2.091 |
Neg logit contributions | |
arcity | -3.668 |
Cage | -3.143 |
Hardware | -3.089 |
Sisters | -3.078 |
τ | -3.014 |
Token | biology |
Token ID | 43592 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
anka | +1.942 |
plom | +1.898 |
agen | +1.894 |
glitter | +1.802 |
interpol | +1.709 |
Neg logit contributions | |
arcity | -2.920 |
Cage | -2.486 |
Sisters | -2.461 |
Hardware | -2.446 |
Drive | -2.378 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
anka | +2.888 |
agen | +2.789 |
plom | +2.789 |
glitter | +2.708 |
interpol | +2.613 |
Neg logit contributions | |
arcity | -4.353 |
Hardware | -3.599 |
Cage | -3.571 |
Sisters | -3.496 |
τ | -3.470 |
Token | conservation |
Token ID | 14903 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
glitter | +4.317 |
anka | +4.288 |
plom | +4.200 |
interpol | +4.121 |
mer | +4.120 |
Neg logit contributions | |
arcity | -5.623 |
Hardware | -4.629 |
Cage | -4.521 |
Drive | -4.481 |
τ | -4.449 |
Token | with |
Token ID | 351 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
arcity | +0.978 |
Cage | +0.804 |
Hardware | +0.798 |
Sisters | +0.777 |
τ | +0.766 |
Neg logit contributions | |
anka | -0.814 |
plom | -0.789 |
agen | -0.770 |
glitter | -0.769 |
interpol | -0.743 |
Token | all |
Token ID | 477 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
arcity | +1.400 |
Hardware | +1.145 |
Cage | +1.139 |
Sisters | +1.105 |
τ | +1.087 |
Neg logit contributions | |
anka | -1.165 |
plom | -1.118 |
glitter | -1.108 |
agen | -1.098 |
interpol | -1.066 |
Token | this |
Token ID | 428 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
arcity | +0.307 |
Hardware | +0.250 |
Cage | +0.248 |
Sisters | +0.241 |
τ | +0.237 |
Neg logit contributions | |
anka | -0.251 |
glitter | -0.240 |
plom | -0.239 |
agen | -0.238 |
interpol | -0.229 |
Token | Al |
Token ID | 978 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
anka | +0.652 |
glitter | +0.625 |
plom | +0.617 |
agen | +0.611 |
interpol | +0.595 |
Neg logit contributions | |
arcity | -0.939 |
Hardware | -0.772 |
Cage | -0.766 |
Sisters | -0.752 |
τ | -0.742 |
Token | J |
Token ID | 449 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
anka | +0.169 |
plom | +0.161 |
agen | +0.159 |
glitter | +0.157 |
interpol | +0.151 |
Neg logit contributions | |
arcity | -0.203 |
Hardware | -0.166 |
Cage | -0.165 |
Sisters | -0.162 |
τ | -0.159 |
Token | our |
Token ID | 454 |
Feature activation | -9.07e-01 |
Pos logit contributions | |
arcity | +0.857 |
Cage | +0.703 |
Hardware | +0.703 |
Sisters | +0.684 |
τ | +0.673 |
Neg logit contributions | |
anka | -0.740 |
plom | -0.711 |
agen | -0.700 |
glitter | -0.690 |
interpol | -0.664 |
Token | g |
Token ID | 70 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
anka | +3.990 |
plom | +3.709 |
agen | +3.675 |
glitter | +3.634 |
interpol | +3.548 |
Neg logit contributions | |
arcity | -4.676 |
Cage | -3.739 |
Hardware | -3.674 |
Sisters | -3.559 |
Drive | -3.540 |
Token | ensen |
Token ID | 18756 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
anka | +2.571 |
agen | +2.502 |
plom | +2.491 |
glitter | +2.406 |
mer | +2.313 |
Neg logit contributions | |
arcity | -2.096 |
Hardware | -1.616 |
Cage | -1.601 |
Sisters | -1.514 |
τ | -1.502 |
Token | and |
Token ID | 290 |
Feature activation | -3.43e-03 |
Pos logit contributions | |
arcity | +0.925 |
Cage | +0.765 |
Hardware | +0.760 |
Sisters | +0.744 |
τ | +0.731 |
Neg logit contributions | |
anka | -0.678 |
plom | -0.653 |
agen | -0.639 |
glitter | -0.639 |
interpol | -0.614 |
Token | Ge |
Token ID | 2269 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
anka | +0.015 |
glitter | +0.015 |
plom | +0.015 |
agen | +0.015 |
interpol | +0.014 |
Neg logit contributions | |
arcity | -0.017 |
Hardware | -0.014 |
Cage | -0.013 |
Sisters | -0.013 |
τ | -0.013 |
Token | za |
Token ID | 4496 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
arcity | +0.290 |
Hardware | +0.241 |
Cage | +0.240 |
Sisters | +0.234 |
τ | +0.229 |
Neg logit contributions | |
anka | -0.236 |
plom | -0.228 |
agen | -0.225 |
glitter | -0.217 |
interpol | -0.209 |
Token | groups |
Token ID | 2628 |
Feature activation | +9.96e-02 |
Pos logit contributions | |
arcity | +0.754 |
Cage | +0.628 |
Hardware | +0.625 |
Sisters | +0.607 |
τ | +0.601 |
Neg logit contributions | |
anka | -0.556 |
plom | -0.541 |
glitter | -0.526 |
agen | -0.525 |
interpol | -0.505 |
Token | @ |
Token ID | 31 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
arcity | +0.509 |
Cage | +0.423 |
Hardware | +0.416 |
Sisters | +0.408 |
τ | +0.399 |
Neg logit contributions | |
anka | -0.433 |
plom | -0.424 |
agen | -0.415 |
glitter | -0.406 |
interpol | -0.389 |
Token | peace |
Token ID | 22988 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
anka | +1.625 |
plom | +1.589 |
agen | +1.528 |
glitter | +1.494 |
interpol | +1.401 |
Neg logit contributions | |
arcity | -2.796 |
Cage | -2.420 |
Hardware | -2.391 |
Sisters | -2.348 |
Drive | -2.307 |
Token | cor |
Token ID | 10215 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
anka | +0.699 |
plom | +0.669 |
agen | +0.633 |
glitter | +0.599 |
interpol | +0.553 |
Neg logit contributions | |
arcity | -2.262 |
Hardware | -2.011 |
Cage | -2.005 |
Sisters | -1.975 |
τ | -1.951 |
Token | ps |
Token ID | 862 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
anka | +2.240 |
plom | +2.223 |
agen | +2.147 |
glitter | +2.036 |
interpol | +1.980 |
Neg logit contributions | |
arcity | -2.962 |
Cage | -2.527 |
Hardware | -2.494 |
RI | -2.426 |
Sisters | -2.422 |
Token | connect |
Token ID | 8443 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
anka | +1.006 |
plom | +0.973 |
agen | +0.941 |
glitter | +0.913 |
interpol | +0.864 |
Neg logit contributions | |
arcity | -1.923 |
Cage | -1.658 |
Hardware | -1.641 |
Sisters | -1.622 |
RI | -1.596 |
Token | . |
Token ID | 13 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
anka | +3.149 |
agen | +2.966 |
plom | +2.950 |
glitter | +2.770 |
interpol | +2.660 |
Neg logit contributions | |
arcity | -5.330 |
conduc | -4.521 |
Cage | -4.465 |
Sisters | -4.380 |
Hardware | -4.361 |
Token | org |
Token ID | 2398 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
anka | +1.523 |
agen | +1.477 |
plom | +1.469 |
glitter | +1.385 |
interpol | +1.320 |
Neg logit contributions | |
arcity | -2.251 |
Cage | -1.884 |
Hardware | -1.861 |
Sisters | -1.841 |
τ | -1.804 |
Token | . |
Token ID | 13 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
anka | +1.478 |
plom | +1.435 |
agen | +1.397 |
glitter | +1.391 |
interpol | +1.336 |
Neg logit contributions | |
arcity | -1.867 |
Cage | -1.552 |
Hardware | -1.536 |
Sisters | -1.505 |
τ | -1.478 |
Token | The |
Token ID | 383 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
arcity | +0.860 |
Cage | +0.729 |
Hardware | +0.724 |
Sisters | +0.709 |
τ | +0.701 |
Neg logit contributions | |
anka | -0.512 |
plom | -0.494 |
agen | -0.479 |
glitter | -0.475 |
interpol | -0.453 |
Token | deadline |
Token ID | 12508 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
arcity | +0.178 |
Hardware | +0.146 |
Cage | +0.145 |
τ | +0.142 |
Sisters | +0.139 |
Neg logit contributions | |
anka | -0.141 |
plom | -0.139 |
glitter | -0.137 |
agen | -0.135 |
interpol | -0.130 |
Token | safety |
Token ID | 3747 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
anka | +0.302 |
glitter | +0.293 |
plom | +0.290 |
agen | +0.285 |
interpol | +0.272 |
Neg logit contributions | |
arcity | -0.442 |
Cage | -0.365 |
Hardware | -0.363 |
Sisters | -0.353 |
τ | -0.353 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
arcity | +1.219 |
Cage | +1.011 |
Hardware | +1.000 |
Sisters | +0.976 |
τ | +0.956 |
Neg logit contributions | |
anka | -1.092 |
plom | -1.060 |
agen | -1.033 |
glitter | -1.022 |
interpol | -0.991 |
Token | It |
Token ID | 632 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
arcity | +0.823 |
Hardware | +0.661 |
Cage | +0.658 |
Sisters | +0.637 |
τ | +0.634 |
Neg logit contributions | |
anka | -0.723 |
plom | -0.697 |
agen | -0.687 |
glitter | -0.684 |
interpol | -0.656 |
Token | 's |
Token ID | 338 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
arcity | +1.030 |
Hardware | +0.826 |
Cage | +0.823 |
Sisters | +0.799 |
τ | +0.783 |
Neg logit contributions | |
anka | -1.092 |
plom | -1.062 |
agen | -1.044 |
glitter | -1.043 |
interpol | -1.008 |
Token | right |
Token ID | 826 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
anka | +0.299 |
glitter | +0.287 |
plom | +0.285 |
agen | +0.281 |
interpol | +0.271 |
Neg logit contributions | |
arcity | -0.422 |
Hardware | -0.349 |
Cage | -0.346 |
Sisters | -0.336 |
τ | -0.335 |
Token | up |
Token ID | 510 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
arcity | +0.108 |
Cage | +0.090 |
Hardware | +0.089 |
Sisters | +0.088 |
τ | +0.086 |
Neg logit contributions | |
anka | -0.078 |
plom | -0.074 |
agen | -0.073 |
glitter | -0.073 |
interpol | -0.070 |
Token | there |
Token ID | 612 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
arcity | +2.942 |
Cage | +2.391 |
Hardware | +2.349 |
Sisters | +2.305 |
τ | +2.241 |
Neg logit contributions | |
anka | -3.014 |
plom | -2.940 |
agen | -2.870 |
glitter | -2.853 |
interpol | -2.771 |
Token | with |
Token ID | 351 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
arcity | +3.350 |
Cage | +2.810 |
Hardware | +2.747 |
Sisters | +2.669 |
RI | +2.617 |
Neg logit contributions | |
anka | -3.015 |
plom | -2.934 |
agen | -2.859 |
glitter | -2.811 |
interpol | -2.757 |
Token | be |
Token ID | 307 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
arcity | +1.746 |
Hardware | +1.440 |
Cage | +1.437 |
Sisters | +1.401 |
τ | +1.385 |
Neg logit contributions | |
anka | -1.324 |
plom | -1.280 |
glitter | -1.262 |
agen | -1.250 |
interpol | -1.201 |
Token | avers |
Token ID | 30400 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
arcity | +1.617 |
Cage | +1.353 |
Hardware | +1.337 |
Sisters | +1.311 |
τ | +1.287 |
Neg logit contributions | |
anka | -1.217 |
plom | -1.180 |
agen | -1.148 |
glitter | -1.142 |
interpol | -1.099 |
Token | and |
Token ID | 290 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
arcity | +1.341 |
Cage | +1.113 |
Hardware | +1.098 |
Sisters | +1.076 |
τ | +1.055 |
Neg logit contributions | |
anka | -1.126 |
plom | -1.093 |
agen | -1.064 |
glitter | -1.056 |
interpol | -1.021 |
Token | was |
Token ID | 373 |
Feature activation | +3.02e-03 |
Pos logit contributions | |
anka | +0.525 |
plom | +0.511 |
agen | +0.506 |
glitter | +0.502 |
interpol | +0.499 |
Neg logit contributions | |
arcity | -0.644 |
Hardware | -0.520 |
Cage | -0.517 |
Sisters | -0.505 |
τ | -0.504 |
Token | available |
Token ID | 1695 |
Feature activation | +4.25e-04 |
Pos logit contributions | |
arcity | +0.016 |
Cage | +0.013 |
Hardware | +0.013 |
Sisters | +0.013 |
τ | +0.013 |
Neg logit contributions | |
anka | -0.012 |
plom | -0.012 |
agen | -0.012 |
glitter | -0.012 |
interpol | -0.012 |
Token | for |
Token ID | 329 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
arcity | +0.002 |
Cage | +0.002 |
Hardware | +0.002 |
Sisters | +0.002 |
τ | +0.002 |
Neg logit contributions | |
anka | -0.002 |
plom | -0.002 |
agen | -0.002 |
glitter | -0.002 |
interpol | -0.002 |
Token | 802 |
Token ID | 33121 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
arcity | +0.382 |
Cage | +0.309 |
Hardware | +0.304 |
τ | +0.298 |
Sisters | +0.297 |
Neg logit contributions | |
anka | -0.321 |
plom | -0.309 |
agen | -0.307 |
glitter | -0.306 |
interpol | -0.299 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
arcity | +1.518 |
Hardware | +1.259 |
Cage | +1.254 |
Sisters | +1.222 |
τ | +1.189 |
Neg logit contributions | |
anka | -1.393 |
plom | -1.361 |
glitter | -1.318 |
agen | -1.309 |
interpol | -1.264 |
Token | 11 |
Token ID | 1157 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
arcity | +0.075 |
Cage | +0.061 |
Hardware | +0.061 |
Sisters | +0.059 |
τ | +0.058 |
Neg logit contributions | |
anka | -0.072 |
plom | -0.070 |
agen | -0.069 |
glitter | -0.069 |
interpol | -0.066 |
Token | g |
Token ID | 70 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
arcity | +2.959 |
Cage | +2.471 |
Hardware | +2.452 |
Sisters | +2.403 |
τ | +2.352 |
Neg logit contributions | |
anka | -2.247 |
plom | -2.182 |
agen | -2.116 |
glitter | -2.111 |
interpol | -2.026 |
Token | devices |
Token ID | 4410 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
arcity | +0.785 |
Cage | +0.640 |
Hardware | +0.630 |
Sisters | +0.617 |
τ | +0.604 |
Neg logit contributions | |
anka | -0.716 |
plom | -0.694 |
agen | -0.682 |
glitter | -0.675 |
interpol | -0.656 |
Token | . |
Token ID | 13 |
Feature activation | +6.13e-02 |
Pos logit contributions | |
arcity | +1.133 |
Cage | +0.941 |
Hardware | +0.932 |
Sisters | +0.914 |
τ | +0.899 |
Neg logit contributions | |
anka | -0.826 |
plom | -0.795 |
agen | -0.779 |
glitter | -0.777 |
interpol | -0.750 |
Token | However |
Token ID | 2102 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
arcity | +0.297 |
Cage | +0.234 |
Hardware | +0.230 |
Sisters | +0.225 |
τ | +0.223 |
Neg logit contributions | |
anka | -0.284 |
plom | -0.274 |
agen | -0.272 |
glitter | -0.268 |
interpol | -0.263 |
Token | , |
Token ID | 11 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
arcity | +0.652 |
Cage | +0.535 |
Hardware | +0.529 |
Sisters | +0.518 |
τ | +0.507 |
Neg logit contributions | |
anka | -0.592 |
plom | -0.575 |
agen | -0.562 |
glitter | -0.559 |
interpol | -0.539 |
Token | Division |
Token ID | 7458 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
arcity | +0.548 |
Cage | +0.452 |
Hardware | +0.451 |
Sisters | +0.438 |
τ | +0.432 |
Neg logit contributions | |
anka | -0.438 |
plom | -0.421 |
agen | -0.413 |
glitter | -0.413 |
interpol | -0.397 |
Token | National |
Token ID | 2351 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
arcity | +0.184 |
Cage | +0.147 |
Hardware | +0.147 |
Sisters | +0.141 |
τ | +0.139 |
Neg logit contributions | |
anka | -0.186 |
plom | -0.179 |
glitter | -0.176 |
agen | -0.175 |
interpol | -0.170 |
Token | Forest |
Token ID | 9115 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
anka | +0.168 |
plom | +0.159 |
glitter | +0.158 |
agen | +0.158 |
interpol | +0.151 |
Neg logit contributions | |
arcity | -0.199 |
Hardware | -0.160 |
Cage | -0.159 |
Sisters | -0.154 |
τ | -0.154 |
Token | — |
Token ID | 851 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
arcity | +1.725 |
Cage | +1.424 |
Hardware | +1.411 |
Sisters | +1.379 |
τ | +1.359 |
Neg logit contributions | |
anka | -1.433 |
plom | -1.378 |
agen | -1.353 |
glitter | -1.350 |
interpol | -1.297 |
Token | swap |
Token ID | 16075 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
anka | +0.217 |
plom | +0.207 |
glitter | +0.206 |
agen | +0.204 |
interpol | +0.197 |
Neg logit contributions | |
arcity | -0.272 |
Cage | -0.223 |
Hardware | -0.223 |
Sisters | -0.216 |
τ | -0.214 |
Token | the |
Token ID | 262 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
arcity | +3.330 |
Cage | +2.759 |
Hardware | +2.729 |
Sisters | +2.672 |
τ | +2.620 |
Neg logit contributions | |
anka | -2.759 |
plom | -2.678 |
agen | -2.612 |
glitter | -2.598 |
interpol | -2.503 |
Token | concrete |
Token ID | 10017 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
arcity | +3.148 |
Cage | +2.636 |
Hardware | +2.611 |
Sisters | +2.557 |
τ | +2.516 |
Neg logit contributions | |
anka | -2.283 |
plom | -2.208 |
agen | -2.150 |
glitter | -2.141 |
interpol | -2.055 |
Token | for |
Token ID | 329 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
arcity | +0.058 |
Hardware | +0.047 |
Cage | +0.047 |
τ | +0.045 |
Sisters | +0.045 |
Neg logit contributions | |
anka | -0.044 |
glitter | -0.042 |
plom | -0.042 |
agen | -0.041 |
interpol | -0.040 |
Token | trees |
Token ID | 7150 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
arcity | +1.612 |
Hardware | +1.331 |
Cage | +1.325 |
Sisters | +1.293 |
τ | +1.283 |
Neg logit contributions | |
anka | -1.191 |
glitter | -1.145 |
plom | -1.144 |
agen | -1.117 |
interpol | -1.081 |
Token | , |
Token ID | 11 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
arcity | +0.085 |
Cage | +0.070 |
Hardware | +0.070 |
Sisters | +0.068 |
τ | +0.067 |
Neg logit contributions | |
anka | -0.066 |
plom | -0.063 |
glitter | -0.062 |
agen | -0.062 |
interpol | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +8.91e-02 |
Pos logit contributions | |
arcity | +0.439 |
Hardware | +0.357 |
Cage | +0.356 |
Sisters | +0.345 |
τ | +0.344 |
Neg logit contributions | |
anka | -0.337 |
glitter | -0.323 |
plom | -0.321 |
agen | -0.312 |
interpol | -0.307 |
Token | from |
Token ID | 422 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
arcity | +1.941 |
Cage | +1.585 |
Hardware | +1.564 |
Sisters | +1.530 |
τ | +1.498 |
Neg logit contributions | |
anka | -1.862 |
plom | -1.810 |
agen | -1.770 |
glitter | -1.760 |
interpol | -1.702 |
Token | antagon |
Token ID | 23775 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
arcity | +0.082 |
Hardware | +0.069 |
Cage | +0.068 |
Sisters | +0.066 |
τ | +0.066 |
Neg logit contributions | |
anka | -0.059 |
plom | -0.057 |
glitter | -0.057 |
agen | -0.055 |
interpol | -0.053 |
Token | istic |
Token ID | 2569 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
anka | +0.556 |
plom | +0.535 |
agen | +0.526 |
glitter | +0.517 |
interpol | +0.492 |
Neg logit contributions | |
arcity | -0.913 |
Cage | -0.773 |
Hardware | -0.766 |
Sisters | -0.746 |
τ | -0.736 |
Token | states |
Token ID | 2585 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
arcity | +0.248 |
Hardware | +0.206 |
Cage | +0.205 |
Sisters | +0.198 |
Drive | +0.197 |
Neg logit contributions | |
anka | -0.172 |
plom | -0.167 |
glitter | -0.165 |
agen | -0.161 |
interpol | -0.155 |
Token | , |
Token ID | 11 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
anka | +0.539 |
plom | +0.525 |
glitter | +0.508 |
agen | +0.506 |
interpol | +0.478 |
Neg logit contributions | |
arcity | -0.899 |
Cage | -0.761 |
Hardware | -0.759 |
Sisters | -0.739 |
Drive | -0.726 |
Token | but |
Token ID | 475 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
arcity | +2.070 |
Cage | +1.760 |
Hardware | +1.747 |
Sisters | +1.712 |
τ | +1.686 |
Neg logit contributions | |
anka | -1.195 |
plom | -1.152 |
agen | -1.114 |
glitter | -1.112 |
interpol | -1.059 |
Token | from |
Token ID | 422 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
arcity | +3.079 |
Cage | +2.561 |
Hardware | +2.526 |
Sisters | +2.487 |
τ | +2.436 |
Neg logit contributions | |
anka | -2.483 |
plom | -2.401 |
agen | -2.351 |
glitter | -2.329 |
interpol | -2.248 |
Token | our |
Token ID | 674 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
arcity | +0.245 |
Hardware | +0.203 |
Cage | +0.203 |
Sisters | +0.196 |
τ | +0.194 |
Neg logit contributions | |
anka | -0.174 |
plom | -0.169 |
glitter | -0.165 |
agen | -0.162 |
interpol | -0.157 |
Token | own |
Token ID | 898 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
arcity | +0.149 |
Cage | +0.124 |
Hardware | +0.123 |
Sisters | +0.119 |
τ | +0.118 |
Neg logit contributions | |
anka | -0.108 |
plom | -0.105 |
glitter | -0.103 |
agen | -0.101 |
interpol | -0.097 |
Token | foreign |
Token ID | 3215 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
arcity | +2.025 |
Cage | +1.690 |
Hardware | +1.671 |
Sisters | +1.641 |
τ | +1.609 |
Neg logit contributions | |
anka | -1.570 |
plom | -1.522 |
agen | -1.485 |
glitter | -1.473 |
interpol | -1.419 |
Token | service |
Token ID | 2139 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
arcity | +0.854 |
Cage | +0.708 |
Hardware | +0.701 |
Sisters | +0.683 |
τ | +0.673 |
Neg logit contributions | |
anka | -0.675 |
plom | -0.655 |
agen | -0.639 |
glitter | -0.638 |
interpol | -0.611 |
Token | support |
Token ID | 1104 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
arcity | +0.006 |
Hardware | +0.005 |
Cage | +0.005 |
Sisters | +0.005 |
τ | +0.005 |
Neg logit contributions | |
anka | -0.004 |
plom | -0.004 |
glitter | -0.004 |
agen | -0.004 |
interpol | -0.004 |
Token | their |
Token ID | 511 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
anka | +0.296 |
plom | +0.282 |
glitter | +0.279 |
agen | +0.276 |
interpol | +0.268 |
Neg logit contributions | |
arcity | -0.403 |
Hardware | -0.330 |
Cage | -0.326 |
Sisters | -0.319 |
τ | -0.317 |
Token | already |
Token ID | 1541 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
anka | +0.832 |
glitter | +0.797 |
plom | +0.790 |
agen | +0.772 |
interpol | +0.745 |
Neg logit contributions | |
arcity | -1.482 |
Hardware | -1.251 |
Cage | -1.247 |
τ | -1.214 |
Sisters | -1.208 |
Token | - |
Token ID | 12 |
Feature activation | +1.32e-03 |
Pos logit contributions | |
arcity | +0.255 |
Cage | +0.214 |
Hardware | +0.213 |
Sisters | +0.208 |
τ | +0.205 |
Neg logit contributions | |
anka | -0.174 |
plom | -0.168 |
glitter | -0.164 |
agen | -0.164 |
interpol | -0.156 |
Token | te |
Token ID | 660 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
arcity | +0.007 |
Cage | +0.006 |
Hardware | +0.006 |
Sisters | +0.005 |
τ | +0.005 |
Neg logit contributions | |
anka | -0.006 |
plom | -0.006 |
agen | -0.005 |
glitter | -0.005 |
interpol | -0.005 |
Token | et |
Token ID | 316 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
arcity | +0.523 |
Cage | +0.438 |
Hardware | +0.433 |
Sisters | +0.429 |
τ | +0.418 |
Neg logit contributions | |
anka | -0.422 |
plom | -0.412 |
agen | -0.401 |
glitter | -0.397 |
interpol | -0.379 |
Token | ering |
Token ID | 1586 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
arcity | +3.675 |
Cage | +3.150 |
Hardware | +3.112 |
Sisters | +3.056 |
τ | +3.012 |
Neg logit contributions | |
anka | -1.910 |
plom | -1.847 |
glitter | -1.775 |
agen | -1.762 |
interpol | -1.694 |
Token | legislative |
Token ID | 10828 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
arcity | +0.759 |
Hardware | +0.643 |
Cage | +0.642 |
Sisters | +0.628 |
τ | +0.623 |
Neg logit contributions | |
anka | -0.395 |
plom | -0.377 |
glitter | -0.371 |
agen | -0.368 |
interpol | -0.351 |
Token | agenda |
Token ID | 8666 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
anka | +0.954 |
glitter | +0.910 |
plom | +0.901 |
agen | +0.881 |
interpol | +0.853 |
Neg logit contributions | |
arcity | -1.896 |
Hardware | -1.585 |
Cage | -1.569 |
τ | -1.553 |
Sisters | -1.527 |
Token | if |
Token ID | 611 |
Feature activation | -2.52e-03 |
Pos logit contributions | |
anka | +0.192 |
plom | +0.186 |
agen | +0.181 |
glitter | +0.180 |
interpol | +0.172 |
Neg logit contributions | |
arcity | -0.274 |
Cage | -0.229 |
Hardware | -0.228 |
Sisters | -0.222 |
τ | -0.219 |
Token | there |
Token ID | 612 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
anka | +0.011 |
plom | +0.010 |
agen | +0.010 |
glitter | +0.010 |
interpol | +0.010 |
Neg logit contributions | |
arcity | -0.013 |
Hardware | -0.011 |
Cage | -0.011 |
Sisters | -0.011 |
τ | -0.011 |
Token | category |
Token ID | 6536 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
arcity | +1.390 |
Cage | +1.130 |
Hardware | +1.129 |
Sisters | +1.103 |
τ | +1.092 |
Neg logit contributions | |
anka | -1.119 |
plom | -1.084 |
glitter | -1.076 |
agen | -1.065 |
interpol | -1.043 |
Token | which |
Token ID | 543 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
arcity | +0.273 |
Cage | +0.227 |
Hardware | +0.225 |
Sisters | +0.221 |
τ | +0.216 |
Neg logit contributions | |
anka | -0.210 |
plom | -0.205 |
agen | -0.200 |
glitter | -0.198 |
interpol | -0.192 |
Token | could |
Token ID | 714 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
arcity | +1.553 |
Hardware | +1.266 |
Cage | +1.250 |
Sisters | +1.217 |
τ | +1.210 |
Neg logit contributions | |
anka | -1.348 |
plom | -1.310 |
glitter | -1.293 |
agen | -1.278 |
interpol | -1.259 |
Token | capture |
Token ID | 8006 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
anka | +0.130 |
plom | +0.128 |
glitter | +0.127 |
agen | +0.126 |
interpol | +0.124 |
Neg logit contributions | |
arcity | -0.166 |
Hardware | -0.138 |
Cage | -0.137 |
Sisters | -0.133 |
τ | -0.131 |
Token | the |
Token ID | 262 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
anka | +0.051 |
glitter | +0.049 |
plom | +0.049 |
agen | +0.048 |
interpol | +0.047 |
Neg logit contributions | |
arcity | -0.065 |
Hardware | -0.053 |
Cage | -0.053 |
Sisters | -0.051 |
τ | -0.051 |
Token | economic |
Token ID | 3034 |
Feature activation | -8.18e-03 |
Pos logit contributions | |
anka | +0.071 |
glitter | +0.069 |
plom | +0.068 |
agen | +0.067 |
interpol | +0.065 |
Neg logit contributions | |
arcity | -0.100 |
Hardware | -0.083 |
Cage | -0.083 |
Sisters | -0.080 |
τ | -0.080 |
Token | middle |
Token ID | 3504 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
anka | +0.034 |
glitter | +0.033 |
plom | +0.033 |
agen | +0.032 |
interpol | +0.032 |
Neg logit contributions | |
arcity | -0.043 |
Hardware | -0.036 |
Cage | -0.036 |
Sisters | -0.035 |
τ | -0.034 |
Token | in |
Token ID | 287 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
anka | +0.199 |
plom | +0.190 |
glitter | +0.187 |
agen | +0.187 |
interpol | +0.179 |
Neg logit contributions | |
arcity | -0.285 |
Cage | -0.239 |
Hardware | -0.238 |
Sisters | -0.232 |
τ | -0.229 |
Token | India |
Token ID | 3794 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
anka | +0.482 |
plom | +0.461 |
glitter | +0.461 |
agen | +0.452 |
interpol | +0.438 |
Neg logit contributions | |
arcity | -0.626 |
Hardware | -0.522 |
Cage | -0.519 |
Sisters | -0.502 |
τ | -0.498 |
Token | . |
Token ID | 13 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
anka | +0.738 |
plom | +0.714 |
agen | +0.697 |
glitter | +0.694 |
interpol | +0.667 |
Neg logit contributions | |
arcity | -0.979 |
Cage | -0.812 |
Hardware | -0.807 |
Sisters | -0.791 |
Drive | -0.772 |
Token | While |
Token ID | 2893 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
anka | +0.199 |
plom | +0.190 |
glitter | +0.187 |
agen | +0.187 |
interpol | +0.181 |
Neg logit contributions | |
arcity | -0.210 |
Hardware | -0.166 |
Cage | -0.165 |
Sisters | -0.158 |
τ | -0.158 |
Token | to |
Token ID | 284 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
arcity | +0.294 |
Cage | +0.235 |
Hardware | +0.231 |
Sisters | +0.226 |
Drive | +0.220 |
Neg logit contributions | |
anka | -0.333 |
plom | -0.321 |
agen | -0.316 |
glitter | -0.312 |
interpol | -0.302 |
Token | sort |
Token ID | 3297 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
arcity | +1.330 |
Cage | +1.106 |
Hardware | +1.103 |
Sisters | +1.076 |
τ | +1.062 |
Neg logit contributions | |
anka | -0.982 |
plom | -0.959 |
glitter | -0.936 |
agen | -0.934 |
interpol | -0.901 |
Token | its |
Token ID | 663 |
Feature activation | -9.77e-02 |
Pos logit contributions | |
arcity | +1.583 |
Cage | +1.331 |
Hardware | +1.318 |
Sisters | +1.291 |
τ | +1.271 |
Neg logit contributions | |
anka | -1.081 |
plom | -1.042 |
agen | -1.016 |
glitter | -1.013 |
interpol | -0.970 |
Token | financial |
Token ID | 3176 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
anka | +0.396 |
glitter | +0.379 |
plom | +0.378 |
agen | +0.370 |
interpol | +0.359 |
Neg logit contributions | |
arcity | -0.535 |
Cage | -0.443 |
Hardware | -0.439 |
Sisters | -0.424 |
τ | -0.424 |
Token | position |
Token ID | 2292 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
arcity | +0.614 |
Cage | +0.498 |
Hardware | +0.495 |
Sisters | +0.480 |
τ | +0.475 |
Neg logit contributions | |
anka | -0.540 |
plom | -0.520 |
glitter | -0.514 |
agen | -0.511 |
interpol | -0.493 |
Token | . |
Token ID | 13 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
anka | +0.519 |
plom | +0.504 |
agen | +0.493 |
glitter | +0.489 |
interpol | +0.469 |
Neg logit contributions | |
arcity | -0.728 |
Cage | -0.607 |
Hardware | -0.603 |
Sisters | -0.588 |
τ | -0.580 |
Token | The |
Token ID | 383 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
arcity | +0.296 |
Cage | +0.239 |
Hardware | +0.238 |
Sisters | +0.230 |
τ | +0.227 |
Neg logit contributions | |
anka | -0.286 |
plom | -0.276 |
agen | -0.270 |
glitter | -0.269 |
interpol | -0.260 |
Token | company |
Token ID | 1664 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
arcity | +0.243 |
Cage | +0.203 |
Hardware | +0.201 |
τ | +0.194 |
Sisters | +0.194 |
Neg logit contributions | |
anka | -0.176 |
plom | -0.169 |
glitter | -0.167 |
agen | -0.166 |
interpol | -0.160 |
Token | has |
Token ID | 468 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
arcity | +1.432 |
Cage | +1.171 |
Hardware | +1.164 |
Sisters | +1.132 |
τ | +1.116 |
Neg logit contributions | |
anka | -1.263 |
plom | -1.227 |
agen | -1.201 |
glitter | -1.197 |
interpol | -1.157 |
Token | warned |
Token ID | 7728 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
arcity | +0.490 |
Hardware | +0.404 |
Cage | +0.404 |
Sisters | +0.391 |
τ | +0.388 |
Neg logit contributions | |
anka | -0.364 |
plom | -0.355 |
glitter | -0.350 |
agen | -0.343 |
interpol | -0.337 |
Token | it |
Token ID | 340 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
arcity | +0.785 |
Hardware | +0.646 |
Cage | +0.642 |
Sisters | +0.621 |
τ | +0.615 |
Neg logit contributions | |
anka | -0.645 |
plom | -0.623 |
glitter | -0.618 |
agen | -0.612 |
interpol | -0.591 |
Token |
|
Token ID | 198 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
arcity | +0.926 |
Cage | +0.720 |
Hardware | +0.708 |
Sisters | +0.692 |
τ | +0.676 |
Neg logit contributions | |
anka | -1.102 |
plom | -1.071 |
agen | -1.056 |
glitter | -1.049 |
interpol | -1.018 |
Token | S |
Token ID | 50 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
arcity | +0.393 |
Cage | +0.290 |
Hardware | +0.274 |
Sisters | +0.272 |
RI | +0.258 |
Neg logit contributions | |
anka | -0.640 |
plom | -0.624 |
agen | -0.618 |
glitter | -0.612 |
interpol | -0.597 |
Token | EC |
Token ID | 13182 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
arcity | +0.499 |
Hardware | +0.398 |
Cage | +0.396 |
Sisters | +0.380 |
Drive | +0.374 |
Neg logit contributions | |
anka | -0.586 |
plom | -0.572 |
agen | -0.562 |
glitter | -0.557 |
interpol | -0.541 |
Token | UN |
Token ID | 4944 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
arcity | +0.300 |
Cage | +0.250 |
Hardware | +0.247 |
Sisters | +0.241 |
τ | +0.236 |
Neg logit contributions | |
anka | -0.245 |
plom | -0.237 |
agen | -0.231 |
glitter | -0.230 |
interpol | -0.222 |
Token | DA |
Token ID | 5631 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
arcity | +0.519 |
Cage | +0.381 |
Hardware | +0.374 |
Sisters | +0.360 |
τ | +0.348 |
Neg logit contributions | |
anka | -0.966 |
plom | -0.946 |
agen | -0.929 |
glitter | -0.925 |
interpol | -0.903 |
Token | S |
Token ID | 311 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
arcity | +0.475 |
Cage | +0.378 |
Hardware | +0.377 |
Sisters | +0.364 |
Drive | +0.357 |
Neg logit contributions | |
anka | -0.498 |
plom | -0.485 |
agen | -0.476 |
glitter | -0.473 |
interpol | -0.460 |
Token | EC |
Token ID | 13182 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
arcity | +1.046 |
Hardware | +0.832 |
Cage | +0.830 |
Sisters | +0.802 |
Drive | +0.783 |
Neg logit contributions | |
anka | -1.101 |
plom | -1.071 |
agen | -1.055 |
glitter | -1.050 |
interpol | -1.019 |
Token | UN |
Token ID | 4944 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
arcity | +0.864 |
Cage | +0.709 |
Hardware | +0.701 |
Sisters | +0.682 |
τ | +0.666 |
Neg logit contributions | |
anka | -0.756 |
plom | -0.731 |
agen | -0.717 |
glitter | -0.710 |
interpol | -0.687 |
Token | D |
Token ID | 35 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
arcity | +0.096 |
Cage | +0.072 |
Hardware | +0.071 |
Sisters | +0.068 |
τ | +0.066 |
Neg logit contributions | |
anka | -0.137 |
plom | -0.133 |
agen | -0.132 |
glitter | -0.131 |
interpol | -0.127 |
Token | � |
Token ID | 127 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
anka | +0.631 |
plom | +0.607 |
agen | +0.602 |
glitter | +0.588 |
interpol | +0.569 |
Neg logit contributions | |
arcity | -0.810 |
Cage | -0.668 |
Hardware | -0.663 |
Sisters | -0.642 |
Drive | -0.632 |
Token | � |
Token ID | 228 |
Feature activation | -4.14e-03 |
Pos logit contributions | |
arcity | +0.759 |
Cage | +0.658 |
Hardware | +0.656 |
Sisters | +0.645 |
Drive | +0.633 |
Neg logit contributions | |
anka | -0.272 |
plom | -0.259 |
agen | -0.249 |
glitter | -0.247 |
interpol | -0.230 |
Token | Charlie |
Token ID | 11526 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
anka | +0.138 |
glitter | +0.135 |
plom | +0.133 |
agen | +0.131 |
interpol | +0.129 |
Neg logit contributions | |
arcity | -0.175 |
Hardware | -0.146 |
Cage | -0.140 |
Drive | -0.139 |
τ | -0.139 |
Token | and |
Token ID | 290 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
arcity | +0.640 |
Cage | +0.527 |
Hardware | +0.526 |
Sisters | +0.512 |
τ | +0.504 |
Neg logit contributions | |
anka | -0.521 |
plom | -0.505 |
agen | -0.494 |
glitter | -0.492 |
interpol | -0.473 |
Token | the |
Token ID | 262 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
arcity | +0.350 |
Hardware | +0.282 |
Cage | +0.279 |
τ | +0.272 |
Sisters | +0.270 |
Neg logit contributions | |
anka | -0.317 |
plom | -0.305 |
glitter | -0.303 |
agen | -0.302 |
interpol | -0.291 |
Token | Chocolate |
Token ID | 24777 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
anka | +0.091 |
plom | +0.083 |
glitter | +0.082 |
agen | +0.081 |
interpol | +0.072 |
Neg logit contributions | |
arcity | -0.348 |
Hardware | -0.304 |
Cage | -0.300 |
τ | -0.299 |
Drive | -0.295 |
Token | Factory |
Token ID | 19239 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
arcity | +0.052 |
Hardware | +0.041 |
Cage | +0.040 |
τ | +0.039 |
Drive | +0.038 |
Neg logit contributions | |
anka | -0.058 |
plom | -0.056 |
agen | -0.056 |
glitter | -0.056 |
interpol | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
anka | +0.210 |
plom | +0.203 |
glitter | +0.200 |
agen | +0.200 |
interpol | +0.193 |
Neg logit contributions | |
arcity | -0.264 |
Hardware | -0.218 |
Cage | -0.216 |
Sisters | -0.209 |
τ | -0.207 |
Token | from |
Token ID | 422 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
arcity | +0.994 |
Hardware | +0.825 |
Cage | +0.818 |
Sisters | +0.793 |
τ | +0.791 |
Neg logit contributions | |
anka | -0.716 |
plom | -0.692 |
glitter | -0.686 |
agen | -0.674 |
interpol | -0.657 |
Token | the |
Token ID | 262 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
anka | +0.184 |
glitter | +0.177 |
plom | +0.177 |
agen | +0.173 |
interpol | +0.169 |
Neg logit contributions | |
arcity | -0.214 |
Hardware | -0.177 |
Cage | -0.173 |
Sisters | -0.168 |
τ | -0.168 |
Token | poverty |
Token ID | 8098 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
arcity | +0.063 |
Hardware | +0.053 |
Cage | +0.052 |
τ | +0.051 |
Sisters | +0.050 |
Neg logit contributions | |
anka | -0.044 |
glitter | -0.043 |
plom | -0.042 |
agen | -0.042 |
interpol | -0.041 |
Token | - |
Token ID | 12 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
anka | +0.266 |
glitter | +0.256 |
plom | +0.253 |
agen | +0.251 |
interpol | +0.246 |
Neg logit contributions | |
arcity | -0.327 |
Hardware | -0.272 |
Cage | -0.267 |
Sisters | -0.258 |
τ | -0.258 |
Token | str |
Token ID | 2536 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
arcity | +0.807 |
Cage | +0.642 |
Hardware | +0.636 |
Sisters | +0.619 |
τ | +0.602 |
Neg logit contributions | |
anka | -0.929 |
plom | -0.904 |
agen | -0.890 |
glitter | -0.884 |
interpol | -0.853 |
Token | that |
Token ID | 326 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
arcity | +0.919 |
Cage | +0.755 |
Hardware | +0.755 |
τ | +0.721 |
Sisters | +0.716 |
Neg logit contributions | |
anka | -0.715 |
glitter | -0.684 |
plom | -0.683 |
agen | -0.668 |
interpol | -0.655 |
Token | you |
Token ID | 345 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
arcity | +0.095 |
Hardware | +0.079 |
Cage | +0.078 |
τ | +0.075 |
Drive | +0.074 |
Neg logit contributions | |
anka | -0.068 |
plom | -0.065 |
glitter | -0.065 |
agen | -0.064 |
interpol | -0.063 |
Token | still |
Token ID | 991 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
arcity | +0.713 |
Hardware | +0.593 |
Cage | +0.591 |
τ | +0.573 |
Sisters | +0.572 |
Neg logit contributions | |
anka | -0.512 |
plom | -0.494 |
glitter | -0.493 |
agen | -0.490 |
interpol | -0.470 |
Token | get |
Token ID | 651 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
arcity | +1.253 |
Cage | +1.042 |
Hardware | +1.038 |
Sisters | +1.010 |
τ | +0.999 |
Neg logit contributions | |
anka | -0.943 |
plom | -0.909 |
glitter | -0.897 |
agen | -0.895 |
interpol | -0.859 |
Token | awesome |
Token ID | 7427 |
Feature activation | -7.84e-03 |
Pos logit contributions | |
anka | +0.755 |
glitter | +0.738 |
plom | +0.717 |
interpol | +0.699 |
agen | +0.697 |
Neg logit contributions | |
arcity | -0.963 |
Hardware | -0.794 |
Cage | -0.786 |
τ | -0.761 |
xtap | -0.740 |
Token | stuff |
Token ID | 3404 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
anka | +0.033 |
glitter | +0.032 |
plom | +0.031 |
agen | +0.031 |
interpol | +0.030 |
Neg logit contributions | |
arcity | -0.042 |
Cage | -0.034 |
Hardware | -0.034 |
τ | -0.033 |
Sisters | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
arcity | +0.804 |
Cage | +0.664 |
Hardware | +0.664 |
Sisters | +0.642 |
τ | +0.636 |
Neg logit contributions | |
anka | -0.606 |
plom | -0.583 |
glitter | -0.573 |
agen | -0.567 |
interpol | -0.549 |
Token | cool |
Token ID | 3608 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
anka | +0.166 |
glitter | +0.161 |
plom | +0.157 |
agen | +0.154 |
interpol | +0.152 |
Neg logit contributions | |
arcity | -0.222 |
Cage | -0.184 |
Hardware | -0.183 |
τ | -0.178 |
Sisters | -0.175 |
Token | experiences |
Token ID | 6461 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
arcity | +0.148 |
Cage | +0.121 |
Hardware | +0.119 |
τ | +0.115 |
Sisters | +0.115 |
Neg logit contributions | |
anka | -0.134 |
glitter | -0.129 |
plom | -0.129 |
agen | -0.126 |
interpol | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -8.30e-02 |
Pos logit contributions | |
anka | +0.582 |
plom | +0.557 |
glitter | +0.552 |
agen | +0.543 |
interpol | +0.527 |
Neg logit contributions | |
arcity | -0.817 |
Hardware | -0.675 |
Cage | -0.674 |
Sisters | -0.651 |
τ | -0.650 |
Token |
|
Token ID | 198 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
anka | +0.387 |
plom | +0.369 |
glitter | +0.367 |
agen | +0.366 |
interpol | +0.354 |
Neg logit contributions | |
arcity | -0.402 |
Hardware | -0.315 |
Cage | -0.315 |
τ | -0.302 |
Sisters | -0.299 |
Token | в |
Token ID | 38857 |
Feature activation | +7.70e-03 |
Pos logit contributions | |
arcity | +0.738 |
Cage | +0.603 |
Hardware | +0.597 |
Sisters | +0.583 |
Drive | +0.570 |
Neg logit contributions | |
anka | -0.644 |
plom | -0.624 |
agen | -0.611 |
glitter | -0.611 |
interpol | -0.589 |
Token | а |
Token ID | 16142 |
Feature activation | +3.34e-03 |
Pos logit contributions | |
arcity | +0.043 |
Cage | +0.036 |
Hardware | +0.035 |
Sisters | +0.034 |
Drive | +0.034 |
Neg logit contributions | |
anka | -0.031 |
plom | -0.029 |
agen | -0.029 |
glitter | -0.029 |
interpol | -0.027 |
Token | т |
Token ID | 20375 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
arcity | +0.017 |
Cage | +0.014 |
Hardware | +0.014 |
Sisters | +0.014 |
Drive | +0.013 |
Neg logit contributions | |
anka | -0.014 |
plom | -0.014 |
agen | -0.014 |
glitter | -0.014 |
interpol | -0.013 |
Token | ь |
Token ID | 45367 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
anka | +0.406 |
plom | +0.392 |
agen | +0.387 |
glitter | +0.383 |
interpol | +0.369 |
Neg logit contributions | |
arcity | -0.509 |
Cage | -0.421 |
Hardware | -0.418 |
Sisters | -0.410 |
Drive | -0.399 |
Token | � |
Token ID | 12466 |
Feature activation | -8.56e-02 |
Pos logit contributions | |
arcity | +0.326 |
Cage | +0.262 |
Hardware | +0.259 |
Sisters | +0.253 |
Drive | +0.246 |
Neg logit contributions | |
anka | -0.339 |
plom | -0.329 |
agen | -0.324 |
glitter | -0.323 |
interpol | -0.313 |
Token | � |
Token ID | 116 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
anka | +0.302 |
plom | +0.288 |
glitter | +0.282 |
agen | +0.281 |
interpol | +0.266 |
Neg logit contributions | |
arcity | -0.508 |
Cage | -0.431 |
Hardware | -0.427 |
Sisters | -0.420 |
Drive | -0.411 |
Token | � |
Token ID | 141 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
anka | +1.969 |
plom | +1.906 |
agen | +1.901 |
glitter | +1.879 |
interpol | +1.823 |
Neg logit contributions | |
arcity | -1.740 |
Cage | -1.357 |
Hardware | -1.342 |
Sisters | -1.307 |
Drive | -1.269 |
Token | � |
Token ID | 227 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
arcity | +0.323 |
Cage | +0.282 |
Hardware | +0.280 |
Sisters | +0.276 |
Drive | +0.272 |
Neg logit contributions | |
anka | -0.105 |
plom | -0.097 |
agen | -0.095 |
glitter | -0.093 |
interpol | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
anka | +1.592 |
plom | +1.529 |
agen | +1.512 |
glitter | +1.512 |
interpol | +1.451 |
Neg logit contributions | |
arcity | -1.446 |
Cage | -1.140 |
Hardware | -1.122 |
Sisters | -1.099 |
Drive | -1.068 |
Token | � |
Token ID | 12466 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
anka | +0.508 |
plom | +0.491 |
agen | +0.484 |
glitter | +0.482 |
interpol | +0.468 |
Neg logit contributions | |
arcity | -0.427 |
Cage | -0.330 |
Hardware | -0.328 |
Sisters | -0.317 |
Drive | -0.309 |
Token | � |
Token ID | 248 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
arcity | +1.221 |
Cage | +1.031 |
Hardware | +1.023 |
Sisters | +1.005 |
Drive | +0.982 |
Neg logit contributions | |
anka | -0.773 |
plom | -0.744 |
glitter | -0.724 |
agen | -0.722 |
interpol | -0.687 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
arcity | +1.303 |
Cage | +1.081 |
Hardware | +1.067 |
Sisters | +1.048 |
τ | +1.026 |
Neg logit contributions | |
anka | -1.116 |
plom | -1.089 |
agen | -1.062 |
glitter | -1.056 |
interpol | -1.023 |
Token | Gordon |
Token ID | 11646 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
arcity | +0.991 |
Cage | +0.769 |
Hardware | +0.766 |
Sisters | +0.749 |
τ | +0.725 |
Neg logit contributions | |
anka | -1.185 |
plom | -1.146 |
agen | -1.133 |
glitter | -1.128 |
interpol | -1.094 |
Token | Reid |
Token ID | 16095 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
arcity | +0.958 |
Hardware | +0.647 |
Cage | +0.636 |
Sisters | +0.592 |
τ | +0.586 |
Neg logit contributions | |
anka | -2.139 |
plom | -2.091 |
agen | -2.067 |
glitter | -2.059 |
interpol | -2.012 |
Token | ( |
Token ID | 357 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
anka | +0.895 |
plom | +0.869 |
agen | +0.848 |
glitter | +0.844 |
interpol | +0.814 |
Neg logit contributions | |
arcity | -1.014 |
Cage | -0.834 |
Hardware | -0.826 |
Sisters | -0.807 |
τ | -0.791 |
Token | 108 |
Token ID | 15711 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
arcity | +0.767 |
Cage | +0.620 |
Sisters | +0.605 |
Hardware | +0.603 |
RI | +0.584 |
Neg logit contributions | |
anka | -0.651 |
plom | -0.625 |
agen | -0.623 |
glitter | -0.605 |
interpol | -0.584 |
Token | ) |
Token ID | 8 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
arcity | +0.370 |
Cage | +0.292 |
Hardware | +0.288 |
Sisters | +0.280 |
τ | +0.273 |
Neg logit contributions | |
anka | -0.408 |
plom | -0.393 |
agen | -0.388 |
glitter | -0.384 |
interpol | -0.372 |
Token | ( |
Token ID | 357 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
anka | +0.777 |
plom | +0.737 |
agen | +0.731 |
glitter | +0.723 |
interpol | +0.700 |
Neg logit contributions | |
arcity | -1.039 |
Hardware | -0.844 |
Cage | -0.841 |
Sisters | -0.815 |
τ | -0.807 |
Token | 20 |
Token ID | 1238 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
anka | +0.145 |
agen | +0.140 |
plom | +0.140 |
glitter | +0.134 |
interpol | +0.131 |
Neg logit contributions | |
arcity | -0.173 |
Cage | -0.138 |
Sisters | -0.135 |
Hardware | -0.134 |
RI | -0.130 |
Token | ) |
Token ID | 8 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
arcity | +0.269 |
Cage | +0.211 |
Hardware | +0.208 |
Sisters | +0.203 |
τ | +0.198 |
Neg logit contributions | |
anka | -0.292 |
plom | -0.281 |
agen | -0.278 |
glitter | -0.274 |
interpol | -0.266 |
Token |
|
Token ID | 198 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
arcity | +1.744 |
Cage | +1.372 |
Hardware | +1.331 |
Sisters | +1.299 |
τ | +1.266 |
Neg logit contributions | |
anka | -2.398 |
plom | -2.363 |
agen | -2.316 |
glitter | -2.299 |
interpol | -2.232 |
Token |
|
Token ID | 198 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
arcity | +0.812 |
Cage | +0.630 |
Hardware | +0.621 |
Sisters | +0.606 |
τ | +0.593 |
Neg logit contributions | |
anka | -0.948 |
plom | -0.919 |
agen | -0.908 |
glitter | -0.900 |
interpol | -0.873 |
Token | near |
Token ID | 1474 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
arcity | +0.458 |
Hardware | +0.380 |
Cage | +0.379 |
Sisters | +0.369 |
τ | +0.365 |
Neg logit contributions | |
anka | -0.352 |
plom | -0.337 |
glitter | -0.333 |
agen | -0.333 |
interpol | -0.316 |
Token | the |
Token ID | 262 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
anka | +0.306 |
plom | +0.296 |
glitter | +0.292 |
agen | +0.289 |
interpol | +0.279 |
Neg logit contributions | |
arcity | -0.329 |
Hardware | -0.270 |
Cage | -0.264 |
Sisters | -0.257 |
τ | -0.255 |
Token | portal |
Token ID | 17898 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
arcity | +0.415 |
Cage | +0.349 |
Hardware | +0.348 |
Sisters | +0.339 |
τ | +0.336 |
Neg logit contributions | |
anka | -0.258 |
plom | -0.249 |
glitter | -0.243 |
agen | -0.241 |
interpol | -0.231 |
Token | to |
Token ID | 284 |
Feature activation | +5.58e-02 |
Pos logit contributions | |
arcity | +1.602 |
Cage | +1.317 |
Hardware | +1.310 |
Sisters | +1.275 |
τ | +1.257 |
Neg logit contributions | |
anka | -1.357 |
plom | -1.315 |
agen | -1.287 |
glitter | -1.286 |
interpol | -1.234 |
Token | R |
Token ID | 371 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
arcity | +0.299 |
Hardware | +0.251 |
Cage | +0.246 |
τ | +0.239 |
Sisters | +0.238 |
Neg logit contributions | |
anka | -0.226 |
plom | -0.222 |
glitter | -0.218 |
agen | -0.214 |
interpol | -0.209 |
Token | al |
Token ID | 282 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
anka | +0.610 |
plom | +0.591 |
agen | +0.584 |
glitter | +0.566 |
interpol | +0.551 |
Neg logit contributions | |
arcity | -0.565 |
Cage | -0.464 |
Hardware | -0.458 |
Sisters | -0.448 |
τ | -0.434 |
Token | n |
Token ID | 77 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
anka | +1.551 |
plom | +1.449 |
agen | +1.447 |
glitter | +1.373 |
interpol | +1.313 |
Neg logit contributions | |
arcity | -2.448 |
Cage | -2.097 |
Hardware | -2.077 |
Sisters | -2.013 |
τ | -1.996 |
Token | oth |
Token ID | 849 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
anka | +1.041 |
plom | +0.994 |
agen | +0.985 |
glitter | +0.950 |
interpol | +0.909 |
Neg logit contributions | |
arcity | -1.051 |
Hardware | -0.867 |
Cage | -0.859 |
Sisters | -0.846 |
τ | -0.818 |
Token |
|
Token ID | 198 |
Feature activation | -9.69e-02 |
Pos logit contributions | |
anka | +0.977 |
plom | +0.946 |
agen | +0.933 |
glitter | +0.931 |
interpol | +0.888 |
Neg logit contributions | |
arcity | -1.095 |
Hardware | -0.900 |
Cage | -0.885 |
Sisters | -0.857 |
τ | -0.851 |
Token |
|
Token ID | 198 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
anka | +0.481 |
plom | +0.465 |
agen | +0.460 |
glitter | +0.456 |
interpol | +0.442 |
Neg logit contributions | |
arcity | -0.444 |
Cage | -0.344 |
Hardware | -0.339 |
Sisters | -0.332 |
τ | -0.325 |
Token | only |
Token ID | 8807 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
anka | +0.559 |
plom | +0.542 |
agen | +0.536 |
glitter | +0.522 |
interpol | +0.503 |
Neg logit contributions | |
arcity | -0.566 |
Cage | -0.454 |
Sisters | -0.441 |
Hardware | -0.441 |
RI | -0.426 |
Token | low |
Token ID | 1877 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
arcity | +0.431 |
Hardware | +0.356 |
Cage | +0.355 |
τ | +0.343 |
Sisters | +0.342 |
Neg logit contributions | |
anka | -0.326 |
plom | -0.312 |
glitter | -0.311 |
agen | -0.305 |
interpol | -0.296 |
Token | - |
Token ID | 12 |
Feature activation | -9.72e-03 |
Pos logit contributions | |
arcity | +1.058 |
Cage | +0.890 |
Hardware | +0.884 |
Sisters | +0.863 |
τ | +0.852 |
Neg logit contributions | |
anka | -0.733 |
plom | -0.707 |
agen | -0.691 |
glitter | -0.690 |
interpol | -0.659 |
Token | lying |
Token ID | 3157 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
anka | +0.042 |
plom | +0.041 |
agen | +0.040 |
glitter | +0.040 |
interpol | +0.038 |
Neg logit contributions | |
arcity | -0.049 |
Cage | -0.041 |
Hardware | -0.040 |
Sisters | -0.040 |
τ | -0.039 |
Token | areas |
Token ID | 3006 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
arcity | +0.955 |
Hardware | +0.783 |
Cage | +0.781 |
τ | +0.755 |
Sisters | +0.751 |
Neg logit contributions | |
anka | -0.766 |
plom | -0.733 |
glitter | -0.729 |
agen | -0.718 |
interpol | -0.698 |
Token | started |
Token ID | 2067 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
anka | +0.081 |
glitter | +0.078 |
plom | +0.078 |
agen | +0.077 |
interpol | +0.074 |
Neg logit contributions | |
arcity | -0.105 |
Cage | -0.088 |
Hardware | -0.087 |
Sisters | -0.085 |
Drive | -0.084 |
Token | flooding |
Token ID | 17448 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
anka | +0.278 |
glitter | +0.271 |
plom | +0.268 |
agen | +0.262 |
interpol | +0.257 |
Neg logit contributions | |
arcity | -0.351 |
Cage | -0.290 |
Hardware | -0.290 |
Sisters | -0.280 |
Drive | -0.277 |
Token | over |
Token ID | 625 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
anka | +0.439 |
plom | +0.416 |
agen | +0.415 |
glitter | +0.413 |
interpol | +0.393 |
Neg logit contributions | |
arcity | -0.603 |
Cage | -0.504 |
Hardware | -0.501 |
Sisters | -0.488 |
τ | -0.483 |
Token | the |
Token ID | 262 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
anka | +0.287 |
glitter | +0.273 |
plom | +0.272 |
agen | +0.268 |
interpol | +0.259 |
Neg logit contributions | |
arcity | -0.336 |
Cage | -0.279 |
Hardware | -0.275 |
Sisters | -0.268 |
τ | -0.265 |
Token | last |
Token ID | 938 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
arcity | +1.390 |
Cage | +1.141 |
Hardware | +1.136 |
Sisters | +1.100 |
τ | +1.095 |
Neg logit contributions | |
anka | -1.155 |
plom | -1.116 |
glitter | -1.105 |
agen | -1.091 |
interpol | -1.057 |
Token | few |
Token ID | 1178 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
anka | +0.924 |
glitter | +0.893 |
plom | +0.871 |
agen | +0.865 |
interpol | +0.846 |
Neg logit contributions | |
arcity | -1.117 |
Hardware | -0.904 |
Cage | -0.896 |
τ | -0.874 |
Sisters | -0.866 |
Token | weeks |
Token ID | 2745 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
arcity | +0.777 |
Cage | +0.615 |
Hardware | +0.609 |
Sisters | +0.584 |
τ | +0.580 |
Neg logit contributions | |
anka | -0.890 |
plom | -0.861 |
glitter | -0.854 |
agen | -0.844 |
interpol | -0.823 |
Token | , |
Token ID | 11 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
arcity | +1.809 |
Cage | +1.494 |
Hardware | +1.457 |
Sisters | +1.441 |
τ | +1.400 |
Neg logit contributions | |
anka | -1.730 |
plom | -1.688 |
agen | -1.641 |
glitter | -1.627 |
interpol | -1.587 |
Token | has |
Token ID | 468 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
arcity | +0.083 |
Hardware | +0.069 |
Cage | +0.068 |
Sisters | +0.066 |
τ | +0.065 |
Neg logit contributions | |
anka | -0.066 |
plom | -0.064 |
glitter | -0.063 |
agen | -0.062 |
interpol | -0.060 |
Token | not |
Token ID | 407 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
anka | +0.285 |
plom | +0.274 |
glitter | +0.273 |
agen | +0.266 |
interpol | +0.259 |
Neg logit contributions | |
arcity | -0.393 |
Hardware | -0.326 |
Cage | -0.324 |
Sisters | -0.314 |
τ | -0.311 |
Token | addressed |
Token ID | 9469 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
arcity | +0.333 |
Hardware | +0.277 |
Cage | +0.276 |
Sisters | +0.268 |
τ | +0.265 |
Neg logit contributions | |
anka | -0.253 |
plom | -0.244 |
glitter | -0.240 |
agen | -0.239 |
interpol | -0.229 |
Token | the |
Token ID | 262 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
arcity | +0.223 |
Hardware | +0.183 |
Cage | +0.182 |
Sisters | +0.177 |
τ | +0.176 |
Neg logit contributions | |
anka | -0.185 |
plom | -0.177 |
glitter | -0.174 |
agen | -0.173 |
interpol | -0.166 |
Token | rumors |
Token ID | 14923 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
anka | +0.785 |
plom | +0.744 |
glitter | +0.743 |
agen | +0.733 |
interpol | +0.699 |
Neg logit contributions | |
arcity | -1.164 |
Hardware | -0.970 |
Cage | -0.963 |
τ | -0.937 |
Sisters | -0.932 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
anka | +1.342 |
plom | +1.282 |
glitter | +1.266 |
agen | +1.264 |
interpol | +1.201 |
Neg logit contributions | |
arcity | -1.817 |
Hardware | -1.510 |
Cage | -1.489 |
Sisters | -1.444 |
τ | -1.442 |
Token | but |
Token ID | 475 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
arcity | +0.699 |
Hardware | +0.582 |
Cage | +0.582 |
Sisters | +0.566 |
τ | +0.560 |
Neg logit contributions | |
anka | -0.477 |
plom | -0.458 |
glitter | -0.450 |
agen | -0.446 |
interpol | -0.430 |
Token | brother |
Token ID | 3956 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
arcity | +0.317 |
Hardware | +0.263 |
Cage | +0.259 |
τ | +0.251 |
Sisters | +0.250 |
Neg logit contributions | |
anka | -0.239 |
plom | -0.228 |
glitter | -0.224 |
agen | -0.221 |
interpol | -0.213 |
Token | Eric |
Token ID | 7651 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
anka | +0.147 |
plom | +0.140 |
agen | +0.139 |
glitter | +0.138 |
interpol | +0.133 |
Neg logit contributions | |
arcity | -0.176 |
Hardware | -0.144 |
Cage | -0.143 |
Sisters | -0.138 |
τ | -0.138 |
Token | backed |
Token ID | 9763 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
arcity | +2.446 |
Cage | +2.027 |
Hardware | +2.008 |
Sisters | +1.964 |
τ | +1.928 |
Neg logit contributions | |
anka | -1.982 |
plom | -1.920 |
agen | -1.873 |
glitter | -1.864 |
interpol | -1.794 |
Token | were |
Token ID | 547 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
anka | +1.143 |
glitter | +1.103 |
plom | +1.101 |
agen | +1.081 |
interpol | +1.078 |
Neg logit contributions | |
arcity | -1.623 |
Hardware | -1.342 |
Cage | -1.335 |
τ | -1.308 |
Sisters | -1.307 |
Token | play |
Token ID | 711 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
anka | +1.047 |
plom | +1.018 |
glitter | +1.017 |
agen | +0.984 |
interpol | +0.978 |
Neg logit contributions | |
arcity | -1.508 |
Hardware | -1.242 |
Cage | -1.238 |
Sisters | -1.199 |
τ | -1.196 |
Token | - |
Token ID | 12 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
arcity | +0.893 |
Cage | +0.769 |
Hardware | +0.764 |
Sisters | +0.751 |
τ | +0.740 |
Neg logit contributions | |
anka | -0.417 |
plom | -0.399 |
agen | -0.386 |
glitter | -0.383 |
interpol | -0.362 |
Token | testing |
Token ID | 33407 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
anka | +0.081 |
plom | +0.077 |
agen | +0.075 |
glitter | +0.074 |
interpol | +0.070 |
Neg logit contributions | |
arcity | -0.183 |
Cage | -0.158 |
Hardware | -0.157 |
Sisters | -0.154 |
τ | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
anka | +2.283 |
glitter | +2.206 |
plom | +2.180 |
agen | +2.176 |
interpol | +2.128 |
Neg logit contributions | |
arcity | -3.498 |
Cage | -2.858 |
Hardware | -2.855 |
Sisters | -2.793 |
τ | -2.790 |
Token | One |
Token ID | 1881 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
anka | +1.486 |
plom | +1.419 |
glitter | +1.415 |
agen | +1.408 |
interpol | +1.370 |
Neg logit contributions | |
arcity | -1.713 |
Hardware | -1.352 |
Cage | -1.330 |
τ | -1.301 |
Sisters | -1.283 |
Token | was |
Token ID | 373 |
Feature activation | +7.41e-03 |
Pos logit contributions | |
anka | +2.581 |
glitter | +2.491 |
plom | +2.477 |
agen | +2.434 |
interpol | +2.378 |
Neg logit contributions | |
arcity | -3.479 |
Hardware | -2.846 |
Cage | -2.795 |
Sisters | -2.739 |
τ | -2.733 |
Token | a |
Token ID | 257 |
Feature activation | +8.50e-03 |
Pos logit contributions | |
arcity | +0.039 |
Hardware | +0.032 |
Cage | +0.031 |
Sisters | +0.030 |
τ | +0.030 |
Neg logit contributions | |
anka | -0.032 |
plom | -0.031 |
glitter | -0.030 |
agen | -0.030 |
interpol | -0.029 |
Token | Thought |
Token ID | 27522 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
arcity | +0.048 |
Hardware | +0.040 |
Cage | +0.040 |
Sisters | +0.039 |
τ | +0.039 |
Neg logit contributions | |
anka | -0.032 |
plom | -0.031 |
glitter | -0.031 |
agen | -0.030 |
interpol | -0.029 |
Token | st |
Token ID | 301 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
arcity | +0.782 |
Hardware | +0.634 |
Cage | +0.618 |
τ | +0.607 |
FACE | +0.600 |
Neg logit contributions | |
anka | -0.746 |
plom | -0.731 |
agen | -0.729 |
glitter | -0.714 |
interpol | -0.686 |
Token | eal |
Token ID | 2287 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
anka | +1.266 |
agen | +1.244 |
plom | +1.237 |
glitter | +1.194 |
interpol | +1.177 |
Neg logit contributions | |
arcity | -0.903 |
Hardware | -0.719 |
Cage | -0.702 |
Sisters | -0.689 |
τ | -0.672 |
Token | cell |
Token ID | 2685 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
anka | +0.932 |
plom | +0.889 |
glitter | +0.887 |
agen | +0.883 |
interpol | +0.871 |
Neg logit contributions | |
arcity | -1.127 |
Hardware | -0.926 |
Cage | -0.919 |
Sisters | -0.886 |
τ | -0.885 |
Token | , |
Token ID | 11 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
anka | +0.618 |
plom | +0.589 |
agen | +0.588 |
glitter | +0.587 |
interpol | +0.576 |
Neg logit contributions | |
arcity | -0.749 |
Hardware | -0.622 |
Cage | -0.619 |
Sisters | -0.596 |
Drive | -0.590 |
Token | as |
Token ID | 355 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
arcity | +0.804 |
Hardware | +0.664 |
Cage | +0.662 |
Sisters | +0.643 |
τ | +0.634 |
Neg logit contributions | |
anka | -0.577 |
plom | -0.551 |
glitter | -0.545 |
agen | -0.544 |
interpol | -0.537 |
Token | shown |
Token ID | 3402 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
anka | +0.660 |
glitter | +0.625 |
plom | +0.619 |
agen | +0.618 |
interpol | +0.618 |
Neg logit contributions | |
arcity | -0.835 |
Hardware | -0.675 |
Cage | -0.668 |
Sisters | -0.646 |
Drive | -0.644 |
Token | in |
Token ID | 287 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
arcity | +0.097 |
Cage | +0.080 |
Hardware | +0.080 |
Sisters | +0.078 |
Drive | +0.077 |
Neg logit contributions | |
anka | -0.066 |
plom | -0.063 |
agen | -0.062 |
glitter | -0.062 |
interpol | -0.060 |
Token | Figure |
Token ID | 11291 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
arcity | +0.246 |
Cage | +0.198 |
Hardware | +0.197 |
Sisters | +0.191 |
τ | +0.187 |
Neg logit contributions | |
anka | -0.238 |
plom | -0.231 |
glitter | -0.227 |
agen | -0.227 |
interpol | -0.221 |
Token | 3 |
Token ID | 513 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
anka | +2.657 |
glitter | +2.555 |
plom | +2.543 |
agen | +2.540 |
interpol | +2.495 |
Neg logit contributions | |
arcity | -3.012 |
Hardware | -2.424 |
Cage | -2.385 |
τ | -2.322 |
Sisters | -2.313 |
Token | . |
Token ID | 13 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
arcity | +1.297 |
Cage | +1.077 |
Hardware | +1.065 |
Sisters | +1.043 |
τ | +1.023 |
Neg logit contributions | |
anka | -1.053 |
plom | -1.022 |
agen | -0.996 |
glitter | -0.991 |
interpol | -0.954 |
Token |
|
Token ID | 198 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
arcity | +0.889 |
Cage | +0.707 |
Hardware | +0.700 |
Sisters | +0.680 |
τ | +0.667 |
Neg logit contributions | |
anka | -0.941 |
plom | -0.912 |
agen | -0.898 |
glitter | -0.893 |
interpol | -0.870 |
Token |
|
Token ID | 198 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
arcity | +0.937 |
Cage | +0.729 |
Hardware | +0.717 |
Sisters | +0.700 |
τ | +0.683 |
Neg logit contributions | |
anka | -1.137 |
plom | -1.107 |
agen | -1.090 |
glitter | -1.083 |
interpol | -1.051 |
Token | Figure |
Token ID | 11337 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
arcity | +0.936 |
Cage | +0.744 |
Sisters | +0.720 |
Hardware | +0.715 |
RI | +0.695 |
Neg logit contributions | |
anka | -1.005 |
plom | -0.976 |
agen | -0.963 |
glitter | -0.953 |
interpol | -0.926 |
Token | a |
Token ID | 257 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
anka | +0.656 |
plom | +0.634 |
glitter | +0.628 |
agen | +0.624 |
interpol | +0.596 |
Neg logit contributions | |
arcity | -0.747 |
Hardware | -0.609 |
Cage | -0.606 |
Drive | -0.582 |
τ | -0.581 |
Token | crater |
Token ID | 32593 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
anka | +0.058 |
glitter | +0.057 |
plom | +0.057 |
agen | +0.055 |
interpol | +0.053 |
Neg logit contributions | |
arcity | -0.101 |
Hardware | -0.085 |
Cage | -0.085 |
τ | -0.082 |
Sisters | -0.082 |
Token | on |
Token ID | 319 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
anka | +1.468 |
glitter | +1.435 |
agen | +1.424 |
plom | +1.394 |
interpol | +1.337 |
Neg logit contributions | |
arcity | -2.128 |
Hardware | -1.800 |
Cage | -1.787 |
τ | -1.713 |
Sisters | -1.708 |
Token | planet |
Token ID | 5440 |
Feature activation | -5.63e-02 |
Pos logit contributions | |
anka | +1.467 |
glitter | +1.429 |
plom | +1.419 |
agen | +1.408 |
interpol | +1.343 |
Neg logit contributions | |
arcity | -1.636 |
Hardware | -1.337 |
Cage | -1.317 |
τ | -1.261 |
Drive | -1.257 |
Token | Venus |
Token ID | 21094 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
anka | +0.245 |
plom | +0.236 |
glitter | +0.235 |
agen | +0.235 |
interpol | +0.222 |
Neg logit contributions | |
arcity | -0.287 |
Hardware | -0.236 |
Cage | -0.234 |
Sisters | -0.226 |
τ | -0.224 |
Token | in |
Token ID | 287 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
arcity | +0.253 |
Hardware | +0.210 |
Cage | +0.208 |
Sisters | +0.200 |
τ | +0.199 |
Neg logit contributions | |
anka | -0.199 |
glitter | -0.191 |
plom | -0.191 |
agen | -0.189 |
interpol | -0.180 |
Token | her |
Token ID | 607 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
anka | +2.182 |
glitter | +2.175 |
plom | +2.109 |
agen | +2.061 |
interpol | +1.985 |
Neg logit contributions | |
arcity | -3.530 |
Hardware | -3.010 |
Cage | -2.933 |
τ | -2.876 |
Drive | -2.868 |
Token | honor |
Token ID | 7522 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
anka | +0.476 |
glitter | +0.446 |
plom | +0.444 |
agen | +0.426 |
interpol | +0.398 |
Neg logit contributions | |
arcity | -1.744 |
Hardware | -1.520 |
Cage | -1.513 |
τ | -1.488 |
Sisters | -1.481 |
Token | .[ |
Token ID | 3693 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
anka | +1.260 |
plom | +1.215 |
agen | +1.186 |
glitter | +1.182 |
interpol | +1.134 |
Neg logit contributions | |
arcity | -1.689 |
Cage | -1.405 |
Hardware | -1.401 |
Sisters | -1.360 |
τ | -1.343 |
Token | 55 |
Token ID | 2816 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
anka | +0.065 |
plom | +0.063 |
agen | +0.061 |
glitter | +0.061 |
interpol | +0.059 |
Neg logit contributions | |
arcity | -0.084 |
Cage | -0.070 |
Hardware | -0.069 |
Sisters | -0.068 |
τ | -0.067 |
Token | ] |
Token ID | 60 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
anka | +0.731 |
plom | +0.710 |
agen | +0.692 |
glitter | +0.689 |
interpol | +0.663 |
Neg logit contributions | |
arcity | -0.894 |
Cage | -0.742 |
Hardware | -0.734 |
Sisters | -0.719 |
τ | -0.706 |
Token | of |
Token ID | 286 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
arcity | +1.166 |
Cage | +0.946 |
Hardware | +0.929 |
Sisters | +0.911 |
τ | +0.890 |
Neg logit contributions | |
anka | -1.211 |
plom | -1.176 |
agen | -1.149 |
glitter | -1.141 |
interpol | -1.105 |
Token | which |
Token ID | 543 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
anka | +0.154 |
plom | +0.150 |
agen | +0.146 |
glitter | +0.146 |
interpol | +0.140 |
Neg logit contributions | |
arcity | -0.181 |
Cage | -0.150 |
Hardware | -0.148 |
Sisters | -0.145 |
τ | -0.142 |
Token | way |
Token ID | 835 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
anka | +1.384 |
plom | +1.340 |
glitter | +1.329 |
agen | +1.304 |
interpol | +1.276 |
Neg logit contributions | |
arcity | -1.749 |
Hardware | -1.436 |
Cage | -1.408 |
τ | -1.378 |
Sisters | -1.376 |
Token | they |
Token ID | 484 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
anka | +0.319 |
plom | +0.307 |
glitter | +0.301 |
agen | +0.301 |
interpol | +0.289 |
Neg logit contributions | |
arcity | -0.391 |
Cage | -0.321 |
Hardware | -0.320 |
Sisters | -0.310 |
τ | -0.308 |
Token | were |
Token ID | 547 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
anka | +1.457 |
glitter | +1.445 |
plom | +1.420 |
interpol | +1.403 |
agen | +1.392 |
Neg logit contributions | |
arcity | -2.073 |
Hardware | -1.739 |
Cage | -1.690 |
τ | -1.677 |
Sisters | -1.650 |
Token | going |
Token ID | 1016 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
anka | +1.376 |
plom | +1.348 |
glitter | +1.341 |
agen | +1.308 |
interpol | +1.306 |
Neg logit contributions | |
arcity | -1.762 |
Hardware | -1.438 |
Cage | -1.417 |
τ | -1.371 |
Sisters | -1.370 |
Token | . |
Token ID | 13 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
anka | +2.883 |
plom | +2.761 |
glitter | +2.717 |
agen | +2.706 |
interpol | +2.628 |
Neg logit contributions | |
arcity | -3.684 |
Hardware | -3.027 |
Cage | -2.998 |
τ | -2.924 |
Sisters | -2.913 |
Token | As |
Token ID | 1081 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
anka | +1.200 |
plom | +1.154 |
agen | +1.143 |
glitter | +1.137 |
interpol | +1.099 |
Neg logit contributions | |
arcity | -1.361 |
Hardware | -1.069 |
Cage | -1.057 |
τ | -1.024 |
Sisters | -1.020 |
Token | part |
Token ID | 636 |
Feature activation | +8.69e-02 |
Pos logit contributions | |
anka | +0.180 |
plom | +0.173 |
glitter | +0.170 |
agen | +0.169 |
interpol | +0.162 |
Neg logit contributions | |
arcity | -0.222 |
Hardware | -0.182 |
Cage | -0.181 |
Sisters | -0.176 |
τ | -0.174 |
Token | of |
Token ID | 286 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
arcity | +0.432 |
Cage | +0.352 |
Hardware | +0.347 |
Sisters | +0.340 |
τ | +0.335 |
Neg logit contributions | |
anka | -0.394 |
plom | -0.380 |
agen | -0.373 |
glitter | -0.372 |
interpol | -0.360 |
Token | the |
Token ID | 262 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
anka | +1.568 |
plom | +1.520 |
glitter | +1.507 |
agen | +1.474 |
interpol | +1.454 |
Neg logit contributions | |
arcity | -1.994 |
Hardware | -1.638 |
Cage | -1.605 |
τ | -1.575 |
Sisters | -1.561 |
Token | do |
Token ID | 466 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
anka | +0.722 |
plom | +0.716 |
glitter | +0.714 |
agen | +0.704 |
interpol | +0.693 |
Neg logit contributions | |
arcity | -0.971 |
Hardware | -0.805 |
Cage | -0.798 |
τ | -0.772 |
Sisters | -0.770 |
Token | two |
Token ID | 734 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
anka | +0.174 |
plom | +0.169 |
glitter | +0.167 |
agen | +0.166 |
interpol | +0.161 |
Neg logit contributions | |
arcity | -0.235 |
Hardware | -0.195 |
Cage | -0.194 |
Sisters | -0.188 |
τ | -0.188 |
Token | main |
Token ID | 1388 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
anka | +0.546 |
glitter | +0.531 |
plom | +0.530 |
agen | +0.522 |
interpol | +0.517 |
Neg logit contributions | |
arcity | -0.464 |
Hardware | -0.360 |
Cage | -0.358 |
τ | -0.343 |
Sisters | -0.335 |
Token | things |
Token ID | 1243 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
arcity | +0.153 |
Cage | +0.115 |
Hardware | +0.114 |
τ | +0.109 |
Sisters | +0.107 |
Neg logit contributions | |
anka | -0.213 |
plom | -0.208 |
glitter | -0.206 |
agen | -0.203 |
interpol | -0.202 |
Token | . |
Token ID | 13 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
anka | +1.486 |
plom | +1.447 |
agen | +1.424 |
glitter | +1.414 |
interpol | +1.369 |
Neg logit contributions | |
arcity | -1.477 |
Hardware | -1.184 |
Cage | -1.183 |
τ | -1.134 |
Sisters | -1.131 |
Token | One |
Token ID | 1881 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
anka | +2.212 |
plom | +2.149 |
agen | +2.137 |
glitter | +2.119 |
interpol | +2.042 |
Neg logit contributions | |
arcity | -2.509 |
Hardware | -2.014 |
Cage | -1.950 |
τ | -1.939 |
Sisters | -1.883 |
Token | is |
Token ID | 318 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
anka | +2.510 |
glitter | +2.450 |
plom | +2.448 |
agen | +2.414 |
interpol | +2.347 |
Neg logit contributions | |
arcity | -4.115 |
Hardware | -3.390 |
Cage | -3.330 |
τ | -3.266 |
Sisters | -3.258 |
Token | to |
Token ID | 284 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
anka | +0.566 |
plom | +0.556 |
glitter | +0.550 |
agen | +0.540 |
interpol | +0.530 |
Neg logit contributions | |
arcity | -0.677 |
Hardware | -0.552 |
Cage | -0.546 |
τ | -0.530 |
Sisters | -0.529 |
Token | create |
Token ID | 2251 |
Feature activation | +4.88e-03 |
Pos logit contributions | |
anka | +0.071 |
plom | +0.070 |
glitter | +0.069 |
agen | +0.068 |
interpol | +0.067 |
Neg logit contributions | |
arcity | -0.099 |
Hardware | -0.082 |
Cage | -0.081 |
τ | -0.079 |
Sisters | -0.079 |
Token | a |
Token ID | 257 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
arcity | +0.026 |
Hardware | +0.022 |
Cage | +0.021 |
τ | +0.021 |
Sisters | +0.021 |
Neg logit contributions | |
anka | -0.020 |
plom | -0.019 |
glitter | -0.019 |
agen | -0.019 |
interpol | -0.019 |
Token | shared |
Token ID | 4888 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
anka | +0.298 |
plom | +0.291 |
glitter | +0.289 |
agen | +0.282 |
interpol | +0.275 |
Neg logit contributions | |
arcity | -0.439 |
Hardware | -0.366 |
Cage | -0.362 |
τ | -0.353 |
Sisters | -0.351 |